2. Corvino’s Case

On the conception of justice as reciprocity at issue, principles of justice are those acceptable to rational, mutually disinterested individuals as parties to a hypothetical contract to establish and distribute the fruits of mutually advantageous cooperation (Corvino Reference Corvino2023: 396, 404). Such fruits constitute the parties’ ‘cooperative surplus’ (Gauthier Reference Gauthier1986: 130, 141). As Corvino (Reference Corvino2023: 399) puts it, a cooperative surplus exists between A and B just when “the wealth produced jointly by A and B is greater than the sum of the individual wealth that A and B would produce working alone”. He distinguishes two variants of justice as reciprocity. On JSR, the parties have full information regarding their circumstances, allowing for unequal bargaining power – David Gauthier’s (Reference Gauthier1986) view being the most prominent example. (Note that, as will become important later on, full information includes the extent of their relative bargaining power.) By contrast, JFR, epitomised by John Rawls’s (Reference Rawls1999a) justice as fairness, withholds certain information from the parties to prevent morally arbitrary factors influencing their reasoning, ensuring equal bargaining power.Footnote ⁴

In Corvino’s (Reference Corvino2023: 396) telling, the non-reciprocity problem assumes future people cannot impact present people’s well-being, preventing them from engaging in mutually advantageous cooperation, thus excluding intergenerational relations from justice as reciprocity. He claims future people can impact present people’s well-being. How? “Future generations”, he writes, “contribute substantially to increasing the well-being of the living generations, by offering the latter a reasonable expectation that their transgenerational actions will be continued” (Corvino Reference Corvino2023: 398). Following Andina (Reference Andina2018: 369–71), transgenerational actions are collective actions that transcend their participants’ lifespans; curing cancer comes to mind (Scheffler Reference Scheffler and Kolodny2013: 26–27), or building a cathedral (Karnein Reference Karnein2022: 278ff). If we, present people, reasonably expect future people will exist, with a certain level of well-being, we can also reasonably expect, the thought goes, that they will contribute to finding a cure, or to finishing the cathedral.

Crucially, the reasonable expectation that our transgenerational actions will be continued through future people’s economic activity allows us to keep the economy “future-oriented and capital-preserving” (Corvino Reference Corvino2023: 400). It justifies, for example, a company investing in R&D with long-term profit timelines, creating more jobs in the present. It is what allows many companies to benefit, now, from ‘value investors’ who buy the company’s stock only because of its expected long-term growth. It makes possible taking on large-scale debt required to finance major infrastructure projects, such as the Channel Tunnel, thanks to future people taking over the debt and paying tunnel fees themselves. And, for many such public works, it is what allows governments to justify investing at all, with benefits to future citizens playing a decisive role in cost-benefit analyses. Present people evidently benefit from these choices being economically viable (Corvino Reference Corvino2023: 398–403). He concludes that future people can impact present people’s well-being, even future people with whom present people will never overlap (Corvino Reference Corvino2023: 402–3). We might add that, if Corvino is right, Rawls (Reference Rawls1999a: 254) was importantly mistaken in claiming that “[i]t is a natural fact that generations are spread out in time and actual economic benefits flow only in one direction”.

Next, he reasons, if future people can positively impact present people’s well-being, then, since present people can also positively impact future people’s well-being, present and future people can engage in mutually advantageous cooperation, establishing a cooperative surplus. They can thus stand in the relation required by justice as reciprocity. Indeed, since both JSR and JFR concern agreements to distribute a cooperative surplus, Corvino takes his argument to show that both can vindicate intergenerational justice by appeal to direct reciprocity (Corvino Reference Corvino2023: 399–400, 404–5). And importantly, he insists, this relation of reciprocity is “synchronic in that [future people] benefit [present people] in the present (i.e. before [future people] take an active role in society or even before they are born)” (Corvino Reference Corvino2023: 397).

