I

In recent decades there has been considerable interest in the questions of whether, and to what extent, we can accept the tradition-bound nature of rationality and the possibility of cultural variation in traditions, while still rejecting the relativistic thesis that no issue between competing traditions is rationally resolvable. Alasdair MacIntyre has argued that it is possible to defuse the threat of relativism even if we grant that reason is inescapably tradition-dependent. Central to his view is the theory of the rationality of traditions, which says that some traditions have a greater claim to our allegiance than others because one tradition can rationally defeat another through the work of inquiry.

However, this theory has been widely criticized.Footnote ¹ Broadly speaking, critics of MacIntyre maintain that his claims about the possibility of rational evaluation across traditions are irreconcilable with his conception of the tradition-dependent nature of rationality. Either his theory of the rationality of traditions has a tradition-independent basis, in which case his conception of rationality is false, or the theory is merely justified within and for the members of a particular tradition (namely, MacIntyre's) and it therefore fails to refute relativism. While these commentators disagree as to which side MacIntyre ultimately comes down on, they all agree on the intractability of the problem itself.

My aim in what follows is to show that this criticism is misguided. MacIntyre does not face the unpromising task of having to choose between self-contradiction, on the one hand, and self-trivialization, on the other, because the supposed tension in his work is merely apparent. Though Lott (Reference Lott2002) has also defended a claim along these lines, I will argue that Lott's response to the problem is exegetically and philosophically inadequate. Given the shortcomings of Lott's account, we have good reason to return to the familiar debate about MacIntyre's project. On the view I will defend, the rationality of traditions is not simply one more theoretical account of rationality. Rather, it is a theory that is implicit in and presupposed by traditions of inquiry generally.

I propose to divide the present essay into several parts. First, I spell out MacIntyre's theory of the rationality of traditions. Next, I describe Herdt's (Reference Herdt1998) assessment of the incoherence in his project. Then, in the third section, I turn aside to consider Lott's response to Herdt. Finally, in the remaining sections, I argue that MacIntyre avoids the apparent dilemma in his work because, in addition to the familiar dichotomy of tradition-dependence and tradition-independence, he proposes a third alternative, which I will follow Herdt in calling ‘tradition-transcendental’ (Reference Herdt1998, 526).

II

Following Mosteller (Reference Mosteller2006, 45), and using the term ‘tradition’ to refer to the collection of historically constituted and socially embodied norms that a given group uses to justify their beliefs, we can distinguish between three key features of MacIntyre's moral epistemology:

1. (i) Rationality is dependent on the resources of traditions;

2. (ii) there is no neutral, tradition-independent way in which to assess the epistemic status of a theory or belief; and

3. (iii) one tradition can rationally defeat another by argument.Footnote ²

Let's begin with (i). To say that rationality is dependent on traditions is to say that a theory or belief can be established as rationally justified or unjustified only according to the standards and principles of one tradition or another. As MacIntyre writes: ‘There is no standing ground, no place for enquiry, no way to engage in the practices of advancing, evaluating, accepting, and rejecting reasoned argument apart from that which is provided by some particular tradition’ (Reference MacIntyre1988, 350).

The implication of this aspect of MacIntyre's view is twofold. First, since every tradition has its own distinctive set of standards for distinguishing justified from unjustified beliefs, there are at least as many different accounts of rationality as there are traditions. MacIntyre calls our attention to this point in Whose Justice? Which Rationality? (WJWR) when he says: ‘since there are a diversity of traditions of enquiry, with histories, there are, so it will turn out, rationalities rather than rationality’ (Reference MacIntyre1988, 9).

Second – and this brings us to (ii) – the notion of a tradition-independent standpoint for inquiry is mistaken. Because traditions provide us with the only means we have for engaging in rational debate, ‘it is an illusion to suppose that there is some neutral standing ground, some locus for rationality as such, which can afford rational resources sufficient for enquiry independent of all traditions’ (Reference MacIntyre1988, 367). This means that given any two (or more) rival traditions, there is no appealing to an external viewpoint in order to adjudicate between them. For MacIntyre, the standing ground for inquiry is always within traditions and never outside of them.Footnote ³

In order to avoid the obviously relativistic implications of this view, MacIntyre also develops his theory of the rationality of traditions, or (iii). In the present essay, my focus will be on his discussion of this theory in Chapter 18 of WJWR, not only because this is perhaps his best and most well-known account of the rationality of traditions, but also because it has been at the center of the scholarly debate about his view in recent years.Footnote ⁴ In WJWR, MacIntyre claims that traditions progress through three stages. All traditions begin from historically contingent commitments. At this stage there is typically unquestioning deference to a set of authoritative beliefs. During the next stage, people begin to examine these beliefs and identify shortcomings. These shortcomings are typically remedied during the third stage. In some cases, however, a tradition is unable to re-establish coherence in its system of beliefs. When this occurs, MacIntyre says that the tradition enters into a period of ‘epistemological crisis’ (Reference MacIntyre1988, 361).

An epistemological crisis arises when a tradition is unsuccessful in its attempts to resolve its own internal problems and difficulties. Characteristically, such a tradition fails according to its own standards of rationality. Sometimes the resolution of such a crisis occurs only when a rival tradition comes along that is able to satisfy three key conditions. It must:

1. (1) Solve the previously intractable problems in the tradition in crisis;

2. (2) Provide a cogent and illuminating explanation as to why the tradition in crisis failed in its own attempts to solve those problems; and

3. (3) Meet both of these conditions in a way that preserves continuity between it and the tradition in the crisis (see MacIntyre Reference MacIntyre1988, 362).

