cPV ■^ -0 OUTLINE Historical Method BY FRED MORROW FLING, Ph.D., P7-ofesior cf European His oiy in the U/iivers.'iy of Neb aska. LINCOLN J. H. MiLLKR. lS9i>. THE LIBRARY OF CONGRESS, Two Copies Received my 2 1903 Copyright Entry ClASfa^ XXc No COPY B. Copyrighted 1899, by y. M. Fling. ,V CONTENTS. CHAPTER. PAGE. I. Introduction 5 II. Sources, Bibliography, and Auxiliary Sci- ences 15 III. External Criticism: Genuineness of the Source 26 IV. External Criticism: Localization of the Source 36 V. External Criticism: Analysis of the Source and Restoration of the Text 49 VI. Internal Criticism: Interpretation of the Source and Value of the Source 63 VII. Internal Criticism: . Establishment of the Facts 75 VIII. Synthetic Operations : Imagining the Facts. Grouping the Facts, and Con- structi Vf Reasoning 87 IX. Synthetic Operations: Environment and the Philosophy of History 100 X. Synthetic Operations: Exposition 113 CHAPTER I. INTRODUCTION. WE shall have better history teaching when we have better trained teachers; and we shall have the trained teachers when the teachers themselves, and those who employ them, realize that history can be taught only by those who have been prepared for the work. As the matter stands to-day, it is the popular belief that any intelligent person may teach a class in history without special training, or, with no other knowledge than general information, may participate in a discussion upon methods of teaching history and what the object of his- torical study is. It is not an uncommon thmg for a college graduate, who has devoted all his time to Greek and Latin, or to science, to have a class in history assigned to him. He may not be particularly pleased with the assignment, but it does not strike him as at all incongruous. While, on the other hand, no good high school principal would assign a class in Greek to a man who had not been trained for that work. What is the reason for this distinction ? It is not far to seek. As history has been taught, and is still taught, in the high school, no spe- cial training is necessary. Any bright man can read over the lesson and hear the class recite it. The large majority of history teachers never engaged in a bit of original research and 6 HISTORICAL METHOD. have no more idea of what constitutes history a science than has the intelligent public outside the school room. This unscientific spirit crops out of all the discussions in teachers' associations and of most of the articles in educational, papers. The reasons so commonly given to justify the study of history can be traced to the same source. Instead of studying history for the same reason that we study botany or chemistry, that is, for its own sake, we must study it for the ethical training it gives, for its power of forming character, and so on ad infinitum. Now, history is not ethics, and it claims a right to an independent existence. It deals with the evolution of man in society, and no further reason need be given to justify the study of history than the necessity of knowing how that evolution has taken place. If the teachers of history can be induced to see how rational this view of their work is, one long step will have been taken toward improving the work. But what is this training that the teacher must have, if better work is to be done and if history is to take its place by the side of Latin, Greek, mathematics, and the sciences as a dis- ciplinary study? She must Itarn what the process is by which an historical narrative is constructed, and she must go through that process herself. She can no more become a satisfactory teacher of history without this training than she could become a successful teacher of chemistry without laboring with her own hands in the laboratorv. It is difficult to HISTORICAL METHOD. " make the teachers realize this, but if the cru- sade for better history teaching is to succeed, they must realize it. This subject of how history is written, I have treated in a general manner elsewhere; it is my intention to treat it now more in detail. But before taking up the successive steps in the con- struction of an historical narrative, I wish to point out the differences between the historical method and the method employed in the nat- ural sciences. In the natural sciences, the so-called method of direct observation is made use of. The ob- ject itself is studied directly either with the naked eye or with the microscope. Not one ob- servation but many are made and under the most favorable circumstances. The observa- tions thus made are recorded at once, and in exact, scientific language, the meaning of which is not ambiguous. But scientific truth is not established by the work of one man. Other scientists must make similar observations and obtain like results before these results can be accepted as fully demonstrated. It is only necessary to recall the controversy over the supposed discovery of a cure for consumption by Dr. Koch, of Berlin, to make clear how ex- acting the scientists are, and how difficult it is to establish a new truth bej^ond the possibility of doubt. Is the historical process similar to this? Not at all. It is quite different. History deals with the past. It may be the past of this morning, of the war with Spain, or of the Persian wars 8 HISTORICAL METHOD. of Greek history, but it is always the past. Yet it is not with the past in an indefinite way that history has to do. It is with the past in which human society has developed and the problem with which it deals is this: How has the present complex, world-society been evolved from the primitive, disconnected beginnings of four or five thousand years ago ? It is its busi- ness to reconstruct the process and to describe the successive steps. When it has done that, it has done its whole duty. But how does it perform this duty ? What is its method? It cannot be the method of direct o})servation, the method of the natural sciences, and the reason is very plain; the objects are not here to be observed. The past can ])e known to us only through its records, technically called the sources. These sources are of two kinds; material remains and traditions. The remains are all of those things that were actually part and parcel of the life of past generations; bodies of men, clothing, weapons, houses, roads, bridges, newspapers, letters, coins, etc. This subdivision of the sources will be better appre- ciated if a list 1)6 m:ide of the material o))jects that will form the sources for the history of our own society. One of the characteristics of modern historical method is the increase in the variety of the source material. Some of our most valuable information is drawn from ma- terial that past generations never thought of putting to such a use. The other main division of source material is tradition. It is of three kinds; oral, written, HISTORICAL METHOD. 9 and pictorial. Pictorial tradition has attained a great significance in our generation and a great value that it did not formerly possess. This change is due to photography. For the historian of the Napoleonic era, the great war scenes by French contemporary artists are of little value; while for the historian of our war with Spain, the snap-shots taken by the camera of a war correspondent will form the most valuable source material. Oral tradition is the least reliable of all. It is the account of an event that has passed from lip to lip and has been handed down from one generation to another. It soon becomes utterly unreliable and worthless, although it may have been very valuable when it came from the lips of the eye witness. The written tradition, upon which the his- torian chiefly relies for his knowledge of the thoughts and acts of men in the past, if it be a source, contains the record of what has been seen or heard by an eye or ear witness. This is the material with which the historian works. He observes it directly, it is true, but what he observes is not the event, not the object, but the record of an observation made upon that object. And what an observation it often is! Made, perhaps, by an incompetent per- son, who, at the time, had no intention of record- ing it, it is onesided and incomplete, and written down so long after the event that what little value it originally had has been materially im- paired, if not wholly destroyed. Add to this, the fact that it is expressed in unscientific 10 HISTORICAL METHOD. language, and some of the difficulties of the problem will be clear. What would be the value to the chemist of a series of experiments, if, — to borrow the illustration of a French writer, — his knowledge of them were based upon the accidental observations of the janitor of the laboratory ? Not only, then, is the his- torian of the remote past unable to observe the events directly, "but it is very rare that the documetits of which he makes use contain ex- act observations. He cannot, moreover, make use of the records of observations scientifically established, that, in the other sciences, may, and often do, take the place of direct observa- tion." His method must be that of indirect observa- tion. He starts with the record and attempts to work his way back to the fact, to see the fact as the observer saw it. The fact is the goal of his eflPorts, not the starting point, as in the work of the natural scientist. The docu- ments that form the starting point for the his- torian are nothing more " than the traces of ps\'chological operations." In order to infer from the document the fact that gave rise to it, the student of history must retrace the whole eeiies of psychological operations that lay be- tween the fact and the written record of the observation, retracing them in the inverse order, beginning with the document. The object of the procedure is to establish the genuineness of the document and the value of the observations. If the document is not genuine, we need not take it into account; and HISTORICAL METHOD. H an observation is practically useless until it has been localized, that is, until we know lohen it was made, tohere it was made, and by whom it was made. From this crucible of criticism, the contents of a document come forth separated into single affirmations, each affirmation bearing the mark of its value. Ihis is the foundation work that places in the hands of the historian observa- tions similar to those possessed by the scien- tist, but seldom, if ever, as exact or as valuable. The work of historical criticism is extremely difficult, hut ahsolutely necessary. It is, however, the portion of method to which the least attention is paid in our colleges, although it is the best developed part of metho7); Lorenz ''Die Geschiclitswissenschaf t " (1886, 1891); of Dolci, the Italian, " Sintesi di scieiiza storica" (1887); and of the Englishman Free- man, ''The Methods of Historical Study" (1886). Up to 1889, these were the most important treatises that had appeared on method. They dealt with the subject in a summary way — many of the works being only pamphlets — and often treated only parts of method instead of the whole. There was need of a work that should gather up these partial results, combine them, and attempt to present them in a systematic and detailed manner. Such a work was puo- lished by Bernheim in 1889. The title is " Lehr- buch der historischen Methode." It contains six hundred pages and describes in detail all the steps in the construction of an historical narra- tive. The book marks an epoch. For the first time a real text-book on method had been pro- duced. In 1897 a more popular work was pub- lished in France by Langlois and Seignobos, en- titled, "Introduction aux etudes historiques." Although the work does not pretend to be an exhaustive treatise like that of Bernheim, yet certain divisions of the subject are dealt with in a much more satisfactory manner and really supplement the work of Bernheim. Besides these two hand-books treating of the whole subject, many monographs, or partial studies, have been published, so that the litera- ture upon method has become one of quite re- spectable size, and can not be neglected . by any serious student of history. 18 HISTORICAL METHOD. But what is the result of all this study by so many centuries of historians ? A conscious operation in the treatment of historical material, an understanding of what has already been accomplished, and a pretty fair appreciation of what remains to be done. As yet, the form in which the results are pre- sented has not been fixed by tradition; but there is a quite ffeneral agreement as to the subject matter and order of arrangement, although there is some disagreement as to the nomencla- ture to be employed. Beruheim, after an introduction dealing with such questions as the definition of history, the relation of history to other sciences, and the pos- sibility of attaining scientific certainty in his- torical study, divides his work into four parts: (1) Quellenkunde, treating of bibliography, source collections, and the auxiliary sciences; (2) Kritik, treating of the genuineness of the sources, their origin and value, of the estab- lishment of historical fact, and the arrange- ment of the facts established: (3) Auffassung, dealing with the interprctati )U and grouping of I'acts, with their physical, psychical, and so- cial environment, and with the philosophy of history; (4) Darstellung, or the formulation of the results obtained in the preceding investi- gation. The grouping of Langlois and Seignobos is somewhat simpler. Their work is divided into three parts: (1) Les connaissances pr^alcihles^ or preliminary knowledge, equivalent to Bern- heim's Quellenkunde; (2) Operations analyt- HISTORICAL METHOD. 19 iques^ embracing criticism, interpretation, and establishment of the facts; (3) Oph^ations syn- tMtiques^ or combination of the facts and con- structive reasoning together with the presenta- tion of results. There is one important difference between the arrangement of Bernheim and that of Lang- lois and Seignobos; in the first, interpretation follows the establishment of the fact; in the last, it precedes it. With that exception, there is substantial agreement in the arrange- ment of the two works. It would be safe to say then, that, whatever title may be given to the parts, a work on method naturally falls into three or four parts; four, if the narrative, or presentation of the re- sults, forms an independent division. A moment's thought will show that all this is nothing more than a careful description of the procedure of the student of history from the time that he selects his subject for investi- gation until he commits the results of this in- vestigation to paper. It is my intention in thia chapter and the following to sketch rapidly the successive steps in this procedure as they are described in the works just referred to. I hope that it may be helpful to teachers that have not access to these works or who would be miable to read them. If they would draw the greatest benefit from this study, let them fol- low the process step by step, investigating some historical topic in accordance with the method described. Let them repeat the proc- ess again and again, and careful scientific work will soon become second nature. 