Criticism of Foucault\'s Epistemology

Chanhyu Lee
Pf. Flynn
Phil 541
Dec 18 2015

Foucault's Epistemology

1. Introduction

In this paper, I attempt to analyze Foucault's epistemological framework in his book, The Order of Things. Although Foucault's known for his associating knowledge with power, I want to primarily focus on epistemology, and not politics or ethics. While this may be the very attitude that he may be critical of, I believe that not 'all' knowledge is reducible to power. Foucault might have agreed that not all knowledge is reducible to power because his methodological approach was archaeological in nature, meaning that he tried to find the conditions of knowledge of particular ages, i.e. the Renaissance, the Classical, and the Modern, and thus showing 'what knowledge truly is'. It may be the case that archaeological knowledge is a different kind of knowledge from that of a knowledge of a particular era because it is not normative, but rather showing 'how' norms are constituted, and thus being meta-epistemological. This can be affirmed by his notion of 'episteme'. He defines episteme this way: if in any given culture and at any given moment, there is always only one episteme that defines the conditions of possibility of all knowledge, whether expressed in a theory or silently invested in a practice. (Foucault 168)
While it is clear that he provides an archaeological account on the conditions of knowledge, it is difficult to say if Foucault has his own version of epistemology because he shies away from providing a prescriptive definition of it. However, I am interested in 'what knowledge is justified' rather than the conditions that make it possible. The conditions of possibility of knowledge are irrelevant on the validity of knowledge; otherwise, the argument would fall into genetic fallacy. I believe Foucault has his own idea of what knowledge is truthful. I interpret him not to be a radical cultural relativist, based on his critical stance on the modern episteme. His value judgment on a particular episteme tells me that he has a standard that is apart from epistemes that make it possible to compare them and to put high value on one. His criticism of representation and his understanding of language can provide materials necessary to understand what his view on knowledge is. In this paper, I will argue that Foucault can be interpreted as an anti-realist, meaning we cannot know the reality of the world outside language, based on his discourse on representation and language. I also argue that his position as an anti-realist is at odds with his archaeological methodology that underlines extra linguistic conditions that make knowledge possible. Finally, I will suggest possible ways to resolve the incompatibility.

2. Limits of Representation

What is representation? Representation is the episteme of the Classical age. It is a way to see things in which ideas correspond to the representation of the object of thought. While the Renaissance sought knowledge in the resemblance of objects, the Classical places identity between abstract thought and being. For example, Descartes' famous Cogito shows representational thought in which "I think" corresponds to "I exist". The problem of representation, according to Foucault, is that it lacks foundation: "The power of representation to connect must instead be sought "outside representation, beyond its immediate visibility, in a sort of behind-the-scenes world even deeper and more dense than representation itself." (239) What this means is that there is no connection between thought and being. How could we know that they are related? It is impossible to know that there is necessary relation between the sign and the signified. The only reason that makes one to believe their relation is the sign's self-referentiality. Foucault notes, "From the Classical age, the sign is the representativity of the representation in so far as it is representable." (65) To avoid infinite regress, the sign in the representational system is self-refenrential, which does not guarantee the truth of its representativity. This problem of representation can be broadened to the modern debate on external reality. Can we know if something exists outside the mind? A realist would say yes that there is a real world that exists outside the mind. In this framework, representation necessarily plays a part because a realist must say he knows the existence of the external reality. The realist is claiming that his knowledge corresponds to the being of the world. An anti-realist may propose a counter argument that it is impossible to compare 'the real world' and 'its representation'. Furthermore, he would claim that 'the real world' is already understood in the context of our language use, and therefore the dichotomy between the 'real' and the 'represented' is false. Is Foucault an anti-realist? I think so, based on his criticism on representation. His idea on the being of language can make this issue of whether he is an anti-realist even clearer.

2. Language

For Foucault, language does not necessarily serves a function of representation. There are hints in his writing that suggest the essence of language is not in denoting other things, but rather is in its own, independent being.

