[Critical review of: R. Bergmuller, R. Johnstone, A. Russell and R. Bshary, Integrating cooperative breeding into theoretical concepts of cooperation, Behav. Process. 76 (2) 2007: 61-72.] In this issue, Bergmuller et al., 2007 ... have provided a valuable review paper, re-establishing cooperative breeding within a theoretical framework in field studies of cooperative breeding vertebrates. I am in agreement with the authors in their outlook and suggestions. In this critical review, I wish to (1) review the costs and benefits approach previously taken in the field, (2) evaluate communal breeding species with regard to the game-theoretical approach promoted by Bergmuller et al. (2007), and (3) consider the reasons why empiricists researching cooperative breeding have generally side-stepped a game-theoretical approach. There is no doubting the great interest in cooperative breeding systems, in particular which individuals care for young, and the distribution of costs and benefits to care to determine whether and how individuals accumulate evolutionary fitness (Heinsohn and Legge, 1999). However, as Bergmuller et al. (2007) point out—this has proceeded relatively independently of game-theoretical approaches to cooperative behaviour.