Next Article in Journal
Contingency, Irony and Morality: A Critical Review of Rorty’s Notion of the Liberal Utopia
Next Article in Special Issue
Rorty, Williams, and Davidson: Skepticism and Metaepistemology
Previous Article in Journal
Evolutionary Medicine and Future of Humanity: Will Evolution Have the Final Word?
Previous Article in Special Issue
Reconsidering Richard Rorty’s Private-Public Distinction
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Taking Rorty’s Irony Seriously

School of Historical and Philosophical Studies, University of Melbourne, Victoria 3010, Australia
Humanities 2013, 2(2), 292-312;
Received: 4 May 2013 / Revised: 26 May 2013 / Accepted: 6 June 2013 / Published: 12 June 2013
(This article belongs to the Special Issue The Legacy of Richard Rorty)


Richard Rorty’s Contingency, Irony and Solidarity (CIS) is an ambitious and provocative, but for many readers a deeply flawed work. This paper argues that many of its apparent flaws can be understood as integral to Rorty’s attempt to write a work of private, post-theoretical irony. The paper’s first section outlines the substantive theoretical claims about language, selfhood and community which Rorty proposes as an antiessentialist alternative to ‘metaphysics’. The second identifies three difficulties—residual dualism, conceptual problems with the public-private distinction, and the work’s self-referential consistency—which constitute serious, but obvious problems for those substantive claims. The third section argues that Rorty’s metaphilosophical discussion of ‘ironist theory’ suggests CIS should be read as a personal work of irony which eschews theoretical ambitions, showing how this is consistent with and provides a motive for accepting the presence of conspicuous difficulties. The final section considers how the work’s metaphilosophical views interact with its substantive theoretical claims. The work’s irony is interpreted as resulting from the tension between these, so as to coexist rather than conflict with Rorty’s enduring commitments to liberalism and pragmatism.

1. Introduction

Richard Rorty’s Contingency, Irony and Solidarity (hereafter CIS) [1] is an ambitious work that proposes both a wide-ranging substantive philosophical picture and corresponding metaphilosophical views about the aims and methods of philosophy. It is also a perplexing work that strikes many of its readers as deeply flawed, either because its central substantive claims appear untenable or because its irony appears to conflict with its theoretical commitments. This paper will argue, however, that many of its apparent flaws can be seen as integral to Rorty’s attempt to produce a work of post-theoretical irony, a work designed to frustrate the expectation of a unified coherent reading rather than seriously to advance a philosophical theory. I attempt to show that to this end Rorty makes use of two unusual but sophisticated dialectic strategies: in addition to setting out theoretical views which he knew to have conspicuous and serious weaknesses—signalling an ironist’s sensitivity to their fragility—CIS has a multi-voiced structure in which its metaphilosophical voice systematically disrupts prima facie theoretical claims. These are obviously bold claims, likely to seem implausible and even disrespectful. My hope is that neither impression will endure. Rather, by taking seriously its stated metaphilosophical views, I hope to show that CIS emerges as a work of far greater sophistication than it initially appears, one occupying a special place in Rorty’s oeuvre as his only fully fledged work of irony.1
The following discussion is divided into four sections. The first outlines some of the substantive theoretical claims Rorty proposes as an antiessentialist alternative to ‘metaphysics’. The second identifies three basic difficulties—‘fault lines’, as I call them—which stand out as serious but obvious threats to the views outlined in the first section. The third section turns to Rorty’s metaphilosophical discussion of what he calls ‘ironist theory’. I show that taking this as a guide to the interpretation of CIS itself suggests that the work should be seen as an idiosyncratic and personal work of irony and allows the presence of conspicuous fault lines to be understood as a deliberate strategy. The final section considers how these metaphilosophical views interact with the work’s various substantive theoretical claims and interprets the work’s irony as a result of the tension between these. By considering the broader context provided by Rorty’s subsequent works, I argue that the irony of CIS coexists rather than conflicts with Rorty’s liberalism and pragmatism, and that many criticisms of CIS fail to register its ironic character—by taking seriously what was intended ironically, rather than taking Rorty’s irony seriously.

2. The Contingency Theory

In many passages of CIS Rorty speaks—as I shall put it—in a theoretical voice, setting out a substantive philosophical position encompassing a complex of views about language/truth, selfhood and community—which I shall refer to as the ‘Contingency Theory’—and various social-political views. As a background for the following discussion, I begin by outlining the Contingency Theory’s opposition to two defining features of what Rorty labels ‘metaphysics’ and by contrasting its role with that of Rorty’s social-political commitments.
The first defining feature of metaphysics is to posit ‘intrinsic natures’ or ‘essences’—i.e., fixed sets of determinate, atemporal properties (or form)—to which a correct understanding of ourselves and the world is held accountable. On this view, as Rorty characterizes it, truth is something ‘out there’ waiting to be discovered. By contrast, the Contingency Theory proposes an alternative picture on which truth is an attribute of sentences, which are ‘human creations’ ([1], p. 5), such that there is nothing ‘out there’ in the world to which we are either epistemologically or morally answerable.2
This alternative picture comprises an antiessentialist conception of language, the self and community. Drawing on Wittgenstein and Davidson, Rorty first recommends thinking of a linguistic vocabulary as a set of disparate tools rather than as a ‘medium’ serving to build up a picture of the world. This conception of language is supposed to liberate us from what Rorty calls the ‘seesaw battles’ between two opposing views—usually ‘realist’ and ‘idealist’ alternatives—to which the traditional subject-object distinction gives rise ([1], p. 11). Rorty then offers a linguistic conception of selfhood, as something made up of the particular vocabularies and narratives each of us uses to describe our lives. Thus the idea of a deep, hard-wired human nature is replaced by that of linguistic software running on biological hardware: ‘There is nothing to people except what has been socialized into them [...]. To be a person is to speak a particular language’—‘human beings are simply incarnated vocabularies’ ([1], pp. 177, 88). A corresponding pattern emerges in Rorty’s conception of human communities, which are to be created through shared vocabularies and narratives that engender a sense of solidarity, rather than being held together by unconditioned universal features. The resultant binds to others are concrete and limited in reach, such that our ability to reason and to empathize with others depends on our ability to identify with them ‘under some description or other’ ([1], p. 191). In this way Rorty’s position becomes explicitly ethnocentric, advocating community-relative conceptions of justification and moral sensitivity.
A second feature Rorty sees as defining metaphysics is the attempt to ‘fuse’ or ‘unify’ the public and private domains ([1], pp. xiii, xv). This would be to integrate the two domains within a ‘more comprehensive philosophical outlook’, or to ‘hold all the sides of our life in a single vision, to describe them with a single vocabulary’ ([1], pp. xiv, xvi). Rorty rejects all such attempts, which he takes to rest on an appeal to human nature, and instead proposes a ‘firm distinction between the private and the public’ ([1], p. 83), so as to allow the ‘incommensurable’ but ‘equally valid’ goals of ‘self-creation’ and the fostering of ‘human solidarity’ to be pursued independently ([1], p. xv).
The relation between the public and private Rorty envisages is embodied in the figure of the ‘liberal ironist’, proposed as an ideal intellectual in place of the metaphysician. The ironist is introduced as someone who ‘faces up to the contingency of his or her own most central beliefs and desires’ ([1], p. xv). As Rorty defines her, the ironist ‘has radical and continuing doubts about the final vocabulary she currently uses’, she ‘realizes that argument phrased in her present vocabulary can neither underwrite nor dissolve these doubts’ and ‘does not think her vocabulary is closer to reality than others’ ([1], p. 73). Accordingly, in private the ironist is sensitive to the contingency of her evaluative vocabulary and is free to pursue self-creation, experimenting in a ‘sprit of playfulness’ and aspiring to ‘an expanding repertoire of alternative descriptions rather than The One Right Description’ ([1], p. 39 f.). But the liberal ironist is further defined by a public side. Despite her awareness of contingency, she allows her public conduct to be guided by the commitment to avoid cruelty wherever possible.3 The liberal ironist is, so to speak, an embodied disjunction: while experimenting playfully with her private self-creation vocabulary, she is just as committed to the public vocabulary of Enlightenment political liberalism as any metaphysician.
It is worth noting that Rorty might be said to hold ‘theoretical’ views in two different senses. The first is exemplified in the Contingency Theory, as just outlined, which characterizes the role of language and its relation to the physical and social world. However, in describing the commitments that the liberal undertakes Rorty might also be said to hold a social or political theory, roughly a set of beliefs about the values and aims that should guide society and interactions with other people. Both kinds of view are philosophically important, and each might be expounded or adhered to with sincerity and conviction. Nonetheless, for the purposes of the following discussion they differ importantly in that the first is concerned with understanding, or how the world is, while the second is concerned with action, or how the world should be. The difference is important here because the two kinds of theory clearly stand in different relations to Rorty’s own writing. Thus the Contingency Theory, which articulates conditions governing the role and validity of literal talk about the world, bears on whether and how Rorty’s own claims make sense. By contrast, his social-political convictions do not impact generally on the interpretation or coherence of his views, bearing instead on whether society allows him to voice them. The main focus of this paper is on Rorty’s theoretical claims in the former sense. However, the final section also considers whether the social-political convictions expressed in CIS are affected by its metaphilosophical views.

