[Forthcoming in American Philosophical Quarterly, January 2007]
Is It Possible to Follow One’s Conscience?
G. F. Schueler, University of New Mexico
People are sometimes urged to follow their consciences. “Always let your conscience be your guide,” Jiminy Cricket advised Pinocchio.1 Such advice apparently presupposes a second order principle or claim, different from first order moral principles, and yet one that has some moral force of its own, independent of that involved in whatever one’s conscience actually tells one to do. We seem to feel that there is something wrong about ‘giving in to temptation’ and going against one’s own moral convictions, no matter what they are, as if it is morally better to act on some moral principle, even a mistaken one, than to ignore one’s moral convictions altogether. Likewise, someone who resists temptation and does what she sincerely thinks morally right, even if she is mistaken, nevertheless still gets some ‘moral credit’, perhaps even gets to some extent excused from blame for her wrong action.
This paper will examine what it is to ‘follow one’s conscience’ in this sense. More specifically it will examine what it is to follow what will be called ‘the principle of conscience’, the principle that simply asserts that it is wrong to violate one’s own moral convictions, that is, the principle that seems presupposed by Jiminy Cricket’s advice to Pinocchio . Since the principle of conscience is apparently a moral principle, it is presumably one on which we should act in appropriate circumstances. It will be argued however that there is a puzzle here. On the most obvious way of formulating this principle, it is very implausible. But more plausible ways of formulating it result in a principle that cannot be followed.
In discussions of ‘following one’s conscience’ the term ‘conscience’, since it is a noun, might seem to be intended to refer to some metaphysically implausible inner mechanism by which moral truths are discovered or communicated (a ‘still small voice’ as one sometimes hears it called, or a ‘moral compass’). But one could doubt the existence of any such mechanism while still finding plausible the associated moral principle or truth. The principle of conscience will be taken to be the claim stated above, which seems prima facie plausible, i.e. the claim that it is always wrong for anyone to violate her own moral convictions or beliefs. To do that is what ‘going against your conscience’ comes to. It is wrong to do something that one believes to be wrong. So, putting this into more transparent form and instantiating to a specific person (you) and a particular conviction (which you can pick from your own moral convictions and so will just be labeled “it is wrong to do X” here), we get
(W) If you believe it is wrong for you to do X then it is wrong for you to do X.
Here “wrong” means “morally wrong”, and of course “have it as a deeply held moral conviction” or something similar could also always substitute in (W) if “believe” seems too thin. Presumably it is safe to assume that you do have some beliefs or convictions about what it is wrong for you to do. Virtually all of us have such convictions. Of course the fact that you have such convictions or beliefs is just a fact about you, but presumably it really is a fact. So apparently the following has some true instances:
(F) You believe it is wrong for you to do X.
But, as any first semester logic student can testify, from (W) and (F) it follows that
(SMC: for “SPECTACTULAR MORAL CONCLUSION”):
It is wrong for you to do X.
So here is a very simple, one-step proof, starting from the principle of conscience as it applies to you, that obviously generalizes to show that all your convictions about what it is wrong for you to do are correct, since clearly this reasoning, if it works at all, will work for any of your beliefs about what it is wrong for you to do. As stated above the label “It is wrong for you to do X” is being used here as a stand-in for your actual moral convictions. And of course a similar ‘proof’ will work for anyone, or at least anyone who has any moral convictions. So the principle of conscience, instantiated to specific people and specific convictions, along with ordinary facts to the effect that we have such convictions, apparently allows us to construct a set of very short but obviously valid proofs establishing the truth of every one of our moral convictions about what it is wrong for us to do.
