Historical criticism is the art of distinguishing the true from the false concerning facts of the past. It has for its object both the documents which have been handed down to us and the facts themselves. We may distinguish three kinds of historical sources: written documents, unwritten evidence; and tradition. As further means of reaching a knowledge of the facts there are three processes of indirect research, viz.: negative argument, conjecture, and a priori argument.
It may be said at once that the study of sources and the use of indirect processes will avail little for proper criticism if one is not guided chiefly by an ardent love of truth such as will prevent him from turning aside from the object in view through any prejudice, religious, national, or domestic, that might trouble his judgment. The rôle of the critic differs much from that of an advocate. He must, moreover, consider that he has to fulfil at once the duties of an examining magistrate and an expert juryman, for whom elementary probity, to say nothing of their oath, makes it a conscientious duty to decide only on the fullest possible knowledge of the details of the matter submitted to their examination, and in keeping with the conclusion which they have drawn from these details; guarding themselves at the same time against all personal feeling either of affection or of hatred respecting the litigants. But inexorable impartiality is not enough; the critic should also possess a fund of that natural logic known as common sense, which enables us to estimate correctly, neither more nor less, the value of a conclusion in strict keeping with given premises. If, moreover, the investigator be acute and shrewd, so that he discerns at a glance the elements of evidence offered by the various kinds of information before him, which elements often appear quite meaningless to the untrained observer, we may consider him thoroughly fitted for the task of critic. He must now proceed to familiarize himself with the historical method, i. e. with the rules of the art of historical criticism. In the remainder of this article we shall present a brief résumé of these rules apropos of the various kinds of documents and processes which the historian employs in determining the relative degree of certainty which attaches to the facts that engage his attention.
There are two kinds of written documents. Some are drawn up by ecclesiastical or civil authority, and are known as public documents; others, emanating from private individuals and possessing no official guarantee, are known as private documents. Public or private, however, all such documents raise at once three preliminary questions:
(1) authenticity and integrity;
Authenticity and Integrity
Does the document which confronts us as a source of information really belong to the time and the author claimed for it, and do we possess it in the shape in which it left that author's hand? There is little or no difficulty in the case of a document printed during the author's lifetime, and given at once a wide distribution. It is otherwise when, as often happens, the document is both ancient and in manuscript. The so-called auxiliary sciences of history, i. e. palæography, diplomatics, epigraphy, numismatics, sigillography, or sphragistics, furnish practical rules that generally suffice to determine approximately the age of a manuscript. In this preliminary stage of research we are greatly aided by the nature of the material on which the manuscript is written, e. g. papyrus, parchment, cotton or rag paper; by the system of abbreviations employed, character of the hand-writing, ornamentation, and other details that vary according to countries and epochs. It is rare that a document claiming to be an original or an autograph, when submitted to such a series of tests, leaves room for reasonable doubt regarding its authenticity or non-authenticity. More frequently, however, ancient documents survive only in the form of copies, or copies of copies, and their verification thus becomes more complicated. We must pass judgment on each manuscript and compare the manuscripts with one another. This comparison enables us, on the one hand, to fix their age (approximately) by the rules of palæography; on the other, it reveals a number of variant readings. In this way it becomes possible to designate some as belonging to one "family", i. e. as transcribed from one original model, and thus eventually to reconstruct, more or less perfectly, the primitive text as it left the author's hand. Such labour (merely preliminary, after all, to the question of authenticity), were every one forced to perform it, would deter most students of historical science at the very outset. It becomes, however, daily less necessary. Men specially devoted to this important and arduous branch of criticism, and of a literary probity beyond suspicion, have published and continue to publish, with the generous aid of their governments and of learned societies, more or less extensive editions of ancient historical sources which place at our disposal, one might almost say more advantageously, the manuscripts themselves. In the prefaces of these scholarly publications all the known manuscripts of each document are carefully described, classified, and often partially represented in facsimile, thereby enabling us to verify the palæographic features of the manuscript in question. The edition itself is usually made after one of the principal manuscripts; moreover, on each page we find an exact summary (sometimes in apparently excessive detail) of all the variant readings found in the other manuscripts of the text. With such helps the authenticity of a work or of a text may be discussed without searching all the libraries of Europe or tiring one's eyes in deciphering the more or less legible handwriting of the Middle Ages.
