INTRODUCTION

TO THE

STUDY OF HISTORY

BY

CH. V. LANGLOIS & CH. SEIGNOBOS

OF THE SORBONNE

AUTHORIZED TRANSLATION BY G. G. BERRY

WITH A PREFACE BY F. YORK POWELL

NEW YORK

HENRY HOLT AND COMPANY

1904

TO THE READER

It is a pleasure to recommend this useful and well-written little bookto English readers. It will both interest and help. There are, forinstance, a few pages devoted to the question of evidence that will bean aid to every one desirous of getting at the truth respecting anyseries of facts, as well as to the student of history. No one can readit without finding out that to the historian history is not merely apretty but rather difficult branch of literature, and that a historybook is not necessarily good if it appears to the literary critic'readable and interesting, ' nor bad because it seems to him 'hard orheavy reading. ' The literary critic, in fact, is beginning to find outthat he reads a history as he might read a treatise on mathematics orlinguistics, at his peril, and that he is no judge of its value or lackof value. Only the expert can judge that. It will probably surprise somepeople to find that in the opinion of our authors (who agree with Mr. Morse Stephens and with the majority of scholars here) the formation andexpression of ethical judgments, the approval or condemnation of CaiusJulius Cæsar, or of Cæsar Borgia, is not a thing within the historian'sprovince. His business is to find out what can be known about thecharacters and situations with which he is engaged, to put what he canascertain before his readers in a clear form, and lastly to consider andattempt to ascertain what scientific use can be made of these facts hehas ascertained. Ethic on its didactic side is outside his businessaltogether. In fact MM. Langlois and Seignobos write for those "whopropose to deal with documents [especially written documents] with aview to preparing or accomplishing historic work in a scientific way. "They have the temerity to view history as a scientific pursuit, and theyare endeavouring to explain to the student who intends to pursue thisbranch of anthropologic science the best and safest methods ofobservation open to him, hence they modestly term their little book "anessay on the method of historic sciences. " They are bold enough to lookforward to a day, as not far distant, when a sensible or honest man willno more dare to write history unscientifically than he would to-day bewilling to waste his time and that of others on observing the heavensunscientifically, and registering as trustworthy his unchecked anduntimed observations. 

Whether we like it or not, history has got to be scientifically studied, and it is not a question of style but of accuracy, of fulness ofobservation, and correctness of reasoning, that is before the student. Huxley and Darwin and Clifford have shown that a book may be goodscience and yet good reading. Truth has not always been found repulsivealthough she was not bedizened with rhetorical adornments; indeed, thevery pursuit of her has long been recognised as arduous but extremelyfascinating. _Toute trouvaille_, as our authors aptly remark, _procureune jouissance_. 

It will be a positive gain to have the road cleared of a mass ofrubbish, that has hindered the advance of knowledge. History must beworked at in a scientific spirit, as biology or chemistry is worked at. As M. Seignobos says, "On ne s'arrête plus guère aujourd'hui à discuter, sous sa forme théologique la théorie de la Providence dans l'Histoire. Mais la tendence à expliquer les faits historiques par les causestranscendantes persiste dans des théories plus modernes où lametaphysique se déguise sous des formes scientifiques. " We shouldcertainly get rid in time of those curious Hegelianisms "under which inlay disguise lurks the old theologic theory of final causes"; or thepseudo-patriotic supposition of the "historic mission (Beruf) attributedto certain people or persons. " The study of historic facts does not evenmake for the popular newspaper theory of the continuous and necessaryprogress of humanity, it shows only "partial and intermittent advances, and gives us no reason to attribute them to a permanent cause inherentin collective humanity rather than to a series of local accidents. " Butthe historian's path is still like that of Bunyan's hero, bordered bypitfalls and haunted by hobgoblins, though certain of his giantadversaries are crippled and one or two slain. He has also his ownfaults to master, or at least to check, as MM. Langlois and Seignobosnot infrequently hint, _e. G. _ "Nearly all beginners have a vexatioustendency to go off into superfluous digressions, heaping up reflexionand information that have no bearing on the main subject. They willrecognise, if they think over it, that the causes of this leaning arebad taste, a kind of naïve vanity, sometimes a disordered mind. " Again:"The faults of historic works intended for the general public . .. Arethe results of the insufficient preparation of the bad literary trainingof the popularisers. " What an admirable criticism there is too of thatpeculiarly German shortcoming (one not, however, unknown elsewhere), which results in men "whose learning is ample, whose monographs destinedfor scholars are highly praiseworthy, showing themselves capable, whenthey write for the public, of sinning heavily against scientificmethods, " so that, in their determination to stir their public, "theywho are so scrupulous and particular when it is a question of dealingwith minutiæ, abandon themselves like the mass of mankind to theirnatural inclinations when they come to set forth general questions. Theytake sides, they blame, they praise, they colour, they embellish, theyallow themselves to take account of personal, patriotic, ethical, ormetaphysical considerations. Above all, they apply themselves with whattalent has fallen to their lot to the task of creating a work of art, and, so applying themselves, those of them who lack talent becomeridiculous, and the talent of those who possess it is spoilt by theiranxiety for effect. "

On the other hand, while the student is rejoicing at the smart rapsbestowed upon the Teutonic offender, he is warned against the error ofthinking that "provided he can make himself understood, the historianhas the right to use a faulty, low, careless, or clogged style. .. . Seeing the extreme complexity of the phenomena he must endeavour todescribe, he has not the privilege of writing badly. But he ought_always_ to write well, and not to bedizen his prose with extra fineryonce a week. "

Of course much that is said in this book has been said before, but I donot know any book wherein the student of history will find such anorganised collection of practical and helpful instructions. There areseveral points on which one is unable to find oneself in agreement withMM. Langlois and Seignobos, but these occur mainly where they aredealing with theory; as far as practical work goes, one finds oneself inalmost perfect concurrence with them. That they know little of the wayin which history is taught and studied in England or Canada or theUnited States is not at all an hindrance to the use of their book. Thestudent may enjoy the pleasure of making his own examples out of Englishbooks to the rules they lay down. He may compare their cautions againstfalse reasoning and instances of fallacy with those set forth in thatexcellent and concise essay of Bentham's, which is apparently unknown tothem. He will not fail to see that we in England have much to learn inthis subject of history from the French. The French archives are not sofine as ours, but they take care to preserve their local and provincialdocuments, as well as their national and central records; they givetheir archivists a regular training, they calendar and make accessibleall that time and fate have spared of pre-revolutionary documents. Wehave not got farther than the provision of a fine central Record Officefurnished with very inadequate means for calendaring the masses ofdocuments already stored and monthly accumulating there, though we havelately set up at Oxford, Cambridge, and London the regular courses ofpalæography, diplomatic, and bibliography, that constitute thepreliminary training of the archivist or historical researcher. We wantmore: we must have county archives, kept by trained archivists. We musthave more trained archivists at the disposal of the Deputy Keeper of theRolls, we must have such means as the _Bibliothèque de l'École desChartes_ for full reports of special and minute investigations anddiscoveries, for hand-lists and the like, before we can be considered asdoing as much for history as the heavily taxed French nation doescheerfully, and with a sound confidence that the money it spends wiselyin science is in the truest sense money saved. 

For those interested in the teaching of history, this book is one of themost suggestive helps that has yet appeared. With a blackboard, a text(such as are now cheap), or a text-book (such as Stubbs or Prothero orGardiner), an atlas, and access to a decent public library and anaverage local museum, the teacher who has mastered its intent shouldnever be at a loss for an interesting catechetical lecture or expositionto a class, whether of adults or of younger folk. 

Not the least practical part of the work of MM. Langlois and Seignoboshas been the consideration they have given to such every-day issues asthe teacher is constantly called upon to face. History cannot safely beneglected in schools, though it is by no means necessary that theUniversities should turn out large bodies of trained historians. It ispossible indeed that the serious study of history might gain were therefewer external inducements at the Universities to lead to the popularityof the History Schools. But in this very popularity there lies a greatopportunity for concerted efforts, not only to better the processes ofstudy, but also to clear off the vast arrears of classification andexamination of the erroneous historic material at our disposition inthis country. 

The historian has been (as our authors hint) too much the ally of thepolitician; he has used his knowledge as material for preachingdemocracy in the United States, absolutism in Prussia, Orleanistopposition in France, and so on (English readers will easily recallexamples from their own countrymen's work): in the century to come hewill have to ally himself with the students of physical science, withwhose methods his own have so much in common. It is not patriotism, norreligion, nor art, but the attainment of truth that is and must be thehistorian's single aim. 

But it is also to be borne in mind that history is an excellentinstrument of culture, for, as our authors point out, "the practice andmethod of historic investigation is a pursuit extremely healthful forthe mind, freeing it from the disease of credulity, " and fortifying itin other ways as a discipline, though precisely how to best use historyfor this purpose is still in some ways uncertain, and after all it is amatter which concerns Pædagogic and Ethic more than the student ofhistory, though it is plain that MM. Langlois and Seignobos have notneglected to consider it. 

One can hardly help thinking, too, that, in schools and places where theyoung are trained, something might be gained by treating such books asPlutarch's Lives not as history (for which they were never intended) butas text-books of ethic, as examples of conduct, public or private. Thehistorian very properly furnishes the ethical student with material, though it is not right to reckon the ethical student's judgment upon thehistorian's facts as history in any sense. It is not an historian'squestion, for instance, whether Napoleon was right or wrong in hisconduct at Jaffa, or Nelson in his behaviour at Naples; that is a matterfor the student of ethic or the religious dogmatist to decide: all thatthe historian has to do is to get what conclusion he can out of theconflict of evidence, and to decide whether Napoleon or Nelson actuallydid that of which their enemies accused them, or, if he cannot arrive atfact, to state probability, and the reasons that incline him to lean tothe affirmative or negative. 

As to the possibility of a "philosophy of history, " a real one, not themockeries that have long been discredited by scientific students, thereader will find some pregnant remarks here in the epilogue and thechapters that precede it. There is an absence of unreasonable optimismin our authors' views. "It is probable that hereditary differences havecontributed to determine events; so that in part historic evolution isproduced by physiological and anthropologic causes. But historyfurnishes no trustworthy process by which it may be possible todetermine the action of those hereditary differences between man andman, " _i. E. _ she starts with races 'endowed' each with peculiaritiesthat make them 'disposed to act' somewhat differently under similarpressure. "History is only able to grasp the conditions of theirexistence. " And what M. Seignobos calls the final problem--_Is evolutionproduced merely by changed conditions?_--must according to him remaininsoluble by the legitimate processes of history. The student may acceptor reject this view as his notions of evidence prompt him to do. M. Seignobos has at all events laid down a basis for discussion insufficiently clear terms. 

As to the composition of the joint work we are told that M. Seignoboshas been especially concerned with the chapters that touch theory, andM. Langlois with those that deal with practice. Both authors havealready proved their competence--M. Seignobos' labours on Modern Historyhave been widely appreciated, while M. Langlois' "Hand-book of HistoricBibliography" is already a standard text-book, and bids fair to remainso. We are grateful to both of them for the pains they have taken to beclear and definite, and for their determination to shirk none of thedifficulties that have met them. They have produced a hand-book thatstudents will use and value in proportion to their use of it, a bookthat will save much muddle of thought and much loss of time, a bookwritten in the right spirit to inspire its readers. We are not bound toagree with all M. Seignobos' dogmas, and can hardly accept, forinstance, M. Langlois' apology for the brutal methods of controversythat are an evil legacy from the theologian and the grammarian, and areapt to darken truth and to cripple the powers of those who engage inthem. For though it is possible that the secondary effect of thesebarbarous scuffles may sometimes have been salutary in deterringimpostors from 'taking up' history, I am not aware of any positiveexamples to justify this opinion. There is this, however, to be said, that fully conscious of their own fallibility, M. Langlois and hisexcellent collaborator have supplied in their canons of criticism andmaxims the best corrections of any mistakes into which they may havefallen by the way. Is not the House of Fame, as the poet tells us, amore wonderful and quaintly wrought habitation than _Domus Dedali_itself? And may not honest historians be pardoned if they are sometimesconfused for a brief moment by the never-ending noise and marvellousmotion of that deceptive mint and treasury, and fatigued by thecontinual trial and examination of the material that issues therefrom?The student will, at least, learn from MM. Langlois and Seignobos tohave no mercy on his own shortcomings, to spare no pains, to grudge noexpenditure of time or energy in the investigation of a carefully chosenand important historical problem, to aim at doing the bit of work inhand so thoroughly that it will not need to be done again. 

It would be unjust to omit here to mention Dr. Bernheim's "Exposition ofHistoric Method, " or _Lehrbuch der historischen Methode_, so justlypraised and used by our authors, but I believe that as an introductionto the subject, intended for the use of English or North Americanstudents, this little volume will be found the handier and morepractical work. Of its value to English workers I can speak fromexperience, and I know many teachers to whom it will be welcome in itspresent form. 

It would have been easy to 'adapt' this book by altering its examples, by modifying its excellent plan, by cutting here and carving there tothe supposed convenience of an imaginary public, but the better part hasbeen chosen of giving English readers this manual precisely as itappeared in French. And surely one would rather read what M. Langlois, an experienced teacher and a tried scholar, thought on a moot point, than be presented with the views of some English 'adaptor' who had readhis book, as to what he would have said had he been an Englishmanlecturing to English students. That the present translator has takenmuch pains to faithfully report his authors, I know (though I have notcompared English and French throughout every page), so that I cancommend his honest work to the reader as I have already commended theexcellent matter that he has been concerned in preparing for a widerpublic than the French original could command. 

F. YORK POWELL. 

ORIEL COLLEGE, OXFORD, _July 1898_. 

CONTENTS

PAGE

TO THE READER v

AUTHORS' PREFACE

What this work is _not_ meant to be--Works on the Philosophyof History 1

What it _is_ meant to be 2

Existing works on Historical Methods--Droysen, Freeman, Daunou, &c. 3

Reasons why the study of method is useful 7

Bernheim's _Lehrbuch_--In what way it leaves room for anotherbook 10

Need of warning to students 11

The general public 13

Distribution of the work between the two authors 13

BOOK I

PRELIMINARY STUDIES

CHAPTER I

THE SEARCH FOR DOCUMENTS

Documents: their nature, use, necessity 17

Utility of _Heuristic_, or the art of discovering documents 18

The difficulties of Heuristic--Ancient times--H. H. Bancroft--Stateof things at the Renaissance 19

Growth of libraries--Collectors--Effects of revolutionaryconfiscation in promoting the concentration and theaccessibility of documents 20

Possible future progress--Need for the cataloguing and indexingof documents 27

Students and bibliographical knowledge--Effect of presentconditions in deterring men from historical work 32

The remedies--Official cataloguing of libraries--Activity oflearned societies--of governments 34

Different kinds of bibliographical works needed by students 37

Different degrees of difficulty of Heuristic in different parts ofHistory--to be kept in view when choosing a subject ofresearch 38

CHAPTER II

"AUXILIARY SCIENCES"

Documents are raw material, and need a preliminary elaboration 42

Obsolete views on the historian's apprenticeship--Mably, Daunou 43

Commonplace and exaggeration on this subject--Freeman--Variousfutilities 45

The scientific conception of the historian'sapprenticeship--Palæography--Epigraphy--Philology--Diplomatic 48

History of Literature--Archæology 51

Criticism of phrase "auxiliary sciences"--The subjects not all_sciences_--None of them auxiliary to the _whole_ of History 52

This scientific conception is of recent growth--The École desChartes--Modern manuals of Palæography, Epigraphy, &c. --List of the chief of them 55

BOOK II

_ANALYTICAL OPERATIONS_

CHAPTER I

GENERAL CONDITIONS OF HISTORICAL KNOWLEDGE

Direct and indirect knowledge of facts 63

History not a science of direct observation--Its data obtainedby chains of reasoning 64

Twofold division of Historical Criticism: _External_, investigatingthe transmission and origin of documents and thestatements in them; _Internal_, dealing with the contentof the statements and their probability 66

Complexity of Historical Criticism 67

Necessity of Criticism--The human mind naturally uncritical 68

SECTION I. --EXTERNAL CRITICISM

CHAPTER II

TEXTUAL CRITICISM

Errors in the reproduction of documents: their frequencyunder the most favourable conditions--Mistakes ofcopyists--"Sound" and "corrupt" texts 71

Necessity of emendation--The method subject to fixed rules 73

Methods of textual criticism: (_a_) original preserved; (_b_) asingle copy preserved, conjectural emendation; (_c_) severalcopies preserved, comparison of errors, families of manuscripts 75

Different degrees of difficulty of textual criticism: its resultsnegative--The "emendation game"--What still remainsto be done 83

CHAPTER III

CRITICAL INVESTIGATION OF AUTHORSHIP PAGENatural tendency to accept indications of authorship--Examplesof false attributions--Necessity of verification--Applicationof internal criticism 87

Interpolations and continuations--Evidence of style 92

Plagiarism and borrowings by authors from each other--Thefiliation of statements--The investigation of sources 93

Importance of investigations of authorship--The extreme ofdistrust to be avoided--Criticism only a means to an end 98

CHAPTER IV

CRITICAL CLASSIFICATION OF SOURCES

Importance of classification--The first impulse wrong--Thenote-book system not the best--Nor the ledger-system--Northe "system" of trusting the memory 101

The system of slips the best--Its drawbacks--Means ofobviating them--The advantage of good "private librarianship" 103

Methods of work vary according to the object aimed at--Thecompiling of _Regesta_ or of a _Corpus_--Classification bytime, place, species, and form 105

Chronological arrangement to be used when possible--Geographicalarrangement best for inscriptions--When thesefail, alphabetical order of "incipit"--Logical order usefulfor some special purposes--Not for a _Corpus_ or for _Regesta_ 107

CHAPTER V

CRITICAL SCHOLARSHIP AND SCHOLARS

Different opinions on the importance and dignity of externalcriticism--It is justified by its necessity--But is onlypreliminary to the higher part of historical work 112

Distinction between "historians" and "critical scholars" [Fr. "_érudite_"]--Expediency, within limits, of the division oflabour in this respect--The exceptional skill acquired byspecialists--Difference of work the corollary of differenceof natural aptitudes 115

The natural aptitudes required for external criticism--Fondnessfor the work, which is distasteful to the creativegenius--The puzzle-solving instinct--Accuracy and itsopposite--"Froude's Disease"--Patience, order, perseverance 121

The mental defects produced by devotion to externalcriticism--Its paralysing effect on theover-scrupulous--Hypercriticism--Dilettantism 128

The "organisation of scientific labour" 135

The harshness of judgment attributed to scholars, not alwaysrightly--Much of it a proper jealousy for historic truth--Badwork nowadays soon detected 136

SECTION II. --INTERNAL CRITICISM

CHAPTER VI

INTERPRETATIVE CRITICISM (HERMENEUTIC)

Internal criticism deals with the mental operations whichbegin with the observation of a fact and end with thewriting of words in a document--It is divided into twostages: the first concerned with what the author meant, the second with the value of his statements 141

Necessity of separating the two operations--Danger of readingopinions into a text 143

The analysis of documents--The method of slips--Completenessnecessary 145

Necessity of linguistic study--General knowledge of a languagenot enough--Particular variety of a language as used at agiven time, in a given country, by a given author--Therule of context 146

Different degrees of difficulty in interpretation 149

Oblique senses: allegory, metaphor, &c. --How to detect them--Formertendency to find symbolism everywhere--Moderntendency to find allusion everywhere 151

Results of interpretation--Subjective inquiries 153

CHAPTER VII

THE NEGATIVE INTERNAL CRITICISM OF THE GOOD FAITHAND ACCURACY OF AUTHORS

Natural tendency to trust documents--Criticism originally dueto contradictions--The rule of methodical doubt--Defectivemodes of criticism 155

Documents to be analysed, and the irreducible elementscriticised separately 159

The "accent of sincerity"--No trust to be placed in impressionsproduced by the form of statements 161

Criticism examines the conditions affecting (1) the compositionof the document as a whole; (2) the making of each particularstatement--In both cases using a previously madelist of possible reasons for distrust or confidence 162

Reasons for doubting good faith: (1) the author's interest;(2) the force of circumstances, official reports; (3) sympathyand antipathy; (4) vanity; (5) deference to publicopinion; (6) literary distortion 166

Reasons for doubting accuracy: (1) the author a bad observer, hallucinations, illusions, prejudices; (2) the author notwell situated for observing; (3) negligence and indifference;(4) fact not of nature to be directly observed 172

Cases where the author is not the original observer of thefact--Tradition, written and oral--Legend--Anecdotes--Anonymousstatements 177

Special reasons without which anonymous statements are notto be accepted: (1) falsehood improbable because (_a_) thefact is opposed to interest or vanity of author, (_b_) the factwas generally known, (_c_) the fact was indifferent to theauthor; (2) error improbable because the fact was too bigto mistake; (3) the fact seemed improbable or unintelligibleto the author 185

How critical operations are shortened in practice 189

CHAPTER VIII

THE DETERMINATION OF PARTICULAR FACTS

The conceptions of authors, whether well or ill founded, arethe subject-matter of certain studies--They necessarilycontain elements of truth, which, under certain restrictions, may sometimes be inferred from them 191

The statements of authors, taken singly, do not rise aboveprobability--The only _sure_ results of criticism are _negative_--Toestablish facts it is necessary to compare differentstatements 194

Contradictions between statements, real and apparent 198

Agreement of statements--Necessity of proving them to beindependent--Perfect agreement not so conclusive asoccasional coincidence--Cases where different observationsof the same fact are not independent--General factsthe easiest to prove 199

Different facts, each imperfectly proved, corroborate eachother when they harmonise 204

Disagreement between documents and other sources ofknowledge--Improbable statements--Miracles--When scienceand history conflict, history should give way 205

BOOK III

_SYNTHETIC OPERATIONS_

CHAPTER I

GENERAL CONDITIONS OF HISTORICAL CONSTRUCTION

The materials of Historical Construction are isolated facts, of very different kinds, of very different degrees ofgenerality, each belonging to a definite time and place, of different degrees of certainty 211

Subjectivity of History 214

The facts learnt from documents relate to (1) living beingsand material objects; (2) actions, individual and collective;(3) motives and conceptions 217

The facts of the past must be imagined on the model of thoseof the present--Danger of error especially in regard tomental facts 219

Some of the conditions of human life are permanent--Thestudy of these provides a framework into which detailstaken from documents are to be fitted--For this purposesystematic lists of questions are to be used, drawn upbeforehand, and relating to the universal conditions of life 224

Outline of Historical Construction--The division oflabour--Historians must use the works of their colleagues andpredecessors, but not without critical precautions 228

CHAPTER II

THE GROUPING OF FACTS

Historical facts may be classified and arranged either accordingto their time and place, or according to their nature--Schemefor the _logical_ classification of general historicalfacts 232

The selection of facts for treatment--The history of civilisationand "battle-history"--Both needed 236

The determination of groups of men--Precautions to beobserved--The notion of "race" 238

The study of institutions--Danger of being misled by metaphors--Thequestions which should be asked 241

Evolutions: operations involved in the study of them--Theplace of particular facts (events) in evolution--Importantand unimportant facts 244

Periods--How they should be defined 249

CHAPTER III

CONSTRUCTIVE REASONING

Incompleteness of the facts yielded by documents--Cautionsto be observed in filling up the gaps by reasoning 252

The argument from silence--When admissible 254

Positive reasoning based on documents--The general principlesemployed must enter into details, and the particularfacts to which they are applied must not be taken inisolation 256

CHAPTER IV

THE CONSTRUCTION OF GENERAL FORMULÆ

History, like every science, needs formulæ by which the factsacquired may be condensed into manageable form 262

Descriptive formulæ--Should retain characteristic features--Shouldbe as concrete as possible 264

Formulæ describing general facts--How constructed--Conventionalforms and realities--Mode of formulating anevolution 266

Formulæ describing unique facts--Principle of choice--"Character"of persons--Precautions in formulatingthem--Formulæ describing events 270

Quantitative formulæ--Operations by which they may beobtained: measurement, enumeration, valuation, sampling, generalisation--Precautions to be observed in generalising 274

Formulæ expressing relations--General conclusions--Estimationof the extent and value of the knowledge acquired--Imperfectionof data not to be forgotten in construction 279

Groups and their classification 282

The "solidarity" of social phenomena--Necessity of studyingcauses--Metaphysical hypothesis--Providence--Conceptionof events as "rational"--The Hegelian "ideas"--Thehistorical "mission"--The theory of the generalprogress of humanity 285

The conception of society as an organism--The comparativemethod--Statistics--Causes cannot be investigateddirectly, as in other sciences--Causation as exhibited inthe sequence of particular events 288

The study of the causes of social evolution must look beyondabstractions to the concrete, acting and thinking men--Theplace of hereditary characteristics in determiningevolution 292

CHAPTER V

EXPOSITION

Former conceptions of history-writing--The ancient andmediæval ideal--The "history of civilisation"--Themodern historical "manual"--The romantic ideal at thebeginning of the century--History regarded as a branchof literature up to 1850 296

The modern scientific ideal--Monographs--Right choice ofsubject--References--Chronological order--Unambiguoustitles--Economy of erudition 303

General works--_A. _ meant for students and specialists--Worksof reference or "repertories" and scientific manuals ofspecial branches of history--Their form and style--Collaborationin their production--Scientific generalhistories 307

_B. _ Works intended for the public--The best kind ofpopularisation--The inferior kind--Specialists who lower theirstandard when they write for the public--The literarystyle suitable for history 311

CONCLUSION

Summary description of the methods of history--The futureof history 316

The utility of history--Not directly applicable to presentconditions--Affords an explanation of the present--Helps(and is helped by) the social sciences--A means of intellectualculture 319

APPENDIX I

THE SECONDARY TEACHING OF HISTORY IN FRANCE

Late introduction of history as a subject of secondaryinstruction--Defective methods employed up to the end of theSecond Empire 325

The reform movement--Questions involved relating to generalorganisation--Choice of subjects--Order of teaching--Methodsof instruction--These questions to be answeredin the way that will make history most useful as a meansof social culture 328

Material aids--Engravings--Books--Methods of teaching 332

APPENDIX II

THE HIGHER TEACHING OF HISTORY IN FRANCE

The different institutions--The Collège de France--TheFaculties of Letters--The École Normale--The École desChartes--The École pratique des hautes Études 335

Reform of the Faculties--Preparation for degrees--TheExamination question--Principles on which it is to besolved--The _Diplôme d'études supérieures_ 340

Influence of the movement on the other institutions--Co-operationof the institutions 345

INDEX OF PROPER NAMES 347

AUTHORS' PREFACE

The title of this work is clear. However, it is necessary to statesuccinctly both what our intention has, and what it has not been; forunder this same title, "Introduction to the Study of History, " verydifferent books have already been published. 

It has not been our intention to give, as Mr. W. B. Boyce[1] has done, asummary of universal history for the use of beginners and readers ofscanty leisure. 

Nor has it been our intention to add a new item to the abundantliterature of what is ordinarily called the "Philosophy of History. "Thinkers, for the most part not professed historians, have made historythe subject of their meditations; they have sought for its "analogies"and its "laws. " Some have supposed themselves to have discovered "thelaws which have governed the development of humanity, " and thus to have"raised history to the rank of a positive science. "[2] These vastabstract constructions inspire with an invincible _a priori_ mistrust, not the general public only, but superior minds as well. Fustel deCoulanges, as his latest biographer tells us, was severe on thePhilosophy of History; these systems were as repugnant to him asmetaphysics to the positivists. Rightly or wrongly (without doubtwrongly), the Philosophy of History, not having been cultivatedexclusively by well-informed, cautious men of vigorous and soundjudgment, has fallen into disrepute. The reader will be reassured--ordisappointed, as the case may be--to learn that this subject will findno place in the present work. [3]

We propose to examine the conditions and the methods, to indicate thecharacter and the limits, of historical knowledge. How do we ascertain, in respect of the past, what part of it it is possible, what part of itit is important, to know? What is a document? How are documents to betreated with a view to historical work? What are historical facts? Howare they to be grouped to make history? Whoever occupies himself withhistory performs, more or less unconsciously, complicated operations ofcriticism and construction, of analysis and synthesis. But beginners, and the majority of those who have never reflected on the principles ofhistorical methodology, make use, in the performance of theseoperations, of instinctive methods which, not being, in general, rational methods, do not usually lead to scientific truth. It is, therefore, useful to make known and logically justify the theory of thetruly rational methods--a theory which is now settled in some parts, though still incomplete in some points of capital importance. 

The present "Introduction to the Study of History" is thus intended, notas a summary of ascertained facts or a system of general ideas onuniversal history, but as an essay on the method of the historicalsciences. 

We proceed to state the reasons why we have thought such a workopportune, and to explain the spirit in which we have undertaken towrite it. 

I

The books which treat of the methodology of the historical sciences arescarcely less numerous, and at the same time not in much better favour, than the books on the Philosophy of History. Specialists despise them. Awidespread opinion is expressed in the words attributed to a certainscholar: "You wish to write a book on philology; you will do much betterto produce a book with some good philology in it. When I am asked todefine philology, I always answer that it is what I work at. "[4] Again, in reference to J. G. Droysen's _Précis of the Science of History_, acertain critic expressed an opinion which was meant to be, and was, acommonplace: "Generally speaking, treatises of this kind are ofnecessity both obscure and useless: obscure, because there is nothingmore vague than their object; useless, because it is possible to be anhistorian without troubling oneself about the principles of historicalmethodology which they claim to exhibit. "[5] The arguments used by thesedespisers of methodology are strong enough in all appearance. Theyreduce to the following. As a matter of fact, there are men whomanifestly follow good methods, and are universally recognised asscholars or historians of the first order, without having ever studiedthe principles of method; conversely, it does not appear that those whohave written on historical method from the logical point of view have inconsequence attained any marked superiority as scholars or historians:some, indeed, have been known for their incompetence or mediocrity inthese capacities. In this there is nothing that need surprise us. Whowould think of postponing original research in chemistry, mathematics, the sciences proper, until he had studied the methods employed in thosesciences? Historical criticism! Yes, but the best way to learn it is toapply it; practice teaches all that is wanted. [6] Take, too, the extantworks on historical method, even the most recent of them, those of J. G. Droysen, E. A. Freeman, A. Tardif, U. Chevalier, and others; the utmostdiligence will extract from them nothing in the way of clear ideasbeyond the most obvious and commonplace truisms. [7]

We willingly recognise that this manner of thinking is not entirelywrong. The great majority of works on the method of pursuing historicalinvestigations and of writing history--what is called _Historic_ inGermany and England--are superficial, insipid, unreadable, sometimesridiculous. [8] To begin with, those prior to the nineteenth century, afull analysis of which is given by P. C. F. Daunou in the seventh volumeof his _Cours d'études historiques_, [9] are nearly all of them meretreatises on rhetoric, in which the rhetoric is antiquated, and theproblems discussed are the oddest imaginable. [10] Daunou makes merryover them, but he himself has shown good sense and nothing more in hismonumental work, which at the present time seems little better, andcertainly not more useful, than the earlier treatises. [11] As to themodern ones, it is true that not all have been able to escape the twodangers to which works of this character are exposed--that of beingobscure on the one hand, or commonplace on the other. J. G. Droysen's_Grundriss der Historik_ is heavy, pedantic, and confused beyond allimagination. [12] Freeman, Tardif, and Chevalier tell us nothing but whatis elementary and obvious. Their followers may still be observeddiscussing at interminable length idle questions, such as: whetherhistory is a science or an art; what are the duties of history; what isthe use of history; and so on. On the other hand, there is incontestabletruth in the remark that nearly all the specialists and historians ofto-day are, as far as method goes, self-taught, with no training exceptwhat they have gained by practice, or by imitating and associating withthe older masters of the craft. 

But though many works on the principles of method justify the distrustwith which such works are generally regarded, and though most professedhistorians have been able, apparently with no ill results, to dispensewith reflection upon historical method, it would, in our opinion, be astrained inference to conclude that specialists and historians(especially those of the future) have no need to make themselvesacquainted with the processes of historical work. The literature ofmethodology is, in fact, not without its value: gradually there has beenformed a treasury of subtle observations and precise rules, suggested byexperience, which are something more than mere common sense. [13] And, admitting the existence of those who, without having ever learnt toreason, always reason well, by a gift of nature, it would be easy to setagainst these exceptions innumerable cases in which ignorance of logic, the use of irrational methods, want of reflection on the conditions ofhistorical analysis and synthesis, have robbed the work of specialistsand historians of much of its value. 

The truth is, that, of all branches of study, history is without a doubtthe one in which it is most necessary for students to have a clearconsciousness of the methods they use. The reason is, that in historyinstinctive methods are, as we cannot too often repeat, irrationalmethods; some preparation is therefore required to counteract the firstimpulse. Besides, the rational methods of obtaining historical knowledgediffer so widely from the methods of all other sciences, that someperception of their distinctive features is necessary to avoid thetemptation of applying to history the methods of those sciences whichhave already been systematised. This explains why mathematicians andchemists can, more easily than historians, dispense with an"introduction" to their subject. There is no need to insist at greaterlength on the utility of historical methodology, for there is evidentlynothing very serious in the attacks which have been made on it. But itbehooves us to explain the reasons which have led to the composition ofthe present work. For the last fifty years a great number of intelligentand open-minded men have meditated on the methods of the historicalsciences. Naturally we find among them many historians, universityprofessors, whose position enables them to understand better than othersthe intellectual needs of the young; but at the same time professedlogicians, and even novelists. In this connection, Fustel de Coulangesleft a tradition behind him at the University of Paris. "Heendeavoured, " we are told, [14] "to reduce the rules of method to veryprecise formulæ . .. ; in his view no task was more urgent than that ofteaching students how to attain truth. " Among these men, some, likeRenan, [15] have been content to insert scattered observations in theirgeneral works or their occasional writings;[16] others, as Fustel deCoulanges, Freeman, Droysen, Laurence, Stubbs, De Smedt, VonPflugk-Harttung, and so on, have taken the trouble to express theirthoughts on the subject in special treatises. There are many books, "inaugural lectures, " "academic orations, " and review-articles, published in all countries, but especially in France, Germany, England, the United States, and Italy, both on the whole subject of methodologyand on the different parts of it. It will occur to the reader that itwould be a far from useless labour to collect and arrange theobservations which are scattered, and, one might say, lost, in thesenumerous books and minor writings. But it is too late to undertake thispleasant task; it has been recently performed, and in the mostpainstaking manner. Professor Ernst Bernheim, of the University ofGreifswald, has worked through nearly all the modern works on historicalmethod, and the fruit of his labours is an arrangement under appropriateheadings, most of them invented by himself, of a great number ofreflections and selected references. His _Lehrbuch der historischenMethode_[17] (Leipzig, 1894, 8vo) condenses, in the manner of German_Lehrbücher_, the special literature of the subject of which it treats. It is not our intention to do over again what has already been done sowell. But we are of opinion that even after this laborious andwell-planned compilation something still remains to be said. In thefirst place, Professor Bernheim deals largely with metaphysical problemswhich we consider devoid of interest; while, conversely, he entirelyignores certain considerations which appear to us to be, boththeoretically and practically, of the greatest importance. In the secondplace, the teaching of the _Lehrbuch_ is sound enough, but lacks vigourand originality. Lastly, the _Lehrbuch_ is not addressed to the generalpublic; both the language in which it is written and the form in whichit is composed render it inaccessible to the great majority of Frenchreaders. This is enough to justify our undertaking to write a book ofour own, instead of simply recommending the book of ProfessorBernheim. [18]

II

This "Introduction to the Study of History" does not claim, like the_Lehrbuch der historischen Methode_, to be a treatise on historicalmethodology. [19] It is a sketch in outline. We undertook itscomposition, at the beginning of the scholastic year 1896-97, in orderthat the new students at the Sorbonne might be warned what the study ofhistory is and ought to be. 

Long experience has taught us the necessity of such warnings. Thegreater part of those who enter upon a career of historical study do so, as a matter of fact, without knowing why, without having ever askedthemselves whether they are fitted for historical work, of the truenature of which they are often ignorant. Generally their motives forchoosing an historical career are of the most futile character. One hasbeen successful in history at college;[20] another feels himself drawntowards the past by the same kind of romantic attraction which, we aretold, determined the vocation of Augustin Thierry; some are misled bythe fancy that history is a comparatively easy subject. It is certainlyimportant that these irrational votaries should be enlightened and putto the test as soon as possible. 

Having given a course of lectures, to novices, by way of "Introductionto the Study of History, " we thought that, with a little revision, theselectures might be made useful to others besides novices. Scholars andprofessed historians will doubtless have nothing to learn from thiswork; but if they should find in it a stimulus to personal reflection onthe craft which some of them practise in a mechanical fashion, thatwould be something gained. As for the public, which reads the works ofhistorians, is it not desirable that it should know how these works areproduced, in order to be able to judge them better?

We do not, therefore, like Professor Bernheim, write exclusively forpresent and future specialists, but also for the public interested inhistory. We thus lay ourselves under an obligation to be as concise, asclear, and as little technical as possible. But to be concise and clearon subjects of this kind often means to appear superficial. Commonplaceon the one hand, obscurity on the other: these, as we have already seen, are the evils between which we have the sorry privilege of choosing. Weadmit the difficulty. But we do not think it insurmountable, and ourendeavour has been to say what we had to say in the clearest possiblemanner. 

The first half of the book has been written by M. Langlois, the secondby M. Seignobos; but the two collaborators have constantly aided, consulted, and checked each other. [21]

PARIS, _August 1897_. 

INTRODUCTION TO THE STUDY OF HISTORY

BOOK I

PRELIMINARY STUDIES

CHAPTER I

THE SEARCH FOR DOCUMENTS (HEURISTIC)

The historian works with documents. Documents are the traces which havebeen left by the thoughts and actions of men of former times. Of thesethoughts and actions, however, very few leave any visible traces, andthese traces, when there are any, are seldom durable; an accident isenough to efface them. Now every thought and every action that has leftno visible traces, or none but what have since disappeared, is lost forhistory; is as though it had never been. For want of documents thehistory of immense periods in the past of humanity is destined to remainfor ever unknown. For there is no substitute for documents: nodocuments, no history. 

In order to draw legitimate inferences from a document to the fact ofwhich it is the trace, numerous precautions are requisite which will beindicated in the sequel. But it is clear that, prior to any criticalexamination or interpretation of documents, the question presents itselfwhether there are any documents at all, how many there are, and wherethey are. If I undertake to deal with a point of history, [22] ofwhatever nature, my first step will be to ascertain the place or placeswhere the documents necessary for its treatment, if any such exist, areto be found. The search for and the collection of documents is thus apart, logically the first and most important part, of the historian'scraft. In Germany it has received the convenient, because short, name of_Heuristik_. Is there any need to prove the capital importance ofHeuristic? Assuredly not. It is obvious that if it is neglected, if thestudent does not, before he sets to work on a point of history, placehimself in a position to command all accessible sources of information, his risk (no small one at the best) of working upon insufficient data isquite unnecessarily increased: works of erudition or history constructedin accordance with the rules of the most exact method have beenvitiated, or even rendered worthless, by the accidental circumstancethat the author was unacquainted with the documents by which those whichhe had within reach, and with which he was content, might have beenillustrated, supplemented, or discredited. The scholars and historiansof to-day, standing, as they do, in other respects on an equality withtheir predecessors of the last few centuries, are only enabled tosurpass them by their possession of more abundant means ofinformation. [23] Heuristic is, in fact, easier to-day than it used tobe, although the honest Wagner has still good grounds for saying:

 "Wie schwer sind nicht die Mittel zu erwerben, Durch die man zu den Quellen steigt!"[24]

Let us endeavour to explain why the collection of documents, once solaborious, is still no easy matter, in spite of the progress made in thelast century; and how this essential operation may, in the course ofcontinued progress, be still further simplified. 

I. Those who first endeavoured to write history from the sources foundthemselves in an embarrassing situation. Were the events they proposedto relate recent, so that all the witnesses of them were not yet dead?They had the resource of interviewing the witnesses who survived. Thucydides, Froissart, and many others have followed this procedure. When Mr. H. H. Bancroft, the historian of the Pacific Coast ofCalifornia, resolved to collect materials for the history of events manyof the actors in which were still alive, he mobilised a whole army ofreporters charged to extract conversations from them. [25] But when theevents to be related were ancient, so that no man then living could havewitnessed them, and no account of them had been preserved by oraltradition, what then? Nothing was left but to collect documents of everykind, principally written ones, relating to the distant past which wasto be studied. This was a difficult task at a time when libraries wererare, archives secret, and documents scattered. About the year 1860, Mr. Bancroft, in California, was in a situation analogous to that of theearlier researchers in our part of the world. His plan was as follows:He was rich; he cleared the market of all documents, printed ormanuscript; he negotiated with financially embarrassed families andcorporations for the purchase of their archives, or the permission tohave them copied by his paid agents. This done, he housed his collectionin premises built for the purpose, and classified it. Theoreticallythere could not be a more rational procedure. But this rapid, Americanmethod has only once been employed with sufficient resources andsufficient consistency to ensure its success; at any other time, and inany other place, it would have been out of the question. Nowhere elsehave the circumstances been so favourable for it. 

At the epoch of the Renaissance the documents of ancient and modernhistory were scattered in innumerable private libraries and ininnumerable depositories of archives, almost all of them inaccessible, not to mention those which lay hidden beneath the soil, their veryexistence as yet unsuspected. It was at that time a physicalimpossibility to procure a list of all the documents serving for theelucidation of a question (for example, a list of all the manuscriptsstill preserved of an ancient work); and if, by a miracle, such a listwas to be had, it was another impossibility to consult all thesedocuments except at the cost of journeys, expenses, and negotiationswithout end. Consequences easy to foresee did, as a matter of fact, ensue. Firstly, the difficulties of Heuristic being insurmountable, theearliest scholars and historians--employing, as they did, not all thedocuments, nor the best documents, but those documents on which theycould lay their hands--were nearly always ill-informed; and their worksare now without interest except so far as they are founded on documentswhich have since been lost. Secondly, the first scholars and historiansto be relatively well-informed were those who, in virtue of theirprofession, had access to rich storehouses of documents--librarians, keepers of archives, monks, magistrates, whose order or whosecorporation possessed libraries or archives of considerable extent. [26]

It is true that collectors soon arose who, by money payments, or by morequestionable expedients, such as theft, formed, with more or less regardfor the interests of scientific study, "cabinets" of collections oforiginal documents, and of copies. But these European collectors, ofwhom there has been a great number since the fifteenth century, differvery noticeably from Mr. Bancroft. The Californian, in fact, onlycollected documents relating to a particular subject (the history ofcertain Pacific states), and his ambition was to make his collectioncomplete; most European collectors have acquired waifs and strays andfragments of every description, forming, when combined, totals whichappear insignificant by the side of the huge mass of historicaldocuments which existed at the time. Besides, it was not, in general, with any purpose of making them generally accessible that collectorslike Peiresc, Gaignières, Clairambault, Colbert, and many others, withdrew from circulation documents which were in danger of being lost;they were content (and it was creditable to do as much as this) to sharethem, more or less freely, with their friends. But collectors (and theirheirs) are fickle people, and sometimes eccentric in their notions. Certainly it is better that documents should be preserved in privatecollections, than that they should be entirely unprotected andabsolutely inaccessible to the scientific worker; but in order thatHeuristic should be made really easier, the first condition is that allcollections of documents should be _public_. [27]

Now the finest private collections of documents--libraries and museumscombined--were naturally, in the Europe of the Renaissance, thosepossessed by kings. And while other private collections were oftendispersed upon the death of their founders, these, on the contrary, never ceased to grow; they were enriched, indeed, by the wreckage of allthe others. The _Cabinet des manuscrits de France_, for example, formedby the French kings, and by them thrown open to the public, had, at theend of the eighteenth century, absorbed the best part of the collectionswhich had been the personal work of the amateurs and scholars of the twopreceding centuries. [28] Similarly in other countries. The concentrationof a great number of historical documents in vast public (orsemi-public) establishments was the fortunate result of this spontaneousevolution. 

The arbitrary proceedings of the Revolution were still more favourable, and still more effective in securing the amelioration of the materialconditions of historical research. The Revolution of 1789 in France, analogous movements in other countries, led to the violent confiscation, for the profit of the state (that is, of everybody), of a host ofprivate archives and collections--the archives, libraries, and museumsof the crown, the archives and libraries of monasteries and suppressedcorporations, and so on. In France, in 1790, the Constituent Assemblythus placed the state in possession of a great number of depositories ofhistorical documents, previously scattered, and guarded more or lessjealously from the curiosity of scholars; these treasures have sincebeen divided among four different national institutions. The samephenomenon has been more recently observed, on a smaller scale, inGermany, Spain, and Italy. 

The confiscations of the revolutionary period, as well as thecollections of the period which preceded it, have both been productiveof serious damage. The collector is, or rather often was, a barbarianwho did not hesitate, when he saw a chance of adding to his collectionof specimens and rare remains, to mutilate monuments, to dissectmanuscripts, to break up whole archives, in order to possess himself ofthe fragments. On this score many acts of vandalism were perpetratedbefore the Revolution. Naturally, the revolutionary procedure ofconfiscation and transference was also productive of lamentableconsequences; besides the destruction which was the result of negligenceand that which was due to the mere pleasure of destroying, theunfortunate idea arose that collections might be systematically_weeded_, those documents only to be preserved which were "interesting"or "useful, " the rest to be got rid of. The task of weeding wasentrusted to well-meaning but incompetent and overworked men, who werethus led to commit irreparable havoc in our ancient archives. At thepresent day there are workers engaged in the task, one requiring anextraordinary amount of time, patience, and care, of restoring thedismembered collections, and replacing the fragments which were thenisolated in so brutal a manner by these zealous but unreflectingmanipulators of historical documents. It must be recognised, moreover, that the mutilations due to revolutionary activity and thepre-revolutionary collectors are insignificant in comparison with thosewhich are the result of accident and the destructive work of time. Buthad they been ten times as serious, they would have been amplycompensated by two advantages of the first importance, on which wecannot lay too much stress: (1) the concentration, in a relatively smallnumber of depositories, of documents which were formerly scattered, and, as it were, lost, in a hundred different places; (2) the opening ofthese depositories to the public. The remnant of historical documentswhich has survived the destructive effects of accident and vandalism isnow at last safely housed, classified, made accessible, and treated aspublic property. 

Ancient historical documents are now, as we have seen, collected andpreserved chiefly in those public institutions which are calledarchives, libraries, and museums. It is true that this does not apply to_all_ existing documents; in spite of the unceasing acquisitions bypurchase and gift which archives, libraries, and museums all over theworld have been making every year for a long time past, there stillexist private collections, dealers who supply them, and documents incirculation. But the exceptions, which in this case are negligeable, donot affect the general rule. Besides, all the ancient documents which, in limited quantity, still range at large, are destined sooner or laterto find their way into the state institutions, whose doors are alwaysopen to let in, but never to let out. [29]

It is to be desired, as a matter of principle, that the depositories ofdocuments (archives, libraries, and museums) should not be too numerous;and we have pointed out that, fortunately, they are now beyondcomparison less numerous than they were a hundred years ago. Could notthe centralisation of documents, with its evident advantages forresearchers, be carried still further? Are there not still collectionsof documents of which it would be hard to justify the separateexistence? Perhaps;[30] but the problem of the centralisation ofdocuments is no longer urgent, now that the processes of reproductionhave been perfected, especially as the inconveniences arising from amultitude of depositories are met by the expedient, now in general use, of allowing the documents to travel: it is now possible for the studentto consult, without expense, in the public library of the city where heresides, documents belonging, say, to the libraries of St. Petersburg, Brussels, and Florence; we now rarely meet with institutions like theArchives Nationales at Paris, the British Museum at London, and theMéjanes Library at Aix-en-Provence, whose statutes absolutely prohibitall lending-out of their contents. [31]

II. It being granted that the majority of historical documents are nowpreserved in public institutions (archives, libraries, and museums), Heuristic would be very easy if only good descriptive catalogues hadbeen drawn up of all the existing collections of documents, if thesecatalogues were furnished with indexes, or if general repertories(alphabetical, systematic, &c. ) had been made relating to them; lastly, if there were some place where it was possible to consult the completecollection of all these catalogues and their indexes. But Heuristic isstill difficult, because these conditions are, unfortunately, still veryfar from being adequately realised. 

Firstly, there are depositories of documents (archives, libraries, andmuseums) whose contents have never been even partially catalogued, sothat no one knows what is in them. The depositories of which we possesscomplete descriptive catalogues are rare; there are many collectionspreserved in celebrated institutions which have only been catalogued inpart, and the bulk of which still remains to be described. [32] In thesecond place, what a variety there is among existing catalogues! Thereare some old ones which do not now correspond to the presentclassification of documents, and which cannot be used withoutreference-tables; there are new ones which are equally based on obsoletesystems, too detailed or too summary; some are printed, others inmanuscript, on registers or slips; some are carefully executed andclear, many are scamped, inadequate, and provisional. Taking printedcatalogues alone, it requires a whole apprenticeship to learn todistinguish, in this enormous mass of confusion, between what istrustworthy and what is not; in other words, to make any use of them atall. Lastly, where are the existing catalogues to be consulted? Most ofthe great libraries only possess incomplete collections of them; thereis no general guide to them anywhere. 

This is a deplorable state of things. In fact, the documents containedin uncatalogued depositories and collections are practicallynon-existent for researchers who have no leisure to work through thewhole of their contents for themselves. We have said before: nodocuments, no history. But to have no good descriptive catalogues ofcollections of documents means, in practice, to be unable to ascertainthe existence of documents otherwise than by chance. We infer that theprogress of history depends in great measure on the progress of thegeneral catalogue of historical documents which is still fragmentary andimperfect. On this point there is general agreement. Père Bernard deMontfaucon considered his _Bibliotheca bibliothecarum manuscriptarumnova_, a collection of library catalogues, as "the most useful and mostinteresting work he had produced in his whole life. "[33] "In the presentstate of science, " wrote Renan in 1848, [34] "nothing is wanted moreurgently than a critical catalogue of the manuscripts in the differentlibraries . .. A humble task to all appearance; . .. And yet theresearches of scholars are hampered and incomplete pending itsdefinitive completion. " "We should have better books on our ancientliterature, " says M. P. Meyer, [35] "if the predecessors of M. Delisle[in his capacity of administrator of the Bibliothèque Nationale atParis] had applied themselves with equal ardour and diligence to thecataloguing of the treasures committed to their care. "

It will be well to indicate briefly the causes and state the exactconsequences of a state of things which has been deplored as long asscholars have existed, and which is improving, though slowly. "I assureyou, " said Renan, [36] "that the few hundred thousand francs a Ministerof Public Instruction might apply to the purpose [of preparingcatalogues] would be better employed than three-quarters of the sum nowdevoted to literature. " It is rare to find a minister, in France orelsewhere, convinced of this truth, and resolute enough to actaccordingly. Besides, it has not always been true that, in order toobtain good catalogues, it is sufficient, as well as necessary, to makea pecuniary sacrifice: it is only recently that the best methods ofdescribing documents have been authoritatively fixed; the task ofrecruiting competent workers--no great difficulty nowadays--would havebeen neither easy nor free from anxiety at an epoch when competentworkers were rarer than they are now. So much for the materialobstacles--want of money and want of men. A cause of another kind hasnot been without its influence. The functionaries charged with theadministration of depositories of documents have not always displayedthe zeal which they now display for making their collections accessibleby means of accurate catalogues. To prepare a catalogue (in the exactand at the same time summary form which is now used) is a laborioustask, a task without joy and without reward. It has often happened thatsuch a functionary, living, in virtue of his office, in the midst ofdocuments which he is at liberty to consult at any moment, and placed ina much more favourable position than the general public for utilisingthe collection without the aid of a catalogue, and making discoveries inthe process, has preferred to work for himself rather than for others, and made the tedious construction of a catalogue a secondary mattercompared with his personal researches. 

Who are the persons that in our own day have discovered, published, andannotated the greatest number of documents? The functionaries attachedto the depositories of documents. Without a doubt this circumstance hasretarded the progress of the general catalogue of historical documents. The situation has been this: the persons who were the best able todispense with catalogues were precisely the persons whose duty it was tomake them. 

The imperfection of descriptive catalogues has consequences whichdeserve our attention. On the one hand, we can never be sure that wehave exhausted all the sources of information; who knows what may beheld in reserve by the uncatalogued collections?[37] On the other hand, in order to obtain the maximum amount of information, it is necessary tobe thoroughly acquainted with the resources furnished by the existingliterature of Heuristic, and to devote a great deal of time topreliminary researches. In point of fact, every one who proposes tocollect documents for the treatment of a point of history begins byconsulting indexes and catalogues. [38] Novices set about this importantoperation so slowly, with so little skill, and with so much effort, asto move more experienced workers to mirth or pity, according to theirdisposition. Those who find amusement in watching novices stumble andstrain and waste their time in the labyrinth of catalogues, neglectingthose which are valuable, and thoroughly exploring those which areuseless, remember that they also have passed through similarexperiences: let every one have his turn. Those who observe with regretthis waste of time and strength consider that, while inevitable up to acertain point, it serves no good purpose; they ask whether somethingmight not be done to mitigate the severity of this apprenticeship toHeuristic, which at one time cost them so dear. Besides, is notresearch, in the present condition of its material aids, difficultenough whatever the experience of the researcher? There are scholars andhistorians who devote the best part of their powers to materialsearches. Certain branches of historical work, relating chiefly tomediæval and modern subjects (the documents of ancient history arefewer, have been more studied, and are better catalogued than theothers), imply not merely the assiduous use of catalogues, not allfurnished with indexes, but also the personal inspection of the wholecontents of immense collections which are either badly catalogued or notcatalogued at all. Experience proves beyond a doubt that the prospect ofthese long searches, which must be performed before the moreintellectual part of the work can be begun, has deterred, and continuesto deter, men of excellent abilities from undertaking historical work. They are, in fact, confronted with a dilemma: either they must work on asupply of documents which is in all probability incomplete, or they mustspend themselves in unlimited searches, often fruitless, the results ofwhich seldom appear worth the time they have cost. It goes against thegrain to spend a great part of one's life in turning over catalogueswithout indexes, or in passing under review, one after another, all theitems which go to form accumulations of uncatalogued _miscellanea_, inorder to obtain information (positive or negative) which might have beenobtained easily and instantaneously if the collections had beencatalogued and if the catalogues had been indexed. The most seriousconsequence of the present imperfection of the material aids toHeuristic is the discouragement which is sure to be felt by many ablemen who know their worth, and have some sense of the due proportion ofeffort and reward. [39]

If it lay in the nature of things that the search for historicaldocuments, in public depositories, must necessarily be as laborious asit still is, we might resign ourselves to the inconvenience: no onethinks of regretting the inevitable expenditure of time and labour whichis demanded by archæological research, whatever the results may prove tobe. But the imperfection of the modern instruments of Heuristic is quiteunnecessary. The state of things which existed for some centuries hasnow been reformed indifferently; there is no valid reason why it shouldnot some day be reformed altogether. We are thus led, after treating ofthe causes and the effects, to say a few words about the remedies. 

The instruments of Heuristic are being continually perfected, before oureyes, in two ways. Every year witnesses an increase in the number ofdescriptive catalogues of archives, libraries, and museums, prepared bythe functionaries attached to these institutions. In addition to this, powerful learned societies employ experts to pass from one depositoryto another cataloguing the documents there, in order to pick out all thedocuments of a particular class, or relating to a special subject: thusthe society of Bollandists caused a general catalogue of hagiographicaldocuments to be prepared by its emissaries, and the Imperial Academy ofVienna catalogued in a similar manner the monuments of patristicliterature. The society of the _Monumenta Germanioe Historica_ has fora long time been conducting vast searches of the same kind; and it wasby the same process of exploring the museums and libraries of the wholeof Europe that the construction of the _Corpus Inscriptionum Latinarum_was lately rendered possible. Lastly, several governments have taken theinitiative in sending abroad persons charged to catalogue, on theirbehalf, documents in which they are interested: thus England, theNetherlands, Switzerland, the United States, and other governments, grant regular subsidies to agents of theirs occupied in cataloguing andtranscribing, in the great depositories of Europe, the documents whichrelate to the history of England, the Netherlands, Switzerland, theUnited States, and the rest. [40] With what rapidity and with whatperfection these useful labours can be conducted, provided that acompetent staff, suitably directed, can be had as well as the money topay it, is shown by the history of the _general catalogue of themanuscripts in the public libraries of France_. This excellentdescriptive catalogue was begun in 1885, and now, in 1897, it extends tonearly fifty volumes, and will soon be completed. The _CorpusInscriptionum Latinarum_ will have been produced in less than fiftyyears. The results obtained by the Bollandists and the Imperial Academyof Vienna are not less conclusive. Assuredly nothing is now lacking, except funds, to secure the speedy endowment of historical study withthe indispensable instruments of research. The methods employed in theconstruction of these instruments are now permanently fixed, and it isan easy matter to recruit a trained staff. Such a staff must evidentlybe largely composed of keepers of archives and professional librarians, but it would also contain unattached workers with a decided vocation forthe construction of catalogues and indexes. Such workers are morenumerous than one would at first be inclined to think. Not thatcataloguing is easy: it requires patience, the most scrupulousattention, and the most varied learning; but many minds are attracted bytasks which, like this, are at once determinate, capable of beingdefinitely completed, and of manifest utility. In the large andheterogeneous family of those who labour to promote the progress ofhistorical study, the makers of descriptive catalogues and indexes forma section to themselves. When they devote themselves exclusively totheir art they acquire by practice, as one might expect, a high degreeof dexterity. 

