Praxis and Agency in Foucault



 This paper examines the consequences for agency that Foucault’s historiographical approach constructs. The analysis begins by explaining the dif- ference between ‘‘legislative history’’ and ‘‘exemplary history,’’ drawing parallels to similar theoretical distinctions offered in the works of Max Weber, J.L. Austin, and Zygmunt Bauman. The analysis continues by reading Habermas’s critique of Fou- cault through the tropological lenses suggested by White [Metahistory. Baltimore, MD: Johns Hopkins University Press, 1973]; it argues that Habermas’s critique misrecognizes the tropes of Foucaultian genealogy. The paper draws implications for education by articulating possibilities for praxis and agency in terms of pedagogy specifically related to the distinction between didactics and modeling. The paper concludes by suggesting that genealogy does not ‘‘play by Hegel’s rules,’’ but rather exemplifies agency in ways that are not recognizable from a modernist perspective.
KEY WORDS: agency, critical theory, didactics vs. modeling, exemplary history, Foucault, genealogy, Habermas, historiography, legislative history, postmodernism

The Nietzschean tradition does not play the game of critique by Hegel’s rules. David Owen

In order to explain the kind of agency that is constructed through Foucault’s historiography, I will contrast Foucaultian genealogy with modernist historical projects. As an example of this contrast, I refer to the famous critique against Foucault leveled by Habermas in The Philosophical Discourse of Modernity: Twelve Lectures (Habermas, 1987). The Foucault/Habermas debate about what counts as ‘‘critical’’ history has been elaborated on several fronts. In 1994, Michael Kelly’s anthology, Critique and Power: Recasting the Foucault/Habermas Debate compiled the pivotal essays by Foucault and Habermas along with eight significant commentaries. This compilation fleshed out the modern/postmodern debate from multiple perspectives including systems theory, pragmatism, and feminism. I also draw from contemporary US social theory (Con- nolly, 1985; Dean, 1994; Owen, 1995) to argue that Foucault’s

Studies in Philosophy and Education 23: 445–466, 2004.
Ó 2004 Kluwer Academic Publishers. Printed in the Netherlands.

genealogical approach presents us with possibilities for praxis and agency1 that are not available in a modernist discourse. In the field of education, the literature on ‘‘new cultural history’’ has contrib- uted significantly to this debate (see, e.g., Popkewitz et al., 2001; DePaepe and Simon, 1996). My analysis is indebted to this litera- ture that forms the basis on which my critical analysis was conceived.

This paper is organized in two parts whose task it is to examine the consequences for agency that Foucault’s historiography constructs. In the first part, I explain the difference between ‘‘legislative history’’ and ‘‘exemplary history’’ by drawing parallels to similar theoretical distinctions offered in the works of Max Weber, J.L. Austin, and Zygmunt Bauman. Then, in the second part of the paper, I read Habermas’s critique of Foucault through the tropological lenses suggested by White (1973) to argue that Habermas’s critique mis- recognizes the tropes of Foucaultian genealogy.2 Throughout the paper, I suggest implications for education as I seek to articulate possibilities for praxis and agency in terms of pedagogy, specifically the distinction between didactics and modeling. I conclude by sug- gesting that genealogy does not ‘‘play by Hegel’s rules,’’ but rather exemplifies agency in ways that are not recognizable from a mod- ernist perspective.


Habermassian notions of praxis and agency have frequently been used as standards against which Foucault’s historiographical work has been evaluated. Against that position, and following Bernstein (1994) and Connolly (1985), I argue that a Foucaultian construc- tion of agency cannot be understood using the same analytic tools as those deployed by modernist (Habermassian or Hegelian) his- torical analyses. Happily, several social philosophers have already
1 In most current literature, the terms praxis and agency are affiliated with a modern critical theory tradition, deriving from the Frankfurt School. However, in this analysis, I bring those two terms into a genealogical discourse where they take on different meanings.

2 I think that White, himself (perhaps ironically) misrecognizes the tropes of genealogy.


provided elegant theoretical mechanisms by which the rhetoric of modernist histories can be distinguished from the rhetoric of Foucault’s historiography. In this section, I briefly summarize five of these different, but related, theoretical distinctions that can be used to understand the difference between modernist history and Foucault’s genealogy vis-a` -vis the construction of praxis and agency. I begin with Max Weber’s (1961) distinction between emissary and exemplary authority. Then I suggest that J.L. Austin’s (1962) linguistic theory, which contrasts what words say with what words do, is useful for understanding aspects of Foucault’s historiography. Third, I refer to Zygmunt Bauman’s (1987) distinction between legislators and interpreters that applies to dif- ferences in modern and postmodern historiographical projects. Finally, both Mitchell Dean (1994) and David Owen (1995) engage explicitly with Foucault’s project and provide illuminating analyses. Dean distinguishes between synthetic philosophies of history and effective history; Owen explicates the differences between leg- islative and exemplary critique. Taken together, these five theoret- ical approaches help to explicate historiographical effects that Depaepe and Simon (1996) have differentiated as ‘‘mirror’’ versus ‘‘lever.’’

These five sets of distinctions provide a variety of perspectives for understanding how genealogy constructs praxis and agency in ways that are unlike modernist historical constructions of praxis and agency. Throughout, I seek to draw connections to pedagogy, specifically to the discursive distinction between didactics and modeling.

(1) Emissary vs. Exemplary Prophets: Max Weber (1961). In his ‘‘The Social Psychology of the World Religions,’’ Weber wrote pri- marily about the constitution of authority within the realm of orga- nized religions. In the context of religious discourse, Weber analyzed the voice of authority as the voice of the prophet. He explained that the exemplary prophet must embody the teachings, whereas the emissary prophet must proclaim the teachings. ‘‘Exemplary prophecy points out the path to salvation by exemplary living, usually by a contemplative and apathetic–ecstatic life. The emissary type of prophecy addresses its demands to the world in the name of a god’’ (Weber, 1961, p. 1394). The exemplary prophet is an embodiment; the emissary prophet is an instrument. The pedagogy of the exem- plary prophet is modeling; the pedagogy of the emissary prophet is didactics.

