Contingency and Pessimism:
In Contingency, Irony and Solidarity, Richard Rorty does two things. First, he summarizes the results of historicist and antifoundationalist heroes from Nietzsche to Donald Davidson: they agree that everything is contingent. Second, he tells us what people who accept that everything is contingent can practice in place of philosophy (that is, metaphysics, the pursuit of necessary principles). On the one hand, they can indulge themselves in the poetic "self-creation" of the "edifying" or "perfectionist" philosophers, as long as they do not pretend to occupy some vantage point from which to survey human nature or any other universal principles; and on the other hand, they can promote the values of a specific intellectual or political tradition, as long as they do not pretend to have reasons for so doing which extend beyond that tradition. But if they have understood the implications of contingency they will not try to integrate or synthesize these two different impulses; they will realize that "there is no way to make both speak a single language."
I do not think Rorty has given any good reasons for accepting this, however; such a compartmentalization of language does not follow from the fact that everything, including our language, is contingent -- at least not on any reasonable interpretation of what "the contingency of language" can amount to. The most unfortunate consequence of this compartmentalization is that it prohibits us from being guided by any positive ideal of the person within society: Rorty thinks that in the absence of a metaphysical account of human nature, reflections on freedom or human potential must take the form of strictly individual achievements. He invokes tradition as the source of moral and political principles; but he rejects the traditional connection between justice and freedom -- or more specifically, the traditional concept of the good man as one whose sense of justice flows from his personal experience of freedom. This rejection is based on the idea that language itself is in the first instance a strictly individual achievement: all language begins as the opaque, non-referential wordplay Rorty calls "metaphor", and gains meaning only through habitual use. Thus the discontinuity between freedom and solidarity is grounded in a discontinuity between creativity and knowledge -- the former being, as he says repeatedly, blind and purposeless (even though the latter is but a fixation of the former). In this way 'contingency' becomes a name for incoherence -- the incoherence of cultural evolution, and of tradition itself as embodied in creative individuals.
The notion of contingency is exploited in a misleading way in Rorty's discussion, so as to offer a series of false dichotomies: between metaphysics and a kind of pessimism which is merely an inversion of metaphysical optimism; between theism and a kind of humanism that ignores the other creatures and natural things around us (and the natural-historical forces within us); and ultimately, between Absolute Knowledge and incoherence. I want to suggest a genuinely non-metaphysical way past all these dichotomies, by calling attention to the discursive structure Rorty himself relies on as an alternative to theory in the classical sense. Rorty characterizes both philosophy and the discourse of social improvement as narrative: on the one hand the perfectionist's redescription of history in her own image; on the other, the socially conscious novel or the utopian historical narrative. As the category that spans science, history and poetry, narrative bears the weight, in Rorty's discussion, of all the coherence and rationality left to us in a post-metaphysical culture. But when he talks about the possibilities of philosophy, Rorty speaks as if the only alternative to theoria were the most subjective and random kind of poetic creation. In this paper I will first identify and criticize Rorty's concept of contingency in meaning, which puts "metaphor" and sheer novelty in place of coherence; and then I will try to show why the coherence of narrative need not be constrained in the ways Rorty insists.
a. Finding Meaning
If Rorty's book is a series of false dichotomies, then the foundation of the book is one of the most venerable false dichotomies: the question as to whether knowledge is a result of discovery or creation. Rorty himself characterizes this as a "seesaw" we need to get off of; but, as with the other dichotomies, he undertakes one alternative (always the side of "reversed Platonism") as a matter of practical rhetoric, even while officially denying the choice. So it is when he favors "the vague, misleading, but pregnant and inspiring thought that truth is made rather than found." The way in which this thought can mislead us is familiar from philosophical history: it puts the human subject in place of God, as a creator ex nihilo. It rejects the pure passive receptivity of naive empiricism by abandoning receptivity altogether. The right way to get off this seesaw, it seems to me, would simply be to look more carefully at the ways in which creation and discovery are interconnected throughout the evolution of language and experience. Writers like Wittgenstein, Gadamer and Goodman could help us with this. Instead, Rorty suggests that language is created in the black hole of the "strong poet’s genius, a product of pure imagination that only later somehow congeals into criteria and other terms of knowledge.
In criticizing his view, I am not saying that Rorty has failed to account for truth really being "out there" -- on the contrary, I agree with him that to demand such an account (in the global sense that would answer to skepticism) is to wrongly treat whole systems of discourse in ways that make sense only for sentences within a system. A sentence can refer to what is "out there" because the language game to which it belongs at once establishes what kinds of things can be said to be (out there) and what counts as referring to them; but the language game itself (which includes the associated beliefs, perceptions and behavior) has no such path of appeal. From this Rorty concludes that we should give up "the notion of language as something which can be adequate or inadequate to the world or to the self." And I agree with this too, if 'adequacy' means the correspondence of language, considered as a mere representational medium, to its objects. But language can be more or less adequate to needs we have which are not of our making; and this functional adequacy is enough to back up our sense of finding meaning in the world.