It breaks with the whole definition of genres as forms adapted to an order of representations and becomes merely a manifestation of a language which has no other law that that of affirming – in opposition to all other forms of discourse – its own precipitous existence; and so there is nothing for it to do but to curve backing a perpetual return upon itself, as if its discourse could have no other content than the expression of its own form. (300)

He also writes, "literature is the contestation of philology (representation): it leads language back from grammar to the naked power of speech, and there it encounters the untamed, imperious being of words." (300) Here, we can clearly see that Foucault views literature as the original and the truthful function of language. Language leads back itself to literature when its original function is contaminated with other pseudo-functions. Foucault goes on to say that "Western culture, the being of man and the being of language have never, at any time, been able to coexist and to articulate themselves one upon the other. Their incompatibility has been one of the fundamental features of our thought. (339) With reservations, I believe his position can be entitled as "linguistic idealism". Would he say that it is possible to go beyond language and reach to truth outside it? I am inclined to say no because that would entail representation. He would at least agree that it is impossible to describe such truth because all descriptions are bound to language, and therefore they would be false or meaningless. This denial of extra-linguistic reality is what constitutes of linguistic idealism.

4. Archaeology

Now, I must turn toward Foucault's archaeological method of investigation in The Order of Things. I do so to point out few problems that may be contradictory to what he has stated in the book. It is crucial to understand what he means by archaeology because his arguments are surrounded by it. Foucault defines his archaeological methodology, "the archaeological level of investigation is the level of thinking in a project which is concerned with what made something possible." (31) Therefore, what he is doing is meta-epistemological in nature. One must note that Foucaultian archaeology is not normative. He is not providing us with a rule that governs conditions of knowledge. Rather, what he is doing is descriptive. He admits that his methodology is a positivistic method, "modalities of order have been recognized, posited, linked with space and time, in order to create the positive basis of knowledge as we find it employed in grammar and philology, in natural history and biology [and] in the study of wealth and political economy. (Foucault xxi) It is also noteworthy that his notion of episteme is also named as 'historical a priori'. The 'a priori' comes from that there are unconscious conditions that makes possible to think in a certain way, similar to transcendental philosophy of Kant. However, Foucault claims that these conditions are not themselves transcendental but are affected by historical, socio-economic events. Therefore, he excludes the 'agent' from the subject of knowledge because it is these external factors that make one possible to think.
However, there is a problem with positivism, that it is also a product of Classical representation of which he is critical. Gary Gutting explains the problem of positivism:

Positivism may maintain that in fact this is all our thought is, that there is nothing beyond the phenomenal realm. But it does so in a context that, in contrast to the Classical age, at least follows the possibility of other construals of thought. Of course, the positivist can try to exclude these other construals (as, for example, meaningless). But how can he do this except by methods (for example, the appeal to a principle of verification) that themselves go beyond the resources of thought limited to the phenomenal realm? (Gutting 185)

Positivism cannot be justified because while Classical representation has at least an axis that is arbitrary and self-referential, it does not even have an axis of thought. Therefore, just as the principle of verification is itself unjustified, positivism is groundless while proclaiming to possess truth. I think Foucault's own archaeological method can be under the same criticism. His claim that his method is purely empirical is not grounded because he does not provide us with a transcendental standard that can give it a foundation. He turned normative knowledge with empirical bases, but these too must be grounded. As long as he is trying to provide us with an "accurate" picture on the nature of knowledge, it is representational because representation is an order of identity and difference. One can say that he is not a positivist because he is not a transcendental thinker, but a historical one. This would mean that his own archaeology would fall into the being of language that is incompatible with the being of man. Then, it would also be problematic for him to claim that his archaeology provides an accurate picture on the reality of knowledge.

