Velleman’s chapter “What Good is a Will?” describes many of the challenges that action theorists grapple with, including how some of Anscombe’s ideas can both solve and create more problems for theories like Bratman’s. He speaks about the nature of the will and what role it serves in the philosophy of action, and while I agree with most of Velleman’s writing and his defense of Anscombe, there is one particular topic he discusses where I believe he does not adequately paint the full picture of intention. In particular, for cases in which an agent doesn’t have a reliable connection to practical knowledge of the outcome, I find that Anscombe’s explanations stray from what I would consider common sense intuition about what it means to do something intentionally. I agree that non-observational knowledge is present with intention in the usual case; however, by defining intention in terms of this knowledge, I believe we lose some of the nuances of what intention is. In this paper, I will argue how it is possible to not have reliable non-observational knowledge about an end result while still achieving that result intentionally. In doing so, I will demonstrate that there are cases where intentionality strays from practical knowledge and that Anscombe’s theory should be qualified to allow for these exceptions.
Assume you are attempting to hit a bullseye on a dartboard that is 7’9¼” away (per regulation, of course). The question we have to answer is: if you do end up hitting the bullseye, did you do it intentionally? According to Anscombe, it depends on how skillful you are with darts, or in other words, how reliable your attempt can be made into knowledge that you indeed hit the bullseye. If you are immensely skilled with darts, then you have “a reliable way of hitting the bull’s-eye, in which case your intention to hit it is reliably connected to that result and will probably amount to knowledge” (Velleman). This would constitute non-observational practical knowledge of hitting the bullseye, so Anscombe would claim that you hit the bullseye intentionally. However, if you have very little skill with playing darts, then your attempt to hit the bullseye cannot be reliably connected to the event of hitting the bullseye, so your hitting the bullseye would only come to be known through observation, hence you did not do it intentionally according to Anscombe. If the probability of success is very close to one, then it is intentional, whereas if your probability of success is close to zero, then hitting the bullseye amounts to the same luck involved in winning the lottery, making it not intentional.
Anscombe does not attempt to explain the sort of Sorites paradox that emerges from making this claim—where in this spectrum of probability would we even place the threshold that defines intentionality—but this aspect is not important since it is inessential for Anscombe’s claim about intentionality and for my qualification of her claim. To avoid this muddy middle ground, I will take the most extreme case, someone who is very unskilled at darts (the kind of friend you would need to leave the room for before their turn), and show how even if they somehow manage to hit the bullseye, they did so intentionally.
Consider the following premises: (1) You do something intentionally if and only if you don’t do it accidentally; and (2) If you aim to do something and achieve that aim, you did not do it accidentally. These lead to the conclusion that (3) if you aim to do something and achieve the desired outcome, even if succeeding in that aim is highly improbable, you do it intentionally—which is the argument that I intend to defend. I introduce this adverb accidentally—a term that doesn’t seem to be very prevalent in the philosophy of action literature—not to obscure the argument, but rather to open the discourse up to an entirely different area of intuition that has the potential to aid us in navigating these difficult questions. The argument, as presented, seems valid, so in order to defend statement 3, I will argue why I believe the premises must be true.
I will prove this biconditional by proving both directions of the implication. The forward direction is trivial, and I presume Anscombe would agree with me. If you do some action intentionally, then it would be incorrect to say that you also did it accidentally; therefore, you must have not done it accidentally. The other direction is slightly more nuanced, but I still think holds. If you did some action not by accident, then it seems like our intuition would say that you did it intentionally. There is some room here for actions that would be classified as neither intentional nor accidental, but I’m not sure what would even fall into that category, so I will assume that every action is either intentional or accidental. For example, if I stepped on your foot, and I said it wasn’t an accident, I can hardly see what else you could say about that situation than that it was on purpose, or intentional.
This claim is likely more controversial, and the one that Anscombe would disagree with. If I aim to do something—even if I have a low chance of succeeding—and I achieve it, was it an accident? Let’s dissect a few concrete examples.
