#3 32.3 Vorobej PubVers. Hybrid Arguments and Moral Relevance MARK VOROBEJ Department of Philosophy McMaster University 1280 Main Street West Hamilton, Ontario L8S 4K1 vorobej@mcmaster.ca Abstract: Some of Jonathan Dancy’s strongest arguments in sup- port of moral particularism depend crucially upon the distinction he draws between four different kinds of relevance relations—favourers, enablers, intensifiers and attenuators. I agree with Dancy that different premises can play fundamentally different structural roles within moral argumentation. However, in this paper I argue that enablers, in- tensifiers and attenuators can be ana- lyzed in terms of the more primitive relevance relation of supplementa- tion (when they are not functioning simply as favourers themselves). This account generates a simpler and more elegant argument in support of moral particularism. Résumé: Certains des plus forts arguments de Jonathan Dancy à l'appui du particularisme moral dé- pendent de façon cruciale sur la dis- tinction qu'il établit entre quatre dif- férents types de relations de perti- nence—les relations qui favorisent, permettent, intensifient et atténuent. Je suis d'accord avec Dancy que différentes prémisses peuvent jouer des rôles structurels fondamentale- ment différents dans l'argumentation morale. Toutefois, je soutiens dans cet article qu’on peut analyser les relations qui permettent, intensifient et atténuent en termes de la relation de pertinence plus primitive de la supplémentation (quand elles ne fonctionnent pas simplement comme une relation qui favorise). Cette ana- lyse nous donne un argument plus simple et plus élégant à l'appui du particularisme moral. Keywords: Jonathan Dancy, hybrid arguments, argument structure, supple- mentation relations, moral particularism. 1. 
 In Ethics Without Principles, Jonathan Dancy draws a distinc- tion between four different “forms of moral relevance”— favourers, enablers, intensifiers and attenuators—that plays a central role in his argument in support of moral particularism (Dancy 2004: 52).1 In this paper, I first provide a more general 



























































 © Mark Vorobej. Informal Logic, Vol. 32, No. 3 (2012), pp. 306-312. Hybrid Arguments and Moral Relevance 307
 account of Dancy’s notion of intensifiers and attenuators, ac- cording to which intensification and attenuation are understood as specific kinds of supplementation relations. I next offer a cor- rection to Dancy’s analysis of enablers, and then argue that the enabling relation can also be analyzed in terms of the more primitive relation of supplementation. This reduces Dancy’s taxonomy of four fundamentally different kinds of relevance relations to just two (i.e., the favouring relation and the supple- mentation relation). Finally, I challenge Dancy’s claim that there are at least two fundamentally different kinds of enablers. The net result of this discussion, then, is a structurally simpler and more elegant account in support of moral particularism that highlights the central role that hybrid arguments—arguments that contain supplementation relations—have played in this im- portant debate. I view the following comments as being friendly amend- ments to Dancy’s account—amendments that allow us to make better sense of his genuine insights into the various roles that different premises can play in moral argumentation. At the same time, I remain neutral, in this paper, about the extent, if any, to which these insights help to build a case in support of moral par- ticularism. My concern in this paper is solely with argument structure and, more specifically, with the question of how premises— either in isolation or in combination with one another—can gen- erate reasons in support of conclusions. I argue that Dancy’s relatively complicated account can be simplified considerably by attending to the manner in which independently irrelevant premises can alter the strength of the reasons that are offered by an argument’s independently relevant premises in support of that argument’s conclusion. 2. Dancy claims that, in the following argument, (A) (1) She is in trouble and needs help. (2) I am the only other person around. So (3) I ought to help her. premise (1) is a favourer since (1), by itself, provides a reason for helping her.2 (2), however, is not a favourer since the fact 































































































































 1 All further references to this work will be cited within the body of this pa- per. 2 Strictly speaking, Dancy denies that practical reasoning has an inferential structure (p. 105). And, strictly speaking, favouring is a relation between a reason and an action; rather than a reason and a proposition (pp. 29, 38). At Mark Vorobej 308
 that I am the only other person around is, by itself, no reason to help her. That is, (2), on its own, is irrelevant to (3). But (2) is not altogether irrelevant within (A), since together (1) and (2) provide a better reason for helping her than does (1) alone. So (2) is relevant to (3) in a different kind of way. (2), Dancy says, is an intensifier. It provides a reason that strengthens the reason for (3) that is already provided by (1) (pp. 41-42).3 Dancy’s claim that (A) includes an intensifier can be gen- eralized as follows. Let’s say that a premise P supplements a premise Q, within an argument A, iff (i) P on its own is irrele- vant to A’s conclusion C, (ii) Q on its own is relevant to C, and (iii) P and Q together provide a stronger reason in support of C than Q alone provides. And that A is a hybrid argument iff A contains at least one supplementation relation. Then we can say that (2) supplements (1) within the hybrid argument (A). This terminology renders precise Dancy’s specific claim that (2) is an intensifier within (A). It’s also easy to generalize this account further in a way that allows for sets of premises to supplement other sets of premises within hybrid arguments of increasing complexity.4 An attenuator, Dancy argues, is the “opposite” of an inten- sifier. In argument (B), for example, (B) (1) She is in trouble and needs help. (2) It’s all her own fault. So (3) I ought to help her. premise (2), while again irrelevant on its own to (3), weakens rather than strengthens the reason that the favourer (1) on its own provides in support of (3) (p.42). Let’s say that a premise P negatively supplements a prem- ise Q, within an argument A, iff clauses (i) and (ii) are as before (in our definition of the supplementation relation), and (iii) P 































































































































