The word 'intention'--not to be confused with 'intension'--and its associated "notion" is at the center of a number of confusions among contemporary philosophers. I am writing to address some issues raised in discussion and not necessarily to come to a solution or answer particular literature.
I take it that we have a common-sense notion of intention, and that this notion, like many in the quotidian repetoire, is many-faceted, complex, sometimes vague, sometimes ambiguous, and thoroughly dim to parsing. I say dim because it is not, I think, either opaque or "basic" (whatever that might mean). I want only to pick a thread or two from that web of concepts.
One supposes that a natural occasion on which to say of someone X that they intended to A is when, upon looking at the action performed, we can concoct a story that will give X a reason
to A. We can do here without a lengthy exposition of what counts as a reason. Let us suppose that what falls under the concept is anything one could be described as wanting to bring about, or some past event(s) that give us an intuitive understanding of a motivation. We shall exclude all of those possibilities that leave such a connection unintelligible. This has the happy effect of excluding the insane from necessarily being held accountable for their reasons in action (nowadays a commonplace notion, though an issue not to be discussed further here); it alsol allows us to elide the issue of motivation, which is not itself the topic under discussion here. Philosophical accounts of that
(the discussion of Hume's supposedly "hydraulic" theory) are supposed to be part of the explanatory structure that is, essentially, an ontologic discovery of the nature of the events we attribute to the will.
We are rather concerned with what constitutes intention--is it an act, a property, etc.?--and the conditions under which something may be intentional. I take it that, on the sense under discussion, the things that may count as intentional are actions
. I will leave the notion of action itself somewhat vague, with the following outline: actions are discrete events--regions of space-time--in which a process appears to be initiated by an entity such as a plant, animal, spirit, robot, etc.; and where such appearance is due to our inability under ordinary circumstances to construct any sort of detailed picture of the causal story
we must admit can be told about what came immediately prior to the event. Reasons, then, take part in a sort of causal story, but not a physicalistic one, as in the above outline.
Now as to the idea that we are able to "concoct a story," as was said earlier, about a reason for an action. If this is to be a condition for an action's being intentional, then we must find out what sort of story is the one we "have in mind." A Kantian might propose the following. The story we are referring to amounts to the construction of a hypothetical imperative under which a reason and an act are tied together. So, in order to obtain B (which we stipulate X already takes as a reason; do not be distracted yet) X does A. The imperative allows us to attribute a willing of the means (A) to the end (B) to X, and intention must then amount to just having a reason
A follower of Wittgenstein and Anscombe might propose this alternative. We must be able, in theory only, to elicit from the agent a response to the question "why did you do A?" other than
(i) "I didn't mean to do that" or (ii) "I didn't realize what I was doing." "No reason" is an acceptable answer on this view; X may say that she "meant to do A" or the like, or even "I don't know" (think of an adolescent), without providing any answer to do with an antecedent desire, want, aim, or reason. (For neither proposal does this imply we should supply one, by fiat, from the subconscious; that would lead us to difficulties with circularity worse than the ones already skirted above.)
The major difficulty between these views (ignoring for now such things as the larger implications of phrasing things in the language of Kant's Groundwork
) is whether an action may be said to be intentional when, all concoctions aside, X refuses to admit, let alone enunciate, an antecedent reason. And I see no reason why we must a priori
define intentional actions as ones undertaken for reasons. As since I have already appealed to ordinary intuitions in order to get the discussion's topic in view, it would be foolhardy to appeal to them in order to solve the present trouble.
I do not want to attempt here to adjudicate between various theories of action in terms of Anscombe's "descriptions" account, though I must admit that I am attracted to it as an account of what is said to be intentional
under normal usage. However I think that the present space is not the appropriate place to take up that fight, as what we are concerned withnow, action, already includes that argument while still being identical with the class of things that may be "intentional."
Let us try another tack. Consider the possibility that thinking
is active. That is, adding two and two (and coming up with a result) is an act on my part, even if I do not articulate any particular thoughts having to do with or relating to my doing so other than the adding itself. (Perhaps one's thoughts wander, and then one thinks "2+2=5 is wrong" and immediately afterward "2+2=4" and then "Orwell had some interesting things to day about language." I am supposing these are discrete thoughts, but even if some theory puts the lie to it the example of the solo thought is, I think, solid enough to stand signage toward the discussion here.) Leaving what thoughts are to intuition, I still assert that each thought is an act. Not all thoughts, however, are intentional. Thus the literary "thoughts unbidden" and the song that is stuck in one's head amount to malfunctions like muscle spasms or perhaps involuntary muscle movements--they can be controlled, but onl by the engagement of the will. That is, only when the apparatus of the mind that does
these processes is used intentionally
Unfortunately now we have only moved the difficulty. This attempt to cash out intention in terms of some feature of mental processes had the advantage of deciding decisively for the Kantians. It has the drawback of leaving the notion of intention to be subsumed under a metaphysical account of the will, and that
will be a source of contention even more difficult to navigate than the present discussion has been.
I do not have any conclusions to offer after this admittedly unsatisfying survey of the landscape. I would point to a pair of thoughts that may prove fruitful. The first is the suggestion that intention is an empty concept applied post hoc
to events because of our natural mechanism of creating a narrative--our personal history and identity. Taken seriously, intention is just what we attribute to agents in virtue of their making intelligible actions in the world. Compare that to the Humean account of the natural laws as making intelligible the patterns of impressions we receive. The second thought is that it is unsatisfactory to appeal to willing or the like in order to have an acceptable account of intention. After all, what was the aim of the above discussion but to explicate the concept in some small respect--and how quickly that turned into an ontological snipe hunt! One wonders how it is possible to stay within the realm of ordinary usage without resorting to the actions of the soul, as perhaps Davidson would allow us to do, and how Descartes would insist we do--for these things merely explain intention's being part of our world and not its nature.