You will notice that I speak of finding meaning, whereas Rorty speaks of finding/making truth. My reason for shifting terms is that I think Rorty draws consequences from the contemporary arguments about truth, which he applies in effect to the concept of meaning; and that the consequences are unnecessarily nihilistic -- that is, skeptical about the extent of the coherence and comprehensiveness possible in our understanding of the spatiotemporal neighborhoods we occupy. To deny that the adoption of a new language game is a discovery of pre-existing truth is just to guard against metaphysical justification of the language game; but to deny that it is a discovery of meaning is to suggest that every innovation is a simply arbitrary (not initially meaningful) step in the evolution of culture. And this does not follow from the abandonment of metaphysics; for it is the coherence of our diachronic involvement in the web of experience and practice -- not metaphysical assumptions about what corresponds to our representations -- that prevents what we say from being arbitrary or subjective. Rorty says
the world does not speak. Only we do. The world can,
once we have programmed ourselves with a language, cause us to hold beliefs. But it cannot propose a language for us to speak.
He intends this as a denial that anything has "an intrinsic nature to be expressed or represented," and he holds that such "intrinsic natures" or essences can only be understood as a result of "a divine creation, the work of someone who had something in mind, who Himself spoke some language in which He described his own project." Such a picture supports the treatment of a whole language system as "true", as if the system could correspond to a part of the divine language inherent in the world order.
Let us grant for the sake of argument that all metaphysics amounts to an elaboration of this picture; and let us agree in abandoning it -- the world is not a book waiting to be read by us. But why say that we have "programmed ourselves with a language"? How could we do this? We would have to know the meaning of the "program" before we could accomplish such a feat; but this is impossible in Rorty's terms. His effort to portray modern philosophy as vainly repeating the search for essences, first with regard to Mind and then Language, depends on emphasizing that all characteristics of minds and languages can be seen as the result of unsurveyable causal sequences that could have been otherwise. In other words, we are "programmed" directly and indirectly by countless factors, of which Rorty delights in offerring the most random examples.
...for all we know, or should care, Aristotle's metaphorical use of ousia, Saint Paul's metaphorical use of agape, and Newton's metaphorical use of gravitas, were the results of cosmic rays scrambling the fine structure of some crucial neurons in their brains. Or, more plausibly, they were the result of some odd episodes in infancy -- some obsessional kinks left in these brains by idiosyncratic traumata. It hardly matters how the trick was done.
Let me emphasize that I have nothing against looking for such causal connections; I have not misunderstood Rorty to be engaged in a reductionism that would privilege the rays, neurons or obsessions over other terms. As he says of Freud, "He just wants to give us one more redescription of things to be filed alongside all the others." Explanation in terms of aesthetic, moral, political (and even religious) terms therefore may carry just as much weight as physical and physiological explanations. But there is a rhetorical difference: speaking in the latter terms conveys the impression that the phenomena we are describing are intrinsically meaningless (the meaning is all "in us"). We should not give too much weight to this impression. We can speak of discovering meaning, not in the sense of finding a ready-made language (or nonlinguistic essences that would call a specific language forth), but of experimentally working out a set of satisfactory responses to the natural and cultural environment in a historical context. We do not think up any language "program" in a worldless imagination or meaningless environment.
b. Significance and Coherence
There are many ways of drawing a distinction between the meaningful and the meaningless. The one dominating Rorty's discussion depends on the connection between meaning and intention: an act is imbued with the agent's meaning, whereas the relation of physical cause to effect is "blind", not leading anywhere, a matter of fate or chance (although neither of these terms quite jibes with causality). The discussion makes use of the phrase "the blind impress all our behavings bear" from a poem by Philip Larkin, in order to suggest that the contingency of our language makes it impossible for us to see the essences or grasp the real meanings of things. Again, the bugaboo is the idea of the world as the embodiment of a divine intentionality. Rorty wants a culture
in which no trace of divinity remained, either in the form of a divinized world or a divinized self. Such a culture would have no room for the notion that there are nonhuman forces to which human beings should be responsible.
Members of this culture should not "derive the meanings of their lives from anything except other finite, mortal, contingently existing human beings."
Among several problems raised by this passage, one seems especially obvious: the notion that we have no responsibilities toward, nor can we derive a sense of significance from, anything nonhuman (because only humans have intentions). I don't know whether obsession with the human/divine contrast (metaphysically understood) has caused Rorty to ignore the thought of our responsibilities toward other species and our environment, or if he would deny such responsibilities. So I will not spend time arguing against an egregious anthropocentrism; but I take these remarks as symptomatic of preoccupation with a narrow notion of meaning and significance. (You may say that I am dwelling on what are merely rhetorical excesses; but these excesses serve to cover up the real philosophical problems.)
Another way of drawing the distinction between the meaningful and the meaningless would be to start from the concept of narrative which Rorty repeatedly urges upon us, as when he says that
What binds societies together are common vocabularies and common hopes. The vocabularies are, typically, parasitic on the hopes -- in the sense that the principal function of the vocabularies is to tell stories about future outcomes which compensate for present sacrifices. 
to sketch a narrative of our own development, our
idiosyncratic moral struggle, which is more finely
textured, far more custom-tailored to our individual case, than the moral vocabulary which the philosophical tradition offerred us. 
