A version of this paper was originally published as "The Place of Phronesis in Postmodern Hermeneutics," Philosophy Today 37 (1993): 298-305; reprinted in The Very Idea of Radical Hermeneutics, ed. Roy Martinez (Atlantic Highlands: Humanities Press), pp. 22-32. Originally presented at the American Philosophical Association, Eastern Division meeting colloquium (1990). Please cite the original publication, which is slightly different from the one presented here.




The Place of Phronesis in Postmodern Hermeneutics

Shaun Gallagher

Canisius College




The conception of paralogy, which Jean-Francois Lyotard develops in The Postmodern Condition, motivates a number of questions concerning justice and the moral life. In this paper I suggest that Lyotard's account fails to provide an adequate answer to these questions, and that a more satisfactory account of justice in paralogy can be developed by exploring the concept of phronesis. John Caputo's "ethics of dissemination," in some respects, leads us in this direction. Although both theorists attempt to develop their accounts in terms of the concept of phronesis, Lyotard reduces phronesis to cleverness, while Caputo elevates it to what he calls "meta-phronesis," a conception of how we are meant to cope under conditions of postmodern paralogy. Caputo's analysis of meta-phronesis, however, is carried out within the framework of a critique of the traditional concept of phronesis. To show how Caputo's account might lend itself to a fuller conception of phronesis, one which is appropriate to a postmodern hermeneutics, I raise some questions about (a) the radical definition of the paralogical situation offered by both Lyotard and Caputo, (b) Caputo's critique of phronesis, and (c) his characterization of meta-phronesis.


Lyotard on Paralogy and Phronesis

Lyotard relying on Thomas Kuhn's conceptions of scientific revolutions, suggests that scientific knowledge involves a search for instabilities rather than consensus. The result of "revolutionary science," Lyotard contends, is a shift in the very concept of reason. In postmodern reason there is no legitimate meta-language that would organize or provide a sense of totality or hierarchy in the scientific field; there is only a plurality of relative and local narratives or language games. This is the situation of paralogy: postmodern knowledge concerns itself with incommensurabilities, undecidables, conditions of incomplete information, fracta, discontinuities, and paradoxes.1 The recourse to metanarratives is closed off -- no more theological systematization, no more philosophy of Spirit, no more hermeneutics, no more Enlightenment ideology of emancipation (see PMC xxiii, 35). Thus Lyotard's famous definition of postmodernism: "an incredulity toward metanarratives" (PMC xxiv). The goal of science is neither consensus attainable within metanarrative, nor the construction of a stabilizing paradigm. All of this is familiar ground to the reader of Lyotard's "report on knowledge."

In The Postmodern Condition Lyotard does attempt, in some limited fashion, to deal with issues of justice under conditions of paralogy. He condemns technocratic arrogance that would confine us to strictly controlled paradigms. In light of the heteromorphous nature of the paralogical situation, one must cope on a local level without recourse to metanarratives. Rather than a traditional metanarrative of justice, Lyotard offers the model of petit narratives, i.e., temporary contracts instead of permanent institutions in areas concerned with professional, emotional, sexual, cultural, familial, political, and international relations (see PMC 66). Justice involves playing games fairly; thus everyone should be granted access to the information they need: "give the public free access to the memory and data banks" (PMC 67). Consequently, Lyotard speaks against what he calls "terrorist" exclusion -- the elimination of or threat to eliminate "a player from the language game one shares with him" (PMC 63).

Lyotard proposes that in the local contexts of language games the players must reach a temporary consensus on the rules that define the game -- a consensus which would be eventually cancelled. But in any attempt to formulate local, temporary prescriptives, do we not need more than what Lyotard calls "meta-arguments" and temporary contracts? If we use computer technology to supply ourselves with information necessary for making "knowledgeable decisions" (PMC 67), would not those decisions, if they would be ethical, entail, as Lyotard himself recommends, more than information and the use of performativity criteria? If students in the postmodern university need to learn how to access information and computer data (see PMC 51), and if they need cleverness, speedy imagination, and the various skills and abilities to brainstorm and to work in teams (PMC 51-52), do they not need something more than these skills if they are to avoid measuring everything according to the performativity criterion? In effect, isn't there something missing in Lyotard's petit theory of justice?

