De-Shalit summarizes the presumptions of contractarians (hitherto referred to as C), which can be used to support "a theory of distributive justice" (p. 87). He claims they hold that people are "rational" and "self-interested", and they have identities apart from society. C’s hold that people have some natural rights, but they are willing to forgo or restrict them for the sake of security and/or economic advantages, and to promote justice. The ultimate goal for the C is to justify social co-operation, according to de-Shalit.
De-Shalit claims that he will concentrate on "the idea of one generation having obligations to future generations, derived from the rationale of a hypothetical agreement" (p. 88). He argues that the contract notion is implicit in the contention that as the past did so for us, so we do so for them, and so we are "committed to maintain the chain" (p. 88). He focuses on the work of two C’s, Gauthier (hitherto referred to as G), who holds the ‘mutual-advantage’ theory, and Rawls (hitherto referred to as R), who subscribes to theories of ‘impartiality’. Both G and R share notions of justice, says de-Shalit, yet he questions whether justice can be obtained via rational agreement in cases where there are conflicts of interest.
To sum up the underpinnings of G’s theories, de-Shalit provides the following list: justice is a necessity and a compromise; it is in one’s self-interest to be just (and one is motivated accordingly), and as co-operation is the most prudent way to get what one wants, it is therefore the most rational; the question of distribution is what one can reasonably expect given one’s original bargaining position. Thus, G allows differences of strength to play a major role in his theory. R, on the other hand, is characterized by the following: justice is based on "an impartial attitude on the part of all relevant persons"; no personal advantages, as one must accept one’s lot, since this is the reasonable result of impartiality; just solutions are not based on personal advantage. Thus, strength of parties has no role in R’s theory.
De-Shalit first discusses G’s ‘mutual advantage’ theory. The allocation of goods is based on bargaining principles, where the agreement in distribution of goods rests on agreement of concessions. One will concede as low as possible, and as equally to others as possible, for maximum gain (g’s ‘minimax principle’). But how can there be "intergenerational bargaining", without social choice (as future generations do not exist), asks de-Shalit (p. 90). The solution is that our interactions with the young and the old today creates a ‘continuous thread’, so that our agreements remain constant from one generation to the next. The whole question of obligations to future generations is sidestepped as contracts are intertwined between existing generations, creating a "chain of generations" and the need to continue it (p. 91).
There are three "circumstances of justice" that de-Shalit argues the C holds, of which he only discusses the first, or ‘equality of power’ (‘moderate scarcity’ and ‘impossible to ascertain’ are ignored in this analysis). He claims the C needs these conditions met, for the theory to be mutually advantageous. He gives a Rawlsian definition of equality of power, and proceeds to criticize how this concept does not apply to distant generations (but, as one will recall, neither does de-Shalit’s own theory). Ironically, it seems that it is the vulnerability of future generations, says Goodin, which gives rise to our obligations. Thus, according to de-Shalit, this first circumstance of justice cannot be met, as there is a vast "asymmetry of power between the generations" (p. 93).
Another equality of power claim is rejected by de-Shalit. This claim holds that future generations can reflect back and criticize past generations. However, this does not eliminate the significant power present generations have to affect future generations (such as by polluting the air and water). Indeed, this is a " ‘different’ sort of equality of power", but it clearly violated the Rawlsian definition of equality (p. 93).
The C poses that we can neutralize and compensate future generations by the fact that some aspects of our lives will be worse off and others will be better off. The superiority of the first generation over the second generation is compensated by the second generation being superior to the third and so on. This line of reasoning, however, does not account for the still inequality of power intergenerationally, according to de-Shalit.
He next analyzes the problems with equating the reason for remaining in the contract to that of entering into the contract. A contract must contain two factors: first, it must have optimal outcome over any other possible contract; and second, it must be better than not having a contract at all. The state of nature in which no one saves results in a Hobbesian conception, says de-Shalit. A contract will entail the preservation of the species, namely, protection from "extravagant predecessors" (p. 95). But why once we enjoy the benefits of the past, ought we remain in the contract? The need to continue the chain does not seem to be present. Reiteration does not seem to work, as we merely want others to concede so that we can gain. We don’t care about them, or whether they are worse off than we are. Nor can we force future generations to stay in it, as there are no ‘intergenerational sanctions’ for violations.