Finally, Corvino argues both JSR and JFR endorse the principle Transgenerational Sufficiency: present people are to leave future people the resources required for “the minimum level of welfare of [future people] necessary for [present people] to have the reasonable expectation that the transgenerational actions in which [present people] are involved will continue in the future” (Corvino Reference Corvino2023: 404). (I will refer to the principle as ‘Transgenerational Sufficiency’ and use ‘TS_R’ for the amount of resources it requires present people to leave for future people.) We must appreciate, he begins, that the relevant bargain “is a different ‘bargain’ than that between contemporaries, as [future people] cannot refuse the ‘offer’ they receive (since they are not yet born)” (Corvino Reference Corvino2023: 404). Future people, that is, have no bargaining power. He then reasons:

In brief, given future people cannot refuse present people’s offer and will simply be glad to receive as much as they can, the principle both present and future people would ‘agree’ to will just be what it is rational for present people to propose, which he claims is Transgenerational Sufficiency.Footnote ⁵

3. Justice as Self-Interested Reciprocity

My claim, again, is that on neither JSR nor JFR does the relation Corvino identifies between present and future people, in which both generations abide by Transgenerational Sufficiency, constitute the relation of reciprocity required by that view and that, while we have strong reasons to believe that JFR does not endorse Transgenerational Sufficiency, JSR may well yet endorse it in a bargain between contemporaries rather than in one between generations. I start in this section with JSR. As we will see, the main problem is that Transgenerational Sufficiency does not require either generation to restrain their pursuit of self-interest in the way JSR requires to count as a genuine principle of justice. After explaining why, I will show how the problem generalizes beyond Corvino’s view to cast doubt on the possibility of any bargaining-theoretic construction of synchronic direct intergenerational reciprocity on JSR, and then explain why TS may nonetheless be chosen in a bargain between contemporaries.

The restraint of self-interest underpins JSR’s notion of direct reciprocity. In a canonical characterization, Brian Barry (Reference Barry1989: 6–7) writes that, on this view, “[j]ustice is simply rational prudence pursued in contexts where the cooperation (or at least forbearance) of other people is a condition of our being able to get what we want. Justice is the name we give to the constraints on themselves that rational self-interested people would agree to” to obtain others’ cooperation and establish a cooperative surplus. And Gauthier (Reference Gauthier1986: 3) – a plausible spokesperson for JSR as a whole – simply identifies “moral principles” (including principles of justice) as those that “constrain the actor pursuing his own interest”.

The role of such constraint on JSR is often presented game-theoretically via the Prisoner’s Dilemma (Gauthier Reference Gauthier1986: 103–4, 116–17; Barry Reference Barry1995: 51). Without rehashing the specifics, key to such situations is that the equilibrium outcome – resulting from each party maximizing their utility given the other’s choice – is Pareto sub-optimal; another available outcome exists better for at least one party (in this case both) and worse for none. This optimal outcome obtains only if both parties cooperate by restraining their utility-maximizing behaviour. Being greater in total utility than the non-cooperative equilibrium, it represents a cooperative surplus. Principles of justice are those that require such restraint. Rational agents do best to adopt a conditional disposition to obey them (Gauthier Reference Gauthier1986: Ch. 6).

That said, as Peter Vanderschraaf (Reference Vanderschraaf2018: 277) notes, not all interactions between parties regulated by JSR resemble a Prisoner’s Dilemma. Generalizing, and informed by an array of theorists of JSR from Hobbes and Hume to the present day, Vanderschraaf proposes a number of necessary conditions for a system of rules, or principles, to count as a theory of JSR, the first four being most relevant here:

1. (1) The principles require each party capable of pursuing their own interests to restrain such pursuit, at least to some extent, in order to advance others’ interests.

2. (2) If no party violates the principles, each party capable of pursuing interests who abides by them must be better off than the non-cooperative baseline.

3. (3) Each party capable of pursuing interests expects that were they to violate the principles, they would lose at least some benefits stemming from others’ restrained behaviour towards them.

4. (4) Each party capable of pursuing interests expects that if they abide by the principles then others will restrain their behaviour towards them as the principles require (adapted from Vanderschraaf Reference Vanderschraaf2018: 275–78).