Needless to say, this is deeply implausible.2 For one thing, not only do these ‘proofs’ not depend on the content of our moral convictions; they don’t depend in any way on how we arrived at these convictions. They ‘prove’ the convictions not only of the moral philosopher who has spent her life seriously reflecting on morality, but also those of the most superficial ditz, who has never read or thought about anything more profound than comic books or video games, not to mention the racist bigot who is convinced that it is wrong for her to allow blacks to vote and the religious zealot who thinks all those who don’t accept her religion should be driven out of the country. They ‘prove’ absolutely everybody’s convictions equally. And even if your convictions are diametrically opposed to those of the person next to you (or across the picket line, or on the other side of the barbed-wire fence), no problems. All are ‘proved’ correct by this reasoning. (The conclusions ‘proved’ here don’t actually contradict each other, of course – unless the agent herself already holds contradictory moral beliefs - since each contains an essential reference to whomever makes it. The ‘you’ of these convictions is just the ‘you’ of premise (F), i.e. whoever has the conviction in question. My thought “I am six feet tall” doesn’t contradict your thought “I am not six feet tall”.)
In fact, if these ‘proofs’ are correct then it is hard to see that any form of so-called ‘moral realism’ could be true. That is because it is hard to see that there could be any objective grounds for any of your moral convictions, or those of anyone else either for that matter. For suppose there were some objective grounds for some of your moral convictions, that is suppose that there were some features of reality (other than the fact that you have these conviction) which made some of your moral convictions true. That would mean that unless you are infallible on such matters some of your moral convictions will be false. And even if you are infallible on such matters, the fact that something other than your convictions makes some of your moral convictions true entails that it is at least logically possible that some of your convictions are false. There would then still be a difference between whatever made your convictions true and the mere fact that you have them. But that means that the argument given above couldn’t possibly prove that your convictions are all true, since it depends only on the fact that you have these convictions. If it is, or can be, the case that some of your convictions are false, independently of the fact that you have them, then the argument above couldn’t actually prove them to be true. So, to put it the other way around, if these alleged ‘proofs’ are correct then there cannot possibly be any objective grounds for anyone’s moral convictions about what it is wrong for her to do.
All this is so overwhelmingly implausible that it very strongly suggests that either the principle of conscience cannot be what it was suggested above that it was (“It is always wrong for anyone to violate her own moral convictions or beliefs.”) or, more plausibly, that we cannot understand it in terms of (W). Another reason for thinking that this latter is the case is that, according to Philippa Foot at least, it is possible to hold that there is independent, objective ground for the truth of moral beliefs, while still accepting the principle of conscience. This is her view, one she also attributes to Aquinas. According to Foot, “Aquinas argues that a man acts badly if he goes against his conscience, whatever it may be that his conscience tells him to do, so that even the erring conscience ‘binds.’”3 Clearly the idea that your conscience can be in error in what it tells you entails that there is some other standard than conscience for right and wrong. So Foot’s view seems to be that you should, or even must, follow your moral convictions (since conscience ‘binds’) and, at the same time, that your moral convictions can be in error. But this just throws into sharp relief the puzzle we encountered above. What happens when in fact one’s moral convictions are mistaken? Suppose that you believe, but mistakenly or erroneously, that it is wrong for you to do Y. It looks as if Foot will have to say, since conscience ‘binds’, that you must not do Y, that is, that your doing Y would be wrong. But then how can it be that your belief that it is wrong for you to do Y is erroneous, if it is wrong for you to do Y?