20 HISTORICAL METHOD. The rest of this chapter will be devoted to whiit Bernheim calls Quellenkunde and Lanorlois and Seii^nobos, Connaisance prkalahles. Sources were defined in the preceding chap- ter, and Bernheim's classification of them under remains and tradition was given. It is clear that, if there are no sources, no history can be written. If a student is desirous of investi- gating a subject, he asks himself the questions: ''Are there any sources? What are they? Where are they?" If there are no sources, the subject, however interesting, can not, of course^ be investigated. Great masses of source material are being destroyed in various ways every day. On a recent tour of investi- gation in France, I learned in two places, at 8t. Martin, on the He do Re, and at Saintes, on the neighboring mainland, that valuable archives, containing sources for which I vras seeking, had recently been destroyed by fire. It is a common thing in the course of an inves- tigation to run across traces of sources that once existed and perhaps exist to-day, l)ut can not be found. Often sources are known to be hidden in private archives, to which access is denied. But even when the student knows that sources exist and where they exist, his work is often rendered diflicult by the fact that his sources are scattered and a use of them would oblige him to make long journeys. His work will be lightened if a government has acquired all of this material and placed it in a central depot. It will be lightened even more if this manu- HISTOKICAL METHOD. 21. script material has been published and he can study it comfortably by his own fireside. While the study of written tradition may thus be made easy, there are certain kinds of source material that can be studied only upon the spot. An exact copy of a manuscript may be studied even m^re satisfactorily than the manuscript itself, but neither photographs of an historical spot, nor descriptions of it, nor both, will do for the student what direct observation will do. But whether he can only study at home or can also go abroad, it behooves the student of history to make the acquaintance of the great source collections that have been published by governments, associations, and individuals. The contemporary histories of Greece and Kome have been carefully edited in ihe original Greek or Latin, and also translated into English. The Greek and Roman inscriptions have been gathered up from every side, carefully restored and published. Hundreds of specialists are en- gaged in making public the Latin sources of the Middle Ages, and the sources of the later periods composed in the language of the vari- ous peoples. Some periods have been thor- oughly worked, while others are still almost virgin soil. So difficult is much of this work, so nice and varied the skill required of the worker, that many men do nothing but this: they simply prepare the sources that others may make use of them. Historical work is be- coming every year more differentiated, and to make it successful the heartiest co-operation must exist among the workers. 22 HISTORICAL MKTHOD. The source collections of which I have boon writing- are made up of complete documents, narratives, etc. There are other source col- lections of a more elementary character, com- posed of short typical documents and of extracts from narrative sources. These are for the use of beginners. The new method of history work has called into existence a large amount of this material. From Harvard Uni- versil}- have come exlraels and documi'iits on United States History; from the University of Pennsylvania, " Original Sources of European History;" from the University of Michigan, sources of English History; from the Univer- sity of Indiana, sources of European History; and from the University of Nebraska, sources of European and American History. But suppose that there are sources and that they are accessible, how does the student learn what they are and where they are? It is the work of bibliography to tell him this. After the subject for investigation has been selected, his first step is to seek for a book that will answer these two questions for him. Such a work is not always to be found. Bibliogra- phy is not in an advanced stage of develop- ment. The larger number of works upon which the student must de{)end are out of date and others are thoroughly unscientific In many of them, no distinction is made between sources and narratives based upon the sources, and, for the most part, when the sources are enumerated there is nothing to indicate their contents nor the value of the contents. The HISTORICAL METHOD. 23 most of this work the advanced student 18 obliged to do for himself. Historical study ■will be much easier when good bibliographies have been prei)ared. Although he may have learned what the sources are that he needs, the student is often in ignorance of the whereabouts of his sources, especially if they consist of rare printed books or manuscripts. Here bibliography might help him, but it seldom does. The large and wealthy libraries ought to have the books and certain archives should contain the manuscripts. But books and manuscripts are not always where they should be, and even when they are they are very often not catalogued. Yet however incomplete these bibliographi- cal aids are, they are all we have and are im- proving rapidly each day. The student that does not know how to make use of them will find himself badly handicapped. A most help- ful little book upon historical bibliography was recently published in Paris. The author is Langlois and the title Manuel de hiUiograjyhie histoinque. When the student, through the use of bibli- ography, has succeeded in reaching the sources, he finds that his work can not go on without the use of one or more auxiliary sciences. It may be a manuscript that he has before him, and it may be incumbent upon him to deter- mine its genuineness before using it. The per- formance of such a task would call for a knowl- edge oi palaeography^ or the science of writing, of diplomatics^ or the science of documents, and 2-t HI TORICAL MKTIIOD. perhaps several otl; rs. If it is known that the document is genuine, the student must at least have a knowledge of the language in which it is written in order to interpret it. For some periods, such a knowledge is not easy to acquire. The investigator in the tields of Grecian, Roman, or Mediaeval History must have a knowledge oi pMlologij^ or the .science of language. He must be acquaint d with all the changes that take place in the meaning of a word in order to understand how it is used at a particular time. When the student comes to criticise his sources, and to determine their value, he tinds that a knowledge of psychol- ogy is necessary; in arranging his facts, he must make use of chronology ; in coml)iniug them, of logic; in forming the background, he is aided by geography^ ethnology^ economics^ and sociology; and in searching for the deeper meanings of historical development, by philos- ophy. These are the most important of* the auxiliary sciences. There are, of course, many others, determined by the peculiar nature of the subject investigated. It would appear, then, that historical inves- tigation is neither easy nor simple. And why should it be? It has to do with the most dif- ficult and complex of subjects — the evolution of man in society. We are just coming to a real- ization of the magnitude of the task to be accomplished in correctly tracing this evolu- tion, and of the only way in which it may be accomplished. The uninitiated are accustomed to sneer at the specialist in history who con- HISTORICAL METHOD. 2o fines himself to a limited field and works it thoroughly. But it is the sneer of ignorance. Such special iztilion in the natural f-ciemes is taken as a milter of course. Wc must learn that the same reasons make specialization im- perative in historical sciences. AVithout spe- cialization, we can not advance. Special study and comprehensive views of history are not irreconcilable things. Every scientific investigator will not only know first hand the results obtained in his own part of the field, but he will know second hand the results obtained in other parts of the field. Speciali- zation can be dangerous only when the special- ist fails to keep in touch with the greater whole of which his work is only a part. If the student, supplied with the necessary knowledge of the auxiliary sciences, has been able, through the aid of bibliography, to find the sources that he seeks, his next step will be to decide how much of these sources can be ad- mitted as evidence on the subject under inves- tigation. To settle that question is the province of Criticism or Kritik. CHAPTER III. EXTERNAL CRITICISM; GENUINENESS OF THE SOURCE. BERNHEIM'S KRITIK, the second division of Method, covers practically the same _^round as Lanojlois, and Seignobos' Op- ^rations analytiques. The subdivisions of the former are External Criticism, Internal Criti- cism, and Critical Arrangement of the Material; of the latter. External Criticism and Internal Criticism. External Criticism, Bernheim subdivides into: Testing the Genuineness, Localization of the Source, and Editing; Langlois and Seignobos, into Criticism of Restoration, Criticism of Ori- gin, Critical Classitication of the Sources, and Criticism of Erudition and the Erudites. The ground covered in both works is practi- cally the same, Bernheim being, of course, more technical and detailed, while Langlois and Seignobos, in their interesting chapter on "La critique d'erudition et les drudits," deal with a subject not treated by Bernheim, or, rather, treat it from a different point of view. In this chapter, I shall consider the first sub- division of External Criticism, the Testing of the Genuineness of the Source. The first question that the historian puts to the sources that he has brought together is *' Are they genuine? Or, subdividing the ques- tion, he asks, "Are they what they appear HISTORICAL METHOD. 27 to be?" (forojery), and "Are they what I think they are?" (self-deception). In the first case, the trouble lies with the source; in the sec- ond case, with the historian. A lack of criti- cism in the first case would lead us to use material that should not be used; ii lack of criticism in the second case, or it may be hypercriticism, would cause us to reject material that should be used. The historian should guard against these two errors. Man is naturally credulous. It is much eas- ier to believe what we hear than to sift the evi- dence in order to find out the truth. This last process is so unnatural that few men will un- dertake it unless it is absolutely necessary. Criticism is often a thankless task, for its re- sults are frequently negative, forcing the his- torian to throw aside as worthless what he has gathered with so much difficulty. The critical attitude toward the sources has been a product of time. Although it has reached its fullest development in our day, there were historians among the Greeks whose atti- tude was in some respects strikingly modern. Speaking of the credulous spirit, Thucydides said (I., 20): "For men receive alike without examination from each other the reports of past events, even though they may have happened in their own country. * * •• With so little pains is the investigation of truth pursued by most men; and they rather turn to views al- ready formed." Referring to his own methods of investigation, he wrote (I., 22): " But with regard to the facts of what was done in this 28 HISTORICAL METHOD. war, I do not presume to state them on hearsay from any chance informant, nor as I thoiiifht probable myself, but those at which I was per- sonally present, and, when informed by others, only after investijyating them accurately in every particular, as far as was possible." Many passao^es from the histories of Polyb- ius (1., 14; XII., 17-22) show that his attitude toward the sources was decidedly critical. But I recall nothing that would indicate that either of these writers carried their skepticism so far as to doubt the genuineness of the material that fell into their hands. They dealt more with what we call to-day Internal Criticism. Even here Thucydides Avas not consistent, but at- tempted to make a rational narrative out of the myths of the Iliad, gravely discussed the reasons for Agamenmon's leadership in the Trojan War, and knew the contents of the sealed letter sent by Pausanias to the Persian king, Xerxes. In a word, the critical method was not thoroughly conscious and scientific. The Greeks left us nothing in the writing of history but the work of Lucian, referred to in the precedhig chapter. The Romans did not accomplish as much as the Greeks, and the man of the Middle Ages was incapable of doing crit- ical work. With the Renaissance, the forward movement began again and from rational criti- cism the scholars of the f ollownig period passed rapidly to h3'percriticism. In the latter part of the seventeenth century, the Jesuit Harduin, dis- turbed by the large amount of forged material that he encountered, went so far as "to deny HISTORICAL METHOD. 