A Psychology of the Stranger

Pervin covers a lot of ground in his useful and timely critique of contemporary trait theory. While applauding the significant advances trait theories have made in the past decade, he takes issue with some of the more imperialistic champions of traits in general and the Big Five trait taxonomy in particular-those who now argue that (a) personality is traits and nothing but traits (e.g., Brody, 1988; A. H. Buss, 1989) and (b) there are five or so basic and universal trait domains (e.g., Digman, 1990; Goldberg, 1993; McCrae & Costa, 1990; McCrae & John, 1992). In Pervin's temperate and eclectic view, traits may prove to be serviceable constructs in accounting for a modicum of personality stability and consistency over time and across situations, but they are too static and superficial to explain personality dynamics, personality pattern and organization, and meaningful personality change. Like Murray (1938) and McClelland (1951) before him, Pervin envisions personality as a complex system or set of systems encompassing a wide range of different constructs. Personality encompasses traits, yes, but also motives, schemas, plans, and presumably other variables that must be invoked to account for human individuality. There are points in Pervin's skeptical review of the empirical evidence for the Big Five trait approach where I think he comes down too hard. First, in reviewing the evidence for heritability of traits, Pervin rightly points out that the twin studies of the 1980s suggest that genetics typically accounts for no more than about 50% of the variability in trait scores, leaving half for the environment. But one should not lost sight of how impressive this consistent finding is, of how strong a case it does indeed make for the genetic underpinning of temperament traits. Further, one must know that research has generally failed to identify substantial effects of shared environments on traits. Rather, environmental effects appear to be nonshared and therefore rather difficult to pinpoint in research, for they are experienced differently for each person (Dunn & Plomin, 1990). In my view, therefore, it is difficult to ignore or paper over the evidence that many traits, including the Big Five dimensions of Extraversion and Neuroticism, are at least moderately heritable and that genetic influences work to promote trait stability over time. A second example of what I think is excessive criticism of the Big Five model is Pervin's assertion that trait research has not surpassed the ".30 barrier" made famous by Mischel (1968). Do personality trait scores predict behavior at anything beyond trivial levels (i.e., r = +.30)? Although I agree that Big Five researchers have tended to neglect the problem of predicting behavior in research, it is probably unfair to intimate that, when it comes to predicting behavior, trait research is still mired in the Mischellian mud of 1968. As Pervin observes in the beginning of his target article, it is difficult to figure out exactly why trait theories have garnered so much enthusiasm in recent years, but one of the reasons is probably that a great deal of research employing aggregation, moderator variables, and careful assessments of traits and behavior in the 1970s and 1980s showed convincingly, in my view, that wellmeasured trait dimensions do indeed predict behavior at significant levels and do so with at least as much efficacy as do the situational factors typically championed by social psychologists and social-learning theorists (Epstein, 1984; Funder & Colvin, 1991; Funder & Ozer, 1983; Kenrick & Funder, 1988; Moskowitz, 1982). As the situationist program began to run out of steam and a new generation of trait research gained vigor and influence in the 1980s, the field of personality psychology became a much more hospitable place for the kind of conceptual consensus that the Big Five now represents. Proponents of the Big Five should not be faulted for seizing a moment made possible by others. Third, Pervin may overstate the case when he avers that trait constructs serve as mere descriptions rather than explanations for behavior. This is a favorite criticism of those who do not like traits, and it points to a problem that trait psychologists have struggled with since the time of Allport (1937). Do traits explain anything? My view is that, even in their most modest incarnations, general trait labels do provide a kind of rudimentary explanation. It is the kind of explanation captured in what Aristotle called formal cause (Rychlak, 1981). A formal-cause explanation accounts for phenomena in terms of a general form or pattern that they take. For example, a wide set of discrete behaviors exhibited by a given person over the course of a weektalking to strangers on the bus, boisterous laughing at a party, asserting oneself confidently in class, making an