Foot’s answer, following Aquinas, is this. From the claim that conscience binds, she says, “It does not follow … that anyone who follows his conscience necessarily acts well: he will also act badly when he intentionally does things that are evil, even if he thinks them good.”4 This entails, as Foot says, that in situations where your conscience tells you something false, that is, where your moral convictions are in error, there is no way for you to act well. If you follow your moral convictions then you act badly, since your convictions were in error. And if you violate your convictions then you also act badly since you violated your convictions. She argues for this with the example of someone who genuinely believes it is right to do something that is in fact horribly evil. “Of a man such as Himmler,” she says, “who seems to have believed that he should gas Jewish men, women and children (and therefore indisputably innocent human beings), Aquinas would have said that so long as he thought as he did he could not but act badly. For either he would spare the lives of those in the concentration camps and so go against his conscience or else he would kill them and so do an evil deed.”5
One way this might be thought to solve the problem of how conscience can both ‘bind’ and be open to error, can be found in the shift in terminology. Does the principle of conscience say, as we have been supposing, that it is wrong to go against your conscience? Or does it say that you do a bad thing to go against your conscience? The difference is this. It is beyond doubt a bad thing for you to amputate one of your own limbs, say an arm, without benefit of anesthesia. It will be horribly painful and you will then be without that limb. But it could be that only by doing this can an even worse eventuality be avoided, if for instance a huge rock has fallen on your arm while you are out hiking in the wilderness, pinning you to the ground, and the only way to escape a slow, painful death by exposure is to remove this arm.6 If this is your situation then you don’t have any good choices, only bad ones: a horribly painful amputation of your arm or slow, perhaps even more painful, death. Can there be any doubt that in such a situation (assuming ‘other things are equal’, that is, that no one will come to help you or the like) the right choice to make is to amputate the arm? That doesn’t mean that somehow this isn’t a very bad thing, amputating one’s arm without anesthesia, only that doing this is the right choice in these (very unhappy) circumstances.
Analogous situations can arise where the good and bad in question are ‘moral’. One might have to choose between, say, breaking a promise and allowing, or even causing, an innocent person to die. Or between breaking a promise to one person and breaking another promise to someone else. It could very well be that while either choice involves one doing something bad or immoral, still one choice is less bad or immoral than the other. This is one straightforward way to understand the Foot-Aquinas doctrine. It makes sense to say, what it seems plausible to take Foot to be saying when she speaks of ‘acting badly’, that all the choices someone has are bad ones, that is, morally bad or immoral ones. That is analogous to what happens in the amputation case, where all the choices are very unhappy ones. But it does not make sense to say that all the choices someone has are wrong ones. To say that someone made the wrong choice in some situation is just to say that there was more to be said for some other choice open to her (or less to be said against it). So it is very unclear what it could even mean to say that in some situation all the choices someone had were wrong ones, unless that just means that, as in the amputation case, all were bad ones. But as the amputation example shows, this is consistent with there being a right choice.7
So, if we understand it in this way, for the Foot-Aquinas doctrine to make sense of the idea that the principle of conscience is consistent with the existence of objective standards of right and wrong, we will need to understand this principle as saying not that it is wrong but that it is a bad thing to violate one’s conscience. That is, we need to change our reading of the principle of conscience from (W) to
(B) If you believe your doing X would be wrong then it is bad for you to do X (that is, your doing X would be a bad thing).
This is an improvement over (W). It allows us to say that one’s conscience can be in error because it doesn’t rule out objective, non-conscience based, standards of good and bad. It allows that, as in Foot’s example of Himmler, someone may do something bad both in following her conscience, if her conscience is in error, and, because of (B), by violating her conscience. But this can’t be the right way to understand the principle of conscience either.
For one thing, even though (B) allows us to make sense of the idea that one’s conscience can be in error, since it allows that there can be other sources of badness than conscience, it doesn’t block the first sort of problem that (W) encountered. In Foot’s Himmler example, since Himmler believed that it would be wrong for him to spare the lives of those in the concentration camps, (B) allows him (or us) to conclude that his sparing the lives of these people would be a bad thing. Indeed, along with the fact that Himmler has the moral beliefs he does, (B) entails that his sparing their lives actually would be a bad thing.8 Even though this is consistent with holding that it would be even worse for him not to spare their lives (and hence that his conscience is in error), do we really want to say that the mere fact that Himmler believes this is wrong makes it, to some degree at least, a bad thing (to be weighed against other considerations)? (B) requires this.
There is another problem here as well. It is very hard to see how anyone could ever use (B) in her actual moral reasoning, that is, in deliberating about what to do. (W) was, we might say, far too easy to use as a practical principle. As was argued above, (W) served, or would serve if we accepted it, simply to establish a whole set of moral truths, even ones diametrically opposed to each other, because it would establish all moral beliefs about what it is wrong for her to do, for the person who has the belief. If you believe it is wrong for you to do something then, according to (W), it really is wrong. That seems very implausible.