The manuscripts once counted and classified, we must examine whether all, even the most ancient, bear the name of the author to whom the work is generally attributed. If it be lacking in the oldest, and be found only in those of a later date, especially if the name offered by the earlier manuscripts differ from that given by later copyists, we may rightly doubt the fidelity of the transcription. Such doubt will often occur apropos of a passage not met in the oldest manuscripts, but only in the more recent, or vice versa. Unless we can otherwise explain this divergency, we are naturally justified in suspecting an interpolation or a mutilation in the later manuscripts. While the authenticity of a work may be proved by the agreement of all its manuscripts, it is possible further to confirm it by the testimony of ancient writers who quote the work under the same title, and as a work of the same author; such quotations are especially helpful if they are rather extensive and correspond well to the text as found in the manuscripts. On the other hand, if one or several of such quoted passages are not met with in the manuscript, or if they be not reproduced in identical terms, there is reason to believe that we have not before us the document quoted by ancient writers or at least that our copy has suffered notably from the negligence or bad faith of those who transcribed it. To these signs of authenticity, called extrinsic because they are based on testimony foreign to the author's own work, may be added certain intrinsic signs based on an examination of the work itself. When dealing with official and public acts care must be taken to see that not only the handwriting, but also the opening and closing formulæ, the titles of persons, the manner of noting dates, and other similar corroborative indications conform to the known customs of the age to which the document is attributed. Amid so many means of verification it is extremely difficult for a forgery to escape detection. Words and phraseology furnish another test. Each century possesses its own peculiar diction, and amid so many pitfalls of this nature it is scarcely possible for the forger to cloak successfully his misdeed. This is also true for the style of each particular author. In general, especially in the case of the great writers, each one has his own peculiar stamp by which he is easily recognized, or which at least prevents us from attributing to the same pen compositions quite unequal in style. In the application of this rule, no doubt, care should be taken not to exaggerate. A writer varies his tone and his language according to the subject of which be treats, the nature of his literary composition, and the class of readers whom he addresses. Nevertheless an acute and practised mind will have little difficulty in recognizing among the various works of a given author certain qualities which betray at once the character of the writer and his style or habitual manner of writing. Another and a surer means for the detection of positive forgery or the alteration of a document is the commission of anachronisms in facts or dates, the mention in a work of persons, institutions, or customs that are certainly of a later date than the period to which it claims to belong; akin to this are plagiarism and the servile imitation of more recent writers.
The critic must now make the best possible use of the written sources at his disposal, i. e. he must understand them well, which is not always an easy matter. His difficulty may arise from the obscurity of certain words, from their grammatical form, or from their grouping in the phrase he seeks to interpret. As to the sense of the individual words it is supremely important that the critic should be able to read the documents in the language in which they were written rather than in translations. Doubtless there are excellent translations, and they may he very helpful; but it is always dangerous to trust them blindly. The scholar who enters conscientiously upon the work of critic will always feel it a strict duty to warn his readers whenever he quotes a text from a translation. It is well known that to interpret a term correctly it is not enough to know its meaning at a particular epoch, which we are accustomed to regard as classic, in the language to which it belongs. We need only open any large Latin lexicon, e. g. Forcellini's or Freund's (especially if we keep in view the corresponding page of the Latin "Glossarium" of Du Cange), to appreciate at once the very remarkable modifications of meaning undergone by Latin terms in different periods of the language, either from the substitution of new meanings for older ones or by the concurrent use of both old and new. In his efforts to fix the age of a text the critic will, therefore, be occasionally obliged to exclude a meaning that had not yet arisen, or had ceased to be in use when the text in question was composed; sometimes he will be left in a condition of uncertainty or suspense, and obliged to abstain from conclusions agreeable enough but unsafe. Again, in order to grasp correctly the sense of a text it becomes necessary to understand the political or religious opinions of the author, the peculiar institutions of his age and country, the general character of his style, the matters which he treats, and the circumstances under which he speaks. These things considered a general expression may take on quite a particular sense which it would be disastrous for the critic to overlook. Often these details can only be understood from the context of the passage under discussion. In general, whenever there is occasion to verify the exactness of a quotation made in support of a thesis, it is prudent to read the entire chapter whence it is taken, sometimes even to read the whole work. An individual testimony, isolated from all its surroundings in an author's work, seems often quite decisive, yet when we read the work itself our faith in the value of the argument based on such partial quotation is either very much shaken or else disappears entirely.