Talcott Parsons, in his introduction to Weber’s The Sociology of Religion, referred to Weber’s two types of prophets as exemplary and ethical.3 He wrote:

The exemplary prophet provides a model for a way of life which can be followed by others, embodying in a religious sense what is defined as a higher level of personal virtue. There is, however, no implication that the standards of this pattern or ‘way’ are binding on any social community as such. Then ethical prophet, on the other hand, imposes demands on certain categories of men in such a way that not only do they have an opportunity, but it is rather their duty to follow his precepts. These precepts in turn are defined, not so much to exhort followers to emulate the prophet’s personal example, as they are to exhort them to conform with an impersonally defined normative order (Parsons, 1922, p. xxxv).

Weber’s distinction can be applied to highlight the exemplary aspect of Foucault’s historiography. Foucault’s rhetorical stance is not emissary because it denies both the expectation of dutiful following and the imposition of a normative order. His histories do not con- clude with exhortations to follow this or that course of action; genealogy implicitly rejects the emissary role.

In addition to the assumption of this rhetorical stance, Foucault also made explicit statements denying the emissary role of the intel- lectual. Using military metaphors, Foucault asserted that it is not the role of the historian to advise about ‘‘the project, tactics and goals’’ of the battle. This stance acknowledges a place for intellectual work in the political activism, but it rejects the traditional role of intel- lectual as advisor or teacher in relation to other people.

The intellectual no longer has to play the role of an advisor. The project, tactics and goals to be adopted are a matter for those who do the fighting. What the intellectual can do is to provide instruments of analysis, and at present this is the historian’s essential role. What’s effectively needed is a ramified, penetrative perception of the present, one that makes it possible to locate lines of weakness, strong points, posi- tions where the instances of power have secured and implanted themselves by a system of organization dating back over 150 years. In other words, a topological and geological survey of the battlefield – that is the intellectual’s role. But as for saying, ‘Here is what you must do!’, certainly not. (Foucault, 1980a, p. 62).

The absence of certain didactic pronouncements in Foucault’s his- toriography has prompted some modernist critics to claim that genealogy does not teach anything, and its critical stance does not make an intellectual contribution to the activist cause. However, as

3 I could find no explanation for why Parsons changed Weber’s terminology from ‘‘emissary’’ to ‘‘ethical.’’


Weber’s distinction helps to clarify, the contribution of genealogy is one of exemplary modeling, not one of emissary didactics.

One example of modeling from Foucault’s work is his iconoclastic treatment of power. For decades, critical history had been colonized by Marxian, Freudian, and Frankfurt School notions of power. However, both implicitly and explicitly, Foucault argued extensively against dialectics and ‘‘the repressive hypothesis’’:
If power were never anything but repressive, if it never did anything but to say no, do you really think one would be brought to obey it? What makes power hold good, what makes it accepted, is simply the fact that it doesn’t only weigh on us as a force that says no; it also traverses and produces things, it induces pleasure, forms knowledge, produces discourse. It needs to be considered as a productive network that runs through the whole social body... (Foucault, 2000, p. 120).

When he casts the notion of power as a productive mechanism that circulates in discourse, Foucault undermines the modernist legislation that power belongs to certain people or groups of people. Foucault’s genealogy speaks of power in a way that is not ‘‘politically correct.’’ Bernstein (1994) calls genealogy ‘‘the rhetoric of disruption’’ (p. 222) because its effect is iconoclastic.

Another example of exemplary critique is Foucault’s Discipline and Punish. This history of punishment disrupts modernist assump- tions when it demonstrates that modern penal systems are at least as cruel and barbaric as medieval torture was. In doing so, genealogy models an audacious kind of thinking that flies in the face of received wisdom and calls into question the will to truth. Bernstein (1994) describes this disruption most effectively:

When Foucault begins Discipline and Punish with the detailed, graphic execution of Damiens... it is a stunning rhetorical device for eliciting conflicting disruptive reactions in the reader. For Foucault knows the reader will react with a sense of the horror to what initially appears to be the barbaric spectacle of gratuitous torture. We are seduced in taking comfort in the realization that ‘our’ methods of punishment, whatever their effects, are much more humane.... It is only gradually that our confidence begins to be undermined as we see what the process of ‘humanization’ involves. For we come to see how ‘the birth of the prison’ is virtually an allegory for the birth of the disciplinary society, the panoptic society of surveillance that makes such effective use of the disciplines that control our bodies (Bernstein, 1994, pp. 224– 225).

Bernstein’s description highlights the effectiveness of Foucault’s his- torical account, not the accuracy of the representation. Bernstein calls our attention not to the formal aspects of argument, nor to the


orthodoxy or heterodoxy of this position relative to an ideological stand. Rather, Bernstein’s account captures the exemplary qualities of Foucault’s historiography.
(2) Stative vs. performative utterances: J.L. Austin (1962). In the first 4 of the 12 William James Lectures he delivered at Harvard in 1955, Austin laid out his famous distinction between performative utterances and statements (or ‘‘constatives’’).4 He said, ‘‘the issuing of a [performative] utterance is the performing of an action – it is not normally thought of as just saying something’’ (Austin, 1962, pp. 6–7). Austin gave the following classic examples of performative utter- ances: ‘‘I take this woman to be my lawful wedded wife,’’ ‘‘I name this ship the Queen Elizabeth,’’ and ‘‘I give and bequeath my watch to my brother.’’ The performative utterance, then, is distinguished from the statement in that the performative does something rather than just say something.

Austin articulated his linguistic theory using sentences as exam- ples, and as such, the theory does not map directly onto an analysis of a historiographical discourse. Perhaps Austin’s theory is of limited use here; nevertheless, his characterization of the performative utterance offers some valuable points of clarification. In this section, I suggest that the primary role of Foucault’s genealogy is to perform a critical intervention into the present, so genealogy behaves more like a performative utterance than a statement or report.