29 the entire foundation of our historical knowl- edo-e, and to reject as forged a long series of historical works and documents: Pindar, Thucydides, Dionysius, Diodorus, Strabo, Jo- sephus, Varro, Livy, Terence, Virgil, Horace, Eusebius, Cassiodorus, etc." A reaction naturally set in against this ex- treme view, leading to the present rational attitude of carefully testing all material and "holding fast that which is good." This is the Bolid foundation of External Criticism, upon which modern historical science is built. But it is not surprising that the man of the Renaissance were led into hypercriticism. The highways and byways of history are slrewn with forgeries. Every kind of source material can point to its famous examples. vSome of these cases are well known to others than the special student of history. A long lisi of forgeries could be made under the hojid of Kemains. This practice of fabricat- ing relics of the past and, for various reasons, passing them off as genuine has been continued down to our own day. Two of the most famous of th-ose perpetrated in the nineteenth century are described by Bernheim; the first was the ]Moabi!e pottery, the second the Sardinian literature, or "Parchments of Arborea." After the discovery, in 1S66, of the Mesa stone with its invaluable inscription, in the land of Moab, there appeared for sale by a dealer in antiquities at Jerusalem certain old Hebrew in- scriptions similar to that on the Mesa stone. In the spring of the year 1872, there appeared 30 HISTORICAL METHOD. at the same place certain pieces of pottery and later in the year vases, urns, etc., with inscrip- tions and drawings, 2,000 pieces in all. The articles were brought to Jerusalem by an Aralj, Selim, who had been in the employ of Euro- pean excavators. The dealer in Jerusalem was charged with fraud, and, in company with those interested, went to the place indicated by Selim and found other articles of the same nature. Although criticism was not silenced, many of the articles were bought, at the advice of Ger- man savants, for the Berlin Museum. Careful criticism has shown that the articles are counter- feits and that the work was probably done by the Arab Selim. The Sardinian forgery is even more interest- ing. In 1863-65, there was published in Italy a series of letters, biographies, poems, and other literary fragments, supposed to have been composed in the island of Sardinia in the period from the eighth to the fifteenth centuries. The original manuscripts were of parchment and paper. The find created a great sensation, for it was not known that such a state of culture had ever existed in Sardinia. The originals, after publication, w^ere deposited in the library at Cagliari. As a heated discussion had arisen in Italy over the genuineness of the material, some of the originals were submitted to the Academy of Sciences at Berlin for criticism. Jaffe in- vestigated the material of the manuscripts and the handwriting; Tobler, the language and literature; Dove, the historical contents. They established, beyond the possibility of a doubt, that the material was forged. HISTORICAL METHOD. 31 The Forged Decretals, the Gift of Constan tine, the poems of Ossian and Chatter ton are forgeries known to every school boy. Marie Antoinette suffered much at the hands of the forger. The historian of the French Revolution who attempts to write the life of *^^his unfortunate woman is confronted at the very thres-hold of his work with the question, " How many of the letters attributed to her were really written by her? " The famous collections of her letters by Feuillet de Conches and Count d' Hunolstein contain a great mass of forgeries. A glance at the introduction to the first volume of the collection by La Rocheterie and DeBeau- court (Paris, 1895) w^ill give some idea of what a Herculean labor the determination of the genuineness of the material may become. In 1895 a work entitled, ''The Journal of a Spy in Paris during the Reign of Terror," pur- porting to have been written in 1791 by one Raoul Hesdin, was printed in London by the reputable firm of John Murray. The editor did not give his name, did not state where the manuscript was found, nor where it could be seen by the skeptical. The wx)rk received little attention on this side of the Atlantic. The American Historical Review (July, 1896) remarked that "the unsatisfactory point about The Journal is that no evidence is given of its authenticity," but no attempt was made to prove by a study of its contents that the work was a forgery. This was successfully undertaken by the English Historical Review in the July number of 1S96. It is a good example •yl HISrORICAL METHOD for the .student nf history to study. The work "was shown to be a forgery. This case is the more interesting as the anon- ymous editor attempted to defend himself by anonymous letters written to the Athenaeum. Although the work is a forgery, it is a clever forgery, and it would be well worth the while of the historical student to give it some study. The absence of the manuscript rendered, of course, the work of detection much more diffi- cult than it otherwise would have been. The question of the authenticity of the so- called ''Casket Letters'' of Mary Queen of Scots is still an unsettled question. The ap- pearance of the M^moires of Talleyrand a few years ago raised a discussion upon their genuineness that lasted for more than a year. The manuscript of Talleyrand was not to be found; it had probably been destroyed. The existing manuscript was a copy made by Ba- court. This gentleman had formerly edited the correspondence between Mirabeau and De Lamarck, and had taken great liberties with it. This rendered the critics suspicious, and they were naturally desirous to know why the origi- nal manuscript had been destroyed and how much of the M6moires was the work of Talley- rand and how much the work of Bacourt. They will probably never know. In 1897, the English Historical Review and the German Historische Zeitschrift contained interesting critical articles on a series of secret reports on the French Revolution published in the Dropmore Papers. HISTORICAL METHOD. 33 The M^moires de Weber on the French Revolution is largely the work of Lolly-ToUen- dal, and it is claimed that the Comte de Segur wrote the Memoires de Besenval. And so the list miofht be continued indefinitely. Besides the injury done by treating forged material as if it were genuine, as great an injury may be done by treating genuine material as if it were forged. Bernheim gives a number of inter- esting illustrations of this kind of error. The mistake is due to ignorance. During the first half of this century, quite a number of mediae- val sources were set aside as forgeries, but have since been recognized as genuine. Enough has been written, I take it, to make clear the necessity of testing the genuineness of sources before using them. It is now in order to say a word about how this is done. Apart from the genius that characterizes the most successful criticism, the indespensable preparation for this work is the acquisition of a fund of detailed knowledge concerning the source material of the period in which the for- gery is supposed to have originated. Such a fund is not the property of the novice, and only the veteran knows how diflicult of acquisition it is, how much time and patience and skill are ex- pended in securing it. The investigation of the genuineness of a source is little more than a series of compari- sons systematically conducted. The suspected source forms a part of the remains that have come down to us from some previous age. If it be genuine it will be in harmony with all the 34 HISTORICAL METHOD. other sources of that period and bear the marks common to all the culture products of that aofe. A simple statement of this fact will make clear that as difficult as it may be to de'ect a forgery, it is even more difficult for a fortifery to escape detection if the critic possesses an adequate knowledge of the period concerned. The critic deals first with the form of the document, with the writing, language, style, and composition. Palaeography and philology have obtained such a development that he would indeed be the prince of forgers who could successfully imitate the language and writing of past ages and deceive the critics. To reproduce successfully the style of a certain man of a certain age, would be even more diffi- cult were it not for the fact that this part of criticism has been but little developed. The success of this sort of thing is due to the ignor- ance of the critic rather than to the skill of the forger. A growth of knowledge and of method will reverse these relations. After dealing with the form of the source, the critic turns to the contents. (1) Do they agree with what we have learned from other genuine sources of the same age and place? (2) Is the writer ignorant of things that a writer of that day would have mentioned ? (3) Is he ac- quainted with events of which he jcspuld not have known at the date of writing? Oiil;hese three questions, the second is the most difficult to answer. If a forger passed unscathed the ordeal of one and three, it would be rather dif- ficult to convict him under two. It is the so- HISTORICAL METHOD. 35 called argument from silence, and is often used in a most unscientific manner. It is almost im- possible for a forger to escape the test of three. Every man is a child of his own time, and it is practically impossible, in dealing with an ear- lier period, to conceal his personality. It was through his knowledge of later events that the forger of " The Journal of a Spy" fell a victim to the critics. If the source passes the tests of outward form and of content, the critic then asks if the in- formation drawn from the source fits naturally into the chain of historical evolution as we know it. A successful answer to this ques- tion can be given, of course, only by a master of the period. Besides these main tests, there are others that may not be decisive in themselves, but that sup- ply us with cumulative evidence. Such tests are found in the peculiar conditions under which the source was discovered, the use, by the forger, of documents or other records that could not hava been known to him at the time when the record was supposed to have been made, and the de- tection of certain prejudices in the source that might explain the object of the forgery. From what has been said, it ought to be clear that clever cases of forgery can be detected only by experienced critics. If the source material stands the test and is clearly genuine, the historian takes the next step in External Criticism by attempting to localize the source, that is, to tell when the ac- count was written, where it was written, and who the author was. CHAPTER IV. EXTERNAL CRITICISM: LOCALIZATION OF THE SOURCE. HAVING decided that the material is gen- uine, the historian has to deal with the further question, "Shall it be admitted as evidence? " A reply to this question is pos- sible only when tlie material has been localized. Kow sources are the results of human activity, either destined originally to serve as proofs of historical events or fitted to serve as such proof because of their origin and existence. The first class of sources constitutes tradition; the second the historian styles remains. If, then, the events are to be restored by means of the remains and tra'litions, it is perfectly clear that the historian must know with certainty to what events the remains belong, and that the tradi- tions actually come down to us from individ- uals who were themselves participants in the events or at least eye-witnesses. "It would be absurd to seek information upon an event in the writings of one who knew nothing about it and w^as not able to know anything about it." The historian must know, then, when the source originated, where it originated, and who the au- thor was. " A document whose author, date, and place of origin are totally unknowable is good for nothing." When these questions have been answered, the source has been localized, and the historian knows whether it may be ad- mitted as evidence or not. The further ques- HISTORICAL METHOD. 37 tion as to what this evidence is worth is dealt with by Internal Criticism. It is evident that the work of localizing the source is closely connected with that of deter- mining its genuineness. A source might claim to be the w-ork of a Frenchman, living in Paris in the year 1794; in proving it to be a forgery, we show that it is the work of an Englishman, living in London, in the year 1895. It is the object of the investigation that marks the differ- ence between the two processes. In the first, we wish to know if we are dealing with a genu- ine document; in the second, if the document, through its origin, could contain the evidence that we seek. A false document, claiming to contain evidence, would, when localized, be of no value; a genuine, but unlocalized, source, might, when localized, prove to be of no value for our investigation. If the document is genuine, and the name of the author, the time and place of writing are all given, there is no need of an investigation. A distinction should, how'ever, be made between the time of writing, and the time of printing; the place of writing and the place of printing, the author of the title page and the real author. If these two sets of facts always agreed, the work of investigation would be rendered much easier. Setting aside, for the present, these latter problems, let us consider the more diffi- cult ones; how is the origin of a source — written tradition, for example— ascertained when there is no title page indicating the author and the time and place of writing? 38 HISTORICAL METHOD, The determination of the date of a source is often a very difficult matter. It is especially difficult when it must be determined by the con- tents of the document controlled by general information. In the first place, we endeavor to locate the source in a general way by a study of form, language, style, and contents. In this way, we place it in a certain century or even generation. Here palaeography and philology are of use to us. If our source is in the form of a manuscript, palaeograph}' tells us it was written in such or such a century. Even the school-boy knows how writing changes from generation to generation. Men of thirty-five, living to-day, have had experience of three gen- erations of penmanship; the style of their fath- ers, the so-called Spencerian style, and the lately introduced upright style. Modifications of a similar nature characterize the whole his- tory of writing; a knowledge of this history enables the palaeographists to locate the manu- script approximately. The history of language lends its aid, and this may be employed with printed sources. Words and expressions are born and die. The philol- ogist tells us that a certain word appeared for the first time in a language in a certain century. If the word appears in the given source, it must have been written after that century. He tells us, also, that a certain word disappeared from a certain language in a given century. If the word appears in the source, the record must have been made before the disappearance of the word. HISTORICAL METHOD. 