Summary of de-Shalit, Chapter 3

Sanjay Lal

Phi 646

In chapter three of Why Posterity Matters, entitled the "Utilitarian theory and the not-yet-born", Avner de-Shalit questions the claim that the utilitarian approach is suitable for discussions about intergenerational justice. Before offering his criticisms of that claim, de-Shalit gives us reasons why he thinks the claim is appealing (pp.67-8). These reasons are: (1) The utilitarian theory is supposedly impartial toward when one exists. (2) The utilitarian theory considers the consequences and effects of an act or policy (and therefore is consistent with how "most" environmental policy is arrived at). (3) The utilitarian theory permits us to adopt our acts to changing circumstances as long as we adhere to the basic principle that the moral act is that which brings about the greatest happiness to the greatest number of people (thus we supposedly have a guarantee that with this theory a just intergenerational policy will be relevant in the future). (4) By not regarding acts as intrinsically good or evil, utilitarianism seems to enable us to consider policies right in one set of circumstances and wrong in another.

de-Shalit’s criticisms of the utilitarian approach to questions of intergenerational justice is broken down under two headings in chapter three: how the approach pertains to questions of population policy (p. 68), and how the approach pertains to questions of intergenerational distribution (p. 77). de-Shalit’s first major criticism is direct toward average utilitarian (p.69-70). de-Shalit sees an "absurdity" follow if one uses the principle of average utility to determine whether or not they should procreate a child who will be happy but somewhat less happy than the average (p. 69). According to de-Shalit, "Average utilitarianism...directs us not to conceive the child in that case, because average happiness would be reduced if we did so." (p. 69) de-Shalit finds that recommendation "misguided and unacceptable" (p. 69) since the child in the considered case is going to be happy as will his/her parents (who already meet the criteria of being at least average in their happiness) the result of the procreation of such a child will be "at least as good as the original state of affairs, if not better."(p. 70)

Furthermore, de-Shalit maintains that average utilitarianism could lead us toward a very cruel and slippery slope like conclusion about population policy: the elimination of unhappy people (p. 70-1). Such a means of increasing average happiness is, de-Shalit maintains, one alternative average utilitarianism "finds difficult to dismiss of reject" (p. 71). de-Shalit writes: "We have seen that, if we do not add unhappy children to our society, then average happiness will increase, assuming that we have happy ones. While this sounds plausible to many, the more radical way of increasing average happiness is by decreasing the number of unhappy people in our society by getting rid of some who are miserable." (p. 71) What’s more to de-Shalit is that since among those who have average happiness in a given society some will be more happier than others (p. 71), and since the discarding of the less happier members of a society may not affect the utility of the more happier ones, average utilitarianism can lead to a slippery slope in which not only unhappy people can be eliminated but so can those less happy than others (p. 71). Such conclusions are unacceptable to de-Shalit.

Total utilitarianism does not fare much better than average utilitarianism, in de-Shalit’s estimate, with regard to questions of population policy (pp.71-ff.). An immediate problem that de-Shalit sees with the total utilitarian approach to population policy is that following the approach, according to de-Shalit, leads to the conclusion of maximizing the number of people on earth—a conclusion out of step with sound population policy (p.71). De-Shalit writes, "If every child that exists is happy (assuming that it is almost always better to live than not to, so that every child has some degree of happiness), then we have to give birth to the maximum number of children." (p.71) In other words, if our aim is to maximize overall happiness among the population there seems to be no better way of fulfilling that aim, as far as de-Shalit is concerned, than through maximizing the overall population. Moreover, the total utilitarian defense that total utilitarianism obligates us, not to produce the greatest number of people, but to bring about the greatest happiness among those who are already existing in the population is, to de-Shalit, a defense that "steps beyond utilitarian limits" (p.73). de-Shalit sees that to be the case because the defense "bars the entry of non-members (to the existing population)" (p. 73) and thus goes against the utilitarian values of universal ethics and impartiality.

de-Shalit turns to questions pertaining to utilitarian theory and intergenerational distribution (p. 77) and reaches conclusions equally negative to the ones he reached when considering utilitarian theory and population policy. Two big problems that de-Shalit sees for utilitarian theory in this context is determining who the affected people are in our attempt to maximize the utility of all who will be affected by our actions, and not having tools for measuring or calculating person utilities (p.77).