What is now the value of a text rightly understood? Every historical statement or testimony naturally suggests two questions: Has the witness in question a proper knowledge of the fact concerning which he is called to testify? And if so, is he altogether sincere in his deposition? On an impartial answer to these questions depends the degree of confidence to be accorded to his testimony.
Concerning the knowledge of the witness we may ask: Did he live at the time when, and in the place where, the fact occurred, and was he so circumstanced that he could know it? Or, at least, are we sure that he obtained his information from a good source? The more guarantees he gives in this respect the more, all else being equal, does he prove himself trustworthy. As to the question of sincerity it is not enough to be satisfied that the witness did not wish to utter a deliberate lie; if it could be reasonably shown that he had a personal interest in warping the truth, grave suspicions would be raised as to the veracity of all his statements. Cases of formal and wilful mendacity in historical sources may be regarded as rare. Much more frequently prejudice or passion secretly pervert the natural sincerity of a man who really respects himself and esteems the respect of others. It is possible, and that with a certain good faith, to deceive both one's self and others. It is the duty of the critic to enumerate and weigh all the influences which may have altered more or less the sincerity of a witness -- personal likes or dislikes, social or oratorical proprieties, self-esteem or vanity, as well as the influences which may affect the clearness of a writer's memory or the uprightness of his will. It by no means follows that the authority of a witness is always weakened by the process described above; often quite the contrary happens. When a witness has overcome influences that usually powerfully affect a man's mind and dissuade him from yielding to the natural love of truth, there is no longer any reason to doubt his veracity. Moreover, when he asserts a fact unfavourable to the religious or political cause which he otherwise defends with ardour; when he thus gains no particular advantage, but on the contrary subjects himself to serious disadvantage; in a word, whenever his statements or avowals are in manifest opposition to his interests, his prejudices, and his inclinations, it is clear that his evidence is far weightier than that of a perfectly disinterested man. Again, the preceding considerations apply not only to the immediate witnesses of the fact in question, but also to all the intermediaries through whom their evidence is transmitted to us. The trustworthiness of the latter must be established as well as that of the authorities to which they appeal.
Given the necessity of observing so much caution in the use of historical texts, it may appear very difficult to reach complete certainty regarding the facts of history. How may we be sure, especially in dealing with ancient times, that our witness presents every desirable guarantee? Often he is scarcely known to us, or quite anonymous. How many facts, once held to be established, have been eliminated from the pages of history. And for how many more must we indefinitely suspend our judgment for lack of sufficiently convincing authority. Historical certitude would indeed be difficult to reach if for each fact we had but one isolated piece of evidence. Full certainty would then be possible only when it could be shown that the character and position of a witness were such as to preclude any reasonable doubt as to the exactness of his statements. But if the veracity of the witness is guaranteed only by negative data, i. e. if we are merely aware no known circumstances warrant us in suspecting carelessness or bad faith, there arises in us a more or less vague belief, such as we easily accord to any quite unknown person who seriously relates an event which he says he has seen, while on our part we have no reason to suppose either that he himself is deceived or that he is deceiving us. Strictly speaking, our belief in such a witness cannot be called a halting faith. On the other hand it differs considerably from a belief that is based on more solid foundations. We shall not, therefore, be much surprised if the occurrence be later described in an entirely different manner, nor shall we object to abandoning our former belief when better informed by more reliable witnesses. Were it otherwise, our passions would be to blame for causing us to hold to a belief, flattering perhaps, but unsupported by sufficient evidence. We frankly admit, therefore, the possibility of a more or less wavering mental adhesion to facts that rest on a single testimony and whose value we are unable properly to appreciate. It is otherwise in the case of facts confirmed by several witnesses placed in entirely different conditions. It is very difficult, nay generally speaking morally impossible, that three, four, or even more persons, not subject to any common influence, should be deceived in the same manner, or should be parties to the same deception. When, therefore, we find a fact established by several statements or narratives taken from different sources, yet all concordant, there is scarcely any further room for reasonable doubt as to the entire truth of the fact. At this stage, however, we must be very certain that the historical sources are truly different. Ten or twenty writers who copy the narrative of an ancient author, without any new source of knowledge at their disposal, in general add nothing to the authority of him from whom they have gleaned their information. They are but echoes of an original testimony, already well known. It may happen, however, and the case is by no means rare, that narratives based on different sources exhibit more or less disagreement. How then shall we form our judgment?