Foucault has been criticized by some historians for documenting his historical claims with ‘‘minor texts’’ or obscure artifacts. Other historians (including some feminists) commend Foucault’s use of little-known documents because by doing so, he calls attention to hitherto marginalized voices (see, e.g. Goldstein, 1994). However, both critics and advocates base their commentaries on assumptions about artifacts as objectively representative data in the historical analysis. Therefore, both sides misrecognize the performative role of genealogy by evaluating genealogy as if it were a statement or rep- resentation of some historical ‘‘truth.’’ These commentaries miss the point of genealogy as history of the present. Representation is not the goal of genealogy; political effectiveness is the goal of genealogy. The targets of genealogical critique are those intractable assumptions in the present that serve to sort people unjustly. Foucault’s histories of

4 Later, beginning in Lecture VI, Austin problematizes that clear distinction and elaborates a more nuanced analysis of speech acts. For the moment, however, in this paper, the earlier rough distinction will suffice.


the insane asylum, the prison, and sexuality are all exemplary polit- ical studies that undermine the naturalness of our assumptions about psychiatry, juridical punishment, and the essential division of the sexes. The purpose of these genealogies is to undermine our current assumptions about what is ‘‘normal.’’
One salient point here is the question of evaluation: What makes one historical account better than another? Modernist historio- graphy – Metahistory notwithstanding – has largely been judged on accuracy; evaluation is based on the completeness of the data sources and the appropriateness of the analytic claims. Using Austin’s dis- tinction between statement and performative, we see that a statement can be evaluated as being either true or false based on whether the statement corresponds with some reality; however, a performative cannot be evaluated as being either true or false. Performatives can be evaluated only on whether they ‘‘go right.’’5 We can take this to mean that performative utterances are to be judged on their effec- tiveness. That is, if someone says, ‘‘I bet it will rain,’’ but refuses to pay off a losing bet, then the performative utterance can be judged ineffective.

An analogous distinction can be made for the evaluation of Foucault’s genealogical approach. Genealogy does not make emis- sary-like statements, therefore, it is not – or should not be – a question of whether Foucault’s historical claims are true or false, except to the extent that glaring falsehoods or inadequate docu- mentation diminish the rhetorical effectiveness of the argument. Genealogy should be judged as a performative. Genealogy should be considered good insofar as it effectively disrupts status quo assumptions about the construction of the subject.

This performative mode does not ignore rigorous scholarship. If the scholarship is trivial, then the genealogy cannot be effective be- cause the work will be dismissed by serious scholars. Performativity is a question of whether the genealogical evidence disrupts old habits of thought that have been limiting what it is possible to think, and that can be effective only to the extent that the argument is relevant in the history of the present. Genealogy does not pretend to be ‘‘scientific’’; it purports to be political. Therefore it does not make sense to

5 For a performative that goes right – a promise that is upheld or a bequeathal that is executed – Austin uses the curious terms ‘‘happy’’ or ‘‘felicitous.’’ Performatives that go wrong – a bet that is not honored or a vow that is broken – are called ‘‘infelicities.’’

evaluate genealogy exclusively on bases of truth and falsehood or analytic coherence.

Connolly (1985) describes the work of genealogy as incitement. Genealogists deploy ‘‘rhetorical devices to incite the experience of discord or discrepancy between the social construction of the self, truth, and rationality’’ (p. 368). Regarding the evaluation of genealogy, Connolly emphasizes that genealogy is not about truth claims:

The rhetorical figures, to use a phrase of Nietzsche’s, incite us to listen to a different claim rather than to accept the findings of an argument; and they proceed in this way because genealogy of the will to truth cannot present itself as a set of truth claims. Genealogy is not a claim to truth (although it functions in an episteme in which established theories of truth are called into question); it exercises a claim upon the self that unsettles the urge to give hegemony to the will to truth (Connolly, 1985, p. 368).

The performative role of genealogy corresponds roughly to Weber’s notion of the exemplary prophet. The performative/exem- plary stance reflects a pedagogical enactment of modeling rather than didactics. The role of the intellectual is to embody a critical attitude, rather than to exhort others to become critical or to legislate a program of critique. Performative critique is itself an act of freedom. At the same time, its rhetorical stance does not demand a following, so the sort of authority embodied in a performative stance is unlike the emissary or stative sort of authority reiterated by a didactic stance.

Woolgar (1986) further reinforces this distinction. In an essay on praxis and discourse, Woolgar characterizes the stative speech act as characteristic of Anglo-Saxon empiricism. He describes Anglo-Saxon empiricism as a discourse in which (1) ‘‘Objects exist independent of the observer’’; (2) There is a distinction between ‘‘what scientists do’’ and ‘‘what people say scientists do’’; and (3) Praxis and discourse are two separate things (p. 310). According to Woolgar, French (Post)Structuralism is distinct from Anglo-Saxon empiricism in the following ways: (1) Foucault’s discourse is ‘‘a whole concatenation of activities, events, circumstances and objects which together make up a particular world-view’’ (p. 312); (2) ‘‘Reality is constituted in and through discourse’’ (p. 312); and (3) Foucault’s view does not pro- hibit us from making the very practical distinction between a thing and what is said about that thing (p. 314). Woolgar’s analysis pro- vides further grounds upon which to draw a distinction between


stative (praxis-is-separate-from-discourse; action-is-different-from- language) and performative (praxis-is-inseparable-from-discourse) theories.
When history is regarded in terms of its performative dimensions, its agentive properties cannot be evaluated according to modern legislative standards. Rather, genealogy shifts agency and praxis from the stative to the performative.
(3) Legislative vs. interpretive authority: Bauman (1987). In his book entitled Legislators and Interpreters, Bauman distinguishes modern intellectual work from postmodern intellectual work along lines that are parallel to those of Weber’s distinction between emis- sary and exemplary prophets. In his description of the legislator role, Bauman also shows how the legislator (like the emissary) appeals to an objective ontology. The legislator/emissary serves as an instru- ment for another ‘‘objective’’ authority. Furthermore, for the legislator, objective knowledge is not available to everyone; rather, objective knowledge is accessible to a select few who have mas- tered the procedural rules of validity and good taste. In Bauman’s words:

