Macdonald, Scott - Ultimate Ends in Practical Reasoning: Aquinas's Aristotelian Moral Psychology and Anscombe's Fallacy

12/17/2010

The Philosophical Review, Vol 50, No 1 Jan 1991

This paper ties to defend the view that humans are rational ultimate ends-seekers against two objections from Elizabeth Anscombe. The view is Aristotle's but more refined by Aquinas, and it is, roughly, that all human activity aims at some (one) ultimate end. Anscombe's objections are given on the first page: first, human actions might have no particular end, and second, to believe that every human action aims at one ultimate end is probably an error in reasoning: a fallacious move from 'every action has some end' to 'some end is had for every action'. (pg31) Aside from the logical fallacy, the author's main task is to account for how the view that multiple intrinsic ends can be admitted to the Aristotelian conception of one ultimate end at which all human action aims. (pg32) Instead of using a reconstruction of Aristotle's argument, author prefers to use Aquinas' throughout the paper, since he believes is more expanded and explicit.

Author considers Aquinas' argument to have four main parts, each of which author goes about refining and defending. (pg34)
[A] Each human action is for the sake of some end
[B] Each human action is for the sake of some ultimate end
[C] There is some single ultimate end for the sake of which all the human actions of an individual human being are done
[D] There is some single ultimate end for the sake of which all human actions of all human beings are done

Over the course of the paper, author refines all the above claims into modified ones:

[A'] Each human action properly so-called, that is, each action resulting from deliberated willing, is for the sake of some end (pg37)
[B'] Each fully rational human action is for the sake of at least one weak ultimate end (pg45)
[C'] There is some single strong ultimate end for the sake of which all the human actions of a fully rational being are done (pg47)
[D'] All fully rational human beings share a purely formal strong ultimate end (pg61)

The paper takes each claim in order and converts it to it's altered state, and then discusses each one. First is to narrow the discussion of human action to human action that is done on a rational basis. Author utilizes a common distinction in Aquinas involving activities associated with humans (e.g. breathing) and actions done by a human qua human-- by virtue of a having reason. (pg35) Thus, Anscombe's first objection, that some human activity is just 'doing what we are doing, just because' is isolated into a different set of activities and author only wants to focus on those done for a reason. Author sketches what it means to perform a human action properly so-called: it is the intellect conceiving of a good, judging it as good, and presenting it to the will as worth doing: 'deliberated willing' (pg36). With this explication, author argues that A' is a conceptual truth talking about a subset of human activity, and author believes this subset is sufficiently large to be interesting (pg38-9). The key point here, which is used as a strategy throughout the paper, is to avoid taking A-D as being empirical generalizations about humans but instead criteria for what rational action consists in for humans (pg39-40). The second main strategy is to insist that Aquinas' arguments are laying out formal criteria, not contentful ones. Which particular ends Aquinas might believe are intrinsically the best are not implicit in A-D. (pg40-1)

In the next section author gives some terminology regarding ends and means: an end is subordinate if it is done for the sake of another end, and such a relation is transitive. (pg42) Author will later give further specifications: subordinate ends can be subordinate-as-constituent or subordinate-as-means (pg52), both of which, author claims, are distinct from the third subordination: instrumental subordination. More terminology: means-ends actions go into a 'tree'; every human action has an end'-tree, and it must contain one 'ultimate' end, meaning an end sought for itself. The argument for each tree having an ultimate end (pg43-4) is that without such an end, the will would not be moved to pursue the subordinate ends or other means. (This might also serve as a reason why an end tree cannot be infinite, without a terminus). An ultimate end (end desirable in itself) is named a 'weak ultimate end', to contrast it with a 'strong ultimate end', which is considered an end or set of ends that fully satisfy a particular human. [Of interest here is that author will talk about creating a strong ultimate end out of deliberation and conscious, rational, discarding of some weak ultimate ends (pg65)] With the argument for the need for at least one end-in-itself (weak ultimate end), author specifies B'.

The next part of the paper seeks to install plausibility for C', and author relies heavily on the formal, criterial aspects of the discussion and also some additional distinctions. For author, a strong ultimate end is considered simple to be an end which fully satisfies the rational human to every extent possible: "it completely fulfills all a human being's rational desires". (pg46-7) However, it does not have to be one thing. It can be an aggregate of many weak ultimate ends, combined together into one (disjunctive?) end. To believe that a strong ultimate end must be one thing is a 'monolithic' conception of the ultimate end, of which, author conjectures, hedonistic utilitarianism is an example. (pg48) However, there can be an aggregate case, which is a collection of weak ultimate ends into some sort of formula. (pg49-50) It is here that author introduces ends that can be ends-in-themselves but also subordinate-as-constituents of a further-- aggregated-- ultimate end. (pg51) "The possibility of an aggregate view of ultimate ends, then, requires us to distinguish three different relations of subordination among ends: instrumental subordination, subordinatino-as-a-constituent-part, and subordination-as-a-means." (pg53)

The worrisome part of Aquinas' argument for the rationality of an ultimate end is the teleological element-- that each being desires its own 'perfection'. But author believes that the case can be made that it is rational to have a strong ultimate end without such teleology. (pg54) The way author goes about this is to try to establish that the human intellect can consider 'the good' in the abstract, or the universal concept of 'the good'. Thus humans can not only consider a particular aim or goal as good-in-itself, but also form a conception of various goods as comprising the (abstract) good. (pg55) We can see how some goods conflict, how some might need weighing, or perhaps even rejecting. Thus a human can conceive of how her collection of good-in-themselves might conflict or might fit together into a coherent and structured ultimate good (using practical reason), the best life (pg56). What is also revealed here is that such an effort will uncover one single strong ultimate end (possibly aggregated), since more than one will mean incoherence or conflict-- a sort of schizophrenia (pg58). Author uses an example of Albert Schweitzer, who may have both wanted to pursue science and also music. Though he chose to pursue science, he could still recognize the pursuit of music as an end-in-itself, but he could not be said to desire it any longer. (pg57-8) Author places the use of practical reason in crafting an aggregate strong ultimate end from multiple ends-in-themselves.

Author introduces one more type of subordinate end, the subordinate-as-specification. (pg59) This subordinate is a determinate specification for a more abstract, indeterminate-ending ultimate end. Author uses 'attain political power' as the ultimate end, and 'win the mayoral race' as a subordinate-as-specification. Author then reviews the three types of ends (strong ultimate, weak ultimate, and instrumental-to-weak-ultimate) and the four subordinate relations (instrumental, -as-means, -as-constituents, -as-specifications) in a prolonged example of a triathelete. (pg60, also see appendix) The upshot of these introductions is that it gives best-life theorists much better tools in talking about the motivations and desires of their human subjects. For instance, it is acceptable to say both that one loves her children for its own sake, but also because it is a constituent of living well (pg63-4).

Author spcifies that rational human agents should all share a formal strong ultimate end-- which would be ether a monolithic or aggregate concept of the individual's most-satisfying good. (pg61-2) The trouble is in figuring out the content. Author believes that practical reasoning isn't just means-ends reasoning, but also reasoning about which ends go into the ultimate one (pg63-4). This is in contrast to philosophers who might argue that there is no rational way to decide between strong ultimate ends. (Frankfurt or Williams?)