a very much workshop draft of Irreducibly Normative Properties by Chris Heathwood [email protected] Those who maintain that normative or evaluative properties cannot be reduced to, or explained in terms of, natural properties have difficulty explaining what these properties are. Stephen Finlay in the Philosophy Compass characterizes the problem as follows: On the nonnaturalists’ view, … reality has ‘brute, inexplicable’ normativity, which cannot be explained in motivational or other natural terms. This inexplicability is twofold: we cannot explain what normativity is in nonnormative language, and neither can we explain why the fundamental normative truths hold (e.g., why the fact that pain hurts counts in favor of preventing it). (Finlay 2007: 24)

I have argued elsewhere that no meta-ethical or meta-normative theory, whether constructivist, reductionist, or both, can explain why the fundamental normative truths hold (Heathwood 2012). In this paper, I attempt to address the other “inexplicability problem” for normative non-naturalism: that of explaining what normativity is in non-normative terms. I don’t claim to be giving a complete characterization of normativity in such terms, an identification of normativity with some natural phenomenon. To do that would be to abandon non-naturalism. Instead, I identify an interesting and informative fact about normative properties that goes some way towards elucidating them in nonnormative terms. It is this: that to attribute a normative or evaluative property to something is, in virtue of the nature of this property, necessarily to commend or condemn that thing. This view characterizes normativity in terms of the natural phenomenon of performing certain familiar speech acts. It explains how a property can be at once descriptive (as, in some trivial sense, all genuine properties must be) and evaluative: in describing things by attributing these properties to them, we can’t help but also be making an evaluation. In merely listing some of the features of reality, we can’t help but get ourselves involved in the business of commending, recommending, condemning, and so forth. In what follows, I further explain the initial problem (§1), I clarify the proposed solution (§2), I describe further explanatory work that the hypothesis can do, thus giving it some independent support (§3), and I defend it against some objections (§4). I mean my topic to cover both normative properties (such

1

as are instantiated when some act ought to be done) and evaluative properties (such as the property of being intrinsically good), including both their “positive” and “negative” kinds. But for simplicity, I sometimes omit reference to all of the relevant categories. It is not my aim here to be giving any arguments for the existence (and instantiation) of irreducibly normative or evaluative properties.1 It is rather to be offering an account of what such properties would be. It is a motivated by an objection to non-naturalism, though, as we will see, critics of non-naturalism can accept the account too, even as part of an argument against non-naturalism. 1. A problem for normative or evaluative non-reductionism: What is normativity or value? Non-reductionists about normativity or value maintain that normative or evaluative properties or facts – facts such as that some action ought to be done, that some state of affairs is intrinsically good, or that someone has a reason to do something – are sui generis.2 They are not identical to any kind of fact that can be expressed in non-normative or non-evaluative terms. They are like dualists about the mind, who maintain that mental facts are not identical to any fact that can be expressed in non-mental terms. Like dualists about the mind, normative or evaluative non-reductionists have difficulty elucidating the nature of their one-of-a-kind properties. Metaethical reductionists, by definition, say exactly what they take normativity or value to be, whether it’s pleasant experiences, being motivated after deliberation, a homeostatic cluster of things that satisfy human needs, or whatever gives to a spectator the pleasing sentiment of approbation. Reductionists thus have an answer to the question of the nature of value or normativity. Many non-reductionists, by contrast, are resigned to their inability to answer, and to accepting the mystery. G.E. Moore (1903: §6) writes,

For such arguments, see Sidgwick 1907, Moore 1903, Shafer-Landau 2003, Huemer 2005, Parfit 2011, and others. 2 I take non-reductionism about these properties to be the view that they are not identical to any properties that can be expressed in non-normative, non-evaluative terms (and are instantiated). Non-naturalism is the view that they are not natural properties (and are instantiated). These are different views, though the differences between them don’t matter much here, and I sometimes slide between them. 1

2

If I am asked ‘What is good?’ my answer is that good is good, and that is the end of the matter. Or if I am asked ‘How is good to be defined?’ my answer is that it cannot be defined, and that is all I have to say about it.”3