The typically modern strategy of intellectual work is one best characterized by the metaphor of the ‘legislator’ role. It consists of making authoritative statements which arbitrate in controversies of opinions and which select those opinions which, having been selected, become correct and binding. The authority to arbitrate is in this case legitimized by superior (objective) knowledge to which intellectuals have a better access than the non-intellectual part of society. Access to such knowledge is better thanks to procedural rules which assure the attainment of truth, the arrival at valid moral judgement, and the selection of proper artistic taste (Bauman, 1987, pp. 4–5).

In contrast to the legislator, Bauman characterizes the interpreter role as a postmodern one in which the major task is translation to facilitate communication.
The typically post-modern strategy of intellectual work is one best characterized by the metaphor of the ‘interpreter’ role. It consists of translating statements, made within one communally based tradition, so that they can be understood within the system of knowledge based on another tradition. Instead of being orientated towards selecting the best social order, this strategy is aimed at facilitating communication between autonomous (sovereign) participants. It is concerned with preventing the distortion of meaning in the process of communication. For this purpose, it pro- motes the need to penetrate deeply the alien system of knowledge from which the translation is to be made..., and the need to maintain the delicate balance between the two conversing traditions necessary for the message to be both undistorted


(regarding the meaning invested by the sender) and understood (by the recipient) (Bauman, 1987, p. 5).
Bauman’s contribution explains the differences in the construction of authority between legislators and interpreters. In a legislative (emissary/stative) mode, knowledge originates from an objective place. The legislator acts as a mouthpiece for objective knowledge, and the legislator accrues legitimacy by mastering procedures of investigation (e.g., research methods). In contrast, the interpreter acts as a conduit for communication.6 Bauman’s interpreter is not the same as Weber’s exemplary prophet; however, the distinction between legislator and interpreter provides another dimension from which to understand genealogy as a particular kind of non-didactic praxis.

Foucault’s historiography makes no pretense to appeal to any sort of objective ontology, so it is not legislative. Rather, history is always interpretive (although not interpretive in the hermeneutical sense of individual readings). Recognizing that all ontological assumptions are products of particular historical circumstances and discursive interpretation, Foucault abandons all hope for objectivity in a ra- tionalist sense. All histories are produced by the confluence of power relations in which they are written; and as such, no histories can avoid a certain kind of presentism. At the same time, the reiteration of historical accounts contributes to the constellations of constraints and possibilities in the present – shaping what it is (im)possible to think. Since objectivity is impossible, and history is a political tool, genealogy is most effective when it takes the liberty of strategic interpretation, and mobilizing the political dimensions of historical writing.

(4) Progressive vs. effective history: Dean (1994). Many modern- ist, especially Marxian, histories begin with the assumption that some groups of people have power and other groups do not have power; then those histories conclude their analyses by affirming that the same people who had power still have power, and the same people who did not have power still do not have power. That sort of history is leg- islative insofar as the historical narrative itself serves as an instrument that teaches about the ‘‘objective’’ realities of power in the world.

6 At the end of the book, Bauman further problematizes the authoritative role of interpretive philosophers. He argues that, in postmodern times, interpretive philos- ophers have also sought to recuperate their legislative authority by establishing procedures for effective communication.

This form of history intends to raise people’s consciousness by informing them of systemic oppression.

In contrast to that legislative approach, Dean (1994) describes the problem addressed by ‘‘critical and effective histories.’’7 He argues that historians have a responsibility to think in new ways. Critical modernist and critical postmodernist historians agree that injustices exist in the world; however, they disagree rather dramatically about the methods of redress. In the spirit of an emissary prophet or leg- islator, the modernist historian acts as a channel for objective knowledge about the world, as if to say ‘‘The truth shall make you free.’’ In contrast, effective history, like exemplary prophets and performative utterances, enacts new ways of thinking.

The problem is how to form oneself, not in the absence, but in the presence of a plurality of codes, and with a multiplicity of means.... As Deleuze showed, the self is folding of the exterior relation, of the governance of others, back on oneself.... It is the folding back of the Outside to form an Inside, a tissue of foldings that creates an Inside deeper and more remote than any notion of consciousness, identity or even the unconscious can capture. After Foucault, it is our responsibility to ourselves to invent new ways of doing this folding, as much as it is our responsibility to invent new ways of knowing and new forms of political participation. These are the things to which Foucault’s thought, his critical and effective histories, can connect us (Dean, 1994, p. 216).

The problematic of critical work in the context of discursive forma- tion is to ‘‘invent new ways’’ of thinking about the relation of the self to the social. To engage in genealogy is to be profoundly suspicious of the workings of power. We are invited to be particularly suspicious of the ways our own thinking has been normalized. Genealogy cautions us to ‘‘trust no one,’’ and that includes the genealogist.

