that even such masters as Freeman, Ranke, Fustel de Coulanges, and Taine, their works battered and riddled by criticism, undergo purgation for their faults of method and their errors of application.

But I must turn to the positive consideration of my subject. M. Langlois, part author of a brilliant manual of historical method, says: "The peculiarity "The peculiarity of historical facts is this, that they are known indirectly by the help of their traces. Historical knowledge is essentially indirect knowledge." In other sciences the facts are directly observed, and the experiment may be repeated ad libitum. Even where the scientist uses the observations of others, these are made by trained observers and they can be repeated at need if suspected of error. The observations which the historian must use, on the other hand, are rarely made by competent persons, or according to any systematic plan. They usually come to him second or third hand, or are the random recollections of bystander or participant set down haphazard, without much bearing on the fact to be elicited, and often not committed to writing till time has dimmed the memory of what actually was done or observed. It isagain I refer to Langlois-as though a chemist were forced to glean his knowledge of a series of experiments from the chance observations of the laboratory janitor, narrated weeks after the event. The historian can rarely interrogate the fact itself; he can only know it from the imperfect, often mendacious, usually erroneous record which has come down to him. documents, no history" is the unvarying rule-interpreting the word document to mean, in the widest sense, any trace left by the fact, whether material remain, oral tradition, or written or pictorial record. New documents turning up may at any moment reclaim for history an epoch or field of human activity before unknown. To cite but one example, it was through the discovery and interpretation of the cuneiform tablets of Assyria and Babylonia that there was made known a field of history of whose existence indeed we were before aware, but whose outlines and features were shrouded in the darkness of documentary poverty.

"No

To the historical investigator the documents are everything. This is so, however, not in the sense of constituting an end in themselves, for they are merely the starting point-the only possible starting pointin the search for historical truth. "No documents, no history;" but this saying by no means implies that the document at once and of itself yields up historical truth. Of all stubborn, intractable things, the document can on occasion show itself the most stubborn, the most intractable! Fustel de Coulanges, who in some respects may be considered the founder of the modern scientific school of French historians, was fond of characterizing history as "the most difficult of sciences." Perhaps there is exaggeration here; we are all prone to magnify our difficulties and to minimize those of our neighbors. But the impression is so widespread that history is a subject which requires no special skill on the part of its votaries-a subject to which any person of mode

rately liberal culture may turn with good chance of success, in case he but know how to write well-that I am tempted to set down somewhat at length the processes to which the document must be subjected before it will yield up its content of truth.

Assume that with bibliography, catalogue, index, and table of contents our document has been hunted to its lair in archives or repository; assume also that it is deciphered, the true text established, and the first formal work of external criticism performed. These processes, though often carried on by the interpreting historian, may well be left to the philologian or other critical scholar. Even so, much remains to be done before pen can be put to paper in the way of narrative or exposition.

First comes the critical investigation of authorship. By whom, and where, and when, was the document composed? It may be a forgery, as are the well known Forged Decretals and Donation of Constantine, and so many others of the documents of the Middle Ages. So late as 1895 there was published in London by a reputable firm a work entitled, “The Journal of a Spy in Paris During the Reign of Terror," which purported to have been written in 1794; but this was soon shown, from a study of its contents, to be an arrant forgery. Again, even if our document is genuine, its author may have borrowed his information, without so stating, from another source. Gordon's "History of the American Revolution," which was long accepted as the most reliable [contemporary] history of the Revolution from a British pen,' was shown a few years ago to be an impudent plagiarism, copied page after page from the English "Annual Register," and possessing no authority as an independent observation of the events. In modern history, where the number of first-hand documents of proved authenticity is so great, work of this sort is less frequently needed; but for ancient and medieval history the working out of the affiliation of sources is an important part of the historian's preliminary labor.