While waiting for the fact to be clearly recognised that the time isopportune for pushing vigorously in every country the construction of ageneral catalogue of historical documents, we may indicate a palliative:it is important that scholars and historians, especially novices, shouldbe accurately informed of the state of the instruments of research whichare at their disposal, and be regularly apprised of any improvementsthat from time to time may be made in them. Experience and accident havebeen for a long time trusted to supply this information; but empiricalknowledge, besides being costly, as we have already pointed out, isalmost always imperfect. Recently the task has been undertaken ofconstructing catalogues of catalogues--critical and systematic lists ofall the catalogues in existence. There can be no doubt that fewbibliographical enterprises have possessed, in so great a degree, thecharacter of general utility. 

But scholars and historians often need, in respect of documents, information not usually supplied by descriptive catalogues; they wish, for example, to know whether such and such a document is known or not, whether it has already been critically dealt with, annotated, orutilised. [41] This information can only be found in the works of formerscholars and historians. In order to become acquainted with theseworks, recourse must be had to those "bibliographical repertories, "properly so called, of all kinds, compiled from very different points ofview, which have already been published. Among the indispensableinstruments of Heuristic must thus be reckoned bibliographicalrepertories of historical literature, as well as repertories ofcatalogues of original documents. 

To supply the classified list of all those repertories (repertories ofcatalogues, bibliographical repertories, properly so called), togetherwith other appropriate information, in order to save students frommistakes and waste of time, is the object of what we are at liberty tocall the "science of repertories, " or "historical bibliography. "Professor Bernheim has published a preliminary sketch[42] of it, whichwe have endeavoured to expand. [43] The expanded sketch bears date April1896: numerous additions, not to speak of revision, would already benecessary, for the bibliographical apparatus of the historical sciencesis being renewed, at the present time, with astonishing rapidity. A bookon the repertories for the use of scholars and historians is, as ageneral rule, out of date the day after it has been completed. 

III. The knowledge of repertories is useful to all; the preliminarysearch for documents is laborious to all; but not in the same degree. Certain parts of history, which have been long cultivated, now enjoythe advantage of having all their documents described, collected, andclassified in large publications devoted to the purpose, so that, indealing with these subjects, the historian can do all that need be doneat his desk. The study of local history does not generally require morethan local search. Some important monographs are based on a small numberof documents, all belonging to the same collection, and of such a naturethat it would be superfluous to look for others elsewhere. On the otherhand, a humble piece of work, such as a modest edition of a text ofwhich the ancient copies are not rare, and are to be found scattered inseveral libraries of Europe, may have involved inquiries, negotiations, and journeys without end. Since the majority of the documents ofmediæval and modern history are still unedited, or badly edited, it maybe laid down as a general principle that, in order to write a really newchapter of mediæval or modern history, it is necessary to have longhaunted the great depositories of original documents, and to have, if wemay use the expression, worried their catalogues. 

It is thus incumbent on every one to choose the subject of his labourswith the greatest care, instead of leaving it to be determined by purechance. There are some subjects which, in the present state of theinstruments of research, cannot be treated except at the cost ofenormous searches in which life and intellect are consumed withoutprofit. These subjects are not necessarily more interesting than others, and some day, perhaps to-morrow, improvements in the aids to researchwill make them easily manageable. It is necessary for the studentconsciously and deliberately to make his choice between differenthistorical subjects depend on the existence or non-existence ofparticular catalogues of documents and bibliographical repertories; onhis relative inclination for desk work on the one hand, and the labourof exploring depositories on the other; even on the facilities he hasfor making use of particular collections. "Is it possible to do work inthe provinces?" Renan asked at the congress of learned societies at theSorbonne in 1889; and gave a very good answer to his own question: "Atleast half one's scientific work can be done at one's own desk . .. Takecomparative philology, for example: with an initial outlay of somethousands of francs, and subscriptions to three or four specialpublications, a student would command all the tools of his trade . .. Thesame applies to universal philosophy . .. Many branches of study can thusbe prosecuted quite privately, and in the closest retirement. "[44]Doubtless, but there are "rarities, specialities, researches whichrequire the aid of powerful machinery. " One half of historical work maynow be done in private, with limited resources, but only half; the otherhalf still presupposes the employment of such resources, in the way ofrepertories and documents, as can only be found in the great centres ofstudy; often, indeed, it is necessary to visit several of these centresin succession. In short, the case stands with history much as it doeswith geography: in respect of some portions of the globe, we possessdocuments published in manageable form sufficiently complete andsufficiently well classified to enable us to reason about them to goodpurpose without leaving our fireside; while in the case of an unexploredor badly explored region, the slightest monograph implies a considerableexpenditure of time and physical strength. It is dangerous to choose asubject of study, as many do, without having first realised the natureand extent of the preliminary researches which it demands; there areinstances of men struggling for years with such researches, who mighthave been occupied to better advantage in work of another character. Asprecautions against this danger, which is the more formidable to novicesthe more active and zealous they are, an examination of the presentconditions of Heuristic in general, and positive notions of HistoricalBibliography, are certainly to be warmly recommended. 

CHAPTER II

"AUXILIARY SCIENCES"

Let us suppose that the preliminary searches, treated of in thepreceding chapter, have been made methodically and successfully; thegreater part, if not the whole, of the documents bearing on a givensubject have been discovered and made available. Of two things one:either these documents have been already subjected to criticalelaboration, or they are in the condition of raw material; this is apoint which must be settled by "bibliographical" researches, which also, as we have already observed, form part of the inquiries which precedethe logical part of the work. In the first case, where the documentshave already gone through a process of elaboration, it is necessary tobe in a position to verify the accuracy of the critical work; in thesecond case, where the documents are still raw material, the studentmust do the critical work himself. In both cases certain antecedent andauxiliary knowledge of a positive kind, _Vor-und Hülfskenntnisse_, asthey are called, are every whit as indispensable as the habit ofaccurate reasoning; for if, in the course of critical work, it ispossible to go wrong through reasoning badly, it is also possible to gowrong out of pure ignorance. The profession of a scholar or historianis, moreover, similar in this respect to all other professions; it isimpossible to follow it without possessing a certain equipment oftechnical notions, whose absence neither natural aptitude nor evenmethod can make good. In what, then, does the technical _apprenticeship_of the scholar or the historian consist? Or, to employ language which, though inappropriate, as we shall endeavour to show, is in more commonuse: what, in addition to the knowledge of repertories, are the"auxiliary sciences" of history?

Daunou, in his _Cours d'études historiques_, [45] has proposed a questionof the same kind. "What studies, " says he, "will the intending historianneed to have gone through, what kinds of knowledge ought he to haveacquired, in order to begin writing a work with any hope of success?"Before him, Mably, in his _Traité de l'étude de l'histoire_, had alsorecognised that "there are preparatory studies with which no historiancan dispense. " But on this subject Mably and Daunou entertained viewswhich nowadays seem singular enough. It is instructive to mark the exactdistance which separates their point of view from ours. "First of all, "said Mably, "study the law of nature, public law, moral and politicalscience. " Daunou, a man of great judgment, permanent secretary to theAcademy of Inscriptions and Belles-Lettres, writing about 1820, dividesthe studies which, in his opinion, constitute "the apprenticeship of thehistorian, " into three classes--literary, philosophical, historical. Onthe "literary" studies he expatiates at great length: to begin with, the historian must "have read with attention the great models. " Whichgreat models? Daunou "does not hesitate" to place in the front rank "themasterpieces of epic poetry;" for "it is the poets who have created theart of narrative, and whoever has not learnt it from them cannot havemore than an imperfect knowledge of it. " He further recommends thereading of modern novels; "they will teach the method of giving anartistic pose to persons and events, of distributing details, ofskilfully carrying on the thread of the narrative, of interrupting it, of resuming it, of sustaining the attention and provoking the curiosityof the reader. " Finally, good historical works should be read:"Herodotus, Thucydides, Xenophon, Polybius, and Plutarch among theGreeks; Cæsar, Sallust, Livy, and Tacitus among the Latins; and amongthe moderns, Macchiavelli, Guicciardini, Giannone, Hume, Robertson, Gibbon, the Cardinal de Retz, Vertot, Voltaire, Raynal, and Rulhière. Not that I would exclude the others, but these will suffice to provideall the styles which are suitable for history; for a great diversity ofform is to be met with in the works of these writers. " In the secondplace come philosophical studies; a thorough mastery of "ideology, morals, and politics" is required. "As to the works from which knowledgeof this kind is to be obtained, Daguesseau has instanced Aristotle, Cicero, Grotius: I should add the best ancient and modern moralists, treatises on political economy published since the middle of the lastcentury, the writings on political science in general, and on itsdetails and application, of Macchiavelli, Bodin, Locke, Montesquieu, Rousseau, Mably, and the most enlightened of their disciples andcommentators. " In the third place, before writing history, "it isevidently necessary to know it. " "A writer will not give the world newinformation on a subject like this unless he begins by making himselfmaster of what is already known of it. " The future historian has alreadymade the acquaintance of the best historical works, and studied them asmodels of style; "it will be to his advantage to read them a secondtime, but endeavouring more particularly to grasp all the facts whichthey contain, and to let them make so deep an impression on his mindthat they may be permanently fixed in his memory. "

These are the "positive" notions which, eighty years ago, wereconsidered indispensable to the general historian. At the same timethere was a confused idea that "in order to acquire a profound knowledgeof particular subjects" there were yet other useful branches of study. "The subjects of which historians treat, " says Daunou, "the detailswhich they occasionally light upon, require very extensive and variedattainments. " He goes on to particularise, observe in what terms: "veryoften a knowledge of several languages, sometimes too some notion ofphysics and mathematics. " And he adds: "On these subjects, however, thegeneral education which we may assume to be common to all men of lettersis sufficient for the writer who devotes himself to historicalcomposition. .. . "

All the authors who, like Daunou, have attempted to enumerate thepreliminary attainments, as well as the moral or intellectual aptitudes, necessary for "writing history, " have either fallen into commonplace orpitched their requirements ridiculously high. According to Freeman, thehistorian ought to know everything--philosophy, law, finance, ethnography, geography, anthropology, natural science, and what not; isnot an historian, in point of fact, likely enough in the course of hisstudy of the past to meet with questions of philosophy, law, finance, and the rest of the series? And if financial science, for example, isnecessary to a writer who treats of contemporary finance, is it less soto the writer who claims to express an opinion on the financialquestions of the past? "The historian, " Freeman declares, "may haveincidentally to deal with any subject whatever, and the more branches ofknowledge he is master of, the better prepared he is for his own work. "True, all branches of human knowledge are not equally useful; some ofthem are only serviceable on rare occasions, and accidentally: "We couldhardly make it even a counsel of perfection to the historian to makehimself an accomplished chemist, on the chance of an occasion in whichchemistry might be of use to him in his study;" but other specialsubjects are more closely related to history: "for example, geology anda whole group of sciences which have a close connection with geology. .. . The historian will clearly do his own regular work better for beingmaster of them. .. . "[46] The question has also been asked whether"history is one of those studies anciently called _umbratiles_, forwhich all that is wanted is a quiet mind and habits of industry, " orwhether it is a good thing for the historian to have mingled in theturmoil of active life, and to have helped to make the history of hisown time before sitting down to write that of the past. Indeed, whatquestions have not been asked? Floods of ink have been poured out overthese uninteresting and unanswerable questions, the long and fruitlessdebating of which has done not a little to discredit works onmethodology. Our opinion is that nothing relevant can be added to thedictates of mere common sense on the subject of the apprenticeship tothe "art of writing history, " unless perhaps that this apprenticeshipshould consist, above everything, in the study, hitherto so generallyneglected, of the principles of historical method. 

Besides, it is not the "literary historian, " the moralising andquill-driving "historians, " as conceived by Daunou and his school, thatwe have had in view; we are here only concerned with those scholars andhistorians who intend to deal with documents in order to facilitate oractually perform the scientific work of history. These stand in need ofa _technical apprenticeship_. What meaning are we to attach to thisterm?

Let us suppose we have before us a written document. What use can wemake of it if we cannot read it? Up to the time of François Champollion, Egyptian documents, being written in hieroglyphics, were, withoutmetaphor, a dead-letter. It will be readily admitted that in order todeal with ancient Assyrian history it is necessary to have learnt todecipher cuneiform inscriptions. Similarly, whoever desires to dooriginal work from the sources, in ancient or mediæval history, will, ifhe is prudent, learn to decipher inscriptions and manuscripts. We thussee why Greek and Latin epigraphy and mediæval palæography--that is, thesum of the various kinds of knowledge required for the deciphering ofancient and mediæval manuscripts and inscriptions--are considered as"auxiliary sciences" to history, or rather, the historical study ofantiquity and the middle ages. It is evident that mediæval Latinpalæography forms part of the necessary outfit of the mediævalist, justas the palæography of hieroglyphics is essential to the Egyptologist. There is, however, a difference to be observed. No one will ever thinkof devoting himself to Egyptology without having first studied theappropriate palæography. On the other hand, it is not very rare for aman to undertake the study of local documents of the middle ages withouthaving learnt to date their forms approximately, and to decipher theirabbreviations correctly. The resemblance which most mediæval writingbears to modern writing is sufficiently close to foster the illusionthat ingenuity and practice will be enough to carry him through. Thisillusion is dangerous. Scholars who have received no regularpalæographical initiation can almost always be recognised by the grosserrors which they commit from time to time in deciphering--errors whichare sometimes enough to completely ruin the subsequent operations ofcriticism and interpretation. As for the self-taught experts who acquiretheir skill by dint of practice, the orthodox palæographic initiationwhich they have missed would at least have saved them much groping inthe dark, long hours of labour, and many a disappointment. 

Suppose a document has been deciphered. How is it to be turned toaccount, unless it be first understood? Inscriptions in Etruscan and theancient language of Cambodia have been read, but no one understandsthem. As long as this is the case they must remain useless. It is clearthat in order to deal with Greek history it is necessary to consultdocuments in the Greek language, and therefore necessary to know Greek. Rank truism, the reader will say. Yes, but many proceed as if it hadnever occurred to them. Young students attack ancient history with onlya superficial tincture of Greek and Latin. Many who have never studiedmediæval French and Latin think they know them because they understandclassical Latin and modern French, and they attempt the interpretationof texts whose literal meaning escapes them, or appears to be obscurewhen in reality perfectly plain. Innumerable historical errors owe theirorigin to false or inexact interpretations of quite straightforwardtexts, perpetrated by men who were insufficiently acquainted with thegrammar, the vocabulary, or the niceties of ancient languages. Solidphilological study ought logically to precede historical research inevery instance where the documents to be employed are not to be had in amodern language, and in a form in which they can be easily understood. 

Suppose a document is intelligible. It would not be legitimate to takeit into consideration without having verified its authenticity, if itsauthenticity has not been already settled beyond a doubt. Now in orderto verify the authenticity or ascertain the origin of a document twothings are required--reasoning power and knowledge. In other words, itis necessary to reason from certain positive data which represent thecondensed results of previous research, which cannot be improvised, andmust, therefore, be learnt. To distinguish a genuine from a spuriouscharter would, in fact, be often an impossible task for the best trainedlogician, if he were unacquainted with the practice of such and such achancery, at such and such a date, or with the features common to allthe admittedly genuine charters of a particular class. He would beobliged to do what the first scholars did--ascertain for himself, by thecomparison of a great number of similar documents, what featuresdistinguish the admittedly genuine documents from the others, beforeallowing himself to pronounce judgment in any special instance. Will nothis task be enormously simplified if there is in existence a body ofdoctrine, a treasury of accumulated observations, a system of resultsobtained by workers who have already made, repeated, and checked theminute comparisons he would otherwise have been obliged to make forhimself? This body of doctrines, observations, and results, calculatedto assist the criticism of diplomas and charters, does exist; it iscalled Diplomatic. We shall, therefore, assign to Diplomatic, along withEpigraphy, Palæography, and Philology, the character of a subjectauxiliary to historical research. 

Epigraphy and Palæography, Philology, and Diplomatic with its adjuncts(technical Chronology and Sphragistic) are not the only subjects ofstudy which subserve historical research. It would be extremelyinjudicious to undertake to deal critically with literary documents onwhich no critical work has as yet been done without making oneselffamiliar with the results obtained by those who have already dealtcritically with documents of the same class: the sum of these resultsforms a department to itself, which has a name--the History ofLiterature. [47] The critical treatment of illustrative documents, suchas the productions of architecture, sculpture, and painting, objects ofall kinds (arms, dress, utensils, coins, medals, armorial bearings, andso forth), presupposes a thorough acquaintance with the rules andobservations which constitute Archæology properly so called and itsdetached branches--Numismatic and Heraldry. 

We are now in a position to examine to some purpose the hazy notionexpressed by the phrase, "the sciences auxiliary to history. " We alsoread of "ancillary sciences, " and, in French, "sciences satellites. "None of these expressions is really satisfactory. 

First of all, the so-called "auxiliary sciences" are not all of them_sciences_. Diplomatic, for example, and the History of Literature areonly systematised accumulations of facts, acquired by criticism, whichare of a nature to facilitate the application of critical methods todocuments hitherto untouched. On the other hand, Philology is anorganised science, and has its own laws. 

In the second place, among the branches of knowledge auxiliary--properlyspeaking, not to history, but to historical research--we mustdistinguish between those which every worker in the field ought tomaster, and those in respect of which he needs only to know where tolook when he has occasion to make use of them; between knowledge whichought to become part of a man's self, and information which he may becontent to possess only in potentiality. A mediævalist should _know_ howto read and understand mediæval texts; he would gain no advantage byaccumulating in his memory the mass of particular facts pertaining tothe History of Literature and Diplomatic which are to be found, in theirproper place, in well-constructed works of reference. 

Lastly, there are no branches of knowledge which are auxiliary toHistory (or even historical research) in general--that is, which areuseful to all students irrespectively of the particular part of historyon which they are engaged. [48] It appears, then, that there is nogeneral answer possible to the question raised at the beginning of thischapter: in what should the technical apprenticeship of the scholar orhistorian consist? In what does it consist? That depends. It depends onthe part of history he proposes to study. A knowledge of palæography isquite useless for the purpose of investigating the history of theFrench Revolution, and a knowledge of Greek is equally useless for thetreatment of a question in mediæval French history. [49] But we may go sofar as to say that the preliminary outfit of every one who wishes to dooriginal work in history should consist (in addition to the "commoneducation, " that is, general culture, of which Daunou writes) in theknowledge calculated to aid in the discovery, the understanding, andthe criticism of documents. The exact nature of this knowledge variesfrom case to case according as the student specialises in one or anotherpart of universal history. The technical apprenticeship is relativelyshort and easy for those who occupy themselves with modern orcontemporary history, long and laborious for those who occupy themselveswith ancient and mediæval history. 

This reform of the historian's technical apprenticeship which consistsin substituting the acquisition of positive knowledge, truly auxiliaryto historical research, for the study of the "great models, " literaryand philosophical, is of quite recent date. In France, for the greaterpart of the present century, students of history received none but aliterary education, after Daunou's pattern. Almost all of them werecontented with such a preparation, and did not look beyond it; some fewperceived and regretted, when it was too late for a remedy, theinsufficiency of their early training; with a few illustriousexceptions, the best of them never rose to be more than distinguishedmen of letters, incapable of scientific work. There was at that time noorganisation for teaching the "auxiliary sciences" and the technique ofresearch except in the case of French mediæval history, and that in aspecial school, the École des chartes. This simple fact, moreover, secured for this school during a period of fifty years a markedsuperiority over all the other French (or even foreign) institutions ofhigher education; excellent workers were there trained who contributedmany new results, while elsewhere people were idly discussingproblems. [50] To-day it is still at the École des chartes that themediævalist has the opportunity of going through his technicalapprenticeship in the best and most complete manner, thanks to thecombined and progressive three-years courses of Romance philology, palæography, archæology, historiography, and mediæval law. But the"auxiliary sciences" are now taught everywhere more or less adequately;they have been introduced into the university curricula. On the otherhand, students' handbooks of epigraphy, palæography, diplomatic, and soforth, have multiplied during the last twenty-five years. Twenty-fiveyears ago it would have been vain to look for a good book which shouldsupply the want of oral instruction on these subjects; since theestablishment of professorships "manuals" have appeared[51] which wouldalmost make them superfluous were it not that oral instruction, based onpractical exercises, has here an exceptional value. Whether a studentdoes or does not enjoy the advantage of a regular drilling in aninstitution for higher education, he has henceforth no excuse forremaining in ignorance of those things which he ought to know beforeentering upon historical work. There is, in fact, less of this kind ofneglect than there used to be. On this head, the success of theabove-mentioned "manuals, " with their rapid succession of editions, isvery significant. [52]

Here, then, we have the future historian armed with the preliminaryknowledge, the neglect of which would have condemned him topowerlessness or to continual mistakes. We suppose him protected fromthe errors without number which have their origin in an imperfectknowledge of the writing and the language of documents, in ignorance ofprevious work and the results obtained by textual criticism; he has anirreproachable _cognitio cogniti et cognoscendi_. A very optimisticsupposition, by the way, as we are bound to admit. We know but too wellthat to have gone through a regular course of "auxiliary sciences, " orto have read attentively the best treatises on bibliography, palæography, philology, and so on, or even to have acquired somepersonal experience by practical exercises, is not enough to ensure thata man shall always be well informed, still less to make him infallible. In the first place, those who have for a long time studied documents ofa given class or of a given period possess, in regard to these, incommunicable knowledge in virtue of which they are able to deal betterthan others with new documents which they may meet with of the sameclass or period; nothing can replace the "special erudition" which isthe specialist's reward for hard work. [53] And secondly, specialiststhemselves make mistakes: palæographers must be perpetually on theirguard not to decipher falsely; is there a philologist who has not somefaults of construing on his conscience? Scholars usually well informedhave printed as unedited texts which had already been published, andhave neglected documents it was their business to know. Scholars spendtheir lives in incessantly perfecting their "auxiliary" knowledge, whichthey rightly regard as never perfect. But all this does not prevent usfrom maintaining our hypothesis. Only let it be understood that inpractice we do not postpone work upon documents till we shall havegained a serene and absolute mastery over all the "auxiliary branches ofknowledge:" we should never dare to begin. 

It remains to know how to treat documents supposing one has successfullypassed through the preliminary apprenticeship. 

BOOK II

ANALYTICAL OPERATIONS

CHAPTER I

GENERAL CONDITIONS OF HISTORICAL KNOWLEDGE

We have already stated that history is studied from documents, and thatdocuments are the traces of past events. [54] This is the place toindicate the consequences involved in this statement and thisdefinition. 

Events can be empirically known in two ways only: by direct observationwhile they are in progress; and indirectly, by the study of the traceswhich they leave behind them. Take an earthquake, for example. I have adirect knowledge of it if I am present when the phenomenon occurs; anindirect knowledge if, without having been thus present, I observe itsphysical effects (crevices, ruins), or if, after these effects havedisappeared, I read a description written by some one who has himselfwitnessed the phenomenon or its effects. Now, the peculiarity of"historical facts"[55] is this, that they are only known indirectly bythe help of their traces. Historical knowledge is essentially indirectknowledge. The methods of historical science ought, therefore, to beradically different from those of the direct sciences; that is to say, of all the other sciences, except geology, which are founded on directobservation. Historical science, whatever may be said, [56] is not ascience of observation at all. 

The facts of the past are only known to us by the traces of them whichhave been preserved. These traces, it is true, are directly observed bythe historian, but, after that, he has nothing more to observe; whatremains is the work of reasoning, in which he endeavours to infer, withthe greatest possible exactness, the facts from the traces. The documentis his starting-point, the fact his goal. [57] Between thisstarting-point and this goal he has to pass through a complicated seriesof inferences, closely interwoven with each other, in which there areinnumerable chances of error; while the least error, whether committedat the beginning, middle, or end of the work, may vitiate all hisconclusions. The "historical, " or indirect, method is thus obviouslyinferior to the method of direct observation; but historians have nochoice: it is the _only_ method of arriving at past facts, and we shallsee later on[58] how, in spite of these disadvantages, it is possiblefor this method to lead to scientific knowledge. 

The detailed analysis of the reasonings which lead from the inspectionof documents to the knowledge of facts is one of the chief parts ofHistorical Methodology. It is the domain of criticism. The sevenfollowing chapters will be devoted to it. We shall endeavour, first ofall, to give a very summary sketch of the general lines and maindivisions of the subject. 

I. We may distinguish two species of documents. Sometimes the past eventhas left a material trace (a monument, a fabricated article). Sometimes, and more commonly, the trace is of the psychological order--a writtendescription or narrative. The first case is much simpler than thesecond. For there is a fixed relation between certain physicalappearances and the causes which produced them; and this relation, governed by physical laws, is known to us. [59] But a psychologicaltrace, on the other hand, is purely symbolic: it is not the fact itself;it is not even the immediate impression made by the fact upon thewitness's mind, but only a conventional symbol of that impression. Written documents, then, are not, as material documents are, valuableby themselves; they are only valuable as signs of psychologicaloperations, which are often complicated and hard to unravel. The immensemajority of the documents which furnish the historian withstarting-points for his reasonings are nothing else than traces ofpsychological operations. 

This granted, in order to conclude from a written document to the factwhich was its remote cause--that is, in order to ascertain the relationwhich connects the document with the fact--it is necessary to reproducethe whole series of intermediate causes which have given rise to thedocument. It is necessary to revive in imagination the whole of thatseries of acts performed by the author of the document which begins withthe fact observed by him and ends with the manuscript (or printedvolume), in order to arrive at the original event. Such is the aim andsuch the process of critical analysis. [60]

First of all we observe the document. Is it now in the same state aswhen it was produced? Has it deteriorated since? We endeavour to findout how it was made in order to restore it, if need be, to its originalform, and to ascertain its origin. The first group of preliminaryinvestigations, bearing upon the writing, the language, the form, thesource, constitutes the special domain of EXTERNAL CRITICISM, orcritical scholarship. Next comes INTERNAL CRITICISM: it endeavours, bythe help of analogies mostly borrowed from general psychology, toreproduce the mental states through which the author of the documentpassed. Knowing what the author of the document has said, we ask (1)What did he mean? (2) Did he believe what he said? (3) Was he justifiedin believing whatever he did believe? This last step brings the documentto a point where it resembles the data of the objective sciences: itbecomes an observation; it only remains to treat it by the methods ofthe objective sciences. Every document is valuable precisely to theextent to which, by the study of its origin, it has been reduced to awell-made observation. 

II. Two conclusions may be drawn from what we have just said: theextreme complexity and the absolute necessity of Historical Criticism. 

Compared with other students the historian is in a very disagreeablesituation. It is not merely that he cannot, as the chemist does, observehis facts directly; it very rarely happens that the documents which heis obliged to use represent precise observations. He has at his disposalnone of those systematic records of observations which, in theestablished sciences, can and do replace direct observation. He is inthe situation of a chemist who should know a series of experiments onlyfrom the report of his laboratory-boy. The historian is compelled toturn to account rough and ready reports, such as no man of sciencewould be content with. [61] All the more necessary are the precautions tobe taken in utilising these documents, the only materials of historicalscience. It is evidently most important to eliminate those which areworthless, and to ascertain the amount of correct observationrepresented by those which are left. 

All the more necessary, too, are cautions on this subject, because thenatural inclination of the human mind is to take no precautions at all, and to treat these matters, which really demand the utmost obtainableprecision, with careless laxity. It is true that every one admits theutility of criticism in theory; but this is just one of those principleswhich are more easily admitted than put into practice. Many centuriesand whole eras of brilliant civilisation had to pass away before thefirst dawn of criticism was visible among the most intellectual peoplesin the world. Neither the orientals nor the middle ages ever formed adefinite conception of it. [62] Up to our own day there have beenenlightened men who, in employing documents for the purpose of writinghistory, have neglected the most elementary precautions, andunconsciously assumed false generalisations. Even now most youngstudents would, if left to themselves, fall into the old errors. Forcriticism is antagonistic to the normal bent of the mind. Thespontaneous tendency of man is to yield assent to affirmations, and toreproduce them, without even clearly distinguishing them from theresults of his own observation. In every-day life do we not acceptindiscriminately, without any kind of verification, hearsay reports, anonymous and unguaranteed statements, "documents" of indifferent orinferior authority? It takes a special reason to induce us to take thetrouble to examine into the origin and value of a document on thehistory of yesterday; otherwise, if there is no outrageous improbabilityin it, and as long as it is not contradicted, we swallow it whole, wepin our faith to it, we hawk it about, and, if need be, embellish it inthe process. Every candid man must admit that it requires a violenteffort to shake off _ignavia critica_, that common form of intellectualsloth, that this effort must be continually repeated, and is oftenaccompanied by real pain. 

The natural instinct of a man in the water is to do precisely that whichwill infallibly cause him to be drowned; learning to swim meansacquiring the habit of suppressing spontaneous movements and performingothers instead. Similarly, criticism is not a natural habit; it must beinculcated, and only becomes organic by dint of continued practice. 

Historical work is, then, pre-eminently critical; whoever enters upon itwithout having first been put on his guard against his instinct is sureto be drowned in it. In order to appreciate the danger it is well toexamine one's conscience and analyse the causes of that _ignavia_ whichmust be fought against till it is replaced by a critical attitude ofmind. [63] It is also very salutary to familiarise oneself with theprinciples of historical method, and to analyse the theory of them, oneby one, as we propose to do in the present volume. "History, like everyother study, is chiefly subject to errors of fact arising frominattention, but it is more exposed than any other study to errors dueto that mental confusion which produces incomplete analyses andfallacious reasonings. .. . Historians would advance fewer affirmationswithout proof if they had to analyse each one of their affirmations;they would commit themselves to fewer false principles if they made it arule to formulate all their principles; they would be guilty of fewerfallacies if they were obliged to set out all their arguments in logicalform. "[64]

_SECTION I. --EXTERNAL CRITICISM_

CHAPTER II

TEXTUAL CRITICISM

Let us suppose that an author of our own day has written a book: hesends his manuscript to the printer; with his own hand he corrects theproofs, and marks them "Press. " A book which is printed under theseconditions comes into our hands in what is, for a document, a very goodcondition. Whoever the author may be, and whatever his sentiments andintentions, we can be certain--and this is the only point that concernsus at present--that we have before us a fairly accurate reproduction ofthe text which he wrote. We are obliged to say "fairly accurate, " for ifthe author has corrected his proofs badly, or if the printers have notpaid proper attention to his corrections, the reproduction of theoriginal text is imperfect, even in this specially favourable case. Printers not unfrequently make a man say something which he never meantto say, and which he does not notice till too late. 

Sometimes it is required to reproduce a work the author of which isdead, and the autograph manuscript of which cannot be sent to theprinter. This was the case with the _Mémoires d'outre-tombe_ ofChateaubriand, for example; it is of daily occurrence in regard to thefamiliar correspondence of well-known persons which is printed in hasteto satisfy the curiosity of the public, and of which the originalmanuscript is very fragile. First the text is copied; it is then set upby the compositor from the copy, which comes to the same thing ascopying it again; this second copy is lastly, or ought to be, collated(in the proofs) with the first copy, or, better still, with theoriginal, by some one who takes the place of the deceased author. Theguarantees of accuracy are fewer in this case than in the first; forbetween the original and the ultimate reproduction there is oneintermediary the more (the manuscript copy), and it may be that theoriginal is hard for anybody but the author to decipher. And, in fact, the text of memoirs and posthumous correspondence is often disfigured byerrors of transcription and punctuation occurring in editions which atfirst sight give the impression of having been carefully executed. [65]

Turning now to ancient documents, let us ask in what state they havebeen preserved. In nearly every case the originals have been lost, andwe have nothing but copies. Have these copies been made directly fromthe originals? No; they are copies of copies. The scribes who executedthem were not by any means all of them capable and conscientious men;they often transcribed texts which they did not understand at all, orwhich they understood incorrectly, and it was not always the fashion, asit was in the time of the Carlovingian Renaissance, to compare thecopies with the originals. [66]

If our printed books, after the successive revisions of author andprinter's reader, are still but imperfect reproductions, it is only tobe expected that ancient documents, copied and recopied as they havebeen for centuries with very little care, and exposed at every freshtranscription to new risk of alteration, should have reached us full ofinaccuracies. 

There is thus an obvious precaution to be taken. Before using a documentwe must find out whether its text is "sound"--that is, in as closeagreement as possible with the original manuscript of the author; andwhen the text is "corrupt" we must emend it. In using a text which hasbeen corrupted in transmission, we run the risk of attributing to theauthor what really comes from the copyists. There are actual cases oftheories which were based on passages falsified in transmission, andwhich collapsed as soon as the true readings were discovered orrestored. Printers' errors and mistakes in copying are not alwaysinnocuous or merely diverting; they are sometimes insidious and capableof misleading the reader. [67]

One would naturally suppose that historians of repute would always makeit a rule to procure "sound" texts, properly emended and restored, ofthe texts they have to consult. That is a mistake. For a long timehistorians simply used the texts which they had within easy reach, without verifying their accuracy. And, what is more, the very scholarswhose business it is to edit texts did not discover the art of restoringthem all at once; not so very long ago, documents were commonly editedfrom the first copies, good or bad, that came to hand, combined andcorrected at random. Editions of ancient texts are nowadays mostly"critical;" but it is not yet thirty years since the publication of thefirst "critical editions" of the great works of the middle ages, and thecritical text of some ancient classics (Pausanias, for example) hasstill to be constructed. 

Not all historical documents have as yet been published in a formcalculated to give historians the security they need, and somehistorians still act as if they had not realised that an unsettled text, as such, requires cautious handling. Still, considerable progress hasbeen made. From the experience accumulated by several generations ofscholars there has been evolved a recognised method of purifying andrestoring texts. No part of historical method has a more solidfoundation, or is more generally known. It is clearly explained inseveral works of popular philology. [68] For this reason we shall here becontent to give a general view of its essential principles, and toindicate its results. 

I. We will suppose a document has not been edited in conformity withcritical rules. How are we to proceed in order to construct the bestpossible text? Three cases present themselves. 

(_a_) The most simple case is that in which we possess the original, theauthor's autograph itself. There is then nothing to do but to reproducethe text of it with absolute fidelity. [69] Theoretically nothing can beeasier; in practice this elementary operation demands a sustainedattention of which not every one is capable. If any one doubts it, lethim try. Copyists who never make mistakes and never allow theirattention to be distracted are rare even among scholars. 

(_b_) Second case. The original has been lost; only a single copy of itis known. It is necessary to be cautious, for the probability is thatthis copy contains errors. 

Texts degenerate in accordance with certain laws. A great deal of painshas been taken to discover and classify the causes and the ordinaryforms of the differences which are observed between originals andcopies; and hence rules have been deduced which may be applied to theconjectural restoration of those passages in a unique copy of a lostoriginal which are certainly corrupt (because unintelligible), or are soin all probability. 

Alterations of an original occurring in a copy--"traditional variants, "as they are called--are due either to fraud or to error. Some copyistshave deliberately modified or suppressed passages. [70] Nearly allcopyists have committed errors of judgment or accidental errors. Errorsof judgment when half-educated and not wholly intelligent copyists havethought it their duty to correct passages and words in the originalwhich they could not understand. [71] Accidental errors when they misreadwhile copying, or misheard while writing from dictation, or when theyinvoluntarily made slips of the pen. 

Modifications arising from fraud or errors of judgment are often verydifficult to rectify, or even to discover. Some accidental errors (theomission of several lines, for example) are irreparable in the case weare considering, that of a unique copy. But most accidental errors canbe detected by any one who knows the ordinary forms: confusions ofsense, letters, and words, transpositions of words, letters, andsyllables, dittography (unmeaning repetition of letters or syllables), haplography (syllables or words written once only where they should havebeen written twice), false divisions between words, badly punctuatedsentences, and other mistakes of the same kind. Errors of these varioustypes have been made by the scribes of every country and every age, irrespectively of the handwriting and language of the originals. Butsome confusions of letters occur frequently in copies of uncialoriginals, and others in copies of minuscule originals. Confusions ofsense and of words are explained by analogies of vocabulary orpronunciation, which naturally vary from language to language and fromepoch to epoch. The general theory of conjectural emendation reduces tothe sketch we have just given; there is no general apprenticeship to theart. What a man learns is not to restore any text that may be put beforehim, but Greek texts, Latin texts, French texts, and so on, as the casemay be; for the conjectural emendation of a text presupposes, besidesgeneral notions on the processes by which texts degenerate, a profoundknowledge of (1) a special language; (2) a special handwriting; (3) _theconfusions (of sense, letters, and words) which were habitual to thosewho copied texts of that language written in that style of handwriting_. To aid in the apprenticeship to the conjectural emendation of Greek andLatin texts, tabulated lists (alphabetical and systematic) of variousreadings, frequent confusions, and probable corrections, have beendrawn up. [72] It is true that they cannot take the place of practicalwork, done under the guidance of experts, but they are of very great useto the experts themselves. [73]

It would be easy to give a list of happy emendations. The mostsatisfactory are those whose correctness is obvious palæographically, asis the case with the classical emendation by Madvig of the text ofSeneca's Letters (89, 4). The old reading was: "Philosophia unde dictasit, apparet; ipso enim nomine fatetur. Quidam et sapientiam ita quidamfinierunt, ut dicerent divinorum et humanorum sapientiam . .. "--whichdoes not make sense. It used to be supposed that words had dropped outbetween _ita_ and _quidam_. Madvig pictured to himself the text of thelost archetype, which was written in capitals, and in which, as wasusual before the eighth century, the words were not separated (_scriptiocontinua_), nor the sentences punctuated; he asked himself whether thecopyist, with such an archetype before him, had not divided the words atrandom, and he had no difficulty in reading: ". .. Ipso enim nominefatetur quid amet. Sapientiam ita quidam finierunt. .. . " Blass, Reinach, and Lindsay, in the works referred to in the note, mention several othermasterly and elegant emendations. Nor have the Hellenists and Latinistsany monopoly; equally brilliant emendations might be culled from theworks of Orientalists, Romancists, and Germanists, now that texts ofOriental, Romance, and Germanic languages have been subjected to verbalcriticism. We have already stated that scholarly corrections arepossible even in the text of quite modern documents, reproducedtypographically under the most favourable conditions. 

Perhaps no one, in our day, has equalled Madvig in the art ofconjectural emendation. But Madvig himself had no high opinion of thework of modern scholarship. He thought that the humanists of thesixteenth and seventeenth centuries were, in this respect, bettertrained than modern scholars. The conjectural emendation of Greek andLatin texts is, in fact, a branch of sport, success in which isproportionate not only to a man's ingenuity and palæographical instinct, but also to the correctness, rapidity, and delicacy of his appreciationof the niceties of the classical languages. Now, the early scholars wereundoubtedly too bold, but they were more intimately familiar with theclassical languages than our modern scholars are. 

However that may be, there can be no doubt that numerous texts whichhave been preserved, in corrupt form, in unique copies, have resisted, and will continue to resist, the efforts of criticism. Very oftencriticism ascertains the fact of the text having been altered, stateswhat the sense requires, and then prudently stops, every trace of theoriginal reading having been obscured by a confused tangle of successivecorrections and errors which it is hopeless to attempt to unravel. Thescholars who devote themselves to the fascinating pursuit of conjecturalcriticism are liable, in their ardour, to suspect perfectly innocentreadings, and, in desperate passages, to propose adventurous hypotheses. They are well aware of this, and therefore make it a rule to draw a veryclear distinction, in their editions, between readings found inmanuscripts and their own restorations of the text. 

(_c_) Third case. We possess several copies, which differ from eachother, of a document whose original is lost. Here modern scholars have amarked advantage over their predecessors: besides being better informed, they set about the comparison of copies more methodically. The objectis, as in the preceding case, to reconstruct the archetype as exactly aspossible. 

The scholars of earlier days had to struggle, as novices have tostruggle now, in a case of this kind, against a very natural and a veryreprehensible impulse--to use the first copy that comes to hand, whatever its character may happen to be. The second impulse is not muchbetter--to use the oldest copy out of several of different date. Intheory, and very often in practice, the relative age of the copies is ofno importance; a sixteenth-century manuscript which reproduces a goodlost copy of the eleventh century is much more valuable than a faultyand retouched copy made in the twelfth or thirteenth century. The thirdimpulse is still far from being good; it is to count the attestedreadings and decide by the majority. Suppose there are twenty copies ofa text; the reading A is attested eighteen times, the reading B twice. To make this a reason for choosing A is to make the gratuitousassumption that all the manuscripts have the same authority. This is anerror of judgment; for if seventeen of the eighteen manuscripts whichgive the reading A have been copied from the eighteenth, the reading Ais in reality attested only once; and the only question is whether it isintrinsically better or worse than the reading B. 

It has been recognised that the only rational procedure is to begin bydetermining in what relation the copies stand to each other. For thispurpose we adopt as our starting-point the incontrovertible axiom thatall the copies which contain the same mistakes in the same passages musthave been either copied from each other or all derived from a copycontaining those mistakes. It is inconceivable that several copyists, independently reproducing an original free from errors, should allintroduce exactly the same errors; identity of errors attests communityof origin. We shall cast aside without scruple all the copies derivedfrom a single manuscript which has been preserved. Evidently they canhave no value beyond what is possessed by their common source; if theydiffer from it, it can only be in virtue of new errors; it would bewaste of time to study their variations. Having eliminated these, wehave before us none but independent copies, which have been madedirectly from the archetype, or secondary copies whose source (a copytaken directly from the archetype) has been lost. In order to group thesecondary copies into _families_, each of which shall represent what issubstantially the same tradition, we again have recourse to thecomparison of errors. By this method we can generally draw up withouttoo much trouble a complete genealogical table (_stemma codicum_) of thepreserved copies, which will bring out very clearly their relativeimportance. This is not the place to discuss the difficult cases where, in consequence of too great a number of intermediaries having been lost, or from ancient copyists having arbitrarily blended the texts ofdifferent traditions, the operation becomes extremely laborious orimpracticable. Besides, in these extreme cases there is no new methodinvolved: the comparison of corresponding passages is a powerfulinstrument, but it is the only one which criticism has at its disposalfor this task. 

When the genealogical tree of the manuscripts has been drawn up, weendeavour to restore the text of the archetype by comparing thedifferent traditions. If these agree and give a satisfactory text, thereis no difficulty. If they differ, we decide between them. If theyaccidentally agree in giving a defective text, we have recourse toconjectural emendation, as if there were only one copy. 

It is, theoretically, much more advantageous to have several independentcopies of a lost original than to have only one, for the mere mechanicalcomparison of the different readings is often enough to removeobscurities which the uncertain light of conjectural criticism wouldnever have illuminated. However, an abundance of manuscripts is anembarrassment rather than a help when the work of grouping them has beenleft undone or done badly; nothing can be more unsatisfactory than thearbitrary and hybrid restorations which are founded on copies whoserelations to each other and to the archetype have not been ascertainedbeforehand. On the other hand, the application of rational methodsrequires, in some cases, a formidable expenditure of time and labour. Some works are preserved in hundreds of copies all differing from eachother; sometimes (as in the case of the Gospels) the variants of a textof quite moderate extent are to be counted by thousands; several yearsof assiduous labour are necessary for the preparation of a criticaledition of some mediæval romances. And after all this labour, all thesecollations and comparisons, can we be sure that the text of the romanceis sensibly better than it would have been if there had been only two orthree manuscripts to work upon? No. Some critical editions, owing to theapparent wealth of material applicable to the work, demand a mechanicaleffort which is altogether out of proportion to the positive resultswhich are its reward. 

"Critical editions" founded on several copies of a lost original oughtto supply the public with the means of verifying the "_stemma codicum_"which the editor has drawn up, and should give the rejected variants inthe notes. By this means competent readers are, at the worst, put inpossession, if not of the best possible text, at least of the materialsfor constructing it. [74]

II. The results of textual criticism--a kind of cleaning andmending--are purely negative. By the aid of conjecture, or by the aid ofconjecture and comparison combined, we are enabled to construct, notnecessarily a good text, but the best text possible, of documents whoseoriginal is lost. What we thus effect is the elimination of corrupt andadventitious readings likely to cause error, and the recognition ofsuspected passages as such. But it is obvious that no new information issupplied by this process. The text of a document which has been restoredat the cost of infinite pains is not worth more than that of a documentwhose original has been preserved; on the contrary, it is worth less. Ifthe autograph manuscript of the Æneid had not been destroyed, centuriesof collation and conjecture would have been saved, and the text of theÆneid would have been better than it is. This is intended for those whoexcel at the "emendation game, "[75] who are in consequence fond of it, and would really be sorry to have no occasion to play it. 

III. There will, however, be abundant scope for textual criticism aslong as we do not possess the exact text of every historical document. In the present state of science few labours are more useful than thosewhich bring new texts to light or improve texts already known. It is areal service to the study of history to publish unedited or badly editedtexts in a manner conformable to the rules of criticism. In everycountry learned societies without number are devoting the greater partof their resources and activity to this important work. But the immensenumber of the texts to be criticised, [76] and the minute care requiredby the operations of verbal criticism, [77] prevent the work ofpublication and restoration from advancing at any but a slow pace. Before all the texts which are of interest for mediæval and modernhistory shall have been edited or re-edited _secundum artem_, a longperiod must elapse, even supposing that the relatively rapid pace of thelast few years should be still further accelerated. [78]

CHAPTER III

CRITICAL INVESTIGATION OF AUTHORSHIP

It would be absurd to look for information about a fact in the papers ofsome one who knew nothing, and could know nothing, about it. The firstquestions, then, which we ask when we are confronted with a document is:Where does it come from? who is the author of it? what is its date? Adocument in respect of which we necessarily are in total ignorance ofthe author, the place, and the date is good for nothing. 

This truth, which seems elementary, has only been adequately recognisedin our own day. Such is the natural [Greek: hakrishia] of man, that thosewho were the first to make a habit of inquiring into the authorship ofdocuments prided themselves, and justly, on the advance they had made. 

Most modern documents contain a precise indication of their authorship:in our days, books, newspaper articles, official papers, and evenprivate writings, are, in general, dated and signed. Many ancientdocuments, on the other hand, are anonymous, without date, and have nosufficient indication of their place of origin. 

The spontaneous tendency of the human mind is to place confidence in theindications of authorship, when there are any. On the cover and in thepreface of the _Châtiments_, Victor Hugo is named as the author;therefore Victor Hugo is the author of the _Châtiments_. In such andsuch a picture gallery we see an unsigned picture whose frame has beenfurnished by the management with a tablet bearing the name of Leonardoda Vinci; therefore Leonardo da Vinci painted this picture. A poem withthe title _Philomena_ is found under the name of Saint Bonaventura in M. Clément's _Extraits des poètes chrétiens_, in most editions of SaintBonaventura's "works, " and in a great number of mediæval manuscripts;therefore _Philomena_ was written by Saint Bonaventura, and "we maygather thence much precious knowledge of the very soul" of this holyman. [79] Vrain-Lucas offered to M. Chasles autographs of Vercingetorix, Cleopatra, and Saint Mary Magdalene, duly signed, and with theflourishes complete:[80] here, thought M. Chasles, are autographs ofVercingetorix, Cleopatra, and Saint Mary Magdalene. This is one of themost universal, and at the same time indestructible, forms of publiccredulity. 

Experience and reflection have shown the necessity of methodicallychecking these instinctive impulses of confiding trust. The autographsof Vercingetorix, Cleopatra, and Mary Magdalene had been manufactured byVrain-Lucas. The _Philomena_, attributed by mediæval scribes now toSaint Bonaventura, now to Louis of Granada, now to John Hoveden, now toJohn Peckham, is perhaps by none of these authors, and certainly not bythe first-named. Paintings in which there is not the least gleam oftalent have, in the most celebrated galleries of Italy, been trickedout, without the least shadow of proof, with the glorious name ofLeonardo. On the other hand, it is perfectly true that Victor Hugo isthe author of the _Châtiments_. The conclusion is, that the most preciseindications of authorship are never sufficient _by themselves_. Theyonly afford a presumption, strong or weak--very strong, in general, where modern documents are concerned, often very weak in the case ofancient documents. False indications of authorship exist, some foistedupon insignificant works in order to enhance their value, some appendedto works of merit in order to serve the reputation of a particularperson, or to mystify posterity; and there are a hundred other motiveswhich may easily be imagined, and of which a list has been drawn up:[81]the "pseudepigraphic" literature of antiquity and the middle ages isenormous. There are, in addition, documents which are forged frombeginning to end; the forgers have naturally furnished them with veryprecise indications of their alleged authorship. Verification istherefore necessary. But how is it to be had? When the apparentauthorship of a document is suspected, we use for its verification thesame method which serves to fix, as far as possible, the origin ofdocuments which are furnished with no indications at all on this head. As the procedure is the same in both cases, it is not necessary todistinguish further between them. 

I. The chief instrument used in the investigation of authorship is the_internal analysis_ of the document under consideration, performed witha view to bring out any indications it may contain of a nature to supplyinformation about the author, and the time and place in which he lived. 

First of all we examine the handwriting of the document. SaintBonaventura was born in 1221; if poems attributed to him are containedin manuscripts executed in the eleventh century, we have in thiscircumstance an excellent proof that the attribution is ill-founded: nodocument of which there exists a copy in eleventh-century handwritingcan be posterior in date to the eleventh century. Then we examine thelanguage. It is known that certain forms have only been used in certainplaces and at certain dates. Most forgers have betrayed themselves byignorance of facts of this kind; they let slip modern words or phrases. It has been possible to establish the fact that certain Phoenicianinscriptions, found in South America, were earlier than a certain Germandissertation on a point of Phoenician syntax. In the case of officialinstruments we examine the formulæ. If a document which purports to be aMerovingian charter does not exhibit the ordinary formulæ of genuineMerovingian charters it must be spurious. Lastly, we note all thepositive data which occur in the document--the facts which are mentionedor alluded to. When these facts are otherwise known, from sources whicha forger could not have had at his disposal, the _bonâ fides_ of thedocument is established, and the date fixed approximately between themost recent event of which the author shows knowledge, and the nextfollowing event which he does not mention but would have done if he hadknown of it. Arguments may also be founded on the circumstance thatparticular facts are mentioned with approval, or particular opinionsexpressed, and help us to make a conjectural estimate of the status, theenvironment, and the character of the author. 

When the internal analysis of a document is carefully performed, itgenerally gives us a tolerably accurate notion of its authorship. Bymeans of a methodical comparison, instituted between the variouselements of the documents analysed and the corresponding elements ofsimilar documents whose authorship was known with certainty, thedetection of many a forgery[82] has been rendered possible, andadditional information acquired about the circumstances under which mostgenuine documents have been produced. 

The results obtained by internal analysis are supplemented and verifiedby collecting all the external evidence relative to the document undercriticism which can be found scattered over the documents of the same orlater epochs--quotations, biographical details about the author, and soon. Sometimes there is a significant absence of any such information:the fact that an alleged Merovingian charter has not been quoted byanybody before the seventeenth century, and has only been seen by aseventeenth-century scholar who has been convicted of fraud, suggeststhe thought that it is modern. 

II. Hitherto we have considered only the simplest case, in which thedocument under examination is the work of a single author. But manydocuments have, at different times, received additions which it isimportant to distinguish from the original text, in order that we maynot attribute to X, the author of the text, what really belongs to Y orZ, his unforeseen collaborators. [83] There are two kinds ofadditions--interpolations and continuations. To interpolate is to insertinto the text words or sentences which were not in the author'smanuscript. [84] Usually interpolations are accidental, due to thenegligence of the copyist, and explicable as the introduction into thetext of interlinear glosses or marginal notes; but there are cases wheresome one has deliberately added to (or substituted for) the author'stext words or sentences out of his own head, for the sake ofcompleteness, ornament, or emphasis. If we had before us the manuscriptin which the deliberate interpolation was made, the appearance of theadded matter and the traces of erasure would make the case clear atonce. But the first interpolated copy has nearly always been lost, andin the copies derived from it every trace of addition or substitutionhas disappeared. There is no need to define "continuations. " It is wellknown that many chronicles of the middle ages have been "continued" byvarious writers, none of whom took the trouble to indicate where his ownwork began or ended. 

Sometimes interpolations and continuations can be very readilydistinguished in the course of the operations for restoring a text ofwhich there are several copies, when it so happens that some of thesecopies reproduce the primitive text as it was before any addition wasmade to it. But if all the copies are founded on previous copies whichalready contained the interpolations or continuations, recourse must behad to internal analysis. Is the style uniform throughout the document?Does the book breathe one and the same spirit from cover to cover? Arethere no contradictions, no gaps in the sequence of ideas? In practice, when the continuators or interpolators have been men of well-markedpersonality and decided views, analysis will separate the original fromthe additions as cleanly as a pair of scissors. When the whole iswritten in a level, colourless style, the lines of division are not soeasy to see; it is then better to confess the fact than to multiplyhypotheses. 

III. The critical investigation of authorship is not finished as soon asa document has been accurately or approximately localised in space andtime, and as much information as possible obtained about the author orauthors. [85] Here is a book: we wish to ascertain the origin of theinformation contained in it, that is, to be in a position to appreciateits value; is it enough to know that it was written in 1890, at Paris, by So-and-so? Perhaps So-and-so copied slavishly, without mentioning thefact, an earlier work, written in 1850. The responsible guarantor of theborrowed parts is not So-and-so, but the author of 1850. Plagiarism, itis true, is now rare, forbidden by the law, and considereddishonourable; formerly it was common, tolerated, and unpunished. Manyhistorical documents, with every appearance of originality, are nothingbut unavowed repetitions of earlier documents, and historiansoccasionally experience, in this connection, remarkable disillusions. Certain passages in Eginhard, a ninth-century chronicler, are borrowedfrom Suetonius: they have nothing to do with the history of the ninthcentury; how if the fact had not been discovered? An event is attestedthree times, by three chroniclers; but these three attestations, whichagree so admirably, are really only one if it is ascertained that two ofthe three chroniclers copied the third, or that the three parallelaccounts have been drawn from one and the same source. Pontificalletters and Imperial charters of the middle ages contain eloquentpassages which must not be taken seriously; they are part of theofficial style, and were copied word for word from chancery formularies. 

It belongs to the investigation of authorship to discover, as far aspossible, the _sources_ utilised by the authors of documents. 

The problem thus presented to us has some resemblance to that of therestoration of texts of which we have already spoken. In both cases weproceed on the assumption that identical readings have a common source:a number of different scribes, in transcribing a text, will not makeexactly the same mistakes in exactly the same places; a number ofdifferent writers, relating the same facts, will not have viewed themfrom exactly the same standpoint, nor will they say the same things inexactly the same language. The great complexity of historical eventsmakes it extremely improbable that two independent observers shouldnarrate them in the same manner. We endeavour to group the documentsinto families in the same way as we make families of manuscripts. Similarly, we are enabled in the result to draw up genealogical tables. The examiners who correct the compositions of candidates for thebachelor's degree sometimes notice that the papers of two candidates whosat next each other bear a family likeness. If they have a mind to findout which is derived from the other, they have no difficulty in doingso, in spite of the petty artifices (slight modifications, expansions, abstracts, additions, suppressions, transpositions) which the plagiaristmultiplies in order to throw suspicion off the scent The two guilty onesare sufficiently betrayed by their common errors; the more culpable ofthe two is detected by the slips he will have made, and especially bythe errors in his own papers which are due to peculiarities in those ofhis accommodating friend. Similarly when two ancient documents are inquestion: when the author of one has copied directly from the other, thefiliation is generally easy to establish; the plagiarist, whether heabridges or expands, nearly always betrays himself sooner or later. [86]

When there are three documents in a family their mutual relationshipsare sometimes harder to specify. Let A, B, and C be the documents. Suppose A is the common source: perhaps B and C copied it independently;perhaps C only knew A through the medium of B, or B knew it only throughC. If B and C have abridged the common source in different ways, theyare evidently independent. When B depends on C, or _vice versâ_, we havethe simplest case, treated in the preceding paragraph. But suppose theauthor of C combined A and B, while B had already used A: the genealogybegins to get complicated. It is more complicated still when there arefour, five, or more documents in a family, for the number of possiblecombinations increases with great rapidity. However, if too manyintermediate links have not been lost, criticism succeeds indisentangling the relationships by persistent and ingenious applicationsof the method of repeated comparisons. Modern scholars (Krusch, forexample, who has made a speciality of Merovingian hagiography) haverecently constructed, by the use of this method, precise genealogies ofthe utmost solidity. [87] The results of the critical investigation ofauthorship, as applied to the filiation of documents, are of two kinds. Firstly, lost documents are reconstructed. Suppose two chroniclers, Band C, have used, each in his own way, a common source X, which has nowdisappeared. We may form an idea of X by piecing together the fragmentsof it which occur imbedded in B and C, just as we form an idea of a lostmanuscript by comparing the partial copies of it which have beenpreserved. On the other hand, criticism destroys the authority of a hostof "authentic" documents--that is, documents which no one suspects ofhaving been falsified--by showing that they are derivative, that theyare worth whatever their sources may be worth, and that, when theyembellish their sources with imaginary details and rhetoricalflourishes, they are worth just nothing at all. In Germany and Englandeditors of documents have introduced the excellent system of printingborrowed passages in small characters, and original passages whosesource is unknown in larger characters. Thanks to this system it ispossible to see at a glance that celebrated chronicles, which are often(very wrongly) quoted, are mere compilations, of no value in themselves:thus the _Flores historiarum_ of the self-styled Matthew of Westminster, perhaps the most popular of the English mediæval chronicles, are almostentirely taken from original works by Wendover and Matthew of Paris. [88]

IV. The critical investigation of authorship saves historians from hugeblunders. Its results are striking. By eliminating spurious documents, by detecting false ascriptions, by determining the conditions ofproduction of documents which had been defaced by time, and byconnecting them with their sources, [89] it has rendered services of suchmagnitude that to-day it is regarded as having a special right to thename of "criticism. " It is usual to say of an historian that he "failsin criticism" when he neglects to distinguish between documents, when henever mistrusts traditional ascriptions, and when he accepts, as ifafraid to lose a single one, all the pieces of information, ancient ormodern, good or bad, which come to him, from whatever quarter. [90]

This view is perfectly just. We must not, however, be satisfied withthis form of criticism, and we must not abuse it. 

