Parfit, Kant and Consequentialism

In my last posting on Parfit I discussed the general problems I think there are with his notion of "contractualist consequentialism" as presented in the third of his 2002 Tanner Lectures. In this posting I want to look at the concluding section of this third lecture in order to see how Parfit brings the argument there to a conclusion and to use the discussion of this conclusion to demonstrate again some reasons for thinking that Parfit stacks the deck rather when he reaches the conclusion that Kant gives the basis for a contractualist view that is substantively consequentialist.


Parfit opens the final section of this lecture with a statement that whilst Kant was not a consequentialist the argument's point is not to examine the reasons why Kant was not a consequentialist but instead to point to some reasons for taking it to be the case that we can develop a consequentialist view by appealing to some elements of Kant's positions. In making this case Parfit appeals to an important distinction. The distinction is between what he terms Kant's "moral beliefs" on the one hand and the "implications of his principles" on the other. The former are what are involved in Kant's not being a consequentialist on Parfit's view, whilst the latter, by contrast, point to consequentialist conclusions. This distinction is an important one, as we shall see in looking at the concluding phase of Parfit's argument.


Parfit's case is built by reference to the principle he extracted from Kant's Formula of Humanity as the first part of it. This principle is what Parfit terms "Kant's Consent Principle" (KCP). The "consent" involved is framed by Parfit as "rational consent" and it is connected to the act consequentialist view that rational consent can be seen as agreement that everyone be treated in ways that would make things go best. If we look at rational consent in this manner, then treating people in a way that implies adoption of an axiological standard of value is equivalent to treating them in ways to which they could rationally consent. One of the points Parfit rushes to make here is, however, meant to assuage a problem Rawls raised against utilitarianism. Rawls indicated that a problem with accepting "classical utilitarianism" was that it required some to  sacrifice themselves for the benefit of others and that such a sacrifice was one that the ones being sacrificed could accept only if they were prepared to sacrifice the very point of being in an ethical space at all (so it is connected to his problem with the "separateness of persons"). However Parfit thinks he has a reply here as if the act in question would impose great burdens on some for the sake of greater benefits for others then this act would not make things go best but instead go worse. This is despite the fact that the person in question might "rationally consent" to the act. On this basis Parfit argues that consequentialists don't have to be seen as utilitarians.


The separation of consequentialism from utilitarianism that this argument involves does, however, require, as a consequence that an optimific act is not one that can be assessed only through the prism of rational consent. Rational consent is thus not a means to determine the rightness of the act if such rightness is standardly viewed in terms of optimific understanding. In which case, whilst Parfit thinks that KCP could be accepted by act consequentialists, it is not sufficient as a standard of rightness for them. 


Parfit, however, takes the argument a different way as he now moves away from the consent principle and back to the principle he earlier termed "Kant's Contractualist Formula" (for which see this earlier posting). The contractualist formula requires that we appeal to principles that everyone could rationally choose and we could not rationally will that everyone accepts act consequentialist principles and so the act consequentialist principle is not one we should act upon. Notably, however, and remarkably, Parfit has here done something that cuts heavily against the grain. He has presented act consequentialism as sufficiently robust that it can provide a defence for the separateness of persons, such a defence that it can be distinguished from utilitarianism, and then, having done this, has used an appeal to contractualist views about universal rational acceptance, to undercut act consequentialism after all. It would, according to standard ways of presenting the moral situation, be rather more apt to say that act consequentialism, like act utilitarianism, is insufficiently robust to prevent appeal to demanding notions of self-denial and then use that to support a view of consent principles that did not view the latter in optimific terms. So Parfit's initial argumentative strategy is, to say the least, very odd.


Parfit refers now to the argument Sidgwick famously used that it would be best, from a utilitarian point of view, if not everyone was a utilitarian. This argument, which shows that utilitarianism lacks transparency, is one that Parfit endorses here against act consequentialism. In doing so he appears to appeal to the view that moral beliefs matter in the sense that if we knew it to be true that it would not make things go best if everyone acted in ways dictated by principles of maximising what would go best, we should prefer it that not everyone had the belief that they should make things go best. Indeed, not only does Parfit appear to make this move, but he also uses it to prefer rule consequentialism to act consequentialism since rule consequentialism requires us to act on principles whose acceptance would make things go best even should this be the view that not everyone believe that they should act in ways that make things go best.


Parfit now constructs an argument that is meant to move from  Kant's "contractualist formula" that we ought to act on principles whose universal acceptance everyone could rationally choose to the conclusion that rule consequentialism is correct on Kantian contractualist grounds. Now there are several stages in this argument which bear close analysis if a response is to be made. The first point to look at is how Parfit views reasons. Parfit introduces and critically interrogates the premise everyone could rationally choose whatever they would have sufficient reason to choose. This premise is interrogated as what we have sufficient reason to choose is argued by Parfit to depend on facts whilst what we rationally choose is, by contrast, based on beliefs. The difference matters since false beliefs give us reasons to choose something that it is not ultimately rational to choose. So the application of Kant's "contractualist formula" requires congruence between beliefs and facts. 


The next premise is to the effect that everyone would have sufficient reason to choose that everyone accept the principles whose acceptance would make things go best assuming the congruence we have established as a requirement for this. Now, in critically evaluating this premise, Parfit points to the "impartial reason-involving" sense of normative reason as the basis for viewing how to evaluate whether or not things have gone best. However, in assessing this claim, Parfit also refers back to his earlier distinction between "moral beliefs" and "principles" since he rejects the objection that such a principle might produce immoral action on the grounds that such an objection illegitimately introduces moral beliefs. There is a problem with this reply since it assumes that the contractualist procedure is thus far not loaded or is effectively neutral between different views. But it is not. In taking our standard in terms of an axiological conception of evaluation we have already adopted a consequentialist standard into our reasoning so it is evident such a standard will be our conclusion. If, however, our argument precisely concerns a question about how "rational consent" is to be understood, for example, then we cannot simply stipulate that it be seen in not merely an impartial sense but in terms of a view of impartial reason that is pre-set in terms of outcomes. Viewing impartial reason in terms of outcomes incorporates a "moral belief" into the "neutral" evaluation. And it does so in a way that is then used to appeal to "principles" as if they are not described in ways that are set against other "principles" where the other principles are then merely demoted to the status of "beliefs".


The reason why Parfit thinks that such a division between "principles" and "beliefs" is stable is because he sees the former as tracking "facts" and making them congruent with "true beliefs" whilst an insistence on "moral belief" simpliciter is simply an argument from what we "already accept". However the reference to "facts" here is one that involves a view of the factual that is itself normatively loaded since it takes factual support to consist in a view about what is axiologically supported and so it is not neutral between views.


Parfit argues that in pursuing this strategy he is following a standard contractualist procedure and he even appeals here to the example of Rawls to support him. Rawls, however, when constructing the original position, does so by building the position out of what he termed "formal constraints of the concept of right" and such constraints are not discussed by Parfit. Particularly important from such constraints was a solution formally argued to the "priority problem" with regard to principles. Parfit stipulates a solution by taking axiological criteria to be supreme, something that cuts precisely against his earlier recognition of the point that there is a case to be made against abstracting from the separateness of persons.


Parfit next goes on to consider this claim by balancing what he regards as the considerations of personal points of view with those of impartial reason. In doing so Parfit stresses the idea that there are non-self-interested personal reasons which, indeed, there are in terms of different conceptions of the good. However, Parfit does not consider this question in terms of different conceptions of the good but only in terms of personal reasons being seen, if they conflict with impartial ones, as self-interested. Parfit does not consider that personal reasons that oppose what he views as "impartial" reason may also be principled in their opposition. Because of this Parfit simply arrives at the conclusion that it is always optimific to follow impartial reason and that we should accept impartial reasons as personal reasons given that this is so. But this is not evident at all if there are impartial reasons that are non-optimific in form as is the case with non-consequentialist views. 


Because Parfit has construed the choice situation in the way he has he cannot avoid the problem of unreasonable burdens that, at the beginning of the discussion, he wished to argue was not a necessary element of consequentialist views. Assuming that one holds a non-optimific view of goodness by means of a conception of the priority of the right over the good one will always be told that ones view of the good is only personal and should be sacrificed optimifically which is to abstract from the separateness of persons in precisely the way Rawls complains of. This is still the consequence of Parfit's preferred rule consequentialism which is why it still incorporates the classic consequentialist move of requiring us to see impartial reasons as always optimific in form. Parfit thinks he can appeal against this since he argues that ultimately he is appealing to Kant's "contractualist formula" and basing rule consequentialism on this. However this is dishonest as the contractualist formula, in application, is always used to bolster the conception that impartial reason is framed in terms of optimific outcomes so it is only a supplementary principle meant as a support for consequentialism and is not, as Parfit pretends, foundational for ethical principles. 


