Something has always bothered me about Parfit’s treatment of Scanlon’s contractualism both in his "Justifiability to Each Other" and in the new Climbing the Mountain. Finally after years of being troubled by this I think I’m starting to be able to put my finger on it.
Parfit starts from various life-saving cases and the principles designed to deal with them. He says that ‘If we think about morality in Scanlon’s way, we cannot appeal to … intuitive beliefs’ about rightness and wrongness of different actions in deciding which of these principles are acceptable and which rejectable.
He continues: ‘According to what we call this [sic] moral beliefs restriction, when we apply Scanlon’s formula, we cannot reject moral principles by appealing to our beliefs about which actions are wrong’
This feature of Scanlon’s view has wide implications’.
Then he goes on to say how the contractualist cannot reject the act-consequentialist principle as we normally do (well, we don’t but let that go) by saying that it is against our moral intuitions.
I think this is quite confused in many ways. First, there is a good reason why contractualist say that moral principles cannot be rejected by appealing to our beliefs about which actions are wrong. This is because beliefs are rarely good reasons to do anything let alone reject principles. Parfit, if anyone, should know this. And, Scanlon of course knows this. Here is how Scanlon formulates the restriction:
‘It would be circular for contractualism to cite, as the reason that people have for objecting to such principles, the fact that they are wrong according to some noncontractualist standard’ (Scanlon 1998, 216, my emphasis).
That is better. Facts seem at least like good candidates for reasons to reject principles. I’m not sure that Scanlon needs the according to some standard addition though. But, what is restricted here is that facts about wrongness of actions cannot be reasons for rejection. Nothing is said or implied about beliefs about wrongness. I’m not sure how Parfit can get this repeatedly wrong.
But, maybe Parfit had something else in mind. Scanlon lists different generic, agent-relative reasons as good reasons for rejection. These include bodily injuries, ability to rely on the assurances of others, and to have control over their lives (p. 204). Maybe Parfit thinks that contractualist should reject the idea that our moral beliefs can have an influence on what we should think about how strong these considerations are as reasons for rejecting different principles.
But, I’m not sure why contractualist should reject that idea. I don’t know why their situation is any different from that of the consequentialists’. They often see that their view initially has some implications that are against our moral intuitions. What they do is they go back to fix their views about well-being, other values, the priority of the first off and then come back with refined principles that better fit our moral judgments. This does not make the view empty if they still hang on to some improved substantial value theory. When they get that right, they can use their view to generate substantial implications and solutions to new cases.
Why can’t the contractualist do the same? If the initial non-rejectable principles do not fit our moral convictions, why can’t she go back and check whether she had the strengths of the personal, agent-relative reasons right and even fix them in the light of the our moral convictions and the aim of getting the normative outcomes of contractualism to fit them. Nothing in this implies that the contractualist starts to take facts about wrongness to be reasons for rejecting moral principles against Scanlon’s (and not Parfit’s restriction). If the consequentialist (even Parfit seems to do this on occasion) is allowed to use the reflective equilibrium to formulate the best version of her view, then so should the contractualist be allowed to do too.
Furthermore, if contractualism and act-consequentialism are normative theories on a par it still seems like an open game to compare both of them (after we have the fixed versions of them from the initial reflective equilibrium) to our moral convictions about right and wrong. So, I don’t see why the contractualist cannot challenge act-consequentialism because it fits our moral intuitions worse than contractualism (if it does) just like Parfit wants.
Jussi,
I think that Derek Parfit’s point is something like this.
When we ask whether some principle is reasonably rejectable we cannot use our moral beliefs about what is wrong to determine this. Why? Because what makes some act wrong is, according to Scanlonian Contractualism, that it is disallowed by some principle we cannot reasonably reject. If our reason for claiming that this principle is not reasonably rejectable is that this act would be wrong, then we would have gone round in a vicious circle. That is why, as I understand things, we cannot use our intuitive moral beliefs about which acts are wrong when we try to determine this by asking whether such acts would be disallowed by some principle that couldn’t reasonably be rejected.