We must not abuse it. The extreme of distrust, in these matters, isalmost as mischievous as the extreme of credulity. Père Hardouin, whoattributed the works of Vergil and Horace to mediæval monks, was everywhit as ridiculous as the victim of Vrain-Lucas. It is an abuse of themethods of this species of criticism to apply them, as has been done, indiscriminately, for the mere pleasure of it. The bunglers who haveused this species of criticism to brand as spurious perfectly genuinedocuments, such as the writings of Hroswitha, the _Ligurinus_, and thebull _Unam Sanctam_, [91] or to establish imaginary filiations betweencertain annals, on the strength of superficial indications, would havediscredited criticism before now if that had been possible. It ispraiseworthy, certainly, to react against those who never raise a doubtabout the authorship of a document; but it is carrying the reaction toofar to take an exclusive interest in periods of history which depend ondocuments of uncertain authorship. The only reason why the documents ofmodern and contemporary history are found less interesting than those ofantiquity and the early middle ages, is that the identity which nearlyalways obtains between their apparent and their real authorship leavesno room for those knotty problems of attribution in which the _virtuosi_of criticism are accustomed to display their skill. [92]

Nor must we be content with it. The critical investigation ofauthorship, like textual criticism, is preparatory, and its resultsnegative. Its final aim and crowning achievement is to get rid ofdocuments which are not documents, and which would have misled us; thatis all. "It teaches us not to use bad documents; it does not teach ushow to turn good ones to account. "[93] It is not the whole of"historical criticism;" it is only one stone in the edifice. [94]

CHAPTER IV

CRITICAL CLASSIFICATION OF SOURCES

By the help of the preceding operations the documents, all thedocuments, let us suppose, of a given class, or relating to a givensubject, have been found. We know where they are; the text of each hasbeen restored, if necessary, and each has been critically examined inrespect of authorship. We know where they have come from. It remains tocombine and classify the materials thus verified. This is the last ofthe operations which may be called preparatory to the work of higher (orinternal) criticism and construction. 

Whoever studies a point of history is obliged, first of all, to classifyhis sources. To arrange, in a rational and convenient manner, theverified materials before making use of them, is an apparently humble, but really very important, part of the historian's profession. Those whohave learnt how to do it possess, on that account alone, a markedadvantage: they give themselves less trouble, and they obtain betterresults; the others waste their time and labour; they are smotheredsometimes under the disorderly mass of notes, extracts, copies, scraps, which they themselves have accumulated. Who was it spoke of those busypeople who spend their lives lifting building-stones without knowingwhere to place them, raising as they do so clouds of blinding dust?

I. Here, again, we have to confess that the first, the natural impulse, is not the right one. The first impulse of most men who have to utilisea number of texts is to make notes from them, one after another, in theorder in which they study them. Many of the early scholars (whose paperswe possess) worked on this system, and so do most beginners who are notwarned beforehand; the latter keep, as the former kept, note-books, which they fill continuously and progressively with notes on the textsthey are interested in. This method is utterly wrong. The materialscollected must be classified sooner or later; otherwise it would benecessary, when occasion arose, to deal separately with the materialsbearing on a given point, to read right through the whole series ofnote-books, and this laborious process would have to be repeated everytime a new detail was wanted. If this method seems attractive at first, it is because it appears to save time. But this is false economy; theultimate result is, an enormous addition to the labour of search, andgreat difficulty in combining the materials. 

Others, well understanding the advantages of systematic classification, have proposed to fit their materials, as fast as collected, into theirappropriate places in a prearranged scheme. For this purpose they usenote-books of which every page has first been provided with a heading. Thus all the entries of the same kind are close to one another. Thissystem leaves something to be desired; for additions will not alwaysfit without inconvenience into their proper place; and the scheme ofclassification, once adopted, is rigid, and can only be modified withdifficulty. Many librarians used to draw up their catalogues on thisplan, which is now universally condemned. 

There is a still more barbarous method, which need not receive more thanpassing mention. This is simply to register documents in the memorywithout taking written notes. This method has been used. Historiansendowed with excellent memories, and lazy to boot, have indulged thiswhim, with the result that their quotations and references are mostlyinexact. The human memory is a delicate piece of registering apparatus, but it is so little an instrument of precision that such presumption isinexcusable. 

Every one admits nowadays that it is advisable to collect materials onseparate cards or slips of paper. The notes from each document areentered upon a loose leaf furnished with the precisest possibleindications of origin. The advantages of this artifice are obvious: thedetachability of the slips enables us to group them at will in a host ofdifferent combinations; if necessary, to change their places: it is easyto bring texts of the same kind together, and to incorporate additions, as they are acquired, in the interior of the groups to which theybelong. As for documents which are interesting from several points ofview, and which ought to appear in several groups, it is sufficient toenter them several times over on different slips; or they may berepresented, as often as may be required, on reference-slips. Moreover, the method of slips is the only one mechanically possible for thepurpose of forming, classifying, and utilising a collection of documentsof any great extent. Statisticians, financiers, and men of letters whoobserve, have now discovered this as well as scholars. 

The method of slips is not without its drawbacks. Each slip ought to befurnished with precise references to the source from which its contentshave been derived; consequently, if a document has been analysed uponfifty different slips, the same references must be repeated fifty times. Hence a slight increase in the amount of writing to be done. It iscertainly on account of this trivial complication that some obstinatelycling to the inferior note-book system. Again, in virtue of their verydetachability, the slips, or loose leaves, are liable to go astray; andwhen a slip is lost how is it to be replaced? To begin with, itsdisappearance is not perceived, and, if it were, the only remedy wouldbe to go right through all the work already done from beginning to end. But the truth is, experience has suggested a variety of very simpleprecautions, which we need not here explain in detail, by which thedrawbacks of the system are reduced to a minimum. It is recommended touse slips of uniform size and tough material, and to arrange them at theearliest opportunity in covers or drawers or otherwise. Every one isfree to form his own habits in these matters. But it is well to realisebeforehand that these habits, according as they are more or lessrational and practical, have a direct influence on the results ofscientific work. Renan speaks of "these points of private librarianshipwhich make up the half of scientific work. "[95] This is not too strong. One scholar will owe a good part of his well-deserved reputation to hismethod of collecting, while another will be, so to speak, paralysed byhis clumsiness in that particular. [96]

After having collected the documents, whether copied _in extenso_ orabridged, on slips or loose leaves, we classify them. On what scheme? Inwhat order? Clearly different cases must be treated differently, and itwould not be reasonable to lay down precise formulæ to govern them all. However, we may give a few general considerations. 

II. We distinguish between the historian who classifies verifieddocuments for the purposes of historical work, and the scholar whocompiles "_Regesta_. " By the words "_Regesta_" and "_Corpus_" weunderstand methodically classified collections of historical documents. In a "_Corpus_" documents are reproduced _in extenso_; in "_Regesta_"they are analysed and described. 

The use of these compilations is to assist researchers in collectingdocuments. Scholars set themselves to perform, once for all, tasks ofsearch and classification from which, thanks to them, the public willhenceforth be free. 

Documents may be grouped according to their date, according to theirplace of origin, according to their contents, according to theirform. [97] Here we have the four categories of time, place, species, andform; by superposing, then, we obtain divisions of smaller extent. Wemay undertake, for example, to make a group of all the documents havinga given form, of a given country, and lying between two given dates(French royal charters of the reign of Philip Augustus); or of all thedocuments of a given form (Latin inscriptions); or of a given species(Latin hymns); of a given epoch (antiquity, the middle ages). We mayrecall, by way of illustration, the existence of a _Corpus InscriptionumGræcarum_, of a _Corpus Inscriptionum Latinarum_, of a _CorpusScriptorum Ecclesiasticorum Latinorum_, the _Regesta Imperii_ of J. F. Böhmer and his continuators, the _Regesta Pontificum Romanorum_ of P. Jaffé and A. Potthast. 

Whatever the division chosen, there are two alternatives: either thedocuments to be placed in this division are dated or they are not. 

If they are dated, as is the case, for example, with the charters issuedfrom the chancery of a prince, care will have been taken to place at thehead of each slip the date (expressed in modern reckoning) of thedocument entered upon it. Nothing is then easier than to group inchronological order all the slips, that is, all the documents, whichhave been collected. The rule is to use chronological classificationwhenever possible. There is only one difficulty, and that is of apractical order. Even in the most favourable circumstances some of thedocuments will have accidentally lost their dates; these dates thecompiler is bound to restore, or at least to attempt to restore; longand patient research is necessary for the purpose. 

If the documents are not dated, a choice must be made between thealphabetical, the geographical, and the systematic order. The history ofthe _Corpus_ of Latin inscriptions bears witness to the difficulty ofthis choice. "The arrangement according to date was impossible, seeingthat most of the inscriptions are not dated. From the time of Smetius itwas usual to divide them into classes, that is, a distinction was made, resting solely on the contents of the inscription, and having no regardto their place of origin, between religious, sepulchral, military, andpoetical inscriptions, those which have a public character, and thosewhich only concern private persons, and so on. Boeckh, although he hadpreferred the geographical arrangement for his _Corpus InscriptionumGræcarum_, was of opinion that the arrangement by subjects, which hadbeen hitherto employed, was the only possible one for a Latin_Corpus_. .. . " [Even those who, in France, proposed the geographicalarrangement] "wished to make an exception of texts relating to thegeneral history of a country, certainly, at any rate, in the case of theEmpire; in 1845 Zumpt defended a very complicated eclectic system ofthis kind. In 1847 Mommsen still rejected the geographical arrangementexcept for municipal inscriptions, and in 1852, when he published theInscriptions of the Kingdom of Naples, he had not entirely changed hisopinion. It was only on being charged by the Academy of Berlin with thepublication of the _Corpus Inscriptionum Latinarum_, that, grown wise byexperience, he rejected even the exceptions proposed by Egger in thecase of the general history of a province, and thought it his duty tokeep to the geographical arrangement pure and simple. "[98] And yet, considering the nature of epigraphic documents, the arrangementaccording to place was the only rational one. This has been amplydemonstrated for more than fifty years; but collectors of inscriptionsdid not come to an agreement on the subject till after two centuries oftentative efforts in different directions. For two centuries collectionsof Latin inscriptions have been made without any perception of the factthat "to group inscriptions according to their subjects is much the samething as to publish an edition of Cicero in which his speeches, treatises, and letters should be cut up and the fragments arrangedaccording to their subject-matter;" that "epigraphic monuments belongingto the same territory mutually explain each other when placed side byside;" and, lastly, that "while it is all but impossible to range inorder of subject-matter a hundred thousand inscriptions nearly all ofwhich belong to several categories; on the other hand, each monument hasbut one place, and a very definite place, in the geographicalorder. "[99]

The alphabetical arrangement is very convenient when the chronologicaland geographical arrangements are unsuitable. There are documents, suchas the sermons, the hymns, and the secular songs of the middle ages, which are not precisely dated or localised. They are arranged in thealphabetical order of their _incipit_--that is, the words with whichthey begin. [100]

The systematic order, or arrangement by subjects, is not to berecommended for the compilation of a _Corpus_ or of _regesta_. It isalways arbitrary, and leads to inevitable repetition and confusion. Besides, given collections arranged in chronological, geographical, oralphabetical order, nothing more than the addition of a good table ofcontents is needed to make them available for all the purposes whichwould be served by a systematic arrangement. One of the chief rules ofthe art of _Corpus_ and _regesta_-making, that great art which has beencarried to such perfection in the second half of the nineteenthcentury, [101] is to provide these collections, whatever the groupingadopted, with a variety of tables and indexes of a kind to facilitatethe use of them: _incipit_ tables in chronological _regesta_ which lendthemselves to such treatment, indexes of names and dates in _regesta_arranged by order of _incipit_, and so on. 

_Corpus_ and _regesta_-makers collect and classify for the use of othersdocuments in which, at any rate in _all_ of which, they have no directinterest, and are absorbed in this labour. Ordinary workers, on theother hand, only collect and classify materials useful for theirindividual studies. Hence certain differences arise. For example, thearrangement by subjects, on a predetermined system, which is so littleto be recommended for great collections, often provides those who arecomposing monographs on their own account with a scheme ofclassification preferable to any other. But it will always be well tocultivate the mechanical habits of which professional compilers havelearnt the value by experience: to write at the head of every slip itsdate, if there is occasion for it, and a heading[102] in any case; tomultiply cross-references and indices; to keep a record, on a separateset of slips, of all the sources utilised, in order to avoid the dangerof having to work a second time through materials already dealt with. The regular observance of these maxims goes a great way towards makingscientific historical work easier and more solid. The possession of awell-arranged, though incomplete, collection of slips has enabled M. B. Hauréau to exhibit to the end of his life an undeniable mastery over thevery special class of historical problems which he studied. [103]

CHAPTER V

CRITICAL SCHOLARSHIP AND SCHOLARS

The sum of the operations described in the preceding chapters(restoration of texts, investigation of authorship, collection andclassification of verified documents) constitutes the vast domain ofexternal criticism, or critical scholarship. 

The public at large, with its vulgar and superficial standards, hasnothing but disdain for the whole of critical scholarship. Some of itsvotaries, on the other hand, are inclined to exalt it unduly. But thereis a happy medium between these extremes of over-appreciation andcontempt. 

The crude opinion of those who pity and despise the minute analysis ofexternal criticism hardly deserves refutation. There is only oneargument for the legitimacy and honourable character of the obscurelabours of erudition, but it is a decisive argument: it rests on theirindispensability. No erudition, no history. "_Non sunt contemnenda quasiparva_, " says St. Jerome, "_sine quibus magna constare nonpossunt_. "[104]

On the other hand, scholars by profession, in their zeal to justifytheir pride in their work, are not content with maintaining itsnecessity; they allow themselves to be carried away into an exaggerationof its merit and importance. It has been said that the sure methods ofexternal criticism have raised history to the dignity of a science, "ofan exact science;" that critical investigations of authorship "enableus, better than any other study, to gain a profound insight into pastages;" that the habit of criticising texts refines or even confers the"historical sense. " It has been tacitly assumed that external criticismis the whole of historical criticism, and that beyond the purgation, emendation, and classification of documents there is nothing left to do. This illusion, common enough among specialists, is too crude to needexpress refutation; the fact is, that it is the psychological criticismwhich deals with interpretation and examines into the good faith andaccuracy of authors that has, better than any other study, enabled us togain a profound insight into past ages, not external criticism. [105] Anhistorian who should be fortunate enough to find all the documentsbearing on his studies already edited correctly, classified, andcritically examined as to authorship, would be in just as good aposition to use them for writing history as if he had performed all thepreliminary operations himself. It is quite possible, whatever may besaid, to have the historical sense in full measure without having ever, both literally and figuratively, wiped away the dust from originaldocuments--that is, without having discovered and restored them foroneself. We need not interpret in the Jewish or etymological sense thedictum of Renan: "I do not think it possible for any one to acquire aclear notion of history, its limits, and the amount of confidence to beplaced in the different categories of historical investigation, unlesshe is in the habit of _handling_ original documents. "[106] This is to beunderstood as simply referring to the habit of going direct to thesources, and treating definite problems. [107] Without doubt a day willcome when all the documents relating to the history of classicalantiquity shall have been edited and treated critically. There will thenbe no more room, in this department of study, for textual criticism orthe investigation of sources; but, for all that, the conditions for thetreatment of general ancient history, or special parts of it, will bethen eminently favourable. External criticism, as we cannot too oftenrepeat, is entirely preparatory; it is a means, not an end; the idealstate of things would be that it should have been already sufficientlypractised that we might dispense with it for the future; it is only atemporary necessity. Theoretically, not only is it unnecessary for thosewho wish to make historical syntheses to do for themselves thepreparatory work on the materials which they use, but we have a right toask, as has been often asked, whether there is any advantage in theirdoing it. [108] Would it not be preferable that workers in the field ofhistory should specialise? On the one class--the specialists--woulddevolve the absorbing tasks of external or erudite criticism; theothers, relieved of the weight of these tasks, would have greaterliberty to devote themselves to the work of higher criticism, ofcombination and construction. Such was the opinion of Mark Pattison, whosaid, _History cannot be written from manuscripts_, which is as much asto say: "It is impossible for a man to write history from documentswhich he is obliged to put for himself into a condition in which theycan be used. "

Formerly the professions of "critical scholar" and "historian" were, infact, clearly distinguished. The "historians" cultivated the empty andpompous species of literature which then was known as "history, " withoutconsidering themselves bound to keep in touch with the work of thescholars. The latter, for their part, determined by their criticalresearches the conditions under which history must be written, but wereat no pains to write it themselves. Content to collect, emend, andclassify historical documents, they took no interest in history, andunderstood the past no better than did the mass of their contemporaries. The scholars acted as though erudition were an end in itself, and thehistorians as if they had been able to reconstruct vanished realities bythe mere force of reflection and ingenuity applied to the inferiordocuments, which were common property. So complete a divorce betweenerudition and history seems to-day almost inexplicable, and it was intruth mischievous enough. We need not say that the present advocates ofthe division of labour in history have nothing of the kind in view. Itis admittedly necessary that close relations should obtain between theworld of historians and that of critical scholars, for the work of thelatter has no reason for existence beyond its utility to the former. Allthat is meant is, that certain analytical and all synthetic operationsare not necessarily better performed when they are performed by the sameperson; that though the characters of historian and scholar may becombined, there is nothing illegitimate in their separation; and thatperhaps this separation is desirable in theory, as, in practice, it isoften a necessity. 

In practice, what happens is as follows. Whatever part of history a manundertakes to study, there are only three possible cases. In the firstthe sources have already been emended and classified; in the second thepreliminary work on the sources, which has been only partially done, ornot at all, offers no great difficulty; in the third the sources are ina very bad state, and require a great deal of labour to fit them foruse. We may observe, in passing, that there is naturally no proportionbetween the intrinsic importance of the subject and the amount ofpreliminary work which must be done before it can be treated: there aresome subjects of the highest interest, for example the history of theorigin and early development of Christianity, which could not beproperly attacked till after the completion of investigations whichoccupied several generations of scholars; but the material criticism ofthe sources of the history of the French Revolution, another subject ofthe first rank, gave much less trouble; and there are comparativelyunimportant problems in mediæval history which will not be solved tillafter an immense amount of external criticism shall have been performed. 

In the two first cases the expediency of a division of labour does notcome in question. But take the third case. A man of ability discoversthat the documents which are necessary for the treatment of a point ofhistory are in a very bad condition; they are scattered, corrupt, anduntrustworthy. He must take his choice; either he must abandon thesubject, having no taste for the mechanical operations which he knows tobe necessary, but which, as he foresees, would absorb the whole of hisenergy; or else he resolves to enter upon the preparatory critical work, without concealing from himself that in all probability he will neverhave time to utilise the materials he has verified, and that he willtherefore be working for those who will come after him. If he adopts thesecond alternative he becomes a critical scholar by profession, as itwere in spite of himself. _A priori_, it is true, there is nothing toprevent those who make great collections of texts and publish criticaleditions from using their own compilations and editions for the writingof history; and we see, as a matter of fact, that several men havedivided themselves between the preparatory tasks of external criticismand the more exalted labours of historical construction: it is enough tomention the names of Waitz, Mommsen, and Hauréau. But this combinationis very rare, for several reasons. The first is the shortness of life;there are catalogues, editions, _regesta_ on a great scale, theconstruction of which entails so much mechanical labour as to exhaustthe strength of the most zealous worker. The second is the fact that, for many persons, the tasks of critical scholarship are not withouttheir charm; nearly every one finds in them a singular satisfaction inthe long run; and some have confined themselves to these tasks whomight, strictly speaking, have aspired to higher things. 

Is it a good thing in itself that some workers should, voluntarily ornot, confine themselves to the researches of critical scholarship? Yes, without a doubt. In the study of history, the results of the division oflabour are the same as in the industrial arts, and highlysatisfactory--more abundant, more successful, better regulatedproduction. Critics who have been long habituated to the restoration oftexts restore them with incomparable dexterity and sureness; those whodevote themselves exclusively to investigations of authorship andsources have intuitions which would not occur to others less versed inthis difficult and highly specialised branch; those who have spenttheir lives in the construction of catalogues and the compilation of_regesta_ construct and compile them more easily, more quickly, andbetter than the man in the street. Thus, not only is there no specialreason for requiring every "historian" to be at the same time an activeworker in the field of critical scholarship, but even those scholars whoare engaged in the operations of external criticism come under differentcategories. Similarly, in a stoneyard there is no point in the architectbeing at the same time a workman, nor have all the workmen the samefunctions. Although most critical scholars have not rigorouslyspecialised so far, and although they vary their pleasures byvoluntarily executing different kinds of critical work, it would be easyto name some who are specialists in descriptive catalogues and indexes(archivists, librarians, and the like), others who are more particularly"critics" (purifiers, restorers, and editors of texts), and others whoare pre-eminently compilers of _regesta_. "The moment it is admittedthat erudition is only valuable for the sake of its results, it becomesimpossible to carry the division of scientific labour too far;"[109] andthe progress of the historical sciences corresponds to the narrower andnarrower specialisation of the workers. It was possible, not very longago, for the same man to devote himself successively to all theoperations of historical inquiry, but that was because he appealed to anot very exacting public: nowadays we require of those who criticisedocuments a minute accuracy and an absolute perfection which presupposereal professional skill. The historical sciences have now reached astage in their evolution at which the main lines have been traced, thegreat discoveries made, and nothing remains but a more precise treatmentof details. We feel instinctively that any further advance must be bydint of investigations of such extent, and analyses of such depth, asnone but specialists are capable of. 

But the best justification of the division of workers into "scholars"and "historians" (and of the distribution of the former among thevarious branches of external criticism) is to be found in the fact thatdifferent persons have a natural vocation for different tasks. One ofthe chief justifications of the institution of higher historicalteaching is, in our opinion, the opportunity afforded the teachers(presumably men of experience) of discerning in the students, in thecourse of their university career, either the germ of a vocation forcritical scholarship, or fundamental unfitness for critical work, as thecase may be. [110] _Criticus non fit, sed nascitur. _ For one who is notendowed by nature with certain aptitudes, a career of technicalerudition has nothing but disappointments in store: the greatest servicethat can be rendered to young men hesitating whether to adopt such acareer or not is to warn them of the fact. Those who hitherto havedevoted themselves to the preparatory tasks of criticism have eitherchosen them in preference to others because they had a taste for them, or else have submitted to them because they knew they were necessary;those who engaged in them by choice have less merit, from the ethicalpoint of view, than those who submitted to them, but, for all that, theyhave mostly obtained better results, because they have worked, not as amatter of duty, but joyfully and whole-heartedly. It is important thatevery one should realise the situation, and, in his own as well as thegeneral interest, embrace the special work which suits him best. 

We now propose to examine the natural aptitudes which fit, and the trulyprohibitory defects which disqualify, for the labours of externalcriticism. We shall, then, devote a few words to the effects produced onthe character by professional habituation to the labours of criticalscholarship. 

The chief condition of success in these labours is to like them. Thosewho are exceptionally gifted as poets or thinkers--that is, those whoare endowed with creative power--have much difficulty in adaptingthemselves to the technical drudgery of preparatory criticism: they arefar from despising it; on the contrary, they hold it in honour, if theyare clear-sighted; but they shrink from devoting themselves to it, forfear of using a razor, as is said, to cut stones. "I have no mind, "wrote Leibnitz to Basnage, who had exhorted him to compile an immense_Corpus_ of unpublished and printed documents relating to the history ofthe law of nations; "I have no mind to turn transcriber. .. . Does it notoccur to you that the advice you give me resembles that of a man whoshould wish to marry his friend to a shrew? For to engage a man in alifelong work is much the same as to find him a wife. "[111] And Renan, speaking of those immense preliminary labours "which have renderedpossible the researches of the higher criticism" and attempts athistorical construction, says: "The man who, with livelier intellectualneeds [than those of the men who performed these labours], should nowaccomplish such an act of abnegation, would be a hero. .. . "[112] AlthoughRenan directed the publication of the _Corpus InscriptionumSemiticarum_, and Leibnitz was the editor of the _Scriptores rerumBrunsvicensium_, neither Leibnitz, nor Renan, nor their peers have, fortunately, had the heroism to sacrifice their higher faculties topurely critical learning. 

Outside the class of superior men (and the infinitely more numerousclass of those who wrongly think themselves such), nearly every one, aswe have already said, finds in the long run a kind of satisfaction inthe minutiæ of preparatory criticism. The reason is, that the practiceof this criticism appeals to and develops two very widespreadtastes--the taste for collecting and the taste for puzzles. The pleasureof collecting is one which is felt not by children only, but by adultsas well, no matter whether the collection be one of various readings orof postage-stamps. The deciphering of rebuses, the solution of smallproblems of strictly definite scope, are occupations which attract manyable minds. Every find brings pleasure, and in the field of eruditionthere are innumerable finds--some lying exposed and obvious, someguarded by all but impenetrable barriers--to reward both those who doand those who do not delight in surmounting difficulties. All thescholars of any distinction have possessed in an eminent degree theinstincts of the collector and the puzzle-solver, and some of them havebeen quite conscious of the fact. "The more difficulties we encounteredin our chosen path, " says M. Hauréau, "the more the enterprise pleasedus. This species of labour, which is called bibliography [investigationsof authorship, principally from the point of view of pseudepigraphy], could not aspire to the homage of the public, but it has a greatattraction for those who devote themselves to it. Yes, it is doubtless ahumble study, but how many others are there which so often compensatethe trouble they give by affording us opportunity to cry Eureka. "[113]Julien Havet, when he was "already known to the learned men of Europe, "used to divert himself "by apparently frivolous amusements, such asguessing square words or deciphering cryptograms. "[114] Profoundinstincts, and, for all the childish or ridiculous perversions whichthey may exhibit in certain individuals, of the highest utility! Afterall, these are forms, the most rudimentary forms, of the scientificspirit. Those who are devoid of them have no place in the world ofcritical scholarship. But those who aspire to be critical scholars willalways be numerous; for the labours of interpretation, construction, andexposition require the rarest gifts: all those whom chance has throwninto the study of history, who desire to do useful work in thatdepartment, but are wanting in psychological tact, or find compositionirksome, will always allow themselves to be fascinated by the simple andcalm pleasures of the preliminary tasks. 

But in order to succeed in critical labours it is not enough to likethem. It is necessary to possess qualifications "for which zeal is nosubstitute. " What qualifications? Those who have asked this questionhave answered vaguely: "Qualifications of the moral rather than theintellectual order, patience, intellectual honesty. .. . " Is it notpossible to be more precise?

There are young students with no _a priori_ repugnance for the laboursof external criticism, who perhaps are even disposed to like them, whoyet are--experience has shown it--totally incapable of performing them. There would be nothing perplexing in this if these persons wereintellectually feeble; this incapacity would then be but onemanifestation of their general weakness; nor yet if they had gonethrough no technical apprenticeship. But we are concerned with men ofeducation and intelligence, sometimes of exceptional ability, who do notlabour under the above disadvantages. These are the people of whom wehear: "He works badly, he has the genius of inaccuracy. " Theircatalogues, their editions, their _regesta_, their monographs swarm withimperfections, and never inspire confidence; try as they may, theynever attain, I do not say absolute accuracy, but any decent degree ofaccuracy. They are subject to "chronic inaccuracy, " a disease of whichthe English historian Froude is a typical and celebrated case. Froudewas a gifted writer, but destined never to advance any statement thatwas not disfigured by error; it has been said of him that he wasconstitutionally inaccurate. For example, he had visited the city ofAdelaide in Australia: "We saw, " says he, "below us, in a basin with ariver winding through it, a city of 150, 000 inhabitants, none of whomhas ever known or will ever know one moment's anxiety as to therecurring regularity of his three meals a day. " Thus Froude, now for thefacts: Adelaide is built on an eminence; no river runs through it; whenFroude visited it the population did not exceed 75, 000, and it wassuffering from a famine at the time. And more of the same kind. [115]Froude was perfectly aware of the utility of criticism, and he was evenone of the first in England to base the study of history on that oforiginal documents, as well unpublished as published; but his mentalconformation rendered him altogether unfit for the emendation of texts;indeed, he murdered them, unintentionally, whenever he touched them. Just as Daltonism (an affection of the organs of sight which prevents aman from distinguishing correctly between red and green signals)incapacitates for employment on a railway, so chronic inaccuracy, or"Froude's Disease" (a malady not very difficult to diagnose) ought to beregarded as incompatible with the professional practice of criticalscholarship. 

Froude's Disease does not appear to have ever been studied by thepsychologists, nor, indeed, is it to be considered as a separatepathological entity. Every one makes mistakes "out of carelessness, ""through inadvertence, " and in many other ways. What is abnormal is tomake many mistakes, to be always making them, in spite of the mostpersevering efforts to be exact. Probably this phenomenon is connectedwith weakness of the attention and excessive activity of the involuntary(or subconscious) imagination which the will of the patient, lackingstrength and stability, is unable sufficiently to control. Theinvoluntary imagination intrudes upon intellectual operations only tovitiate them; its part is to fill up the gaps of memory by conjecture, to magnify and attenuate realities, and to confuse them with theproducts of pure invention. Most children distort everything byinexactitude of this kind, and it is only after a hard struggle thatthey ever attain to a scrupulous accuracy--that is, learn to mastertheir imagination. Many men remain children, in this respect, the wholeof their lives. 

But, let the psychological causes of Froude's Disease be what they may, another point claims our attention. The man of the sanest andbest-balanced mind is liable to bungle the simplest kinds of criticalwork if he does not allow them the necessary time. In these mattersprecipitancy is the source of innumerable errors. It is rightly saidthat patience is the cardinal virtue of the scholar. Do not work toofast, act as if there were always something to be gained by waiting, leave work undone rather than spoil it: these are maxims easy enough topronounce, but not to be followed in practice by any but persons ofcalm temperament. There are nervous, excitable persons, who are alwaysin a hurry to get to the end, always seeking variety in theiroccupations, and always anxious to dazzle and astonish: these maypossibly find honourable employment in other careers; but if theyembrace erudition, they are doomed to pile up a mass of provisionalwork, which is likely to do more harm than good, and is sure in the longrun to cause them many a vexation. The true scholar is cool, reserved, circumspect. In the midst of the turmoil of life, which flows past himlike a torrent, he never hurries. Why should he hurry? The importantthing is, that the work he does should be solid, definitive, imperishable. Better "spend weeks polishing a masterpiece of a score ofpages" in order to convince two or three among the scholars of Europethat a particular charter is spurious, or take ten years to constructthe best possible text of a corrupt document, than give to the press inthe same interval volumes of moderately accurate _anecdota_ which futurescholars will some day have to put through the mill again from beginningto end. 

Whatever special branch of critical scholarship a man may choose, heought to be gifted with prudence, an exceptionally powerful attentionand will, and, moreover, to combine a speculative turn of mind withcomplete disinterestedness and little taste for action; for he must makeup his mind to work for distant and uncertain results, and, in nearlyevery case, for the benefit of others. For textual criticism and theinvestigation of sources, it is, moreover, very useful to have thepuzzle-solving instinct--that is, a nimble, ingenious mind, fertile inhypotheses, prompt to seize and even to guess the relations of things. For tasks of description and compilation (the preparation of inventoriesand catalogues, _corpus_ and _regesta_-making) it is absolutelynecessary to possess the collector's instinct, together with anexceptional appetite for work, and the qualities of order, industry, andperseverance. [116] These are the aptitudes required. The labours ofexternal criticism are so distasteful to those who lack these aptitudes, and the results obtained are, in their case, so small in comparison withthe time expended, that it is impossible for a man to make too sure ofhis vocation before entering upon a career of critical scholarship. Itis pitiful to see those who, for want of a wise word spoken in dueseason, lose their way and vainly exhaust themselves in such a career, especially when they have good reason for believing that they might haveemployed their talents to better advantage in other directions. [117]

II. As critical and preparatory tasks are remarkably well suited to thetemperament of a very large number of Germans, and as the activity ofGerman erudition during the present century has been enormous, it is toGermany that we must go for the best cases of those mental deformationswhich are produced, in the long run, by the habitual practice ofexternal criticism. Hardly a year passes but complaints are heard, inand about the German universities, of the ill effects produced onscholars by the tasks of criticism. 

In 1890, Herr Philippi, as Rector of the University of Giessen, forciblydeplored the chasm which, as he said, is opening between preparatorycriticism and general culture: textual criticism loses itself ininsignificant minutiæ; scholars collate for the mere pleasure ofcollating; infinite precautions are employed in the restoration ofworthless documents; it is thus evident that "more importance isattached to the materials of study than to its intellectual results. "The Rector of Giessen sees in the diffuse style of German scholars andin the bitterness of their polemical writings an effect of the habitthey have contracted of "excessive preoccupation with littlethings. "[118] In the same year the same note was sounded, at theUniversity of Bâle, by Herr J. V. Pflugk-Harttung. "The highest branchesof historical science are despised, " says this author in his_Geschichtsbetrachtungen_[119]: "all that is valued is microscopicobservations and absolute accuracy in unimportant details. The criticismof texts and sources has become a branch of sport: the least breach ofthe rules of the game is considered unpardonable, while conformity tothem is enough to assure the approval of connoisseurs, irrespectively ofthe intrinsic value of the results obtained. Scholars are mostlymalevolent and discourteous towards each other; they make molehills andcall them mountains; their vanity is as comic as that of the citizen ofFrankfort who used complacently to observe, 'All that you can seethrough yonder archway is Frankfort territory. '"[120] We, for our part, are inclined to draw a distinction between three professional risks towhich scholars are subject: dilettantism, hypercriticism, and loss ofthe power to work. 

To take the last first: the habit of critical analysis has a relaxingand paralysing action on certain intelligences. Men, of naturally timiddispositions, discover that whatever pains they take with their criticalwork, their editing or classifying of documents, they are very apt tomake slight mistakes, and these slight mistakes, as a result of theircritical education, fill them with horror and dread. To discoverblunders in their signed work when the time for correction is past, causes them acute suffering. They reach at length a state of morbidanxiety and scrupulosity which prevents them from doing anything atall, for fear of possible imperfections. The _examen rigorosum_ to whichthey are continually subjecting themselves brings them to a standstill. They give the same measure to the productions of others, and in the endthey see in historical works nothing but the authorities and the notes, the _apparatus criticus_, and in the _apparatus criticus_ they seenothing but the faults in it which require correction. 

_Hypercriticism. _--The excess of criticism, just as much as the crudestignorance, leads to error. It consists in the application of criticalcanons to cases outside their jurisdiction. It is related to criticismas logic-chopping is to logic. There are persons who scent enigmaseverywhere, even where there are none. They take perfectly clear textsand subtilise on them till they make them doubtful, under the pretext offreeing them from imaginary corruptions. They discover traces of forgeryin authentic documents. A strange state of mind! By constantly guardingagainst the instinct of credulity they come to suspect everything. [121]It is to be observed that in proportion as the criticism of texts andsources makes positive progress, the danger of hypercriticism increases. When all the sources of history have been properly criticised (forcertain parts of ancient history this is no distant prospect), goodsense will call a halt. But scholars will refuse to halt; they willrefine, as they do already on the best established texts, and those whorefine will inevitably fall into hypercriticism. "The peculiarity of thestudy of history and its auxiliary philological sciences, " says Renan, "is that as soon as they have attained their relative perfection theybegin to destroy themselves. "[122] Hypercriticism is the cause of this. 

_Dilettantism. _--Scholars by profession and vocation have a tendency totreat the external criticism of documents as a game of skill, difficult, but deriving an interest, much as chess does, from the very complicationof its rules. Some of them are indifferent to the larger questions--tohistory itself, in fact. They criticise for the sake of criticism, and, in their view, the elegance of the method of investigation is much moreimportant than the results, whatever they may be. These _virtuosi_ arenot concerned to connect their labours with some general idea--tocriticise systematically, for example, all the documents relating to aquestion, in order to understand it; they criticise indiscriminatelytexts relating to all manner of subjects, on the one condition of beingsufficiently corrupt. Armed with their critical skill, they range overthe whole of the domain of history, and stop wherever a knotty probleminvites their services; this problem solved, or at least discussed, theygo elsewhere to look for others. They leave behind them no coherentwork, but a heterogeneous collection of memoirs on every conceivablesubject, which resembles, as Carlyle says, a curiosity shop or anarchipelago of small islands. 

Dilettanti defend their dilettantism by sufficiently plausiblearguments. To begin with, say they, everything is important; in historythere is no document which has not its value: "No scientific work isbarren, no truth is without its use for science . .. ; in history there isno such thing as a trivial subject;" consequently, "it is not the natureof the subject which makes work valuable, but the method employed. "[123]The important thing in history is not "the ideas one accumulates; it isthe mental gymnastics, the intellectual training--in short, thescientific spirit. " Even supposing that there are degrees of importanceamong the data of history, no one has a right to maintain _a priori_that a document is "useless. " What, pray, is the criterion of utility inthese matters? How many documents are there not which, after being longdespised, have been suddenly placed in the foreground by a change ofstandpoint or by new discoveries? "All exclusion is rash; there is noresearch which it is possible to brand beforehand as necessarilysterile. That which has no value in itself may become valuable as anecessary means. " Perhaps a day may come when, science being in a sensecomplete, indifferent documents and facts may be safely thrownoverboard; but we are not at present in a position to distinguish thesuperfluous from the necessary, and in all probability the line ofdemarcation will never be easy to trace. This justifies the most specialresearches and the most futile in all appearance. And, if it come to theworst, what does it matter if there is a certain amount of work wasted?"It is a law in science, as in all human effort, " and indeed in all theoperations of nature, "to work in broad outlines, with a wide margin ofwhat is superfluous. "

We shall not undertake to refute these arguments to the full extent inwhich this is possible. Besides, Renan, who has put the case for bothsides of the question with equal vigour, definitively closed the debatein the following words: "It may be said that some researches are uselessin the sense of taking up time which would have been better spent onmore serious questions. .. . Although it is not necessary for an artisanto have a complete knowledge of the work he is employed to execute, itis still to be desired that those who devote themselves to speciallabours should have some notion of the more general considerations whichalone give value to their researches. If all the industrious workers towhom modern science owes its progress had had a philosophicalcomprehension of what they were doing, how much precious time would havebeen saved!. .. It is deeply to be regretted that there should be such animmense waste of human effort, merely for want of guidance, and a clearconsciousness of the end to be pursued. "[124]

Dilettantism is incompatible with a certain elevation of mind, and witha certain degree of "moral perfection, " but not with technicalproficiency. Some of the most accomplished critics merely make a tradeof their skill, and have never reflected on the ends to which their artis a means. It would, however, be wrong to infer that science itself hasnothing to fear from dilettantism. The dilettanti of criticism who workas fancy or curiosity bids them, who are attracted to problems not bytheir intrinsic importance, but by their difficulty, do not supplyhistorians (those whose work it is to combine materials and use them forthe main purposes of history) with the materials of which the latterhave the most pressing need, but with others which might have waited. Ifthe activity of specialists in external criticism were exclusivelydirected to questions whose solution is important, and if it wereregulated and guided from above, it would be more fruitful. 

The idea of providing against the dangers of dilettantism by a rational"organisation of labour" is already ancient. Fifty years ago it wascommon to hear people talking of "supervision, " of "concentratingscattered forces;" dreams were rife of "vast workshops" organised on themodel of those of modern industry, in which the preparatory labours ofcritical scholarship were to be performed on a great scale, in theinterests of science. In nearly all countries, in fact, governments(through the medium of historical committees and commissions), academies, and learned societies have endeavoured in our day, much asmonastic congregations did of old, to group professed scholars for thepurposes of vast collective enterprises, and to co-ordinate theirefforts. But this banding of specialists in external criticism for theservice and under the supervision of competent men presents greatmechanical difficulties. The problem of the "organisation of scientificlabour" is still the order of the day. [125]

III. Scholars are often censured for pride and excessive harshness inthe judgments which they pass on the labours of their colleagues; andthese faults, as we have seen, are often attributed to their excessive"preoccupation with little things, " especially by persons whose attemptshave been severely judged. In reality there do exist modest and kindlyscholars: it is a question of character; professional "preoccupationwith little things" is not enough to change natural disposition in thisrespect. "Ce bon monsieur Du Cange, " as the Benedictines said, wasmodest to excess. "Nothing more is required, " says he, in speaking ofhis labours, "but eyes and fingers in order to do as much and more;" henever blamed any one, on principle. "If I study it is for the pleasureof studying, and not to give pain to any one else, any more than tomyself. "[126] It is, however, true that most scholars have nocompunction in exposing each other's mistakes, and that their austerezeal sometimes finds expression in harsh and overbearing language. Barring the harshness they are quite right. Like physicians, chemists, and other members of learned and scientific professions, they have akeen appreciation of the value of scientific truth, and it is for thisreason that they make a point of calling offenders to account. They arethus enabled to bar the door against the tribe of incapables andcharlatans who once infested their profession. 

Among the youths who propose to devote themselves to the study ofhistory there are some in whom the commercial spirit and vulgar ambitionare stronger than the love of science. These are apt to say tothemselves: "Historical work, if it is to be done according to the rulesof method, requires an infinite amount of labour and caution. But do wenot see historical writings whose authors have more or less seriouslyviolated the rules? Are these authors thought any the less of on thisaccount? Is it always the most conscientious writer who enjoys thehighest consideration? Cannot tact supply the place of knowledge?" Iftact really could supply the place of knowledge, then, as it is easierto do bad work than good, and as the important thing with these peopleis success, they might be tempted to conclude that it does not matterhow badly they work as long as they succeed. Why should not things go inthese matters as they do in life, where it is not necessarily the bestmen that get on best? Well, it is due to the pitiless severity of thecritics that calculations of this kind would be as disastrous as theyare despicable. 

Towards the end of the Second Empire there was in France no enlightenedpublic opinion on the subject of historical work. Bad books ofhistorical erudition were published with impunity, and sometimes evenprocured undeserved rewards for their authors. It was then that thefounders of the _Revue Critique d'histoire et de littérature_ undertookto combat a state of things which they lightly deemed demoralising. With this object they administered public chastisement to those scholarswho showed lack of conscience or method, in a manner calculated todisgust them with erudition for ever. They performed sundry notableexecutions, not for the pleasure of it, but with the firm resolve toestablish a censorship and a wholesome dread of justice, in the domainof historical study. Bad workers henceforth received no quarter, andthough the _Revue_ did not exert any great influence on the public atlarge, its police-operations covered a wide enough radius to impressmost of those concerned with the necessity of sincerity and respect formethod. During the last twenty-five years the impulse thus given hasspread beyond all expectation. 

It is now a matter of great difficulty to impose on the world ofscholars, in matters connected with their studies, or at least to keepup the deception for any length of time. In the case of the historicalsciences, as well as the sciences proper, it is now too late to found anew error or to discredit an old truth. It may be a few months, possiblya few years, before a bungled experiment in chemistry or a scampededition is recognised as such; but inexact results, though temporarilyaccepted under reserve, are always sooner or later, and generally verysoon, discovered, denounced, and eliminated. The theory of theoperations of external criticism is now so well established, the numberof specialists thoroughly versed in them is now so great in everycountry, that, with rare exceptions, descriptive catalogues ofdocuments, editions, _regesta_, monographs, are scrutinised, dissected, and judged as soon as they appear. It is well to be warned. It will forthe future be the height of imprudence to risk publishing a work oferudition without having first done everything possible to make itunassailable; otherwise it will immediately, or after brief delay, beattacked and demolished. Not knowing this, certain well-meaning personsstill show themselves, from time to time, simple enough to enter thelists of critical scholarship insufficiently prepared; they are filledwith a desire to be useful, and are apparently convinced that here, asin politics and elsewhere, it is possible to work by extemporised andapproximate methods without any "special knowledge. " They are sorryafterwards. The knowing ones do not take the risk; the tasks of criticalscholarship have no seductions for them, for they are aware that thelabour is great and the glory moderate, and that the field is engrossedby clever specialists not too well disposed towards intruders. They seeplainly there is no room for them here. The blunt uncompromising honestyof the scholars thus delivers them from undesirable company of a kindwhich the "historians" proper have still occasionally to put up with. 

Bad workers, in fact, on the hunt for a public less closely criticalthan the scholars, are very ready to take refuge in historicalexposition. The rules of method are here less obvious, or, rather, notso well known. While the criticism of texts and sources has been placedon a scientific basis, historical synthesis is still performedhaphazard. Mental confusion, ignorance, negligence--faults which standout so clearly in works of critical scholarship--may in historicalworks be disguised up to a certain point by literary artifices, and thepublic at large, which is not well educated in this respect, is notshocked. [127] In short, there is still, in this department, a certainchance of impunity. This chance, however, is diminishing, and a day willcome, before so very long, when the superficial writers who makeincorrect syntheses will be treated with as little consideration as isnow received by those who show themselves unscrupulous or unskilful inthe technique of preparatory criticism. The works of the most celebratedhistorians of the nineteenth century, those who died but yesterday, Augustin Thierry, Ranke, Fustel de Coulanges, Taine, and others, arealready battered and riddled with criticism. The faults of their methodshave already been seen, defined, and condemned. 

Those who are insensible to other considerations ought to be moved tohonesty in historical work by the reflection that the time is now past, or nearly so, when it was possible to do bad work without having tosuffer for it. 

_SECTION II. --INTERNAL CRITICISM_

CHAPTER VI

INTERPRETATIVE CRITICISM (HERMENEUTIC)

I. When a zoologist describes the form and situation of a muscle, when aphysiologist gives the curve of a movement, we are able to accept theirresults without reserve, because we know by what method, by whatinstruments, by what system of notation they have obtained them. [128]But when Tacitus says of the Germans, _Arva per annos mutant_, we do notknow beforehand whether he took the right method to inform himself, noreven in what sense he used the words _arva_ and _mutant_; to ascertainthis a preliminary operation is required. [129] This operation isinternal criticism. 

The object of criticism is to discover what in a document may beaccepted as true. Now the document is only the final result of a longseries of operations, on the details of which the author gives us noinformation. He had to observe or collect facts, to frame sentences, towrite down words; and these operations, which are perfectly distinct onefrom another, may not all have been performed with the same accuracy. Itis therefore necessary to _analyse_ the product of the author's labourin order to distinguish which operations have been incorrectlyperformed, and reject their _results_. _Analysis_ is thus necessary tocriticism; all criticism begins with analysis. 

In order to be logically complete, the analysis ought to reconstruct_all_ the operations which the author must have performed, and toexamine them _one by one_, to see whether each has been performedcorrectly. It would be necessary to pass in review all the successiveacts by which the document was produced, from the moment when the authorobserved the fact which is its subject up to the movements of his handby which he traced the letters of the document; or, rather, it would benecessary to proceed in the opposite direction, step by step, from themovements of the hand back to the observation. This method would be solong and so tedious that no one would ever have the time or the patienceto apply it. 

Internal criticism is not, like external criticism, an instrument usedfor the mere pleasure of using it;[130] it yields no immediatesatisfaction, because it does not definitively solve any problem. It isonly applied because it is necessary, and its use is restricted to abare minimum. The most exacting historian is satisfied with an abridgedmethod which concentrates all the operations into two groups: (1) theanalysis of the contents of the document, and the positiveinterpretative criticism which is necessary for ascertaining what theauthor meant; (2) the analysis of the conditions under which thedocument was produced, and its negative criticism, necessary for theverification of the author's statements. This twofold division of thelabour of criticism is, moreover, only employed by a select few. Thenatural tendency, even of historians who work methodically, is to readthe text with the object of extracting information directly from it, without any thought of first ascertaining what exactly was in theauthor's mind. [131] This procedure is excusable at most in the case ofnineteenth-century documents, written by men whose language and mode ofthought are familiar to us, and then only when there is not more thanone possible interpretation. It becomes dangerous as soon as theauthor's habits of language or thought begin to differ from those of thehistorian who reads him, or when the meaning of the text is not obviousand indisputable. Whoever, in reading a text, is not exclusivelyoccupied with the effort to understand it, is sure to read impressionsof his own into it; he is struck by phrases or words in the documentwhich correspond to his own ideas, or agree with his own _a priori_notion of the facts; unconsciously he detaches these phrases or words, and forms out of them an imaginary text which he puts in the place ofthe real text of the author. [132]

II. Here, as always in history, method consists in repressing the firstimpulse. It is necessary to be penetrated by the principle, sufficientlyobvious but often forgotten, that a document only contains the ideas ofthe man who wrote it, and to make it a rule to begin by understandingthe text by itself, _before_ asking what can be extracted from it forthe purposes of history. We thus arrive at this general rule of method:the study of every document should begin with an analysis of itscontents, made with the sole aim of determining the real meaning of theauthor. 

This analysis is a preliminary operation, distinct and independent. Experience here, as in the tasks of critical scholarship, [133] hasdecided in favour of the system of slips. Each slip will contain theanalysis of a document, of a separate part of a document, or of anepisode in a narrative; the analysis ought to indicate not only thegeneral sense of the text, but also, as far as possible, the object andviews of the author. It will be well to reproduce verbally anyexpressions which may seem characteristic of the author's thought. Sometimes it will be enough to have analysed the text mentally: it isnot always necessary to put down in black and white the whole contentsof a document; in such cases we simply enter the points of which weintend to make use. But against the ever-present danger of substitutingone's personal impressions for the text there is only one realsafeguard; it should be made an invariable rule never on any account tomake an extract from a document, or a partial analysis of it, withouthaving _first_ made a comprehensive analysis[134] of it mentally, if noton paper. 

To analyse a document is to discern and isolate all the ideas expressedby the author. Analysis thus reduces to _interpretative criticism_. 

Interpretation passes through two stages: the first is concerned withthe literal, the second with the real meaning. 

III. The determination of the literal meaning of a document is alinguistic operation; accordingly, Philology (in the narrow sense) hasbeen reckoned among the auxiliary sciences of history. To understand atext it is first necessary to know the language. But a _general_knowledge of the language is not enough. In order to interpret Gregoryof Tours, it is not enough to know Latin in a general way; it isnecessary to add a special study of the particular kind of Latin writtenby Gregory of Tours. 

The natural tendency is to attribute the same meaning to the same wordwherever it occurs. We instinctively treat a language as if it were afixed system of signs. Fixity, indeed, is a characteristic of the signswhich have been expressly invented for scientific use, such asalgebraical notation or the nomenclature of chemistry. Here everyexpression has a single precise meaning, which is absolute andinvariable; it expresses an accurately analysed and defined idea, onlyone such idea, and that always the same in whatever context theexpression may occur, and by whatever author it may be used. Butordinary language, in which documents are written, fluctuates: each wordexpresses a complex and ill-defined idea; its meanings are manifold, relative, and variable; the same word may stand for several differentthings, and is used in different senses by the same author according tothe context; lastly, the meaning of a word varies from author to author, and is modified in the course of time. _Vel_, which in classical Latinonly has the meanings _or_ and _even_, means _and_ in certain epochs ofthe middle ages; _suffragium_, which is classical Latin for _suffrage_, takes in mediæval Latin the sense of _help_. We have, then, to learn toresist the instinct which leads us to explain all the expressions of atext by their classical or ordinary meanings. The grammaticalinterpretation, based on the general rules of the language, must besupplemented by an historical interpretation founded on an examinationof the particular case. 

The method consists in determining the special meaning of the words inthe document; it rests on a few very simple principles. 

(1) Language changes by continuous evolution. Each epoch has a languageof its own, which must be treated as a separate system of signs. Inorder to understand a document we must know the _language of thetime_--that is, the meanings of words and forms of expression in use atthe time when the text was written. The meaning of a word is to bedetermined by bringing together the passages where it is employed: itwill generally be found that in one or other of these the remainder ofthe sentence leaves no doubt as to the meaning of the word inquestion. [135] Information of this kind is given in historicaldictionaries, such as the _Thesaurus Linguæ Latinæ_; or the glossariesof Du Cange. In these compilations the article devoted to each word is acollection of the passages in which the word occurs, accompanied byindications of authorship which fix the epoch. 

When the author wrote in a dead language which he had learnt out ofbooks--this is the case with the Latin texts of the earlier middleages--we must be on our guard against words used in an arbitrary sense, or selected for the sake of elegance: for example, _consul_ (count, earl), _capite census_ (censitary), _agellus_ (grand domain). 

(2) Linguistic usage may vary from one region to another; we have, then, to know the _language of the country_ where the document waswritten--that is, the peculiar meanings current in the country. 

(3) Each author has his own manner of writing; we have, then, to studythe _language of the author_, the peculiar senses in which he usedwords. [136] This purpose is served by lexicons to a single author, asMeusel's _Lexicon Cæsarianum_, in which are brought together all thepassages in which the author used each word. 

(4) An expression changes its meaning according to the passage in whichit occurs; we must therefore interpret each word and sentence not as ifit stood isolated, but with an eye to the general sense of the context. This is the _rule of context_, [137] a fundamental rule ofinterpretation. Its meaning is that, before making use of a phrase takenfrom a text, we must have read the text in its entirety; it prohibitsthe stuffing of a modern work with _quotations_--that is, shreds ofphrases torn from passages without regard to the special sense given tothem by the context. [138]

These rules, if rigorously applied, would constitute an exact method ofinterpretation which would hardly leave any chance of error, but wouldrequire an enormous expenditure of time. What an immense amount oflabour would be necessary if, in the case of _each_ word, we had todetermine by a special operation its meaning in the language of thetime, of the country, of the author, and in the context! Yet this is thelabour demanded by a well-made translation: in the case of some ancientworks of great literary value it has been submitted to; for the mass ofhistorical documents we content ourselves, in practice, with an abridgedmethod. 

All words are not equally subject to variations of meaning; most of themkeep a fairly uniform meaning in all authors and in all periods. We maytherefore be satisfied to study specially those expressions which, fromtheir nature, are liable to take different meanings: first, ready-madeexpressions which, being fixed, do not follow the evolution of the wordsof which they are composed; secondly, and chiefly, words denoting thingswhich are in their nature subject to evolution; classes of men (_miles_, _colonus_, _servus_); institutions (_conventus_, _justitia_, _judex_);usages (_alleu_, _bénéfice_, _élection_); feelings, common objects. Inthe case of all words of such classes it would be imprudent to assume afixed meaning; it is an absolutely necessary precaution to ascertainwhat is the sense in which they are used in the text to be interpreted. "These studies of words, " said Fustel de Coulanges, "have a greatimportance in historical science. A badly interpreted term may be thesource of serious error. "[139] And, in fact, simply by a methodicalapplication of interpretative criticism to a hundred words or so, hesucceeded in revolutionising the study of the Merovingian epoch. 

IV. When we have analysed the document and determined the literalmeaning of its phrases, we cannot even yet be sure that we have reachedthe real thoughts of the author. It is possible that he may have usedsome expressions in an oblique sense; there are several kinds of caseswhere this occurs: allegory and symbolism, jests and hoaxes, allusionand implication, even the ordinary figures of speech, metaphor, hyperbole, litotes. [140] In all these cases it is necessary to piercethrough the literal meaning to the real meaning, which the author haspurposely disguised under an inexact form. 

Logically the problem is very embarrassing: there is no fixed externalcriterion by which we can make sure of detecting an oblique sense; inthe case of the hoax, which in the present century has become a branchof literature, it is an essential part of the author's plan to leave noindication which would betray the jest. In practice we may be morallycertain that an author is not using an oblique sense wherever his primeobject is to be understood; we are therefore not likely to meet withdifficulties of this kind in official documents, in charters, and inhistorical narratives. In all these cases the general form of thedocument permits us to assume that it is written in the literal sense ofthe words. 

On the other hand, we must be prepared for oblique senses when theauthor had other interests than that of being understood, or when hewrote for a public which could understand his allusions and read betweenthe lines, or when his readers, in virtue of a religious or literaryinitiation, might be expected to understand his symbolisms and figuresof speech. This is the case with religious texts, private letters, andall those literary works which form so large a part of the documents onantiquity. Thus the art of recognising and determining hidden meaningsin texts has always occupied a large space in the theory of_hermeneutic_[141] (which is Greek for interpretative criticism), and inthe _exegesis_ of the sacred texts and of classical authors. 

The different modes of introducing an oblique sense behind the literalsense are too varied, and depend too much on special circumstances, forit to be possible to reduce the art of detecting them to definite rules. Only one general principle can be laid down, and that is, that when theliteral sense is absurd, incoherent, or obscure, or in contradictionwith the ideas of the author or the facts known to him, then we ought topresume an oblique sense. 

In order to determine this sense, the procedure is the same as forstudying the language of an author: we compare the passages in which theexpressions occur in which we suspect an oblique sense, and look to seewhether there is not one where the meaning may be guessed from thecontext. A celebrated instance of this procedure is the discovery of theallegorical meaning of the Beast in the Apocalypse. But as there is nocertain method of solving these problems, we never have a right to saywe have discovered all the hidden meanings or seized all the allusionscontained in a text; and even when we think we have found the sense, weshall do well to draw no inferences from a necessarily conjecturalinterpretation. 

