What follows is the second part of the post that I started here. Also, parts of this post were copied from a paper I am working tentatively entitled “Contractualism, Basic Aggregation, and Varied Ills Aggregation: A Defense.”
0. We will once again start with contractualist principles (slightly revised from last time):
CW: an act, α, is wrong if some person has grounds for a reasonable rejection of a system of action guiding principles that includes [edit 5/29] any principle that allows for the performance of α.
CR: an act, β, is right if no person has grounds for a reasonable rejection of a system of action guiding principles that includes any principle that allows for the performance of β.
RR: agent x has grounds for a reasonable rejection of a principle γ if and only if γ causes agent x to “fare worse” than any other agent and there is another principle, θ, that does not require any agent to sacrifice as much as x does under γ.
IR: Only the interests of individual people can ground reasonable rejections.
1. Let me remind of you of the basic aggregation case:
BA: Imagine that you are in a boat that is slowly filling with water. Equidistant from you are two other boats. In the boat to your left is A. A’s boat is quickly filling with water. Her boat will shortly go under and she will die if you do not go to save her. In the boat to your right are B and C. Their boat is filling with water at the same pace as A’s boat. Their boat will also go under shortly, and they will both die if you don’t go save them. You only have time to save either A or B&C because of your leaky boat.
Moreover, let me remind of Scanlon’s argument for why you are obligated to save B&C and let A die:
Scanlon argues that if you were to save A both B and C would have grounds for a reasonable rejection. C could say, e.g., that you were not giving her due respect if you saved A. In other words, you would be neglecting important reasons for action that are derived from your obligations to her. Scanlon argues that this is true because if the case were different such that you had a choice between saving A or B (and C was not present), it would be permissible to save either A or B. In other words, you wouldn’t have decisive reason to save either of them. However, once C enters the picture things change. It is no longer permissible to save A because of the reasons that derive from C. You would be wronging C by saving A. Rahul Kumar puts it this way: there are two stages of deliberation. In the first stage, you can either go left or right. A and B neutralize each other; i.e., they cancel each other out. However, C ‘tips the scales’ such hat you must go right. After the decision has been made to go right, there is another decision, and that is whether you should save B along with C. Since it will not be any harder for you to save B, you ought to do it.
2. Now I will move on to the three objections I introduced last time. I will start with the second objection first. I wrote the following about this objection last time:
The second attack comes from Jens Timmermann, who argues that Scanlon’s argument falters early on. Specifically, when Scanlon concludes that it would be permissible to save either A or B in the case where you can either save A or B (i.e., the case where C is not present). Timmermann writes,
“The two processes or states of affairs of A’s being saved or B’s being saved may be equivalent; but it does not follow that if you went ahead and saved either A or B you would be justified either way… This is indeed the case with two identical objects, say two precious vases; but in the case of human lives it matters greatly which way the decision goes– to A as well as to B. Their claims have the same strength, but their lives are not interchangeable. The conclusion for the case of equal claims cannot therefore be that anything goes– quite the opposite: nothing goes” (107, 108).
Timmermann is arguing here that it is not permissible to save either A or B. In other words (it is curious that he doesn’t put it this way), he thinks that if you save A, B will have a reasonable rejection of your action, and vice versa.
Timmermann’s argument here is that you are not permitted to save either A or B when C is not present. On contractualism, this is false. To see this, let’s analyze what one needs in order to ground a reasonable rejection (this is expressed by RR above). In short, agent A needs to fare worse under the principle, Y, and there must be an alternative principle such that, under the alternative principle, no one fares as badly as X does under Y. Under this principle, neither A nor B can reasonably reject a principle that allows you to save the other person. Let’s say you decide to save A. B does not have a reasonable rejection of the principle you act upon (let’s call this X) when you save A. She meets the first conjunct; i.e., she fares worse under that principle. However, she fails to meet the second conjunct because there is no alternative principle where no one fares as badly as she does under X. The other relevant principle would mandate that you save B. However, under that principle, A fares just as badly as B does under X. Thus, B cannot reasonably reject the principle you act upon when you save A. Likewise, A cannot reasonably reject the principle you act upon when you save B. Consequently, Timmermann’s objection fails.
3. Let’s now consider the first objection I described last time. I wrote the following on that occasion:
The first attack comes from Michael Otsuka (Alastair Norcross makes a similar point), who argues that Scanlon’s argument implicitly appeals to aggregation. He writes, “Scanlon[’s] argument for saving the greater number considers C’s claim in combination with B’s claim so that they together tip the balance in favour of saving B and C” (291). In other words, Otsuka argues that Scanlon’s argument violates the individualist restriction by aggregating B and C’s claim.