What is meaningful could be said to be that which fits into one of our narratives -- into some implicit or explicit story-line or poetic argument with which we are involved directly or vicariously. The meaningless would then be that which has no such place in our story, and so is just an excrescence.
This way of drawing the distinction would cut across the intentional/causal boundary which seems to lie behind Rorty's exclusion of the nonhuman; and thus it would avoid the misleading suggestion that we can only find what we (or other humans) have made. It is precisely the "fine texture" of vital narrative structures that backs up our sense of discovering unavoidable themes in life, because we feel constrained in the "invention" as well as the "application" of these structures. (Even in the grip of an obsession, my fantasies keep revealing features of the realities they butt up against.)
So when Rorty says that "Any seemingly random constellation of such things [the color of a leaf; the feel of a piece of skin] can set the tone of a life", he is being misleading. The thing that becomes a motif in my life does so for a reason I can usually explain to you -- autobiographies are intelligible. The fact that Rorty says "seemingly random" saves his statement from being blatantly false -- as it is, it underscores the motive of self-expression in autobiographical narrative, as contrasted with the aim of generality in some metaphysically-minded way of interpreting a life in terms of universal ideals. But autobiography is not composed solely of proper names (anymore than metaphysics is free of particular examples or metaphors); zeal to eliminate metaphysical pretensions should not make us embrace a dichotomy between universality and privacy. This false dichotomy so pervades Rorty's discussion that it is as if he had said that the elements of our lives are all random; that is, "really" random -- rather than taking randomness as relativized to a particular perspective (e.g. that of a biographer who has not yet grasped the significance of some detail). It is as if Rorty replaced the cosmic Order with a cosmic Disorder.
Rorty praises Nietzsche, Freud and Bloom for their "common strategy": "to substitute a tissue of contingent relations ... for a formed, unified, present, self-contained substance, something capable of being seen steadily and whole." Again, the easy part is to give one more kick in the ribs to the Metaphysics of Presence: neither poems, nor lives, nor worlds should be viewed as "self-contained substances." But what about the alternative, the tissue of relations? That it stretches into the past and future is certain; and that it is not seen with a steady gaze is easy to admit as well. But must we also deny that it lacks any kind of form or unity, that is, coherence?
Consider once more the model of narrative. It is sequential; it involves surprise, reversal of fortunes, change of perspectives; it starts from details which are seemingly random. And it manages, by this very process, to make us see things which initially looked like "one damn thing after another" as contributing to and partaking of a larger meaning. This meaning has nothing to do with an eternal intentionality; it is just the ordinary kind of meaning that makes lives, families, cultures (and their environments) hang together in the imperfect ways they do. Such coherence is temporal -- understood by beings thrown into the world with needs and desires that affect their perceptions and expectations (of the world, themselves and each other) -- and by that very token meaningful. There is no such thing as a future for us which is approached (or a past which is traced) otherwise than in the context of a set of anticipations. Thus to relegate form and coherence to the sphere of our subjective projection (as opposed to what is really out there) is to commit the very error Rorty is against. This error is embodied most clearly in Rorty's doctrine of knowledge as "literalized metaphor."
c. Metaphor and Vision
I have already alluded to Rorty's unabashed embrace of the Bloomian "strong poet", meaning roughly: one so obsessed with her own identity and language as to deny the influence of her predecessors (by "misreading" them). This denial epitomizes the discontinuity which Rorty posits between what is known and what is created, or as he would say, between old language and new. Rorty's "ideally liberalized state" is inhabited by those who see their language, conscience and highest hopes "as contingent products, as literalizations of what were once accidentally produced metaphors." The word 'accidentally' here corresponds to the "blind" relation between events which renders meaning as nothing but our projection. Pure creativity is ultimately indistinguishable from randomness.
The hope of such a [strong] poet is that what the past tried to do to her she will succeed in doing to the past: to make those very causal processes which blindly impressed all her own behavings, bear her impress.
So the poet, "the vanguard of the species" , does not really interpret or reveal; she only gestures and projects.
A more moderate way of rendering the contingency of language might be to say: although we find ourselves situated in an unsurveyable causal nexus, we select and interpret those features of our background which best fit into the story we are trying to tell of ourselves. And we find ourselves already in various stories, both as narrators and as characters. The story as such both precedes and follows the selection of details -- there is a circular relation of whole and parts here. What makes for a "best fit" can be explained either externally, i.e. in terms of causal relations between events described and their descriptions, or internally, in terms of the themes and horizons of the narrative itself.
But that means that the "impress" is not a blind one, regarded in either direction. A series of gradations runs from the countless causal connections we filter out or never notice in our narration, to the contextual details we tacitly acknowledge, to the central influences we claim as part of our destiny. All these influences could, in one sense or another, have been otherwise; but given the way things have turned out, we need to interpret them in a limited range of ways, with certain features in the foreground. As Wittgenstein remarked, in order to stress that even rituals are not arbitrary, opaque gestures: "If the flea developed a ritual, it would be concerned with the dog."