On the other hand, it is necessary to guard against the temptation tolook for allegorical meanings everywhere, as the neo-Platonists did inPlato's works and the Swedenborgians in the Bible. This attack of_hyper-hermeneutic_ is now over, but we are not yet safe from theanalogous tendency to look for allusions everywhere. Investigations ofthis kind are always conjectural, and are better calculated to flatterthe vanity of the interpreter than to furnish results of which historycan make use. 

V. When we have at length reached the real sense of the text, theoperation of positive analysis is concluded. Its result is to make usacquainted with the author's conceptions, the images he had in his mind, the general notions in terms of which he represented the world tohimself. This information belongs to a very important branch ofknowledge, out of which is constituted a whole group of historicalsciences:[142] the history of the illustrative arts and of literature, the history of science, the history of philosophical and moraldoctrine, mythology and the history of dogmas (wrongly called religiousbeliefs, because here we are studying official doctrines withoutinquiring whether they are believed), the history of law, the history ofofficial institutions (so far as we do not inquire how they were appliedin practice), the assemblage of popular legends, traditions, opinions, conceptions (inexactly called beliefs) which are comprised under thename of folk-lore. 

All these studies need only the external criticism which investigatesauthorship and origin and interpretative criticism; they require onedegree less elaboration than the history of objective facts, andaccordingly they have been earlier established on a methodical basis. 

CHAPTER VII

THE NEGATIVE INTERNAL CRITICISM OF THE GOOD FAITH AND ACCURACY OFAUTHORS

I. Analysis and positive interpretative criticism only penetrate as faras the inward workings of the mind of the author of a document, and onlyhelp us to know his ideas. They give no direct information aboutexternal facts. Even when the author was able to observe them, his textonly indicates how he wished to represent them, not how he really sawthem, still less how they really happened. What an author expresses isnot always what he believed, for he may have lied; what he believed isnot necessarily what happened, for he may have been mistaken. Thesepropositions are obvious. And yet a first and natural impulse leads usto accept as true every statement contained in a document, which isequivalent to assuming that no author ever lied or was deceived; andthis spontaneous credulity seems to possess a high degree of vitality, for it persists in spite of the innumerable instances of error andmendacity which daily experience brings before us. 

Reflection has been forced on historians in the course of their work bythe circumstance of their finding documents which contradicted eachother; in such cases they have been obliged to doubt, and, afterexamination, to admit the existence of error or mendacity; thus negativecriticism has appeared as a practical necessity for the purpose ofeliminating statements which are obviously false or erroneous. But theinstinct of confidence is so indestructible that it has hithertoprevented even those professionally concerned from systematising theinternal criticism of statements in the same way as the externalcriticism which deals with the origin of documents has beensystematised. Historians, in their works, and even theoretical writerson historical method, [143] have been satisfied with common notions andvague formulæ in striking contrast with the precise terminology of thecritical investigation of sources. They are content to examine whetherthe author was roughly _contemporary_ with the events, whether he was anocular _witness_, whether he was _sincere_ and _well-informed_, whetherhe knew the truth and desired to tell it, or even--summing up the wholequestion in a single formula--whether he was _trustworthy_. 

This superficial criticism is certainly better than no criticism at all, and has sufficed to give those who have applied it the consciousness ofincontestable superiority. But it is only a halfway-house between commoncredulity and scientific method. Here, as in every science, thestarting-point must be methodical doubt. [144] All that has not beenproved must be temporarily regarded as doubtful; no proposition is tobe affirmed unless reasons can be adduced in favour of its truth. Applied to the statements contained in documents, methodical doubtbecomes _methodical distrust_. 

The historian ought to distrust _a priori_ every statement of an author, for he cannot be sure that it is not mendacious or mistaken. At the bestit affords a presumption. For the historian to adopt it and affirm itafresh on his own account implies that he regards it as a scientifictruth. To take this decisive step is what he has no right to do withoutgood reasons. But the human mind is so constituted that this step isoften taken unconsciously (cf. Book ii. Chap. I. ). Against thisdangerous tendency criticism has only one means of defence. We must notpostpone doubt till it is forced upon us by conflicting statements indocuments; we must _begin_ by doubting. We must never forget theinterval which separates a statement made by any author whatsoever froma scientifically established truth, so that we may continually keep inmind the responsibility which we assume when we reproduce a statement. 

Even after we have accepted the principle and resolved to apply thisunnatural distrust in practice, we tend instinctively to free ourselvesfrom it as soon as possible. The natural impulse is to perform thecriticism of the whole of an author, or at least of the whole of adocument, in the lump; to divide authorities into two categories, thesheep on the right, the goats on the left; on the one side trustworthyauthors and good documents, on the other suspected authors and baddocuments. Having thus exhausted our powers of distrust, we proceed toreproduce without discussion all the statements contained in the "gooddocument. " We consent to distrust suspected authors such as Suidas orAimo, but we affirm as established truth everything that has been saidby Thucydides or Gregory of Tours. [145] We apply to authors thatjudicial procedure which divides witnesses into admissible andinadmissible: having once accepted a witness, we feel ourselves bound toadmit all his testimony; we dare not doubt any of his statements withouta special reason. Instinctively we take sides with the author on whom wehave bestowed our approval, and we go so far as to say, as in the lawcourts, that the burden of proof rests with those who reject validtestimony. [146]

The confusion is still further increased by the use of the word_authentic_, borrowed from judicial language. It has reference to theorigin only, not to the contents; to say that a document is authentic ismerely to say that its origin is certain, not that its contents are freefrom error. But authenticity inspires a degree of respect which disposesus to accept the contents without discussion. To doubt the statements ofan authentic document would seem presumptuous, or at least we thinkourselves bound to wait for overwhelming proof before we impeach thetestimony of the author. 

II. These natural instincts must be methodically resisted. A document(still more a literary work) is not all of a piece; it is composed of agreat number of independent statements, any one of which may beintentionally or unintentionally false, while the others are _bonâ fide_and accurate, or conversely, since each statement is the outcome of amental operation which may have been incorrectly performed, while otherswere performed correctly. It is not, therefore, enough to examine adocument as a whole; each of the statements in it must be examinedseparately; _criticism_ is impossible without _analysis_. 

Thus internal criticism conducts us to two general rules. 

(1) A scientific truth is not established by _testimony_. In order toaffirm a proposition we must have special reasons for believing ittrue. It may happen in certain cases that an author's statement is asufficient reason for belief; but we cannot know that beforehand. Therule, then, will be to examine each separate statement in order to makesure whether it is of a nature to constitute a sufficient reason forbelief. 

(2) The criticism of a document is not to be performed _en bloc_. Therule will be to _analyse_ the document into its elements, in order toisolate the different statements of which it is composed and to examineeach of them separately. Sometimes a single sentence contains severalstatements; they must be separated and criticised one by one. In a sale, for example, we distinguish the date, the place, the vendor, thepurchaser, the object, the price, and each one of the conditions. 

In practice, criticism and analysis are performed simultaneously, and, except in the case of texts in a difficult language, may proceed _paripassu_ with interpretative analysis and criticism. As soon as weunderstand a phrase we analyse it and criticise each of its elements. 

It thus appears that _logically_ criticism comprises an enormous numberof operations. In describing them, with all the details necessary forthe understanding of their mechanism and the reasons for theiremployment, we are likely to give the impression of a procedure too slowto be practicable. Such an impression is inevitably produced by everyverbal description of a complicated process. Compare the time occupiedin describing a movement in fencing with that required to execute it;compare the tedium of the grammar and dictionary with the rapidity ofreading. Like every practical art, criticism consists in the habit ofperforming certain acts. In the period of apprenticeship, before thehabit is acquired, we are obliged to think of each act separately beforeperforming it, and to analyse the movements; accordingly we perform themall slowly and with difficulty; but the habit once acquired, the acts, which have now become instinctive and unconscious, are performed withease and rapidity. The reader must therefore not be uneasy about theslowness of the critical processes; he will see later on how they areabridged in practice. 

III. The problem of criticism may be stated as follows. Given astatement made by a man of whose mental operations we have noexperience, and the value of the statement depending exclusively on themanner in which these operations were performed; to ascertain whetherthese operations were performed correctly. The mere statement of theproblem shows that we cannot hope for any direct or definitive solutionof it; we lack the essential datum, namely, the manner in which theauthor performed the mental operations concerned. Criticism thereforedoes not advance beyond indirect and provisional solutions, and does nomore than furnish data which require a final elaboration. 

A natural instinct leads us to judge of the value of statements by theirform. We think we can tell at a glance whether an author is sincere or anarrative accurate. We seek for what is called "the accent ofsincerity, " or "an impression of truth. " This impression is almostirresistible, but it is none the less an illusion. There is no externalcriterion either of good faith or of accuracy. "The accent of sincerity"is the appearance of conviction; an orator, an actor, an habitual liarwill put more of it into his lies than an undecided man into hisstatement of what he believes to be the truth. Energy of affirmationdoes not always mean strength of conviction, but sometimes onlycleverness or effrontery. [147] Similarly, abundance and precision ofdetail, though they produce a vivid impression on unexperienced readers, do not guarantee the accuracy of the facts;[148] they give us noinformation about anything but the imagination of the author when he issincere, or his impudence when he is the reverse. We are apt to say of acircumstantial narrative: "Things of this kind are not invented. " Theyare not invented, but they are very easy to transfer from one person, country, or time to another. There is thus no external characteristic ofa document which can relieve us of the obligation to criticise it. 

The value of an author's statement depends solely on the conditionsunder which he performed certain mental operations. Criticism has noother resource than the examination of these conditions. But it is nota case of reconstructing all of them; it is enough to answer a singlequestion: did the author perform these operations correctly or not? Thequestion may be approached on two sides. 

(1) The critical investigation of authorship has often taught us the_general_ conditions under which the author operated. It is probablethat some of these influenced each one of the operations. We oughttherefore to begin by studying the information we possess about theauthor and the composition of the document, taking particular pains tolook in the habits, sentiments, and personal situation of the author, orin the circumstances in which he composed, for all the reasons whichcould have existed for incorrectness on the one hand, or exceptionalaccuracy on the other. In order to perceive these reasons it isnecessary to be on the lookout for them beforehand. The only method, therefore, is to draw up a general set of questions having reference tothe possible causes of inaccuracy. We shall then apply it to the generalconditions under which the document was composed, in order to discoverthose causes which may have rendered the author's mental operationsincorrect and vitiated the results. But all that we shall thusobtain--even in the exceptionally favourable cases in which theconditions of origin are well known--will be _general_ indications, which will be insufficient for the purposes of criticism, for criticismmust always deal with each separate statement. 

(2) The criticism of particular statements is confined to the use of asingle method, which, by a curious paradox, is the study of the_universal_ conditions under which documents are composed. Theinformation which is not furnished by the general study of the authormay be sought for by a consideration of the necessary processes of thehuman mind; for, since these are universal, they must appear in eachparticular case. We know what are the cases in which men in general areinclined to alter or distort facts. What we have to do in the case ofeach statement is to examine whether it was made under suchcircumstances as to lead us to suspect, from our knowledge of the habitsof normal humanity, that the operations implied in the making of it wereincorrectly performed. The practical procedure will be to draw up a setof questions relating to the habitual causes of inaccuracy. 

The whole of criticism thus reduces to the drawing up and answering oftwo sets of questions: one for the purpose of bringing before our mindsthose general conditions affecting the composition of the document, fromwhich we may deduce general motives for distrust or confidence; theother for the purpose of realising the special conditions of eachstatement, from which special motives may be drawn for distrust orconfidence. These two sets of questions ought to be drawn up beforehandin such a form as may enable us to examine methodically both thedocument in general and each statement in particular; and as they arethe same for all documents, it is useful to formulate them once for all. 

IV. The critical process comprises two series of questions, whichcorrespond to the two series of operations by which the document wasproduced. All that interpretative criticism tells us is what the authormeant; it remains to determine (1) what he really believed, for he maynot have been sincere; (2) what he really knew, for he may have beenmistaken. We may therefore distinguish a _critical examination of theauthor's good faith_, by which we seek to determine whether the authorof the document lied or not, and a _critical examination_ of his_accuracy_, by which we seek to determine whether he was or was notmistaken. 

In practice we rarely need to know what an author believed, unless weare making a special study of his character. We have no direct interestin the author; he is merely the medium through which we reach theexternal facts he reports. The aim of criticism is to determine whetherthe author has reported the facts correctly. If he has given inexactinformation, it is indifferent whether he did so intentionally or not;to draw a distinction would complicate matters unnecessarily. There isthus little occasion to make a separate examination of an author's goodfaith, and we may shorten our labours by including in a single set ofquestions all the causes which lead to misstatement. But for the sake ofclearness it will be well to discuss the questions to be asked in twoseparate series. 

The questions in the first series will help us to inquire whether wehave any reason to distrust the sincerity of a statement. We ask whetherthe author was in any of those situations which normally incline a manto be insincere. We must ask what these situations are, both asaffecting the general composition of a document, and as affecting eachparticular statement. Experience supplies the answer. Every violation oftruth, small or great, is due to a wish on the part of the author toproduce a particular impression upon the reader. Our set of questionsthus reduces to a list of the motives which may, in the general case, lead an author to violate truth. The following are the most importantcases:--

(1) The author seeks to gain a practical advantage for himself; hewishes to deceive the reader of the document, in order to persuade himto an action, or to dissuade him from it; he knowingly gives falseinformation: we then say the author has an interest in deceiving. Thisis the case with most official documents. Even in documents which havenot been composed for a practical purpose, every interested statementhas a chance of being mendacious. In order to determine which statementsare to be suspected, we are to ask what _can_ have been the general aimof the author in writing the document as a whole; and again, what canhave been his particular purpose in making each of the separatestatements which compose the document. But there are two naturaltendencies to be resisted. The first is, to ask what interest the authorcould have _had_ in lying, meaning what interest should _we_ have had inhis place; we must ask instead what interest can he have _thought_ hehad in lying, and we must look for the answer in his tastes and ideals. The other tendency is to take sole account of the _individual_ interestof the author; we ought, however, to remember that the author may havegiven false information in order to serve a _collective_ interest. Thisis one of the difficulties of criticism. An author is a member at oneand the same time of several different groups, a family, a province, acountry, a religious denomination, a political party, a class insociety, whose interests often conflict; we have to discover the groupin which he took most interest, and for which he worked. 

(2) The author was placed in a situation which compelled him to violatetruth. This happens whenever he has to draw up a document in conformitywith rule or custom, while the actual circumstances are in some point orother in conflict with rule or custom; he is then obliged to state thatthe conditions were normal, and thus make a false declaration in respectof all the irregularities. In nearly every report of proceedings thereis some slight deviation from truth as to the day, the hour, the place, the number or the names of those present. Most of us have observed, ifnot taken part in, some of these petty fictions. But we are too apt toforget them when we come to criticise documents relating to the past. The _authentic_ character of the documents contributes to the illusion;we instinctively make _authentic_ a synonym of _sincere_. The rigidrules which govern the composition of every authentic document seem toguarantee sincerity; they are, on the contrary, an incentive to falsify, not the main facts, but the accessory circumstances. From the fact of aperson having signed a report we may infer that he agreed to it, but notthat he was actually present at the time when the report mentions him ashaving been present. 

(3) The author viewed with sympathy or antipathy a group of men (nation, party, denomination, province, city, family), or an assemblage ofdoctrines or institutions (religion, school of philosophy, politicaltheory), and was led to distort facts in such a manner as to representhis friends in a favourable and his opponents in an unfavourable light. These are instances of a general bias which affects all the statementsof an author, and they are so obvious that the ancients perceived themand gave them names (_studium_ and _odium_); from ancient times it hasbeen a literary commonplace for historians to protest that they havesteered clear of both. 

(4) The author was induced by private or collective vanity to violatetruth for the purpose of exalting himself or his group. He made suchstatements as he thought likely to give the reader the impression thathe and his possessed qualities deserving of esteem. We have therefore toinquire whether a given statement may not be influenced by vanity. Butwe must take care not to represent the author's vanity to ourselves asbeing exactly like our own vanity or that of our contemporaries. Different people are vain for different reasons; we must inquire whatwas our author's particular vanity; he may have lied in order toattribute to himself or his friends actions which we should considerdishonourable. Charles IX. Falsely boasted of having organised theMassacre of St. Bartholomew. There is, however, a kind of vanity whichis universal, and that is, the desire to appear to be a person ofexalted rank playing an important part in affairs. We must, therefore, always distrust a statement which attributes to the author or his groupa high place in the world. [149]

(5) The author desired to please the public, or at least to avoidshocking it. He has expressed sentiments and ideas in harmony with themorality or the fashion of his public; he has distorted facts in orderto adapt them to the passions and prejudices of his time, even thosewhich he did not share. The purest types of this kind of falsehood arefound in ceremonial forms, official formulæ, declarations prescribed byetiquette, set speeches, polite phrases. The statements which come underthis head are so open to suspicion that we are unable to derive fromthem any information about the facts stated. We are all aware of this sofar as relates to the contemporary formulæ of which we see instancesevery day, but we often forget it in the criticism of documents, especially those belonging to an age from which few documents have comedown to us. No one would think of looking for the real sentiments of aman in the assurances of respect with which he ends his letters. Butpeople believed for a long time in the humility of certainecclesiastical dignitaries of the middle ages, because, on the day oftheir election, they began by refusing an office of which they declaredthemselves unworthy, till at last comparison showed that this refusalwas a mere conventional form. And there are still scholars who, like theBenedictines of the eighteenth century, look in the chancery-formulæ ofa prince for information as to his piety or his liberality. [150]

In order to recognise these conventional declarations there are twolines of general study to be pursued: the one is directed to the author, and seeks to discover what was the public he addressed, for in one andthe same country there are usually several different publics, each ofwhich has its own code of morals or propriety; the other is directedtowards the public, and seeks to determine its morals or its manners. 

(6) The author endeavoured to please the public by literary artifices. He distorted facts in order to embellish them according to his ownæsthetic notions. We have therefore to look for the ideal of the authoror of his time, in order to be on our guard against passages distortedto suit that ideal. But without special study we may calculate on thecommon kinds of literary distortion. Rhetorical distortion consists inattributing to persons noble attitudes, acts, sentiments, and, aboveall, words: this is a natural tendency in young boys who are beginningto practise the art of composition, and in writers still in asemi-barbarous stage; it is the common defect of the mediævalchroniclers. [151] Epic distortion embellishes the narrative by addingpicturesque details, speeches delivered by the persons concerned, numbers, sometimes names of persons; it is dangerous, because theprecision of the details produces an illusive appearance of truth. [152]Dramatic distortion consists in grouping the facts in such a way as toenhance the dramatic effect by concentrating facts, which in realitywere separate, upon a single moment, a single person, or a single group. Writing of this kind is what we call "truer than the truth. " It is themost dangerous form of distortion, the form employed by artistichistorians, by Herodotus, Tacitus, the Italians of the Renaissance. Lyrical distortion exaggerates the intensity of the sentiments and theemotions of the author and his friends: we should remember this when weattempt to reconstruct "the psychology" of a person. 

Literary distortion does not much affect archives (though instances ofit are found in most charters of the eleventh century); but itprofoundly modifies all literary texts, including the narratives ofhistorians. Now, the natural tendency is to trust writers more readilywhen they have talent, and to admit statements with less difficulty whenthey are presented in good literary form. Criticism must counteract thistendency by the application of the paradoxical rule, that the moreinteresting a statement is from the artistic point of view, [153] themore it ought to be suspected. We must distrust every narrative whichis very picturesque or very dramatic, in which the personages assumenoble attitudes or manifest great intensity of feeling. 

This first series of questions will yield the _provisional_ result ofenabling us to note the statements which have a _chance_ of beingmendacious. 

V. The second series of questions will be of use in determining whetherthere is any reason to distrust the accuracy of a statement. Was theauthor in one of those situations which cause a man to make mistakes? Asin dealing with good faith, we must look for these conditions both asaffecting the document as a whole, and as affecting each of theparticular statements in it. 

The practice of the established sciences teaches us the conditions of anexact knowledge of facts. There is only one scientific procedure forgaining knowledge of a fact, namely, _observation_; every statement, therefore, must rest, directly or indirectly, upon an observation, andthis observation must have been made correctly. 

The set of questions by the aid of which we investigate theprobabilities of error may be drawn up in the light of experience, whichbrings before us the most common cases of error. 

(1) The author was in a situation to observe the fact, and supposed hereally had observed it; he was, however, prevented from doing so by someinterior force of which he was unconscious, an hallucination, anillusion, or a mere prejudice. It would be useless, as well asimpossible, to determine which of these agencies was at work; it isenough to ascertain whether the author had a tendency to observe badly. It is scarcely possible in the case of a particular statement torecognise that it was the result of an hallucination or an illusion. Atthe most we may learn, either from information derived from othersources or by comparison, that an author had a _general_ propensity tothis kind of error. 

There is a better chance of recognising whether a statement was due toprejudice. In the life or the works of an author we may find the tracesof his dominant prejudices. With reference to each of his particularstatements, we ought to ask whether it is not the result of apreconceived idea of the author on a class of men or a kind of facts. This inquiry partly coincides with the search for motives of falsehood:interest, vanity, sympathy, and antipathy give rise to prejudices whichalter the truth in the same manner as wilful falsehood. We thereforeemploy the questions already formulated for the purpose of testing goodfaith. But there is one to be added. In putting forward a statement hasthe author been led to distort it unconsciously by the circumstance thathe was answering a question? This is the case of all statements obtainedby interrogating witnesses. Even apart from the cases where the personinterrogated seeks to please the proposer of the question by giving ananswer which he thinks will be agreeable to him, every question suggestsits own answer, or at least its form, and this form is dictatedbeforehand by some one unacquainted with the facts. It is thereforenecessary to apply a special criticism to every statement obtained byinterrogation; we must ask what was the question put, and what were thepreconceptions to which it may have given rise in the mind of the personinterrogated. 

(2) The author was badly situated for observing. The practice of thesciences teaches us what are the conditions for correct observation. Theobserver ought to be placed where he can see correctly, and should haveno practical interest, no desire to obtain a particular result, nopreconceived idea about the result. He ought to record the observationimmediately, in a precise system of notation; he ought to give a preciseindication of his method. These conditions, which are insisted on in thesciences of observation, are never completely fulfilled by the authorsof documents. 

It would be useless, therefore, to ask whether there have been chancesof inaccuracy; _there always have been_, and it is just this thatdistinguishes a _document_ from an _observation_. It only remains tolook for the obvious causes of error in the conditions of observation:to inquire whether the observer was in a place where he could not see orhear well, as would be the case, for example, with a subordinate whoshould presume to narrate the secret deliberations of a council ofdignitaries; whether his attention was greatly distracted by thenecessity for action, as it would be on the field of battle, forexample; whether he was inattentive because the facts had littleinterest for him; whether he lacked the special experience or generalintelligence necessary for understanding the facts; whether he analysedhis impressions badly, or confused different events. Above all, we mustask when he _wrote down_ what he saw or heard. This is the mostimportant point: the only exact observation is the one which isrecorded immediately it is made; such is the constant procedure in theestablished sciences; an impression committed to writing later on isonly a recollection, liable to be confused in the memory with otherrecollections. _Memoirs_ written several years after the facts, often atthe very end of the author's career, have introduced innumerable errorsinto history. It must be made a rule to treat _memoirs_ with specialdistrust as second-hand documents, in spite of their appearance of beingcontemporary testimony. 

(3) The author states facts which he could have observed, but to whichhe did not take the trouble to attend. From idleness or negligence hereported details which he has merely inferred, or even imagined atrandom, and which turn out to be false. This is a common source oferror, though it does not readily occur to one, and is to be suspectedwherever the author was obliged to procure information in which he tooklittle interest, in order to fill up a blank form. Of this kind areanswers to questions put by an authority (it is enough to observe howmost official inquiries are conducted in our own day), and detailedaccounts of ceremonies or public functions. There is too strong atemptation to write the account from the programme, or in agreement withthe usual order of the proceedings. How many accounts of meetings of allkinds have been published by reporters who were not present at them!Similar efforts of imagination are suspected--sometimes, it is thought, clearly recognised--in the writings of mediæval chroniclers. [154] Therule, then, will be to distrust all narratives conforming too closely toa set formula. 

(4) The fact stated is of such a nature that it could not have beenlearnt by observation alone. It may be a hidden fact--a private secret, for example. It may be a fact relating to a collectivity, and applyingto an extensive area or a long period of time; for example, the commonact of a whole army, a custom common to a whole people or a whole age, astatistical total obtained by the addition of numerous items. It may bea comprehensive judgment on the character of a man, a group, a custom, an event. Here we have to do with propositions derived from observationsby synthesis or inference: the author can only have arrived at themindirectly; he began with data furnished by observation, and elaboratedthem by the logical processes of abstraction, generalisation, reasoning, calculation. Two questions arise. Does it appear that the author hadsufficient data to work upon? Was he accurate, or the reverse, in hisuse of the data he had?

On the probable inaccuracies of an author, general indications may beobtained from an examination of his writings. This examination will showus how he worked: whether he was capable of abstraction, reasoning, generalisation, and what were the mistakes he was in the habit ofmaking. In order to determine the value of the data, we must criticiseeach statement separately; we must imagine the conditions under whichthe author observed, and ask ourselves whether he was able to procurethe necessary data for his statement. This is an indispensableprecaution in dealing with large totals in statistics and descriptionsof popular usages; for it is possible that the author may have obtainedthe total he gives by a process of conjectural valuation (this is theordinary practice in stating the number of combatants or killed in abattle), or by combining subsidiary totals, all of which were notaccurate; it is possible that he may have extended to a whole people, awhole country, a whole period, that which was true only of a small groupknown to him. [155]

VI. These two first series of questions bearing on the good faith andthe accuracy of the statements in the document are based on thesupposition that the author has observed the fact himself. This is afeature common to all reports of observations in the establishedsciences. But in history there is so great a dearth of direct_observations_, of even moderate value, that we are obliged to turn toaccount documents which every other science would reject. [156] Take anynarrative at random, even if it be the work of a contemporary, it willbe found that the facts observed by the author are never more than apart of the whole number. In nearly every document the majority of thestatements do not come from the author at first hand, but arereproductions of the statements of others. Even where a general relatesa battle in which he commanded, he does not communicate his ownobservations, but those of his officers; his narrative is in a largemeasure a "second-hand document. "[157]

In order to criticise a second-hand statement it is no longer enough toexamine the conditions under which the author of the document worked:this author is, in such a case, a mere agent of transmission; the trueauthor is the person who supplied him the information. The critic, therefore, must change his ground, and ask whether the informantobserved and reported correctly; and if he too had the information fromsome one else (the commonest case), the chase must be pursued from oneintermediary to another, till the person is found who first launched thestatement on its career, and with regard to him the question must beasked: Was he an accurate observer?

Logically such a search is not inconceivable; ancient collections ofArab traditions give lists of their successive guarantors. But, inpractice, lack of documents nearly always prevents us from getting asfar as the observer of a fact; the observation remains anonymous. Ageneral question then presents itself: How are we to criticise ananonymous statement? It is not only "anonymous documents" with which weare concerned, where the composition as a whole is the work of anunknown author; even when the author is known, this question arises withrespect to each statement of his drawn from an unknown source. 

Criticism works by reproducing the conditions under which an authorwrote, and has hardly anything to take hold of where a statement isanonymous. The only method left is to examine the general conditions ofthe document. We may inquire whether there is any feature common to allthe statements of a document indicating that they all proceed frompersons having the same prejudices or passions: in this case thetradition followed by the author is biassed; the tradition followed byHerodotus has both an Athenian bias and a Delphic bias. In respect ofeach fact derived from such a tradition we must ask whether it has notbeen distorted by the interest, the vanity, or the prejudices of thegroup concerned. We may even ignore the author, and ask whether therewas anything likely to make for or against correct observation, commonto all the men of the time and country in which the observation musthave been made: for example, what means of information, and whatprejudices, had the Greeks of Herodotus' time with respect to theScythians. 

The most useful of all these general inquiries has reference to thatmode of transmitting anonymous statements which is called _tradition_. No second-hand statement has any value except in so far as it reproducesits source; every addition is an alteration, and ought to be eliminated. Similarly, all the intermediary sources are valueless except as copiesof the original statement founded directly on observation. The criticneeds to know whether this transmission from hand to hand has preservedor distorted the original statement; above all, whether the traditionembodied in the document was _written_ or _oral_. Writing fixes astatement, and ensures its being transmitted faithfully; when astatement is communicated orally, the impression in the mind of thehearer is apt to be modified by confusion with other impressions; inpassing from one intermediary to another the statement is modified atevery step, [158] and as these modifications arise from different causes, there is no possibility of measuring or correcting them. 

Oral tradition is by its nature a process of continual alteration; hencein the established sciences only written transmission is accepted. Historians have no avowable motive for proceeding differently, at anyrate when it is a case of establishing a particular fact. We musttherefore search documents for statements derived from oral tradition inorder that we may suspect them. We rarely have direct information as tostatements being thus derived; authors who borrow from oral traditionare not anxious to proclaim the fact. [159] There is thus only anindirect method, and that is to ascertain that written transmission wasimpossible; we may then be sure that the fact reached the author only byoral tradition. We have therefore to ask the question: In this periodand in this group of men was it customary to commit to writing facts ofthis kind? If the answer is negative, the fact considered rests on oraltradition alone. 

The most striking form of oral tradition is _legend_. It arises amonggroups of men with whom the spoken word is the only means oftransmission, in barbarous societies, or in classes of little culture, such as peasants or soldiers. In this case it is the whole group offacts which is transmitted orally and assumes the legendary form. Thereis a legendary period in the early history of every people: in Greece, at Rome, among the Germanic and Slavonic races, the most ancientmemories of the people form a stratum of legend. In periods ofcivilisation popular legends continue to exist in reference to eventswhich strike the imagination of the people. [160] Legend is exclusivelyoral tradition. 

When a people has emerged from the legendary period and begun to commitits history to writing, oral tradition does not come to an end, but onlyapplies to a narrower sphere; it is now restricted to facts which arenot registered, whether because they are by their nature secret, orbecause no one takes the trouble to record them, such as privateactions, words, the details of events. Thus arise _anecdotes_, whichhave been named "the legends of civilised society. " Like legends theyhave their origin in confused recollections, allusions, mistakeninterpretations, imaginings of all kinds which fasten upon particularpersons and events. 

Legends and anecdotes are at bottom mere popular beliefs, arbitrarilyattached to historical personages; they belong to folk-lore, not tohistory. [161] We must therefore guard against the temptation to treatlegend as an alloy of accurate facts and errors out of which it ispossible by analysis to extract grains of historical truth. A legend isa conglomerate in which there may be some grains of truth, and which mayeven be capable of being analysed into its elements; but there is nomeans of distinguishing the elements taken from reality from those whichare the work of imagination. To use Niebuhr's expression, a legend is "amirage produced by an invisible object according to an unknown law ofrefraction. "

The crudest analytical procedure consists in rejecting those details inthe legendary narrative which appear impossible, miraculous, contradictory, or absurd, and retaining the rational residue ashistorical. This is how the Protestant rationalists of the eighteenthcentury treated biblical narratives. One might as well amputate themarvellous part of a fairy tale, suppress Puss in Boots, and keep theMarquis of Carabas as an historical character. A more refined but noless dangerous method is to compare different legends in order to deducetheir common historical basis. Grote[162] has shown, with reference toGreek tradition, that it is impossible to extract any trustworthyinformation from legend by any process whatever. [163] We must make upour minds to treat legend as a product of imagination; we may look in itfor a people's conceptions, not for the external facts in that people'shistory. The rule will be to reject every statement of legendary origin;nor does this apply only to narratives in legendary form: a narrativewhich has an historical appearance, but is founded on the data oflegend, the opening chapters of Thucydides for example, ought equally tobe discarded. 

In the case of written transmission it remains to inquire whether theauthor reproduced his source without altering it. This inquiry formspart of the critical investigation of the sources, [164] so far as it canbe pursued by a comparison of texts. But when the source has disappearedwe are reduced to internal criticism. We ask, first of all, whether theauthor can have had exact information, otherwise his statement isvalueless. We next put to ourselves the general question: Was the authorin the habit of altering his sources, and in what manner? And in regardto each separate second-hand statement we ask whether it has theappearance of being an exact reproduction or an arrangement. We judge bythe form: when we meet with a passage whose style is out of harmony withthe main body of the composition, we have before us a fragment of anearlier document; the more servile the reproduction the more valuable isthe passage, for it can contain no exact information beyond what wasalready in the source. 

VII. In spite of all these investigations, criticism never succeeds indetermining the parentage of all the statements to the extent of findingout who it was that observed, or even recorded, each fact. In most casesthe inquiry ends in leaving the statement anonymous. 

We are thus confronted with a fact, observed we know not by whom norhow, recorded we know not when nor how. No other science accepts factswhich come in such a condition, without possibility of verification, subject to incalculable chances of error. But history can turn them toaccount, because it does not, like the other sciences, need a supply offacts which are difficult to ascertain. 

The notion of a _fact_, when we come to examine it precisely, reduces toan affirmative judgment having reference to external reality. Theoperations by which we arrive at such a judgment are more or lessdifficult, and the risk of error is greater or smaller according to thenature of the realities investigated and the degree of precision withwhich we wish to formulate them. Chemistry and biology need to discernfacts of a delicate order, rapid movements, transient states, and tomeasure them in exact figures. History can work with facts of a muchcoarser kind, spread over a large extent of space or time, such as theexistence of a custom, of a man, of a group, even of a people; and thesefacts may be roughly expressed in vague words conveying no idea ofaccurate measurement. With such easily observed facts as these to dealwith, history can afford to be much less exacting with regard to theconditions of observation. The imperfection of the means of informationis compensated by a natural faculty of being satisfied with informationwhich can easily be obtained. 

Documents supply little else besides ill-verified facts, subject to manyrisks of falsehood or error. But there are some facts in respect ofwhich it is very difficult to lie or be mistaken. The last series ofquestions which the critic should ask is intended to distinguish, in themass of alleged facts, those which by their nature are little subject tothe risk of alteration, and which are therefore very probably correct. We know what, in general, are the classes of facts which enjoy thisprivilege; we are thus enabled to draw up a list of questions forgeneral use, and in applying them to any particular case we ask whetherthe fact in question comes under any of the heads specified in advance. 

(1) The fact is of a nature to render falsehood improbable. A man liesin order to produce an impression, and has no motive to lie in a casewhere he believes that the false impression would be of no use, or thatthe falsehood would be ineffectual. In order to determine whether theauthor was in such a situation there are several questions to be asked. 

(_a_) Is the fact stated manifestly prejudicial to the effect which theauthor wished to produce? Does it run counter to the interest, thevanity, the sentiments, the literary tastes of the author and his group;or to the opinions which he made a point of not offending? In such acase there is a probability of good faith. But in the application ofthis criterion there is danger; it has often been wrongly used, and intwo ways. One of these is to take for a confession what was meant for aboast, as the declaration of Charles IX. That he was responsible for theMassacre of St. Bartholomew. Or again, we trust without examination anAthenian who speaks ill of the Athenians, or a Protestant who accusesother Protestants. But it is quite possible the author's notions of hisinterest or honour were very different from ours;[165] or he may havewished to calumniate fellow-citizens who did not belong to his ownparty, or co-religionists who did not belong to his own sect. Thiscriterion must therefore be restricted to cases where we know exactlywhat _effect_ he wished to produce, and in what _group_ he was mainlyinterested. 

(_b_) Was the fact stated so obviously known to the public that theauthor, even if tempted to falsehood, would have been restrained by thecertainty of being detected? This is the case with facts which are easyto verify, which are not remote in point of time or space, which applyto a wide area or a long period, especially if the public had anyinterest in verifying them. But the fear of detection is only anintermittent check, opposed by interest whenever the author has anymotive for deceiving. It acts unequally on different minds--strongly onmen of culture and self-control who understand their public, feebly inbarbarous ages and on passionate men. [166] This criterion, therefore, isto be restricted to cases where we know what idea the author had of hisreaders, and whether he was dispassionate enough to keep them in mind. 

(_c_) Was the fact stated _indifferent_ to the author, so that he had notemptation to misrepresent it? This is the case with facts of a generalkind, usages, institutions, objects, persons, which the author mentionsincidentally. A narrative, even a false one, cannot be composedexclusively of falsehoods; the author must localise his facts, and needsto surround them with a framework of truth. The facts which form thisframework had no interest for him; at that time every one knew them. Butfor us they are instructive, and we can depend on them, for the authorhad no intention of deceiving us. 

(2) The fact was of a kind to render error improbable. Numerous as thechances of error are, still there are facts so "big" it is hard to bemistaken about them. We have, then, to ask whether the alleged fact waseasy to ascertain: (_a_) Did it cover a long period of time, so that itmust have been frequently observed? Take, for example, the case of amonument, a man, a custom, an event which was in progress for aconsiderable time. (_b_) Did it cover a wide area, so that many peopleobserved it?--as, for example, a battle, a war, a custom common to awhole people. (_c_) Is it expressed in such general terms thatsuperficial observation was enough to discover it?--as the mereexistence of a man, a city, a people, a custom. Facts of this large andgeneral kind make up the bulk of historical knowledge. 

(3) The fact was of such a nature that it would not have been statedunless it was true. A man does not declare that he has seen somethingcontrary to his expectations and habits of mind unless observation hascompelled him to admit it. A fact which seems very improbable to the manwho relates it has a good chance of being true. We have, then, to askwhether the fact stated was in contradiction with the author's opinions, whether it is a phenomenon of a kind unknown to him, an action or acustom which seems unintelligible to him; whether it is a saying whoseimport transcends his intelligence, such as the sayings of Christreported in the Gospels, or the answers made by Joan of Arc to questionsput to her in the course of her trial. But we must guard against judgingof the author's ideas by our own standards: when men who are accustomedto believe in the marvellous speak of monsters, of miracles, of wizards, there is nothing in these to contradict their expectations, and thecriterion does not apply. 

VIII. We have at last reached the end of this description of thecritical operations; its length is due to the necessity of describingsuccessively operations which are performed simultaneously. We will nowconsider how these methods are applied in practice. 

If the text be one whose interpretation is debatable, the examination isdivided into two stages: the first comprises the reading of the textwith a view to the determination of the meaning, without attempting todraw any information from it; the second comprises the critical study ofthe facts contained in the document. In the case of documents whosemeaning is clear, we may begin the critical examination on the firstreading, reserving for separate study any individual passages ofdoubtful meaning. 

We begin by collecting the _general_ information we possess about thedocument and the author, with the special purpose of discovering theconditions which may have influenced the production of the document--theepoch, the place, the purpose, the circumstances of its composition; theauthor's social status, country, party, sect, family, interests, passions, prejudices, linguistic habits, methods of work, means ofinformation, culture, abilities, and mental defects; the nature of thefacts and the mode of their transmission. Information on all thesepoints is supplied by the preparatory critical investigation ofauthorship and sources. We now combine the different heads, mentallyapplying the set of general critical questions; this should be done atthe outset, and the results impressed on the memory, for they will needto be present to the mind during the remainder of the operations. 

Thus prepared, we attack the document. As we read we mentally analyseit, destroying all the author's combinations, discarding all hisliterary devices, in order to arrive at the facts, which we formulate insimple and precise language. We thus free ourselves from the deferenceimposed by artistic form, and from all submission to the author'sideas--an emancipation without which criticism is impossible. 

The document thus analysed resolves into a long series of the author'sconceptions and statements as to facts. 

With regard to each statement, we ask ourselves whether there is aprobability of their being false or erroneous, or whether, on the otherhand, there are exceptional chances in favour of good faith andaccuracy, working through the list of critical questions prepared forparticular cases. This list of questions must be always present to themind. At first it may seem cumbersome, perhaps pedantic; but as it willbe applied more than a hundred times in each page of the document, itwill in the end be used unconsciously. As we read a text, all thereasons for distrust or confidence will occur to the mindsimultaneously, combined into a single impression. 

Analysis and critical questioning will then have become a matter ofinstinct, and we shall have acquired for ever that methodicallyanalytical, distrustful, not too respectful turn of mind which is oftenmystically called "the critical sense, " but which is nothing else thanan unconscious _habit_ of criticism. 

CHAPTER VIII

THE DETERMINATION OF PARTICULAR FACTS

Critical analysis yields in the result a number of conceptions andstatements, accompanied by comments on the probability of the factsstated being accurate. It remains to examine how we can deduce fromthese materials those particular historical facts which are to form thebasis of scientific knowledge. Conceptions and statements are twodifferent kinds of results, and must be treated by different methods. 

I. Every conception which is expressed in writing or by any illustrativerepresentation is in itself a definite, unimpeachable fact That which isexpressed must have first been present in the mind of some one--if notin that of the author, who may have reproduced a formula he did notunderstand, then in the mind of the man who originated the formula. Theexistence of a conception may be learnt from a single instance andproved from a single document. Analysis and interpretation are thussufficient for the purpose of drawing up the complete list of thosefacts which form the basis of the history of the arts, the sciences, orof doctrines. [167] It is the task of external criticism to localisethese facts by determining the epoch, the country, the author of eachconception. The duration, geographical distribution, origin, andfiliation of conceptions belong to historical synthesis. Internalcriticism has nothing to do here; the fact is taken directly from thedocument. 

We may advance a step farther. In themselves conceptions are nothing butfacts in psychology; but imagination does not create its objects, ittakes the elements of them from reality. Descriptions of imaginary factsare constructed out of the real facts which the author has observed inhis experience. These elements of knowledge, the raw material of theimaginary description, may be sought for and isolated. In dealing withperiods and with classes of facts for which documents arerare--antiquity, for example, and the usages of private life--theattempt has been made to lay under contribution works of literature, epic poems, novels, plays. [168] The method is legitimate, but onlywithin the limits of certain restrictions which one is very apt toforget. 

(1) It does not apply to social facts of a psychological order, themoral or artistic standards of a society; the moral and æstheticconceptions in a document give at most the individual standards of theauthor; we have no right to conclude from these to the morals or theæsthetic tastes of the age. We must at least wait till we have comparedseveral different authors of the same period. 

(2) Descriptions even of physical facts and objects may be products ofthe author's imagination. It is only the _elements_ of them which weknow to be certainly real; all that we can assert is the separateexistence of the irreducible elements, form, material, colour, number. When the poet speaks of golden gates or silver bucklers, we cannot inferthat golden gates and silver bucklers ever existed in reality; nothingis certain beyond the separate existence of gates, bucklers, gold, andsilver. The analysis must therefore be carried to the point ofdistinguishing those elements which the author must necessarily havetaken from experience: objects, their purpose, ordinary actions. 

(3) The conception of an object or an action proves that it existed, butnot that it was common; the object or action may have been unique, orrestricted to a very small circle; poets and novelists are fond oftaking their models from an exceptional world. 

(4) The facts yielded by this method are not localised in space or time;the author may have taken them from a time or country not his own. 

All these restrictions may be summarised as follows: before drawing anyinference from a work of literature as to the state of the society inwhich the author lived, we should ask ourselves what would be the worthof a similar inference as to contemporary manners drawn from a modernnovel. 

With the facts yielded by conceptions we may join those indifferentfacts of an obvious and elementary character which the author hasstated almost without thinking. Logically we have no right to call themcertain, for we do sometimes meet with men who make mistakes aboutobvious and elementary facts, and others who lie even on indifferentmatters. But such cases are so rare that there is not much danger inadmitting as certain facts of this kind which are supported by a singledocument, and this is how we deal, in practice, with periods of whichlittle is known. The institutions of the Gauls and Germans are describedfrom the unique texts of Cæsar and Tacitus. Facts so easy to discoverare forced upon the authors of descriptions much as realities are forcedupon poets. 

II. On the other hand, a statement in a document as to an objective factis never enough to establish that fact. The chances of falsehood orerror are so many, the conditions which gave rise to the statement areso little known, that we cannot be sure that none of these chances hastaken effect. The critical examination provides no definitive solution;it is indispensable if we are to avoid error, but it is insufficient toconduct us to truth. 

Criticism can _prove_ no fact; it only yields probabilities. Its end andresult is to decompose documents into statements, each labelled with anestimate of its value--worthless statement, statement open to suspicion(strong or weak), statement probably (or, very probably) true, statementof unknown value. 

Of all these different kinds of results one only is definitive--_thestatement of an author who can have had no information on the fact hestates is null and void_; it is to be rejected as we reject anapocryphal document. [169] But criticism here merely destroys illusorysources of information; it supplies nothing certain to take their place. The only sure results of criticism are _negative_. All the positiveresults are subject to doubt; they reduce to propositions of the form:"There are chances for or against the truth of such and such astatement. " Chances only. A statement open to suspicion may turn out tobe true; a statement whose truth is probable may, after all, be false. Instances occur continually, and we are never sufficiently wellacquainted with the conditions under which the observation was made to_know_ whether it was made ill or well. 

In order to obtain a definitive result we require a final operation. After passing through the ordeal of criticism, statements presentthemselves as probable or improbable. But even the most probable ofthem, taken by themselves, remain mere probabilities: to pass from themto categorical propositions in scientific form is a step we have noright to take; a proposition in a science is an assertion not open todebate, and that is what the statements we have before us are not. It isa principle common to all sciences of observation not to base ascientific conclusion on a single observation; the fact must have beencorroborated by several independent observations before it is affirmedcategorically. History, with its imperfect modes of acquiringinformation, has less right than any other science to claim exemptionfrom this principle. An historical statement is, in the most favourablecase, but an indifferently made observation, and needs otherobservations to corroborate it. 

It is by combining observations that every science is built up: ascientific fact is a centre on which several different observationsconverge. [170] Each observation is subject to chances of error whichcannot be entirely eliminated; but if several observations agree, thiscan hardly be in virtue of a common error: the more probable explanationof the agreement is that the observers have all seen the same realityand have all described it correctly. Errors are personal and tend todiverge; it is the correct observations that agree. 

Applied to history, this principle leads to a last series of operations, intermediate between purely analytical criticism and the syntheticoperations--the comparison of statements. 

We begin by classifying the results yielded by critical analysis in sucha way as to bring together those statements which relate to the samefact. The operation is facilitated mechanically by the method of slips. Either each statement has been entered on a separate slip, or else asingle slip has been assigned for each fact, and the differentstatements relating to it entered upon the slip as met with in thecourse of reading. By bringing the statements together we learn theextent of our information on the fact; the definitive conclusion dependson the relation between the statements. We have, then, to studyseparately the different cases which may occur. 

III. Most frequently, except in contemporary history, the documents onlysupply a single statement on a given fact. In such a case all the othersciences follow an invariable rule: an isolated observation is notadmitted into science; it is quoted (with the observer's name), but noconclusions are drawn from it. Historians have no avowable motive forproceeding otherwise. When a fact is supported by no more than thestatement of a single man, however honest he may be, historians oughtnot to assert it, but to do as men of science do--give the reference(Thucydides states, Cæsar says that . .. ); this is all they have a rightto affirm. In reality they all retain the habit of stating facts, as wasdone in the middle ages, on the _authority_ of Thucydides or of Cæsar;many are simple enough to do so in express terms. Thus, allowingthemselves to be guided by natural credulity, unchecked by science, historians end by admitting, on the insufficient presumption afforded bya unique document, any statement which does not happen to becontradicted by another document. Hence the absurd consequence thathistory is more positive, and seems better established in regard tothose little known periods which are represented by a single writer thanin regard to facts known from thousands of documents which contradicteach other. The wars of the Medes known to Herodotus alone, theadventures of Fredegonda related by none but Gregory of Tours, are lesssubject to discussion than the events of the French Revolution, whichhave been described by hundreds of contemporaries. This is adiscreditable state of things which cannot be ended except by arevolution in the minds of historians. 

IV. When we have several statements relating to the same fact, they maycontradict each other or they may agree. In order to be certain thatthey really do contradict each other, we have to make sure that they doactually relate to the same fact. Two apparently contradictorystatements may be merely parallel; they may not relate exactly to thesame moment, the same place, the same persons, the same episodes of anevent, and they may be both correct. [171] We must not, however, inferthat they confirm each other; each comes under the category of uniquestatements. 

If the contradiction is real, at least one of the statements is false. In such cases it is a natural tendency to seek to reconcile them by acompromise--to split the difference. This peace-making spirit is thereverse of scientific. A says two and two make four; B says they makefive. We are not to conclude that two and two make four and a half; wemust examine and see which is right. This examination is the work ofcriticism. Of two contradictory statements, it nearly always happensthat one is open to suspicion; this should be rejected if the competingstatement has been judged very probably true. If both are open tosuspicion, we abstain from drawing any conclusion. We do the same ifseveral statements open to suspicion agree together as against a singlestatement which is not suspected. [172]

V. When several statements agree, it is still necessary to resist thenatural tendency to believe that the fact has been demonstrated. Thefirst impulse is to count each document as one source of information. Weare well aware in matters of every-day life that men are apt to copyeach other, that a single narrative often serves the turn of severalnarrators, that several newspapers sometimes happen to publish the samecorrespondence, that several reporters sometimes agree to let one oftheir number do the work for all. We have, in such a case, severaldocuments, several statements--have we the same number of observations?Obviously not. When one statement reproduces another, it does notconstitute a new observation, and even if an observation were to bereproduced by a hundred different authors, these hundred copies wouldamount to no more than one observation. To count them as a hundred wouldbe the same thing as to count a hundred printed copies of the same bookas a hundred different documents. But the respect paid to "historicaldocuments" is sometimes stronger than obvious truth. The same statementoccurring in several different documents by different authors has anillusory appearance of multiplicity; an identical fact related in tendifferent documents at once gives the impression of being established byten agreeing observations. This impression is to be distrusted. Anagreement is only conclusive when the agreeing statements represent_observations_ which are independent of each other. Before we draw anyconclusion from an agreement we must examine whether it is an agreementbetween _independent_ observations. Two operations are thus required. 

(1) We begin by inquiring whether the statements are independent, or arereproductions of one and the same observation. This inquiry is partlythe work of that part of external criticism which deals with theinvestigation of sources;[173] but that investigation only touches therelations between written documents, and stops short when it hasdetermined which passages of an author are borrowed from other authors. Borrowed passages are to be rejected without discussion. But the samework remains to be done in reference to statements which were notcommitted to writing. We have to compare the statements which relate tothe same fact, in order to find out whether they proceeded originallyfrom different observers, or at least from different observations. 

The principle is analogous to that employed in the investigation ofsources. The details of a social fact are so manifold, and there are somany different ways of looking at the same fact, that two independentobservers cannot possibly give completely coincident accounts; if twostatements present the same details in the same order, they must bederived from a common observation; different observations are bound todiverge somewhere. We may often apply an _a priori_ principle: if thefact was of such a nature that it could only be observed or reported bya single observer, then all the accounts of it must be derived from asingle observation. These principles[174] enable us to recognise manycases of different observations, and still more numerous cases ofobservations being reproduced. 

There remains a great number of doubtful cases. The natural tendency isto treat them as if they were cases of independent observation. But thescientific procedure would be the exact reverse of this: as long as thestatements are not proved to be independent we have no right to assumethat their agreement is conclusive. 

It is only after we have determined the relations between the differentstatements that we can begin to count them and examine into theiragreement. Here again we have to distrust the first impulse; the kind ofagreement which is really conclusive is not, as one would naturallyimagine, a perfect similarity between two narratives, but an occasionalcoincidence between two narratives which only partially resemble eachother. The natural tendency is to think that the closer the agreementis, the greater is its demonstrative power; we ought, on the contrary, to adopt as a rule the paradox that an agreement proves more when it isconfined to a small number of circumstances. It is at such points ofcoincidence between diverging statements that we are to look forscientifically established historical facts. 

(2) Before drawing any conclusions it remains to make sure whether the_different_ observations of the same fact are entirely _independent_;for it is possible that one may have influenced another to such a degreethat their agreement is inconclusive. We have to guard against thefollowing cases:--

(_a_) The different observations have been made by the same author, whohas recorded them either in the same or in different documents; specialreasons must then be had before it can be assumed that the author reallymade the observation afresh, and did not content himself with merelyrepeating a single observation. 

(_b_) There were several observers, but they commissioned one of theirnumber to write a single document. We have to ascertain whether thedocument merely gives the statements of the writer, or whether the otherobservers checked his work. 

(_c_) Several observers recorded their observations in differentdocuments, but under similar conditions. We must apply the list ofcritical questions in order to ascertain whether they were not allsubject to the same influences, predisposing to falsehood or error;whether, for example, they had a common interest, a common vanity, orcommon prejudices. 

The only observations which are certainly independent are those whichare contained in different documents, written by different authors, whobelonged to different groups, and worked under different conditions. Cases of perfectly conclusive agreement are thus rare, except inreference to modern periods. 

The possibility of proving an historical fact depends on the number ofindependent documents relating to it which have been preserved, and thepreservation of the documents is a matter of chance; this explains theshare which chance has in the formation of historical science. 

The facts which it is possible to establish are chiefly those whichcover a large extent of space or time (sometimes called _general_facts), customs, doctrines, institutions, great events; they were easierto observe than the others, and are now easier to prove. Historicalmethod is not, however, essentially powerless to establish facts ofshort duration and limited extent (those which are called _particularfacts_), such as a saying, a momentary act. It is enough that severalpersons should have been present when the fact occurred, that theyshould have recorded it, and that their writings should have come downto us. We know what were the words which Luther uttered at the Diet ofWorms; we know that he did not say what tradition puts in his mouth. This concurrence of favourable conditions becomes more and more frequentwith the organisation of newspapers, of shorthand writers, and ofdepositories of documents. 

In the case of antiquity and the middle ages historical knowledge islimited to general facts by the scarcity of documents. In dealing withcontemporary history it is possible to include more and more particularfacts. The general public supposes the opposite of this; it issuspicious about contemporary facts, with reference to which it seescontradictory narratives circulating, and believes without hesitationancient facts, which it does not see contradicted anywhere. Itsconfidence is at its greatest in respect of that history which we havenot the means of knowing, and its scepticism increases with the means ofknowledge. 

VI. _Agreement between documents_ leads to conclusions which are not allof them definitive. In order to complete and rectify our conclusions wehave still to study _the harmony of the facts_. 

Several facts which, taken in isolation, are only imperfectly proved, may confirm each other in such a manner as to produce a collectivecertainty. The facts which the documents present in isolation havesometimes been in reality sufficiently near each other to be connected. Of this kind are the successive actions of the same man or of the samegroup of men, the habits of the same group at different epochs separatedby short intervals, or of similar groups at the same epoch. It is nodoubt possible that one of several analogous facts may be true andanother false; the certainty of the first does not justify thecategorical assertion of the second. But yet the harmony of several suchfacts, each proved imperfectly, yields a kind of certainty; the facts donot, in the strict sense of the word, prove, but they _confirm_[175]each other. The doubt which attached to each one of them disappears; weobtain that species of certainty which is produced by theinterconnection of facts. Thus the comparison of conclusions which areseparately doubtful yields a whole which is morally certain. In anitinerary of a sovereign, the days and the places confirm each otherwhen they harmonize so as to form a coherent whole. An institution or apopular usage is established by the harmony of accounts, each of whichis no more than probable, relating to different times and places. 

This method is a difficult one to apply. The notion of harmony is a muchvaguer one than that of agreement. We cannot assign any precise generalrules for distinguishing facts which are sufficiently connected to forma whole, the harmony of whose parts would be conclusive; nor can wedetermine beforehand the duration and extent of that which may be takento form a whole. Facts separated by half a century of time and a hundredleagues of space may confirm each other in such a way as to establish apopular usage (for example, among the ancient Germans); but they wouldprove nothing if they were taken from a heterogeneous society subject torapid evolution (take, for example, French society in 1750, and again in1800, in Alsace and in Provence). Here we have to study the relationbetween the facts. This brings us to the beginnings of historicalconstruction; here is the transition from analytical to syntheticoperations. 

VII. But it remains to consider cases of discordance between factsestablished by documents and other facts established by other methods. It happens sometimes that a fact obtained as an historical conclusion isin contradiction with a body of known historical facts, or with the sumof our knowledge of humanity founded on direct observation, or with ascientific law established by the regular method of an establishedscience. In the first two cases the fact is only in conflict withhistory, psychology, or sociology, all imperfectly established sciences;we then simply call the fact _improbable_. If it is in conflict with atrue science it becomes a _miracle_. What are we to do with animprobable or miraculous fact? Are we to admit it after examination ofthe documents, or are we to pass on and shelve the question?

_Improbability_ is not a scientific notion; it varies with theindividual. Each person finds improbable what he is not accustomed tosee: a peasant would think the telephone much more improbable than aghost; a king of Siam refused to believe in the existence of ice. It isimportant to know who precisely it is to whom the fact appears to beimprobable. Is it to the mass who have no scientific culture? For these, science is more improbable than miracle, physiology than spiritualism;their notions of improbability are worthless. Is it to the man whopossesses scientific culture? If so, we have to deal with that whichseems improbable to a scientific mind, and it would be more accurate tosay that the fact is contrary to the results of science--that there isdisagreement between the direct observations of men of science and theindirect testimony of the documents. 

How is this conflict to be decided? The question has no great practicalinterest; nearly all the documents which relate miraculous facts arealready open to suspicion on other grounds, and would be discarded by asound criticism. But the question of miracles has raised such passionsthat it may be well to indicate how it affects the historian. [176]

The general tendency to believe in the marvellous has filled withmiraculous facts the documents of nearly every people. Historically theexistence of the devil is much better proved than that of Pisistratus:there has not been preserved a single word of a contemporary ofPisistratus saying that he has seen him; thousands of "ocular witnesses"declare they have seen the devil; few historical facts have beenestablished by so great a number of independent testimonies. However, wedo not hesitate to reject the devil and to accept Pisistratus. For theexistence of the devil would be irreconcilable with the laws of all theestablished sciences. 