These are MacIntyre's conditions for the rational defeat of a tradition. In his view, the rival tradition that fulfills them justifies its standards of rationality and claims to truth as superior to the standards and claims of the tradition in crisis.

It is worth noting that MacIntyre does not make the implausible claim that there will be actual acknowledgment of defeat simply because the rationality of traditions requires it (see MacIntyre Reference MacIntyre1988, 365). Nor does he assert that the members of a tradition in crisis must actually recognize the failure of their tradition in order for it to be defeated (see MacIntyre Reference MacIntyre and Deutsch1991, 117). He does seem to suggest, however, that the vindication of a tradition depends on the ability of its adherents to avoid epistemological crises in resolving their own problems, while at the same time providing successful explanations of the crises in other traditions. As he writes: ‘It is in respect of their adequacy or inadequacy in their responses to epistemological crises that traditions are vindicated or fail to be vindicated’ (Reference MacIntyre1988, 366; see also MacIntyre Reference MacIntyre and Deutsch1991, 117). Elsewhere MacIntyre grants, more cautiously, that we can assert the rational superiority of one tradition over another even if both are ‘more or less successful’ (Reference MacIntyre and Cunningham2009, 35).Footnote ⁵

III

Can MacIntyre's conditions for rational defeat be sustained? As I mentioned earlier, most commentators agree that even though MacIntyre wants to maintain both the tradition-bound nature of rationality and the possibility of rational assessment across traditions, he cannot have it both ways. Either there are no tradition-independent standards of rationality – in which case (i) and (ii) are true and (iii) is false – or there are such standards – in which case (i) and (ii) are false and (iii) is true. Whereas the first option implies that MacIntyre fails to defuse the threat of relativism, the second implies that his position is inconsistent.

Herdt (Reference Herdt1998) comes down on the side of the second option. According to Herdt, the difficulty for MacIntyre is that his theory of the rationality of traditions must be universally valid in order to refute relativism, yet universal validity disproves tradition-dependence. MacIntyre can only have the one at the cost of the other, and since he clearly takes his theory to show that the ineluctability of traditions does not entail relativism, his account is inconsistent. Especially telling in this regard, says Herdt, is that the rationality of traditions is strikingly similar to many of the Enlightenment theories that he criticizes. For just as Enlightenment thinkers ‘took refuge in universality and impersonality, ’ so MacIntyre also attempts to provide a ‘universally valid … solution’ that amounts to ‘a universal standard or principle of rationality’ (Herdt Reference MacIntyre1988, 526, 535). Thus Herdt asks: ‘Is MacIntyre's rationality of traditions perhaps just a new Enlightenment method?’ (535). In her view, the answer is yes. Because it purports to hold across all traditions and to govern the conflicts between them, his rationality of traditions amounts to precisely the kind of tradition-independent theory that (i) and (ii) reject.

Herdt notes that MacIntyre could perhaps try to defend himself against this charge by saying that his point is not to provide actual standards of rationality. Instead, his point is to spell out a procedure for assessing rival traditions. Because the standards to be used in applying this procedure will be the standards of particular traditions, he is not committed to any tradition-independent norms. However, Herdt rejects this line of reasoning on the grounds that MacIntyre's procedure can be used to adjudicate inter-traditional conflicts and so ‘does provide a standard not limited to a particular tradition, even if only at a general, procedural level’ (Reference Herdt1998, 535).

There are two important ways, according to Herdt, in which his procedure is universally valid and hence tradition-independent. First, it functions as a general criterion of rationality. MacIntyre claims that the members of a failing tradition must transfer their allegiance to the rival tradition that successfully resolves their epistemological crisis. Why suppose that the shift in allegiance is rational? To answer this question, says Herdt, we cannot simply appeal to the strength of the theories within the rival tradition, as the strength of these theories cannot explain why rationality requires the members of the failing tradition to accept the standards of their rival. The only way we can make sense of this requirement is on the basis of MacIntyre's tradition-independent theory. ‘Without this standard, which transcends the conflicting traditions, ’ Herdt thinks, ‘the shift of loyalty would be irrational’ (Reference Herdt1998, 536).

Second, the rationality of traditions possesses the same universal and tradition-independent characteristics as other procedural approaches. Herdt mentions three: (1) it is not limited in its range of application to a particular tradition; (2) it transcends specific difficulties; and (3) it can be spelled out without respect to any concrete disputes. As a general theory of rational debate between traditions, the theory possesses these characteristics because it purports to be valid in all cases. Indeed it is precisely because it possesses them that MacIntyre is able to reject relativism. Yet the notion that the theory offers ‘shared criteria for the superiority of a tradition’ about which ‘all are expected to agree’ is in tension with his rejection of the possibility of a standing ground outside of traditions (Herdt Reference Herdt1998, 537). Because the rationality of traditions supplies universally valid criteria of theory choice, it is not limited to a particular tradition.

