Particularism, Pattern and Rule-following

全文

(1)《國立政治大學哲學學報》第二十三期（2010 年 1 月）頁 79-116 ©國立政治大學哲學系. Particularism, Pattern and Rule-following Peter Shiu-Hwa Tsu The Australian National University, Ph.D. Candidate in Philosophy. Abstract Frank Jackson, Michael Smith and Philip Pettit, in their co-authored paper “Ethical Particularism and Pattern” argue on behalf of the principlists that there must be a unifying meaning pattern governing the term “rightness”. And if there is a unifying meaning pattern, let’s say X, that governs the use of the term “rightness”, then we can get a true moral principle of the following form: X is right. Particularism, a doctrine which denies the existence of any true moral principles, would thus be falsified. In defense of particularism, I will critique four arguments invoked in support of the claim that there is such a pattern: the conceptual competence argument, the normativity of meaning argument, the consistency argument, and the universalizability argument. I contend that none of these arguments work. In the end of my paper, I argue that even if there is a pattern of the term “rightness”, it will not help the principlists to establish their claim that there are true moral principles, for a meaning pattern of rightness has to be distinguished from a criterion of rightness.. Keywords: particularism, principlism, rule-following, the normativity of meaning, universalizability, Frank Jackson, Michael Smith, Philip Pettit.

(2) 80 NCCU Philosophical Journal Vol. 23. Particularism, Pattern and Rule-following I. Introduction A common metaphysical picture of morality has it that morality is made up of a true and coherent set of moral principles. It follows from this picture that if one negates the existence of moral principles, one negates morality altogether. For without moral principles, it seems that there would be no standards against which the moral status of actions can be determined. In pursuance of this common metaphysical picture of morality, one chief concern of normative ethics has been to formulate basic moral principles that govern the moral terrain. It is generally believed that in basic moral principles lies the ultimate source of moral truths. The principlists, though arguing among themselves over what the correct basic moral principles are and over the number of them, all tacitly agree that a major part of normative ethics is built upon the articulation of the basic moral principles and their application to practical moral issues. While the heated debate is continuing about the correct formulation and application of the basic moral principle(s), the common metaphysical picture underlying it has not received proper attention---not until the appearance of the contemporary particularists. Contrary to the principlists, the particularists argue that morality does not depend upon codification into a true and coherent set of moral principles. In this view, general principles fail to capture the complex-.

(3) Particularism, Pattern and Rule-following 81. ity and uniqueness of particular circumstances. 1 Exceptions to principles are common and exceptions to exceptions are not unusual. 2 In other words, there are no exceptionless principles of the sort which the principlists have in mind. The particularists believe that the moral status of an action is not determined by moral principles; instead it always relies on the particular configuration of its contextual features. In this article, I will examine the debate between the principlists and the particularists with special focus on the question of whether there is any true moral principle. In particular, I will examine the argument from pattern the principlists often appeal to to establish their claim that there are true moral principles. I will argue that there is no strong reason for us to believe that it is sound. Although this would not refute principlism completely, as it may appeal to some other independent arguments for support, nevertheless, if I am right about this, it would be indicative of a reason for us to be less confident about the truth of principlism and more confident about the truth of particularism.. II. The Principlists’ Argument The principlists’ argument from pattern, as Jackson, Smith, and Pettit (henceforth the Canberrans) 3 understand it, is roughly as follows: P1: If the meaning of the term “rightness” is governed by a pattern / rule (or if there is a unifying feature of the term “rightness”), then there are true moral principles P2: The meaning of the term “rightness” is governed by a pattern / rule (or there is a unifying feature of the term “rightness”) C: There are true moral principles.. 1. Martha Nussbaum, Love’s Knowledge, (N.Y.: Oxford University Press, 1990), p. 69. Darin Davis, Rules and Vision: Particularism in Contemporary Ethics, (St. Louis University Philosophy Ph.D. Thesis, 2004), p. 1. 3 I follow Jonathan Dancy in calling Frank Jackson, Philip Pettit, and Michael Smith the Canberrans. 2.

(4) 82 NCCU Philosophical Journal Vol. 23. It is not hard to understand why the principlists believe in P1. It seems quite straightforward that if the meaning of the term “rightness” is governed by a meaning rule, whatever it is, we can get a true moral principle of the following schemata: “X is right”, in which X is regarded as the pattern for rightness or its unifying feature. For instance, the Canberrans point out that for utilitarians, “if the ethical conclusion is that X is right, the pattern for rightness is given by the rubric: X is right if and only if X maximizes expected happiness.” And similarly, non-utilitarians might hold other patterns such as given by: “X is right if and only if X is what an agent who exemplified all the virtues would do; or by: X is right if and only if X satisfies a certain weighted sum of prima facie duties better than any alternative to X.” 4 With regard to P2, things are a bit more complicated. There are four major reasons why the principlists hold that P2 is true. They are the thesis of conceptual competence, the thesis of the normativity of meaning, the thesis of consistency and the thesis of universalizability. In what follows I will argue that none of these provide conclusive reasons for thinking P2 is true. After I argue that these theses fail to support P2, I will argue that even if P2 is true, it does not follow that there are true moral principles as there is a distinction between the meaning of rightness and the criterion of rightness.. III. Conceptual Competence First, on behalf of the principlists, the Canberrans argue that P2 must be true, for if the meaning of rightness were not rule-governed, then we could never be competent with this concept. Let’s call this thesis the thesis of conceptual competence. Namely, the thesis of conceptual competence says that for anyone to be competent with the concept of rightness at all, there must be a pattern that governs its meaning. For without a pattern, the only way. 4. Frank Jackson, Philip Pettit, and Michael Smith. “Ethical Particularism and Pafferns”. In Brad Hooker and Margaret Little (Eds.), Moral Particularism, p.80..

(5) Particularism, Pattern and Rule-following 83. we could learn the concept of rightness is to acquaint ourselves with all the occasions in which it applies correctly to individual actions. But as the number of right actions is infinite, we, as finite beings, could never learn the concept of rightness in such a way. 5 And this amounts to a reductio of the claim that there is no pattern that governs the meaning of rightness. For the only way we could learn the concept of rightness is via learning its meaning-rule, or so the Canberrans assume. 6 But the Canberrans have not canvassed all the plausible means by which we can come to acquire the concept of rightness, so the particularists argue. Indeed, citing Roschian prototype theory of learning, Dancy argues against the classical theory of learning the Canberrans assume. 7 Roughly, the contrast between the classical theory and the prototype theory lies in the fact that the former claims that every concept is governed by a meaning-rule that defines its necessary and sufficient conditions whereas the latter claims that a concept is not governed by a meaning rule that delimits the boundaries a concept can apply; rather, a concept has some paradigmatic examples with which the members of the same concept share some family resemblances. 8 In fact, the Canberrans acknowledge the above objection along a similar line, which goes roughly as follows: There need not be any pattern for us to be competent with the concept of rightness. “A diet of examples, or putative examples, can give us understanding of a term, can allow us to grasp a concept, without its being the case that there is a. 5. Darin Davis, Rules and Vision: Particularism in Contemporary Ethics, p. 87; see also Paul Boghossian “The Rule-Following Considerations” in Alexander Miller and Crispin Wright (eds.), Rule-Following and Meaning, (Chesham: Acumen 2002), p. 141. 6 Kihlbom argues that it rings similar to Hare’s view on moral learning. For Hare claims that teaching is always the teaching of a moral principle. But there is a difference here. Hare merely claims that in the situations that are exactly the same in their morally relevant features, the same moral predicate applies. But this is quite compatible with there being no unifying feature between different types of situations in which the same moral predicate applies. See Ulrik Kihlbom, Ethical Particularism: An Essay On Moral Reasons, (Stockholm: Stockholm University 2002), p. 85. 7 Jonathan Dancy, Ethics Without Principles, (Oxford: Oxford University Press 2004), p. 111; see also his “Can the Particularist Learn the Difference between Right and Wrong?”, in K. Brinkmann (eds.), The Proceedings of the Twentieth World Congress of Philosophy, vol. 1: Ethics (Bowling Green, OH: Philosophy Documentation Center 1999), pp. 59-72. 8 See Paul Horwich, “Meaning as Use”, Ch 3, in Meaning, (Oxford: Oxford University Press 1999), p. 52..