39 But often neither palaeography nor philoloo^y, nor even style can do more than locate tho source in the first or second half of a century. How can the date be fixed more definitely ? Here we must depend largely upon the con- tents of the source. References to events, known to us from other sources, prove that the record was made after the events took place or that it arose at the time of the events. That is shown by the manner in which the events are referred to. Writing upon the events of July, 1TS9, Bailly said, "If M. Barrere had been lis- tened to, many things accomplished by time and accident would not have happened, the revolu- tion would have been less complete; but we should have been saved from the anarchy to which the constitution has been exposed and is Btill exposed (Today, 23d of February, 1792)." Although the work is printed in the form of a diary, kept from day to day in 17S9, it is evi- dent, from remarks like the above, that it was written several years later. Brissot's M^moires offer an excellent oppor- tunity for a study upon the date of writing. Even in the first volume, dealing with his early life, there are repeated references to events that took place in the last years of his life. On one page he refers to "my pamphlet of the month of October, written against the factions of Marat and Robespierre," and to "the choices made by the sections of Paris for the National Convention." So every record made by an eye-witness, but made some time after the events, is likely to 40 HISTORICAL METHOD. supply some such clue, as the above, to the date of writing. The use of such expressions as "up to the present time," "at the date of writing," or references to the results of certain acts that are being described, are helpful to the historian. The failure to mention events that the wit- ness undoubtedly would have mentioned had he known of them is also helpful. This is the so- called argument from silence. There are great dangers connected with its use. The reasoning is, " Because the witness does not mention this event, the event never took place." For this reasoning to be valid, it is clear that all of the events must have been recorded and the records preserved. If the witness did not record all the facts, or if any of the records have been lost, the reasoning would be false. It would seem to be evident, then, that this argument can be employed only in certain clearly delined cases, namely, when (1) the " witness desired to note systematically all the facts of a certain kind and was acquainted with them all; and (2) when the fact, if it had existed, would have made such an impression upon the mind of the witness that he would have been forced to re- cord it." Sometimes a single reference is suflScient to fix the date; often the procedure is more diffi- cult, and the historian must determine the limits within which the record was made. The one limit is called the terminus post quem^ or the limit after which the source must have originated: the other limit is the terminus ante HISTORICAL METHOD. ^1 queni^ or the date before which the record must have been made. The following excel- lent illustration is given by Bernheim of the search for these two limits: One of the annals describing the period of Charlemagne and his predecessors, and written contemporaneously, treats the years from 741-791. It is seen at once that it was not written year by year, l)e- cause frequent references are made to later events. Assuming that they were not put in at a later date, we may make use of them to tix the terminxLS post quern. The latest event men- tioned that may be used in this way happened in the year 1781. In speaking of the Duke of Bavaria,Tassilo, who had been conquered and had promised submission, the annalist writes, "But the promises that he had made he did not keep long (Sed non diu promissiones quas fecerat conservavit). It is evident the writer knew of the subsequent revolt of Tassilo in 1788 This is the terminus post quern. There is but one reference that gives assist- ance in establishing the terminus ante quein. In 785 the annalist writes: "And then all Saxony was subdued " (et tunc tota Saxonia subjugata est). He would hardly have writ- ten like this had he known of the breaking away of all Saxony from the rule of Charle- magne in 793. This, then, is the terminus ante quern. The work was written, if this reason- ing be sound, between 788 and 793. The determination of the place of origin is often more difficult than the fixing of the date of a source. The place where the record was 42 HISTORICAL METHOD. found, and the imprint, maj or may not help ua in the investi<^ation. A manuscript written on the island of St. Helena may be discovered in the United States, and the place given on the title page may have been intended to mislead the censor, when a censorship of the press ex- isted. In dealing with French works printed before the Revolution, it is never safe to accept without investigation the place of publication given on the title page. The writing, if it be a manuscript, or the language may aid us, in case that we have the original language. The subject matter furnishes the most valua- ble evidence. How this material may be used will be best shown by another example from Bernheim. In the early part of this century, there was discovered in the monastery of St Michael at Liineburg a few sheets of parch- ment manuscript containing annals for the years 1057-1130. iSleither the name of the author, nor time and place of writing, were given. The part from 1100 on was clearly the work of a contemporary. Where was it writ- ten ? The handwriting was of the twelfth cen- tury, but showed no local characteristics. The same was true of the language, that was the Latin of the twelfth century. The place of discovery might point to Lower Saxony as the place of origin, bat not without further proof. An examination of the contents showed that the part from 1100 on bore the stamp of unity. Saxon events are treated in great de- tail, while events taking place in the rest of Germany, even when important, are simply HISTORICAL METHOD. 43 mentioned or not referred to at all. Changes in the bishops of different bishopries occupy much space, and the writer is especially inter- ested in the bishoprics of Magdeburg, Bremen, Halberstadt, and Merseburg (Saxon bishop- rics). The most of the princes, whose deaths are mentioned, are Saxon, and the writer as- sumes that when he refers simplv to the "Markgraf Rudolf" or the "Graf Friedrich," the reader will understand him. The deaths in the f imily of the Count of Stade are given regularly, and the writer as- sumes that his readers are acquainted with even these relatively unimportant lords. ' ' Udo comes" (Count Udo) is the regular form of reference. So great is the interest in this family that in the midst of the account of the struggle between Henry IV. and his sons, the annalist breaks off his narrative to note that '"Count Linderus, .with the surname of Udo, was taken sick, was brought to the cloister of Rosenfeld, and died there." The mention of this cloister in connection with the Count of Stade is an important clue. Investigation shows that the cloister of Rosenfeld is lo- cated in the land of the Count of Stade, that it was founded by the Counts of Stade. Who, then, would be so much interested in the Counts of Stade as a monk in the cloister of Rosenfeld, and who wrote his annals for the circle of readers about him? And a notice from the year 113C> points unmistakably to the cloister of Rosenfeld as the place where the annals originated. " Cono abbas obiit" (the 44 HISTORICAL METHOD. abbot Kuno died) runs the record. Only in the monastery where the annals were written could a reference like that— a reference that did not orive name of the monastery over which Kuno presided— be understood. From other sources, we learn that Kuno was the abbot at the head of the cloister until 1130. This was clearly the place where the annals were written. The determination of the authorship of a source is of the greatest importance, Not that we may simply know the name of the author, do we seek this information, but that we may know what kind of a person he is and what his position in society is. Only in this way can we determine what his testimony is worth. This information might be made use of even when we did not know the writer's name. The most common means of determining the authorship of a source is to compare it with other sources. Here the knowledge of time and place of origin is of value, as it enables us to limit the body of sources with which we work. If we have a manuscript and know it is gen- uine, we may compare it with other manu- scripts of the same period. In modern history, where distinguished men have left large quan- tities of manuscript material behind them, their hand-writing is well known, and it is easy to locate a newly discovered manuscript. In the Middle Ages, the work is more difficult, for there is less individuality in the hand- writing and less material for comparison. HISTORICAL METHOD. 45 To determine the authorship of a source by a comparison of its style with that of other con- temporary works is a diiScult undertaking. All the world is acquainted with the attempt to prove in this way that "Bacon wrote Shake- speare's plays. But this sort of thing is gener- ally unscientific in two w'ays. In the first place, the investigators forget that all the writers of a given generation will have much in common, and when we prove that an anonymous work has, in the matter of style, much in common with the work of a known writer, it may he possible to establish further that these common things are found in the works of all other writers of that period. In the second place, the inves- tigators have lost sight of the fact that an au- thors style changes; it changes as he grows older, as he treats different subjects, as he ad- dresses different classes of readers. Much time has V)een wasted in purposeless work of this kind, and although some progress has been made, this part of method is in a very unsatis- factory condition. It often happens, when handwriting and style fail to give definite results, that the au- thorship may be settled in other ways. Fre- quently references made by the writer to him- self, to his interests, occupations, position in life, and persons with whom he is associated point clearly to some known persons whose sur- roundings correspond to those indicated in the sources. Such a piece of work could, of course, be successfully carried out only by a historian possessing a large fund of information oa the 46 HISTORICAL METHOD. period of histoiy to which the source belonged. Sometimes we are aided in determining the authorship of a work by references to it in other works, where the reference is coupled with the author's name. At times these works give exact quotations that are found literally in the anonymous work. As was pointed out above, the important thing in the determination of the authorship of a source is not simply to learn the author's name, but the author's personality. To know that the writer of a certain source was an un- known A or B, and to know nothing else, would profit us little. If the source furnishes us abundant information upon the personality of the writer, it is of no value to know his name, unless the person be well known, and a knowl- edge of this name will enable us to obtain else- where further information about him. An example from Bernheim will illustrate the methods by which the authorship of anony- mous sources may be determined. One of the most important sources for the close of the ninth century in Germany is the chronicle of Regino, the Abbott of Priim. This is continued from 907-967 by an unknown writer. He evidently worked in the sixties, making use, at first, of other annals and, later, writing more independently and treating the subject more in detail. From the interest that he betrays for the cloister of St. Maximin at Trier, it is evident that he belongs to this cloister; the events enumerated are such as only a resident would be likely to take note of. HISTORICAL METHOD. 47 Now the resident of a cloister engaged in literary- work, could have been none other than a monk. This tirst inference gains support from the fact that the first part of the chronicle was written in St. Maximiu where Rogino took refuge after being expelled from Priim. Among the few persons of the cloister named by the writer of the chronicle, one, Adalbert by name, is especially prominent. In 961 it is stated, that at the instigation of the Arch- bishop of Mainz, "of whom Adalbert might have expected something better," the monk was sent as a wandering preacher to Russia. He was fitted out for the journey by the king. In 962, he returned from his bootless mission, passing through great dangers and receiving a most hearty welcome at home. The writer is so well informed upon the adventures of Adal- bert and speaks of them with so much feeling that he must have been on intimate terms with him, or he must have been Adalbert himself. For this last presumption there is considerable evidence drawn from what we know about Adal- bert from other sources. In 966, he was made Abbot at Weissenburg, and in 968 became Arch- bishop of Magdeburg. From his career, it is evident that he was an educated man. The writer of the chronicle shows by his language and the character of his narrative that he pos- sessed a culture not common in that day. The chronicle mentions the transfer of Adalbert to Weissenburg and breaks off with the year 968, the year when he was raised to the archbishopric. It may be said, then, "with the greatest prob- 48 HISTORICAL METHOD. ability, if not with certainty, tliat Adalbert was the writer of the continuation of Regino's chronicle." Such are the problems to be solved in the lo- calization of a source, and such are some of the methods of solving them. CHAPTER V. EXTERNAL, CRITICISM: ANALYSIS OF THE SOURCE AND RESTORATION OF THE TEXT. THE work of External Criticism is not com- pleted when the source has been shown to be genuine and has been localize I. It still remains for the student of history to ana- lyze his sources and, in some cases, even to en- deavor to restore the printed or written text, corrupted by copyists. The need of text anal3'sis is self-evident. In the first place, all of the events recorded by a witness have not, as a rule, been directly ob- served by him. Not all parts of his record are equally valuable and the first-hand evidence can be separated from the derived only by anal- ysis. In the second place, as we shall see later, historical facts are established by the agree- ment among independent witnesses. It is of the first importance, then, that the independ- ence of the witnesses should be established, and this is done by studying the relation of one source to another. I shall consider, then, (1) the analysis of a single source; and (2) the an- alysis of the relationship existing among sev- eral sources. Failure to analj^ze their sources and to dis- tinguish between what the witness knows first hand and what he has derived from others is one of the characteristics of the uncritical his- torian. Having decided that the work as a whole 50 HISTORIC AL METHOD. is genuine, and that it was written by a s;on- temporary who lived in the midst of the event/? described, the oreneral inference is made that all the evidence contained in the record must be source material. This is, of course, as a rule, a false inference, but it is surprising how long it has taken historical science to get beyond it. Thucydides wrote the history of the Pelo- ponnesian war. The work is preceded by an in- troduction in which he deals with the history of Greece up to his own day. It is self-evident that Thucydides could have Avitnessed but a small I)art of the events that he recorded; for the events of his own day, he obtained his informa- tion largely from eye witnesses, while for the past he was dependent upon written and oral tradition. It is necessary (1) to analyze the work and, if possible, to separate Thucydides' personal knowledge from his information de- rived from other sources; and (2) to learn, if possible, what the other sources were. Neither of these operations can be success- fully carried out. For although Thucydides, in referring to his methods of work, states (I., 22) that he gave "the facts of what was done in the war only after investigating them accurately in every particular, as far as possi- ble," he seldom, if ever, gives his source of in- formation. The necessity of proof was not realized in his dsiy.