If I think Braelyn is a good person, I think this is so in virtue of her having certain descriptive features, like being kind or generous. And similarly, it seems, for other evaluative or normative features (Braelyn being morally required to refrain from injuriously striking Herro when he has minorly offended her, Braelyn having reason to tie her shoes, etc.). Meno-like, we might draw out and precisify our intuitions here.
(1) The ‘in virtue of’ refers to a kind of non-causal metaphysical determination or dependency (sometimes called “grounding” by philosophers). In this, it is in the same broad category as a thing being red in virtue of it being crimson (or it being crimson making it the case that it is red). Such determination or dependency does not happen across time and is not causal (e.g., it is not of the same type as my painting the object red making it red).
(2) Though we might initially take it that the determination here, like in the crimson/red case, is total. But in this we are not being true to the content of our intuitions upon relevant inquiry. To see this, let’s simplify our case and concern ourselves with Braelyn being a good person to some extent or in some particular way in virtue of her being a kind person (maybe overall Braelyn is not a good person because, despite her kindness, she is the world’s worst liar). This way, we have simply Braelyn’s kindness determining her being a good person (in some respect, to some extent). Now: if Braelyn’s kindness (or a person’s kindness generally) were to do its work only when certain background conditions obtain (as lit matches start fires only in the presence of oxygen), would our intuition change to the effect that kindness itself does not do the relevant determining (the determining being total, as in the crimson/red case)? When I think the matter over in this way, my intuition about kindness itself doing the determining does not change. Similarly when I suppose that the kindness does all of the determining work. So our intuition, precisified in an important way, is that Braelyn’s kindness either partially or fully determines her goodness.
It might help to imagine what some of the background conditions might be. Here is a possibility. Perhaps part of what goes into someone being a good person is it being appropriate for others to admire them. And perhaps such appropriateness — itself a normative property — is itself partially or fully determined by certain motivational and functional facts about human psychology. If our brains were wired more like the brains of cats or octopuses, perhaps we would not have the capacity to admire anything and so it would not be appropriate for any given person to admire Braelyn’s for being kind. On this kind of story, a whole, whole lot more than Braelyn being kind does work in determining Braelyn’s goodness!
(3) Do our intuitions here commit us to the view that, if Braelyn’s kindness only partially determines her goodness, her goodness is nevertheless fully determined by some larger set of non-normative or descriptive features or conditions? I think the answer here is ‘No’. Suppose, again, that it being appropriate to admire Braelyn for her kindness, as well as Braelyn’s being kind, is part of what goes into fully determining Braelyn being a good person. We can coherently suppose — against the particular story outlined above — that it being appropriate to admire kindness in another person (or perhaps even in oneself) is a brute feature of the world (and hence not itself determined by anything). I don’t think this view is correct, but the coherence of this scenario shows that: (a) we cannot follow out our intuitions about a case like Braelyn’s being a good person and get to the view that such broadly normative features are fully determined by non-normative or descriptive features (e.g., features of the natural world revealed by observation and science) and (b) we don’t have the same partial-or-full determination intuition that we have in the case of Braelyn’s goodness regarding all normative features (in particular, we don’t have it with regard to normative appropriateness — but maybe as well not with other normative features, especially the features that appear to be basic).
(4) We can abstract from a thesis of full determination of the normative by the descriptive (or non-normative) to a thesis of necessary covariation — i.e., a so-called “supervenience” thesis. But we cannot make a similar move from a partial determination thesis. So we cannot, on the basis of our intuitions about the dependence of (certain) normative features on (certain) descriptive features, justify a supervenience thesis (e.g., “no normative difference without descriptive difference”). Contrary to what most metaethicists believe, our intuitions about (certain) normative features being dependent on or determinated by descriptive features do not support either a full determination or a weaker necessary covariation (or supervenience) thesis regarding the relationship between the normative and the descriptive. What we have is apparent determination, it being open whether partial or full, of some (but not all) normative features by descriptive features. More work is required to determine just which normative features appear to be this way. Once this is determined, we have the relevant bits of “basic intuitive data” that a good metaethical theory should explain (or perhaps explain away with a convincing story about why our initial intuitive data here is inaccurate).