The following essay is based loosely upon informal remarks presented to the Sixth International Summer School in History of Science at Uppsala, Sweden (7-13 June 1998). The presentation provoked a general discussion on the last day of the meeting, whose theme was "The Structure of Knowledge: Classifications of Science and Learning since the Renaissance."
During the past week, we have been given a wonderful guided tour of five centuries of thought about the organization of knowledge. Paula Findlen demonstrated the role played by specific Early Modern physical locations (symbolic and real) such as "memory theaters," anatomical theaters and botanical gardens, in preparing the way for the emergence of the idea of the modern scientific discipline. "Discipline" here referred not just to a distinctly delineated arena of knowledge, but also to the control of behavior required by those who inhabited these spaces. Robert Darnton examined the structure of knowledge represented in Diderots Encyclopédie and in subsequent encyclopedist projects, and he linked these changing intellectual structures to the transformation of social structure and political power in eighteenth-century France. Nicholaas Rupke explored the work of Alexander von Humboldt, who quite literally mapped all kinds of knowledge (not just geographic) obtained from his nineteenth-century world travels. Rupke then presented his own "geography" of Humboldt himself, describing the multiplicity of biographical "Humboldts" who have been created in different places at different times. Finally, Sven Widmalm analyzed the "system" of knowledge that emerged in the modern university of the late nineteenth and early twentieth centuries. This system, which was simultaneously national and international, was increasingly seen less as a philosophical reality, and more as a professional tool, expediting the expansion in scope and prestige of twentieth-century science and scientists.
Throughout the week, all four lecturers utilized the concept of the structure of knowledge as a lens through which to examine the history of science. What I would like to do now is to replace this lens with a mirror: a mirror that we hold up, not to the past, but to ourselves. By doing so, we can consider the significance of structures of knowledge in defining ourselves and our own scholarly activity as historians of science. The structure of knowledge under which we currently operate is, for the most part, still strongly governed by the system of modern disciplines described by Widmalm, the system whose evolution was traced by the previous three speakers. How does this disciplinary structure shape our own work and our self-identity? My preliminary conclusion is that, paradoxically, we historians of science suffer from both too much discipline, and too little discipline.
Too much discipline
The structure of knowledge in the History of Science was first codified in 1913, when George Sarton constructed the first "Analytical Bibliography" for the first volume of Isis. 1 From the very beginning, Sarton seems to have struggled with a plenitude of disciplines, and he constructed a tripartite system to attempt to accommodate them all. The first part of his bibliography, which he considered the most important, was the chronological systemization of the field, drawing its structure from the discipline of history. In fact, the first twelve sections (7 pages of entries) of this chronological system were not explicitly defined chronologically, but instead by a mixture of region, nation and era: Antiquity, Cuneiform Civilization, Egypt, Classical Antiquity, Greece, Rome, Byzantium, The Middle Ages, India, Islam, Orient, and Extreme-Orient. A century-by-century classification followed, and this section constituted the largest portion of the whole (20 pages).
The second part of the bibliography (10 pages) was a subject-driven classification for articles not easily classified chronologically. Here, Sarton explicitly drew upon the organization of scientific knowledge as constructed by Comte and Ostwald. Modifying these precedents slightly for his own purposes, he separated into distinct categories articles on Mathematics, Mechanics, Astronomy, Physics, Chemistry, Technology, Biology, Mineralogy and Geology, Botany, Zoology and Anatomy, Medicine, and Psychology. Sarton emphasized that - unlike the system of Comte, for example - his system had only "practical" value, and was of no philosophical interest.
The final part of Sartons bibliography (5 pages) was made up of an eclectic mix of "disciplines connexes que lhistorien de la science ne peut ignorer, mais qui sont cependant bien distinctes de son domaine de travail habituel." Entries consisted of: Prehistory; Anthropology and Ethnology; Archeology, Museums and Collections; Science and Art; Science and Occultism; Science and Religion; and Science and Philosophy.
From the very beginning, then, History of Science was "all over the map" of the modern disciplines. To constitute the field, Sarton was compelled to draw upon the structure and resources of History, of Science and of other disciplines in the humanities and social sciences. In spite of Sartons synthetic vision, and his privileging of the historical (his chronological index), the titles of most of that sections entries indicate that historians of science, when defining a research topic, were more strongly influenced by the modern disciplinary classification of the sciences than by the categories of the historian or the eclectic approach of the last section of the bibliography. 2 Considering the scientific background of many historians of science at that time, perhaps this is not surprising. It was also a useful strategy (whether conscious or not) for claiming the authority and prestige of science itself for the new field.