This objection can be put very succinctly: Otsuka argues that Scanlon’s argument violates IR, which blocks aggregative considerations from grounding a reasonable rejection. Otsuka asks us to imagine that we are weighing the agents’ claims on a metaphorical scale. He then writes,
The pairwise comparer places A on the left-hand scale and B on the right-hand scale. Then the pairwise comparer removes B from the right-hand scale and places C on the right-hand scale (while leaving A on the left-hand scale). In both cases, the scales are evenly balanced. Scanlon also place[s] A on the left-hand scale and B on the right-hand scale, at which point the scales balance evenly. Then [he] tip[s] the scales to the right by placing C on the right-hand scale along-side B. It follows from this method of tipping the balance that the Scanlon argument appeals to ‘reasons corresponding to the claims of groups of individuals.’
I think that Otsuka clearly misinterprets the argument. To see this, let’s make the type of aggregation disallowed by IR more explicit:
NAP: One must not appeal to, inter alia, ‘reasons corresponding to the claims of groups or individuals’ when attempting to ground a reasonable rejection to a moral principle.
Otsuka claims that Scanlon’s argument violates NAP (which is a sub-principle of IR). It is not the case that Scanlon’s argument violates NAP, however. Let’s take up Otsuka’s analogy, except now we will give numerical values to the weights of the individual’s claims.
Let’s say that when we put A on the left-hand scale it dips by a value of 10. When we put B on the right-hand side, her claim balances out A’s because it also has a value of 10. However, once we put C’s claim on the right-hand side, that side becomes the dominating side because of C’s individual claim of 10. It is certainly the case that the inclusion of B’s claim on the right-hand side makes a difference. But it is not the case that C’s complaint about the principle where you save A appeals to ‘reasons corresponding to the claims of groups or individuals.’ C can reasonably complain that that principle ignores her claim to be saved because it treats the scale as if she is not present at all. In other words, it treats the scale like it is balanced. C’s complaint is akin to what her complaint would be if she were the only one in danger and you decided to do nothing. It merely appeals to her individual reasons to be saved. Her germane complaint is that you have failed to notice the value of her life. Surely B’s presence makes such a complaint possible, but that doesn’t mean that C is violating NAP when she reasonably rejects the principle you act upon when you save A over B&C.
4. À propos the final objection, I wrote the following last time:
[Timmermann] argues that A will have grounds for a reasonable rejection of you saving B&C. Timmerman argues that, “A’s complaint will be that he never stood a chance” (ibid). Moreover, he argues that B can also complain because “the value of his life has allegedly been neutralized by the value of someone else’s life on a distant island, he is saved just because he happens to e on the same island as C” (ibid). He sums up his objection by stating, “B and C, like A and B previously [i.e., in his first objection], are considered not equal individuals, but interchangeable objects; which is precisely what persons are not” (ibid). Hence, he argues that you would be treating A like an object if you saved B&C. In fact, you would be treating B like an object as well. You would not be respecting them as people.
This can be broken down into three claims. The first is that when C is not present we are being unfair to A and B when we conclude that it is permissible to save either. This claim was refuted in section 2 above.
The second claim is that “A’s complaint will be that he never stood a chance.” I take this to mean two things. First, it means that A’s complaint is that there is no way that it is permissible to save her under contractualism. Second, it means that A’s complaint is that we do not respect her claims as a person. I think that the former is true. Given the situation, A cannot be permissibly saved. However, it is false to say that one is committed to saying the second reading is true because the first is. It is the case that we give due weight to A’s claims. The very fact that it would be permissible to save A when C is not present is proof of this.
The third and final claim made here is that B would have grounds for a reasonable complaint because “the value of his [B’s] life has allegedly been neutralized by the value of someone else’s life on a distant island, he is saved just because he happens to be on the same island as C.” It seems clear that Timmermann is misinterpreting the argument here. Let’s break this down into two separate claims. The first says that B can reasonably complain because her claims are being neutralized by A’s. However, B cannot make such a complaint because B cannot justify this complaint to A. That is, it would be unreasonable for B to say to A that you ought to save her because A will suffer the most under that principle and there is no alternative principle that will cause less suffering for any other agent. In this way, contractualism is relational. That is, contractualism holds that it is not only important what claims such and such an individual can make, but it is also important how those claims’ strength relate to other individuals’ claims. This is highly attractive feature of contractualism, and B must keep it in mind when she is making complaints. If she does, then she will realize that, when it comes to whom you should save, A’s claim neutralizes her own claim. Thus, she cannot reasonably complain about A’s neutralization of her claims.
The second part of Timmermann’s claim is that B can complain that the only reason she is being saved is because she happens to be on the same island as C. This is not true. B has grounds for the exact same reasonable rejection that C does of the principle you act upon when you save A. That is, it wrong to interpret the argument as saying that B does not have grounds for a reasonable rejection if you save A. B does have such grounds for precisely the same reasons C does. You would be ignoring B’s claim if you saved A because you would be acting like B is not present. It is clearer to break down the claims like we did. That is, it is easier to say that A’s claims and B’s claims neutralize each other, and C’s claims tip the scale. However, we could have just as easily said that A’s claims and C’s claims neutralize each other, and B’s claims tip the scale. In other words, B has grounds for a reasonable rejection of you saving A in the same way C does.