One example of an attempt to invent new ways of thinking comes from the second volume of the History of Sexuality where Foucault reconsiders the conventional wisdom about the relationship of Greek morality to Judeo-Christian morality. The supposition that Foucault attacks is a commonly held one, namely, that classical Greek sexual mores were relatively unrestricted, and Judeo-Christian mores are repressive. As a way of challenging this supposition, Foucault writes the history of sexuality in a way that emphasizes the complicated principles that governed classical Greek sexual practices. He writes, ‘‘[A]s much as we like to credit the Greeks with a great liberty of morals, the representation of sexual acts that they suggest in their

7 Foucault used the term ‘‘effective history’’ after Nietzsche’s Wirkungsgeschichte.

written works – and even in their erotic literature – seems to have been characterized by a good deal of reserve’’ (Foucault, 1990, p. 39). Then he goes on to document the highly structured rules pertaining to classical Greek sexual morality, explaining in meticulous detail that classical Greek sexuality was governed by aphrodisia, chresis, and enkrateia. This historical account disabuses us of the notion that Judeo-Christian sexuality was more repressive or rule-governed that that of the Greeks:

If one wanted to assign an origin to those few great themes that shaped our sexual morality (the idea that pleasure belongs to the dangerous domain of evil, the obli- gation to practice monogamous fidelity, the exclusion of partners of the same sex), not only would it be a mistake to attribute them to that fiction called ‘Judeo-Christian’ morality, it would be a bigger mistake to look behind them for the timeless operation of prohibition, or the permanent form of law. (Foucault, 1990, p. 251).

Insofar as the history of sexuality destabilizes routine assumptions about progressive morality, it is an example of a critical and effective history. After reading the second volume of The History of Sexuality, it is very difficult for us to continue to believe that Greek sexual mores were more libertarian than Judeo-Christian mores. Moreover, it compels us to rethink the inevitability of sexual rules of behavior as they are currently practiced.

In summary, to legislate new ways is praxis to bestow agency; to perform new ways is praxis to assume agency. To be an emissary of knowledge is to defer agency; to take on the responsibility of per- forming new ways of knowing and new forms of political participa- tion is to exemplify agency. The former is characterized by a didactic pedagogy/praxis; the latter is characterized by an exemplary peda- gogy/praxis. For purposes of clarifying Foucault’s historiographical project, I have, to this point, overstated these distinctions. Later, this strict distinction will be problematized in a consideration of legisla- tive performance.

(5) Legislative vs. exemplary critique: Owen (1995). Modernist critical histories are designated as forms of ‘‘legislative’’ critique be- cause the intellectual work dictates the possibilities for agency, autonomy, and transformation. Owen’s (1995) contrastive notion of ‘‘exemplary’’ critique helps to explain why the absence of legislated agency is not the same as the absence of agency:

Genealogy cannot legislate autonomy for us, it recognizes no grounds on which such an act of legislation could be secured, but it can (and does) exemplify its commitment

to the value of autonomy in the form of its reflection on our present (Owen, 1995, p. 492).
As a form of exemplary critique, genealogy is itself an act of transgression. Foucault’s work upsets the Kantian notion of critique. Invoking Kant’s ‘‘What is it possible to think?’’ Foucault rejoins: ‘‘What is it impossible to think?’’ (1970, p. xv). Genealogy is not a reflection but an invention that gestures toward the impossible: ‘‘Undoubtedly, no form of reflection yet developed, no established discourse, can supply its [transgression’s] model, its foundation, or even the riches of its vocabulary’’ (Foucault, 1998, p. 77). Exemplary critique pushes the limits of what it is possible to think in order to point to the horizon of what it is not possible to think.

When it comes to the question of ‘‘new ways,’’ Owen then explains the necessarily perspectival qualities of genealogy. The perspectival account in genealogy arises from a commitment to the historicity of all knowledge. Perspectival is another way of saying ‘‘historically specific.’’ Sharing historiographical commitments with perspectival- ists (e.g., Hayden White), genealogy eschews both the quest for objectivity and the pursuit of universality. In this case, ‘‘historical’’ means there is no knowledge that is true in all times and in all places; there is no ‘‘God’s eye’’ point of view. Rather, all knowledge is dependent on historical context; all knowledge is a product of his- torically specific circumstances. If all knowledge is a product of his- torically specific circumstances, then genealogy is, too. ‘‘Genealogy situates itself as a product of the historical process it is engaged in investigating; its perspectives are immanent to the historical process and acknowledge their partiality’’ (Owen, 1995, p. 502). Genealogy recognizes itself as having been shaped by power relations of the present, and therefore necessarily perspectival (or, even, presentist). An ethos of critique, then, ‘‘is an exercise in which extreme attention to what is real is confronted with the practice of a liberty that simultaneously respects this reality and violates it’’ (Foucault, 1997, p. 311).

A perspectival commitment is evident in Foucault’s history of Enlightenment. Exactly two hundred years after the publication of Kant’s ‘‘Was Ist Aufkla ̈ rung?’’, Foucault (1984/1997) used the same title for his own essay, ‘‘What is Enlightenment?’’ In it, Foucault ironically reiterates Kant’s characterization of Enlightenment as an ‘‘ethos of critique.’’ He writes, ‘‘[Kant’s] critique is, in a sense, the handbook of reason that has grown up in Enlightenment; and,


conversely the Enlightenment is the age of the critique’’ (Foucault, 1997, p. 308). By characterizing Enlightenment this way, Foucault is writing a history of the present. The essay challenges the present assumption that modern reason is the same as Enlightenment reason. Against continuity, Foucault’s genealogy characterizes the relation- ship between Enlightenment and modernity as a historical break. In order to document this break, Foucault mobilized an array of his- torical referents to analyze a profound disjunction between Enlight- enment (roughly culminating with Kant’s work) and modernity.8 In The Order of Things, Foucault writes:

on the archaeological level, we see that the system of positivities was transformed in a whole-sale fashion at the end of the 18th century and beginning of the 19th century. Not that reason made any progress: it was simply that the mode of being of things, and of the order that divided them up before presenting them to the understanding, was profoundly altered. (Foucault, 1970, p. xxii).
In contrast to the Enlightenment ethos of critique, Foucault calls modernity the heroization of the present: ‘‘Modernity is not a phe- nomenon of sensitivity to the fleeting present; it is the will to ‘heroize’ the present... This heroization is ironic, needless to say’’ (Foucault, 1997, p. 310).