In addition to these perceived problems for utilitarian theory and intergenerational distribution, de-Shalit proceeds to offer at least one more by respectively considering supposed problems for objective and subjective utilitarianism with regard to the issue under discussion (p. 79-84). If we follow objective utilitarianism—in which we relate utility to some object that we believe worth maximizing—we could very well, according to de-Shalit, be dictating our values, desires, and norms to future generations (p. 79).

de-Shalit writes, "If we contemporaries decide what ‘utility’ stands for, what the objects of maximization are, then we will, in effect, be imposing our tastes and values on future persons." (p. 79) de-Shalit never tells us why such "imposing" is necessarily a bad thing or how one can get around its seemingly inevitable nature, but seems to suppose that such "imposing", in some way, goes against reasonable moral attitudes toward future generations.

de-Shalit turns his attention to subjective utilitarianism (p. 81)—utility that is related to the fulfillment of people’s desires—and finds this kind of utilitarianism to also provide an inadequate approach to issues of intergenerational distribution. de-Shalit writes, "(I)f our utilitarian follows subjective utilitarianism she must acknowledge the fact that she has no tools whatsoever to predict the tastes of the not-yet-born, and therefore has no means of deciding about utilities." (p. 81) The distinction some utilitarians have offered between tastes and preferences—in which tastes remain constant over time but preferences don’t—in an attempt to get around the supposed problem of not having tools to predict the tastes of the not-yet-born will not work as far as de-Shalit is concerned (p. 82-4). de-Shalit gives examples from the sociology of religion to show that tastes do, in fact, change over time (p. 82). Since religious affiliation is not subject to the demands of price controls and other aspects of market forces, de-Shalit’s argument goes, it is more accurate to think of religious participation as a taste than a preference. However, examples like the decrease in religious Jews from 50 years ago as well as the dramatic decrease in the birth rate of the Roman Catholic community in Quebec indicate to de-Shalit that taste does not remain constant over time and thus no tools exist for predicting what the tastes of the not-yet-born will be (p. 82-3).

de-Shalit concludes his criticisms of the utilitarian approach to intergenerational justice and distribution by posing a dilemma to advocates of the utilitarian approach (p.84-86). Since there will always be more people in the future than exist in the present and since the utilitarian approach directs to act in a way which brings about the greatest good for the greatest number of people the approach is either too demanding on those living in the present or is too irresponsible toward those who will live in the future. For it is only by sacrificing to the fullest measure possible can a present generation ensure that they are acting in a way that will bring about the greatest good for the greatest number of people (all those yet-to-be-born). If the utilitarian claims that only the greatest good for the greatest number of presently living humans is what utilitarianism directs us toward, obviously we—according to de-Shalit—have no obligation under utilitarianism to act toward justice in regard to those yet born.

 

Comments by John Nolt

It should be noted that de-Shalit expresses a common misunderstanding of utilitarianism when he says on p.67 that according to utilitarianism it is sufficient justification of a policy that it produces more happiness than pain, for this is not the same as maximizing utility. There many be many policies that produce more happiness than pain. What utilitarianism recommends is that we choose the one with the greatest balance of happiness over pain.

Sanjay speaks of de-Shalit's conviction that averaging utilitarianism leads down a "slippery slope" to elimination of the unhappy or the less happy. I don't see this as a slippery slope. It is just a straightforward implication of averaging utilitarianism in situations where eliminating the less happy does not decrease the happiness of the remaining people.

Sanjay writes that "An immediate problem that de-Shalit sees with the total utilitarian approach to population policy is that following the approach, according to de-Shalit, leads to the conclusion of maximizing the number of people on earth—a conclusion out of step with sound population policy." Note, however, that this conclusion does not follow from total utilitarianism alone, but only from total utilitarianism given certain possible scenarios. Here is a numerical example (which also illustrates the difference between total and average utilitarianism):

Suppose (ideal assumption) that there are two and only two differences in utility between any population of n individuals and a population of n+1:
    1. In the population of n+1 there is an additional positive quantity X of utility that is not there in the population of n (because the life of the extra person is worth living)
    2. But in the population n+1 there is also an additional collective negative utility Y (as a result, say, of competition for scarce resources) such that:
a X + Y > 0 (Total utility is greater when we add a person.)

b X + Y is less than the average utility in population n (Average utility is reduced when we add a person)

 So long as the average utility in population n is positive, these assumptions can be satisfied -- i.e., 0 < (X + Y) < average utility in population n.

