Right-making features: Do they speak?

I have been reading and submitting ideas, gaining feedback in the last few weeks. This has led me to decide that I am going to focus upon a metaphysical idea of what a principle is, and how it makes actions right or wrong. This idea is a big part of Dancy, but it is something which has also been discussed by a little by Shafer-Landau in his paper moral-rules. I am going to explore an aspect of this idea which I find both interesting and slippery.

If actions are made right by their non-moral features, or the non-moral features of the situation they occur within, then these features are right-making. It seems easy to point to cases where a non-moral feature of an action makes that action right: the generosity of the action of buying a hot drink for a homeless person in the wintertime makes the action good. It is harder to point to a non-moral feature of situations which are right-making more generally. Generously giving out the stolen goods you have acquired rather than returning them to their owners is not a better act because it is generous.

Dancy makes much of this kind of distinction. In Ethics Without Principles He discusses the difference between two interpretations of holism: the thesis in the theory of reasons that a feature that is a reason in one case may be no reason at all, or an opposite reason, in another. The first interpretation might be of the kind he calls a 'Brandom-style approach. Robert Brandom gives a holistic outline of reasons which Dancy calls 'non-monotonic. Brandom gives these conditions:

1. If I strike this dry, well-made match, then it will light. (p -> q)
2. If p and the match is in a very strong electromagnetic field, then it will not light. ((p&r)->¬q)
3. If p and r and the match is in a Faraday cage, then it will light. ((p&q&r)->q)
4. If p and r and s and the room is evacuated of oxygen, then it will not light. ((p&q&r&s&t)->¬q)

This set of conditionals offers a different approach to how reasons might operate together than Dancy sees himself as advocating. Dancy believes this to be a different kind of holism than his, because each combination of features (p, p&r, p&r&s, p&r&s&t) speaks in favour of the match lighting or not lighting. Dancy wants to say that a single feature may in one situation be a reason in favour of one outcome in one situation, and in another situation be a reason against the same thing occurring. In one case it is right to do X because of feature Y, and another case it is wrong to do X because of Y. When the features add up they retain their identity, but they might change their weight and force.

Interesting distinction. My first impression is that it might not quite add up, but I am taking Dancy with a pinch of salt at the moment.

Guides and Standards

Continuing the search for principles of moral philosophy and what these things might be I have (as mentioned in the last post) recently been reading the excellent Principled Ethics. I have previously looked at McKeever and Ridge's first description of what principles might be; standards (I wrote about them here). They have more distinctions to add, and I will now move on to their second and third suggestion of what a principle might be: guides, and action guiding standards.

Where standards are the exceptionless rules we might follow if we had a godlike perspective, rules like the principle of utility or the categorical imperative which apply always, perhaps even necessarily, guides are more like rules of thumb. If we were explaining to someone how to win at the stock market we might either pass on the obvious but vacuous standard "buy low and sell high", or give the practical advice of spreading their investments and knowing what the average price is for each share bought. Ethics in practice are performed in a limited time frame, by imperfect agents acting in non-ideal situations. Because the bigger picture is hard to even know, and harder to predict, it is useful perhaps to see principles as less like standards and more like guides. Like rules of thumb.

The authors begin to talk about how one might construe Ross and his system of prima facie duties as a system of guides. I suppose their intent was to highlight the way in which his system allows rules to override each-other, leading one to occasionally break one principle because of consideration of another. They put a disclaimer in, to highlight that Ross may not have intended his ideas in this way, and I feel this was wise. Ross did not have this kind of idea in mind at all. There is a huge gulf between 'useful rules of thumb' and the fundamental set of principles he described. His method was to 'boil down' certain principles we think of as part of morality until only those which could not be split further remained. He certainly did not think of these as the kinds of rules which allowed exception, in the way a rule of thumb might when the standard was consulted.

McKeever and Ridge continue with an interesting combination of their first two ideas, and this gives us action guiding standards (p9).  They consider two ways this might take shape, one 'uninteresting' version in which the only standards accepted are practical, and one which makes use of an ideal virtuous agent. The virtuous agent would be able to ascent to the higher level of critical awareness of a standard to assess the rightness of the guide they use. The authors this ability is necessary for situation in which the agent is able to forsake their virtue in-order to (for example) achieve the greatest happiness for the greatest number.

I wonder if the uninteresting version is as uninteresting, and I wonder if the virtuous agent view is actually much to do with virtue. The 'uninteresting' view is interesting because I like the idea that some guides for action may also be exceptionless standards. I suppose the authors see that this kind of crossover would be rare due to the complexity of putting standards into practice, and in this I certainly see their point. It struck me that a principle like 'do not lie' if it were understood as complicated and involving clauses which specified action may be exceptionless. If it were sufficiently complicated, it would be both a standard and a guide... but of course as finite beings, perhaps we would not be able to grasp such complexity.