3. Three Major Fault Lines

Rorty’s work unsurprisingly attracts criticism from many quarters, such as defenders of realism, universal rationalism and orthodox analytic philosophy. But no matter how sympathetic one is to Rorty, it is difficult to avoid the impression that the views put forward in CIS’s theoretical voice are deeply problematic. This section identifies three—as I shall call them—major fault lines running through Rorty’s discussion of the Contingency Theory. It will transpire that these fault lines are all surprisingly obvious, prompting the question of why Rorty was apparently indifferent to them.

3.1. Residual Dualism

While rejecting the idea of language as a ‘medium’ and emphasizing the need to get off the ‘seesaw’ of dualist debates, many of Rorty’s own views continue to be shaped by dualist alternatives. The starting point for his discussion of language ([1], pp. 1–7), for example, is a contrast between the view that truth is ‘made’ and that truth is ‘found’ which clearly parallels traditional opposition of a subject-dependent option (appearances, representations, the for-us) and an objective option (reality, things-in-themselves, the in-itself). Rather than moving beyond this contrast, however, Rorty relies on it constantly in criticizing the idea of accountability to the world (truth as found) while emphasizing the constructed or creative character of the human world (truth as made).4 Another example is Rorty’s view that rational persuasion is vocabulary-relative and limited to the ‘interior of a language game’ ([1], p. 47). This view underlies both Rorty’s claim not to be ‘arguing’ for his own position and his conception of morality as a community-wide notion based on the sharing of language-games and vocabularies. However, in thinking of languages as closed systems (of inferential relations) based on the use of corresponding sets of concepts or vocabularies, this view defines an inside-outside structure that parallels the traditional image of a mind looking out on the world. Finally, Rorty’s language-based conception of the self at least gives the impression of reducing the human to a particular kind of representation, one paralleling the traditional ‘veil of ideas’ in its independence from the surrounding world. Rorty would presumably respond that the Wittgensteinian notion of language-games he relies on breaks with the idea that language is representational while reflecting the embedding of language use in practical situations and the wider world (cf. [5], pp. 4, 10 (§§7,23)). Nonetheless, his conception of the self seems to make language a ‘medium’ of the kind he rejects, i.e., something mediating the way in which humans are connected with their world. Moreover, a ‘vocabulary’ surely remains, to reappropriate Rorty’s own words, a ‘medium out of which beliefs and desires are constructed’, meaning that ‘we are still able to ask questions about language of the same sort as we asked about consciousness’ ([1], p. 11).
These views are surprising in two ways. First, because in each of the three above respects, Rorty’s position looks like a linguistic reworking of (transcendental) idealism. It might initially seem that he makes only a minimal concession by accepting that truth is ‘made’ in the sense that languages are human artefacts ([1], p. 7). However, given the link Rorty assumes between justification and vocabularies, together with his language-based conception of the self and communities, this concession has far-reaching consequences, implying that our grasp of the world, ourselves and others is pervasively constructed. Second, the above views are surprising because we might reasonably have expected Rorty, as an avowed nondualist, to move beyond bipolar options such as the made/found alternative and to reconceive (‘redescribe’) our epistemological situation in a way acknowledging how truth, for example, is indebted to both sides, as a product of both our perspective on the world and the way the world itself is.5
Particularly puzzling in the latter respect is Rorty’s insistence that justification has a socially articulated inside and outside, which is difficult to reconcile with a ‘Davidsonian’ rejection of language as a medium. Such an inside-outside delineation assumes that different sets of concepts, or ‘conceptual schemes’, play a distinguished (transcendental) role in connecting knowledge and beliefs to the world in different ways. Davidson’s rejection of the idea of conceptual schemes and any such distinction is part of the background for his claim to have ‘erased the boundary between knowing a language and knowing our way around the world generally’ ([6], p. 107; cf. [1], p. 15). Yet this rejection, which Rorty endorses, surely applies equally to Rortian ‘vocabularies’ and their supposed role in defining the limits of justification ([7]; cf. [8], pp. 5 ff.). A further problem arises through Rorty’s appropriation of Davidson’s account of linguistic competence in terms of ‘passing theories’ indexed to individual speakers and particular times. While this seems congenial to Rorty’s idea of developing an idiosyncratic personal lexicon, Davidson’s is a general claim about a speaker’s abilities that provides no basis for a public/private distinction of the kind Rorty postulates. Moreover, Rorty’s view that communities are characterized by a shared vocabulary and values appears to conflict with Davidson’s claim that language is not a ‘clearly defined shared structure’ ([6], p. 107; my italics).6 Thus Davidson’s views undermine both the language-game internalism and the ethnocentric notion of justification that Rorty advocates.
The preceding considerations bear directly on Rorty’s own project in CIS. To begin with, in general terms his conception of language should tell us what it is to make literal claims about the world. It is also clearly important that a linguistic idealist position of the kind hinted at above would be conducive to a denial of accountability to the world. Indeed it is difficult to avoid the impression that Rorty has set out a view of representation and its role in human life that underlies his pragmatist and ironist approach to philosophy.7 One difficulty, however, is that endorsing such a position—as Rorty appears to—not only upholds a basic framework closely paralleling traditional dualist alternatives, but makes a standard dualist move (antirealism). This means that rather than liberating us from dualist questions and ‘seesaw battles’, Rorty is simply sitting on one end of the seesaw. The alternative would be to redescribe the human situation in a framework that moves beyond dualist alternatives in a way I have suggested that Rorty does not. Had he done so, he might have succeeded in his stated aim of getting away from ‘seesaw battles’. But the cost of doing this would be to give up the idealist view of language, self and community that provides an apparent justification for his antifoundationalism.

3.2. The Public-Private Distinction

Despite its proclaimed cardinal importance to the work, there is no sustained discussion in CIS of how the public-private distinction is to be understood, leaving unclear how the distinction is intended and whether it can do its intended work.
Rorty usually talks about the distinction as a split in one’s ‘vocabulary’: ‘the ironist’s final vocabulary can be and should be split into a large private and a small public sector’ ([1], p. 100, cf. pp. 92, 120). However, it remains unclear how this split is to be made. For example, it is not clear whether everyone’s public vocabulary is to comprise the same set of core terms. Occasionally Rorty links this expectation with the ‘metaphysician’ as opposed to the ironist ([1], p. 92 f.); yet elsewhere he claims that equivalent ‘public’ terms, marking the same thin moral commitments and allowing argumentative discourse, should form a subset of everyone’s vocabulary (cf. [1], p. 93, xiv). Similarly, it is not clear where the boundary between the public and the private is supposed to lie. Often Rorty clearly has in mind political communities or society at large. Sometimes public contexts are to include (non-political) philosophical theories’ claims to intersubjective rather than merely individual validity.8 And ultimately—most noticeably in his discussions of Nabokov and Orwell—Rorty extends his use of the term ‘public’ to all interpersonal interaction, including at the one-to-one level.9 A further source of unclarity is that Rorty does not cash out the distinction in practical and psychological terms, as his view of the interconnection between vocabularies, language-games and selfhood appears to require. He does not, that is, tell us which language-games or practices count as private and which as public. For example, are ‘private’ language-games just those we play alone, in introspective and reclusive activities such as daydreaming or writing a diary? Similarly, at a psychological level how are we assumed to partition our mental life into public and private sides? Would we not feel pressure to bring our public and private attitudes into accord? (And, if our private and public sides are free to diverge—suppose I were outwardly conformist, but a rebel in ‘private’—how does this differ from delusion or superficiality?)
Beyond matters of clarity, it remains doubtful that the distinction could play its intended role of allowing our public and private sides to flourish independently. Since being a particular self involves various practical and social roles, and characteristic patterns of behaviour towards others, it seems difficult to see the suggestion that selfhood might be realized or properly developed ‘privately’ in abstraction from a ‘public’ side as coherent.

