Ethics without principles
Jonathan Dancy interviewed by Richard Marshall.
Jonathan Dancy is a groovy moral philosopher who writes books about moral particularism and is always wondering what reasons count. He’s written many books about this because he reckons we ought to work out whether we think things like moral principles actually exist. He says no empirical enquiry is sufficient to establish relevance; the matter requires judgement in a way that lies beyond observation and inference. Which makes you think, which is what his philosophical jive is all about!
3:AM: You’ve been a top philosopher for quite some time. What have been the satisfactions of being one so far, and what would you say to someone who questioned the relevance of philosophy?
Jonathan Dancy: I don’t feel like a top philosopher at all. If I am at the top, there must be a great many others up there with me. But I do think that my philosophical life has been far more successful than I ever expected, and the notion of success at issue here is grounded in what I take to be a common aim, or at least hope, of intellectuals, that their ideas will make a difference to the way in which others think, or change the intellectual agenda in some way. So the satisfaction for me comes when I see the ways in which my efforts have affected the contents of ethics courses, of readers, and of textbooks, and when I see that the possibility of particularism in ethics seems now to be taken seriously in a way that it never was before.
As for those who question the relevance of philosophy, they first have to say what it is not relevant to. There are normally two answers to this question, relevant to them and relevant to practical life. That it is not relevant to their concerns is something I find sad but inevitable; I don’t see that everyone should be interested in what we do. The idea that it is not relevant to practical life seems, by contrast, to depend upon a rather limited conception of what practical life may be. Just to take an example close to home: surely the question whether there are any moral principles is pretty close to practice, since if we accept that there are none, this is likely to affect how we think and thereby what we go on to do.
3:AM: You make a striking point in an article responding to an idea by Christopher Hookway which gets at one area of your philosophical interests. That there are considerations implicated in believing something that are not in themselves reasons for believing something but are nevertheless ‘enabling conditions’ giving us the right to believe something is something you have been arguing for the last twenty years. So what is the significance of this argument? What alternative view are you against here? Can you perhaps give us an example?
JD: The distinction between the things (or people) who do the work and those that don’t themselves do the work but enable others to do it has widespread implications. A fairly trivial example is the distinction between the role of the actors in a theatre and the role of the backstage staff. The role of the latter is to enable the former to play their role. We just need a distinction between roles. More generally, though, there is a tendency in the theory of reasoning to see every relevant consideration as playing the same role, that of a premise.
But my view is that in order to understand reasoning, we need to distinguish between the different sorts of roles that considerations can play. In this, I find myself in agreement with Stephen Toulmin’s suggestions in his The Uses of Argument. Just as a principle of inference should be distinguished from a premise, so should considerations that affect the ability of other such to stand as reasons for our conclusion.
Here is an example of the sort of thing I mean. That the gun was found in the butler’s cupboard is, let us say, a reason to believe him to be the guilty party. That it did not appear that we were intended to find it plays, however, a different role. It is not a further reason to believe the butler to be guilty, but a consideration such that, had it not been the case, the location of the gun would have been a reason to believe the butler innocent. That is, if it looks like a plant, the fact that we found the gun in the butler’s cupboard is now a reason on the other side.
In moral philosophy, I used the distinction between favourers and enablers to stop a certain form of anti-particularist argument. Suppose I offer a counterexample to a suggested moral principle of the form ‘All F actions are wrong’, my example being an action that is F and G, and not wrong. A common response is to say that the relevant principle is really ‘All actions that are F and not G are wrong’.
My response to this is, first, that this process might go on indefinitely, and second, that this supposes that being not-G has the same sort of relevance to the wrongness of the action as does being F; but it might be that being not-G is not itself a wrong-making feature, but simply a feature whose presence is required for the Fness of the action to make it wrong. If so, being not-G is an enabler, rather than a wrong-maker. It enables the Fness of the action to make it wrong, without itself counting against the action. So the supposed principle ‘All F actions are wrong’, is false, and there is no true replacement ‘All actions that are F and G are wrong’.
3:AM: In your book Moral Reasons you argue for a realist conception of ethics, which is the claim that there are facts of the matter in ethics. And then you attack consequentialism because you find consequentialists have far too narrow a view of reasons relevant for moral judgments. I guess these include your ‘enabling conditions’ you just spoke about. But what would you say to the policy maker who says that the reasons for moral action on your account are too diffuse to be pinned down and measured and so lead to the consequence that public policy makers, condemned to work with mechanical algorithms are thus condemned to work outside of morality?
JD: I don’t see that policy makers are condemned to work with mechanical algorithms. I do allow that the process of moral reasoning, whether in public affairs or in private life, is much more complex than traditional accounts have allowed. But that does not lead to pessimism, since I think that we also have the rational capacity to cope with the complexities that face us. But talk of pinning reasons down and measuring them makes me pretty uneasy.