At first glance it might seem that Lyotard supplies, in the interviews of Just Gaming, the answers to these questions by developing a theory of phronesis. Lyotard, while not looking for a meta-discourse on virtue, indicates, obliquely, that under the condition of paralogy, we require a "sensitivity to differences" and an "ability to tolerate the incommensurable" (PMC xxv). He explicitly turns to a theory of phronesis to develop his discussion of justice in the paralogical situation.2 The Aristotelian concept of phronesis serves his purpose because, contrary to a Platonic conception of justice, it does not depend on a theoretical metanarrative for its legitimacy. Phronesis is a purely prescriptive judging without appeal to theoretical criteria. "This is, after all, what Aristotle calls prudence. It consists in dispensing justice without models. It is not possible to produce a learned discourse upon what justice is" (JG 26; also see p. 43). Phronesis involves a dialectics which requires judging "case by case," "because each situation is singular" and there are no external criteria to guide judgment (JG 27). From case to case the mean is redefined in practical wisdom. "When one says: in every instance, choose the mean, it means, for Aristotle, that his mean cannot be determined in itself, that is, outside of the situation in which we find it" (JG 27). Thus, the prescriptions one gets through phronesis are what Lyotard might call "dangling prescriptives" (JG 59) -- i.e., they are not grounded on theoretical descriptions, but are developed "case by case."

In contrast to a clear Aristotelian distinction, however, Lyotard's concept of phronesis seems to be nothing more than cleverness. This is precisely the case to the extent that Lyotard would describe the prescriptives of phronesis as absolutely dangling, i.e., without any anchor. Lyotard goes so far in this direction as to deny an essential Aristotelian dimension of phronesis, namely, that phronesis depends upon education or on hexis, a habit of virtue, an ethos.3 Although both Lyotard and Aristotle would agree that there is no theory or method that explicitly defines phronesis, and that phronesis is not a trained ability, they would disagree about the proper educational background for phronesis. Phronesis, according to Aristotle, requires an education, a knowledge of particulars which comes from experience (NE 1141b17; 1142a14) -- and this includes understanding (which involves learning) and good sense. Phronesis also requires virtue; the person who wants phronesis must develop the right habits, an ethos developed over time. This formation of the right ethos in experience -- I call this the educational backdrop of phronesis -- is precisely what Lyotard denies. In effect, Lyotard would hold that the person with phronesis plays a good game -- a just game -- quickly without preparation, and only in the immediacy of the paralogical situation.4

Clearly, without the ethos, without a backdrop of educational experience, what Lyotard calls phronesis is nothing more than what Aristotle would call cleverness. He ignores Aristotle's clear contrast between these two concepts (see NE: 1144a25ff), as well as Aristotle's denial that "quickness of mind" is sufficient for the excellence in deliberation which is phronesis (1142b5-16). He equates phronesis with speedy imagination, "the capacity to actualize the relevant data for solving a problem 'here and now', and to organize that data into an efficient strategy" (PMC 51).

For Lyotard, phronesis is simply the ability to play the game with inventiveness, to play "master strokes" (JG 61). Justice is nothing other than this. "Justice consists in working at the limits of what the rules permit, in order to invent new moves, perhaps new rules, and therefore new games" (JG 100). In the end, inventiveness seems to be the sole measure of "goodness." For Lyotard's postmodern individual, action is not informed or educated by an understanding of goodness, or by a concern for seeking the good in any other sense than understanding and seeking after novel rules and games.

Caputo on Paralogy and Meta-phronesis

There is a clear contrast between Lyotard's conception of phronesis and the model of meta-phronesis proposed by Caputo. While Lyotard, in the first instance, seems to recognize a place for phronesis within postmodern hermeneutics, as we have seen he reduces phronesis to mere cleverness. In contrast, in Caputo's radical hermeneutics, we confront, first, an explicit critique of phronesis. Caputo seems to allow no room for phronesis within the paralogical situation. I want to show, however, that, appearances to the contrary, Caputo, much more than Lyotard, does recognize a place for genuine phronesis in the paralogical situation, and thereby supplies what Lyotard lacks; a more adequate conception of how we might cope with the paralogical condition of postmodernity.