3.3. Performative Consistency

Rorty’s insistence on distinguishing the public and private raises the question of his own work’s intended standpoint. On the one hand, it seems intended as a ‘private’ work, as ‘ironist philosophers are private philosophers’ whose ‘work is ill-suited to public purposes, of no use to liberals qua liberals’ ([1], p. 94 f.). Yet CIS is clearly a ‘public’ work in the sense of aspiring to political relevance—one of its principal aims is to defend liberal political culture from erosion by ironist thinking, to ‘think of philosophy as in the service of democratic politics’.10 It also appears to be public in the sense of claiming intersubjective validity as a theoretical work. Hence the work faces a dilemma of its own making: either its aims are too public to be a work of private irony; or, if read as a work of private irony, consistency would require giving up its claims to public relevance.
If we accept that CIS embraces both public and private aspects, the question arises of how it differs from the metaphysician’s project of ‘unifying’ the two in a ‘more comprehensive philosophical outlook’ or a ‘single vocabulary’. A first response to this might be that the two aspects of his work are independent or disjunctive. But this seems wrong, as Rorty’s view of self-creation is clearly attuned to his political ideals, linked by the aim of maximizing personal freedom. Accordingly, on the one hand, society is held together by ‘little more than a consensus that the point of social organization is to let everybody have a chance at self-creation’ ([1], p. 84). On the other hand, at the individual level, to ‘see one’s language, one’s conscience, one’s morality […] as contingent products […] is to adopt a self-identity which suits one for citizenship in such an ideally liberal state’ ([1], p. 61).
A second response might be that the public and private aspects of his work are not ‘unified’, but linked in some other way, which Rorty refers to as ‘reconciliation’ or ‘accommodation’ (cf. [1], pp. 68, 141). Yet there is a further complication in that the two aspects appear to be calibrated to a certain conception of what it is to be properly human. Rorty is careful to avoid suggesting that any particular vocation or life-content be considered paradigmatic, but while using the poet as a paradigm of selfhood he lapses into suggesting that all of us are poets deep down. Thus he explains that ‘the distinctively human, as opposed to animal, portion of each human life [is] the use for symbolic purposes of every particular person, object, situation, event, and word encountered later in life’ ([1], p. 36). Similarly, the ‘strong poet’s’ urge to demonstrate his individuality is presented as ‘merely a special form of an unconscious need everyone has’ ([1], p. 43). So while Rorty would no doubt deny relying on a conception of human ‘nature’ that ‘unifies’ the public and private aspects of his position, there is at least a danger that the difference is merely one of labels.
These three fault lines constitute obvious targets for standard philosophical criticism. Indeed, I suspect—although I am not claiming to have established here—that each of the fault lines just outlined could be shown to render the position set out by CIS untenable as a theoretical position. What I want to emphasize here, however, is that each occurs where key features of Rorty’s own exposition should lead them to be expected. They stand out as obvious difficulties striking at the heart of Rorty’s project.11 It is therefore surprising that Rorty appears not to anticipate them. One possibility, of course, is that he was genuinely unaware of these difficulties, in which case CIS would be—as it is often taken to be—a seriously intended but deeply flawed work. Another possibility is that Rorty was aware of these difficulties and at least accepted their presence in his text. Perhaps he didn’t want the Contingency Theory to be taken too seriously and intended to invite such criticisms so as to foster an ironist’s ‘radical and continuing doubts’ about his vocabulary. In this case, precisely the vulnerability of the views put forward by Rorty’s theoretical voice would befit them to developing an ironist sensitivity in the reader, a sense that even what appear to be the theoretical underpinnings of ironism remain open to challenge and revision. Despite its obvious initial implausibility, the next two sections will set out how this second possibility is both consistent with and plausibly motivated by Rorty’s apparent intention to write a book of post-theoretical irony.