To recognize this is to overcome the "relativist predicament" Rorty scorns, but of which he is, in my view, still a victim. He wants to avoid the scheme/content dogma so forcefully criticized by Donald Davidson ; but he is not content with Davidson's explicit criticisms. Instead he says that one will cease to be afflicted with the dogma "if one accepts Davidson's claim that new metaphors are causes, but not reasons, for changes of belief." While Rorty notes that Davidson "cannot be held responsible" for his interpretation and extrapolation of Davidson's views on metaphor, comparing this interpretation with the original reveals how the contingency of a holistic, coherence-based view can be turned into incoherence and blindness.
Rorty fixes on the idea that new beliefs are stated in terms which are originally metaphors; and that a belief, the holding of a proposition to be (literally) true, cannot be compared with any metaphorical meaning -- because there is no such thing. "To have a meaning is to have a place in a language game. Metaphors, by definition, do not." And yet, according to Rorty, all literal meaning is just congealed metaphor. Thus Rorty (and allegedly Davidson) "see language as we now see evolution, as new forms of life constantly killing off old forms -- not to accomplish a higher purpose, but blindly." At this point we can treat the reference to "higher purpose" as a red herring (although we could, if we wish, translate it into narrative terms); but we must consider how the "blindness" at the root of language (the blindness of the strong poet) is based on a peculiar notion of linguistic creativity as bereft of cognition, because of an imagined discontinuity between old language and new.
I see Rorty's view as a distortion of Davidson's in three ways.
First, contrary to Rorty's emphasis on discontinuity ("nothing in existence prior to the metaphor's occurrence is sufficient to understand the metaphorical use"), Davidson holds that metaphorical use "depends entirely on the ordinary meaning." Rorty, while acknowledging the obvious point that metaphors could not affect us if we did not recognize the words as belonging to our language, yet wants to describe the words-asmetaphors as "unfamiliar noises" which do not cross the line "between stimulus and cognition" until such time as the metaphors die -- meaning that people begin to treat them as terms "in a pattern of justification and belief." But how and why they begin to do this is inexplicable on his account: successful metaphors are "idiosyncracies which just happen to catch on with other people." This discontinuity between old and new causes Rorty to view as an "embarrassment" Davidson's straightforward acknowledgment of the cognitive basis for metaphorical use (more on this shortly).
Second, Davidson insists that "Novelty is not the issue", contrary to Rorty's emphasis on metaphor as what does not yet have a familiar place in the language game. Davidson says explicitly: "In its context a word once taken for a metaphor remains a metaphor on the hundredth hearing, while a word may easily be appreciated in a new literal role on a first encounter." When metaphors get old and die they do not necessarily turn literal in rigor mortis; they may simply become cliches, or terms of convenience which we are too lazy to replace, but which we still regard as having a secondary application distinct from the primary or "literal" one. Furthermore, we are surrounded by language creation which involves no metaphor at all. Thus Rorty seems to me mistaken when he claims to think
with Davidson, of the literal-metaphorical distinction as the distinction between old language and new
language rather than in terms of a distinction between words which latch on to the world and those which do not.
The fact is that Davidson wants to avoid this very dichotomy. He wants to explain meaning by reference to truth-conditions, that is, to beliefs and intentions, and in general to illuminate the interdependence of linguistic and non-linguistic terms -- but not by trying to explicate any metaphysical latching-on. And far from relying on mere novelty (the flip side of the opaque Humean concept of "custom"), he derides
the kind of theory that tries to derive the literal meaning of each sentence from a "standard" use. Since the literal meaning operates as well when the use is absent as when it is present, no convention that operates only in "standard" situations can give the literal meaning.
Third, and most importantly, although Davidson is against attributing any "cognitive content" to metaphors, he says "Metaphor does lead us to notice what might not otherwise be noticed..." In fact, this idea, which for Rorty is just part of Davidson's "outdated rhetoric", is actually the basis of his view; for him the question is "how the metaphor is related to what it makes us see." Davidson disagrees with Nelson Goodman on this question. Whereas Goodman thinks that what metaphor reveals -- which one might call its meaning -- can be spelled out in terms of reference or "application", i.e. as a contrast between different sortings of some set of features, Davidson insists that metaphor is to be viewed in terms of its function of inciting such different sortings and comparisons -- so that no new meaning of the word enters with the metaphor. Thus, far from denying the cognitive aspect of its use, the denial of metaphorical meaning is meant to show a connection between linguistic creativity and cognition. Metaphor is based both on what we already understand, and on a range of possibilities for future understanding, some of which we must prospectively grasp as soon as the metaphor begins to work on us.
Davidson gives two reasons for denying "the thesis that associated with a metaphor is a cognitive content" : one is that much of what metaphors show us "is not propositional in character" ; the other is that it is "not finite in scope" (i.e. the audience is free to -- indeed, is encouraged to -- pursue an indefinite number of unspecified connections). Goodman plausibly replies to the second point that sometimes the intended scope is quite limited, and that incitement to notice relations is not what specifically distinguishes metaphor. But this disagreement seems mostly technical; Davidson and Goodman both seem to be describing the same situations, and not to be concerned with poetic self-expression per se. The same thing goes for the first point: Goodman has a plausible idiom in which 'knowing' is used in such a way as to be broader than the "propositional" in Davidson's sense -- it involves more, he says, than holding true beliefs. The importance of these other factors is not in question, only a decision as to whether 'knowing' is to be used strictly as a correlate of propositions, or whether it will refer as well to the activities of the knower which are required for the production, selection and interpretation of propositions.