(6) 84 NCCU Philosophical Journal Vol. 23. pattern exemplified by the examples, namely, the pattern whose grasp underlies our ability to say of new cases whether or not they fall under the concept.” 9 But the Canberrans also argue that there is more than meets the eye and it is merely an illusion that there is no pattern that we latch on to when we learn a concept. The source of illusion originates in the thought that “in the case of family resemblance concepts, new cases often call for decision.” 10 This thought certainly has a strong grip on us. Take the concept of game for example. As Wittgenstein has vividly illustrated, there does not appear to be any unifying feature of all the activities we call games. And as this is so, when we face a new kind of activity, judgment is called for to determine whether it is a game or not. The concept of rightness is similar. If the concept of rightness had clear-cut boundaries of meaning, then no decision would be called for in the borderline cases. But apparently this is not the case for decision is indeed called for there. But again, we need to be reminded of the fact that this thought is the source of illusion, according to the Canberrans, and should be debunked. The Canberrans argue that in the particularists’ construal of the concept of rightness, it is actually not analogous to the concept of game, because in those family resemblance cases, “it [might] be difficult to spot or state the pattern, but the fact that, given a large enough diet of examples, we can say of some new case whether or not it is, say, a game (or, perhaps that it is indeterminate whether it is or not) shows that there is a pattern we can latch on to; our ability to project shows that we have discerned the complex commonality that constitutes the pattern.” But according to the Canberrans, the particularists argue that. 9. Frank Jackson, Philip Pettit, and Michael Smith, “Ethical Particularism and Patterns” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 87; see also Michael Smith, The Moral Problem, (Malden: Blackwell 1995) p. 55. 10 Frank Jackson, Philip Pettit, and Michael Smith, “Ethical Particularism and Patterns” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), pp. 87-88..

(7) Particularism, Pattern and Rule-following 85. there is not even a highly disjunctive commonality or pattern that unites the right acts when described in descriptive terms. 11 Two comments are in place here. First, the Canberrans’ construal of Wittgenstein’s example of game is mistaken, as Jay Garfield correctly pointed out. For Wittgenstein certainly would not think that all the activities we call games have a disjunctive commonality. According to Garfield, Wittgenstein argues that “no finite disjunctions could do and infinite disjunctions are neither well-formed in human natural language nor, qua disjunctions, learnable.” And to argue that there must be a disjunctive commonality to all the members of a family resemblance concept is question-begging in this context, as the particularists argue with Wittgenstein that we can still grasp the concept of rightness despite the fact that there is no commonality that unifies its members. 12 If commonality is construed as disjunctive, then the particularists need not deny that there might be such a disjunctive commonality amongst all the individual members of a concept. What they deny is a unifying commonality. Second, why think that there must be a pattern that governs the meaning of game given that there is indeterminacy in borderline cases and some judgment is called for? The Canberrans think that they have a reply for this. They argue: “We can all agree that there are cases where it is indeterminate whether or not some concept or term applies [but] this is consistent with there being a pattern because it can be indeterminate whether or not a pattern is exemplified.” 13 (emphasis added) And the Canberrans obviously think that the above statement is true of the concept of rightness. To some this may sound like a metaphysical cop-out. But I think under a more charitable reading, what the Canberrans are claiming is that the particularists have not given us any conclusive reason for thinking that there is no pattern. This is certainly right, given that our. 11. Frank Jackson, Philip Pettit, and Michael Smith, “Ethical Particularism and Patterns” in Brad Hooker and Margaret Little (eds.), Moral Particularism, p. 83. 12 Jay Garfield, “Particularity and Principle” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 190, footnote 25. 13 Frank Jackson, Philip Pettit, and Michael Smith, “Ethical Particularism and Patterns” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 88..

(8) 86 NCCU Philosophical Journal Vol. 23. learning process can be given a pattern interpretation. But neither have the Canberrans shown us that there must be a pattern either. The existence of a pattern seems to be merely a conjecture. Logical consistency is too weak to establish the existence of a pattern. The claim that there is indeterminacy in some cases, though consistent with the claim that there is a pattern, is also consistent with there being no pattern at all. The Canberrans do not provide any reason for us to believe in the one rather than the other. Some might think that Williamson’s epistemicist view about vagueness might lend some independent support for the existence of a pattern here. It might well be argued that in cases where it is indeterminate how many grains of sands it takes to make a heap of sands, this merely reveals our ignorance of the metaphysical fact about the number of sands it takes to make a heap rather than metaphysical indeterminacy in the concept of heap itself. And indeed, there are some independent reasons to think that the indeterminacy is merely epistemic rather than metaphysical. One might be motivated by the strong belief in the principle of bi-valence for instance. 14 However, it seems to me that Williamson’s view does not really help the Canberrans qua principlists. Even if in the borderline cases there is a matter of a fact about whether the concept of rightness applies, it does not establish the existence of a pattern. For the particularists might well accept that there is a determinate fact about whether the concept of rightness applies in the borderline cases. Nevertheless, it is still open to the particularists to question whether in all the cases, including the borderline cases and the non-borderline cases, in which the concept of rightness applies there is a unifying feature of rightness. So even if there is a determinate fact in the borderline cases, this does not provide any stronger reason to believe that there is a unifying pattern rather than that there is none.. 14. Timothy Williamson, Vagueness, (New York: Routledge, 1996), ch. 7..

(9) Particularism, Pattern and Rule-following 87. IV. Normativity of Meaning There is another thesis the Canberrans advance to support their claim that there must be a meaning-rule that governs the meaning of rightness. It is the thesis of the normativity of meaning. Roughly it says that for there to be a distinction between correct and incorrect uses of a term, there must be a meaning rule. Note that this need not beg the question against Kripke’s sceptic. 15 For the meaning rule can have reference to a shared “form of life” in the same linguistic community. Namely, an appeal to the “skeptical solution”, 16 in Kripke’s terminology, is available to the Canberrans. Yet, according to the Canberrans, the particularists, or at least many of them, impressed (perhaps excessively) with the indeterminacy of meaning in the borderline cases, argue that there is no meaning rule that governs the meaning of rightness. This does indeed smack of what many particularists, McDowell in particular, in fact argue. Citing Wittgenstein’s example of an individual extending a series of numbers, McDowell seems to be making the skeptical point that there is really no fact of the matter about which rule the individual is following. No appeal to the facts about the individual’s psychological mechanism could justify the claim that the correct move after 996, 998, and 1000 is 1002 instead of 1004. In a similar vein, no rule can be said to govern the meaning of rightness. 17 McDowell’s discussion of the hard cases seems to further confirm the impression that he argues that no rules govern the application of the concept of rightness in the hard cases. He contends: “If one is convinced that one is in the right on a hard case, one will find oneself say-. 15. Saul Kripke, Wittgenstein on Rules and Private Language, (Oxford: Blackwell 1982), chapter 2. 16 Ibid, chapter 3. 17 John McDowell, “Virtue and Reason” in Mind, Value & Reality (Massachusetts: Harvard University Press, 2002), p. 59; see also his “Non-Cognitivism and Rule-Following” in Mind, Value & Reality, (Massachusetts: Harvard University Press 2002), pp. 206-207..

