Are moral facts in any sense reducible to natural facts?

Yes, assuming that moral facts exist, they are reducible to natural facts – or put differently, there are no irreducibly moral facts. The combination of global supervenience with intensionalism entails that moral facts are wholly reducible to natural facts, and those two premisses seem compelling. In this essay, I first briefly define terms and discuss the question’s assumption that moral facts exist, before demonstrating that moral naturalism follows directly from the conjunction of supervenience and intensionalism. After discussing and rebutting some objections to the soundness and validity of this argument, I explore two potential implications of moral naturalism. Finally, I conclude that whilst moral facts are indeed reducible to natural facts, this does not mean that the former can be deduced from the latter, and furthermore creates some difficulties in accounting for the normative force of morality.

• Definition of moral facts and natural facts; assumptions of the question

  • Natural facts: true descriptive propositions about the state of the physical world, of the sort that can be discovered through scientific inquiry. In this essay, I shall assume that there are no supernatural facts, but the argument presented below can be reformulated to account for them and show that moral facts are reducible to a combination of natural & supernatural facts.

  • Moral facts: true normative propositions about ethics. Some examples of statements which may be moral facts are “Liars always deserve our blame”, and “More serotonin in the brain is good”, and “The policeman’s arresting the perpetrator of the robbery that occurred on Broad Street yesterday evening was right.”.

  • Reducible: we can say that a certain moral fact $M$ (e.g. “The policeman’s arresting the perpetrator of the robbery that occurred on Broad Street yesterday evening was right.”) is reducible to a natural fact if the predicate in $M$ (“was right”) necessarily picks out exactly the same objects as some predicate about purely descriptive properties does.

  • The phrasing of the question assumes that moral facts of this sort exist – that is, it assumes the falsity of both non-cognitivism (which asserts that moral judgements do not express beliefs and so are not truth-apt) and error theory (which argues that moral judgements are propositional statements but are never true). An alternative way of phrasing the question would be to ask “Are there any irreducibly moral facts?”, to which the answer would be no, since those two theories claim there are no moral facts at all and (as I will show), even if there are moral facts they must be reducible to natural ones.

• Present Jackson’s argument

  • Supervenience: moral facts supervene on natural facts, meaning that there can be no change in moral facts without some natural facts changing.

    • This seems intuitive – it would be strange if, e.g., a particular act of murder were impermissible in this world but somehow permissible in another where all the natural facts of the matter were identical.

  • Intensionalism: if predicate P and predicate Q necessarily co-extend, then they ascribe the same property.

  • Consider the set of all actions which are right, $\mathbf{A}$, with size $k$. Each action $A_{n} \in \mathbf{A}$ has a certain set of descriptive properties $D_{n}$ – for example, the time and place of its doing, by whom it was done, the state of a village across the world from where it was done, etc. We can conjoin each of these properties to form a predicate $P_{n}$ which action $A_{n}$ satisfies: “has descriptive properties $D_{n}^{1}$, $D_{n}^{2}$ , … and $D_{n}^{m}$”.

  • Ranging over the set of predicates ${ P_{1},P_{2},\ldots,\ P_{k}}$, we can now construct a disjunctive predicate $P^{*}$ which will be satisfied by every right action: “satisfies $P_{1}$ or $P_{2}$ or … or $P_{k}$”.

  • By construction, if $a$ is right then it satisfies $P^{*}$. And by Supervenience, satisfying $P^{*}$ is sufficient for an action $a$ to be right, because it is not possible for $a$ to have all the descriptive properties of some right action (which it must, to satisfy $P^{*}$) without itself being right.

  • So, the predicate $P^{*}$ necessarily co-extends with the predicate “is right”, and therefore by Intensionalism the property of being right is the same as the property of satisfying $P^{*}$.

  • This shows that moral facts are reducible to natural facts in the sense that for any posited moral property (e.g. rightness, wrongness, goodness, worthiness, etc), there exists some natural property which it is identical to.

• Objections to soundness and validity

  • Criticism of intensionalism: perhaps the predicates P and Q can necessarily co-extend yet not ascribe the same property. For example, “is a three-angled polygon” and “is a three-sided polygon” co-extend but on first glance seem to ascribing different properties. If we examine this intuition more carefully, though, it fades – e.g. is there a third property ascribed by the predicate “is a triangle”, and a fourth ascribed by “is a closed figure where the sum of the number of angles and the number of sides is 6”, and so on?

  • Criticism of infinite disjunction: perhaps $P^{*}$ does not ascribe a genuine descriptive property, because it is an infinite disjunction. But consider the property of being a colour, or of being bald. Both of these might be formulated as infinite disjunctions, and we would have no qualms about their legitimacy as descriptive properties.

• Implications of moral naturalism

  • We have shown that moral facts are reducible to natural facts. However, this does not imply that by observing natural facts we can deduce moral facts. In order to arrive at $P^{*}$ we need to know all the descriptive properties of each right action, which in turn requires having access to us the set of right actions $\mathbf{A}$. So constructing $P^{*}$ not only demands a granularity of knowledge about the world which is impossible to attain in practice, but also (and more crucially) requires prior knowledge about the very moral facts we might hope to deduce.

  • Moral naturalism does create a challenge for how we explain the force of moral reasons, though. It seems like natural facts are not the sort of objects which can give rise to normative obligations / reasons to act – yet we would like morality to have some kind of practical authority. Perhaps we ought not to worry so much about this, though. If a Humean view of rationality is correct (i.e. one has prudential reason to do just the things that promote one’s own desire satisfaction), then prudential reasons must be grounded in natural facts (since desire satisfaction is a property of the physical world). This seems entirely plausible. Similarly, what is implausible about moral reasons also being grounded in natural facts?

Bibliography

Jackson, F. (1998). From Metaphysics to Ethics, Oxford: Oxford University Press. (Chapter 5). https://doi.org/10.1093/0198250614.003.0005

Streumer, B. (2008). Are There Irreducibly Normative Properties? Australasian Journal of Philosophy, 86(4), 537–561. https://doi.org/10.1080/00048400802215349

Enoch, D., 2011. Taking morality seriously: a defense of robust realism. Oxford University Press, Oxford. (Chapter 1). https://doi.org/10.1093/acprof:oso/9780199579969.003.0001