Terjemah

Chapter Eleven

The Political Theory of Recognition



















This chapter will of necessity offer only a truncated treatment of the question as only a whole book could do justice to the prob- lems it addresses. This is because of the com- plexity and breadth of questions of political philosophy today and also because most of the criticisms that have been raised against Honneth’s ethics of recognition have tar- geted the implications of his model for politi- cal philosophy.



This chapter will therefore have to be selec- tive. It is organised in three sections, each dedicated to a specific issue. The first section deals with Honneth’s account of the liberal- ism/communitarianism debate. Because the writings that Honneth specifically dedicated to political philosophy first centred on that debate, this initial problem allows me to schematically present his official response. This section, however, must also briefly show in what sense the theory of recognition is nei- ther a liberal, nor a communitarian position.



The second section addresses one of the most serious objections raised against Honneth’s theory of recognition: its alleged weakness in dealing with problems arising from economic







injustice. Focusing on this problem enables us to present another one of Hon- neth’s explicit proposals in political philosophy, with a focus, this time, on the critique of political economy.



The third section deals with the question of identity. This is probably the area that is most often targeted by critiques of recognition, and mostly in direct connection with problems of political philosophy.







Section 1. The critique of liberalism





Recognition and the liberalism/communitarianism debate



The Struggle for Recognition was finished by Honneth at a time when he was actively involved in introducing the debate between liberalism and commu- nitarianism in Germany. This historical overlap had a strong influence on the final shape of Struggle for Recognition. As with moral philosophy, Honneth explicitly presented his theory of recognition as an alternative, “between Aris- totle and Kant”, to the two famous positions that were dividing the major English-speaking scene at the time. A passage at the end of the book made the point clearly and succinctly:



The line of argument that we have been following in the reconstruction of the model of recognition points to a position that does not seem to fit clearly into either of these two alternatives. Our approach departs from the Kan- tian position in that it is concerned not solely with the moral autonomy of human beings but also with the conditions for their self-realisation in gen- eral. Hence, morality understood as the point of view of universal respect, becomes one of several protective measures that serve the general purpose of enabling a good life. But in contrast to those movements that distance themselves from Kant, this concept of the good should not be conceived as the expression of substantive values that constitute the ethos of a concrete tradition-based community. Rather, it has to do with the structural elements of ethical life, which, from the general point of view of the communicative enabling of self-realisation, can be normatively extracted from the plural- ity of all particular forms of life. (. . .) Our recognition-theoretic approach stands in the middle between a moral theory going back to Kant, on the one hand, and communitarian ethics, on the other. It shares with the former the







interest in the most general norms possible, norms which are understood as conditions for specific possibilities; it shares with the latter, however, the orientation towards human self-realisation as the end.1

At a superficial level, Honneth’s position in political philosophy seems in fact to have greatly shifted. The early Marxist position seems to gradually make way for a more mainstream liberal one, culminating in a passage of the book written with Fraser, where Honneth explicitly aligns the ethics of recognition with the “teleological liberalism” of Rawls and Raz.2 In that same passage, however, the third “classical” reference is Hegel again. Given the role that Hegel plays for many alternatives to liberalism, as the most decisive first ref- erence point, notably because of his critique of rights-based approaches to justice and social-contract types of arguments, it is clear that Honneth means something quite vague by “liberalism” in this passage. Liberalism in this particular page is only the name for any position premised on the notion of equal treatment of all and the acknowledgment of every individual’s right to autonomy.



By contrast, the basic idea underpinning Honneth’s alternative position in political philosophy is the idea of the “social conditions of individual auton- omy”.3 Honneth has held on to this most fundamental intuition throughout his work. Most significantly, he has never abandoned it, even in his most recent writings, those which, for a superficial view of his development, could signal a shift towards liberalism. For example, it is an idea that he keeps returning to in the book with Fraser. We encountered it in Honneth’s second major read- ing of Hegel (Suffering from Indeterminacy), presented in 2000.

The long quote at the beginning of this chapter already gives a good sum- mary of what is entailed in this idea. We also encountered a particularly vivid expression of it in the chapter on social theory (chapter 9), when we noted Honneth’s particular way of interpreting the difference between Hobbes and Rousseau:





unlike political philosophy, social philosophy no longer seeks outs the con- ditions of a correct or just social order, but instead attempts to ascertain the limitations that this new form of life imposes on human self-realisation.4

The emphasis on the social conditions of justice remains a distant yet power- ful echo of a typically “Left-Hegelian” critique of politics. In this tradition, the critique of politics in fact involves two separate problems. Both problems turn around the problematic relation between society and politics: the first prob- lem relates to the social origin of normative principles; the second is the prob- lem of the adequacy of normative principles to social reality. One of the most significant examples of such a line of critique is Marx’s twofold critique of idealist philosophy and of normative political theory. Already in his earliest writings, for example in his critique of Hegel’s political philosophy the young Marx had connected the idealism of mainstream philosophy and politics to the inverted state of real society:5 the alienation of humanity in contemporary society brings about a false realisation of it in the political and the juridical. The reality of the social therefore directly contradicts that illusory realisation. This is Marx’s famous opposition between the citizen and the bourgeois.6 The ideal, philosophical expression, which is also the justification, of this inversion in reality occurs through a theoretical inversion that constitutes the essence of all idealisms: taking the norms for the real. Honneth’s intervention in political philosophy can be seen as a distant, mediated, repetition of a similar critical stance towards normative political philosophy. Marx saw in normative politi- cal philosophy an idealistic repetition, and thus an ideological justification, of a real political inversion: the satisfaction in the realm of the norms of needs that are unfulfilled in social reality. Similarly, Honneth presses contemporary political philosophy on the articulation of its normative claims with social reality, in terms both of the origin of the political norms and principles in a social reality where suffering and injustice are omnipresent, as well as on the realisability of normative principles within social reality.



These two critical insights—the social origin of political claims and the social applicability of political principles—centred around the problematic link between politics and social life, are also at the core of Honneth’s reception of the liberalism/communitarianism debate in the early 1990s, in “The Limits of Liberalism” (1991) and in the introduction to the edited collection Honneth published to present the key texts of the debate (Kommunitarismus, 1993). In retracing the conceptual stages of the debate, Honneth is careful not to reduce it to the ontological and methodological problems of atomism versus holism. As he notes, Rawls’ reply to Sandel allowed him to maintain his procedural- ism even after he accepted the ontological point that subjective identity can be formed only within a community of values and a strong conception of the good. Indeed, Rawls was able to turn the objection back to the communitar- ians by emphasising the argument at the heart of his model, the idea that



The legal guarantee of personal autonomy is not something which stands in the way of the intersubjective process of personal identity formation, but rather, conversely, first makes it feasible in society.7

In other words, granting the ontological premise about the intersubjective char- acter of subjective formation not only does not refute, but even strengthens the case for the methodological and normative priority of rights and liberties. We can note that this leap, from the ontological to the normative, could be seen to be a problem for Honneth himself, since he is typically an author who derives normative consequences from a substantive model of the subject. This could mean that he would have identified in others a problem that in fact also plagues his own model. Zurn’s critique of the anthropological basis of Honneth’s own theory of justice articulates precisely that criticism. Honneth, however, is well aware of the difficulty for himself, and answers it by histori- cising the anthropological argument: the social conditions of subjective iden- tity have arisen as a result of normative differentiation and henceforth create a framework, which plays simultaneously a constitutive and a normative role.



Honneth identifies more serious objections to the theory of justice in Taylor’s and MacIntyre’s critiques. The point here is no longer to argue directly from the ontological to the normative, to reject the proceduralist method on the basis of a “communitarian” view of the subject, since the two standpoints (the ontology of the subject and the norms of politics) must be strictly separated and liberals can show that their normative position can remain largely unaf- fected by ontological questions. Indeed, liberals can take an “intersubjective” view of the subject on board without major changes to the overall construct. Rather, the core question concerns the problem of the relation between the conception of subjective self-realisation and its realisation in a community. If the subject can truly achieve full autonomy only through sharing common value references because subjective self-realisation relies on meta-subjective, social frameworks of meaning, then liberal theories of justice have a huge problem to confront because they refuse, methodologically, to discuss such ethical frames of reference. They agree about the intersubjective dependency of the subject, but fail to translate this into the type of normative consequence that communitarians emphasise: namely, that the subject can therefore realise herself or himself only within communities for which this goal of self-reali- sation and what it entails already exist as shared values, which implies that these values therefore exist prior to the individual’s perception of them. For example, the communitarians note that legally guaranteed liberties cannot be exercised by the subject if the subject is not the member of a community that makes this exercise meaningful and socially possible. In other words, there are irreducible social preconditions to the individual quest for self-realisa- tion, whatever the actual content of self-realisation might be, depending on the social and historical context. But liberal theories cannot account for this moment of the social precondition of individual self-realisation, because of their neutrality towards ethical values.



Given that the liberalist tradition insists that normative status may not be granted to any specific ethical value, it is not possible within the framework

of such theories to develop the idea of a community that is integrated in terms of a notion of ethical life, even though, it is precisely this which we evidently have to presuppose when trying to explain the process of individual realisation of freedom.8

In his introduction to the Kommunitarismus volume, as he introduces Sandel and McIntyre’s positions, Honneth puts the same argument in negative form: more is needed for a society to function than just the collection of legally entrenched individual evaluative preferences; for society to be morally inte- grated, its underlying normative principles require the sharing of a basic nor- mative horizon. Without this horizon, subjects will not be motivated to fulfil and respect the principles. A collection of legally protected individuals does not yet make a society. The sharing of a fundamental normative horizon is also required as a precondition of social life; and because individuals develop their autonomy within the framework of social life, it is also a condition of a free individual life.9





Is Honneth a communitarian?



But this communitarian argument sounds very close to the general orienta- tion of Honneth’s thinking in social and political philosophy. Honneth’s own insistence on the intersubjective preconditions of individual freedom would seem to make him a close ally of the communitarians. In the passage from The Struggle for Recognition cited earlier, he was criticising the narrowness of the liberal paradigm precisely by holding up against it the principle of: “mak- ing possible the good life” (die Ermöglichung des guten Lebens). As we saw in chapter 7, Aristotle’s key premise that the whole precedes the part in a sense encapsulates Honneth’s own basic ontological position in social theory.

Even more significant is the overlap between Honneth’s philosophical anthro- pology and that of Charles Taylor. It is quite telling that Taylor was invited to write the preface to the English translation of Social Action and Human Nature, the study in philosophical anthropology, which was decisive in determin- ing the course of Honneth’s later thought. Taylor gave the young Honneth and Joas an illustrious example to follow, and a decisive theoretical support, in their attempt to anchor the critique of modern society in philosophical- anthropological arguments. Honneth later paid his intellectual debt to Tay- lor in the postface to the German translation of Negative Freedom.10 The basic idea underpinning both theories is the same, namely the idea of “the com- municative presuppositions of all processes of self-realisation”.11 As we saw in chapter 4, it is this fundamental insight which explains why Honneth finds Taylor’s model the right one to follow in order to ground critique philosophi- cally, even ahead of Habermas.12 Indeed, if one recalls the initial sections of Taylor’s famous essay on recognition, in which he laid out the basic premises of his political theory, one would find a number of major overlaps: the idea that “a crucial feature of human life is its fundamentally dialogical character”; that, as a result, “misrecognition (. . .) can inflict a grievous wound, saddling its victims with a crippling self-hatred”; or the idea that the pre-modern notion of “honour” has been replaced by a new notion of normative importance for the individual, the notion of dignity, which sunders itself between a univer- sal moment, the “politics of equal dignity”, and a particularistic moment, the politics of difference”.13



However, despite Honneth’s thematic closeness to Taylor, I would argue that his position in actual fact differs quite substantively from him and from the communitarians more broadly.14 The crucial difference between a Honneth- ian and a communitarian position in political philosophy concerns the inter-

pretation of the notion of “social preconditions of subjective identity”, that is to say, the way in which the reference to a common value horizon necessary for conceptualising both individual autonomy and social life is construed. There are two fundamental ways of understanding this “shared value hori- zon”. That divergence leads to major differences in the political outlook, both on the theoretical and practical levels.