I think that generally we all recognise the same reasons, and that disagreement arises when we assess their relative strength. Here talk of measuring may seem appropriate, but I would think that we estimate the strength of reasons rather than measure it. (There is, after all, no unit of measurement for reasons.) Such estimation is not easy, but politicians expect to differ from each other on such matters and do not expect (and never have expected) to be able somehow to prove that they are right.
3:AM: You used to argue that moral knowledge is of what in particular cases is a moral reason for what. There are no principles for moral reasons. (or any reasons actually). Basic moral knowledge is of contingent particular truths. But we know these a priori. Each new experience has to be worked out and judged but the judgment delivered is a priori. Is this right? Can you say something more about this and why you argued that this kind of judgment in the moral sphere is analogous to how we know what is relevant to what. Linked to what I asked in the previous question, but from a slightly different angle, does your view imply that we should be learning how to make judgments sensitive to these reasons rather than applying some cost accountancy model of measuring worth?
JD: Yes, in Ethics without Principles I did argue that our basic moral knowledge is of what is a reason for what in a particular case, and that such knowledge is a priori knowledge of contingent particular truths.
My reasoning was this. If we allow my starting point, that what is a reason in one case may not be a reason in another, the question whether it is a reason here seems to have to be a contingent one (since it might have been otherwise and would have been if the case had been relevantly different), and particular. But once all the empirical facts are in, we still have work to do to establish what features are reasons for what. So our decision on the latter front must be reached a priori, since it lies beyond anything that experience can inform us about.
The same argument applies to our decisions on another rational matter, what is relevant to what (in a particular case). No empirical enquiry is sufficient to establish relevance; the matter requires judgement in a way that lies beyond observation and inference. And if something can be deprived of its relevance by changes elsewhere, it must be a contingent matter whether it is here relevant or not. So yes, I do think that our sense of relevance can be trained, and that it should be, and I don’t think that there is a possible unit of measurement for relevance (or worth) that would be capable of doing most of the work for us.
3:AM: Your view is a kind of moral particularism and if it is, how do you answer the argument that if we acquire all our moral concepts on a case by case consideration its difficult how we could ever learn anything like a generally realizable morality? You discuss this and other issues in your book that has the title that announces what for some sounds like a contradiction: Ethics Without Principles?
JD: I don’t think I would say that we acquire all our moral concepts on a case by case basis. My view is rather that we acquire our moral knowledge on a case by case basis, starting, as I said above, from knowledge of particular reasons. It doesn’t follow from this alone that we cannot build up a ‘generally realisable morality’, even if what is meant by this that we cannot build up a principle-based ethic. Ross, after all, thought that we do start from particular cases, but that what we discern in such cases was capable of revealing to us the truth of certain prima facie principles, by a procedure called intuitive induction.
I don’t say that this is impossible, but I do say that every suggested Rossian prima facie principle is subject to counter-example, and that I know of no real need for moral principles. Much of my time has been spent arguing that principles are at best redundant. In saying this I take myself to be, in a way, defending the possibility of moral distinctions. For if there are no available principles, or not enough to do all the work, some would say that this is the death of ethics. Not I, however. My view is that this wouldn’t matter, since morality is in no need of principles.
3:AM: Did Christopher Peacocke change your mind? I was impressed by your remarks when writing about his book The Realm of Reason that twenty years of hard fought argumentation was severely challenged by his book. How hard is it to change your mind in philosophy, not on a minor detail but really on an argument which, as you wrote then, was something that a substantial amount of your philosophical career had been dedicated to developing and defending?
JD: Peacocke did not change my mind, but he did make it clear to me that I had more work to do. That is not much damage, for it is not as if I didn’t really know this already. One can, however, conveniently forget such things and carry on regardless, and this was what I had been doing. And an occasional shock is nothing but salutary.
But your question does raise interesting issues about philosophical commitment. Many philosophers take up positions which they defend with vigour. I am one such. But when I ask myself whether I believe that my position is correct, I don’t really know how to reply. I have committed myself to these things, but belief seems to require more than that sort of commitment. Still, whether I believe them or not, abandoning them would be a blow. Of course one can abandon an argument without abandoning its conclusion.
And indeed I did lose one of my favourite arguments when Michael Ridge and Sean McKeever persuaded me that holism in the theory of reasons does not actually entail particularism in ethics. To preserve my conclusion, I had to find another route from my premise to my conclusion. But all this is everyday stuff.