IV

Lott (Reference Lott2002) has suggested one way in which we might respond to Herdt. According to Lott, we must be careful to distinguish between two different senses in which a claim can be ‘valid.’ In one sense, validity has to do with justification. To say that claims cannot be justified from a tradition-independent point of view is to say that they are never universally valid in this first sense. In another sense, however, validity can be understood in terms of scope. Saying that claims can be universally valid in this second sense is to assert that their truth is not limited to a particular tradition. So understood, a claim is universally valid insofar as it is about the way the world is for everyone.

Lott argues that, for MacIntyre, a claim can be universally valid in the second sense of validity even though it is not universally valid in the first sense. In his words: ‘a theory might be rationally justified only given certain assumptions and standards of reason particular to a certain tradition, yet that theory might be a theory about how things are for all people, regardless of their tradition’ (Reference Lott2002, 334).Footnote ⁶ This is because of the difference between truth and justification: while the latter depends on historically contingent conditions, the former does not. A theory that is rationally justified only for the members of certain traditions at particular times and in particular places may still be a theory about the way things are in all possible times and places.

Herdt's mistake, according to Lott, is to overlook this important distinction. Herdt criticizes MacIntyre for making universally valid claims that are impossible by his own lights. In fact, however, MacIntyre only says that his claims are true. He does not assert that they are justified from a tradition-independent point of view: ‘although MacIntyre's position does claim for itself universal validity in the second sense … MacIntyre does not claim that his position has universal validity in the first sense’ (Lott Reference Lott2002, 335). Thus, while MacIntyre may be wrong about the universal applicability of his theory, we cannot say that his position is inconsistent. Though his theory of the rational competition between traditions purports to be universal in scope, it has the same justificatory basis as all other theories.

It should be noted that this response to Herdt allows us to resolve both of her main objections to MacIntyre's project. First, it implies that there is no problem in having to appeal to the rationality of traditions in order to decide issues of contention between traditions. Such an appeal is perfectly consistent with MacIntyre's conception of the tradition-bound nature of rationality because the theory is situated within a particular tradition. Second, Lott's interpretation also implies that the theory does not presuppose tradition-independent standards despite possessing a number of universal characteristics. Though MacIntyre takes it to be universal in scope, the standards on which it rests are internal to his preferred tradition.

Still, Lott's reading is exegetically infelicitous. Consider After Virtue. In this text, MacIntyre argues that the Aristotelian tradition is able to satisfy precisely those conditions for the rational defeat of a tradition that he later identifies as such in WJWR. Specifically, he argues that Aristotelianism not only resolves the intractable problems of what he dubs the ‘Enlightenment project’ but also can explain why this project had to fail (Reference MacIntyre1984, 51). Surely, the similarity between this argument and his theory of the rationality of traditions is not coincidental. Yet if Lott is correct and the theory depends on MacIntyre's tradition, then the argument cannot be rationally efficacious for anyone who belongs to a different tradition. Hence the implication is that After Virtue is little more than a defense of Aristotelianism directed at other Aristotelians, which seems implausible. As R Scott Smith points out, ‘[MacIntyre] has written After Virtue? … in such a way as to argue for much more than just an exposition and defense of his particularistic viewpoint’ (Reference Smith2003, 103).Footnote ⁷

Moreover, Lott's tradition-dependent interpretation is philosophically unsupportable. In response to an imagined objection from Herdt, Lott argues that MacIntyre's rationality of traditions can be justified self-referentially – that is, ‘in the same way that the account describes.’ This means that the theory is supported according to the process of development specified in the theory itself. Lott puts this point as follows:

It is far from clear, however, that self-referential justification does the philosophical work that Lott needs it to do. The difficulty with the tradition-dependent interpretation of the rationality of traditions is that it seems unable to account for the ability of the members of rival traditions to engage in meaningful debate. Suppose that Thomism satisfies MacIntyre's conditions for rational defeat. In this kind of situation, MacIntyre claims that rationality compels the members of the failing tradition to shift their loyalty to his tradition. Yet it is hard to see why they should do so if his theory about the rational justification of traditions is itself rationally justified only for those who already accept the standards of the Thomistic tradition. Indeed, it is not clear that any serious philosophical encounter has taken place. Insofar as the theory depends on assumptions particular to the Thomistic tradition, it seems that Thomists have not actually provided their rivals with any reason to accept their standpoint. Does the notion of self-referential justification allow us to avoid this familiar problem? It seems not.

This is because it is one thing to point out that a tradition-dependent theory can be justified on its own terms, quite another to claim that the members of rival traditions have reason to accept the theory as justified. Establishing the first claim does not amount to proving the second. The members of a rival tradition can admit that a theory about the rational vindication of traditions is vindicated through the process specified in theory itself. That does not mean they have reason to accept the theory. All this shows is that it is possible to provide the theory and tradition with a circular justification. In other words, MacIntyre's rationality of traditions can be justified within his tradition and his tradition can be supported according to the criteria spelled out in the theory. But such a justification will have no force for anyone who refuses to step into the circle.

Moreover, if Lott is correct, MacIntyre's opponents could have their own theories of rational competition between traditions (cf. Mosteller Reference Mosteller2006, 69). Indeed, their theories could even be vindicated in precisely the way Lott describes. Given that rationality is tradition-bound, there can be no sound a priori argument to rule out this possibility. So we appear to move from a question about alternative and incompatible sets of standards to a question about alternative and incompatible theories regarding the justification of sets of standards. Yet the essential issue remains unchanged: we still need a method for deciding between rival conceptions of justification.