Second, when we are using the reflective equilibrium method, we use our intuitive moral beliefs as data or reasons for or against rejecting moral principles or theories. If some suggested moral theory or principle could not be part of some reflective equilibrium, because we have no intuitive moral beliefs which go together with this theory or principle, then we have reason to reject the suggested moral theory or principle. So, it is not a mistake, as you say, to appeal to moral or normative beliefs when we reason about moral principle or theories.
When we ask whether contractualism is a good moral theory, we may appeal to or use or intuitive moral beliefs to determine this. But, when we do contracutalist moral reasoning, then we cannot, for the reason given above, use these particular normative beliefs. Again, here is why. When we do contractualist moral reasoning, at least if it is the Scanlonian kind, it would be circular to use our beliefs about what is wrong when we ask whether some possible principle is reasonably rejectable or not. If whether some act is wrong determines whether some principle (dis-)allowing is reasonably rejectable, then it couldn’t be that facts about the reasonableness of rejecting these principles make such acts wrong. But, this is precisely what the contracutalist wants to claim. So, the Scanlonian contractualist must say that, when we try to determine whether the relevant kind of principle is reasonably rejectable or not, then we cannot appeal to or use our intuitive beliefs about wrongness. We must use other normative beliefs, like for instance beliefs about what would be bad for particular people, or what would count as great burdens.
I hope that this is helpful.
Sven,
unfortunately not really. You just repeated the same, rather weak objection as most other people.
1. According to Scanlon (well his old view), that an act is forbidden by the non-rejectable principles does not make acts wrong but rather is what it is for them to be wrong. Parfit’s contractualism wants to argue for the ‘making-view’.
2. Scanlon’s deontic constraint – facts about wrongness are not reasons to reject the proposed first-order principles – makes the account a non-circular account of wrongness. Ridge’s Saving Scanlon paper is very good on this. Stratton-Lake has another way of arguing for the same conclusion.
3. I’m not convinced by your second point. Say that act-consequentialism implies that hanging the scapegoat is the right thing to do for the sheriff. The reason I take there to be for rejecting act-contractualism in the reflective equilibrium test is not that *I believe that killing the innocent is wrong* but rather *that killing the innocent is wrong*.
4. Notice that in the way I described the use of the reflective equilibrium to formulate the best version of contractualism I no-where said that wrongness of an action would be a reason to reject a principle. In this sense wrongness of an action does not determine which actions are forbidden by the non-rejecteble principles (which are supposed to determine which acts are wrong). I agree that this would be circular. But, no contractualist has ever thought this.
The reasons to reject suggested action-guiding and justification-providing principles are limited to agent-relative reasons in contractualism – wrongness is not that kind of a consideration or a reason.
But, I don’t see why contractualist cannot say that if she has first assigned some relative strengths to different agent-relative reasons and gets untintuive results from the test, she can use her intuitions about wrongness to reassess whether she was right about the relative strengths of those reasons.
Consequentialists do the same thing in their axiology all the time – why can’t contractualists do the same? Doing this does not make contractualism a ‘spare wheel’ if we don’t think that there are substantial limits to what things can plausibly be good agent-relative reasons to reject principles, if contractualism can offer substantial moral conclusions in new cases and challenge some moral intuitions we have about familiar cases, if it can offer an account of the reasons not to do wrong actions, and so on.
In addition, the argument
1. I believe that torture is wrong.
2. For this reason, the principle that forbids torturing is not rejectable
3. Therefore, torture is wrong
is not circular but just a really bad argument. Neither of the first two premises assumes the conclusion.
The circular argument would be:
1. Torture is wrong.
2. For this reason, the principle that forbids torturing is not rejectable.
3. Therefore, torturing is wrong.
Jussi,
Thanks for these responses.
First, contrary to how you took it, I did not intend my claims to be objections to Scanlonian Contractualism. I think Scanlon’s theory is very plausible, but it is interesting to see what kinds of belief we can use in his suggested kind of moral reasoning.
