Behavioral Ecology and Sociobiology

, Volume 6, Issue 3, pp 187–197

Testing the ‘agonistic buffering’ hypothesis

I. The dynamics of participation in the triadic interaction
  • David Milton Taub

DOI: 10.1007/BF00569200

Cite this article as:
Taub, D.M. Behav Ecol Sociobiol (1980) 6: 187. doi:10.1007/BF00569200


Data are presented from a 15-month study on triadic male-infant interactions (‘agonistic buffering’) among wild Barbary macaques, and the ‘agonistic buffering’ hypothesis reevaluated. The sociometrics of triadic interactions derived from the distribution of 535 interactions among individually known adult and subadult males showed that there were significant individual male differences in the frequency of initating and/or receiving triadic interactions, but there were no such differences between the high and low ranking male subgroups (Tables 2–4).

Males did not choose each other equally for triadic encounters. Each male had a different set of three other males (out of 11 possible) that he preferred to approach for a triadic encounter; and each male received triadic approaches essentially from only three males (Tables 4 and 5). Each male showed a specific preference for which infant to select for participation in the triad, and in fact, males who preferred one another for triads also preferred to use the same infant in these encounters. These males that preferred each other and the same infant both had a special care-taking relationship with that preferred infant (Tables 6–7).

Possessing an infant per se seemed to be irrelevant in whether a male would approach or be approached for a triad. In 78% of all triads, males separated immediately after termination of the triad. When the ‘agonistic buffering’ hypothesis is reexamined against the data in this study, it appears that it cannot adequately accommodate the thesis that it serves to regulate dominant/subordinate relations among males. Rather males choose to participate with each other in ‘agonistic buffering’ because, and by means of, a shared, common, and special care-taking relationship with the same infant. The indiscriminate use of the terminology ‘agonistic buffering’ to describe multiple-male/infant interactions in this species should be dropped.

Copyright information

© Springer-Verlag 1980

Authors and Affiliations

  • David Milton Taub
    • 1
  1. 1.Department of Comparative Medicine, Bowman Gray School of MedicineWake Forest UniversityWinston-SalemUSA
  2. 2.Yemassee Primate CenterYemasseeUSA