So whilst Parfit concludes by saying he has not argued that we should become rule consequentialists but only that Kantian and contractualist premises support rule consequentialist conclusions this is not what his argument shows. What his argument shows is, instead, that if contractualist and consent principles are interpreted axiologically then they support consequentialism. But that is only because they have been pre-defined in such a way that consequentialism is their inevitable focus. Parfit assumes that the only response to this argument is to favour "moral beliefs" over "principles" but, as I have argued, this division is founded on an incorporation of a "belief" into his account of "principles" in the first place.

Parfit and "Rawls's Formula"

In my previous posting on the third of Parfit's 2002 Tanner Lectures I looked at the account Parfit gave of the Rawlsian view of deliberative rationality. In this posting I'm going to look further at the third 2002 lecture in order to concentrate now on the treatment Parfit gives there of what he terms "Rawls's Formula", which is Parfit's way of describing the methodological appeal Rawls makes to the "veil of ignorance".


The appeal to the "veil of ignorance" is as part of Rawls' discussion of the original position and is a way in which he moves from the initial situation to the original position. Parfit views it as a "revision" of the basic contractualist conception of rational agreement and he analyses the various defences Rawls gives for the appeal to it and offers what he takes to be a problem with the different versions of the "veil" that Rawls is said to present. Parfit mentions amongst the justifications Rawls gives for appealing to the veil of ignorance that it enables us to move away from contingent facts that apply to us at present and which may influence the way we currently view the situation but which should not, impartially, affect our understanding of just distributions. The way Parfit understands this is that there is no "threat advantage" in the situation, a point that we noted in the previous posting was utilised by Rawls to argue against contemporary Hobbesian views. Our ignorance of who we are in the situation removes any such advantage though Parfit, somewhat oddly, also assumes that this means that "everyone's well-being" is taken into account when, instead, it means that the outcome that is arrived at is one that is not based on considerations of well-being but instead is formally construed.


The difference between these understandings of the veil of ignorance becomes clearer when Parfit begins to assess the general claim that Rawls makes to the effect that the contract doctrine offers a general alternative to utilitarianism. Parfit is surprised that Rawls feels able to make such a claim given, as Parfit puts it, that an appeal to "a combination of self-interested rationality and impartiality" tends to produce utilitarianism. Rawls, however, has already rejected appeals to self-interest on the grounds of the formal constraints of right and his view of impartiality is not one that is shaped by appeal to well-being as Parfit assumes hence the notion of impartiality favoured by Rawls does not tend to lead to utilitarian conclusions.


The reason why Parfit thinks differently about the appeal to the veil of ignorance than I am suggesting is correct is due to how he views Rawls' consideration of the notion of "average" utility which Rawls discusses in section 27 of A Theory of Justice. Rawls here provides a description of why the notion of "average" utility is preferable to the "classical" notion of utilitarianism and imagines arriving at the former through recourse to a device of contractualist reasoning. The way this goes is that Rawls views the notion of "average" utility as conjoined to a conception of having an "equal chance" of being any individual within the society though he also adds to this the assumption that all individuals have "similar preferences" whether or not they belong to the same society. Given these assumptions there is a general rationale existent for the "average" principle of utility. Parfit points out, in a footnote, that there is, even given these assumptions, something to be said to favour taking the "interests" of the worst off most in the situation. In fact, Rawls goes further, pointing out, in the process, that the situation of choice involved in the original position is quite different to any other choice given that the uncertainties it involves are so great and adding that it would be rational, given such uncertainties, to go so far in weighting the position of the worst off that the principle of average utility ends up (even given the "equal chances" idea) of being practically identical with the difference principle.


Parfit pays insufficient attention to this point remaining fixed on the claim that there is something of a case for the principle of "average" utility assuming the preference for criteria of well-being are tantamount for us. In making this point Parfit draws on the way that the argument of section 27 of Theory develops by progressing from the "equal chances" conception of the veil of ignorance to a "full" veil of ignorance in which there is no knowledge provided of the place one has in society. In fact, Parfit fails to note here that it is not only the place one has in society that is affected by moving to the "full" veil of ignorance. It is also the case that we no longer assume that individuals have "similar preferences" whether or not they belong to the society. We do not claim any knowledge of such preferences when we move to the full veil which is another way of saying that questions of well-being (understood through the prism of preferences) are not assumed in advance to guide our decision of the right principles to be chosen.


Parfit points to the account Rawls gives in section 28 of Theory where Rawls gives a further argument for moving away from an "equal chances" conception of the veil of ignorance and where Rawls advances the argument that "there seem to be no objective grounds in the initial situation for assuming that one has an equal chance of turning out to be anybody". In stating this point Rawls refers to the way that the principle of average utility appears to presuppose a principle of insufficient reason as the ground for how probabilities are to be assessed and such an appeal is merely one to "as-if probabilities" that effectively are only ways of stipulating the principle of average utility and are not means of arguing for it as a principle to be adopted. Further the conception that is at work in the "equal chances" view of the veil of ignorance that is preferred by the advocate of average utility itself requires a very stark conception of the person which Rawls terms that of "bare" personhood so that the persons in question lack determinate characters (which is one of the ways the "equal chance" notion works out). By contrast Rawls posits a conception of determinate personhood as the basis of choice so that there really are "interests and ends" of the persons in question even if the nature of them is unknown to the parties involved. So the form of the veil of ignorance that supports the "equal chances" conception, whilst not as full as that which Rawls advocates, is more drastic in its treatment of the persons engaged in forming the contractual agreement.


Parfit, however, objects to the claim that there are no "objective" grounds for favouring the "equal chances" version of the veil of ignorance on the grounds that it appears to treat the original position as an actual state rather than an hypothetical one. This is incorrect as we have seen. It is not a question of treating the original position as an "actual" state, it is rather one of specifying what assumptions are involved in the relative ways the thickness of the "veil" is presented as. The "equal chances" formula requires adoption of a particular conception of probability that is itself not neutral between principles to be considered and which is, further, thicker than Rawls' version of the "veil" in another way, in the way, that is, that it treats the persons in the contractual situation since it renders them mere vessels for utility maximisation. Thus there are two ways in which the "equal chances" view of the veil of ignorance is philosophically problematic: firstly, it assumes a form of probability assessment that is a covert introduction of the principle to be argued for and secondly it pictures the contracting parties in a way that is part of the general utilitarian conception of ignoring the separateness of persons.


Indeed returning to the way the path to the principle of average utility was earlier rendered by Rawls in section 27 of Theory shows that it was precisely by assuming these points that the principle was taken to be available for consideration in the first place. Rawls wrote there: "if we waive the problem of interpersonal conceptions of utility, and if the parties are viewed as rational individuals who have no aversion to risk and who follow the principle of insufficient reason in computing likelihoods....then the idea of the initial situation leads naturally to the average principle" (my emphasis). It is only by not following through that it is these controversial assumptions that support the advocacy of the principle of average utility under the "equal chances" view of the veil that Parfit can assume that the argument for the "equal chances" view of the veil is as good as that for the "no knowledge" view of the veil (which latter is, as we have seen, mis-named).


Parfit assumes that the two versions of the veil of ignorance are equal in their impartiality. In fact, however, they express two quite different conceptions of impartiality since the "equal chances" version of the veil of ignorance takes impartiality to imply that persons are treated rightly when they are viewed only in terms of interests and desires whilst the other version of the veil of ignorance takes it that there is more to people than this. So different conceptions of the person are related to different views of impartiality and the one that supports the "equal chances" formula is one that is framed to lead us to the principle of average utility and to its formula of "impartiality".