For the historian the solution of the problem is obvious. [177] Theobservations whose results are contained in historical documents arenever of equal value with those of contemporary scientists; we havealready shown why. The indirect method of history is always inferior tothe direct methods of the sciences of observation. If its results do notharmonise with theirs, it is history which must give way; historicalscience, with its imperfect means of information, cannot claim to check, contradict, or correct the results of other sciences, but must ratheruse their results to correct its own. The progress of the directsciences sometimes modifies the results of historical interpretation; afact established by direct observation aids in the comprehension andcriticism of documents. Cases of stigmata and nervous anæsthesia whichhave been scientifically observed have led to the admission as true ofhistorical narratives of analogous facts, as in the case of the stigmataof certain saints and the possessed nuns of Loudun. But history cannotaid the progress of the direct sciences. It is kept at a distance fromreality by its indirect means of information, and must accept the lawsthat are established by those sciences which come into immediate contactwith reality. In order to reject one of these laws new directobservations are necessary. Such revolutions are possible, but they mustbe brought about from within. History has no power to take theinitiative in them. 

The solution is not so clear in the case of facts which do not harmonisewith a body of historical knowledge or with the sciences, still in theembryonic stage, which deal with man. It depends on the opinion we formas to the value of such knowledge. We can at least lay down thepractical rule that in order to contradict history, psychology, orsociology, we must have very strong documents, and this is a case whichhardly ever occurs. 

BOOK III

_SYNTHETIC OPERATIONS_

CHAPTER I

GENERAL CONDITIONS OF HISTORICAL CONSTRUCTION

The criticism of documents only yields isolated facts. In order toorganise them into a body of science it is necessary to perform a seriesof synthetic operations. The study of these processes of historicalconstruction forms the second half of Methodology. 

The mode of construction cannot be regulated by the ideal plan of thescience we desire to construct; it depends on the materials we have atour disposal. It would be chimerical to formulate a scheme which thematerials would not allow us to carry out; it would be like proposing toconstruct an Eiffel tower with building-stones. The fundamental defectof philosophies of history is that they forget this practical necessity. 

I. Let us begin by considering the materials of history. What is theirform and their nature? How do they differ from the materials of othersciences?

Historical facts are derived from the critical analysis of thedocuments. They issue from this process in the form to which analysishas reduced them, chopped small into individual statements; for asingle sentence contains several statements: we have often accepted someand rejected others; each of these statements represents a fact. 

Historical facts have the common characteristic of having been takenfrom documents; but they differ greatly among themselves. 

(1) They represent phenomena of very different nature. From the samedocument we derive facts bearing on handwriting, language, style, doctrines, customs, events. The Mesha inscription furnishes factsbearing on Moabite handwriting and language, the belief in the godChemosh, the practices belonging to his cult, the war between theMoabites and Israel. Thus the facts reach us pell-mell, withoutdistinction of nature. This mixture of heterogeneous facts is one of thecharacteristics which differentiate history from the other sciences. Thesciences of direct observation choose the facts to be studied, andsystematically limit themselves to the observation of facts of a singlespecies. The documentary sciences receive the facts, already observed, at the hands of authors of documents, who supply them in disorder. Forthe purpose of remedying this disorder it is necessary to sort the factsand group them by species. But, for the purpose of sorting them, it isnecessary to know precisely what it is that constitutes a _species_ ofhistorical facts; in order to group them we need a principle ofclassification applicable to them. But on these two questions of capitalimportance historians have not as yet succeeded in formulating preciserules. 

(2) Historical facts present themselves in very different degrees ofgenerality, from the highly general facts which apply to a whole peopleand which lasted for centuries (institutions, customs, beliefs), down tothe most transient actions of a single man (a word, a movement). Hereagain history differs from the sciences of direct observation, whichregularly start from particular facts and labour methodically tocondense them into general facts. In order to form groups the facts mustbe reduced to a common degree of generality, which makes it necessary toinquire to what degree of generality we can and ought to reduce thedifferent species of facts. And this is what historians do not agreeabout among themselves. 

(3) Historical facts are localised; each belongs to a given time and agiven country. If we suppress the time and place to which they belong, they lose their historical character; they now contribute only to theknowledge of universal humanity, as is the case with facts of folk-lorewhose origin is unknown. This necessity of localisation is also foreignto the general sciences; it is confined to the descriptive sciences, which deal with the geographical distribution and with the evolution ofphenomena. It obliges the historian to study separately the factsbelonging to different countries and different epochs. 

(4) The facts which have been extracted from documents by criticalanalysis present themselves accompanied by a critical estimate of theirprobability. [178] In every case where we have not reached completecertainty, whenever the fact is merely probable--still more when it isopen to suspicion--criticism supplies the fact to the historianaccompanied by a label which he has no right to remove, and whichprevents the fact from being definitively admitted into the science. Even those facts which, after comparison with others, end by beingestablished, are subject to temporary exclusion, like the clinical caseswhich accumulate in the medical reviews before they are consideredsufficiently proved to be received as scientific facts. 

Historical construction has thus to be performed with an incoherent massof minute facts, with detail-knowledge reduced as it were to a powder. It must utilise a heterogeneous medley of materials, relating todifferent subjects and places, differing in their degree of generalityand certainty. No method of classifying them is provided by the practiceof historians; history, which began by being a form of literature, hasremained the least methodical of the sciences. 

II. In every science the next step after observing the facts is toformulate a series of questions according to some methodicalsystem;[179] every science is composed of the answers to such a seriesof questions. In all the sciences of direct observation, even if thequestions to be answered have not been put down in advance, the factswhich are observed suggest questions, and require them to be formulatedprecisely. But historians have no discipline of this kind; many of themare accustomed to imitate artists, and do not even think of askingthemselves what they are looking for. They take from their documentsthose parts which strike them, often for purely personal reasons, andreproduce them, changing the language and adding any miscellaneousreflections which come into their minds. 

If history is not to be lost in the confusion of its materials, it mustbe made a rule to proceed here, as in the other sciences, by way ofquestion and answer. [180] But how are the questions to be chosen in ascience so different from the others? This is the fundamental problem ofmethod. The only way to solve it is to begin by determining theessential characteristic of historical facts by which they aredifferentiated from the facts of the other sciences. 

The sciences of direct observation deal with _realities_, taken in theirentirety. The science which borders most closely on history in respectof its subject-matter, descriptive zoology, proceeds by the examinationof a real and complete animal. This animal is first observed, as awhole, by actual vision; it is then dissected into its parts; thisdissection is _analysis_ in the original sense of the word ([Greek:hanalhyein], to break up into parts). It is then possible to put the partstogether again in such a way as to exhibit the structure of the whole;this is _real_ synthesis. It is possible to watch the _real_ movementswhich are the functions of the organs in such a way as to observe themutual actions and reactions of the different parts of the organism. Itis possible to compare _real_ wholes and see what are the parts inwhich they resemble each other, so as to be able to classify themaccording to real points of resemblance. The science is a body ofobjective knowledge founded on _real_ analysis, synthesis, andcomparison; actual sight of the things studied guides the scientificresearcher and dictates the questions he is to ask himself. 

In history there is nothing like this. One is apt to say that history isthe "vision" of past events, and that it proceeds by "analysis": theseare two metaphors, dangerous if we suffer ourselves to be misled bythem. [181] In history we see nothing real except paper with writing onit--and sometimes monuments or the products of art or industry. Thehistorian has nothing before him which he can analyse physically, nothing which he can destroy and reconstruct. "Historical analysis" isno more real than is the vision of historical facts; it is an abstractprocess, a purely intellectual operation. The analysis of a documentconsists in a _mental_ search for the items of information it contains, with the object of criticising them one by one. The analysis of a factconsists in the process of distinguishing _mentally_ between itsdifferent details (the various episodes of an event, the characteristicsof an institution), with the object of paying special attention to eachdetail in turn; that is what is called examining the different "aspects"of a fact, --another metaphor. The human mind is vague by nature, andspontaneously revives only vague collective impressions; to impartclearness to these it is necessary to ask what individual impressions goto form a given collective impression, in order that precision may beattained by a successive consideration of them. This is an indispensableoperation but we must not exaggerate its scope. It is not an objectivemethod which yields a knowledge of real objects; it is only a subjectivemethod which aims at detecting those abstract elements which compose ourimpressions. [182] From the very nature of its materials history isnecessarily a subjective science. It would be illegitimate to extend tothis intellectual analysis of subjective impressions the rules whichgovern the real analysis of real objects. 

History, then, must guard against the temptation to imitate the methodof the biological sciences. Historical facts are so different from thefacts of the other sciences that their study requires a differentmethod. 

III. Documents, the sole source of historical knowledge, giveinformation on three categories of facts:

(1) _Living beings and material objects. _ Documents make us acquaintedwith the existence of human beings, physical conditions, products of artand industry. In all these cases physical facts have been broughtbefore the author by physical perception. But we have before us nothingbut intellectual phenomena, facts seen "through the author'simagination, " or, to speak accurately, mental images representative ofthe author's impressions--images which we form on the _analogy_ of theimages which were in his mind. The Temple at Jerusalem was a materialobject which men saw, but we cannot see it now; all we can now do is toform a mental image of it, analogous to that which existed in the mindsof those who saw and described it. 

(2) _Actions of men. _ Documents relate the actions (and words) of men offormer times. Here, too, are physical facts which were known to theauthors by sight and hearing, but which are now for us no more than theauthor's recollections, subjective images which are reproduced in ourminds. When Cæsar was stabbed the dagger-thrusts were seen, the words ofthe murderers were heard; we have nothing but mental images. Actions andwords all have this characteristic, that each was the action or the wordof an individual; the imagination can only represent to itself_individual_ acts, copied from those which are brought before us bydirect physical observation. As these are the actions of men living in asociety, most of them are performed simultaneously by severalindividuals, or are directed to some common end. These are collectiveacts; but, in the imagination as in direct observation, they alwaysreduce to a sum of individual actions. The "social fact, " as recognisedby certain sociologists, is a philosophical construction, not anhistorical fact. 

(3) Motives and conceptions. Human actions do not contain their owncause within themselves; they have _motives_. This vague word denotesboth the stimulus which occasions the performance of an action, and the_representation_ of the action which is in the mind of a man at themoment when he performs it. We can imagine motives only as existing in aman's mind, and in the form of vague interior representations, analogousto those which we have of our own inward states; we can express themonly by words, generally metaphorical. Here we have _psychic_ facts, generally called feelings and ideas. Documents exhibit three kinds ofsuch facts: (_a_) motives and conceptions in the authors' minds andexpressed by them; (_b_) motives and ideas attributed by the authors tocontemporaries of theirs whose actions they have seen; (_c_) motiveswhich we ourselves may suppose to have influenced the actions related inthe documents, and which we represent to ourselves on the model of ourown motives. 

Physical facts, human actions (both individual and collective), psychicfacts--these form the objects of historical knowledge; they are none ofthem observed directly, they are all _imagined_. Historians--nearly allof them unconsciously and under the impression that they are observingrealities--are occupied solely with images. 

IV. How, then, is it possible to imagine facts without their beingwholly imaginary? The facts, as they exist in the historian's mind, arenecessarily subjective; that is one of the reasons given for refusing torecognise history as a science. But subjective is not a synonym ofunreal. A recollection is only an image; but it is not therefore achimera, it is the representation of a vanished reality. It is true thatthe historian who works with documents has no personal recollections ofwhich he can make direct use; but he forms mental images on the model ofhis own recollections. He assumes that realities (objects, actions, motives), which have now disappeared, but were formerly observed by theauthors of the documents, resembled the realities of his own day whichhe has himself seen and which he retains in his memory. This is thepostulate of all the documentary sciences. If former humanity did notresemble the humanity of to-day, documents would be unintelligible. Starting from this assumed resemblance, the historian forms a mentalrepresentation of the bygone facts of history similar to his ownrecollection of the facts he has witnessed. 

This operation, which is performed unconsciously, is one of theprincipal sources of error in history. The things of the past which areto be pictured in imagination were not wholly similar to the things ofthe present which we have seen; we have never seen a man like Cæsar orClovis, and we have never experienced the same mental states as they. Inthe established sciences it is equally true that one man will work onfacts which another has observed, and which he must therefore representto himself by analogy; but these facts are defined by precise termswhich indicate what invariable elements ought to appear in the image. Even in physiology the notions which occur are sufficiently clear andfixed for the same word to evoke in the minds of all naturalists similarimages of an organ or a movement. The reason is that each notion whichhas a name has been formed by a method of observation and abstraction inthe course of which all the characteristics which belong to the notionhave been precisely determined and described. 

But in proportion as a body of knowledge is more nearly concerned withthe invisible facts of the mind, its notions become more confused andits language less precise. Even the most ordinary facts of human life, social conditions, actions, motives, feelings, can only be expressed byvague terms (_king_, _warrior_, _to fight_, _to elect_). In the case ofmore complex phenomena, language is so indefinite that there is noagreement even as to the essential elements of the phenomena. What arewe to understand by a tribe, an army, an industry, a market, arevolution? Here history shares the vagueness common to all the sciencesof humanity, psychological or social. But its indirect method ofrepresentation by mental images renders this vagueness still moredangerous. The historical images in our minds ought, then, to reproduceat least the essential features of the images which were in the minds ofthe direct observers of past facts; but the terms in which theyexpressed their mental images never tell us exactly what these essentialelements were. 

Facts which we did not see, described in language which does not permitus to represent them in our minds with exactness, form the data ofhistory. The historian, however, is obliged to picture the facts in hisimagination, and he should make it his constant endeavour to constructhis mental images out of none but correct elements, so that he mayimagine the facts as he would have seen them if he had been able toobserve them personally. [183] But the formation of a mental imagerequires more elements than the documents supply. Let any one endeavourto form a mental representation of a battle or a ceremony out of thedata of a narrative, however detailed; he will see how many features heis compelled to add. This necessity becomes physically perceptible inattempts to restore monuments in accordance with descriptions (forexample, the Temple at Jerusalem), in pictures which claim to berepresentations of historical scenes, in the drawings of illustratednewspapers. 

Every historical image contains a large part of fancy. The historiancannot get rid of it, but he can take stock of the real elements whichenter into his images and confine his constructions to these; they arethe elements which he has derived from the documents. If, in order tounderstand the battle between Cæsar and Ariovistus, he finds itnecessary to make a mental picture of the two opposing armies, he willbe careful to draw no conclusions from the general aspect under which heimagines them; he will base his reasonings exclusively on the realdetails furnished by the documents. 

V. The problem of historical method may be finally stated as follows. Out of the different elements we find in documents we form mentalimages. Some of these, relating entirely to physical objects, arefurnished to us by illustrative monuments, and they directly representsome of the physical aspects of the things of the past. Most of them, however, including all the images we form of psychic facts, areconstructed on the model either of ancient representations, or, morefrequently, of the facts we have observed in our own experience. Now, the things of the past were only partially similar to the things of thepresent, and it is precisely the points of difference which make historyinteresting. How are we to represent to ourselves these elements ofdifference for which we have no model? We have never seen a company ofmen resembling the Frankish warriors, and we have never personallyexperienced the feelings which Clovis had when setting out to fightagainst the Visigoths. How are we to make our imagination of facts ofthis kind harmonise with the reality?

Practically, what happens is as follows. Immediately on the reading of asentence in a document an image is formed in our minds by a spontaneousoperation beyond our control. This image is based on a superficialanalogy, and is, as a rule, grossly inaccurate. Any one who searches hismemory may recall the absurd manner in which he first represented tohimself the persons and scenes of the past. It is the task of history torectify these images gradually, by eliminating the false elements one byone, and replacing them by true ones. We have seen red-haired people, bucklers, and Frankish battle-axes (or at least drawings of theseobjects); we bring these elements together, in order to correct ourfirst mental image of the Frankish warriors. The historical image thusends by becoming a combination of features borrowed from differentexperiences. 

It is not enough to represent to oneself isolated persons, objects, andactions. Men and their actions form part of a whole, of a society and ofa process of evolution. It is, therefore, further necessary to representto oneself the relations between different men and different actions(nations, governments, laws, wars). 

But in order to imagine relations it is necessary to have a conceptionof collectivities or wholes, and the documents only give isolatedelements. Here again the historian is obliged to use a subjectivemethod. He imagines a society or a process of evolution, and in thisimaginary framework he disposes the elements furnished by the documents. Thus, whereas biological classification is guided by the objectiveobservation of physical units, historical classification can only beeffected upon subjective units existing in the imagination. 

The realities of the past are things which we do not observe, and whichwe can only know in virtue of their resemblance to the realities of thepresent. In order to realise the conditions under which past eventshappened, we must observe the humanity of to-day, and look for theconditions under which analogous events happen now. History thus becomesan application of the descriptive sciences which deal with humanity, descriptive psychology, sociology or social science; but all thesesciences are still but imperfectly established, and their defects retardthe establishment of a science of history. 

Some of the conditions of human life are, however, so necessary and soobvious that the most superficial observation is enough to establishthem. These are the conditions common to all humanity; they have theirorigin either in the physiological organisation which determines thematerial needs of men, or in the psychological organisation whichdetermines their habits in matters of conduct. These conditions cantherefore be provided for by the use of a set of general questionsapplicable to all the cases that may occur. It is with historicalconstruction as with historical criticism--the impossibility of directobservation compels the use of prearranged sets of questions. 

The human actions which form the subject-matter of history differ fromage to age and from country to country, just as men and societies havediffered from each other; and, indeed, it is the special aim of historyto study these differences. If men had always had the same form ofgovernment or spoken the same language, there would be no occasion towrite the history of forms of government or the history of languages. But these differences are comprised within limits imposed by the generalconditions of human life; they are but varieties of certain modes ofbeing and doing which are common to the whole of humanity, or at leastto the great majority of men. We cannot know _a priori_ what was themode of government or the language of an historical people; it is thebusiness of history to tell us. But that a given people had a languageand had a form of government is something which we are entitled toassume, before examination, in every possible case. 

By drawing up the list of the fundamental phenomena which we may expectto find in the life of every individual and every people, we shall havesuggested to us a set of general questions which will be summary, butstill sufficient to enable us to arrange the bulk of historical facts ina certain number of natural groups, each of which will form a specialbranch of history. This scheme of general classification will supply thescaffolding of historical construction. 

The set of general questions will only apply to phenomena of constantoccurrence: it cannot anticipate the thousands of local or accidentalevents which enter into the life of an individual or a nation; it will, therefore, not contain all the questions which the historian must answerbefore he can give a complete picture of the past. The detailed study ofthe facts will require the use of lists of questions entering more intodetail, and differing according to the nature of the events, the men, orthe societies studied. In order to frame these lists, we begin bysetting down those questions or matters of detail which are suggested bythe mere reading of the documents; but for the purpose of arrangingthese questions, often indeed for the purpose of making the listcomplete, recourse must be had to the systematic _a priori_ method. Among the classes of facts, the persons, and the societies with which weare well acquainted (either from direct observation or from history), welook for those which resemble the facts, the persons, or the societieswhich we wish to study. By analysing the scheme of arrangement used inthe scientific treatment of these familiar cases we shall learn whatquestions ought to be asked in reference to the analogous cases which wepropose to investigate. Of course the model must be chosenintelligently; we must not apply to a barbarous society a list ofquestions framed on the study of a civilised nation, and ask with regardto a feudal domain what agents corresponded to each of our ministers ofstate--as Boutaric did in his study of the administration of Alphonse ofPoitiers. 

This method of drawing up lists of questions which bases all historicalconstruction on an _a priori_ procedure, would be objectionable ifhistory really were a science of observation; and perhaps some willthink it compares very unfavourably with the _a posteriori_ methods ofthe natural sciences. But its justification is simple: it is the onlymethod which it is possible to employ, and the only method which, as amatter of fact, ever has been employed. The moment an historian attemptsto put in order the facts contained in documents, he constructs out ofthe knowledge he has (or thinks he has) of human affairs a scheme ofarrangement which is the equivalent of a list of questions--unless, perhaps, he adopts a scheme which one of his predecessors hasconstructed in a similar manner. But when this work has been performedunconsciously, the scheme of arrangement remains incomplete andconfused. Thus it is not a case of deciding whether to work with orwithout an _a priori_ set of questions--we must work with such a set inany case--the choice merely lies between the unconscious use of anincomplete and confused set of questions and the conscious use of aprecise and complete set. 

VI. We can now sketch the plan of historical construction in a way whichwill determine the series of synthetic operations necessary to raise theedifice. 

The critical analysis of the documents has supplied thematerials--historical facts still in a state of dispersion. We begin by_imagining_ these facts on the model of what we suppose to be theanalogous facts of the present; by combining elements taken from realityat different points, we endeavour to form a mental image which shallresemble as nearly as possible that which would have been produced bydirect observation of the past event. This is the first operation, inseparable in practice from the reading of the documents. Consideringthat it will be enough to have indicated its nature here, [184] we haverefrained from devoting a special chapter to it. 

The facts having been thus imagined, we _group_ them according toschemes of classification devised on the model of a body of facts whichwe have observed directly, and which we suppose analogous to the body ofpast facts under consideration. This is the second operation; it isperformed by the aid of systematic questions, and its result is todivide the mass of historical facts into homogeneous portions which weafterwards form into groups until the entire history of the past hasbeen systematically arranged according to a general scheme. 

When we have arranged in this scheme the facts taken from the documents, there remain gaps whose extent is always considerable, and is enormousfor those parts of history in regard to which documents are scanty. Weendeavour to fill some of these gaps by _reasoning_ based on the factswhich are known. This is (or should be) the third operation; itincreases the sum of historical knowledge by an application of logic. 

We still possess nothing but a mass of facts placed side by side in ascheme of classification. We have to condense them into _formulæ_, inorder to deduce their general characteristics and their relation to eachother. This is the fourth operation; it leads to the final conclusionsof history, and crowns the work of historical construction from thescientific point of view. 

But as historical knowledge, which is by nature complex and unwieldy, isexceptionally difficult to communicate, we still have to look for themethods of expounding historical results in appropriate form. 

VII. This series of operations, easy to conceive in the mind, has neverbeen more than imperfectly performed. It is beset by materialdifficulties which theories of methodology do not take into account, butwhich it would be better to face, with the purpose of discoveringwhether they are after all insurmountable. 

The operations of history are so numerous, from the first discovery ofthe document to the final formula of the conclusion, they require suchminute precautions, so great a variety of natural gifts and acquiredhabits, that there is no man who can perform _by himself_ all the workon any one point. History is less able than any other science todispense with the division of labour; but there is no other science inwhich labour is so imperfectly divided. We find specialists in criticalscholarship writing general histories in which they let theirimagination guide them in the work of construction;[185] and, on theother hand, there are constructive historians who use for their workmaterials whose value they have not tested. [186] The reason is that thedivision of labour implies a common understanding among the workers, andin history no such understanding exists. Except in the preparatoryoperations of external criticism, each worker follows the guidance ofhis own private inspiration; he is at no pains to work on the same linesas the others, nor does he pay any regard to the whole of which his ownwork is to form a part. Thus no historian can feel perfectly safe inadopting the results of another's work, as may be done in theestablished sciences, for he does not know whether these results havebeen obtained by trustworthy methods. The most scrupulous go so far asto admit nothing until they have done the work on the documents overagain for themselves. This was the attitude adopted by Fustel deCoulanges. It is barely possible to satisfy this exacting standard inthe case of little-known periods, the documents relating to which areconfined to a few volumes; and yet some have gone so far as to maintainthe dogma that no historian should ever work at second hand. [187] This, indeed, is what an historian is compelled to do when the documents aretoo numerous for him to be able to read them all; but he does not sayso, to avoid scandal. 

It would be better to acknowledge the truth frankly. So complex ascience as history, where facts must ordinarily be accumulated by themillion before it is possible to formulate conclusions, cannot be builtup on this principle of continually beginning afresh. Historicalconstruction is not work that can be done with documents, any more thanhistory can be "written from manuscripts, " and for the same reason--theshortness of time. In order that science may advance it is necessary tocombine the results of thousands of detail-researches. 

But how are we to proceed in view of the fact that most researches havebeen conducted upon methods which, if not defective, are at least opento suspicion? Universal confidence would lead to error as surely asuniversal distrust would make progress impossible. One useful rule, atany rate, may be stated, as follows: The works of historians should beread with the same critical precautions which are observed in thereading of documents. A natural instinct impels us to look principallyfor the conclusions, and to accept them as so much established truth; weought, on the contrary, to be continually applying analysis, we ought tolook for the facts, the _proofs_, the fragments of documents--in short, the materials. We shall be doing the author's work over again, but weshall do it very much faster than he did, for that which takes up timeis the collection and combination of the materials; and we shall acceptno conclusions but those we consider to have been proved. 

CHAPTER II

THE GROUPING OF FACTS

I. The prime necessity for the historian, when confronted with the chaosof historical facts, is to limit the field of his researches. In theocean of universal history what facts is he to choose for collection?Secondly, in the mass of facts so chosen he will have to distinguishbetween different groups and make subdivisions. Lastly, within each ofthese subdivisions he will have to arrange the facts one by one. Thusall historical construction should begin with the search for a principleto guide in the selection, the grouping, and the arrangement of facts. This principle may be sought either in the external conditions of thefacts or in their intrinsic nature. 

The simplest and easiest mode of classification is that which is foundedon external conditions. Every historical fact belongs to a definite timeand a definite place, and relates to a definite man or group of men: aconvenient basis is thus afforded for the division and arrangement offacts. We have the history of a period, of a country, of a nation, of aman (biography); the ancient historians and those of the Renaissanceused no other type. Within this general scheme the subdivisions areformed on the same principle, and facts are arranged in chronologicaland geographical order, or according to the groups to which theyrelate. As to the selection of facts to be arranged in this scheme, fora long time it was made on no fixed principle; historians followed theirindividual fancy, and chose from among the facts relating to a givenperiod, country, or nation all that they deemed interesting or curious. Livy and Tacitus mingle accounts of floods, epidemics, and the birth ofmonsters with their narratives of wars and revolutions. 

Classification of facts by their intrinsic nature was introduced verylate, and has made way but slowly and imperfectly. It took its riseoutside the domain of history, in certain branches of study dealing withspecial human phenomena--language, literature, art, law, politicaleconomy, religion; studies which began by being dogmatic, but graduallyassumed an historical character. The principle of this mode ofclassification is to select and group together those facts which relateto the same species of actions; each of these groups becomes thesubject-matter of a special branch of history. The totality of factsthus comes to be arranged in compartments which may be constructed _apriori_ by the study of the totality of human activities; thesecorrespond to the set of general questions of which we have spoken inthe preceding chapter. 

In the following table we have attempted to provide a general scheme forthe classification[188] of historical facts, founded on the nature ofthe _conditions_ and of the _manifestations_ of activity. 

I. MATERIAL CONDITIONS. (1) _Study of the body_: _A. _ Anthropology(ethnology), anatomy, and physiology, anomalies and pathologicalpeculiarities. _B. _ Demography (number, sex, age, births, deaths, diseases). (2) _Study of the environment_: _A. _ Natural geographicalenvironment (orographic configuration, climate, water, soil, flora, andfauna). _B. _ Artificial environment, forestry (cultivation, buildings, roads, implements, &c. ). 

II. INTELLECTUAL HABITS (not obligatory). (1) _Language_ (vocabulary, syntax, phonetics, semasiology). Handwriting. (2) _Arts_: _A. _ Plasticarts (conditions of production, conceptions, methods, works). _B. _ Artsof expression, music, dance, literature. (3) _Sciences_ (conditions ofproduction, methods, results). (4) _Philosophy and Morals_ (conceptions, precepts, actual practice). (5) _Religion_ (beliefs, practices). [189]

III. MATERIAL CUSTOMS (not obligatory). (1) _Material life_: _A. _ Food(materials, modes of preparing, stimulants). _B. _ Clothes and personaladornment. _C. _ Dwellings and furniture. (2) _Private life_: _A. _Employment of time (toilette, care of the person, meals). _B. _ Socialceremonies (funerals and marriages, festivals, etiquette). _C. _Amusements (modes of exercise and hunting, games and spectacles, socialmeetings, travelling). 

IV. ECONOMIC CUSTOMS. (1) _Production_: _A. _ Agriculture andstock-breeding. _B. _ Exploitation of minerals. (2) _Transformation, Transport and industries_:[190] technical processes, division of labour, means of communication. (3) _Commerce_: exchange and sale, credit. (4)_Distribution_: system of property, transmission, contracts, profit-sharing. 

V. SOCIAL INSTITUTIONS. (1) _The family_: _A. _ Constitution, authority, condition of women and children. _B. _ Economic organisation. [191] Familyproperty, succession. (2) _Education and instruction_ (aim, methods, _personnel_). (3) _Social classes_ (principle of division, rulesregulating intercourse). 

VI. PUBLIC INSTITUTIONS (obligatory). (1) _Political institutions_: _A. _Sovereign (_personnel_, procedure). _B. _ Administration, services (war, justice, finance, &c. ). _C. _ Elected authorities, assemblies, electoralbodies (powers, procedure). (2) _Ecclesiastical institutions_ (the samedivisions). (3) _International institutions_: _A. _ Diplomacy. _B. _ War(usages of war and military arts). _C. _ Private law and commerce. 

This grouping of facts according to their nature is combined with thesystem of grouping by time and place; we thus obtain chronological, geographical, or, national sections in each branch. The history of aspecies of activity (language, painting, government) subdivides into thehistory of periods, countries, and nations (history of the ancient Greeklanguage, history of the government of France in the nineteenthcentury). 

The same principles aid in determining the order in which the facts areto be arranged. The necessity of presenting facts one after anotherobliges us to adopt some methodical rule of succession. We may describesuccessively either all the facts which relate to a given place, orthose which relate to a given country, or all the facts of a givenspecies. All historical matter can be distributed in three differentkinds of order: _chronological_ order, _geographical_ order, that kindof order which is governed by the nature of actions and is generallycalled _logical_ order. It is impossible to use any of these ordersexclusively: in every chronological exposition there necessarily occurgeographical or logical cross-divisions, transitions from one country toanother, or from one species of facts to a different species, andconversely. But it is always necessary to decide which shall be the mainorder into which the others enter as subdivisions. 

It is a delicate matter to choose between these three orders; our choicewill be decided by different reasons according to the subject, andaccording to the public for whom we are working. That is to say, it willdepend on the method of exposition; it would take up too much space togive the theory of it. 

II. When we come to the selection of historical facts for classificationand arrangement, a question is raised which has been disputed withconsiderable warmth. 

Every human action is by its nature an individual transient phenomenonwhich is confined to a definite time and a definite place. Strictlyspeaking, every fact is unique. But every action of a man resemblesother actions of the same man, or of other members of the same group, and often to so great a degree that the whole group of actions receivesa common name, in which their individuality is lost. These groups ofsimilar actions, which the human mind is irresistibly impelled to form, are called habits, usages, institutions. These are merely constructionsof the mind, but they are imposed so forcibly on our intellect that manyof them must be recognised and constantly employed; habits arecollective facts, possessing extension in time and space. Historicalfacts may therefore be considered under two different aspects: we mayregard either the individual, particular, and transient elements inthem, or we may look for what is collective, general, and durable. According to the first conception, history is a continuous narrative ofthe incidents which have happened among men in the past; according tothe second, it is the picture of the successive habits of humanity. 

On this subject there has been a contest, especially in Germany, betweenthe partisans of the history of civilisation (_Kulturgeschichte_)[192]and the historians who remain faithful to ancient tradition; in Francewe have had the struggle between the history of institutions, manners, and ideas, and political history, contemptuously nicknamed"battle-history" by its opponents. 

This opposition is explained by the difference between the documentswhich the workers on either side were accustomed to deal with. Thehistorians, principally occupied with political history, read ofindividual and transient acts of rulers in which it was difficult todetect any common feature. In the special histories, on the contrary(except that of literature), the documents exhibit none but generalfacts, a linguistic form, a religious rite, a rule of law; an effort ofimagination is required to picture the man who pronounced the word, whoperformed the rite, or who applied the rule in practice. 

There is no need to take sides in this controversy. Historicalconstruction in its completeness implies the study of facts under bothaspects. The representation of men's habits of thought, life, and actionis obviously an important part of history. And yet, supposing we hadbrought together all the acts of all individuals for the purpose ofextracting what is common to them, there would still remain a residuewhich we should have no right to reject, for it is the distinctivelyhistorical element--the circumstance that a particular action was theaction of a given man, or group of men, at a given moment. In a schemeof classification which should only recognise the general facts ofpolitical life there would be no place for the victory of Pharsalia orthe taking of the Bastille--accidental and transient facts, but withoutwhich the history of Roman and French institutions would beunintelligible. 

History is thus obliged to combine with the study of general facts thestudy of certain particular facts. It has a mixed character, fluctuatingbetween a science of generalities and a narrative of adventures. Thedifficulty of classing this hybrid under one of the categories of humanthought has often been expressed by the childish question: Is history ascience or an art?

III. The general table given above may be used for the determination ofall the species of habits (usages or institutions) of which the historymay be written. But before applying this general scheme to the study ofany particular group of habits, language, religion, private usages, orpolitical institutions, there is always a preliminary question to beanswered: Whose were the habits we are about to study? They were commonto a great number of individuals; and a collection of individuals withthe same habits is what we call a _group_. The first condition, then, for the study of a habit is the determination of the group which haspractised it. At this point we must beware of the first impulse; itleads to a negligence which may ruin the whole of our historicalconstruction. 

The natural tendency is to conceive the human group on the model of thezoological species--as a body of men who all resemble each other. Wetake a group united by a very obvious common characteristic, a nationunited by a common official government (Romans, English, French), apeople speaking the same language (Greeks, ancient Germans), and weproceed as if all the members of this group resembled each other atevery point and had the same usages. 

As a matter of fact, no real group, not even a centralised society, is ahomogeneous whole. For a great part of human activity--language, art, science, religion, economic interests--the group is constantlyfluctuating. What are we to understand by the group of those who speakGreek, the Christian group, the group of modern science? And even thosegroups to which some precision is given by an official organisation, States and Churches, are but superficial unities composed ofheterogeneous elements. The English nation comprises Welsh, Scotch, andIrish; the Catholic Church is composed of adherents scattered over thewhole world, and differing in everything but religion. There is nogroup whose members have the same habits in every respect. The same manis at the same time a member of several groups, and in each group he hascompanions who differ from those he has in the others. A French Canadianbelongs to the British Empire, the Catholic Church, the group ofFrench-speaking people. Thus the different groups overlap each other ina way that makes it impossible to divide humanity into sharply distinctsocieties existing side by side. 

In historical documents we find the contemporary names of groups, manyof them resting on mere superficial resemblances. It must be made a rulenot to adopt popular notions of this kind without criticising them. Wemust accurately determine the nature and extent of the group, asking: Ofwhat men was it composed? What bond united them? What habits had they incommon? In what species of activity did they differ? Not till after suchcriticism shall we be able to tell what are the habits in respect ofwhich the group in question may be used as a basis of study. In order tostudy intellectual habits (language, religion, art, science) we shallnot take a political unit, the nation, but the group consisting of thosewho shared the habit in question. In order to study economic facts weshall choose a group united by a common economic interest; we shallreserve the political group for the study of social and political facts, and we shall discard _race_[193] altogether. 

Even in those points in which a group is homogeneous it is not entirelyso; it is divided into sub-groups, the members of which differ insecondary habits; a language is divided into dialects, a religion intosects, a nation into provinces. Conversely, one group resembles othergroups in a way that justifies its being regarded as contiguous withthem; in a general classification we may recognise "families" oflanguages, arts, and peoples. We have, then, to ask: How was a givengroup sub-divided? Of what larger group did it form a part?

It then becomes possible to study methodically a given habit, or eventhe totality of the habits belonging to a given time and place, byfollowing the table given above. The operation presents no difficultiesof method in the case of those species of facts which appear asindividual and voluntary habits--language, art, sciences, conceptions, private usages; here it is enough to ascertain in what each habitconsisted. It is merely necessary to distinguish carefully between thosewho originated or maintained habits (artists, the learned, philosophers, introducers of fashions) and the mass who accepted them. 

But when we come to social or political habits (what we callinstitutions), we meet with new conditions which produce an inevitableillusion. The members of the same social or political group do notmerely habitually perform _similar_ actions; they influence each otherby _reciprocal_ actions, they command, coerce, pay each other. Habitshere take the form of _relations_ between the different members; whenthey are of old standing, formulated in official rules, imposed by avisible authority, maintained by a special set of persons, they occupyso important a place in life, that, to the persons under theirinfluence, they appear as external realities. The men, too, whospecialise in an occupation or a function which becomes the dominatinghabit of their lives, appear as grouped in distinct categories (classes, corporations, churches, governments); and these categories are taken forreal existences, or at least for organs of various functions in a realexistence, namely, society. We follow the analogy of an animal's body sofar as to describe the "structure" and the "functions" of a society, even its "anatomy" and "physiology. " These are pure metaphors. By thestructure of a society we mean the rules and the customs by whichoccupations and enjoyments are distributed among its members; by itsfunctions we mean the habitual actions by which each man enters intorelations with the others. It may be convenient to use these terms, butit should be remembered that the underlying reality is composed entirelyof habits and customs. 

The study of institutions, however, obliges us to ask special questionsabout persons and their functions. In respect of social and economicinstitutions we have to ask what was the principle of the division oflabour and of the division into classes, what were the professions andclasses, how were they recruited, what were the relations between themembers of the different professions and classes. In respect ofpolitical institutions, which are sanctioned by obligatory rules and avisible authority, two new series of questions arise. (1) Who were thepersons invested with authority? When authority is divided we have tostudy the division of functions, to analyse the _personnel_ ofgovernment into its different groups (supreme and subordinate, centraland local), and to distinguish each of the special bodies. In respect ofeach class of men concerned in the government we shall ask: How werethey recruited? What was their official authority? What were their realpowers? (2) What were the official rules? What was their form (custom, orders, law, precedent)? What was their content (rules of law)? What wasthe mode of application (procedure)? And, above all, how did the rulesdiffer from the practice (abuse of power, exploitation, conflictsbetween executive agents, non-observance of rules)?

After the determination of all the facts which constitute a society, itremains to find the place which this society occupies among the totalnumber of the societies contemporary with it. Here we enter upon thestudy of international institutions, intellectual, economic, andpolitical (diplomacy and the usages of war); the same questions apply asin the study of political institutions. A study should also be made ofthe habits common to several societies, and of those relations which donot assume an official form. This is one of the least advanced parts ofhistorical construction. 

IV. The outcome of all this labour is a tabulated view of human life ata given moment; it gives us the knowledge of a _state_ of society (inGerman, _Zustand_). But history is not limited to the study ofsimultaneous facts, taken in a state of rest, to what we may call the_statics_ of society. It also studies the states of society at differentmoments, and discovers the differences between these states. The habitsof men and the material conditions under which they live change fromepoch to epoch; even when they appear to be constant they do not remainunaltered in every respect. There is therefore occasion to investigatethese changes; thus arises the study of successive facts. 

Of these changes the most interesting for the work of historicalconstruction are those which tend in a common direction, [194] so that invirtue of a series of gradual differentiations a usage or a state ofsociety is transformed into a different usage or state, or, to speakwithout metaphor, cases where the men of a given period practise a habitvery different from that of their predecessors without any abrupt changehaving taken place. This is _evolution_. 

Evolution occurs in all human habits. In order to investigate it, therefore, it is enough to turn once more to the series of questionswhich we used in constructing a tabulated view of society. In respect ofeach of the facts, conditions, usages, persons invested with authority, official rules, the question is to be asked: What was the evolution ofthis fact?

This study will involve several operations: (1) the determination of thefact whose evolution is to be studied; (2) the fixing of the duration ofthe time during which the evolution took place (the period should be sochosen that while the transformation is obvious, there yet remains aconnecting link between the initial and the final condition); (3) theestablishing of the different stages of the evolution; (4) theinvestigation of the means by which it was brought about. 

V. A series, even a complete series, of all the states of all societiesand of all their evolutions would not be enough to exhaust thesubject-matter of history. There remains a number of unique facts whichwe cannot pass over, because they explain the origin of certain statesof society, and form the starting-points of evolutions. How could westudy the institutions or the evolution of France if we ignored theconquest of Gaul by Cæsar and the invasion of the Barbarians?

This necessity of studying unique facts has caused it to be said thathistory cannot be a science, for every science has for its object thatwhich is general. History is here in the same situation as cosmography, geology, the science of animal species: it is not the abstract knowledgeof the general relations between facts, it is a study which aims at_explaining_ reality. Now, reality exists but once. There has been but asingle evolution of the world, of animal life, of humanity. In each ofthese evolutions the successive facts have not been the product ofabstract laws, but of the concurrence, at each moment, of severalcircumstances of different nature. This concurrence, sometimes calledchance, has produced a series of accidents which have determined theparticular course taken by evolution. [195] Evolution can only beunderstood by the study of these accidents; history is here on the samefooting as geology or palæontology. 

Thus scientific history may go back to the accidents, or events, whichtraditional history collected for literary reasons, because they struckthe imagination, and employ them for the study of evolution. We may thuslook for the facts which have influenced the evolution of each one ofthe habits of humanity. Each event will be arranged under its date inthe evolution which it is supposed to have influenced. It will thensuffice to bring together the events of every kind, and to arrange themin chronological and geographical order, to have a representation ofhistorical evolution as a whole. 

Then, over and above the _special_ histories in which the facts arearranged under purely abstract categories (art, religion, private life, political institutions), we shall have constructed a concrete _general_history, which will connect together the various special histories byexhibiting the main stream of evolution which has dominated all thespecial evolutions. None of the species of facts which we study apart(religion, art, law, constitutions) forms a closed world within whichevolution takes place in obedience to a kind of internal impulse, asspecialists are prone to imagine. The evolution of a usage or of aninstitution (language, religion, church, state) is only a metaphor; ausage is an abstraction, abstractions do not evolve; it is only_existences_ that evolve, in the strict sense of the word. [196] When achange takes place in a usage, this means that the men who practise ithave changed. Now, men are not built in water-tight compartments(religious, juridical, economic) within which phenomena can occur inisolation; an event which modifies the condition of a man changes hishabits in a great variety of respects. The invasion of the Barbariansinfluenced alike language, private life, and political institutions. Wecannot, therefore, understand evolution by confining ourselves to aspecial branch of history; the specialist, even for the purpose ofwriting the complete history of his own branch, must look beyond theconfines of his own subject into the field of general events. It is themerit of Taine to have asserted, with reference to English literature, that literary evolution depends, not on literary events, but on facts ofa general character. 

The general history of individual facts was developed before the specialhistories. It contains the residue of facts which have not found a placein the special histories, and has been reduced in extent by theformation and detachment of special branches. As general facts areprincipally of a political nature, and as it is more difficult toorganise these into a special branch, general history has in practicebeen confounded with political history (_Staatengeschichte_). [197] Thuspolitical historians have been led to make themselves the champions ofgeneral history, and to retain in their constructions all the generalfacts (migrations of peoples, religious reforms, inventions, anddiscoveries) necessary for the understanding of political evolution. 

In order to construct general history it is necessary to look for allthe facts which, because they have produced changes, can explain eitherthe state of a society or one of its evolutions. We must search for themamong all classes of facts, displacements of population, artistic, scientific, religious, technical innovations, changes in the _personnel_of government, revolutions, wars, discoveries of countries. 

That which is important is that the fact should have had a decisiveinfluence. We must therefore resist the natural temptation to dividefacts into great and small. It goes against the grain to admit thatgreat effects may have had small causes, that Cleopatra's nose may havemade a difference to the Roman Empire. This repugnance is of ametaphysical order; it springs from a preconceived opinion on thegovernment of the world. In all the sciences which deal with anevolution we find individual facts which serve as starting-points forseries of vast transformations. A drove of horses brought by theSpanish has stocked the whole of South America. In a flood a branch ofa tree may dam a current and transform the aspect of a valley. 

In human evolution we meet with great transformations which have nointelligible cause beyond an individual accident. [198] In the sixteenthcentury England changed its religion three times on the death of asovereign (Henry VIII. , Edward VI. , Mary). Importance not to be measuredby the initial fact, but by the facts which resulted from it. We mustnot, therefore, deny _a priori_ the action of individuals and discardindividual facts. We must examine whether a given individual was in aposition to make his influence strongly felt. There are two cases inwhich we may assume that he was: (1) when his action served as anexample to a mass of men and created a tradition, a case frequent inart, science, religion, and technical matters; (2) when he had power toissue commands and direct the actions of a mass of men, as is the casewith the heads of a state, an army, or a church. The episodes in a man'slife may thus become important facts. 

Accordingly, in the scheme of historical classification a place shouldbe assigned for persons and events. 

VI. In every study of successive facts it is necessary to provide anumber of halting-places, to distinguish beginnings and ends, in orderthat chronological divisions may be made in the enormous mass of facts. These divisions are _periods_; the use of them is as old as history. Weneed them, not only in general history, but in the special branches ofhistory as well, whenever we study an extent of time long enough for anevolution to be sensible. It is by means of events that we fix theirlimits. 

In the special branches of history, after having decided what changes ofhabits are to be considered as reaching deepest, we adopt them asmarking _dates_ in the evolution; we then inquire what event producedthem. The event which led to the formation or the change of a habitbecomes the beginning or the end of a period. Sometimes these boundaryevents are of the same species as the facts whose evolution we arestudying--literary facts in the history of literature, political factsin political history. But more often they belong to a different species, and the special historian is obliged to borrow them from generalhistory. 

In general history the periods should be divided according to theevolution of several species of phenomena; we look for events which markan epoch simultaneously in several branches (the Invasion of theBarbarians, the Reformation, the French Revolution). We may thusconstruct periods which are common to several branches of evolution, whose beginning and whose end are each marked by a single event. It isthus that the traditional division of universal history into periods hasbeen effected. The sub-periods are obtained by the same process, bytaking for limits events which have produced consequences of secondaryimportance. 

The periods which are thus constructed according to the events are ofunequal duration. We must not be troubled by this want of symmetry; aperiod ought not to be a fixed number of years, but the time occupied bya distinct phase of evolution. Now, evolution is not a regular movement;sometimes a long series of years passes without notable change, thencome moments of rapid transformation. On this difference Saint-Simon hasfounded a distinction between _organic_ periods (of slow change) and_critical_ periods (of rapid change). 

CHAPTER III

CONSTRUCTIVE REASONING

I. The historical facts supplied by documents are never enough to fillall the blanks in such schemes of classification and arrangement as wehave been considering. There are many questions to which no directanswer is given by the documents; many features are lacking withoutwhich the complete picture of the various states of society, ofevolutions and events, cannot be given. We are irresistibly impelled toendeavour to fill up these gaps. 

In the sciences of direct observation, when a fact is missing from aseries, it is sought for by a new observation. In history, where we havenot this resource, we seek to extend our knowledge by the help ofreasoning. Starting from facts known to us from the documents, weendeavour to reach new facts by inference. If the reasoning be correct, this method of acquiring knowledge is legitimate. 

But experience shows that of all the methods of acquiring historicalknowledge, reasoning is the most difficult to employ correctly, and theone which has introduced the most serious errors. It should not be usedwithout the safeguard of a number of precautions calculated to keep thedanger continually before the mind. 

(1) Reasoning should never be combined with the analysis of a document. The reader who allows himself to introduce into a text what the authorhas not expressly put there ends by making him say what he neverintended to say. [199]

(2) Facts obtained by the direct examination of documents should neverbe confused with the results obtained by reasoning. When we state a factknown to us by reasoning only, we must not allow it to be supposed thatwe have found it in the documents; we must disclose the method by whichwe have obtained it. 

(3) Unconscious reasoning must never be allowed; there are too manychances of error. It will be enough to make a point of putting everyargument into logical form; in the case of bad reasoning the majorpremiss is generally monstrous to an appalling degree. 

(4) If the reasoning leaves the least doubt, no attempt must be made todraw a conclusion; the point treated must be left in the conjecturalstage, clearly distinguished from the definitively established results. 

(5) It is not permissible to return to a conjecture and endeavour totransform it into a certainty. Here the first impression is most likelyto be right. By reflection upon a conjecture we familiarise ourselveswith it, and end by thinking it better established; while the truth is, we are merely more accustomed to it. This is a frequent mishap withthose who devote themselves to long meditation on a small number oftexts. 

There are two ways of employing reasoning, one negative, the otherpositive; we shall examine them separately. 

II. The negative mode of reasoning, called also the "argument fromsilence, " is based on the absence of indications with regard to afact. [200] From the circumstance of the fact not being mentioned in anydocument it is inferred that there was no such fact; the argument isapplied to all kinds of subjects, usages of every description, evolutions, events. It rests on a feeling which in ordinary life isexpressed by saying: "If it were true, we should have heard of it;" itimplies a general proposition which may be formulated thus: "If analleged event really had occurred, there would be some document inexistence in which it would be referred to. "

In order that such reasoning should be justified it would be necessarythat every fact should have been observed and recorded in writing, andthat all the records should have been preserved. Now, the greater partof the documents which have been written have been lost, and the greaterpart of the events which happen are not recorded in writing. In themajority of cases the argument would be invalid. It must therefore berestricted to the cases where the conditions implied in it have beenfulfilled. 

(1) It is necessary not only that there should be now no documents inexistence which mention the fact in question, but that there shouldnever have been any. If the documents are lost we can conclude nothing. The argument from silence ought, therefore, to be employed the morerarely the greater the number of documents that have been lost; it is ofmuch less use in ancient history than in dealing with the nineteenthcentury. Some, desiring to free themselves from this restriction, aretempted to assume that the lost documents contained nothing interesting;if they were lost, say they, the reason was that they were not worthpreserving. But the truth is, every manuscript is at the mercy of theleast accident; its preservation or destruction is a matter of purechance. 

(2) The fact must have been of such a kind that it could not fail to beobserved and recorded. Because a fact has not been recorded it does notfollow that it has not been observed. Any one who is concerned in anorganisation for the collection of a particular species of facts knowshow much commoner those facts are than people think, and how many casespass unnoticed or without leaving any written trace. It is so withearthquakes, cases of hydrophobia, whales stranded on the shore. Besides, many facts, even those which are well known to those who arecontemporary with them, are not recorded, because the officialauthorities prevent their publication; this is what happens to thesecret acts of governments and the complaints of the lower classes. Thissilence, which proves nothing, greatly impresses unreflectinghistorians; it is the origin of the widespread sophism of the "good oldtimes. " No document relates any abuse of power by officials or anycomplaints made by peasants; therefore, everything was regular andnobody was suffering. Before we argue from silence we should ask: Mightnot this fact have failed to be recorded in any of the documents wepossess? That which is conclusive is not the absence of any document ona given fact, but silence as to the fact in a document in which it wouldnaturally be mentioned. 

The negative argument is thus limited to a few clearly defined cases. (1) The author of the document in which the fact is not mentioned hadthe intention of systematically recording all the facts of the sameclass, and must have been acquainted with all of them. (Tacitus soughtto enumerate the peoples of Germany; the _Notitia dignitatum_ mentionedall the provinces of the Empire; the absence from these lists of apeople or a province proves that it did not then exist. ) (2) The fact, if it was such, must have affected the author's imagination so forciblyas necessarily to enter into his conceptions. (If there had been regularassemblies of the Frankish people, Gregory of Tours could not haveconceived and described the life of the Frankish kings withoutmentioning them. )

III. The positive mode of reasoning begins with a fact established bythe documents, and infers some other fact which the documents do notmention. It is an application of the fundamental principle of history, the _analogy_ between present and past humanity. In the present weobserve that the facts of humanity are connected together. Given onefact, another fact accompanies it, either because the first is the causeof the second, or because the second is the cause of the first, orbecause both are effects of a common cause. We assume that in the pastsimilar facts were connected in a similar manner, and this assumption iscorroborated by the direct study of the past in the documents. From agiven fact, therefore, which we find in the past, we may infer theexistence of the other facts which were connected with it. 

This reasoning applies to facts of all kinds, usages, transformations, individual incidents. We may begin with any known fact and endeavour toinfer unknown facts from it. Now the facts of humanity, having a commoncentre, man, are all connected together, not merely facts of the sameclass, but facts belonging to the most widely different classes. Thereare connections, not merely between the different facts relating to art, to religion, to manners, to politics, but between the facts of religionon the one hand and the facts of art, of politics, and of manners on theother; thus from a fact of one species we may infer facts of all theother species. 

To examine those connections between facts on which reasonings may befounded would mean tabulating all the known relations between the factsof humanity, that is, giving a full account of all the empirical laws ofsocial life. Such a labour would provide matter for a whole book. [201]Here we shall content ourselves with indicating the general rulesgoverning this kind of reasoning, and the precautions to be takenagainst the most common errors. 

The argument rests on two propositions: one is general, and is derivedfrom experience of human affairs; the other is particular, and isderived from the documents. In practice, we begin with the particularproposition, the historical fact: Salamis bears a Phoenician name. Wethen look for a general proposition: the language of the name of a cityis the language of the people which founded it. And we conclude:Salamis, bearing a Phoenician name, was founded by the Phoenicians. 

In order that the conclusion may be certain, two conditions arenecessary. 

(1) The general proposition must be accurately true; the two facts whichit declares to be connected must be connected in such a way that the oneis never found without the other. If this condition were completelysatisfied we should have a _law_, in the scientific sense of the word;but in dealing with the facts of humanity--apart from those physicalconditions whose laws are established by the regular sciences--we canonly work with empirical laws obtained by rough determinations ofgeneral facts which are not analysed in such a manner as to educe theirtrue causes. These empirical laws are approximately true only when theyrelate to a numerous body of facts, for we can never quite know how fareach is necessary to produce the result. The proposition relating to thelanguage of the name of a city does not go enough into detail to bealways true. Petersburg is a German name, Syracuse in America bears aGreek name. Other conditions must be fulfilled before we can be surethat the name is connected with the nationality of the founders. Weshould, therefore, only employ such propositions as go into detail. 

(2) In order to employ a general proposition which goes into detail, wemust have a detailed knowledge of the particular fact; for it is nottill after this fact has been established that we look for an empiricalgeneral law on which to found an argument. We shall begin, then, bystudying the particular conditions of the case (the situation ofSalamis, the habits of the Greeks and Phoenicians); we shall not workon a single detail, but on an assemblage of details. 

Thus, in historical reasoning it is necessary to have (1) an accurategeneral proposition; (2) a detailed knowledge of a past fact. It is badworkmanship to assume a false general proposition--to suppose, forexample, as Augustin Thierry did, that every aristocracy had its originin a conquest. It is bad workmanship, again, to found an argument on anisolated detail (the name of a city). The nature of these errorsindicates the precautions to be taken. 

(1) The spontaneous tendency is to take as a basis of reasoning those"common-sense truths" which form nearly the whole of our knowledge ofsocial life. Now, the greater part of these are to some extent false, for the science of social life is still imperfect. And the chief dangerin them lies in the circumstance that we use them unconsciously. Thesafest precaution will be always to formulate the supposed law on whichwe propose to base an argument. In every instance where such and such afact occurs, it is certain that such and such another fact occurs also. If this proposition is obviously false, we shall at once see it to beso; if it is too general, we shall inquire what new conditions may beintroduced to make it accurate. 

(2) A second spontaneous impulse leads us to draw consequences fromisolated facts, even of the slightest kind (or rather, the idea of eachfact awakens in us, by association, the idea of other facts). This isthe natural procedure in the history of literature. Each circumstance inthe life of an author supplies material for reasoning; we construct byconjecture all the influences which could have acted upon him, and weassume that they did act upon him. All the branches of history whichstudy a single species of facts, isolated from every other species(language, arts, private law, religion), are exposed to the same danger, because they deal with fragments of human life, not with comprehensivecollections of phenomena. But few conclusions are firmly establishedexcept those which rest on a comprehensive body of data. We do not makea diagnosis from a single symptom, but from a number of concurrentsymptoms. The precaution to be taken will be to avoid working with anisolated detail or an abstract fact. We must have before our mindsactual men, as affected by the principal conditions under which theylived. 

We must be prepared to realise but rarely the conditions of a certaininference; we are too little acquainted with the laws of social life, and too seldom know the precise details of an historical fact. Thus mostof our reasonings will only afford presumptions, not certainties. But itis with reasonings as with documents. [202] When several presumptionsall point in the same direction they confirm each other, and end byproducing a legitimate certitude. History fills up some of its gaps byan accumulation of reasonings. Doubts remain as to the Phoenicianorigin of various Greek cities, but there is no doubt about the presenceof the Phoenicians in Greece. 

CHAPTER IV

THE CONSTRUCTION OF GENERAL FORMULÆ

I. Suppose we had methodically arranged all the historical factsestablished by the analysis of documents, or by reasoning; we shouldpossess a systematised inventory of the whole of history, and the workof construction would be complete. Ought history to stop at this point?The question is warmly debated, and we cannot avoid giving an answer, for it is a question with a practical bearing. 

Critical scholars, who are accustomed to collect all the facts relatingto their speciality, without any personal preference, are inclined toregard a complete, accurate, and objective collection of facts as theprime requisite. All historical facts have an equal right to a place inhistory; to retain some as being of greater importance, and reject therest as comparatively unimportant, would be to introduce the subjectiveelement of choice, variable according to individual fancy; historycannot sacrifice a single fact. 

