For those interested, the current draft of my diss. proposal is here (.pdf). Over the next few weeks, I want to tease out the last remaining threads of argument that I haven't quite developed, namely those that draw out implications from my model for reasons-explanations of action. Today I'll be talking about reasons.
As discussed previously here, here, and here, and, of course, in more depth in the proposal, a successful reasons-explanation of action has to accomplish three things:
- It must account for the connection between reasons and action in the agent;
- It must give reasons that the agent genuinely finds objectively worthy; and
- It must fit the action within the pattern implied by (1) and (2).
An analysis of (2) thus comes for free as the means (action) is connected to the ends (reasons) by the former being an way of achieving the value of the latter. Thus, reasons are, always, valuable. For them to be objectively valuable, though, it has to be true that the ends are not purposes (which are subjective, in my view) but goals: that is, things in the world, such as states of affairs, which one aims one's action towards as a means.
Finally, the pattern is what I've characterized as a pseudopattern -- that is, a class of patterns -- called efficacious attainment of value (EAV). As long as the means is appropriate to the end, and the end is objectively valuable, then the pattern of the action (means) and reasons (end) counts as EAV.
The implications I want to be able to draw out of this model relate to a number of longstanding problems in various areas of philosophy. The first is the idea that reasons for action can't be psychological and must instead be real (i.e., goals, not purposes). The reason for this turns on the requirement of objective value. Approximately, some ψ is subjectively valuable if it seems objectively valuable to me; some ψ is objectively valuable if it actually is valuable. (These can, of course, coincide; I can correctly take the world to be as it seems to be.) Since a successful reasons-explanation of action has to show us that the reasons were objectively valuable, it has to be impossible that the reasons fail to exhibit objective value. (This sounds a little trite, I know. But I'm going somewhere with it.) But, psychological states can fail to exhibit objective value. This leaves us with two possibilities: either (a) the psychological states that fail to exhibit objective value are not really reasons, or (b) no psychological states are reasons. If (a), then the psychological states that are reasons are either (i) those that correctly track objective value or (ii) those that actually instantiate objective value. (ii) is a non-starter, I think, for it presumes a distinction within our psychological states that needlessly clutters the ontology, i.e., there would have to be a whole set of psychological states that actually instantive objective value, and a whole other set, just as complete, that did not. If (i), however, then there doesn't seem to be any need for the psychological states. That is, they seem to be playing a practical role analogous to the epistemological role played by good ol' sense-data, of wonderfully convenient intermediaries between the entirely separate spheres of mind and world. Since there's no good grounds to maintain this harsh separation, it follows that there's no need for these intermediaries. So, we may as well abandon (a) and go, instead, to (b).
I also want to claim that reasons are complex, not simple, and plural, i.e., that when I φ, I do so for a constellation of reasons (complex, not simple) and that any given reason can favour multiple actions, and any action can be favoured by multiple reasons (plural). I think this follows pretty straightforwardly. A given means can be appropriate to multiple ends; hence, one action can be favoured by many different reasons. Similarly, it's entirely possible that a given end can be attained efficaciously by multiple means; hence, one reason can favour multiple actions. Finally, there's no grounds to suspect that when I φ, I must do so for exactly one goal. Consider that any goal I achieve could, in principle, be part of a long series of goals (call this a "project"); thus, to achieve the first goal is to, at least in part, achieve the super-goal of the project.
Next week, I'll discuss arguments regarding the implications for agency and reasoning. (Which will be really fun, as I know almost nothing about the relevant literature. Woo-hoo!)