Those who obey such principles, he suggests, stand in a relation of “reciprocal restraint” (Vanderschraaf Reference Vanderschraaf2018: 305). As such, I propose the following: two parties stand in the relation of direct reciprocity required by JSR only if they each abide by principles satisfying (at least) these four conditions, and where they each count as the relevant ‘other’ with respect to each other.

Transgenerational Sufficiency meets none of these conditions when the parties are present and future people. It fails (1) because it requires neither present nor future people to restrain their pursuit of self-interest to advance the others’ interests. Plainly, it does not require future people to restrain their pursuit of self-interest to benefit present people. Corvino would disagree, however, with the claim that Transgenerational Sufficiency does not require present people to restrain their self-interest to benefit future people; he claims that the principle requires present people to leave TS_R for future people “instead of acting selfishly and pursuing the maximization of individual [or ‘generational’] utility” (Corvino Reference Corvino2023: 404, emphasis added). However, leaving TS_R for future people just is what maximizes present people’s generational utility, independent of future people’s actions. As Corvino insists, present people benefit from their reasonable expectation of future people’s economic participation before future people are born, or at least before future people “take an active role in society” (Corvino Reference Corvino2023: 397), including by continuing present people’s transgenerational actions. The benefit to present people, then, does not depend on what future people actually end up doing.Footnote ⁶ Present people can either leave TS_R for future people, or not. If they do, they benefit from the relevant expectation (and, of course, Corvino (Reference Corvino2023: 404–5) assumes that present people benefit more from leaving TS_R than they lose in foregone consumption). However, if they do not leave TS_R, they do not benefit in this way. Transgenerational Sufficiency does not require them to restrain their pursuit of self-interest.

It follows that Transgenerational Sufficiency also fails (2). If no party violates the principle, then present people leave TS_R for future people. This makes present people no better off than the non-cooperative baseline (all else equal). For, on JSR, the non-cooperative baseline consists in parties maximizing their utility; as we have just seen, present people maximize their utility by leaving TS_R – they leave TS_R either way. Consequently, future people are no better off either.

Furthermore, Transgenerational Sufficiency fails both (3) and (4) because neither present nor future people can have the respective relevant expectation. First consider the expectations of present people. Transgenerational Sufficiency does not require future people to restrain their conduct towards present people and so there are no relevant benefits for present people to expect to lose (3), nor can they expect their abidance to lead to future people’s relevant restraint towards them (4). Now consider future people’s expectations. By the time future people choose whether or not to leave TS_R for their successors, present people have already either left or not left TS_R for them. Thus, future people will not expect to receive any less benefit from present people were they to choose not to leave TS_R themselves (3), nor can they expect any further relevant restraint from present people towards them (4).

So, in abiding by Transgenerational Sufficiency, present and future people would not stand in the relation of direct reciprocity required by JSR. Even if leaving TS_R were not, as I have claimed, in present people’s unrestrained self-interest, Transgenerational Sufficiency would still fail (3) and (4) because the explanation of why it fails these conditions in the preceding paragraph did not appeal to that claim. Indeed, it would still fail (1) since it does not require future people to restrain their self-interest, whereas (1) requires this of each party. And even if one rejects the conception of direct reciprocity I have attributed JSR, the fact Transgenerational Sufficiency fails Vanderschraaf’s conditions with respect to present and future people shows it is not a genuine principle of intergenerational justice on JSR.

Moreover, the fact that Transgenerational Sufficiency does not constrain present people’s pursuit of self-interest reveals a dilemma for Corvino’s characterization of the relevant bargaining problem, one that in turn will illuminate the more general lesson for theorizing synchronic direct intergenerational reciprocity on JSR flagged above.Footnote ⁷ Recall that a (Nash) bargaining problem consists of a feasible set of outcomes the parties could agree to, each assigning a utility to each party, with one such outcome representing the non-agreement point – the utilities each party receives if they fail to agree. The Pareto frontier is the subset of feasible outcomes that are Pareto optimal. A (determinate) bargaining solution specifies the uniquely rational outcome on the Pareto frontier the parties could agree to (Thrasher Reference Thrasher2014: 684–85; Vanderschraaf Reference Vanderschraaf2023: Chs 3–4). On at least one standard interpretation, each party initially proposes the outcome that maximizes their share of the cooperative surplus, after which through a process of exchanging concessions the parties rationally converge on the bargaining solution (Gauthier Reference Gauthier1986: 133).

