I’m a contractualist. There – I’ve said it. My supervisor Brad Hooker is the rule-consequentialist. You might think that we have endless debates about which of the closely resembling views is right. Unfortunately we have better things to do. But, I do want to explain why his main argument based on the moral status of animals against contractualism fails to convert me into a rule-consequentialist.
Hooker’s argument is that rule-consequentialism can make sense of the idea that morality imposes constraints on how we should treat animals whereas contractualism cannot do so. I want to claim that if rule-consequentialism can make sense of these constraints, then so can contractualism, whereas if contractualism cannot make sense of these constraints, then rule-consequentialism cannot either. When it comes to animals, we are on the same boat.
To see this, I think it is worth distinguishing between two questions:
1. What constraints does theory X place on how we should treat animals?
2. What does theory X tells us about why we should respect those constraints?
I accept that if a moral theory gives an answer to either 1 or 2 that conflicts with our moral convictions, then this is a serious fault in the theory. However, I want to suggest that rule-consequentialists and contractualists will have to give the same answers to these questions and so these questions cannot be decisive between the views.
Start from 1. According to rule-consequentialism, an act is wrong if it is forbidden by the set of principles which, if adopted by almost everyone, would have the best consequences in terms of over-all well-being. It’s important to note that the well-being of every sentient being whose life can go better or worse counts. This implies that presumably a rule that forbids torturing animals for mild amusement of humans has better overall consequences than the allowing rule. Thus, we seem to get the right sort of constraints. There may be a problem that the ideal rule becomes to demanding if it requires maximizing the well-being of all animals. But, leave that aside. Let us accept that rule-consequentialism gives a right answer to 1.
What about contractualism? Well, it says that an action is wrong if it is forbidden by a set of principles that no-one could reasonably reject. Reasonable rejection, furthermore, is a function of the strongest objections individuals could make to alternative sets of principles. The set to which the strongest objection is weakest is the non-rejectable one.
The question then is, could the sets of principles that do not place constraints on the treatment of animals be reasonably rejected? I think they could be. First, I think they could be rejected by many humans. Animals and their well-being is an important part of the personal life, essential concerns and projects of many humans and so for them living in a world were animals are mistreated would constitute a personal burden. Now, you might complain that this route to rejection is too indirect.
In that case, I propose that the animals could reasonably reject the allowing principles too. There is a clear sense in which their lives would go worse and be burdened due to the considered principles. They would have reason to reject these principles as a result. These reasons would also be of the very same sort for which humans can reject principles. To deny these as reasons for rejection would be a clear instance of speciesism. Of course, the animals are not able to present their objections in the imagined hypothetical negotiations, but I don’t see how this is here or there. Many humans couldn’t or wouldn’t either. If it helps, you can imagine trustees that do it for the animals. If these reasons for rejection are allowed, then it seems to me that contractualism can give an intuitive answer to 1 – a similar answer as rule-consequentialism. So, 1 cannot be the decider.
Many people, including Brad, think that 2 is the crucial question. Contractualists want to say that the reason why we should follow the non-rejectable principles is that we have reason to be able to justify our actions impartially to others. So, why not mistreat the bunny you see in the garden? Leaving neighbours aside, presumably the idea is that you could not justify the mistreatment to the bunny. That sounds fine to me. But, Brad says that ‘justification to a rabbit, or a dog, or even a whale, manifestly does not make sense’. Brad doesn’t really explain why he thinks this – we seem to just have a clash of intuitions. What do others think? Anyway, let’s accept the view that giving justification to animals does not make sense. Then contractualism would fail with respect to the question 2.
How do rule-consequentialists cope with the question 2? You might think that they do well. They could be thought to be able to say that the reason to follow the constraints created in question 1 is to bring about more general well-being. In that case, the rule-consequentialist might be better off. Unfortunately, the rule-consequentialist cannot say this. This is because of the classic problem of rule-consequentialism. There will be cases in which one can either bring about the maximum amount of well-being or follow the ideal code but not both.
For instance, one could bring about very small benefits to a large enough number of humans or avoid harming one bunny seriously. Assume that avoiding harming the one bunny seriously is what the ideal code requires whereas the other option maximizes well-being. Now, why does rule-consequentialist say we need to follow the ideal code? It cannot say that because this maximizes well-being. Brad’s solution to avoid this problem is to say that acting on the ideal rules is ‘impartially defensible’. To me, this strikes as another way of saying that it is ‘impartially justifiable’. But, to whom do we need to be able to impartially justify our actions in the case under consideration?
Well, presumably to the bunny (or her trustee or the people who care about her) who would be mistreated for the sake of the small benefits to many. But, rule-consequentialists have already said that justifying actions to a bunny (i.e., being able to defend them) does not make sense. So, rule-consequentialism seems to fail with the question 2 too. A rule-consequentialist could say that it does make sense to want to act in a defensible way when the defence is given to a bunny (or the trustee or the people who care about bunnies), but then so could a contractualist. So, I still fail to see the difference.