4. A Work of Irony

There are also many passages in CIS which speak in a metaphilosophical voice, setting out Rorty’s view of the aims and methods of what he calls ‘ironist theory’. This section considers what the implications would be of taking this metaphilosophical voice to guide the interpretation of CIS itself. The overall interpretation of the work will involve further complications that remain to be discussed in the next section. However, as Rorty’s work explicitly situates itself in the tradition of ironist discourse, understanding the implications of these passages is clearly important in working towards a complete understanding of CIS.
Rorty’s ironist, to recall, is someone who is aware of the contingency of, and entertains doubts about, their own vocabulary and beliefs. Given his principled freedom from external determination, Rorty suggests that the ironist’s ‘generic task’ is ‘to create the taste by which he will be judged […] to be able to sum up his life in his own terms’ ([1], p. 97).
The ironist theorist is a particular kind of ironist: he is an ironist with a philosophical background, which Rorty identifies as the ‘Plato-Kant canon’, against which to carry out the task of finding his own terms. The ironist theorist seeks ‘a redescription of that canon which will cause it to lose the power it has over him’ ([1], p. 97). Ironist theory thus has what might be called a therapeutic function: ‘The goal of ironist theory is to understand the metaphysical urge, the urge to theorize, so well that one becomes entirely free of it.’12
A particular problem for ironist theorists is how to deal with the contingency of their own position. Their theories characteristically take the form of a synoptic narrative or a theoretical framework used to describe the philosophical tradition (the ‘Plato-Kant canon’). Theories of this kind, according to Rorty, appear to suggest that the ‘realm of possibility is now exhausted’ and that there is no ‘dialectical space left’ for the ironist’s own view to be recontextualized ([1], p. 104). As a result, the ironist theorist risks looking foolish in the way, Rorty believes, Hegel did by presenting his own work as the final episode in the history of philosophy. To do this is to ‘fancy oneself’ or set oneself up as the Last Philosopher ([1], p. 106). The problem for the ironist theorist is therefore how to be consistent as an ironist, i.e., how to demonstrate an awareness of possible changes to one’s own vocabulary while putting forward a theoretical framework. Rorty describes this problem—‘how to finitize while exhibiting a knowledge of one’s own finitude’—as ‘the problem of ironist theory’ ([1], p. 104 f.).
To understand how Rorty thinks this can be done—how to avoid the folly of setting oneself up as the Last Philosopher and solve ‘the problem of ironist theory’—I want to piece together various clues offered by his discussions of Nietzsche, Proust, Heidegger and Derrida (in Chapters 4–6). Note that in doing this I am not defending Rorty’s readings of these authors and will not be discussing their exegetic virtues or otherwise. The value of these discussions, for my present purpose, lies simply in what they tell us about the kind of book Rorty was trying to write, not what they tell us about the authors he is discussing.
Nietzsche and Proust are introduced as authors who succeeded in the ironist’s ‘generic task’ of creating the taste by which they would be judged. Both are to have demonstrated an awareness of contingency by redescribing themselves and other people—in Nietzsche’s case via their books—in the pursuit of autonomy. However, Rorty thinks that Proust was more consistent in his recognition of contingency, as Nietzsche sometimes affiliated himself with something larger than his personal perspective, something ‘sublime’ such as the destiny of Europe or the Will to Power. While taking this affiliation to be what distinguishes Nietzsche as a ‘theorist’ rather than a novelist, Rorty thinks it also means that Nietzsche is less able than Proust—as a ‘non-theorist’ ([1], p. 105)—to acknowledge the possibility of being redescribed himself.13 We might find this comparison odd. After all, it is obvious that a novelist has no need of a larger theoretical perspective, and it seems inappropriate to congratulate him for solving a problem he never faces. Despite this incongruity, there is a lesson to be drawn from Rorty’s comparison. What it brings out is that consistent ironism involves recognizing contingency looking forwards as well as backwards, and hence accepting that one’s own position is open to subsequent redescription by others.
A second way of demonstrating one’s ironist credentials is attributed to Heidegger. According to Rorty, Heidegger made use of a complex linguistic strategy in order to write a history of metaphysics in a way that distanced him from that history while preventing his own redescription by others. On the one hand, Heidegger recognized the unavoidability of using the traditional vocabulary of metaphysics in writing its history. On the other hand, in developing his own views Heidegger sought to exploit the poetic resonance of terms (‘elementary words’), developing rich webs of association that defy reformulation. In this way the language of metaphysics was deployed as a ‘bracketed’ vocabulary, retaining the ‘seriousness of its finality’ ([1], p. 112), yet constantly transcended in a richer perspective that ‘poetize[s] philosophical language by letting the phonemes matter’ ([1], p. 116). In using this linguistic strategy, Rorty explains, Heidegger ‘wanted to construct a vocabulary which would both constantly dismantle itself and constantly take itself seriously’ ([1], p. 112). Now, although Rorty thinks Heidegger goes astray in mistaking idiosyncratic personal resonances for universally binding ‘elementary words’, he is conspicuously impressed by what he takes to be Heidegger’s method. Clearly this is not because Heidegger succeeds (let us assume) in making his position un-re-describable—for, as his discussion of Nietzsche and Proust clarified, Rorty thinks a consistent ironist should embrace the possibility of being redescribed. Rather, I suggest, it is because Heidegger exemplifies the attitude a Rortian ironist should have towards vocabularies. In ‘bracketing’ the language of metaphysics, Heidegger shows us how one can have an ambivalent provisional commitment to a given vocabulary which is simultaneously in question—how we can have a vocabulary which is being simultaneously ‘dismantled’ and taken seriously.
Rorty sees the most consistently ironist approach to theory in Derrida’s The Postcard: From Socrates to Freud and Beyond. In this work, according to Rorty, Derrida deals with the history of philosophy in a manner analogous to Proust. In contrast to the theoretical and transcendentalist tenor of his earlier works, Derrida here ‘privatizes his philosophical thinking’ by writing about his philosophical predecessors in an idiosyncratic manner, developing personal associations and making ‘private jokes’ ([1], p. 125). In doing this, Rorty further explains, Derrida ‘simply drops theory’ ([1], p. 125) and has no pretence to fitting his predecessors into a determinate framework or yielding determinate results—‘there is nothing propositional to be taken away from the experience of reading it’ ([1], p. 135, cf. p. 130). As it has no take home message, it is inherently unsuited to play any public role, leading Rorty to conclude that the best thing to do with such a book is to ‘sit back and enjoy it’ ([1], p. 133). Against this background he praises Derrida for having invented a new kind of writing, one which is beyond the philosophy-literature contrast and stands as ‘the end product of ironist theorizing’, constituting ‘the only solution to the self-referential problem [...] of how to distance one’s predecessors without doing exactly what one has repudiated them for doing’ ([1], p. 125).
In the light of these various clues, what kind of book should we expect CIS itself to be? Note, first of all, that it does not quite meet the definition for a work of ironist theory. For CIS is not itself a book about the ‘Plato-Kant canon’; rather, all its main reference authors are themselves ironist theorists. So perhaps it would be better described as a book about ironist theory or about ironist theorists.
It might also be argued that Rorty does not offer the same kind of theory as the ironist theorist, i.e., a narrative or theory applying to the philosophical tradition. Whether or not that is the case, it is not difficult to see that the problems Rorty diagnoses result for the ironist from the attempt to advance any kind of theory. This is because to put forward a ‘theory’ of any kind is to propose a framework that is not supposed to be ‘redescribed’ by others at will. To concede that a theory can be redescribed in this way would undermine its claim to validity. Put another way, it is part of the force (illocution) of proposing a ‘theory’ to resist redescription and to lay claim to assent from others. This means there is an inherent tension between the act of advancing a theory of any kind and the requirement that ironists embrace the possibility of subsequent redescription. Rorty is therefore right, in discussing Derrida, to suggest that the only way to solve the ‘self-referential problem’ is to ‘drop theory’. However, this is because what Rorty called ‘the problem of ironist theory’ is in fact a problem which cannot be solved by theory—it is a problem generated precisely by an ironist’s attempt to have a theory.
In that case, should CIS be thought of as a work of ‘theory’? On the one hand, it seems not. Rorty has made clear that he thinks the very attempt to theorize is problematic for the ironist and himself presents ironist theory as passé, literally something of the past.14 On the other hand, given the lengthy presentation of philosophical views of language, the self and community (the Contingency Theory), as well as the role of poetry, literature, etc., how could CIS be anything but a work of theory? It might seem that Rorty is in the embarrassing position of being an ironist advancing a theory, while telling us that consistent ironist writing eschews theory.
I want to suggest that Rorty’s blushes might be spared if his praise of certain features in Derrida and Heidegger is taken as a guide to what he himself was trying to do. Rorty’s high praise for the mode of writing he attributes to Derrida suggests that this might be the model for Rorty’s own practice. This would make CIS a more private book than it initially seems to be, such that it is intended primarily as a book of personal responses to authors. Rorty would be writing about philosophers simply because these are the authors he has contingently encountered in his own past.15 Such an emphasis on the personal perspective coheres with Rorty’s sometimes rather idiosyncratic ‘redescriptions’ of the authors he discusses, as well as his expressions of sympathy for ‘strong readings’ and of indifference to the ‘real meaning of books’ ([1], pp. 160n, 80). For the same reason Rorty would no doubt be unconcerned by any charge that his metaphilosophical readings of Nietzsche, Heidegger and Derrida are inaccurate. Rorty’s interest appears to be not how these authors are best read by their own lights, but simply to narrate his own personal responses to them.
How, in that case, should one account for the presence of a large amount of philosophical theory (in particular the Contingency Theory) in CIS? After all, Rorty’s Derrida ‘simply drops theory’. It is here, I suggest, that Rorty’s Heidegger comes in. What Rorty takes from Heidegger is the idea of using, while simultaneously ‘bracketing’ a theoretical vocabulary. Precisely because he is writing from his own contingent standpoint, Rorty accepts—with what he in one context calls ‘ironic resignation’ ([1], p. 76)—the inevitability of using the vocabulary and frameworks of reference of existing discourse. In this respect, like Heidegger, he accepts the contingencies of his own situation (his own facticity). To emulate the bracketing he discerns in Heidegger, Rorty needs some way of exhibiting the ironist’s ambivalent attitude towards her own vocabulary. However, he cannot do this in the ‘Heideggerian’ way by poetizing his philosophical terms, as this would be to resist redescription and, in this sense, to fail as an ironist. So Rorty needs some other way of destabilizing the theoretical vocabulary he is using, some other way of having this vocabulary ‘both constantly dismantle itself and constantly take itself seriously’ ([1], p. 112).
This takes us back to the major fault lines of Section 3. There I suggested that it is odd that Rorty should have failed to anticipate and respond to what appear to be rather obvious tensions in the Contingency Theory. However, I think we are now in a position to see why the presence of such fault lines might be integral to Rorty’s achieving his overall aims. For, if many of its central claims are manifestly faulted, his theory cannot be thought of as coherent and enduring. Moreover, Rorty might plausibly have ‘failed’ to anticipate these problems simply because he intended them to be noticed, because he was inviting us to notice them. Destabilizing his own theoretical vocabulary in this way would amount to withholding full endorsement of the Contingency Theory and in this sense—one analogous to Husserlian epoché—‘bracketing’ that theory.16 It would also allow Rorty to mark his own openness to future revisions, as the ironist attitude requires, and so to avoid presenting himself as the Last Philosopher.17
Taking the work’s metaphilosophical voice as a guide to its interpretation thus suggests that CIS should be read as a post-theoretical work of irony, a personal work similar to the approach Rorty attributes to Derrida. This would imply that Rorty’s commitment to the Contingency Theory was somewhat ambivalent, allowing us to make sense of the thought that he might have exploited the presence of conspicuous fault lines in his own theory so as to expose its fragility. In an ironist perspective the Contingency Theory might then be seen as what Rorty later (in discussing Nabokov) refers to as a ‘self-consuming artifact’, one of ‘the systems of general ideas which ironist theorists construct in order to attack the very idea of a general idea’ ([1], p. 154).