- Incidentally he tells us that he suffered from the plague (II., 48 ), com- manded in Thrace (IV., 104) and was exiled (V., 26); but these statements are notmade for the purpose of showing us where he obtained his information. HISTORICAL METHOD. 51 It ought to be clear by this time that the fact that Thucydides wrote the "History of the Peloponnesian War " does not justify the his- torian in accepting the whole work as the result of his observations. It is evidently composed of material of unequal value. How unsci- entifically much of the work has been done in the past will be realized when it is stated that the question as to whether the Boeotians mi- grated from Thessaly to Boeotia in early times has been often settled in the past by a quota- tion from Thucydides (I., 12). When it is remembered that if there ever was such a migration, it must have taken place several hundred years before his day and that the event must have been without a written record for many generations, it will be easy to com- prehend the desperate straits in which the his- torian finds himself who cites Thucydides on such a point and really believes that he has proved anything by the citation. A vast amount of time and labor have been expended on the analysis of the Greek and Ro- man historians. After a careful examination of all the attempts to analyze the sources of Roman history. Dr. Carl Peter (Zur Kritik der Quellen der alteren Romischen Geschichte, Halle, 1879) concludes that, for the most part, such work can lead to nothing definite (page 166). The same remark would apply to the larger part of the written traditions on the his- tory of Greece. Some of the possibilities, however, are shown in Kirehhoff's " Thuky dides und sein Urkundenmaterial, Berlin 52 HISTORICAL METHOD. 1895," especially in the study of the truce be- tween Athens and Sparta (IV., 118, 119). What our sources are for early Roman history m&j be learned from Pelham (Outlines of Ro- man History, N. Y., 1893, page 3), namely, the tradition as established in the time of Livy and Dionysius (about first century, B. C). Under these circumstances, one would be justi- fied in saying that we know practically noth- ing about the first few centuries of Roman his- tory. It is source analysis that has led to these results. They are negative, it is true, but the acceptance of negative results in the place of unscientific and impossible constructions repre- sents a distinct step in advance. " Most historians," says Seignobos, " refrain from rejecting a legend till its falsity has been proved, and if by chance no document has been preserved to contradict it, they adopt it pro- visionally. This is how the first five centuries of Rome are dealt with. This method, unfor tunately still too general, helps to prevent his- tory from being established as a science." But more satisfactory results are obtained in the study of periods nearer our own time. The period of the French Revolution is a verita- ble TumTnelplatz for untrained historians. No- where does the neglect of source analysis lead to more disastrous results. For the events of 1789 the Moniteur and the Archi/ces parlemen- taires are commonly referred to as sources. There are copies of the Moniteur dealing with the events from May 5, 1789, but the publica- tion of the paper did not begin until Novem- HISTOKICAL METHOD. 53 ber of that year. Several years later the desire to make the file complete for the revolu- tion led the editors to publish the back num- bers from May 5 until the real publication began. This port'on of the paper can, in no sense, be called a source; it is a second-hand compilation. An analysis of the material con- tained in it shows that other newspapers (Mira- beau's Courrier de Provence)^ m^moires (Bailly), and contemporary histories {Histoire de la revolution yar deux amis de la liberty) were made use of it. Ranke has an interesting study on the Moniteur in his "Revolution- skriege." He there points out that the compi- lation for the year 1789 is composed of two parts; the second part, dealing with the events happening outside the Assembly, is taken al- most bodily from the history by "Two Friends of Liberty," referred to above. It is clear, then, that, instead of using the Momteui\ we should go back to the source used by the com- pilers of the Moniteur. Even here the need of source analysis will still be felt, for Flammer- mont tells us, in his work, '''La jour nee du iJf, juillet, 1789, Paris, 1892," that ifor this great event the work has "no original value." The authors utilized the most of the accounts by eye witnesses that had been published when they composed their history, but as we have the same works at our disposal we pass on to them, and begin anew the task of analysis. This one example ought to be sufficient to establish the necessity of source analysis. The state of the Archives parlementaires is 54 HISTORICAL, METHOD. even worse than that of the Moniteur. The portion of the work devoted to 1789 was com- piled about thirty years ago and the chief source was the Moniteur! The work was done at the expense of the French oroverninent. It is now being done over again by M. Brette. That is a good example of the loss of time and money resulting from unscientific work. More than that, the work has been often read by those who did not know its character and the generalizations based upon it are often un- sound. A good criticism is found in Brette, " Zonna>ve ; ca'N it, "founded on the nature of the conditions and of the manifestations of activity." This questionnaire Q,oxi\w[\'S>\k\& following groups: I., Material Conditions; 11,, Intellectual Habits (not obligatory ) ; III., Material Customs (not obligatory); IV., Economic Customs; V., So- cial Institutions; VI., Public Institutions (obli- gatory). This method of grouping facts according to their nature may be combined with the lirst method of chronological and geographical grouping. Thus we might have the history of Greek art in the time of Pericles. But a scheme that disposes of the facts com- mon to many men and persisting through one or more centuries does not dispose of all the facts. There still remain the acts and words peculiar to certain individuals. What shall be done with them? What is the part that the individual plays in historical development? Is the life of society controlled by fixed laws and is the individual a helpless atom? These are HISTORICAL METHOD. 93 questions that divide the historians of the con- tinent to-day, and in Germany they wage a war that is anything but merry. It is the old ques- tion of necessity and free will. But, as Seignobos sa3^s, one may not take sides here. Both general and particular facts must be taken into account. History is explan- atory of the real, and the real happens but once. There is but one evolution of society. In this evolution "the facts that succeed one another have been the product not of abstract laws, but of the conjuncture, at each moment, of many facts of different kinds. This conjunc- ture, sometimes called chance, has produced a series of accidents which have determined the particular march of the evolution. The evolu- tion is intelligible only by the stud}^ of these accidents; history is here on the same footing with geology and palaeontology." The histor}^ of Roman institutions would be unintelligible without a knowledge of the bat- tle of Pharsalus. " History is thus obliged to combine the stud}' of general facts with the study of cer- tain particular facts." This mixed character — half science, half narrative of adventures — has often given rise to the question, " Is history a science or an art? " There are two kinds of facts, then, to be grouped: general facts and particular facts. I shall treat them in order. In dealing with the general facts that treat of habits, manners, and ousioni-;, institutions, lan- guage, religion, etc. , after deciding what habit 94 HISTORICAL METHOD. we ^hall study, it is necessary, first of all, to de- termine the group to which the habit belongs. The natural tendency is to assume that a group is made up of like units. Because a group of people talk the same language we are apt to think that the members of the group have everything else in common. A minute's reflec- tion would make clear the falsity of this infer- ence, "for no real group, not even a central- ized society, is a homogeneous entity. What is the group of people that talk Greek, the Chris- tian group, the group of modern science? The English nation consists of Gauls, Scots, and Irish J the Catholic church consists of the faith- ful scattered throughout the entire world and differing in everything except religion.'' The Swiss are united in government, but are divided in language (French, German, Italian) and in religion (Catholic and Protestant). Think of the bewildering way in which the groups, made up of individuals with one or more habits in common, overlap in the United States! We must kno.w, then, what people compose the group; by what bonds they are united; what activities they have in common, and in what they differ. This study will show us for what a group may serve. For the study of language, religion, etc., we would not select a national group. But even when a group has some habit in common, the group is not homogeneous; there are subdivisions. Language is divided into dialects and religion into sects. It is nec- essary to determine the subdivision of each grouD. HISTORICAL METHOD. 95 When all the habits of a society have been studied, the society as a whole must be examined in its relation to other societies of the same time. "This is the study of international insti- tutions, intellectual, economic, and political (diplomacy and the usages in war). * * * To all this it would be necessary to add the study of habits common to many societies and relations that do not take on an official form.'* All this gives us nothing more than a descrip- tion of society in repose. History, however, treats of society in motion, evolving. It is necessary to trace the manner in which these institutions change. The steps in this process are: (1) The choice of the fact whose evolution is to be traced; (2) the period of the evolution; (3) the successive steps; and (4) the means by which the evolution has been brought about. The particular facts, the accidents of history, still remain to be treated. They are " the facts that have acted upon the evolution of each of the habits of humanity." All of these facts taken together, classed by order of time and country, would bind together the special histo- ries of the institutions and give a picture of the "ensemble of historic evolution." But all of the facts can not be described. Which shall be chosen? Those without which the evolution can not be described. The fact in itself may have been small; the effect produced may have been decisive, and the effect is the all-important thing. Both in special histories (the study of habits) and in general histories (the study of decisive 96 HISTORICAL METHOD. accidents) it is necessary to mark the stages in the evolution, to divide it into periods. This is done by means of events. For the special history, an event that has produced the for- mation or the chani^e of a habit becomes the commencement or end of a period. Here the event is orenerally of the same species as those that form the object of the study, while in oren- eral history the periods embrace the evolution of several kinds of facts. The Migrations, the Renaissance, the Ref- ormation, the French Revolution were all- embracing in their effect on society. The periods thus formed are of unequal length. For evolution is not regular and a period of slow uneventful evolution is often followed by an age of rapid, dramatic trans- formation. This rapid, fragmentary presentation of the grouping of the facts is necessarily unsatisfac- tory. The most exhaustive treatise would leave but abstract conceptions in the mind of the reader when unaccompanied by the study of typical cases. If we would learn how to group material, we must not only try our own hands at it, but we must study the works of successful his- torians. Seignobos' '''' Sistoire politique de V Europe contemjporaine^''* is an excellent exam- ple, because it enables us to see how well he applies his own theory. The preface contains a discussion of the kinds of classification — log- * The work is being translated and wili be pabltshed by Henry Holt & Co. HISTORICAL mi:ti!