Suppose, by contrast, that I came to be persuaded that there was a need for principles of some type or other, in the sort of way that I have been officially denying for years. I would definitely have to try to find some way to come to terms with this, and I would probably console myself with the thought that at least I had forced people to do more than assume the necessity of principles, and actually to work out why they are so indispensable in ethics (when they seem not to be elsewhere, e.g. in aesthetics). I think I could come to think of this as an acceptable second best.
3:AM: You ask a pretty cool question which will help us understand some more about our views on the relation of reasons to acting. You ask in a recent paper; ‘Is it possible to run for the reason that the train is leaving when one doesn’t know that the train is leaving?’ So can we, and what does your argument show us?
JD: Can we run for the reason that p when we do not know that p? John Hyman and Jennifer Hornsby are convinced that we cannot, and I am convinced that we can. My view is that we cannot run for the reason that p if we do not believe that p, but that knowledge is not necessary. It is quite hard to work out how to argue about this. Take a case where I take myself to know what the bus timetable is, and in that light am running to the bus stop so as not to miss the bus. Actually I don’t know this, because I have misremembered what is written in the timetable which I only consulted last week. Since then, however, the bus times have been changed, so that in fact I am right in thinking that the bus is about to leave. I would say, if asked what my reason for running is, that it is that the bus is about to leave. Hyman and Hornsby would say that this cannot be my reason, since I do not know it to be true; all I have is a luckily true belief. I think they owe us a reason for saying this, and I have not really seen one yet. But they certainly have some interesting things to say against my position, which may yet lead me to change some aspects of the position I defended in Practical Reality.
3:AM: What’s the difference between the right and the good and why should it matter?
JD: There are various positions to be considered. Some people think that the best action is always the right one, and others don’t. If the latter are in the right of it, there must be some difference between being right and being best. Even if the best action is always the right one, it might be that we are dealing here with two different concepts, and two different properties. G. E. Moore did say in Principia Ethica that it was there is only one concept here, but he later abandoned that view in favour of the view that there are two concepts and two properties, but that every action that has one property has the other, and so every instance of one concept is an instance of the other.
Those who think that the best action is always the right one can still allow that the action is right because it is best, and not best because it is right. Others think that though we do have a (prima facie) duty to make things go as well as possible, we have other duties which sometimes take precedence, so that we can be morally required to make things worse than we might have. On this view, the good and the right can conflict. This is significant because the basic intuition that underlies consequentialism is a sort of philosophical boggle, the thought that it cannot be one’s duty to do anything other than the best one can. So the possibility of a non-consequentialist position is dependent, it seems, on supposing otherwise.
3:AM: An example of the great subtleness of your thought is the distinction you make between the prospective use of a moral principle to guide action, on the one hand, and its retrospective use to appraise the way an agent governed herself. You have a cool example to show why this is very important when you consider why we shouldn’t kill a person so her organs can be used to save the lives of multiple other individuals. Can you say why this is important especially in the light of considerations about intentions and moral action.
JD: I am not quite sure what this question is aiming at. But one relevant fact is that assessing an action that has been done must be a very different activity from evaluating the merits of different alternative possibilities when deciding how to act. For when we are deciding how to act, there is as yet no actual action there to evaluate. All that deliberation can do is to tell us, not which action to do, but how to act – that is, it tells us that our action, when we do it, must have certain qualities, be of a certain sort.
Similarly, a reason to act is not a reason to do this or that particular action, but only a reason to act in a certain way, or to do an act of a certain sort. So the prospective question is quite different from the retrospective question. Intentions, too, can only be intentions to act in a certain way; one cannot intend a particular action. One can of course ask of an action, once it is done, whether it was intentional or not. But this last is not a question about the action, understood as a particular, but only about whether the agent’s behaviour was intentional in some respects.
3:AM: You’ve a general worry that there’s an orthodoxy current at the moment that doesn’t make a distinction between reasons for action and reasons for why we ought to do something. Is that right? So how interested have you been in work of xphi and others who have looked at what the folk actually think about moral reasons and whether their intuitions square with those of philosophers?
JD: Yes, I think there is a distinction between two relations: the first is the favouring relation that relates considerations and actions (or beliefs, come to that, or emotions – any response, really) and the other is the making-right relation, which is a relation between consideration and rightness. To be a rightmaker is not the same as to be a favourer; being right is not favoured by anything.
Nonetheless a consideration that favours an action may succeed in making it right. It is all a bit complicated at this point and I am not at all sure that I have seen my way through it. I’m not sure, either, whether this issue is much connected to the results of experimental philosophy. I haven’t been as interested in these results as I should have been, probably, but in the present case I don’t see that one can learn much by asking the folk what they think or how they would respond to certain scenarios. The existence and nature of a subtle distinction like this one is not something about which I think the folk would have much to say, and if they did pronounce on the matter (which I would say is very improbable) I would not take their views to be of much significance. It is one thing to see how the folk operate a distinction that they understand, and another to see whether they can grasp a distinction in the first place.
