At a party given for its reopening last year, the Museum of Modern Art in New York served champagne to five thousand guests. An alert reporter for the Times, Charlotte Curtis, noted that there were eighty cases, which, she informed her readers, amounted to 960 bottles or 7,680 three-ounce drinks. Somehow through this detail the Museum’s party at once becomes alive; a fashionable New York occasion. One sees the crush, the women eyeing each other’s clothes, the exchange of greetings, and feels the gratifying sense of elegance and importance imparted by champagne—even if, at one and a half drinks per person, it was not on an exactly riotous scale. All this is conveyed by Miss Curtis’ detail. It is, I think, the way history as well as journalism should be written. It is what Pooh-Bah, in The Mikado, meant when, telling how the victim’s head stood on its neck and bowed three times to him at the execution of Nanki-Poo, he added that this was “corroborative detail intended to give artistic verisimilitude to an otherwise bald and unconvincing narrative.” Not that Miss Curtis’ narrative was either bald or unconvincing; on the contrary, it was precise, factual, and a model in every way. But what made it excel, made it vivid and memorable, was her use of corroborative detail.
Pooh-Bah’s statement of the case establishes him in my estimate as a major historian or, at least, as the formulator of a major principle of historiography. True, he invented his corroborative detail, which is cheating if you are a historian and fiction if you are not; nevertheless, what counts is his recognition of its importance. He knew that it supplies verisimilitude, that without it a narrative is bald and unconvincing. Neither he nor I, of course, discovered the principle; historians have for long made use of it, beginning with Thucydides, who insisted on details of topography, “the appearance of cities and localities, the description of rivers and harbors, the peculiar features of seas and countries and their relative distances.”
Corroborative detail is the great corrective. Without it historical narrative and interpretation, both, may slip easily into the invalid. It is a disciplinarian. It forces the historian who uses and respects it to cleave to the truth, or as much as he can find out of the truth. It keeps him from soaring off the ground into theories of his own invention. On those Toynbeean heights the air is stimulating and the view is vast, but people and houses down below are too small to be seen. However persuaded the historian may be of the validity of the theories he conceives, if they are not supported and illustrated by corroborative detail they are of no more value as history than Pooh-Bah’s report of the imagined execution.
It is wiser, I believe, to arrive at theory by way of the evidence rather than the other way around, like so many revisionists today. It is more rewarding, in any case, to assemble the facts first and, in the process of arranging them in narrative form, to discover a theory or a historical generalization emerging of its own accord. This to me is the excitement, the built-in treasure hunt, of writing history. In the book I am working on now, which deals with the twenty-year period before 1914 (and the reader must forgive me if all my examples are drawn from my own work, but that, after all, is the thing one knows best), I have been writing about a moment during the Dreyfus Affair in France when on the day of the reopening of Parliament everyone expected the Army to attempt acoup d’état. English observers predicted it, troops were brought into the capital, the Royalist pretender was summoned to the frontier, mobs hooted and rioted in the streets, but when the day had passed, nothing had happened; the Republic still stood. By this time I had assembled so much corroborative detail pointing to a coup d’état that I had to explain why it had not occurred. Suddenly I had to stop and think. After a while I found myself writing, “The Right lacked that necessary chemical of a coup—a leader. It had its small, if loud, fanatics; but to upset the established government in a democratic country requires either foreign help or the stuff of a dictator.” That is a historical generalization, I believe; a modest one, to be sure, but my size. I had arrived at it out of the necessity of the material and felt immensely pleased and proud. These moments do not occur every day; sometimes no more than one a chapter, if that, but when they do they leave one with a lovely sense of achievement.
I am a disciple of the ounce because I mistrust history in gallon jugs whose purveyors are more concerned with establishing the meaning and purpose of history than with what happened. Is it necessary to insist on a purpose? No one asks the novelist why he writes novels or the poet what is his purpose in writing poems. The lilies of the field, as I remember, were not required to have a demonstrable purpose. Why cannot history be studied and written and read for its own sake, as the record of human behavior, the most fascinating subject of all? Insistence on a purpose turns the historian into a prophet—and that is another profession.
To return to my own: Corroborative detail will not produce a generalization every time, but it will often reveal a historical truth, besides keeping one grounded in historical reality. When I was investigating General Mercier, the Minister of War who was responsible for the original condemnation of Dreyfus and who in the course of the Affair became the hero of the Right, I discovered that at parties of the haut monde ladies rose to their feet when General Mercier entered the room. That is the kind of detail which to me is worth a week of research. It illustrates the society, the people, the state of feeling at the time more vividly than anything I could write and in shorter space, too, which is an additional advantage. It epitomizes, it crystallizes, it visualizes. The reader can see it; moreover, it sticks in his mind; it is memorable.