<^:»d. 1»7 ical. chronological, and geographical — a ;d tha advantages of each. In the work, the three kinds of classifications are combined. In the first part, tlie geograph- ical order is followed and the interior history of each state is studied separately and succes- sively; in the second part, the logical order is employed and the political phenomena common to the diti'erent European societies are grouped together; in the third part, the chronological order is used, and the exterior relations of the states considered. The volume deserves a care- ful stud}^ as a successful attempt at scientific grouping. The picture formed by grouping the facts Mould be much less complete, if we had only the material that criticism furnished us. In this material, there are many gaps. These gaps become noticeable during the work of grouping the facts, and the historian endeavors to meet this difliculty by constructive reason- ing. " We set out from the facts made known to us by the sources, in order to infer new facts. If the reasoning is correct, this method of ob'aining know edge is legitimate." It is, however, a dangerous method, if not employed with the greatest care. Seignobos makes the following valuable suggestions con- cerning the control of the method: (1) Never mix up reasoning with the analy- sis of a source; (2) never confound the facts drawn directly from the sources with facts ob- tained by reasoning; (3 ) never reason unconsci- ously, but Dut the argument into logical form 98 HISTORICAL METHOD. and the fallacy is easily detected; (4) if there is the least doubt about the soundness of the reasoning, draw no conclusions; (5) never at- tempt to turn a conjecture (No. 4) into a cer- tainty by dwelling upon it. Too long reflec- tion upon a few sources renders the conjecture familiar and at last plausible. The chances are that the first impression is correct. There are two ways of employing construc- tive reasoning: negative and positive. Nega- tive reasoning, or the "argument from silence," has already been dealt with. Positive reason- ing starts with a fact found in the sources and infers a fact not found there. A good illustra- tion is given l)y Bernheim; we find in a certain document, dated May 10, that the Bishop of "Wormes signs himself, " Wormatiensiselectus;" a document, dated May 16, bears the signature, " Wormatiensis episcopus." From these facts we infer the additional facts, that between the 10th and lOtli the bishop elect was consecrated. We infer more. We know that it was custom- ary for such a ceremony to fall on Sunday or a festival day; computation shows that the l'2th. was Sunday, and we infer that the bishop was confirmed May 12. That this positive reasoning may be exact, it is necessary: (1) that the general proposition should be exact, that is, "the two facts that it assumes to be bound together ought to be of such a nature that the first is never found with- out the second;" of this we ma}^ be certain only when we operate with detailed propositions; (2) "That the general proposition may be de- tailed, the particular historical fact must itself HISTORICAL MKTHOD. 99 be known in detail." The conditions of reliable positive construction are rarely realized. " We know too little about the laws of social life and too rarely the precise details of an historical fact." These are the steps in the synthetic opera- tions that are included in the group to which this chapter is devoted. Having treated the Imagining of the Facts, the Grouping of the Facts, and Constructive Reasoning, I shall con- sider in the next chapter the Environment and the Philosophy of History. CHAPTER IX. SYNTHETIC OPERATIONS: ENVIRONMENT AND THE PHILOSOPHY OF HISTORY. THERE was a time, and that not long ago, when a work on method would have been complete without the treatment of such questions as Environment and the Philosophy oF History. But that day is past. The histor- ian of to-day realizes that it is not only neces- sary to consider each event as a link in a chain of events — if he would understand the particu- lar event — but that he must also possess a knowledge of the phyp^cal, psychical, and social conditions that form the environment of the events. But the sciences dealing with these conditions are in a formative state and can furnish only scanty assis'ance. Anthropo- geography, anthropology, ethnology, individual and social psychology, and sociology will trans- form historical work when they themselves have reached a more advanced stage of develop- ment. Under the influence of these sciences, synthetic historical work will, in the future, become scientific in its turn, and another im- portant field will be rescued from dilettanteism. The influence of geography upon the de- velopment of society is recognized })y histor- ians and, as a rule, every history of a people is prefaced by a chapter upon the geography of the country. But this a rather poor make- shift. It is almost equivalent to the presenta- HISTOKICAL METHOD. 101 tion of crude material to be worked over by the readers. The question that interests the student of history is " W^hat influence did the ofeo^raphy of the country have upon the his- tory of its people?" That question can not be answered by a simple description of the natural features of the country; it can not be answered once for all by an introductory chap- ter. The study of the relation of man to his geoofraphical environment must ^o hand in hand with the description of the acts that were conditioned by that environment. In dealing with this subject, it must never be forgotten (1) that the historic races did not originate in the environment in which we lind them, and (2) that man is not passive clay to be moulded by his physical surroundings. No attempt should be made to explain the brilliant history of the Greeks from the geog- raphy of Gresce alone. There is no way of de- termining how long the Greeks had been in Greece previous to the recorded beginnings of their history. . It is very certain, however, that when they migrated to this country the people bore with them tribal characteristics, in- herited from ancestors, who had been for thou- sands of years subject to natural influences in other places. How much, then, that we find in Greek character is due to the environment in Greece and how much to the earlier environ- ment of the ancestors of the Greeks, we shall never be able to determine. Suppose, for an in- stant, that the records that made it possible to explain the presence of the negroes in our south- 102 HISTORICAL METHOD. ern states were lost. What success would the historian have that attempted to explain the characteristics of these people from the en- vironment in which they find themselves to-day? Human beings, moreover, are not like chem- ical atoms; the same external causes, acting on different human aggregates produce unlike effects. To one people, a sea would be a bar- rier; to another, it is the threshold to a new world. The character of a people must, then, be always counted with. Certain natural con- ditions are capable of producing such and such effects if the right people is brought into con- tact with them. This power of reaction differs not only from people to people but in the same people from time to time. How unwise it is to attribute too great an influence to natural en- vironment is nowhere better illustrated than in the history of Greece. The same sea and sky, the same mountains and transparent atmos- phere, but how different the results to-day! If the physical environment of the Greek has not changed, the social environment certainly has. One of the common fallacies encountered in the consideration of this matter of the relation of geography to the evolution of man in soci- ety is the belief that man emancipates himself by degrees from the influence of his physical environment. According to this theory, the barbarian is more dependent upon nature than the man of civilization. This statement of the case does not make clear the true situation. The savage is bound to nature by few and slen- der bonds; the civilized man by many and HISTORICAL METHOD. 103 strong bonds. The latter makes more use of nature than the former. He has a greater va- riety of resources; when one fails him the others serve him. The farmer. who plants but a single crop and sees it perish from lack of moisture is no less dependent upon nature than the savage, who, living from the natural rice of the swamp, is driven to the verge of starva- tion by the first wind that strips the plants. These two limitations made, it is certain that geographical environment plays a vastly im- portant role in human history. It affects both the conditions and the acts of men. It affects their bodies through climate and their minds through startling natural phenomena. It im- pels men of the North to the warm lands of the South and controls the direction of the move- ment l)y river valleN-s (n;itural highways) or checks it by high mountains. It makes im- possible the development of a high civilization upon islands of the ocean (lack of space); it enriches and develops science by the struggle with nature, dict-ites man's clothing and even his social organization (social and political di- vision of the inhabitants of a desert). All of these things are not history, but they make history intelligible. For however great the psychical development of man may be in the future, it will always rest upon a physical foundation, and this physical side must inevita- bly link him to his geographical environment. But if the historian turns on the one side to the geographer for aid, he turns on the other to the psychologist. Historical acts are noth- 104 HISTORICAL METHOD. ing else than the "expression of human feel- ing, conceiving and willing, the activities of that psychophysical unit that we call the soul or the mind." Simmel (Die Probleme der Geschitsphilosophie. Berlin, 1S92, p. 33) calls psychology the " apriori of historical science." It is clearly important, then, that the historian should understand something nbout the psychi- cal conditions under which the individual or society — in part or in whole — act. Not that such general knowledge will enable him to de- termine what particular psychological fact hap- pened at a given time and place, but that he will be able to tell what psychological facts could not have taken place under given condi- tions and he will be helped in the interpreta- tion of the facts. The common use of collective terms such as the state, the church, society, culture, corpora- tions has tended to obscure the fact that all historical acts are the result of the feeling, conceiving, and willing, of individuals. How- ever important social psychology may be it should never lead us to undervalue the impor- tance for the historian of a knowledge of the psychology of the individual. Such knowledge has always been possessed and applied, in some degree, by historians. It was, however, "an instinctive knowledge of the universal identity of human feeling, think- ing, and willing," that the ordinary man makes use of in his attempts to understand the acts of others; and furthermore an empirical knowl- edge of their own mental life, combined with HISTORICAL METHOD. 105 the empirical knowledg-e of the soul life of oth- ers drawn from reading and experience. This knowledge was employed, for the most part, in suppl^dng motives for acts when these motives were not given by tradition. Such work is most difficult. Its successful accom- plishment depends upon the ability to put one's self in the place of the historical i)ors()n- age and to feel and think as he felt and thought. The fact thai like outward acts are often «iiie to different inward motives renders the attempt to infer these motives a very delicate operation. Robespierre favored the Hebertists and they attached themselves to him. An Italian noble- man had his enemies in his power and instead of destroying them he dismissed them with gifts; they felt insulted and planned to take his life. But the empirical knowledge of ps3^cholog- ical conditions should be widened and deepened by the scientitic study of the mind, and not by the study of the sound mind only, but also of the diseased mind. The whole attitude toward certain classes of phenomena, such as religious exaltation and hallucinations, has been changed by psychical research. All the historical proc- esses of interpretation, combination, and repro- duction are conditioned by mental laws and the study of these processes can never lead to the best results if the laws are not taken into account. It has been shown in a previous chapter that the work of interpretation and combination not only calls for a knowledge of the individual 106 HISTORICAL METHOD. psychical, but also of the social-psychical con- ditions, or the conditions of mind having their roots in the relations of men with one another. Whether we look upon these conditions as the manifestations of a social mind (Volksgeist) matters little; the important thing is that the living together of men in society produces re- sults that are not simply the mechanical total of individual sensations and thoughts; there is an additional something characteristic of the whole. Ko better illustration can be given than the language of a people. It is a product of the social spirit. All have contributed to its growth, some consciously, others unconsciously, Itut of all it may be said " Was er webt, das weiss kein Weber" (The weaver knows not what he weaves. ) The national consciousness, although it exists only in the sensations and conceptions of individuals, yet constitutes a peculiar whole and exercises a peculiar influ- ence. The consciousness that the same general conception of the fatherland lives in the minds of millions of other men and women preserves and even increases the patriotism of the indi- vidual. Since, then, social relations call forth peculiar psychical effects, these effects may rea- sonably constitute matter for investigation and the fiekl may be set aside under the head of social psychology. Although the science has been born it is still an infant. So little has been accomplished that the historian is obliged to do for himself the work that will be done in the future by an auxiliary science. Through his own investiga- HISTORICAL METHOD. 107 tions, he must win for himself the necessary knowledge of the social-psychical conditions. He must appreciate the distinctions of time and locality when dealing with the past. Difficult as it is to appreciate justly the social-psychical conditions of contemporaries who ma}^ be di- rectly observed — like the French, Germans, and Italians — it becomes infinitely more diffi- cult to deal successfully with past ages tliat can be studied only indirectly through the sources. Only through a knowledge of the social- psychical conditions can the historian determine what is peculiar to the individual and what the common property of the age in which he lived. Who would undertake to speak with authority of the work of a great artist without having first acquainted himself with the condition of that particular art in the age in which the artist lived? There is no commandment of good his- torical work that is more frequently violated than the commandment that the writer shall ac- quaint himself with the spirit of the age in which the events that he would narrate took place. And what wonder? The man who un- dertakes to familiarize himself with the social- psychical conditions before describing the events that w^ere conditioned by them often finds that life is too short for the completion of his task. A knowledge of the geographical, the indi- vidual-psychical and social-psychical conditions is not all that constitutes an acquaintance with man's environment. Every individual born into an advanced society finds himself surrounded 108 HISTORICAL METHOD. by the vast accumulations inherited from past ages. Probably the great superiority of the civilized man over the savage is due in a very large degree to this fact. Imagine the child of cultured parents transferred immediately after birth to the care of African negroes and reared in their midst. A little reflection will show that the wide chasm between his real life and the life that he might have led was due to the absence of culture accumulations among the Africans. How great a blunder Buckle committed in fail- ing to take into account the culture conditions can be readily seen. While it is true that the culture products are the results of historical events, they slioukl, nevertheless, be treated as independent factors in all historical problems, because no historical development has ever taken place that has not been influenced by some existing culture conditions. These culture conditions act, for the most part, almost like geographical conditions, since they are not subject to important changes through the arbitrary acts of individuals or of particular generations. Some, it is true, are more changeable than others. The constitu- tion of a state is more easily changed than the language of the people. The culture condi- tions, moreover, do not influence all alike. "The sun shines equally upon the just and the unjust, the educated and the ignorant, the rich and the poor; " the literature of a people exerts a powerful influence upon a few. a slight in- fluence upon many. Then again the influence of all the culture products is not the same. HISTORICAL METHOD. 