Right here an important distinction is necessary. The various narratives of a fact often exhibit a perfect harmony as to substance, their divergence appearing only in matters of detail upon which information was had with greater difficulty. In such cases the partial disagreement of the witnesses, far from lessening their authority regarding the principal fact serves to confirm it; disagreement of this kind shows on the one hand an absence of collusion, and on the other a reliance of witnesses on certain sources of information common to all. There is, however, an exception. It may happen that several writers, whose veracity we are otherwise justified in suspecting, agree in narrating with much precision of detail a fact favourable to their common likes and dislikes. They either report it as eye-witnesses or they declare that they reproduce faithfully the narrative of such witnesses. In dealing with writers of this character the critic must examine carefully all their statements, down to the minutest detail; often a very insignificant circumstance will reveal the deception. We may recall here the ingenious questioning by which Daniel saved the life and reputation of Susanna (Dan., xiii, 52-60). Similar means are often employed with success in the law courts to overthrow clever systems of defence built up by culprits, or to convict a party who has suborned false witnesses in the interest of a bad cause. Occasionally such measures might be advantageously applied in the conduct of historical examinations. Let us suppose that there exists a conflict of opinion about the substance of a fact, and that it has been found impossible to reconcile the witnesses. it is clear that they disagree. At this point, evidently, we must cease to insist on their absolute value and weigh them one against the other. Keeping always in view the circumstances of time, place, and personal position of the different witnesses, we must seek to ascertain in which of them the conditions of knowledge and veracity appear to predominate; this examination will determine the measure of confidence to be reposed in them, and, consequently, the degree of certainty or probability that attaches to the fact they narrate. Frequently, though no indispensable preliminary of mental conviction, a careful comparison of more or less discordant versions of a fact or an event will reveal in the rejected witnesses the very sources or causes of their errors, and thereby exhibit in much clearer light the complete solution of problems whose data seemed at first sight confused and contradictory.
To hang a man, a clever examining magistrate does not always need one line of his writing. Silent witnesses have often convicted a criminal more efficaciously than positive accusers. The most insignificant object left by him on the scene of his crime, another found in his possession, an uncommon degree of prodigality, a hundred other equally trifling tokens, lay bare very often the most ingeniously planned schemes for avoiding detection by the law. Even so in the science of history. Here nothing is negligible or unimportant. Monuments of architecture, objects of plastic art, coins, weapons, implements of labour, household utensils, material objects of every kind may in one way or another furnish us precious information. Certain classes of historical sources have long since attained the dignity of special auxiliary sciences. Such are heraldry, or armorial science; glyptics, which deals with engraved stones; ceramics, or the study of pottery in all its epochs. To these we may add numismatics, sigillography, and especially linguistics, not so much for a surer interpretation of the texts as for procuring data from which may be conclusively established the origins of peoples and their migrations. Archæology, in its broadest sense, comprises all these sciences; in its most restricted sense it is confined to objects which are beyond their scope. Truly it is a vast province that here spreads out before the historical pioneer, and he needs much erudition, acumen, and tact to venture therein. Fortunately, as with manuscripts and inscriptions, it is no longer necessary for the historical student to possess a thorough knowledge of all these auxiliary sciences before entering on his proper task. For most of them there exist excellent special works in which we may easily find any archæological details needful in the discussion of an historical question. It is to these works and to the advice of men learned in such matters that we must have recourse in order to solve the two preliminary questions regarding all evidence, written and unwritten: that of authenticity or provenance, and that of meaning, i. e., in archæological remains, the use to which the objects discovered were once put. In dealing with unwritten evidence these questions are more delicate; similarly the rules for our guidance are much more difficult, both to formulate and to apply. It is here, particularly, that shrewdness and acumen, and the prophetic insight that comes of long practice, offer help more important by far than the most exact rules. It is only by dint of observation and comparison that we learn eventually to distinguish with accuracy. These preliminaries once satisfied, we enter on the task of historical criticism properly speaking. Through it these precious relics of the past are called to shed light on certain writings, to confirm their evidence, to reveal a fact not committed to them; more frequently they furnish a sure basis of conjecture whence eventually follow discoveries of great importance. Here, however, and it cannot be repeated too often, the path of the historical student is perilous indeed. The misadventures of amateur archæologists, whether in the matter of pretended discoveries or in dissertations based on them, have provoked no little raillery, not only among severely just professional critics, but also among romancers and dramatic writers. As already stated, it is especially by the judicious use of conjecture that we obtain from these silent witnesses such information as it is in their power to furnish. For more specific treatment of this powerful but delicate instrument of historical criticism we refer the reader to a subsequent section of this article: CONJECTURE IN HISTORY.