But if we change to (B), we are in something like the opposite position. To see the issue here we need to recall the distinction between the considerations that get ‘weighed up’ in practical deliberation, that is, the principles or facts (or whatever they are) that serve as the premises of practical reasoning, and the final, ‘all things considered’ judgment that is the conclusion, the judgment on which one acts (assuming no weakness of will or other similar glitch). To mark this distinction let’s call the facts or principles that get weighed up in deliberation, ‘considerations’ and the conclusion at which practical reasoning arrives, the ‘final’ or ‘all-things-considered’ judgment. This distinction applies both to practical reasoning generally and to specifically moral reasoning, where the final judgment is about what morality requires. Words such as “good” and “bad” make most sense describing the considerations that precede making a final judgment, not in that judgment itself. That is because it might be good for you to do something while still being better, even much better, for you not to do it. Similarly, it might be bad for you to do something and yet even worse for you not to do it (as in the amputation case, or where a lie is required to save an innocent life).
But words such as “should”, “ought”, “wrong” and “right” can be used in either all-things-considered judgments or in the considerations that lead someone to make such a judgment.9 It was the fact that “wrong” is typically used in the all-things-considered sense that led to the trouble, for (W), about how it could be consistent with objective moral considerations, since it cannot be wrong, all-things-considered, to do X and at the same time wrong, all-things-considered, not to do X. But presumably “wrong” can also be used in the ‘consideration’ (or pro tanto) sense. For instance when we say it is wrong to lie or wrong to break a promise, those principles are frequently not taken as simply and finally deciding the issue whenever they apply. After all one can sometimes avoid lying only by breaking a promise. In fact one can sometimes only avoid breaking one promise by breaking another. So in principles of this sort it would seem we have to read “wrong” as “a bad thing” (i.e. something wrong about it or pro tanto wrong). That is, we have to take the wrongness of lying or promise breaking as a consideration in our practical reasoning but not one that always decides the issue, i.e. that tells us what to do all things considered.
Now what about (B), the principle that says “If you believe your doing X would be wrong then it is bad for you to do X.”? If we try reading the “wrong” in the antecedent here as “a bad thing”, then this principle says that if you believe your doing X would be a bad thing then it is bad for you to do X. But on this reading, (B) no longer looks like a plausible way to understand the principle of conscience. That is because, as we have seen, sometimes doing a bad thing, even a morally bad thing such as breaking a promise or telling a lie, is the right thing to do all things considered, i.e. if the only alternative involves doing something even worse. Suppose you are in such a situation. For instance suppose that you have to choose between telling a very serious lie and allowing an innocent person to be killed. In such a case, reading (B) as saying that if you believe your doing X would be a bad thing then it is bad for you to do X both gives you no help and in any case does not say what we want it to say. It gives no help because in such a situation it says both that doing X, e.g. telling the lie, would be a bad thing and also that not doing X, not telling the lie and so allowing the innocent person to be killed, would be a bad thing. Since you believe that both your alternatives are morally bad, the principle (read this way) applies to both sides of the dilemma, so to speak, and so simply mirrors the conflict between your two bad alternatives. It thus gives you no help at all in deciding what you (morally) should do.
But in any case (B), read this way, now doesn’t say what we want it to say because it might be that in a situation such as the one we are imagining, where both alternatives are morally bad, one of the alternatives would be worse, perhaps even much worse, than the other. That would seem to be the case if one has to choose between telling a lie and allowing an innocent person to be killed. And of course the agent might perfectly well realize this and so judge that one alternative is morally much worse than the other. But that judgment is a further judgment, one that wouldn’t be included in (B) under this way of reading it. Read in the way we are considering, (B) only says that if you believe something is bad for you to do then it is bad for you to do it. And as we have seen (B), read in this way, applies to both alternatives in this situation equally. Yet if we want (B) to capture the principle of conscience, it is the further judgment, that one alternative would be better and the other worse, that (B) should cover. Otherwise you could do what you regarded as much worse than the alternative and not, on this reading of (B), violate the principle of conscience, at least not any more than if you had acted on the other, less bad alternative. But that can’t be right. If you judge that one alternative is much worse than the other then surely you violate your conscience if you act on that alternative.