As a separate argument Parfit considers the claim that utilitarianism conflicts with our strongest moral beliefs and that this is a distinct argument for rejection of it and to favour Rawls' principles of justice. However, against this argument Parfit states that if we use the appeal to our beliefs to favour one of the versions of the veil of ignorance we cannot also use the veil to produce a rationale for these beliefs. This fails to grasp the point of the reference to the veil of ignorance and suggests some misunderstandings on Parfit's point of Rawls' procedure. The point of appeal to the veil of ignorance is to move away, as Parfit earlier recognised, from contingent features of existing situations in order that consideration of correct principles can be undertaken. Does this entail that there are no constraints involved in the choice of the principles in question? No: Rawls is explicit in appealing to the constraints of right in the situation. He incorporates such features as, for example, the publicity condition. The point is not that such conditions are equivalent to or presented as arguments for the principles of justice. It is rather that they provide us with a way in which the criteria for what would count as a preferable conception of justice can be specified. And if some account of a principle emerges which violates some of these conditions (which include what Rawls terms the "circumstances" of justice) then this is an argument against them.


The final argument that Parfit presents against the formula of the veil of ignorance Rawls has favoured returns to the reasons Rawls gives for taking the position of the worst off into particular account. Parfit formulates this as an argument that requires us to "maximise the minimum level" and thus terms it the "Maximin Argument".  There is lack of specificity in terms of how Rawls understands the worst-off since there is no one specific way that definitely defines them and Parfit suggests that this group should be understood in egoistic terms despite the point that this offends against specific constraints of right on the grounds that Rawls mentions at one time a view of a "representative" person of this group which Parfit argues offends equally against specific constraints of right. However the key problem here is not so much how the group is to be defined as to understand how addressing their situation should be understood. Parfit typically takes it that we address their situation by means of application of a consequentialist process of welfare distribution whereas this is contrary to the point and process of the difference principle. The latter points not to welfare increments but instead to the conditions under which self-respect and the general primary goods (which include conditions of character formation) are to be enhanced. So the considerations Parfit applies to the "Maximin Argument" are of the wrong sort.


Finally Parfit looks at the Maximin Argument in terms of the selection of moral principles rather than in relation to principles of justice which means that the assessment of the argument is simply understood in a way that is not apt for it. One of the reasons why this odd element of Parfit's account of the Maximin Argument is followed appears to be due to an overall claim on Parfit's part that the difference between Rawslian reasoning and that of utilitarians is much less than first appears. A reason for this appears to be Rawls' remark in section 87 of Theory that, in the "initial situation", parties are not taken to have any particular ethical motivation but instead to decide solely on the basis of what seems best calculated "to further their interests" and this passage is meant to bolster the peculiar way Parfit understands the original position. The point here is one that Rawls is making about recognising the intuitive idea of rational prudential choice but Rawls immediately adds that the formal conditions on principles and the veil of ignorance mean that it necessarily also includes moral features and we have seen that his criticism of the average principle of utility further pointed to a conception of the persons involved in the contractual situation as having ends in addition to interests even though we did not know what they were. This passage therefore does not support Parfit's conception of the original position.

Parfit and Kant on Impartiality (III)

The published Chapter 14 of On What Matters corresponds to the account given in the 2008 version of the work which I treated previously. In many respects the analysis and response to the discussion thus is much of a piece with that given earlier. In order to make the case for a response to Parfit's published version worthwhile, however, I have chosen to focus here on what I now view to be the key methodological device used in Parfit's discussion of impartiality. This is the recourse to a process of imaginative identification as the way in which impartiality is to be understood. The importance of stress on this will become clear when I make subsequently manifest how it ensures that Parfit does not seriously respond to the Kantian view that he takes himself to be replying to in this chapter.


The chapter opens, as previous versions of it did, with a statement of Kant's rejection of a principle that Parfit, along with others, has identified with the "Golden Rule" where the rejected rule incorporates already a notion of imaginative identification as it asks us to treat others as we would wish to be treated by them. Kant rejects this rule and indicates a number of problems with it. The nature of the specific problems that Kant poses with the rule are not identical, however, with how Parfit takes Kant's objections to go. Kant objects that the rule is not fit to be viewed as providing a universal law and he gives 2 reasons why this is so: a) it does not contain the ground of duties toward oneself or of love toward others; b) it does not contain the ground of duties owed to others. The second part of the first objection (concerning the duty of love towards others) is restated on the basis that some would agree to not be benefited by others if only it follows that they are not required to benefit these others. Hence this point is to the effect that the nature of the reciprocity recognised in the rule is insufficient to ground beneficence. The second objection is supported with a reductio proof to the effect that the rejected rule provides no ground for a judge sentencing a criminal since the criminal can ask the judge whether he would like to be punished. In both cases Kant suggests that the process of identification that the rejected rule seems to require is insufficient to ground the duty in question.


Parfit's discussion opens with a description of why the objection to the attempt to ground beneficence on the rejected rule would fail according to Kant. Parfit indicates that the rejected rule is not faulty in the way indicated since the rule does not lapse simply because one is not inclined to ask for the help of others. The reason why Parfit thinks he can make this move is that the rejected rule requires that we help others if we would wish to be helped whereas the exemption suggested in the objection Kant makes is that someone might only say, not, that they don't want help but that they would be willing to forego it if they are thereby freed from obligation towards others. The reason why Parfit thinks this is a reply to Kant appears initially mysterious. After all, isn't the denial of such help thus the basis for saying that I have renounced the need for help and therefore have a ground for asking others to do the same? I think that Parfit fails to see this point as he is held captive to a picture, a picture that runs through the whole of this chapter and the picture is one in which the imaginative appeal of the need of others is taken to provide not only a binding ground of obligation but to have a kind of obvious basis as such a ground, one that no one could really be obtuse enough to deny (except, as we shall see later, in cases where the other person is radically different to oneself).


Parfit is, however, wrong to think that he has replied to Kant's objection. The objection holds since the rejected rule requires and mandates only reciprocity of treatment on the basis of the imagined identification and the person Kant considers has simply requested that others extend to him the courtesy he would to them of refusing to ask for help. This plea is entirely in accord with the rejected rule and is at variance with the Law of Nature formula since the law of nature formula requires not imaginative identification but instead the construction of a world governed by the maxims that are stated in the aim underlying the specific policies of given actions. Here it would be an aim of generalised non-beneficence and this is why Kant views the maxim here as falling foul of the Law of Nature formula in a way it does not fall foul of the rejected rule. It falls foul precisely because such a generalised maxim of indifference could not be willed universally as to will it would undercut a basic condition of agency which requires the help of others to prosecute innumerable policies and their accompanying aims so that the maxim of non-beneficence is in tension with the general conditions of agency, not with imaginative relations to others.


Parfit's next consideration of the case of benevolence states that the rejected rule does not refer to conditions of willing as it does not view people generally as being absolute monarchs or dictators who can make others do what we will, a remark that indicates a complete failure to understand the basis of reference to willing in Kant's case. It is not that Kant wishes us to imagine that any one is an absolute monarch, it is rather that he wishes us to view our will as setting laws that apply to others in the same way as natural laws do, that is, as publicly acknowledged and consistently binding of their action. When so viewed some types of maxim conflict with conditions of agency and this shows the problem with them. Parfit subsequently shifts ground and argues that the rejected rule can also incorporate understanding of this is if it is stated in a form that says something like, "do as you would be done by". However this is simply not equivalent to a recognition of willing as it is rather again an imaginative relation to others in terms of how each of us treat others and requires us, as in the standard form, to regulate our conduct by how we would wish to be treated, thus beginning from a kind of egoistic starting point and requiring that standpoint to be checked by this imagination of others treating us as we have treated them. But this is not the same as thinking about internal requirements of willing and their coherence which is what Kant is concerned with.


The next step in Parfit's treatment of benevolence is to view the rejected rule as having the resource that he thinks is behind Kant's discussion of willing, namely of capturing a notion of rational choice. Then the "Golden" rule becomes: "We ought to treat others only in ways in which we would rationally be willing to be treated by others". This reformulation has the same methodological form as the classic version of the rule since it proceeds exactly in the way just described in the previous paragraph with the addition that now "rational" willing is presented as a form of willing that prudently involves a check on its own exercise for the benefit of the one willing. This converts the "Golden" rule into a kind of instrumental guide for conduct but such a guide is not equivalent to an appeal to the conditions of rational willing as outlined in the Law of Nature argument. 