The dilemma concerns the choice of non-agreement point. For Corvino (Reference Corvino2023: 419) “[t]he point of non-agreement … corresponds to transgenerational sufficiency for [future people]”. While he does not specify explicitly, presumably this non-agreement point gives present people the utility they would receive from consuming what is left given they leave TS_R.Footnote ⁸ This choice of non-agreement point indeed results in Transgenerational Sufficiency being accepted – on Corvino’s assumption, that is, that future people cannot refuse.Footnote ⁹ For, given future people cannot refuse, present people’s initial claim to their maximum share of the cooperative surplus will prevail, and TS_R is stipulated to be, recall, “the minimum necessary” present people need give future people “to get the maximum possible” from them. However, attributing TS_R to future people at the non-agreement point concedes that, without agreement, present people already decide to leave TS_R. Once again, we see Transgenerational Sufficiency does not require present people to restrain their self-interest. And since it also does not require future people to restrain their self-interest, this choice of non-agreement point effectively concedes that no bargain is needed. That is the first horn. The second is simple. Any non-agreement point that does not involve present people leaving TS_R misrepresents the relevant bargaining situation. A central tenet of JSR is that, in the absence of cooperation, the parties maximize utility. As I have argued, present people maximize utility by transferring future people TS_R.

Importantly – and here now is the more general lesson – this dilemma illuminates a worry for the possibility of any bargaining-theoretic construction of synchronic direct intergenerational reciprocity on JSR. The fact that present people benefit synchronically – which, recall, implies they benefit before future people are born or at least take an active role in society, and thus that present people benefit regardless of what future people then do – means that future people cannot be granted bargaining power. If future people did have bargaining power, we would no longer have a theory of JSR, for then present people would lack full information as to the power they hold over future people by virtue of their temporal position (recall that full information includes information concerning the extent of one’s relative bargaining power). However, if future people have no bargaining power, then the fact that present people benefit synchronically will ensure that any principle requiring the bargaining solution that emerges will fail to constrain present people’s pursuit of self-interest, and thus will not constitute a genuine principle of justice on JSR.

Why? On the one hand, since the size of the share of the surplus present people receive does not depend on what future people do, this means that, independent of any agreement, present people maximize their self-interest by establishing the surplus and keeping their maximum possible share of it. On the other hand, since future people lack bargaining power, the bargaining solution of any agreement they reach will likewise give present people their maximum possible share of the surplus. So any principle designed to produce the outcome corresponding to this solution will not require present people to restrain their pursuit of self-interest. The principle will thus not count as a genuine principle of intergenerational justice on JSR, and will not place different generations abiding by it in the relation of direct reciprocity the theory requires.Footnote ¹⁰

That said, even if Transgenerational Sufficiency does not constitute a principle of direct intergenerational reciprocity, Corvino’s argument nonetheless illuminates an important role that concern for the continuation of our transgenerational actions might play in theories of JSR.Footnote ¹¹ If, as he rightly claims, it is in the interest of present people to pass on TS_R to future people, then Transgenerational Sufficiency might emerge as a constraint on the principles that present people would agree to when bargaining among themselves (although note that just because it would be in present people’s best collective interest to agree to a principle does not guarantee that it would be accepted in the bargain). For example, principles that generate benefits to present people at the expense of failing to ensure future people receive TS_R will produce a smaller cooperative surplus for present people than those that provide the same benefits but do ensure that future people receive TS_R, all else equal. While developing the implications of this thought would require careful further investigation, the upshot may well be that Transgenerational Sufficiency would be accepted in an intragenerational bargain as a principle specifying duties regarding future people that are owed to present people by their contemporaries. Indeed, construing the view as such would place it among the good company of others who have justified duties regarding, but not owed to, future people (Hubin Reference Hubin1976; Vanderheiden Reference Vanderheiden2006; Mazor Reference Mazor2010; Gheaus Reference Gheaus2016; Meijers Reference Meijers2023). Of course, if this is right, then at least from the vantage point of concern for near-term future generations, we should agree with Corvino’s broader point that the non-reciprocity problem is not as serious an obstacle for theories of JSR than it has seemed.Footnote ¹²

