Writing at page 174 of his book, Our Knowledge of Fact and Value, Hall makes the point that
Our moral and aesthetic judgments are not about our emotions; they are renditions of our emotions, having the same object as our emotions have, saying (in part) exactly what our emotions say. They are expressions of emotions.
Here, it seems to me, is the crux of Hall's claims about value and our knowledge of it. He seems to want to say that valuing is just the expression of our emotions and that emotion language is like descriptive language (contra Hume) in that both are directed at objects, both refer and thus both types of language say something that is determinable from the facts. But there are certain differences in the logic he notes, for descriptive statements are true or false based on the adequacy with which they depict the facts which our perceptions capture for us while:
Evaluative sentences in conventional language receive whatever probability they have from their truthfulness to emotions.
He adds
I need not remark that the 'truthfulness' to which I refer is not correctness of depiction but faithfulness of translation.
In this way, Hall sets out to explode the Humean account which severs fact from value. After indulging in some psychologizing about why we tend to rely on linguistic usages which downplay the emotional content of moral claims (we often seem to want to couch moral claims in purely rational or evidentiary statements), Hall proceeds to suggest that, since he has stated that all of our linguistic usages, including the merely descriptive, carry emotional content (either favorable, unfavorable or indifferent), there must be some means of differentiating between the claims which our competing emotions about the things we can depict have on us:
Some method of determining relative probabilities must be found if we are to accept emotions as furnishing the basis of our knowledge of value.
This strikes me as friendly to a point I've made elsewhere in these discussions, i.e., that moral valuing gets its special character, which seems to confer on it an authority over and above that which we accord to the other kinds of valuing we do, from its concern to assess actions in full. In so doing, it has a controlling influence on all our other assertions of value. What I mean by this is that actions are expressions of value judgments and can be assessed, either as the physical events needed to procure a desired object/objective (assessed for their instrumental value) or as a proxy for the desired object/objective (the value of the thing pursued conferring value on the action which is implemented to pursue it). But this doesn't get at every aspect of the action being assessed.
Insofar as an action is also an expression of an acting agent, this is an incomplete picture of an action. What the agent is thinking, wanting, planning, etc. all matter in the assessment of the act. And this is just to say that, when we want to look at an action in its entirety, what we have to take into account, besides the role and relative efficiency of the action in doing what it's supposed to do, is that which underlies it, what it's intended to do. This means that the action cannot be fully assessed without consideration of its intentions and intentions are peculiar kinds of things. They are not particular mental features, I would argue, but just whatever is going on in the agent relative to the action when the agent decides to act, and does act.
Intentions are the various considerations of the agent, which he or she attends to and relies on, in initiating the action. A full assessment of the action then cannot occur without inclusion of an assessment of the intentions.
But, as an indefinite thing of this type, the intention looks like nothing more than a momentary snapshot of the state of mind of the acting agent. So, if valuing actions in full means also valuing intentions, then it must also mean valuing the agents who intend.
To the extent that the idea of moral valuation hinges on this notion of assessing actions in full, however this is accomplished (and I have my own views on this, described elsewhere in these discussions), it also provides the basis for according to moral claims their apparent extra authority (i.e., a basis for why we think that a good or bad moral judgment supersedes and should control other value judgments that we make). Here Hall's move to discuss finding "some method of determining relative probabilities" with regard to the truth, or lack thereof, of competing value claims promises to be interesting.
As soon as I get a chance I hope to move on and finish the book. Any effort to explain moral valuing, and valuing in general, in a way that preserves its capacity to be rationally discussed and argued about in the way that seems intuitively right, is always going to be worth looking at I think.