That the two are not equivalent becomes apparent as Parfit looks at cases in which beneficence is required. To sharpen the  "Golden" rule in his new understanding of it Parfit asks that we apply it to conditions that might not be actual for us so that, for example, in thinking about cases of starvation we should imagine not if we would be willing to do without food now where we may have plentiful ways of getting some in the future and perhaps have been eating too much, but in some quite different case where we really are like those who are starving. Similarly, it is not sufficient, on Parfit's revised "Golden" rule, for a racist to deny others the use of resources on the grounds that they are not a member of the affected race as the revised rule requires that this racist place himself in the position of those who he is harming. This makes clear that identification with the plight of those harmed is to the fore and that this often requires imagination to be stretched as when it may be necessary to imagine that one is a member of a different race, lives elsewhere or even belongs to a different gender. This expanded procedure leads to Parfit viewing the "Golden" rule next as stating that we should treat others as we would rationally wish to be treated ourselves "if we were going to be in these other people's positions, and would be relevantly like them". 


Parfit next looks at the objection Kant makes using a reductio argument to the idea that the rejected rule could treat of duties owed to others with the case of punishment used to suggest a problem. Parfit gives a formulation of the rejected rule that he thinks would face this problem and this produces a further reformulation of it on his part so it now becomes: "we ought to treat other people as we would rationally be willing to be treated if we were going to be in the positions of all of these people, and would be relevantly like them". On this basis the judge can now reject the case the criminal makes for why he should not be punished as the judge takes a wider identification than was at first in view and includes the victims of the crime (for example) in his imaginative extension of his identity. On these grounds the rule can provide a basis for punishment. Notice, however, that the question of the scope of identification is carefully here limited. Imagine further that the criminal had a number of dependants and that their need is greater than that of his victims. Should this be taken into account by the judge in his imaginative identification thus giving the criminal a renewed case? Or is there perhaps something wrong with the whole idea that what matters here is imaginative identification?


The last point that Parfit considers is Kant's suggestion that the rejected rule does not cover duties to oneself. Parfit acknowledges that if the rule does not do this it could have serious consequences and that generally it is likely it has been formulated in a way that does not appear to recognise this point given that we do not tend to ignore our own well-being. In acknowledging this point Parfit simply extends the scope of imaginative identification further so that when we consider how to treat others we are one of the "everyone" who would be affected. Notably, however, this appears to make us simply equivalent to others and, indeed, to render all others commensurable with each other.


Parfit considers the "Golden" rule and Kant's Formula of Universal Law as alike in two respects. Firstly, they both appeal to rational grounds of choice. Secondly, they both take everyone to matter in a way that involves some basic moral egalitarianism. Now, the point I wish to make is that the ground of the "Golden" rule on Parfit's defence of it is quite different, in both these respects, from the ground of Kant's procedure for identification of the supreme principle of morality. In regard to rational grounds of choice the "Golden" rule defines this in terms of needs and benefits and understands these in terms of imaginative processes of identification with others. With regard to moral egalitarianism it further assumes a basic commensurability of moral agents. By contrast, Kant's procedure is to view the ground of universality in terms of conditions of rational agency, not in terms of needs and benefits but instead in terms of what the conditions are for any identification of an agent that has any needs or benefits. Secondly, moral egalitarianism is defined not in terms of commensurability of agents where such commensurability abstracts from the separateness of persons but instead identification of persons as basic moral units.


Parfit views the different questions posed methodologically by appeal to these two notions in a way that is different to how I would put them. According to him the "Golden" rule asks how I would like it if something of a certain sort was done to me. By contrast, the formulas of Kant are alleged to ask "what if everyone did that"? In fact whilst I agree that the "Golden" rule has an implicitly egoistic starting point I don't think that it contrasts with the Kantian procedure primarily just through that being its question whilst Kant asks instead of some kind of universal consequence of others doing certain things. Rather it is the case that the advocates of the "Golden" rule adopt the procedure of asking me to view actions in terms of the consequences to me of certain things becoming universal. Whilst Kant's examples also often appear to do this the basic question underlying them is asking something else in addition: namely, what would happen to agency, my own included, if x, y, or z became universal. So it is not a question of a consequence that is primarily at issue for Kant but of a status being respected, the status, that is, of agency itself.


The "Golden" rule asks questions about points of view for Parfit, the points of view of others and myself in certain situations and assesses rights and wrongs through trying to arrive at a way of harmonising my point of view with others. Kant's procedure is different as he instead wishes to ask us what the conditions of agency themselves require. When Parfit subsequently invokes the idea of the "Impartial Observer" beloved of Adam Smith he is not leaving the ground that he has occupied previously of defence of the "Golden" rule as this "impartial" observer is an external way of visualising the conditions of imaginative identification meant to remove specific peculiarities from how I may view something. By thus abstracting from my own case I am supposed to still be able to carry out imaginative identification without having this any longer involve even partial reference to my own self. This abstractive sense of imagination is still one in which an idea or image of a self is meant to fill the role that for Kant is filled by a general argument about agency itself.


Parfit assumes that a procedure of imaginative identification has more motivational power than the kind of argument Kant uses. Some of the reasons why Parfit thinks this become clear as he motivates some objections to Kant. The first objection is the one he terms the "rarity objection" which essentially consists in imagining cases in which I can ensure that someone else suffers a burden or fails to get a benefit unjustly for my own egoistic reasons due to it being the case that our situation ensures that my actions don't subsequently rebound on me. This rarity objection is a form of maxim-fiddling which is meant to get past Kant's Law of Nature formula on the grounds that if others don't and can't discover what I have done they will not subsequently affect me in the way I have affected an other here. This, however, simply fails to view the maxim in question as one that has been put to Kant's test at all since if it had been it would be a public, universally known law and given that it would there would be no way for concealment to take place. Such a publicity criterion is key to the statement of the Law of Nature formula and failure to bear it in mind ensures that Parfit states an objection that simply does not apply.


A different type of objection is made by Parfit to Kant when he states what he terms the "high stakes objection", based on the problem that in some situations acting against another may well be overwhelmingly plausible to me given that failing to do so would have seriously adverse consequences for me. Parfit is right to think that such circumstances are ones in which terrible temptation is placed before people and that frequently they weaken when this is the case. However, whilst this is true, it is still not sufficient to say that the Law of Nature formula does not apply in the case. It still governs the case and that it does is not affected by the difficulties of the person in question. They are still asked to take the universal law into account, however difficult it is for them psychologically to do so. In similar cases, other moral rules of similar scope would be similarly difficult but would not, due to this alone, have to be contravened.


The third and final objection Parfit makes concerns what he calls the "non-reversibility objection" in which we are in a situation with regard to others that is asymmetric. The example Parfit refers to is again the case of the racist who wishes to deny services to others of other races, something that, as Parfit puts it, is universal "in his social world". Parfit seems to assume that because the maxim of such conduct is currently widely "acted upon" that it poses some special challenge to Kant's account and that the better process of response is the one motivated by the extension of imaginative identification. However it is neither the case that some special problem is stated here nor that the case is better dealt with by appeal to imaginative identification. The reason it is not a special problem is that the wide currency of a practice in some given social circle is not the same as its actual universality and nor would even an actual universality be the same as a normative universality. It is the latter which is at issue and it is not defined by processes of exclusion in which others are arbitrarily treated as undeserving of recognition as rational beings. The reason the extension of imaginative identification is not preferable to the appeal to conditions of rational agency is that the required identification may simply be declared too difficult or, even worse, the racist or sexist might simply state that in such cases they would rightly accept their own inferiority. Appeal to universality of agential requirements does not allow such a response.


In conclusion Parfit considers cases of "defences" of Kant which view the Kantian procedure on the pattern of imaginative identification so that it becomes the case that Kant is treated as if he was formulating "a greatly inflated version of the Golden Rule". This is not the point of Kant's procedure as Nagel suggests when he asks us to think we are in everyone else's position or when T.C. Williams asks us to take the standpoint of the impartial observer. Such methodological moves are alien to the spirit and point of the Kantian method which does not require us to be responsive to others' needs on the basis of sympathy. The ground of benevolence is not found there for Kant. It is, rather, found in conditions of agency itself. It is the point of Kant's objection to the "Golden" rule that this ground is not identified by it as what is morally crucial.


Parfit fails to see this is and this failure is the basis of the concluding move of the chapter, a move in which he argues for an assimilation of Kant's procedure with that of Thomas Scanlon. Scanlon, like Kant, asks the question, "what general principles of action could we all will" and this appears to be viewed by Parfit, at the end of this long chapter, as a question like the one that motivated the "Golden" rule on his construal of it. As we have seen it is not the case that the methodology of the "Golden" rule is the same as that of Kant and nor is it the case that what is salient for the one is equivalent to what is salient to the other. So when Parfit concludes the chapter with his formulation of a "Kantian Contractualist Formula" it remains to ask whether contractualism is to be understood by Parfit after the image of imaginative identification that has guided his account of impartiality.

