Blog post

Foucault's Immanent Contradictions

From Habermas to Honneth, critics have been keen to portray Foucault as a paradox-prone thinker. Thomas Lemke argues that we should embrace the recurring contradictions in Foucault's thought as symptoms rather than inherent problems.

Thomas Lemke22 February 2019

Foucault's Immanent Contradictions

This is an excerpt from the introduction to Thomas Lemke's Foucault's Analysis of Modern Governmentality: A Critique of Political Reason.​​​​​ 

Were not Foucault’s critics right to fault him for the contradictions immanent in his work? Did they not accurately describe the theoretical incoherence of calling for political resistance on the basis of a neutral conception of power? Was it not necessary to dissolve these aporias, contradictions and paradoxes in one direction or another? It seems Foucault had only two possibilities. According to the first line of reasoning, he overcame the problem and affirmed the validity of his neutral conception of power by giving up on critical ambitions: he came to advocate theoretical relativism, no longer seeking to distinguish between better or worse, greater or lesser freedom, or more or less just forms of power. Alternatively, Foucault made his political motives and normative value judgements clear – that is, he gave up on the neutrality he had professed – so that the critical standards of his theoretical engagement became manifest and available for political mobilization. Either-or.

When I first began working on Foucault, I accepted that his analysis of power lacked (self-) reflection. Yet at the same time, I recognized that many reasons existed to call such a diagnosis into question. My engagement with Foucault’s ‘paradoxes’ followed a course that proved paradoxical in its own right: the more obvious and manifest the ‘problem’ became, the more I asked myself whether it really posed a problem. There were at least three reasons for my mounting scepticism.

To start with, it is highly improbable that such manifest contradiction escaped Foucault, a subtle thinker attuned to the political consequences of theory. On the contrary – as Habermas has noted (among others) – Foucault was well aware of the paradoxes, even if he never changed his position with regard to them. The question, then, was why he did not abandon his contradictory outlook in order to resolve the problem in one way or another. Can this bearing really be reduced to ‘professing irrationalism’,[1]or did Foucault have something else in mind – and if so, what? If we accept that Foucault consciously operated with contradictions, what did he want to achieve?

A second reason, connected to the first, involves the kind of critique levelled at Foucault. On the one hand, I felt that contradictions in his work were identified correctly and accurately, on the other hand, I was left with the impression that seeking them out proves unproductive and negative; doing so follows a rationalistic strategy focused on insufficient theoretical reflection and intellectual misprision. Critics intone lamentations about a lack of self-reflection, then sound a call for a coherent theoretical position to resolve the problem. From this perspective, ‘disquieting contradictions’ represent the product of ‘inconsistency’[2] or, alternately, ‘the result of Foucault’s deficient reflection on the normative conditions of his own writings’[3] – that is, a mistake or shortcoming of his theoretical work.

The third aspect concerns the apparent randomness of the political positions that Foucault adopted. Even though critics propose similar diagnoses, the assessment of his overall politics does not present a clear picture. Foucault has been labelled a ‘young conservative’;[4] for others, he represents nihilism or anarchism.[5] Axel Honneth views Foucault as standing close to the positivism of systems theory; in contrast, Mark Poster holds that he continued the tradition of Western Marxism ‘by other means’.[6] Notwithstanding comparable accounts of problems, then, readers have ironically enough not arrived at a uniform classification of Foucault’s writings. Quite the opposite. Their status remains unclear – or, better: contradictory.