That modernity was a developmental outgrowth of Enlightenment is still a truism in some intellectual circles. Working to dislodge ‘‘reason’’ from its modernist affiliation with instrumental rationality (a project eerily reminiscent of the Frankfurt School), Foucault characterized Enlightenment reason as something utterly different from modern reason. By writing history in this way, Foucault renders the modernist assumptions about reason less natural and less inevi- table. In genealogy the limits of reason are revealed to be conven- tional products of particular historical circumstances. When assumptions are revealed as conventional rather than natural, they lose their grip on thought. Genealogy thereby casts a spell to trans- form things from inevitable to changeable.

8 In a 1976 interview, Foucault (2000) downplays the emphasis on discontinuity in The Order of Things. He says, ‘‘My problem was not at all to say ‘Voila` , long live discontinuity, we are in the discontinuous and a good thing, too,’ but to pose the question ‘How is it that at certain moments and in certain orders of knowledge, there are these sudden take-offs, these hastenings of evolution, these transformations which fail to correspond to the calm, continuist image that is normally accredited?’’ (p. 114).

In this example about Enlightenment, an ‘‘attitude of perpetual critique’’ is a way of calling attention to the necessarily perspectival qualities of genealogy. Foucault recognized that a certain kind of presentism is inevitable; we can see only through our presentist lenses. So rather than pretend to be able to compensate for a perspectival view – by methodological rigor, for example – we are invited to write history in a way that ‘‘simultaneously respects this reality and violates it’’.


In this section I summarize Habermas’s critique of Foucault’s his- toriography and read that critique through the tropological lenses suggested by Hayden White.
Habermas accused Foucault of being contradictory by invoking the very thing he intended to dispense with, namely the philosophy of the subject:

In his basic concept of power, Foucault has forced together the idealist idea of transcendental synthesis with the presuppositions of an empiricist ontology. This approach cannot lead to a way out of the philosophy of the subject, because the concept of power that is supposed to provide a common denominator for the con- trary semantic components has been taken from the repertoire of the philosophy of the subject itself (Habermas, 1985/2000, p. 274).

Basically, this is an analytical critique. It focuses on the logical as- pects of argument and finds an inconsistency.9 The upshot of Haber- mas’s critique comes when he says, ‘‘genealogical historiography emerges from its cocoon as precisely the presentistic-, relativistic-, cryptonormative illusory science that it does not want to be’’ (Haber- mas, 1987/2000, p. 276, emphases in original).

If we invoke White’s tropological historiography, we can say that Habermas treats Foucault’s history as if it were a formal argument (as opposed to White’s other narrative tropes, namely explanation by emplotment or explanation by ideological implication). However, to treat Foucault’s history exclusively in terms of its formal structure is to ignore the other narrative and rhetorical features of the argument.

9 Interestingly, elsewhere Habermas deploys a tropological critique of Derrida’s work (see ‘‘Excursus on Leveling the Genre Distinction between Philosophy and Literature,’’ 1985/2000, pp. 185–210).


For example, White identifies the trope of ‘‘explanation by emplot- ment’’ as characteristic of historical writing; explanation by emplot- ment can be executed in any one of four dramatic modes: Romance, Comedy, Tragedy, and Satire. Habermas’s own work tends to weave Romance with Tragedy insofar as his historical analysis rejects uto- pianism (therefore Tragic), but extends a theory of communicative action that offers possibilities for the development of agency and autonomy (therefore Romantic).

Few people would disagree that Foucault’s work tends to execute the Satirical trope, and his genealogies are infused with irony. As White asserts, Satire ‘‘presupposes the ultimate inadequacy of the visions of the world dramatically represented in the genres of Ro- mance, Comedy, and Tragedy alike’’ (White, 1973, p. 10, emphasis in original). Furthermore, White declares that irony is nihilistic in that it does not spur political activism in the same way as an ‘‘inspirational’’ or ‘‘motivational’’ speaker does.

Irony tends to dissolve all belief in the possibility of positive political actions. In its apprehension of the essential folly or absurdity of the human condition, it tends to engender belief in the ‘madness’ of civilization itself and to inspire a Mandarin-like disdain for those seeking to grasp the nature of social reality in either science or art (White, 1973, p. 38).

So even if Habermas were to extend his critique of Foucault based on White’s schema of narrative tropes, he would still probably con- clude that Satire and Irony could not serve as politically active critical contributions to historiography. If, according to Habermas, it is the role of critical intellectuals to provide theoretical mechanisms for successful completion of modernity, then, presumably, Satire and Irony would not serve such a cause. In fact, Habermas regards the ironic turn in Foucault’s history to be a logical flaw:

Whereas, according to Foucault’s diagnosis, the human sciences submit to the ironic movement of scientistic self-mastery and end up in an unsalutary objectivism (or better yet – come to an end therein), a no less ironic fate overtakes genealogical historiography; it follows the movement of a radically historicist extinction of the subject and ends up in an unholy subjectivism (Habermas, 1985/2000, p. 276).

I would like to address Habermas’s critique on two fronts, first regarding his construction of rationality, and second regarding the effectiveness of irony.
1. In these examples, both Habermas and (to a lesser extent) White construct reason in formal analytic terms. Habermas focuses almost


exclusively on the formal, logical aspects of Foucault’s historiogra- phy, and White regards Irony as anti-scientific.10 In these examples, Satire and Irony get classified outside the limits of critical rationality, and therefore not contributory to the project of ‘‘rationalization of the lifeworld.’’

However, the limits of reason can be extended to include Satire and Irony without overextending the limits of reason past the point where reason is indistinguishable from non-reason. Sandra Harding (1992) makes an analogous argument about objectivity. She says that if we do not consider historical and contextual factors as part of scientific discovery, then we are censoring a large part of the available knowledge, and therefore we are being reductionist rather than objective. In other words, if we reduce scientific knowledge about radioactivity to its laws of physics, without considering the envi- ronmental and militaristic implications of those laws, then we have drawn ideological limitations around knowledge in a non-objective way. Similarly, in the case of reason, if we limit the definition of reason to its analytic propositions, without considering the other narrative and tropological features of language, then we are being reductionist about reason. In Foucault’s historiography, the value of reason is not diminished by the inclusion of irony; rather, the value of reason is fortified by constructing reason in a more robust way.