5. Consequences of Irony

The metaphilosophical voice in CIS plausibly provides a guide to the overall interpretation of the work. However, it is clearly not the only factor, and in deciding how to interpret its importance it should be considered how this voice interacts with the various theoretical and socio-political claims that are articulated both in CIS itself and in the broader context of Rorty’s work.
Before going any further, however, an obvious challenge to the approach developed in the last two sections should be addressed. To suggest that Rorty put forward theoretical views he knew to be seriously flawed and intended to be recognized as such might seem absurd and/or a wilful and disrespectful misreading of the work. If Rorty intended CIS to be read in this way, one might object, why didn’t he just say so? One straightforward response to this is simply that he does, that the metaphilosophical views developed in the work are surely intended to shed light on its own practice. Admittedly, this would be a circuitous route to revealing his intentions, but a direct statement—something like ‘this book is a work of self-consuming irony’—would presumably ruin the fun, depriving the work of its intellectual and aesthetic sophistication. (Recall Rorty’s comment that ‘straightforward, unselfconscious, transparent prose [is] precisely the kind of prose no self-creating ironist wants to write’; [1], p. 89.) A second straightforward response might be to dismiss the challenge by pointing out that Rorty, as the champion of ‘strong’ readings, would be in no position to object to a supposed misreading. This is perhaps true. But if it remains merely an ad hominem response it carries no weight. For the fact that Rorty thinks liberties can be taken in interpreting other authors is not in itself a justification for doing the same with his texts.
There is a somewhat deeper reason that prevents the ironist reading developed above being rejected as a misinterpretation of Rorty’s intentions, due to the way the presence of irony complicates the interpretation of a text.18 To shed light on these complications I want to appropriate two thoughts from Paul de Man’s insightful essay ‘The Concept of Irony’ [12].19 The first is that the overall mood of a text is altered by the presence of irony, which introduces a pervasive destabilization of its comprehensibility. Once we know that a text is being put forward in an ironic spirit, it becomes difficult to disentangle irony from any intended literal content. We are forced to ask how two such aspects of the text might be distinguished, and then to concede that the suspicion of irony renders text-immanent evidence unsuitable as a criterion. There is a pervasive breakdown in the attitude of trust we might otherwise bring to a text, as when we must decide whether to believe the habitual liar or the boy who cried wolf too often. In de Man’s terms—adopted from Friedrich Schlegel—irony is a ‘permanent parabasis’, where ‘parabasis’ is taken to mean ‘the interruption of a discourse by a shift in the rhetorical register’ ([12], pp. 179, 178). To say that parabasis is permanent means simply that such shifts can occur anywhere in the text, and it is this possibility which ‘interrupts, disrupts, profoundly the inner mood’ of the text in question ([12], p. 179). From the reader’s perspective knowing that a text is ironic in mood thus introduces a principled opacity with respect to any intended literal content.20
According to this line of thought, by self-consciously positioning itself as an ironist discourse, CIS cannot claim to have a discernable core of literal content for correct interpretation to pick out. Rorty thus gives up the entitlement to claim that his intentions have been misunderstood: even if his text is interpreted in a manner other than its author intended (which is obviously possible), the adoption of an ironic stance breaches the conditions for reliable communication, so that the notion of misinterpretation is no longer in point—at least for any reading that stands in a suitably intelligible relation to the evidence of the text itself. Note that none of this amounts to, or is intended as, an objection to Rorty. On the contrary, Rorty’s own disavowal of the idea that a text has a real meaning—along with his own somewhat freewheeling appropriations of Davidson, Nietzsche, Freud, Derrida etc.—provide evidence that Rorty himself recognized and embraced the destabilization of texts that irony brings. And far from recoiling from the extreme conclusion that an ironist text yields no determinate content, it was precisely because ‘there is nothing propositional to be taken away from the experience of reading it’ that Rorty singles out Derrida’s Postcard for particular praise ([1], p. 135).
Of course, the preceding considerations also mean that the reading developed here cannot claim to be directly tracking Rorty’s true intentions. So in what sense can it claim to be correct? As with any serious exegesis, my aim here is to remain accountable to the textual evidence and make the best sense possible of it. The present reading is thus not intended as an ironic redescription or a strong (re)interpretation of CIS, but aims to read that work by its own lights. The point of denying that Rorty could claim to have been misinterpreted is simply that he too would remain accountable to the textual data, including complications due to a metaphilosophical voice that renders the text’s overall claims unclear by questioning the status of theory. This is not to deny that Rorty had particular intentions in writing the book, but merely to insist that any such intentions do not simply trump or outweigh the textual evidence.
To see what that means in the present case, consider two possible scenarios. The first would be that Rorty intended CIS to be read in just the way I am suggesting. In this case, despite the complications due to the introduction of the metaphilosophical voice, the textual evidence would have succeeded in conveying his genuine intentions. This best case scenario contrasts with a second, worst case scenario. Suppose Rorty did not intend the metaphilosophical voice to guide interpretation of the work or for CIS to be read as a post-theoretical work of irony.21 While this might provide prima facie grounds to question the reading suggested here, we would nonetheless be entitled to hold Rorty accountable to the text as it stands. We might, for example, ask why his discussion is complicated by the presence of what I have called the metaphilosophical voice. If Rorty intended his theoretical views to be taken at face value, why discourse on the desiderata and difficulties of ironist theory at length, and why align his own position (the ‘liberal ironist’) to this discourse? Further, it would then be extremely surprising that the metaphilosophical voice—against his own intentions—sustains and even encourages a detailed ironist reading of the kind set out in Section 3. All the more so, as the possibility of such a reading is easily avoided—or rather is avoided by default. For, without the deliberate and approving thematization of ironist theory, the reader would have no grounds to doubt that Rorty’s theoretical claims are being asserted literally. Finally, on this second scenario, the difficulties inherent in the Contingency Theory would indeed constitute acute problems rather than potentially forming part of a more sophisticated overall dialectic. Hence, even in this ‘worst case’ scenario, Rorty’s intentions should not be assumed simply to trump the textual evidence, as his success in realizing those intentions must be judged on the evidence of the textual artefact he produced. (Just as my intention to produce a proof of Fermat’s last theorem or a painting in correct central perspective would not suffice to achieve these aims.)
Although the ironic mood of CIS implies that it cannot reliably convey Rorty’s specific intentions, considering the two possibilities just sketched does provide an indirect reason for favouring the reading proposed here. For, if Rorty succeeded in writing the book he intended to, then we would expect his intentions to be inscribed in the textual evidence. Hence, as just outlined, assuming that Rorty intended to write a work of post-theoretical irony allows us to see him as having succeeded in this way, whereas key features of his text would otherwise be left in need of explanation.
Independently of Rorty’s intentions, however, the textual evidence itself provides direct reasons for thinking that the present reading, characterized by taking seriously the metaphilosophical voice in CIS, best makes sense of the text. To begin with, as I have been suggesting, various features of Rorty’s text—not least the presence of this voice—would otherwise be difficult to explain. Disregarding the implications of this metaphilosophical voice would in turn yield only a partial and arbitrarily selective reading of the text. Further, while the reading proposed here might seem unusual (‘counterintuitive’), it is important to notice that taking Rorty’s theoretical claims to be advocated seriously is not an obviously better alternative. One reason for this is that his theoretical claims are themselves deeply problematic (cf. Section 3 above). Another is that Rorty’s theorizing would, in that case, conflict with his metaphilosophical pronouncements and one would be left asking why he himself had failed to write an ironist work. By contrast, the present reading has the advantage of allowing us to see Rorty as doing what he tells us philosophical writers should do, and so to understand why his theoretical and metaphilosophical voices belong together in the same book.
