Reflections on generativity and flourishing: a response to Snow’s Kohlberg Memorial Lecture

View correction statement:

Erratum

Most readers will recognize more than a grain of truth in Professor Nancy Snow’s retelling of Clarence Darrow’s humorously negative take on parental generativity. Yet, of course, there is also a positive side, which is, I hope, more dominant. I propose that during the first seasons of our life we are nurtured by our parents, but during the latter seasons of our life we are nurtured by our children. Generative parents interact with their offspring in ways that offer valuable support for their children’s development but, concurrently, children also have a way of contributing to their parents’ development, a way of requiring us to become more mature adults. In Erikson’s terminology, there is a generative ‘cogwheeling’ between the lives of children and parents, such that child–parent interactions create opportunities that can foster the psychosocial development of both children and adults, including generative competencies to care for others.

From Professor Snow’s title, ‘Generativity and Flourishing,’ to the closing comments on ‘our visions of a better future,’ her proposal is provocative and insightful. I cannot help but be sympathetic to her general thesis—that generativity is a necessary but not sufficient condition for Aristotelian flourishing and general thriving. Nevertheless, there are several points worth clarification. I will limit my comments largely to Professor Snow’s psychology.

Most helpfully, Snow presents ‘conceptualizations and a working definition’ of generativity, in which she nicely captures Erik Erikson’s evolving portrait of generativity as she traces his ripening statements from the 1950s to the 1990s and concludes with Joan Erikson’s insightful comments on generativity. Joan’s final words are especially fitting, given that her intellectual contribution to Eriksonian psychosocial theory is often overlooked (Snarey & Bell, 2006). Although Erik’s writings were wide ranging, he is best known for this pioneering stage model of the life cycle, which was created in generative collaboration with his partner, sociologist Joan M. Erikson (Friedman, 1999, 215–220; cf. Erikson & Erikson, 1950; Erikson, Erikson, & Kivnick, 1986; J. M. Erikson, 1988, 1997).

Professor Snow also notes that Joan ‘apparently’ considered ‘generativity as a life stage.’ In contrast, Snow decided to call generativity the ‘preferred mode of being’ of adulthood in order to remain ‘neutral among interpretations of generativity as a life stage, urge, instinct, or desire.’ Although different interpretations are possible, I believe that Snow’s definition is largely in agreement with the Eriksons’ theory. To illustrate, consider the Eriksons’ matrix of human development, which first appeared in Childhood and Society (1950, 1963). The eight psychosocial stages are listed along the bottom horizontal axis, the parallel age periods appear along the left vertical axis and the names of the stages are indicated along a diagonal. Thus, in the bottom-left corner of the matrix, the Eriksons placed the stage of Trust versus Mistrust during the age-period of Infancy and one cell up and one cell to the right is the stage of Autonomy versus Shame and Doubt during Toddlerhood and so forth. When I first saw this matrix as an undergraduate student, I was struck by the fact that 56 of the 64 cells in the matrix were empty! Then I learned that that these cells remind us that all of the eight psychosocial tasks were present throughout the entire life cycle, even though each stage emerged as a person’s psychosocial preoccupation or primary developmental task in an ordered sequence. So, just as it is useful to know that the ‘preferred mode of being’ or primary developmental preoccupation during adulthood is Generativity versus Stagnation, it is also helpful to recognize another point of agreement between the Eriksons and Snow—generativity is also a ‘mode of being’ across the entire life span (cf. Poole & Snarey, 2011).

Each of Eriksons’ eight stages has benefited from analysis by other researchers who subsequently made distinctions within each stage and usually accompanied these distinctions with an empirical measure. First John Kotre (1984), and later I, attempted to do this for generativity, but it is important to note that we both were building on the fact that the Eriksons had implicitly distinguished between three interrelated types of generativity. We called these: (1) biological generativity (birth); (2) parental generativity (childrearing); and (3) societal generativity (cultural parents).¹ Biological generativity links Intimacy (Stage 6) to a full-bodied Generativity (Stage 7) as a couple expands their mutual interests and invests themselves ‘in that which is being generated and cared for together’ (Erikson & Erikson, 1997, p. 67). ‘Parental generativity, in turn, is a tendon connecting biological and societal generativity’ (Snarey & Clark, 1998, p. 47).

In my research, the explicit hypothesis (and, I believe, the Eriksons’ implicit premise) was the following: caring for children and adolescents prepares adults to care for the growth of younger adults outside the family and to contribute to the well-being of the larger community across the generations. Snow is understandably cautious on this point. In her words,‘this appears to be problematic. … For example, many people who are not biological parents are capable of generativity in the sense that they can and do nurture children who are not biologically their own. And people who have not had children can be [societally] generative in the sense of being able to teach skills and [to pass on] the culture.’ The Eriksons were sensitive to this possible alternative understanding of their position and addressed it over time.