The same is true, verbally though not visually, of a statement by President Eliot of Harvard in 1896 in a speech on international arbitration, a great issue of the time. In this chapter I was writing about the founding tradition of the United States as an anti-militarist, anti-imperialist nation, secure within its own shores, having nothing to do with the wicked armaments and standing armies of Europe, setting an example of unarmed strength and righteousness. Looking for material to illustrate the tradition, I found in a newspaper report these words of Eliot, which I have not seen quoted by anyone else: “The building of a navy,” he said, “and the presence of a large standing army mean … the abandonment of what is characteristically American.… The building of a navy and particularly of battleships is English and French policy. It should never be ours.”
How superb that is! Its assurance, its conviction, its Olympian authority—what does it not reveal of the man, the time, the idea? In those words I saw clearly for the first time the nature and quality of the American anti-militarist tradition, of what has been called the American dream—it was a case of detail not merely corroborating but revealing an aspect of history.
Failing to know such details, one can be led astray. In 1890 Congress authorized the building of the first three American battleships and, two years later, a fourth. Shortly thereafter, in 1895, this country plunged into a major quarrel with Great Britain, known as the Venezuelan crisis, in which there was much shaking of fists and chauvinist shrieking for war. Three years later we were at war with Spain. She was no longer a naval power equal to Britain, of course, but still not negligible. One would like to know what exactly was American naval strength at the time of both these crises. How many, if any, of the battleships authorized in 1890 were actually at sea five years later? When the jingoes were howling for war in 1895, what ships did we have to protect our coasts, much less to take the offensive? It seemed to me this was a piece of information worth knowing.
To my astonishment, on looking for the answer in textbooks on the period, I could not find it. The historians of America’s rise to world power, of the era of expansion, of American foreign policy, or even of the Navy have not concerned themselves with what evidently seems to them an irrelevant detail. It was hardly irrelevant to policymakers of the time who bore the responsibility for decisions of peace or war. Text after text in American history is published every year, each repeating on this question more or less what his predecessor has said before, with no further enlightenment. To find the facts I finally had to write to the Director of Naval History at the Navy Department in Washington.
My point is not how many battleships we had on hand in 1895 and ’98 (which I now know) but why this hard, physical fact was missing from the professional historians’ treatment. “Bald and unconvincing,” said Pooh-Bah of narrative without fact, a judgment in which I join.
When I come across a generalization or a general statement in history unsupported by illustration I am instantly on guard; my reaction is, “Show me.” If a historian writes that it was raining heavily on the day war was declared, that is a detail corroborating a statement, let us say, that the day was gloomy. But if he writes merely that it was a gloomy day without mentioning the rain, I want to know what is his evidence; what made it gloomy. Or if he writes, “The population was in a belligerent mood,” or, “It was a period of great anxiety,” he is indulging in general statements which carry no conviction to me if they are not illustrated by some evidence. I write, for example, that fashionable French society in the 1890s imitated the English in manners and habits. Imagining myself to be my own reader—a complicated fugue that goes on all the time at my desk—my reaction is of course, “Show me.” The next two sentences do. I write, “The Greffulhes and Breteuils were intimates of the Prince of Wales, le betting was the custom at Longchamps, le Derby was held at Chantilly, le steeplechase at Auteuil and an unwanted member was black-boulé at the Jockey Club. Charles Haas, the original of Swann, had ‘Mr’ engraved on his calling cards.”
Even if corroborative detail did not serve a valid historical purpose, its use makes a narrative more graphic and intelligible, more pleasurable to read, in short more readable. It assists communication, and communication is, after all, the major purpose. History written in abstract terms communicates nothing to me. I cannot comprehend the abstract, and since a writer tends to create the reader in his own image, I assume my reader cannot comprehend it either. No doubt I underestimate him. Certainly many serious thinkers write in the abstract and many people read them with interest and profit and even, I suppose, pleasure. I respect this ability, but I am unable to emulate it.
My favorite visible detail in The Guns of August, for some inexplicable reason, is the one about the Grand Duke Nicholas, who was so tall (six foot six) that when he established headquarters in a railroad car his aide pinned up a fringe of white paper over the doorway to remind him to duck his head. Why this insignificant item, after several years’ work and out of all the material crammed into a book of 450 pages, should be the particular one to stick most sharply in my mind I cannot explain, but it is. I was so charmed by the white paper fringe that I constructed a whole paragraph describing Russian headquarters at Baranovici in order to slip it in logically.