So it would seem we have to read the term “wrong” in the antecedent of (B) as referring to your final, all-things-considered moral judgment about what you should do, not merely to the considerations that you weigh up in coming to that final judgment. Understood in this way, what (B) says is that if you believe that, all-things-considered, your doing X would be wrong, then it is bad for you to do X (that is how we are understanding Foot’s claim that you would ‘act badly’ in doing X). This is more plausible than the other reading of (B) we were just considering since now (B) doesn’t just mirror the considerations used in deciding what to do when you face two bad alternatives but says it is wrong not to follow your final, all-things-considered moral judgment about what you should do. And surely that is what we want the principle of conscience to say. But read in this way, (B) cannot be used in moral reasoning at all.
That is because, on this reading, (B) only applies after the agent has finished her moral reasoning and come to a final, all-things-considered judgment about what she morally ought to do. All-things-considered moral judgments (as the term is being used here at least) are not distinguished by their form or their content, any more than conclusions are in other contexts. They are distinguished by their place in an agent’s moral reasoning, that is, by their place as conclusions of this reasoning. So on this way of understanding (B), and hence on this way of understanding the principle of conscience, it only applies after the agent has come to a conclusion about what she morally should do all-things-considered. So it could not possibly be used in arriving at this conclusion.
(W), as a way of understanding the principle of conscience, raised two problems. It allowed a (surely spurious) ‘proof’ that all our moral convictions about what it is wrong for us to do are flatly true, and it allowed no sense in which one’s conscience can be in error. (B), as we are now reading it, avoids the second of these problems, though not a variant of the first (since it entails that things believed bad are actually bad). Moreover, it is not usable in moral reasoning at all. So (B) seems to entail consequences every bit as implausible as those that followed from (W).
Jonathan Dancy offers a reading of the principle of conscience that seems to avoid the sorts of problems that (W) and (B) encounter. Dancy agrees that the injunction to “follow our conscience” … “must not be taken to mean that our moral beliefs about our own actions are always true, indeed infallible.”10 So presumably he would agree that (W) cannot be the right way to understand the principle of conscience, and for essentially the reasons given above. Instead of opting for (B), however, Dancy says, “Perhaps the right structure to assign to the thought that one ought not to do what one thinks it wrong to do is that of a complex prohibition - a prohibition on thinking an action wrong, but doing it all the same.”11 This is an example of what Dancy calls a ban on “detachment”, that is, a ban on engaging in exactly the sort of reasoning that we saw above (W) would allow us to engage in, reasoning that uses (W) together with a description of one of our moral convictions so as to ‘detach’ the consequent to reach a conclusion to the effect that the action in question really is wrong for whomever has that conviction. And since, as we have seen, (B) allows a similar sort of detachment (of a conclusion to the effect that performing the action in question would be bad) presumably Dancy would reject it as well. Dancy’s idea is that when you have the moral conviction that it is wrong to do X, that does not show that in fact it is wrong for you to do X. Rather it shows that it is wrong for you to do X while at the same time believing this wrong.
(W) is this: If you believe it is wrong for you to do X then it is wrong for you to do X. What Dancy seems to be saying is that taking the principle of conscience in this way, as (W) does, mistakes where the ‘wrongness’ is, that is, really, mistakes the logical form of the principle of conscience. John Broome has pointed out that an analogous mistake can happen in formulating principles of rationality12. Here is one of his examples. Suppose you believe, for no reason at all (or indeed, against all the evidence), that
(s) The world was made in six days.
This sentence obviously entails
(l) The world was made in less than a week.
Because of this entailment of (l) by (s) it seems that
(E) If you believe (s) then you ought to believe (l).
But since you do believe (s), by modus ponens (E) gives us at once
(STC: for “Spectacular Theoretical Conclusion”)
You ought to believe the world was made in less than a week.