Hi Jussi,
I was wondering how you differentiate between contractualism as you define it, and consequentialism in general, as you define it.
If i can paraphrase, you define contractualism as the set of [available?] principles to which the weakest objections can be brought. So there is a strong sense of objectivity built into this mechanism – there must be [fairly] solid criteria by which we weight what a reasonable objection is? If we allow for too much subjectivism of criteria, the assessment of strength of objection becomes meaningless.
With this in mind is it not also arguable that the most objective measurements of an outcome, or principle set, are the most tangible harms [and benefits] it is likely to result in? Do we not venture into *more* subjective terrain the further we move from ‘basic’ harms such as pain, bodily damage and death (and even instances of these can be subjectively described as goods for some individuals), to ideas such as perhaps character, virtue or honesty?
With that in mind then, does this account of contractualism not risk collapsing into consequentialism itself, for if it relies on:
(1) measurable [fairly] objective harms [and benefits] to assess the strength of objections, and:
(2) we assume [justifiably?] that rational actors would want to minimalise harms [and maximise benefits]?
Then doesn’t this version of contractualism say something along the lines of:
(3) That the [only] set of [available] principles that no-one could reasonably reject, is the set which maximises good outcomes, and minimalises bad ones?
Appreciate your thoughts.
Richard,
that’s a good question. I think there is actually two questions here. So, the first issue is whether contractualism collapses into (rule-)consequentialism. I don’t think this happens. The reason for this is that rule-consequentialists aggregate the objections (and benefits) to the principles interpersonally. That is, small objections based on the burdens individuals come to experience amount to a big objection when summed up.
Contractualists are not supposed to allow this. For them the burdens and objections can only be aggregated intrapersonally and compared pair-wise between individuals. In a sense the resulting principles have the best consequences – they create the smallest personal burdens but this is not the same sense as when rule-consequentialists say principles are best, i.e., ones that create the largest amount of aggregate well-being.
The second question is can contractualists give any objective criteria for comparing the personal objections people have to the codes. This seems to depend on your general account of reasons, their objectivity, and our abilities to make reason-judgments. At least we can describe better and more reliable ways of making reason-judgments. It is a further question how much disagreement there would be, should we consider one party in such disagreements to be mistaken, and so on. These are the sort of questions that a defender of any moral theory has to struggle with.
Jussi,
A few ideas to throw out:
Your argument that RC is no better than off than contractualism with respect to question 2 relies on familiar objections about the potential of RC to result in suboptimal outcomes. But that’s an objection to RC’s answer to 1, not an objection to what RC says in response to 2. In other words, your worry is that RC places no constraints — or at least no non-overridable constraints — on the mistreatment of animals. But maybe I’m misunderstanding — I’m not sure.
Second, this disagreement may not be best captured in terms of question 2. I.e., the worry that consequentialists would likely have about contractualism’s implications about the treatment of animals have to do with the kind of (moral) reasons contractualism provides for how we should treat animals. The contractualist seems to be saying, for instance, that inflicting gratuitous pain on animals is wrong because they (or their trustees) have reason to reject gratuitous pain. In contrast, RC is apt to say that something’s being gratuitously painful to animals is a reason not to do it. It’s the state of affairs that provides the reason, not any actual or conceivable attitude animals or their trustees have toward the pain. What contractualism misses (its critics might say) is the unmediated nature of the reasons to not engage in this kind of mistreatment of animals.
Finally, I’m interested in this matter of ‘justification to’ animals: Not sure where I stand, and I’m curious both (a) why contractualism is so closely associated with justification-to and consequentialists generally not associated with it, and (b) whether, after all the idealizations, qualifications, etc. that contractualists typically introduce, justification-to doesn’t turn out to be extentionsionally equivalent to ‘justification’.
Hi Jussi. Nice post. But what if someone were to argue as follows? That x can reasonably F entails that x can F. A bunny cannot reject any principles (in order to reject a principle, one must first understand it etc.). Therefore, a bunny cannot reasonably reject any principles.
Thanks for the comments guys. Here’s few thoughts:
Michael,
I do think that in these cases I have in mind RC gets the right result – the suboptimal choice fits our intuitions about what to do. What RC seems to leave unexplained is why we should act in this way. This would make it to be a question over 2. This I think assumes that RC does not collapse into AC as Brad argues. If it does then RC will give the unintuitive answers in which case you are right that we are back to question 1.
I like the way you put the problem on the table in the second paragraph. I think here contractualism and RC do equally well too. Many contractualists, including Scanlon (and me), accept the first-order reasons such as animals pain. They just want to claim that there is additional strong reason to act justifiably. There is a question of whether the first reason isn’t already sufficient. It might well be.