In my mind I am currently thinking of this as the complexity problem. A related problem for principles seems to be the non-translatability problem. Perhaps the two are linked: principles do not seem to be the kinds of things we can express with a simple rule (though we try to anyway). When we do express them so, they have hidden, unmentioned clauses which we take to be 'obvious' if one understands the principle properly. It strikes me that this 'understanding properly' is highly complex, and so I don't think it is inconceivable that principles are complex. I need to respond to Dancy on this point however, and am not ready yet.

As for the virtuous agent view, I find it misunderstands virtue. I realise the authors are talking about a virtuous agent in the sense that a utilitarian might call her virtuous who acts in ways which generally promote the greatest happiness. Given this meaning of 'virtuous agent' it does make sense to say that one might forsake their 'virtue' in order to promote the greatest good. However this is to take the idea of virtue as merely the internalisation of acts which generally promote the greatest happiness. Virtue on this view of it is the rule of thumb, and not something which is good in itself.

What I have against this view is perhaps just pre-critical thought. I like to think that virtue is intrinsically good, and that an agent who exemplifies the virtues (perhaps not Aristotle, but similar) would thereby be good in some sense which transcends utility. I need to consider this a bit further to develop it into a criticism.

A Better Grip of the Principle

I have finally managed to get my hands on McKeever and Ridge's book Principled Ethics, and am both quite excited and a bit worried by their approach. The reason I am quite excited is that it mirrors what I am already trying to do. The reason I am a bit worried is because I fear they may have already done what I was trying to do, and made my line of thought redundant. I suppose I will simply have to study in depth and see if I can pick up on anything they have missed!

The sensibly begin their work by separating out terms in the way I had previously intended to, and in particular getting straight (to them!) what is meant by a moral principle. They initially have 6 divisions into the term which they use for analysis (p8-14). Additionally they want to talk about the scope of operation for moral principles. They discuss hedged and un-hedged principles (p21) and he way in which a principle can be moral, non-moral or intra moral. I wanted to here discuss the distinctions they makes regarding principles, though I will later return to the interesting topic of scope. I seems both must be important if we want to talk about principles.

They begin by saying that one might simply see principles as standards. The authors characterise this sense of 'principle' as being something which provides sufficient condition for the application of a moral concept (p7). These are supposed to be necessary truths which might found any contingent principles, and are exemplified by notions such as the categorical imperative or the principle of utility.

Part of their discussion of this particular characterisation is to do with an argument about supervenience. Even particularists might generally agree that moral facts about a case supervene upon the non-moral facts. If there is a physical difference in the facts of the case then this difference may make a moral difference to the case as well. According to McKeever and Ridge this alone means that we might generally describe moral concepts in physical terms, and so that a comprehensive physical description of the possible world in which that case occurs (including all the facts) would give a sufficient condition for the application of a moral concept. Of course if this were true, principles as standards would be physically true.

They mention this only to flesh out what they mean by 'generalisation'. They do not mean a comprehensive physical description of the physical world, as this would be a very very long description indeed. Also many things in such a 'supervenience function' are irrelevant to moral thought, just as many physical facts seem irrelevant in any case.

So a standard is less that a supervenience function (it is shorter) and must be useful if it is to qualify as a principle. This interesting distinction is useful, and one I intend to make use of.

The idea of discounting supervenience function's as candidates for moral principles seems a bit strange to me. An example of a principle, like the principle of maximum utility, seems like a physical law. It is a generalised proposition which can be applied in a uniform way (if we could know the consequences, of course!) and would arguable give us an idea of what course of action would be correct. The idea that a comprehensive description of the physical world might qualify as the same kind of thing seems bizarre, and obviously quite different from this. For a start it is much longer, and far more complicated, and may come as a string of data capturing the position and momentum of fundamental particles.

If we are to consider both the short, and the nearly infinite as candidates for principles, then why simply rule out a complete description of the given possible world? Why not just use a complete physical description of a given case?

I notice that the authors make use of the notion of a 'moral concept' which is separate from a moral principle. This term is used throughout the chapter, and I am not certain what it means. As I understand particularism, it is largely a negative thesis, and I am not sure that the sort of morality it might promote requires anything so clear as a concept. At present I struggle to see what could be a moral concept, other than something like a principle like "do not lie" or "do not kill". Perhaps further reflection will shed some light upon this, but it is a possible problem with the discussion.

I will add to this tomorrow with some thoughts about their next distinction,; principles as guides.