It might, however, be objected that to diagnose a pervasive destabilization of the text is both an exaggeration and a mistake. Why think that Rorty’s metaphilosophical discourse on ironist theory impacts on the Contingency Theory’s claims? To be sure, that is how it looks, but perhaps this interpretive move is self-defeating. For it seems to assume that the metaphilosophical voice is privileged, itself inexplicably exempted from the mood of irony it generates. Yet, in that case, why not adopt a Rylean distrust of metaphilosophical reflection and instead privilege and preserve Rorty’s theoretical voice?22
This brings me to the second line of thought I want to draw from de Man, namely, his view of the means by which a pervasive destabilization of a text’s comprehensibility—a ‘permanent parabasis’—is affected. The shifts in ‘rhetorical register’ that most interest de Man are due to the use of different and conflicting voices within a work, one of which is a serious theoretical voice setting out a body of theoretical commitments capable in principle of being taken literally (unironically). The latter feature is crucial: as irony is a distortion of the literal mood, there needs to be an uncorrupted voice, so to speak, which irony modifies (cf. note 15). On de Man’s view, it is the presence in the narrative structure of divergent voices, which interfere with and disrupt each other, that generates the overall mood of irony. As an example of the use of a multi-voiced structure to produce irony, he discusses the ironic recontextualization of Fichte’s idealism in Schlegel’s novel Lucinde. Schlegel’s novel includes an apparently philosophical reflection, formulated in Fichte’s terminology, which is recognizable as a thinly veiled ‘reflection on the very physical questions involved in sexual intercourse’ ([12], p. 168). By speaking the language of Fichte’s theory in narrative contexts in which it is clearly out of place, Schlegel sets up equivocation, signals distance and mocks the gravitas of that theory.
The interest of de Man’s (Schlegelian) view here is that—as the preceding sections have tried to make clear—a similar multi-voiced structure can be discerned in CIS. The voice which sets out the Contingency Theory—Rorty’s conceptions of language, the self and community—is clearly a theoretical voice, analogous to that of Fichte in de Man’s example. A second voice is the metaphilosophical voice that sets out Rorty’s views on the aims and evolution of ironist theory. Given that Rorty saw the solution to ‘the problem of ironist theory’ in Derrida’s ‘dropping’ of theory and the shift to a new genre of writing centring on idiosyncratic ‘private jokes’ yielding ‘nothing propositional’, these two voices are obviously in tension. The presence of the metaphilosophical voice clearly interrupts and disrupts the theoretical voice and engenders the work’s overall ironic mood.
Two important lessons can be drawn from de Man’s model. First, it makes clear that the generation of an ironic mood does not require the metaphilosophical voice (itself inexplicably exempted from irony) to be privileged. Rather, a deliberately staged antagonism between a plurality of voices within the text suffices to generate the reciprocal interruption or disruption required for ‘permanent parabasis’. To be sure, without the metaphilosophical voice we would take Rorty’s theoretical views as being advanced literally, but rather than being privileged, this voice needs only to produce conflict with the theoretical voice, to send out mixed messages about how to take the latter’s content. Second, de Man’s model sheds light on what happens if we fail to register a suitably structured text as having an ironic mood. Such failure involves not realizing that the work is intended to have a multi-voiced structure with deliberate antagonism between different voices. In this case we will read the text as being intended literally, presumably recognizing the various lines of thought presented, and perhaps seeing them as in unintentional conflict. In other words, because the irony generated by a multi-voiced narrative structure is parasitic on and modifies a prima facie serious theoretical voice, failure to appreciate an intended ironic mood leaves us taking the theoretical voice’s claims at face value. (Just as we might be left with a literal understanding of a joke we fail to ‘get’.)
As de Man’s model is quite distinctive, a couple of clarifications are perhaps necessary. First, because it involves conflict between them, this model requires that different voices, or different lines of thought, are distinguishable. This does not, however, mean that the theoretical and metaphilosophical voices are independent. Indeed, their ability to be in tension is due to the fact that they should be systematically connected, as each defines conditions of adequacy for the other (the metaphilosophical voice articulates what is or should be done in philosophical theory). In terms of the analogy, the concern is with two distinguishable voices engaged in a dialogue about the practice of doing philosophy—a dialogue in which they should agree, but in the case of irony do not. Second, the suggestion here is not that the metaphilosophical voice, considered in isolation, cancels itself out through its own irony, nor that the text recovers a literal mood by this means (such that the theoretical voice emerges from the ruins, so to speak, as the literally advocated core of the work). Some people, of course, might argue that an ironist metaphilosophical voice is incoherent or self-cancelling. However, even if that were the case, it would not constitute any reason to think the theoretical voice’s claims are literally asserted and would instead suggest merely that the attempt to be an ironist has failed. On de Man’s model what matters in generating an ironic mood at the text level is simply that there is text-internal evidence that casts doubt on the literal force of the theoretical voice. Returning to the analogy, the idea is that two voices, each of which is at least prima facie intelligible, can be discerned as part of a dialogue, but that the dialogue does not yield a net overall result due to their disagreement. Finally, a failure on the reader’s part to discern irony does not mean that the text reverts to the mode of literal assertion. The reader takes the theoretical voice literally in this case only as a result of a significant misunderstanding, of failing to grasp the subtlety and overall point of the text. Analogously, I might take literally the words of someone who congratulates me ironically on my tact or my culinary skills, but in doing so I would clearly miss the point of the utterance. (Similarly, my failure to understand a joke clearly does not make it into a literal assertion.)
An example will show how the second lesson of de Man’s model affects the overall interpretation of CIS. Many objections to Rorty’s work centre on the thought that his irony is incompatible with his social and political views. Thus it seems odd that while the ironist is characterized by ongoing experimentation with vocabularies, liberal political values are supposed to elicit an adherence no less stable than that required by metaphysics. One response to this is the objection that the suspension of irony in the public sphere looks ad hoc and arbitrary. Another is to accept this move, but object that it leaves Rorty, in one critic’s words, with a dualist split between ‘moral commitment […] and the detachment of the ironist […] two perspectives which are not merely distinct but basically contradictory’ ([15], p. 27). Alternatively, it is sometimes felt that the ‘liberal ironist’ is an impossible juxtaposition, as the ironist’s habitual unorthodoxy rules out solidarity with others.23 Rorty’s attempt to head off such objections by introducing the public-private distinction fares little better. In addition to the conceptual problems with this distinction (Section 3.2), it seems no less ad hoc and problematic than the original irony/liberalism tension, and is itself sometimes suspected of being a metaphysical postulate that Rorty ‘inexplicably exempts from his otherwise potent irony’.24