(10) 88 NCCU Philosophical Journal Vol. 23. ing, as one’s arguments tail off without securing acceptance, ‘You simply aren’t seeing it’, or ‘But don’t you see?’” 18 And he goes on to rebut the principlists’ claim that “it is possible, in principle, to spell out a universal formula that specifies, in unproblematic terms, the conditions under which the concept one intends is correctly applied.” 19 For in those hard cases, the rule seems to run out and “the issue seems to turn on that appreciation of the particular instance whose absence is deplored in ‘You simply aren’t seeing it’”. 20 If what is said above is what McDowell, or the particularists in general, mean, namely, that there is no “universal formula” that governs the correct application of a concept, then the Canberrans are right in asserting that for the particularists not only those borderline hard cases require judgment but every new case calls for judgment too. And of course, this does not sound plausible. Take the concept of redness for example. In the non-borderline cases, there seems to be obviously a pattern that governs the meaning of redness. For in those cases, unlike the borderline cases, there is a clear distinction between the correct application of a term and the incorrect one. Little, if any, judgment is called for to see whether the concept applies or not. 21 Indeed, as the Canberrans point out, a wholesale meaning skepticism is too high a price for the particularists to pay. 22 The particularists, if they embrace meaning skepticism, would obviously have difficulty explaining the clear-cut distinction between the correct application of a concept and the incorrect one in the non-borderline cases. However, I don’t think the particularists embrace meaning skepticism. In fact, McDowell quite explicitly stresses that he should not be taken as expressing skepticism. Rather, what he tries to do is to shift the ground of our confidence for the claim that there is a particular. 18. John McDowell, “Non-Cognitivism and Rule-Following” in Mind, Value & Reality, (Massachusetts: Harvard University Press 2002), p. 209. 19 Ibid, p. 209. 20 Ibid, p. 210. 21 H.L. A. Hart also argues that in the “plain case”, the recognition of instances of a pattern seems unproblematic or “automatic”. See his The Concept of Law, (Oxford: Clarendon Press, 1961), p. 123. 22 Frank Jackson, Philip Pettit, and Michael Smith, “Ethical Particularism and Patterns” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 88..

(11) Particularism, Pattern and Rule-following 89. meaning rule we are following from individual psychology to the whole community. 23 And nothing of what he has said “cast[s] any doubt on the idea that the correctness of a move, [in the case of an individual extending a series of numbers], can be proved---so that it is compulsory to go on like that.” 24 “The point is just that we should not misidentify the perspective from which this necessity is discernible.” 25 But one thing to be noted here is that as I pointed out earlier the Canberrans need not deny that the normativity of the meaning of a term is essentially grounded in the shared “form of life” rather than individual mental states. 26 What they argue is that as long as there is a rule that governs the meaning of rightness such as, let’s say, “maximizing happiness”, then no matter what it is that provides the most solid ground for the rule, there is a corresponding true moral principle such as “maximizing happiness is right”. In view of the principlists’ claim above, one cannot help but wonder whether the particularists could coherently maintain that there is a meaning rule grounded by the community, a hard-edged rule with no fuzzy boundaries. For if the particularists were to admit that there is such a meaning rule, let’s say X, that governs the meaning of rightness, then the particularists would have to admit inconsistently that there is a true moral principle in the form of “X is right.”, thereby giving up particularism. However, I don’t think there would necessarily be inconsistency involved in the particularists’ position if they were to embrace a hard-edged rule grounded by the community. For instance, suppose that the community has all along been following the quus rule. Isn’t the quus rule just as much hard-edged as the plus rule in that neither of them allows for fuzziness in their applications? To illustrate, if according to the quus rule, the correct move for an individual to make after 996, 998, and 1000 is 1004 instead of 1002, then if the individual were to make the 1002 move, the individual would be clearly making a mis-. 23. John McDowell, “Non-Cognitivism and Rule-Following” in Mind, Value & Reality (Massachusetts: Harvard University Press 2002), p. 205. 24 Ibid, p. 208. 25 Ibid, p. 208. 26 Frank Jackson, From Metaphysics to Ethics, (Oxford: Oxford University Press 1998), p. 137..

(12) 90 NCCU Philosophical Journal Vol. 23. take in that she deviates from the meaning rule the community approves of. There is nothing indeterminate here in the case. 27 In a similar vein, one could argue that the gright rule is just no less hard-edged than the right rule, if the right rule is hard-edged at all, in that the gright rule allows no fuzziness in its application. But presumably, the gright kind of rules are not the sort of true moral principles the principlists wish to establish and hence they are not the sort of creatures the particularists would have to set their faces against. So admitting that there is a hard-edged rule approved by the community does not necessarily commit the particularists to any sort of inconsistency. Problems of inconsistencies only arise for the particularists if the actual rule we follow in our actual community, namely the right rule instead of the gright rule, is hard-edged for that matter. But apparently this is not the case if we take a look at the linguistic practice in our own linguistic community. If someone applies “fruit” to tomato, then it would take a stickler to insist that she misapplies the word “fruit” 28 Similarly, if someone applies the word “right” to an act of abortion, it would take a hard-core religious fanatic to insist that she misapplies the word right. For the particularists, the actions the word “right” applies to have no common unifying feature but are connected with each other by a network of similarities. Here, if one holds the view that the distinction between correct and incorrect uses of a term has to take the form of a hard-edged rule, as the Canberrans apparently do, then one might suspect that in rejecting a hard-edged rule that governs the meaning of rightness, the particularist is susceptible to the charge of being a Humpty Dumpty, who holds an “anything goes” attitude towards the meaning of the term rightness. Of course, this is absurd and particularism would certainly be rendered implausible if it had this implication. But the particularists could coherently argue that there is a distinction between correct and incorrect uses of a term without subscrib-. 27. Saul Kripke makes essentially the same point in his Wittgenstein on Rules and Private Language, (Oxford: Blackwell 1982), p. 82. 28 As many regard tomatoes as a kind of vegetable. Huw Price makes a similar point in his manuscript “Expressivism, Pluralism and Representationalism---A New Bifurcation Thesis”, p. 18..