It might be objected that this argument overlooks the fact that self-referential justification is not necessarily self-supporting. According to Lott, since ‘a key aspect of the rational justification of MacIntyre's theory will be his ability to write a history that accounts for the problems and solutions of MacIntyre's own tradition as well as rival traditions, ’ it is always possible that his theory will ‘cease to be rationally justifiable’ (Reference Lott2002, 336–337). But to say that the theory may no longer be justified on its own terms is simply to say that the members of MacIntyre's tradition may find themselves without adequate resources for deciding between traditions. MacIntyre claims, after all, that the rationality of traditions is their best account of rational conflict resolution thus far. So, if the theory is no longer justified by their own lights, then, faced with the decision to give their allegiance to one particular tradition rather than another, they will have no choice but to make an arbitrary leap of faith. Yet it is precisely this kind of irrationalism that MacIntyre criticizes in the work of Kuhn and Kierkegaard (see [Reference MacIntyre1977] Reference MacIntyre2006, 16, Reference MacIntyre1984, 39). Thus, I take it that we have good reason to reconsider his theory altogether. It is to this task that I now turn.

V

Toward the end of her article, Herdt proposes an alternative way for MacIntyre to avoid the charge of self-contradiction. His best chance for restoring coherence to his project, she argues, is to defend (iii) as presupposed by all traditions. Such a theory of rationality would be what Herdt calls ‘tradition-transcendental’ because it would neither be tied to a specific tradition nor justified in a tradition-independent sense. Instead, it would be embedded in all traditions. Herdt puts the point this way: ‘While it is true that the theory would have no independent existence in some never-never land, it would be implicit in practices of enquiry generally, not dependent on a particular tradition’ (Reference Herdt1998, 538). Because the theory would be embodied in traditions, it would not contradict MacIntyre's claim that rationality is inescapably tradition-bound. In addition, it would avoid the justificatory problems that plague Lott's account, since every tradition would be committed to admitting defeat when the conditions for rational defeat were met.

What is particularly striking about Herdt's proposed solution to the problem is that even though she ‘does not think that MacIntyre regards his own position as offering such a third alternative, ’ she recognizes that he articulates the alternative at one point in Chapter 18 of WJWR (Reference Herdt1998, 524). She quotes the following passage in support of this claim:

Herdt is surely right to suggest that the passage gives expression to – in her words, ‘hints at’ – a view very much like the one she describes as tradition-transcendental (Reference Herdt1998, 538). Like Herdt, MacIntyre refers to a theory of rationality that is inherent in practices of inquiry.

But Herdt is wrong to think that MacIntyre does not actually mean to offer the type of alternative that he describes in this passage. Consider what he says earlier in the same paragraph:

This passage suggests that MacIntyre is not simply hinting at an alternative in the passage Herdt cites. Instead, he is stating the program that he intends to follow in the remainder of the chapter.

Of course, MacIntyre only refers in these quotations to the particular traditions that he considers in WJWR – Aristotelianism, Thomism, the early modern Scottish tradition, and liberalism. Nonetheless, it is implausible to suppose he means to develop a theory of rationality that is limited to these traditions. First, throughout Chapter 18, MacIntyre refers to the features of all traditions, such as when he says: ‘Every tradition, whether it recognizes the fact or not, confronts the possibility that at some future time it will fall into a state of epistemological crisis’ (Reference MacIntyre1988, 364). Second, were his theory limited to the traditions that he considers in WJWR, it would hardly supply an adequate response to the relativist challenge. Though it would imply that rational competition is possible between some few traditions, relativists could still maintain that all other traditions have an equal claim to our allegiance. We have good reason, therefore, to think that MacIntyre sees his theory of rationality as embodied in traditions generally.Footnote ⁸

VI

Central to the rationality of traditions is the idea that the concept of truth is implicit in tradition-constituted inquiry. Consider the following passage:

Here MacIntyre makes two related claims. First, he asserts that rival traditions share the same concern with truth. Second, he says the reason that we can resolve disagreements between rival traditions is that traditions are oriented toward the truth. Why suppose that traditions share this concern? And in what sense does it allow for rational inquiry between traditions? These questions will be the focus of the next two sections.

Let's begin with the first question. In some places, MacIntyre suggests that traditions make claims to truth as a matter of empirical fact, as in the following passage: ‘within every major cultural and social tradition we find some distinctive view of human nature and some distinctive conception of the human good presented as – true’ (Reference MacIntyre, Horton and Mendus1994, 295; cf. MacIntyre [Reference MacIntyre1994] Reference MacIntyre2006, 54–55). In other places, he defends the stronger claim that truth is a precondition of inquiry. Clearly this claim is contentious. At a minimum, it is in conflict with the view of Nietzscheans who deny that the goal of inquiry is final truth (see MacIntyre Reference MacIntyre1990, 66). So MacIntyre needs to defend his claim against the standpoint of what he calls ‘the Nietzschean tradition’ (Reference MacIntyre, Horton and Mendus1994, 298). It is clear that he cannot do so, however, by appeal to his rationality of traditions. Because this theory requires a conception of truth that Nietzsche and his followers reject, MacIntyre can only rely on it at the cost of begging the question against them.