Erikson speculated in 1950 that achieving societal generativity is more difficult without the prior experience of parenting children. But nine years later he made a point of acknowledging in the context of a discussion on parenting that ‘there are people who, from misfortune or because of special or genuine gifts in other directions, do not apply this drive to offspring but to other forms of altruistic concern and creativity, which may absorb their kind of parental responsibility’ (1959, p. 103). Erik and Joan, in their 1981 article on generativity in the Harvard Educational Review, further emphasized that ‘generativity can find expression in productivity or creativity as well as procreativity’ (p. 255). Nevertheless, they held firm to their idea that parent-like forms of altruism then become necessary for adult development. Why? In part because they believed that ‘the only way to counteract the possible effects of a necessary deprivation is [through] a kind of sublimation’ (Erikson & Erikson, 1981, p. 263).

Snow also argues that generativity is an Aristotelian-type of virtue. Erikson spoke of virtues and vices as well. He identified the particular virtue that develops out of the successful resolution of each of the eight psychosocial crises or developmental turning points. For instance, an excess of despair over generativity at Stage 7 gives rise to the vice of ‘rejectivity,’ but a resolution characterized by a surplus of generativity over despair gives rise to the virtue of ‘care.’ But here it is useful to clarify that, although Erikson originally used the term ego ‘virtues,’ he later used the term ‘strengths’ or the phrase ‘virtues or strengths’ (cf. Erikson, 1964, 1968, 1975). He made this modification because he had not intended the word ‘virtue’ to be understood in an overly moralistic sense. Rather, he was using the term in an Old English sense in which ‘virtue’ primarily meant ‘inherent strength,’ as in the expression, ‘the whisky has lost its virtue’ because it has stood out too long (Evans, 1966).² A similar use of ‘virtue’ as meaning ‘strength’ is seen in Erikson’s excerpt from the Gospel of Mark’s account of a sick woman who came up behind Jesus in a large crowd and touched his cloak, in the hope that she would be healed (1981, 327). Erikson, quoting a Lutheran translation, tells us that Jesus, ‘immediately [knew] in himself that virtue had gone out of him’ (Mark 5:29–34, italics added). Of course, ego strengths still are legitimately related to types of moral will or moral courage, so the distinction is one of degree rather than of kind.

Rather than understanding generativity as a virtue, I have described a ‘generative ethic of care’ (Snarey, 1993, 1998; cf. Hawkins & Dollahite, 1997). Briefly, a generative ethic of care is an ethical position that first considers the effects on and for the next generation. An ethic of generative care overcomes the ambivalence associated with irreversible obligations through an inclusive concern for the next generation. In some sense, of course, Erikson also took this perspective. Generativity is an ‘ethical orientation,’ he states, that ‘makes the difference between adulthood and adolescence’ (1975, p. 207). This is seen clearly when he recounts the course of ethical development in his book, Dimensions of a New Identity (1974). He writes that after you have learned ‘what you care to do and who you care to be’ (identity), and ‘whom you care to be with’ (intimacy), you are ready to learn ‘what and whom you can take care of’ (generativity) (1974, p. 124). Again we see that an ethic of care is a constant throughout the life cycle, even though it is during middle adulthood that it should emerge as the dominant ethical perspective.

Professor Snow invited me to interject my own ‘studies of generativity in fathers’ in order to address ‘the multitude of factors that contribute to and detract from various types of generativity.’ In How Fathers Care for the Next Generation: A Four-Decade Study (1993), I reported that father’s parental generativity is significantly predicted by several boyhood and concurrent background variables, but that the ability of these various boyhood and adulthood variables to account for variations in the men’s parental generativity differed between the total quantity of all childrearing activities versus specific types of childrearing.

When predicting the total quantity of parental generativity, the characteristics of the women in the men’s lives generally accounted for the largest amount of unique variance. Here are three suggestive longitudinal examples: (1) Fathers who were highly involved in childrearing were significantly more likely to have been raised by mothers who were less educated than most of the mothers and, concurrently, were also more likely to be married to wives who were relatively more educated; (2) Marital affinity was another important positive predictor. High involvement in childrearing was closely related to marital contentment and commitment. Thus it seems that, as Eriksonian theory would predict, men who had positively resolved the prior psychosocial task of Stage 6, ‘Intimacy versus Isolation,’ were better prepared to meet the psychosocial challenge of Stage 7, ‘Generativity versus Stagnation’; (3) Finally, a wife’s employment outside the home was also a significant predictor of father’s childrearing participation. The longitudinal findings showed that wives’ employment provided an immediate motivation for fathers to increase their childrearing participation, but once this was set in motion, the fathers’ boyhood backgrounds predict their preferred types of participation.