2. Habermas makes a remarkable statement distinguishing archaeology from genealogy:

Under the stoic gaze of the archaeologist, history hardens into an iceberg covered with the crystalline forms of arbitrary formations of discourses... Under the cynical gaze of the genealogist, the iceberg begins to move. (Habermas, 1987, p. 252; emphases in original)
In this excerpt, we can see that, ironically, Habermas explicitly rec- ognizes genealogy as effective history. History that effects change (‘‘moves icebergs’’) is precisely the definition of exemplary critique. Legislative critique constructs agency in terms of propositional statements and legislative assertions. This aspect is decidedly missing from Foucault’s historiography. That does not imply that agency is missing from Foucault’s historiography. Even according to

10 In other places, Habermas’s analysis engages with other aspects of the lifeworld, including ‘‘prepositional, illocutionary, and intentional components’’ 1985/2000, p. 343.

Habermas, evidently, genealogy can move icebergs. Therefore, Fou- cault’s historiography exemplifies agency.

Thoroughly suspicious, even pessimistic, genealogy exemplifies agency because it is a strategic act of political engagement. If, at times, Foucault’s analyses appear determinative, it is not because genealogy denies agency. Rather it is because the historiography does not legislate agency. A thoroughly suspicious attitude grants us critical leverage, and genealogy is an act of transgression. If we are not suspicious, if we trust in legislation to provide license for agency, then we have no chance of discovering ways in which legislation may also be complicit with existing power hierarchies.

One possibility for reproductive complicity in legislative histories becomes apparent when we seek to address the following peda- gogical questions: In any given history, who is (implicitly or explicitly) authorized to teach whom about what? Are ‘‘oppressed’’ people always cast in the role of student rather than teacher? How can we tell the difference between autonomous voices and depen- dent, colonized voices? A legislative portrayal of autonomy and agency in historiography has confounding implications that Cruik- shank (1999) calls ‘‘the will to empower.’’ The will to empower reiterates an asymmetric relationship of authority between those who are regarded as autonomous and those who are regarded as dependent. When praxis means legislating the means by which autonomy can be achieved, an essential hierarchy is reiterated. Reiteration is the means of discursive support for the system of reasoning that perpetuates those hierarchies. Repetition reinforces the standard.


To draw a sharp distinction between legislative and exemplary hi- story is to clarify, indeed to legislate, the ways in which praxis and agency in Foucault’s historiography are different from praxis and agency in Habermas’s historical project. To this extent the distinction has some pedagogical value in a didactic sense. However, from the standpoint of exemplary critique, the distinction takes on other problematic dimensions. One of those problems is raised by Bauman, namely the assumption of ‘‘meta-professional authority.’’ To return to Bauman’s words:


While the post-modern strategy entails the abandonment of the universalistic ambitions of the intellectuals’ own tradition, it does not abandon the universalistic ambitions of the intellectuals towards their own tradition; here, they retain their meta-professional authority, legislating about the procedural rules which allow them to arbitrate controversies of opinion and make statements intended as binding. The novel difficulty, however, is how to draw the boundaries of such community as may serve as the territory for legislative practices. (Bauman, 1987, p. 5)

In this excerpt, Bauman calls our attention to the legislative dimen- sions of interpretation. Similarly, insofar as genealogy is written in order to incite, it may be said to entail implicit legislative demands. At one level of discourse analysis, the distinction between legislative and exemplary is fairly clear-cut. At another level, the distinction becomes much blurrier.

J.L. Austin recognized this blurriness as well. At the end of his linguistics course, Austin pointed out a problem with the strict dis- tinction between kinds of speech acts. After meticulously docu- menting his famous stative/performative distinction, Austin did not omit irony. Remarkably, at the conclusion of his last lecture, Austin pointed to the strangeness of what might be called a performative contradiction that he himself executed when he conducted his lin- guistic analysis in lecture style. He wrote:

In these lectures, then, I have been doing two things which I do not altogether like doing. These are: (1) producing a program, that is, saying what ought to be done rather than doing something; (2) lecturing. (Austin, 1962, p. 163)

Exactly the same kind of accusation can be applied to the analysis in this paper. To draw a strict analytical distinction between legislative and exemplary critique is a legislative speech act. Furthermore, there are implicit aspects to the rhetoric that assume authority in ways similar to legislation.

Insofar as genealogy can be said to work towards a particular vision of agency and historiography, it cannot be free of some of the emissary qualities that are characteristic of legislative history. How- ever, the subject/text relation that is established in an exemplary hi- story is nonetheless different from the subject/text relation established in a legislative history.

Many genealogies target the fear-of-not-being-normal as a crucial disciplinary technology. In a way, the critical part of the text is not what you write, but where you write it from: ‘‘to live one’s life as a work of art.’’ To assume a place of subjective freedom places a heavy ethical burden on the subject positions of both reader and


writer. Critical history can no longer pass the ethical buck off to some institution or structure that is beyond anyone’s control. Rec- ognizing the magnitude of this ethical burden means recognizing that everything is dangerous; recognizing that everything is dangerous means adopting a fairly modest – even pessimistic – view of the intellectual’s capacity to change the world. In genealogy, there is an enactment of agency in the exemplification of new ways of thinking. At the same time, this agency recognizes its perspectival limits, and is reluctant to make promises about its redemptive or emancipatory capacities.

