ABSTRACT
We can now reverse the procedure; we can go downstream, and, once we have covered the domain of discursive formations and statements, once we have outlined their general theory, we can proceed to possible domains of application. We can examine what use is served by this analysis that I have rather solemnly called ‘archaeology’. Indeed, we must: for, to be frank, as they are at the moment, things are rather disturbing. I set out with a relatively simple problem: the division of discourse into great unities that were not those of œuvres, authors, books, or themes. And with the sole purpose of establishing them, I have set about constructing a whole series of notions (discursive formations, positivity, archive), I have defined a domain (statements, the enunciative field, discursive practices), I have tried to reveal the specificity of a method that is neither formalizing nor interpretative; in short, I have appealed to a whole apparatus, whose sheer weight and, no doubt, somewhat bizarre machinery are a source of embarrassment. For two or three reasons: there exist already enough methods for describing and analysing language (langage) for it not to be presumptuous to wish to add another. And, anyway, I was suspicious of such unities of discourse as the ‘book’ and the ‘œuvre’ because I suspected152them of not being as immediate and self-evident as they appeared: is it reasonable to replace them by unities that one has established with so much effort, after so much groping, and in accordance with principles so obscure that it has taken hundreds of pages to elucidate them? And are the things that all these instruments finally delimit, those ‘discourses’ whose identity they map out, the same as those figures (called ‘psychiatry’, or ‘political economy’, or ‘Natural History’) for which I empirically set out, and which have provided me with a pretext for developing this strange arsenal? It is now of the utmost importance that I should measure the descriptive efficacy of the notions that I have tried to define. I must discover whether the machine works, and what it can produce. What, then, can this ‘archaeology’ offer that other descriptions are unable to provide? What are the rewards for such a heavy enterprise?