For all periods the document must be localized as to time and place. A document which cannot be dated with more or less exactness is comparatively valueless. In Madison's journal of the debates in the Federal convention of 1787 is given a draft for a constitution, marvelously like the plan eventually adopted, which purports to be the one introduced by Charles Pinckney, of South Carolina. If the document is actually of the date given it, the chief credit for our Federal constitution must be ascribed to Pinckney. In the record of the proceedings of the convention, however, Pinckney's plan is practically ignored, and the discussion is based entirely on other drafts submitted. The presumptive evidence afforded by this fact is strengthened by what we now know of the history of the plan recorded by Madison. Madison for some reason did not copy out Pinckney's plan at the time of its introduction, and did not secure a copy of it until 1819. In that year the official journal of the convention was published, and it contained Pinckney's plan as supplied to the editor of

עד

the journal by the family. This Madison at that time included in his notes without examination. The draft in Madison's journal, then, instead of being a contemporaneous copy, dates from a period thirty years later. It seems, moreover, to have been written out from memory, and is largely influenced by the constitution actually adopted. As evidence of what Pinckney's plan was in 1787, the document posBesses no value; and any claim for Pinckney's influence on the constitution, based on this document, is worthless.

When we have our document localized in time and place, and know the personality of its author, we must still make sure that our interpretation of its lanpguage is the one its author meant it to bear. Where documents are few and in tongues no longer living, a great deal of emphasis needs be placed on this phase of the historian's task. Much of the pre-eminence which Fustel de Coulanges enjoyed among students devoting themselves to the early Middle Ages was due to the care and conscientiousness with which he went through the extant documents from the fourth to the ninth centuries, weighing without preconceived theory the force of word and phrase, and interpreting them in the light of the context, and of the usage of the author's time, place, and individual habit. When we read in Tacitus's description of the early Germans, Arva per annos mutant, we easily translate, Each year they shift the fields." But what are we to understand by " fields?" and what by "shifting?" Is this a mere field rotation, the community leading a settled life?' or is there annual migration of the community as a whole? Do we have individual, or communal ownership of land? or something which can scarcely as yet be called "ownership" at all? Is an Is an aristocrat or a democratic regime implied in the arrangement? Much, it will be seen in fact the determination of practically the whole trend of medieval popular development, both for the Continent and for England-hinges upon the interpretation of these words and their context. In view of the uncertainty of the interpretation, it is not surprising that for a generation or more historians have ranged themselves in rival schools, under the names of Romanists and Germanists.

"

For modern history, where the student suffers not from a paucity but a plethora of documents, and where the linguistic difficulties are less, the interpretative criticism, though still important, is less vitally so. The criticism of the good faith and accuracy of the author, on the other hand, loses no whit of its importance. It is comparatively easy, in most modern documents, to make out what the author meant to say; but it is less easy always to determine whether what he says is true. Every separate statement must be examined with respect to good faith and accuracy. The author may have had personal, party, class, or national ends to further, by concealing or distorting the truth. The need of such criticism in the case of documents dealing with the causes and progress of the present great European war is especially noteworthy. Again, who to-day would accept unquestion

ingly the statements of a Republican newspaper concerning the aims and policies of the Democratic party, or an account by either Protestant or Catholic divine of the beliefs and practices of his opponents? The monkish chroniclers of the Middle Ages inevitably portrayed men and events from the monkish standpoint. Vanity, too, a desire to magnify one's own exploits, may enter as a distorting factor—as in the memoirs of the famous Cardinal De Retz, and the boasts of King Charles IX, infamously false, of having organized the Massacre of St. Bartholomew. Allowance also must be made for literary artifice, for the desire to please the public, and for official formulas. The phrase, servant of the servants of God," found in most papal bulls, does not necessarily convey an idea of special humility on the part of the Pope using it, nor, on the other hand, should the formulas in Carolingian documents, filched from Roman imperial chancelleries, be allowed to deck out that mock empire with the attributes of departed glory.