In his book, Radical Hermeneutics, Caputo adopts a model that is akin to Lyotard's concept of paralogy. Both Caputo and Lyotard speak of a new conception of rationality, which Caputo calls "post-metaphysical" or "post-modern" rationality.5 Caputo, like Lyotard, adopts Kuhn's distinction between "normal" science and "revolutionary" science. They both want to emphasize the transgressive nature of revolutionary knowledge as it functions outside of the constraints of normalized paradigms. Paralogy is expressed here as a "conflict of paradigms" which cannot be resolved by a common metatheory (RH 217-218). Caputo reiterates Kuhn's point: there are no "meta-paradigmatic criteria in terms of which paradigms can be judged and directly compared with each other ..." (RH 218). In such instances of paralogy, reason "is forced to cut its own path, to break with established habits, to think in a radical, ground breaking way. I would say that at this point reason is put in play with the matter at hand .... At this point, where no established way has been charted, there is a certain free play of scientific rationality ... a creative transformation of old signs ..." (RH 219).

Again, like Lyotard, Caputo identifies and critiques a certain arrogance and terrorism in attempts to hold normal science together: "normal science flourishes because of its authoritarian practices .... Science flourishes because of the violence with which it enforces its paradigm." Caputo, like Lyotard, insists "that all such violence must be seen as violence, that every paradigm is a fiction, a contingency, a way of laying things out which cannot claim absolute status or immunity from reform" (RH 220).

If things are as we say, in flux, in undecidable drift and slippage, and if reason is to respond to things, to keep up a correspondence with them ... reason must play it loose, be capable of unexpected moves, of paradigm switches, of following up unorthodox suggestions. (RH 229)

In facing up to the paralogical situation, Caputo raises the same concern about justice that Lyotard does: the games must be played fairly, and no one should be excluded (RH 228). Likewise, Caputo rejects the performativity criterion (see RH 221), and strongly objects to the demands of technological efficiency, warning of "the danger of an expanding technology which [transforms] the world and man's relationship to the world, [transforms] our ideas of art and man and of the city, the polis and its institutions" (RH 232). There is a clear kinship between Lyotard's thinking on the performativity-based university and Caputo's sharp critique of precisely the same institution (cf. PMC 47-53; RH 228-235).

Notwithstanding such agreement, we find, between Caputo and Lyotard, a clear contrast in their search for a meta-virtue. Caputo outlines an "ethics of dissemination," i.e., a way to talk about how we can cope with the paralogical situation. Faced with paralogy, action "takes its point of departure not from fixed points of reference and steady principles (as in metaphysical ethics) and not from the nomos issued by a primordial ethos (as in eschatological ethics) but precisely from the dissemination of principles and primordial epochs. It is precisely from the breakdown of standpoints and resting points of all sorts that we begin to act" (RH 238). Taking one's point of departure from the flux of paralogy rather than from the fixed logos of metaphysics requires "humility" (RH 258), but also compassion (RH 259), and a strong sense of suspicion (RH 260). In the end, however, one needs some kind of meta-virtue that tells us when to be humble, or compassionate, when to practice a hermeneutics of suspicion and when to "trust in the dynamics of the agora" (RH 261). Caputo calls this meta-virtue "civility" or "meta-phronesis" (RH 262). Meta-phronesis is Caputo's answer to all of the questions we posed to Lyotard. Meta-phronesis is something more than what we find expressed in temporary contracts; it is a prerequisite for making our information-based decisions knowledgeable and good; it is more than academic cleverness and speedy imagination. Meta-phronesis is both "the skill to cope with competing paradigms," and "the virtue of knowing how to like and live with the dissemination of ethos"(RH 262).

Caputo wins this concept of meta-phronesis only in a critical battle with Gadamer's notion of phronesis -- in Caputo's view, a battle against an enduring metanarrative, a long metaphysical tradition extending back through medieval Christianity to the Greeks. Caputo develops his concept of meta-phronesis only in contrast to the failure of phronesis in the face of paralogy. But what, precisely, is the difference between postmodern meta-phronesis and traditional phronesis?

Caputo argues against the adequacy of Gadamer's notion of phronesis for the paralogical situation. He acknowledges the connection between Gadamer's notion of phronesis and Heidegger's notion of Verstehen (RH 109), and thus the role phronesis plays as a model for hermeneutical application (RH 110). He also recognizes the important role phronesis could play within the framework of the established paradigms of normal science. "The originality of Gadamer's conception lies in his insistence that both practical reason and scientific reason need to be understood in terms of phronesis .... But there are definite limitations to such a project from the point of view of a radical hermeneutic. The hermeneutic conception of phronesis presupposes an existing schema, a world already in place. It is the virtue of applying or appropriating a preexisting paradigm" (RH 210-211). Outside of established paradigms, however, in the paralogical situation, phronesis cannot operate. "Craft and skill and a sense of what the situation demands all founder at that point when the situation as a totality is not organized, when the horizons are skewed, when the whole trembles, when things are ruptured, decentered, disseminated ..." (RH 211). Here, Caputo goes on to claim, "there is no question of applying an agreed-upon standard, and Gadamer tends to ignore the subversion of hermeneutic phronesis by a diversity of power plays" (RH 261).