Rorty has the narrowest and least plausible way of using
'knowing': he seems to assume it will inevitably be connected with "notions like 'reality', 'real essence', 'objective point of view', and 'the correspondence of language and reality'." When he insists that a fragment of Yeats "acquires a place in people's practices", but does not become "part of what they know" , he is opting for the tight connection between 'knowing' and 'believing'; whereas many people, especially poetry-lovers, are inclined to use 'know' in Goodman's broader sense.
Goodman and Davidson agree that the knowledge in question is not "contained" (represented or encoded) in the metaphor, nor in the set (finite or not) of features to which it applies or calls attention. And they agree in generally accepting the "inscrutability of reference", in the sense of viewing substantive, linguistic and epistemic terms as interdependent parts of holistic systems -- thus ruling out any question of language-independent essences, or of One True Language or cognitive system. But what they most importantly agree on is this: there is something for us to see in connection with the use of metaphor. This is why Rorty is justified in citing "Hesse's claim that it is new metaphors which have made intellectual progress possible." Metaphors are one way in which we expand our perspective; but such expansion is as far as possible from Rorty's view of "blind evolution".
a. Self-discovery in the community
Rorty's "poeticized culture"
would be a culture which, precisely by appreciating that all touchstones are [merely cultural] artifacts, would take as its goal the creation of ever more various and multicolored artifacts.
Would I be wrong to paraphrase this as a goal of novelty for novelty's sake? Put this together with Rorty's Nietzschean emphasis on self-creation -- the claim that "strong philosophers... are interested in dissolving inherited problems rather than in solving them," and that consequently what they really do is only to change "the way we talk, thereby changing what we want to do and what we think we are" -- and what you have is an ideal of change with no continuity -- the cultural ideal behind his picture of discontinuous language. In this view, every time we change, who we think we are is radically independent of who we have been. We are like the young rebel who cannot recognize his parents' influences in the very efforts he makes to deny them. But Rorty accepts one arbitrary restriction on this anarchy in advance; we must be tolerant, and sensitive to the pain of others -- even though we do not think there is any reason for being that way. Rorty has to erect an arbitrary barrier between the doctrine of language as metaphor (which is supposed to be the insight of the "ironist theory" of Wittgenstein, Heidegger, Derrida, et.al.). and his social concerns, because the former recognizes no constants of human nature, no rationality and no coherence.
Of course this talk of strong poets and philosophers is polemic, not theory; but what kind of polemic? One must be in dire straits for a polemic of "change for the sake of change" to be prudent. One must be desperate in order to speak at all, if one thinks that the best outcome of speaking is that one will produce nothing but "idiosyncracies which just happen to catch on with other people"; for in that case one has no reason to expect that this random catching-on will be more in accord with one's intentions (e.g. to avoid cruelty) than with their contraries. We need not have any nostalgia for Absolute Truth in order to prefer a polemic which will be rational and coherent -- which will use metaphors to enlighten, and which will tell a story of where we should be heading by reflecting on the good and bad aspects of where we have been, and on the capacities of the terrain in our neighborhood for supporting the roads we would build in order to extend it.
Rorty says that
in abandoning the traditional notion of truth, Nietzsche did not abandon the idea of discovering the causes of our being what we are. He did not give up the idea that an individual might track home the blind impress all his behavings bore. He only rejected the idea that this tracking was a process of discovery.
Discovery concerns "a truth which was out there (or in here) all the time." But again, the dead metaphysical horse carries us into unfortunate linguistic turns: 'discovery' should not only be used to connote obsession with a notion of truth "out there all the time"; for once we have established that "all the time" is a grammatical characterization, a statement about the role of a proposition in a context, there is nothing either wrong or misleading about our saying that the positron, the Mandelbrot set, or the form of constitutional democracy was discovered by the instigators of the relevant practices and language.
Ordinary language countenances talk of discovering
possibilities. Rorty's insistence that new possibilities are simply created, as if out of thin air or "blindly", suggests the attitude of change as blindly willed, rather than participated in understandingly, which was analyzed as a kind of congenital disease in Heidegger's Being and Time. His ideal of "resoluteness" is proposed in opposition to this tendency, as part of an argument for thinking in terms of a wide historical
context (unified by the theme of "Being"). He urges us to consider possibility as "higher" than actuality; and he analyzes blindness to possibility as a failure of patience, i.e. as distraction by the latest thing. This does not mean that our possibilities are "out there", as destinations to which the tracks are already laid down; indeed he prefers the metaphor of dead ends, as a symbol of contingency. But he does insist that the most finely detailed and productive undertakings of possibilities have the character of a thoughtful response.
