Summary of Part 1
Foucault begins The Archaeology of Knowledge by considering two competing tendencies in the work of contemporary historians. Histories who study ideas and knowledge—how different concepts emerge and spread—tend to study discontinuity: the sudden appearance of new, unexpected ideas. In contrast, historians who study more traditional topics like material things and political events, tend to study continuity: how a stable structure or narrative underlies a series of events. Foucault says both trends are the effect of an underlying problem, which he calls the problem of the “document.” Once upon a time, historians treated documents as records of historical events. It was “inert material through which [the historian] tries to reconstitute what men have done or said.” Now, Foucault argues, it is instead necessary to understand the documents themselves and how the documents came to be. This means we are not studying ideas or events and using documents as records of them. Instead, documents are their own “monuments” that need to be excavated and analyzed.
If documents are monuments themselves worthy of study, rather than records of other monuments like war, then Foucault says we are dealing not with history but “archaeology”: the study of discursive monuments. He outlines four consequences of this change from history to archaeology and from the study of events to the study of discourse. First, it’s hard to know in advance what “series” or chain of related things need to be studied. When you’re studying war, you can study all the battles in the war. But when you’re studying the history of an idea, there is no boundary or limit set in advance. Historians have to figure out not just what they are studying, but which materials form the archive of their study. This has led to very different time horizons for historical study, such as the long period in traditional history or the short periods in the history of ideas.
The second and third consequence of the shift Foucault has noticed are related: history is increasingly “discontinuous,” and this is also why it cannot be “total.” It used to be that historian avoided discontinuities or gaps in their archives or the stories they were telling. Now, they exploit discontinuity, cutting across different scales of study and blending traditionally distinct archives like philosophy and medicine. If discontinuity is now the rule rather than the exception of historical analysis, Foucault thinks it needs to be better theorized. He also wants to better theorize what it means that it is no longer possible to have a total history of a civilization as a whole, as if there could be some master plot that a given society follows in its development. Such a history tends to supply one principle that organizes all the events in a timeline; for instance, a society might be seen to become increasingly more rational. In contrast, Foucault wants to talk about a “general” history of “dispersion,” which means not unifying events under a single principle, but exploring all the different relations between very different and simultaneously unfolding timelines and domains.
The fourth consequence, hinted at in all the above, is that history has reached a crisis in methodology. How to proceed in building an archive when your object of study is not as neatly bounded as the past? How to connect different histories in your story, such as the history of medicine and the history of philosophy? As historians answer these questions, they will also begin working in fields related to, but usually distinct from, history. Historians will have to talk with linguists, sociologists, psychologists, and so on, and in doing so, might find they are studying similar things, such as the development of new knowledge systems or social formations. But at the same time, this new archaeology is distinct from what has been called the philosophy of history. It is not trying to tell us what history is, but asking how different documents came to be able to say different things at different points in history.
Foucault presents The Archaeology of Knowledge as a response to the urgent question of how to do a history of discontinuity attuned to the status of documents over and above the mere events recorded in them. But first, he acknowledges the resistance of historians—or people in general—to accepting discontinuity. He thinks this is because people like coherent, simple narratives, which also give a special place for “the sovereignty of consciousness.” That means the idea that history is created by the agency of individuals, according to their intentions and manifest in their actions. If, instead, we are looking at how the statements in documents change over time, we are looking at something indifferent to any one individual. That means we decenter human agency in historical analysis, and that can be scary for people who think human agency is the center of the world.
For Foucault, Karl Marx was the first—and until now really the only—person to conceive of history beyond the terms of individual agency. Rather than tell the story of kings or geniuses, Marx told the story of classes of people and successions of “modes of production,” or the ways in which a society’s economy was organized. This is a kind of “impersonal” history, because it is a history of groups and structures rather than persons. Foucault is inspired by this decentering of “anthropology” or the study of people. At the same time, he wants to move this insight further into a more explicit methodology about how to study history when we give up on the illusion of individuals being its cause and essence. How do you do a history without biography?
In the final pages of his Introduction, Foucault explains that in some ways The Archaeology of Knowledge is a theory of the method he had used in previous historical works, including Madness and Civilization and The Order of Things. These books were, on the surface, histories of ideas and the sciences, looking at how modern ideas of mental illness and taxonomy that emerged from the Renaissance to the present. In these books, he had practiced a method of looking at the document as a monument to be studied in its own right, decentering the human or individual agency of subjects, and exploiting, rather than ignoring, gaps or discontinuities in the archive. Throughout The Archaeology of Knowledge, Foucault will refer to these other books as “concrete” examples of his larger method. But in this book, he will above all aim to provide a systemic explanation of this method and why it is different from the histories of ideas that other people are currently writing.
Analysis of Part 1
The first sentence of The Archaeology of Knowledge reveals a lot about Foucault’s ambitions. He writes of “historians” in general and identifies two major trends that characterize their work from the past generation. In doing so, he asserts his own authority to assess not only the work of this or that historian, but of the discipline as a whole. This confidence can be quite surprising if you have not read Foucault before. He writes with a liveliness and fierceness in his prose that carry his argument forward. But along the way, he provides very few examples of the work he is seemingly attacking. It is the power of his generalization that we are invited to accept it without having it filled in with specifics.
Where specifics do come in to this Introduction is at the end, when Foucault reminds us he has just written three histories himself in the past decade: Madness and Civilization, The Birth of the Clinic, and The Order of Things. There are many effects that come from naming his works. For one, after speaking generally of historians but mentioning none by name, it is as if Foucault is saying he is the only historian worth mentioning. For another, it is a real indication of his qualifications that he has written on such a range of subjects in such incisive ways over the course of such a short period of time. It suggests there is a true method to his madness that makes it possible to accomplish so much so quickly. Foucault presents The Archaeology of Knowledge as a sort of meta-history. He will speak in generalities at times, because the specifics and concrete details are in the histories he has already written. They are like practices of the theory that Archaeology provides.
Although he mentions no historians by name, Foucault does imply that he is joining an intellectual tradition through the names he drops, including Karl Marx and Friedrich Nietzsche. These are big names, and each belongs to a man who founded a whole new school of thought. In a similar way, Foucault seems to be suggesting he is inaugurating his own school of historiography. It is perhaps for this reason he must call it “archaeology” to distinguish it even in name from the “history” of his peers. It is as if, after The Archaeology of Knowledge, history (or at least historiography, the study of history) will never be the same again.
But it would be against the grain of Foucault’s argument to suggest he will be the inventor of this line of thinking, in the same way that Marx gave birth to Marxism. For above all, Foucault is saying that individuals don’t have that sort of power to shape discourse. There are larger and more impersonal, but unspoken, systems of rules that regulate what can be said or written at a given time. Thus, at the same time that Foucault positions himself in a special place to argue for a new way of doing history, he is demoting the role of individuals like himself in the first place.
Perhaps it would be best to call Foucault ambivalent in The Archaeology of Knowledge: split between different goals and aspirations. He writes with confidence as an individual at the same time he attacks the “sovereignty of consciousness.” He argues for the impersonality of discourse at the same time he writes his own history of it, as a single person. For Foucault, such ambivalence is a resource rather than a threat. As he will argue with greater detail in Part 4, the aim of discourse analysis is not to reduce complexity or resolve contradiction or eliminate ambivalence. Rather, it is to positively map out the tensions and latent conflicts within a field of discourse. Just as we do not expect coherency from discourse, we should not expect complete coherency from a person, like Foucault.