Both the communitarians and Honneth ground their normative outlooks in an ontology of the human subject, in a view of the human subject’s radical social, or intersubjective, dependency. Communitarians interpret this ontological fact by saying that any subjective attempt at self-realisation occurs through the mobilisation of the meta-subjective norms of the community in which the subject was socialised. Taylor’s “strong evaluations”, for example, through which the subject interprets, evaluates and organises his or her intentions and desires, arise within a communal, social and cultural pre-given world, the horizon of values of the community in which the subject has been socialised. This remains the case even when the “strong evaluations” are creative, or transformative, and overstep the boundaries of the original value horizon. Even then, the new values remain bound up with the community’s horizon inasmuch as they overstep, or rather overstretch, precisely that horizon and could be articulated only in that precise normative vocabulary. This is the ground, as Honneth recalls, of Taylor’s rejection of the Habermasian proce- duralist approach in normative questions:



Since, as human beings, we cannot avoid understanding ourselves in the light of strong evaluations, an external position from which we could nor- matively define a specific procedure that would transcend cultures, can in principle not be reached by us; on the contrary, any such definition is itself always already tied to an overarching understanding of the good life that stems from the normative traditional horizon (normativen Traditionszusam- menhang) of the particular culture to which we belong.15

Honneth of course agrees that selves develop through procedures of sociali- sation that are thoroughly determined by the relationship to others and their embeddedness in a particular culture. But Honneth prefers to emphasise a dif- ferent implication of this intersubjective view of subjective formation. Rather than the embeddedness in a specific cultural tradition, Honneth emphasises the essential vulnerability of the subject as a result of intersubjective depen- dency. To say it in summary form: whilst the communitarians stress the com- munity side in the individual-community dialectic, Honneth stresses the individual one. The normative consequence of such difference in emphasis relates to the aims of social philosophy: if intersubjective vulnerability rather than communality is the core concern, the decisive dimensions to study are the structures through which individuals can achieve self-realisation through intersubjective dependency, not so much the subject’s ties to the commu- nity.16 In other words, the core concept is that of individual self-realisation, not whether community is itself a primary good.



Most of the other differences between Honneth and communitarian positions derive from this initial difference in emphasis and from the implications that flow from it for the critical project.

The first major difference is theoretical, and concerns the problem that plagues contemporary political philosophy: how to establish normative guidelines for an ethic, that is to say, for collective representations of the good, given the irreducibility of value pluralism in modern societies. In Habermas, the initial “ontological” argument about the intersubjective dependency of the subject leads to a formal-universalistic, proceduralist solution: justice is synonymous with equal participation in normative discourse. Honneth shares Haber- mas’ ontological presupposition, even if he aims to expand communication beyond linguistic understanding. He also shares Habermas basic direction in the application of that insight to political theory. The spheres of recognition characterise precisely the formal principles that result from a more fleshed out anthropological approach to communication. On that basis, even if the content of their philosophical anthropologies differ, Honneth and Habermas share the same methodological approach to the question of value pluralism and the problem of relativism that is entailed in it. Both insist on the meta- subjective normative horizons framing subjective self-realisation, but the problem of relativism is avoided because the principles that now arise from that insight have a formal universal import. They can be shown to apply uni- versally, in formal or structural terms, beyond variations in content (the con- tent of rights, of types of social recognition).



By contrast, from Honneth’s perspective, liberals and communitarians are mired in a “theoretical cul-de-sac” on that issue. Both camps end up agreeing, despite their differences that “well integrated communities play a constitutive part in the realisation of individual freedoms”.17 However, because they draw a contextualist conclusion from that idea, namely that there is therefore no possibility of adjudicating on conceptions of the good life from outside these communities, they can no longer discriminate between models of commu- nity, between freedom enhancing and freedom-curtailing cultural practices. For Honneth, the exemplary case of such methodological self-contradiction in contemporary political philosophy, is Michael Walzer’s hermeneutic theory of justice, as his critical review published in 1991 argues:



in the writings of Walzer, a discrepancy is opened up that is typical of the contemporary intellectual situation: a rhetoric of particularism prevents the articulation of the remaining universalistic motives which one would pre- cisely have to use in order to buttress the defence of a cultural pluralism that one aims to defend so decisively.18

The conundrum is the same for the liberals since their ethical neutrality also prevents them from making normative decisions regarding conceptions of the good life. Typical here is Rawls’ own development, which led him from a uni- versalistic to a historically situated account of justice, a shift that runs the risk of falling into the contextualist trap:

As a result, both sides find themselves in the same dilemma. They no lon- ger have any supra-contextual criterion with which to distinguish justifiably between morally acceptable and morally objectionable concepts of the col- lective good. (. . .) Yet, both sides are at the same time all the more dependent on such a criterion because in the mean time they widely agree that without any link to value convictions, there is an inability to clarify the conditions under which individual freedom is realised.19

For Honneth, Habermas’ approach is the only one that enables us to avoid falling into the contextualist trap. Despite all other differences with Haber- mas, Honneth’s approach is deeply inspired by him (and this can surely be related to their underlying “Left-Hegelianism”): the moral principles of mod- ern societies are sought in formal structures arising from the basic anthropo- logical-philosophical premise of the intersubjective dependency of subjective identity.



These theoretical questions are directly linked to more practical questions, in particular to the question of the model of social critique. What Habermas’ solu- tion demonstrates is both the need and the possibility of maintaining a valid reference to a principle of “transcendence within immanence” in normative political theory. Such a prerequisite is of course vital for critical theory. And of course, it constitutes a theoretical imperative that has featured prominently, from the very beginning, in all of Honneth’s writings, from his initial critique of Foucault, for example, to the critiques, fifteen years later, of Gadamer and McDowell. Throughout, the fundamental concern has remained the same: any consistent normative-critical moment in social philosophy requires the passage to some context-transcending viewpoint, and the main theoretical difficulty for contemporary social philosophy is in how to found and articu- late this moment.20

The preservation of a moment of “transcendence within immanence”, how- ever, also preserves a certain radicality in the political implications of social theory. This will not sound very convincing against the background of the strong continuity between Habermas’ and Honneth’s models of political theory. However, it can also be argued that, already in Habermas, but even more so for Honneth, the distance from liberal and communitarian method- ologies can be related not first and foremost to conceptual or methodologi- cal issues, such as those just highlighted, but primarily to the implications of these approaches for social critique.21



Regarding communitarian positions, the implication of a political theory based on the notion of a struggle for recognition is a deep suspicion towards the notion of community. The theory of recognition intends to maintain the irreducibility of conflict and the central role played by struggle in the theory of society. For Honneth as much as for communitarians the idea that social inte- gration requires the reference to supra-individual value horizons is central, especially in reference to the third sphere of recognition. But a crucial feature of such value horizons in his model is that they are contested, and secondly, that they are contested not primarily on cultural, but on social grounds. Or rather, more precisely, social conflicts around norms or values that are cul- tural in content, in fact have a moral basis inasmuch as they trigger claims of injustice and demands of redress. As was shown earlier, Honneth’s specific approach to cultural conflicts is to highlight that they are just as equally social conflicts, conflicts along and about structures of domination.

This insistence on the conflictual structure of the social in general, and by repercussion on the “fragmented” (or rather: split, torn, as the German term “zerrissen” denotes) character of value horizons, thus gives a different mean- ing to the notion of “the good life”. For communitarians, the shared value horizon must be largely uncontested because for them it is a genetic and logical precondition of any subjective capacity of ethical self-articulation. In Honneth on the other hand, the dialectic between subjective identity and communal value horizon is fraught with tension. More often than not, the relation is an unhappy one, leading notably to what he describes as social pathologies, until a successful struggle for recognition is able to modify the framework and make it possible for specific claims of recognition to be taken into account. From Honneth’s perspective, communitarianism therefore runs the risk of advocating conservative positions, or at least of being unable to address transformative practices adequately. The explicit political opinions of the communitarian theorists are not the issue here. Honneth’s critical position targets equally traditionalist and more progressive accounts. In all cases, the difference in the interpretation of the motto of “the social preconditions of the good life” is grounded in the view that a critical theory of society must take its cue from the unsatisfied social demands, rather than from the socialised agents’ reliance upon communal values and norms. In other words, the her- meneutic moment that is constitutive of all types of communitarian accounts of politics, whatever their explicit political allegiance might otherwise be, always threatens to end up in a form of theoretical conservatism.22 On the contrary, Honneth’s ethics of recognition is concerned with the possibility of giving a normative justification of social movements aimed at social transfor- mation. The reference to communal value horizons in this case can only be a “dissensual” one, that is to say, one that acknowledges the necessity of ethical integration but adds immediately that the latter is only ever achieved through the misrecognition or denial of recognition of large sets of social interests.

What’s at issue here is not just the deduction of an image of politics from the concept of society. One of the major dimensions of social injustice stems from the fact that social claims are not simply unrepresented in the public sphere, but that they are not even representable, that is to say, that they are repressed from the political field, mainly because of the ideological foreclosure of that field. This is already one of Honneth’s main arguments against Habermas’ discourse ethics. A general point arising from such criticism is that normative theories of justice that overlook the structural aspect of social domination add to the injury of empirical injustice the insult of its meta-theoretical repres- sion. Such theories reinforce, or entrench, empirical mechanisms of injustice, even if their explicit intentions are progressive. This means very simply that conceptual and methodological issues, in particular the concept of society or the diagnosis of modern democracy with which one operates, beyond their theoretical import, have direct practical, political significance. From that per- spective, Habermas who himself highlights the problematic nature of the link between theory and practice in political liberalism, would himself be account- able to such critical argument. Indeed, this is the crucial point at which Hon- neth’s and Habermas’ political theories part ways. Honneth’s insistence on retaining a notion of conflict, even in a transformed, neo-Hegelian rather than neo-Marxist mode, within his theory of society, grants him a critical vantage point on the political theory that derives from discourse ethics. Despite the latter insisting on the necessity of linking the normative account of justice to a substantive theory of modern society, it seems liable to the reproach of repro- ducing within itself, at the theoretical level, the repression of unmet social demands in the political processes that is one of the central mechanisms of

contemporary injustice.23







From that point of view, the implications of Honneth’s model of the struggle for recognition for political theory seem to come close to Jacques Rancière’s own definition of politics as a rupture of the social consensus that is performed by agents demanding the practical acknowledgement of their equality. Rancière does not explicitly refer to ‘recognition’, although the notion sometimes slips through. But his radical egalitarian principle and his emphasis on “dissensus” as the core mechanism of politics are formally close to Honneth’s emphasis on the experience of injustice as the negative locus of political action.24 His own criticism of Habermas’ is similar to the one developed by Honneth in 1990.

Despite all the common features that seem to bring Honneth close to commu- nitarian positions, and especially Taylor, the emphasis on dissensus makes their normative theories of justice largely incompatible. This is mainly because Honneth bases his own account on a negativistic methodology, whereby the norm of justice is defined primarily as the abolition of injustice. On that model, the principles of a just society are those normative guidelines, that is, extensions, corrections, complements, transformations of existing principles of justice, for which those who suffer from injustice would, and sometimes do, struggle if they are in social and cultural contexts that enable them do so. The three spheres of recognition provide the grammar to analyse the differ- ent types of claims around which social movements are articulated. But they have a methodological status that is radically different from the normative principles of other political theories. They are not tied reflectively to specific historical political traditions, nor are they reconstructions of the necessary social principles to which all reasonable modern social agents would have to agree. They point to the normative dimensions that subjects suffering from social injustice implicitly rely on when they engage in social struggle against injustice. As such, these principles make no sense separated from the negative experience of injustice and the attempt at addressing the causes of injustice.





These principles thus achieve the link between the normative and the empiri- cal in a very specific way. In communitarianism, the normative principles are deduced from a philosophical anthropology emphasising the intersubjective preconditions of subjective realisation, and they are hermeneutically tied to distinctive communities, by reference to each community’s specific notion of the good life. This always threatens to rob normative political philosophy of the possibility of critique because the hermeneutic methodology does not provide valid access to the context-transcending norms on which to ground critique. Honneth’s principles, by contrast, are indeed drawn from philosoph- ical anthropology, but this initial theoretical foundation is also complemented by the normative reconstruction of real historical struggles against injustice and leads to an image of the normative conditions of subjective identity that remains formal. Such a methodology amounts to an anthropology of modern subjectivity that provides norms for the critique of existing social contexts. Crucially, however, the social incarnation of these norms is not relativistically contextualised in reference to specific social contexts, but sought in the real social movements and in the social pathologies as documented by critical soci- ology. In that way, the danger of a de-contextualised social critique is avoided since the formal norms receive their substantial content, for example, the spe- cific content of rights claim, only in reference to the specific social, historical situation. At the same time, however, the danger of a dilution of normativity and thus of critique, as a result of contextualism, is avoided. Instead, the social theorist can always translate the formal model into the search for the “norma- tive surplus” which, from within the immanence of social life, points towards its transformation on the basis of what in it, is unacceptable.