Understandable as many of these criticisms are, they result from the attempt to read CIS as speaking in a single voice, as intending a coherent overall position. As such, they arguably presuppose a failure to register properly the mood of irony in CIS and the means used to attain this. On the reading offered here, once we appreciate the ironic mood of CIS, the requirement for overall coherence or reconciliation presupposed by many objections is simply misplaced. Rather than being reconciled, in the sense of being brought into reflective equilibrium, the tensions between irony and politics—the private and the public—can be acknowledged as forming part of the work’s internal multi-voiced structure. Indeed, independently of any consideration of irony, the persistence of such objections is somewhat surprising, as Rorty clearly disavows a need for the kind of unity they envisage. The whole point of the public-private distinction, to recall, was to oppose the metaphysician’s project of ‘unifying’ the two domains in a ‘more comprehensive philosophical outlook’ or a ‘single vocabulary’. This message is generalized in Rorty’s conception of language, which likens vocabularies to tools (for specific, unconnected purposes), as opposed to pieces of a jigsaw puzzle that can be pieced together to form a single complete picture ([1], p. 11). The clear implication is that Rorty’s vocabulary is supposed to remain internally fractured, rather than form a unified whole. So while others might independently seek to reconcile private irony with public liberalism coherently, any such attempt clearly departs markedly from Rorty’s own stated aims and in this respect fails as criticism.
Such criticisms of the relation between Rorty’s irony and his politics are, however, consistent with the above suggestion about what happens when readers fail to grasp the ironic mood of CIS. In this case, I said, claims set out by the work’s theoretical voice will be taken at face value. The same can be expected to occur with social-political views set out in CIS. And sure enough, whatever failings they discern in the book, none of Rorty’s critics doubt the sincerity of his advocacy of political liberalism (no one thinks, for example, that these claims are undermined by his ironism). There is a sense, then, in which Rorty’s book has a fail-safe design: if we miss his irony we are left with his politics, and the worst that can happen is that we find him a mediocre theorist of liberalism. In this way CIS performatively proves Rorty’s claim that ironism is not ‘intrinsically hostile’ to liberal values (cf. [1], p. xv) by providing an example of high ironist writing suspended above a safety net of liberal avowals.
This, I think, also allows us to understand better the relationship between Rorty’s ironism and his pragmatism. Both labels stand for a general antimetaphysical—i.e., antiessentialist or antifoundational—attitude that views current commitments as contingent products open to future revision, and both see no need—either intellectual or practical—for an overarching unity in one’s vocabulary. So, on the one hand, irony does not threaten the antimetaphysical stance of pragmatism; on the other hand, if we miss the irony of CIS, we are left with Rorty’s pragmatism. Ironist theory and writing is a highbrow indulgence that coexists rather than conflicting with his pragmatism.25 One reason we are ‘left with’ commitments to pragmatism and political liberalism is that CIS occupies a unique position in Rorty’s oeuvre. First, it is the only work of his in which irony features prominently, as this theme postdates the Mirror of Nature and disappears in his subsequent writings, which focus on defending his views on philosophical, social and cultural issues in a more conventional academic manner. In the broader context of these works Rorty’s enduring commitments, his pragmatism and his liberalism, for example, feature without irony. Further, as his later works were predominantly short, CIS remained the only work of Rorty’s with the subject matter, length and organization to undertake such a complex project of ironist writing. As an indulgent work of private irony, it remains the exception.
Finally, what is the point of an ironist reading of CIS of the kind set out here? After all, the end result appears to be that the book lacks overall coherence and puts forward no positive theoretical message. One point requires clarification and emphasis. To say that CIS lacks overall coherence does not mean it exhibits ‘incoherence’, which is both unintentional and marks a failed attempt to speak in a single voice. Rather, the textual evidence suggests that CIS is deliberately ironic and multi-voiced in its architecture and is designed to frustrate the expectation of a unified coherent reading. This is clearly consistent with the state of mind Rorty attributes to ironists, who are ‘never quite able to take themselves seriously’ due to an awareness of the ‘contingency and fragility of their final vocabularies’ ([1], p. 73 f.). Borrowing Sartre’s term, Rorty fittingly describes this state as ‘meta-stable’, as the ironist finds herself in a condition which—just as Sartrean bad faith—is able to endure despite internal contradiction. Rorty’s practice thus mirrors his view that the ironist, unlike the metaphysician, can live without a unified ‘coherent’ philosophical outlook.
All the same, what is the point of such a reading? Rorty himself faces the same question about his interpretation of Derrida’s Postcard. Having conceded that it is of ‘no use’ in yielding propositional conclusions, he suggests simply that we ‘have to see whether we can find a use for it’ over time ([1], p. 135). In other words, Rorty leaves the question open. Does this mean ironist writing is useless? And why should we want to be ironists? It is not hard to come up with an answer to such questions: The ironist’s mindset is characterized by intellectual modesty and flexibility, curiosity, sensitivity to and tolerance of other viewpoints, etc. It cultivates qualities that are conducive to pluralistic liberal political culture, perhaps more so than the metaphysician’s conviction that his personal beliefs and values are timeless and universal. Although Rorty appears to share these views (cf. [1], pp. 61, 80), they presumably answer the wrong question, as these are public uses, whereas ironism is for him a private not a public matter. Whatever uses ‘we’ might find for ironist writings should therefore be not collective-we uses, but we-as-individuals uses. In this respect an individual might, for example, find their life enriched or improved by the intellectual openness and moral sensitivity to alterity that ironist theory encourages, or by the liberation it brings from the mindset of absolute commitment. But whether and how an individual benefits from ironist discourse ultimately remains a contingent matter—we-as-individuals have to wait and ‘see whether we can find a use for it’. In that case, however, Rorty has good reason to leave the question open: if the private and individual uses of ironism are contingent and many, there will simply be no common denominator.
The implication that CIS lacks a collective-we function might be taken to count against an ironist reading of the work. One reason for thinking this is that CIS also claims to have a public side. Another is that in the context of Rorty’s later work it might be felt that the public function of philosophy emerges as more important—after all, I have already conceded that CIS is Rorty’s only work of irony.26 Why would someone interested in prioritizing public affairs over the private write a book of post-theoretical irony? A simple answer to this is: because of the threat (highlighted by Rorty, [1], p. xv) that ironist writing appears to pose to liberal public discourse. Writing an ironist work of the kind I have suggested CIS is could then be seen as also having the social-political function of eliminating this apparent threat. After all, Rorty’s discussion of ironist discourse presents it as culminating in the abandonment of theoretical aspirations and as limited in its application to the private sphere. Further, CIS itself might serve as a counterexample, demonstrating (as pointed out above) that ironist writing is not ‘intrinsically hostile’ to liberal politics. Finally, having written one book of this kind, there would be no need to keep repeating the exercise—as the supposed threat to liberal politics would already have been eliminated. In this way, reading CIS as an ironist work would be consistent not only with the text-internal evidence, but with the broader context of Rorty’s work and particularly its subsequently less ironist and more public orientation.
However, my aim here has not been to assess the philosophical virtues and value of ironist discourse itself, but to consider its importance in interpreting CIS. In this respect the point of an ironist reading is simply to allow this work to be recognized as much richer and more sophisticated than might otherwise be thought. Appreciating this complexity is of some relevance to many critical responses to the work since—as just set out—these often result from a failure to take seriously the work’s ironism. Beyond this, however, appreciating the work’s sophistication as a work of irony parallels the appreciation of a complex work of literature or visual art. And if we look to his assessment of Derrida one last time, this again seems to have been Rorty’s intention: to produce an erudite work of ironist philosophical writing, the ultimate aim of which is that we should just ‘sit back and enjoy it’.