4. Justice as Fair Reciprocity

Now let us consider JFR, which, recall, denies the parties certain information about their circumstances to establish equal bargaining power. By far the most prominent version is Rawls’s (Reference Rawls1999a) justice as fairness; indeed Corvino (Reference Corvino2023: 404) cites Rawls as the exemplar of JFR. I will argue that, on Rawls’s view, the parties would not choose Transgenerational Sufficiency. Consequently, given the tight conceptual link Rawls draws between reasonable acceptability to the parties and reciprocity between them, the relation between present and future people Corvino identifies – in which, again, both generations abide by Transgenerational Sufficiency – is not the relation of reciprocity required by Rawls’s view.Footnote ¹³ I will then generalize this result to non-Rawlsian versions of JFR, where the headline will be that, given Corvino’s case for Transgenerational Sufficiency assumes unequal bargaining power, Transgenerational Sufficiency will not be chosen on views of JFR given their uniform stipulation of equal bargaining power. Finally, I will show how this latter assumption mitigates the extent to which synchronicity, in Corvino’s sense, can influence the content of any principle of intergenerational distributive justice on JFR.

For Rawls (Reference Rawls2005: 17), reciprocity “is not the idea of mutual advantage” found in JSR. Instead, citizens stand in the relation of reciprocity only if they each gain from abiding by principles that are reasonable for all to accept (Rawls Reference Rawls2005: 16).Footnote ¹⁴ In justice as fairness more specifically, the test of reasonable acceptance is whether the principle would be agreed to in the original position, for the original position’s stipulation of deliberative equality between the parties, established by the veil of ignorance, is designed to reflect citizens’ status as moral equals (Rawls Reference Rawls2005: 22–23). To see why Transgenerational Sufficiency would not be chosen in the original position, first note that because the reasoning Corvino attributes the parties – which, recall, he claims holds true equally of JSR and JFR – assumes both that the parties know which generation they are in and a stark difference in bargaining power between generations, this reasoning does not apply to Rawls’s view. Behind the veil, the parties do not know their generation, effectively ensuring equal bargaining power between generations (Rawls Reference Rawls1999a: 121). Indeed, as Rawls acknowledges, since all parties are behind the veil, “each is convinced by the same arguments” such that “we can view the agreement in the original position from the standpoint of one person selected at random” and so “the parties have no basis for bargaining in the usual sense” (Rawls Reference Rawls1999a: 120). We thus need to look beyond the reasoning Corvino attributes the parties to Rawls’s own account of such reasoning.

To decide upon a principle of intergenerational distributive justice, the parties must agree, Rawls claims, on the rate of savings each generation must put aside for the next. Savings, here, are understood as “net additions to real capital of all kinds” (Rawls Reference Rawls1999c: 107), while the rate represents the “fraction of the social product” to be saved (Rawls Reference Rawls and Kelly2001: 160). More specifically, they must decide on a schedule of savings rates applying a different rate to each stage of advance, with poorer (richer) generations required to follow a lower (higher) rate. Given the parties are assumed to be contemporaries, and so initially will not agree to save at all, Rawls (in his later work) stipulates that they must choose a schedule they would want all previous generations to have followed (Rawls Reference Rawls1999a: 254–55; Reference Rawls2005: 274). He proposes the following deliberative procedure: for each level of advance, the parties ask themselves what members of adjacent generations would together deem a “fair” and “reasonable” rate, “balancing how much they would be willing to save for their more immediate descendants against what they would feel entitled to claim of their more immediate predecessors” to “arrive at the estimate that seems fair from both sides”, and then piece together these judgements (Rawls Reference Rawls1999a: 255–56).