(13) Particularism, Pattern and Rule-following 91. ing to the existence of a hard-edged pattern. The criterion of the correct application of a term should be distinguished from a hard-edged pattern. Only when one supposes that the distinction between correct and incorrect uses has to take the form of a hard-edged meaning rule can one take the particularists to task. But this supposition is not well-grounded, as evidenced by the application of many family resemblance concepts. Take the concept “fruit” again for example. The members of this concept are connected by a network of similarities. It would certainly be a misuse of the term “fruit”, if someone uses it to refer to something that does not share anything in common with any members of a concept, let’s say, a rock. And in the borderline cases whether a concept is correctly applied is very much determined by the linguistic community. Very often, some flexibility is allowed in the borderline cases. If someone tells me tomato is a fruit rather than a vegetable, I don’t think he misapplies the word “fruit”. To say that there is a meaning rule that determines absolutely whether it is correct or not to apply such a word seems to make our linguistic practice more rigid than it is. 29 What is said about the use of the term fruit can also be said about the term rightness. Certainly, there is no common feature of rightness formulated in the form of a hard-edged meaning rule that governs the meaning of rightness if we follow Wittgenstein’s sound advice “Don’t think but look!” Nevertheless, this does not mean that there is no distinction between the correct and the incorrect uses of the term rightness. If someone applies the term right to an action (token) that is obviously wrong, he certainly misapplies the term. And in those borderline cases, flexibility in the application of a term is allowed. Even a self-claimed anti-particularist Bernard Gert admits (or argues indeed) that thinking that there is one unique correct answer when neither of the contrary moral judgments is not irrational in the borderline cases would be a sign of moral arrogance. 30. 29. Richard Hare, Freedom and Reason, (Oxford: Oxford University Press 1963), p. 14. Bernard Gert, “Moral Arrogance and Moral Theories”, Nous Supplementary Volume, Philosophical Issues, 15, Normativity, 2005, pp. 368-385.. 30.

(14) 92 NCCU Philosophical Journal Vol. 23. V. Rationality Requirement: The Thesis of Consistency Some principlists argue that there must be a meaning rule that governs the meaning of rightness; otherwise, we will not be using the word consistently. This thought is related to a specific conception of rationality epitomized in the writing of Hare. 31 Hare argues that part of the meaning of rightness lies in its descriptive meaning. And this fact requires us to consistently apply a moral term to any other acts that are similar in relevant descriptive aspects just like it requires us to apply a purely descriptive term to anything that is similar in relevant descriptive aspects. If not, we are simply inconsistent with our use of the word. And as rationality requires consistency, it is not rational for us to apply a concept differently. So it seems that rationality means following a rule consistently. But McDowell argues that this is due to a prejudiced conception of rationality. 32 Rationality does not require the existence of a meaning rule. As I said earlier, an individual could have been following a gright rule. It would be consistent with the gright rule for the individual to call action A right while calling action B wrong even if action A and action B are exactly the same in their morally relevant aspects, just as it would be consistent with the quus rule that the individual makes the 1004 move after 996, 998, and 1000. But again, we need to be reminded of the fact that McDowell is not making a skeptical point. Rationality does indeed require us to follow the plus rule rather than the quus one consistently when we are extending a series of numbers. It is just that there is no objective platonic fact that determines what rule it is that we should be following to meet the requirement of rationality. From what McDowell calls the “platonic perspective”, the following of the quus rule is just as rational as the. 31. John McDowell, “Non-Cognitivism and Rule-Following” in Mind, Value & Reality (Massachusetts: Harvard University Press 2002), p. 202. 32 John McDowell, “Virtue and Reason” in Mind, Value & Reality, (Massachusetts: Harvard University Press 2002), p. 58..

(15) Particularism, Pattern and Rule-following 93. following of the plus rule. 33 It is only when looking from a community’s perspective can we say that making the 1004 move after 996, 998 and 1000 is irrational. Similarly, from the platonic perspective, the following of the gright rule is just as rational as the following of the right rule. It is only when looking from a community’s perspective can we say that the following of the gright rule is irrational. But again, this would suffice for the purpose of the principlists. The Canberrans might well admit that there is no objective perspective from which rationality requires the following of a rule. But as long as from the community’s perspective, rationality requires that we as members of the same shared form of life ought to follow the meaning rule in our community consistently, 34 then it would be enough to establish the claim that without such a rule there would be no distinction between rationality and lunacy in our community. But as I see it, the requirement of rationality does not really help to show that the meaning of rightness is governed by a unifying meaning rule. It may well be true that rationality requires us to follow a rule consistently as Hare suggests. But this is compatible with there being no unifying meaning rule that governs the meaning of rightness. Namely, the particularists might well admit that if we apply the word right to an act of telling a white lie, then rationality would require us to apply the word right to any act that is exactly similar to this one in morally relevant aspects. If we call a second action not right that has exactly the same morally relevant features as those of the first, then we deviate from the meaning rule that governs the meaning of rightness in this type of case and can be rightly accused of being irrational by our community’s standard. Let us call the meaning rule that governs the meaning of rightness in this type of case D1. In a similar vein, one could argue that there is a meaning rule, let’s say D2, that governs the meaning of rightness in a type of case of fulfilling one’s promise, and. 33. Simon Blackburn makes the same point in Spreading the Word, (New York: Oxford University Press 1984), p. 79. 34 Of course, this might fly in the face of the possibility of a James Joycean idiolect. But this charge is not available to the Canberrans as they argue that there must be a (exceptionless) meaning rule that we follow. If the meaning rule could have exceptions, this would amount to saying that the moral principle could have exceptions; principlism would thus lose its distinction from particularism..

(16) 94 NCCU Philosophical Journal Vol. 23. another meaning rule, let’s say, D3, that governs the meaning of rightness in a type of case of saving a drowning child. Although rationality requires that one follows D1 consistently in the case of telling a white lie, D2 in the case of fulfilling one’s promise, D3 in the case of saving a drowning child, it does not show that rightness has a unifying pattern that governs its meaning in these three different types of cases. Maybe there is a unifying feature of D1, D2, and D3, and in that sense the meaning of rightness can be codified into the unifying feature, but rationality in itself does not show that. So it does not seem to be the case that the requirement of consistency or rationality would help the principlists establish their claim that there is a meaning rule that governs the meaning of rightness.. VI. Rationality Requirement Continued: The Thesis of Universalizability Following the thread of the argument from the last section, some might argue that the principlists wouldn’t have to go so far as to claim that there is a unifying feature of D1, D2, and D3 to establish their claim that there are true moral principles. The existence of D1, D2 and D3 in themselves are sufficient to establish that there are true moral principles. And the existence of D1, D2 and D3 just seems to follow from the thesis of universalizability. Roughly, the thesis of universalizability says that a moral verdict must be universalizable in the sense that the same moral verdict should be applied to all actions that have exactly the same morally relevant features. If we admit this, we must admit the existence of a meaning rule that governs the meaning of rightness when it is applied to each individual type of actions. I will discuss two kinds of particularists’ replies in what follows. I think neither of them are successful and I will offer a more promising reply on behalf of the particularists..