I would like to thank Sebastian Gardner, James Tartaglia and the anonymous referees for helpful comments on earlier versions of this paper.

Conflict of Interest

The author declares no conflict of interest.


  1. Richard Rorty. Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press, 1989. [Google Scholar]
  2. Richard Rorty. Philosophy and the Mirror of Nature. Princeton: Princeton University Press, 1979. [Google Scholar]
  3. Richard Rorty. Philosophy and Social Hope. London: Penguin, 1999. [Google Scholar]
  4. Richard Rorty. “The Very Idea of Human Answerability to the World: John McDowell’s Version of Empiricism.” In Truth and Progress: Philosophical Papers. Cambridge: Cambridge University Press, 1998, Volume 3. [Google Scholar]
  5. Ludwig Wittgenstein. Philosophical Investigations. Oxford: Blackwell, 2001. [Google Scholar]
  6. Donald Davidson. “A Nice Derangement of Epitaphs.” In Truth, Language, and History. Oxford: Clarendon, 2005. [Google Scholar]
  7. Donald Davidson. “On the Very Idea of a Conceptual Scheme.” In Inquiries into Truth and Interpretation. Oxford: Clarendon, 1984. [Google Scholar]
  8. Richard Rorty. “The World Well Lost.” In Consequences of Pragmatism. Minneapolis: University of Minnesota Press, 1982. [Google Scholar]
  9. Donald Davidson. “The Social Aspect of Language.” In Truth, Language, and History. Oxford: Clarendon, 2005. [Google Scholar]
  10. Michael Dummett. “The Social Character of Meaning.” In Truth and other Enigmas. Cambridge: Harvard University Press, 1980. [Google Scholar]
  11. Edmund Husserl. Ideas. General Introduction to Pure Phenomenology. Translated by William Ralph Boyce Gibson. London: Allen & Unwin, 1976. [Google Scholar]
  12. Paul de Man. “The Concept of Irony.” In Aesthetic Ideology. Mineapolis: University of Minnesota Press, 1996. [Google Scholar]
  13. Richard Rorty. “Wittgenstein and the Linguistic Turn.” In Philosophy as Cultural Politics: Philosophical Papers. Cambridge: Cambridge University Press, 2007, Volume 4. [Google Scholar]
  14. Gilbert Ryle. The Concept of Mind. Chicago: University of Chicago Press, 1949. [Google Scholar]
  15. John Horton. “Irony and Commitment. An Irreconcilable Dualism of Modernity.” In Richard Rorty. Critical Dialogues. Edited by Matthew Festenstein and Simon Thompson. Cambridge: Polity, 2001. [Google Scholar]
  16. Nancy Fraser. “Solidarity or Singularity? Richard Rorty between Romanticism and Technocrary.” In Reading Rorty. Edited by Alan Malachowski. Oxford: Blackwell, 1990. [Google Scholar]
  17. Thomas McCarthy. “Private Irony and Public Decency: Richard Rorty’s New Pragmatism.” Critical Inquiry 16, 2 (1990): 355–70. [Google Scholar] [CrossRef]
  18. Daniel Conway. “Taking Irony Seriously: Rorty’s Postmetaphysical Liberalism.” American Literary History 3, 1 (1991): 198–208. [Google Scholar] [CrossRef]
  19. Sterling Lynch. “Romantic Longings, Moral Ideas and Democratic Priorities: On Richard Rorty’s use of the Distinction Between the Private and the Public.” International Journal of Philosophical Studies 15, 1 (2007): 97–120. [Google Scholar] [CrossRef]
  20. James Tartaglia. “Did Rorty’s Pragmatism Have Foundations? ” International Journal of Philosophical Studies 18, 5 (2010): 607–27. [Google Scholar] [CrossRef]
  • 1I return to the question of whether the reading suggested here fails to respect Rorty’s intentions in Section 5 below.
  • 2This parallels Rorty’s earlier conclusion in the Mirror of Nature that there is no such thing as a privileged representation of the world ([2], p. 212).
  • 3Rorty defines a liberal as someone who thinks ‘cruelty is the worst thing they do’ ([1], p. 74).
  • 4Rorty was later careful to disavow the found/made contrast (e.g., [3], p. xvii f.). However, he clearly continued to believe that the ‘linguistic turn’ had achieved ‘a turn away from the very idea of human answerability to the world’ ([4], p. 142 f.).
  • 5As Heidegger’s view of disclosure clearly aims to. This failure is all the more notable, given Rorty’s acknowledgement of the ‘platitude that the world may cause us to be justified in believing a sentence true’ ([1], p. 5; my italics) as opposed to simply causing belief in their truth.
  • 6See Davidson’s [9] defence of the primacy of the idiolect, as opposed to Dummett’s [10] emphasis on the primacy of a linguistic community.
  • 7It is, therefore, particularly interesting that when his attention turns to political liberalism Rorty writes: ‘Nothing requires us to first get straight about language, then about belief and knowledge, then about personhood, and finally about society.’ ([1], p. 55) After all, this appears to be what he has done in the preceding three chapters of CIS.
  • 8See Rorty’s discussion of Heidegger: ‘Heidegger thought he knew some words which had, or should have had, resonance for everybody in modern Europe’, whereas Rorty views such resonances as ‘a private and idiosyncratic matter’ ([1], p. 118 f.).
  • 9[1], p. 143. cf. p. 120, where the contrast is glossed simply as ‘the distinction between duty to self and duty to others’.
  • 10[1], p. 196. Liberal political culture is to be provided with ‘an improved self-description’, allowing it to live up to ‘its own self-image’ ([1], pp. 52, 60).
  • 11Is it plausible that Rorty (given his training) failed to notice the proximity of his position to traditional idealism? Or that, having made the public-private distinction central to his position, he was unaware of the need to explicate that distinction and of its implications for his own work?
  • 12[1], p. 96 f. This parallels Rorty’s approach in the Mirror of Nature, in which the therapeutic motivation was explicit ([2], p. 7).
  • 13Cf. ‘novels are a safer medium than theory for expressing one’s recognition of the relativity and contingency of authority figures’ ([1], p. 107)—significantly, Rorty is not denying that theory can express such a recognition; novels are simply a ‘safer medium’.
  • 14‘Ironist theory ran its course…’ ([1], p. 120; for similar retrospective references cf. pp. 136 and 97).
  • 15‘The past, for the ironist, is the books which have suggested that there might be such a thing as an unironizable vocabulary’ ([1], p. 98). A parallel claim about the background of analytic discourse is made in the Mirror of Nature ([2], p. 8).
  • 16Husserl’s epoché preserves cognitive access to intentional structures, but suspends or ‘neutralizes’ judgements concerning the existence of objects and states of affairs referred to ([11], p. 306 f. (§109)).
  • 17It would be a way, in Rorty’s earlier idiom, of ‘continuing the conversation of the West’ ([2], p. 394) by stimulating or provoking objections from others.
  • 18In saying this I am not suggesting Rorty would have rejected the above reading. Indeed, I conject that Rorty’s own ethos of playful pluralism would have led him to acknowledge open-mindedly the possibility of an ironist reading of CIS, even if this wasn’t his own guiding intention.
  • 19De Man’s own discussion is based on a reading of Friedrich Schlegel’s conception of irony and reframes Schlegel’s concerns in terms of whether irony is a ‘trope’. Here I shall interpret de Man’s claims more generically—hence ‘appropriation’—while hopefully respecting his underlying motives.
  • 20Our assumption that assertion is the default illocution is challenged so that it is unclear how to take sentence-level contents. As de Man ([12], p. 167) puts it: ‘what is at stake in irony is the possibility of understanding […], the possibility of deciding on a meaning or on a multiple set of meanings’.
  • 21This seems to be implied by a 2006 discussion of therapeutic readings of the Tractatus as a self-consuming artefact. While accepting that such readings capture Wittgenstein’s intentions, Rorty there states that he himself has no interest in approaches that ‘explode illusions of sense from within’ and that for him ‘the point of reading philosophy books is not self-transformation but rather cultural change’ ([13], p. 169). The latter claim is clearly difficult to reconcile with the emphasis on (‘private’) ‘self-creation’ in CIS, and these comments could plausibly be read as a change in his metaphilosophical position. However, if this was already Rorty’s view 17 years earlier, then, the above considerations apply. At the end of the paper I suggest one way in which writing a work of post-theoretical irony might form part of a longer term project in which the social-cultural orientation is predominant.
  • 22For example: ‘The descriptions given by philosophers of their own objectives and their own procedures have seldom squared with their actual results or their actual manners of working’ ([14], p. 321).
  • 23Cf. [15], p. 22—Rorty is further suspected of an antidemocratic elitism (cf. [15], p. 22 and [16], p. 309) and of severing the connection between politics and theory ([16], pp. 315 ff.; [17], p. 366 f.).
  • 24As Conway ([18], p. 200) puts it. Similarly, Lynch ([19], p. 107) criticizes the public-private distinction as an ‘attempt to formulate (perhaps inadvertently) a timeless a priori moral principle [...]. This is, of course, inconsistent with his anti-essentialism.’
  • 25Cf. [1], p. 87. As Tartaglia [20] shows, Rorty’s pragmatism faces similar difficulties to his ironism in coherently balancing apparent theoretical commitments with a metaphilosophical emphasis on non-commitment to theoretical claims.
  • 26Note, however, that it is debatable whether later work should be relied on in interpreting CIS. The fact that, just as any philosophical text, CIS had the potential to be developed in different directions (e.g., in a more ironist direction) should make us wary of assimilating it to the perspective of subsequent works.

Share and Cite

MDPI and ACS Style

Inkpin, A. Taking Rorty’s Irony Seriously. Humanities 2013, 2, 292-312.

AMA Style

Inkpin A. Taking Rorty’s Irony Seriously. Humanities. 2013; 2(2):292-312.

Chicago/Turabian Style

Inkpin, Andrew. 2013. "Taking Rorty’s Irony Seriously" Humanities 2, no. 2: 292-312.

Article Metrics

Back to TopTop