EXAMPLE: Consider a population of 1000 (on an isolated island, say), with an average utility of 10000 units. (This makes the total utility 10,000,000 units). Suppose that one more person adds X = 9999 units of utility and the rest lose 1 each for a total of Y = -1000.

Thus the total utility is increased by (X + Y) = 8999 when we add a person.

But average utility decreases. Average utility in n+1 is now 9999. To see this without calculation, note that each person of the original n loses 1, so their average is now 9999. We add another person whose life is worth 9999. This gives a total average of 9999.

If our ideal assumption holds over the whole range of values, we get absurd results on either form of utilitarianism:

 Sanjay notes that de-Shalit writes, "If every child that exists is happy (assuming that it is almost always better to live than not to, so that every child has some degree of happiness), then we have to give birth to the maximum number of children." (p.71) But we should note that this conclusion doesn't follow unless we assume that the birth of this child causes no collective loss of utility to others (the negative utility Y in the example above).

Sanjay observes that "Two big problems that de-Shalit sees for utilitarian theory in this context is determining who the affected people are in our attempt to maximize the utility of all who will be affected by our actions, and not having tools for measuring or calculating person utilities." With regard to the first of these problems, here is de-Shalit's argument to refute this alleged implication of utilitarianism (see pp. 77-8):

1 The morality of an action depends only on the utilities of the people who would exist if the action were performed (reductio assumption)

2 If I were to allow a beggar to starve, the beggar would as a result of my action not exist.

So 3 The morality of the action of allowing a beggar to starve does not depend on the utility of the

beggar (which is absurd). Thus, says de-Shalit, we should reject 1.

But de-Shalit ignores the fact that the beggar would have existed though my action caused him to starve; what Parfit and other utilitarians intend by 1 is that the morality of an act depends only on the utilities of all those who would ever have existed if the action were performed. De-Shalit's objection therefore misses the point.

De-Shalit seems to take Parfit's depletion vs. conservation example to show that according to utilitarianism we have no duties to future generations (which of course not what Parfit concludes at all). (74) Here's an effort at analysis of de-Shalit's argument:

1 No matter what policy we enact, after a 200 years entirely different people would exist than if we had chosen any other policy. (Nonidentity thesis)

2 We cannot harm people who will never exist.

  So 3 No matter what policy we enact, we can’t harm anyone who would have lived more than 200 years later had we chosen a different policy. (1,2)

4 We can’t harm a person by choosing a policy the only alternatives to which insure that the person doesn’t exist.

So 5 No matter what policy we enact, we can’t harm anyone who then lives more than 200 years later. (1,4)

So 6 No matter what policy we enact, we can’t harm anyone who then lives more than 200 years later or who would have lived more than 200 years later had we chosen any other policy. (3,5)

7 We have duties only to people we can harm.

So 8 We have no duty to anyone who lives (or might have lived) more than 200 years after we do. (6,7)

So 9 We have no duties to future generations (after 200 years). (8)

 But premise 4 is false; in horrendous situations it is better not to have lived (unless we assume the value of life itself to be infinite) and so it is in fact worse for folks that they live. There is also a question about 7: we have duties also to people whom we can benefit. Also the move from 8 to 9 is invalid. We need not suppose that morality is person-regarding.

A few additional observations:

I took de-Shalit's definition of objective utilitarianism to be as follows:

Objective utilitarianism: we ought to maximize happiness conceived as real, universal or objective happiness; satisfaction of "perfectly prudent preferences" De-Shalit says that Parfit shows that Q applies only to same-number decisions. (75) This is wrong. Parfit restricts Q to same-number cases and wonders how to generalize it: The Same Number Quality Claim (Q): If in either of two possible outcomes the same number of people would ever live, it would be worse if those who live are worse off, or have a lower quality of life, that those who would have lived. (360) What Parfit shows is only that we can't interpret "worse off" either exclusively as on average (on pain of hells or mere addition paradoxes) or exclusively as in total (on pain of the repugnant conclusion)

De-shalit claims that the notion of non-person-regarding happiness makes no sense because there is no happiness without people. (76) This is bogus. A non-person-regarding theory simply takes no account of whether the same people or different people exist in alternatives to choices. It need not assume happiness apart from people.

Note de-Shalit's misuse of the term "argument" at the top of 77.
 

John Nolt's Course Listings
John Nolt's Home Page