 the same time, “for ease of reference” (p. 38), Dancy doesn’t object to dis- cussing these issues within a familiar argumentative premise-conclusion framework. A “favourer,” I believe, is what argumentation theorists stan- dardly refer to as a convergent premise. 3 My goal in this paper is not to challenge Dancy’s interpretation of how rea- sons operate within any particular passage, but to make better sense of his overall approach. Some have argued, for example, that there is really only one (complex) reason operating in those cases where Dancy sees a reason operating in conjunction with something else that is not a reason of the same kind. Others have argued that there are really two (or more) reasons of the same kind, that are operating independently of each other, in some of the scenarios that Dancy considers. Brad Hooker, for example, argues in both of these ways in (Hooker 2008: 18-23). I will assume, however, that Dancy’s claims about how we should individuate reasons, within the particular pas- sages under consideration, are correct. 4 For further discussion of hybrid arguments, see chapter six of (Vorobej 2006).
 Hybrid Arguments and Moral Relevance 309
 and Q together provide a weaker reason in support of C than Q alone provides. And that A is a negative hybrid iff A contains at least one negative supplementation relation. Then we can say that (2) negatively supplements (1) within the negative hybrid (B).5 This again captures what Dancy means in claiming that (2) is an attenuator within (B).6 Intensifiers and attenuators, then, play parallel roles within different kinds of supplementation re- lations. So we can think of intensifiers and attenuators as two specific instances of the more general phenomenon of supple- mentation. In general, then, a hybrid argument contains an independ- ently irrelevant premise (i.e., a premise that on its own is irrele- vant to the argument’s conclusion C) that alters the strength of the support that some independently relevant premise (i.e., some premise on its own) provides to C.7 An intensifier is an inde- pendently irrelevant premise that strengthens an argument’s overall support for C. An intensifier, that is, supplements some other independently relevant premise. In (A), for example, (1) and (2) together provide more support for (3) than (1) provides on its own. So (2) is an intensifier. An attenuator is an independently irrelevant premise that weakens an argument’s overall support for its conclusion. An attenuator, that is, negatively supplements some other independ- ently relevant premise. In (B), for example, (1) and (2) provide less support for (3) than (1) provides on its own. So (2) is an at- tenuator. 3. Dancy is much more interested in the distinction between fa- vourers and enablers, however.8 Dancy claims that, in the fol- lowing argument, 



























































 5 I discuss negative hybrids on p. 306 of (Vorobej 2006). 6 On my reading of (B), (2) is neither positively relevant nor negatively rele- vant to (3). So it might be better to read (2) as follows: Whatever trouble she may be in is her own fault. This more clearly gives you no reason to help her —since on this reading (2) doesn’t say that she is in trouble—and no reason not to help her. 7 Here I am restricting my focus to cases in which a single premise supple- ments some other single premise. 8 Amongst other things, Dancy uses the “main distinction” (p.73) between favourers and enablers to reveal an error in W.D. Ross’s “general metaphysi- cal picture” of duty (p. 45), to explain the difference between resultance and supervenience (p. 87), and to explain why (full) atomism is false (p. 94). Mark Vorobej 310
 (C) (1) I promised to do it. (2) My promise was not given under duress. (3) I am able to do it. (4) There is no greater reason not to do it. So (5) I ought to do it. premise (1) is a favourer since the fact that I promised to do something is a reason to do it. (3), however, is not a favourer since the mere fact that I am able to do something is not, by it- self, a reason to do that thing. Rather, (3) is an enabler because “in the absence of (3) … (1) would give me no reason to act” (p. 40, my italics). Read literally, however, this last claim is incorrect. If we simply removed (3) from (C), then (1) would continue to pro- vide a reason in favour of performing the act in question. There- fore, what Dancy must mean is that if (3) was false, then (1) would provide no reason in support of (5). And that’s simply because one can’t have any reason to perform an act that one is not able to perform. If this is what it means for (3) to be an enabler, then en- ablers are not logically primitive reasons since the role of (3) within (C) can be captured by first of all noting that the negation of (3), on its own, is negatively relevant to (5) in the following sense: if (3) is false (i.e., if the negation of (3) is true) then (5) must be false as well.9 And if this is right then, arguably, (3) supplements (1) within (C). (3) on its own is irrelevant to (5). But (3) strengthens the case that (1) offers in support of (5)— since the truth of (3) eliminates one of the possible ways in which the argumentative support for (5), that is provided by (1), can be undermined or defeated. Dancy claims that (2) is also an enabler within (C). “If my promise had been given under duress,” Dancy writes, “I would have no reason to keep it” (p.39). This seems right, and again this amounts to the claim that the negation of (2), on its own, is negatively relevant to (5) in the same sense; namely, if (2) is false then (5) must be false as well. And so if (2) is false then, again, (1) would provide no reason in support of (5). And that’s simply because one can’t have any reason to fulfill a promise that was made under duress. However, it’s not so clear in this case how (2) itself func- tions within (C), and Dancy doesn’t address this issue. There seem to be two plausible options. First, since (2), unlike (3), ac- tually refers to the making of a promise, (2), like (1), is a fa- vourer within (C). (That is, the fact that I’ve freely made a 



























