Parfit and Kant on Impartiality (II)

The version of Parfit's account of impartiality that was given in Climbing the Mountain was treated here. In one respect the version that appears in the 2008 version of On What Matters is vulnerable to an objection that was given to the earlier version in a completely unchanged way. This is that Parfit persists in On What Matters in stating the Silver Rule to which Kant states an objection in the Groundwork as if it was equivalent to the Golden Rule, something that he is, it is true, far from alone in doing. Since in the earlier posting I stated at some length the problems I take there to be with this I won't revisit that point here.


The reason why the whole question of the "Golden" Rule is discussed is because of Kant's account of benevolence in the second part of the Groundwork which arises as one of the examples to which the Formula of Universal Law is related. The rule that Kant objects to as a potential alternative to the Formula of Universal Law is rejected by him for three different reasons. Firstly, the rejected rule does not contain the "ground" of duties to oneself, secondly it does not contain the ground of duties "of love" towards others and thirdly it does not contain the ground of "duties" owed to others. The duties of "love" include benevolence and Kant indicates here that the rejected rule does not lead to a duty of benevolence since, on its basis, one can say that one won't help others as one does not require help from them.


The rejected rule is one that Parfit states provides help to others on the basis of the need the others possess and this would be sufficient on its basis to provide an obligatory basis for one helping others. However the rule as Parfit states it is simply that we treat others "as we would want others to treat us" and Kant's point is that if we reject the need for help we can thereby reciprocally deny the need to supply it. This point appears not to be grasped clearly by Parfit. Parfit seems to be arguing, firstly, that the objection misses its target which it does not and secondly that it would apply to FUL which it would not. FUL does not focus on need but on the ground of obligation and the ground of obligation is one that requires universalisation of maxims in accordance with formal rules. These latter do not permit the exemption begged by the one who does not wish to be helped.


Parfit goes on to consider Kant's argument to this effect treating it as a requirement of rationality. When seen this way Parfit appears then to think that the rejected rule would make the same requirement but in doing so he fails to note the reciprocity stated in the rejected rule which is what allows the begged exemption to be stated and which is not involved in FUL in terms of needs. In an attempt to capture some of the force of this idea Parfit restates the rejected rule as follows: "We ought to treat others only in ways in which we would rationally be willing to be treated by others". However this restated form of the rejected rule is one that is then given an objection by Parfit in terms of a different type of begged exemption. Given the reformulation the begged exemption this time does not refer to need but instead is formulated in racist terms with the example being someone who is willing to universally treat those of another race in a way that denies them treatment of a sort that would not be denied to himself. The point that Parfit makes in response is that such a begged exemption misunderstands the reformulated rule which asks us to treat others as we would be treated by them were we in their position.


This additional requirement leads Parfit to restate the rejected rule in a new way: "We ought to treat others only in ways in which we would rationally be willing to be treated, if we were going to be in these other people's positions, and would be relevantly like them".  This point brings in a form of imaginative identification and in so doing is meant to counteract the racist move. Having reached this point Parfit assumes that he has reached a formula that is not subject to Kant's objection that the rule be rejected because it does not supply a basis for duties of "love" towards others and turns instead to Kant's objection that the rule does not cover duties "owed" to others. That objection was stated in a form that was appropriate to a duty of right since Kant stated that we need a formula that does not lead - absurdly - to the situation that a convicted criminal can argue with his judge that the judge would not wish to be imprisoned and thus has no right to imprison others. Parfit concedes that certain kinds of view of the rejected rule would have this result. However the right interpretation of the rule would not have this effect and in making this point Parfit revises the rule again: "We ought to treat other people as we would rationally be willing to be treated if we were going to be in the positions of all of these people, and would be relevantly like them".  This rule is meant to rule out the absurd result but has the effect that the nature of its application is much harder to see.


The judge is apparently to be asked, in response, to commit the act of imaginative identification not merely of the state of the criminal but of those affected by the criminal's acts. Due to this the judge would not be sympathetic to the criminal's plea. Seen this way, however, it would be the case that the rule has to be viewed as much more extensive than the initial formulation suggested and that might well lead us to the view that it is no longer the same rule that Kant objected to. 


The final objection that Kant made to the rejected rule was that it does not disclose the basis of duties to oneself. Parfit first suggests that it is not meant to do this and so this is not an objection to it but concedes that failure to describe such duties might also distort our sense of obligations to others. In order to meet this objection, however, Parfit makes a move that appears questionably consequentialist. It is to reformulate the rejected rule as follows: "We ought to treat everyone as we would rationally be willing to be treated if we were going to be in all of these people's positions, and would be relevantly like them". This response is meant to cover the objection based on duties to oneself by treating oneself, as Mill famously put it, as "one and no more than one" which is precisely the grounds on which Rawls stated that utilitarianism abstracts from the separateness of persons. After all, one is not simply equivalent to others and to treat one in this way is to view obligations as an occasion for maximisation. In fact Parfit does state an awareness of this point but does not attempt to reformulate further the rejected rule in order to meet it.


Rather than do this Parfit looks again at FUL and at Onora O'Neill's view that it is intended to pick out the intuitive idea that we should not treat ourselves as worthy of special treatment. Parfit does specifically mention Rawls' separateness of persons objection and concedes that one person's burdens are not compensated by another's benefits. 


Parfit next considers some objections that can be made against universal impartial principles beginning with the Rarity Objection that applies to actions that would be "too rare" to have significant effects on others. This includes cases in which some might be punished for crimes they have not committed on the maxim "let others be punished for my crimes", a maxim that Parfit thinks could pass the Law of Nature test. In appearing to think this Parfit reveals a lack of understanding of the test since the only way this could be taken to be a rule universally applicable is if one could will it applying universally so that others could likewise state it something that would provide a clear contradiction in the will. Parfit seems to think that this is an insufficient response since he thinks this is only a question of a calculated risk not of a real Law of Nature despite affecting to consider the application of the Law of Nature formula. Parfit thus does not really consider the application of the Law of Nature formula in this case at all.


A different objection that Parfit considers is what he terms the High Stakes Objection where performing some kind of act would be undertaken due to it being the case that not to perform would have unusually high consequences for the one in question. So, for example, unless I steal a particular drug from another who also needs it, I'll die. In such a case there is a particularly strong incentive to perform the act in question and this undercuts the appeal to the Law of Nature formula. In fact although it might well be particularly difficult to follow the advice of the Law of Nature formula here this example does not undercut it since, again, application of the maxim in question as a universal law of nature would also render me vulnerable in the same way and thus still be problematic as a universal law to will.


The next point Parfit makes is that the rejected rule is more direct in its impartiality than is the Law of Nature formula. However it is not obvious this is so. The rejected formula requires an act of imaginative identification which fails to include a real sense of agency in it whereas the Law of Nature formula, by contrast, in making its appeal directly to the conditions of agency, could be said to be more direct in its appeal. If what is meant with the claim of indirectness here is that the Law of Nature formula is less directly about the others affected the point is that it is only the ground of the duty, it is not the duty itself which is to the others directly but not directly about them in ground.


The next objection Parfit states is what he calls the Non-Reversibility one which is meant to say that in some cases there is a lack of symmetry between self and others that the Law of Nature formula does not capture. So: "we may know that, even if everyone did these things to others, no one would do these things to us". However this point views the universality tests as if they are meant to seen in terms of actual effects on us rather than as applying to the world in which we would live through the laws that would apply. Such laws would apply to how others related to us in general even if not in specific terms with regard to a given act. Seen like this I fail to see how a non-reversibility claim can be made. Parfit assumes that the racist case cannot be ruled out on the basis of the Law of Nature formula but this is only because he does not think of the racist formula as one that can be used by all races against each other and that this is what is precisely not being willed by the racist (who is therefore "begging an exemption").


Similarly Parfit rightly points out that in many cases privileged people are acting on maxims that allow them to be privileged in appropriate ways. However the point is one not of whether privilege as such is to be viewed as given but of whether behaviour is to rightly based on the ground of duty expressed in the Law of Nature formula. And this does not allow begging of exemption on the part of groups any more than on the part of individuals. So Parfit's examples of treatments of women or slaves are not counter cases to the Law of Nature as treating other groups as inferior on the basis of certain characteristics is not relating to them as rational beings who themselves have the capacity to state and govern themselves by universal laws. It is here that the wrong of such maxims resides.