Against this very reasonable view there is nothing to be urged except amaterial difficulty; this, however, is enough, for it is the practicalmotive of all the sciences: we mean the impossibility of acquiring orcommunicating complete knowledge. A body of history in which no factwas sacrificed would have to contain all the actions, all the thoughts, all the adventures of all men at all times. It would form a total whichno one could possibly make himself master of, not for want of materials, but for want of time. This, indeed, applies, as things are, to certainvoluminous collections of documents: the collected reports ofparliamentary debates contain the whole history of the variousassemblies, but to learn their history from these sources would requiremore than a lifetime. 

Every science must take into consideration the practical conditions oflife, at least so far as it claims to be a real science, a science whichit is possible to know. Any ideal which ends by making knowledgeimpossible impedes the establishment of the science. 

Science is a saving of time and labour, effected by a process whichprovides a rapid means of learning and understanding facts; it consistsin the slow collection of a quantity of details and their condensationinto portable and incontrovertible formulæ. History, which is moreencumbered with details than any other science, has the choice betweentwo alternatives: to be complete and unknowable, or to be knowable andincomplete. All the other sciences have chosen the second alternative;they abridge and they condense, preferring to take the risk ofmutilating and arbitrarily combining the facts to the certainty of beingunable either to understand or communicate them. Scholars have preferredto confine themselves to the periods of ancient history, where chance, which has destroyed nearly all the sources of information, has freedthem from the responsibility of choosing between facts by depriving themof nearly all the means of knowing them. 

History, in order to constitute itself a science, must elaborate the rawmaterial of facts. It must condense them into manageable form by meansof descriptive formulæ, qualitative and quantitative. It must search forthose connections between facts which form the ultimate conclusions ofevery science. 

II. The facts of humanity, with their complex and varied character, cannot be reduced like chemical facts to a few simple formulæ. Like theother sciences which deal with life, history needs descriptive formulæin order to express the nature of the different phenomena. 

In order to be manageable, a formula must be short; in order to give anexact idea of the facts, it must be precise. Now, in the knowledge ofhuman affairs, precision can only be obtained by attention tocharacteristic details, for these alone enable us to understand how onefact differed from others, and what there was in it peculiar to itself. There is thus a conflict between the need of brevity, which leads us tolook for concrete formulæ, and the necessity of being precise, whichrequires us to adopt detailed formulæ. Formulæ which are too short makescience vague and illusory, formulæ which are too long encumber it andmake it useless. This dilemma can only be evaded by a perpetualcompromise, the principle of which is to compress the facts by omittingall that is not necessary for the purpose of representing them to themind, and to stop at the point where omission would suppress somecharacteristic feature. 

This operation, which is difficult in itself, is still furthercomplicated by the state in which the facts which are to be condensedinto formulæ present themselves. According to the nature of thedocuments from which they are derived, they come to us in all thedifferent degrees of precision: from the detailed narrative whichrelates the smallest episodes (the battle of Waterloo) down to thebarest mention in a couple of words (the victory of the Austrasians atTestry). On different facts of the same kind we possess an amount ofdetails which is infinitely variable according as the documents give usa complete description or a mere mention. How are we to organise into acommon whole, items of knowledge which differ so widely in point ofprecision? When facts are known to us from a vague word of generalimport, we cannot reduce them to a less degree of generality and agreater degree of precision; we do not know the details. If we add themconjecturally we shall produce an historical novel. This is whatAugustin Thierry did in the case of his _Récits mérovingiens_. Whenfacts are known in detail, it is always easy to reduce them to a greaterdegree of generality by suppressing characteristic details; this is whatis done by the authors of abridgements. But the result of this procedurewould be to reduce history to a mass of vague generalities, uniform forthe whole of time except for the proper names and the dates. It would bea dangerous method of introducing symmetry, to bring all facts to acommon degree of generality by levelling them all to the condition ofthose which are the most imperfectly known. In those cases, therefore, where the documents give details, our descriptive formulæ should alwaysretain the characteristic features of the facts. 

In order to construct these formulæ we must return to the set ofquestions which we employed in grouping the facts, we must answer eachquestion, and compare the answers. We shall then combine them into ascondensed and as precise a formula as possible, taking care to keep afixed sense for every word. This may appear to be a matter of style, butwhat we have in view here is not merely a principle of exposition, necessary for the sake of being intelligible to the reader, it is aprecaution which the author ought to take on his own account. The factsof society are of an elusive nature, and for the purpose of seizing andexpressing them, fixed and precise language is an indispensableinstrument; no historian is complete without good language. 

It will be well to make the greatest possible use of concrete anddescriptive terms: their meaning is always clear. It will be prudent todesignate collective groups only by collective, not by abstract names(royalty, State, democracy, Reformation, Revolution), and to avoidpersonifying abstractions. We think we are simply using metaphors, andthen we are carried away by the force of the words. Certainly abstractterms have something very seductive about them, they give a scientificappearance to a proposition. But it is only an appearance, behind whichscholasticism is apt to be concealed; the word, having no concretemeaning, becomes a purely verbal notion (like the soporific virtue ofwhich Molière speaks). As long as our notions on social phenomena havenot been reduced to truly scientific formulæ, the most scientific coursewill be to express them in terms of every-day experience. 

In order to construct a formula, we should know beforehand what elementsought to enter into it. We must here make a distinction between generalfacts (habits and evolutions) and unique facts (events). 

III. General facts consist in actions which are often repeated, and arecommon to a number of men. We have to determine their _character_, _extent_, and _duration_. 

In order to formulate their character, we combine all the features whichconstitute a fact (habit, institution) and distinguish it from allothers. We unite under the same formula all the individual cases whichgreatly resemble each other, by neglecting the individual differences. 

This concentration is performed without effort in the case of habitswhich have to do with forms (language, handwriting), and in the case ofall intellectual habits; those who practised these habits have alreadygiven them expression in formulæ, which we have only to collect. Thesame holds of these institutions which are sanctioned by expresslyformulated rules (regulations, laws, private statutes). Accordingly thespecial branches of history were the first to yield methodical formulæ. On the other hand, these special branches do not go beyond superficialand conventional facts, they do not reach the real actions and thoughtsof men: in language they deal with written words, not the realpronunciation; in religion with official dogmas and rites, not with thereal beliefs of the mass of the people; in morals with avowed precepts, not with the effective ideals; in institutions with official rules, notwith the real practice. On all these subjects the knowledge ofconventional forms must some day be supplemented by a parallel study ofthe real habits. 

It is much more difficult to embrace in a single formula a habit whichis composed of real actions, as is the case with economic phenomena, private life, politics; for we have to find in the different actionsthose common characteristics which constitute the habit; or, if thiswork has already been done in the documents, and condensed into aformula (the most common case), we must criticise this formula in orderto make sure that it really represents a homogeneous habit. 

The same difficulty occurs in constructing the formula for a group; wehave to describe the characteristic common to all the members of thegroup and to find a collective name which shall exactly designate it. Indocuments there is no lack of names of groups; but, as they have theirorigin in usage, many of them correspond but ill to the real groups; wehave to criticise these names to fix their precise meaning, sometimes tocorrect their application. 

This first operation should yield formulæ expressive of the conventionaland real characteristics of all the habits of the different groups. 

In order to fix the precise _extent_ of a habit we shall seek the mostdistant points where it appears (this will give the area ofdistribution), and the region where it is most common (the centre). Sometimes the operation takes the form of a map (for example the map ofthe _tumuli_ and the _dolmens_ of France). It will also be necessary toindicate the groups of men who practised each habit, and the sub-groupsin which it was most pronounced. 

The formula should also indicate the _duration_ of the habit. We shalllook for the extreme cases, the first and the last appearance of theform, the doctrine, the usage, the institution, the group. But it willnot be enough to note the two isolated cases, the earliest and the mostrecent; we must ascertain the period in which it was really active. 

The formula of an evolution ought to indicate the successive variationsin the habit, giving in each case precise limits of extent and duration. Then, by comparing all the variations, it will be possible to determinethe general course of the evolution. The general formula will indicatewhen and where the evolution began and ended, and the nature of thechange which it effected. All evolutions present common features whichenable them to be divided into stages. Every habit (usage orinstitution) begins by being the spontaneous act of several individuals;when others imitate them it becomes a usage. Similarly social functionsare in the first instance performed by persons who undertake themspontaneously, when these persons are recognised by others they acquirean official status. This is the first stage; individual initiativefollowed by general imitation and recognition. The usage becomestraditional and is transformed into an obligatory custom or rule; thepersons acquire a permanent status and are invested with powers ofmaterial or moral constraint. This is the stage of tradition andauthority; very often it is the last stage, and continues till thesociety is destroyed. The usage is relaxed, the rules are violated, thepersons in authority cease to be obeyed; this is the stage of revolt anddecomposition. Finally, in certain civilised societies, the rule iscriticised, the persons in authority are censured, by the action of apart of the subjects a rational change is effected in the composition ofthe governing body, which is subjected to supervision; this is the stageof reform and of checks. 

IV. In the case of unique facts we cannot expect to bring severaltogether under a common formula, for the nature of these facts is tooccur but once. However, it is imperatively necessary to abridge, wecannot preserve all the acts of all the members of an assembly or of allthe officers of a state. Many individuals and many facts must besacrificed. 

How are we to choose? Personal tastes and patriotism give rise topreferences for congenial characters and for local events; but the onlyprinciple of selection which can be employed by all historians in commonis that which is based on the part played in the evolution of humanaffairs. We ought to retain those persons and those events which havevisibly influenced the course of an evolution. We may recognise them byour inability to describe the evolution without mentioning them. The menare those who have modified the state of a society either by thecreation or the introduction of a habit (artists, men of science, inventors, founders, apostles), or as directors of a movement, heads ofstates, of parties, of armies. The events are those which have broughtabout changes in the habits or the state of societies. 

In order to construct a formula descriptive of an historical person, wemust take particulars from his biography and his habits. From hisbiography we shall take those facts which determined his career, formedhis habits, and occasioned the actions by which he influenced society. These comprise physiological conditions (physique, temperament, state ofhealth), [203] the educational influences, the social conditions to whichhe was subject. The history of literature has accustomed us toresearches of this kind. 

Among the habits of a man it is necessary to determine his fundamentalconceptions relating to the class of facts in which his influence wasfelt, his conception of life, his knowledge, his predominating tastes, his habitual occupations, his principles of conduct. From these details, in which there is infinite variety, an impression is formed of the man's"character, " and the collection of these characteristic featuresconstitutes his "portrait, " or, to use a favourite phrase of the day, his "psychology. " This exercise, which is still held in great esteem, dates from the time when history was still a branch of literature; it isdoubtful whether it can ever become a scientific process. There isperhaps no sure method of summing up the character of a man, even in hislifetime, still less when we can only know him indirectly through themedium of documents. The controversies relative to the interpretation ofthe conduct of Alexander are a good example of this uncertainty. 

If, however, we take the risk of seeking a formula to describe acharacter, there are two natural temptations against which we mustguard: (I) We must not construct the formula out of the person'sassertions in regard to himself. (2) The study of imaginary personages(dramas and novels) has accustomed us to seek a logical connectionbetween the various sentiments and the various acts of a man; acharacter, in literature, is constructed logically. This search forcoherency must not be transferred to the study of real men. We are lesslikely to do so in the case of those whom we observe in their lifetime, because we see too many characteristics in them which could not enterinto a coherent formula. But the absence of documents, by suppressingthose characteristics which would have checked us, encourages us toarrange the very small number of those which remain in the form of astage-character. This is why the great men of antiquity seem to us tohave been much more logical than our contemporaries are. 

How are we to construct a formula for an event? The imperative need ofsimplification causes us to combine under a single name an enormous massof minute facts which are perceived in the lump, and between which wevaguely feel that there is a connection (a battle, a war, a reform). The facts which are thus combined are such facts as have conduced to acommon result. That is how the common notion of an event arises, andthere is no more scientific conception to put in its place. Facts, then, are to be grouped according to their consequences; those which have hadno visible consequences disappear, the others are fused into a certainnumber of aggregates which we call events. 

In order to describe an event, it is necessary to give preciseindications (I) of its character, (2) of its extent. 

(I) By the character of an event we mean the features which distinguishit from every other event, not merely the external conditions of dateand place, but the manner in which it occurred, and its immediatecauses. The following are the items of information which the formulashould contain. One or more men, in such and such mental states(conceptions, motives of the action), working under such and suchmaterial conditions (locality, instrument), performed such and suchactions, which had for their result such and such a modification. Forthe determination of the motives of the actions, the only method is tocompare the actions, firstly, with the declarations of those whoperformed them; secondly, with the interpretation of those who witnessedtheir performance. There is often a doubt remaining: this is the fieldof party polemics; every one attributes noble motives to the actions ofhis own party and discreditable motives to those of the opposite party. But actions described without any indication of motive would beunintelligible. 

(2) The extension of the event will be indicated both in space (theplace where it happened, and the region in which its immediate effectswere felt) and in time, the moment when its realisation began, and themoment when the result was brought about. 

V. Descriptive formulæ relating to characters, being merely qualitative, only give an abstract idea of the facts; in order to realise the placethey occupied in reality, quantity is necessary. It is not a matter ofindifference whether a given usage was practised by a hundred men or bymillions. 

For the purpose of introducing quantity into formulæ we have at ourdisposal several methods, of various degrees of imperfection, which helpus to attain the end in view with various degrees of precision. Arrangedin descending order of precision they are as follows:--

(1) _Measurement_ is a perfectly scientific procedure, for equal numbersrepresent absolutely identical values. But a common unit is necessary, and that can only be had for time and for physical phenomena (lengths, surfaces, weights). Figures relating to production and sums of money arethe essential elements in the statement of economic and financial facts. But facts of the psychological order remain inaccessible to measurement. 

(2) _Enumeration_, which is the process employed in statistics, [204] isapplicable to all the facts which have in common a definitecharacteristic which can be made use of for counting them. The factswhich are thus comprehended under a single number do not all belong tothe same species, they may have in common but a single characteristic, abstract (crime, lawsuit) or conventional (workman, lodging); thefigures merely indicate the number of cases in which a givencharacteristic is met with; they do not represent a homogeneous whole. Anatural tendency is to confuse number with measurement, and to supposethat facts are known with scientific precision because it has beenpossible to apply number to them; this is an illusion to be guardedagainst, we must not take the figures which give the number of apopulation or an army for the measure of its importance. [205] Still, enumeration yields results which are necessary for the construction offormulæ relating to groups. But the operation is restricted to thosecases in which it is possible to know all the units of a given specieslying within given limits, for it is performed by first ticking off, then adding. Before undertaking a retrospective enumeration, therefore, it will be well to make sure that the documents are complete enough toexhibit all the units which are to be enumerated. As to figures given indocuments, they are to be distrusted. 

(3) _Valuation_ is a kind of incomplete enumeration applying to aportion of the field, and made on the supposition that the sameproportions hold good through the whole of the field. It is an expedientto which, in history, it is often necessary to have recourse whendocuments are unequally abundant for the different divisions of thesubject. The result is open to doubt, unless we are sure that theportion to which enumeration was applied was exactly similar to theremainder. 

(4) _Sampling_ is a process of enumeration restricted to a few unitstaken at different points in the field of investigation; we calculatethe proportion of cases (say 90 per cent. ) where a given characteristicoccurs, we assume that the same proportion holds throughout, and ifthere are several categories we obtain the proportion between them. Inhistory this procedure is applicable to facts of every kind, for thepurpose of determining either the proportion between the different formsor usages which occur within a given region or period, or the proportionwhich obtains, within a heterogeneous group, between members belongingto different classes. This procedure gives us an approximate idea of thefrequency of facts and the proportion between the different elements ofa society; it can even show what species of facts are most commonlyfound together, and are therefore probably connected. But in order thatthe method may be employed correctly it is necessary that the samplesshould be representative of the whole, and not of a part which mightpossibly be exceptional in character. They should therefore be chosen atvery different points, and under very different conditions, in orderthat the exceptions may compensate each other. It is not enough to takethem at points which are _distant_ from each other; for example, on thedifferent frontiers of a country, for the very circumstance ofsituation on a frontier is an exceptional condition. Verification may behad by following the methods by which anthropologists obtain averages. 

(5) _Generalisation_ is only an instinctive process of simplification. As soon as we perceive a certain characteristic in an object, we extendthis characteristic to all other objects which at all resemble it. Inall human concerns, where the facts are always complex, we makegeneralisations unconsciously; we attribute to a whole people the habitsof a few individuals, or those of the first group forming part of thepeople which comes within our knowledge; we extend to a whole periodhabits which are ascertained to have existed at a given moment. This isthe most active of all the causes of historical error, and one whoseinfluence is felt in every department, in the study of usages and ofinstitutions, even in the appreciation of the morality of a people. [206]Generalisation rests on a vague idea that all facts which are contiguousto each other, or which resemble each other in some point, are similarat all points. It is an unconscious and ill-performed process ofsampling. It may therefore be made correct by being subjected to theconditions of a well-performed process of sampling. We must examine thecases on which we propose to found a generalisation and ask ourselves. What right have we to generalise? That is, what reason have we forassuming that the characteristic discovered in these cases will occur inthe remaining thousands of cases? that the cases chosen resemble theaverage? The only valid reason would be that these cases arerepresentative of the whole. We are thus brought back to the process ofmethodical sampling. 

The right method of conducting the operation is as follows: (1) We mustfix the precise limits of the field within which we intend to generalise(that is, to assume the similarity of all the cases), we must determinethe country, the group, the class, the period as to which we are togeneralise. Care must be taken not to make the field too large byconfusing a part with the whole (a Greek or Germanic people with thewhole Greek or Germanic race). (2) We must make sure that the factslying within the field resemble each other in the points on which wewish to generalise, and therefore we have to distrust those vague namesunder which are comprehended groups of very different character(Christians, French, Aryans, Romans). (3) We must make sure that thefacts from which we propose to generalise are representative samples, that they really belong to the field of investigation, for it doeshappen sometimes that men or facts are taken as specimens of one groupwhen they really belong to another. Nor must they be exceptional, as isto be presumed in all cases when the conditions are exceptional; authorsof documents tend to record by preference those facts which surprisethem, hence exceptional cases occupy in documents a space which is outof proportion to their real number; this is one of the chief sources oferror. (4) The number of samples necessary to support a generalisationis the greater the less ground there is for supposing a resemblancebetween all the cases occurring within the field of investigation. Asmall number may suffice in treating of points in which men tend to beara strong resemblance to each other, either by imitation and convention(language, rites, ceremonies), or from the influence of custom andobligatory regulations (social institutions, political institutions incountries where the authorities are obeyed). A large number is requisitefor facts where individual initiative plays a more important part (art, science, morality), and sometimes, as in respect of private conduct, allgeneralisation is as a rule impossible. 

VI. Descriptive formulæ are in no science the final result of the work. It still remains to group the facts in such a way as to bring out theircollective import, it still remains to search for their mutualrelations; these are the general conclusions. History, by reason of theimperfection of its mode of acquiring knowledge, needs, in addition, apreliminary operation for determining the bearing of the knowledgeacquired. [207]

The work of criticism has supplied us with nothing but a number ofisolated remarks on the value of the knowledge which the documents havepermitted us to acquire. These must be combined. We shall therefore takea whole group of facts entered under a common heading--a particularclass of facts, a country, a period, an event--and we shall summarisethe results yielded by the criticism of particular facts so as to obtaina general formula. We shall have to take into consideration: (1) theextent, (2) the value of our knowledge. 

(1) We shall ask ourselves what are the blanks left by the documents. Byworking through the scheme used for the grouping of facts it is easy todiscover what are the classes of facts on which we lack information. Inthe case of evolution, we notice which links are missing in the chain ofsuccessive modifications; in the case of events, what episodes, whatgroups of actors are still unknown to us; what facts enter or disappearfrom the field of our knowledge without our being able to trace theirbeginning or end. We ought to construct, mentally at any rate, atabulated scheme of the points on which we are ignorant, in order tokeep before our minds the distance separating the knowledge we have froma perfect knowledge. 

(2) The value of our knowledge depends on the value of our documents. Criticism has given us indications on this point in each separate case, these indications, so far as relating to a given body of facts, must besummarised under a few heads. Does our knowledge come originally fromdirect observation, from written tradition, or from oral tradition? Dowe possess several traditions of different bias, or a single tradition?Do we possess documents of different classes or of one single class? Isour information vague or precise, detailed or summary, literary orpositive, official or confidential?

The natural tendency is to forget, in construction, the results yieldedby criticism, to forget the incompleteness of our knowledge and theelements of doubt in it. An eager desire to increase to the greatestpossible extent the amount of our information and the number of ourconclusions impels us to seek emancipation from all negativerestrictions. We thus run a great risk of using fragmentary andsuspicious sources of information for the purpose of forming generalimpressions, just as if we were in possession of a complete record. Itis easy to forget the existence of those facts which the documents donot describe (economic facts, slaves in antiquity), it is easy toexaggerate the space occupied by facts which are known to us (Greek art, Roman inscriptions, mediæval monasteries). We instinctively estimate theimportance of facts by the number of the documents which mention them. We forget the peculiar character of the documents, and, when they allhave a common origin, we forget that they have all subjected the factsto the same distortions, and that their community of origin rendersverification impossible; we submissively reproduce the bias of thetradition (Roman, orthodox, aristocratic). 

In order to resist these natural tendencies, it is enough to pass inreview the whole body of facts and the whole body of tradition, beforeattempting to draw any general conclusion. 

VII. Descriptive formulæ give the particular character of each smallgroup of facts. In order to obtain a general conclusion, we must combinethese detailed results into a general formula. We must not comparetogether isolated details or secondary characteristics, [208] but groupsof facts which resemble each other in a whole set of characteristics. 

We thus form an aggregate (of institutions, of groups of men, ofevents). Following the method indicated above, we determine itsdistinguishing characteristics, its extent, its duration, its quantityor importance. 

As we form groups of greater and greater generality we drop, with eachnew degree of generality, those characteristics which vary, and retainthose which are common to all the members of the new group. We must stopat the point where nothing is left except the characteristics common tothe whole of humanity. The result is the condensation into a singleformula of the general character of an order of facts, of a language, areligion, an art, an economic organisation, a society, a government, acomplex event (such as the Invasion or the Reformation). 

As long as these comprehensive formulæ remain isolated the conclusion isincomplete. And as it is no longer possible to fuse them into highergeneralisations, we feel the need of comparing them for the purpose ofclassification. This classification may be attempted by two methods. 

(1) We may compare together similar categories of special facts, language, religions, arts, governments, taking them from the whole ofhumanity, and classifying together those which most resemble each other. We obtain families of languages, religions, and governments, which wemay again classify and arrange among themselves. This is an abstractkind of classification; it isolates one species of facts from all theothers, and thus renounces all claim to exhibit causes. It has theadvantage of being rapidly performed and of yielding a technicalvocabulary which is useful for designating facts. 

(2) We may compare real groups of real individuals, we may takesocieties which figure in history and classify them according to theirsimilarities. This is a concrete classification analogous to that ofzoology, in which, not functions, but whole animals are classified. Itis true that the groups are less clearly marked than in zoology; nor isthere a general agreement as to the characteristics in respect of whichwe are to look for resemblances. Are we to choose the economic or thepolitical organisation of the groups, or their intellectual condition?No principle of choice has as yet become obligatory. 

History has not yet succeeded in establishing a scientific system ofcomprehensive classification. Possibly human groups are not sufficientlyhomogeneous to furnish a solid basis of comparison, and not sharplyenough divided to be treated as comparable units. 

VIII. The study of the relations between simultaneous facts consists ina search for the connections between all the facts of different specieswhich occur in a given society. We have a vague consciousness that thedifferent habits which are separated by abstraction and ranged underdifferent categories (art, religion, political institutions), are notisolated in reality, that they have common characteristics, and thatthey are closely enough connected for a change in one of them to bringabout a change in another. This is a fundamental idea of the _Esprit desLois_ of Montesquieu. This bond of connection, sometimes called_consensus_, has received the name of _Zusammenhang_ from the Germanschool. From this conception has arisen the theory of the _Volksgeist_(the mind of a people), a counterfeit of which has within the last fewyears been introduced into France under the name of "âme nationale. "This conception is also at the bottom of the theory regarding the soulof society which Lamprecht has expounded. 

After the rejection of these mystical conceptions there remains a vaguebut incontrovertible fact, the "solidarity" which exists between thedifferent habits of one and the same people. In order to study it withprecision it would be necessary to analyse it, and a connecting bondcannot be analysed. It is thus quite natural that this part of socialscience should have remained a refuge for mystery and obscurity. 

By the comparison of different societies which resemble or differ fromeach other in a given department (religion or government), with theobject of discovering in what other departments they resemble or differfrom each other, it is possible that interesting empirical results mightbe obtained. But, in order to _explain_ the _consensus_, it is necessaryto work back to the facts which have produced it, the common causes ofthe various habits. We are thus obliged to undertake the investigationof causes, and we enter the province of what is called _philosophical_history, because it investigates what was formerly called the_philosophy_ of facts--that is to say, their permanent relations. 

IX. The necessity of rising above the simple determination of facts inorder to _explain_ them by their _causes_, a necessity which hasgoverned the development of all the sciences, has at length been felteven in the study of history. Hence have arisen systematic philosophiesof history, and attempts to discover historical laws and causes. Wecannot here enter into a critical examination of these attempts, whichthe nineteenth century has produced in so great number; we shall merelyindicate what are the ways in which the problem has been attacked, andwhat obstacles have prevented a scientific solution from being reached. 

The most natural method of explanation consists in the assumption that atranscendental cause, Providence, guides the whole course of eventstowards an end which is known to God. [209] This explanation can be but ametaphysical doctrine, crowning the work of science; for thedistinguishing feature of science is that it only studies efficientcauses. The historian is not called upon to investigate the first causeor final causes any more than the chemist or the naturalist. And, infact, few writers on history nowadays stop to discuss the theory ofProvidence in its theological form. 

But the tendency to explain historical facts by transcendental causessurvives in more modern theories in which metaphysic is disguised underscientific forms. The historians of the nineteenth century have been sostrongly influenced by their philosophical education that most of them, sometimes unconsciously, introduce metaphysical formulæ into theconstruction of history. It will be enough to enumerate these systems, and point out their metaphysical character, so that reflectinghistorians may be warned to distrust them. 

The theory of the rational character of history rests on the notion thatevery real historical fact is at the same time "rational"--that is, inconformity with an intelligible comprehensive plan; ordinarily it istacitly assumed that every social fact has its _raison d'être_ in thedevelopment of society--that is, that it ends by turning to theadvantage of society; hence the cause of every institution is sought forin the social need it was originally meant to supply. [210] This is thefundamental idea of Hegelianism, if not with Hegel, at least with thehistorians who have been his disciples (Ranke, Mommsen, Droysen, inFrance Cousin, Taine, and Michelet). This is a lay disguise of the oldtheological theory of final causes which assumes the existence of aProvidence occupied in guiding humanity in the direction of itsinterests. This is a consoling, but not a scientific _a priori_hypothesis; for the observation of historical facts does not indicatethat things have always happened in the most rational way, or in the waymost advantageous to men, nor that institutions have had any other causethan the interest of those who established them; the facts, indeed, point rather to the opposite conclusion. 

From the same metaphysical source has also sprung the Hegelian theory ofthe _ideas_ which are successively realised in history through themedium of successive peoples. This theory, which has been popularised inFrance by Cousin and Michelet, has had its day, even in Germany, but ithas been revived, especially in Germany, in the form of the historicalmission (_Beruf_) which is attributed to peoples and persons. It willhere be enough to observe that the very metaphors of "idea" and"mission" imply a transcendental anthropomorphic cause. 

From the same optimistic conception of a rational guidance of the worldis derived the theory of the continuous and necessary _progress_ ofhumanity. Although it has been adopted by the positivists, this ismerely a metaphysical hypothesis. In the ordinary sense of the word, "progress" is merely a subjective expression denoting those changeswhich follow the direction of our preferences. But, even taking the wordin the objective sense given to it by Spencer (an increase in thevariety and coordination of social phenomena), the study of historicalfacts does not point to a _single_ universal and continuous progress ofhumanity, it brings before us a _number_ of partial and intermittentprogressive movements, and it gives us no reason to attribute them to apermanent cause inherent in humanity as a whole rather than to a seriesof local accidents. [211]

Attempts at a more scientific form of explanation have had their originin the special branches of history (of languages, religion, law). By theseparate study of the succession of facts of a single species, specialists have been enabled to ascertain the regular recurrence of thesame successions of facts, and these results have been expressed informulæ which are sometimes called laws (for example, the law of thetonic accent); these are never more than empirical laws which merelyindicate successions of facts without explaining them, for they do notreveal the efficient cause. But specialists, influenced by a naturalmetaphor, and struck by the regularity of these successions, haveregarded the evolution of usages (of a word, a rite, a dogma, a rule oflaw), as if it were an organic development analogous to the growth of aplant; we hear of the "life of words, " of the "death of dogmas, " of the"growth of myths. " Then, in forgetfulness of the fact that all thesethings are pure abstractions, it has been tacitly assumed that there isa force inhering in the word, the rite, the rule, which produces itsevolution. This is the theory of the development (_Entwickelung_) ofusages and institutions; it was started in Germany by the "historical"school, and has dominated all the special branches of history. Thehistory of languages alone has succeeded in shaking off itsinfluence. [212] Just as usages have been treated as if they wereexistences possessing a separate life of their own, so the succession ofindividuals composing the various bodies within a society (royalty, church, senate, parliament) has been personified by the attribution toit of a will, which is treated as an active cause. A world of imaginarybeings has thus been created behind the historical facts, and hasreplaced Providence in the explanation of them. For our defence againstthis deceptive mythology a single rule will suffice: Never seek thecauses of an historical fact without having first expressed itconcretely in terms of acting and thinking individuals. If abstractionsare used, every metaphor must be avoided which would make them play thepart of living beings. 

By a comparison of the evolutions of the different species of factswhich coexist in one and the same society, the "historical" school wasled to the discovery of solidarity (_Zusammenhang_). [213] But, beforeattempting to discover its causes by analysis, the adherents of thisschool assumed the existence of a permanent general cause residing inthe society itself. And, as it was customary to personify society, aspecial temperament was attributed to it, the peculiar genius of thenation or the race, manifesting itself in the different socialactivities and explaining their solidarity. [214] This was simply anhypothesis suggested by the animal world, in which each species haspermanent characteristics. It would have been inadequate, for in orderto explain how a given society comes to change its character from oneepoch to another (the Greeks between the seventh and the fourthcenturies, the English between the fifteenth and the nineteenth), itwould have been necessary to invoke the aid of external causes. And thetheory is untenable, for all the societies known to history are groupsof men without anthropological unity and without common hereditarycharacteristics. 

In addition to these metaphysical or metaphorical explanations, attemptshave been made to apply to the investigation of causes in history theclassical procedure of the natural sciences: the comparison of parallelseries of successive phenomena in order to discover those which alwaysappear together. The "comparative method" has assumed several differentforms. Sometimes the subject of study has been a detail of social life(a usage, an institution, a belief, a rule), defined in abstract terms;its evolutions in different societies have been compared with a view todetermine the common evolution which is to be attributed to one and thesame general cause. Thus have arisen comparative philology, mythology, and law. It has been proposed (in England) to give precision to thecomparative method by applying "statistics"; this would mean thesystematic comparison of all known societies and the enumeration of allthe cases where two usages are found together. This is the principle ofBacon's tables of agreement; it is to be feared that it will be no morefertile in results. The defect of all such methods is that they apply toabstract and partly arbitrary notions, sometimes merely to verbalresemblances, and do not rest on a knowledge of the whole of theconditions under which the facts occur. 

We can conceive a more concrete method which, instead of comparingfragments, should compare wholes, that is entire societies, either thesame society at different stages of its evolution (England in thesixteenth, and again in the nineteenth century), or else the generalevolution of several societies, contemporary with each other (Englandand France), or existing at different epochs (Rome and England). Such amethod might be useful negatively, for the purpose of ascertaining thata given fact is not the necessary effect of another, since they are notalways found together (for example, the emancipation of women andChristianity). But positive results are hardly to be expected of it, forthe concomitance of two facts in several series does not show whetherone is the cause of the other, or whether both are joint effects of asingle cause. 

The methodical investigation of the causes of a fact requires ananalysis of the conditions under which the fact occurs, performed so asto isolate the necessary condition which is its cause; it presupposes, therefore, the complete knowledge of these conditions. But this isprecisely what we never have in history. We must therefore renounce theidea of arriving at causes by direct methods such as are used in theother sciences. 

As a matter of fact, however, historians often do employ the notion ofcause, which, as we have shown above, is indispensable for the purposeof formulating events and constructing periods. They know causes partlyfrom the authors of documents who observed the facts, partly from theanalogy of the causes which we all observe at the present day. The wholehistory of events is a chain of obviously and incontrovertibly connectedincidents, each one of which is the determining cause of another. Thelance-thrust of Montgomery is the cause of the death of Henry II. ; thisdeath is the cause of the accession to power of the Guises, which againis the cause of the rising of the Protestants. 

The observation of causes by the authors of documents is limited to theinterconnection of the accidental facts observed by them; these are, intruth, the causes which are known with the greatest certainty. Thushistory, unlike the other sciences, is better able to ascertain thecauses of particular incidents than those of general transformations, for the work is found already done in the documents. 

In the investigation of the causes of general facts, historicalconstruction is reduced to the analogy between the past and the present. Whatever chance there is of finding the causes which explain theevolution of past societies must lie in the direct observation of thetransformations of present societies. 

This is a branch of study which is not yet firmly established; here wecan only state the principles of it. 

(1) In order to ascertain the causes of the solidarity between thedifferent habits of one and the same society, it is necessary to lookbeyond the abstract and conventional form which the facts assume inlanguage (dogma, rule, rite, institution), and attend to the realconcrete centres, which are always thinking and acting men. Here onlyare found together the different species of activity which languageseparates by abstraction. Their solidarity is to be sought for in somedominating feature in the character or the environment of the men whichinfluences all the different manifestations of their activity. We mustnot expect the same degrees of solidarity in all the species ofactivity; there will be most of it in those species where eachindividual is in close dependence on the actions of the mass (economic, social, political life); there will be less of it in the intellectualactivities (arts, sciences), where individual initiative has freerplay. [215] Documents mention most habits (beliefs, customs, institutions) in the lump, without distinguishing individuals; and yet, in one and the same society, habits vary considerably from one man toanother. It is necessary to take account of these differences, otherwisethere is a danger of explaining the actions of artists and men ofscience by the beliefs and the habits of their prince or theirtradesmen. 

(2) In order to ascertain the causes of an evolution, it is necessary tostudy the only beings which can evolve--men. Every evolution has for itscause a change in the material conditions or in the habits of certainmen. Observation shows us two kinds of change. In the one case, the menremain the same, but change their manner of acting or thinking, eithervoluntarily through imitation, or by compulsion. In the other, the menwho practised the old usage disappear and are replaced by others who donot practise it; these may be strangers, or they may be the descendantsof the first set of men, but educated in a different manner. Thisrenewing of the generations seems, in our day, to be the most activecause of evolution. It is natural to suppose that the same holds good ofthe past; evolution has been slower, the more exclusively eachgeneration has been formed by the imitation of its forerunners. 

There is still one more question to ask. Are men all alike, differingmerely in the _conditions_ under which they live (education, resources, government), and is evolution produced solely by changes in these_conditions_? Or are there groups of men with _hereditary differences_, born with tendencies to different activities and with aptitudes leadingto different evolutions, so that evolution may be the product, in partat least, of the increase, the diminution, and the displacement of thesegroups? Taking the extreme cases, the white, black, and yellow races ofmankind, the differences in aptitude are obvious; no black people hasever developed a civilisation. It is thus probable that smallerhereditary differences may have had their share in the determination ofevents. If so, historical evolution would be partly produced byphysiological and anthropological causes. But history provides us withno sure means of determining the action of these hereditary differencesbetween men; it goes no further than the conditions of their existence. The last question of history remains insoluble by historical methods. 

CHAPTER V

EXPOSITION

We have still to study a question whose practical interest is obvious:What are the forms in which historical works present themselves? Theseforms are, in fact, very numerous. Some of them are antiquated; not allare legitimate; the best have their drawbacks. We should ask, therefore, not only what are the forms in which historical works appear, but alsowhich of these represent truly rational types of exposition. 

By "historical works" we mean here all those which are intended tocommunicate results obtained by the labour of historical construction, whatever may be the nature, the extent, and the bearing of theseresults. The critical elaboration of documents, which is treated of inBook II. , and which is preparatory to historical construction, isnaturally excluded. 

Historians may differ, and up to the present have differed, on severalessential points. They have not always had, nor have they all now, thesame conception of the end aimed at by historical work; hence arisedifferences in the nature of the facts chosen, the manner of dividingthe subject, that is, of co-ordinating the facts, the manner ofpresenting them, the manner of proving them. This would be the place toindicate how "the mode of writing history" has evolved from thebeginning. But as the history of the modes of writing history has notyet been written well, [216] we shall here content ourselves with somevery general remarks on the period prior to the second half of thenineteenth century, confining ourselves to what is strictly necessaryfor the understanding of the present situation. 

I. History was first conceived as the narration of memorable events. Topreserve the memory and propagate the knowledge of glorious deeds, or ofevents which were of importance to a man, a family, or a people; suchwas the aim of history in the tune of Thucydides and Livy. In addition, history was early considered as a collection of precedents, and theknowledge of history as a practical preparation for life, especiallypolitical life (military and civil). Polybius and Plutarch wrote toinstruct, they claimed to give recipes for action. Hence in classicalantiquity the subject-matter of history consisted chiefly of politicalincidents, wars, and revolutions. The ordinary framework of historicalexposition (within which the facts were usually arranged inchronological order) was the life of a person, the whole life of apeople, or a particular period in it; there were in antiquity but fewessays in general history. As the aim of the historian was to please orto instruct, or to please and instruct at the same time, history was abranch of literature: there were not too many scruples on the score ofproofs; those who worked from written documents took no care todistinguish the text of such documents from their own text; inreproducing the narratives of their predecessors they adorned them withdetails, and sometimes (under pretext of being precise) with numbers, with speeches, with reflections, and elegances. We can in a manner seethem at work in every instance where it is possible to compare Greek andRoman historians, Ephorus and Livy, for example, with their sources. 

The writers of the Renaissance directly imitated the ancients. For them, too, history was a literary art with apologetic aims or didacticpretensions. In Italy it was too often a means of gaining the favour ofprinces, or a theme for declamations. This state of affairs lasted along time. Even in the seventeenth century we find, in Mézeray, anhistorian of the ancient classical pattern. 

However, in the historical literature of the Renaissance, two noveltiesclaim our attention, in which the mediæval influence is incontrovertiblymanifest. On the one hand we see the retention of a form of expositionwhich was unusual in antiquity, which was created by the Catholichistorians of the later ages (Eusebius, Orosius), and which enjoyedgreat favour in the Middle Ages, --that which, instead of embracing onlythe history of a single man, family, or people, embraces universalhistory. On the other hand there was introduced a mechanical artifice ofexposition, having its origin in a practice common in the mediævalschools (the gloss), which had far-reaching consequences. The customarose of adding notes to printed books of history. [217] Notes have madeit possible to distinguish between the historical narrative and thedocuments which support it, to give references to sources, todisencumber and illustrate the text. It was in collections of documents, and in critical dissertations, that the artifice of annotation was firstemployed; thence it penetrated, slowly, into historical works of otherclasses. 

A second period begins in the eighteenth century. The "philosophers"then began to conceive history as the study, not of events for their ownsakes, but of the habits of men. They were thus led to take an interest, not only in facts of a political order, but in the evolution of thearts, the sciences, of industry, and in manners. Montesquieu andVoltaire personified these tendencies. The _Essai sur les moeurs_ isthe first sketch, and, in some respects, the masterpiece of history thusconceived. The detailed narration of political and military events wasstill regarded as the main work of history, but to this it now becamecustomary to add, generally by way of supplement or appendix, a sketchof the "progress of the human mind. " The expression "history ofcivilisation" appears before the end of the eighteenth century. At thesame time German university professors, especially at Göttingen, werecreating, in order to supply educational needs, the new form of thehistorical "manual, " a methodical collection of carefully justifiedfacts, with no literary or other pretensions. Collections of historicalfacts, made with a view to aid in the interpretation of literary texts, or out of mere curiosity in regard to the things of the past, hadexisted from ancient times; but the medleys of Athenæus and AulusGellius, or the vaster and better arranged compilations of the MiddleAges and the Renaissance, are by no means to be compared with the"scientific manuals" of which the German professors then gave themodels. These professors, moreover, contributed towards the clearing upof the vague, general notion which the philosophers had of"civilisation, " for they applied themselves to the organisation of thehistory of languages, of literatures, of the arts, of religions, of law, of economic phenomena, and so on, as so many separate branches of study. Thus the domain of history was greatly enlarged, and scientific, thatis, simple and objective, exposition began to compete with therhetorical or sententious, patriotic or philosophical ideals ofantiquity. 

This competition was at first timid and obscure, for the beginning ofthe nineteenth century was marked by a literary renaissance whichrenovated historical literature. Under the influence of the romanticmovement historians sought for more vivid methods of exposition thanthose employed by their predecessors, methods better adapted to strikethe imagination and rouse the emotions of the public, by filling themind with poetical images of vanished realities. Some endeavoured topreserve the peculiar colouring of the original documents, which theyadapted: "Charmed with the contemporary narratives, " says Barante, "Ihave endeavoured to write a consecutive account which should borrow fromthem their animation and interest. " This leads directly to the neglectof criticism, and to the reproduction of whatever is effective from theliterary point of view. Others declared that the facts of the past oughtto be recounted with all the emotions of a spectator. "Thierry, " saysMichelet, praising him, "in telling us the story of Klodowig, breathesthe spirit and shows the emotion of recently invaded France. .. . "Michelet "stated the problem of history as the resuscitation of integrallife in the inmost parts of the organism. " With the romantic historiansthe choice of subject, of plan, of the proofs, of the style, isdominated by an engrossing desire to produce an effect--a literary, nota scientific ambition. Some romantic historians have slid down thisinclined plane to the level of the "historical novel. " We know thenature of this species of literature, which flourished so vigorouslyfrom the Abbé Barthélemy and Chateaubriand down to Mérimée and Ebers, and which some are now vainly attempting to rejuvenate. The object is to"make the scenes of the past live again" in dramatic picturesartistically constructed with "true" colours and details. The obviousobject of the method is that it does not provide the reader with anymeans of distinguishing between the elements borrowed from thedocuments and the imaginary elements, not to mention the fact thatgenerally the documents used are not all of the same origin, so thatwhile the colour of each stone may be "true" that of the mosaic isfalse. Dezobry's _Rome au siècle d'Auguste_, Augustin Thierry's _Récitsmérovingiens_, and other "pictures" produced at the same epoch wereconstructed on the same principle, and are subject to the same drawbacksas the historical novels properly so-called. [218]

We may summarise what precedes by saying that, up to about 1850, historycontinued to be, both for historians and the public, a branch ofliterature. An excellent proof of this lies in the fact that up tillthen historians were accustomed to publish new editions of their works, at intervals of several years, without making any change in them, andthat the public tolerated the practice. Now every scientific work needsto be continually recast, revised, brought up to date. Scientificworkers do not claim to give their works an immutable form, they do notexpect to be read by posterity or to achieve personal immortality; it isenough for them if the results of their researches, corrected, it maybe, and possibly transformed by subsequent researches, should beincorporated in the fund of knowledge which forms the scientificheritage of mankind. No one reads Newton or Lavoisier; it is enough fortheir glory that their labours should have contributed to theproduction of works by which their own have been superseded, and whichwill be, sooner or later, superseded in their turn. It is only works ofart that enjoy perpetual youth. And the public is well aware of thefact; no one would ever think of studying natural history in Buffon, whatever his opinion might be of the merits of this stylist. But thesame public is quite ready to study history in Augustin Thierry, inMacaulay, in Carlyle, in Michelet, and the books of the great writerswho have treated historical subjects are reprinted, fifty years afterthe author's death, in their original form, though they are manifestlyno longer on a level with current knowledge. It is clear that, for many, form counts before matter in history, and that an historical work isprimarily, if not exclusively, a work of art. [219]

II. It is within the last fifty years that the scientific forms ofhistorical exposition have been evolved and settled, in accordance withthe general principle that the aim of history is not to please, nor togive practical maxims of conduct, nor to arouse the emotions, butknowledge pure and simple. 

We begin by distinguishing between (1) monographs and (2) works of ageneral character. 

(1) A man writes a monograph when he proposes to elucidate a specialpoint, a single fact, or a limited body of facts, for example the wholeor a portion of the life of an individual, a single event or a series ofevents between two dates lying near together. The types of possiblesubjects of a monograph cannot be enumerated, for the subject-matter ofhistory can be divided indefinitely, and in an infinite number of ways. But all modes of division are not equally judicious, and, though thereverse has been maintained, there are, in history as in all thesciences, subjects which it would be stupid to treat in monographs, andmonographs which, though well executed, represent so much uselesslabour. [220] Persons of moderate ability and no great mental range, devoted to what is called "curious" learning, are very ready to occupythemselves with insignificant questions;[221] indeed, for the purpose ofmaking a first estimate of an historian's intellectual power, a fairlygood criterion may be had in the list of the monographs he haswritten. [222] It is the gift of seeing the important problems, and thetaste for their treatment, as well as the power of solving them, which, in all the sciences, raise men to the first rank. But let us suppose thesubject has been rationally chosen. Every monograph, in order to beuseful--that is, capable of being fully turned to account--shouldconform to three rules: (1) in a monograph every historical fact derivedfrom documents should only be presented accompanied by a reference tothe documents from which it is taken, and an estimate of the value ofthese documents;[223] (2) chronological order should be followed as faras possible, because this is the order in which we know that the factsoccurred, and by which we are guided in searching for causes andeffects; (3) the title of the monograph must enable its subject to beknown with exactitude: we cannot protest too strongly against thoseincomplete or fancy titles which so unnecessarily complicatebibliographical searches. A fourth rule has been laid down; it has beensaid "a monograph is useful only when it exhausts the subject"; but itis quite legitimate to do temporary work with documents which one has atone's disposal, even when there is reason to believe that others exist, provided always that precise notice is given as to what documents havebeen employed. 

Any one who has tact will see that, in a monograph, the apparatus ofdemonstration, while needing to be complete, ought to be reduced to whatis strictly necessary. Sobriety is imperative; all parading of eruditionwhich might have been spared without inconvenience is odious. [224] Inhistory it often happens that the best executed monographs furnish noother result than the proof that knowledge is impossible. It isnecessary to resist the desire which leads some to round off withsubjective, ambitious, and vague conclusions monographs which will notbear them. [225] The proper conclusion of a good monograph is thebalance-sheet of the results obtained by it and the points leftdoubtful. A monograph made on these principles may grow antiquated, butit will not fall to pieces, and its author will never need to blush forit. 

(2) Works of a general character are addressed either to students or tothe general public. 

_A. _ General works intended principally for students and specialists nowappear in the form of "repertories, " "manuals, " and "scientifichistories. " In a repertory a number of verified facts belonging to agiven class are collected and arranged in an order which makes it easyto refer to them. If the facts thus collected have precise dates, chronological order is adopted: thus the task has been undertaken ofcompiling "Annals" of German history, in which the summary entry of theevents, arranged by dates, is accompanied by the texts from which theevents are known, with accurate references to the sources and the worksof critics; the collection of the _Jahrbücher der deutschen Geschichte_has for its object the elucidation, as far as is possible, of the factsof German history, including all that is susceptible of scientificdiscussion and proof, but omitting all that belongs to the domain ofappreciation and general views. When the facts are badly dated, or aresimultaneous, alphabetical arrangement must be employed; thus we haveDictionaries: dictionaries of institutions, biographical dictionaries, historical encyclopædias, such as the _Realencyclopædie_ ofPauly-Wissowa. These alphabetical repertories are, in theory, just asthe _Jahrbücher_, collections of proved facts; if, in practice, thereferences in them are less rigorous, if the apparatus of textssupporting the statements is less complete, the difference is withoutjustification. [226] _Scientific manuals_ are also, properly speaking, repertories, since they are collections in which established facts arearranged in systematic order, and are exhibited objectively, with theirproofs, and without any literary adornment. The authors of these"manuals, " of which the most numerous and the most perfect specimenshave been composed in our days in the German universities, have noobject in view except to draw up minute inventories of the acquisitionsmade by knowledge, in order that workers may be enabled to assimilatethe results of criticism with greater ease and rapidity, and may befurnished with starting-points for new researches. Manuals of this kindnow exist for most of the special branches of the history ofcivilisation (languages, literature, religion, law, _Alterthümer_, andso on), for the history of institutions, for the different parts ofecclesiastical history. It will suffice to mention the names ofSchoemann, of Marquardt and Mommsen, of Gilbert, of Krumbacher, ofHarnack, of Möller. These works are not marked by the dryness of themajority of the primitive "manuals, " which were published in Germany ahundred years ago, and which were little more than tables of subjects, with references to the books and documents to be consulted; in themodern type the exposition and discussion are no doubt terse andcompact, but yet not abbreviated beyond a point at which they may betolerated, even preferred by cultivated readers. They take away thetaste for other books, as G. Paris very well says:[227] "When one hasfeasted on these substantial pages, so full of facts, which, with alltheir appearance of impersonality, yet contain, and above all suggest, so many thoughts, it is difficult to read books, even books ofdistinction, in which the subject is cut up symmetrically to fit in witha preconceived system, is coloured by fancy, and is, so to speak, presented to us in disguise, books in which the author continually comesbetween us and the spectacle which he claims to make intelligible to us, but which he never allows us to see. " The great historical "manuals, "uniform with the treatises and manuals of the other sciences (with theadded complication of authorities and proofs), ought to be, and are, continually improved, emended, corrected, brought up to date: they are, by definition, works of science and not of art. 

The earliest repertories and the earliest scientific "manuals" werecomposed by isolated individuals. But it was soon recognised that asingle man cannot correctly arrange, or have the proper mastery over avast collection of facts. The task has been divided. Repertories areexecuted, in our days, by collaborators in association (who aresometimes of different nationalities and write in different languages). The great manuals (of I. Von Müller, of G. Gröber, of H. Paul, andothers) are collections of special treatises each written by aspecialist. The principle of collaboration is excellent, but oncondition (1) that the collective work is of a nature to be resolvedinto great independent, though co-ordinated, monographs; (2) that thesection entrusted to each collaborator has a certain extent; if thenumber of collaborators is too great and the part of each too limited, the liberty and the responsibility of each are diminished or disappear. 

_Histories_, intended to give a narrative of events which happened butonce, and to state the general facts which dominate the whole course ofspecial evolutions, still have a reason for existence, even after themultiplication of methodical manuals. But scientific methods ofexposition have been introduced into them, as into monographs andmanuals, and that by imitation. The reform has consisted, in every case, in the renunciation of literary ornaments and of statements withoutproof. Grote produced the first model of a "history" thus defined. Atthe same time certain forms which once had a vogue have now fallen intodisuse: this is the case with the "Universal Histories" with continuousnarrative, which were so much liked, for different reasons, in theMiddle Ages and in the eighteenth century; in the present centurySchlosser and Weber in Germany, Cantù in Italy, have produced the lastspecimens of them. This type has been abandoned for historical reasons, because we have ceased to regard humanity as a whole, bound together bya single evolution; and for practical reasons, because we haverecognised the impossibility of collecting so overwhelming a mass offacts in a single work. The Universal Histories which are stillpublished in collaboration (the Oncken collection is the best type ofthem), are, like the great manuals, composed of independent sections, each treated by a different author; they are publishers' combinations. Historians have in our days been led to adopt the division by states(national histories) and by epochs. [228]

_B. _ There is in theory no reason why historical works intendedprincipally for the public should not be conceived in the same spirit asworks designed for students and specialists, nor why they should not becomposed in the same manner, apart from simplifications and omissionswhich readily suggest themselves. And, in fact, there are in existencesuccinct, substantial, and readable summaries, in which no statement isadvanced which is not tacitly supported by solid references, in whichthe acquisitions of science are precisely stated, judiciouslyexplained, their significance and value clearly brought out. The French, thanks to their natural gifts of tact, dexterity, and accuracy of mind, excel, as a rule, in this department. There have been published in ourcountry review-articles and works of higher popularisation in which theresults of a number of original works have been cleverly condensed, in away that has won the admiration of the very specialists who, by theirheavy monographs, have rendered these works possible. Nothing, however, is more dangerous than popularisation. As a matter of fact, most worksof popularisation do not conform to the modern ideal of historicalexposition; we frequently find in them survivals of the ancient ideal, that of antiquity, the Renaissance, and the romantic school. 

The explanation is easy. The defects of the historical works designedfor the general public--defects which are sometimes enormous, and have, with many able minds, discredited popular works as a class--are theconsequences of the insufficient preparation or of the inferior literaryeducation of the "popularisers. "

A populariser is excused from original research; but he ought to knoweverything of importance that has been published on his subject, heought to be up to date, and to have thought out for himself theconclusions reached by the specialists. If he has not personally made aspecial study of the subject he proposes to treat, he must obviouslyread it up, and the task is long. For the professional populariser thereis a strong temptation to study superficially a few recent monographs, to hastily string together or combine extracts from them, and, in orderto render this medley more attractive, to deck it out, as far as ispossible, with "general ideas" and external graces. The temptation isall the stronger from the circumstance that most specialists take nointerest in works of popularisation, that these works are, in general, lucrative, and that the public at large is not in a position todistinguish clearly between honest and sham popularisation. In short, there are some, absurd as it may seem, who do not hesitate to summarisefor others what they have not taken the trouble to learn for themselves, and to teach that of which they are ignorant. Hence, in most works ofhistorical popularisation, there inevitably appear blemishes of everykind, which the well-informed always note with pleasure, but with apleasure in which there is some touch of bitterness, because they alonecan see these faults: unacknowledged borrowings, inexact references, mutilated names and texts, second-hand quotations, worthless hypotheses, imprudent assertions, puerile generalisations, and, in the enunciationof the most false or the most debatable opinions, an air of tranquilauthority. [229]

On the other hand, men whose information is all that could be desired, whose monographs intended for specialists are full of merit, sometimesshow themselves capable, when they write for the public, of graveoffences against scientific method. The Germans are habitual offenders:consider Mommsen, Droysen, Curtius, and Lamprecht. The reason is thatthese authors, when they address the public, wish to produce an effectupon it. Their desire to make a strong impression leads them to acertain relaxation of scientific rigour, and to the old rejected habitsof ancient historiography. These men, scrupulous and minute as they arewhen they are engaged in establishing details, abandon themselves, intheir exposition of general questions, to their natural impulses, likethe common run of men. They take sides, they censure, they extol; theycolour, they embellish; they allow themselves to be influenced bypersonal, patriotic, moral, or metaphysical considerations. And, overand above all this, they apply themselves, with their several degrees oftalent, to the task of producing works of art; in this endeavour thosewho have no talent make themselves ridiculous, and the talent of thosewho have any is spoilt by their preoccupation with the effect they wishto produce. 

Not, let it be well understood, that "form" is of no importance, orthat, provided he makes himself intelligible, the historian has a rightto employ incorrect, vulgar, slovenly, or clumsy language. A contemptfor rhetoric, for paste diamonds and paper flowers, does not exclude ataste for a pure and strong, a terse and pregnant style. Fustel deCoulanges was a good writer, although throughout his life he recommendedand practised the avoidance of metaphor. On the contrary we see no harmin repeating[230] that the historian, considering the extreme complexityof the phenomena he undertakes to describe, is under an obligation notto write badly. But he should write _consistently_ well, and neverbedeck himself with finery. 

CONCLUSION

I. History is only the utilisation of documents. But it is a matter ofchance whether documents are preserved or lost. Hence the predominantpart played by chance in the formation of history. 

The quantity of documents in existence, if not of known documents, isgiven; time, in spite of all the precautions which are taken nowadays, is continually diminishing it; it will never increase. History has atits disposal a limited stock of documents; this very circumstance limitsthe possible progress of historical science. When all the documents areknown, and have gone through the operations which fit them for use, thework of critical scholarship will be finished. In the case of someancient periods, for which documents are rare, we can now see that in ageneration or two it will be time to stop. Historians will then beobliged to take refuge more and more in modern periods. Thus historywill not fulfil the dream which, in the nineteenth century, inspired theromantic school with so much enthusiasm for the study of history: itwill not penetrate the mystery of the origin of societies; and, for wantof documents, the beginnings of the evolution of humanity will alwaysremain obscure. 

The historian does not collect by his own observation the materialsnecessary for history as is done in the other sciences: he works onfacts the knowledge of which has been transmitted by former observers. In history knowledge is not obtained, as in the other sciences, bydirect methods, it is indirect. History is not, as has been said, ascience of observation, but a science of reasoning. 

In order to use facts which have been observed under unknown conditions, it is necessary to apply criticism to them, and criticism consists in aseries of reasonings by analogy. The facts as furnished by criticism areisolated and scattered; in order to organise them into a structure it isnecessary to imagine and group them in accordance with theirresemblances to facts of the present day, an operation which alsodepends on the use of analogies. This necessity compels history to usean exceptional method. In order to frame its arguments from analogy, itmust always combine the knowledge of the particular conditions underwhich the facts of the past occurred with an understanding of thegeneral conditions under which the facts of humanity occur. Its methodis to draw up special _tables_ of the facts of an epoch in the past, andto apply to them sets of _questions_ founded on the study of thepresent. 