66

Criticism for accuracy is a different matter from that for good faith. Here we need inquire whether the authors statements suffer from hallucination or prejudices, whether he was really in a position to know the facts which he relates, whether he was attentive to them, or through lack of interest or distraction (as through need for action on a field of battle) he may have observed inaccurately; whether, above all, he wrote down his observations at the time or some years after the fact. Memoirs written after a number of years are responsible for the introduction into our histories of many errors. To this source, in large part, are due the conflicting claims as to the authorship of the Northwest Ordinance of 1787. To a similar defect of memory was due the claim, advanced by Walt Whitman's friends, that Leaves of Grass" was composed prior to the poet's reading of Emerson's works, when Whitman himself, before the impairment of his mind in later life, had expressly confessed to Emerson's influence. To all the foregoing causes of error must be added that which comes from a constitutional incapacity for accuracy-a defect of mental constitution analogous to color-blindness, which from one of its most noted victims has been styled 'Froude's disease." This eminent writer, though conscientious and industrious to a high degree-though convinced of the necessity of basing history on documents, and aware of the utility of criticism was unable to treat any subject with more than an approach to accuracy. The classical example of his defect is the following, from his account of Adelaide, Australia, a city which he had visited personally: "We saw below us," he writes, basin with a river winding through it, a city of 150,000 inhabitants, none of whom had ever known or will ever know one moment's anxiety as to the recurring regularity of his three meals a day." The facts, as unctuously pointed out by Froude's critics, are these: Adelaide is built on an eminence, not in a basin; there is no river running through it; the population when Froude visited it did not exceed 75,000

"

in a

persons; and at that very time it was suffering from a famine! This instance can be paralleled-not quite so neatly, perhaps, but still in all that affects the principle-by many cases among authors of documents as well as secondary writers. The reality of the defect can scarcely be doubted by anyone who has wrestled with an honest but incapable witness; nor, I think, will teachers of experience need much argument to convince them that an incapacity for in a certain class of minds, is a mournful

accuracy, reality.

I have sketched at such length the critical processes to which the scientific historian must submit his materials because of an impression which prevailsfounded, it must be confessed, upon the too common habit of some historians-that all that is necessary, when you find a statement with respect to a particular fact, especially if the statement is in a contemporary source, is to transfer it unquestioningly to your narrative. In reality, however, the true historical student will accept the statement only after he has tested and tried it for error and bad faith. Then, and then only, may he properly proceed to consider the relation of that particular statement to others, derived from other sources, and bearing on the same fact or event. It must not be forgotten that the testimony of one witness, even of unimpeachable veracity, is not enough to establish more than the probability of the fact. Corroborative testimony, drawn from other observations, is needed before the fact can be accepted as conclusively proved. Often such corroborative testimony will be wholly wanting; often the corroboration will be only partial; frequently tested statements will be completely at variance and flatly contradictory. May the historian in the latter case accept the testimony which agrees with preconceived theory and reject that which contradicts? This, in brief, is what John Fiske, that prince of historical popularizers, was wont to do; this, for example, was his method of dealing with the alleged 1497 voyage of Americus Vespucius. Such procedure, though venial perhaps in the popularizer, is inexcusable in the scholar. The true course is that indicated by Justin Winsor, himself easily first among critical writers on early American history. torical truth," says he, "is reached by balancing everything, and not by assimilating that which easily suits." Even so we must be on our guard. If one witness testifies that two and two make four, and another that two and two make five, we may not split the difference and say the sum is four and a half. One witness is right and the other is wrong. If we can conclusively or with a fair degree of probability say which is right, well and good. If not, we can only point out the conflict, and leave the question in the uncertainty which must inevitably, through lack of decisive evidence, attach to so much of our historical knowledge.