The first example I will point to is the dartboard example. Presume that your highly uncoordinated friend walks up to the dartboard in the Rose and Crown and surprises everyone by getting the first bullseye of their entire life. I claim that this was not an accident. Consider an alternate case where this same friend walks up to the dartboard, aims for the center, and unfortunately misses by a few inches. They are disappointed for a few seconds until you mention to them that they actually hit the triple twenty, which turns out to be worth more points than the bullseye. It would obviously be weird if they consequently claimed to have hit the triple twenty intentionally; after all, they weren’t even aiming for it. There is a quality about them hitting the triple twenty—namely the accidental nature of it—that does not appear when they hit the bullseye. In order to claim that hitting the bullseye is also an accident, Anscombe would need to concede that hitting the bullseye and hitting the triple twenty are identical, which they are clearly not.
Here is another example that can help delineate the function of “accidentally” in these cases. Consider a rock climber who has been trying to climb one singular route for the past year. The route’s difficulty and length are such that the climber has never made it to the top despite hundreds of attempts. Now, what happens if, on some random Tuesday afternoon, they miraculously navigate through all of the various challenges of the climb, and make it to the very top? Anscombe would have to claim that they made it to the top by accident since they didn’t have a reliable connection between their trying and the knowledge of their making it to the top. But this doesn’t feel right, nor does it explain the situation in an accurate manner. There is something inherently non-accidental about trying and succeeding to make it to the top that is clearly distinct from the accidental nature of something like going to the top of the wrong climb.
A third example is the 10 carbon copies story that has appeared in class time and time again. If you are trying to make 10 carbon copies and end up actually making all of them, even if you didn’t know that you were succeeding while you were doing it, it would be peculiar to say that you made those 10 copies accidentally, which mirrors the intuition of the previous two examples.
In one of Velleman’s footnotes, he hints at his agreement with me in this regard: “Whether something is done intentionally depends on the reasons for and against doing it—especially against. An unreliable but lucky marksman may not be credited with hitting the bullʼs-eye intentionally, but he may be blamed for hitting the president intentionally, simply by virtue of trying to hit the president and despite his lack of a reliable method for doing so…On this point, among others, I believe that Anscombe is mistaken” (Velleman). He concedes that in certain cases, it seems that intentionality can still be present without having the proper reliability of knowledge that Anscombe requires; however, this concession is contingent upon there being the right kinds of “reasons for and against doing it.” He only thinks the lucky marksman kills the president intentionally because he had good enough reason not to. I go one step further than Velleman because this line of reasoning applies in the case of the lucky darts player as well. I see no reason that intention in this regard should be reliant upon the gravity of a certain action—that buying a car is somehow more intentional than buying a pencil simply because the action carries more weight. As is made apparent by Knobe’s experiments, we tend to tie the adverb intentionally with ideas of accountability and responsibility; however, this is specific to side effect cases, and I do not believe that the kind or relative weight of these reasons matters for intentionality in cases like the lucky darts player. In playing darts and assassinating a president, there are numerous reasons for and against doing them, so according to Velleman’s reasoning, they should both be counted as intentional.
I’ve argued why I believe premises 1 and 2 to be truthful, which leads me to believe the conclusion that statement 3 must be true: that succeeding in a task, even a highly improbable one, means you completed that task intentionally. What are the implications of this conclusion? Firstly, note that in these cases, there exists no non-observational practical knowledge about the result, which means that in certain circumstances, it is possible to do something intentionally without this kind of knowledge. I would agree with Anscombe that in normal cases of intention, there is this distinct quality of personal non-observational knowledge; however, this cannot be the only defining characteristic of intention, since I believe that these cases exist that are intentional yet don’t have it. I will end with a brief note on the difference between intention and intentionally. If it is true that you can intentionally make 10 carbon copies or intentionally hit the bullseye, is it necessarily true that you can intend to do these things. Anscombe derives intending from intentionally acting, so she would need to say yes even if she disagrees with the antecedent of that claim. McCann would also say yes since he draws no distinction between aims and intentions. Although I agree with Bratman that intentions are subject to a different standard of coordination than aims, I think this standard only applies for conflicting intentions, not for intentions that can fail. In this sense, I believe that you can also intend to hit the bullseye, although exploring this topic would require another entire paper. But, I have already exceeded the recommended length, so I’m off to go win the lottery intentionally.
J. David Velleman, “What Good is a Will?” Manuel Vargas and Gideon Yaffe, eds., Rational and Social Agency, Oxford University Press, 2014 pp. 83-105.