To Rorty this smacks of theology. But the real fault he finds is that Heidegger thought his exemplification of personal freedom -- the creation of his own language and recasting of the philosophical tradition in his own image -- had anything at all to do with society at large. Rorty's prescription is to keep our responsibility toward ourselves, to self-creation, separate from our responsibilities toward others, and to accept the fact that there is no way to make both speak a single language. The denial that human possibilities can be discovered is in the service of this prescription.
It is even easier to criticize Heidegger's politics than to attack the Metaphysics of Presence; but it does not follow that the kind of perfectionism Heidegger (and countless others) pursued should be kept separate from our public concerns. Nor does it seem reasonable even to expect that this could be accomplished. Rorty says of past philosophers that they were out to "inform us what we really are, what we are compelled to be by powers not ourselves." The fault in this lies with the way they thought of the relation between what or how we already are and what we should become: they described the possibilities they envisioned as if they were somehow already actualized in some higher realm or predestined future, instead of just taking possibilities per se more seriously, as Heidegger's slogan suggests. They spoke as if their vision already contained all the important details and were not subject to crucial modifications in the course of being realized.
But there are worse errors. The fact is that they did "inform" us (give us form), to the extent that we became their ideas of humanity and lived their stories. (To quote Rorty, "They reveal us because they made us.") And as Rorty wants to insist, this was subject to "powers not ourselves" -- whether we think of these as causes in an established descriptive framework, or as lacunae that may only be intelligible in some future version. We may also say that the influence of the "strong" on the "weak" was subject to constraints inherent in the narratives already shared by both of them (although it may have been impossible to say in advance which features of the old structure would be considered necessary from the standpoint of the new). Thus there is no way I can see of rigorously distinguishing between the public conditions for working out social concerns and the equally public conditions for working out private concerns (i.e. the perfectionism which I pursue against the background of all the other vocabularies and stories at work in my head); the language games do overlap, even though they are not converging on a predestined form.
Rorty sees private perfectionism as irrelevant and possibly dangerous to liberal ideals, because he sees it as fundamentally "ironist", i.e. cognizant of its basis in pure creativity or will, unrestrained by universal principles. But we need not adopt this view of ourselves as self-caused divinities, dictators or strong poets; and if we do not, we have no reason to isolate our need for self-expression from our respect for others in society. The resoluteness with which we adhere to the major themes of our lives is precisely not based on a naked "I gotta be me", but on what Heidegger calls "repetition", or as I would say, on faithful retelling of cultural and personal themes -- that is, on the plausibility of the most comprehensive narratives we can construct from the elements of our historical situation.
Irony in Rorty's sense is not a hindrance but rather a precondition for the weaving of vital narratives, which progress by surprise and expansion of perspective rather than by adherence to the systematic unity of an external perspective. Rorty shrugs off this contrast between system and narrative: "Ironist theory ran its course in the attempt to achieve [the synthesis of the public and private] through narrative rather than system."  But this is only because he insists on speaking of "theory" where it is no longer appropriate, and taking people like Heidegger as examples; whereas on my view Heidegger is simply a bad example, who failed to live up to his intimations, and to his quite reasonable descriptions of how lives can be more rather than less coherent.
The notion that the unity of the story I can tell about myself and the unity of the story we can tell about our culture are strictly different sorts of unities, is an unsupported dogma. Its weakness is apparent as soon as one considers it in its own terms (as soon as one considers the actual processes of such storytelling), rather than in terms of some metaphysical attempts to deduce these unities from theoretical principles. But this rather theoretical dogma is bound up with an even more intriguing empirical one, to which I now turn. It is the notion that the only available way of grounding our social discourse in the Western liberal tradition is by reference to the avoidance of cruelty and humiliation. It says, in other words, that there is no positive, only a negative goal for a liberal society:
I do not think that we liberals can now imagine a future of "human dignity, freedom and peace." That is, we cannot tell ourselves a story about how to get from the actual present to such a future.
b. Pain and Love
To sum up: Rorty wants to keep the Romantic exaggeration of making over finding, which he supports with a bad argument about metaphor and its relation to knowledge, because he wants us to abandon the attempt to find connections between what we say about the world when we are concerned with our own places in it, and what we say when we are concerned with the welfare of others. He emphasizes the nihilistic aspects of "ironist theory" (philosophy as practiced by the demolishers of metaphysics), so that we will be sure to keep philosophy in its effete and ineffectual place. His reliance on distinctions and strategies which he officially rejects is motivated by this dogma:
The sort of autonomy which self-creating ironists like Nietzsche, Derrida or Foucault seek is not the sort of thing that could ever be embodied in social institutions. Autonomy is not something which all human beings have within them and which society can release by ceasing to repress them. It is something which certain particular human beings hope to attain by selfcreation, and which a few actually do.
I cannot meet this claim head-on because, as already indicated with regard to Heidegger, I do not take these German and French heroes as exemplars of coherence (whatever their contributions to our understanding of the relations between coherence and correspondence, meaning and truth, etc.). Once again we must resist a bad dichotomy: between taking 'human nature' in the metaphysical sense of a soul or eternal form, and denying the vast and rich range of commonalities we have with the overwhelming majority of humans in historical times. I do not take the distinction between the strong and the weak as crucial here; and indeed it seems that Rorty sometimes forgets the important point he makes in connection with Freud, that
nobody is dull [i.e. weak] through and through, for there is no such thing as a dull unconscious. What makes Freud more useful and more plausible than Nietzsche is that he does not relegate the vast majority of humanity to the status of dying animals. For Freud's account of unconscious fantasy shows us how to see every life as a poem... 