(17) Particularism, Pattern and Rule-following 95. One typical response from the particularists is to say that the thesis of universalizability is useless for action-guiding because we cannot capture exactly what the morally relevant aspects of the action are. To illustrate this point, take the case of telling a white lie for instance. If an action of telling a white lie is right, the thesis of universalizability does not say that any action of telling a white lie is right; rather it says that any action that is similar to the first action in morally relevant aspects is right. Given the nuances and complexities that might be involved in the contexts of telling a white lie, it seems extremely difficult to formulate the morally relevant aspects. For instance, it might be thought that telling this particular white lie is right because doing so would help prevent hurting others’ feelings. Yet, there are circumstances in which doing so is not allowed even if it would help prevent hurting others’ feelings. Doctors in many cases have the obligation to tell their patients the truth even if doing so would hurt their patients’ feelings. And this shows that the morally relevant aspects, in virtue of having which the particular white lie derives its rightness, are not really the features involved in preventing hurting others’ feelings. For not all white lies that involve the features of avoiding hurting others’ feelings are right. The morally relevant aspects are more subtle than those features. Given the complexities of the moral situations we might encounter, the particularists are usually quite skeptical about whether anyone is capable of formulating the morally relevant aspects precisely. However, for the thesis of universalizability to be useful in terms of guiding actions, the morally relevant aspects must be formulated precisely. Otherwise, it is just useless. Look at the following imagined conversation between a puzzled student and his master: Student: Master, what are the morally relevant aspects that make all morally right white lies morally right and set them apart from the morally wrong ones? Master: That is a good question. All the white lies that have features XYZ would have the property of moral rightness and features XYZ are the features that set all the morally right white lies apart from the morally wrong ones. Student: What are features XYZ?.

(18) 96 NCCU Philosophical Journal Vol. 23. Master: I am sorry to tell you that I cannot formulate them. I trust that if readers have the same intuitions as I do, the master does not really provide helpful guidance to the student. If so, I think that it is not implausible to conclude with the particularists that if the thesis of universalizability is to provide guidance to the puzzled, the morally relevant aspects must be precisely formulated; otherwise, the thesis of universalizability is just useless for that matter. But the principlists may well argue that for an agent to be guided by the thesis of universalizability, she need not necessarily be able to formulate the morally relevant aspects. As long as she is capable of applying the right predicate to every action that is similar to the first action of telling a white lie in morally relevant aspects, she can be said to be guided by the thesis of universalizability implicitly despite the fact she is not capable of formulating what those morally relevant aspects are. However, to be fair to the particularists, we need to distinguish two contexts in which the thesis of universalizability is meant to be action-guiding. One is the context in which the agent makes a moral judgment based on a pre-established true moral verdict and the universalized principle derived from it. In this sort of context, the principlists might well be right that in being guided by the universalized principle, one need not be able to formulate the morally relevant features. To illustrate this with an analogy, a chicken sexer might well be guided by a principle to tell male chickens from female ones although when asked what the relevant features are according to which he makes his judgment, he might be completely at a loss. Similarly, a virtuous agent might be guided by a principle that sets the right actions apart from the wrong ones although he is not capable of articulating it. But there is another sort of context in which the thesis of universalizability is meant to be action-guiding. It is the kind of context in which a puzzled student asks for moral advice, as illustrated by the above imagined conversation between the student and his master. It is in this sort of context the particularists claim that the thesis of universalizability cannot be action guiding without a clear formulation of the morally relevant aspects of an action. It should be noted, however, that quite a number of principlists are not reluctant to bite the bullet and argue that despite the complexity.

(19) Particularism, Pattern and Rule-following 97. of the moral situations one might encounter, the morally relevant aspects can nevertheless be produced. But, as I see it, once the morally relevant aspects are identified by the principlists, it will be no wonder if the particularists carry on with generating counterexamples to the claim that the same moral predicate applies to all actions that have exactly those morally relevant aspects. As Little rightly comments, “there’s something not [just] a little farcical about measuring dialectic success in terms of who can outlast whom---those who want to refine the principles or those who want to find exceptions.” 35 Such a debate, presumably, will not be very fruitful. The real problem with the particularists, as it seems to me, does not lie in their claim that the morally relevant features cannot be precisely formulated, as the principlists contend, but rather in the fact that even if the morally relevant aspects cannot be precisely formulated and hence the thesis of universalizability cannot be action-guiding, it does not show that there is no meaning rule. The particularists simply miss the target. For even if we cannot formulate the morally relevant aspects of an action, let’s say, action A, for the universalized principle derived from it to be useful, there is nevertheless a meaning rule which says that we must apply the same moral predicate we apply to action A to any action with exactly the same morally relevant features as those of action A’s. Of course, this might not be very useful in terms of action guiding if the morally relevant features are left unspecified. But the action guiding aspect of the moral principle should not be confused with the ontological aspect of the moral principle. Even if a moral principle is not very useful in action guiding, this does not mean that it is not true.. 35. Margaret Little, “Moral Generalities Revisited” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 279..

(20) 98 NCCU Philosophical Journal Vol. 23. VII. Falsifying the Thesis of Universalizability A more germane response from the particularists is to argue that the thesis of universalizability is false. If it is false, then one could call one action right but the other action with exactly the same morally relevant features wrong. Namely, if the thesis of universalizability is false, it provides us with no good reason to think that there is a meaning rule that requires us to apply the same moral predicate to all actions that have exactly the same morally relevant features. But as this is quite counterintuitive, were particularism incompatible with the thesis of universalizability, this would seem more like a reason to reject particularism rather than one to reject the thesis of universalizability. 36 So how can the particularists’ claim gain at least some initial plausibility? Citing an example from Melville’s novel, Billy Budd, Winch argues on behalf of the particularists that killing Billy Budd is right for Vere but wrong for himself because the same set of morally relevant considerations struck them differently as a result of their different personal characters. And hence it seems that the thesis of universalizability is false because two actions that have exactly the same morally relevant features could still have different moral properties if they strike the two persons in different ways. 37 For this argument to succeed, however, the personal character should not be included in the morally relevant features. For if Vere and Winch’s different personal characters were included in the morally relevant features of the action, then this would not falsify the thesis of universalizability. For the thesis of universalizability says that the same moral predicate must apply to all actions that have exactly the same morally relevant features; in the case of Billy Budd, if personal character were included in the morally relevant features of the action, Vere’s action of killing Billy Budd would have morally relevant features dif-. 36. Roger Crisp, “Particularizing Particularism” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 42. 37 Peter Winch, “The Universalizability of Moral Judgments”, Ethics and Action (London: Routledge and Kegan Paul 1972), pp. 151-170..

(21) Particularism, Pattern and Rule-following 99. ferent from those of Winch’s action of killing Billy Budd as Vere and Winch have different personal characters. And if the two actions do turn out to have contrary moral properties, this would not falsify the thesis of universalizability as they have different morally relevant features. For to falsify the thesis of universalizability, we need to find a case in which the two actions have exactly the same morally relevant features but nevertheless have contrary moral properties. So are there any good reasons to exclude personal character from the morally relevant features in the case of Billy Budd? The particularists typically argue that there are two sorts of reasons to support the claim that personal character should not be included in the morally relevant features in the particular context of Billy Budd. One is the reason from self-discovery. The other is the reason from triviality. I am going to argue in what follows that neither of these is compelling. First, Raz argues that if personal character were included in the morally relevant features, then Winch would not have undergone a process of self-discovery 38 when he imagined himself making the moral decision. However, Raz argues that this does not square with the moral phenomenology. For Winch did discover something about himself, that he could not bring himself to hang Billy Budd. If he had regarded his personal character as one of those morally relevant features that ought to be included in moral considerations, then there would be nothing new he discovered about himself when he imagined himself having to make the moral decision. But here, I think we need to make a distinction between there being a reason for the agent to Ø and the agent having a reason to Ø . It may well be true that Winch’s own personal character was not among his reasons (or the morally relevant features he takes into consideration) to Ø . But nevertheless, if personal character would make a difference to the moral properties of the action in the case of Billy Budd, as Winch clearly indicates, it should nevertheless be treated as a reason or a morally relevant feature which plays a role in the determination of the. 38. Joseph Raz, “The Truth in Particularism” in Brad Hooker and Margaret Little (eds.), Moral Particularism, (Oxford: Oxford University Press 2000), p. 76; see also Stephen Darwall, Philosophical Ethics, (Boulder: Colorado Westview Press 1998), p. 7 for a similar idea on how making a moral decision involves a process of self-discovery..