Similarly, Derek Parfit (2006: 330) writes, If words like ‘reason’ and ‘ought’ neither refer to natural features, nor express our attitudes, what could they possibly mean? Non-reductive realists, as I have conceded, do not give helpful answers to these questions.4

Many non-reductionists are content to claim that, when it comes to this irreducible normativity, you simply get it or you don’t, and we cannot say much to explain it. But this plays into the hands of critics of non-reductive realism, who find such kinds of fact too mysterious to believe in. It would thus be an advantage for non-reductionists if they could say something to help us understand these facts. Note that the problem for normative non-reductionists here is more acute than the corresponding problem for dualists about the mind. Although dualists about the mind may not be able to say much to explain the nature of phenomenal consciousness, they can at least point to the fact that we are all intimately acquainted with it anyway, through introspection, an empirical way of knowing. They can also point out (if they are not also epiphenomenalists) that mental facts figure in causal explanations of non-mental phenomena. Non-naturalists about normativity, on the other hand, hold that normative facts are neither empirically observable nor casually efficacious. Skepticism about them is thus a more reasonable worry.5 2. A solution: irreducibly normative properties as essentially commendatory properties We use words to describe reality, but we do many other things with them as well. By uttering certain words in the right context, we can thank someone, make an offer, or condemn an act. Speech acts are a familiar, natural phenomenon. Typically, and perhaps even in all other cases, which kind of speech act (beyond assertion) a person performs with which (declarative) utterances is a contingent, conventional matter. But what is interesting and By ‘What is good?’, Moore surely means, What is goodness? He of course has substantive, informative answers to the question, What things are good? 4 See also … 5 Non-reductive naturalists in metaethics – those who hold that normative properties are not identical to any other kind of property but are nonetheless causally efficacious and thus empirically discoverable – may be thought to occupy a position similar to (non-epiphenomenal) dualism about the mind: they may not be able to say much as to what irreducible normativity is, but they may try to support belief in it through such empirical means. 3

3

distinctive about normative or evaluative properties, I claim, is that if a person attributes one to something, she can’t help but also be commending or condemning the thing. Normative and evaluative properties, if irreducible, have this special feature: if someone says sincerely that something in the world has one of these properties, he, of necessity, due to the nature of these properties and not any contingent conventions, involves himself in more than mere assertion, more than mere description of the world. The nature of the property is such that it makes him either commend or condemn, praise or criticize, speak positively or negatively, speak for or against. We can describe this as the view of normative properties as essentially commendatory properties. This hypothesis, if true, should go some way towards assuaging critics of normative non-reductionism who are mystified as to what these irreducibly normative properties are supposed to be. We are all familiar with commending and condemning; we all do it, no matter our metaethical predilections. These irreducibly normative properties are interesting, according to our hypothesis, because they are inherently such that they make us do it, whether we want to or not, whenever we merely attribute one of these properties to something. That is something substantive and interesting about their nature; it distinguishes them from ordinary properties; and it distinguishes them from other properties whose nature and existence is contested in philosophy, such as modal, mental, or mathematical properties. When I say that these properties make us commend or condemn, I don’t mean ‘make’ in a causal sense, as when a parent, concerned with politeness, makes his child commend a friend. The relationship is rather a constitutive one. In attributing a normative property, we thereby commend. The view of normative properties as necessarily commending properties should not be confused with any form of motivational judgment internalism, the view that normative judgment necessarily involves motivational pro-attitudes of some kind (at least for some class of judgers). It is no part of the view here that when we assert that some act ought to be done, and thereby, according to the hypothesis, commend it, we necessarily have some positive conative or desirelike attitude towards the act. A person can commend something even when he has no positive conative attitudes or feelings towards the thing, just as a person can thank someone or apologize to someone even when they don’t feel thankful or apologetic.6 Some clarification here. We can distinguish differing grades of such speech act. Take apology. There is insincere apology, as when someone is being sarcastic. No genuine apology occurs in insincere apology. Among sincere apologies, we can distinguish high-grade apologies, in which the apologizer feels genuine remorse, from a lower-grade variety, in which there are no such feelings, but a genuine apology still occurs. My view is that a person who sincerely (e.g., not sarcastically) attributes an evaluative property to something thereby commends it. If he lacks 6

4

The idea that making a normative judgment inherently involves an act of commending or condemning is often associated with non-cognitivism.7 But because the view denies that this is all that is involved, and holds that we are also attributing an objective property to something, and so asserting a proposition, problems that vex pure non-cognitivist theories are avoided. Consider, for example, the embedding problem. Pure non-cognitivists hold that expressions like ‘is good’ are used to express certain positive conative attitudes towards things, or perhaps simply to commend these things, and that this exhausts their meaning. Familiar problems arise when they try to explain the meaning of expressions in which evaluative terms are embedded, such as negations, questions, or conditionals. It is difficult to devise a theory that identifies which attitudes utterances of these kinds express in a way that preserves the right logical relations between them and the attitude expressed by an utterance of the form ‘x is good’. The hypothesis of normative properties as essentially commendatory properties faces no such problems. All it claims is that when we attribute goodness to something, we can’t help but be commending it. It isn’t committed to any claim about what we are doing when we, say, ask whether something is good (beyond the idea that we are asking whether it has an essentially commendatory property). Moreover, because, unlike on noncognitivism, there is, on this view, a genuine property of goodness, we can say that when we ask whether something is good, we are simply asking whether the thing has a certain property. Which property? The property that is such that, whenever we attribute it to something, we are necessarily commending it. Pure non-cognitivist theories may be less popular these days than hybrids of these and cognitivist theories, and the view defended here is in some ways similar to these hybrid views. A comparison thus might be useful. According to a simple version of hybridism, when we make a sincere utterance of the form, “x is good,” we do two categories of things: (i) we attribute the property of goodness to x, thus asserting the proposition that x is good and thus expressing that belief; and (ii) we express a desire-like attitude towards x. The view defended here does not include anything like (ii). First, what goes on in normative judgment beyond mere belief, on the view here, is not a desirelike attitude, but (at least when the judgment is expressed) an act of commendation or condemnation, which, as emphasized above, need involve no conative attitudes. Second, it’s part of the view here that this “something in addition to mere belief” happens (a) of necessity and (b) due to the nature of the property attributed. Some would classify as a hybrid theory a theory claiming

positive conative or emotional states towards the thing, this might signal or constitute that something less than fully ideal is going on, but genuine commendation has still occurred. 7 See, e.g., Hare 1952: ch. 5.

5

merely that the “something in addition to mere belief” (whether an attitude or what have you) occurs only typically.8 Nonetheless, perhaps my view counts as a form of the hybrid theory. But if it does, it is only, at least as far as it goes here, a hybrid theory of normative utterance rather than, as is standard, a hybrid theory of normative thought. Interestingly, this helps it avoid some of the problems that hybrid theories of normative thought inherent from their pure non-cognitivist forebears. Consider one example.9 Suppose I accept that P and that if P, then Q (where ‘P’ and ‘Q’ stand for simple normative sentences, and ‘acceptance’ is a term that covers both the belief and the conative state, if hybridism is true). Given certain other background facts, I am rationally required to accept the conclusion that Q. This acceptance, for the hybrid theorist about normative thought, involves both a belief and a conative state. Suppose next that, as rationality requires, I believe whatever one believes when one accepts that Q. But suppose I fail to have whatever conative state is involved, according to the hybridist, in accepting that Q. The hybridist needs to get the result that this is irrational, but it is an irrationality located in the conative states, not the beliefs (since I have the belief that rationality requires). It is a challenge to devise an adequate account of this conative irrationality. Moreover, whatever account the hybridist gives would seem to be one available to a pure non-cognitivist, thus potentially undercutting the advantages the hybrid theory is supposed to enjoy over pure non-cognitivist theories. This isn’t to say that hybrid theories can’t meet the challenge. But the view outlined here, despite its similarity to hybridism, doesn’t even face it. The view here simply says that in sincerely uttering that P, one is commending the subject of this proposition. If one also sincerely utters that if P, then Q, then one is, to be sure, committed to believing that Q. But one is not committed to commending – to performing an act of commendation towards – the subject of Q. This follows from the uncontroversial fact that one isn’t rationally required to utter that Q. We’re not in general required to say what we believe. And if one does happen to utter that Q, then one will, according to my view, automatically be commending the subject of Q. Thus there is no issue concerning some subject who utters that Q but fails to commend, as there is the issue for standard hybrid theories concerning the subject who has the belief but not the required conative state. Finally, hybrid theories of normative thought are sometimes motivated by a desire to inject genuine normativity into a naturalistic realist metaethic.10 But the view defended here is motivated instead by a desire to be able to explain, to some extent, what normativity might be if it is irreducible and non-natural. They See Copp 2001, including as described in Copp 2009: 171. Due to Mark Schroeder (2009: 268-71), though the presentation above is simplified. 10 See Copp 2001. 8 9

6

thus begin from quite different motivations, even if they end up in spots that are in some ways similar. 3. Further work the hypothesis can do The view of normative or evaluative properties as essentially commendatory or condemnatory not only helps non-naturalists explain the nature of these properties, it does further interesting work: (a) it provides for an account of the “queerness” of normative or evaluative properties, one superior to other accounts; (b) it helps us characterize non-reductionism in ethics in more than merely a negative way; and (c) it explains (again, better than some other accounts) why the open-question argument works, assuming, as nonreductionists typically believe, that it does work. a. Characterizing queerness My view makes for an alternative, and I believe superior, characterization of the sense in which irreducibly normative properties are unusual. J.L. Mackie famously claims normative properties to be unacceptably “queer” (1977: 38-42). One central aspect of this queerness, Mackie suggests, is their power to motivate anyone acquainted with them (1977: 40). Mackie is here suggesting that nonreductionists about value may be committed to some kind of motivational judgment internalism. But, as many point out, this isn’t so. Non-naturalists can, and many do, plausibly reject motivational judgment internalism.11 Nonetheless, irreducibly normative properties, if they exist, do seem to have some kind of power lacked by ordinary properties. In my view, it is not their ability to make us comply with the facts in which they figure, or even just to make us want to. Their power is not a causal power at all. It is instead their ability to make anyone who merely attributes one to something thereby to be engaging in a practice that goes beyond mere description – the practice of evaluating, of commending or condemning. It does this whether the attributor likes it or not. It can indeed seem rather odd that simply reporting, as disinterested scientists do, that some object in the world has some property can force one into the business of making an evaluation, but this, I am claiming, is what normative and evaluative properties do. And this is the sense in which they are queer. What is queer is not the notion that normative language involves one in making recommendations, commendations, etc. I’m not sure any metaethical view would deny that it does this. What is queer is that, when it comes to irreducibly normative properties, one can’t help but get oneself involved in the 11

E.g., Shafer-Landau (2003: ch. 6), … .

7

business of making a recommendation simply by stating a fact.12 It is that these properties are at once descriptive and evaluative: the facts involving them both describe the world and evaluate it, and thus force us to engage in acts of evaluation merely by describing things. I believe this account of normativity’s queerness confirms Mackie’s contention that these properties or facts “would be entities … of a very strange sort, utterly different from anything else in the universe” (Mackie 1977: 38). It is not our business here to investigate the question whether irreducible normativity’s queerness is enough, or even any, reason to doubt its existence. I am not here arguing for metaethical non-naturalism. We are rather addressing an objection to it, the objection that non-naturalism has no plausible account of what these non-natural irreducibly normative properties are supposed to be. As this section illustrates, it is an account that even those who reject non-naturalism can accept, and indeed may even welcome, as it may help underwrite an argument against non-naturalism (the argument from queerness). Mackie’s argument from queerness (or the part of it we are focusing on here) seems to many to fail, because it falsely encumbers non-naturalism with something like motivational judgment internalism. If the account of queerness provided by my view is correct, however, opponents of non-naturalism have an argument from queerness that at least avoids this defect. Our hypothesis may thus be something of a double-edged sword for non-naturalists, helping them to shed light on the nature of their mysterious properties, though in doing so, replacing this mystery with acquaintance of something we ought to be suspicious of. I don’t believe this is any reason to think that the hypothesis is mistaken, though it does admittedly pull against its original motivations. b. Characterizing non-reductionism The hypothesis of evaluative properties as essentially commendatory properties allows non-reductionists in ethics to characterize their view positively. Non-reductionism is typically characterized only negatively, as the view that normative properties fail to be identical to any properties that can be expressed in non-normative terms. But if the hypothesis here is right, we can add to this the further thesis that normative properties are essentially commendatory. Nonreductionists can tell us not just what evaluative properties aren’t, but what they are. Although I have been focusing on non-naturalist non-reductionism, the problem of explaining irreducible normativity confronts non-reductive naturalism as well. Non-reductive naturalists can, if they wish, adopt our Note that the ‘irreducible’ here is important. If reductionism is true, one apparently can attribute normative properties without evaluating, since on this view normative properties are just natural properties, which one can always attribute without evaluating. 12

8

hypothesis, too. The hypothesis provides a way to characterize the broad genus of non-reductive theories in a positive way. c. Explaining why the open-question argument works The theory of normative and evaluative properties as essentially commendatory or condemnatory can also explain why the open-question argument, the most important argument for irreducible normativity or value, succeeds (if it does). According to one way of putting the argument, no evaluative property is identical to any property that can be expressed in nonevaluative terms because no evaluative predicate means the same as any nonevaluative predicate. And no evaluative predicate means the same as any nonevaluative predicate because, according at least to all appearances, it is not conceptually confused or self-contradictory to claim, of something to which the relevant non-evaluative predicate applies, that the relevant evaluative predicate doesn’t apply. Both steps of this argument merit, and have received, extensive examination. That is not our aim here. Our aim is rather to ask, if this argument succeeds, why does it? Why will any attempt to identify the evaluative with the non-evaluative founder in this way? According to Darwall, Gibbard, and Railton (1992: 117), the open-question argument is compelling because Attributions of goodness appear to have a conceptual link with the guidance of action … . Our confidence that the openness of the open question does not depend upon any error or oversight may stem from our seeming ability to imagine, for any naturalistic property R, clear-headed beings who would fail to find appropriate reason or motive to action in the mere fact that R obtains (or is seen to be in the offing). Given this imaginative possibility, it has not been logically secured that R is action-guiding (even if, as a matter of fact, we all do find R psychologically compelling). And this absence of a logical or conceptual link to action shows us exactly where there is room to ask, intelligibly, whether R really is good.13

Darwall, Gibbard, and Railton want to explain the power of the open-question argument essentially by appeal to motivational judgment internalism. Their view is that any judgment involving a normative or evaluative concept entails a motive to action on the part of the judger.14 But no naturalistic concept is like this. For any naturalistic concept, there are possible beings who judge that the concept applies to something, but feel no motive to pursue it, or respond in any other way towards it.15 I have corrected what I believe is a typo in the original text, changing a ‘P’ to an ‘R’. Or perhaps this needs to be qualified to apply to communities. … 15 The exegesis here is a little complicated, since Darwall, Gibbard, and Railton write of beings who would fail to find appropriate reason or motive to action in the mere fact that R obtains. Perhaps ‘motive’ is put in to clarify that by ‘reason’, they mean motivating reason. I suspect that this is indeed their aim (Miller (2003: 20-4) agrees). But perhaps by ‘reason’, Darwall, Gibbard, and Railton mean normative reason. If so, then … . 13 14

9

But motivational judgment internalism is rather uncertain – and definitely less certain, in my view, than the reality of the open-question phenomenon it is here meant to explain. (It may also be worth noting that motivational judgment internalism is probably unfriendly to non-naturalism, and so if this is our explanation of the open-question phenomenon, it may move us instead to noncognitivism [or even a form of naturalism compatible with motivational judgment internalism, such as some form of subjectivism].) I believe that the open-question phenomenon is better explained by the hypothesis of evaluative properties as necessarily commendatory. Attributions of goodness have a link not with the guidance of action but with the making of commendations or criticisms. When we attribute goodness to something, we can’t help but be commending it. Not so for any natural property. Even if we attribute some non-evaluative property that is evaluatively relevant – e.g., is a natural property in virtue of which something is intrinsically good – to make the attribution of the non-evaluative property is not necessarily to involve oneself in an act of commending. Since no natural property is essentially commendatory, no normative property can be identified with any natural property. Attributing a non-evaluative property to something may of course contingently involve oneself in an act of commending. If it is common ground between us that pleasure is good, then, in describing a hike as pleasant, I might thereby be commending it. But this utterance’s being a commendation is contingent upon the background evaluative claim. Only that claim is essentially commendatory. My account of the open-question phenomenon is at home with R.M. Hare’s formulation of the open-question argument, which he takes to be an improvement upon Moore’s original version: our attack upon naturalistic definitions of ‘good’ was based upon the fact that if it were true that ‘a good A’ meant the same as ‘an A which is C’ [where ‘C’ is some non-evaluative term], then it would be impossible to use the sentence ‘An A which is C is good’ in order to commend A’s which are C; for this sentence would be analytic and equivalent to ‘An A which is C is C’. Now it seems clear that we do use sentences of the form ‘An A which is C is good’ in order to commend A’s which are C … . (Hare 1952: 90-1)

Hare, of course, drew a different conclusion than Moore. That is not our concern here. Here, we began with the objection to non-naturalism that it cannot helpfully explain what normative or evaluative properties are. I proposed an explanation: they are essentially commendatory properties. I then endeavored to catalog other phenomena that this hypothesis, if true, would explain. One of these is the open-question phenomenon.

10

4. Some objections a. An uninformative tautology One objection to the theory of normative properties as essentially commendatory is that it sheds no light on normativity, because it is covertly tautologous. It is covertly tautologous because the best account of what it is to commend something is that it is to attribute a positive normative property to it. Thus, my view would ultimately be saying no more than that the normative properties are those such that when you attribute one to something, you can’t help but be attributing a normative property to it. But the account of commending on which this objection relies is doubtful. For we often commend without attributing normative or evaluative properties to things. I might commend some product simply by pointing out that it is inexpensive, but inexpensiveness is a mere descriptive property. We can even commend without attributing a property at all, as when we give something a “thumbs up,” or someone a “high five.” Finally, surely a committed normative nihilist can engage in the business of commending. Upon receiving a request for advice, she might say this: “As a nihilist, I don’t believe anything has any normative properties; thus, I don’t think there is anything you ought to do. Nonetheless, I recommend that you make a donation of $20.” This nihilist is commending a certain course of action, but she appears to have succeeded in refraining from attributing any normative properties to it. One might wonder why the nihilist is commending the action. She might reply that she simply finds herself drawn to it, or some feature of it, in a way that she regards as ultimately non-rational.16 b. Commending the bad A second objection is similar to an objection to both non-cognitivism and motivational judgment internalism. Imagine a cadre of devils interested in discovering what would we be bad precisely to bring that about. One devil says to another, “I recommend that you take that plan of action, since it would be very bad indeed.” My theory commits me to saying that, in attributing badness to the [What if I’m wrong and to commend just is to attribute a positive normative property? Some would say that this wouldn’t be fatal. For some would say that one can shed light on phenomena even when this sort of circularity is involved – that some circularities are informative (see, e.g., Carroll 1999: 238 and also this passage from Sharon Ford “Possible objections include that [my account] sounds like a circular definition. This is unavoidable when talking about fundamentals since there is no way to describe them in terms more basic than themselves. But it is an acceptable, informative circularity…”). There is also the related issue of explaining why merely attributing a property can get one to commend. A tempting explanation appeals to the thesis that to commend simply is to attribute a positive normative property.] 16

11

act, the devil is condemning it. But in fact he is attributing badness to it precisely to commend it. One response to this kind of case claims that the devil isn’t really judging the act to be bad, but is instead using an “inverted commas” sense of ‘bad’ (Hare 1952; Smith 1994). According to this idea, the devil doesn’t really think that the act in question would be bad – he’s recommending it after all! – and what he is really saying is something like this: “I recommend that you take that plan of action, since it would be what most people call ‘very bad’.” Since such a remark does not evidently involve the devil in attributing actual badness to anything, if this is what his original remark really means, it would be no counterexample to my hypothesis. The “inverted commas reply” is an interesting strategy for both noncognitivists and motivational judgment internalists, who need it to be that the devil is not literally judging that the act is bad, given that he has none of the negative attitudes or motivations these theories may require that he have. But this reply is a non-starter for normative realists who want to accept the view of normative properties as essentially commendatory. Non-cognitivists don’t believe in normative properties.17 Their theory is most fundamentally a theory of normative thought. But my account is for those who do believe in normative properties. Given that these normative or evaluative properties are real, they are there for the devil to attribute to things. Let us suppose that the course of action that the devil is recommending really does have the evaluative property of being bad. The devil, we can suppose, can come to know this as well as I can, and can thus correctly attribute it to the act in question, and, given his interests, can mean thereby to be recommending it. But I have to say that he is condemning it. Fortunately, two other replies to the devil objection are more promising. According to the more straightforward reply, the devil’s remark involves him in a conflict of speech acts. Any assertion, whether about phenomena naturalistic or normative, can be used to perform just about any kind of speech act, given the right conventions and context. My view is thus compatible with the idea that, given their conventions and interests, these devils can attribute negative evaluative properties to things in order to commend these things. What my view is committed to is that, despite their efforts, the devils can’t help but also be condemning these bad things by attributing badness to them. Hence, the conflict of speech acts. The original objection assumed that, from the fact that the devil is clearly commending the action, we can conclude that he is not also condemning it. But this assumption is hasty. If one is tempted to make it, that may be because one is assuming that in order for a person to qualify as performing certain speech acts, 17

At least not straightforwardly. But I have to ignore this complication here.

12

he needs to have certain attitudes. In particular, it might be thought that in order for the devil to be condemning the act in question, he must disapprove of it, which, I agree, he does not. But that is a mistaken view of these speech acts. As noted earlier, one need not have any negative attitudes towards some act in order to be condemning it with some speech act. A person might, in rote fashion, judge some act to be wrong without being able to muster any negative conative attitudes against it. He is still condemning the act. Likewise, it is not implausible to hold that the devil is doing the same when he judges that the act in question would be bad. A second reply to the devil objection, even if unnecessary, is worth investigating. According to this reply, features of the devil’s utterance and context serve to cancel the condemnation that would otherwise be present by the attribution of badness to the action. Such pragmatic cancellation is common, but there may be reasons to doubt whether my theory can allow it for evaluative attributions. For the theory claims that the speech act involved is not conventional, but connected to the essential nature of the property attributed. Two options present themselves: arguing that cancelability is nonetheless allowed by the theory, or explicitly revising the theory to allow for it. A revised theory would admittedly be less elegant and perhaps have an ad hoc flavor. It would look like this: normative properties are those such that, when someone attributes one to something, she necessarily involves herself in commending the thing, so long as no features of context cancel or override this commendation. Reminded of the notions of prima facie rightness and of epistemic defeasibility, one might say that these properties have prima facie or defeasible commending force necessarily built into them. Though less elegant, this theory avoids the devil objection without the commitment of the previous reply. If it could be shown that other speech acts behave in this defeasible fashion, it would enjoy some independent support. c. Unknowing attributions Suppose your favorite property is, appropriately enough, intrinsic goodness, although I don’t know this. You tell me that a certain thing, g, has your favorite property. I report to a third party that g has your favorite property, though, again, I don’t know what this property is. In reporting this to the third party, have I attributed intrinsic goodness to g? It would seem so. Am I thereby commending g? Not obviously so. The three main possible replies to this objection are the following: (i) that in reporting to the third party that g has your favorite property, I am not in fact attributing intrinsic goodness to g, despite the fact that intrinsic goodness is your favorite property; (ii) that in reporting that g has your favorite property, I am in fact commending g, although I could have no idea that I am; and (iii) the