Nonetheless, this strong emphasis in practice upon the structure of the scientific disciplines had certain drawbacks. It enabled and encouraged an internal, intellectual history approach to the topic, an approach that often spoke more to scientists than to other scholars. It also created, then widened, the division between the histories of science, technology and medicine, recreating the professional divisions that separated scientists from engineers and doctors. Only recently have historians of science begun to mine the rich potential of topics represented by the third section of Sartons system of classification - topics such as the history of collecting, the history of science and art, the history of the place of science. These topics (well-represented in this weeks lectures) cross, transcend, or sometimes simply ignore the older boundaries of the field, the boundaries of the scientific disciplines themselves. It is also the case that many sessions at recent History of Science Society (HSS) meetings could just as easily have been placed on the program for the Society for the History of Technology (SHOT), and vice-versa. Still, there remain strong forces at work encouraging historians of science to identify themselves with a particular scientific discipline - to be a "Historian of Physics," "Historian of Biology," etc. - and even stronger forces keep HSS and SHOT from merging into a single society. As long as these forces remain at work, they will present obstacles to the execution of certain kinds of histories of science. It may be that today, the drawbacks outweigh the advantages of this strong dependence upon the disciplines of science.
Too little discipline
While it may be the case that the History of Science has "suffered" in some ways from its strong dependence upon the disciplinary structure of science, it also suffers, paradoxically, from a lack of discipline. We are not scientists, after all, nor have we committed ourselves to the construction of an identity as "regular" historians. Almost a century after its founding, our field remains ambiguously placed within the university. We continue to struggle to discover and to demonstrate to others just WHERE we belong. We are between disciplines, between science and history, and we transgress the boundaries of each. Robert Darnton described how problematic such boundary-crossers can be. To call someone a "dog" or a "son of a bitch" in eighteenth-century France was a powerful invective, because of the ambiguous and unclassifiable character of dogs. Neither indoor animals nor outdoor animals, they existed on the periphery of both worlds, and this is what made them suspect. Like Darntons dogs, we are neither "inside" nor "outside" the discipline of science. Indeed, within the context of the so-called "Science Wars" (perhaps better named "History of Science Wars"), a few of us have even been cursed as "sons of bitches"! Perhaps this is appropriate? Historians, on the other hand, seem more likely to curse us with benign neglect rather than outright hostility, and it is not clear which is preferable.
Many historians of science find this "in-between-ness" an advantage rather than a disadvantage; an opportunity, not a problem. Some even seek out additional disciplinary boundaries to transgress, for example the boundary between history and fiction. The question of the relationship between historical writing and fictional writing was raised during our discussion of Nicholaas Rupkes talk on the many different "Humboldts" that biographers (including himself) have created. Can they all be "historical fact"? If not, are they all then necessarily "fiction"? Rupke asserted that, in his opinion, all of these Humboldts were historically legitimate, and he confessed that he constructed his own Humboldt simply to suit himself, not to achieve some ideally "objective" historical biography. What made his Humboldt historical fact rather than fiction was Rupkes dedication to the craft of history, his ability to convince others, in the terms of our own historical culture, that his Humboldt was real. This "confession" was clearly problematic for some in the audience, specifically, for the graduate students and younger scholars who, one might have guessed, would have been more willing than their staid elders to accept such an approach. But in fact, it makes sense that scholars-in-training should be most concerned with the lines that demarcate their chosen field of study: they, after all, are still in the process of finding their own place within the structure of knowledge. The freedom of transgression is more typically expressed by those who can afford to ignore boundaries; those whose reputation, within the prescribed limits of the field, is already well established. From the lofty heights of a perch in a well-endowed chair, the boundaries are fully visible, and it is easy to imagine an intellectual journey that crosses them at will. From the ground, however, where the borders arent always as evident, it can be harder to find ones way.
This observation leads to a general conclusion concerning the implications of this brief and informal discussion of the role of disciplines in the History of Science. I believe the importance of this kind of discussion lies, not so much in its significance for scholarship, but for pedagogy. It would be great if the academic world constituted a border-free landscape of ideas, a sort of "European Union" of the mind, in which scholars could wander at will, and where their ideas would constitute a common currency freely exchanged among all sorts of thoughtful persons. But this is not a realistic vision. When it comes to getting and keeping a job, and thriving in the real world of the academy, disciplines and boundaries matter. We owe it to our graduate students to provide them with a map of this terrain, in order to enable them to find their own way. Where does the History of Science lie within the structure of knowledge today? While this brief essay has by no means answered that question, I hope that it has provided a context and an impetus to stimulate the dialogue necessary to arrive at that answer.