Is Honneth a liberal?



Again, it is this emphasis on the way in which the normative principles relate to social reality that is at the heart of the difference between Honneth’s and liberal theories of justice.



In terms of content, as noted at the outset, Honneth’s position is very close to that of Rawls and other liberal political theorists. In their great majority, all contemporary political philosophers share a similar political aim: to offer normative justifications of a social-democratic model of contemporary soci- ety. They all agree that the principle of universal equality is the fundamental foundation of modern societies. In particular, all agree that this principle is tied essentially to the possibility of individual self-respect, one of the essential primary goods whose fair distribution defines a just social order. It is inter- esting to highlight a particularly telling moment in the liberal literature: the fact that the defining reference in that field, Rawls’ theory of justice, famously emphasises the normative primacy of self-respect, making it one of the essen- tial primary goods. In a famous passage of the Theory of Justice, Rawls seemed to anticipate the German discussions on the conceptual roots of practical iden- tity: “Without self-respect, nothing may seem worth doing”.25



Once again, however, beyond the great overlaps in terms of content, the dif- ferences in the methodological approaches have crucial political-theoretical, and indeed, practical significance. And once again, the key question is how the different theories negotiate the difficult connection between normative discussions about the principles of the just social order and the reality of his- torical societies.



Rawls, and with him most other liberal political philosophers, consider the normative construction of principles of justice as a reflective reconstruction of the actual normative foundations of real Western democratic societies, or indeed, of one society in particular, American society.26 This methodol-

ogy implies that one considers contemporary Western democracies (or rather the USA) as providing an incarnation of the principles of justice that can be arrived at through “reflective equilibrium”. Indeed, this is what “reflective equilibrium” entails: that the political conception of justice only articulates in a purified manner “firmly held convictions” that are already operative.27

Consequently, one might well say that Rawls’ principles, notably the second, egalitarian one, have a utopian dimension that can lead to a radical critique of existing societies. But the radical transformation that seems to be implied in the image of a full realisation of the two principles is in fact only the trans- formation of a reality that is implicitly taken as already just, as containing the norms of its own transformation. However “radical” one might take Rawlsian liberalism to be, it is premised on the idea that social reality deep-down is already just.28



By contrast, the three “principles” corresponding to the three spheres of rec- ognition are not assumed to be substantive principles already realised in Western democracies. This is a point where the emphasis on the formality of Honneth’s concept of Sittlichkeit is crucial. The spheres of recognition rep- resent a typology of normative structures through which subjects can make sense of their social experience. And this, mainly when they experience injus- tice because according to the pragmatist principle, it is in the failure of action that reflection is possible on the normative expectations that were always already presupposed in it. The three spheres of recognition therefore can be called normative principles, and Honneth does characterise them in this way, in “The Limits of Liberalism” for example.29 But the key methodological dif- ference between substantive principles seen to be embodied in real existing societies and formal principles that structure justice claims changes their criti- cal and political status significantly.

In the first case, since the principles are seen to be already at play in exist- ing societies, the political output of political theory is to measure reality to its own standard, to approximate reality to itself. A substantive historical narrative underpins this view of modern politics, namely that contempo- rary societies are realising for good a process of social rationalisation that is synonymous with modernisation. In particular, the principle of toleration, which arose as a problem with the Reformation and whose solution was presented already by Locke in the 18th century, is on the way to being fully realised.30 Western societies, on that model, are therefore on their way towards solving the conundrum of the rational society. This is so since, according to the normative model developed in Political Liberalism, they on the one hand manage to enshrine the individual rights of “rational” subjects, whilst suc- cessfully overcoming the paradox necessarily created by this very principle. A paradox necessarily emerges from the enshrining of individual rights, since their “comprehensive doctrines” are by definition incompatible (since they are essentially individual) and thus risk making the society of such “liber- ated” individuals impossible. Rawls’ solution is specific to him, but the way of presenting the problem, as the conundrum of reconciling individual rights with the value pluralism it implies as well as with social integration, is com- mon to all liberal writers. Implied in this way of putting the problem, how- ever, is the view that Western democracies, at least normatively, are on their way to realising the truth of modern politics, which is itself nothing but the political and social image of rationality itself.31 Rawls, for example, explicitly endorses the idea of a reciprocal influence between the rationalisation of indi- viduals and the rationalisation of society.

More or less implicitly, political liberalism sees itself as the normative reflec- tion that accompanies a real social and historical trend that is synonymous with the realisation of reason and freedom. As a result, there is a tendency in political liberalism to underplay empirical injustices because the real, norma- tive achievement of these societies acts as a counter-objection to the critique of injustice. The liberal philosopher can always say: yes, these injustices exist, but look how far we’ve progressed; the norms for an abolition of these injus- tices are already at play and only the irrationality of individuals and com- munities prevents their full actualisation. This justificatory view of Western societies is all the more problematic when the model is compared to other forms of society. In that case, the toleration that reigns within is no longer maintained, and liberalism that was neutral about “comprehensive views” within becomes a comprehensive view in its own rights, sure of its good right to impose itself on others. Many suspect that it is not just a coincidence that the same word, liberalism, is used both in the esoteric, academic discourse of egalitarian, well-intentioned political theory, and in the much more sinis- ter discourses of contemporary neo-imperialist Realpolitik. Beyond the many differences that make any identification between the two impossible, what unites them is the belief in the moral superiority of Western democracies, and the structural blindness to the injustices that proliferate in them and that they impose, as real societies, on others.





By contrast, the ethics of recognition, even though it also extracts its norma- tive principles from a progressive view of modernity, and is premised on a “legitimating” vision of modernity, is not forced to adopt such justificatory attitude towards historically existing social situations. Claims of recognition along the three formal axes described by Honneth can be formulated and redeemed in any number of ways.



The decisive difference lies with the negativistic methodology. Honneth never fails to emphasise this aspect of his theory.32 This is not just a meth- odological point. The critical and political status of his theory depends on it. Methodological negativism touches first of all the normative reconstruction of modernity. Rawls’ constructivist approach quickly showed its real metal by acknowledging it was a “device of representation” in fact relying on the normative reconstruction of already existing principles. Such a method is built on the assumption that the principles it “constructs” are already implic- itly embodied in the reality it normatively describes, and that it is therefore intrinsically unable to transcend its social context. A negativistic methodol- ogy, by contrast, argues in this way: in order to know what normative prin- ciples structure social life, one looks at the claims that have historically been

raised in real claims of injustice, and in real attempts to change the social order on that basis. These claims and the struggles they inspire negatively, point to the normative presuppositions that are injured by social structures. In this case, the inscription of the principles in the real existing societies has a very different status. Yes, individuals can, under favourable circumstances, rely on these principles to make their claims. In this sense, one can say that the principles already exist. But they exist as grounds for the rejection of injustice, not as realisations of a potential for full rationality. They are possibilities of struggle and dissent in the name of an experience of injustice, not the already achieved figure of justice. On that model, one can even talk of moral prog- ress, of a certain accumulation of normative advances, without falling under the criticism just formulated. It is not (just) that these advances are empiri- cally always under threat. Rather, the negativistic method ensures that such advances are not taken to be principles that are already fully substantiated. The theory remains agnostic about the final image of justice. It can paint an ideal, conceptual end point of full recognition (when each and everyone is both fully individuated and full integrated), but it does not anchor it in a particular context. In that sense, the Kantian strand in Honneth’s theory of recognition is not to be associated with the normativism of the second Kritik, but rather with the writings on history, where Kant attempted to devise a non-metaphysical teleological account of the realisation of equality.33



The fundamental methodological difference between the ethics of recognition and liberalism as a result of Honneth’s use of a negativistic approach is well captured by a late passage in Honneth’s second rejoinder to Fraser in 2001. The recent date of that passage confirms that Honneth has in fact not budged so much on his basic critical attitude towards liberalism, and remains commit- ted to a classical, “Left-Hegelian” position:

We do not simply seek to apply what we take to be well-grounded norma- tive principles to a given social order in order to arrive at judgements about morally justified corrections or improvements. Rather, social reality must be described in a way that shows how norms and principles considered justi-

fied could already have become socially valid.34



The task of the critical theorist is not to adjudicate from above on an external social reality, on the grounds that such empirical reality is at odds with itself since the principles of justice are already in operation. Rather, the political cri- tique of a given social reality consists in uncovering potentialities for norma- tive improvement within its immanence. Social reality then is indeed at odds with itself but the tensions that plague that reality are not between its own norms and its reality, but rather between different forces within itself.



As a result, another crucial element to contrast the ethics of recognition and political liberalism is the place the critique of social pathologies takes in the overall model. For a negativistic methodology these experiences take centre stage: it is in the experiences of social suffering that the normative principles undergirding modern societies negatively appear, either in a historical sense, in the reconstruction of the normative core of modern society, or in a criti- cal sense, when it comes to uncovering new applications of those principles. The “political theory” that grows out of an approach so strongly reliant upon sociological insights looks at politics as the institutionalisation of normative demands by social groups. The political, then, is defined as the clash between normative claims and counter-claims, between justifications of domination and denunciations of domination. Such theory does not have much to say on other questions of political philosophy, for example about the structure of the state, or the paradoxes of democratic sovereignty. Its strength consists in focusing normative political philosophy back on the substantial link between politics and social life. The old critique of political philosophy from the point of view of real social suffering is thus reawakened in a valid way.

Of course, Rawls’ second principle seems to be well capable of answering fac- tually-based objections by pointing precisely to the fact that it offers a highly restrictive justification of inequalities that is clearly not met in real situations. But the point of the contrast and the objection it entails is not a crude fac- tual one. The objection consists rather in saying that the pathologies that are produced by the social-economic order prove that the very language with which social inequality is discussed in political liberalism is inadequate, as a theoretical vocabulary. Social pathologies show that there is not much sense in talking of basic rights and liberties when the subjective identities are so deeply threatened by the social structure that they can no longer function. The gap that opens up between the academic language and the reality it purports to reconstruct is too wide. In a society where socially induced pathologies are immense, the vocabulary of a distribution of inequalities makes no sense; it is the wrong grammar. It might well be formally consistent, but its formal consistency is an empty victory. The crucial issue is the sociological relevance of the normative principles.



Things are in fact even worse if one entertains the idea that the crude factual objection, which proposes that reality is not as the principles of justice say, might not be that crude after all. Is it so certain, after all, that even in the “hap- piest” periods of the Fordist consensus, in the most flourishing times of the most integrated welfare-states, the language of the distribution of social and economic inequalities was not already seriously underestimating the amount of injustice? This becomes even more acute if, as do Renault or Bernstein, we replace the critique of liberalism in its Marxist background and remember that the battle against liberalism was a battle to impose the recognition of the equal normative importance of socially caused suffering. This, however, is precisely the background of Honneth’s own position. In other words, the eth- ics of recognition carry out in the changed theoretical and empirical context of the late 20th century the critique of liberal rights and the defence of social rights. It does so by highlighting, against the letter of Marx, the normative significance of modern law. But the emphasis on the social element makes it the direct heir of the socialist critiques of liberalism.



In the end, what makes the ethics of recognition incompatible with a liberal position, despite the essential agreement on the content of the principles high- lighted, is the divergence on how to rate the impact of social inequality on subjectivity. The point here is not to oppose an alleged naïve anthropological view of socialised individuals, the “atomistic” liberal one, with a more “com- prehensive”, “intersubjectivistic” one. Honneth does not believe that this is the correct way to highlight the weaknesses in political liberalism. The point, rather, concerns the method used to articulate the normative principles. Rawls’ and other egalitarian solutions explicitly attempt to respond to the charge that

liberal liberties are only “formal”, by integrating a strong reference to social and economic equality. But the main premise of liberalism is maintained even in these egalitarian versions of liberalism, to wit, the primacy of liberty over equality, of rights of liberty over social rights. The ethics of recognition, by contrast, puts a very different emphasis on inequality, because it takes a very different view of the impact of inequality on social individuals. The primacy of rights, of negative freedoms in liberalism is coherent from a normative per- spective only if one agrees with the assumption that an unequal distribution of “social values” remains external to the subject’s identity.35 The inequality counts as an injustice only when the distribution is not fair (when, for exam- ple, in Rawls’ model, it is not “to the advantage of all”). The fact that inequal- ity is treated in quantitative terms, as something to be distributed, means that it cannot be seen as the possible source of a moral injury. The “social bases of self-respect” accordingly are themselves “distributed” together with the other primary goods, like money. On that model, it is unacceptable to be deprived of those bases, not because it actually harms the individual, but because one lacks a resource that is otherwise given to others.

