For the purposes of later work on meta-ethics (a long series explaining my meta-ethical theory, pluralistic moral reductionism), it will be helpful for me to summarize Peter Railton’s moral reductionism for my audience. To do so, I’ll be summarizing some of chapter 9 in Miller’s Introduction to Contemporary Metaethics.
I previously explained the differences between cognitivism and non-cognitivism, naturalism and non-naturalism, and naturalistic reductionism vs. naturalistic non-reductionism in my post on Cornell realism. Now I discuss one kind of naturalistic moral reductionism, the kind developed by Peter Railton in the 1980s and 90s.
Railton’s approach to moral theory
Railton is a methodological naturalist in that he claims to do philosophy with methods that are continuous with the sciences. He is also a metaphysical naturalist in that he thinks only natural things exist – things made of quarks and so on.
His strategy is “to postulate a realm of facts in virtue of the contribution they would make to the a posteriori explanation of features of our experience.” Specifically, he plans to offer reforming definitions for moral terms. For example, one might define ‘morally right’ as standing in for some natural property N (say, ‘contributes to the well-being of conscious creatures’). Railton doesn’t mean to claim that when normal English speakers say ‘morally right’ they mean to call up property N. Rather, he seeks to find reductive definitions for moral terms that fit the world and explain certain features of our experience.
And because he offers a reductionistic view of morality, Railton does not need to argue (with the Cornell realists) that moral properties pull their own weight in explanations. Instead, the natural properties to which moral properties reduce already pull their own weight in scientific explanations.
When offering reductionistic accounts, one might offer a vindicative reduction or an eliminative reduction. Consider the case of ‘polywater’, which in the 60s was thought to be a special form of water. It turned out that polywater was just normal water that had been contaminated by poorly cleaned lab equipment. Here, the reduction of polywater to “water containing impurities from poorly cleaned lab equipment” was an eliminative reduction; it called for an abandonment of the concept of polywater.
Other times, a reduction leaves enough of our original concept in tact that we find it tolerable to keep the term as it was. The reduction of water to H2O is one example. We once thought that water was one of the fundamental elements of the world, but it turned out that water is itself made of more fundamental elements. Still, the reduction of water to H2O “reinforces, rather than impugning, our sense that there really is water.” This reduction merely showed us more clearly what water really is in the natural world. If we had discovered that water was merely air with a few impurities in it (say, fire), we may have found that reduction as being eliminative, and abandoned our use of the concept of ‘water.’
Railton’s project is to see whether his reduction of moral properties to natural properties can be vindicative rather than eliminative. His criteria for success are (1) that the natural properties to which moral terms are reduced can play a successful role in explaining things (an explanatory role), and (2) that these natural properties can engage people motivationally “in the ways characteristic of moral properties” (a normative role).
Railton begins with a reductive account of non-moral value: that something can be “desirable for someone, or good for him.”
Hobbes can be thought of as offering a reforming definition for non-moral goodness such that “good for X” means “desired by X.” Sex is (non-morally) good for me if I desire sex. Etc.
But Railton resists such a crude reduction of non-moral goodness:
My [actual desires] frequently reflect ignorance, confusion, or lack of consideration, as hindsight attests. The fact that I am now constituted that I desire something which, had I better knowledge of it, I would wish I had never sought, does not seem to recommend it to me as part of my good.
This suggests a different reforming definition for non-moral goodness. Suppose we say that “G is good for X” means that X would desire G if X was fully informed and ideally rational.
But this doesn’t seem quite right, either. Consider Joe, a philosopher lost in Chicago. A map would be good for Joe, but Joe wouldn’t desire a map if he was fully informed and ideally rational because then he would already know everything about the geography of Chicago.
So let’s try again. Railton suggests that “G is good for X” means that a fully informed, perfectly rational version of X would want the non-ideal X to want G in his circumstances. Joe’s fully informed and perfectly rational self would have no need for a map of Chicago, but it seems a fully informed and perfectly rational Joe would want the non-ideal Joe to have a map if the non-ideal Joe were lost in Chicago.
Thus, Railton proposes:
An individual’s good consists in what he would want himself to want, or to pursue, were he to contemplate his present situation from a standpoint fully and vividly informed about himself and his circumstances, and entirely free of cognitive error or lapses of instrumental rationality.
This is Railton’s ‘full-information analysis’ of what is non-morally good for a person.
Note also that this account leads to some predictions. As Miller puts it:
Firstly, since a person’s good is determined by what her fully informed and ideally rational self would want her to want given her actual circumstances, we would expect that, ceteris paribus, an individual will generally be a better judge of what is good for her than other people; secondly, that one’s knowledge of one’s good should increase as one’s informational base increases with experience; thirdly, since an agent’s ideal self will take into account what her non-ideal self is like as well as her actual circumstances, we would expect that people with ‘similar personal and social characteristics’ will tend, in similar circumstances, to have similar values; and fourthly, what is good for one person will tend to approach what is good for another ‘in those areas of life where individuals are most alike in other regards…
We haven’t done enough science to know if these predictions are correct, but they are certainly widely-held views.