(22) 100 NCCU Philosophical Journal Vol. 23. moral status of the action. Namely, the fact that Winch has a certain sort of personal character was among the reasons contributing to the determination of the moral status of the action despite the fact that he himself did not see it as a reason. If so, the Billy Budd case is not really incompatible with the thesis of universalizability, because for the thesis of universalizability to hold, this construal of there being a reason for the agent to Ø is all that is needed. In the case of Billy Budd, the thesis of universalizabilty would say, as I said earlier, that for anyone, if she had the character of Winch, the action of hanging Billy Budd would not be right for her whereas for anyone, if she had the character of Vere, the same sort of action would be right for her. Some particularists argue that if one counts the character of the person as a morally relevant feature in the case of Billy Budd, the thesis of universalizability would be made trivially true. For if Vere finds it right to hang Billy Budd, then it is merely trivially true that anyone who has the same character as that of Vere would find it right to hang Billy Budd too. It is trivially true in the sense that it is unlikely that anyone except Vere himself could have his own personal character. So what we got from the thesis of universalizability might be nothing more than the tautological statement that if Vere finds it right to hang Billy Budd, he finds it right to hang Billy Budd. Although this statement is trivial, it is no argument against its truth simply by calling it nasty names. If particularism has to hang its hat, so to speak, on the rejection of the thesis of universalizability, then particularism would seem to be a lost cause. However, I don’t think that the particularists need to reject the thesis of universalizability for it is not incompatible with particularism. For the meaning rule derived from the thesis of universalizabilty merely requires that we must apply the same moral predicate to every action that has exactly the same morally relevant features as those of the first action. But this says nothing about whether there are any two actions that have exactly the same morally relevant features. Whether it is true that there must be a second action that has exactly the same morally relevant features as those of the first has been.

(23) Particularism, Pattern and Rule-following 101. questioned by Dancy and Nussbaum. 39 For the second action, though having the morally relevant features same as those of the first, might have further morally relevant features. To exclude this possibility, the set of the morally relevant features of an action must be expanded to include any features that might come to be morally relevant in other contexts. But as any feature can be morally relevant when given the right context (indeed including the color of your shoelace according to the particularists), once the expansion is allowed, there is really no stopping point and the set of morally relevant features of an action is likely to include the totality of its features. When the set of the morally relevant features of an action is expanded thus wide, then it is not clear whether any other action could have a set of morally relevant features exactly the same as the one of the first action. If there is not, then it is not clear whether what the thesis of universalizability establishes is a moral principle or merely a moral verdict. The difference between the two lies in the fact that a moral principle is a generalization about the moral status of an action type whereas a moral verdict, in my construal, can be about the moral status of a single action token.. VIII. Irrelevance of Meaning Pattern Despite the particularists’ efforts to show that there is no unifying pattern, they are nevertheless open to the possibility that there can be a meaning rule that governs the meaning of rightness. In other words, the meaning rule can be characterized as one that allows fuzzy boundaries yet nevertheless retains the normativity of its meaning. It is also one which allows merely family resemblances amongst the members of a concept without there being a unifying feature. And the meaning rule is by and large determined by the community. But it would be most unfortunate if particularism has to hang its hat on the communitarian assumption, begging an important question against the latter Davidson, who argues that no appeal to conventions, rules or communities is necessary for there to be a distinction between. 39. Jonathan Dancy, Moral Reason, (Oxford: Blackwell 1993), p. 78; Martha Nussbaum, Love’s Knowledge, (New York: Oxford University Press 1990), p. 72..

(24) 102 NCCU Philosophical Journal Vol. 23. correct and incorrect uses of a term. Even a soft-edged rule that allows family resemblances is not necessary. What really determines the correct use of a term is its role in the process of communication. 40 Having said this, I don’t think the particularists would need to hang their hat on this. For even if the Canberrans’ claim is true that there is a unifying hard-edged pattern that governs the meaning of rightness, it might still be the case that there are no true moral principles of the sort the principlists have in mind. For a moral principle is not equivalent to a meaning rule about the meaning of rightness or wrongness; rather, it is a general criterion of what is right or what is wrong. To see this, let’s think about the moral debate between people who hold different moral principles. If we admit that there is a substantive moral debate between, let’s say, the consequentialists and the deontologists, then it seems to me much more plausible to see them as arguing over the general criterion of rightness rather than merely over the meaning of rightness. If this were the other way around, then there would be no way to distinguish them from two lexicographers debating the meaning of the term rightness. But of course, this does not seem plausible. In general, people do not take out guns and knives when their uses of a term differ. They are much more likely to do so when the criterion of rightness they live by clash. 41 So it seems that there is a distinction between the meaning of the term rightness and the criterion of rightness. If this makes sense, it seems that there can still be an issue about the general criterion of what is right and wrong even if both disputants agree on the meaning of the term rightness or wrongness. Say if they agree that “right” just means “morally permissible”, there is still an issue about what the unifying feature is of all the things that are morally permissible. The particularists may well agree with the principlists that the right actions are those actions that have the property of moral permissibility but still maintain that there is no unifying feature of the ac-. 40. Donald Davidson, Essay 18“Communication and Convention”, Inquiries into Truth and Interpretation, (Oxford: Oxford University Press 2001) pp. 265-280; “A Nice Derangement of Epitaphs” Trut, Language and History,(Oxford: Oxford University Press 2005), pp. 89-107. 41 I owe this point to Denis Robison’s talk at the AAP conference held in Australian National University 2006..

(25) Particularism, Pattern and Rule-following 103. tions that are morally permissible. So if one admits that the meaning of rightness is governed by a meaning pattern, it doesn’t seem to follow this that there is a general criterion of rightness (or a moral principle). Against the claims of the particularists, some might argue that if the meaning of rightness is governed by the meaning pattern of moral permissibility, then we may get a true moral principle which says that whatever action is morally permissible is right. Although this principle is somewhat trivial, nevertheless, again, calling it nasty names is no argument against its truth. Once this is admitted, particularism seems to be a lost cause. However, there is an important clarificatory question to be asked here, that is, whether this sort of moral principle is the kind of moral principle the principlists have in mind when they seek to establish the existence of true moral principles. If it is not, then the particularists need not set their faces against such creatures, for the particularits are only interested in taking issue with those moral principles the principlists actually have in mind. And indeed, if only we take a look at the history of ethics, it would take us no time to realize that taking the above mentioned principle to be the sort of principle the principlists have striven to establish would seem to be a caricature of the principlists’ enterprise. What are the real moral principles the principlists have in mind? To illustrate, they should include at least the following: the Kantian kind of categorical imperatives, the Millian kind of principle of utility, the ideal observer theory, the Hursthouse kind of virtue principle, Scanlonian kind of contract theory, the Richardsonian kind of specified principle, the Gertian kind of commonsensical principles or the Rossian kind of prima facie principles. None of these theorists are content with a moral principle which merely says that a morally permissible action is right. Moreover, if it were really the case that the trivial principle mentioned above is the sort of true moral principle the principlists have in mind, we might say that the principlists could easily win the debate by conjuring up an even more trivial moral principle which says that for any action, if it is right, it is right. But this does not sound plausible. For this sort of principle, if it is a principle at all, does not capture the sort of thing the principlists have in mind at all if again we remind ourselves of the history of the principlists’ attempt to establish their.

