This study evaluates a prominent functional explanation for play aggression, the practice hypothesis, employing data from a non-Western culture, the Zapotec of Oaxaca, Mexico. The practice hypothesis proposes that play aggression, sometimes called rough-and-tumble play, has evolved primarily as a “safe” way for young individuals to practice and perfect adult fighting and/or hunting skills. Ethological observations of 3–8-year-old Zapotec children (n = 48) form the basis for investigating serious and play aggression, and for evaluating predictions stemming from the practice hypothesis. Play aggression in Zapotec children is superficially similar to real aggression, but can be differentiated from aggression in a number of ways, such as by facial expression, nature of outcomes, and amount of behavioral variability. Rates of play aggression were not significantly different between Zapotec boys and girls; however, mixed-sex dyads during play aggression occurred less often than same-sex dyads. Play partners were closer in age than chance would predict, with initiators tending to be slightly older than recipients, except when boys initiated play aggression toward girls. Injuries during play aggression occurred very infrequently (in about 1% of the episodes), and children were more likely to stay together following play than following serious aggression. The Zapotec data are discussed in relation to predictions regarding play partner choice, sex differences, the overall nature (design) of play aggression (in relation to Zapotec adult aggression), as well as in relation to phylogenetic and cross-cultural considerations. It is concluded that most of the findings are consistent with the practice hypothesis. It is also suggested that because play and serious aggression can be distinguished and apparently stem from different underlying motivations, these behavioral patterns not be confounded as is sometimes done in observational studies of children's aggression.
No takes yet. Share an insight, caveat, or question.
Douglas P. Fry (1990) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: