ISSA Proceedings 2014 – The Very Idea Of Ethical Arguments
Abstract: If non-cognitivism is true, an ethical argument cannot be a sequence of propositions as traditionally understood. I take steps towards developing an account of ethical argument that, as far as it goes, is, I believe, compatible with a particular version of non-cognitivism, namely expressivism.
Keywords: attitudinal consistency, attitudinal relevance, attitudinal validity, cogency, expressivism, non-cognitivism, propositionalism.
I begin with an account of non-cognitivism:
According to non-cognitivism, there are no moral facts or truths…. Moral judgements don’t attempt to, and don’t ever, state facts. Their purpose isn’t to describe any sort of moral reality. Instead, they serve as expressive vehicles, primarily giving vent to our emotions, prescribing courses of action, or expressing our non-cognitive commitments. As such, they aren’t the sort of things fit to be considered either true or false. (Shafer-Landau, 2003, p. 18)
[F]or… non-cognitivists, there is nothing that can make moral judgments true – no moral facts or moral reality that they could possibly correctly represent, nothing they are true of (ibid., p. 20, note 8).
Starting from the idea that there is no moral reality that agents are trying to appreciate or depict in their moral judgements, non-cognitivists have analyzed such judgements as the expression of non-cognitive states (ibid., p. 153).
This last point is worth emphasizing. Non-cognitivists don’t start with the claim that moral judgments are expressive vehicles; rather, their expressive analysis of moral judgments is their alternative to the view that the purpose of moral judgments is to describe some sort of moral reality, and is motivated by their metaphysical claim that there is no such reality.
If, as non-cognitivism holds, moral judgments (or ethical judgments – I will use these terms interchangeably) are neither true nor false, then they aren’t propositions as traditionally understood, for as traditionally understood a proposition is either true or false, and this is the view I will take here.
If ethical judgments aren’t propositions, then ethical arguments aren’t arguments in what Woods, Irvine, and Walton (2004) call “the narrow sense,” namely “sequences of propositions, one of which is the argument’s conclusion, the rest of which are the argument’s premisses” (p. 2). There are more than a few textbooks which take arguments as such to be propositional. If arguments as such are propositional, then ethical arguments are impossible if noncognitivism is true. On one view, this is a problem for textbooks that take arguments as such to be propositional; on another view, it’s a problem for non-cognitivism.
The philosopher Michael Smith takes it to be a problem for non-cognitivism. In a critique of the non-cognitivism of moral irrealists (who deny that there are belief-independent moral truths), he says that “the whole business of moral argument and moral reflection only makes sense on the assumption that moral judgments are truth-assessable” (Smith, 1993, p. 403). Now it certainly seems that moral judgments are truth-assessable, for it makes sense to say of a moral judgment such as ‘slavery is wrong’ that it is true that slavery is wrong, or that it is false that slavery is wrong. Thus, the philosopher Simon Blackburn, who is a highly sophisticated non-cognitivist, speaks of what he calls “the propositional grammar of ethics” (Blackburn, 1985, p. 6). If moral judgments don’t express propositions, then their propositional grammar is misleading. Suppose it is misleading, and that non-cognitivism is true. Are ethical arguments nevertheless possible? In this paper, I will develop an account of ethical argument that, as far as it will go, is, in my view, compatible with what I will take to be a particular version of non-cognitivism, namely expressivism.
2. Expressivism and propositionalism
According to Parfit (2011), moral expressivists hold that “[w]hen we claim that some act is wrong, we are not intending to say something true, but are expressing our disapproving attitude toward such acts” (p. 380). In more general terms, expressivism, as I will understand it, holds that the utterance of a moral judgment is the expression of an attitude.[i] (I will make this characterization more precise below.) Further, since expressivism (as I understand it) is a version of non-cognitivism, its analysis of (the utterance of) moral judgments is motivated by the (non-cognitivist) claim that there are no moral facts and no moral reality that moral judgments purport to describe.
I will follow Alex Grzankowski (2012) in taking attitudes to be intentional mental states and in taking a mental state (or other phenomenon) to be intentional “if (and only if) it is about something” (p. 4). Some, if not all, attitudes are propositional. Propositional attitudes are intentional mental states which have propositions for their objects (ibid., p. 5).
According to Grzankowski, “theorists interested in intentional states have focused almost exclusively on [propositional attitudes], some even explicitly maintaining that all intentional states are propositional attitudes” (ibid., p. 1). If attitudes are intentional mental states, and if all intentional states are propositional attitudes, then all attitudes are propositional; if this is the case, and if the (sincere) utterance of a moral judgment is the expression of an attitude, then the object of the attitude expressed is a proposition. But expressivists (qua non-cognitivists) will say (and here I make more precise my initial characterization of expressivism) that the attitude expressed in the sincere utterance of a moral judgment is neither true nor false, and so its object is not a proposition as traditionally understood. Accordingly, expressivists must either hold that not all attitudes are propositional, or grant that all attitudes are propositional but claim that the object of a propositional attitude, though it must be a proposition, need not be a proposition as traditionally understood. I won’t consider the second of these options, but I will say something about the first.
If not all attitudes are propositional, then some attitudes are intentional states whose objects are not propositions. Thus, to take an example of Grzankowski’s, if liking is a non-propositional attitude, and “if a subject likes Sally … the object of his attitude is not a proposition concerning Sally, nor does his standing in a liking relation to Sally depend upon a propositional attitude” (ibid., p. 5).
The view that, on the contrary, all attitudes are propositional, Grzankowski calls propositionalism. Propositionalists hold that “the most fundamental objects of the attitudes are propositions” (ibid., pp. 2-3). Grzankowski distinguishes two versions of propositionalism. Version A holds that “[f]or every attitudinal relation between a subject and a non-propositional object, there is a propositional attitude or attitudes (of that subject’s) in terms of which it can be analysed” (ibid., p. 7). Version B holds that “[f]or every attitudinal relation between a subject and a non-propositional object, there are propositional attitudes (of that subject’s) upon which it supervenes” (ibid.).
Grzankowski challenges both versions of propositionalism. He argues that “there are attitudes that relate individuals to non-propositional objects and do so not in virtue of relating them to propositions” (ibid., p. 1). Examples of such attitudes “include loving, liking, hating, and fearing, though there are probably many more” (ibid.). Expressivists will say that the sincere utterance by a subject of a positive (negative) moral evaluation of a non-propositional object expresses a positive (negative) attitude (of the subject’s) towards the object. For expressivists who think that not all attitudes are propositional (and I will mean all and only such expressivists when I speak of expressivists hereafter) the philosophical issue (following Grzankowski) is whether, for every such attitudinal relation between a subject and a non-propositional object, (a) there is a propositional attitude (of that subject’s) in terms of which the relation can be analyzed, or (b) there are propositional attitudes (of that subject’s) upon which the relation supervenes.
Expressivists must reject (a), for it is tantamount to analyzing away (positive and negative) non-propositional attitudes (cf. Grzankowski, 2012, p. 10).
What about (b)? It is a special case of Grzankowski’s second version of propositionalism. He explains that on this version, for S to V y, where ‘V’ is a psychological verb such as ‘like’ or ‘fear’ and “‘y’ is a non-that-clause noun phrase” (ibid., p. 6), S’s bearing “some or other propositional attitude relation to a proposition concerning y … is sufficient for his V-ing y” (ibid., p. 8). Grzankowski argues that “propositionalists cannot meet this sufficiency requirement” (ibid., p. 10). No doubt “Jim wouldn’t like Jackie if he didn’t think she existed,” but his thinking she exists obviously isn’t sufficient for his liking of her. Nor, as counterexamples will show, is his believing “that Jackie is nice,” or his liking “that Jackie is kind” (ibid., p. 11). A similar strategy is available to expressivists. Suppose that S disapproves of factory farming. Presumably she wouldn’t do so if she didn’t believe that factory farming is practised, but this belief isn’t sufficient for her disapproval. Nor would be her believing that factory farming is cruel: she might not disapprove of cruelty, or she might but nevertheless approve of factory farming all things considered. (Here and below I take ‘cruel’ to mean ‘causing pain or suffering’; cf. The New Oxford Dictionary of English, 1998.)
Suppose, however, that S does believe that factory farming is cruel, and for this reason disapproves of it. Then her non-propositional attitude of disapproving of factory farming is a consequence of her having a propositional attitude. It is also a consequence of her believing (dispositionally if not occurrently) that her belief that factory farming is cruel is a reason for her to disapprove of it. Is the latter belief sufficient for her disapproval? Not necessarily: she might believe that she has reason to disapprove of factory farming but not do so – or so an expressivist might elect to argue. But suppose that S’s believing that her belief that factory farming is cruel is a reason for her to disapprove of it is sufficient for her disapproval. Then it is possible for a subject to be “in a non-propositional attitude in virtue of being in a propositional attitude state (or states)” (Grzankowski, 2012, p. 8). Does it follow that a moral judgment the sincere utterance of which by a subject is the expression of such a non-propositional attitude (of the subject’s) is a true-or-false proposition? Expressivists can argue that this does not follow. For (i) it does not follow (expressivists can argue) that there are “moral facts or [a] moral reality that [such a moral judgment] could possibly correctly represent, [something it] could be true of” (Shafer-Landau, 2003, p. 20, note 8). (ii) Nor does it follow that in uttering such a judgment a subject would be “trying, but failing, to describe” something (ibid., p. 20). Expressivists can argue for (i) because it is their denial of there being a moral reality that motivates their interpretation of (the utterance of) a moral judgment, not their interpretation of (the utterance of) a moral judgment that motivates their denial of there being a moral reality (cf. ibid., p. 153). They can argue for (ii) because they are not error theorists: they do not hold that moral judgments have truth-values but, because there are no moral facts, are false. (cf. Brink, 1999, p. 588).
3. Towards an expressivist account of ethical argument
I will take an argument to be an ethical argument just in case it has an ethical conclusion. On a different view, an argument is an ethical argument just in case it has an ethical conclusion and at least one ethical premise. On the view I’m taking, an ethical argument may have one or more ethical premises, but this isn’t necessary for it to be an ethical argument. Consider the following argument:
Factory farming is morally reprehensible because it causes animals to suffer.
This argument apparently depends upon a claim to the effect that a human practice which causes animals to suffer is morally reprehensible. A claim to this effect may be considered to be a tacit premise of the argument, or it may be considered to be a background assumption relative to which the stated premise is positively relevant to the conclusion. The view that an argument is an ethical argument just in case it has an ethical conclusion leaves open both of these interpretations.
An ethical conclusion, or an ethical premise, is an ethical sentence. The ethical sentences with which I will be concerned will be what I will call simple ethical sentences. A simple ethical sentence, I wish to stipulate, is a sentence that has, or is analyzable as having, exactly one ethical predicate, in the grammatical sense, which it predicates of exactly one term. A sentence of this sort evaluates the extension of the term of which its ethical predicate is predicated. I will refer to the thing(s) comprising this term’s extension as the object(s) evaluated by the sentence. The sentence ‘cruel practices are wrong’ is a simple ethical sentence in my stipulated sense. It predicates the grammatical ethical predicate ‘are wrong’ of the term ‘cruel practices,’ whose extension comprises all such practices. The sentence evaluates cruel practices, and so, in my usage, such practices are the objects it evaluates. I would add that all this remains true, mutatis mutandis, if the sentence’s ethical predicate is taken to be its logical predicate, namely ‘wrong.’
Expressivists are not at liberty to take an ethical argument, as here defined, to be a sequence of propositions, but they can take an ethical argument to be a sequence of sentences, one of which is the argument’s conclusion, the rest of which are the argument’s premises and are put forward as reasons for accepting the ethical sentence that is the argument’s conclusion.
When may a person be said by an expressivist to accept an ethical sentence? Here is a possible answer. A person, S, accepts ethical sentence E, at time t, just in case at time t S holds towards the object(s) evaluated by E an attitude of the type that, on an expressivist interpretation, would (defeasibly) be taken to be expressed by an utterance of E. (‘Defeasibly,’ because, for one thing, an utterance of an ethical sentence might be insincere.) Suppose, then, that an expressivist takes this to be what it is for a person to accept an ethical sentence, and also takes an ethical argument to be one in which the premises are put forward as reasons for accepting the ethical sentence that is the argument’s conclusion. Then she may say (and I think should say) that the premises of an ethical argument are put forward as reasons for holding an attitude of the type that, on an expressivist interpretation, would (defeasibly) be taken to be expressed by an utterance of that sentence.
Consider again the argument that factory farming is morally reprehensible because it causes animals to suffer. The arguer treats the premise that factory farming causes animals to suffer as a reason for accepting the conclusion that factory farming is morally reprehensible. Expressivists can say that for the arguer to treat the premise as a reason for accepting the conclusion is for her to have a certain attitude towards the fact (as the arguer takes it to be) that factory farming causes animals to suffer: it is for the arguer to be unfavourably disposed towards this feature of factory farming. The arguer might express this attitude propositionally by saying that this feature of factory farming (namely, the fact that it causes animals to suffer) matters – it’s morally relevant; more specifically, it counts against factory farming.
This is an ethical attitude. Can expressivists say that ethical attitudes admit of justification? I believe they can, and that their best option would be to accept a reflective-equilibrium account of ethical justification – an account that accommodates the expressivist thesis that (sincere) utterances of ethical judgments express attitudes. On such an account, the test for justification will be how well a person’s ethical attitudes fit with one another and with her related non-ethical beliefs. A good fit will require consistency, and so a reflective-equilibrium expressivist will require an account of attitudinal consistency. Here is such an account. An attitude pair is consistent if there is a possible world in which both attitudes are fulfilled at the same time, and inconsistent otherwise. Thus, the attitude of favouring execution for murder is consistent with the attitude of opposing execution for manslaughter because there is a possible world in which execution is the punishment for murder but not for manslaughter. In contrast, the attitude of opposing execution for murder is inconsistent with the attitude of favouring Felix’s execution for murder because there is no possible world in which there are no executions for murder and Felix is executed for murder. There is more to be said about what a reflective-equilibrium expressivism would look like, or could look like, but I won’t say more about this here. Instead, I will apply the account of attitudinal consistency that I have just presented to the following argument.
1. All cruel practices are wrong.
2. Factory farming is a cruel practice.
3. Factory farming is wrong.
Assume that at time t S accepts 1 and therefore has a negative attitude towards all cruel practices; more specifically, let us suppose, S disapproves of such practices. S also accepts 2, and 2 is true. But S rejects 3: his attitude towards factory farming is one of non-disapproval, but not one of indifference; rather, he approves of factory farming.
On these assumptions, at time t S disapproves of all cruel practices but approves of a particular practice which he believes, correctly, is cruel. Is there a possible world in which these attitudes are both fulfilled? This depends on whether there is a possible world in which factory farming is practised but is not cruel. Suppose that it is conceptually impossible for factory farming not to be cruel; then premise 2 is necessarily true, and there is no possible world in which factory farming is practised but is not cruel. On this assumption, there is no possible world in which there are no cruel practices but there is a practice of factory farming, and so there is no possible world in which the attitude of disapproving of all cruel practices and the attitude of approving the practice of factory farming are both fulfilled. Thus, if S were to accept the premises of Argument (2) but reject the conclusion because he approved of factory farming, then, if premise 2 is necessarily true, there would be an inconsistency in his attitudes. If S accepts the premises of Argument (2), and if premise 2 is necessarily true, then S cannot, on pain of attitudinal inconsistency, reject the conclusion if he does so because he approves of factory farming.
4. The account continued
4.1 Attitudinal validity
The preceding example shows that it is possible for an expressivist to have a concept of what might be called attitudinal validity. Such a concept might be defined as follows for an ethical argument with at least one ethical premise (as well as an ethical conclusion) and with at least one true-or-false non-ethical premise and no non-ethical premise that is neither true nor false. Such an argument is attitudinally valid for S at time t if at time t S cannot, on pain of attitudinal inconsistency, both accept the argument’s premises and reject the conclusion. This condition is satisfied if and only if S’s rejection of the conclusion would be a consequence of his having an attitude inconsistent with an attitude his holding of which explains his acceptance of the (or an) ethical premise of the argument.
Let us apply this account of attitudinal validity to Argument (2). If at time t S were to reject the argument’s conclusion because he approved of factory farming, this attitude of his would be inconsistent with an attitude (disapproval of all cruel practices) his holding of which explains (on our previous assumptions) his acceptance of the argument’s ethical premise (all cruel practices are wrong). Thus, S could not, on pain of attitudinal inconsistency, both accept the argument’s premises and reject the conclusion, and so the argument is attitudinally valid for S at time t. But the attitudinal inconsistency would arise only given our assumption that the argument’s non-ethical premise (factory farming is a cruel practice) is a necessary truth, and this fact prompts the following question: for an ethical argument to be attitudinally valid for a subject at a time, must it have at least one true-or-false non-ethical premise that is necessarily true? The answer is no. Consider the following argument:
1. Execution for a conviction of murder is always wrong.
2. Felix has been executed for a conviction of murder.
3. Felix’s execution was wrong.
Assume that at time t S accepts 1: she disapproves of execution for a murder conviction. She also accepts 2, and 2 is true. Suppose that S were to reject 3 because she approves of Felix’s having been executed for his murder conviction. A world in which this attitude is fulfilled is one in which Felix has been convicted of murder and executed. A world in which the attitude of disapproving execution for a murder conviction is fulfilled is one in which there are no such executions (and never have been). Since there is no possible world in which these attitudes are co-fulfilled, they are inconsistent. Thus, S could not, on pain of attitudinal inconsistency, accept the premises of Argument (3) but reject the conclusion if her rejection of the conclusion were a consequence of her approving of Felix’s having been executed for his murder conviction. Thus, Argument (3) is attitudinally valid for S at time t. This analysis assumes the truth of premise 2, but premise 2 is not a necessary truth. Thus, for an ethical argument to be attitudinally valid for a subject at a time, it need not have at least one true-or-false non-ethical premise that is necessarily true.
In the preceding discussion, I have assumed the possibility of a person’s rejecting the conclusion of some ethical argument (with an ethical premise) because he holds an attitude inconsistent with an attitude his holding of which explains his (assumed) acceptance of the ethical premise. But is this a possibility – logically speaking? Could it be, for example, that at time t a person disapproves of all cruel practices, yet approves of a particular practice which he believes, correctly, to be cruel? Suppose it could not. Then it would not be possible for S at time t both to accept the premises of Argument (2) and also to reject the conclusion as a consequence of his having an attitude inconsistent with an attitude his holding of which explains his acceptance of the argument’s ethical premise; hence, on my proffered account of attitudinal validity, Argument (2) would be attitudinally valid for S a time t. And likewise in any such case.
4.2 Attitudinal relevance
An expressivist account of ethical argument will require an account of when the premise(s) of an ethical argument are (positively) relevant to the conclusion. Plainly, this will not be an account of (positive) propositional relevance; rather, it will be an account of what I will call (positive) attitudinal relevance. I will give such an account in a moment. First, however, recall our earlier stipulation that S accepts ethical sentence E at time t just in case S holds towards the object(s) evaluated by E an attitude of the type that, on an expressivist interpretation, would (defeasibly) be taken to be expressed by an utterance of E (e.g., an attitude of disapproval).
Now let ‘E’ be an ethical sentence and let ‘P’ be a true-or-false non-ethical sentence. If S accepts E at time t, she then has a certain attitude towards the object(s) evaluated by E. If she has this attitude because she believes P, then for her P is positively attitudinally relevant to E. An expressivist might add that if S accepts E and believes that she does so because she believes P, then she regards (her belief that) P as her reason for accepting E.
Consider, for example, the following sentences: (1) Factory farming is cruel. (2) Factory farming is wrong. For an expressivist, a sincere utterance of 2 would be the expression of a negative attitude towards factory farming. If S accepts 2 she has such an attitude, and if she believes 1 and accepts 2 because she believes 1, then for her 1 is positively attitudinally relevant to 2.
Next, consider Argument (2) once again:
1. All cruel practices are wrong.
2. Factory farming is a cruel practice.
3. Factory farming is wrong.
On the present account of attitudinal relevance, for S at time t the premises of Argument (2) are jointly positively attitudinally relevant to the conclusion if S accepts the conclusion because she accepts premise 1 and believes premise 2.
To take the account a step further, consider the following example:
1. Life imprisonment for murder is a more effective deterrent than the death penalty.
2. The death penalty has resulted in the execution of wrongly convicted persons.
3. Life imprisonment for murder is morally preferable to the death penalty.
Counter considerations to 3:
a. Life imprisonment for murder is much more costly than the death penalty.
b. The death penalty is a better fit for the crime of murder than the death penalty.
S accepts premises 1 and 2 as true (possibly after doing some research). Each inclines him to some degree to favour life imprisonment for murder more than the death penalty. Thus, for S each is positively attitudinally relevant to 3, since he would hold this attitude if he accepted 3 and did so because (or partly because) he accepted 1 and 2. S also accepts as true counterconsideration (a), and for him it is negatively attitudinally relevant to 3 because it makes him less inclined to favour life imprisonment for murder over the death penalty (and thus to accept 3) than he would be given just (his acceptance of) premises 1 and 2. S doesn’t accept counterconsideration (b) but for him it is nevertheless negatively attitudinally relevant to 3 because he believes that if he did accept (b) he would be still less inclined, and perhaps on balance disinclined, to favour life imprisonment for murder over the death penalty. Upon reflection, he accepts 3 because he accepts premises 1 and 2 as true and because for him (I shall assume) they outweigh counterconsiderations (a) and (b).
An expressivist account of ethical argument will, I shall suppose, include an account of what it is for an ethical argument to be cogent. Here I will suggest an expressivist account of cogency (just) for what I will call a Type 1 ethical argument, namely an ethical argument with at least one ethical premise and at least one true-or-false non-ethical premise and no non-ethical premise that is neither true nor false. A Type 1 ethical argument is cogent for S at time t if at time t:
(a) S is justified in accepting the argument’s ethical premise(s);
(b) S is epistemically justified in accepting as true the argument’s non-ethical premise(s);
(c) the argument is attitudinally valid for S
(d) for S, his acceptance of the premises would be sufficient, but not conclusive, reason for him to accept the conclusion.
Condition (a): S is justified in accepting the argument’s ethical premise(s) at time t if, for each such premise, he is justified by a reflective equilibrium test in holding an attitude of the type that, on an expressivist interpretation, would (defeasibly) be taken to be expressed by an utterance of the premise.
Condition (d) is satisfied at time t if and only if (i) were S to accept the conclusion at time t he would do so because he accepted the premises (in which case for him the premises would be positively attitudinally relevant to the conclusion) or because he accepted the premises and for him they were not outweighed at time t by any counterconsiderations then known to him; and (ii) the argument is not attitudinally valid for S at time t (so that for him his acceptance of the premises would not be conclusive reason to accept the conclusion).
I have said nothing about the vexed problem of how, or whether, “expressivists can make sense of sameness of meaning [of an ethical sentence] in asserted and unasserted contexts” (Shafer-Landau 2003, pp. 23-4). (An example of the latter would be the occurrence of an ethical sentence as the antecedent/consequent of a conditional sentence.) Nor have I said anything about the no less vexed problem of how, or whether, expressivists can differentiate between the attitudes expressed in ethical utterances of, for example, the following forms: ‘x is right,’ ‘x is permissible,’ ‘x is supererogatory.’ (Cf. ibid., pp. 24-25). In these and no doubt other respects, my proffered expressivist account of ethical argument is incomplete. Moreover, I do not claim that, even just as far as it goes, it is an adequate account of ethical argument. My interest, rather, is in whether, as far as it goes, it is compatible with expressivism, hence an account that expressivists are free to give, and I believe it is.
I am indebted to Geoff Goddu and David Hitchcock, whose very helpful comments on the version of this paper that I presented at the ISSA 2014 conference led me to rethink and revise the account I gave therein of attitudinal consistency and the account I gave of attitudinal validity. In light of a further comment by David Hitchcock, I also revised my account of when a person may be said by an expressivist to accept an ethical sentence.
i. I take prescriptivism to be a different version of non-cognitivism. Prescriptivism holds that moral judgments have a prescriptive meaning and a descriptive meaning, and that in virtue of their prescriptive meaning they prescribe or guide conduct. Prescriptivists can allow that prescriptions express attitudes, and expressivists can allow that attitudes can be expressed in the form of prescriptions, so there can be common ground between prescriptivists and expressivists
Blackburn, S. (1985). Errors and the phenomenology of value. In T. Honderich (Ed.), Morality and Objectivity. A Tribute to J.L. Mackie (pp. 1-20). London; Boston: Routledge & Kegan Paul.
Brink, D. (1999). Moral realism. In R. Audi (Ed.), The Cambridge Dictionary of Philosophy. Cambridge: Cambridge University Press.
Grzankowski, A. (2012). Not all attitudes are propositional. European Journal of Philosophy,1-18. Wiley Online Library: doi/10.1111/j.1468-0378.2012.00534.x.
Parfit, D. (2011). On what matters, Vol. 2. Oxford: Oxford University Press.
Shafer-Landau, R. (2003). Moral Realism: A Defence. Oxford: Clarendon Press.
Smith, M. (1993). Realism. In P. Singer (Ed.), A Companion to Ethics (pp. 399-410). Oxford: Blackwell Publishers Ltd.
Woods, J., Irvine, A., & Walton, D. (2004). Argument:critical thinking, logic, and the fallacies. Toronto: Pearson/Prentice Hall.