3:AM: Isn’t there a danger in your views defending judgment, thick concepts and the like that morals become fixed by the prejudices of an in-group. I recall Ernest Gellner worrying that consequentialism was thought of as being the ethics for the parvenu, the outsider who hadn’t internalized the old-boy culture (I think his target was Michael Oakeshot). Given that so much of modern industrial society makes us all strangers don’t your arguments make us all foreigners to morality because there is no single dominant culture in play?
JD: I do think that there are non-disentanglable thick concepts, but I don’t think it impossible to abandon a thick concept for the reason that the distinctions that it draws are no longer ones that one feels comfortable with. Of course we are brought up with a way of looking at things, much of which is expressed using thick concepts, and it may be hard to escape from that viewpoint, but I would not say it is impossible. So we are not locked into the ethical perspective we have inherited from our parents, or that is common to our social group. And the (supposed) fact that there is no single dominant moral culture in contemporary (Western) society does not prevent people from having, and sharing with others, fairly ferocious moral opinions. So I don’t see that danger either.
3:AM: In your introductory essay to Blackwell’s Companion to Epistemology you make remarks that suggest that you feel that there are genuine distinctions to be made between analytic and continental philosophy. Is this right, and if so, what are they?
JD: Actually those remarks were a little tongue in cheek. Bernard Williams used to say that the distinction between analytic and continental philosophy is a cross-classification, like the distinction between Japanese cars and sports cars. I don’t think that is quite the right way to look at it. Those who run this distinction tend to have certain writers in mind on both sides, say Quine at one end and Derrida at the other; and there are indeed huge differences in style between these two. But most philosophy is at neither extreme.
I remember a book by Julius Kovesi, Moral Notions, which argued that the supposed distinction between ‘good’ and ‘yellow’ was actually a matter of degree, and that many concepts lay between these two outliers. By choosing the two things at either end of the spectrum one gives the impression that there is a sort of chasm between them, and that everything must fall on one side or the other, when in fact there is no such chasm, and all we need to do is to plot the relevant differences carefully. The same applies to the differences between Quine and Derrida.
3:AM: I notice that you are partly responsible for Parfit writing his two volume On What Matters. He was supposed to write some comments to contributions to your book Reading Parfit but he went out of control and just wouldn’t stop writing! So, given that you are (in a strange way) its midwife, what do you make of On What Matters and would you defend it against naturalist challenges such as Philip Kitchers’s?
JD: I haven’t read Parfit’s On What Matters since it came out, though I have read many parts of it, or versions of many parts, many times over the years. I’ll have to rectify this soon, because I will be teaching this book at UT Austin next Spring semester. Whatever one thinks of the book, I think it hardly deniable that Parfit is a genius, extraordinarily talented and inventive. One does not have to agree with him to say this.
But I do think that he is right about non-naturalism. It is one thing to show that non-naturalism is a live option, and another to show that it is the truth. The general question how there can be instantiations of non-natural properties and relations in a material world seems to me to be one without much of a bite. I am helped in thinking this by a sense that the favouring relation is unlikely to be naturalisable. Of course this does nothing to assuage those who feel that the very idea of a non-natural property or relation is a nonsense. But philosophers choose which attacks to respond to and which simply to leave aside.
3:AM: When not reading and writing philosophy, are there books that you have found inspirational or enlightening?
JD: The trouble with this question is that, though I know that it sounds a bit limited, the sort of inspiration and enlightenment I seek and recognise is going to be broadly philosophical. So anything I mention is going to be semi-philosophical in nature: Montaigne, for instance, or Sir Thomas Browne – two people whose minds seem so much better stocked than my own. But these are something of an acquired taste.
3:AM: And finally, are there five books you would recommend to the cool readers here at 3:AM wanting to learn more about all this stuff?
JD: In general it seems to me preferable to read the great classics of the subject, or shorter books that have stood the test of time pretty well, rather than the latest efforts of professional philosophers. Among the classics are Hume’s Enquiry Concerning the Principles of Morals, Kant’s Groundwork of the Metaphysics of Morals (but one really needs help with this), and Mill’s Utilitarianism. More recent short books are G. E. Moore’s Ethics, A. C. Ewing’s Teach Yourself Ethics, and Bernard Williams’ Morality. (I would love to write a book of this last sort myself, but it is not an easy task.)
ABOUT THE INTERVIEWER
Richard Marshall is still biding his time.
First published in 3:AM Magazine: Friday, August 31st, 2012.