Archeology is the term Foucault gives to his method, which seeks to describe discourses in the conditions of their emergence and transformation rather than in their deeper, hidden meaning, their propositional or logical content, or their expression of an individual or collective psychology. Archeological analysis studies discourse only at its level of positive existence, and never takes discourse to be a trace or record of something outside of itself. In his discussion of the archive, Foucault writes that the term 'archeology' marks the distance necessary for the historian to be able to describe the archive with any clarity. This distance is not just a methodological requirement, but a crucial and pervasive feature of the history that the archeological method tries to describe: a history defined throughout by difference. 'Archeology' also has strong connotations of positivity; Foucault's method always describes only the positive, verifiably extant aspect of discourse, as one might describe a physical artifact or 'monument.'
The archive is usually taken to be the total set of collected texts from a given period (or for history altogether). Foucault describes the archive in terms of the conditions of the possibility of its construction, thus changing it from a static collection of texts to a set of relations and institutions that enable statements to continue to exist (i.e., to become part of an archive). Thus, for Foucault the archive is not a set of things or even a set of statements, but rather a set of relations: it is 'the general system of the formation and transformation of statements.'
Discourse is the object of Foucault's history. It is extremely wide-ranging and variable, tending to cross over almost every traditional historical unity (from the book to the spirit of an age); but it does so only because it has a very specific level of existence that has never before been analyzed in and of itself. This level is defined in a way similar to that of the statement (the basic element of discourse) and that of the enunciative function (the function by which discourse operates), as an aspect of language that captures its emergence and transformation in the active world. The analysis of discourse rigorously ignores any fundamental dependence on anything outside of discourse itself; discourse is never taken as a record of historical events, an articulation of meaningful content, or the expression of an individual or collective psychology. Instead, it is analyzed strictly at the level of 'things said,' the level at which statements have their 'conditions of possibility' and their conditions of relation to one another. Thus, discourse is not just a set of articulated propositions, nor is it the trace of an otherwise hidden psychology, spirit, or encompassing historical idea; it is the set of relations within which all of these other factors gain their sense (their conditions of possibility).
This is a key term in Foucault's attempt to make his method consistent as a theoretical structure. The long, central chapter on 'The Enunciative Function' serves to describe a specific, so far unrecognized level of existence for signs: Foucault calls this level the statement. In trying to define the statement, however, Foucault ends up defining the enunciative function by which the level of the statement operates. We have generally analyzed pieces of language based on their content (whether this is a proposition, an expression of a psychology, or both) or based on their material existence (their appearance once, at a specific time and place). If we analyze a statement in terms of the enunciative function, we seek to describe the discursive conditions under which it could be said, rather than the grammatical, propositional, or strictly material conditions under which it could be formulated. Thus, an enunciation always involves a position from which something is said; this position is not defined by a psychology, but by its place within (and its effect on) a field of discourse in all its complexity. The enunciative function, then, designates that aspect of language by which statements relate to other statements.
The set of relations between discursive positivity, knowledge, and science that archeological analysis examines at the threshold of epistemologization (see above) is the episteme. The episteme is not itself a form of knowledge, and it has no general content in and of itself; it is not a world- view or 'a slice of history common to all branches of knowledge' in a given period. The term refers only to a level of relations involving knowledge and science as they emerge within a discursive positivity; these relations are various and shifting, even for a single period.
The positivities (see above) that constitute discursive formations and relations form a 'historical a priori, a level of historical language which other modes of analysis depend on but fail to address. Discourse functions at the level of 'things said;' thus, any analysis of the formal structure, hidden meaning, or psychological traces of discourse take the level of discourse itself for granted, as a kind of raw material that is difficult to recognize due to its operation at the level of existence itself. It is important to note that the historical a priori constituted by the positivity of discourse is not an a priori in the usual sense of a formal philosophical principle. Instead, the historical a priori is simply a feature of the level of discourse as opposed to other levels of analysis; it does not remain stable as a single principle with a single content, but rather shifts with the transformations of the positivities themselves.
Foucault's opposes two terms for knowledge: connaissance refers to a specific corpus of knowledge or a discipline (it is knowledge as an object, known by a removed subject); savoir, at least for Foucault, refers to a kind of knowledge that is underlying but explicit and describable. Foucault's method treats knowledge in the sense of savoir, as 'the conditions that are necessary in a particular period for this or that type of object to be given to connaissance' as something that is known. In short, 'knowledge,' as the major focus of Foucault's method refers to the discursive conditions of possibility for what we generally understand as objective or subjective 'knowledge.' At one point in 'Science and Knowledge,' Foucault describes the archeological method in these terms: 'Instead of exploring the consciousness/knowledge (connaissance)/science axis (which cannot escape subjectivity), archeology explores the discursive practice/knowledge (savoir)/science axis.'