It may therefore seem that the disagreement between MacIntyre and Nietzsche is rationally irresolvable. That is not the case. MacIntyre can, and in fact does, attempt to defend his claim about the need for a conception of truth in inquiry by employing the same argumentative strategy that Aristotle and Aquinas use in defending the most fundamental standards of rationality.Footnote ⁹ For Aristotle and Aquinas, anyone who rejects the principle of non-contradiction is unable to engage in meaningful discussion. Skeptics do not come to a rationally motivated conclusion so much as they simply place themselves outside of the realm of rational debate. In MacIntyre's view, Nietzschean perspectivists do something similar.

Perspectivists deny that claims are true or false in a traditional sense. They hold that the theses of one tradition can be true without implying that the theses of its rivals are false (see for example MacIntyre Reference MacIntyre1988, 352). This is problematic, according to MacIntyre, because rational inquiry requires that rival standpoints can be understood as advancing logically incompatible claims: otherwise it makes no sense to ask which claims are true, let alone to try to find out which standpoint is superior. In this kind of situation, the parties involved are simply talking about different things (see MacIntyre [Reference MacIntyre1994] Reference MacIntyre2006, 60–61). Thus, since perspectivism denies that the truth-claims of rival moral traditions should be understood as logically incompatible, it follows that perspectivists are unable to participate in meaningful philosophical debate. As MacIntyre writes, they are ‘by their stance excluded from the possession of any conception of truth adequate for systematic rational enquiry’ (Reference MacIntyre1988, 368). Since perspectivists cannot engage in such inquiry, they cannot be said to realize anything by way of advancing reasons. Instead, they simply pass beyond the limits of intelligible discourse.Footnote ¹⁰

In this way, MacIntyre, like Aristotle and Aquinas, attempts to defend his view by spelling out the limits of rational debate. Just as the skeptic overlooks the importance of the principle of non-contradiction for intelligible discussion, so the Nietzschean perspectivist, in his view, ‘fails to recognize how integral the conception of truth is to tradition-constituted forms of enquiry’ (MacIntyre Reference MacIntyre1988, 367). There is no reason to suppose, then, that MacIntyre is unable to defend the presuppositions of his theory of rationality by argument. By appealing to the limits of rational debate, he can and does attempt to justify his claim that we cannot dispense with the notion of final truth.

VII

At this point we can turn to our second question, which concerns the relationship between the notion of truth and the rationality of traditions. Recall that earlier I suggested this theory is one that every tradition has regardless of its particular theses or standards. It should now be noted that the theory is not simply presupposed by traditions as MacIntyre happens to understand them. Sure enough, MacIntyre concedes that his ‘concept of a tradition, together with the criteria for its use and application, is itself one developed from within one particular tradition-based standpoint’ (Reference MacIntyre, Horton and Mendus1994, 295). But if his point were merely about traditions of the kind he prefers, then it would at best only be justified for those who share his view of traditions. His opponents would have no reason to accept it. Worse, since they could conceivably have their own rationally justified conception of traditions, MacIntyre would need to show that we can adjudicate between alternative conceptions.

That he does not supply any argument to this effect suggests that (iii) is not merely, by definition, inherent in traditions as he characterizes them. Instead, as I hope will emerge in what follows, MacIntyre's point is that (iii) is always and already embedded in truth-seeking practices of inquiry. Central to his account is the idea that there is a connection between claims to truth and justification. Claiming that our current beliefs are true commits us to asserting that the evidence does not and will not support a discrepancy between these beliefs and the world. As MacIntyre writes: ‘To claim truth for one's present mindset and the judgments which are its expression is to claim that this kind of inadequacy, this kind of discrepancy, will never appear in any possible future situation’ (Reference MacIntyre1988, 358; see also MacIntyre [Reference MacIntyre1994] Reference MacIntyre2006, 66). The reason for this is straightforward: were a discrepancy between our beliefs and the world to arise, we could no longer justifiably sustain the claim to truth for our current beliefs. Since one characteristic indication of such discrepancy is incoherence, the ability to avoid incoherence in our current beliefs is one measure of what we are justified in claiming as true.

The point is worth spelling out. To make a claim to truth is to assert that the mind has achieved a relationship of adequacy to the world that rules out the possibility of incoherence. For the mind is wholly adequate in this way, says MacIntyre, only insofar as ‘the expectations which it frames … are not liable to disappointment’ (Reference MacIntyre1988, 356). And we avoid disappointment in our tradition-induced expectations only if there never comes a time at which we are required by our own standards to admit a lack of correspondence between our beliefs about the world and the world as we later come to understand it. This is why MacIntyre says: ‘To claim that some thesis is true is … to claim for all possible times and places that it cannot be shown to fail to correspond to reality’ (Reference MacIntyre1988, 363). He is unequivocal about the significance of this point for his rationality of traditions: ‘The implications of this claim made in this way from within a tradition are precisely what enable us to show how the relativist challenge is misconceived’ (363–364).

The most important implication is that when a tradition lapses into epistemological crisis, its claims to truth and standards of rationality are called into doubt. In this kind of situation, MacIntyre writes,

Sometimes the members of the tradition will find that they can resolve their problems through the development of new theories or concepts. In other cases, however, they may persistently fail to remedy the incoherences in their current beliefs despite their best philosophical efforts. Eventually, lack of resolution may force them to admit that the problems are insoluble within the limits of their own tradition. Thus, since claims to truth cannot justifiably be sustained in the face of incoherence, the implication is that they may, sooner or later, have to concede the falsity of the central theses of their tradition (see MacIntyre Reference MacIntyre1988, 364).