Parfit fails to grasp this and thus thinks that a different type of rule to the Law of Nature requirement is needed to address specifically disadvantaged groups. This is why he refers to the kinds of imaginative identification that was involved in his reformulation of the rejected rule. This is taken to be preferable to Kant's insistence on a first-personal reference in the Formula of Universal Law. However the point of such reference is to bring out that it is a question of willing that is at issue and this is not involved as clearly in Parfit's reformulation of the "Golden" Rule. Parfit prefers to the Law of Nature formula a version of Moral Belief that is stated on the grounds of a proposal from Thomas Scanlon as follows: "It is wrong for us to  act on some maxim unless everyone could rationally will it to be true that everyone believes that such acts are morally permitted". This reference to moral belief has the advantage of bringing in a kind of publicity requirement although the way it does it is not notably preferable to the form of publicity involved in mandating laws to hold as laws of nature. Parfit thinks that the formulation from Scanlon addresses the situation of oppressed groups better but I have seen no reason to accept this.


Parfit's argument concludes with a suggestion that the best way of viewing the situation of rational willing is in terms of a further revision that is expressed as the Kantian Contractualist formula as follows: "Everyone ought to follow the principles whose universal acceptance everyone could rationally will". This formula assimilates Kant to Scanlon but leaves open all the questions about the relationship between the two.

Parfit and Kant on Impartiality

In my last posting on Parfit I looked at Chapter 9 of Climbing the Mountain, the first manuscript-length version of what has recently been published as On What Matters. In this posting I'm going to look at Chapter 10 of Climbing the Mountain where Parfit addresses the notion of impartiality, including, thereby, the relationship between the Formula of Universal Law and the "Golden Rule". 


This chapter opens with a description of Kant's view of beneficence indicating that the duty to beneficence is conceived of by Kant as a duty we owe to others on the grounds that we expect such conduct from others towards ourselves. This way of capturing the duty of beneficence leads Parfit to invoke the "Golden Rule" of the Gospels, famously formulated as do unto others as you would they did unto you. Kant is generally taken to have rejected this rule, a rejection based on citation of a footnote in the Groundwork:

"Let it not be thought that the trivial quod tibia non vis fiery etc. can serve as the benchmark or principle here. For it is, though with various limitations, just derived from the latter; it can be no universal law, for it does not contain the ground of duties to oneself, not of duties of love to others (for many a man would gladly agree that others should not benefit him if only he might be exempt from showing them beneficence), finally not of owed duties to one another; for the criminal would argue on this ground against the judges who punish him, and so on." (Ak. 4: 430n)

A few points are worth making about this citation, the first of which is that Kant does not here cite, as Parfit supposes, the Golden Rule but, rather, the so-called "Silver" Rule, which is a negative formulation to the effect, do not do unto others as you would not have them do unto you. This point is significant since the problems listed with the Silver Rule by Kant may well not transfer simply to the Golden Rule and, arguing on this basis, it has been suggested by some, that Kant therefore did not mean any direct criticism of the Golden Rule at all.


Kant introduces the cited footnote just after discussing the false promising maxim in relation to the Formula of Humanity, hence, not in connection with the Formula of Universal Law at all. So, when Kant claims that the Silver Rule is "just derived" from the latter, he presumably means by "the latter" the reference to consideration of the ends that are contained in the person of others. The subsequent point is that the Silver Rule is not a "universal" law which, given its negative formulation, is not that surprising a point. It will only operate to forbid conduct, not to promote it and when Kant goes on to the consideration of the third example in relation to the Formula of Humanity (to do with cultivation of talents) he introduces precisely considerations for thinking that there is a positive as well as a negative sense to this formula, thereby demarcating it from the Silver Rule. Kant next amplifies the point about the lack of universality of the Silver Rule pointing out that it is not possible to arrive at the "ground" of a number of duties from it. This argument repeats the point about the essentially negative character of the Silver Rule as it is expansive duties that cannot be arrived at on its basis including duties to oneself and duties of love. Both these kinds of duties are explored not in the Groundwork but in the Doctrine of Virtue where duties of love cover duties with regard to beneficence, gratitude and sympathy so the point clearly again is that such duties cannot be derived from a purely negative formula such as the Silver Rule. Similarly, with duties towards others generally, the Silver Rule is open to the kinds of problems Kant lists with regard to both maxims of indifference and the conduct of the criminal towards the judge.


Unfortunately Parfit conflates the actual rule Kant cites and distances himself from (the Silver Rule) with the rule from the Gospels (the Golden Rule) that Kant does not give and does not discuss. This is a shame, not least since, as the Golden Rule is stated in a positive form, it is not as evidently open to the same riposte Kant sets out against the Silver Rule. Parfit in fact goes further than pursuing the case Kant sets out against the Silver Rule as if it were an argument against the Golden Rule, he also turns Kant's case against another formula that was not at issue for Kant in the citation given, namely, the Formula of Universal Law. Textually problematic as this is Parfit opens with an account of the maxim of indifference and states that if people wish to be helped then they should also wish to help others. Having put the point this way as a case that the Golden Rule can enable to be stated against the maxim of indifference, Parfit argues that the maxim of indifference could be formulated in accord with the Formula of Universal Law although it cannot accord with the Golden Rule. This claim is problematic on a number of levels, not least that Kant at no point relates the Formula of Universal Law to duties but only refers the Formula of the Law of Nature to them. Not only is this so but the maxim of indifference is specifically rejected when related to the Formula of the Law of Nature on the grounds that it could not be "willed" to hold everywhere as a law of nature (the contradiction in the will test). This point is part of showing that the Law of Nature formula has two dimensions and that one of these involves a sense of consistency of willing, not merely of conception.


Parfit does note that Kant discusses this riposte of contradiction in willing but rejects it as inadequate on the grounds that it involves people wanting to be helped whereas it is possible that there could be people who never want to be helped and that with regard to them the reference to universal law will be insufficient. There are further problems with these remarks including the point that wishing to be helped is not an optional extra in human life even though there are situations in which some may be convinced they will never require help. As such the discussion involved in Kant's account of beneficence with regard to universal law includes a sense of what it is that has to apply to finite rational beings due to the nature of their finitude.


Parfit again recognises this point and formulates a response with regard to what he terms "rational" willing. Having recognised it he proceeds to reformulate the Golden Rule in such a way that it includes a similar recognition as appears in Kant's discussion of universal laws. This reformulation on Parfit's part renders it as follows: "We ought to treat others in ways in which, if we had the choice and were rational, we would choose that others treat us". This formulation is condensed further into a reference to what we would rationally choose. Having reformulated the Golden Rule in this way Parfit has significantly altered its scope and understanding. This is furthered when he extends application of the new rule not just to the "actual world" but also to imaginary cases.


Having reformulated and extended the Golden Rule in the ways indicated Parfit next considers cases that might be thought to present difficulties for it including one in which a racist claims to be applying the Golden Rule when refusing admission to his hotel to anyone not of his race. This can be done if the racist assumes only that the rule applies simply to those of different races and not to any of his own race and thus endorses such treatment as applicable universally given that his own race is always in the favourable position. In response Parfit claims that the racist in question has misunderstood the Golden Rule since the rule requires that one relate to others as if one might be in their position. So application of the Golden Rule requires imagining changes not just in other things but also, potentially, in ourselves so that we were as the others we treat now are. This gives the third formulation of the "Golden Rule" that Parfit has: "We ought to treat others only in ways in which we would rationally choose that we ourselves be treated, if we were going to be in these other people's positions, and if we would also be relevantly like them".


This conception of the Golden Rule involves a form of imaginative identification and in so doing is well on the way to being something like an impartial spectator model of moral reasoning. It is clear, however, that the reasoning and the model has reached a point very far removed from the formulation of the Silver Rule Kant criticised in the Groundwork.


Returning to the Groundwork text Parfit refers to the example of the criminal arguing with the judge concerning punishment that Kant gives towards the conclusion of the footnote discussing the Silver Rule. Parfit suggests that the rule Kant is here rejecting is more like a rule that states we ought to treat each other person "only in ways in which we would choose that we ourselves be treated if we were going to be in this person's position". This formulation, the one Parfit imagines Kant to be actually reading into the Silver Rule is like the third formula of the Golden Rule that Parfit has given in the sense that it involves a form of imaginative identification. It however does not include the sense of rational willing that Parfit has built into his third formula of the Golden Rule. Parfit indicates that the kind of identification he takes Kant to be involved in the rule he objects to is one that it is, indeed, right to reject. The reason Parfit rejects it, however, is that it implies a kind of egoistic position. In response Parfit gives another kind of formulation of the Golden Rule that cites it as stating that we "ought to treat other people as we would rationally choose that we be treated if we were going to be in the positions of all of these people, and would be relevantly like them". This extension of the process of imaginative identification clearly makes it impersonal since, if we are to be asked to think of what it would be like to be in the positions of "all" of them then no particular characteristics can be at issue (unlike in the earlier formula where Parfit was precisely building in particular characteristics to forestall the racist move).