The argument that inequalities must be arranged in such a way that they are “to the advantage of all” seems to be a very strong restriction to inequalities, certainly one that would apply a strong critical norm to real societies. But social and economic inequalities can be deemed to be acceptable at all only in an image of society where they don’t really impact on subjectivities. This is possible only for a liberal vision of the modern subject according to which the guaranteeing of basic rights and liberties provides the sufficient conditions for self-realisation. Social and economic status then gives some further content to a definition of self made possible by the liberties, but in a sense remains exter- nal to the subject who is first of all a free subject. On that model, social and economic inequalities are purely quantitative, and have no qualitative dimen- sion to them. They do divide society into separate groups and classes, but not in a way that is detrimental to individuals’ self-respect. For example, as long as the subject is able to take part in some “social unions”, a participation that the rights and liberties enshrined in the first principle ensure, the “union of social unions” remains possible, indeed it is even said that it can be a just social order, despite existing inequalities.36 In other words, the recognition of formal equality is sufficient to ensure self-respect or self-esteem to everyone, despite the existence of social and economic inequalities. The key critical argu- ment is not just that Rawls fails to notice the difference between self-respect and self-esteem, a distinction that appears clearly through Honneth’s spheres of recognition. The real problem in this confusion is not a theoretical failure, but its practical, political implication: by failing to distinguish clearly between self-respect and self-esteem, Rawls implicitly argues that self-respect always ensures the possibility of self-esteem. As long as one is ensured of one’s basic rights and liberties allowing one to engage in one’s socially determined activi- ties, self-esteem naturally follows. By making self-respect the first of the “pri- mary goods” and giving it at one point a negative characterisation, Rawls seemed very close to Honneth. But this impression dissipates quickly if one reads the remainder of the paragraph of The Theory of Justice dedicated to self- respect: “It normally suffices that for each person there is some association (one or more) to which (the individual) belongs and within which the activi- ties that are rational for him are publicly affirmed by others”.37 This is all that is needed for inequality to not lead to a destruction of self-esteem and thus to become acceptable. In other words, a society can be fully just even with a great amount of social contempt. All that is required is that those unfortunate enough to have missed out in the natural distribution of abilities and capaci- ties can find solace in unions of peers where their lesser achievements find their due recognition. If that is the case, the contempt shown by other “social unions” will matter little to them.

From the point of view of the ethics of recognition, such a version of the social division of labour is simply too optimistic and indeed naïve. It fails to take into account the reality of social domination that comes with the division of the social field in groups with diverging value models. It ignores the results of critical sociology and the history of social movements, about the impact of the hierarchical division of society on the members of the dominated groups.

It does not see that social domination can lead to moral injury, and that this is a major characteristic of injustice. Injustice is not just the unfair distribu- tion of goods to which all are entitled. More deeply, injustice has to do with the destruction of the moral basis of self-realisation. Unfair distribution, on that model, is unjust primarily because it undermines the sense of equality that is achieved in true recognition. And so, the very language of unfair dis- tribution is inadequate since the point of injustice is qualitative (moral), not quantitative.



Finally, this critical focus on the implicit image of society underpinning lib- eralism points to another critical perspective from the vantage point of the theory of recognition. The lack of consideration for the reality of social domi- nation leads to a pacified view of society and of the division of social labour, where the cooperation between individuals and groups is postulated and seen as unproblematic. Accordingly, Rawls argues that the underlying normative principle “implicit in the public culture of a democratic society”38 is that of a “fair system of cooperation amongst free and equal persons”. Such a concep- tion of society explicitly rejects the structuring effects of class struggle and group domination. Once again, the normative-empirical distinction could be brought forward to reject this criticism as simplistic and ill-informed. But this rejoinder would fail to heed the full force of the argument about social domination: namely, that the structural role played by social domination in all forms of society, including and especially in modern societies, renders the very distinction between the empirical and the normative suspicious, since one of the most pernicious impacts of domination consists precisely in affect- ing the normative order. In other words, the reality of social domination needs to make its impact in theory and the normative itself, lest theory reproduces, willingly or not, the structures of domination. This was, for example, one of the main criticisms of the early Honneth against Habermas’ discourse eth- ics: the failure to see the problematic anchoring of his democratic theory in social reality reverberated into the theory itself, and requested a major correc- tion. Honneth’s ethics of recognition contest any political theory that would

argue in isolation from the results of social theory. This is clearly a distant yet unmistakable faithfulness to Marx’s own critique of the liberal philosophy of his time.







Section 2. The critique of political economy



The objections to recognition theory from the point of view of economic injus- tice, the problem of “redistribution”, have been articulated most strongly by Nancy Fraser in a series of important articles published throughout the 1990s,39 and finally in her long confrontation with Honneth, in their joint publication, Redistribution or Recognition (2003). Simon Thompson’s book on “The Political Theory of Recognition” already presents a comprehensive treatment of the debate between Fraser and Honneth. Instead of attempting to offer an alter- native coverage of the exchange, I will instead focus on a number of points that seem crucial for a full understanding of the scope of Honneth’s political theory, including on the question of economic injustice.40 I use another refer- ence as guideline, namely the long article dedicated by Christopher Zurn to Honneth’s accounts of economic injustice, despite the fact that Fraser’s argu- ments provide the background reference for Zurn’s criticisms.41 The reason for this is twofold: Zurn provides a wonderful synthesis of all the arguments in this debate; and secondly, his emphasis on a key article by Honneth, the

1993 “Democracy as Reflexive Cooperation”, allows us to present some of the most striking, and insufficiently acknowledged, features of Honneth’s politi- cal theory.





Democracy as social cooperation



Zurn launches his Fraserian critique of what he sees as recognition-theory’s reductionistic stance on the economy with a reading of a crucial article pub- lished by Honneth in German in 1993, immediately after the publication The Struggle for Recognition, which was first translated and published in English in 1998: “Democracy as Reflexive Cooperation”.42 At first glance, it could seem surprising to direct a critique of Honneth’s stance on economic injustice through this text since it is concerned primarily with questions of political phi- losophy. In it, Honneth argues that Dewey’s writings on democracy provide a fruitful alternative approach to the question of the normative foundations of democracy compared with the main contemporary contenders: liberal- ism, proceduralism and republicanism. The reason why Zurn chooses this text in particular to cast new light on the “redistribution versus recognition” debate stems from the specificity of Dewey’s solution to the political problem. Dewey, wilfully pursuing the old Hegelian-Marxist intuition via new, prag- matist, philosophical means, argues that a vibrant democracy relies on a “fair and just division of labour”. Honneth’s explicit reappropriation of Dewey’s model, therefore, gives invaluable clues as to his own approach to the prob- lem of the division of labour, and thus the first insights into his approach to the relation between recognitive and economic relations.

In the introduction to the article, in order to justify the retrieval of Dewey’s democratic theory, Honneth recapitulates the problem that had guided his earlier forays into political philosophy: the key question, he argues once more, is the issue of the social foundation of political participation, the social foundations of democracy. Democracy relies on the requirement of individual participation in the process of will-formation, yet many contemporary mod- els of democracy offer insufficient or indeed inexistent conceptual analyses that explain and normatively clarify how individuals are motivated, and from which point of view, to participate in the debates over the issues concerning their community. As we saw in the previous section, this is especially the case for liberalism.





Dewey’s solution, as Honneth reconstructs it, is clearly indebted to Marx: democracy for Dewey is not so much grounded in, but is rather identified with, social cooperation. Inasmuch as individuals through their interactions in the activities of social life always already collaborate and are thus forced to deliberate amongst each other, they are already engaged implicitly in a pro- cess, which the political movement makes explicit and reflexive. Democracy, therefore, as the normative ideal of modern politics, in the end designates a certain state of society, in which social cooperation is fully developed, rather than just a set of institutions or a kind of deliberative procedure. This solution, therefore, solves the problem of individual participation, since according to it, the participation in social life is already in nuce a participation in that society’s reflexive moment of political will-formation. But it is clear that the specific problem solved with Dewey’s emphasis on social cooperation in fact points well beyond the specialised debates of political theory, and aims instead at an expansive vision of society as a whole, in other words, at the possibility of individual flourishing on the basis of healthy social relations, in which partici- pation in democratic life is only one of the dimensions of social cooperation. In other words, Honneth finds in Dewey a perfect illustration of the kind of concerns defining his vision of “social philosophy”, a type of philosophy, to quote once again the key passage from the “Pathologies of the Social” article, that does not “seek out the conditions of a correct or just social order, but instead attempts to ascertain the possibilities that a form of life entails for human self-realisation”.



The key argument justifying this identification of social cooperation and democratic politics, the argument that plays a central role in the evaluation of Honneth’s approach to economic problems, is encapsulated in the specifi- cally pragmatist sense of the notion of “reflexivity”. In his early writings, however,43 as Honneth shows, Dewey does not clearly identify the specific reflexivity that is inherent in politics. Instead, he directly identifies democratic politics and cooperative society, thus repeating the reduction of politics that is also at play in Marx. What is missing, in both the early Dewey and Marx, is a separate analysis of the political moment in its specific role and its spe- cific structure. Inspired directly by Hegel’s notion of Sittlichkeit, but failing to heed Hegel’s careful description of the State in its multiple relations to soci- ety, Dewey brings together without sufficient mediation, in a kind of Rous- seauian reprise, individual autonomy and political sovereignty. The element that brings them together is the individual’s participation in the division of labour:



Because each member of society contributes, on the basis of a division of labour, through her own activities to the maintenance of society, she rep- resents a ‘vital embodiment’ of the end of society. For that reason, she is entitled not just to a part of the freedom made socially possible; rather, as an individual she always possesses the entire sovereignty through which all jointly as a people become the sovereign bearer of power.44

Despite the serious shortcoming that the lack of mediation between society and politics represents, the decisive role played by the “division of labour” in these early writings is already highly significant. As in Hegel, the functional- ist aspect of the economic organisation and the corresponding instrumentality of economic activity from the individual point of view, are only superficially disconnected from ethical life. In fact, in both Hegel and Dewey, the divi- sion of labour is itself an essential moment of ethicality: not only indirectly, because it allows the community to reproduce itself materially, but much more directly and importantly, because the inscription of individual activity within the overall organisation of social life is the properly ethical condition for the individual’s participation in political life.

In later writings,45 Dewey acknowledges the separate moment of deliberation marking the specificity of politics by contrast with social cooperation. But this acknowledgement in no way weakens the fundamental intuition that democ- racy in the end designates a “social ideal” just as much as a specifically politi- cal one. Now, however, it is the pragmatist understanding of reflexivity that allows Dewey to maintain his key, Hegelian-Marxian, intuition.



The intersubjective and reflexive definition of truth in the pragmatist tradition is well-known. Truth-seeking procedures, notably in the sciences, are seen as extensions of those everyday procedures that we put into play when some of our basic implicit assumptions are held in check by disruptions of experience and their failure to eventuate. Truth-seeking procedures are thus reflexive processes inquiring into, and correcting, the implicit assumptions that have been proven wrong. And this process, for the pragmatists, is best engaged in via a community of inquiry. In other words, it is in the essence of scien- tific inquiry to be “reflexive cooperation”. In his later writings, Dewey simply finishes the circle that had led from society to science, back to social life. As Honneth reconstructs his argument:



in social cooperation, the intelligence of the solution to emerging problems increases to the degree to which all those involved could, without constraint and with equal rights, exchange information and introduce reflections.46

This then, leads to an idea of democratic deliberation which, as reflexive, is now relatively separate from the immanence of social life: democratic deliber- ation, like scientific debate, is that reflexive moment where the community of “inquirers” attempts to solve as one community a problem that has emerged in an area that concerns everyone.