His

There is one process, however, which in the hands of the skilled historian will often avail to settle conflicts in statements and clear up obscurity. This is what has been styled constructive reasoning. On its

negative side this gives us the "argument from silence," in accordance with which we infer, from the absence of statements where we might legitimately expect them, the non-existence of the fact. On its positive side constructive reasoning gives us the argument from analogy," likewise legitimate when properly used; and the argument from the harmony of the facts. The latter, perhaps, constitutes the most frequent and effective application of reason to the final determination of fact. A train of events, the life of a man, the body of usages which we call an institution-each constitutes a whole of which the parts are interrelated. Every fact definitely and conclusively established, no matter how trivial or unmeaning it may seem, can rightly be made a test with respect to other facts of the same connection. The. question of whether the English position at the battle of Hastings was or was not defended by a palisade or fixed defense of some sort, seems quite disproportionate to the amount of ink which was shed over it by the partisans of Mr. Freeman and his critic, Mr. Round. But on that question hinges our whole conception of that battle, so momentous in English history; on it, too, depends in large part Freeman's vaunted reputation for accuracy. The main outlines of history, doubtless, are pretty thoroughly estab lished; but many of the details-which, like the warm flesh clothing the human frame, give expression and character to our persons-remain to be elicited. The multiplication, therefore, of available historical materials the unceasing flood of regesta, calendars of state papers, reports of historical societies, local histories, and historical dissertations-a flood which so dismays the soul of the litterateur-brings no regret to the historical scholar. Each fact ascertained is like a new piece fitted into one of those sectional picture puzzles. It not only contributes its part to the representation of the whole, but it enables us to fit in other blocks, the meaning of which before was in doubt.

In looking back over this fragmentary discussion it occurs to me that the main impression left upon your minds may be that Sainte Beuve's ironical de scription of history, as a fable agreed upon," may after all be right. So far as history is not merely a method but a body of fact, the impression may not be wholly unwarranted. Certainly the facts believed by the public are often doubtful, and long after the his torian has exploded myth and legend they linger (to his disgust) in popular narrative and school textbook. So far as concerns the scientific historian, however, this uncertainty as to the facts exists only in spite of his efforts and not because of any easy credulity on his part. His attitude is that of methodical doubt; the historian," says Seignobos, "ought to distrust a priori every statement of an author." It is only the unscientific historian who presents sur mises as facts, and states as definitely proved that which at best is only probable. The proper attitude is that indicated by Rénan, in the preface to one of his books. "Every phrase," said he, "must be ac companied by a 'perhaps.' I believe I have made a

2

colonies. Rather do the newer writers present her as a harassed mother of many children, to whose family had suddenly been added several more youthful and dependent members, without corresponding increase of income having been obtained for their maintenance. If they could be taken care of until the proper time, they would contribute greatly to the family prosperity and reputation; but the problem of their immediate maintenance was a great and puzzling one.

In her perplexity, Great Britain turned to the now well-to-do thirteen continental colonies, and, without particular thought or concern, informed them that for the future they must supply the funds for about a third of their own protection against Indian, French, and outside dangers. Renewed investigation of this period appears to have revealed no evidence that Great Britain regarded itself as guilty of any startling innovation in demanding that British troops be used in the defence of colonial frontiers, or that the money which the colonists were to contribute should be raised with greater regularity by means of a parliamentary tax rather than with irregularity through the older method of requisitions or requests of colonial governors for voluntary grants by colonial assemblies.

Thus, to the British mind, defense rather than taxation, was the chief object and content of the acts of 1764 and 1765. To those of the colonists to whom fell the leadership in affairs of this sort, however, taxation appeared at once the chief and all-important feature, for that taxation was, to the colonial mind, taxation without representation," and as such, a violation of the most fundamental "right of an Englishman or of an English-American colonist.