The operations which must necessarily be performed in order to pass fromthe inspection of documents to the knowledge of the facts and evolutionsof the past are very numerous. Hence the necessity of the division andorganisation of labour in history. It is requisite, on the one hand, that those specialists who occupy themselves with the search fordocuments, their restoration and preliminary classification, shouldco-ordinate their efforts, in order that the preparatory work ofcritical scholarship may be finished as soon as possible, under the bestconditions as to accuracy and economy of labour. On the other hand, authors of partial syntheses (monographs) designed to serve as materialsfor more comprehensive syntheses ought to agree among themselves to workon a common method, in order that the results of each may be used by theothers without preliminary investigations. Lastly, workers of experienceshould be found to renounce personal research and devote their wholetime to the study of these partial syntheses, in order to combine themscientifically in comprehensive works of historical construction. And ifthe result of these labours were to bring out clear and certainconclusions as to the nature and the causes of social evolution, a trulyscientific "philosophy of history" would have been created, whichhistorians might acknowledge as legitimately crowning historicalscience. 

Conceivably a day may come when, thanks to the organisation of labour, all existing documents will have been discovered, emended, arranged, andall the facts established of which the traces have not been destroyed. When that day comes, history will be established, but it will not befixed: it will continue to be gradually modified in proportion as thedirect study of existing societies becomes more scientific and permits abetter understanding of social phenomena and their evolution; for thenew ideas which will doubtless be acquired on the nature, the causes, and the relative importance of social facts will continue to transformthe ideas which will be formed of the societies and events of thepast. [231]

II. It is an obsolete illusion to suppose that history suppliesinformation of practical utility in the conduct of life (_Historiamagistra vitæ_), lessons directly profitable to individuals and peoples;the conditions under which human actions are performed are rarelysufficiently similar at two different moments for the "lessons ofhistory" to be directly applicable. But it is an error to say, by way ofreaction, that "the distinguishing feature of history is to be good fornothing. "[232] It has an indirect utility. 

History enables us to understand the present in so far as it explainsthe origin of the existing state of things. Here we must admit thathistory does not offer an equal interest through the whole extent oftime which it covers; there are remote generations whose traces are nolonger visible in the world as it now is; for the purpose of explainingthe political constitution of contemporary England, for example, thestudy of the Anglo-Saxon witangemot is without value, that of the eventsof the eighteenth and nineteenth centuries is all-important. Theevolution of the civilised societies has within the last hundred yearsbeen accelerated to such a degree that, for the understanding of theirpresent form, the history of these hundred years is more important thanthat of the ten preceding centuries. As an explanation of the present, history would almost reduce to the study of the contemporary period. 

History is also indispensable for the completion of the political andsocial sciences, which are still in process of formation; for the directobservation of social phenomena (in a state of rest) is not a sufficientfoundation for these sciences--there must be added a study of thedevelopment of these phenomena in time, that is, their history. [233]This is why all the sciences which deal with man (linguistic, law, science of religions, political economy, and so on) have in this centuryassumed the form of historical sciences. 

But the chief merit of history is that of being an instrument ofintellectual culture; it is so in several ways. Firstly, the practiceof the historical method of investigation, of which the principles havebeen sketched in the present volume, is very hygienic for the mind, which it cures of credulity. Secondly, history, by exhibiting to us agreat number of differing societies, prepares us to understand andtolerate a variety of usages; by showing us that societies have oftenbeen transformed, it familiarises us with variation in social forms, andcures us of a morbid dread of change. Lastly, the contemplation of pastevolutions, which enables us to understand how the transformations ofhumanity are brought about by changes of habits and the renewal ofgenerations, saves us from the temptation of applying biologicalanalogies (selection, struggle for existence, inherited habits, and soon) to the explanation of social evolution, which is not produced by theoperation of the same causes as animal evolution. 

APPENDICES

APPENDIX I

THE SECONDARY TEACHING OF HISTORY IN FRANCE

I. The teaching of history is a recent addition to secondary education. Formerly history was taught to the sons of kings and great persons, inorder to give them a preparation in the art of governing, according tothe ancient tradition, but it was a sacred science reserved for thefuture rulers of states, a science for princes, not for subjects. Thesecondary schools which have been organised since the sixteenth century, ecclesiastical or secular, Catholic or Protestant, did not admit historyinto their plan of study, or only admitted it as an appendage to thestudy of the ancient languages. This was the tradition of the Jesuits inFrance; it was adopted by the University of Napoleon. 

History was only introduced into secondary education in the nineteenthcentury, under the pressure of public opinion; and although it has beenallotted more space in France than in England, or even in Germany, ithas continued to be a subsidiary subject, not taught in a special class(as philosophy is), nor always by a special professor, and counting forvery little in examinations. 

Historical instruction has for a long time felt the effects of themanner in which it was introduced. The subject was imposed by theauthorities on teachers trained exclusively in the study of literature, and could find no suitable place in a system of classical educationbased on the study of forms, and indifferent to the knowledge of socialphenomena. History was taught because it was prescribed by theprogramme; but this programme, the sole motive and guide of theinstruction, was always an accident, and varied with the preferences, oreven the personal studies of those who framed it. History formed part ofthe social conventions; there are, it was said, names and facts "ofwhich it is not permissible to be ignorant"; but the things of whichignorance was not permitted varied greatly, from the names of theMerovingian kings and the battles of the Seven Years' War to the SalicLaw and the work of Saint Vincent de Paul. 

The improvised staffs which, in order to carry out the programme, had tofurnish impromptu instruction in history, had no clear idea either ofthe reasons for such instruction, or of its place in general education, or of the technical methods necessary for giving it. With this lack oftradition, of pedagogic preparation, and even of mechanical aids, theprofessor of history found himself carried back to the ages beforeprinting, when the teacher had to supply the pupil with all the factswhich formed the subject-matter of instruction, and he adopted themediæval procedure. Armed with a note-book in which he had written downthe list of facts to be taught, he read it out to the pupils, sometimesmaking a pretence of extemporising; this was the "lesson, " thecorner-stone of historical instruction. The whole series of lessons, determined by the programme, formed the "course. " The pupil was expectedto write as he listened (this was called "taking notes") and to composea written account of what he had heard (this was the _rédaction_). Butas the pupils were not taught how to take notes, nearly all of them werecontent to write very rapidly, from the professor's dictation, a roughdraft, which they copied out at home in the form of a _rédaction_, without any endeavour to grasp the meaning either of what they heard orwhat they transcribed. To this mechanical labour the most zealous addedextracts copied from books, generally with just as little reflection. 

In order to get the facts judged essential into the pupils' heads, theprofessor used to make a very short version of the lesson, the "summary"or "abstract, " which he dictated openly, and caused to be learnt byheart. Thus of the two written exercises which occupied nearly the wholetime of the class, one (the summary) was an overt dictation, the other(the _rédaction_) an unavowed dictation. 

The only means adopted to check the pupils' work was to make them repeatthe summary word for word, and to question them on the _rédaction_, thatis to make them repeat approximately the words of the professor. Of thetwo oral exercises one was an overt, the other an unavowed repetition. 

It is true the pupil was given a book, the _Précis d'histoire_, [234] butthis book had the same form as the professor's course, and instead ofserving as a basis for the oral instruction, merely duplicated it, and, as a rule, duplicated it badly, for it was not intelligible to thepupil. The authors of these text-books, [235] adopting the traditionalmethods of "abridgments, " endeavoured to accumulate the greatestpossible number of facts by omitting all their characteristic detailsand summarising them in the most general, and therefore vague, expressions. In the elementary books nothing was left but a residue ofproper names and dates connected by formulæ of a uniform type; historyappeared as a series of wars, treaties, reforms, revolutions, which onlydiffered in the names of peoples, sovereigns, fields of battle, and inthe figures giving the years. [236]

Such, down to the end of the Second Empire, was historical instructionin all French institutions, both secular and ecclesiastical--with a fewexceptions, whose merit is measured by their rarity, for in those days aprofessor of history needed a more than common share of energy andinitiative to rise above the routine of _rédaction_ and summary. 

II. In recent times the general movement of educational reform, whichbegan in the Department and the Faculties, has at last extended tosecondary instruction. The professors of history have been emancipatedfrom the jealous supervision which weighed on their teaching under thegovernment of the Empire, and have taken the opportunity to make trialof new methods. A system of historical pedagogy has been devised. It hasbeen revealed with the approbation of the Department in the discussionsof the society for the study of questions of secondary education, in the_Revue de l'enseignement secondaire_, and in the _Revue universitaire_. It has received official sanction in the _Instructions_ appended to theprogramme of 1890; the report on history, the work of M. Lavisse, hasbecome the charter which protects the professors who favour reform intheir struggle against tradition. [237]

Historical instruction will no doubt issue from this crisis ofrenovation organised and provided with a rational pedagogic andtechnical system, such as is possessed by the older branches ofinstruction in languages, literature, and philosophy. But it is only tobe expected that the reform should be much slower than in the case ofthe higher instruction. The _personnel_ is much more numerous, and takeslonger to train or to renew; the pupils are less zealous and lessintelligent; the routine of the parents opposes to the new methods aforce of inertia which is unknown to the Faculties; and theBaccalaureate, that general obstacle to all reform, is particularlymischievous in its effect on historical instruction, which it reduces toa set of questions and answers. 

III. It is now possible, however, to indicate what is the direction inwhich historical instruction is likely to develop in France[238] and thequestions which will need to be solved for the purpose of introducing arational technical system. Here we shall endeavour to formulate thesequestions in a methodical table. 

(1) _General Organisation. _--What object should historical instructionaim at? What services can it render to the culture of the pupil? Whatinfluence can it have upon his conduct? What facts ought it to enablehim to understand? And, consequently, what principles ought to guide thechoice of subjects and methods? Ought the instruction to be spread overthe whole duration of the classes, or should it be concentrated in aspecial class? Should it be given in one-hour or two-hour classes?Should history be distributed into several cycles, as in Germany, so asto cause the pupil to return several times to the same subject atdifferent periods of his studies? Or should it be expounded in a singlecontinuous course, beginning with the commencement of study, as inFrance? Should the professor give a complete course, or should he selecta few questions and leave the pupil to study the others by himself?Should he expound the facts orally, or should he require the pupils tolearn them in the first instance from a book, so as to make the course aseries of explanations?

(2) _Choice of Subjects. _--What proportion should be observed betweenhome and foreign history? between ancient and contemporary history?between the special branches of history (art, religion, customs, economics) and general history? between institutions or usages, andevents? between the evolution of material usages, intellectual history, social life, political life? between the study of particular incidents, of biography, of dramatic episodes, and the study of the interconnectionof events and general evolutions? What place should be assigned toproper names and dates? Should we profit by the opportunities affordedby legends to arouse the critical spirit? or should we avoid legends?

(3) _Order. _--In what order should the subjects be attacked? Shouldinstruction begin with the most ancient periods and the countries withthe most ancient civilisations in order to follow chronological orderand the order of evolution? or should it begin with the periods and thecountries which are nearest to us so as to proceed from the better knownto the less known? In the exposition of each period, should thechronological, geographical, or logical order be followed? Should theteacher begin by describing conditions or by narrating events?

(4) _Methods of Instruction. _--Should the pupil be given general formulæfirst or particular images? Should the professor state the formulæhimself or require the pupil to search for them? Should formulæ belearnt by heart? In what cases? How are images of historical facts to beproduced in the pupils' minds? What use is to be made of engravings? ofreproductions and restorations? of imaginary scenes? What use is to bemade of narratives and descriptions? of authors' texts? of historicalnovels? To what extent ought words and formulæ to be quoted? How arefacts to be localised? What use is to be made of chronological tables?of synchronical tables? of geographical sketches? of statistical andgraphic tables? What is the way to make comprehensible the character ofevents and customs? the motives of actions? the conditions of customs?How are the episodes of an event to be chosen? and the examples of acustom? How is the interconnection of facts and the process of evolutionto be made intelligible? What use is to be made of comparison? Whatstyle of language is to be employed? To what extent should concrete, abstract, and technical terms be used? How is it to be verified that thepupil has understood the terms and assimilated the facts? Can exercisesbe organised in which the pupil may do original work on the facts? Whatinstruments of study should the pupil have? How should school-books becompiled, with a view to giving the pupil practice in original work?

For the purpose of stating and justifying the solutions of all theseproblems, a special treatise would not be too much. [239] Here we shallmerely indicate the general principles on which a tolerable agreementseems to have been now reached in France. 

We no longer go to history for lessons in morals, nor for good examplesof conduct, nor yet for dramatic or picturesque scenes. We understandthat for all these purposes legend would be preferable to history, forit presents a chain of causes and effects more in accordance with ourideas of justice, more perfect and heroic characters, finer and moreaffecting scenes. Nor do we seek to use history, as is done in Germany, for the purpose of promoting patriotism and loyalty; we feel that itwould be illogical for different persons to draw opposite conclusionsfrom the same science according to their country or party; it would bean invitation to every people to mutilate, if not to alter, history inthe direction of its preferences. We understand that the value of everyscience consists in its being true, and we ask from history truth andnothing more. [240]

The function of history in education is perhaps not yet clearly apparentto all those who teach it. But all those who reflect are agreed toregard it as being principally an instrument of social culture. Thestudy of the societies of the past causes the pupil to understand, bythe help of actual instances, what a society is; it familiarises himwith the principal social phenomena and the different species of usages, their variety and their resemblances. The study of events and evolutionsfamiliarises him with the idea of the continual transformation whichhuman affairs undergo, it secures him against an unreasoning dread ofsocial changes; it rectifies his notion of progress. All theseacquisitions render the pupil fitter for public life; history thusappears as an indispensable branch of instruction in a democraticsociety. 

The guiding principle of historical pedagogy will therefore be to seekfor those subjects and those methods which are best calculated toexhibit social phenomena and give an understanding of their evolution. Before admitting a fact into the plan of instruction, it should be askedfirst of all what educational influence it can exercise; secondly, whether there are adequate means of bringing the pupil to see andunderstand it. Every fact should be discarded which is instructive onlyin a low degree, or which is too complicated to be understood, or inregard to which we do not possess details enough to make itintelligible. 

IV. To make rational instruction a reality it is not enough to develop atheory of historical pedagogy. It is necessary to renew the materialaids and the methods. 

History necessarily involves the knowledge of a great number of facts. The professor of history, with no resources but his voice, a blackboard, and abridgments which are little better than chronological tables, is inmuch the same situation as a professor of Latin without texts ordictionary. The pupil in history needs a repertory of historical factsas the Latin pupil needs a repertory of Latin words; he needscollections of _facts_, and the school text-books are mostly collectionsof _words_. 

There are two vehicles of facts, engravings and books. Engravingsexhibit material objects and external aspects, they are usefulprincipally for the study of material civilisation. It is some timesince the attempt was first made in Germany to put in the hands of thepupil a collection of engravings arranged for the purposes of historicalinstruction. The same need has, in France, produced the _Albumhistorique_, which is published under the direction of M. Lavisse. 

The book is the chief instrument. It ought to contain all thecharacteristic features necessary for forming mental representations ofthe events, the motives, the habits, the institutions studied; it willconsist principally in narratives and descriptions, to whichcharacteristic sayings and formulæ may be appended. For a long time itwas endeavoured to construct those books out of extracts selected fromancient authors; they were compiled in the form of collections oftexts. [241] Experience seems to indicate that this method must beabandoned; it has a scientific appearance, it is true, but is notintelligible to children. It is better to address pupils in contemporarylanguage. It is in this spirit that, pursuant to the _Instructions_ of1890, [242] collections of _Historical Readings_ have been compiled, ofwhich the most important has been published by the firm of Hachette. 

The pupils' methods of work still bear witness to the late introductionof historical teaching. In most historical classes methods still prevailwhich only exercise the pupils' receptivity: the course of lectures, thesummary, reading, questioning, the _rédaction_, the reproduction ofmaps. It is as if a Latin pupil were to confine himself to repeatinggrammar-lessons and extracts from authors, without ever doingtranslation or composition. 

In order that the teaching may make an adequate impression, it isnecessary, if not to discard all these passive methods, at least tosupplement them by exercises which call out the activity of the pupil. Some such exercises have already been experimented with, and othersmight be devised. [243] The pupil may be set to analyse engravings, narratives, and descriptions in such a way as to bring out the characterof the facts: the short written or oral analysis will guarantee that hehas seen and understood, it will be an opportunity to inculcate thehabit of using only precise terms. Or the pupil may be asked to furnisha drawing, a geographical sketch, a synchronical table. He may berequired to draw up tables of comparison between different societies, and tables showing the interconnection of facts. 

A book is needed to supply the pupil with the materials for theseexercises. Thus the reform of methods is connected with the reform ofthe instruments of work. Both reforms will progress according as theprofessors and the public perceive more clearly the part played byhistorical instruction in social education. 

APPENDIX II

THE HIGHER TEACHING OF HISTORY IN FRANCE

The higher teaching of history has been in a great measure transformed, in our country, within the last thirty years. The process has beengradual, as it ought to have been, and has consisted in a succession ofslight modifications. But although a rational continuity has beenobserved in the steps taken, the great number of these steps has notfailed, in these last days, to astonish, and even to offend, the public. Public opinion, to which appeal has been made in favour of reforms, hasbeen somewhat surprised by being appealed to so often, and perhaps it isnot superfluous to indicate here, once more, the general significanceand the inner logic of the movement which we are witnessing. 

I. Before the last years of the Second Empire, the higher teaching ofthe historical sciences was organised in France on no coherentsystem. [244]

There were chairs of history in different institutions, of differenttypes: at the Collège de France, in the Faculties of Letters, and in the"special schools, " such as the École normale supérieure and the Écoledes chartes. 

The Collège de France was a relic of the institutions of the _ancienrégime_. It was founded in the sixteenth century in opposition to thescholastic Sorbonne, to be a refuge for the new sciences, and had theglorious privilege of representing historically the higher speculativestudies, the spirit of free inquiry, and the interests of pure science. Unfortunately, in the domain of the historical sciences, the Collège deFrance had allowed its traditions to be obliterated up to a certainpoint. The great men who taught history in this illustrious institution(J. Michelet, for example), were not technical experts, nor even men oflearning, in the proper sense of the word. The audiences which theyswayed by their eloquence were not composed of students of history. 

The Faculties of Letters formed part of a system established by theNapoleonic legislator. This legislator, in creating the Faculties, by nomeans entertained the design of encouraging scientific research. He hadno great love for science. The Faculties of Law, of Medicine, and so on, were intended by him to be professional schools supplying society withthe lawyers, physicians, and so on, which it needs. But three of thefive Faculties were unable, from the beginning, to perform the partallotted them, while the other two, Law and Medicine, successfullyperformed theirs. The Faculties of Catholic Theology did not train thepriests needed by society, because the State consented to the educationof the priests being conducted in the diocesan seminaries. The Facultiesof Sciences and of Letters did not train the professors for secondaryeducation, the engineers, and so on, needed by society, because theywere here met by the triumphant competition of "special schools"previously instituted: the École normale, the École polytechnique. TheFaculties of Catholic Theology, of Sciences, and of Letters weretherefore obliged to justify their existence by other modes of activity. In particular, the professors of history in the Faculties of Letterscould not undertake the instruction of the young men who were destinedto teach history in the _lycées_. Deprived of these special pupils, theyfound themselves in a situation analogous to that of those charged withhistorical instruction at the Collège de France. They too were not, as arule, technical experts. For half a century they carried on the work ofhigher popularisation in lectures delivered to large audiences ofleisured persons (since much abused), who were attracted by the force, the elegance, and the pleasing style of their diction. 

The function of training the future teachers for secondary education wasreserved for the École normale supérieure. Now at this epoch it was anadmitted principle that to be a good secondary teacher it is necessaryfor a man to know, and sufficient to know perfectly, the subject he ischarged to teach. The one is certainly necessary, but the other is notsufficient: knowledge of a different, of a higher, order is no lessindispensable than the regular "scholastic" equipment. At the Écolethere was never any question of such higher knowledge, but, inaccordance with the prevailing theory, preparation was made forsecondary teaching simply by imparting it. However, as the École normalehas always been excellently recruited, the system in vogue has notprevented it from numbering among its former pupils men of the firstorder, not only as professors, thinkers, or writers, but even ascritical scholars. But it must be recognised that they made their wayfor themselves, in spite of the system, not thanks to it, after, notduring, their pupilage, and principally when they had the advantage, during a stay at the French School at Athens, of the wholesome contactwith documents which they had not enjoyed at the Rue d'Ulm. "Does it notseem strange, " it has been said, "that so many generations of professorsshould have been turned out by the École normale incapable of utilisingdocuments?. .. Formerly, in short, students of history, on leaving theÉcole, were not prepared either to teach history, which they had learnedin a great hurry, or to investigate difficult questions. "[245]

As for the École des chartes, which was founded under the Restoration, it was, from a certain point of view, a special school like the others, designed in theory to train those useful functionaries, archivists andlibrarians. But professional instruction was early reduced to a strictminimum, and the École des chartes was organised on a very originalplan, with a view to provide a rational and complete apprenticeship forthe young men who proposed to study mediæval French history. The pupilsof the École des chartes did not follow any course of "mediævalhistory, " but they learnt all that is necessary for doing work on thesolution of the still open questions of mediæval history. Here alone, invirtue of an accidental anomaly, the subjects which are preliminary andauxiliary to historical research were systematically taught. We havealready had occasion to note the effects of this circumstance. [246]

This was the state of affairs when, towards the end of the SecondEmpire, a vigorous reform movement set in. Some young Frenchmen hadvisited Germany; they had been struck by the superiority of the Germanuniversity system over the Napoleonic system of Faculties and specialschools. Certainly France, with its defective organisation, had producedmany men and many works, but it now began to be held that "in all kindsof enterprises the least possible part should be left to chance, " andthat "when an institution proposes to train professors of history andhistorians, it ought to supply them with the means of becoming what itintends them to be. "

M. V. Duruy, minister of Public Education, supported the partisans of arenaissance of the higher studies. But he did not think it practicableto interfere, for the purpose either of remodelling, of fusing, or ofsuppressing them, with the existing institutions, --the Collège deFrance, the Faculties of Letters, the École normale supérieure, theÉcole des chartes, all of which were consecrated by the services theyhad rendered, and by the lustre they received from the eminent men whohad been, or were, connected with them. He changed nothing, he added. Hecrowned the somewhat heterogeneous edifice of existing institutions bythe creation of an "École pratique des hautes études, " which wasestablished at the Sorbonne in 1868. 

The École pratique des hautes études (historical and philologicalsection) was intended by those who founded it to prepare young men forresearch of a scientific character. It was not meant to be subservientto the interests of the professions, and there was to be nopopularisation. Students were not to go there to learn the resultsobtained by science, but, for the same purpose which takes the chemicalstudent to the laboratory, to be initiated into the technical methods bywhich new results can be obtained. Thus the spirit of the newinstitution was not without some analogy to that of the primitivetradition of the Collège de France. It was endeavoured to do there, forall the branches of universal history and philology, what had long beendone at the École des chartes for the limited domain of French mediævalhistory. 

II. As long as the Faculties of Letters were satisfied to be as theywere (that is, without students), and as long as their ambition did notgo beyond their traditional functions (the holding of public lectures, the conferring of degrees), the organisation of the higher teaching ofthe historical sciences in France remained in the condition which wehave described. When the Faculties of Letters began to seek a newjustification for their existence and new functions, changes becameinevitable. 

This is not the place to explain why and how the Faculties of Letterswere led to desire to work more actively, or rather in other ways thanin the past, for the promotion of the historical sciences. M. V. Duruy, in inaugurating the École des hautes études at the Sorbonne, haddeclared that this young and vigorous plant would thrust asunder the oldstones; and, without a doubt, the spectacle of the fruitful activity ofthe École des hautes études has contributed not a little to awaken theconscience of the Faculties. On the other hand the liberality of thepublic authorities, which have increased the _personnel_ of theFaculties, which have built palaces for them, and liberally endowed themwith the materials required by their work, has imposed new duties onthese privileged institutions. 

It is about twenty-five years since the Faculties of Letters began totransform themselves, and during this period their progressivetransformation has occasioned changes in the whole fabric of the higherteaching of historical science in France, which up to that time hadremained unshaken, even by the ingenious addition of 1868. 

III. The first care of the Faculties was to provide themselves withstudents. This was not, to be sure, the main difficulty, for the Écolenormale supérieurs (in which twenty pupils are admitted every year, chosen from among hundreds of candidates) was no longer sufficient forthe recruiting of the now numerous body of professors engaged insecondary education. Many young men who had been candidates (along withthe pupils of the École normale supérieure) for the degrees which giveaccess to the scholastic profession, were thrown on their own resources. Here was an assured supply of students. At the same time the militarylaws, by attaching much-prized immunities to the title of _licencié èslettres_, were calculated to attract to the Faculties, if they preparedstudents for the licentiate, a large and very interesting class of youngmen. Lastly, the foreigners (so numerous at the École des hautesétudes), who come to France to complete their scientific education, andwho up to that time were surprised to have no opportunity of profitingby the Faculties, were sure to go to them as soon as they found theresomething analogous to what they had been accustomed to find in theGerman universities, and the kind of instruction they wanted. 

Before students in any great number could be taught the way to theFaculties, great efforts were necessary and several years passed; but itwas after the Faculties obtained the students they desired that the realproblems presented themselves for solution. 

The great majority of the students in the Faculties of Letters have beenoriginally candidates for degrees, for the licentiate, and for_agrégation_, who entered with the avowed intention of "preparing" forthe licentiate and for _agrégation_. The Faculties have not been able toescape the obligation of helping them in this "preparation. " But, twentyyears ago, examinations were still conceived in accordance with ancientformulæ. The licentiate was an attestation of advanced secondary study, a kind of "higher baccalaureate"; for the _agrégation_ in the classes ofhistory and geography (which became the real _licentia docendi_), thecandidates were required to show that they "had a very good knowledge ofthe subjects they would be charged to teach. " Henceforth there was adanger lest the teaching of the Faculties, which must, like that of theÉcole normale supérieure, be preparatory for the examinations for thelicentiate and for _agrégation_, should be compelled by the force ofcircumstances to assume the same character. Note that a certainemulation could not fail to arise between the pupils of the Écolenormale and those of the Faculties in the competitions for _agrégation_. The _agrégation_ programmes being what they were, this emulation seemedlikely to have the result of engaging the rival teachers and studentsmore and more in school work, not of a scientific kind, equally devoidof dignity and real utility. 

The danger was very serious. It was perceived from the first by thoseclear-sighted promoters of the reform of the Faculties, MM. A. Dumont, L. Liard, E. Lavisse. M. Lavisse wrote in 1884: "To maintain that theFaculties have for their chief object the preparation for examinationsis to substitute drill for scientific culture: this is the seriousgrievance which able men have against the partisans of innovation. .. . The partisans of innovation reply that they have seen the drawbacks ofthe new departure from the beginning, but that they are convinced that amodification of the examination-system will follow the reform of highereducation; that a reconciliation will be found between scientific workand the preparation for examinations; and that thus the only grievancetheir opponents have against them will fall to the ground. " It is onlydoing justice to the foremost champion of reform to acknowledge that hewas never tired of insisting on the weak point; and in order to convinceoneself that the _examination question_ has always been considered thekey-stone of the problem of the organisation of higher education inFrance, it is only necessary to look through the speeches and thearticles entitled "Education and Examinations, " "Examinations andStudy, " "Study and Examinations, " &c. , which M. Lavisse has collected inhis three volumes published at intervals of five years from 1885onwards: _Questions d'enseignement national, Études et étudiants, Apropos de nos écoles. _

Thus the question of the reform of the examinations connected withhigher education (licentiate, _agrégation_, doctorate) has been placedon the order of the day. It was then in 1884; it is still there in 1897. But, during the interval, visible progress has been made in thedirection which we consider the right one, and now a solution seemsnear. 

IV. The old examination-system required candidates for degrees to showthat they had received an excellent secondary education. As it condemnedthose candidates, students receiving higher instruction, to exercises ofthe same kind as those of which they had already had their fill in the_lycées_, it was a simple matter to attack it. It was defended feebly, and has been demolished. 

But how was it to be replaced? The problem was very complex. Is it anywonder that it was not solved at a stroke?

First of all, it was important to come to an agreement on thispreliminary question: What are the capacities and what is the knowledgestudents should be required to give proof of possessing? Generalknowledge? Technical knowledge and the capacity of doing originalresearch (as at the École des chartes and the École des hautes études)?Pedagogic capacity? It came gradually to be recognised that, consideringthe great extent and variety of the class from which the students aredrawn, it is necessary to draw distinctions. 

From candidates for the licentiate it is enough to require that theyshould give proof of good general culture, permitting them at the sametime, if they wish, to show that they have a taste for, and someexperience in, original research. 

From the candidates for _agrégation_ (_licentia docendi_) who havealready obtained the licentiate, there will be required (1) formal proofthat they know, by experience, what it is to study an historicalproblem, and that they have the technical knowledge necessary for suchstudies; (2) proof of pedagogic capacity, which is a professionalnecessity for this class. 

The students who are not candidates for anything, neither for thelicentiate nor for _agrégation_, and who are simply seeking to obtainscientific initiation--the old programmes did not contemplate theexistence of such a class of students--will merely be required to provethat they have profited by the tuition and the advice they havereceived. 

This settled, a great stride has been made. For programmes, as we know, regulate study. By virtue of the authority of the programmes historicalstudies in the Faculties will now have the threefold character which itis desirable that they should have. General culture will not cease to beheld in honour. Technical exercises in criticism and research will havetheir legitimate place. Lastly, pedagogy (theoretical and practical)will not be neglected. 

The difficulties begin when it is attempted to determine the testswhich, in each department, are the best, that is, the most conclusive. On this subject opinions differ. Though no one now contests theprinciples, the modes of application which have hitherto been tried orsuggested do not meet with unanimous approval. The organisation of thelicentiate has been revised three times; the statute relating to the_agrégation_ in history has been reformed or amended five times. Andthis is not the end. New simplifications are imperative. But what is theimportance of this instability--of which, however, complaints begin tobe heard[247]--if it is established, as we believe it is, that progresstowards a better state of things has been continuous through all thesechanges, without any notable retrogression?

There is no need to explain here in detail the different transitorysystems which have been put into practice. We have had occasion tocriticise them elsewhere. [248] Now that most of what we objected to hasbeen abolished, what is the use of reviving old controversies? We shallnot even mention the points in which the present system seems to us tobe still capable of improvement, for there is reason to hope that itwill soon be modified, and in a very satisfactory manner. Let it sufficeto say that the Faculties now confer a new diploma, the _Diplômed'études supérieures_, which all the students have a right to seek, butwhich the candidates for _agrégation_ are obliged to obtain. Thisdiploma of higher studies, analogous to that of the École des hautesétudes, the _brevet_ of the École des chartes, and the doctorate inphilosophy at the German universities, is given to those students ofhistory who, qualified by a certain academical standing, have passed anexamination in which the principal tests are, besides questions on the"sciences" auxiliary to historical research, the composition and thedefence of an original monograph. Every one now recognises that "theexamination for the diploma of studies will yield excellent fruit, ifthe vigilance and conscientiousness of examiners maintain it at itsproper value. "[249]

V. To sum up, the attractions of preparation for degrees have broughtthe Faculties a host of students. But, under the old system ofexaminations for the licentiate and for _agrégation_, preparation fordegrees was a task which did not harmonise very well with the work whichthe Faculties deemed suitable for themselves, useful to their pupils, and advantageous to science. The examination-system has therefore beenperseveringly reformed, not without difficulty, into conformity with acertain ideal of what the higher teaching of history ought to be. Theresult is that the Faculties have taken rank among the institutionswhich contribute to the positive progress of the historical sciences. Anenumeration of the works which have appeared under their auspices duringthe last few years would, if necessary, bear witness to the fact. 

This evolution has already produced satisfactory results, and willproduce more if it goes on as well as it has begun. To begin with, thetransformation of historical instruction in the Faculties has broughtabout a corresponding transformation at the École normale supérieure. The École normale has also, for two years, been awarding a "_Diplômed'études_"; original researches, pedagogic exercises, and generalculture are encouraged there in the same degree as by the new Faculties. It now differs from the Faculties only in being a close institution, recruited under certain precautions; practically it is a Faculty likethe others, but with a small number of select students. Secondly, theÉcole des hautes études and the École des chartes, both of which will beinstalled at the end of 1897, in the renovated Sorbonne, have stilltheir justification for existence; for many specialists are representedat the École des hautes études which are not, and doubtless never willbe, represented in the Faculties; and, in the case of the studiesbearing on mediæval history, the body of converging instruction given atthe École des chartes will always be incomparable. But the oldantagonism between the École des hautes études and the École des charteson the one hand, and the Faculties on the other, has disappeared. Allthese institutions, lately so dissimilar, will henceforth co-operate forthe purpose of carrying on a common work in a common spirit. Each ofthese retains its name, its autonomy, and its traditions; but togetherthey form a whole: the historical section of an ideal University ofParis, much vaster than the one which was sanctioned by the law in 1896. Of this "greater" University, the École des chartes, the École deshautes études, the École normale supérieure, and the whole body ofhistorical instruction given by the Faculty of Letters, are nowpractically so many independent "_instituts_. "

INDEX OF PROPER NAMES

Abd-el-Kader, 282

Aimo, 158

Alexander the Great, 272

Alphonse of Poitiers, 227

Altamira, R. , 328

Anselme, Père, 303

Ariovistus, 222

Aristophanes, 171

Aristotle, 44

Athenæus, 300

Bacon, Francis, 291

Bancroft, H. H. , 19, 20, 22, 31

Barthélemy, Abbé, 301

Bast, F. J. , 78

Bédier, J. , 85, 112

Bernheim, E. , 6, 7, 10, 13, 38, 56, 74, 91, 99, 100, 156, 182, 198, 237, 297

Blanchère, R. , de la, 319

Blass, F. , 74, 78, 79, 89, 92

Bodin, Jean, 44

Boeckh, A. , 107, 152

Böhmer, J. F. , 106

Bollandists, Society of, 35

Bonaventura, St. , 88, 90

Bouché-Leclercq, A. , 158

Boucherie, A. , 113

Bourdeau, L. , 275

Boutaric, E. , 227

Boyce, W. B. , 1

Bréquigny, L. G. O. F. De, 106

Broglie, E. De, 29

Brugière de Barante, A. G. P. , 301

Brunetière, F. , 113

Buchez, P. J. B. , 1

Bühler, G. , 56

Cæsar, Julius, 44, 194, 197, 218, 220, 222, 245

Cagnat, R. , 57

Cantù, C. , 311

Carlyle, Thomas, 132, 230, 303

Champollion, F. , 48

Charles IX. Of France, 168, 186

Chasles, M. , 88

Chateaubriand, F. A. De, 72, 301

Chemosh, the god, 212

Chérot, H. , 7

Chevalier, U. , 5, 7

Chladenius, J. M. , 6

Cicero, 44, 108

Cleopatra, 88, 248

Clovis, 158, 220, 223, 301

Cobet, C. C. , 78

Coulanges, Fustel de, 1, 9, 10, 64, 140, 144, 148, 149, 150, 158, 170, 215, 216, 230

Cournot, A. A. , 246, 249

Cousin, V. , 286, 287

Curtius, G. , 230, 314

Daniel, Père, 303

Daremberg, C. V. , 308

Darius Hystaspes, 151

Daunou, P. C. F. , 5, 6, 43, 47, 54, 55

Delisle, L. , 23, 97

Deloche, J. E. M. , 148

Demosthenes, 171

Dezobry, C. L. , 302

Droysen, J. G. , 3, 5, 7, 10, 106, 156, 158, 286, 314

Du Cange, C. Du F. , 105, 136, 148

Dumont, A. , 341

Duruy, V. , 327, 338, 339

Ebers, G. , 301

Edward VI. Of England, 249

Egger, E. , 108

Eginhard, 94

Ephorus, 298

Eusebius, 298

Feillet, A. , 162

Feugère, L. , 105, 136

Fisher, H. A. L. , 125

Flaubert, G. , 5, 32, 304, 319

Flint, R. , 2, 6, 8, 285

France, A. , 319

Fredegonda, 197

Freeman, E. A. , 5, 7, 10, 46

Froissart, Jean, 19

Froude, J. A. , 125, 126

Geiger, W. , 56

Gellius, Aulus, 300

Georgisch, P. , 106

Giannone, Pietro, 104

Gibbon, E. , 44

Gilbert, Gustav, 309

Giry, A. , 57

Glasson, E. , 149

Goethe, J. W. Von, 19, 319

Gow, J. , 75

Graux, C. , 123

Gregory of Corinth, 78

Gregory of Tours, 144, 146, 158, 180, 198, 256

Gröber, G. , 57, 310

Grote, G. , 183, 310

Grotius, Hugo, 44

Guicciardini, Francesco, 44

Guiraud, P. , 230

Hagen, H. , 78

Hardouin, Père, 99

Harnack, A. , 309

Havet, Julien, 12, 56, 97, 123, 128

Havet, Louis, 12

Hauréau, B. , 84, 111, 118, 123

Hegel, G. W. F. , 286

Henry VIII. Of England, 249

Henry II. Of France, 292

Henry, V. , 289

Herodotus, 44, 171, 179, 197

Horace, 99

Hoveden, John, 88

Hroswitha, 99

Hugo, Victor, 88, 89

Hume, D. , 44

Jaffé, P. , 106

Jameson, J. F. , 136

Jerome, St. , 112

Jesus Christ, 188

Joan of Arc, 188

John, King of England, 187

Jullian, C. , 297

Krumbacher, K. , 309

Kuhn, E. , 56

Lacombe, T. , 2, 233, 241, 277, 288

Lamprecht, K. , 230, 247, 284, 290, 314

Langlois, Ch. V. , 19, 38, 111, 135, 192, 345

Lasch, B. , 68

Laurent, F. , 285

Lavisse, E. , 134, 328, 333, 337, 341, 342

Lavoisier, A. L. , 302

Leibnitz, G. W. , 121, 122

Lee, Sidney, 308

Le Moyne, Père, 7

Lenglet de Fresnoy, N. , 6

Leonardo da Vinci, 88, 89

Liard, L. , 335, 341

Lindner, T. , 81

Lindsay, W. M. , 78, 79, 84

Livy, 44, 178, 180, 233, 297, 298

Locke, John, 44

Loebell, J. W. , 180

Lorenz, O. , 10

Loudun, the nuns of, 208

Louis VIII. Of France, 187

Louis of Granada, 88

Luard, H. R. , 98

Luther, Martin, 203

Mably, G. B. De, 43, 44

Macaulay, Lord, 303

Macchiavelli, N. , 44

Madvig, J. N. , 78

Mariani, L. , 4

Marquardt, J. , 309

Marselli, N. , 2

Mary Magdalene, St. , 88

Mary, Queen, 249

Matthew of Paris, 98

Matthew of Westminster, 97

Mayr, J. Von, 274

Mérimée, P. , 301

Mesha Inscription, the, 212

Meusel, H. , 148

Meyer, E. , 158

Meyer, P. , 29

Mézeray, F. E. De, 298

Michelet, J. , 230, 271, 286, 287, 301, 303, 327, 336

Möller, W. , 309

Mommsen, T. , 108, 118, 230, 286, 309, 314

Monod, G. , 100, 144, 297, 302

Montesquieu, C. De S. , 44, 257, 284, 299

Montfaucon, Père Bernard de, 29

Montgomery, Gabriel de, 292

Mortet, Ch. And V. , 11

Mourin, E. , 302

Müller, I. Von, 56, 74, 310

Mylaeus, 6

Napoleon I. , 26, 282

Newton, Isaac, 302

Niebuhr, B. G. , 158, 182

Nietzsche, F. , 319

Nitzsch, C. W. , 180

Oncken, W. , 311

Orosius, 298

Ossian, 91

Otto I. , 175

Paris, G. , 309

Patrizzi, Francesco, 6

Pattison, Mark, 115

Paul, H. , 75, 310

Pauly, A. , 308

Pausanias, 74

Peckham, John, 88

Peiresc, N. F. C. De, 22

Pflugk-Harttung, J. Von, 10, 130

Philippi, A. , 129

Piaget, A. , 91

Pisistratus, 207

Plato, 153

Plutarch, 44, 297

Polybius, 44, 279, 297

Potthast, A. , 106

Prou, M. , 57

Ranke, L. , 140, 286

Raynal, J. , 44

Reinach, S. , 75, 79

Renan, E. , 9, 29, 30, 40, 105, 114, 119, 122, 132, 134, 183

Retz, Cardinal de, 44, 162, 169

Rilliet, A. , 162

Robertson, J. M. , 115, 241

Robertson, W. , 44

Rocholl, R. , 285

Rousseau, J. J. , 44

Rulhière, C. C. De, 44

Saglio, E. , 308

Saint-Simon, C. H. De, 251

Sallust, 44

Sanchoniathon, 91

Schiller, J. C. F. Von, 162

Schlosser, F. C. , 311

Schoemann, G. F. , 309

Séguier, J. F. , 109

Seignobos, Ch. , 11, 66, 196, 257, 333, 334

Seneca, 78

Sforza, Ludovico, 282

Sickel, T. Von, 56

Simmel, G. , 217

Smedt, Père de, 10, 156, 207, 254

Spencer, Herbert, 287

Stephen, Leslie, 308

Stubbe, W. , 10

Suetonius, 94

Suger, Abbot, 170

Suidas, 158

Sully, M. , 169

Surville, Clotilde de, 91

Tacitus, 44, 141, 144, 171, 177, 194, 233, 256

Taine, H. A. , 140, 143, 247, 286

Tardif, A. , 5, 7, 156

Taylor, J. , 75

Thierry, Augustin, 98, 140, 230, 259, 265, 301, 302, 303

Thomas, A. , 73

Thucydides, 19, 44, 158, 183, 197, 297

Tobler, A. , 75

Tschudi, J. H. , 162, 171

Turenne, H. De la T. D'A. , 162

Vercingetorix, 88

Vergil, 84, 99

Vertot, R. A. De, 44

Villemarqué, H. De, 181

Vincent de Paul, St. 326

Voltaire, F. M. A. De, 44, 299

Vrain-Lucas, 88

Waitz, G. , 110, 118

Wallace, A. R. , 207

Waltzing, J. P. , 108

Wattenbach, W. , 73

Wauters, A. C. , 106

Weber, G. , 311

Wegele, F. X. Von, 122, 297

Wendover, Roger de, 98

Wissowa, G. , 308

Wittekind, 175

Wright, T. , 84

Xenophon, 44

Zumpt, A. W. , 108

FOOTNOTES:

[1] W. B. Boyce, "Introduction to the Study of History, Civil, Ecclesiastical, and Literary, " London, 1894, 8vo. 

[2] For example, P. J. B. Buches, in his _Introduction à la science del'histoire_, Paris, 1842, 2 vols. 8vo. 

[3] The history of the attempts which have been made to understand andexplain philosophically the history of humanity has been undertaken, asis well known, by Robert Flint. Mr. Flint has already given the historyof the Philosophy of History in French-speaking countries: "HistoricalPhilosophy in France and French Belgium and Switzerland, " Edinburgh andLondon, 1893, 8vo. It is the first volume of the expanded re-edition ofhis "History of the Philosophy of History in Europe, " publishedtwenty-five years ago. Compare the retrospective (or historical) part ofthe work of N. Marselli, _La scienza della storia_, i. , Torino, 1873. 

The most important original work which has appeared in France since thepublication of the analytical repertory of R. Flint is that of P. Lacombe, _De l'histoire considérée comme science_, Paris, 1894, 8vo. Cf. _Revue Critique_, 1895, i. P. 132. 

[4] _Revue Critique d'histoire et de littérature_, 1892, i. P. 164. 

[5] _Revue Critique d'histoire et de littérature_, 1888, ii. P. 295. Cf. _Le Moyen Age_, x. (1897), p. 91: "These books [treatises on historicalmethod] are seldom read by those to whom they might be useful, amateurswho devote their leisure to historical research; and as to professedscholars, it is from their masters' lessons that they have learnt toknow and handle the tools of their trade, leaving out of considerationthe fact that the method of history is the same as that of the othersciences of observation, the gist of which can be stated in a few words. 

[6] In accordance with the principle that historical method can only betaught by example, L. Mariani has given the humorous title _Corsopratico di metodologia della storia_ to a dissertation on a detail inthe history of Fermo. See the _Archivio della Società romana di storiapatria_, xiii. (1890), p. 211. 

[7] See an account of Freeman's work, "The Methods of Historical Study, "in the _Revue Critique_, 1887, i. P. 376. This work, says the critic, isempty and commonplace. We learn from it "that history is not so easy astudy as many fondly imagine, that it has points of contact with all thesciences, and that the historian truly worthy of the name ought to knoweverything; that historical certitude is unattainable, and that, inorder to make the nearest approach to it, it is necessary to haveconstant recourse to the original sources; that it is necessary to knowand use the best modern historians, but never to take their word forgospel. That is all. " He concludes: Freeman "without a doubt taughthistorical method far better by example than he ever succeeded in doingby precept. "

Compare _Bouvard et Pécuchet_, by G. Flaubert. Here we have twosimpletons who, among other projects, propose to write history. In orderto help them, one of their friends sends them (p. 156) "rules ofcriticism taken from the _Cours_ of Daunou, " such as: "It is no proof toappeal to rumour and common opinion; the witnesses cannot appear. Rejectimpossibilities: Pausanias was shown the stone swallowed by Saturn. Keepin mind the skill of forgers, the interest of apologists andcalumniators. " Daunou's work contains a number of truisms quite asobvious, and still more comic than the above. 

[8] Flint (ibid. P. 15) congratulates himself on not having to study theliterature of _Historic_, for "a very large portion of it is so trivialand superficial that it can hardly ever have been of use even to personsof the humblest capacity, and may certainly now be safely confined tokindly oblivion. " Nevertheless, Flint has given in his book a summarylist of the principal works of this kind published in French-speakingcountries from the earliest times. A more general and complete account(though still a summary one) of the literature of this subject in allcountries is furnished by the _Lehrbuch der historischen Methode_ of E. Bernheim (Leipzig, 1894, 8vo), pp. 143 _sqq. _ Flint (who was acquaintedwith several works unknown to Bernheim) stops at 1893, Bernheim at 1894. Since 1889 the _Jahresberichte der Geschichtswissenschaft_ havecontained a periodical account of recent works on historicalmethodology. 

[9] This seventh volume was published in 1844. But Daunou's celebrated_Cours_ was delivered at the Collège de France in the years 1819-30. 

[10] The Italians of the Renaissance (Mylæns, Francesco Patrizzi, andothers), and after them the writers of the last two centuries, ask whatis the relation of history to dialectic and rhetoric; to how many lawsthe historical branch of literature is subject; whether it is right forthe historian to relate treasons, acts of cowardice, crimes, disorders;whether history is entitled to use any style other than the sublime; andso on. The only books on _Historic_, published before the nineteenthcentury, which give evidence of any original effort to attack the realdifficulties, are those of Lenglet de Fresnoy (_Méthode pour étudierl'histoire_, Paris, 1713), and of J. M. Chladenius (_AllgemeineGeschichtswissenschaft_, Leipzig, 1752). The work of Chladenius has beennoticed by Bernheim (ibid. P. 166). 

[11] He has not always shown even good sense, for, in the _Coursd'études historiques_ (vii. P. 105), where he treats of a work, _De__l'histoire_, published in 1670 by Père Le Moyne, a feeble production, to say the least, bearing evident traces of senility, he expresseshimself as follows: "I cannot adopt all the maxims and preceptscontained in this treatise; but I believe that, after that of Lucian, itis the best we have yet seen, and I greatly doubt whether any of thosewhose acquaintance we have still to make has risen to the same height ofphilosophy and originality. " Père H. Chérot has given a sounder estimateof the treatise _De l'histoire_ in his _Étude sur la vie et lesoeuvres du P. Le Moyne_ (Paris, 1887, 8vo), pp. 406 _sqq. _

[12] Bernheim declares, however (ibid. P. 177), that this little workis, in his opinion, the only one which stands at the present level ofscience. 

[13] Flint says very well (ibid. P. 15): "The course of Historic hasbeen, on the whole, one of advance from commonplace reflection onhistory towards a philosophical comprehension of the conditions andprocesses on which the formation of historical science depends. 

[14] By P. Guiraud, in the _Revue des Deux Mondes_, March 1896, p. 75. 

[15] Renan has said some of the truest and best things that have everbeen said on the historical sciences in _L'Avenir de la science_ (Paris, 1890, 8vo), written in 1848. 

[16] Some of the most ingenious, some of the most logical, and some ofthe most widely applicable observations, on the method of the historicalsciences, have so far appeared, not in books on methodology, but in thereviews--of which the _Revue Critique d'histoire et de littérature_ isthe type--devoted to the criticism of new works of history anderudition. It is a very useful exercise to run through the file of the_Revue Critique_, founded, at Paris, in 1867, "to enforce respect formethod, to execute justice upon bad books, to check misdirected andsuperfluous work. "

[17] The first edition of the _Lehrbuch_ is dated 1889. 

[18] The best work that has hitherto been published (in French) onhistorical method is a pamphlet by MM. Ch. And V. Mortet, _La Science del'histoire_ (Paris, 1894, 8vo), 88 pp. , extracted from vol. Xx. Of the_Grande Encyclopédie_. 

[19] One of us, M. Seignobos, proposes to publish later on a completetreatise of Historical Methodology, if there appears to be a public forthis class of work. 

[20] It cannot be too often stated that the study of history, as it isprosecuted at school, does not presuppose the same aptitudes as the samestudy when prosecuted at the university or in after life. Julien Havet, who afterwards devoted himself to the (critical) study of history, foundhistory wearisome at school. "I believe, " says M. L. Havet, "that theteaching of history [in schools] is not organised in such a manner as toprovide sufficient nourishment for the scientific spirit. .. . Of all thestudies comprised in our school curricula, history is the only one inwhich the pupil is not being continually called upon to verifysomething. When he is learning Latin or German, every sentence in atranslation requires him to verify a dozen different rules. In thevarious branches of mathematics the results are never divorced fromtheir proofs; the _problems_, too, compel the pupil to think through thewhole for himself. Where are the _problems_ in history, and whatschoolboy is ever trained to gain by independent effort an insight intothe interconnection of events?" (_Bibliothèque de l'École des chartes_, 1896, p. 84). 

[21] M. Langlois wrote Book I. , Book II. As far at Chapter VI. , thesecond Appendix, and this Preface; M. Seignobos the end of Book II. , Book III. , and the first Appendix. Chapter I. In the second book, Chapter V. Of the third book, and the Conclusion, were written incommon. 

[22] In practice one does not as a rule resolve to treat a point ofhistory before knowing whether there are or are not documents inexistence which enable it to be studied. On the contrary, it is theaccidental discovery of a document which suggests the idea of thoroughlyelucidating the point of history to which it relates, and thus leads tothe collection, for this purpose, of other documents of the same class. 

[23] It is pitiable to see how the best of the early scholars struggledbravely, but vainly, to solve problems which would not even have existedfor them if their collections had not been so incomplete. This lack ofmaterial was a disadvantage for which the most brilliant ingenuity couldnot compensate. 

[24] "How hard it is to gain the means whereby we mount to the sources"(Goethe, _Faust_, i. 3). 

[25] See C. V. Langlois, _H. H. Bancroft et Cie. _, in the _Revueuniversitaire_, 1894, i. P. 233. 

[26] The earlier scholars were conscious of the unfavourable characterof the conditions under which they worked. They suffered keenly from theinsufficiency of the instruments of research and the means ofcomparison. Most of them made great efforts to obtain information. Hencethese voluminous correspondences between scholars of the last fewcenturies, of which our libraries preserve so many precious fragments, and these accounts of scientific searches, of journeys undertaken forthe discovery of historical documents, which, under the name of _Iter_(_Iter Italicum_, _Iter Germanicum_, &c. ), were formerly fashionable. 

[27] We may remark, in passing, a delusion which is childish enough butvery natural, and very common among collectors: they all tend toexaggerate the intrinsic value of the documents they possess, simplybecause they themselves are the possessors. Documents have beenpublished with a sumptuous array of commentaries by persons who hadaccidentally acquired them, and who would, quite rightly, have attachedno importance to them if they had met with them in public collections. This is, we may add, merely a manifestation, in a somewhat crude form, of a general tendency against which it is always necessary to guard: aman readily exaggerates the importance of the documents he possesses, the documents he has discovered, the texts he has edited, the personsand the questions he has studied. 

[28] See L. Delisle, _Le Cabinet des manuscrits de la Bibliothèquenationale_, Paris, 1868-81, 3 vols. 4to. The histories of ancientdepositories of documents, which have been recently published inconsiderable number, have been modelled on this admirable work. 

[29] Many of the ancient documents still in circulation are the proceedsof ancient thefts from state institutions. The precautions now takenagainst a recurrence of such depredations are stringent, and, in nearlyevery instance, as effective as could be desired. 

As to modern (printed) documents, the rule of legal deposit [compulsorypresentation of copies to specified libraries], which has now beenadopted by nearly all civilised countries, guarantees their preservationin public institutions. 

[30] It is known that Napoleon I. Entertained the chimerical design ofconcentrating at Paris the archives of the whole of Europe, and that, for a beginning, he conveyed to that city the archives of the Vatican, the Holy Roman Empire, the crown of Castile, and others, which later onthe French were compelled to restore. Confiscation is now out of thequestion. But the ancient archives of the notaries might be centralisedeverywhere, as in some countries they are already, in publicinstitutions. It is not easy to explain why at Paris the departments ofForeign Affairs, of War, and of Marine preserve ancient papers whosenatural place would be at the Archives Nationales. A great many moreanomalies of this kind might be mentioned, which in certain casesimpede, where they do not altogether preclude, research; for the smallcollections, whose existence is not required, are precisely those whoseregulations are the most oppressive. 

[31] The international exchange of documents is worked in Europe(without charge to the public) by the agency of the various ForeignOffices. Besides this, most of the great institutions have agreementswith each other for mutual loans; this system is as sure and sometimesmore rapid in its operation than the diplomatic system. The question oflending original documents for use outside the institution where theyare preserved has of late years been frequently mooted at congresses ofhistorians and librarians. The results so far obtained are eminentlysatisfactory. 

[32] These are sometimes large collections of formidable bulk; it ismore natural to undertake the cataloguing of small accumulations whichdemand less labor. It is for the same reason that many insignificant butshort cartularies have been published, while several cartularies of thehighest importance, being voluminous, have still to be edited. 

[33] See his autobibliography, published by E. De Broglie, _Bernard deMontfaucon et les Bernardins_, ii. (Paris, 1891, 8vo), p. 323. 

[34] E. Renan, _L'Avenir de la science_, p. 217. 

[35] _Romania_, xxi. (1892), p. 625. 

[36] In the passage quoted above. 

[37] Mr. H. H. Bancroft, in his Memoirs, entitled "Literary Industries"(New York, 1891, 16mo), analyses with sufficient minuteness somepractical consequences of the imperfection of the methods of research. He considers the case of an industrious writer proposing to write thehistory of California. He easily procures a few books, reads them, takesnotes; these books refer him to others, which he consults in the publiclibraries of the city where he resides. Several years are passed in thismanner, at the end of which he perceives that he has not a tenth part ofthe resources in his hands; he travels, maintains correspondences, but, finally despairing of exhausting the subject, he comforts his conscienceand pride with the reflection that he has done much, and that many ofthe works he has not seen, like many of those he has, are probably ofvery slight historic value. As to newspapers and the myriads of UnitedStates government reports, all of them containing facts bearing onCalifornian history; being a sane man, he has never dreamed of searchingthem from beginning to end: he has turned over a few of them, that isall; he knows that each of these fields of research would afford alabour of several years, and that all of them would fill the better partof his life with drudgery. As for oral testimony and manuscripts, hewill gather a few unpublished anecdotes in chance conversations; he willobtain access to a few family papers; all this will appear in his bookas notes and authorities. Now and again he will get hold of a fewdocumentary curiosities among the state archives, but as it would takefifteen years to master the whole collection, he will naturally becontent to glean a little here and there. Then he begins to write. Hedoes not feel called upon to inform the public that he has not seen_all_ the documents; on the contrary, he makes the most of what he hasbeen able to procure in the course of twenty-five years of industriousresearch. 

[38] Some dispense with personal search by invoking the assistance ofthe functionaries charged with the administration of depositories ofdocuments; the indispensable search is, in these cases, conducted by thefunctionaries instead of by the public. Cf. _Bouvard et Pécuchet_, p. 158. Bouvard and Pécuchet resolve to write the life of the Duke ofAngoulême; for this purpose "they determined to spend several days atthe municipal library of Caen to make researches. The librarian placedgeneral histories and pamphlets at their disposal. .. . "

[39] These considerations have already been presented and developed inthe _Revue universitaire_, 1894, i. P. 321 _sqq. _

[40] It is well known that, since the opening of the Papal Archives, several governments and learned societies have established Institutes atRome, the members of which are, for the most part, occupied incataloguing and making known the documents of these archives, inco-operation with the functionaries of the Vatican. The French School atRome, the Austrian Institute, the Prussian Institute, the PolishMission, the Institute of the "Goerresgesellschaft, " Belgian, Danish, Spanish, Portuguese, Russian, and other scholars, have performed, andare performing, cataloguing work of considerable extent in the archivesof the Vatican. 

[41] Catalogues of documents sometimes, but not always, mention the factthat such and such a document has been edited, dealt with critically, utilised. The generally received rule is that the compiler mentionscircumstances of this kind when he is aware of them, without imposing onhimself the enormous task of ascertaining the truth on this head[sic] inevery instance where he is ignorant of it. 

[42] E. Bernheim, _Lehrbuch der historischen Methode_, 2nd ed. , pp. 196-202. 

[43] C. V. Langlois, _Manuel de Bibliographie historique_: I. _Instruments bibliographiques_, Paris, 1896, 16mo. 