And, it just seems that, in contractualist moral reasoning of the Scanlonian kind, we cannot determine whether to believe that some act-disallowing principle is reasonably rejectable by consulting our intuitive beliefs about which acts are wrong. The noteworthy thing about this feature of contractualist moral reasoning is precisely that, unlike other kinds of moral reasoning using the reflective equilibrium method, we cannot use our intuitive beliefs about wrongness in this kind of contractualist moral reasoning.
Second, it does not matter whether we take the Scanlonian contractualist formula to be about what wrongness is or what makes things wrong. Either way principles (dis)allowing acts cannot be rejected in our contracutalist reasoning because these acts intuitively seem right or wrong.
Third, of course we don’t think that it would be wrong to hang somebody because we believe that it is wrong. But, that we intuitively believe that it would be wrong to hang somebody is something that we use in moral reasoning using the reflective equilibrium method. We cannot say that we are using facts about what is right and wrong, but only assumptions or beliefs, because part of what we are trying to decide in this kind of moral reasoning is which moral beliefs can be part of reflective equilibrium. If there is some moral belief that has that feature, then it is in this respect more justified than some other moral beliefs that lack it.
So, if we were to use the reflective equilibrium method to argue against act-utilitarianism, we would use the apparent wrongness of hanging somebody to please others as a reason to mistrust act-utilitarian principles. If we claim to appeal to the actual wrongness of hanging somebody to please others in ARGUING against act-utilitarians, then we are, it seems, begging the question in favor of some kind of non-utilitarian view.
Fourth, I didn’t say that you took the wrongness of some action-principle to give us contractualist reasons for rejecting this principle. Since you said that you didn’t understand what Parfit’s point was supposed to be, or what is reasonable about the claims he is making, I was pointing out the reasons why, in contracutalist reasoning, we cannot use our beliefs about the wrongness of certain possible acts. It seems that you miss-read me.
You wrote:
“But, I don’t see why contractualist cannot say that if she has first assigned some relative strengths to different agent-relative reasons and gets untintuive results from the test, she can use her intuitions about wrongness to reassess whether she was right about the relative strengths of those reasons.”
Our imagined contracutalist cannot do this precisely because she believes, or claims, that the wrongness of some act either consists in, based on, or provided by, the fact that this act is disallowed by principles that, because of some apparent facts about agent-relative reasons, cannot be reasonably rejected. Again, this is the difference between contractualist moral reasoning and other kinds of moral reasoning. The contractualist claims about what wrongness is, or what makes possible acts wrong, seems to prevent her, in other words, from doing what you suggest that she could be doing. However, if (1)we engage in contracutalist reasoning well, 2) we establish that some act would be disallowed by some principle that cannot be reasonably rejected, AND (3) this nevertheless does NOT seem to be a wrongful act, then we can take this as an objection to this kind of contractualism. We could do this since we would then not need to be assuming that contractualism describes the nature of, or what gives something the feature of, wrongness.
Sven,
you write:
“it just seems that, in contractualist moral reasoning of the Scanlonian kind, we cannot determine whether to believe that some act-disallowing principle is reasonably rejectable by consulting our intuitive beliefs about which acts are wrong.”
Why? What is the difference to say consequentialist moral reasoning or to virtue ethicist or to Rossian pluralism? Why can consequentialists make changes in their opinions about which things are valuable on the basis of what they think is wrong but the contractualist cannot reassess her beliefs about agent-relative reasons to reject principles on the same basis? I don’t get this. If you do this, you are not rejecting some principle *because* it allows wrongful actions but rather because of some agent-relative reasons you were mistaken about.
As far as I see, nothing in what you say even starts to address this issue. Keep in mind that in the same way as contractualism, consequentialism is either an account of what wrongness is or what makes things wrong. According to them what wrongness is or what makes things wrong, is not acting in a way that has the most valuable consequences. They use their beliefs about wrongness to get matching views about what things have most value. There just are no relevant differences.
Begging the question in a lawful way by the way just is the art of reaching the reflective equilibrium. You cannot get a non-question begging argument against any view from that method. Some people, the intuitionists, think this is bad and we need some other justification for our moral beliefs than coherence. Other people, the coherentists, think it is ok – if the circle is large enough, it’s not vicious.
