The surest way to better understand grandparenting – or most anything else about families – is to heed Coall & Hertwig's (C&H's) call for interdisciplinary mutualism. They are exactly right: Pitting evolutionary and social science as contenders for center stage gets us nowhere. Progress will come from recognizing that the two approaches emanate from distinct corners of a single – and woefully incomplete – puzzle.
Advances will require continuing efforts to connect distant dots – ultimate and proximate forces – to plumb the interplay between, say, kin selection's primal tug and the exigencies of hedonics and budgets, played out in the context of real-world culture and institutions.
Fortunately, more work exists along these lines than the few harbingers of hope C&H mention. What they aptly call a “motivational engine” was added to evolutionary models more than a decade ago, when economists replaced the standard maximand of reproductive success with something just as fundamental but more immediate – individual well-being (i.e., “utility”), which includes feelings for other family members (e.g., Alger & Weibull, forthcoming; Bergstrom Reference Bergstrom1997). This refinement retains the essence of kin selection (i.e., caring about one's genetic legacy), but fleshes out the details about how much help is provided and when. Help is subject to diminishing marginal utility: the better off the recipient becomes, the smaller the impact of additional help; the worse the predicament, the more help matters. Diminishing marginal utility is a simple but powerful idea.
For example, grandparents and would-be grandparents have been found to be most likely to assist their adult children with housing down payments when the children: (a) wanted to have a child of their own but (b) were concerned that their present living quarters were too small to accommodate one (Cox & Stark Reference Cox and Stark2005). The first reason is arguably tied up with the ultimate motive of fitness; the second, with the proximate motive of marginal utility: All else equal, the smaller the adult child's living space, the more grandparental assistance matters.
Diminishing marginal utility can explain C&H's finding that grandparenting apparently confers less widespread or pronounced benefits for “low risk” families compared to their “high risk” counterparts, where teenage pregnancy or maternal depression occur. Large-scale help would be less consequential, and hence less prevalent, for grandchildren who are already well off. Evolution determines the capacity for altruism; but proximate factors like diminishing marginal utility determine whether and how this capacity is expressed.
Diminishing marginal utility can also help reconcile seemingly disparate motives for kin assistance. Consider altruism and exchange, which C&H – and just about every other researcher – cast as competing motives. The logic of helping behavior can be simplified and enriched by considering how the two might coexist and when one or the other might dominate.
For instance, imagine that Aaron and Ben are colleagues in a firm. At a company picnic by a lake, Aaron hears Ben's cries for help, realizes he is drowning, and prepares to risk his life to save him. Fast-forward to a time long after Ben's rescue. The urgency of altruistic transfers has fallen; Ben is no longer someone at death's door – he is now just someone who works down the hall. Aaron might still help Ben, but only quid pro quo: the same people, but different circumstances and hence different motives for helping.
This portmanteau approach has the counterintuitive implication that the worse off recipients are, the more likely the introduction of public transfers will crowd out private giving. Return to our example and suppose a town lifeguard were to save Ben, thereby letting altruist Aaron off the hook. To cast things in money terms, imagine that social insurance aids impoverished Ben, thereby supplanting Aaron's help. But if Ben were already well off enough that the relevant motive were exchange, increases in Ben's income would only serve to strengthen his bargaining position with Aaron, thereby generating “crowding in.”
The contrasting income effects can explain why evidence for crowding out is scarce for Europe and the United States but not for developing countries, where private assistance matters more for addressing basic needs (Cox & Fafchamps Reference Cox, Fafchamps, Schultz and Strauss2008).
How does a proximate motive like altruism mesh with the ultimate influence of kin selection? If Aaron were Ben's grandfather, it is easy to see how they would dovetail. And generational altruism in the opposite direction – say, care that adult children provide to their parents – could be reconciled by appealing to the fitness benefits of grandparenting, as in “The sooner I can get Mom back on her feet, the more she can do for my kids.”
But what about instances of elder care with no apparent fitness advantage, such as an adult child who, amid demands of work and childrearing, makes time for her frail and indigent mother? Some upstream transfers appear to fit this description (e.g., Perozek Reference Perozek1998; Sloan et al. Reference Sloan, Picone and Hoerger1997), yet they pose a difficulty for kin selection that has attracted surprisingly little notice in the evolutionary literature. A gene that impelled the provision of such care would eventually be out-competed by fitness-enhancing ones. The same goes for the evolution of cultural practices (remember the Shakers?).
What about empathy, a motive deeply ingrained yet focused on today's cues, rather than tomorrow's fitness? Surely the care described above would qualify as such – the key question is how it could prevail in an evolutionary environment. If, in the authors' words, “empathy-based acts are allocated in accordance with kin and reciprocal altruism theory” (target article, sect. 4.3, para. 5), they had better not be allocated too strictly or we are back to square one. One could speculate that unrequited elder care is somehow tied up with the forces of group selection – group-beneficial norms against preventable suffering, say – and provided out of fear of altruistic punishment. Or, that it is a by-product of deep-rooted inclinations to reciprocate, or perhaps a signal of willingness to cooperate with non-kin. Finding the evolutionary basis for this expression of empathy appears to require connecting dots that for the moment appear exceedingly distant. But it is likely a feasible task, and certainly one that would be worthwhile.
