Meta-ethics: Jackson’s Moral Functionalism

by Luke Muehlhauser on April 23, 2011 in Ethics

Previously, I explained Cornell realism (a variety of naturalistic moral non-reductionism) and Railton’s moral reductionism (1, 2). Now, I explain Frank Jackson’s version of moral reductionism, called ‘moral functionalism’, by summarizing some of chapter 9 in Miller’s Introduction to Contemporary Metaethics.

Recall that Railton’s moral reductionism offers reforming definitions for moral terms that are to be evaluated on a posteriori grounds. He did not propose an analytic connection between moral predicates and naturalistic predicates.

In contrast, Frank Jackson’s moral reductionism does claim to discern analytic connections between moral predicates and non-moral naturalistic predicates. It does so by way of a ‘network-style analysis’ of the content of moral judgments.

Network-style analysis is a form of conceptual analysis. Miller explains:

Suppose, for example, that we are in the business of giving an analysis of the content of judgements about colours: our judgements about red, blue, orange, and so on. [Michael] Smith suggests (i) that there all sorts of platitudes about colours, and (ii) that ‘an analysis of a concept is successful just in case it gives us knowledge of all and only the platitudes [relevant to mastery of that concept]‘.

A platitude is relevant to the mastery of a term when it captures the inferential and judgmental dispositions vis-à-vis the term of those who have mastery of it…

In the case of colour terms, for example, the platitudes will include: (a) platitudes about the way colour terms are learned by linking them with features of the world, for example, “To teach someone what the word for red means, say the word for red while showing them some red things”; (b) platitudes which connect colour experiences with colours, for example, “Redness causes us, under certaincircumstances, to have experiences of redness”; (c) corrective platitudes, for example “To see something’s colour, look at it in daylight”;and (d) platitudes about similarity relations between colours, for example, “Red is more similar to orange than to yellow”, “Orange is more similar to yellow than to blue”, and so on.

A proposed analysis of the content of colour judgements will thus be inadequate to the extent that it fails to capture this set of platitudes, the set of judgements and inferences licensed by our mastery of colour concepts themselves. And likewise for any proposed analysis of the content of moral judgements: just as there are platitudes surrounding our use of colour terms there are also platitudes surrounding our use of moral terms, and any proposed account of moral judgement will be inadequate insofar as it fails to capture these platitudes.

Smith has proposed four platitudes surrounding our use of moral terms:

  1. Practicality (e.g. “If someone judges my A-ing to be right then ceteris paribus I will be disposed to A.”)
  2. Objectivity (e.g. “When I say that A-ing is right and you say that A-ing is not right, then at least one of us must be incorrect.”)
  3. Supervenience (e.g. “Acts with the same natural features must have the same moral features as well.”)
  4. Substance (e.g. “Right acts express equal concern in some way”)
  5. Procedure (e.g. “We try to figure out what is right by employing the method of reflective equilibrium.”)

A network-style analysis of a discourse attempts to capture the platitudes of that discourse. Let’s illustrate how this works by returning to color discourse.

First, we rewrite the platitudes of the discourse in property name style. So “red is more similar to orange than to blue” becomes “the property of being red is more similar to the property of being orange than it is to the property of being blue.”

We do this for all colors, and then we conjoin the totality of platitudes as a relational predicate ‘T’ that is true of all the color properties: T[r g b ...], where ‘r’ is the property of being red, ‘g’ is the property of being green, and so on.

Next, we replace the property names with free variables so that we get T[x y z ...].

For the rest, I’ll let Miller explain:

So, if the platitudes are ‘the property of redness causes us, under certain conditions, to have experiences of the property of redness’ and ‘the property of redness is more similar to the property of orangeness than the property of blueness’ and . . ., then T(x, y, z) will be ‘(x causes us, under certain conditions, to have experiences of x) & (x is more similar to y than z) &. . . ‘. Then, if there actually are colours, there is a unique set of properties related to the world and to each other in exactly the way that the conjunction of platitudes says there are. In other words, if there actually are colours, then it is true that:

Then the property of being red can be defined in the following manner: the property of being red is the x such that:

This is a reductive analysis of redness: it defines it in purely non-colour vocabulary, since no colour vocabulary appears on the right-hand side of the definition.

Against such network-style analyses, Smith has proposed the Permutation Problem. Miller offers a rebuttal, but I won’t consider it here.

How does Jackson offer a network-style analysis of moral discourse? I won’t cover the details here, but let me switch from Miller to Lenman, who sums things up: “Jackson seeks to describe a reductive analysis of ethical terms that understands them as picking out the properties that play a certain role in the conceptual network determined by mature folk morality.” In other words, morality is defined in terms of what folk morality would be if it was “mature,” whatever that means.

Lenman then summarizes an objection from Stephen Yablo:

The trouble is that this looks unpromising as a reductive analysis as there is a term in it that appears decidedly evaluative in character, “mature”. “Mature” after all had better not just apply to any old terminal point our ethical development may happen to take us to but only a terminus we could arrive at by good reasonable ethical discussion and argument. One possible answer in the spirit of the wider theory might be to say that “mature” picks out whatever plays the “mature” role in mature folk morality. But that seems circular. After all maturity in current folk morality is plausibly pretty contested. There are many ways morality could develop to a more settled state and we are liable to differ about which of them we count as maturation and which we would describe less favourably. So if I say mature folk morality is mature by the descriptive standard specified by the best candidates for mature folk morality a central question seems to be getting begged. Or, as we might say, echoing Moore, left open.

For the details of Jackson’s program, see his 1998 book From Metaphysics to Ethics. For a more recent discussion, see Brad Streumer’s “Are normative properties descriptive properties?

Previous post:

Next post:

{ 3 comments… read them below or add one }

John D April 25, 2011 at 3:57 am

Felt strange to see this without any comments so far…

Just wanted to say I’m really enjoying this series. I’ve just been writing about WLC’s views on morality over on my blog recently, and I thought it was interesting how he only considers non-natural realism as a viable alternative to theistic metaethics. Now I’m not as opposed to non-naturalism as I reckon you are (I’m probably even more of a metaethical pluralist) but it’s still nice to see the more recondite naturalistic theories getting a decent airing.

Sorry I don’t have anything more substantive to add. Would have liked some more on Miller’s responses alright since I don’t have a copy of his book. Must check out Streumer’s criticisms when I get a chance.


Garren April 25, 2011 at 8:41 am

I found reading about functionalism in general helpful:

‘Functionalism is the doctrine that what makes something a thought, desire, pain (or any other type of mental state) depends not on its internal constitution, but solely on its function, or the role it plays, in the cognitive system of which it is a part.’

So functionalism about moral terms seeks not so much to explain what ‘goodness,’ ‘justice,’ ‘fairness,’ etc. mean as individual terms as explain all terms in the evaluative and normative domain by how they work together in a causal system. At least, this is what I’m understanding after an hour of reading around a bit.

Can’t say I’m a big fan of this approach overall, though there does seem to be some value in using it to account for the wide variety of moral terms once some key terms are understood in a less holistic way.


Luke Muehlhauser April 25, 2011 at 11:05 am

John D,

I’m a metaethical pluralist, too, but yeah: undetectable supernatural and non-natural properties don’t fit in my pluralism.


Leave a Comment