[44] E. Renan, _Feuilles détachées_ (Paris, 1892, 8vo), pp. 96 _sqq. _

[45] vii. P. 228 _sqq. _

[46] E. A. Freeman, _The Methods of Historical Study_ (London, 1885, 8vo), p. 45. 

In France geography has long been regarded as a science closely relatedto history. An _Agrégation_, which combines history and geography, exists at the present day, and in the _lycées_ history and geography aretaught by the same professors. Many people persist in asserting thelegitimacy of this combination, and even take umbrage when it isproposed to separate two branches of knowledge united, as they say, bymany essential connecting links. But it would be hard to find any goodreason, or any facts of experience, to prove that a professor ofhistory, or an historian, is so much the better the more he knows ofgeology, oceanography, climatology, and the whole group of geographicalsciences. In fact, it is with some impatience, and to no immediateadvantage, that students of history work through the courses ofgeography which their curricula force upon them; and those students whohave a real taste for geography would be very glad to throw historyoverboard. The artificial union of history with geography dates back, inFrance, to an epoch when geography was an ill-defined and ill-arrangedsubject, regarded by all as a negligeable branch of study. It is a relicof antiquity that we ought to get rid of at once. 

[47] "Historiography" is a branch of the "History of Literature;" it isthe sum of the results obtained by the critics who have hitherto studiedancient historical writings, such as annals, memoirs, chronicles, biographies, and so forth. 

[48] This is only true under reservation; there is an instrument ofresearch which is indispensable to all historians, to all students, whatever be the subject of their special study. History, moreover, ishere in the same situation as the majority of the other sciences: allwho prosecute original research, of whatever kind, need to know severalliving languages, those of countries where men think and work, ofcountries which, from the point of view of science, stand in theforefront of contemporary civilisation. 

In our days the cultivation of the sciences is not confined to anysingle country, or even to Europe. It is international. All problems, the same problems, are being studied everywhere simultaneously. It isdifficult to-day, and to-morrow it will be impossible, to find a subjectwhich can be treated without taking cognisance of works in a foreignlanguage. Henceforth, for ancient history, Greek and Roman, a knowledgeof German will be as imperative as a knowledge of Greek and Latin. Questions of strictly local history are the only ones still accessibleto those who do not possess the key to foreign literatures. The greatproblems are beyond their reach, for the wretched and ridiculous reasonthat works on these problems in any language but their own are sealedbooks to them. 

Total ignorance of the languages which have hitherto been the ordinaryvehicles of science (German, English, French, Italian) is a diseasewhich age renders incurable. It would not be exacting too much torequire every candidate for a scientific profession to be at least_trilinguis_--that is, to be able to understand, fairly easily, twolanguages besides his mother-tongue. This is a requirement to whichscholars were not subject formerly, when Latin was still the commonlanguage of learned men, but which the conditions of modern scientificwork will henceforth cause to press with increasing weight upon thescholars of every country. [*]

[*] Perhaps a day will come when it will be necessary to know the mostimportant Slavonic language; there are already scholars who are settingthemselves to learn Russian. The idea of restoring Latin to its oldposition of universal language is chimerical. See the file of the_Phoenix, seu nuntius latinus universalis_ (London, 1891, 4to). 

The French scholars who are unable to read German and English arethereby placed in a position of permanent inferiority as compared withtheir better instructed colleagues in France and abroad; whatever theirmerit, they are condemned to work with insufficient means ofinformation, to work badly. They know it. They do their best to hidetheir infirmity, as something to be ashamed of, except when they make acynical parade of it and boast of it; but this boasting, as we caneasily see, is only shame showing itself in a different way. Too muchstress cannot be laid upon the fact that a practical knowledge offoreign languages is auxiliary in the first degree to all historicalwork, as indeed it is to scientific work in general. 

[49] When the "auxiliary sciences" were first inserted in the curriculaof French universities, it was observed that some students whose specialsubject was the French Revolution, and who had no interest whatever inthe middle ages, took up palæography as an "auxiliary science, " and thatsome students of geography, who were in no way interested in antiquity, took up epigraphy. Evidently they had failed to understand that thestudy of the "auxiliary sciences" is recommended, not as an end initself, but because it is of practical utility to those who devotethemselves to certain special subjects. See the _Revue universitaire_, 1895, ii. P. 123. 

[50] On this point note the opinions of T. Von Sickel and J. Havet, quoted in the _Bibliothèque l'École des chartes_, 1896, p. 87. In 1854the Austrian Institute "für österreichische Geschichtsforschung" wasorganised on the model of the French École des chartes. Anotherinstitution of the same type has lately been created in the "Istituto distudi superiori" at Florence. "We are accustomed, " we read in England, "to hear the complaint that there is not in this country any institutionresembling the École des chartes" (_Quarterly Review_, July 1896, p. 122). 

[51] This is a suitable place to enumerate the principal "manuals"published in the last twenty-five years. But a list of them, ending at1894, will be found in Bernheim's _Lehrbuch_, pp. 206 sqq. We will onlyrefer to the great "manuals" of "Philology" (in the comprehensive senseof the German "Philologie, " which includes the history of language andliterature, epigraphy, palæography, and all that pertains to textualcriticism) now in course of publication: the _Grundriss farindo-arischen Philologie und Altertumskunde_, edited by G. Bühler; the_Grundriss der iranischen Philologie_, edited by W. Geiger and E. Kuhn;the _Handbuch der classichen Altertumswissenschaft_, edited by I. VonMüller; the _Grundriss der germanischen Philologie_, edited by H. Paul, the second edition of which began to appear in 1896; the _Grundriss derromanischen Philologie_, edited y G. Gröber. In these vast repertoriesthere will be found, along with a short presentment of the subject, complete bibliographical references, direct as well as indirect. 

[52] The French "manuals" of MM. Prou (Palæography), Giry (Diplomatic), Cagnat (Latin Epigraphy), and others, have diffused among the public theidea and knowledge of the auxiliary subjects of study. New editions haveenabled, and will enable, them to be kept up to date--a very necessaryoperation, for most of these subjects, though now settled in the main, are being enriched and made more precise every day. Cf. _supra_, p. 38. 

[53] What exactly are we to understand by this "incommunicableknowledge, " of which we speak? When a specialist is very familiar withthe documents of a given class or period, associations of ideas areformed in his brain; and when he examines a new document of the sameclass or species, analogies suddenly dawn upon him which would escapeany one of less experience, however well furnished he might be with themost perfect repertories. The fact is, that not all the peculiarities ofdocuments can be isolated; there are some which cannot be classifiedunder any intelligible head, and which, therefore, cannot be found inany tabulated list. But the human memory, when it is good, retains theimpression of these peculiarities, and even a faint and distant stimulussuffices to revive the apprehension of them. 

[54] _Supra_, p. 17. 

[55] This expression, which frequently occurs, needs explanation. It isnot to be taken to apply to a _species_ of facts. There are nohistorical facts in the sense in which we speak of chemical facts. Thesame fact is or is not historical according to the manner in which it isknown. It is only the mode of acquiring knowledge that is historical. Asitting of the Senate is a fact of direct observation for one who takespart in it; it becomes historical for the man who reads about it in areport. The eruption of Vesuvius in the time of Pliny is a geologicalfact which is known historically. The historical character is not in thefacts, but in the manner of knowing them. 

[56] Fustel de Coulanges has said it. Cf. _supra_, p. 4, note 1. 

[57] In the sciences of observation it is the fact itself, observeddirectly, which is the starting-point. 

[58] _Infra_, ch. Vii. 

[59] We shall not treat specially of the criticism of material documents(objects, monuments, &c. ) where it differs from the criticism of writtendocuments. 

[60] For the details and the logical justification of this method seeSeignobos, _Les Conditions psychologiques de la connaissance enhistoire_, in the _Revue philosophique_, 1887, ii. Pp. 1, 168. 

[61] The most favourable case, that in which the document has been drawnup by what is called an ocular "witness, " is still far short of theideal required for scientific knowledge. The notion of _witness_ hasbeen borrowed from the procedure of the law-courts; reduced toscientific terms, it becomes that of an _observer_. A testimony is anobservation. But, in point of fact, historical testimony differsmaterially from scientific observation. The observer proceeds by fixedrules, and clothes his report in language of rigorous precision. On theother hand, the "witness" observes without method, and reports inunprecise language; it is not known whether he has taken the necessaryprecautions. It is an essential attribute of historical documents thatthey come before us as the result of work which has been done withoutmethod and without guarantee. 

[62] See B. Lasch, _Das Erwachen und die Entwickelung der historischenKritik im Mittelalter_ (Breslan, 1887, 8vo). 

[63] Natural credulity is deeply rooted in indolence. It is easier tobelieve than to discuss, to admit than to criticise, to accumulatedocuments than to weigh them. It is also pleasanter; he who criticisesdocuments must sacrifice some of them, and such a sacrifice seems a deadloss to the man who has discovered or acquired the document. 

[64] _Revue philosophique_, l. C. , p. 178. 

[65] A member of the _Société des humanistes français_ (founded at Parisin 1894) amused himself by pointing out, in the _Bulletin_ of thissociety, certain errors amenable to verbal criticism which occur invarious editions of posthumous works, especially the _Mémoiresd'outre-tombe_. He showed that it is possible to remove obscurities inthe most modern documents by the same methods which are used inrestoring ancient texts. 

[66] On the habits of the mediæval copyists, by whose intermediateagency most of the literary works of antiquity have come down to us, seethe notices collected by W. Wattenbach, _Das Schriftwesen imMittelalter_, 3rd ed. (Berlin, 1896, 8vo). 

[67] See, for example, the _Coquilles lexicographiques_ which have beencollected by A. Thomas, in _Romania_, xx. (1891), pp. 464 _sqq. _

[68] See E. Bernheim, _Lehrbuch der historischen Methode_, 2nd ed. , pp. 341-54. Also consult F. Blass, in the _Handbuch der klassischenAltertumswissenschaft_, edited by I. Von Müller, I. , 2nd ed. (1892), pp. 249-89 (with a detailed bibliography); A. Tobler, in the _Grundriss derromanischen Philologie_, I. (1888), pp. 253-63; H. Paul, in the_Grundriss der germanischen Philologie_, I. , 2nd ed. (1896), pp. 184-96. 

In French read the section _Critique des textes, in Minerva, Introduction à l'étude des classiques scolaires grecs et latins_, by J. Gow and S. Reinach (Paris, 1890, 16mo), pp. 50-65. 

The work of J. Taylor, "History of the Transmission of Ancient Books toModern Times" (Liverpool, 1889, 16mo), is of no value. 

[69] This rule is not absolute. The editor is generally accorded theright of unifying the spelling of an autograph document--provided thathe informs the public of the fact--wherever, as in most moderndocuments, the orthographical vagaries of the author possess nophilological interest. See the _Instructions pour la publication destextes historiques_, in the _Bulletin de la Commission royale d'histoirede Belgique_, 5th series, vi. (1896); and the _Grundsätze für dieHerausgabe von Actenstücken zur neueren Geschichte_, laboriouslydiscussed by the second and third Congresses of German historians, in1894 and 1895, in the Deutsche _Zeitschrift für Geschichtswissenschaft_, xi. P. 200, xii. P. 364. The last Congresses of Italian historians, heldat Genoa (1893) and at Rome (1895), have also debated this question, butwithout result. What are the liberties which it is legitimate to take inreproducing autograph texts? The question is more difficult than isimagined by those who are not professionally concerned with it. 

[70] Interpolations will be treated of in chapter iii p. 92. 

[71] The scribes of the Carlovingian Renaissance and of the Renaissanceproper of the fifteenth century endeavoured to furnish intelligibletexts. They therefore corrected everything they did not understand. Several ancient works have been in this manner irretrievably ruined. 

[72] The principal of these are, for the classical languages, besidesthe above-mentioned work of Blass (_supra_, p. 74, note), the_Adversaria critica_ of Madvig (Copenhagen, 1871-74, 3 vols. 8vo). ForGreek, the celebrated _Commentatio palæographica_ of F. J. Bast, published as an appendix to an edition of the grammarian Gregory ofCorinth (Leipzig, 1811, 8vo), and the _Variæ lectiones_ of Cobet(Leiden, 1873, 8vo). For Latin, H. Hagen, _Gradus ad criticen_ (Leipzig, 1879, 8vo), and W. M. Lindsay, "An Introduction to Latin TextualEmendation based on the Text of Plautus" (London, 1896, 16mo). Acontributor to the _Bulletin de la Société des humanistes français_ hasexpressed, in this publication, a wish that a similar collection mightbe compiled for modern French. 

[73] Cf. _Revue Critique_, 1895, ii. P. 358. 

[74] Quite recently our scholars used to neglect this elementaryprecaution, in order, as they said, to avoid an "air of pedantry. " M. B. Hauréau has published, in his _Notices et extraits de quelquesmanuscrits latins de la Bibliothèque nationale_ (vi. P. 310), a piece ofrhythmic verse, "De presbytero et logico. " "It is not unedited, " sayshe; "Thomas Wright has already published it. .. . But this edition is verydefective; the text is occasionally quite unintelligible. We have, therefore, considerably amended it, making use, for this purpose, of twocopies, which, it most be conceded, are neither of them faultless. .. . "The edition follows, with no variants. Verification is impossible. 

[75] "Textual emendation too often misses the mark through want ofknowledge of what may be called _the rules of the game_" (W. M. Lindsay, p. V. In the work referred to above). 

[76] It has often been asked whether _all_ texts are worth the troubleof "establishing" and publishing them. "Among our ancient texts, " saysM. J. Bédier, referring to French mediæval literature, "which ought weto publish? Every one. But, it will be asked, are we not alreadystaggering under the weight of documents?. .. The following is the reasonwhy publication should be exhaustive. As long as we are confronted bythis mass of sealed and mysterious manuscripts, they will appeal to usas if they contained the answer to every riddle; every candid mind willbe hampered by them in its flights of induction. It is desirable topublish them, if only to get rid of them and to be able, for the future, to work as if they did not exist. .. . " (_Revue des Deux Mondes_, February15, 1894, p. 910). All documents ought to be catalogued, as we havealready pointed out (p. 31), in order that researchers may be relievedof the fear that there may be documents, useful for their purposes, ofwhich they know nothing. But in every case where a summary analysis of adocument can give a sufficient idea of its contents, and its form is ofno special interest, there is nothing gained by publishing it _inextenso_. We need not overburden ourselves. Every document will beanalysed some day, but many documents will never be published. 

[77] Editors of texts often render their task still longer and moredifficult than it need be by undertaking the additional duty ofcommentators, under the pretext of explaining the text. It would be totheir advantage to spare themselves this labour, and to dispense withall annotation which does not belong to the "apparatus criticus" proper. See, on this point, T. Lindner, _Ueber die Herausgabe vongeschichtlichen Quellen_, in the _Mittheilungen des Instituts fürösterreichische Geschichtsforschung_, xvi. , 1895, pp. 501 _sqq. _

[78] To realise this it is enough to compare what has hitherto been doneby the most active societies, such as the Society of the _MonumentaGermaniæ historica_ and the _Istituto storico italiano_, with what stillremains for them to do. The greater part of the most ancient documentsand the hardest to restore, which have long taxed the ingenuity ofscholars, have now been placed in a relatively satisfactory condition. But an immense amount of mechanical work has still to be done. 

[79] R. De Gourmont, _Le Latin mystique_ (Paris, 1891, 8vo), p. 258. 

[80] See these alleged autographs in the _Bibliothèque nationale_, nouv. Acq. Fr. , No. 709. 

[81] F. Blass has enumerated the chief of these motives with referenceto the pseudepigraphic literature of antiquity (pp. 269 _sqq. _ in thework already quoted). 

[82] E. Bernheim (_Lehrbuch_, pp. 243 _sqq. _) gives a somewhat lengthylist of spurious documents, now recognised as such. Here it will beenough to recall a few famous hoaxes: Sarchoniathon, Clotilde deSurville, Ossian. Since the publication of Bernheim's book severalcelebrated documents, hitherto exempt from suspicion, have been struckoff the list of authorities. See especially A. Piaget, _La Chronique deschanoines de Neuchâtel_ (Neuchâtel, 1896, 8vo). 

[83] When the modifications of the primitive text are the work of theauthor himself, they are "alterations. " Internal analysis, and thecomparison of different editions, bring them to light. 

[84] See F. Blass, ibid. , pp. 254 _sqq. _

[85] As a rule it matters little whether the _name_ of the author has orhas not been discovered. We read, however, in the _Histoire_ _littérairede la France_ (xxvi. P. 388): "We have ignored anonymous sermons:writings of this facile character are of no importance for literaryhistory when their authors are unknown. " Are they of any more importancewhen we know the authors' names?

[86] In very favourable cases the examination of the plagiarist'smistakes has made it possible to determine even this style ofhandwriting, the size, and the manner of arrangement of the manuscriptsource. The deductions of the investigation of sources, like those oftextual criticism, are sometimes supported by obvious palæographicalconsiderations. 

[87] The Investigations of Julien Havet (_Questions mérovingiennes_, Paris, 1896, 8vo) are regarded as models. Very difficult problems arethere solved with faultless elegance. It is also well worth while toread the memoirs in which M. L. Delisle has discussed questions oforigin. It is in the treatment of these questions that the mostaccomplished scholars win their triumphs. 

[88] See the edition of H. R. Luard (vol. I. , London, 1890, 8vo) in the_Rerum Britannicarum medii ævi scriptores_. Matthew of Westminster's_Flores historiarum_ figure in the Roman "Index, " because of thepassages borrowed from the _Chronica majora_ of Matthew of Paris, whilethe _Chronica majora_ themselves have escaped censure. 

[89] It would be instructive to draw up a list of the celebratedhistorical works, such as Augustin Thierry's _Histoire de la Conquête del'Angleterre par les Normands_, whose authority has been completelydestroyed after the authorship of their sources has been studied. Nothing amuses the gallery more than to see an historian convicted ofhaving built a theory on falsified documents. Nothing is more calculatedto cover an historian with confusion than to find that he has falleninto the error of treating seriously documents which are no documents atall. 

[90] One of the crudest (and commonest) forms of "uncritical method" isthat which consists in employing as if they were documents, and placingon the same footing as documents, the utterances of modern authors onthe subject of documents. Novices do not make a sufficient distinction, in the works of modern authors, between what is added to the originalsource and what is taken from it. 

[91] See a list of examples in Bernheim's _Handbuch_, pp. 283, 289. 

[92] It is because it is necessary to subject documents of mediæval andancient history to the most searching criticism in respect of authorshipthat the study of antiquity and the middle ages passes for more"scientific" than that of modern times. The truth is, that it is merelyhampered by more preliminary difficulties. 

[93] _Revue philosophique_, 1887, ii. P. 170. 

[94] The theory of the critical investigation of authorship is nowsettled, _ne varietur_; it is given in detail in Bernheim's _Lehrbuch_, pp. 242-340. For this reason we have had no scruple in dismissing itwith a short sketch. In French, the introduction of M. G. Monod to his_Études critiques sur les sources de l'histoire mérovingienne_ (Paris, 1872, 8vo) contains elementary considerations on the subject. Cf. _RevueCritique_, 1873, i. P. 308. 

[95] Renan, _Feuilles détachées_, p. 103. 

[96] It would be very interesting to have information on the methods ofwork of the great scholars, particularly those who undertook long tasksof collection and classification. Some information of this kind is to befound in their papers, and occasionally in their correspondence. On themethods of Du Cange, see L. Feugère, _Étude sur la vie et les ouvragesde Du Cange_ (Paris, 1858, 8vo), pp. 62 _sqq. _

[97] See J. G. Droysen, _Grundriss der Historik_, p. 19: "Criticalclassification does not exclusively adopt the chronological point ofview. .. . The more varied the points of view which criticism uses togroup materials, the more solid are the results yielded by converginglines of inquiry. "

The system has now been abandoned of grouping documents in a _Corpus_ orin _regesta_, as was done formerly, because they have the commoncharacteristic of being unedited, or possibly for the exactly oppositereason. At one time the compilers of _Analecta, Reliquiæmanuscriptorum_, "treasuries of _anecdota_, " _spicilegia_, and so on, used to publish all the documents of a certain class which had thecommon feature of being unedited and of appearing interesting to them;on the other hand, Georgisch (_Regesta Chronologico-diplomatica_), Bréquigny (_Table chronologique des diplômes, chartes et actes imprimésconcernant l'histoire de France_), Wauters (_Table chronologique deschartes et diplômes imprimés concernant l'histoire de Belgique_), havegrouped together all the documents of a certain species which had thecommon character of having been printed. 

[98] J. P. Waltzing, _Recueil général des inscriptions latines_(Louvain, 1892, 8vo), p. 41. 

[99] Ibid. When the geographical order is adopted, a difficulty arisesfrom the fact that the origin of certain documents is unknown; manyinscriptions preserved in museums have been brought there no one knowswhence. The difficulty is analogous to that which results, forchronological _regesta_, from documents without date. 

[100] Here the only difficulty arises in the case of documents whose_incipit_ has been lost. In the eighteenth century Séguier devoted agreat part of his life to the construction of a catalogue, in thealphabetical order of the _incipit_, of the Latin inscriptions, to thenumber of 50, 000, which had at that time been published: he searchedthrough some twelve thousand works. This vast compilation has remainedunpublished and useless. Before undertaking work of such magnitude it iswell to make sure that it is on a rational plan, and that thelabour--the hard and thankless labour--will not be wasted. 

[101] See G. Waitz, _Ueber die Herausgabe und Bearbeitung von Regesten_, in the _Historische Zeitschrift_, xl. (1878), pp. 280-95. 

[102] In the absence of a predetermined logical order, and when thechronological order is not suitable, it is sometimes an advantage toprovisionally group the documents (that is, the slips) in thealphabetical order of the words chosen as headings (_Schlagwörter_). This is what is called the "dictionary system. "

[103] See Langlois, _Manuel de bibliographie historique_, i. P. 88. 

[104] This argument is easy to develop, and often has been, recently byM. J. Bédier, in the _Revue des Deux Mondes_, February 15, 1894, pp. 932_sqq. _

There are some who willingly admit that the labours of erudition areuseful, but ask impatiently whether "the editing of a text" or "thedeciphering of a Gothic parchment" is "the supreme effort of the humanmind, " and whether the intellectual ability implied by the practice ofexternal criticism does or does not justify "all the fuss made overthose who possess it. " On this question, obviously devoid of importance, a controversy was held between M. Brunetière, who recommended scholarsto be modest, and M. Boucherie, who insisted on their reasons for beingproud, in the pages of the _Revue des langues romanes_, 1880, vols. Iand ii. 

[105] There have been men who were critics of the first water whereexternal criticism alone was concerned, but who never rose to theconception of higher criticism, or to a true understanding of history. 

[106] Renan, _Essais de morale et de critique_, p. 36. 

[107] "If it were only for the sake of the severe mental discipline, Ishould not think very highly of the philosopher who had not, at leastonce in his life, worked at the elucidation of some special point"(_L'Avenir de la science_, p. 136). 

[108] On the question whether it is necessary for every one to do "allthe preliminary grubbing for himself, " cf. J. M. Robertson, "Buckle andHis Critics" (London, 1895, 8vo), p. 299. 

[109] Renan, _L'Avenir de la science_, p. 230. 

[110] A university professor is in a very good position for discouragingand encouraging vocations; but "it is by personal effort that the goal(critical skill) must be attained by the students, as Waitz well said inan academic oration; the teacher's part in this work is small. .. . "(_Revue Critique_, 1874, ii. P. 232). 

[111] Quoted by Fr. X. Von Wegele, _Geschichte der deutschenHistoriographie_ (München, 1885, 8vo), p. 653. 

[112] Renan, ibid. , p. 125. 

[113] B. Hauréau, _Notices et extraits de quelques manuscrits latins dela Bibliothèque nationale_, i. (Paris, 1890, 8vo). P. V. 

[114] _Bibliothèque de l'École des chartes_, 1896, p. 88. Compareanalogous traits in the interesting intellectual biography of theHellenist, palæographer, and bibliographer, Charles Graux, by E. Lavisse(_Questions d'enseignement national_, Paris, 1885, 18mo, pp. 265_sqq. _). 

[115] See H. A. L. Fisher in the _Fortnightly Review_, Dec. 1894, p. 815. 

[116] Most of those who have a vocation for critical scholarship possessboth the power of solving problems and the taste for collecting. It is, however, easy to divide them into two categories according as they showa marked preference for textual criticism and investigation ofauthorship on the one hand, or for the more absorbing and lessintellectual labours of collection on the other. J. Havet, a past-masterin the study of erudite problems, always declined to undertake a generalcollection of Merovingian royal charters, a work which his admirersexpected from him. In this connection he readily admitted his "want oftaste for feats of endurance" (_Bibliothèque de l'École des chartes_, 1896, p. 222). 

[117] It is common to hear the opposite of this maintained, namely, thatthe labours of critical scholarship (external criticism) have thisadvantage over other labours in the field of history that they arewithin the range of average ability, and that the most moderateintellects, after a suitable preliminary drilling, may be usefullyemployed in them. It is quite true that men with no elevation of soul orpower of thought can make themselves useful in the field of criticism, but then they must have special qualities. The mistake is to think thatwith good will and a special drilling every one without exception can befitted for the operations of external criticism. Among those who areincapable of these operations, as well as among those who are fitted forthem, there are both men of sense and blockheads. 

[118] A. Philippi, _Einige Bemerkungen über den philologischenUnterricht_, Giessen, 1890, 4to. Cf. _Revue Critique_, 1892, i. P. 25. 

[119] J. Von Pflugk-Harttung, _Geschichtsbetrachtungen_, Gotha, 1890, 8vo. 

[120] Ibid. , p. 21. 

[121] Cf. _supra_, p. 99. 

[122] Renan, _L'Avenir de la science_, p. Xiv. 

[123] _Revue historique_, lxiii. (1897), p. 320. 

[124] Renan, ibid. , pp. 122, 243. The same thought has been more thanonce expressed, in different language, by E. Lavisse, in his addressesto the students of Paris (_Questions d'enseignement national_, pp. 14, 86, &c. ). 

[125] One of us (M. Langlois) proposes to give elsewhere a detailedaccount of all that has been done in the last three hundred years, butespecially in the nineteenth century, for the organisation of historicalwork in the principal countries of the world. Some information hasalready been collected on this subject by J. Franklin Jameson, "TheExpenditures of Foreign Governments in behalf of History, " in the"Annual Report of the American Historical Association for 1891, " pp. 38-61. 

[126] L. Feugère, _Étude sur la vie et les ouvrages de Du Cange_, pp. 55, 58. 

[127] Even the specialists in external criticism themselves, when theydo not take the line of despising all synthesis _a priori_, are almostas easily dazzled as anybody else by incorrect syntheses, by a show of"general ideas, " or by literary artifices, in spite of theirclear-sightedness where works of critical scholarship are concerned. 

[128] The sciences of observation do, however, need a species ofcriticism. We do not accept without verification results obtained byanybody, but only results obtained by those who know how to work. Butthis criticism is made once for all, and applies to the author, not tohis works; historical criticism, on the contrary, is obliged to dealseparately with every part of a document. 

[129] Cf. _supra_, book ii. Chap. I. P. 67. 

[130] Cf. _supra_, p. 122. 

[131] Taine appears to have proceeded thus in vol. Ii, _La Révolution_, of his _Origines de la France contemporaine_. He had made extracts fromunpublished documents and inserted a great number of them in his work, but it would seem that he did not first methodically analyse them inorder to determine their meaning. 

[132] Fustel de Coulanges explains very clearly the danger of this_method_: "Some students begin by forming an opinion . .. And it is nottill afterwards that they begin to read the texts. They run a great riskof not understanding them at all, or of understanding them wrongly. Whathappens is that a kind of tacit contest goes on between the text and thepreconceived opinions of the reader; the mind refuses to grasp what iscontrary to its idea, and the issue of the contest commonly is, not thatthe mind surrenders to the evidence of the text, but that the textyields, bends, and accommodates itself to the preconceived opinion. .. . To bring one's personal ideas into the study of texts is the subjectivemethod. A man thinks he is contemplating an object, and it is his ownidea that he is contemplating. He thinks he is observing a fact, and thefact at once assumes the colour, and the significance his mind wishes itto have. He thinks he is reading a text, and the words of the text takea particular meaning to suit a ready-made opinion. It is this subjectivemethod which has done most harm to the history of the Merovingianepoch. .. . To read the texts was not enough; what was required was toread them before forming any convictions. .. . " (_Monarchie franque_, p. 31). For the same reason Fustel de Coulanges deprecated the reading ofone document in the light of another; he protested against the custom ofexplaining the _Germania_ of Tacitus by the barbaric laws. In the _Revuedes questions historiques_, 1897, vol. I, a lesson on method, _Del'analyse des textes historiques_, is given apropos of a commentary byM. Monod on Gregory of Tours: "The historian ought to begin his workwith an exact analysis of each document. .. . The analysis of a text . .. Consists in determining the sense of each word and eliciting the truemeaning of the writer. .. . Instead of searching for the sense of each ofthe historian's words, and for the thought he has expressed in them, he[M. Monod] comments on each sentence in the light of what is found inTacitus or the Salic law. .. . We should understand what analysis reallyis. Many talk about it, few use it. .. . The use of analysis is, by anattentive study of every detail, to elicit from a text all that is init; not to introduce into the text what is not there. "

After reading this excellent advice it will be instructive to read M. Monod's reply (in the _Revue historique_); it will be seen that Fustelde Coulanges himself did not always practise the method he recommended. 

[133] Cf. _supra_, p. 103. 

[134] The work of analysis may be entrusted to a second person; this iswhat happens in the case of _regesta_ and catalogues of records; if theanalysis has been correctly performed by the compiler of _regesta_, there is no need to do it over again. 

[135] Practical examples of this procedure will be found in Deloche, _LaTrustis et l'antrustion royal_ (Paris, 1873, 8vo), and, above all, inFustel de Coulanges. See especially the study of the words _marca_(_Recherches sur quelques problêmes d'histoire_, pp. 322-56), _mallus_(ibid. , 372-402), _alleu_ (_L'Alleu et le domaine rural_, pp. 149-70), _portio_ (ibid. , pp. 239-52). 

[136] The theory and an example of this procedure will be found inFustel de Coulanges, _Recherches sur quelques problêmes d'histoire_ (pp. 189-289), with reference to the statements of Tacitus about the Germans. See especially pp. 263-89, the discussion of the celebrated passage onthe German mode of culture. 

[137] Fustel de Coulanges formulates it thus: "It is never safe toseparate two words from their context; this is just the way to mistaketheir meaning" (_Monarchie franque_, p. 228, note I). 

[138] This is how Fustel de Coulanges condemns this practice: "I am notspeaking of pretenders to learning who quote second-hand, and at mosttake the trouble to verify whether the phrase they have seen quotedreally occurs in the passage indicated. To verify quotations is onething and to read texts quite another, and the two often lead toopposite results" (_Revue des questions historiques_, 1887, vol. I. ). See also (_L'Alleu et le domaine rural_, pp. 171-98) the lesson given toM. Glasson on the theory of the community of land: forty-five quotationsare studied in the light of their context, with the object of provingthat none of them bears the meaning M. Glasson attributed to it. We mayalso compare the reply: Glasson, _Les Communaux et le domaine rural al'époque franque_, Paris, 1890. 

[139] All that is original in Fustel de Coulanges rests on hisinterpretative criticism; he never did personally any work in externalcriticism, and his critical examination of authors' good faith andaccuracy was hampered by a respect for the statements of ancient authorswhich amounted to credulity. 

[140] A parallel difficulty occurs in the interpretation of illustrativemonuments; the representations are not always to be taken literally. Inthe Behistun monument Darius tramples the vanquished chiefs under foot:this is a metaphor. Mediæval miniatures show us persons lying in bedwith crowns on their heads: this is to symbolise their royal rank; thepainter did not mean that they wore their crowns to sleep in. 

[141] A. Boeckh, in the _Encyclopædie und Methodologie derphilologischen Wissenschaften_, second edition (1886), has given atheory of _hermeneutic_ to which Bernheim has been content to refer. 

[142] The method of extracting information on external facts from awriter's conceptions forms part of the theory of constructive reasoning. _See_ book iii. 

[143] For example, Père de Smedt, Tardif, Droysen, and even Bernheim. 

[144] Descartes, who came at a time when history still consisted in thereproduction of pre-existing narratives, did not see how to applymethodical doubt to the subject; he therefore refused to allow it aplace among the sciences. 

[145] Fustel de Coulanges himself did not rise above this kind oftimidity. With reference to a speech attributed to Clovis by Gregory ofTours, he says: "Doubtless we are unable to affirm that these words wereever pronounced. But, all the same, we ought not to affirm, incontradiction to Gregory of Tours, that they were not. .. . The wisestcourse is to accept Gregory's text" (_Monarchie franque_, p. 66). Thewisest, or rather the only scientific course, is to admit that we knownothing about the words of Clovis, for Gregory himself had no knowledgeof them. 

[146] Quite recently, E. Meyer, one of the most critically experthistorians of antiquity, has in his work, _Die Entstehung desJudenthums_ (Halle, 1896, 8vo), revived this strange juridical argumentin favour of the narrative of Nehemiah. M. Bouché-Leclercq, in aremarkable study on "The Reign of Seleucus II. (Callinicus) andHistorical Criticism" (_Revue des Universités du Midi_, April-June1897), seems, by way of reaction against the hypercriticism of Niebuhrand Droysen, to incline towards an analogous theory: "Historicalcriticism, if it is not to degenerate into agnosticism--which would besuicidal--or into individual caprice, must place a certain amount oftrust in testimony which it cannot verify, as long as it is not flatlycontradicted by other testimony of equal value. " M. Bouché-Leclercq isright as against the historian who, "after having discredited all hiswitnesses, claims to put himself in their place, and sees with theireyes something quite different from what they themselves saw. " But whenthe "testimony" is insufficient to give us the scientific knowledge of afact, the only correct attitude is "agnosticism, " that is, a confessionof ignorance; we have no right to shirk this confession because chancehas permitted the destruction of the documents which might havecontradicted the testimony. 

[147] The "Memoirs of Cardinal de Retz" furnish a conclusive instance:the anecdote of the ghosts met by Retz and Turenne. A. Feillet, whoedited Retz in the _Collection des Grands Écrivains de la France_, hasshown (vol. I. P. 192) that this story, so vividly narrated, is falsefrom beginning to end. 

[148] A good example of the fascination exerted by a circumstantialnarrative is the legend respecting the origin of the League of the threeprimitive Swiss cantons (Gessler and the Grütli conspirators), which was_fabricated_ by Tschudi in the sixteenth century, became classical onthe production of Schiller's "William Tell, " and has only beenextirpated with the greatest difficulty. (See Rilliet, _Origines de laConféderation suisse_, Geneva, 1869, 8vo. )

[149] Striking example of falsehoods due to vanity are to be found inabundance in the _Économies royales_ of Sully and the _Mémoires_ ofRetz. 

[150] Fustel de Coulanges himself went to the formulæ of theinscriptions in honour of the emperors for a proof that the peoplesliked the imperial _régime_. "If we read the inscriptions, the sentimentwhich they exhibit is always one of satisfaction and gratitude. .. . Seethe collection of Orelli, the most frequent expressions are. .. . " And theenumeration of the titles of respect given to the emperors ends withthis strange aphorism: "It would show ignorance of human nature to seenothing but flattery in all this. " There is not even flattery here;there is nothing but formulæ. 

[151] Suger, in his life of Louis VI. , is a model of this type. 

[152] The _Chronicon Helveticum_ of Tschudi is a striking instance. 

[153] Aristophanes and Demosthenes are two striking examples of thepower great writers have of paralysing critics and obscuring facts. Nottill the close of the nineteenth century has any one ventured torecognise frankly their lack of good faith. 

[154] For example, the account of the election of Otto I. In the _GestaOttonis_ of Wittekind. 

[155] For example, the statistics on the population, the commerce, andthe wealth of European countries given by the Venetian ambassadors ofthe sixteenth century, and the descriptions of the usages of the Germansin the _Germania_ of Tacitus. 

[156] It would be interesting to examine how much of Roman orMerovingian history would be left if we rejected all documents but thosewhich represent direct observation. 

[157] It will be seen why we have not separately defined and studied"first-hand documents. " The question has not been raised in the propermanner in historical practice. The distinction ought to apply to_statements_, not to documents. It is not the document which comes to usat first, second, or third hand; it is the statement. What is called a"first-hand document" is nearly always composed in part of second-handstatements about facts of which the author had no personal knowledge. The name "second-hand document" is given to those which, like the workof Livy, contain nothing first-hand; but the distinction is too crude toserve as a guide in the critical examination of statements. 

[158] There is much less modification where the oral tradition assumes aregular or striking form, as is the case with verses, maxims, proverbs. 

[159] Sometimes the _form_ of the phrase tells its own tale, when, inthe midst of a detailed narrative, obviously of legendary origin, wecome across a curt, dry entry in annalistic style, obviously copied froma written document. That is what we find in Livy (see Nitzsch, _Dierömische Annalistik_, Leipzig, 1873, 8vo), and in Gregory of Tours (seeLoebell, _Gregor von Tours_, Leipzig, 1868, 8vo). 

[160] The events which strike the popular imagination and aretransmitted by legend are not generally those which seem to us the mostimportant. The heroes of the _chansons de gestes_ are hardly knownhistorically. The Breton epic songs relate, not to the great historicalevents, as Villemarqué's collection led people to believe, but toobscure local episodes. The same holds of the Scandinavian sagas; forthe most part they relate to quarrels among the villagers of Iceland orthe Orkneys. 

[161] The theory of legend is one of the most advanced parts ofcriticism. Bernheim (in his _Lehrbuch_, pp. 380-90) gives a good summaryand a bibliography of it. 

[162] "History of Greece, " vols. I. And ii. Compare Renan, _Histoire dupeuple d'Israël_, vol. I. (Paris, 1887, 8vo), Introduction. 

[163] And yet Niebuhr made use of the Roman legends to construct atheory, which it was afterwards necessary to demolish, of the strugglebetween the patricians and the plebeians; and Curtius, twenty yearsafter Grote, looked for historical facts in the Greek legends. 

[164] See _supra_, pp. 93 _sqq. _

[165] Cf. _supra_, p. 166. 

[166] It is often said, "The author would not have dared to write thisif it had not been true. " This argument does not apply to societies in alow state of civilisation. Louis VIII. Dared to write that John Lacklandhad been condemned by the verdict of his peers. 

[167] See above, p. 153. Similarly, the particular facts which composethe history of forms (palæography, linguistic science) are directlyestablished by the analysis of the document. 

[168] Primitive Greece has been studied in the Homeric poems. Mediævalprivate life has been reconstructed principally from the _chansons degestes_. (See C. V. Langlois, _Les Traditions sur l'histoire de lasociété française au moyen âge d'après les sources littéraires_, in the_Revue historique_, March-April, 1897. )

[169] Most historians refrain from rejecting a legend till its falsityhas been proved, and if by chance no document has been preserved tocontradict it, they adopt it provisionally. This is how the first fivecenturies of Rome are still dealt with. This method, unfortunately stilltoo general, helps to prevent history from being established as ascience. 

[170] For the logical justification of this principle in history see C. Seignobos, _Revue Philosophique_, July-August 1887. Complete scientificcertitude is only produced by an agreement between observations made ondifferent _methods_; it is to be found at the junction of two differentpaths of research. 

[171] This case is studied and a good example given by Bernheim, _Lehrbuch_, p. 421. 

[172] It is hardly necessary to enter a caution against the childishmethod of counting the documents on each side of a question and decidingby the majority. The statement of a single author who was acquaintedwith a fact is evidently worth more than a hundred statements made bypersons who knew nothing about it. The rule has been formulated longago: _Ne numerentur, sed ponderentur_. 

[173] Cf. _supra_, p. 94. 

[174] It is hardly possible to study here the special difficulties whicharise in the application of these principles, as when the author, wishing to conceal his indebtedness, has introduced deviations in orderto put his readers off the scent, or when the author has combinedstatements taken from different documents. 

[175] Here we merely indicate the principle of the method ofconfirmation; its applications would require a very lengthy study. 

[176] Père de Smedt has devoted to this question a part of his_Principes de la critique histoire_ (Paris, 1887, 12mo). 

[177] The solution of the question is different in the case of thesciences of direct observation, especially the biological sciences. Science knows nothing of the possible and the impossible; it onlyrecognises facts which have been correctly or incorrectly observed:facts which had been declared impossible, as the existence of aerolites, have been discovered to be genuine. The very notion of a miracle ismetaphysical; it implies a conception of the universe as a whole whichtranscends the limits of observation. (See Wallace, "Miracles and ModernSpiritualism. ")

[178] See above, p. 194. 

[179] In the experimental sciences an hypothesis is a form of questionaccompanied by a provisional answer. 

[180] Fustel de Coulanges saw the necessity of this. In the preface tohis _Recherches sur quelques problêmes d'histoire_ (Paris, 1885, 8vo) heannounces his intention of presenting his researches "in the form whichall my works have, that is, in the form of questions which I ask myself, and on which I endeavour to throw light. "

[181] Fustel de Coulanges himself seems to have been misled by them:"History is a science; it does not imagine, it only sees" (_Monarchiefranque_, p. 1). "History, like every science, consists in a process ofdiscerning facts, analysing them, comparing them, and noting theirconnections. .. . The historian . .. Seeks facts and attains them by theminute observation of texts, as the chemist finds his in the course ofexperiments conducted with minute precision" (Ibid. , p. 39). 

[182] The subjective character of history has been brought out intostrong relief by the philosopher G. Simmel, _Die Probleme derGeschichtsphilosophie_ (Leipzig, 1892, 8vo). 

[183] This has been eloquently put by Carlyle and Michelet. It is alsothe substance of the famous expression of Ranke: "I wish to state howthat really was" (_wie es eigentlich gewesen_). 

[184] Cf. Pp. 219-23. 

[185] Curtius in his "History of Greece, " Mommsen in his "History ofRome" (before the Empire), Lamprecht in his "History of Germany. "

[186] It will be enough to mention Augustin Thierry, Michelet, andCarlyle. 

[187] See P. Guiraud, _Fustel de Coulanges_ (Paris, 1896, 12mo), p. 164, for some very judicious observations on this subject. 

[188] The classification of M. Lacombe (_De l'histoire considérée commescience_, chap. Vi. ), founded on the motives of actions and the wantsthey are intended to satisfy, is very judicious from the philosophicalpoint of view, but does not meet the practical needs of historians; itrests on abstract psychological categories (economic, reproductive, sympathetic, ambitious, &c. ), and ends by classing together verydifferent species of phenomena (military institutions along witheconomics). 

[189] Ecclesiastical institutions form part of the government; in Germanmanuals of antiquities they are found among institutions, while religionis classed with the arts. 

[190] Modes of transport, which are often put under commerce, form aspecies of industry. 

[191] Property is an institution of mixed character, being at onceeconomic, social, and political. 

[192] For the history and biography of this movement see Bernheim, _Lehrbuch_, pp. 45-55. 

[193] It is no longer necessary to demonstrate the nullity of the notionof _race_. It used to be applied to vague groups, formed by a nation ora language; for race as understood by historians (Greek, Roman, Germanic, Celtic, Slavonic races) has nothing but the name in commonwith race in the anthropological sense--that is, a group of menpossessing the same hereditary characteristics. It has been reduced toan absurdity by the abuse Taine made of it. A very good criticism of itwill be found in Lacombe (ibid. , chap. Xviii. ), and in Robertson ("TheSaxon and the Celt, " London, 1897, 8vo). 

[194] There is no general agreement on the proper place in history ofretrograde changes, of those oscillations which bring things back to thepoint from which they started. 

[195] The theory of chance as affecting history has been expounded in amasterly manner by M. Cournot, _Considérations sur la marche des idéeset des événements dans les temps modernes_ (Paris, 1872, 2 vols. 8vo). 

[196] Lamprecht, in a long article, _Was ist Kulturgeschichte_, published in the _Deutsche Zeitschrift für Geschichtswissenschaft_, NewSeries, vol. I. , 1896, has attempted to base the history of civilisationon the theory of a collective soul of society producing "social-psychic"phenomena common to the whole society, and differing from period toperiod. This is a metaphysical hypothesis. 

[197] The expression _national_ history, introduced in the interests ofpatriotism, denotes the same thing. The history of the nation meanspractically the history of the State. 

[198] See Cournot, ibid. , i. P. Iv. 

[199] We have already (p. 143) treated of this fault of method. 

[200] The discussion of this argument, which was formerly much used inreligious history, was a favourite subject with the earlier writers whotreated of methodology, and still occupies a considerable space in the_Principes de la critique historique_ of Père de Smedt. 

[201] This is what Montesquieu attempted in his _Esprit des Lois_. In acourse of lectures at the Sorbonne, I have endeavoured to give a sketchof such a comprehensive account. --[Ch. S. ]

[202] See p. 204. 

[203] Michelet has discredited the study of physiological influences bythe abuse which he has made of it in the last part of his "History ofFrance"; it is, however, indispensable for the understanding of a man'scareer. 

[204] On the subject of statistics, a method which is now perfected, agood summary with a bibliography will be found in the _Handwörterbuchder Staatswissenschaften_, Jena, 1890-94, Ia. 8vo. And two goodmethodical treatises, J. Von. Mayr, _Theoretische Statistik_ and_Bevölkerungsstatistik_, in the collection of Marquardsen and Seydel, Freiburg, 1895 and 1897, Ia. 8vo. 

[205] As is done by Boardeau (_l'Histoire et les Historiens_, Paris, 1888, 8vo), who proposes to reduce the whole of history to a series ofstatistics. 

[206] A good example will be found in Lacombe, _De l'Histoire ConsidéréeComme Science_, p. 146. 

[207] We have thought it useless to discuss here the question whetherhistory ought, in accordance with the ancient tradition, to fulfil yetanother function, whether it ought to pass judgment on men and events, that is to supplement the description of facts by expressions ofapprobation or censure, either from the point of view of a moral ideal, general or particular (the ideal of a sect, a party, or a nation), orfrom the practical point of view, by examining, as Polybius did, whetherhistorical actions were well or ill adapted to their purpose. Anaddition of this kind could be made to any descriptive study: thenaturalist might express his sympathy with or his admiration for ananimal, he might condemn the ferocity of the tiger, and praise thedevotion of the hen to her chickens. But it is obvious that in history, as in every other subject, judgments of this kind are foreign toscience. 

[208] Comparison between two facts of detail belonging to very differentaggregates (for example the comparison of Abd-el-Kader with Jagurtha, ofNapoleon with Sforza) is a striking method of exposition, but not ameans of reaching a scientific conclusion. 

[209] This system is still followed by several contemporary authors, theBelgian jurist Laurent in his _Études sur l'histoire de l'humanité_, theGerman Rocholl, and even Flint, the English historian of the philosophyof history. 

[210] Thus Taine, in _Les origines de la France Contemporaine_, explainsthe origin of the privileges of the _ancien régime_ by the servicesformerly rendered by the privileged classes. 

[211] A good criticism of the theory of progress will be found in P. Lacombe, _De l'histoire Considérée Comme Science_. 

[212] See the very clear declarations of one of the principalrepresentatives of linguistic science in France, V. Henry, _Antinomieslinguistiques_, Paris, 1896, 8vo. 

[213] See above, p. 284. 

[214] Lamprecht, in the article quoted, p. 247, after having comparedthe artistic, religious, and economic evolutions of mediæval Germany, and after having shown that they can all be divided into periods of thesame duration, explains the simultaneous transformations of thedifferent usages and institutions of a given society by thetransformations of the collective "social soul. " This is only anotherform of the same hypothesis. 

[215] The historians of literature, who began by searching for theconnection between the arts and the rest of social life, thus gave thefirst place to the most difficult question. 

[216] For the earlier epochs, consult good histories of Greek, Roman, and mediæval literature which contain chapters devoted to "historians. "For the modern period, consult the Introduction of M. G. Monod to vol. I. Of the _Revue historique_; the work by F. X. V. Wegde, _Geschichteder deutschen Historiographie_ (1885), relates only to Germany, and ismediocre. Some "Notes on History in France in the Nineteenth Century"have been published by C. Jullian as an Introduction to his _Extraitsdes historiens français du xixe siècle_ (Paris, 1897, 12mo). Thehistory of modern historiography has still to be written. See thepartial attempt by E. Bernheim, _Lehrbuch_, pp. 13 _sqq. _

[217] It would be interesting to find out what are the earliest printedbooks furnished with notes in the modern fashion. Bibliophiles whom wehave consulted are unable to say, their attention never having beendrawn to the point. 

[218] It is clear that the romantic methods which are used for thepurpose of obtaining effects of local colour and "revising" the past, often puerile in the hands of the ablest writers, are altogetherintolerable when they are employed by any others. See a good example(criticism of a book of M. Mourin by M. Monod) in the _Revue Critique_, 1874, ii. Pp. 163 _sqq. _

[219] It is a commonplace, and an error all the same, to maintain theexact opposite of the above, namely, that the works of critical scholarslive, while the works of historians grow antiquated, so that scholarsgain a more solid reputation than historians do: "Père Daniel is nowread no longer, and Père Anselme is always read. " But the works ofscholars become antiquated too, and the fact that not all the parts ofthe work of Père Anselme have yet been superseded (that is why he isstill read), ought not to deceive us: the great majority of the workswritten by scholars, like those of researchers in the sciences proper, are provisional and doomed to oblivion. 

[220] "It is in vain that those professionally concerned try to deceivethemselves on this point; not everything in the past is interesting. ""Supposing we were to write the Life of the Duke of Angoulême, " saysPécuchet. "But he was an imbecile!" answers Bouvard; "Never mind;personages of the second order often have an enormous influence, andperhaps he was able to control the march of events. "--G. Flaubert, _Bouvard et Pécuchet_, p. 157. 

[221] As persons of moderate ability have a tendency to preferinsignificant subjects, there is active competition in the treatment ofsuch subjects. We often have occasion to note the simultaneousappearance of several monographs on the same subject. It is not rare forthe subject to be altogether devoid of importance. 

[222] Interesting subjects for monographs are not always capable ofbeing treated: there are some which the state of the sources puts out ofthe question. This is why beginners, even those who have ability, experience so much embarrassment in choosing subjects for their firstmonographs, when they are not aided by good advice or good fortune, andoften lose themselves in attempting the impossible. It would be verysevere, and very unjust, to judge any one from the list of his _first_monographs. 

[223] In practice it is proper to give at the beginning a list of thesources used in the whole of the monograph (with appropriatebibliographical information as to the printed works, and in the case ofmanuscripts, a mention of the nature of the documents and theirshelf-marks); besides, each special statement should be accompanied byits proof: the exact text of the supporting document should be quoted, if possible, so that the reader may be in a position to verify theinterpretation; otherwise an analysis of it should be given in a note, or, at the least, the title of the document, with its shelf-mark, orwith a precise indication of the place where it was published. Thegeneral rule is to put the reader in a position to know the exactreasons for which such and such conclusions have been adopted at eachstage of the analysis. 

Beginners, resembling ancient authors in this respect, naturally do notobserve all these rules. Frequently, instead of quoting the text or thetitles of documents, they refer to these by their shelf-mark, or by thetitle of the general collection in which they are printed, from whichthe reader can learn nothing as to the nature of the text adduced. Thefollowing is another mistake of the crudest kind, and yet of frequentoccurrence: Beginners, and persons of little experience, do not alwaysunderstand why the custom has been introduced of inserting footnotes; atthe bottom of the pages of the books they have they see a fringe ofnotes; they think themselves bound to fringe their own books in the sameway, but their notes are adventitious and purely ornamental; they do notserve either to exhibit the proof or to enable the reader to verify thestatements. All these methods are inadmissible, and should be vigorouslydenounced. 

[224] Almost all beginners have an unfortunate tendency to wander offinto superfluous digressions, to amass reflections and pieces ofinformation which have no relevance to the main subject; they wouldrecognise, if they reflected, that the causes of this tendency are badtaste, a kind of naïve vanity, sometimes mental confusion. 

[225] We meet with declarations like the following: "I have been longfamiliar with the documents of this period and this class. I have animpression that such and such conclusions, which I cannot prove, aretrue. " Of two things one: either the author can give the reasons for hisimpression, and then we can judge them, or he cannot give them, and wemay assume that he has none of serious value. 

[226] This difference has a tendency to disappear. The most recentalphabetical collections of historical facts (the _Realencyclopædie derclassischen Altertumswissenschaft_ of Pauly-Wissowa, the _Dictionnairedes antiquités_ of Daremberg and Saglio, the _Dictionary of NationalBiography_ of Leslie Stephen and Sidney Lee) are furnished with asufficiently ample apparatus. It is principally in biographicaldictionaries that the custom of giving no proofs tends to persist; seethe _Allgemeine deutsche Biographie_, &c. 

[227] _Revue Critique_, 1874, i. P. 327. 

[228] The custom of appending to "histories, " that is to narratives ofpolitical events, summaries of the results obtained by the specialhistorians of art, literature, &c. , still persists. A "History ofFrance" would not be considered complete if it did not contain chapterson the history of art, literature, manners, &c. , in France. However, itis not the summary account of special evolutions, described at secondhand from the works of specialists, which is in its proper place in ascientific "History"; it is the study of those general facts which havedominated the special evolutions in their entirety. 

[229] It is hard to imagine what it is possible for the most interestingand best established results of modern criticism to become, in the handsof negligent and unskilful popularisers. The persons who know most ofthese possibilities are those who have occasion to read the improvised"compositions" of candidates in history examinations: the ordinarydefects of inferior popularisation are here pushed sometimes to anabsurd length. 

[230] Cf. _supra_, p. 266. 

[231] We have spoken above of the element of subjectivity which it isimpossible to eliminate from historical construction, and which has beenmisinterpreted to the extent of denying history the character of ascience: this element of subjectivity which troubled Pécuchet (G. Flaubert, _Bouvard et Pécuchet_, p. 157) and Sylvestre Bonnard (A. France, _Le crime de Silvestre Bonnard_, p. 310), and which causes Faustto say:

 "Die Zeiten der Vergangenheit Sind uns ein Buch mit sieben Siegeln. Was ihr den Geist der Zeiten heisst, Das ist im Grund der Herren eigner Geist, In dem die Zeiten sich bespiegeln. "

["Past times are to us a book with seven seals. What you call the spiritof the times is at bottom your own spirit, in which the times aremirrored. "--Goethe, _Faust_, i. 3. ]

[232] A saying attributed to a "Sorbonne professor" by M. De laBlanchère (_Revue Critique_, 1895, i. P. 176). Others have declaimed onthe theme that the knowledge of history is mischievous and paralyses. See F. Nietzsche, _Unzeitgemässe Betrachtungen_, II. _Nutzen undNachtheil der Historie für das Leben_, Leipzig, 1874, 8vo. 

[233] History and the social sciences are mutually dependent on eachother; they progress in parallel lines by a continual interchange ofservices. The social sciences furnish a knowledge of the present, required by history for the purpose of making representations of thefacts and reasoning from documents. History gives the information aboutevolutions which is necessary in order to understand the present. 

[234] The same institution has been adopted in German-speaking countriesunder the name of _Leitfaden_ (guiding-thread), and in English-speakingcountries under the name of _Text-book_. 

[235] We must make an exception of Michelet's _Précis de l'histoiremoderne_, and do Duruy the justice to acknowledge that in hisschool-books, even in the first editions, he has endeavoured, oftensuccessfully, to make his narratives both interesting and instructive. 

[236] For a criticism of this method, see above, p. 265. 

[237] The most complete, and probably the most accurate, account of thestate of the secondary teaching of history after the reforms has beengiven by a Spaniard, R. Altamira, _La Enseñanza de la historia_, 2ndedition, Madrid, 1895, 8vo. 

[238] We are here treating only of France. But, in order to dispel anillusion of the French public, we may remark that historical pedagogy isstill less advanced in English-speaking countries, where the methodsused are still mechanical, and even in German-speaking countries, whereit is hampered by the conception of patriotic teaching. 

[239] I have endeavoured, in a course of lectures at the Sorbonne, to doa part of this work. --[Ch. S. ]

[240] Let it be noted, however, that to the question put to thecandidates for the modern Baccalaureate in July 1897, "What purpose isserved by the teaching of history?" eighty per cent. Of the candidatesanswered, in effect, either because they believed it, or because theythought it would please, "To promote patriotism. "--[C. V. L. ]

[241] This is what has been produced in Germany under the name of_Quellenbuch_. 

[242] The same pedagogic theory will be found in the preface to my_Histoire narrative et descriptive des anciens peuples de l'Orient_, Supplement for the use of professors, Paris, 1890, 8vo. --[Ch. S. ]

[243] I have treated this question in the _Revue universitaire_, 1896, vol. I. --[Ch. S. ]

[244] On the organisation of higher education in France at this epochand on the first reforms, see the excellent work of M. L. Liard, _l'Enseignement supérieur en France_, Paris, 1888-94, 2 vols. 8vo. 

[245] E. Lavisse, _Questions d'enseignement national_, p. 12. 

[246] Cf. _supra_, p. 55. 

[247] _Revue historique_, lxiii. (1897), p. 96. 

[248] See the _Revue internationale d'enseignement_, Feb. 1893; the_Revue universitaire_, June 1892, Oct. And Nov. 1894, July 1895; and the_Political Science Quarterly_, Sept 1894. 

[249] _Revue historique_, l. C. P. 98. I have developed elsewhere what Ihave here contented myself with stating. See the _Revue internationalede l'enseignement_, Nov. 1897. --[C. V. L. ]