(26) 104 NCCU Philosophical Journal Vol. 23. favored moral principles. And hence, at the risk of repeating myself here, I don’t think that the particularists would have to deny that there can be a true moral principle like the above mentioned one if it is worthy of the name “principle” at all. In fact, a more powerful and more relevant objection the principlists can raise to the claims of the particularists is that if the meaning of rightness can be reduced to not just a trivial ethical pattern such as that of moral permissibility, but instead to a substantive non-ethical descriptive pattern, let’s say, “maximizing happiness” in the mature folk theory as described by Jackson in his book From Metaphysics to Ethics, then it seems to follow that there must be a true moral principle which claims that “maximizing happiness is right”. Indeed, this objection is motivated by Jackson’s moral functionalism. 42 However, one comment is appropriate here. I think that we need to distinguish two interpretations of the copula “is”. The sentence “maximizing happiness is right” can be interpreted in two different ways, depending on how one interprets the copula “is”. On the first interpretation, the meaning of right is equivalent to the meaning of maximizing happiness. The “is” expresses a relation of identity. Namely, the property of maximizing happiness and the property of rightness stand for the same property. On the second interpretation, the action of maximizing happiness has the property of rightness. The “is” is attributive. The action should not be identified with its property of rightness just as a rose should not be identified with its property of redness. The action doesn’t mean right just as the rose doesn’t mean red. I think that the Canberrans are facing a dilemma here. Only the first interpretation can illustrate their claim that the meaning of rightness can be reduced to a non-ethical pattern. However, it is clearly the second interpretation the utilitarians have in mind when they strive to establish the truth of the principle of utility. But it is not a dilemma the Canberrans cannot address. Indeed, they might well argue that if our meta-ethical inquiry reveals that there. 42. Frank Jackson, From Metaphysics to Ethics, (Oxford: Oxford University Press 1998), p. 145..

(27) Particularism, Pattern and Rule-following 105. is a relationship of type identity between the moral property of rightness and the property of maximizing happiness, as suggested by the first interpretation, then any action that has the property of maximizing happiness must have the moral property of rightness. In short, there is a true moral principle which says that maximizing happiness is right. Indeed, as Hussain and Sha pointed out, “One cannot claim that right just means maximizes utility without it following that if an action maximizes utility, then it is right.” 43 So there seems to be nothing wrong in seeing different principlists’ job as being to offer competing theories for the meaning of rightness if it is the case that their theories offer a substantive non-ethical pattern that governs the meaning of rightness. For once the non-ethical pattern is established, it would seem that a corresponding true moral principle will be established. However, if it were really the case that the different normative theories can be seen as competing accounts for non-ethical pattern that governs the meaning of rightness, then it would imply that when different principlists argue with each other over the moral status of an action, they are not really disagreeing with each other. For as Hare correctly pointed out, there cannot be moral disagreement without agreement on the meaning of the moral terms. Since consequentialists’ theory of the meaning of rightness is different from the deontologists’, they cannot really engage in a moral debate. According to this picture, when Francis Kamm, an anti-consequentialist, argues with Shelly Kagan, a consequentialist, over the moral status of killing one to save five in the Trolley case, they merely talk past each other, for they have different theories of the meaning of rightness. But were one to ask either of them whether their debate is merely linguistic, I think it is very unlikely that they would submit that there is no substantive moral issue between them. This is not to say that they cannot be mistaken about what they are doing. They might be really talking past each other when they think they are engaged in a substantive moral debate. But this claim would take a very compelling argument to justify, given the strong contrary intuition and evidence that. 43 Nadeem Hussain and Nishi Sha, “Misunderstanding Meta-Ethics”, in Russ Shafer-Landau (eds.), Oxford Studies in Meta-Ethics, (Oxford: Oxford University Press 2006), p. 269..

(28) 106 NCCU Philosophical Journal Vol. 23. what we mean by right is roughly the same, at least for people living in the same community. The process of indoctrination of the moral terms guarantees this more or less. 44 So far, I have not seen very compelling arguments against this produced by the Canberrans. And in fact, I don’t think the Canberrans can speak for the principlists here. For the principlists need not deny anything I have just mentioned above. Typically, they agree among themselves that there is a substantive moral debate between different principlists. So they also agree among themselves that their uses of the term rightness agree, for otherwise they would have to admit self-defeatedly that their moral debate is futile. But for their uses of the term rightness to agree, they cannot smuggle their favored moral theories into the meaning of rightness. But this is in fact what the Canberrans take the principlists to be doing. The Canberrans take the principlists to agree that they have no shared meaning of rightness. And this is why the Canberrans think that the principlists are offering competing accounts of which unifying meaning pattern is the right one. For once the unifying meaning pattern is established, then it just follows that there is the corresponding moral principle. And hence, they see it as the particularists’ job to deny that there is a unifying meaning pattern. Namely, the Canberrans see the debate between the principlists and the particularists as focused on whether there is a unifying meaning pattern. But by claiming that the principlists’ job is to establish a certain unifying meaning pattern, it seems to me that the Canberrans are putting the cart before the horse. It seems to me that if there is any substantive pattern, it must be borne out as the fruit of substantive moral debates. For without these moral debates, there is no way to clarify what the ultimate substantive pattern that governs the meaning of rightness is. Similarly, Sayre-McCord argues that the nature of the property rightness, and indeed the moral kind, is discovered through normative theories. 45. 44. Stephen Schiffer, “Meaning and Value”, The Journal of Philosophy, Vol. 87, No. 11, 1990, p. 610. 45 Geoffrey Sayre-McCord, “Good on Twin Earth”, Philosophical Issues, Vol. 8, Truth, 1997, pp. 291-292..

(29) Particularism, Pattern and Rule-following 107. But again, for the moral debates to be possible, there should be a shared meaning of rightness between different principlists. So it seems to me that the Canberrans have a dilemma. For on the one hand, they want to claim that what the different principlists have in common is that they all strive to establish their favored unifying meaning pattern of rightness (implying that they currently share no common meaning pattern of rightness). But on the other hand, if this were really so, moral debates would be impossible and there would be no way for the principlists to establish their favored unifying meaning pattern of rightness. In short, if it were really the case that the principlists are arguing for their favored unifying meaning pattern, the principlists would be defeating their own purpose. So it seems to me that the Canberrans’ characterization of the principlists’ task as offering competing accounts for the meaning pattern of rightness is not correct. 46 More plausibly, I think what the different principlists have in common is their belief in the existence of a general criterion of rightness. On my construal of the principlists’ task, there is no need for the principlists to deny that they share a common meaning pattern of rightness and hence substantive moral debates are possible between themselves. This common meaning pattern of rightness must be kept neutral such that it does not prejudge in favor of or against any particular normative theory. “Moral permissibility” and “the thing to do” are both plausible candidates. (But as I said earlier, even if it may follow from these meaning patterns that there are true moral principles such as “A morally permissible action is right”, they are not the sort of true principles the principlists have in mind. And hence, the particularists need not deny their existence if they are worthy of the name “moral principle” at all.) It may well be objected by the Canberrans that they would certainly agree with me that the principlists’ task is not offering competing accounts for the current folk’s meaning pattern of rightness for the reasons I have indicated, i.e., reasons that there would be no real moral debates without agreement on the meaning of rightness, but neverthe-. 46. Interestingly, Michael Smith expresses his doubt too when he discusses non-subjective definitional naturalism in The Moral Problem, (Oxford: Blackwell 1994), p. 43; I suspect he parts company with the other two Canberrans here..