In another case the process failed. I had read that the Kaiser’s birthday gift to his wife was the same every year: twelve hats selected by himself which she was obliged to wear. There you see the value of corroborative detail in revealing personality; this one is worth a whole book about the Kaiser—or even about Germany. It represents, however, a minor tragedy of The Guns, for I never succeeded in working it in at all. I keep my notes on cards, and the card about the hats started out with those for the first chapter. Not having been used, it was moved forward to a likely place in Chapter 2, missed again, and continued on down through all the chapters until it emerged to a final resting place in a packet marked “Unused.”
A detail about General Sir Douglas Haig, equally revealing of personality or at any rate of contemporary customs and conditions in the British officer corps, did find a place. This was the fact that during the campaign in the Sudan in the nineties he had “a camel laden with claret” in the personal pack train that followed him across the desert. Besides being a vivid bit of social history, the phrase itself, “a camel laden with claret,” is a thing of beauty, a marvel of double and inner alliteration. That, however, brings up another whole subject, the subject of language, which needs an article of its own for adequate discussion.
Having inadvertently reached it, I will only mention that the independent power of words to affect the writing of history is a thing to be watched out for. They have an almost frightening autonomous power to produce in the mind of the reader an image or idea that was not in the mind of the writer. Obviously they operate this way in all forms of writing, but history is particularly sensitive because one has a duty to be accurate, and careless use of words can leave a false impression one had not intended. Fifty percent at least of the critics of The Guns commented on what they said was my exposé of the stupidity of the generals. Nothing of the kind was in my mind when I wrote. What I meant to convey was that the generals were in the trap of the circumstances, training, ideas, and national impulses of their time and their individual countries. I was not trying to convey stupidity but tragedy, fatality. Many reviewers understood this, clearly intelligent perceptive persons (those who understand one always are), but too many kept coming up with that word “stupidity” to my increasing dismay.
This power of words to escape from a writer’s control is a fascinating problem which, since it was not what I started out to discuss, I can only hint at here. One more hint before I leave it: For me the problem lies in the fact that the art of writing interests me as much as the art of history (and I hope it is not provocative to say that I think of history as an art, not a science). In writing I am seduced by the sound of words and by the interaction of their sound and sense. Recently at the start of a paragraph I wrote, “Then occurred the intervention which irretrievably bent the twig of events.” It was intended as a kind of signal to the reader. (Every now and then in a historical narrative, after one has been explaining a rather complicated background, one feels the need of waving a small red flag that says, “Wake up, Reader; something is going to happen.”) Unhappily, after finishing the paragraph, I was forced to admit that the incident in question had not irretrievably bent the twig of events. Yet I hated to give up such a well-made phrase. Should I leave it in because it was good writing or take it out because it was not good history? History governed and it was lost to posterity (although, you notice, I have rescued it here). Words are seductive and dangerous material, to be used with caution. Am I writer first or am I historian? The old argument starts inside my head. Yet there need not always be dichotomy or dispute. The two functions need not be, in fact should not be, at war. The goal is fusion. In the long run the best writer is the best historian.
In quest of that goal I come back to the ounce. The most effective ounce of visual detail is that which indicates something of character or circumstance in addition to appearance. Careless clothes finished off by drooping white socks corroborate a description of Jean Jaurès as looking like the expected image of a labor leader. To convey both the choleric looks and temper and the cavalry officer’s snobbism of Sir John French, it helps to write that he affected a cavalryman’s stock in place of collar and tie, which gave him the appearance of being perpetually on the verge of choking.
The best corroborative detail I ever found concerned Lord Shaftesbury, the eminent Victorian social reformer, author of the Factory Act and child-labor laws, who appeared in my first book, Bible and Sword. He was a man, wrote a contemporary, of the purest, palest, stateliest exterior in Westminster, on whose classic head “every separate dark lock of hair seemed to curl from a sense of duty.” For conveying both appearance and character of a man and the aura of his times, all in one, that line is unequaled.
Novelists have the advantage that they can invent corroborative detail. Wishing to portray, let us say, a melancholy introspective character, they make up physical qualities to suit. The historian must make do with what he can find, though he may sometimes point up what he finds by calling on a familiar image in the mental baggage of the reader. To say that General Joffre looked like Santa Claus instantly conveys a picture which struck me as peculiarly apt when I wrote it. I was thinking of Joffre’s massive paunch, fleshy face, white mustache, and bland and benevolent appearance, and I forgot that Santa Claus wears a beard, which Joffre, of course, did not. Still, the spirit was right. One must take care to choose a recognizable image for this purpose. In my current book I have a melancholy and introspective character, Lord Salisbury, Prime Minister in 1895, a supreme, if far from typical, product of the British aristocracy, a heavy man with a curly beard and big, bald forehead, of whom I wrote that he was called the Hamlet of English politics and looked like Karl Marx. I must say that I was really rather pleased with that phrase, but my editor was merely puzzled. It developed that he did not know what Karl Marx looked like, so the comparison conveyed no image. If it failed its first test, it would certainly not succeed with the average reader and so, sadly, I cut it out.