We can assess whether justice as fairness endorses Transgenerational Sufficiency by evaluating it as a candidate for the just savings principle.Footnote ¹⁵ More specifically, the principle under consideration is that, at each stage of advance, a generation must save enough to ensure the next generation receives the level of resources constituting TS_R at that stage of advance (and assuming the difference principle is followed intragenerationally). A number of considerations suggest the parties would reject it. First, to reiterate, in the deliberative procedure outlined above adjacent generations share a position of equality; the parties, after all, do not know their generation and must place themselves in the shoes of both. The assumption of unequal bargaining power that is the lynchpin of the case for Transgenerational Sufficiency cannot be made here.

Second, reasoning from this position of equality, the parties would deem Transgenerational Sufficiency irrational. Consider a well-off generation G_n in which all live flourishing lives. Suppose their ability to flourish stems from having resources far exceeding TS_R, and that, likewise, the next generation G_n+1 will endure significant hardship if they inherit merely TS_R (perhaps G_n was atypically transgenerationally self-sufficient, if you will). In evaluating Transgenerational Sufficiency, the parties ask themselves whether TS_R constitutes an acceptable balance between what they would be willing to save were they G_n and what they would feel entitled to claim were they G_n+1. Assuming diminishing marginal importance of resources, the gain to G_n+1 of receiving more than TS_R will (potentially significantly) outweigh the loss to G_n of providing it. Plausibly, given Rawls’s insistence that the parties not invoke the difference principle when constructing the savings schedule, we should agree with Steven Wall (Reference Wall2003: 87) that “from the standpoint of mutually disinterested rational self-interest … The parties in the original position would agree to a savings schedule that imposed small costs on one generation if it yielded large benefits to an adjacent generation.” If so, the parties here would only agree to a rate requiring G_n to pass on more than TS_R.

Third, the parties also have good grounds to deem Transgenerational Sufficiency unfair, and thus to reject it according to the deliberative procedure. In the case above, it seems unfair for G_n to merely leave TS_R, especially since a small sacrifice from G_n could lead to large gains for G_n+1. Moreover, TS_R will likely not match the degree to which the savings rate ought to fairly increase with wealth across generations. Rawls (Reference Rawls1999a: 255) attributes the parties the belief that “in a wealthier society greater savings may reasonably be expected since the real burden of saving is less”; in other words, that a fair rate should monotonically increase with wealth (at least until a just basic structure has been established). To be sure, Corvino (Reference Corvino2023: 405–6) claims the quantity of goods constituting TS_R has and will continue to increase over time in line with the increased “degree of transgenerationality” of contemporary societies. But the degree of transgenerationality is not the right sort of metric to which to peg the savings rate. Being a contingent empirical matter highly sensitive to the extent to which the viability of present people’s economic activities depends on future people’s expected behaviour, the degree of transgenerationality is not guaranteed to monotonically increase with wealth. Even if we grant that it does, there is no guarantee it will increase as quickly as intuitively (from the parties’ perspective, given the beliefs Rawls attributes them) the just savings rate ought to. Further, even if in practice it usually does increase as quickly, Transgenerational Sufficiency would nonetheless not be robustly fair across the range of possibilities that parties in the original position must rationally consider.Footnote ¹⁶