109 The form of the state affects all in much the same manner, while the influence of lancjuao^e, art, and science differs from individual to indi- vidual and from group to group. The consideration of the culture conditions has been neglected in the past together with the other elements of the environment. In certain epochs their influence has been so great that the historian could not fail to count with them. But the treatment of economic conditions in connection with the French Revolution, of art in the Age of Pericles, of literature in the per- iod of the Renaissance and of religion in the per- iod of the Reformation does not satisfy the just demands of this element in the environmen*^. The ideal of the historian — perhaps an impos- sible ideal —must be to consider the culture conditions as acting continuously and regularly, not spasmodically, upon the historical evolu- tion, and to trace their iniluence not only upon the events but their mutual influence on each other. Such are the elements of the environment in which the historical events take place. With this environment the historian must acquaint himself and under the influence of it his nar- rative must l)e written. 1 he labor of the great historians can be appreciated only by those who realize how much time must be given to the simple eft'ort to reach a point of vantage from which the event may be seen in its true light. From the historical environment the step is but a short one to the philosophy of history. 110 HISTORICAL METHOD. It should be noted first of all that there is a dis- tinct difference hetyveenj?/uIowj)/iical history — a narrative with philosophical reflections — and the philosophy of history. The first may con- tain a sweeping, comprehensive view of uni- versal history, but so long as it retains its descriptive character it falls short of the phil- osophy of history. For the philosophy of history deals not with the description of histor- ical events, but with the consideration of the universal and fundamental conditions and pro- cesses upon which the historical development rests. If the historical events are introduced at all, it should be simply as illustrative material. The content of the philosophy of history con- sists of a distinct group of problems, and these problems must evidently have to do with his- tory or the evolution of man in society. But what are these problems? An examination of the history of the philosophy of history from Augustine to Lotze makes clear that all the problems that have presented themselves may be grouped under two main heads: (1) How is the historical evolution brought about ? and (2) What are the results and what the significance of the historical evolution? In other words, it is the business of the philosophy of history to investigate the factors of historical evolution and the value of the results of the evolution. The analysis of the factors leads to the three groups of general conditions that have already been treated. The relation of these factors to one another and to the historical evolution must, if possible, be determined. In the fur- HISTORICAL METHOD. m ther analysis of the factors, a new set of prob- lenis presents itself. Is this psychical be- in.o:, the individual of history, a free beino-2 Can any freedom of the Avill exist in a societr where evolution is controlled by natural law? Is this belief in the freedom of the will simply self deception and is the individual absolutely dependent upon external powers or forces? And here we rise to the problem of problems. v> hat IS the motive force in all history? Is it the struo^orie between orood and evil that has been gainer on since the fall of man and will continue until the last jud^^ment ? (So thou'o-ht the philosophers of the Middle A^es.) Is it the hand of a personal God, who by rewards and punishment leads man on to ever higher desti- nies? Or is it the divine idea that has been placed in the germ in the soul of man, to be de- veloped ororanically in history ? Is it the man- ifestation of the God idea itself? Is history simply the unfolding of the immanent world- spirit? Are the natural laws only the form in which theinner, spontaneous will impulses out- wardly realize themselves? Do natural laws alone control history ? Or is it all accident? These problems concerning the factors of evolution lead naturally to the problems deal- ing with the value of the results of evolution. Can we prove that one of the results has been th-e perfecting of man and the improvino- of his condition? If so, has this progress been regu- lar and universal? Have all the social de- ments been equally active and equally de- veloped, or has the evolution been onesided? 112 HISTORICAL METHOD. Are all the elements capable of participating in progress, the moral and artistic equally with the intellectual ? Are all peoples called to take part in this progress or are there a chosen few? Can we even say that only certain classes in certain peoples are the sharers of this culture ? Finally, what is the measure of progress or of regress? The problems of the factors and values stated above have been answered again and again but no satisfactory solution, no solution that does justice to all the conditions of the problem, has yet been presented. Many of the failures, up to the present time, have been due to bad method. The most of the work has been done by men defective in historical training, who have not hesitated to do violence to the facts in order to justify their theories. The philosophy of history in the future must rest on the science of history and grow out of it. It will develop as our knowledge of history develops and its aim will be to comprehend historical facts as regarded from the most universal point of view, that of general human evolution, that of humanity itself. CHAPTER X. SYNTHETIC OPERATIONS: EXPOSITION. IN the precedinor chapters, I have described the process by which an historian discov- ers the source material related to his sub- ject and obtains the use of it; I have explained the critical examination to which the material, when found, must be submitted to determine its genuineness, to localize, and to analyze it; 1 have indicated the method for determining the value of the sources admitted as evidence and for establishing the facts contained in the sources; iinall}-, I have shown how the facts must be imagined and grouped, the gaps in the evidence filled in by constructive reasoning, the ]ihysica\ psychical, and social environment con- structed and the factors and processes of his- torical development (philosophy of history) be understood. It remains for the historian to communicate to others the results of his research. This last to])i;^ is treated by Bernheim under the head L)arstellung and by Seignobos in the two chap- ters entitled Construction des formules ghierales and li.r-position. In his two chapters, Seignobos really treats more topics than are embraced in Bernheim's Darstellung. The first chapter deals with the subject matter of I.^arstellung and adds a few words on the philosophy of history; the chapter on Exposition is devoted 114 HISTORICAL METHOD. to the question of scientific form in the narra- tive and is not treated by Bernheim. The problem of Expodtloii or Darstellung is by no means simply a question of style, although, as Seignobos says, " il n'y a pas d'hislorien complct sans une bonne langue," and the reason is that " pour atteindre des faits aussi f uyants que les fails sociaux, une langue feruie et precise est uu instrument indispens- able." But the need of a good command of language, of a power to use exact, scientific ex- pressions in dealing with facts as elusive as social facts, is not the topic to be emphasized in this chapter; we have to do here with a question of a more difficult nature, namely, how may the results of the investigation be communicated, in as correct a manner as possible^ to others? Not all the results can be commuuicaied. However limited the topic of investigation, not all the results of that investigation can possibly be presented in all their fullness of detail. An historian who attempted to communicate all the facts that he h id found concerning the life of Napoleon would never find readers. It is a practical question. Obliged to choose be- tween '* being complete and unknowable or of being knowable and incomplete," historical S3'^nthesis naturally decided in favor of the latter. If not all the results of the investigation can be communicated, it follows that there must be condensation and this condensation must be performed in such a manner that the narrative will, as far as possible, correspond to the le- HISTORICAL METHOD. 115 ality as it appeared to the investigator. The relation of the narrative to the mass of con- ceptions contained in the views of the man who has seen the evidence first-hand has been compared by Bernheim to the relation of the piano arrangement of a great orchestral work to the work itself. The idea is easily grasped; the execution of the idea is unusually difficult. To condense, to omit unimportant details, to retain the right proportions in the condensed material, is a thing calling for an infinite amount of skill. The selection of the material must depend upon the theme. Details omitted from a uni- versal history would tind place in the history of a state, of a province, or of an individual. In a church history, one kind of material would be emphasized, in an industrial history another. A good historian may learn much in the matter of composition from the artist, for the good historical narrative is characterized by boldness of execution and subordination of details. The most helpful thing that has yet been written on condensation is Seignobos' excellent chapter entitled Constuction desfarmules g^€- rales. "History," he writes, "to become a science, must elaborate the raw material found in the facts. It should condense the facts into descriptive formulae, both qualitative and quan- titative. It should search for the relations be- tween the facts, relations that form the last con- clusion of every veritable science." Historical facts, human facts can not be reduced to a few simple formulae like chemical formulae, but 116 HISTORICAL METHOD. "history, as well as all sciences of life, has need of des<.'riptive formulae to express the character of the different phenomena." The formula should be short that it may be manageable; precise, that it may give an exact it'ea of the fcict. Yet brevity and precision conflict. To obtain brevity, we must eliminate details, while characteristic details alone give precise knowl- edge of human events. A compromise is nec- essary; all that is not strictly necessary must be suppressed, but the work of suppression must cease when it leads to the sacrifice of char- acteristic traits. If the demands of precision are lost sight of, "all history is reduced to a mass of vague generalities, uniform for all time, with the exception of some proper names and dates." In constructing these formulae one would do well to "employ as often as possible concrete and descriptive terms; their meaning is always clear." That is to say, "collective groups should be descril)ed by collective names and not by abstract substantives (as monarchy, state, democracy, reformation, revolution)." When a word or a group of words constituting a formula is employed, there should be no un- certainty as to the meaning that attaches to them. What different meanings attach to the w )rd monarch when applied to Clovis, Louis XL, Louis XIV., Louis XVI., and Louis Philip! This misunderstanding may be avoided b}' a description of the term when first used. Such a device may mar the artistic unity of the uarraiive, but the historian is primarily a scieu- HISTORICAL METHOD. llT tist and only secondarily an artist. Belloc's "Danton" (1899) is a good illustration of the compulsion that the historian feels to make his general terms convey a precise meaning. Tac- itus would have won the eternal gratitude of students of the Middle Ages had he but in- dicated the meaning that attached to abstract Latin substantives when used to describe prim- itive German life. After dividing his formulae into two classes, qualitative and quantitative, Seignobos subdi- vides his qualitative formulae into (1) those descriptive of general facts (habits and evolu- tions) and (2) unique facts (events). "■ General facts consist of acts often repeated and common to ujany men. Their character^ extent^ and duration must be determined." To determine the character, the common traits of a usage or institution are drawn from many individual cases and expressed in a for- mula; the individual variations are neglected. Serfdom in a certain period (jf the Middle Ages, the city life of a like period, might be treated in this wa)^ If the usage is that of language, laws, regulations, etc., it should never be for- gotten that formulae of this kind express only su- perficial facts; "in language the written words, not the })ronunc;ation; in religion, the dogmas and official rites, not the real beliefs of the mass of the people. * * * In all of these cases the knowledge of conventional formulae should be supplemented some day by a study of the real habits." To determine the extent of a habit, one de- 118 HISTORICAL METHOD termines the area of its distribution and the point where it is most practiced; for its dura- tion, the time of its first and last appearance and the epoch of its greatest activity must be noted. In the case of unique facts, many can not be united under the same formula, and it is neces- sary to decide what facts shall be sacrificed. Personal taste shoulJ not determine the choice of facts to be retained. There is but one stand- ard that may be employed, and that is the role played by the fact in human affairs. " Persons and events that have clearly influenced the march of evolution must be preserved. The mark by which one recoo;nizes them is that the evolution could not be described without mak- ing mention of them." In constructing the formula for an individual we must draw our traits from his biography and habits; from his biography we learn the physiological, educational, and social influences under which he lived; from his habits we form an idea of his conception of life, his dominant tastes, his habitual actions, and his rules of con- duct. From all these details, w^e form a portrait or formula of the individual. To construct the formula for an event, we must fix its character and extent. The charac- ter consis's of the traits that distinguish this event from all others. The formula should contain the following points: one or more indi- viduals, impelled by certain motives, working in the midst of certan material conditions (lo- HISTORICAL METHOD. 