In some of these cases, reason may also force the members of the failing tradition to admit defeat with respect to truth. At what point will that occur? It will occur when a rival tradition is able to establish itself as more adequate to the world on their very own terms. For when this happens, their own standards will require them to concede that the rival tradition is closer to the truth than their own tradition. Indeed it is precisely this kind of situation that MacIntyre's conditions for rational defeat are designed to capture.Footnote ¹¹ After all, what the members of the rival tradition that satisfies these conditions will have established is twofold. First, they will have shown that their tradition succeeds in avoiding incoherence at exactly those points that prevent the failing tradition from sustaining its claims to truth. Second, they will have proven that they can explain why the failing tradition is bound to remain inadequate. Thus, given that the members of the failing tradition are committed to achieving the truth, they will find themselves compelled to transfer their allegiance to the rival tradition.

The reason for this is not that they recognize some tradition-independent criterion over and above their claims to truth. Nor is it that a shift in loyalty is called for by the standards of MacIntyre's preferred tradition. Rather, it is because truth implies adequacy to the world, and when the conditions for rational defeat are met, the standards of the failing tradition will require them to admit both that they cannot realize the truth within their tradition and that the rival tradition is better equipped to realize the truth on their own terms.

Consider: when a rival tradition fulfills MacIntyre's conditions for rational defeat, the members of the failing tradition will find themselves in the following situation. On the one hand, since the rival tradition will have resolved their previously intractable problems, they will be forced to concede that it ‘does not itself suffer from the same defects of incoherence or resourcelessness’ (MacIntyre Reference MacIntyre1988, 365). On the other hand, they will also have to acknowledge that since it can provide them with a compelling explanation of their failure, the rival tradition can account for ‘why, given the structures of enquiry within that tradition, the crisis had to happen as it did’ (364–365; cf. MacIntyre Reference MacIntyre and Cunningham2009, 35). But that is just to say that under these circumstances rationality will require the members of the failing tradition to acknowledge the superiority of the rival tradition. For, as we have seen, anyone who makes a claim to the truth of their present beliefs is committed to the view that there will never be a discrepancy between their present beliefs and reality. And precisely what the rival tradition will have established is that there is a discrepancy of this kind.Footnote ¹²

As a result, the members of the failing tradition will have to concede the inferiority of their tradition in respect of truth. In other words, they will be forced to acknowledge ‘the claim to truth for what have hitherto been their own beliefs has been defeated’ (MacIntyre Reference MacIntyre1988, 365). For they are entitled to claim that the beliefs of their tradition are true only if these beliefs do not suffer from inadequacy and the logically incompatible beliefs of other traditions do. Yet what the adherents of the rival tradition will have established is that precisely the opposite is the case. In particular, they will have shown that their tradition is adequate in precisely those ways that the failing tradition is inadequate and also that they can explain how and why the failing tradition is inescapably deficient on its own terms. Thus, reason will require the members of the failing tradition to admit that the rival tradition is nearer to the truth than their tradition.

Read in this light, MacIntyre is not advancing tradition-dependent or tradition-independent standards but rather is articulating and shedding light on standards that are already presupposed by our practices of inquiry. His rationality of traditions is based on standards or conditions for rational defeat that everyone has by virtue of engaging in a truth-seeking form of inquiry. This does not mean, of course, that every tradition must explicitly endorse the conditions. But anyone who belongs to a tradition is at least implicitly committed, whether they recognize it or not, to acknowledging the superiority of a rival tradition and to shifting their allegiance to its standards and claims to truth when the conditions are met.

It may be helpful to restate the thesis I am defending in a slightly different way. Habermas (Reference Habermas1993, 100–101) argues that MacIntyre's theory of rationality runs aground on the dilemma we noticed earlier between self-contradiction and relativism. I am instead suggesting that MacIntyre avoids the dilemma altogether because his view is more Habermasian than even Habermas himself realizes (cf. Herdt Reference Herdt1998, 538–539). Just as Habermas takes his communicative rationality to be embedded in practices of communication aimed at understanding, so MacIntyre takes his rationality of traditions to be inherent in inquiries oriented toward the truth.Footnote ¹³ When the conditions for rational defeat are met, no one who aims at truth can deny the rational inferiority of their tradition without lapsing into incoherence.

VIII

Let's consider several objections to my proposal. First, MacIntyre's account of rationality implies that whether it is irrational to espouse inconsistent beliefs is susceptible to variation from one tradition to another. It is not hard to come up with an example of such variation. One commentator argues, for instance, that it is ‘rather doubtful whether Nietzsche himself would have insisted that standards must be rational or that coherence be included as a criterion’ (Williams Reference Williams1993, 128). Yet the rationality of traditions appears to rely on the principle of non-contradiction. Thus, if this principle is only present in some traditions and not others, then it seems that MacIntyre fails to refute relativism.