But more important to me than the question of human differences is the claim that the kind of ideals generated in connection with self-discovery/creation are incommensurable with the language of public policy; and that consequently we can only say what to avoid, rather than what to pursue, in our own narratives of society. Here once more I see an inversion of metaphysical dogma lurking in the post-metaphysical rhetoric, which I formulate as follows: pain is more real than love.
Rorty does not, so far as I have noticed, speak of love, only of kindness; but he does speak of accepting an ever-greater range of human differences "as included in the range of 'us'." For him the only basis of this acceptance is horror and pity: "For all we share with all other human beings is the same thing we share with all other animals -- the ability to feel pain." Of course this is more rhetorical excess; he is not talking about DNA or higher linguistic capacities, but about our sharing sparks of the divine. The reason I want to bring in the concept of love is that in my view love is the way in which we accept and affirm others just the way they are -- that is, precisely without regard to any "core self" or other common characteristic. Love is blind, partly in the sense of Rorty's discussion: it is especially able to select features and make connections in accord with its own impulses, without pretense of objectivity or logic. But it also enables us, on occasion, to see things we would otherwise miss.
Perhaps love is too fickle -- assuming that we cannot turn
it from the realm of becoming toward that of eternal forms -- to serve as a principle in the construction of social narratives. It has always been the job of the philosopher, the poet and the prophet (as well as the martyr) to deny what seems to be the common sense or "realistic" view of human nature, namely that it is motivated more surely and effectively by pain and fear than by anything else. The classical philosopher says: love is more real than pain; it defeats pain (as well as all manner of ugliness and meaninglessness) through transfiguring it, seeing its "true" or "higher" meaning. Of course this is not ordinary love, with its variability, but the Higher Love.
Rorty is against this old idea. Many of us today believe only in the power of ordinary love, which although contingent, can sometimes be as beautiful and healing as we need it to be. But Rorty is pessimistic:
Faced with the nonhuman, the nonlinguistic, we no longer have an ability to overcome contingency and pain by appropriation and transformation, but only the ability to recognize contingency and pain.
His noteworthy reversion to an epistemic vocabulary ('recognize') here suggests that in this context the appearance/reality distinction is to be enforced in a particularly strict way: pain is real; seeing it another way is illusion. But let us look past this outdated rhetoric, to the ways in which we actually negotiate with pain and ugliness in our world. First of all, how can Rorty maintain the separation between recognition and appropriation in the above passage? Can't I recognize a pain as my own, even (as Wittgenstein argued) in someone else's body or life? And in that sense I can transform it -- not only by sublimating it, but also in the sense in which the pain remains pain though it becomes more than just pain (as my pain, or our pain, it may serve new and more far-reaching functions, opening up new possibilities and metaphors). Of course there is an ambiguity in speaking of transforming pain through redescription. Certainly it is deplorable when we deny or callously seek to justify the pain of others (or ourselves) by such "transformation". Philosophers should provide neither defense nor comfort for the Christian Scientist who denies medical aid to a child, preferring to "see the child as perfect." But some kinds of pain and humiliation can, under the right conditions, evolve into or become part of something else, and language can be part of that process.
This brings us back to the dog and the flea. If language is part of a causal evolutionary network, then there is no reason why we cannot "adopt the language" of the threatening world, as host and parasite adopt each other's ways, or as defense mechanisms and their secondary effects become self-perpetuating factors in the psychic economy. Here it does seem better to replace talk of the world speaking to us with Heidegger's metaphor of language/world bespeaking us -- not because one wants a secular version of a destined order, but because one wants to recognize the full extent of our contingency.
All I am arguing is that the way we respond to pain, and to those "deeper needs" which are only satisfied through something like love, are as intertwined as any other strands in the fabric of behavior and experience. We "tell stories about future outcomes which compensate for present sacrifices" in the context of our own lives just as much as in the context of our culture. In both cases the "compensation" always remains in doubt. But we can no more imagine pain eliminating love than love eliminating pain -- in either case the creatures who could be that way would no longer be recognizable as us.
Contingency does not require pessimism, although it may temper optimism. My view is that the elimination of metaphysics leaves most of our language, conscience and highest hopes pretty much as they were. As Wittgenstein thought that solipsism, when thought out, concides with realism, I think that perfectionism can be made to concide with pragmatism: I think that we can translate most of the great metaphysical themes into historical, narrative terms, without diminishing our appreciation of the possibilities of creative understanding.
 Rorty, Richard Contingency, Irony and Solidarity (hereafter CIS), Cambridge University Press, Cambridge 1989, p. xv
 CIS p.53
 The term is Harold Bloom's (see below p.7). In this context the revolutionary scientist or any other innovator is included as ultimately a kind of poet.