So (E), together with the fact that you believe (s), lets us ‘prove’ that you ought to believe (l), in spite of the fact that you believe (s) for no reason at all. And, Broome adds, since, trivially, every sentence entails itself, the relevant generalization of (E) will allow us to argue from the fact that you believe something to the conclusion that you ought to believe it for absolutely every belief you have. Needless to say this is deeply implausible, and for very much the same reason (W) is.
The solution, according to Broome, is to see that what (E) says, which is roughly that if you believe something then you ought to believe what it entails, is just false. In fact you may not have any reason at all to believe what it entails. There is a ‘normative requirement’ here all right but its form is different from (E). What is true is not that belief in (s) gives you reason to believe (l) but that since (s) entails (l), belief in (s) commits you to (l). This is what Robert Audi calls a coherence principle and, as he says, it might be purely negative, ruling out as incoherent a belief in the first of these claims without a belief in the second.13 (E) is hypothetical in form (‘If you believe (s) then you ought to believe (l)’). So if it were correct, together with the fact that you do indeed believe (s) we could validly ‘detach’ the conclusion that you ought to believe (l). That is what causes the problem. So the correct principle won’t allow such detachment; hence can’t be hypothetical in form. It will simply express the fact that belief in (s) commits you to (l) and so (borrowing Broome’s phrasing) will be roughly:
(R) You ought to see to it that, if you believe (s) then you believe (l).
Here the ‘ought’ covers the whole hypothetical, not just the consequent. So (R) has the form (O(if Bs then Bl)). In Broome’s terms, in the correct principle the ‘ought’ has wide, rather than narrow scope.
Put another way, the correct principle, (R), expresses a normative requirement on belief (or believers) but does not allow the conclusion that a belief in (s) gives one reason to believe (l) or that one ought to believe (l). A similar point applies to intentions and so to practical reasoning. If you intend to drink the last cold beer in the house and know that doing so requires opening the refrigerator door, then you ought to open that door. But it would be a mistake to take this ‘ought’ as ‘narrowly’ covering only the consequent of this sentence, as if it were hypothetical in form. That would mean that if you do indeed have this intention and bit of knowledge it would follow that you ought to open the refrigerator door. And that would be so even if opening that door would be a terrible idea (since you just watched your enemy wire it to a bomb). So, again, the correct principle here would seem to have to use a ‘wide scope’ ought and say something to the effect that you ought to see to it that if you intend to do A and believe doing B is required for doing A, then you do (or at least intend to do) B. That is a normative requirement you can satisfy either by intending to do B (if you intend to do A) or else by giving up the intention to do A (once you see what a bad idea doing B is).
Dancy’s suggestion is that the principle of conscience should be understood in a somewhat similar way, that is as having a form where the term ‘wrong’ has ‘wide scope’ in that it governs the whole sentence and so does not allow valid ‘detachment’ of a narrow scope judgment about the wrongness or badness of that action as both (W) and (B) did. Presumably this would be:
(D) It is wrong to both believe that some possible action of yours would be wrong and yet do it anyway.
Here the logical form of (D) is such that the first ‘wrong’ governs the whole following conjunction. If (D) or something similar is the correct way of understanding the principle of conscience then it is of a different logical form than (W). It is not a hypothetical with the ‘wrong’ applying only to the consequent. So (D) does not allow the sort of ‘detachment’ that made (W) so implausible.
(D) may well be the correct way to understand the principle of conscience. But it still leaves us with a puzzle. Notice that the first ‘wrong’ in (D) is an interpretation of Dancy’s claim that the principle of conscience expresses “a prohibition”. There are two reasons for this interpretation. The first is that the principle of conscience certainly seems to be a moral principle. That is clearly how Foot and Aquinas take it (not to mention Jiminy Cricket) and it seems intuitively correct. The second is that, unlike (R) for instance (or the failed (E)), it seems very implausible to try to understand the prohibition of which Dancy speaks as a matter of logical coherence. Someone who believes something, such as (s), but denies or fails to believe what it entails, such as (l), exhibits a form of incoherence. So would someone who intends to do something that she believes requires some other action but fails to intend to perform that other action. The normative requirement contained in (R) is obviously grounded on the logical (or semantic) relations among the contents of the beliefs described and similarly for the analogous requirement for intentions.