13

counterexample succeeds, but points not to rejecting the theory wholesale but modifying it in light of the example. Option (i) strikes me as the least plausible. ‘Attributes property x’ seems to be an extensional rather than an intensional context. Option (ii) does not strike me as implausible. Indeed, we can imagine other examples of unknowing commendation. A person who is just learning English and who repeats sincerely to an audience, “Martin Luther King was a great man,” not knowing what he is saying, is still probably commending Martin Luther King. If this speaker can unknowingly commend with this remark, why can’t I unknowingly commend when I say that g has your favorite property?18 If option (iii) were taken, the modified theory might say roughly this: that normative properties are like this: when you knowingly attribute one to something, you necessarily commend or condemn it. But as with the final reply in the previous subsection, I’m not sure I’m convinced that a modification from the original, simpler hypothesis is necessary. ***** I believe that seeing normative and evaluative properties as essentially commendatory and condemnatory properties is a worthwhile avenue of research for those tempted to believe that there may be irreducibly normative and evaluative properties.19

[Or is the following (similar) example better? I walk by some room and hear someone say, “Bob is gressible.” The person speaking is knowledgeable and trustworthy, so I come to believe (in some sense) that Bob is gressible, though I don’t know what ‘gressible’ means. Later on, I tell someone, “Hey, Bob is gressible.” It would seem that I just asserted that Bob is gressible. Now imagine a speaker even less competent than me. He walks by a room and hears someone say, “Bob is good.” He comes to believe that Bob is good, though doesn’t know what ‘good’ means. Then later he tells someone, “Hey, Bob is good.” He seems to have performed a commendation, albeit unknowingly, just as I earlier performed an assertion. We should thus accept unknowing commendations anyway.] 19 Thanks, so far, to Michael Rubin, Christian Coons, David Barnett … . 18

14

References Carroll, Noël 1999. Philosophy of Art: A Contemporary Introduction. London: Routledge. Copp, David 2001. Realist-Expressivism: A Neglected Option for Moral Realism. Social Philosophy and Policy 18: 1-43. Copp, David 2009. Realist-Expressivism and Conventional Implicature. Oxford Studies in Metaethics 4: 167-202. Darwall, Stephen, Allan Gibbard, and Peter Railton 1992. Toward Fin de siècle Ethics: Some Trends. The Philosophical Review 101: 115-189. Finlay, Stephen 2007. Four Faces of Moral Realism. Philosophy Compass 2: 820-49. Gibbard, Allan. Wise Choices, Apt Feelings. … Hare, R.M. 1952. The Language of Morals. Oxford: Oxford University Press. Heathwood, Chris 2012. Could Morality Have a Source? Journal of Ethics & Social Philosophy 6: 1-19. Huemer, Michael 2005. Ethical Intuitionism. New York: Palgrave. Mackie, J.L. 1977. Ethics: Inventing Right and Wrong. New York: Penguin. Miller, Alexander 2003. An Introduction to Contemporary Metaethics. Cambridge: Polity Press. Moore, G.E. 1903. Principia Ethica. Cambridge: Cambridge University Press. Parfit, Derek 2006. Normativity. Oxford Studies in Metaethics 1: 325-380. Parfit, Derek 2011. On What Matters. Oxford: Oxford University Press. Schroeder, Mark 2009. Hybrid Expressivism: Virtues and Vices. Ethics 119: 257309. Shafer-Landau, Russ 2003. Moral Realism. Oxford: Oxford University Press. Sidgwick, Henry 1907. The Methods of Ethics, Seventh Edition. London: MacMillan. Smith, Michael 1994. The Moral Problem. Oxford: Blackwell.

15