The Enlightenment project aimed to found a new political order based on universal principles of justice. According to Kant, it summons citizens to “emerge” from intellectual immaturity, defined as “the inability to use one’s own understanding without the guidance of another.” For Kant, reason could be considered a universal tool, “the use [of which] anyone may make as a man of learning,” that would free mankind from injustice (Kant 1784, emphasis added). Yet today, modern reason with its claims for rationality and universalism is no longer seen as universal, but as solely representing—at best—the values of one particular culture (Gray 1995, 123–125; Habermas 1992, 8; Taylor 1995, 27): multiculturalism is a fact. It has recently become sadly commonplace to speak of our contemporary era as one of post-truth, where what is presented as truth by some is attacked by others as a specific cultural opinion at best, or as fake news in the worst of cases.

Such a criticism undermines the very foundation of the modern polis. In this limited sense, we are now in a postmodern era. Indeed, once we admit a multiplicity of worldviews, moral guiding norms also become multivocal. As a result, the articulation of an ethical standpoint “that does not shy away from knocking down the ‘parish walls’” (Benhabib 1992, 228) seems to be beyond reach. Paradoxically, the criticism of the modern paradigm of reason seems to end in the impossibility of any constructive social criticism. The treatment of this problem in the literature is twofold.

In the first strand of thought, that adopts the “multicultural vision,”¹ modern “universal reason” and rationality are criticized for imposing a coercive “norm of intelligibility” (Butler 1991, 162; 2006, 23–24), for forcing “behavioral norms of respectability,” for ranking people into hierarchical categories (Young 1990, 122–155), and for serving as an exclusionary tool that proceeds through categorization and essentialism (Mouffe 2000, 29–30). This strand of thought also defends the abandonment of an external ethical point altogether, since it is always a particular viewpoint falsely posited as external and objective (Fish 1997, 2258). Within such a perspective, the ethical standpoint, if it exists, remains embedded within each culture, that is it is never external to the specific community it guides but solely relative to it (Walzer 1988, 20, 33–34; Taylor 1991, 96). This form of cultural relativism is comparable to the current mistrust of science which, for the sake of my argument, is considered just another viewpoint (Latour 2004, 227; Habermas 1992, 49). However, while this position claims to be more inclusive, it thwarts social criticism along the lines of a culturally transversal inquiry that depends on the possibility of abstracting contextual occurrences into broader patterns of injustice (Fraser and Nicholson 1989).

The second strand of thought aims to accommodate the multicultural predicament without abandoning some form of universal rationality (Habermas 1998; Rawls 2005). Habermas specifically tackles postmodernism which he defines as “the radical criticism of reason” (1987, 86). He proposes instead to contextualize the modern concept (Habermas 1992, 142), noting that the move from traditional to post-metaphysical philosophy means that, instead of reason, contemporary philosophical currents “converge towards the point of a theory of rationality” (Habermas 1984a, 2, emphasis in the original; see also Schnädelbach 1998, 5–6). Besides instrumental reason, he also tries to present as evidence what he defines as “communicative rationality” (Habermas 1992, 50, 139; Cooke 2003, 283) and its capacity to achieve a context-transcending truth, yet immanent in the inter-subjective experience and in specific social interactions (Habermas 1992, 139–142). Habermas indeed argues that intercultural encounters can lead to fused horizons and common truth (Habermas 1992, 138; Cooke 2011, 480), thus directly challenging the frequent claims of multicultural thinkers for the incommensurability of cultures.²

Both Habermas and Rawls offer political models that limit the scope of universality to the use of universally acceptable reasons in the political domain (Habermas 1998; Rawls 2005), thus reducing reason to its procedural capacity of giving justifications, for the sake of reaching a consensus.³ Yet both Rawls and Habermas are in turn criticized for not giving voice to the plurality of worldviews in the public sphere and for imposing too strong a norm on the political. These models are all the more inadequate as the current trend seems to lean toward a “post-secular” era, in which both cultural and religious groups tend to re-politicize, or politicize their claims (Eisenstadt 2008; Baumeister 2011, 222).⁴

In his analysis of “post-metaphysical thinking,” Habermas traces a pervasive continuity in modern philosophy from Hegel, Nietzsche, and Heidegger to French post-structuralists such as Foucault and Derrida who so strongly influence the current “postmodern mood” (Habermas 1992, 140; 1981; see also Benhabib 1992, 14–15). According to McCarthy (1987, xiv), “in Habermas’s dialogue with French post-structuralism, Foucault is the preferred partner. More than any other radical critic of reason, Foucault opens up a field of investigation for social research.” Foucault specifically tackles Habermas’s focus on communication and the ideal speech condition, supposedly devoid of relations of power, as utopian (F1997e, 298), pointing to this proposition as perhaps being too Kantian after all.

Yet if Foucault offers a “radical criticism of reason,” this better qualifies the first period of his work, preceding what the literature has called the “ethical turn” (Bernauer & Rasmussen 1987; Andrieu 2004). A first indication of that change can be seen in two versions of the same text, published six years apart. In 1978, Foucault speaks of “reason – the despotic enlightenment” (F1991a, 12);⁵ in line with his analyses of power, he presents modern scientific and technical rationality as a source of domination and questions its universal validity. In 1984, by contrast, he refers to reason as both “a despot and an enlightenment” (DE II #361, 1587), showing that at this crucial period, he has retrieved other forms of rationality, recognizing positive sides to reason after all.

In another late article, he offers a roughly sketched alternative to Habermas’s communicative reason that is, however, far from a radical criticism: here, reason is a “principle of criticism and of permanent creation of ourselves in our autonomy” (TFR 44, emphasis added). While his early writings convey a picture of modern entrapment (Chowers 2004, 165), toward the end of his life he evolved to a position where a certain activity of thought can lead to “the constitution of the self as an autonomous subject” (TFR 42).

Foucault’s position on reason and rationality is, then, far more complex than the postmodern claim for “the crumbling away of reason”⁶:

I am not prepared to identify reason entirely with the totality of rational forms which have come to dominate – at any given moment, in our own era and even very recently – in types of knowledge, forms of technique, and modalities of government or domination; realms where we can see all the major applications of rationality (…) For me, no given form of rationality is actually reason. (…) I can see multiple transformations, but I cannot see why we should call this transformation a “collapse of reason.” Other forms of rationality are created endlessly. (F1998d, 448–449)

One might be tempted here to conclude that Foucault distinguishes between reason as a capacity of the mind, and the practical manifestations of this capacity, the “form of rationality” that reason takes, in a given time and place. This distinction is correct if one adds that Foucault also refuses to attribute any essential meaning to “reason” per se:

I don’t believe one can speak of an intrinsic notion of ‘rationalization’ without on the one hand positing an absolute value inherent in reason, and on the other taking the risk of applying the term empirically in a completely arbitrary way. I think one must restrict one’s use of this word to an instrumental and relative meaning. (F1991b 79, emphasis added)

Reason is only the general term that covers the variet...