Such an argument, however, is not yet sufficient to justify the link that is sup- posed to be maintained between democratic deliberation, in its relative auton- omy from social life, and the strong social ideal inspired by Hegel and Marx. The problem of linking politics and social life from the pragmatist perspective amounts to the following question: whilst the community of the “involved” is not problematic in scientific inquiry, it becomes the core problem in social phi- losophy. It is not difficult to see why democratic procedures can be described as reflexive deliberations over problems arising in social life. The whole ques- tion is: who is involved in those problems and in what capacity? Why should these problems arising in social life involve all members of social life, and even more pointedly, why those members in their involvement in the division of labour? The response to this question is crucial if the division of labour and the strong social ideal depicted in Dewey’s early writings are to retain their significance. It would be perfectly conceivable, for example, to hold a simi- lar, reflexive, version of democracy, for example a Habermasian one, without making the political moment rest strongly on the division of labour. In brief, it is clear how democracy can be described as reflexivity; the whole difficulty, and originality, of Dewey, is to make it a reflexive cooperation, where coop- eration is not just cooperation at the political level, but political deliberation based on social cooperation.



Once again, the missing link is provided by a pragmatist conceptual scheme:



Social action unfolds in forms of interaction whose consequences in the simple case affect only those immediately involved; but as soon as those not involved see themselves affected by the consequences of such interaction, there emerges from their perspective a need for joint control of the corre- sponding actions either by their cessation or by their promotion.47

On that model, political procedures are called for to coordinate and regulate the consequences of actions that originate at first from particular parts of soci- ety, but which can be seen in fact to affect all members of society. Politics then is truly a reflexive moment where society attempts to solve its own internal problems. The division of labour comes into play in this scheme as soon as the argument is given a normative twist and a specifically democratic version of politics is sought: for all individuals to be involved in the reflexive process of political deliberation, they must already see how they are indirectly affected by the actions in which they are not directly involved. This, the Hegelian- Marxian idea of division of labour ensures, since it shows how individual activity is essentially related to, indeed defined by, its place within the overall social organism. Put negatively: a society is not truly democratic, even if its political procedures formally are, if social agents cannot see how the actions of others relate to them, and how their actions are related to others. Once again, democratic politics is rooted in a democratic society. As a result:



only a fair and just form of a division of labour can give each individual member of society a consciousness of cooperatively contributing with all others to the realisation of common goals. It is only the experience of partici- pating, by means of an individual contribution, in the particular tasks of a group, which in its turn cooperates with all the other groups of a community through the division of labour, that can convince the single individual of the necessity of a democratic public.48

We verify here that the direct problem tackled by Honneth in the 1993 article is indeed that of “the moral foundations of democracy”, but that it in fact points to a much broader conception of society at large, to an expansive “social ideal”. Basically, for Dewey, and clearly for Honneth also, there is no sense in talking of democratic politics separate from the problem of democratic soci- ety. In other words, democracy can only be achieved via social transformation. This is exactly the same conclusion that The Struggle for Recognition reached one year earlier. In the meantime, a highly charged concept of the division of labour has been shown to be necessary: political freedom requires justice in the division of labour; through the division of labour, each individual activity is defined in its social and indeed in its potential political significance.





Critique of Honneth’s account of economic injustice



Zurn’s critique of Honneth’s reductionist stance on the question of economic injustice refers to Honneth’s double argument according to which a fair and just division of labour is the condition for a true democracy, and a condition also of individual flourishing. On the one hand, Zurn acknowledges the great insights that the return to Dewey enables Honneth to develop within contemporary political philosophy: by grounding politics in the ideal of richly articu- lated and diversified social life, he avoids the monolithic solutions of other rights-, value-, or identity-orientated political models; he emphasises the link between plurality in social life and the vibrancy of democracy; and finally he can indeed “emphasise the importance of greater economic equality for a healthy democracy in a way that competing theories do not”.49 At the same time, though, Honneth’s Deweyan solution leads to a truncated view of eco- nomic activity, precisely because of the strong link that is from now on estab- lished between economic distribution and recognition. The “consciousness of cooperatively contributing to the realisation of common goals” not only pro- vides the platform that enables individuals to take part in democratic delib- eration. It also provides recognition of the individual’s contribution to society, it gives the individual his or her social value. Honneth thus finds in Dewey, after Mead, a direct confirmation of his third sphere of recognition.50 The prob- lem, however, is that from now on Honneth seems to approach the economy the wrong way around. Because the ideal of a fair and just division of labour has provided such a powerful model for an alternative, more expansive, and basically more radical, image of politics, from now on the economy seems to be analysed by Honneth from that angle alone. Before the famous exchange with Fraser, Zurn finds in a 2001 article the explicit shift to this position:

the rules organising the distribution of material goods derive from the degree of social esteem enjoyed by social groups, in accordance with insti- tutionalised hierarchies of value, or a normative order. (. . .) Conflicts over distribution . . . are always symbolic struggles over the legitimacy of the sociocultural dispositive that determines the value of activities, attributes and contributions. (. . .) In short, it is a struggle over the cultural definition of what it is that renders an activity socially necessary and valuable.51





In this text, the culturalist reduction of the economy seems to be complete. Zurn’s other critical points all rely on the identification of this shift, and the resulting conclusion that, basically, it seems that for Honneth the economy is explained through culture. Honneth is thus accused of reducing “distributive injustices to injustices in underlying evaluative patterns”, and so to propound a naïve view of the causes of economic injustice. The latter is often not to be explained through reference to the social-cultural value system, but rather to the economic imperatives, which function to a large extent independently of status questions.52 Honneth seems to ignore the basic fact that the explana- tion of phenomena specific to the economic order ought to be in categories of instrumental, not communicative or normative, rationality, in causal, not moral terms.



Furthermore, if the theoretical analysis conflates phenomena belonging to different orders, its practical relevance is seriously in doubt as it risks advocating practical solutions that fail to address the real causes of injus- tice, or even worse, advocates solutions that in fact compound the injustice, because, for example, of negative feedback effects it is not able to take into consideration.53



Finally, Zurn seems to put the finger on what seems to be the undecided nature of the third sphere in Honneth’s construct.54 Honneth can continue to hold on to the 1993 model of reflexive cooperation only if he gives a very abstract notion of “work”, as designating any socially significant individual activ- ity. This, however, makes him incapable of distinguishing between different types of cooperative association (from bowling clubs to factory floors, as Zurn says) and their significance for allowing individuals to take part in “reflexive cooperation”. But then the theory becomes so abstract as to be empirically and practically useless when it comes to analysing real forms of injustice, notably in terms of the transformations that would be necessary in different forms of social association, to challenge the distributive patterns (bowling clubs and factory floors, for example, would be significant in very different ways for

that matter). In terms of the analysis of social injustice, Honneth can explain economic injustice in terms that are adequate to its actual economic aspect (for example, the political-economic factors explaining the reality of low wages) only if his definition of work, as that which the recognitive structure is sup- posed to reward, is so abstract as to be useless empirically and strategically. For example, to explain capital mobility, Honneth would have to accept the main political-economic reason explaining the low level of wages, as being the result of an order of recognition. In order to defend this thesis, he would then have to describe the “political-economic” factors as a relation of “recognition” between, say, labour and capital. But this would lead to such an abstract and simplistic description of the complex reality of contemporary capitalistic pro- cesses, as to be without any real value analytically, and leading once again to useless or even counter-productive conclusions when it comes to addressing the question of the redress of injustice.





Economic injustice as pathology of recognition



Against this devastating attack, the first line of defence consists in granting to Fraser and Zurn at first that recognition theory, qua social theory, is not sufficient to account for the specificity of economic action as opposed to other types of social action, but that it is extremely useful, perhaps irreplaceable, to account for the experience of economic injustice, qua social experience. This is the line taken by Emmanuel Renault in the chapter dedicated to the economic institutions of injustice in L’Expérience de l’Injustice and in other recent writ- ings.55 As Renault writes,

It is clear that, on its own, a theory of recognition is incapable of produc- ing a theory of capitalism, but it never intended to do that anyway. How- ever, by relying on theories elaborated by the sociology of work and the economic sciences, it can nevertheless engage in the analysis of the effects of recognition produced by the institutions of salaried work and the capitalist market.56



Instead of reasoning through the causes, as Fraser and Zurn propose to do, making the critique of injustice methodologically dependent on the analy- sis of the causes of injustice, Renault shows quite emphatically how fruitful recognition theory can be for a critique of economic injustice, inasmuch as it provides a powerful analytical tool to analyse the effects of contemporary economic processes on individuals and communities. On that reading, recog- nition is the most appropriate concept to use in order to describe the experi- ence of injustice as social experience.



Such a claim is easy to accept at least for what is the explicit content of the third sphere in the latest version of the model. As Honneth writes in the 2001 issue of Theory, Culture and Society on recognition:



The rules organising the distribution of material goods (notably wages— JPD) derive from the degree of social esteem enjoyed by social groups, in accordance with institutionalised hierarchies of value.57

The labour market produces specific injustices, whereby some forms of work, some statuses attached to specific professions, are not sufficiently recognised, or not recognised for their proper social value, and this injustice is reflected in the wages. An unjust scale of wages, or the absence of wage compensation (in the case of care work and house work) are directly analysable in terms of recognitive injustice. In any case, this is quite precisely how it is experienced by those who feel their wage, or the absence of a wage, represents a form of social injustice.



Following Renault, a similar argument can be also readily accepted in relation to commodities and services markets. These markets also have direct recogni- tive effects. The price that the market puts on products is a reflection of the value that society attaches to them. As Renault remarks, in Marx, for example, one of the structural conditions of exchange value is the “social validation” of use-value.58 If a product is not seen as being socially useful, it will not be exchanged. This, however, is directly linked to work: for the work of an indi-

vidual to be part of social labour and take place within the division of labour

it has to be recognised as being socially valid, as producing socially validated products. In the capitalistic system, this occurs through the exchange of the products of labour. The price of a product is therefore a more direct than indi- rect recognition of the value of that individual’s activity.



Finally, and still following Renault, the other great economic institution: the capitalist firm, can also be shown to produce injustices that need to be anal- ysed as injustices of recognition. This type of critical diagnosis of the sphere of production from the perspective of pathologies of recognition is also con- ducted by Honneth in great detail in two recent texts: in “Organised Self-Real- isation” (2004), and in an article written with Martin Hartmann: “Paradoxes of Capitalism” (2006).



Drawing on a wealth of recent sociological and psychological literature on the impact of new methods of production and management, Honneth shows how the paradoxical developments of individualisation and recognition in the neo-liberal, or post-fordist, models of economic organisation introduces new tensions and contradictions into the spheres of recognition. This is because the new mode of production, accompanied by new modes of consumption and presentation of self, exploit the subjective potentials liberated by the wel- fare period in order to increase productivity and profit. Working subjects are thereby more efficiently mobilised for the good of the company in particular and the overall system in general. Their entire idiosyncratic selves, with their specific qualities, creativity and desires are thus put in the service of produc- tion. Recognition becomes paradoxical because the instrumental exploitation of it renders it impossible, or even reverses it into a pathological force.59



Recognition, and the greater individuality that comes with it, thus becomes a “factor of production” itself.60 In a period where mass consumption has to a large extent been saturated, the new frontier for increases in productivity,

and thus for profits, was to be found in the intensification of work, which the flexibilisation of work processes and the destructuration of the firm aimed to achieve. Productivity was sought in reorganisation and new forms of management. Recognition plays a decisive part in ensuring this intensifica- tion. The greater autonomy granted to workers in the post-fordist company can be analysed in terms of recognition: at all levels of the hierarchy, many more workers are asked to involve themselves totally in the quality of the production (the quality of the product, but also the efficiency of productive processes), and are promised greater autonomy and recognition of their con- tribution in return. However, the recognition thus granted to the post-fordist workers is flawed. This leads to the emergence of new forms of suffering, of a truly alarming level.





Whether one considers the situation of individuals in the labour market in general, or in the contemporary firm, recognition theory thus appears par- ticularly well adapted for providing an analytical grid to address contempo- rary experiences of injustice linked to the economic system. In other words, despite the harsh criticisms expressed against it on the economic question, it does seem to be well placed as a critical theory, that is, as a theory sub- stantively connecting its conceptual and normative claims to social reality. In particular, by focusing on the importance of the experience of work for con- temporary individuals, as one of the main axes in their lives through which they can develop their subjectivity, or, in pathological cases, find their lives deeply affected, the theory of recognition proves particularly well placed to deal with one of the most important areas of socially induced suffering in contemporary societies. It proves this by showing first of all, in its concep- tual apparatus, how important work is normatively for modern subjectivity. And it proves this secondly, by providing an adequate theoretical grammar to describe contemporary pathologies of work.