Every student of history must eventually face a problem very embarrassing for a conscientious scholar. Facts appear which have left no trace in any writing or contemporary monument. Buried in obscurity for centuries they suddenly appear in full publicity and are accepted as incontrovertible. Everyone repeats the story, often with minute detail, though no one is able to offer any credible evidence of the trustworthiness of the current statement or narrative. It is then said that such facts rest on the evidence known as oral or popular tradition. What degree of confidence is due to this popular tradition? Its originators are quite unknown to us as are also the many intermediaries who have passed it down to the time when we are first cognizant of it. How may we obtain a guarantee of the veracity of the original witnesses and then of their successors? Perhaps a rather natural comparison will help us to a clear solution of this question. We may note at once a striking analogy between tradition concerning the past and public rumour about present events. There are in both cases numberless intermediary and anonymous witnesses, concordant as to the substance of the facts, but as to the details often quite contradictory of one another; in both cases also there is an identical ignorance concerning the original witnesses; in both cases, finally, many instances in which the current information was verified and many others in which it was found to be altogether false. Let us suppose the case of a prudent man deeply interested in knowing precisely what is happening in a distant country; one who, moreover, takes much pains to be well informed. What does he do when he learns by public rumour of an important event said to have occurred in the place in which he is interested? Does he accept blindly every detail thus bruited abroad? On the other hand, does he pay no attention whatever to rumour? He does neither. He gathers eagerly the various narratives current and compares them with one another, notes their points of agreement, and their elements of divergence. Nor does he conclude in haste. He suspends his judgment, seeks to procure official reports, writes to his friends who are on the spot to learn from them reliable news, i. e. confirmation of the facts on which men agree, solutions of the difficulties which arise from discordant versions of the event. Possibly he has no confidence in the persons charged with drawing up the official reports; possibly, too, he cannot correspond with his friends, owing to the interruption of communications by reason of war or other causes. In a word, if such a man found himself dependent on public rumour alone he would remain indefinitely in a state of doubt, content with a more or less probable knowledge until some more certain source of information offered.
Why should we not deal similarly with popular tradition? It appeals in just this way to our attention and we have the same motives for mistrusting it. More than once it has been helpful to judicious critics and pointed the way to important discoveries which they would never have made with the sole aid of written documents or monuments. Let us look at the matter in another way. Have not all students of historical documents come frequently across the same peculiar, one might say capricious admixture of true and false which meets us at every step in the case of popular traditions? It would be equally rash on the one hand to reject all tradition and place faith only in written testimony or contemporary monuments, and on the other to accord to tradition an implicit confidence merely because it was not formally contradicted by other historical data, though it received from them no confirmation. The historian should collect with care the popular traditions of the countries and epochs he is treating, compare them with one another, and determine their value in the light of other information scientifically acquired. Should this light, too, eventually fail him, he must wait patiently until fresh discoveries renew it, content in the meantime with such measure of probability as tradition affords. In this way the already acquired historical wealth will be retained, yet no danger run of exaggerating its value, or, finally, of casting suspicion on its trustworthiness by incorporating with it false or doubtful statements.
THE NEGATIVE ARGUMENT
The negative argument in history is that which is drawn from the silence of contemporary or quasi-contemporary documents concerning a given fact. The great masters of historical science have often used it with success in their refutation of historical errors, sometimes long intrenched in popular belief. It is to be noted that on such occasions they have always held firmly to two principles: first, that the author whose silence is invoked as a proof of the falsity of a given fact, could not have been ignorant of it had it really occurred as related; second, that if he were not ignorant of the fact, he would not have failed to speak of it in the work before us. The greater the certainty of these two points, the stronger is the negative argument. Whenever all doubt in regard to them is removed, we are quite right in holding that a writer's silence concerning a fact in question is equivalent to a formal denial of its truth. There is nothing more rational than this process of reasoning; it is daily employed in our courts of justice. How often is a legal line of attack or defence broken by purely negative evidence. Honourable men are brought before a judicial tribunal who would certainly, in the hypothesis of their truth, have knowledge of the facts alleged by one of the contending parties. If they affirm that they have no knowledge of them, their depositions are rightly considered positive proofs of the falsity of the allegations. Now, evidence of this kind does not differ substantially from the negative argument in the above conditions. In one case, it is true, the witnesses formally state that they know nothing, while in the other we learn as much from their silence. Nevertheless this silence, in the given circumstances, is as significant as a positive assertion.