Finally, merely leaving TS_R will sometimes be insufficient to perform either or both of the two core purposes Rawls attributes the just savings principle, and that he insists the parties “keep in mind” (Rawls Reference Rawls1999a: 256). The first is “to make possible the conditions needed to establish and preserve a just basic structure over time” (Rawls Reference Rawls and Kelly2001: 159), requiring each generation to maintain the just institutions they inherited, create new ones needed, and ensure each citizen enjoys circumstances sufficient to effectively exercise their equal basic liberties (Rawls Reference Rawls1999a: 252–54). The second is “to secure a social world that makes possible a worthwhile life for all its citizens” (Rawls Reference Rawls1999c: 107) by maintaining a “social minimum” (Rawls Reference Rawls1999a: 244–45). There is no guarantee Transgenerational Sufficiency will be sufficient for these tasks. Whether it is sufficient remains a complex and open empirical question.Footnote ¹⁷ But we should note that it is consistent with each citizen having merely the minimum resources to participate in the economy that none, let alone all, meet the social minimum. Indeed Corvino (Reference Corvino2023: 405) acknowledges that TS_R is not pegged to “universal human needs”. And TS_R is consistent with society collectively lacking the resources to preserve, or to create new, just institutions. We see many societies today in which, despite widespread economic participation, such goals are not met. In particular, there is evidence to suggest that democracies – and, I am assuming, just institutions more broadly – are more likely to endure in societies with greater per-capita income (Przeworski et al. Reference Przeworski, Alvarez, Cheibub and Limongi1996; Przeworski Reference Przeworski2005), which a greater savings rate can in turn increase (Gosseries Reference Gosseries2001: 321–22).Footnote ¹⁸ All these considerations give us strong reason to believe that Transgenerational Sufficiency would not be chosen by the parties on Rawls’s version of JFR, and so the relation Corvino identifies is not the relation of reciprocity the view requires.

That said, we should note again here that the reasoning Rawls attributes his parties differs from Corvino’s case for Transgenerational Sufficiency.Footnote ¹⁹ Whereas for the latter the parties are motivated purely to choose principles that are in their own interests, Rawls requires them to choose principles that are both in their interests and that bring about a just basic structure over time. As we have just seen, the fact that Rawls does so is partly why Transgenerational Sufficiency would not be chosen on his view. But we might think that this difference reveals that Transgenerational Sufficiency and the just savings principle play importantly different normative roles; whereas the former constitutes a direct solution to the non-reciprocity problem, the latter is an attempt to circumvent it by introducing an additional goal beyond the pursuit of self-interest. We must therefore consider whether Transgenerational Sufficiency would be chosen on non-Rawlsian versions of JFR that do not employ this strategy.

Turning to such views, recall that what unifies the category of JFR is that each view deprives the parties certain information to eliminate the influence of morally arbitrary considerations on their reasoning (Rawls Reference Rawls1999a: 11, 118). Which generation in time one belongs to is such a consideration; it would not be fair for any particular generation to enjoy a bargaining advantage merely due to their temporal location. So, even if non-Rawlsian variants of JFR eschew many of the specific features of the original position, they still deprive parties of the knowledge of their generation and ensure equal bargaining power between generations. Whether the parties are to assume they are all contemporaries – as Rawls (Reference Rawls1999a: 254) insists, in what he calls the ‘present time of entry’ interpretation – or whether they are to assume they belong to different generations – the ‘general assembly’ model (Rawls Reference Rawls1999a: 120; Heyd Reference Heyd, Gosseries and Meyer2009; Attas Reference Attas, Gosseries and Meyer2009) – the parties must consider what is acceptable to them should they turn out to be G_n+1 rather than G_n.Footnote ²⁰ But with equal bargaining power, G_n+1 will not agree to G_n passing on “the minimum necessary to get the maximum possible from” G_n+1, as Corvino characterizes TS_R, and thus will not agree to Transgenerational Sufficiency. Indeed, the reasoning above concerning why the parties would find the principle irrational on Rawls’s view when G_n is very well-off but G_n+1 will not be if they merely inherit TS_R generalizes to JFR as a whole.Footnote ²¹

The broader lesson here for theorizing synchronic direct intergenerational reciprocity within JFR is this. To stand in a relation of reciprocity, the parties must abide by principles reasonably acceptable to all. Such principles must be acceptable in a deliberative situation in which present and future people enjoy equal bargaining power. But this very assumption of equal bargaining power prevents the synchronic aspect of the relation from influencing the content of the principle. Again, synchronicity, in Corvino’s sense, ensures that present people benefit regardless of what future people then do. In practice, this establishes a power asymmetry. But in the theory of JFR, this power asymmetry cannot be reflected in the principle of justice it generates, for no party knows which end of the asymmetry they will end up on. This holds true for both present time of entry and general assembly models of the relevant deliberative situation.