This critical nexus: pathologies of work as specifically social pathologies, and the centrality of these pathologies in modern society—this nexus has always been at the heart of Honneth’s thinking, right from the beginning.61 With the development of the recognition paradigm, notably through the various shifts

in the interpretation of the “third sphere”, it could appear as though work as a concrete activity gradually receded into the background. The third sphere of recognition seems to give only a partial entry point into contemporary pathol- ogies of work, by focusing solely on the social-psychological, the recognition of one’s contribution to the division of labour at large. However, with his lat- est studies on the paradoxes of capitalism, Honneth has reconnected with his initial project of a “critical conception of work”. His diagnosis of the patholo- gies of the post-fordist economy in a neo-liberal regime has regained some of the diagnostic wealth of his earlier writings on work.



Zurn acknowledges at several points in his article that “Honneth is surely right about how most distributive harms are experienced by individuals”,62 but he contrasts such phenomenological accuracy with another type of social- theoretical relevance, namely the relevance of arguments that link experiences of injustice to their systemic, “political-economic” causes. As previous chap- ters have shown profusely, Honneth’s original position in social theory arises precisely from the rejection of such severing of social theory from social expe- rience. This was already one of the main points of contention with Habermas, for example. This fundamental methodological constant has several dimen- sions. It translates first of all into a series of critical arguments against func- tionalist models, targeting both the theoretical and the practical implications of those models. More positively, it leads to a series of “thick”, substantive theses of social-theoretical nature. How do these two lines unfold in the case of the analysis of economic systems? After all, those systems seem to be pre- cisely the type of social fields that demand the introduction of functionalist elements. What would an “action-theoretic” account of markets look like?





Theory of recognition and economic theory



From the perspective of Honneth’s insistence on retaining the link between experience and theory, Fraser and Zurn’s dualistic models of society seem to repeat the sociological abstraction already uncovered in Habermas. Despite their assurances to the contrary, Fraser and Zurn’s insistence on analytic dualism, on the allegedly strictly analytical distinction between social-cultural versus systemic social reproduction, threatens to end up in an ontological dualism in which economic processes would be reified. As Hon- neth remarks in response to Fraser, in her eagerness to disprove him on the issue of redistribution, she puts forward theses and implicitly accepts presup- positions that prove in the end to be counter-productive for her own pur- poses. If one insists too much on the autonomy and complexity of markets and capitalistic processes more generally, one robs oneself of the means to conduct an immanent critique of them. This is true first of all in terms of the norms of critique. If as Zurn argues following Fraser, “market imperatives simply dictate the choices”, that is, if it is assumed that the logic of economic processes is purely instrumental, the search for the maximisation of profits, then the critique of economic injustice can only be external. But then, where do the norms of critique come from; how are they justified; how can they be shown to relate to “pre-scientific praxis”, and so on? One seems to be reduced to the type of externalist critical standpoint that Critical Theory by definition rejects.



In terms of social theory, if, as Fraser insists, her dualism is only perspec- tival, how can she reject Honneth’s recognitive approach on the basis that it ignores the alleged systemic operational logic of markets? As Honneth remarks, this seems to reify the “perspectivalism” into an ontological thesis. In brief, critical theorists who want to follow Habermas too closely on these issues seem to be caught between the rock of their system-theoretical assump- tions regarding the economy, and the hard place of their normative inclina- tions, which remain indispensable for analysing social phenomena in terms of injustice.



These critical remarks, however, do not by themselves recommend recogni- tion theory as a better social theory for describing modern economic reality. Again, the question is raised: what would be an “action-theoretic”, “recogni- tion-theoretic” account of markets?



Already the argument that recognition has become a “productive factor” is not just a normative claim but also a claim that has explanatory power. If we generalise on the basis of this example, we see where the proposal is heading: a “recognition-theoretic” account of the economy would be one that would show that recognition is not sufficient to fully explain economic processes, but is itself a constitutive element in them. Again, the distinction between expres- sion and constitution is crucial here. An extreme version of recognition theory







would be fully “expressivist”.63 It would see all social institutions, including the economic ones, as more or less perfect “expressions” of recognitive rela- tions. There is no doubt that some of Honneth’s texts flirt with this position.64

According to this model, all social institutions, including the economic ones, are constrained by a normative order which precedes and transcends them. In that sense, institutions can be said to “express” relations of recognition. Against this, Renault for example shows what a “constitutive” theory of rec- ognition in relation to institutions would look like: institutions are not ( just) expressions of pre-social relations of recognition; institutions also produce their own types of recognitive relations. For Renault, this emphasis on the institutional moment was limited to the problem of the critique of injustice: specific institutions produce specific forms of injustice. However, it seems dif- ficult to hold that economic institutions produce specific forms of recogni- tion, but to limit the ontological status of these forms of recognition to that of effects. It is much more likely that the recognitive effects that institutions produce contribute, in their own way, to a certain extent, to the functioning of those institutions. This, in fact, seems to be Honneth’s position in his final rejoinder to Fraser:



I continue to assume that even structural transformations in the economic sphere are not independent of the normative expectations of those affected, but depend at least on their tacit consent.65

This was quite exactly Honneth’s alternative position to Habermas’ dualistic analysis of society and the economy in The Critique of Power. Already in his dissertation, Honneth had diagnosed the theoretical and practical difficulties that a vision of the economy as a norm-free sub-system of society harbours:

in the case of both symbolic and material reproduction the integration of the accomplishments of action takes place on the way toward the formation of normatively constructed institutions. This formation is the result of a process of communication realised in the form of understanding of struggle between social groups.66

However “systematic” the integration of individual actions might appear to be, this integration, whether in the areas of symbolic or material reproduction, always involves the intervention of institutions, which are themselves concre- tions (constitutive or expressive) of recognitive relations.



The most fundamental argument at stake here, which Honneth is forced to deal with, the one notably that makes an expressivist-recognitive theory of the economy inadequate (the analysis of the economy in pure recognitive terms), is the notion of the complexity of action integration as a result of the unpredictability and impenetrability of the nexus of unintended conse- quences. Basically, markets cannot be reasonably presumed to be organised through any wilful notion of social action. This acknowledgement of the inde- pendent, system-like behaviour of economic processes is precisely at the heart of Habermas’ hypothesis of a decoupling of subsystems from the lifeworld:



Survival imperatives require a junctional integration of the lifeworld, which reaches right through the symbolic structures of the lifeworld and therefore cannot be grasped without further ado from the perspective of participants.67

The “invisible hand” of market mechanisms is too “hidden” to be made sense of by the participants, and yet it does allow an integration of individual actions that would otherwise be impossible. As a result, two types of action integration, one symbolic, the other material, one through communication, the other via non-linguistic steering media, must be postulated for modern societies.



As we know, Honneth’s fundamental intuition in social theory is one he shares with Hans Joas, and the one that has been, from his very first texts, the main inspiration behind his critical and constructive work: a thorough, unashamed, action-theoretic stance, rejecting all functionalist and systemic arguments. How can such a stance deal with the problem of the complexity of modern society, which makes a communicative approach to the complex- ity and apparent functional independence of economic systems untenable? Is an action-theoretical stance in social theory irremediably condemned to committing a basic social-theoretic fallacy, and to propounding embarrassing “empirical distortions”? Joas has shown that this serious objection is based on a misunderstanding about the scope and meaning of an action-theoretical stance in social theory: the latter does not deny the existence of unintended, unplanned consequences of individual action. Rather, it only refuses to gener- alise the consequence drawn from them to the analysis of society as a whole, and maintains that social action, qua action, remains both theoretically rel- evant, and indeed necessary for an adequate approach to social movements and democratic theory.68 An action-theoretic stance in social theory does not deny unintended consequences of action; it denies that social action remains foreign to, and powerless in front of, them. Quoting Charles Taylor, Joas insists that:



Making intelligible ‘in terms of action’ means the attempt to relate in a trans- parent way all the unplanned ‘systems’ of consequences of actions to the real actions of real actors. ‘It is certainly not the case that all patterns stem from conscious action, but all patterns have to be made intelligible in relation to conscious action’.69

And in support of his claim, Joas quotes precisely Dewey’s theory of the division of labour as a typical sophisticated version of social theory integrating the notion of unintended consequences without denying the







possibility of social action, without recourse to functionalism or methodologi- cal individualism.70



Honneth argues along similar lines in his final discussion of Habermas in Cri- tique of Power, but already interprets the action-theoretic approach in a “strug- gle for recognition” sense, even before that model has been developed. His argument starts in the negative. What Habermas says of material reproduc- tion is in fact already true of communicative action:



the cultural integration of social groups takes place through an entire com- plex of communicative actions which are not able to be surveyed as such by members of groups.71

If the impossibility of actively coordinating individual action was the reason behind a system-theoretical approach to the economy, the same would have to abide for culture and social integration, as they too are unintended out- comes of communicative processes. But with the distinctions made explicit by Joas, one does not have to bite the system-theoretical bullet: it is one thing to acknowledge the impossibility of a fully intended functional coordination of action, another to exclude all normative dimensions from the mechanisms of action coordination. Rather, with Honneth, an alternative image of society can be presented, one that acknowledges its system-like appearance, on account of its complexity, but refuses to radically separate domains of action, and therefore finds a normative component in all of them. This alternative image, then, is one where indeed there are functional types of action coordination, but where, also, relations of recognition, and notably relations of power, play a decisive, “constitutive” role. We can see why the Dewey article provided such an excellent entry point for assessing Honneth’s theory of economics. With Dewey’s social-democratic solution to the problem of “unintended con- sequences” (through the ideal of a division of labour where each can see his or her actions affecting and being affected by those of others), Honneth had in fact given a preliminary, action-theoretic, intersubjectivistic response to func- tionalist reservations, and not just an answer to a question of strict political theory.





As early as 1988, Honneth saw very clearly what this solution through a “thick” interpretation of the division of labour entails. Already then he had given significant indications about the relationship between the system-like dimension of action coordination, and the communicative-normative (later: recognitive) dimension:



both spheres of reproduction require mechanisms that so unite the particu- lar processes of communication or cooperation in a complex that (. . .) they are able to fulfil the corresponding functions of symbolic reproduction or material reproduction. In both cases, mechanisms of this kind represent institutions in which the respective accomplishments of action are norma- tively institutionalised, that is, under the constraint of the action orienta- tions of subjects that are stored up in the lifeworld, while their execution is sanctioned by the degree of autonomy of a society found in democratic agreements or under authoritatively bound orders.72

The institutional moment that Honneth refers to in this passage is the one through which recognition intervenes constitutively in economic action. Rec- ognition here means normatively regulated social interactions. For him, these interactions are always also asymmetrical because they are based on a specific balance of power between the groups.

Honneth here distinguishes two such moments of normative regulation. First, there can be a coordination of economic actions (of the “accomplishments of action”) only under the constraint of institutionalised “mechanisms” that reflect the state of group interactions (in fact, of their conflict, since power is unequally shared) at a given time. Accordingly, pure, strategic, atomistic indi- vidual action, the aggregate of which, according to the neo-classical model, constitutes the economic system, is pure abstraction. Instead the social-philo- sophical insight nourished by the intersubjectivistic premise and a commu- nicative approach to society insists on the fact that economic actions have an irreducibly “cultural” dimension, if by that is meant, as the Critique of Power argues, the symbolic group-specific filtering of social action, and more pre- cisely, the socially specific filtering of a given state of the division of labour. In clear terms, economic processes, as social realities, are always partly



“constituted” by the interactions of the groups in presence, because the latter act on the basis of their respective cultural worlds as well as within the frame- work constituted by the power relations existing between those worlds. The institutional dimension that unavoidably frames economic action qua action, as it concretises the asymmetrical relations of the different social groups at a given time, always introduces a normative dimension into it.



Secondly, the “execution” of the “accomplishments of action”, in other words, the end-result of action coordination, the overall economic action as it actually takes place at the level of society, is subject to a second normative “control”: the reflexive level of politics, in which group struggle, in the case of demo- cratic politics, finds a second, more reflexive, institutionalised expression. In clear terms: there is always a political element in political economy.