Fortunately, MacIntyre can avoid this problem. The reason, as we have seen, is that his account of rationality is compatible with the view that certain standards are present in all traditions. Like the conditions for the rational defeat of a tradition, MacIntyre claims that the laws of logic are shared by all traditions: ‘traditions … agree in according a certain authority to logic both in their theory and in their practice. Were it not so, their adherents would be unable to disagree in the way in which they do’ (Reference MacIntyre1988, 351). Granted, it is one thing to make this claim and something else to support it. But MacIntyre can defend the claim in the same way that he defends his understanding of the role of truth in inquiry: namely, by arguing that consistency is a precondition of intelligibility.

A second worry is that the tradition-transcendental view fails to rescue MacIntyre from self-contradiction. Recall that MacIntyre appears to face a dilemma between incoherence and relativism because he denies that there is a neutral way to assess the epistemic status of a theory or belief. The tradition-transcendental view initially seems to offer a way between the horns of this dilemma. However, conditions presupposed by all practices of inquiry are neutral between traditions. Hence a critic might argue that MacIntyre contradicts himself after all.

This objection looks tempting at first glance, but it is mistaken. The problem is that the critic conflates two distinct senses of the term ‘neutral.’ In one sense, saying that there are no neutral standards of rationality means that there is no view from nowhere, no standpoint outside all traditions. When MacIntyre says that reason is not neutral in this sense, he means that we always rely on the rational resources of some tradition or another. We can also understand neutrality in terms of the common ground between traditions. In this sense, to say that certain standards are neutral is to say that they are shared across traditions. Such standards are not independent of traditions altogether, but they are neutral between traditions in the sense that they are accepted by and present in every tradition. The critic is appealing to this second way of understanding neutrality when she claims that MacIntyre's conditions for rational defeat are neutral among traditions.

However, though MacIntyre denies that we can retreat to a neutral standpoint outside of traditions, he does not deny that we can withdraw to a neutral standpoint within them. On a tradition-transcendental reading, MacIntyre does not contradict himself because he holds that the conditions for rational defeat are always and already implicit in traditions, without claiming that they are justified from a supra-traditional standpoint. As Herdt remarks, ‘the standing ground is indeed within traditions, not outside of them, but within them necessarily’ (Reference Herdt1998, 538). The idea that the rationality of traditions is inherent in traditions of necessity may strike some readers as surprising, but there is no conflict here for MacIntyre. All that the idea amounts to is the claim that we can defend the theory dialectically by making skeptics aware of the way they already presuppose it. There is no suggestion that the theory is based on an appeal to rationality-as-such.

Here is another way to put the point: what MacIntyre objects to in the Enlightenment is not the notion of tradition-transcendence but rather the notion of tradition-independence. It is this latter notion that he has in mind when he criticizes the Enlightenment view that ‘rational method appeals to principles undeniable by any fully reflective rational person’ (Reference MacIntyre1988, 353). At the same time, there is a sense in which his method also appeals to principles all rational persons are compelled to accept. His rationality of traditions presupposes the laws of logic, after all, and according to MacIntyre, ‘no one who understands the laws of logic can remain rational while rejecting them’ (Reference MacIntyre1988, 4). What is the difference? For MacIntyre, it is that, whereas many Enlightenment thinkers attempted to take leave of traditions altogether, he merely appeals to a viewpoint that is present within all traditions. Though his theory is tradition-transcendent, it is not an Enlightenment method in the sense that he finds problematic.

A third objection comes from a passage toward the beginning of Chapter 18 of WJWR. In this passage, MacIntyre says that shared standards are unable to resolve inter-traditional conflicts.Footnote ¹⁴ His rationality of traditions is able to resolve such conflicts. Hence it looks as though there is a problem here for my view. Since the tradition-transcendental interpretation implies that the conditions for rational defeat are shared across traditions, it appears to be incompatible with the claim that shared standards are too weak to decide issues of contention between traditions.

The problem with this objection is the assumption that MacIntyre means to deny the effectiveness of all shared standards. Were this assumption correct, it seems that the best he could hope for would be to defend his theory as tradition-dependent, since he rejects the notion of tradition-independent standards. Yet we have seen that the tradition-dependent interpretation faces serious philosophical problems and is exegetically infelicitous. Only a few pages after denying the effectiveness of shared standards, moreover, MacIntyre appears to claim that his theory of rationality is shared across all traditions. It is hard to see that this claim is compatible with the tradition-dependent interpretation of his view. Though it is possible that MacIntyre simply did not recognize these issues, a more charitable reconstruction of the passage under review is that he is only talking about shared standards of a particular kind.

What kind is that? On my reading, he is referring in this passage to first-order standards – norms or epistemic principles that we use to justify our beliefs. Tellingly, immediately after claiming that shared standards are unable to resolve inter-traditional conflicts, he writes: ‘It may therefore seem to be the case that we are confronted with … a number of traditions … among which we can have no good reason to decide in favor of any one rather than of the others’ (Reference MacIntyre1988, 351). So his point in denying the effectiveness of shared standards is clearly to motivate the relativist challenge, the view that all sets of standards are equally defensible or indefensible (see MacIntyre Reference MacIntyre1988, 352). His strategy for responding to this challenge is to develop second-order standards of reasoning – standards that we use to justify other standards. I am referring, of course, to his conditions for rational defeat. These conditions are second-order standards because they are designed to help us decide between alternative sets of standards. Yet, since MacIntyre is explaining the motivation behind the relativist challenge in the passage under review, it is natural to read his claim about the ineffectiveness of shared standards as a claim about first-order standards. Hence there is no inconsistency with my view. Though MacIntyre thinks that traditions lack adequate overlap in some standards, he does not think that they lack adequate overlap in all standards.