Material repeatability is a defining characteristic of the statement. It is also a kind of paradox: if we identify a single statement solely on the basis of its specific material existence, that statement will never be truly repeatable (it will be a different statement with each new articulation); but if we identify a statement solely on the basis of what it 'means' (i.e., its propositional content), that statement can be repeated ad infinitum, without regard to the differences in its material, time-space coordinates. The aspect of articulated language that Foucault designates 'statement,' however, lies between these two poles. Its material coordinates are important, but not absolutely binding. Two sentences printed at different times (even, in some cases, with different words) may be identical as statements, and two sentences with exactly the same content (i.e., the same words) may constitute two different statements. 'Material repeatability' refers to the first of these two possibilities, in which the statement is both materially based and repeatable.
The uvre is the set of all texts attributed to a single author. Along with smaller unities like the book and broader ones like the idea of historical development, the uvre is one of the received ideas that Foucault's work intends to challenge (see section two). The uvre is an artificial notion that does not bear up under scrutiny. The idea that a set of texts is unified by virtue of being the expression of a single subject misses the diversity of ways in which those texts relate to their 'author.' The attribution of a text to an author has a different function in documents published posthumously than in documents approved for publication during the author's lifetime; a survey filled out by the author differs in this manner from a novel or a contract. Foucault will eventually replace the idea of the unifying author with a range of actual and possible subject positions from which statements can be made (see section five). These subject positions are defined within the enunciative field, and are independent of any actual person; anyone can write from these positions, and any one author can write from multiple positions (see section eight).
In the chapter entitled 'Rarity, Exteriority, Accumulation' (see section eleven), Foucault begins to use the term 'positivity' to designate an approach to discourse that excludes anything lying beneath it or hidden within it. For archeology, discourse is to be described only on the level of its basic, operative existence, its existence as a set of emerging and transforming statements (and relations between statements). In this sense, archeology addresses only the 'positivities' of discourse. Further on, Foucault uses 'positivity' almost always in noun form, as a catch-all term for statements, discursive formations, or sub-formations like sciences; any one of these (or any set of relations between them) is a positivity.
The statement is the basic unit of discourse, and therefore the basic unit analyzed in the archeological method. The statement has, however, no stable unit; depending on the conditions in which it emerges and exists within a field of discourse, and depending on scope of the 'field of use' in which it is to be analyzed, anything from a scientific chart to a sentence to a novel can be a statement. This makes the statement difficult to define in and of itself, and Foucault ends up defining it not in terms of a stable unit (like the sentence), but in terms of a specific field of function and a corresponding level of the analysis of signs. The enunciative function defines the level at which the statement operates; at issue is how a set of signs emerges and functions in relation to a field of other statements. The level of analysis by which we can describe the statement lies between the analysis of grammar and propositional content on the one hand, and the fact of pure materiality on the other; the analysis of statements works at the level of the active life of language as it functions in a discourse. This in-between status of the statement, in which it is neither just content nor just material, gives statements the definitive quality of 'material repeatability' (see below).
Early on in the Archeology, Foucault repeatedly mentions the analysis of thresholds as one of the key elements in his method. The term gains more specificity, however, in the second-to-last chapter, 'Science and Knowledge.' A threshold, in basic terms, is the point at which a discursive formation is transformed (or transforms itself). Thus, we can speak of the threshold of emergence or of disappearance for a given discourse. In regard to those sub-formations known as sciences, we can identify a series of specific thresholds: positivity, epistemologization, scientificity, and formalization (see section fourteen). Foucault notes that archeological analysis describes the transformations of scientific discourse primarily in terms of the threshold of epistemologization (that is, at the level at which a discursive positivity takes the position of knowledge). Crucially, thresholds are not absolutely tied down to chronology; a threshold is not necessarily a single point in time. Neither is the threshold at which a discourse changes necessarily the threshold for the transformation of its statements, objects, concepts, strategies, or subject- positions. And neither is the series of thresholds for scientific discourses a regular one: thresholds may occur out of sequence, or all at once, and some may not occur at all.
Although the analysis of statements does not take account of anything beyond the level of discursive relations, and though it dismisses any notion of a secret, hidden, or unspoken meaning inherent in articulated language, it has at some point to confront the fact that only some things are said out of a much larger set of things that could be said. Thus, part of the conditions of the emergence of statements involves 'exclusions, limits, or gaps' that define what cannot be said (or cannot be said explicitly). It is crucial, however, to recognize that archeology does not recognize the unsaid as a set of 'meanings concealed in what is formulated.' Archeology simply describes the conditions of the emergence of statements, including those conditions that exclude other possible articulations. In this sense, the factors that demarcate the said from the unsaid are simply the factors that make the said possible.