But violations of conscience involve no such incoherence. Someone who does something she believes to be morally wrong, say in the face of some temptation, is not exhibiting any incoherence. Or if she is it would seem to have to be in a quite different sense than that which grounds (R). What the principle of conscience says is not that in doing something she believes to be morally wrong she does something that involves her in incoherence. It says that she does something morally wrong. So the prohibition of which Dancy speaks would seem to have to be a moral one.
This leaves (D) with the same puzzle that (B) involved. (D) declares to be morally wrong the performance of an action the agent herself believes to be morally wrong. As we saw when discussing (B) though, the belief here must be an ‘all-things-considered’ belief about the moral status of the action. It cannot be a merely that the action is believed to be ‘pro tanto’ wrong, or every situation where an agent does what she regards as the ‘lesser of two evils’ will involve a violation of her conscience. So (D) applies only after the agent has formed her all-things-considered judgment about the moral status of the action. And the ‘after’ here is a matter of logic. (D) applies only to actions about which the agent has already formed an all-things-considered moral judgment. It follows from this that if (D) is the right way to understand the principle of conscience then that principle cannot be used by the agent to which it applies to decide whether what she is doing is morally wrong.
When Jiminy Cricket advised Pinocchio to follow his conscience, he seemed to be offering a piece of moral advice, something Pinocchio should follow and should follow on moral grounds. If the argument of this paper has been correct however it is very difficult to see how this advice can be any good. The moral principle that would have to be the basis of this advice seems to involve a dilemma. Understood in the most straightforward way, as a hypothetical to the effect that if you believe some act to be wrong then you do wrong in performing that act, the principle is pretty clearly just false. It would commit anyone who accepted it to some deeply implausible positions. Alternatively, understood in such a way as to avoid these implausible commitments, for instance in that way Dancy suggests, it applies only after the agent has formed her all-things-considered moral judgment and so cannot be used by the agent to which it applies to decide what she morally ought to do. Of course only three ways of understanding the principle of conscience have been considered here. Is there another way?14
1 In the movie, Pinocchio, Walt Disney Productions, 1940.
2 Jonathan Dancy makes a similar point, which will be discussed further below. See Jonathan Dancy, Practical Reality (Oxford, Oxford University Press, 2000), p. 59.
3 Philippa Foot, “Moral Relativism”, reprinted in her Moral Dilemmas (Oxford, Oxford University Press, 2002), p. 28.
4 Foot, ibid.
5 Foot, ibid.
6 For a description of how he managed to do just this see Aron Ralston’s book, Between a Rock and a Hard Place (Atria Books, New York, 2004).
7 It is also possible of course that all one’s choices are equally bad, but even then it would seem not that all one’s choices were wrong but that none were either right or wrong.
9 As Foot herself has pointed out. See Philippa Foot, “Moral Realism and Moral Dilemma”, reprinted in Foot, 2002.
10 Dancy, p. 59.
11 Dancy, ibid.
12 See John Broome, “Normative Requirements” Ratio, 12 (1999), pp. 398-419. The example below is from John Broome, “Reasons”, in R. Jay Wallace, Michael Smith, Samuel Scheffler and Philip Pettit (eds.), Reason and Value: Themes from the Moral Philosophy of Joseph Raz (Oxford, Oxford University Press, 2004). Thanks are due to an anonymous referee for American Philosophical Quarterly for pointing out the parallel between Broome’s views and Dancy’s.
13 See Robert Audi, Practical Reasoning and Ethical Decision (London, Routledge, 2006) , p. 165.
14 Earlier versions of this paper were read at New Mexico State University and at the University of New Mexico. This version has benefited from the comments and questions of members of both audiences. It has also benefited greatly from numerous excellent suggestions of an anonymous referee for American Philosophical Quarterly