The notion of reciprocity is explored by de-Shalit, as a possible reason for joining and remaining in a contract. Reciprocity comes in two basic forms: mutual advantage or fair play. Regarding the former, which entails mutual concessions for mutual benefit, we might recognize that this is good for both parties existing at the same time. However, we can’t apply this consistently to future generations, as they cannot benefit us, and we can only benefit them (and thus it is not dual directional). Remember, rational behavior entails that we act in our self-interest, and thus there doesn’t seem to be any benefit for our maintaining a contract with future generations.
The latter notion of fair play seems too abstract, according to de-Shalit. This notion dictates that we ought give something to the future as there are those who gave to us. De-Shalit wonders if this implies that were we to have been treated unjustly by our past that we ought to treat the future unjust as a matter of fairness. "The deeds of previous generations", says de-Shalit, are not necessarily the foundations for our "obligations to posterity" (p. 99). Further, why can’t I instead, ponders de-Shalit, reciprocate past deeds to my contemporary (ies), instead of someone in the future? He concludes that the notion of fair play does not create a strict obligation to the future only.
De-Shalit moves on to discuss R’s ‘Saving Principle’ (or SP for short). He begins by giving a brief description of the context of the ‘original position’, in which participants stand behind a ‘veil of ignorance’ so that the "situation of the least advantages will be maximized", commonly known as the ‘maximin strategy’ (p. 99). The main benefits to this C method include equal liberty for all and that the difference principle (or DP) justifies inequalities. De-Shalit observes that R’s recognizes environmental concerns only insofar as we "harm future generations" but they "in general lie outside the scope of the theory of justice" (pp. 99-100). The ‘saving principle’ is proposed to "compensate future generations" from the damage we do to the environment (p. 100). Saving by advancing "policies designed to improve the standard of like of later generations of the least advantaged", is accepted as it prevents the "social minimum" from falling below a justifiable level (p. 100).
Concern for one’s offspring seems to be the motivation behind the SP. this psychological motivation provides "the reason for entering the original position in the first place", says R (p. 101). But this notion violates impartiality when applied intergenerationally, says de-Shalit. There are three elements that need to be examined: primary goods; the motivation assumption; and, whether the extension of the OP can possibly include future generations. He discusses this first element in some detail. De-Shalit claims that C must consider what principles future people would reasonably accept, lest they veto them and not accept the contract. The C does not consider personal preferences or tastes, so we do not require knowledge of the real interests of future persons. The ‘abstract’ person is considered to be one who "knows that her ability to achieve whatever she wants or desires is greater if she has more primary goods", according to R (p. 101). Primary goods are universal, in that we all can be said to need them. What, however, are they based on? Two concepts of primary goods are dismissed by de-Shalit. The first is empirically based, ‘whatever people consider them to be’, but this violates neutrality and does not allow desires to change over time (as they do given change in the environment, technology, and social structure). The second is that primary good are based on "the idea of the person" and one’s ‘moral personality’ (p. 102). R provides a list of primary goods, but De-Shalit criticizes it as not transcending temporal considerations, as it imposes a specific idea of the good upon future generations. Ultimately, de-Shalit says that the notion of primary goods is dependent on a "specific temporal and social environment", and thus will not "be appropriate to the circumstances of future life" (p. 102).
He next considers the problems of the psychological motivation R proposes, which is to account for the knowledge that people in the OP are contemporaries. R uses the model of the family "to further the welfare of the nearest descendants" and de-Shalit notes that R needs this sense of the community to account for our obligation, as does himself. This, however, results in R considering something of value beyond his original assumptions, and thus he is not merely imposing what is just, but also what is good. Further, there is no guarantee that this attitude of the value of family life will be held by future generations and even worse, by all contemporaries and it is thus not universilizable. It is not clear if we should abandon the notion of individuality in favor of the concept of the family. It is also unclear whether we ought consider other institutions, such as nations. R may succeed in arguing that "obligations to contemporaries with respect to future people exist" but this is "a far weaker claim than to claim obligations to future people", according to de-Shalit (p. 107).