[book-strip index="1" style="buy"]

In this light, the ‘problem’ of Foucault shifted, and critique fell back on his critics. Revisiting their interpretations, I found a host of misunderstandings, misreadings and prejudices. A significant portion of secondary literature consists of longwinded repetitions and commonplaces, imputations made solely for polemical reasons, unproductive modes of engagement, and countless oversimplifications and caricatures. A core problem of much criticism is that it operates with a conception of power from which Foucault explicitly sought to distance himself; readers attack him on the basis of what he revealed to be problematic and wished to leave behind.[7] These critics work with a negative conception focused on constraint, repression, domination and so on. Yet Foucault wished to demonstrate the political and historical limitations of this very model. As such, it is not Foucault’s works so much as those of his critics that prove singularly self-referential and sterile: his putative lack of self-reflection means, in fact, that he did not share their basic theoretical assumptions. For this reason, I switched to ‘defending’, ‘justifying’ and ‘legitimating’ Foucault’s theoretical strategy – and looked for problems elsewhere.

Such an approach ultimately proved unsatisfying, too, for it missed the novel and interesting aspects of Foucault’s work. Switching sides did nothing to change the ‘problematic’ as a whole. Two reasons made this conclusion inevitable. First, it was doubtful what theoretical advantage the position offered: did not my line of argument prove just as ‘rationalistic’ as that of Foucault’s critics inasmuch as it situated problems only in the realm of theories and ideas? Was I not doing the same thing as the critics I wanted to refute when I tried to show that Foucault had not, in fact, become entangled in contradictions – that they ‘actually’ resulted from how others interpreted him? In seeking to disprove critics, I implicitly accepted their premises – that is, (a) the idea that theory must be free of contradiction; (b) that Foucault’s apparent contradictions held only in terms of thought; and (c) that these contradictions discredited any and all political praxis.

The matter opened on to a second problem, perhaps of even greater magnitude. I had sought to promote a ‘true’, ‘correct’ and ‘accurate’ interpretation against hasty and distorted readings. In so doing, I failed to see how I was playing along in the same ‘game of truth’ that Fraser, Habermas and others had already rehearsed – that I was acknowledging they were right by accepting their rules and seeking to refute them. Precisely inasmuch as I had been looking for the one and only ‘truth’ – the ‘real Foucault’ – I could not do justice to what was actually at stake. To be sure, it was right – and important – to work against inadequate interpretations, and there could be no doubt that misreadings of Foucault’s works abound. Still, it was not enough just to defend the ‘real Foucault’ against misguided interpreters. Such a strategy of argument presents two key disadvantages: on the one hand, it focuses on ‘securing’ a certain theoretical terrain – terms and concepts – and, on the other, it continues to move within the same critical horizon it means to escape. In my effort to remain true to Foucault’s intuitions by defending them against critics, I had failed to see that this kind of piety is ‘the most touching of treasons’.[8] The path did not lead forward, either: it simply turned the problem upside down.

I had to start again. I had overlooked the most obvious and concrete difficulty: a real problem that I had wanted to eliminate altogether without treating it properly. I had set about seeking out hidden truths and obscure points of reference without taking the real difficulty seriously, searching for what others had overlooked or disregarded – mistakes and oversights – instead of working with what they had found. My point of departure needed to change. Instead of viewing the difficulties I encountered (which others had met with, too) in terms of an intellectual shortcoming, insufficient familiarity with the texts or temporary incomprehension – as obstacles that would vanish in the course of reaching a ‘correct’ understanding – I needed to make problems and points of trouble the starting point and vector of my own work. After all, Foucault presents difficulties for others, too. As such, they possess something like an ‘objective’ status: if so many authors with different theoretical priorities and political orientations agree on the matter, there really is something in Foucault’s writings that provokes or produces problems.[9]

It became clearer and clearer to me that aporias, contradictions and paradoxes do not represent a defect or mistake in Foucault’s work – something to be explained away or resolved. On the contrary, they constitute its theoretical significance and define its ‘problematic’.

Perhaps the question whether Foucault’s critics are right or not is a matter of secondary interest – or simply idle. Whether justified or not, criticism exists. It amounts to a (discursive) fact, and the existence, or positivity, of this fact is where I had to start. Surely, I was ‘on the right side’: misunderstandings and misreadings of Foucault are legion. However, the point is not to be, or prove oneself, right. Nor is the decisive and final question whether Foucault ‘really’ contradicts himself or not – or whether interpreters have read him ‘correctly’ or not. The key item is the empirical reality that his works have been perceived as contradictory.