Sources of corroborative detail must of course be contemporary with the subject. Besides the usual memoirs, letters, and autobiographies, do not overlook novelists and newspapers. The inspired bit about the ladies rising to their feet for General Mercier comes from Proust as do many other brilliant details; for instance, that during the Affair ladies had “A bas les juifs” printed on their parasols. Proust is invaluable not only because there is so much of him but because it is all confined to a narrow segment of society which he knew personally and intimately; it is like a woman describing her own living room. On the other hand, another novel set in the same period, Jean Barois by Roger Martin du Gard, considered a major work of fiction on the Affair, gave me nothing I could use, perhaps because visual detail—at least the striking and memorable detail—was missing. It was all talk and ideas, interesting, of course, but for source material I want something I can see. When you have read Proust you can see Paris of the nineties, horse cabs and lamplight, the clubman making his calls in white gloves stitched in black and gray top hat lined in green leather.
Perhaps this illustrates the distinction between a major and a less gifted novelist which should hold equally true, I believe, for historians. Ideas alone are not flesh and blood. Too often, scholarly history is written in terms of ideas rather than acts; it tells what people wrote instead of what they performed. To write, say, a history of progressivism in America or of socialism in the era of the Second International by quoting the editorials, books, articles, speeches, and so forth of the leading figures is easy. They were the wordiest people in history. If, however, one checks what they said and wrote against what actually was happening, a rather different picture emerges. At present I am writing a chapter on the Socialists and I feel like someone in a small rowboat under Niagara. To find and hold on to anything hard and factual under their torrent of words is an epic struggle. I suspect the reason is that people out of power always talk more than those who have power. The historian must be careful to guard against this phenomenon—weight it, as the statisticians say—lest his result be unbalanced.
Returning to novels as source material, I should mention The Edwardians by V. Sackville-West, which gave me precise and authoritative information on matters on which the writers of memoirs remain discreet. Like Proust, this author was writing of a world she knew. At the great house parties, one learns, the hostess took into consideration established liaisons in assigning the bedrooms and each guest had his name on a card slipped into a small brass frame outside his door. The poets too serve. Referring in this chapter on Edwardian England to the central role of the horse in the life of the British aristocracy, and describing the exhilaration of the hunt, I used a line from a sonnet by Wilfrid Scawen Blunt, “My horse a thing of wings, myself a god.” Anatole France supplied, through the mouth of a character in M. Bergeret, the words to describe a Frenchman’s feeling about the Army at the time of the Affair, that it was “all that is left of our glorious past. It consoles us for the present and gives us hope of the future.” Zola expressed the fear of the bourgeoisie for the working class through the manager’s wife in Germinal, who, watching the march of the striking miners, saw “the red vision of revolution … when on some somber evening at the end of the century the people, unbridled at last, would make the blood of the middle class flow.” In The Guns there is a description of the retreating French Army after the Battle of the Frontiers with their red trousers faded to the color of pale brick, coats ragged and torn, cavernous eyes sunk in unshaven faces, gun carriages with once-new gray paint now blistered and caked with mud. This came from Blasco IbáñnTez’s novel The Four Horsemen of the Apocalypse. From H. G. Wells’s Mr. Britling Sees It Through I took the feeling in England at the outbreak of war that it contained an “enormous hope” of something better afterward, a chance to end war, a “tremendous opportunity” to remake the world.
I do not know if the professors would allow the use of such sources in a graduate dissertation, but I see no reason why a novelist should not supply as authentic material as a journalist or a general. To determine what may justifiably be used from a novel, one applies the same criterion as for any nonfictional account: If a particular item fits with what one knows of the time, the place, the circumstances, and the people, it is acceptable; otherwise not. For myself, I would rather quote Proust or Sackville-West or Zola than a professional colleague as is the academic habit. I could never see any sense whatever in referring to one’s neighbor in the next university as a source. To me that is no source at all; I want to know where a given fact came from originally, not who used it last. As for referring to an earlier book of one’s own as a source, this seems to me the ultimate absurdity. I am told that graduate students are required to cite the secondary historians in order to show they are familiar with the literature, but if I were granting degrees I would demand primary familiarity with primary sources. The secondary histories are necessary when one starts out ignorant of a subject and I am greatly in their debt for guidance, suggestion, bibliography, and outline of events, but once they have put me on the path I like to go the rest of the way myself. If I were a teacher I would disqualify anyone who was content to cite a secondary source as his reference for a fact. To trace it back oneself to its origin means to discover all manner of fresh material from which to make one’s own selection instead of being content to re-use something already selected by someone else.