When predicting the types of parental generativity, I distinguished between three varieties (parental support of social-emotional development vs intellectual-academic development vs physical-athletic development) during two different decades (childhood vs adolescence). In an interesting contrast, the greatest amount of unique variances in each of the different types of childrearing generativity was generally accounted for by the characteristics of the study fathers’ own fathers. Furthermore, it turned out that fathers’ childrearing practices involved both modeling—replicating or repeating specific aspects of the positive model of fathering they received and reworking—revising or rectifying specific aspects of the unsatisfactory fathering they received.

In terms of modeling, the following three longitudinal patterns suggest some of the ways the study men adopted or employed their own fathers’ positive model of paternal generativity: (1) Those whose fathers had been relatively better educated tended as adults to provide significantly more support for their own children’s social-emotional development during their childhood years; (2) Those whose fathers had been employed in relatively higher status and more complex blue-collar jobs tended to provide significantly more support for their own children’s social-emotional development during their adolescence; and (3) Those who grew up in a home in which their father and mother worked together well to provide their children with a cohesive home atmosphere tended as fathers themselves to provide significantly more care and support for their children’s social-emotional development during the adolescent years.

In terms of reworking, the following are three longitudinal patterns by which men countered or reworked their own fathers’ shortcomings: (1) Those whose relationships with their fathers had been distant or nonnurturant tended, as adults themselves, to provide significantly above-average levels of care for their children’s social-emotional and intellectual-academic development during the adolescent decade; (2) Those whose fathers had inconsistently or inadequately supervised their boyhood physical well-being tended, as fathers themselves, to provide significantly higher levels of support for their children’s physical-athletic development during their childhood years; and (3) Those whose fathers had used physical punishment or threats of physical punishment, which instilled fear in them as boys, tended as fathers themselves to provide significantly higher levels of positive care for their children’s physical-athletic development.

These findings demonstrated that modeling and reworking are not mutually exclusive, as it is often assumed, but in a sense the Eriksons’ already knew that both modeling and reworking processes influence the course of generativity. In his analysis of Martin Luther’s life, Young Man Luther (1958), Erikson told the story of how the excessively harsh, but attentive, care provided by the boy’s ambitious father, miner Hans Luder, subsequently contributed, primarily by a process of reworking, to Luther’s theologically generative reinterpretation of his heavenly father as a God who was both genuinely just and unconditionally caring. Hans Luder’s model of ambition, however, also likely contributed to Luther’s determined rise to religious leadership.

In closing, I must mention that a ‘Larry story’ came to mind as I read Professor Snow’s essay. (Everyone who worked with Lawrence Kohlberg came away with a collection of memorable stories and I was no exception.) The story provides a useful conclusion to these reflections. In the early 1980s, when he and I were co-authoring an article on ego development, I remember sitting in his living room and discussing possible revisions of the manuscript. We turned to my section on Erik Erikson and I complained that Erikson was at times inconsistent and hard to pin down. That was all it took to launch Larry into a story about one of his first conversations with Erikson. Apparently Erikson was visiting Kohlberg and Larry took the opportunity to ask Erik to clarify himself on a number of points. I paraphrase: ‘You said such and such in Childhood and Society [in 1950], but then such and such in Identity and the Life Cycle [in 1959] and then still something different in Gandhi’s Truth [in 1969]. As soon as Erikson explained what he meant, as best he could, Larry would hit him with another apparent contradiction and then another and another. Finally, Erikson raised his hands in the air and said, ‘Guilty as charged. I am just a big bull shitter.’ Both Larry and I, of course, burst into laughter.

Kohlberg’s point remains relevant today. Erikson had an ability to deploy developmental concepts with varied meanings and levels of precision. This was due to his evolving psychosocial insights and, no doubt, to the divergent needs of his different audiences. So I remain happy to follow Larry’s advice to cut him some slack and I am confident that he would give us similar advice regarding Nancy Snow’s work at the thorny interface between developmental psychology and virtue ethics.

Notes

1. Kotre (1984) also made a further distinction between two types of societal generativity—technical and cultural.

2. When the audio-video tape recording of Richard Evans’ (1966) interview with Erikson was transcribed and published (Evans, 1967), the word ‘whisky’ was changed to ‘medicine.’ Of course, in either case, ‘if it had stood around too long’ one could say that ‘it had lost its virtue’ (p. 17).