As before, thanks for this responses to my reply to your post.
You ask why we cannot use beliefs about which possible acts, or which possible act-permitted or act-forbidding principles, are wrong when we engage in contractualist reasoning. I shall make another attempt at explaining why cannot do so, and will use my understanding of what contractualist moral reasoning is in doing this.
I take us to be engaging in contractualist moral reasoning if we try to determine whether certain possible acts would be wrong using Scanlonian Contractualism as one of our assumed premises. When we in such a way use Scanlon’s formula in some piece of reasoning, then this prevents us from using claims about the wrongness possible act-allowing or act-disallowing, principles in determining whether the acts which are (dis)allowed by these principles are wrong or right. This kind of moral reasoning is supposed to determine which acts are wrong using things other than the wrongness of possible principles of actions or the wrongness of certain possible acts.
Here is another way of making this point. On one understanding of what Scanlonian contractualist reasoning consist in, we can reason about which possible acts are wrong by asking whether people possibly affected by such acts would have any reasonable objections to our acting in this way. If they would, then—-according to Scanlon’s view—-this possible act is wrong. In giving their objections, these potentially affected people couldn’t say that they object to possible act since it would be wrong for us to act in this way. That—i.e. whether this act, or principles that allow it, is wrong—–is precisely what we are trying to find out by asking whether they would have any reasonably objection to our doing this thing! However, if these potentially affected people were to say that they object to this possible act because it would impose some great burden upon them, then—since this is a reasonable ground for objecting to principles that allow such acts—this would be something we could use in our contractualist moral reasoning.
If the wrongness of some possible act affects whether people potentially affected by this act have self-interested reasons to object to principles that allow this act, then it is unclear why we would use claims about their self-interested reasons to determine whether this possible act is wrong. But, the contractualist claim is just that we can use our beliefs about what people could object to on self-interested grounds when we try to determine which acts, or act-allowing principles, are wrong.
Do you still find the Parfitian claim about how we cannot use our moral beliefs in contractualist moral reasoning mysterious?
Reading my own posts above, I just noticed that, since I wrote them so hastily, they contain a number of annoying typos and grammatical errors. Sorry about that!
I do. First, everything you say there is compatible with the use of the reflective equilibrium method within contractualism in the way I described above. As I said, beliefs about wrongness or wrongness are not reasons to reject principles and nothing in the use of our moral intuitions I suggested implied that they would be. I don’t know how many times I need to repeat this. You are just talking past my view and so is Parfit of Scanlon’s.
Consider this argument:
‘You ask why we cannot use beliefs about which possible acts, or which possible act-permitted or act-forbidding principles, are wrong when we engage in the rule-consequentialist reasoning. I shall make another attempt at explaining why cannot do so, and will use my understanding of what rule-consequentialist moral reasoning is in doing this.
I take us to be engaging in rule-consequentialist moral reasoning if we try to determine whether certain possible acts would be wrong using Hookerian rule-consequentialism as one of our assumed premises. When we in such a way use Hooker’s formula in some piece of reasoning, then this prevents us from using claims about the wrongness possible act-allowing or act-disallowing, principles in determining whether the acts which are (dis)allowed are right or wrong. This kind of moral reasoning is supposed to determine which acts are wrong using things other than the wrongness of possible principles of actions or the wrongness of certain possible acts.
Here is another way of making this point. On one understanding of what Hookerian rule-consequentialist reasoning consist in, we can reason about which possible acts are wrong by asking whether the principles that forbid them had the ideal aggregated consequences in terms of people’s well-being with some priority to the well-being of the worst-off. If they had, then—-according to Hooker’s view—-this possible act is wrong. In assessing what people’s well-being consists of and how much priority we should give for the worst-off, we couldn’t deny that well-being was not just pleasure or argue that we should give some priority to the worst-off because we thought that otherwise the ideal rules would allow wrong acts. That—i.e. whether this act, or principles that allow it, is wrong—–is precisely what we are trying to find out by asking which principles have the ideal consequences!”
