Larry Temkin's Rethinking the Good may be described as a book about the values of alternative outcomes and the trade-offs between them. Much of it is about large scale outcomes, the kinds of social matters that utilitarianism was once thought most effective in handling. But while there is much reflection about aggregate outcomes here, core normative issues are most centrally at issue. Rethinking the Good does not directly rethink good – the property goodness – or the good, the kinds of things which may be said to be good. What it does is examine our thinking about what outcomes are good in the context of practical reasoning. The implications of doing so shall, Temkin suggests, force us to rethink the property of goodness and the things that are good.
The most striking feature of the book as a whole is the length to which Temkin goes to make things clear. The reader is guided throughout, is seldom at a loss, and knows what's going on at each and every step. The main text of the book covers five hundred and nineteen pages. In addition, there are fifty-two pages of appendices. The last appendix is a handy fifteen page set of chapter summaries. This section is a helpful place for the merely curious reader to start. Certain topics arise several times in the book. Parfit's Mere Addition Paradox is one example. Each time it is reintroduced, it is carefully laid out. This feature of the book allows readers just interested in one or two topics to encounter them well-informed.
All the topics of Rethinking the Good hinge on a central matter, a kind of puzzle which Temkin discovered many years ago. Call it ‘Temkin's problematic’. Several of his previous articles addressed the puzzle and aspects related to it. And at numerous points it becomes clear that much of his thinking on it has been refined over many years through his talks, discussions with other ethicists, comments on previous drafts, and a great deal of correspondence. Recognizing that I can't do full justice to the problematic here, let alone its many implications, I'll briefly sketch it. Temkin previously thought that its casualty was the Axiom of Transitivity, that ‘all-things-considered better than’ is a transitive relation. That is, ‘the transitivity of “better than” entails that if, all things considered, A is a better outcome than B, and B is a better outcome than C, then, all things considered, A is a better outcome than C’ (9). Most everyone is apt to go along with this claim, often believing it is obvious or analytic, or something along such lines. And numerous examples will not shake our belief in the Axiom. So, for example, if I judge that a vacation in Miami is better than a vacation in Savannah, and that a vacation in Savannah is better than a vacation in Milwaukee, then it seems that I am committed to holding that a vacation in Miami is better than a vacation in Milwaukee.
So far, no problem. But once we consider two highly plausible principles about outcomes, it comes into view. Briefly stated, they are: ‘Trade-offs between Quality and Number are Sometimes Desirable’ and ‘Trade-offs between Quality and Number are sometimes Undesirable Even When Vast Numbers are at Stake’. The first principle, which Temkin calls The First Standard View, has historically been best represented by utilitarianism. This principle would recommend, for example, that a poor country strongly ruled by a highly cultured aristocracy transition to a liberal democracy which is open to foreign investment. The majority of people would improve in freedom and in their standard of living, and their gain would be worth a diminution of the highly cultured and refined life of the aristocrats. The second principle, The Second Standard View would ask, for example, that the majority of people sacrifice a small amount of well-being in exchange for saving a small minority from experiencing incredible suffering. To use Temkin's own Lollipops for Life case, The Second Standard View prescribes that each of us refrain from one lick of a lollipop in order to save one person from suffering ‘unbearable agony for fifty years before eventually dying a slow, lonely, torturous death’ (34), even if utilitarianism informs us that in the alternative case, if we each had that additional lollipop lick, total utility would be greater.
What connects these two principles in Temkin's problematic are what he calls ‘Spectrum Arguments’. Consider a nine item spectrum of cases that he employs:
A, 10 people go through life without two arms and a leg
B, 50 people go through life without one arm, a leg, and a hand
C, 200 people go through life without an arm and a leg
D, 800 people go through life without an arm and a foot
E, 2,000 people go through life without an arm
F, 6,000 people go through life without a hand
G, 20,000 people go through life without three fingers
H, 100,000 people go through life without a finger
I, 400,000 people go through life without the top third of one finger (50).