In ways like this, Railton’s ‘full-information’ reduction of non-moral goodness can play a genuine role in explaining features of our experience, satisfying the first criterion. What about the second criterion? Railton poses the issue:
It does seem to me to capture an important feature of the concept of intrinsic value to say that what is intrinsically valuable for a person must have a connection with what he would find in some degree compelling or attractive, at least if he were rational and aware. It would be an intolerably alienated conception of someone’s good to imagine that it might fail in any such way to engage him.
Railton argues that his account of non-moral goodness does motivate people in the required way. He gives the example of Beth, a successful and happy accountant who wants to become a writer. When she has saved enough money, she quits her job to be a writer. However, what she writes fails to impress editors and publishers. Beth did not know it, but she lacks the skills and temperament needed to be a writer. But if Beth were fully informed and ideally rational, this “Beth+” would want the non-ideal Beth to lose her desire to become a writer.
In general, we can think of it like this: If you discovered an advisor who had the same basic psychological makeup and values as you do, but was perfectly rational and fully informed, you would take their advice seriously. Think of how often we say “If I could speak now to my younger self…”
So Railton says his ‘full-information’ account of non-moral goodness plays both the explanatory and normative role required for his reduction of non-moral goodness to be vindicative rather than eliminative.
Railton’s account of moral rightness
Railton uses the same strategy to offer a reforming definition of moral rightness and then argue that his account plays a genuine explanatory and normative role. He begins by considering what seems to be distinctive about moral norms:
Moral evaluation seems to be concerned most centrally with the assessment of conduct or character where the interests of more than one individual are at stake… [it] assesses actions or outcomes in a peculiar way: the interests of the strongest or most prestigious party do not always prevail, purely prudential reasons may be subordinated, and so on. More generally, moral resolutions are thought to be determined by criteria of choice that are non-indexical and in some sense comprehensive. This has led a number of philosophers to seek to capture the special character of moral evaluation by identifying a moral point of view that is impartial, but equally concerned with all those potentially affected.
…moral norms reflect a certain kind of rationality, rationality not from the point of view of any particular individual, but from what might be called a social point of view.
Railton proposes an idealized concept of social rationality in terms of
what would be rationally approved of were the interests of all potentially affected individuals counted equally under circumstances of full and vivid information.
Now Railton is ready to offer his reforming definition of ‘moral rightness’:
x is morally right if and only if x would be approved of by an ideally instrumentally rational and fully informed agent considering the question ‘How best to maximize the amount of non-moral goodness?’ from a social point of view in which the interests of all potentially affected individuals were counted equally.
In other words, moral rightness is a matter of what is instrumentally rational from an idealized social point of view.
Next, Railton must show that this account plays a genuine explanatory role and a genuine normative role.
The explanatory role
Consider Lonnie, who feels miserable and lethargic in a foreign country. He comes to desire milk, but unbeknownst to Lonnie his lethargy is due to the fact that he is dehydrated, and drinking milk will only make the problem worse. Of course, Lonnie+ would want Lonnie to desire water rather than milk, and so water but not milk is non-morally good for Lonnie (on Railton’s account of non-moral goodness). If Lonnie goes ahead and drinks the milk, we can explain his resulting misery in terms of not doing what is non-morally good for him – in terms of not doing what he woud desire himself to do if he was ideally rational and fully informed.
In the same way, Railton says we can explain things like social unrest in terms of a society not doing what is ‘morally right’ (that is, ‘instrumentally rational from an idealized social point of view’):
Just as an individual who significantly discounts some of his interests will be liable to certain sorts of dissatisfaction, so will a social arrangement – for example, a form of production, a social or political hierarchy, etc. – that departs from social rationality by significantly discounting the interests of a particular group that have a potential for dissatisfaction and unrest.
Of course, a society might believe it is doing what will best satisfy the interests of all those potentially affected, but be wrong about that. This is why the idealized (perfectly rational, fully informed) social point of view matters. Societies can cause social unrest even without meaning to do so.
The normative role
It must be noted that Railton rejects that the idea that moral obligation must always entail personal rational motivation:
On the present account rational motivation is not a precondition of moral obligation. For example, it could truthfully be said that I ought to be more generous even though greater generosity would not help me to promote my existing ends, or even to satisfy my objective interests. This could be so because what it would be morally right for me to do depends upon what is rational from a point of view that includes, but is not exhausted by, my own.
Railton illustrates this with an analogy to logical ‘oughts.’ Here is Miller:
Plausibly, I ought not to believe both a proposition P and some other proposition or propositions that entail its negation, not-P. But can it be said that simply in virtue of the truth of this ‘ought’ statement I have a reason to seek out and rectify all logical contradictions from my belief set? Think of the amount of intellectual effort this would require: I’d have to check my existing belief set for consistency (no easy matter) and ensure that every time I acquired a new belief it was consistent with my pre-existing belief set.
Logical oughtness does not entail that we always have a reason to adhere to logical oughts. Rather, it is enough that, as Railton says, “we are often concerned with whether our thinking is warranted in a sense that is more intimately connected with its truth-conduciveness than with its instrumentality to our peculiar personal goals.”
In the same way, we can account for moral facts without seeing morality as something it can’t be: “rationally compelling no matter what one’s ends.” Instead, perhaps we can “ask how we might change the ways we live so that moral conduct would more regularly be rational given the ends we actually will have.”
In the next post, I’ll examine some objections to Railton’s moral reductionism.