But, I think the RCist is the same situation. She too should accept the first-order reason of particular animal’s pain. If RC is going to do any work, this view too will need to say something about additional reasons to follow the rules – whether this refers to aggregate social good or impartial defensibility is another issues.
The last issue is interesting and would deserve investigation. My hunch is that contractualists begin from the intuitions about how we wrong people and take the wrongness of actions to be derivative of that. This leads to the thought that justification to people is basic – we wrong the people we cannot justify our actions to. Consequentialists seem to be taking the opposite route.
Campbell,
that’s fine. I think there is a question of in what sense we should understand ‘can’ here in ‘can reasonably reject’. My charitable reading is ‘would have reason to if…’. I’m not sure a requirement for actually being able to understand or other such things follow this. I might need to though give the bunny quite unbunnyish characteristics when I think about the hypothetical rejection of the principles.
Jussi
Very interesting post. I have a couple of questions/observations.
Regarding the claim that there is a problem with RC in that in a specific instance following the rule may result in a less then optimal maximization of well-being, I am wondering why this is a problem for RC? It would seem that this is a recognizable and acceptable consequence of adopting RC over AC or contractualism. We adopt a rule (or principle) because in general following the rule will maximize overall well-being while recognizing that in a specific instance following the rule may not or even will not accomplish this goal. Furthermore, if we accept the classical objection against RC as defeating RC then are we not collapsing into AC and that isn’t that what RC is trying to avoid?
A more serious problem for RC and contractualism is what rules, or principles, to adopt. What counts as a strong reason for accepting a rule or principle for one person may be a weaker reason for another person and how are we to decide between these conflicting claims? If we rely on our intuitions, then it seems that this will not resolve the issue in that one person’s intuitions may support adopting a rule while another’s intuitions may count against adopting that rule and how are we to decide without relying on our intuitions which would seem to beg the question. I have a problem in general with relying heavily on intuitions (or convictions) because they seem to be non-starters; they do not take us very far if there is disagreement over the status of, or relevant justificatory role of the intuition itself.
John,
that’s very good. I agree that it is not a problem as such for RC that it in some cases comes apart from AC. This is supposed to be one of its advantages and Hooker has an ingenious argument for how it is possible based on the internalisation costs of too complicated rules.
The problems only begin from when the views come apart. The RCist takes aggregate social good to be so fundamental that the basic rules should be selected on the basis of how much aggregate social good they bring about. However, in the cases under consideration, we could bring about more aggregate social good by not following the rules but rather by doing what AC tells us. For these cases, the RCists need to find something that would rationally motivate agents not to bring about the maximal amount of the fundamental good – aggregate well-being. I think the move to impartial justifiability is wise. But, then that seems to be taken to be fundamentally important in the same way as contractualists take it to be.
I think there is two parts to the second question – one metaphysical and one epistemic. To come clean, I’m a realist about the reasons to reject principles in the same way as I am realist about other reasons. There is something ‘out there’ to get our judgments about reasons right about. Our intuitions are hopefully pretty reliable in getting information about how the reasons pan out. They are even better if they are subjected to procedures of critical reflection.
I don’t know however from where else we could begin except our intuitions. But, I do see the problem for anti-realist views about reasons in which something seeming to be a reason and something being a reason collapse. Timmons does argue in his Scanlon paper that this leads the contractualist to objectionable forms of relativism.
Jussi
Thanks for your thoughtful response.
You write: “There is something ‘out there’ to get our judgments about reasons right about. Our intuitions are hopefully pretty reliable in getting information about how the reasons pan out.” How do you get rid of the ‘hopefully’ re the reliability of our intuitions? This seems to be the point of contention. If our intuitions are reliable sources of information of some objective and independently real normative structure that we simply have to uncover then we do not need reasons. The intuitions themselves would be epistemically sufficient to warrant our normative claim. If we need reasons that are somehow warranted independently of our intuitions so as to provide epistemic justification for our normative claims then our intuitions are not epistemically sufficient to be reliable sources of a sufficient amount of information of that reality (assuming that your realistic position is correct).
The example I have in mind is Singer’s use of thought experiments to evoke an intuition in his audience, but once this intuition is summoned he moves to a reason based approached to ultimately provide the epistemically needed framework that provides the justification for the normative claim (which is what the intuition evokes) that he is trying to defend. Intuitions may be necessary (although I am not sure that they are) to provide a final accounting of our normative claims but they are not sufficient.