That’s your argument against Hooker. I bet I could do the same for any moral theory whatsover. Why is it an objection against contractualism but not the other views?
Jussi,
I shall read your post carefully, and if I have anything to say about it, I will do so. But, here is just a quick thing. The claim that we cannot use our intuitive moral beliefs while we are engaged in contractualist moral reasoning is not, as far as I can see, necessarily intended as an objection to this kind of reasoning. Why would it be an objection?
Good question. According to Parfit, the objection is that not being able to use our moral intuitions within the contractualist moral reasoning implies that we would not be able to argue against act-consequentialism on the grounds that it has unintuitive moral consequences. He writes “If we all accepted Scanlon’s formula [and the moral belief restriction it implies], Act Consequentialists could dismiss these appeals to our moral intuitions”.
My argument was in the original post that we can give a role to our moral intuitions within contractualism without loosing our ability to argue against act-consequentialism, contractualism, or any other ethical view on the grounds that they have counter-intuitive consequences. And, we can do this without making contractualism circular or empty.
Thanks for this clarification. I shall have to re-read your first post now that I know what your aim was.
Thanks for this clarification. I shall have to re-read your first post now that I know what your aim was.
Jussi,
your post raises an interesting side-issue. What exactly makes it the case that since we have an intuitive belief that F is wrong, then any principle (P) allowing F must (pro tanto) be discarded or revised? (1) The fact that I believe, or that it is generally believed, that F is wrong provides a pro tanto reason for discarding P: but psychological facts can hardly be reasons for anything, though they may be rough guides; (2) The fact that F is wrong gives a reason for discarding P: in a sense, this is obvious (that the table is brown gives a reason for rejecting the proposition that this table is red). But it is also a reason we are not allowed to appeal to, since we are seeking an independent reason why a principle allowing F, i.e., telling us F is not wrong should be discarded. Also, it would be a decisive reason not to believe that F is allowed, whereas we clearly here are talking of pro tanto reasons for belief. (3) The fact that allowing F would not cohere with our intuitive belief that F is wrong provides a reason to reject P. Here what matters is some objective relation of coherence between our beliefs. But of course one could say the opposite: your intuitive belief that F is wrong doesn’t cohere with my principle allowing F, so much the worse for your belief. Why should either lack of coherence matter more? Because that F is wrong is more INTUITIVE? Perhaps. But intuitiveness is hard to assess, comes in degrees, is a very tricky matter overall. Also, I do wonder whether all the work in this third option is done by the objective relation of coherence or by (again) the quality of the psychological fact of intuitive belief. Since reflective equilibrium is a very common case of normative epistemology, we should be able to explain its mechanics in terms of reasons for belief. This is however surprisingly hard to do.
Francesco,
that’s great. It’s just the kind of issue that I’ve been thinking about currently. First, we need to revise your question slightly.
This is because the answer to your question is nothing as Rawling and McNaughton argued. I might believe that murdering is wrong (I do occasionally). The principle that forbids lying allows one to murder. This does not however make it the case that we should reject the principle against lying even if it allows murdering which is wrong. The solution to this problem is to replace ‘allows’ with ‘authorises’. So, the question is what makes it the case that if I’m convinced that F is wrong, then any principle that authorises doing F must be disrecarded?
I’m working on a paper called ‘Externalist Moral Foundationalism’ where I argue for something like your 2 against most intuitionists who would go for 1 and coherentists who would go for 3. Much of it is top secret currently so I cannot tell much about it here. But, maybe we can do this over email at some point.
Oh yeah – one more observation that I think I can make now. In your discussion of (3) you ask why should either one of the coherence to just as coherent but mutually inconsistent belief sets matter more. The thought you entertain is that because ‘F is wrong’ is more INTUITIVE. I take it that giving this reply gets you from (3) to something more like (1). Some beliefs have psychological features (Sidgwick, Ross, Stratton-Lake, Crisp, and so on list them) that provide justification for those beliefs themselves so that they license rejecting incompatible principles. This is the picture I want to reject. I don’t think phenomenological features of certain beliefs can have warranting force. But, that’s all to be argued for.