Sidgwick on Equality
In comments here, Pablo quotes the following from Sidgwick:
"[T]he very indefiniteness of all hedonistic calculations, which was sufficiently shown in Book ii., renders it by no means unlikely that there may be no cognisable difference between the quantities of happiness involved in two sets of consequences respectively; the more rough our estimates necessarily are, the less likely we shall be to come to any clear decision between two apparently balanced alternatives. In all such cases, therefore, it becomes practically important to ask whether any mode of distributing a given quantum of happiness is better than any other. Now the Utilitarian formula seems to supply no answer to this question: at least we have to supplement the principle of seeking the greatest happiness on the whole by some principle of Just or Right distribution of this happiness. The principle which most Utilitarians have either tacitly or expressly adopted is that of pure equality--as given in Bentham's formula, "everybody to count for one, and nobody for more than one." And this principle seems the only one which does not need a special justification; for, as we saw, it must be reasonable to treat any one man in the same way as any other, if there be no reason apparent for treating him differently." (The Methods of Ethics, 7th ed., London, 1907, pp. 416-417)
And then Pablo said of it:
"It seems to me that, contrary to Rawls's interpretation, Sidgwick is not claiming here that utilitarianism requires one to choose the more equal distribution of good when the total aggregate of good is maximal, but instead that, as a matter of contingent fact, those philosophers who have been utilitarians have also embraced a distributive principle which assigns (subordinate) moral significance to equality."
---
I had understood the Sidgwick quote somewhat differently. I'd be interested to know what others think.
I presume that his "as we saw" comment refers back to his argument on p380 (7th Ed.), where he says:
"the self-evident principle strictly stated must take some such negative form as this; 'it cannot be right for A to treat B in a manner in which it would be wrong for B to treat A, merely on the ground that they are two different individuals, and without there being any difference between the natures or circumstances of the two which can be stated as a reasonable ground for difference of treatment.' [the principle's effect is] merely to throw a definite onus pro-bandi on the man who applies to another a treatment of which he would complain if applied to himself; but Common Sense has amply recognised the practical importance of the maxim: and its truth, so far as it goes, appears to me self-evident. [...] the principle just discussed [is] more or less clearly implied in the common notion of 'fairness' or 'equity'"
So I take his overall argument on equality to be something like this:
If one treats two people differently, one must have some justification in mind (this follows from his argument on p381 about parts and wholes, and how they sum). Now imagine that one has a choice of three actions:
1) Person A gets huge benefit, person B gets very little.
2) Person B gets huge benefit, person A gets very little.
3) Persons A and B get equally middling sized benefits.
Further imagine that the total value realised in each is equal. Sidgwick's claim is that choosing (1) or (2) would require some further justification as to why A and B should be treated differently. So by elimination we should choose (3).
(For much of the book he tries his best to describe the theories in an impartial manner, and so often adopts the tone of a disinterested onlooker. I take this to explain why, in the passage Pablo quotes, he states his claim as "most utilitarians have..." and not "the most plausible claim is...".)
utilitarianism and equality
Thanks for this, Al.
1. I’m in broad agreement with your reconstruction of Sidgwick’s argument; I just don’t think it’s incompatible with the interpretation I made of it in my earlier comment. My point was to emphasize that, for Sidgwick, the adoption of a principle of equality to break ties between rival distributions of good is not dictated by utilitarian considerations and is not itself part of the utilitarian theory. As Sidgwick writes in the summary of contents, “We also require a supplementary Principle for Distribution of happiness: the principle of Equality is prima facie reasonable. “ (p. xxxiv; bold italics added) What I left out in my comment, and what you emphasize in yours, is that for Sidgwick there are in fact good reasons for preferring in such cases equality over any alternative distribution of good. But, to return to my point, for Sidgwick these are not utilitarian reasons. So those who do subscribe to a moral theory that requires agents (i) to maximize the good and (ii) to equalize it when any of two or more different alternative distributions would maximize it are best described as quasi-consequentialists, rather than strict consequentialists.
2. It is pertinent to quote the relevant passages from Rawls in A Theory of Justice:
“The correct distribution [on the utilitarian view] is that which yields the maximum fulfilment. […] In itself no distribution of satisfaction is better than another except that the more equal distribution is to be preferred to break ties.” (p. 23)
“A classical utilitarian, on the other hand, is indifferent as to how a constant sum of benefits is distributed. He appeals to equality only to break ties.” (p. 67)
(Both passages appear in the original 1971 as well as in the revised 1999 editions.)
In the light of this textual evidence, I’m no longer sure I was right when I claimed that, for Rawls, the more equal distribution is to be preferred on utilitarian grounds when aggregate good is maximal. So even if my interpretation of Sidgwick was right, what I said in the fragment you quote might still be inaccurate.
James Mill on equality
It would be interesting to trace down the relevant quotes on this subject by the main classical utilitarians. I don't recall this matter being explicitly discussed by Bentham or the Mills, though I do remember something James Mill said in a different context which could serve as an alternative way of specifying what the utilitarian should do when choosing between different outcomes whose aggregate good is maximal. Here's the quote:
"it seems impossible that such equality should ever exist. How is it to be established? Or by what criterion is it to be ascertained? If there is no such criterion, it must, in all cases, be the result of chance. If so, the chances against it are as infinite to one. The idea, therefore, is wholly chimerical and absurd." ('Government', para. 65)
On this line of reply, the very assumption that there could be ties to break is denied as having an infinitesimally small chance of ever being realized. I don't think this proposal would work, however, and for two reasons. First, a complete moral theory should be able to specify what ought or ought not to be done for any possible and not merely actual situation. Even if one grants that events with infinitesimally small probabilities never actually obtain, they are certainly possible and therefore complete moral theories should have a way of handling them (remember that we are discussing utilitarianism as a standard of rightness rather than as a decision procedure). Secondly, it is debatable whether goodness is, in fact, infinitely subdividable. If intrinsically good states supervene on natural states, and if there is only a finite quantity of matter, energy and consciousness in the Universe which cannot be divided ad infinitum, then it appears that there is a limit to how small the smallest quantity of good could possibly be. And if this is so, it is not true that the chances that two outcomes are perfectly equally good are infinitesimally small.