The readings of the Golden Rule that involve the kinds of imaginative identification Parfit has in mind are formulated to prevent the kind of appeal the criminal is imagined to be capable of making working on Kant's conception of the Silver Rule. Parfit does, however, simply now adopt the impersonal reading of the Golden Rule and appears just to drop his earlier interpretation of imaginative identification. This is meant to show that punishment is quite capable of being morally justified. 


Parfit next considers the objection Kant makes to the effect that the Silver Rule provides no basis for duties towards oneself and recognises that failure to do this will ensure that others appear always to have priority over oneself, something that is implausible. Given this point Parfit further refines the Golden Rule to meet this point and arrives at another formulation of it. This formulation states: "We ought to treat everyone as we would rationally choose that we be treated if we were going to be in all of these people's positions, and would be relevantly like them". This point is meant to indicate that we are part of the "everyone" in question though, notably, it abstracts from the separateness of persons in precisely the manner Rawls famously objected to utilitarianism for doing.


Parfit next returns to comparing the "Golden Rule" in the formula in which he has now arrived at with the Formula of Universal Law and points out that both involve appeals to claims about what it would be rational for people to choose and include the basic egalitarian point that everyone matters equally (though the way the latter point is recognised in Parfit's "Golden Rule" is quite different from how it is recognised in the Formula of Humanity). The difference between them that Parfit registers concerns the manner in which thought experiments are conducted. The "Golden Rule" is meant to address the question, on Parfit's view, "what if that was done to me" rather than, as he takes to be required on Kant's Formula of Universal Law, "what if everyone did that"? 


Parfit takes the use of Kant's Law of Nature Formula to require reference to different possible worlds and to apply the laws in question as if they were ones everyone followed in these worlds. This differs from another kind of test Parfit attributes to Kant in which we ask what would happen if everyone had certain kinds of moral beliefs. Parfit next invokes the notion of an ideal impartial observer who is not involved in any events and judges as it were from the standpoint of the "universe" (as Sidgwick puts it in The Methods of Ethics) and he contrasts this with the "Consent Principle" that he previously formulated as the first part of the Formula of Humanity.


Parfit confesses, however, that the "Golden Rule" in his account of it faces the objection that it might lead us to ignore the fact that, in the "actual" world, ideal redistributions of benefits and burdens could produce irretrievable losses for some that, at least for them, are not compensated and that this matters when we are considering distributive justice. Due to recognition of this point Parfit thinks that the "Golden Rule" is, as he puts it, "theoretically" inferior to the Impartial Observer Formula and to the "Consent Principle" he earlier stated as the first part of the Formula of Humanity. The point Parfit should here have noted however is that the "Golden Rule" is not the only one of these principles open to this objection and, if it were, it would be not "theoretically" but "practically" inferior to the other formulas.


Parfit next considers some questions that he thinks create difficulties for Kant's discussion of universal laws. The first point considered here is that wrong-doing often involves acts that can only be rarely performed, a point Parfit formulates as the "Rarity Objection" and which is given an example in the case of someone who is willing to let another be punished for a crime that they committed. The question that is asked with regard to this person is whether their maxim falls foul of the Law of Nature Formula. Parfit suggests that the universalisation of the maxim of the person in question is plausible though, in making this point, he does nothing to test the maxim with regard to the criterion Kant uses in the Groundwork in terms of types of contradiction. In failing to refer to the tests in question Parfit simply pre-judges his case against the Formula of the Law of Nature. Similarly Parfit considers an egoistic maxim here and imagines a case where we have an egoist who could live in a world of other egoists, thereby ignoring the points made about this case when Kant discusses consistency of willing.


A much better case is made by Parfit when he moves from the discussion of rarity to one of "high stakes" and imagines cases where the perceived benefit of maxims for those considering them is very great compared to the alternatives. Such cases do show a ground for action in which it is hard to imagine people being persuaded to act in terms of what would be best overall given their own good is so great to them at the time but they do not count against a formal process in which abstraction from questions of the good is required. So again Parfit builds into his case assumptions that do not allow considerations of Kant's point.


The next objection Parfit considers is that the Law of Nature Formula is insufficiently impartial or not impartial in the "right way" unlike the "Golden Rule". The reason for this claimed difference is that Kant's rule does not build in a specific reference to the actor in question. Kant does, however, fail to do this for a specifically good reason that Parfit recognises which is to abstract away from the consideration that the egoist wishes to hold to. However Parfit suggests that this abstraction fails to recognise a key problem, that he terms "non-reversibility" where it is possible that we do to others something that no one is able to do to us in the same way (for whatever reason). This is, in fact, a problem much like the earlier racist suggestion Parfit considered and shows still that the case is being considered continuously in relation to material benefits that are not relevant to Kant's formula with conditions that do not allow for universality of application so, once again, beg the question against Kant's formula.


Parfit seems to think this "non-reversibility" claim is an important one simply due to the realities that something like it may well be involved in many cases of social injustice. So, for example, if men believe rape of women justified this may be because they are not women and not capable of being treated in the same way. However, even should such beliefs be actually held in such situations they are simply not relevant to the consideration of Kant's formula since, in holding to such beliefs the actors in question are not subjecting themselves to the standards of universality Kant is stating.


Parfit considers a response of this sort to his cases of "non-reversibility" but does so through Thomas Nagel's suggestion of a form of imaginative identification much like the one Parfit himself earlier built into the "Golden Rule". In responding to this point Parfit cites the response Kant made to the maxim of indifference in which Kant pointed to the contradiction in the will as being to do with what it would state about the one who formulated this maxim in a world governed by it. This point is taken by Parfit to indicate an illicit partiality in the formula rather than a consideration of the cases of all as in his own "Golden Rule". However the point surely is that cases of "non-reversibility" do not require the introduction of imaginative identification of the sort Parfit has invoked any more than Kant thought of such as involved in the response to the maxim of indifference. It was rather a question of reaching the maxim in question and formulating it as a universal rule that did the trick. Similarly, assuming something like a principle of "non-reversibility" is at work in much wrong-doing let's articulate its general assumption as something like the following: "I will take advantage of the ways in which others are differently vulnerable to myself in order to have mastery over them" and then think of its universal application. Are there are any cases in which this could not be applied to the disadvantage of the one formulating the maxim? It is enough to ask the question to point to the answer.


Parfit does not consider a response like the one I have given but he does mention Rawls' "veil of ignorance" notion only to suggest that Kant does not include reference to anything like it but, in so doing, he fails to discuss the ways in which Rawls interpreted Kant's typic of universal law in order to show considerations of the "veil of ignorance" type were at work there, thus, once again, he merely assumes his point. Similarly, Parfit considers Thomas Scanlon's interpretation which produces an account of universality that asks us to consider what everyone could rationally will only in order to rebut it by mentioning that Kant does not explicitly state things this way which is no answer to Scanlon. 


Parfit subsequently proposes to revise Kant's formulations in order to meet his notion of "non-reversibility" which is done by importing considerations of the sort Scanlon was concerned with when he referred to what everyone could rationally will. Having done so Parfit concludes the chapter with what he terms is the "Kantian Contractualist Formula" and which is stated as: "Everyone ought to follow the principles whose universal acceptance everyone could rationally will". Such a formula assimilates Kant to Scanlon.

Parfit and Kant on Universal Law

When I last looked at Parfit's 2002 Tanner Lectures, the "Ur-text" of On What Matters, it was to discuss the first of those lectures, and, not least, to examine how there Parfit interpreted Kant's Formula of Humanity. In this posting I'm going to look at Parfit's second 2002 lecture where he turns, by contrast, to the discussion of Kant's Formula(s) of Universal Law.