Phronesis in the Paralogical Situation

In my view, there are three difficulties with Caputo's rejection of phronesis as the meta-virtue required for coping with paralogical flux. They concern, first, his characterization of the paralogical situation; second, his account of phronesis; and third, his description of meta-phronesis.

First, both Lyotard and Caputo offer an overly radical description of the paralogical situation. Paralogy involves the shifting flux of paradigms, where recourse to external standards or established traditions is denied. Between paradigms, where there are no norms to supply answers, where there are no frameworks or fixed orders, but only ambiguity and undecidability, the postmodern, postmetaphysical person must respond without the support of phronesis. On this radical model of paralogy, incommensurability is made absolute.

But isn't this radical characterization of paralogy overplayed? It would seem that neither Lyotard nor Caputo want to admit that whatever situation we find ourselves in, no matter how paralogical, it is always a hermeneutical situation. Beyond the first weeks of life, we are never thrown into an absolutely unfamiliar situation. There is always some basis on which to interpret that which falls outside of any paradigm, simply because we are always situated, always located at some already (and to some degree) familiar locale. Our past, our traditions, our practical interests always condition our situation, so that whatever temporary contract or consensus we agree to, whatever new paradigm we invent, it will never be absolutely without precedent. Here it is clear that Lyotard's reduction of phronesis to cleverness is tied to his exaggerated conception of the paralogical situation. If Lyotard were to allow for the necessary backdrop of educational experience in his conception of phronesis (or, in other words, if he were to admit that prescriptions are never absolutely dangling), he would then be able to recognize that the paralogical situation is also a hermeneutical situation.6

On the other hand, Caputo, between the lines of his overly radical description of paralogy, obliquely acknowledges such hermeneutical constraints. "At this point, where no established way has been charted, there is a certain free play of scientific rationality, a free repetition of possibilities which the old system harbored, a creative transformation of old signs -- let us say, a releasing of the signs which make up the old system" (RH 219, emphasis added). This does not mean that we have recourse to external standards, pre-established rules, or eternally fixed frameworks as ready-made solutions. But it does mean that we do have access to experience and existing information, that our situation has already been structured by certain internal, but now unworkable standards, now preempted rules, and frameworks which have become unfixed. Our creation of new games is never ex nihilo, but always a creative transformation. Paralogy is a situation of uncertainty; but while we do not operate out of a position of certainty (in any Cartesian or Husserlian sense), still, we do not operate outside of a hermeneutical situation within which we find some degree of familiarity.

Second, phronesis, as either Aristotle or Gadamer explains it, is not the mechanical application of preestablished rules to "a world already in place."7 Phronesis is precisely the virtue that one can fall back on within a hermeneutical situation which is uncertain. Caputo mistakenly equates phronesis with a skill or craft which founders when the situation is paralogical (RH 211, 216). But phronesis is neither a method, nor a piece of techne. Caputo certainly knows this from his study of Heidegger and Gadamer.8 He cites Gadamer's description of phronesis as "a mode of knowledge [an eidos gnoseos] that could no longer be based in any way on a final objectifiability in the sense of science .... [A] knowledge within the concrete situation of experience."9 Such knowledge, which phronesis is, cannot be summarized by words like 'skill' or 'craft'. To reduce phronesis to a skill would be to equate it with cleverness. For Aristotle, as well as for Gadamer, phronesis, which is neither cleverness nor techne, is to be relied upon precisely on those occasions when no formula is available in advance, in those situations where we must act kata ton orthon logon (according to right reason).10

Although Caputo does recognize the educational backdrop of phronesis which "does not come in a flash but is slowly nurtured through years of training" (RH 220) it is this lack of speed that Caputo criticizes. In effect, he defines phronesis as conservative because of its slowness; it cannot keep pace with postmodern flux. But to speed phronesis up would be to reduce it to speedy cleverness. Aristotle notes that phronesis requires time, both to be developed and to be applied (Politics: 1269a15-18). If meta-phronesis needs to be fast phronesis (which Lyotard tries to get by jettisoning the educational backdrop), then Caputo would have to jettison the limitations of the hermeneutical situation. But if Caputo would jettison neither the educational backdrop (which, for meta-phronesis, could easily be cast in terms of his discussion of "the conversation of mankind" [see RH 263]) nor the hermeneutical situation, what, then, is the difference between phronesis and meta-phronesis?