On account of these two dimensions, it is an abstraction descriptively (social- theoretically) and a mistake practically, to evacuate from the analysis of eco- nomic action the asymmetrical relations of power between groups, which the theory of recognition reframes as struggle for recognition.



This solution, where recognition is co-constitutive of economic action, is basi- cally the solution that Honneth brings again into the debate with Fraser, when he asserts that “even structural transformations in the economic sphere are not independent of the normative expectations of those affected, but depend at least on their tacit consent”. This is a fascinating aspect of the final page of his second rejoinder to Fraser on the issue of redistribution.



There is little sense in merely appealing to the importance of capitalist imperatives without considering how changes in normative expectations and action routines have paved the way for social negotiations about the scopes of these imperatives.73

Even in the case of economic processes, Honneth reflects, some basic relations of recognition are necessary for the system to function at all, even as sys- tem. Even economic systems function on the basis of fundamental normative agreements between the different social agents; and these agreements, like all other, are only fragile positions of equilibrium that can always be reopened for challenge through politics. To give just one small example, it is in fact not written in the book of nature how much a business should return on invest- ment for the market to acknowledge it as “profitable” and for the financial institutions to rate it accordingly, with all the direct implications attached to such ratings. The current accepted rate (at 15 or even 20%), which time and again has proven to be unsustainable for normal business operations, has become the norm in the current financial world only because of the gradual might acquired by shareholders and capital over managers and workers.74



Honneth’s approach to economics is highly original in social theory today, especially given the almost unquestioned acceptance of the purely systemic nature of markets in post-Habermasian critical theory.75 In fact, his intuition can find strong support in contemporary economic theory, not in the hege- monic neo-classical models, but in some of its most developed alternatives. These “heterodox” schools take precisely as one of their main presuppositions the intuition tentatively put forward by Honneth. The main school of eco- nomic thought that comes to mind is American institutionalism, the approach to economics that represented the mainstream at the time when economic theory was institutionalised in the United States.76 Another major “hetero- dox” economic school that emphasised the role of institutional mediations in

economic processes, and thus attempted to highlight the potentials for social negotiation in the economic processes themselves, is regulation theory. Whilst there are ways of reading regulation theory as being very close to Honneth, despite its declared functionalism, the overlap becomes incontrovertible with one of its most significant spin-offs: the school of conventions developed in France by writers like Michel Aglietta, one of the main theorists in regulation theory in the 1970s, who heavily influenced Boltanski’s and Chiapello’s key work of new sociology, The New Spirit of Capitalism. Currently, a major con- stellation is emerging between economic theorists, social theorists and other social scientists.77 What unites all these different strands is precisely the idea so clearly articulated by Honneth that the most influential institution of con- temporary societies, the market, can only work because it is embedded in the other networks of society, and relies on the socially constituted norms, beliefs and values of the individuals it throws in its mix.







Section 3. The politics of recognition as identity politics?



The last major concern regarding Honneth’s theory of recognition is his pro- posal to ground a normative theory, indeed a theory of normativity, that is to say, a theory specifying the content, logic and efficacy of norms, on a thick theory of subjective identity, informed by a set of theses stemming from dis- ciplines outside of philosophy. At its most basic, this concern simply relates to the image of the subject that arises from Honneth’s take on Hegelian recognition. For post-structuralists especially, it is too antagonistic (Oliver), or too rigid (Alexander Düttman). A similar concern is shared by authors like Patchen Markell who operate within an Arendtian paradigm. For all these writers, the images of the subject and of intersubjective relations drawn by Honneth deny the structural contingency and frailty that are constitutive of them.



For readers located in Honneth’s own camp, that of contemporary Critical Theory, the concerns relate more to the epistemology and methodology of a critical theory project. Writers like Nancy Fraser, Maeve Cooke, Nikolas Kompridis, Andreas Kalyvas, or Christopher Zurn criticise the very attempt of grounding critical claims on the reconstruction of the conditions of sub- jective identity.78 This repeats, all other things being equal, Habermas’ own problematic stance. Honneth’s method is thus said to remain “foundational- ist”, badly universalist, awkwardly fallibilist, teleological and perfectionist, and so on.79

One specific aspect of this general unease regarding Honneth’s reliance on a thick theory of subjective identity relates to the political dimension of the problem. For all the critics of Honneth listed above, the conceptual, method- ological and epistemological difficulties inherent in his theory of the subject come to a head in the negative political implications of that proposal. For most of them, there is something deeply disturbing, truly pernicious about it. The suspicion is that a foundationalist grounding of critique in a thick theory of the subject that overly dramatises the structural dependency of the subject on acknowledgement by others, leads to a reductive, or indeed authoritarian and reifying conception of politics, one that fails the methodological, normative and conceptual requirements of a genuinely emancipatory politics. Of course, Honneth’s intentions are not in question. Rather, as always in these debates, the theory is evaluated against a series of criteria that are deemed neces- sary for a contemporary theory of politics focused on emancipation. When that theory is shown to fail to fulfil one or several of these taken for granted criteria, the conclusion follows that, despite all good intentions, the theorist in fact fails the anti-authoritarian test.



Let us mention only two examples of this type of critique. In both cases, the strategy is similar: first a conceptual critique of Honneth’s notions of recogni- tion and identity; followed by the analysis of the disastrous consequences that follow from that ill-conceived notion.



The first typical example is Patchen Markell’s Arendtian criticism.80 Markell formulates it in the terms of agency and non-sovereignty. The theory of recog- nition, by “misrecognising” the essential open-endedness, “frailty” and con- tingency of human interactions, plays into the hands of the very structures it purports to combat, that is, structures of injustice which originate precisely in that refusal to acknowledge the frailty and contingency of human action, that is to say, the impossibility of full control, of sovereignty. This general argu- ment is specified in terms of the relationship between identity and recogni- tion. Here, Markell articulates in the most potent form a concern that is shared by most authors sceptical of Honneth’s take on recognition, whether or not they operate within an Arendtian paradigm: by making subjective agency dependent on the conferral of recognition by outside powers, Honneth does not see that he is in fact reproducing the very logic of authoritarian schemes which operate precisely by claiming and exercising a monopoly of recogni- tive power, of normative authority, over the dominated.81 Recognition then,

despite all good intentions, must be said to be “complicit with injustice”82 inasmuch as it aims for the same type of sovereign domination that structures unequal social orders.



A formally similar criticism can be found in Düttmann, a good representative of a critique of Honneth from a deconstructive angle. Again, the critique starts with the rejection of the concept of recognition. The latter is said once again to “misrecognise” the truth of recognition, as it “unifies, normalises and disci- plines” what is indeterminate, eventful and of the order of the “in-between”. As a result of this, Honneth, like the other great theorists of recognition, is accused of implicitly relying on institutions that secure this normalising of identity: the state, the institutions and the police.83

By contrast with post-structuralist and Arendtian critics, other critical theo- rists who are closer to Honneth in terms of the references they share tend to put more weight on the restrictions that the focus on recognition imposes to the critical project itself. However, even they are concerned with the political implications of his version of critical theory. The classical objection is of course that demands for recognition and many movements that use the vocabulary of recognition are, as a matter of fact, less than progressive. But there are other concerns. Zurn, for example, articulates most eloquently the Fraserian con- cern with the counter-productive effects of a critique in terms of recognition: according to him, its weak explanatory footing may well lead to advocating measures that would in the end be detrimental for those the theorist attempts to help. For Kompridis, Fraser and Cooke, the unconvincing grounding of critique on recognition, with its inherent remnants of foundationalism, tele- ologism and perfectionism, is a methodological mistake that can lead to “sec- tarianism”, that is, the risk that the theorist unjustifiably favours a form of life over others and restricts the range of possible future options.







Of course each of the authors mentioned here deserves particular attention and each of the criticisms they develop could be the object of a specific response. There is not sufficient room for this here. Moreover, it is possible to make a number of general remarks in defence of Honneth’s proposal to ground critique on “the conditions of positive self-relation”, and also to defend the political implications of this proposal.



Even if each of the authors mentioned above expresses specific concerns about Honneth’s model, it is not inaccurate to claim that, in general terms, most of his critics share a similar disquiet with his attempt to ground political claims in a thick theory of identity, basically, with his way of grounding politics on anthropological claims.



The first problem with some of these criticisms is that they are not self-consis- tent. Whilst some of them berate Honneth for grounding normative critique on a thick model of subjectivity, that same criticism does not seem to apply to them. Sometimes, the critique of Honneth’s reliance on a thick conception of identity is followed in the next page by an alternative conception, which is often just as thick, but often also far less well grounded in extra-philosophical sciences.



A more serious problem is that in many cases, the meaning of identity and the exact status of the Honneth’s anthropological line of argument are not approached with complete accuracy or in full fairness. The criticisms often do not do justice to his specific project.



The image of recognition and of a politics of recognition most of the critics listed above rely on is the following: an individual or a group already has a formed identity (gay, African-American, woman), finds it is unrecognised or ill-recognised in the existing social-cultural order, and therefore makes demands on the broader social world to enforce the recognition of the valu- able aspect of that identity.84 On that reading, the “politics of recognition” relates to the collective actions that arise when specific social groups want

to argue for the validity of claims relating to their particular identity. This is problematic because it reifies identity in a number of different ways: by assuming that identity is fixed and monolithic; by failing to perceive the per- niciousness of identity conferred by dominating groups; and by projecting an ideal of full identity-recognition that smothers creativity and contingency.



The trouble is that such a vision of recognition as recognition of identity is quite far from Honneth’s own model. In Honneth, recognition is not recogni- tion of an already formed identity, of identity as “fait accompli”, as Markell puts it, or as “presupposition” and “result”, as Düttman puts it, but rather a dynamic, indeed an open-ended process that is the condition of identity. The emphasis is not on the damage to a subject or a group when their pre-existing identity is not acknowledged, nor on an alleged end state of fixed identities all recognising and thus stultifying each other. Rather the emphasis in Hon- neth’s model is on the utter dependency of subjects and groups on recognitive relations for their very identity to be at all possible. The difference is quite significant: on the first model recognition is conceived as a good that causes an injustice when absent; an identity has not received its due. On the second model, recognition is not primarily a good, but a quasi-transcendental condi- tion of subjectivity itself.



Indeed, Honneth only rarely uses the word “identity”. Instead he says “posi- tive self-relation”, “self-realisation” or “integrity”. The central concepts in his model do not focus solely or primarily on self-ascription and the ascription of difference by others and by comparison with others, as the concept of identity implies. The moments of ascription and self-ascription are of course present and are undeniably important.85 But they do not exhaust the depth of Hon- neth’s normative approach to identity through recognition. The emphasis is different; the exact normative meaning and import of the key concepts he employs apart from identity are different from those of identity.

The main factor explaining the widespread misreading of Honneth, prob- ably relates to the different contexts of his reception. In the English-speaking world, especially in North America, the reference to the concept of recognition is always predetermined by the debates in political theory around multicul- turalism and Charles Taylor’s decisive intervention in them. As a result of this, most of the time Honneth is criticised in the same breath as Taylor, with- out acknowledgment of the key differences between their two projects.86 This confusion between the two positions is especially striking when the concept of recognition is criticised for its reliance on a misguided conception of iden- tity, and its relation to politics. Undeniably, as was noted in chapter 10, the two positions have a lot in common, at least at a superficial level. The idea that recognition answers a vital human need, and that demands for recognition in the modern world can be of three different kinds, those ideas are shared by Taylor and Honneth. But the key inspirations at the heart of their respective models, in particular, their diverging readings of Hegel, make their models also very different.



Taylor reads Hegel as one of the great philosophers to have seen the central, constitutive place of the phenomenon of expression in human affairs. There can be symbolic meaning for human beings, that is, a truly human world beyond the strictures of first nature, only because human beings are able to develop means of expression beyond mere functional processes and utilitar- ian concerns. Through expression, communities but also individual subjects are able to articulate who they truly are, develop a true, authentic identity. Identity in Taylor is strongly connected with authenticity, to the point of being almost synonymous with it.87 Recognition is required in that model because without it, a subject (individual or collective) cannot be who it truly is. There seems to be some justification in arguing that in this model, despite

his references to intersubjective dependency, Taylor does conceptualise identity as pre-existing and recognition as re-cognition, as confirmation ex post of an already existing instance.