Though it is far from novels, I would like to say a special word for Who’s Who. For one thing, it is likely to be accurate because its entries are written by the subjects themselves. For another, it shows them as they wish to appear and thus often reveals character and even something of the times. H. H. Rogers, a Standard Oil partner and business tycoon of the 1890s, listed himself simply and succinctly as “Capitalist,” obviously in his own eyes a proud and desirable thing to be. The social history of a period is contained in that self-description. Who would call himself by that word today?
As to newspapers, I like them for period flavor perhaps more than for factual information. One must be wary in using them for facts, because an event reported one day in a newspaper is usually modified or denied or turns out to be rumor on the next. It is absolutely essential to take nothing from a newspaper without following the story through for several days or until it disappears from the news. For period flavor, however, newspapers are unsurpassed. In the New York Times for August 10, 1914, I read an account of the attempt by German officers disguised in British uniforms to kidnap General Leman at Liège. The reporter wrote that the General’s staff, “maddened by the dastardly violation of the rules of civilized warfare, spared not but slew.”
This sentence had a tremendous effect on me. In it I saw all the difference between the world before 1914 and the world since. No reporter could write like that today, could use the word “dastardly,” could take as a matter of course the concept of “civilized warfare,” could write unashamedly, “spared not but slew.” Today the sentence is embarrassing; in 1914 it reflected how people thought and the values they believed in. It was this sentence that led me back to do a book on the world before the war.
Women are a particularly good source for physical detail. They seem to notice it more than men or at any rate to consider it more worth reporting. The contents of the German soldier’s knapsack in 1914, including thread, needles, bandages, matches, chocolate, tobacco, I found in the memoirs of an American woman living in Germany. The Russian moose who wandered over the frontier to be shot by the Kaiser at Rominten came from a book by the English woman who was governess to the Kaiser’s daughter. Lady Warwick, mistress for a time of the Prince of Wales until she regrettably espoused socialism, is indispensable for Edwardian society, less for gossip than for habits and behavior. Princess Daisy of Pless prattles endlessly about the endless social rounds of the nobility, but every now and then supplies a dazzling nugget of information. One, which I used in The Zimmermann Telegram, was her description of how the Kaiser complained to her at dinner of the ill-treatment he had received over the Daily Telegraph affair and of how, in the excess of his emotion, “a tear fell on his cigar.” In the memoirs of Edith O’Shaughnessy, wife of the First Secretary of the American Embassy in Mexico, is the description of the German Ambassador, Von Hintze, who dressed and behaved in all things like an Englishman except that he wore a large sapphire ring on his little finger which gave him away. No man would have remarked on that.
In the end, of course, the best place to find corroborative detail is on the spot itself, if it can be visited, as Herodotus did in Asia Minor or Parkman on the Oregon Trail. Take the question of German atrocities in 1914. Nothing requires more careful handling because, owing to post-war disillusions, “atrocity” came to be a word one did not believe in. It was supposed because the Germans had not, after all, cut off the hands of Belgian babies, neither had they shot hostages nor burned Louvain. The results of this disbelief were dangerous because when the Germans became Nazis people were disinclined to believe they were as bad as they seemed and appeasement became the order of the day. (It strikes me that here is a place to put history to use and that a certain wariness might be in order today.) In writing of German terrorism in Belgium in 1914 I was at pains to use only accounts by Germans themselves or in a few cases by Americans, then neutral. The most telling evidence, however, was that which I saw forty-five years later: the rows of gravestones in the churchyard of a little Belgian village on the Meuse, each inscribed with a name and a date and the legend “fusillé par les Allemands.” Or the stone marker on the road outside Senlis, twenty-five miles from Paris, engraved with the date September 2, 1914, and the names of the mayor and six other civilian hostages shot by the Germans. Somehow the occupations engraved opposite the names—baker’s apprentice, stonemason, garçon de café—carried extra conviction. This is the verisimilitude Pooh-Bah and I too have been trying for.
The desire to find the significant detail plus the readiness to open his mind to it and let it report to him are half the historian’s equipment. The other half, concerned with idea, point of view, the reason for writing, the “Why” of history, has been left out of this discussion although I am not unconscious that it looms in the background. The art of writing is the third half. If that list does not add up, it is because history is human behavior, not arithmetic.
Harper’s Magazine, July 1965.