Here is what Temkin says about this particular spectrum of cases: ‘As described, I think most would agree that for “near” members of the preceding handicap spectrum (A and B, B and C, C and D, etc.), the First Standard View would apply, so that trade-offs between quantity and number would be appropriate’ (51). But considering the whole spectrum leads him to say: ‘If “better than” is transitive, this entails that A is better than I. But most would, I think, reject this. Appealing to the Second Standard View, many would claim that the difference in severity between A's and I's handicaps is so great, that for such outcomes trade-offs between quality and number are undesirable’ (Ibid.). Temkin's problematic, then, is seen when we apply the Axiom of Transitivity to a spectrum of cases in which at one end of the spectrum The First Standard View applies, and at its other end The Second Standard View is in force.
Where Temkin's old view was that transitivity was in jeopardy, he now holds that we have a number of ways to deal with the puzzle. And incidentally, the normative dimension of the outcome at issue need not be well-being, but may be freedom, equality or some other property with value. So Temkin's problematic, as applied to freedom, affords us four possible solutions: ‘one must reject the First Standard View for Freedom, the Second Standard View for Freedom, the plausible assumption that there could be a spectrum of alternatives whose members were related in the manner described, or the view that “all-things-considered better than” is a transitive relation. Here, as before, something has to go, but for most advocates of freedom, giving up any of these views will not be easy’ (62).
One of the most attractive features of Rethinking the Good is Temkin's openness to a variety of solutions to the problematic. This book is an exemplar of Russell's ideal that philosophy should be in the business of considering theoretical possibilities, not laying down what must be. And I think a careful consideration of Temkin's problematic will have us agree that it is unclear which way lies the solution. But it might be wondered whether there is a problem at all. Reading through the book it takes some work to see that there is one. This leads us to recognize another admirable quality of Temkin's effort: he takes great pains to avoid any misunderstanding of the problematic. Given all of the feedback he's had about it, he is extraordinarily responsive to the many ways in which people might think the problem may be dismissed. And in fact, a great deal of the book is spent on this very task. To illustrate this, Temkin gives two chapters, consisting of seventy pages, to exploring the nature of transitivity. But the depth of this thoroughness and the extent of groundclearing needed to frame the problem gives rise to a suspicion. If it requires so much work to set up and clarify the problem, perhaps the strong intuitions upon which the problem is supposed to be based are not that strong, after all.
This is not to suggest there is no problem there at all. There does seem to be one. Careful reflection on it, I think, will lead us to a recognition that we are not equipped to deal with it at this particular time. On the first page of his introduction, Temkin admits this: ‘all plausible theories of practical reasoning must give an adequate account of the assessment and ranking of outcomes. Unfortunately, an adequate account has not yet been given; nor is one on the horizon’ (3). This puts Rethinking the Good in a peculiar position. Its focus is a problem that we must recognize we don't know how to address, let alone possess the means of solving. In sticking to his goal of exploring the problem and considering possible routes to solving it, Temkin refrains from offering the kind of grand normative theory that might have some hope of making sense of it all.
What can we take away from this book? There are many things, but I will be brief. It shows quite clearly that our grasp of how to treat large scale outcomes, and the trade-offs they involve, is tenuous at best. Next, we come to appreciate that in normative theory about large outcomes some form of pluralism is likely right. No normative monism can possibly match the various, and often divergent, intuitions we have about outcomes. Lastly, the confidence which was once shown in utilitarian reasoning about big outcomes, and the subsequent confidence by others in Rawlsian reasoning about them, was, to more of an extent than previously thought, a function of a lack of alternatives considered, not an indication that they were somehow inevitable ways of thinking about large scale outcomes. Rethinking the Good is a book which, long after we have put our quest for a supreme principle aside, and after many years of rolling up our sleeves and doing the work to devise viable pluralisms, we can revisit to test how well we've done.
