After I was awarded the doctorate in Philosophy in 1957 and served my obligatory military duty [in the form of six months of active duty, to be followed by five and a half years of National Guard meetings], my first posting in what would be a half century career as a university professor called for me to teach European History to Freshmen and Sophomores at Harvard. Inasmuch as my only previous encounter with the subject had been Mr. Wepner's Modern European History course in Forest Hills High School, you will not be surprised to hear that I learned a good deal on the job. One thing that fascinated me about the discipline of History was the enormous difference in the sheer quantity of data available to historians of different periods. Since the course in which I was teaching ran from Caesar to Napoleon, I was able to get a pretty good sense of the range of challenges confronting historians of Imperial Rome, the Merovingian and Carolingian eras, the High Middle Ages, the Renaissance, the Reformation, and the seventeenth and eighteenth century English and French revolutions.
I quickly formed the idea that there are two sorts of historians: those who have too little data, and those who have too much. The theoretical and explanatory problems facing these two groups of historians are entirely different, with consequences, I eventually concluded, of a far-reaching ideological nature. In the category of those with too little data, I put F-L Ganshof, Marc Bloch, and Henri Pirenne [knowledgeable readers will immediately recognize that my knowledge of historiography is utterly out of date], all historians of what is usually called the Middle Ages. In the category of those with too much data I put Georges Lefebvre and Alfred Cobban, whose sparkling debate about the nature of the French Revolution was one of the most brilliant moments of mid-twentieth century European historiography.
Ganshof, Bloch, Pirenne, and their colleagues were forced to reconstruct eight centuries of European history from a scattering of documents, artifacts, structures, and disinterred remains, rather like a paleontologist elaborating a dinosaur skeleton from a leg bone and a few teeth. There was no question of selection. They used every scrap of evidence available, and pounced on each new discovery eagerly for whatever it could reveal about a time all but hidden from their view. I recall the shock with which I realized that the famous Pirenne Thesis [dating the economic decline of Western Europe from the seventh century closing of Mediterranean trade by the expansion of Islam rather than from the fifth century barbarian invasions of Rome] rested on a chance remark by Gregory of Tours and a few bits of evidence from sixth century trading records. As I read the magisterial accounts of the development of feudalism by Ganshof and Bloch, I often thought of the annual Statistical Abstract of the United States, a copy of which sat on my shelves. The historians of medieval Europe would have sold their souls for one page of that thousand page compilation of data for a single year of the eight hundred year span from 476 to 1276 A.D.! Bloch could not have said with any specificity how many hectares were under cultivation in Burgundy in the eleventh century, or what the population was in the Loire valley in 915 A. D.
The problem confronting Cobban, Lefebvre and their fellow students of the French Revolution was exactly the opposite. So much documentary material had accumulated and was readily available that even the most indefatigable historian could not hope to look at it all. This was borne in upon me when I began to prepare my lectures in Social Sciences 2 on the causes of the French Revolution. The course met in section twice a week, and once a week all 180 students, with their six instructors, gathered for a public lecture by one of us. Since my five colleagues were all brilliant, up and coming Assistant Professors of History, there was a good deal of pressure on me to give a decent account of myself. [One of my colleagues went on to become the Provost of Yale and President of the University of Chicago; a second was appointed the Librarian of Congress; a third became a distinguished professor at Princeton. I was way over my head!] As I rummaged around in Widener Library for some bits of data to flesh out my story of the revolution, I came upon mentions of the cahiers that the royal government solicited from the its subjects in preparation for the meeting of the National Assembly, a gathering of the three great estates [the clergy, the nobility, and the bourgeoisie.] These "notebooks" are individual accounts of grievances, desires, and demands from every corner of society, and they are all preserved to this day in the Bibliotheque Nationale. They are self-evidently an invaluable resource for any historian attempting to form a picture of France in 1789. But the archive is vast, and there is simply no way that one can read it all. What is more, the cahiers are only one of many such archival resources. Alongside them are the annual reports of the Royal Intendants, government officials charged both with reporting on the state of agriculture and manufacture throughout France and also with bringing the latest innovations in modern scientific production to rather backward and out-of-touch provincial estates and monasteries.
Faced with this embarras de richesse, historians are forced to choose what they will investigate. And inevitably, unavoidably, those choices are ideologically shaped and directed. Lefebvre is customarily referred to as a "Marxist historian," but that should not mislead anyone to imagine that he was a party hack pushing a line laid down in Moscow. When one reads his great work, La Révolution Franҫaise, one finds anything resembling partisan pleading confined to a brief Introduction and an equally brief Conclusion. The intervening six hundred pages are a meticulously detailed account of the events leading up to and constituting the revolution. But that account necessarily expresses a series of choices from the virtually limitless array of data available, and it is in those choices that Lefebvre's Marxism finds expression.
I have written of these ideologically guided choices in my tutorial "The Study of Society." There, my focus is on index numbers and the like, but I first became aware of the problem fifty-five years ago as a young Instructor in Philosophy and General Education, frantically preparing to lecture to a group of undergraduates on the causes of the French Revolution.
If I knew a great deal more than I do, I think it would be great fun to give a course devoted to a series of case studies of the methodology of historiography with a focus on the implications for that discipline of the quantity of original data available.