Caputo first calls meta-phronesis "civility" (RH 262). This concept appears at the midpoint of his discussion of the "conversation of mankind" as a model for the ethics of dissemination. With respect to the fair play required to keep the conversation going, one requires not only a "trust in the dynamics of the agora" and a distrust of metanarratives, but also "a Socratic vigilance which insists on keeping the game fair" (RH 261-263). In this regard, however, Caputo dismisses the concept of phronesis and proposes his concept of civility, "which is a kind of meta-phronesis" (RH 262). The term 'civility' in this context seems to derive from Michael Oakeshott and Richard Rorty: "Hermeneutics views [the participants of the conversation] as united in what [Oakeshott] calls a Societas -- persons whose paths through life have fallen together, united by civility rather than by a common goal, much less by a common ground."11 In effect, on the model of the conversation of mankind, civility constitutes the virtue of a society that copes well with the paralogical condition. But is civility anything more than phronesis, or what Rorty terms "the practical wisdom necessary to participate in a conversation."12 The very term 'meta-phronesis' seems to indicate that Caputo has phronesis in mind, but a phronesis capable of operating in the paralogical situation. Is meta-phronesis anything more than phronesis?

Caputo himself refers us to his own earlier analysis in which he clearly argues for the place of phronesis in paralogical situations of uncertainty.13 There he states that phronesis is a "concrete moral insight," a "judgment as to what is to be done here and now" in the concrete and ambiguous moral situation.14 Indeed, rather than being determined by previously set rules, phronesis enables us to work out the rules for the situation. Caputo develops the concept of "applying cases to moral principles."15 But this concept of application, in which the universal gets determined by the particular, "case by case" as Lyotard says, is also to be found in Gadamer's description of phronesis. Gadamer writes, concerning the concept of application modeled on phronesis, that "it does not mean first understanding a given universal in itself and then afterward applying it to a concrete case." Rather one can only try to understand the universal on the basis of the particular hermeneutical situation that one is in.16 Is phronesis here anything less than meta-phronesis?

Caputo, in attempting to elevate phronesis to meta-phronesis, needs to move in a direction opposite to Lyotard who reduces phronesis to cleverness by denying the educational backdrop which phronesis requires and by speeding phronesis up. Where Lyotard, in reference to the paralogical situation, speaks of only a formal sense of justice, equal access to information, and cleverness, Caputo pushes further towards a meta-virtue that depends on more than access and cleverness. But this means that, in spite of his critique of phronesis, Caputo's conception of meta-phronesis can only be phronesis again.

Phronesis has a place within the radical hermeneutics required in the postmodern paralogical situation. Insofar as phronesis cannot be specified outside the situation, which always remains hermeneutical, and insofar as it is "impoverished in the abstract ... and acquires texture only in the application" (RH 216), and "takes on meaning only in the concrete situation" (RH 110), phronesis does not lend itself to metanarratives. It does not attempt to legitimize a particular discourse, position, or paradigm. Phronesis is not a super-discourse or a set of universal rules, but a sub-discourse that can only have a place within and between every language-game and conversation, and can only operate differently according to local situations. If phronesis differs from one context to another, it is not something that can be speeded up by neglecting its educational backdrop or by denying the limitations imposed by the hermeneutical situation. As such it cannot be reduced to information, rules of the game, learnable techniques, skill, proficiency, quickness of mind, or shrewd guessing. These latter concepts and practices could adequately serve a society organized by the performativity principle, but would not promote paralogy or move us towards justice.





1. Jean-Francois Lyotard. The Postmodern Condition: A Report on Knowledge, trans. Geoff Bennington and Brian Massumi, (Minneapolis: University of Minnesota Press, 1984), p. 60; hereafter cited as 'PMC'.