119 cality, instruments), performed certain acts, and the acts produced a certain modification of society. The extent of the event should show the region where it occurred and that af- fected by it, together with the moment when the action began and that when it was finished. The formulae of quality should be supple- mented by those of quantity. The five methods employed in formulating quantity as given by Seignobos are (1) measurement (psychological facts can not be measured), (2) enumeration, (3) evaluation, (4) sampling, (5) generalization. They decrease in exactness from the first to the last. There are certain dangers to be guarded against in the employment of each method. In the second, it should be noted that the method of statistics applies to fac's that have in common a definite character of which use is made that the facts may be counted. These facts, however, are not homogeneous and may have but one thing in common (crimes, suicides, workmen, strikes). The danger is that the statistician may believe that he has described the the facts with scientific precision, when he has only counted them. Evaluation is an enumeration covering a por- tion of the field and based upon the assumption that the same proportion holds good in the rest of the field of investigation. The results are unreliable if it is not known that the portion examined is exactly similar to the other por tions of the field. 120 HISTORICAL MKTIIOD. "Sampling," that consists in makino; an enu- meration of units taken from different portions of the Held of investigation, is of value when the samples are representative of ihe whole. They should be taken from very ditierent points and from groups living under very dif- ferent conditions, that the exceptions may bal- ance one another. " Generalization is only an instinctive method of simplification." It is unconsciously applied in dealing with all complex human events. It is an unconscious "sampling." It may be ren- dered correct by submitting it to the conditions of "sampling." To generalize correctly one must (1) indicate the field of generalization (country, group, class, epoch); (2) be sure that all the facts generalized upon are similar in all the points concerned; (3) be certain that the cases selected are types, and (4) take care that the cases considered are more numerous as the points of resemblance are less numerous. The descriptive formulae, qualitative and quantitative, do not represent the last stage of the synthetic operations. Still larger groups, more general -formulae must be constructed. In forming groups more and more general, the procedure is the same as that described above. At each step in advance some of the character- istics of the smaller groups are dropped until at last only universal human characteristics re- main. In this manner, the formulae for a lan- guage, a religion, a society, or an event are constructed. When this condensation can be carried no ftirther, the attempt may be made to HISTORICAL, METHOD. 1"21 classify the groups by comparison. The two methods of classification suggested by Seig- nobos are (1) comparison of similar categories of special facts, such as languages, religions, and arts, and (2) comparison of "real groups of real individuals. " The first is ' ' an abstract classi- fication that isolates one species of facts from all others;" the second is "a concrete classifi- cation similar to the classifications in zoology, when not the functions but the animal forms are clavSsified." The difiiculty with this last classification is due to the disagreement as to the characters that should constitute the basis of resemblance: shall they be political, eco- nomic, intellectual, or religious? Upon this point no agreement has been reached. The problem of problems still remains un- solved: How to classify all of the groups or formulae and thus construct a grand ensemble embracing all human society. Some historians say that it is impossible, but it continues to be an ideal worth striving for. It is clear that these groups are not isolated in reality and that a change in one brings about a change in the others. If there is unity (Zusammenhang), it will be possible in time to construct the for- mula for this unity. Having formulated the results of the investi- gation it simply remains for the historian to commit these formulae to paper. Here we touch one of the weak points in historical work. The fact that the writing of history has so often been left to men with no scientifio training, men whose main purpose was to write 122 HISTORICAL METHOD. to entertain, and who repeatedly sacrificed the truth in their effort to please — sacrificed the truth because it was commonplace and unat- tractive — this fact has made it difficult for the historical narrative to take on a scientific form. It has hecn said too often that a liistorical work sh(Hild be a work of art. The thing may- be ab.-oliitely impossible. The completeness and attractiveness of the work depend upon the quantity of the sources and the character of the sources. Both of these things are beyond the control of the investigator. The first de- mand made upon the historian is to tell the truth, to tell us exactly what he knows and what he does not know. If there are any gaps in the evidence, it is his business to point them out. We should remember that he is not an in- fallible authority speaking from inspiration, but just a plain fallible man who should be required to prove ever}^ statement that he makes. The demand for proof is not made by the general public; it must be made by the body of historical students. W hat right have untrained men, who have not mastered the subject of which they speak, what moral right have these men to publish histories for the education of the multitude? None whatever. It is simply a business proposition. These popular his- tories in four or five large volumes are the dime novels of historical literature. It should be the business of teachers and writers of history to put an end to the existence of such works by creating a taste for something better. This work may be done in two ways: ( 1 ) by prepar- HISTORICAL METHOD. 123 ing histories that are at the same time popular and sound (Adams' Civilization of the Middle Ages), (2) by training in historical study in the schools. The teachers of history must be stu- dents of history, and the hovs and girls must be taught what proof is in history just as they are taught what proof is in mathematics and the other sciences. Having learned what histor- ical proof is they must be trained to give it themselves and to demand it of others. The historical narrative must, then, take on a scientific form when it is written for students. The sources of information must be indicated and evidence exactly cited in support of gen- eral statements. The writer must do his work in such a way that the reader may be able to control his every statement. Less time will be wasted when this rule is rigorously fol- lowed. What would we think of an investi- gator in chemistry who gave only results, ma le no mention of the processes by which results were reached, and carefully destroyed all traces of his methods as soon ; s his work was accomplished? We t^hould hardly credit him with common sense. And yet that is just tiie course that many historians have pursued in the past, and that many are pursuing to-day. Many of the instructors in our colleges by their irrational methods of instruction are cul- tivating that sort of thing with their students, and until these teachers develop a scientific conscience this state of things is likely to continue. The task that I set myself in the opening 1^-J^ HISTOEICAL METHOD. chapter has been accomplished. I then st:ited my belief that there would never be better teaching of history until there are more stu- dents of history among the teachers. It was that conviction that lead to this attempt to present in a brief outline the substance of the ' method of historical research as found in the works of Bernheim and of Langlois and Seig- nobos. Jf it opens the eyes of any teacher to the necessity of this training and leads them on to study the works that I have so constantly cited, I shall have done all that I hoped to do. EUROPEAN HISTORY STUDIES. FRED MORROW FLING, Ph.D., Professor of European History. University of Nebraska. EDITOR. A monthly publication. Subscription 40 cents. Extracts from tne sources; translations from the writers of the period studied. Vol. III. The French Revolution. No. I. Sept., 1599. Absolutism in the Church: The Consti- tution. No. II. Oct., 1899. Stru<-gle against Ab.solutism in the xr^ TTT XT ,on„ ^<-'liurch; Certificates of Confssion. No- III. Nov., 1899. The Church and the Philosophers. No. I\ . Dec, 1899. Reform Edicts; Turgot S^' vt" it?.-' Iq?^- The Administration of Finance; Necker. No. VI. Feb., 1900. The Notables. No. yil. March, 19J0. The King and the Parliaments. No. \ III. April, 1900. The Age of Pamphlets. No. IX. May, 19.0. The Elections to the States General No. X.June, 1900. The Cahiers of Complaints. Vol. II. Civilization of the Middle Ages. J898-1899. No. I. Christian and Pagan. No. II. The Teutonic Barbarians. No. III. Selections from the Koran No IV Chivalry and the Mode of Warfare. No. V. Feudalism' No.M. Monasticism. No. VII. The Jews of Angevin England. No. \III. Kise of Cities. No. IX. The Trades of Pans. No. X. Mediaeval Science. Bou Id in do h, 6o cents; for introduction, 50 cents net. Vol. I, Greek and Roman Civilization. '^^'' p«Ltu?,vf-;„ "^4^ ?°™^J'^ Age. No. II. The Athenian Constitution. No. III. Spartan Life. No. IV. Alexander's Mode of Warfare. No. V. The Achaen League. No VI Roman Constitution No. VII. Roman Life of the First Punic War. No. VIII. Roman Life of the JugMrthine Period. N .. IX. Roman Life Under the Empire No X Roman Laws. Bou nd in clotii, 60 cents; for introduction. 50 cents net. BOOKS ON METHOD Studies in European and American History. F. M. Fling and H. W. Caldwell. .^■^ ^5°^ °/ °'*'^'" ?°*i, P''g^^ setting forth the principles, methods and advantages of the -'Source Method." Containing a so studies 01 such subjects as The Grecian Period. The Romans, Absolute Monarchies, The French Revolution, Early Virginia History Early Massachusetts Laws, and others. Cloth. Price, $1.00. Outlines of Historical Method. F. M. Fling, Ph.D. The Leaflets All the Leaflets named above are kept in stock. Single conv 5 cents each. Ten or more copies of one number, 4 cents each Ten or more subscriptions to the current volume, or ten or more sets lic^^ SamXVf°thTf.^'J,^r'''^' '^'^^ *° °°^ ^'^'iress aTsO cen s eaen. bamples of the Leaflets sent free to any address. Address J. H. MILLER, Publisher, Lincon, Neb. No. I. Sept., 1899. No. II. Oct., 1899. No III. Nov., 1899. No. IV. Dec, 1899. No. V. Jan , 1900. No. VI. Feb. 1900. No. VII. March , 1900. No. VIII April, 1900. No. IX. May, 1900. No. X. June, 19 0. AMERICAN HISTORY STUDIES. H. W. CALDWELL, A. M., Professor of American History, University of Nebrask' A monthly publication. Subscription 40 cents. Extracts from the sources of American History, Early Laws, Treaties, State Papers, Letters, Speeches, etc Vol. III. Territorial Development: Expansion. Territorial Boundaries. First National Boundaries. The Northwest Territory. Acquisition of Louisiana. Purchase of Florida. Annexation of Texas. Conquest of California and New Mexico. Alaska and Hawaii. West Indies and the Pailippines. Vol. II. Some American Legislators. 1898-1899. No. I. G.illatin. No. II J. Q .\da ns. No. III. Clay. No. IV. Webster. No. V. Calhoun. No. VI. Sumner. No. VII. Dout,'las. No. VIII. Seward. No. IX. Chase. No. X. Bla.ne. Vol. I. A Survey of American History. 1897- 1898. No. I. Founding of thp Colonies. No. II Develop- ment of Union among the Colonies. No. III. Cause of tiie Revolution. No. IV. Formaiion of the Constitution. No. V, Growth of Nationality. No. VI. Slavery (1). No VII. Sin very (-) No. VIII Civil War and Reconstruction. No. IX. Foreign Relations. No. X. Industrial Develop- ment. [Extra. Early Colonial Laws. 5 cents. J BOOKS ON METHOD Studies in European and American History. F. M. Fling and H. W. Caldwell. A Dook of over 300 pages setting forth the principles, methods, and advantages of the "Source Method." Containing also studies on such subjects as The Grecian Period. The Romans. Absolute Monarchies, The French Revolution, Early Virginia History, Early Massacnusetts Laws, and others. Cloth. Price, $1 00. Outlines of Historical Method. F. M. FUNG. Ph.D. This volume contains some of the strongest papers yet published in English on Method in History. It treats clearly the subjects of E.sternal and Internal Criticism of Sources and Synthetio Operations. Cloth. Price. 60 cents. The Leaflets All the Leaflets named above are kept in stock. Single copy, h cents each. Ten or more copies of one number, 4 cents each. Ten or more subscriptions to the current volume, or ten or more sets of any previous volume unbound, sent to one address at 30 cent* each. Samples of the Leaflets sent free to any address. Address J. H. MILLER. Publisher, Lincon, Neb. ■^v^ 810 -m .^^ s^\.:% '9-, ,-o- v'v- x>^- . \^ . . ^,<^' '/■ .^^^' o5 -n*. ' \> * ^ ° " / ^ :<■ \^ °<. : 1 1 1 ■ .,>>^^' '^>. ■>*, ■%r V ^'■I Deacidified using the Bookkeeper proce; A' Neutralizing agent: Magnesium Oxide ^c 0^ '^■^ ^ Treatment Date: APR £002 N,'^^ ■''^. ^ .^ ./^. ^^'^''- PreservationTechnologie o'^~- '^ ' ''* ^^ of'' '^C *W°'"-D LEADER IN PAPER PRESERVATIC V^ . . ,. 'V'- * s 1 ^ ■ ■;Ji s ' « / ■ 111 Thomson Park Drive ' ^ '^ ^V s ""^./T^K. ''^ CranberryTownship. PA 16066 ,-^^' \0 < »4 O fv ^. -TV - : '^ '7* ,.00. . J w» ^ ii '^^^0 .^^v % 4 A ,0 ' " * ^" ^x. '^