 9 Notice that clause (ii) in the definition of our two supplementation relations refers only to positive relevance. That is, Q on its own supports the claim that the argument’s conclusion C is true. But a proposition Q is negatively rele- vant to C iff the truth of Q supports the claim that C is false. Hybrid Arguments and Moral Relevance 311
 promise gives me some reason to keep that promise.) Or second, (2) supplements (1) within (C). (That is, the mere fact that a promise was not made under duress does not provide any addi- tional reason, above and beyond the reason provided by the act of promising itself, to keep that promise. However, it does strengthen the reason that arises out of the act of promising, since it eliminates one of the possible ways in which your duty to keep the promise can be defeated.) For our purposes, how- ever, it doesn’t matter which is the correct or more plausible reading, since in neither case does (2) embody any logically primitive reason of a sort that we have not encountered previ- ously. Yes, the negation of (2) is negatively relevant to (5). But this establishes only that either (2) supplements (1), or (2) is it- self a favourer. 4. Finally, Dancy argues that (4) is also an enabler within (C), but an enabler “of a rather different style” from (2) and (3) (p. 92). This seems to suggest that (4) performs an altogether different kind of enabling function within (C).10 I will argue, however, that (4) does not provide us with any reason to posit any addi- tional distinctive kind of primitive enabling relation. Dancy argues that (4) does not enable (1) to favour (5) since “(1) would have favoured (5) … even if something else had more strongly favoured not doing (5); that one promised can be some reason to act even if there is greater reason not to” (p.40). Arguably, this reasoning is correct. The negation of (4) is negatively relevant to (5) in the following sense: if (4) is false then (5) must be false as well. That is, if there is a greater reason not to do something, then it can’t be true that I ought to do it. Nonetheless, the negation of (4) weakens but does not destroy (1) altogether as a reason. (1) can still provide some support for (5) even though we know, on independent grounds, that (5) can- not be true. As Dancy says, a promise can still favour an action that, on balance, should not be performed. So, if (4) is false, then (1) remains a favourer but is no longer strong enough “to ground an actual duty” (p. 92). However, what does this say about the role of (4) itself within (C)? Again, this is complicated by the fact that Dancy doesn’t address the question of whether (4), on its own, is rele- vant to (5). Again, there are two options. First, suppose that (4) is relevant, on its own, to (5). Then (4), like (1), is a favourer. Or 



























































 10 After describing (4) as an enabler, Dancy struggles inconclusively, on (p. 93), with how to make sense of this claim. Mark Vorobej 312
 second, suppose that (4) is irrelevant, on its own, to (5). Then, arguably, (4) supplements (1) since it eliminates one of the pos- sible ways in which the argumentative support for (5), that is provided by (1), can be weakened. Dancy focusses, then, on the following asymmetry. (1) survives as a (weakened) reason in the presence of the negation of (4); whereas (1) does not survive as any kind of reason at all in the presence of either the negation of (2) or the negation of (3). Now, even if we assume that Dancy is right about this, it doesn’t follow that any of (2), (3) or (4) themselves perform any kind of distinctive “enabling” function within argument (C). On the contrary, (2), (3) and (4) are most plausibly viewed as either favourers, just like (1), or independently irrelevant premises that supplement (1). In conclusion, the relevance relations that obtain within arguments (A), (B) and (C) can be captured adequately in terms of the two primitive relations of favouring and supplementa- tion.11 References Dancy, J. (2004). Ethics Without Principles. Oxford: Oxford University Press. Hooker, B. (2008). “Moral Particularism and the Real World.” In M. Lance, M. Potrc & V. Strahovnik (Eds.), Challenging Moral Particularism, pp. 12-30, Ch. 2. New York: Rout- ledge. Vorobej, M. (2006). A Theory of Argument. Cambridge: Cam- bridge University Press. 



























































 11 I thank an anonymous referee for helpful comments on an earlier draft of this paper.