The role of the critical intellectual in genealogical approaches is not one of forecasting, it is one of recasting memory; as Adorno said, ‘‘All reification is a forgetting.’’ The ethical position of Foucault’s historiography is one of ‘‘pessimistic activism,’’ 11 in which more faith rests on the unforeseen and uncontrolled democratic possibili- ties for the future than on the ability of intellectuals to solve social problems. At the same time, the critical leverage of genealogical critique is historically limited; genealogies can be characterized as critical only insofar as they challenge a significant level of com- monsense assumptions at a given historical moment. Whenever the Foucaultian historiography becomes formalized and instrumental- ized, critical intellectuals will be called upon to interrupt that ap- proach, too.


Austin, J.L. (1962). How to do things with words. Cambridge, MA: Harvard Uni- versity Press.
Bauman, Z. (1987). Legislators and interpreters: On modernity, post-modernity and intellectuals. Ithaca, NY: Cornell University Press.
Bernstein, R. (1994). Foucault: Critique as philosophical ethos. In M. Kelly (Ed), Critique and power: Recasting the Foucault/Habermas debate (pp. 211–241). Cambridge, MA: MIT Press.
Connolly, W.E. (1985 August). Taylor, Foucault, and otherness. Political theory, 13(3), 365–376.
11 This phrase is from Foucault’s most famous ‘‘Everything is dangerous’’ quo- tation, which, however, is seldom quoted in its entirety: ‘‘My point is not that everything is bad, but that everything is dangerous, which is not exactly the same as bad. If everything is dangerous, then we always have something to do. So my po- sition leads not to apathy but to a hyper- and pessimistic activism’’ (from ‘‘On the Genealogy of Ethics’’).
Cruikshank, B. (1999). The will to empower: Democratic citizens and other subjects. Ithaca, NY: Cornell University Press.
Dean, M. (1994). Critical and effective histories: Foucault’s methods and historical sociology. New York: Routledge.
Depaepe, M. & Simon, F. (1996). Paedagogica Historica: Lever or mirror in the making of the history of education? Paedagogica Historica, XXXII(2), 421–450. Foucault, M. (1970). The order of things: An archaeology of the human sciences. New
York: Random House.
Foucault, M. (1980a). Body/Power. In C. Gordon (Ed),
Power/Knowledge: Selected
interviews and other writings 1972–1977 (pp. 55–62). New York: Pantheon. Foucault, M. (1980b). Prison talk. In Power/Knowledge: Selected interviews and other
writings, 1972–1977 (pp. 37–54). New York: Pantheon.
Foucault, M. (1984/1997). What is enlightenment? In P. Rabinow (Ed.),
Foucault: Ethics, subjectivity and truth. (Volume One of The Essential Works of
Michel Foucault 1954–1984) (pp. 304–319). New York: The New Press. Foucault, M. (1989/1996). Talk show. In Sylve`re Lotringer (Ed), Foucault live:
Collected interviews, 1961–1984 (pp. 133–145). New York: Semiotext(e). Foucault, M. (1990). The use of pleasure (trans. R. Hurley). History of sexuality, Vol.
2. New York: Random House.
Foucault, M. (1998). Preface to transgression. In J.D. Faubion (Ed.),
Michel Fou-
cault: Aesthetics, method, and epistemology. (Volume Two of The essential works of
Michel Foucault 1954–1984). pp. 69–87. New York: The New Press.
Foucault, M. (2000). Truth and power. In J.D. Faubion (Ed) trans. R. Hurley et al.
Michel Foucault: Power Essential works of Foucault 1954–1984, Vol. 3 (pp. 111–
133), New York: The New Press.
Freire, P. (1994).
Pedagogy of hope: Reliving pedagogy of the oppressed, trans. Robert
R. Barr. New York: Continuum.
Giroux, H. (1998). Education in unsettling times: Public intellectuals and the promise

of cultural studies. In D. Carlson & M. Apple (Eds), Power/knowledge/pedagogy: The meaning of democratic education in unsettling times (pp. 41–60). Boulder, CO: Westview Press.
Goldstein, J. (Ed) (1994). Foucault and the writing of history. Cambridge, MA: Blackwell Publishers.
Habermas, J. (1985). Questions and counterquestions. In R.J. Bernstein (Ed), Habermas and Modernity (pp. 192–216). Cambridge, UK: Polity Press.
Habermas, J. (1987). The philosophical discourse of modernity: Twelve lectures, trans. F.G. Lawrence. Cambridge, MA: MIT Press.
Harding, S. (1992). After the neutrality ideal: Science, politics, and ‘strong objec- tivity’. Social Research, 59(3), 567–587.
Owen, D. (1995 November). Genealogy as exemplary critique: Reflections on Fou- cault and the imagination of the political. Economy and Society, 24(4), 489–506.
Parsons, T. (1922). Introduction. In M. Weber (Ed), Sociology of religion, (pp. xix- lxvii), trans. E. Fischoff. Boston: Beacon Press.
Popkewitz, T.S., Franklin, B.M. & Pereyra, M.A. (Eds) (2001). Cultural history and education: Critical essays on knowledge and schooling. New York: RoutledgeFalmer. Weber, M. (1946). The social psychology of the world religions. In Talcott Parsons, Edward Shils, Kaspar D. Naegele & Jesse R. Pitts (Eds), Theories of society: Foundations of modern sociological theory (pp. 1385–1402). New York: The Free
Press. (reprinted from From Max Weber: Essays in sociology. Hans A. Gerth and C. Wright Mills [Trans. and Eds.]. pp. 267–301. New York: Oxford University Press.)
Weber, M. (1961). The social psychology of the world religions. In T. Parsons, E. Shils, K.D. Naegele & J.R. Pitts (Eds), Theories of society: Foundations of modern sociological thought (pp. 1385–1402). New York: The Free Press.
White, H. (1973). Metahistory. Baltimore, MD: Johns Hopkins University Press. White, H. (1987). The content of the form: Narrative discourse and representation.
Baltimore, MD: Johns Hopkins University Press.
Woolgar, S. (1986). On the alleged distinction between discourse and
praxis. Social
Studies of Science, 16, 309–317.
Michigan State University 116I Erickson Hall
East Lansing
MI 48824-5047