Thus, the difficulty of situating Foucault politically does not follow from inept or faulty assessments of his work; instead, it is both the result and the goal of the same.[10] Foucault’s critics have also read him ‘correctly’. Their objections should not be viewed in terms of truth content so much as ‘symptoms’ – hence the need for a symptomatic reading when approaching them.

Foucault's Analysis of Governmentality is currently 30%-off through our website.


[1] Habermas, The Philosophical Discourse of Modernity (1990), 278.

[2] Taylor, ‘Connolly, Foucault, and Truth’ (1985), 381.

[3] Honneth, ‘Afterword to the Second German Edition’ (1997a), xxvi.

[4] Habermas, ‘Modernity versus Postmodernity’ (1981), 13.

[5] Marti, Michel Foucault (1988), 149; Fink-Eitel, Foucault zur Einführung (1989), 120–2.

[6] Honneth, The Critique of Power (1997b), 195–202; Honneth, ‘Foucault and Adorno: Two Forms of the Critique of Modernity’ (1995b), 131; Poster, Foucault, Marxism and History (1984), 39–40.

[7] This is William E. Connolly’s main objection to Taylor’s interpretation of Foucault: ‘I contend that those such as Taylor who seek to dismiss fundamental features of the project by showing it to be incoherent will find it more difficult to make that charge stick once they are not allowed to precede their critiques of Foucaultian genealogy by a translation of it into the very formulations it seeks to interrogate’. Connolly, ‘Taylor, Foucault, and Otherness’ (1985), 369; see also Rouse, ‘Power/ Knowledge’ (1994), 104–5. For Taylor’s response, see Taylor, ‘Connolly, Foucault, and Truth’ (1985).

[8] Foucault, 1984o, ‘The Art of Telling the Truth’ [1994], 147.

[9] Régis Debray’s observation goes in the same direction: ‘Foucault’s thinking confronts us with the following question: can there be a Foucauldian politics or not?’. Quoted in Kammler, Michel Foucault: Eine kritische Analyse seines Werks (1986), 192. It does not matter that Debray’s response – like that of most of Foucault’s interpreters – proves negative; whether the answer turns out to be positive or negative, the question itself is more important: even Foucault’s most vehement critics must presuppose it and acknowledge that it accompanies his work as a whole.

[10] Foucault himself put the matter in a nutshell in response to an interviewer’s question about his political stance: ‘I think I have in fact been situated in most of the squares on the political checkerboard, one after another and sometimes simultaneously: as anarchist, leftist, ostentatious or disguised Marxist, nihilist, explicit or secret anti-Marxist, technocrat in the service of Gaullism, new liberal, etc. An American professor complained that a crypto-Marxist like me was invited to the U.S.A., and I was denounced by the press in Eastern European countries for being an accomplice of the dissidents. None of these descriptions is important by itself; taken together, on the other hand, they mean something. And I must admit that I rather like what they mean. It’s true that I prefer not to identify myself and that I’m amused by the diversity of the ways I’ve been judged and classified. Something tells me that by now a more or less approximate place should have been found for me, after so many efforts in such various directions; and since I obviously can’t suspect the competence of the people who are getting muddled up in their divergent judgments, since it isn’t possible to challenge their inattention or their prejudices, I have to be convinced that their inability to situate me has something to do with me. And no doubt fundamentally it concerns my way of approaching political questions’. Foucault, 1984g, ‘Polemics, Politics and Problematizations: An Interview with Michel Foucault’, 383–4.

Foucault's Analysis of Modern Governmentality
Lemke offers the most comprehensive and systematic account of Michel Foucault’s work on power and government from 1970 until his death in 1984. He convincingly argues, using material that has only ...