There are, nevertheless, some who claim that a negative argument can never prevail against a formal text. But this assertion is not even admissible respecting a contemporary text. If the writer to whom it belongs does not offer an absolute and incontestable guarantee of knowledge and veracity, his authority may be very much weakened or even destroyed by the silence of a more reliable and more prudent writer. It often happens in courts of law that the deposition of an eye or ear-witness is questioned, or even rejected, in view of the deposition of some other witness, equally well-placed to see and hear all that occurred, but who yet declares that he neither saw anything nor heard anything. Mabillon was certainly wrong in maintaining that the negative argument could never be used unless one had before him all the works of all the authors of the time when the event happened. On the contrary, a single work of a single author may in certain cases furnish a very sound negative argument. Launoy, on the other hand, is equally wrong in maintaining that the universal silence of writers for a period of about two centuries furnishes a sufficient proof of the falsity of facts not mentioned by them; it is quite possible that no author of this period was morally bound by the nature of his subject-matter to state such facts. In this case the silence of such authors is by no means equivalent to a denial. But, it is objected, in order to raise a doubt as to a fact related by later writers, have not the best critics often relied on this universal silence of historians for some considerable time? This is true, but the epoch in question was one already carefully studied and conscientiously described by several historians. Moreover, the disputed fact, if true, would necessarily have been so public, and such, in kind and importance, that neither ignorance nor wilful omission could be posited for all these historians. We have here, therefore, the two conditions needed to make inexplicable the silence of these authors; consequently, the negative argument loses none of its strength, and is powerful in proportion to the number of silent witnesses. Of course, this line of argument does not apply in the case of some obscure detail, which may easily have been unknown to, or little remarked by some contemporary authors and quite neglected by others; nor, more particularly, does it apply to an epoch of which few monuments are extant, especially few historical writings. In the latter case, the fact of a universal silence on the part of all writers for a considerable period, may, indeed, weaken the certainty of a fact; in reality we do no more than ascertain thereby the absence of all positive evidence in its favour, other than a tradition of uncertain origin. However, once the lack of information is admitted, it is not permissible to advance a step further and present the silence of documents as proof of the falsity of the fact. Their silence in this case is not the negative argument as described above.
The rule laid down in the preceding paragraphs seems to lack no element of precision and practical advantage. But in applying it to ancient times some caution is necessary. In an age of widespread publicity like our own, no important event can occur in any part of the civilized world without being immediately known everywhere and to all. Its principal details, indeed, are at once so fixed in the memory of all interested parties that they will not easily be effaced within a long period. It is astonishing to see how easily some modern writers forget that the former conditions of mankind were very different. They seek to establish an irrefutable negative argument on the hypothesis that a given public fact of importance could not have been unknown to a certain person of education and refinement who lived shortly afterwards. Such writers might learn to be more cautious by recalling a series of curious historical facts. It is enough to remind our readers that when St. Augustine was created auxiliary Bishop of Hippo (391) he did not know, on his own avowal, that the sixth canon of the Council of Nice (325) forbade any consecration of this kind.