Honneth on the other hand, uses Hegel as the first thinker (in time and in importance) to have developed the systematic consequences arising from the intersubjective dependency of the subject. We find ourselves here at one of the points where a minimal acknowledgement of the specific genealogy of Honneth’s model plays a directly theoretical role. Honneth’s theory of recog- nition emerges not from the expressivist and hermeneutic traditions, in which the question of dialogue between existing symbolic worlds is the decisive one. It emerges from the post-Kantian and post-Hegelian traditions, in which the core problem is that of the transformation of transcendental questions in a “detranscandantalised”, that is historicised and “socialised”, context. In that second model, the intersubjective, indeed the social and historical depen- dency of the subject is much more radical, as it were. As a result, the norma- tive import of recognition is very different: recognition is no longer a good that is due to an already existing instance with normatively justified claims; it is the condition of normative life itself, the condition for the subject to exist at all, if by subject we understand a human being who can lead a minimally human life, a life, that is, where she or he can be minimally subject of his or her actions. Recognition is not just a good that is due, it becomes the condition of moral life itself.

The argument is decidedly transcendental in spirit, if not in form. The core conceptual arguments are backed by psychological and “philosophical- anthropological” arguments (not just for Honneth, but also for all the major German authors who inspired his shift to recognition, foremost among them, Habermas and Tugendhat), but such a methodological move is a necessary one in a “detranscendentalised” context. It is a mistake therefore to blame Honneth for ignoring one of the alleged golden rules of contemporary practi- cal philosophy, namely the avoidance of ethical preferences, of perfectionist arguments, as Fraser most spectacularly does, echoing many other North- American critics.88 “Self-realisation”, “integrity”, “positive self-relations” do









not have the same theoretical status, are not placed at the same level, as other “primary goods”. To put it as starkly as possible, we could almost say that they are not goods at all, but conditions of a life in which something like pri- mary goods might be possible. They are formal conditions of any good life. It is therefore misplaced to accuse Honneth of naively defining the just through the good, or of putting the latter above the former, with much heavy weather then being easily made about him not having the means to justify his own chosen good. Integrity through recognition is not a good at all, but the condi- tion for any conception of the good life. Yes, the argument is psychological or “philosophical-anthropological”, in other words, “post-transcendental”. But strictly speaking, it is not an “ethical” argument, rather it is an argument about the condition of possibility of anything ethical.



The ironic outcome of such misunderstandings is that often the critics of Hon- neth propound theses that are in fact very close to his own. This is especially the case with all the writers who oppose the frailty and contingency of human action to his concept of identity, or emphasise the unstable, open aspects of struggles for recognition against what they perceive as Honneth’s overly rigid model of it. For example, when Oliver characterises oppression as “denying the oppressed access to internal life”,89 she is pointing quite exactly to what lies at the heart of Honneth’s concern: precisely, the conditions of possibility for such an internal life, and thus, negatively, the damage done to individuals when these conditions are not met. When this is then turned into a positive prescription—“the oppressed must learn to be actional and create their own meaning”90—we have a good formulation of the programme behind Hon- neth’s politics of recognition.

The important difference between Honneth and his critics on this point is that often the latter do not seem to find that the social obstacles that prevent the oppressed from “creating their own meaning” are worthy of study by theory. Systems of domination and injustice are pointed to and represent an impor- tant reference point in the general political-theoretical discussions, but they are not seriously included in the theoretical effort. As a result, the language









of these critics, whether they are of liberal, poststructuralist or other convic- tion, is often couched in terms of “ought” and “must”. In these accounts, it seems that theory can content itself with the description of the ideal of a dem- ocratic state or a democratic society without having to worry about the social conditions in which the normative prescriptions will take place. Honneth’s approach to politics, by contrast, in good “Left-Hegelian” fashion, hones pre- cisely on the issue of the social conditions that make political participation possible or impossible. In other words, from the perspective of Honneth’s approach to politics, most of the models presented as better alternatives lack a serious engagement with the problem of social domination. From that per- spective, it is simply too easy to conduct political theory separate from a seri- ous engagement with critical sociology.



In fact, the real problem with Honneth’s very distinctive “political theory of recognition” might be not so much that there is, as it were, “too much iden- tity” in it, but perhaps not enough.91 To say this is not meant to suggest that we should return to a Taylorian type of politics of recognition. The ambigu- ity in Honneth’s use of notions such as “integrity”, “positive self-relation”, or even “self-realisation”, is that nowhere does he give a systematic account of the way in which the different features of subjective identity come together to form what is called “identity”. This concept, however, would seem quite important in his overall approach, as a diagnostic, or clinical, concept that would be essential for the political aspect of his theory.



Honneth’s original approach to contemporary political philosophy consists in questioning models, notably political liberalism, that operate without ref- erence to the social.92 If indeed the special perspective granted by a “social philosophy” approach to political issues consists in defining the normative stance from the perspective of “pathologies”, that is, individual and collective

forms of ill-being, then the concept of identity seems irreplaceable because







many “social pathologies” are simply to be described as pathologies affecting the identity of individuals and/or groups. This is what is meant by saying that identity can also be a clinical or diagnostic concept: when social forms are pathogenic, they make people suffer in their identity. When they are extremely pathological, social forms simply damage individuals and commu- nities by destroying their identity. Some highly dysfunctional workplaces can bring people to the brink of madness; the alienation suffered by discriminated groups can lead to serious damage to their “selves”, and so on.



Identity in this sense is thus a descriptive, “ontological”, and, if it is used for critical purposes, a clinical concept. It designates the psychological reality underneath the cultural connotation of identity, whereas the cultural take on identity is usually the primary one in contemporary political theory. Such a concept of identity is therefore different from the concept of identity that is dis- cussed when Honneth is critiqued in the same breath as Taylor (even though Taylor also grounds his politics of recognition in this psychological meaning of identity). As can be seen, the normativity inherent in identity taken in this psychological, one is tempted to say, deep-psychological, sense, is different from the one of assumed in most discussions on identity, especially in debates on multiculturalism. Taken in this sense, identity is not just an ontological but also a normative notion because it designates the necessary conditions of indi- vidual well-being, not just a rightful claim, a “good” amongst others.93 Once again, it is the condition for anything like a good life, and so for anything like a “good” to make sense. Or even, given that for Honneth identity is very close to integrity, the psychological concept of identity is in fact synonymous with the good life itself.

What then is political about identity taken as a psychological concept? And what is the problem with Honneth’s approach to this? As said earlier, nowhere does Honneth systematically describe in what way “integrity”, “self-realisa- tion” differ from “identity”. In particular, he does not show in sufficient detail how the three spheres of recognition, which designate the three conditions of self-realisation, are brought together to form a subjective identity (a specific,









individual type of self-realisation), and what is specifically political about this. The only theoretical aspect Honneth discusses in this regard concerns the possibility that different normative claims, based on different types of rec- ognition, might clash; for example, claims raised from within the intimacy of the family versus “legal” claims based on the demand for self-respect. But he is not interested in what seems to be a central problem for any politics of recognition namely, what is political about identity, even when it is mainly understood as a psychological, or rather, as a social-psychological, rather than as a cultural, concept?



The grounding of the theory of recognition in Hegel and Mead suggests an obvious answer to this question. In both cases, it is the reference to the broad- est horizon of what we could call a “generalised” community that completes the process of subjective formation. For both philosophers, it is only through this reference that the different features of the self can be brought together. In Hegel, the dialectic of the self brings together the different ontological and normative features of “being a subject” (a family member, a moral subject, a “bourgeois”, that is, a participant in the division of labour, and so on), when the socialised self is made to realise that her activities, even when they occur in local spheres, always already entail a “universal” element. True singularity for Hegel is the unity of particularity and universality: fully developed iden- tity is the identity of a subject who has enlarged her perspective beyond the narrow circle of the co-participants directly concerned by his or her actions, and indeed beyond self-interest. Indeed, as we saw in the second section of this chapter, this is quite precisely the Hegelian model of democracy that Dewey reformulated from his pragmatist perspective. In Mead, the dialectic of the “I” and the “Me” forces the self to consider itself not just from the exist- ing established social perspective, that is, from the perspective of an existing “generalised other”, but also from the perspective of the generalised other of a “better community” (Habermas), that is, the community of all those truly affected by social action.



In all these (Hegelian) models, what completes the formation of identity is in fact the political moment, that is, the moment where the “universal”, the community takes a reflexive stance upon itself. For all the thinkers mentioned, the self becomes a concrete self when it oversteps its egocentric boundar- ies and somehow takes the universal perspective upon itself and its social






world. It seems that the logic of Honneth’s argument and the core refer- ences that underpin his thinking should have taken him to that conclusion. In that understanding, the politics of recognition does not refer so much to the grounding of politics in normative claims based on identity, but rather to the fact that subjective identity is completed in politics. It is not so much that politics depends on identity, but rather identity on politics.94



From the perspective of a critical theory programme, it is important to develop the argument that subjective identity is “completed”, as it were, in the politi- cal moment, first of all because it grants the notion of identity a clinical, critical dimension that is different from the notion of identity implied in mainstream debates on multiculturalism.95 Basically, this insight into the political moment of subjective formation makes it possible to develop a conceptual frame in which individual pathologies, as documented by clinical psychology, can be used for the purpose of social critique inasmuch as the latter also addresses pathologies of the political.96 In particular this approach puts the emphasis on the fact that the problem of the obstacle to political participation is not just a psychological, but indeed a substantially political problem. If one takes seriously the approach to politics via the social conditions of justice, then one must also take seriously the problem of social arrangements that impact on subjectivities in such a way as to make politics impossible for entire groups of individuals. And indeed, the overcoming of such social conditions becomes a

fundamental condition of democracy, in other words, one of, if not the, cen-











This is a serious issue for contemporary political phi- losophy which seems to restrict its tasks mostly to the analysis of the norms of democratic politics, and rarely deigns to lower its theoretical gaze to take into account the actual exclusion from politics of large masses of individuals in real-existing “liberal” democracies.



Moreover, the idea that subjective identity “completes” itself in politics provides a useful analytical tool to observe power struggles between social groups. As we have seen, the idea of society as a complex order of recognition entails in particular a dynamic vision of society as the fragile, always contest- able result of a normative consensus amongst the different groups making up the social. The consensus can always be renegotiated because it is never a symmetrical, perfectly equitable one. Some groups exercise more power than others and consequently, the normative agreement only reflects a particular stage in the power relations. As we saw, this gives the normative agreement at the core of the social order an ambiguous status for the dominated groups. The normative nature of the “moral contract” underpinning the social order means that this contract can be reopened at any time, but it also makes their domination and situation of injustice more liveable for them because it pro- vides a basic line of justification for that order. From the perspective of the groups exercising social domination, the fact that the fundamental structure of social interactions can be described in normative terms grants them the opportunity to justify their domination.

This is where the political moment comes in, if by that we understand, follow- ing Hegel, Mead and Dewey, specific conceptions of communal life as univer- sal reflexive moments immanent within social life organised in its different spheres. To say that the process of identity formation is completed in politics then points to the process of idealisation and normative justification through which individuals from specific social groups, and those social groups them- selves as collectives, attempt to complete and perfect the image they have of themselves in coherent, morally acceptable ways, by recourse to a certain image of the social order, in the form of an ideal communal project. Dominating groups and the individuals belonging to them tell stories that explain and







justify their domination as natural and just. Indeed, the attempt to present social domination as naturally grounded has a paradoxical implication for subjective and social narratives: often, it leads to the denial of the necessity to give a narrative, the rejection of the need to make explicit the ideal, societal project underpinning the existing order. Nothing is more political than an attitude that pretends to be detached from politics, if the individual and the group with such an attitude benefit from the existing social organisation. This might be one of the dimensions of the massive depoliticisation witnessed in Western democracies. Depoliticisation can be analysed as an implicit political project on the part of those with privileges who refuse to change anything to the existing order. In Hegelian terms, a perfectly particularistic identity might often be one that is not just simply not-universal, but one that has actively refused the passage to the universal, and so made the universal choice to refuse to universalise (that is, justify and explain). Things are very different for dominated groups. When they are not able to challenge their domination, dominated individuals need to tell their own stories in order to explain their disadvantage and make the experience of injustice liveable. This is basically what Honneth has in mind with the notion of cultural action. But when a struggle for recognition flares up, it is because dominated groups have been able to project a different image of the community, one in which domination is reduced, equality better realised. In this case, collective and subjective iden- tities have been able to expand through psychological and cultural processes that are political through and through since they revolve around an image of the community.