Knut Olav Skarsaune (Inland Norway University of Applied Sciences), "Metaethics as Conceptual Engineering"
Forthcoming in Analytic Philosophy
By Knut Olav Skarsaune
Plus or minus twelve years ago, I wrote a PhD dissertation in Metaethics. Wanting to do it properly, I read up on the literature, and what I read puzzled me. On the one hand, almost everybody seemed to agree that metaethical theories are supposed to answer a set of descriptive, present tense questions, about the semantics, psychology, metaphysics and epistemology of moral/ethical/normative judgments. But on the other hand, many of the arguments in the literature seemed to have less to do with what our judgments are like, in these respects, than with what they should be like. Take for example metaphysical worries about nonnaturalism, or Sharon Street’s Darwinian Dilemma, or David Enoch’s argument from disagreement. Likewise when people I met at conferences explained why they liked expressivism, say, or naturalism. It seemed to me that they gave interesting reasons to speak an expressivist or naturalist language going forward, but I couldn’t see how those reasons would support the claim that ordinary speakers of English are already doing so.
To be sure, there were people in the literature who put their theories forward as (gentle) revisions of ordinary thought and talk, such as David Lewis, Peter Railton, and (Wise Choices, Apt Feelings) Allan Gibbard. But that only compounded the problem, because nobody seemed to care whether a given theory was put forward in a descriptive or revisionary mode. Everybody seemed to treat descriptive and revisionary views as competing solutions to the same underlying philosophical problem, which (I thought) couldn’t be given by the descriptive questions, since the revisionary views don’t claim to answer those questions.
So I holed up in my PhD student lair and pondered this problem. I concluded that the main question in metaethics is indeed what kind of normative judgments to make, going forward. But that raises a puzzle. If we evaluate metaethical theories as policies about what kinds of judgments to make, then it is not clear why they are in competition with each other. Why not make expressivist judgments and constructivist judgments and nonnaturalist judgments, and so on? This is not an issue on the traditional, descriptive conception of metaethics (as there is presumably one best analysis of ordinary normative thought and talk), nor even on the gently revisionist approach (as there is presumably a smallest revision that lacks whatever feature of ordinary thought and talk that needs to be done away with).
The answer I propose is that metaethical theories are competing designs for a limited tract of psychological real estate. Normative judgments play distinctive action-and-attitude-supervising roles, and you can’t simultaneously use different sorts of judgments in the same action-and-attitude-supervising role. My advisors liked this idea, so I included it in my dissertation.
The road from dissertation chapter to article has been long and winding. I was warned (by Stephen Schiffer at a job market practice session): “You know what philosophers are like. They HATE being told that they’re confused about what they’re doing. You’ll get mountains of shit presenting this.” He was on to something, judging by peer reviewer comments. They did hate it. They thought I was too quick to conclude that Street’s Darwinian Dilemma or Enoch’s argument from disagreement don’t work as intended if we’re asking descriptive questions, too quick to dismiss appeals to the Principle of Charity, and/or too quick and ambitious in general.
So the paper has slowly made its way through the circles of peer review hell. It has evolved and (hopefully) matured in the process, even though the basic view is the same. Matti Eklund took an interest in it, and readers familiar with his work will notice his influence many places. David Plunkett and Tristram McPherson helped me see that I don’t need to pick fights unnecessarily: since what I care about is developing the positive account, I can just say that I am describing one possible metaethical “project” and leave others to go their merry way with their more traditional projects.
And of course, “conceptual engineering” has become a popular topic in its own right. Back then I hadn’t heard of it, but when I did, it was clear that I had stumbled onto a special case of a more general idea. The metaethical case is special, though. Suppose we say that to take a conceptual engineering approach to the philosophy of X is to ask, not what our concept of X is (or what X-ness is), but what concept of X we should be using (or which property in the X-vicinity is most worthy of study). In the case of metaethics, the X includes should and most worthy themselves, making the engineering questions circular. My answer to that worry is long and involved, so I won’t try to summarize it here. You can find it in the somewhat pretentious, twelve years in the making, still dear to my heart, finally off my chest, good riddance, paper, here.
Marcus, I would be very interested in Knut's "meta"-reaction to the piece linked below, only in part because it took about 4 years to write, and I was wondering if I will have another 8 to go: "Why we should : an introduction by memoir into the implications of the Egalitarian Revolution of the Paleolithic, or, Anyone for cake?" Abstract attempted thus: It argues that normative impulses to "should" are worldbuilding and thus give survival value to people who are thus organised by the discussion that they work in, compared to those who are not so shouldly organised. Here "discussion" as the otherside of the coin to "argument". This moral urge or worldbuilding urge is more important "in itself" than any detail of morality which are thus mere outcomes of social process, and thus can vary over time & geography, like any arbitrary fashion. https://www.academia.edu/40978261/Why_we_should_an_introduction_by_memoir_into_the_implications_of_the_Egalitarian_Revolution_of_the_Paleolithic_or_Anyone_for_cake
The imperative is a drive, the moral urge itself has no morality, even if we would like it too (moral imperatives etc). I feel this position falls outside the known categories used in metaethics, but I am not a scholar but a mere librarian.