2. See Jean-Francois Lyotard and Jean Loup Thebaud, Just Gaming, trans. Wlad Godzich (Minneapolis: University of Minnesota Press, 1985), pp. 26ff, 95; hereafter cited as 'JG'.

3. In response to Jean-Loup Thebaud's suggestion in this regard, Lyotard responds: "I am not even sure that one can say that [i.e., that the person of phronesis 'has been educated, because there is a habit, because there is a pedagogy of the soul']" (JG 26).

4. In Aristotle's description of phronesis this immediacy is mediated in something like a circle: one is good if one does the good thing; but one does the good thing only if one is good. Lyotard denies the second part of this circle: "It is not because the judge is virtuous that he will judge justly, but if he does judge justly one will be able to say that he is virtuous" (JG 26).

5. John D. Caputo, Radical Hermeneutics: Repetition, Deconstruction, and the Hermeneutic Project (Bloomington: Indiana University Press, 1987), pp. 209-213; hereafter cited as 'RH'.

6. For a description of the hermeneutical situation, see Hans-Georg Gadamer, Truth and Method, 2nd rev. ed., revised translation by Joel Weinsheimer and Donald G. Marshall (New York: Crossroad Press, 1989), pp. 301-302. In Gadamer's terms, knowledge of a situation is always imperfect knowledge gained within the situation. The model for such knowledge is phronesis (see Ibid., pp. 313ff).

7. RH, 211; see Gadamer, Truth and Method, pp. 317-318, 324.

8. The connection between Aristotle's notion of phronesis and Heidegger's hermeneutics of circumspection (Umsicht) has recently been clarified by the publication of Heidegger's "Aristoteles-Einleitung" in Dilthey-Jahrbuch fur Philosophie un Geschichte der Geisteswissenschaften, (Gottingen: Vandenhoeck & Ruprecht, Vol 6 [1989], 228-274). For a good commentary, see Rudolf A. Makkreel, "The Genesis of Heidegger's Phenomenological Hermeneutics and the Rediscovered 'Aristotle Introduction of 1922'," Man and World 23 [1990], 305-320.

9. Gadamer, Philosophical Hermeneutics, trans. D. Linge (Berkeley: University of California Press, 1976), pp.201-202; cited in Caputo, RH, 109.

10. Aristotle, Nicomachean Ethics 1138b25. See Alasdair MacIntyre's comments on this text, in After Virtue (Notre Dame, IN: University of Notre Dame Press, 1984), pp. 152-154.

11. Richard Rorty, Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979), p. 318.

12. Ibid., p. 372. To the extent that civility or meta-phronesis is explicated within the framework of the conversation of mankind it may in fact be something less than phronesis. For a discussion of why the conversation of mankind is an inadequate model for postmodern hermeneutics see my Hermeneutics and Education (Albany: SUNY Press, 1992), pp. 307-311, 350-351.

13. RH, 305n.8; see, John Caputo, "Prudential Insight and Moral Reasoning," Proceedings of the American Catholic Philosophical Association 58 (1984), 50-55.

14. "Prudential Insight," p. 53.

15. On this point, Caputo cites Robert Henle, S.J., "Prudence and Insight in Moral and Legal Decisions," Proceedings of the American Catholic Philosophical Association 56 (1982), p. 29. This concept has recently been taken up by Charles Allen, who contends that phronesis "presupposes that particulars in their full particularity are capable of making sense in a way that universals cannot fully anticipate and that furthermore affects the way in which universals are to be actualized in that instance." See "The Primacy of Phronesis: A Proposal for Avoiding Frustrating Tendencies in Our Conceptions of Rationality," Journal of Religion 69 (1989), p. 366.

16. Gadamer, Truth and Method, p. 341; see p. 324. Caputo disagrees with none of this. He writes, in his 1984 essay: "we will, indeed we all too often, encounter cases which are not covered by a fixed norm. The practical intellect finds itself in a new situation, where it cannot fall back upon an explicit rule which covers the current case. This is not an irrational state of affairs in which we are forced to act capriciously. On the contrary, it is an example of our moral rationality at its best, in its finest hour. For in this case practical reason is thrown back on its own resources and must summon up all its powers of insight to make sense of the situation at hand. (In the same way, in a moment of scientific crisis, in a Kuhnian revolutionary crisis, the physical scientist requires the same phronesis and insight and is not acting irrationally at all.)" ("Prudential Insight," p. 54).