5. Transcendental Philosophy

Is there an alternative way to read Foucault's work that may avoid this problem of positivism? Beatrice Han suggests that this is a problem that Foucault was aware of. Han suggests an explanation; "Foucault makes the leap of understanding the historical a priori in explicitly political terms, through his development of the concept of 'power-knowledge', where power and knowledge are seen as inseparably intertwined." (Han 132) Kelly notes that, "for Han, this is response to the failure to find a plausible historical a priori at the level of discourse itself-examining knowledge endogenously will never reveal what makes something count as true at a particular historical conjecture." (Kelly 95) When, Han responds to Gary Gutting's criticism that Foucault's position cannot be read as transcendental, she notes that Foucault is under framework of transcendental philosophy, "Therefore Foucault ends up in the difficult position of claiming for the historical a priori an efficacy which is excluded by archaeology's very theoretical premises, hence the "strange notion of regularities which regulate themselves", which, as I suggested in my book, repeats the structure of the empirico-transcendental double and shows Foucault's inability." (Han 6) If the conditions of possibility of knowledge are replaced with power, then the conditions do not become empirical, but rather of agency. This may have advantage over archaeological method, but I think it pushes him into a position that he was critical of: anthropology. It bears a problem because power is not of natural origin, but has to be transcendental to provide knowledge a priori conditions. However, it would be antithetical to his criticism of "Man" because power clearly has axis in man. Therefore, Foucault is in a difficult position that if he wants to keep conditions of possibility of knowledge as variable historical factors, he must abandon agency.

6. Man and His Doubles

Is Foucault's position self-refuting that he falls under "the structure of the empirico-transcendental doublet"? This is a question that I also had when reading him. While he is critical of the modern episteme that grounds the conditions of possibility of knowledge in labor, life, and language, he also does the same thing because his archaeology places extra-linguistic conditions that regulate knowledge. Let us turn to his critique of man and his doublet. It is clear that Foucault is critical of philosophy that is devoid of transcendental ground, when he says, "How it (positivism and Marxism) has attempted to exorcise the naïve discourse of a truth reduced wholly to the empirical, and the prophetic discourse which with similar naivete promises at last the eventual attainment by man of experience." (321) This could be interpreted that they need to have a transcendental foundation that can give them grounds. This may be why Foucault maintains the notion of historical a priori. However, he does not want to give it a universal status. In other words, he does not want to be outside of history. This may be his strong endorsement of nominalism and the being of language. But, there is also the problem that he cannot provide a priori conditions within un-universal status. Therefore, I believe his position falls into empirico-transcendental doublet.
7. Possible Solutions
Is there a possible solution that can evade the problem of empirico-transcendental doublet? I believe so. If we turn to the more traditional philosophy, there is a way beyond the paradox. It would be senseless to deny all the empiricities and to maintain transcendental a priori only. In other words, any kind of idealism is susceptible to modern critique of representation that it needs something beyond it. However, relying on empirical science that denies any true transcendental conditions of knowledge would be false because it becomes self-refuting that empirical science needs an axis that must ground it, and the axis cannot be man that exists in nature. The axis, or in other words, the normative foundation, must exist in the metaphysical realm. I am calling for the restoration of metaphysical realism that there must exist a normative realm outside the natural realm of phenomena. Only this way can empirical sciences be grounded but not in man. Man becomes the knower of the normative realm that is intuitively grasped. Therefore, man does not become a center of empirico-transcendental doublet that the transcendental a priori is not a condition in man, as in Kant, but rather something external. For example, free will, which itself is a metaphysical notion that it evades deterministic chain of the universe, would place man above empiricities because it enables man to become the knower, but it too has conditions that are external to nature, that free will cannot be the product of nature. Then, man's status as knowing subject and known object can be coherent.





Bibliography

Foucault, Michel. The Order of Things: An Archaeology of the Human Sciences. New York: Pantheon, 1971. Print.

Gutting, Gary. Michel Foucault's Archaeology of Scientific Reason. Cambridge [England: Cambridge UP, 1989. Print.

Han, Beatrice. Foucault's Critical Project: Between the Transcendental and the Historical. Stanford, Calif.: Stanford UP, 2002. Print.

Han, Beatrice. "Reply to Gary Gutting's Review of Foucault's Critical Project: Between the Transcendental and the Historical." Http://privatewww.essex.ac.uk/~beatrice/Gutting _answer_ 2003-05.pdf. University of Essex. Web. 19 Dec. 2015.

Kelly, Mark. "Béatrice Han, Foucault's Critical Project, Trans. Edward Pile (Stanford, CA: Stanford University Press, 2002). ISBN 0-8047-3709." Foucault Studies 1 (2004): 92-97. Print.










1