How are you interpreting ‘aggregate?’ If it is collective aggregation over the whole population (say 100 people. Imagine that we can increase the aggregate well-being of the whole by 1000 units. The resulting distribution may be uneven to the point of one person having, say, 90% of the total increase and the others sharing the remaining 10%. The overall aggregate well-being of the group as a whole is increased even though some (most) individual members’ well-being are minimalized. I do not think many RC’s would accept this interpretation so it seems that one of the rules justified by consequence-based argument must be justice-as-fairness based regarding distribution within the whole population so that the shares are more even distributed. Is it possible that on RC grounds one might argue for ‘veil of ignorance’ based type of contractualism where one of the rules grounded in RC is that when we make decisions affecting aggregate distribution we must imagine that we do not know how we will be individually affected by the distributive outcome when we make our decision regarding how shares are to be distributed?
I am beginning to ramble….
John,
that’s good. I had in mind a picture of two different kinds of reasons. There certainly are practical reasons and there are also theoretical reasons to have beliefs about the practical reasons. I cannot imagine that we could do without the one or there otehr.
The reasons to reject principles in the contractualist framework seem to belong to the first, practical class. Then there is a question of what theoretical reasons there are to have particular beliefs about what reasons precisely they are. I think intuitions come here as *some* reasons for certain beliefs but certainly not as the final word.
It is true that rule-consequentialists can do various moves to fit our distributive intuitions better. Diminishing marginal utility already does some work. I think Brad Hooker is inclined towards some version of prioritarianism where the well-being of the worst-off counts for more. They could also accept that equality or fairness of the distribution has intrinsic value which needs to be taken into account when we consider which rules have the best consequences. I haven’t seen that anyone would have tried to connect RC with Rawlsian contractualism in the way you suggest but maybe it is possible just by saying that Rawlsian distribution has intrinsic value.
Jussi
To move back to the issue of how best to understand our obligations to animals (assuming we have some) I would like to suggest a problem (not original) for the contractualist that appears to be as serious as the one you poised relative to RC. You wrote. “What about contractualism? Well, it says that an action is wrong if it is forbidden by a set of principles that no one could reasonably reject. Reasonable rejection, furthermore, is a function of the strongest objections individuals could make to alternative sets of principles. The set to which the strongest objection is weakest is the non-rejectable one.’
I want to pick up on the issue raised by Campbell regarding a bunny’s ability to reject a principle. It seems that part of the contractualist position is that one must be able into a contract with others. In order to do so one must do so knowingly and freely and should not be coerced or deceived into entering a contact. To pick up on an idea from Fried (sp?) a contract is a form of promising. Derived from this is the idea that we have obligations only to those we have entered into contracts with (or made promises too). It seems that animals, or at least most of them, cannot make a promise therefore they cannot enter into a relationship that is defined by duties and obligations. Now I may enter into an agreement with my neighbor not to mistreat her dog if her dog takes a poop in my yard if she agrees to pick up the poop on a timely basis. But I do not have any obligation that is derived from and dependant on the inherent nature of the dog derived from this agreement. I am not recognizing any inherent property of the dog that entails that I make a promise to my neighbor to treat her dog in a certain manner. I do however recognize an inherent property in my neighbor, her ability to enter into agreements knowingly and freely that results in me making a promise that affects how I treat her dog. The problem is that I can understand my promise to my neighbor to treat her animal in a certain manner but can I extend this concept to cover how I should treat all animals. I think not and that poses the problem for me on how to understand how a contractualist can explain how we have any general obligations to treat animals in a certain manner. RC does not have this problem because the RCer can make a rule that says we should treat the interests of all sentient beings equally, etc. and this will enable us to understand how we should treat all sentient beings.
It seems to me that this whole discussion goes astray from the moment it asserts “It’s important to note that the well-being of every sentient being whose life can go better or worse counts.” This seems to me to be exactly what arguments about our moral obligations to animals are trying to establish. It’s not settled what “sentience” is or how it confers moral standing. And, heaven forbid we were able to answer this conclusively, we’re going to be hopelessly entangled in questions about whether to save the antelope from the cheetahs.
I think it requires a much more contextual understanding of moral obligation and the real world in order to make sense of these questions. Is it OK to eat animals if they are raised free of suffering? Does it matter how rich you are and thus what options you have to nourish yourself without eating meat? What is the standing of cultural traditions in which meat-eating is part of religious ritual? And, yes, what about the antelope and the cheetahs?
Clearly in the (post-enlightenment Western) world as we know it, our alternative views about the morality of humans eating cats, humans eating ruminants, and cats eating ruminants, is now primarily rooted in the specific institutional arrangements by which the categories of “pets” “food” and “nature” are maintained. Actual rumination (pun intentional) on necessary and unnecessary suffering is generally absent. The point to the work of Singer et al. has been to change the visibility of the key terms and challenge the categories. It is hard to see how contractualism, act consequentialism or rule consequentialism would have any substantive disagreement on implementation if the underlying point – that the suffering of animals is morally repugnant insofar as it is unnecessary — were granted. The extension of Humean empathy seems more than adequate for this task.