 CIS p. 10
 CIS p. 4
 CIS p. 21
 CIS p. 17
 CIS p. 39
 CIS p. 45
 CIS p.86
 CIS p. 32
 Of course digressions, mistranscriptions and other excrescences do appear in most narratives; and since we always view life (unlike a novel) without closure, we may never be certain whether an apparent excrescence will turn out to have hidden significance.
 CIS p. 37
 CIS p. 41
 cf. Bloom, Harold The Anxiety of Influence Oxford Univ. Press, New York 1973. I am by no means denying the value of Bloom's insights for criticism, nor their pertinence to the reading of
 CIS p. 61
 CIS p. 29
 CIS p. 20
 Wittgenstein, Ludwig "Remarks on Frazer's The Golden Bough", trans. by Manser, p.8
 Davidson, Donald "On the Very Idea of a Conceptual Scheme" in Inquiries into Truth and Interpretation, Clarendon Press, Oxford 1984
 CIS p. 50
 CIS p. 18
 CIS p. 19
 "Hesse and Davidson on Metaphor" in Proceedings of the Aristotelian Society Supplementary Volume LXI 1987 (hereafter PAS), p.289 (Rorty's paper and Hesse's response referred to in footnote 43 are grouped under the heading "Unfamiliar Noises".) Rorty is right if you take 'sufficient' in this quote strictly enough -- but then this will hold for most literal uses as well.
 "I depend on the distinction between what words mean and what they are used to do. I think metaphor belongs exclusively to the domain of use. It is sometimes brought off by the imaginative employment of words and sentences and depends entirely on the ordinary meanings of those words and hence on the ordinary meanings of the sentences they comprise." Davidson, Donald, "What Metaphors Mean" in op.cit. p.247
 PAS p.295
 CIS p. 37
 "Pragmatism, Davidson and Truth" in Truth and Interpretation edited by Ernest LePore, Basil Blackwell, 1986 (hereafter PDT)
 PDT p. 252.
 This is especially true in science and technology. 'Charm' and 'color' are not metaphors in physics, anymore than 'byte' and 'modem' in technology. Often acronyms and proper names are used where one could also have used ordinary words which would be deemed metaphors in the same application.
 CIS p. 28
 Davidson, "Communication and Convention" in op.cit. p. 275
 ibid. p. 257
 PDT p.354
 Davidson op.cit. p. 261
 Davidson p.262, quoted by Rorty in CIS p.18
 Davidson p. 263
 Goodman, Nelson Of Minds and Other Matters Harvard Univ. Press, Cambridge 1984 pp. 72-74
 Davidson's own example of calling someone a pig seems to me a good counterexample -- it doesn't invite an exploration of relations, nor is its scope particularly open-ended. To insult someone is not to try to get him to see something. Sometimes metaphor has an even less definite function -- it is just there for the sake of variation (to produce some small amusement, perhaps). And not only does dead metaphor sometimes remain metaphor, as Davidson rightly points out; but much of what is cited as dead metaphor (e.g. Davidson's "mouth of a bottle") was never "live" (in the sense of being surprising, needing interpretation, or inciting exploration) in the first place! It was simply a way of referring to features of things that hadn't been referred to before, which everyone understood immediately or with the briefest of explanations.
 Goodman, Nelson Ways of Worldmaking Hackett, Indianapolis 1978 (cf., e.g. p. 19)
 CIS p. 75 (reading 'of' for second 'and' in text)
 PAS p.294
 Rorty refers to Hesse, Mary, "The Explanatory Function of Metaphor", in Revolutions and Reconstructions in the Philosophy of Science Indiana Univ. Press, Bloomington 1980.
But Hesse has herself criticized Rorty's view of metaphor, in particular his restriction of the cognitive function to "literal language. She likens his view to "logical empiricist attempts to deal with scientific theories, attempts which neglected the dynamics of history and theory-change." ("Tropical Talk: The Myth of the Literal", in PAS p.298) Rejecting, as I have, Rorty's insistence on the discontinuity between "old" and "new", she supposes instead "that the messy transition stages are in many respects the norm of human communication."(p.304) And she emphasizes the cognitive dimension of metaphor even when it is quite ambiguous; for "ambiguity is not total -- we do know how to respond appropriately to tropical talk, we do not flounder about in morasses of uncommunication until we miraculously come upon the cleared space of the literal."(p.309)
 CIS p. 53-54
 CIS p. 20
 CIS p. 27
 CIS p. 26
 CIS p. 117
 CIS p. 120
 CIS p. 182
 CIS p. 65
 CIS p. 35
 CIS p. 192
 CIS p. 177. Rorty's focus on pain and violence is illuminated by a distinction Paul Ricoeur draws between the ethics of virtue, driven by the principle of self-esteem, and moral obligation, based on respect. He claims that it is violence that requires us to move beyond the narrative-teleological dimension of virture, to the "deontological" dimension of obligation. But he also insists: "respect does not abolish self-esteem, but includes it." ("Humans as the subject matter of philosophy", in The Narrative Path, edited by T. Peter Kemp and David Rasmussen, MIT Press, Cambridge 1989)
 CIS p. 40