Parfit opens the second of his lectures with a discussion of the nature of Kant's references to "maxims" in his account of the rightness of acts. Parfit lists some statements that are described by Kant as "maxims" and then proceeds to describe what he terms Kant's "stated criterion of strict duties" which is: "it is wrong to act on maxims that could not be universal laws" though this seems to be based on Ak. 4: 424 where Kant in fact, by referring to at least two types of contradiction, refers to notions of "conception" and "willing" to distinguish them, a distinction that Parfit appears to make nothing of. Parfit looks instead at the notion that what is tested by the reference to universal laws is the "permissibility" of certain maxims (not their "possibility" of being experienced which is what Kant discusses). The notion of "permissibility" is not, in any case, taken by Parfit to be a very helpful way of understanding Kant's notion of universalisation and he raises questions about maxim-formation that have been circulating in the literature for some time in terms of the "test" of universality being either under-determined or over-determined. 


In the process of raising these points Parfit criticises ways of understanding universal laws that have been made by Onora O'Neill which culminates in a switch from the initial "stated criterion of strict duties" to a different "actual criterion of strict duties" which latter states that: "it is wrong to act on maxims whose being universally accepted, or believed to be permissible, would make it impossible for anyone successfully to act upon them". This criteria has the advantage that it appears to make questions of publicity emerge as key to the way universal laws are meant to work. However, it is not this, but instead the notion of "success" that Parfit fixes on, arguing, as he does, that the alleged maxim of coercing others wherever this would benefit me, even if adopted universally, would not be universally impossible to successfully achieve. This objection strikes me as a pretty odd one since such a situation would, if it was the basis of all known and avowed maxims, make rather a large number of actions impossible of success and is the basic reason why, in the Doctrine of Right, Kant can provide an argument for the need for a sovereign power.


Parfit does not consider such a case, however, preferring instead to look at Kant's arguments against lying and stealing which are, however, themselves interpreted by Parfit in ways that are pretty eccentric since he focuses primarily on the question of the egoist, which was not obviously on Kant's agenda at all here. Parfit makes a better point with regard to lying promises since it is here that he sees that the practice of promising is conditional on the notion that keeping them is the general rule to be observed. This leads Parfit to begin to consider the reference to universality in a different way to previously as based not merely on an account of maxims in relation to acts but also in relation to beliefs concerning acts. However Parfit still has difficulties with the kind of criteria that the reference to universal laws seems to involve since, as he can easily point out, there are cases where lying promises might seem to be required. 


The first lesson derived from the discussion by Parfit emerges at this point which is to the effect that relating maxims permissibility to a standard of universal successful accomplishment appears to make the universality criteria either much too weak or much too strong. It is only after this point has been stated that Parfit actually states Kant's Formula of Universal Law which is, however, given in a form that Kant never explicitly gives. The Formula in Parfit's telling emerges as follows: "It is wrong to act on some maxim unless we could also rationally will it to be true that this maxim is a universal law". Shortly after giving this formula Parfit also adds the "law of nature formula" and what he terms "the Permissibility Formula" which latter is said to state: "It is wrong to act on some maxim unless we could also rationally will it to be true that everyone is morally permitted to act in this way".  This latter notion is said to be derived from Thomas Scanlon and is freely admitted by Parfit to be one that is not generally recognised.


Nor does Parfit stop here, however, since after giving the "permissibility formula" he adds that Kant's basic assumption in this alleged formula refers to effects of beliefs people hold and thus depends on what Parfit calls a "Moral Belief Formula", something that he implicitly appealed to earlier in the lecture but which is now explicitly presented as stating: "It is wrong to act on some maxim unless we could also rationally will it to be true that everyone believes such acts to be permissible". This formula is, however, implicitly conservative since it appears to require recognition of actual beliefs and thus not to test them. 


As often, however, Parfit moves on to a different question than the one that seems obvious since his question about the alleged formulas given does not turn on any particular type of objection to their coherence but instead he goes back again to the question of the status of "maxims" for Kant. Here Parfit makes the point that when Kant speaks of "maxims" it seems that different types of things can be at issue. Thus, on some occasions it appears that Kant is using "maxim" to indicate the "policy" on which someone is acting, on others to the relationship between "policy" and "aim". Having made this point Parfit returns to the general problem of having a view of maxims that provides a serious wrong-making criteria and indicates that many maxims have a "mixed" status in the sense that it is not always bad to act on them nor always good. This is termed by Parfit the "mixed maxims objection" to the accounts of maxims that have been discussed by him up to this point. In response to it Parfit suggests a revision of Kant's Formula(s). The point of the revision is, however, a rather odd one. Having indicated that his basic problem is with "mixed maxims" Parfit suggests that the way that maxims are referred to should not be by means of policies or aims but instead by what persons are "intentionally doing". In fact, in stating this, Parfit seems unusually poorly informed since making the understanding of maxims refer to "intentions" is a proposal that has been widely considered and it has provoked serious objections from a range of philosophers.


Nor, in any case, is it obvious that what Parfit goes on to do is to discuss a revision that is really based on this notion of "intentions". It is instead Parfit's move to begin discussing rational willing and this appears to involve some sense of instrumental coherence, a notion that requires no explicit reference to "intentions" at all. 


Parfit states that Kant's Formula(s) of universal law work best when 3 conditions are satisfied, namely, that it would be (a) possible to act on the maxim; (b) the effects of the maxim would be much the same however many people acted on it; (c) these effects could be randomly or equally distributed between people. Understanding Kant's formulas in this way brings them close to considerations that are at work in standard decision theories and in games theories (much as Parfit invokes in Reasons and Persons). Parfit's claim is that the problems that arise for the Kantian formulas are related to failure of one of these conditions. The only failure Parfit explicitly considers is (c) which leads to burdens being imposed only on select groups. In raising this point Parfit gets to the notion of "impartiality" and indicates that there is no guarantee of impartial consideration in weighing universal acceptability. The example of this given is one that I find extremely unpersuasive though and appears to me to be a prime example of what is often termed "maxim-fiddling". 


The example is of a racist who takes it that there would be no problem of universalising maxims of segregation and unequal treatment. But in allowing the racist this claim Parfit neglects to conceive of the world that would therefore be being willed which would be a world in which there would therefore be no problem with the racist himself suffering the same treatment whenever they were in the minority. In failing to conceive of the general test in this way Parfit has simply fiddled the maxim in the racist's favour.


Parfit eventually gets to a version of this response when he considers Thomas Nagel's account of the maxims concerning benevolence which is the classic counterpart of the problem with contradiction of the will that I have transposed here to the case of the racist. Nagel appears to view the benevolence example as requiring us to be placed in "everyone else's position" but this is too strong and all that is needed is precisely what Parfit agrees is found in the discussion of benevolence in the Groundwork which is the failure of rational willing even in the case of the one willing non-beneficence. It is due to such failure that Rawls, as Parfit cites, invokes his idea of the "veil of ignorance" as a way of seeing the problem here. It is not that Kant has to have directly supposed that we do need to invoke this veil, as Parfit wrongly assumes. It is rather that the veil is an alternative way to make the same point in the same spirit.


In response to the discussion at this point Parfit invokes another way of understanding Kant's formulas of universal law in terms that recognises the question about impartiality in an explicit form. This leads Parfit to give what he terms the "Formula of Universally Willed Acts" which is as follows: "An act is wrong unless it could be rationally willed by everyone". This formula is an extension of the Consent Principle that Parfit found to be the first part of Kant's Formula of Humanity. It faces at least the problem Parfit himself mentions which is that Kant does not refer in this way to "what everyone could rationally will" though it is evident he does require some sense of rational willing to be understood. However Parfit concludes the second lecture with 2 final formulas. The first is meant to build in a reference to beliefs and produces what Parfit terms "Kant's Contractualist Formula" which is: "We ought to act on the principles whose universal acceptance everyone could rationally will". This version brings Kant into the vicinity of Thomas Scanlon though this point is not explored in this lecture.


Finally Parfit concludes with a formula that explicitly brings in a reference to Rawls' notion of the "veil of ignorance" and is stated simply as "Rawls's Formula" and is given as follows: "We ought to act on the principles that it would be rational for everyone to choose, as the principles that we would all accept, if no one knew anything about themselves or their circumstances".


Parfit concludes the second 2002 lecture with the claim that the third lecture will be devoted to consideration of the contractualist formula though it surely also considers its relationship to Rawls's Formula and indicates a resolution of some sort of the two formulas. However, before looking at the third and final 2002 lecture I will first consider the ways in which the discussion of the second 2002 lecture becomes transformed and reworked in Climbing the Mountain and the pre-publication version of On What Matters to become, eventually, Chapters 12 to 14 of the published book.