You have
Access
The surest way to better understand grandparenting – or most anything else about families – is to heed Coall & Hertwig's (C&H's) call for interdisciplinary mutualism. They are exactly right: Pitting evolutionary and social science as contenders for center stage gets us nowhere. Progress will come from recognizing that the two approaches emanate from distinct corners of a single – and woefully incomplete – puzzle.
Advances will require continuing efforts to connect distant dots – ultimate and proximate forces – to plumb the interplay between, say, kin selection's primal tug and the exigencies of hedonics and budgets, played out in the context of real-world culture and institutions.
Fortunately, more work exists along these lines than the few harbingers of hope C&H mention. What they aptly call a “motivational engine” was added to evolutionary models more than a decade ago, when economists replaced the standard maximand of reproductive success with something just as fundamental but more immediate – individual well-being (i.e., “utility”), which includes feelings for other family members (e.g., Alger & Weibull, forthcoming; Bergstrom Reference Bergstrom1997). This refinement retains the essence of kin selection (i.e., caring about one's genetic legacy), but fleshes out the details about how much help is provided and when. Help is subject to diminishing marginal utility: the better off the recipient becomes, the smaller the impact of additional help; the worse the predicament, the more help matters. Diminishing marginal utility is a simple but powerful idea.
For example, grandparents and would-be grandparents have been found to be most likely to assist their adult children with housing down payments when the children: (a) wanted to have a child of their own but (b) were concerned that their present living quarters were too small to accommodate one (Cox & Stark Reference Cox and Stark2005). The first reason is arguably tied up with the ultimate motive of fitness; the second, with the proximate motive of marginal utility: All else equal, the smaller the adult child's living space, the more grandparental assistance matters.
Diminishing marginal utility can explain C&H's finding that grandparenting apparently confers less widespread or pronounced benefits for “low risk” families compared to their “high risk” counterparts, where teenage pregnancy or maternal depression occur. Large-scale help would be less consequential, and hence less prevalent, for grandchildren who are already well off. Evolution determines the capacity for altruism; but proximate factors like diminishing marginal utility determine whether and how this capacity is expressed.
Diminishing marginal utility can also help reconcile seemingly disparate motives for kin assistance. Consider altruism and exchange, which C&H – and just about every other researcher – cast as competing motives. The logic of helping behavior can be simplified and enriched by considering how the two might coexist and when one or the other might dominate.
For instance, imagine that Aaron and Ben are colleagues in a firm. At a company picnic by a lake, Aaron hears Ben's cries for help, realizes he is drowning, and prepares to risk his life to save him. Fast-forward to a time long after Ben's rescue. The urgency of altruistic transfers has fallen; Ben is no longer someone at death's door – he is now just someone who works down the hall. Aaron might still help Ben, but only quid pro quo: the same people, but different circumstances and hence different motives for helping.
This portmanteau approach has the counterintuitive implication that the worse off recipients are, the more likely the introduction of public transfers will crowd out private giving. Return to our example and suppose a town lifeguard were to save Ben, thereby letting altruist Aaron off the hook. To cast things in money terms, imagine that social insurance aids impoverished Ben, thereby supplanting Aaron's help. But if Ben were already well off enough that the relevant motive were exchange, increases in Ben's income would only serve to strengthen his bargaining position with Aaron, thereby generating “crowding in.”
The contrasting income effects can explain why evidence for crowding out is scarce for Europe and the United States but not for developing countries, where private assistance matters more for addressing basic needs (Cox & Fafchamps Reference Cox, Fafchamps, Schultz and Strauss2008).
How does a proximate motive like altruism mesh with the ultimate influence of kin selection? If Aaron were Ben's grandfather, it is easy to see how they would dovetail. And generational altruism in the opposite direction – say, care that adult children provide to their parents – could be reconciled by appealing to the fitness benefits of grandparenting, as in “The sooner I can get Mom back on her feet, the more she can do for my kids.”
But what about instances of elder care with no apparent fitness advantage, such as an adult child who, amid demands of work and childrearing, makes time for her frail and indigent mother? Some upstream transfers appear to fit this description (e.g., Perozek Reference Perozek1998; Sloan et al. Reference Sloan, Picone and Hoerger1997), yet they pose a difficulty for kin selection that has attracted surprisingly little notice in the evolutionary literature. A gene that impelled the provision of such care would eventually be out-competed by fitness-enhancing ones. The same goes for the evolution of cultural practices (remember the Shakers?).
What about empathy, a motive deeply ingrained yet focused on today's cues, rather than tomorrow's fitness? Surely the care described above would qualify as such – the key question is how it could prevail in an evolutionary environment. If, in the authors' words, “empathy-based acts are allocated in accordance with kin and reciprocal altruism theory” (target article, sect. 4.3, para. 5), they had better not be allocated too strictly or we are back to square one. One could speculate that unrequited elder care is somehow tied up with the forces of group selection – group-beneficial norms against preventable suffering, say – and provided out of fear of altruistic punishment. Or, that it is a by-product of deep-rooted inclinations to reciprocate, or perhaps a signal of willingness to cooperate with non-kin. Finding the evolutionary basis for this expression of empathy appears to require connecting dots that for the moment appear exceedingly distant. But it is likely a feasible task, and certainly one that would be worthwhile.
ACKNOWLEDGMENTS
I thank Tracy Vietze and Ingela Alger for comments. Support from the National Institutes on Child Health and Human Development (R01-HD045637) is gratefully acknowledged.