In the presentation of the colonial interpretation of representation many teachers make a most important error of fact. One searches in vain the literature of the revolutionary period for a single suggestion that the colonists were willing to accept representation in the British parliament. Their constant and repeated assertion is that the bodies in which they always have been represented, and by which alone they ever have been taxed, are the assemblies of the thirteen colonies; and since to the English-born whatever was customary" thereby became "constitutional," the colonists claimed that their own colonial assemblies were the only bodies in which they could "" constitu

2 For the more recent interpretation of British colonial policy, see G. L. Beer, "British Colonial Policy, 1754-1765" (Macmillan, 1907), 252-316; O. M. Dickerson, "American Colonial Government, 1696-1765" (Clark, Cleveland, 1912), 320-356; W. T. Root, Relations of Pennsylvania with the British Government, 1696-1765" (Appleton, 1912), 389396; C. M. Andrews, "The Colonial Period" (Henry Holt, 1912), 229-253; C. L. Becker, "Beginnings of the American People" (Houghton-Mifflin, 1915), 202-224; Edw. Channing, "History of the United States, III" (Macmillan, 1912), 29-51; C. E. Carter, "Great Britain and the Illinois Country, 1763-1774" (Amer. Hist. Association, 1910), 26

tionally" be represented, and by which they could constitutionally" be taxed."

It is to be regretted that no fairly full reprint of the argument of the Stamp act controversy is easily available to high school teachers. The most extensive collection easily obtainable by all is that found in Hart's "Contemporaries," Vol. II, pp. 381-382 and Briefer extracts from a slightly wider

range of materials are found in Caldwell & Persinger's Source History." The resolutions of the Stamp act congress are easily available in MacDonald's "Select Charters," or in the abridged combinations of his three documentary collections, the "Documentary Source Book." Single or scattered speeches are found in nearly all of the standard source collections, but sufficient material for a really profitable study of the argument of the period can be obtained only by combining that found in several or all of them. It is the writer's belief that the next series of source books to appear will contain more extensive extracts covering a few of the great controversial or formative periods of our history, rather than scattered and unrelated extracts from the whole field of American history.

2. SECOND BRITISH ATTEMPT AT PARLIAMENTARY TAXATION: THE TOWNSHEND ACTS CONTROVERSY, 1767-1770. OUTLINE.-Refusal of New York to obey the Quartering act, and continued colonial resistance to the Sugar act, assembly, renewal of attempt at taxation by Revenue and 1766-1767; parliamentary act suspending the New York Tea acts, and strengthening of trade acts by Customs Commissioners act, 1767; British proposal to use revenue from taxes to pay salaries of royal officials as well as for defence of colonies; colonial avoidance of mobs and riots, substitution of Massachusetts "circular letter" for congress as method of united resistance, and appeal to "nat ural rights of man," as well as to constitutional rights of Englishmen, 1768; renewal of the attempt to establish an Anglican bishop in the colonies, 1767-1769; British opening of the southwest by the Fort Stanwix treaty of 1768, be ginning of colonial settlement of eastern Tennessee and Kentucky, and renewal of plans for "westward state mak ing," 1769; British attempt to suppress Massachusetts cir cular letter, sending of troops to Boston, and revival of old "treason" act against the colonists, 1768-1769; the Vir ginia resolutions of protest, and the (1 Boston massacre," 1769-1770; parliamentary repeal of Townshend Revenue act with exception of duty on tea, and continuation of attempted enforcement of unrepealed acts of 1764-1769.

The period of the controversy over the Townshend acts is passed over very lightly by the great majority of text-books of American history. In reality, it is one of very great significance. In the first place, it

3 Resolutions of the Stamp Act Congress, 1775, and the Massachusetts Circular Letter of 1768 (MacDonald, "Docu mentary Source Book," 137-138, 147-150; Caldwell and Persinger, "Source History United States," 170, 178).

4 J. A. James, "Readings in American History" (ScribHistory of the American Nation" (Appleton, 1914), ner's, 1914), 126-137; A. C. McLaughlin, "Readings in the 52-56; D. S. Muzzey, "Readings in American History" (Ginn, 1915), 111-120; W. M. West, "Source Book in American History" (Allyn & Bacon, 1913), 369-380.