(30) 108 NCCU Philosophical Journal Vol. 23. less maintain that the principlists’ task can still be seen as offering competing accounts for the mature folk’s meaning pattern of rightness. In that case, the principlists are not talking past each other when they are engaged in a moral debate because they all operate on the same current folk’s meaning pattern of rightness. And when the true moral principle bears out as the fruit of the moral debates, it would be definitive of the mature folk’s meaning pattern of rightness. So there is nothing wrong in seeing the debate between the particularists and the principlists as focused on whether there is a mature folk’s meaning pattern of rightness. For if there is, it must indicate that there must be a true moral principle. Is there any reason for thinking that there is? I haven’t seen any compelling arguments to indicate there is. The conceptual competence argument cannot provide support here. For the mature folk’s meaning pattern might be very different from the current folk’s meaning pattern which we are currently competent with. The argument from normativity of meaning cannot work either. For as I have argued, there can still be a distinction between correct and incorrect applications of a term without the existence of a unifying pattern, be it current folk’s or mature folk’s. And the argument from the thesis of consistency does not show that there must be a unifying pattern. For as I have shown, there can still be consistency in the use of the concept of rightness in different types of cases without any unifying pattern. And the thesis of universalizability is simply irrelevant here as what it establishes might not be a unifying pattern or a rule that governs the moral status of an action type but might merely be a verdict about the moral status of a single action token. Perhaps the only plausible reason, at least prima faciely, left for thinking that there must be a unifying pattern is that it just follows from the fact that there must be a true general criterion of rightness that is definitive of the mature folk’s unifying pattern. But this is exactly the issue at stake between the principlists and the particularists; making this presupposition would be begging the question against the particularists. Indeed, as Ridge and McKeever rightly point out, “Moral philosophy has long presupposed that morality is principled with Kantians, consequentialists and others arguing about which principles are correct..

(31) Particularism, Pattern and Rule-following 109. Unfortunately, this presupposition rarely receives a defense.” 47 In view of the challenges from the particularists, “we cannot take the fact that moral philosophy is dominated by principled approaches as evidence that morality really is principled.” 48 To take stock, I don’t think that the unifying pattern argument is relevant to the debate between the particularists and the principlists. If the unifying pattern is trivial, it is not clear whether the principlists themselves would endorse a pattern of this kind. If not, then the particularists need not oppose the existence of the unifying pattern. On the other hand, if the pattern is substantive, we need to ask whether the unifying pattern is the current folk’s or the mature folk’s. If it is the current folk’s pattern, then in order to find out what the pattern is, we must see the principlists’ task as offering competing accounts for the current folk’s meaning patterns of rightness. But as I have argued, it is wrong to see their task in that way because it would imply that they currently have no shared meaning of rightness, and thus further implying that no real moral disagreement is possible between different principlists. This is an implication few principlists themselves would accept. Finally, if the unifying pattern is the mature folk’s, there seems to be no compelling reasons for believing that there must be such a unifying pattern except perhaps by appealing to the belief that there must be a true moral principle definitive of it. But this would be begging the question against the particularists. To sum up, the principlists’ task is hard to make plausible if it is construed as a quest for a unifying meaning pattern, be it trivial or substantive, current folk’s or mature folk’s.. IX. Conclusion Finally, let me briefly summarize what I have done in this article. Firstly, I have argued that the principlists have provided no compelling evidence for showing that there is a unifying meaning rule that governs. 47. Sean McKeever and Michael Ridge, Principled Ethics: Generalism as a Regulative Ideal, (Oxford: Oxford University Press 2006), p. 1. 48 Ibid, p. 1..

(32) 110 NCCU Philosophical Journal Vol. 23. the meaning of rightness. None of the theses of conceptual competence, of normativity of meaning, of consistency and of universalizability provide compelling evidence for believing in the principlists’ claim that there is a unifying meaning rule. Secondly, I have argued that even if there is a meaning rule that governs the meaning of rightness, it does not follow that there are true moral principles. For the true moral principle that is established by the ethical unifying pattern of moral permissibility is not the sort of true moral principles the principlists wish to establish. It is rather doubtful the principlists themselves would grant it the status of a moral principle. Only when the unifying meaning pattern is substantive and non-ethical can it be said with some certainty that it follows from this pattern that there are corresponding true moral principles. However, this line of reasoning is not available to the principlists because to take this line of reasoning, the principlists would have to see their task as offering competing theories for the unifying meaning pattern of rightness, thus implying that they currently don’t have any shared meaning of rightness and there is really no moral disagreement amonst themselves. But this implication is something the principlists wouldn’t accept. A final resort for the principlists is to argue that the meaning of rightness the principlists are after is not the current folk’s meaning of rightness but the mature folk’s meaning of rightness. So they might agree that they share current folk’s meaning of rightness but still are offering competing accounts for the mature folk’s meaning of rightness. If it turns out that there is a substantive unifying pattern that governs the mature folk’s meaning of rightness, then it just follows that there must be corresponding true moral principles. Although this might sound plausible, what is not clear is how the principlists come to determine the unifying pattern that governs the mature folk meaning of rightness. It seems to me that the mature folk meaning of rightness could only be determined by the true general criterion of rightness converged upon after critical reflection and serious moral debates. But is there a general criterion of rightness? This is exactly the issue at stake between the principlists and the particularists. The principlists cannot maintain that there must be a mature folk uni-.

(33) Particularism, Pattern and Rule-following 111. fying meaning pattern of rightness without presupposing that there must be a true general criterion of rightness, thus without begging the question against the particularists. Given what is said above, I don’t think that arguing for the existence of the unifying meaning pattern, be it trivial or substantive, current folk’s or mature folk’s, will in any way help the principlists to establish their claim that there are true moral principles. 49. 49. Many thanks to the two anonymous reviewers for their comments and suggestions for revisions. One reviewer suggests that a forceful defense of particularism would require us to consider issues in metaphysics, anthropology, and natural theology. I agree. However, these issues are far more complicated than I can deal with in this paper. They can be topics for future research. I would also like to thank the audience at the Australasian Philosophy Postgraduate Conference held in Sydney University 2008, and the audience at the Ethics, Application, Theory: Postgraduate Ethics Conference held in Melbourne University 2007 for many stimulating discussions. I would like to thank especially Jeanette Kennett, Daniel Star, and Daniel Stoljar for providing many helpful comments on the earlier drafts of this paper. Finally, I would like to express my thanks to Taiwan’s Ministry of Education for its financial support during the time when this paper was written..