His last consideration is that of who participates in the OP. Were it future people, then given the psychological motivation, it seems that only some future people will be considered (such as those with ‘family ties’). If only contemporaries were in the OP, then "an anti-environmental" or "no-saving policy" would be adopted as "it would be better for contemporaries not to save" (p. 108).
De-Shalit discusses how Jane English and DAJ Richards try to salvage the C theory. English says we can open "the original position to all persons ‘through history’ " and claims the DP in R’s theory implies saving (p. 109). When welfare is declining, the DP says we would save for the sake of the least advantaged. De-Shalit says that this implies we save for the sake of only the least advantaged of our future generations, and that the DP cannot account for saving for all future generations. English takes another stab at this and claims that in states of progress, one would want to raise the level of the least advantages, and thus the rich would be required to invest for future generations. But de-Shalit questions the motivation of these rich, and why one would feel motivated beyond their original position. (I wonder, however, if they would indeed be motivated, as they do not know they are rich in the OP.)
Richard contends that those in the OP must consider all generations, as which generation one is in is presumably ‘behind the veil’. De-Shalit thinks this entails leaving the not yet born out of the decision making process, when they ought to take part. He further claims that these decisions will affect whether one lives or not, but I say so would any policy decision, and it is not clear how that matters. Ought the C theory claim that it is our duty to ensure that those who do exist, exist as well as possible, whomever they might be? If this is the case, however, the C is possibly reduced to Parfit’s "Theory 7", in which only quality matters, which we have already seen refuted by the two hells.
Comments by John Nolt
Carrie's summary is in most respects quite thorough, but I think it does not fully clarify the point of de-Shalit's discussion of mutual advantage theories. In his criticism of mutual advantage theories, I see de-Shalit as arguing that there are four disanalogies with traditional same-generation contractarian theories that undermine intergenerational theories:
Carrie accurately reports de-Shalit's account of Rawls' difference principle, but de-Shalit himself does not get it quite right. (I learned this from a conversation with David Reidy.) Rawls' difference principle is the idea that unequal distribution is permissible only if it is to everyone’s advantage, and within schemes that accomplish this we should choose the one that provides the greatest benefit to the least advantaged.
Note that the reason why concern for one's offspring justifies Rawls' saving principle is that Rawls assumes the original position choosers to be heads of families—an assumption not justified by his supposedly impartial method.
Carrie rightly questions de-Shalit's criticism of Jane English who claims that "in states of progress, one would want to raise the level of the least advantaged, and thus the rich would be required to invest for future generations." But de-Shalit says, "She must show that the rich have a motivation to invest in a project that will bear fruit only a generation or more in the future." (109) Carrie aply replies: "I wonder, however, if they would indeed [need to] be motivated, as they do not know they are rich in the OP [original position]."
I think Carrie misses the point of de-Shalit's criticism of Richards (p. 110). Richards advocates extending Rawl's veil of ignorance to cover future generations, so that those behind the veil do not know whether they will live now or in the future. De-Shalit's objection to this idea is not that "this entails leaving the not yet born out of the decision making process," as Carrie says, but rather that it raises once again Parfit's question of whether a theory ought to be person-affecting. Richards wants to include behind the veil only those who will actually live. De-Shalit says that this is unfair because out decisions affect who will be born, so we must include all who might live. But this, thinks de-Shalit, leads to absurdity. I take the thought experiment on pp. 110-11 to be, among other things, an expression of the following argument:
So 3 The veil must prevent the choosers from knowing whether they are actual or merely possible futurepeople in intergenerational decisions. (1,2)
So 6 The veil of ignorance leads to absurdities in the intergenerational case. (3,4,5)