A final objection concerns the role of the concept of truth in my interpretation. The worry is that tradition-transcendental standards of reasoning amount to little more than rules or principles of truth-talk. In our language, justification is relative but truth is absolute. Though we may be justified in endorsing false beliefs in some contexts, we hold that truth is timeless and does not vary from one context to another. But that is at most a fact about our language-game of truth. So isn’t the rationality of traditions ultimately based on pragmatic rather than transcendental considerations? Even granting that truth-talk commits us to the theory, shouldn’t we say that the conditions for rational defeat are, in a sense, contingent rather than necessary?Footnote ¹⁵

That depends on whether there are intelligible alternatives to the language-game of truth. If not, then there is no meaningful sense in which our commitment to the conditions for rational defeat could have been otherwise. As I read MacIntyre, truth-talk is indispensable – we have no choice but to use the concept of truth as we ordinarily do. Part of the reason for this is that, as we considered earlier, perspectivists cannot reject the notion of final truth without simply passing beyond the limits of rational inquiry.

But the deeper reason is that there is no way to make sense of rational inquiry without the concept of truth. To attempt to remove this concept from inquiry is, in MacIntyre's view, to risk becoming unintelligible to oneself as well as others. According to MacIntyre, philosophers who deny that truth is the aim of inquiry must either narrate their achievement in inquiry or else refrain from doing so. However, if they supply a narrative of progress, then they effectively presuppose a teleological account of inquiry as oriented toward truth. And if they do not supply such a narrative, then we have no intelligible account of what they have accomplished through their work. So, either way, their position seems unsupportable (see MacIntyre [Reference MacIntyre1990] Reference MacIntyre2006, 177). This argument strongly suggests that MacIntyre thinks there is no removing truth from inquiry.Footnote ¹⁶

The argument also makes clear the problem with the objection. The mistake that the critic makes is to assume that we commit ourselves to tradition-transcendental standards only as a consequence of the particular function that truth has in our language. These standards arise, so the thought goes, from a decision to use the term ‘truth’ in a way that is distinct from the term ‘justification.’ For MacIntyre, however, our language-game of truth is not merely one language-game among others. It is the only possible language-game. Without the concept of truth, we cannot make anything of inquiry. As MacIntyre says: ‘Practices of rational justification are … only fully intelligible as parts of all those human activities which aim at truth’ ([Reference MacIntyre1994] Reference MacIntyre2006, 58). Thus, like the conditions for rational defeat, MacIntyre holds that the concept of truth has a tradition-transcendental basis.

IX

We are now in a position to return to Herdt and Lott. If my arguments have been on target, then Herdt is wrong to suggest that the issues between competing traditions must be decided by direct appeal to MacIntyre's conditions for rational defeat. Instead, MacIntyre thinks the members of a failing tradition are compelled by their own standards to recognize the superiority of the tradition that satisfies these conditions. In any such case, the members of the rival tradition will have succeeded in showing that their tradition is more capable of realizing the truth that the members of both traditions seek. And because they will have done so according to the failing tradition's own understanding of what prevents it from maintaining its claims to truth, it will be irrational for its members to remain members of the failing tradition. The concept that bridges the gap between these two traditions is the concept of final truth that is implicit in both of them, not (iii).

Herdt is also wrong to think that the universal characteristics of MacIntyre's rationality of traditions are incompatible with his account of rationality. What we have seen here suggests that MacIntyre can consistently advance a theory of rationality that is universal in its range of application even though he rejects the Enlightenment view of reason as independent of traditions. His claim in (iii) does not presuppose the neutral standing ground outside of traditions that (i) and (ii) deny because the standing ground is embodied in traditions generally. Nor should this come as a surprise. Herdt also recognizes the ability of the tradition-transcendental reading to rescue MacIntyre from her criticism. At one point she even refers to it as his ‘best hope’ for maintaining both his tradition-bound conception of rationality and his denial of relativism (Reference Herdt1998, 524).

The tradition-transcendental view also allows us to avoid the difficulties associated with Lott's tradition-dependent interpretation. Since it appeals to a conception of rationality that is presupposed by all traditions, there is no need for self-referential justification. Instead, we save MacIntyre's account as a refutation of relativism by ruling out the possibility of meta-level conflict with respect to his theory. On this reading, the denial of (iii) is not even a legitimate possibility because anyone who makes claims to truth implicitly affirms (iii), and anyone who rejects the conception of theoretical inquiry as teleologically ordered toward the truth risks becoming unintelligible. Thus, despite the controversial nature of MacIntyre's project, it seems to me that the rationality of traditions is neither inconsistent nor trivial.

But then why have most commentators thought differently? I suspect the reason is that it seems odd to say the rationality of traditions is tradition-transcendent when MacIntyre is such an outspoken critic of the Enlightenment. It is bizarre because such a view seems to imply that MacIntyre is something of a neo-Kantian, at least with respect to this theory. At most, though, I think MacIntyre can be accused of supplying an idiosyncratic critique of the Enlightenment, one that allows for at least some transcendental approaches. He cannot be accused of developing a theory that is self-undermining.