CONJECTURE IN HISTORY
Conjecture or hypothesis occurs in history when the study of documents leads us to suspect, beyond the facts which they directly reveal, other facts, so closely related to them that from a knowledge of the former we may proceed to that of the latter. Such facts are most frequently related as cause and effect. Let an important event happen. How shall we explain it? How was it brought about? Evidently by another fact or a group of other facts which constitute its cause or sufficient reason. These new facts are revealed in no historical documents, or at least no one has hitherto perceived them. At once the investigator sees that here it is possible to discover more than is known from the extant documents. With this hope he begins to read extensively, to set afoot various researches, to interrogate in every sense a great many works and all the monuments relating to the fact with which he has been keenly impressed, to study the persons concerned in it, or the age in which it took place; all this in order to recover the often almost invisible thread which connects this fact with details that were originally unnoticed or set aside as unimportant. Absorbed in intense meditation, sometimes made needless through a sudden illuminating insight which reveals at once the right path, he seeks with earnestness the truth that the positive evidence before him still withholds; he passes from one hypothesis to another; he calls to his aid all the treasures of his memory; thus reinforced he turns again to the study of the documents, and collects with minute care every hint or indication that may avail to demonstrate their accuracy or falsity. From such close verification it sometimes appears that the path first struck out was misleading and must be abandoned; often the investigator is led by this hard toil to modify more or less his original ideas; on the other hand, he sometimes meets with striking confirmation of them. Feeble rays which seemed at first quite uncertain grow in power and number until they seem a torch that pours a flood of light before which all uncertainty must vanish. In this way, also, many new aspects are revealed to the enraptured eyes of the investigator and make known to him a vast field of knowledge of the highest interest.
As already stated conjecture enables us to conclude from effect to cause, but it may also follow an inverse method and help us to conclude from cause to effect. This process, however, is generally less reliable in historical research, and calls for more caution and reserve than when it is applied to physical facts. In the latter case the agents are necessary causes; once their mode of operation is known it is possible to predict with almost absolute certainty their results in given conditions, and conjecture avails us merely to arouse the idea of an effect certain to follow, but which we have not yet seen produced. Moreover, generally speaking, in the physical sciences it is easy to imagine a variety of methods by which an hypothesis may be tried and its accuracy verified. In historical science the situation is not quite the same. It deals largely with the moral laws that regulate the actions of free beings, and these are far from being as invariable in their application as physical laws. Much caution is therefore requisite before risking any judgment as to what a man must have done in given circumstances, all the more as his acts may have been influenced by the free acts of others, or by a number of accidental circumstances now unknown to us, but which may have notably modified in a given case the ideas and ordinary sentiments of the person in question. Prudence is not less necessary when the hypothesis is principally based on analogy; i. e. when, to complete our knowledge concerning a fact, certain details of which are not known to us from historical documents, we have recourse to another fact strikingly similar to the one under consideration and conclude thence, in favour of the first, to a similarity of details that are known to us with certainty only in respect of the second fact. Nevertheless we must not reject absolutely this method of investigation; skilfully treated it may render valuable service. A conjecture appeals to the mind all the more convincingly when it solves at once a number of problems hitherto obscure and lacking correlation. Frequently enough, a given hypothesis, taken separately, yields only slight probability. On the other hand, full certitude often results from the moral convergence of several plausible solutions, all of which point in the same direction. Let it be added that in historical research we shall not easily obtain too many hints nor exceed the limit in verification; also that we must be ever watchful against our own preconceptions that easily tempt us to exaggerate the strength of a conclusion favourable to our hypothesis. Nor must we refuse to consider the arguments that tend to weaken or eliminate the latter. On the contrary, it is precisely these arguments that we must study with most care and sift in every sense so that, given their truth, we may abandon opportunely our too seductive conjecture, or at least modify it, again and again if needful, until eventually it acquire such accuracy and precision as to satisfy the most exacting, and be admitted by all as a scientific acquisition both new and solid. A final recommendation, meant to forewarn against the seductions of historical conjecture certain adventurous and inexperienced writers, will not be out of place here. Let them not yield to an illusion only too common among their kind, namely that by their imaginative power and their genius they are destined to advance notably the cause of historical science without acquiring by hard and painful schooling that large and varied and accurate knowledge which men call erudition. Not every learned historian makes brilliant discoveries on the basis of lucky hypotheses; but learning is generally requisite for such discoveries. In historical scholarship, as in all other walks of life, toil and patience are the usual price of success.
