Plausibly, in certain sorts of cases X having reason to A (where A is either the performance of an action or the having of an attitude), the specific or contributory valence of A-ing for X is not a function of X’s A-ing realizing or promoting some outcome O. Rather it is a function of X’s A-ing being an appropriate or fitting response to some condition, feature or circumstance C. I’ll call these reasons appropriate-response type reasons (with the larger, catch-all category being “non-teleological” reasons). We can, for such reasons, rig up an outcome that plausibly might explain the normative valence. For example, we could say that [X-A-ing-when-C-is-present] or [X-A-ing-when-X-registers-that-C-is-present] is agent-neutrally valuable or beneficial to X, thus making it such that X has reason to A in virtue of realizing or promoting one of these valuable or beneficial outcomes. However, there is this element that is not captured by the “consequentializing” strategy: the valence (for X) is supposed to attach to X’s A-ing — but not due to X’s A-ing being related (including identical to) some condition (the outcome) that itself has some normative or evaluative valence. So it seems plausible that this sort of case of X having reason to A is non-teleological and, in this, not fully or truly “consequentializable” (unless we weaken the conditions for “consequentializability” such that the “original” normative character of these sorts of reasons need not be captured in the move).
Perhaps no practical reasons (X having reason to A) are like this, in which case, all practical reasons are teleological (leaving, perhaps, only reasons of appropriate or fitting attitude in the category of non-teleological reasons). But I suspect this is not the case. I suspect that deontological practical reasons — the reasons of moral requirement — are reasons of appropriate or fitting response to particular sorts of condition or circumstance. In which case such reasons (again, more precisely, cases of X having reason to A) are similar to reasons of fitting attitude (such as reasons to blame, admire, etc.). To have a name, call these reasons (non-teleological reasons) of appropriate response.
It is important that these reasons are (or would be) actual reasons. They are not overall normative valences (e.g., the ought-valence or the has-sufficient-reason valence), even if, in many contexts, they carry the day and yield corresponding overall valences and or verdicts on overall ranking, they need not (if it is appropriate for X to respond to C by A-ing, then perhaps typically X ought or is required to respond to C by A-ing). To some degree, such reasons of appropriate response compete with other considerations (other reasons of appropriate response, purposes of instrumental pursuit or teleological reasons) and hence are not only specific (to the condition C) but also at least potentially contributory without being decisive.
I suspect that recognizing this class of reasons is essential to getting the right explanation for the distinctive features of normative, including moral, requirement (and for other fitting response, such as fitting attitudes). Though we might, as Portmore does, with a lot of work, explain the same features via some very fancy “consequentializing” moves (treating all reasons as teleological, albeit some in a very complicated way), I don’t think such moves do the explanatory work very well — or through things that we have independent reason to think are true (independently, we should think that apparent reasons of appropriate response are just that, not extra-tricky teleological reasons). And, as noted, the independence of such valence over the valence of any outcomes to be realized or promoted is simply not captured by the consequentializing move.
Portmore offers this reason for consequentializing all of the practical reasons: all practical reasons are explained by (not merely associated with) reasons to desire the overall world-outcomes associated with each option in any practical choice.
But this idea is false. The two normative items here (reasons for action, reasons to desire the world in which the action is taken) are always explained by some third normative element. Sometimes (in the teleological reasons case), both of the things are caused by X having most reason to intrinsically desire or value various items that are at stake (and X’s A-ing doing an overall better job at “maximizing utility” than her other options); other times (in the appropriate-response reasons case), both of the things are explained by it being appropriate for X to A as a response to C (and by other reasons X has against A-ing not outweighing). So Portmore’s correlation between X’s reasons to A (or X having most reason to A) and X’s reasons to desire the world in which she A’s (or X having most reason to desire this option-associated outcome) is not explanatory — in either direction, either of the two main cases. And so I’m doubtful that there are very good reasons (even high-level theoretical ones) to consequentialize non-teleological, appropriate-response-type practical reasons (or appropriate-response-type reasons generally).