THE A PRIORI ARGUMENT
Historical criticism has at its disposition one other source of truth, the a priori argument, a delicate weapon, indeed, but very useful when confided to a well-trained hand. As used in history, this argument is based on the intrinsic nature of a fact, leaving aside for the time being all evidence for or against it. In presence of the fact thus bared of all extrinsic relations the a priori process undertakes to show that it does or does not conform to the general laws which regulate the world. These laws fall into three principal classes. The first comprises fundamental or metaphysical laws, e. g. the principle of contradiction, according to which there cannot co-exist in the same subject elements absolutely contradictory of one another, also the principle of causality, according to which no being exists without a cause or sufficient reason for its existence. The second class includes physical laws which govern the phenomena of the world of nature and the activity of the beings which compose it. To this class also belong the laws which govern spiritual natures and faculties that are independent, or in as far as they are independent, of the action of free will. The third class, finally, comprises the moral laws that govern the activity of free beings, considered as such. No one who has acquired, under good guidance, a little experience of the human heart, will deny the existence of this class of laws, i. e. that in given conditions and under certain influences we can forecast in free beings certain habitual activities. Thus, one well-ascertained moral law is that no man will love and follow evil for itself, save only when it appears to him in the guise of good; another such law is that a man, unless he be a monster of perversity, will naturally tell the truth if he have absolutely no interest in lying.
In what way, now, can these three classes of laws, rightly considered, help us to pronounce on the truth of an historic fact? First, if the fact in question present absolutely contradictory and irreconcilable details it must evidently be rejected without further examination. However, it must be clearly proved that there really is such absolute and irreconcilable contradiction between details presented for simultaneous acceptance. It is important, moreover, to ascertain with certainty whether the contradiction affects the substance of the fact, or only accidental circumstances wrongly connected with it in the imagination of the witness, as frequently happens with popular traditions. In such cases it is only details that need to be rejected, precisely as is done when dealing with more or less conflicting testimonies. Physical impossibility, i. e. manifest opposition between well known laws of nature and an historical statement, is also a conclusive argument against the acceptance of such a statement. Non-believers to the contrary notwithstanding, the possibility of miraculous intervention never seriously troubles at this point the judgment of Catholic critics. They know quite well when to admit, in a particular case, such a possibility. Nor are these cases very frequent. They are also aware that for the acceptance of miracles they must require a far greater amount of evidence than when it is question of purely natural facts. We have in the Catholic process of canonization (see BEATIFICATION AND CANONIZATION) an excellent example of the manner in which the proof of miracles is handled by the tribunal which Catholics most respect. It may not be superfluous to add that prudence suggests a certain hesitation or reserve when the physical impossibility of a fact is in question. The laws of nature are not all so thoroughly understood that we run no danger of confounding a strange or new fact with one utterly impossible. The treatment of moral laws is something more delicate, since they are less absolute in application than physical laws. The mysteries of liberty are even more hidden than those of material nature. Consequently, before asserting the moral impossibility of a fact it is well to consider attentively whether there be not some circumstance, however trivial, which may have accidentally exercised on a given person an influence capable of making him act in a manner opposed to the habitual current of his ideas and sentiments. Such exceptions to moral laws, very rare in the multitude, appear more frequently among individuals. Care must be taken, however, not to admit them without grave reason. It is in support of, or in opposition to a conjecture that the a priori argument is mostly used; frequently enough conjecture is confounded with it. Indeed, it is often through the effort to reproduce mentally what certain persons in given conditions must have done, that we finally hit on what they did do; the next step is the collection of more precise evidence such as may confirm and establish quite satisfactorily the truth that we first saw with the eye of the imagination. We should always remember, however, that mere possibility or non-repugnance must not be considered the equivalent of positive probability, any more than mere ignorance of the causes of a fact is equivalent to its improbability, still less its impossibility, when it is sufficiently attested by direct evidence. Superficial or passionate minds are very much exposed to this kind of confusion.
In formulating, as has been done above, the proper rules for the guidance of the mind in its search after historical truth, it should be repeated that the mind must bring to this pursuit certain preliminary qualities and dispositions indicated at the beginning of this article, the first and most essential of which is a sincere and constant love of truth. Nothing can take the place of this sentiment. It is the rule of rules, the vital and efficient principle in all the processes of criticism. Without it they are quite sterile.
DE SMEDT, Principes de La critique historique (Liège, Paris, 1884); BERNHEIM, Lehrbuch der historischen Methode (Leipzig, 1894); LANGLOIS et SEIGNOBOS, Introduction aux études historiques (Paris, 1899). BUTLER, The Modem Critical and Historical School, its methods and tendencies. Dublin Review (London. 1898).
The Catholic Encyclopedia, Volume IV
Nihil Obstat. Remy Lafort, Censor
Imprimatur. +John M. Farley, Archbishop of New York