Punishment as a Means of Competition: Implications for Strong Reciprocity Theory

Abstract

Strong negative reciprocity, that is, sanctions imposed on norm violators at the punisher’s own expense, has powerful cooperation-enhancing effects in both real-life and experimental game situations. However, it is plausible that punishment may obtain alternative roles depending on social context and the personality characteristics of participants. We examined the occurrence of punishing behavior among 80 subjects in a strongly competitive Public Goods game setting. Despite the punishment condition, the amount of the contributions decreased steadily during the game. The amount of contributions had no significant effect on received and imposed punishments. The results indicate that certain social contexts (in this case, intensive competition) exert modifying effects on the role that punishment takes on. Subjects punished each other in order to achieve a higher rank and a financially better outcome. Punishment primarily functioned as a means of rivalry, instead of as a way of second-order cooperation, as strong reciprocity suggests. These results indicate the need for the possible modification of the social conditions of punishment mechanisms described by the strong reciprocity theory as an evolutionary explanation of human cooperation.

Punishment: one tool, many uses

“…While most empirical evidence suggests that this is the case, in many experimental settings there are non-negligible levels of punishment directed towards people who did nothing wrong (e.g. Abbink and Herrmann 2011 ; Abbink and Sadrieh 2009 ; Dawes et al 2007 ; Paál and Bereczkei 2015 ; Raihani and McAuliffe 2012 ; Wu et al 2009 ) or towards those who cooperated more than the punisher (Anderson and Putterman 2006 ; Cinyabuguma et al 2006 ; de Melo and Piaggi 2015 ; Falk et al 2005 ; Goette et al 2012 ; Herrmann et al 2008 ; Irwin and Horne 2013 ; Pfattheicher et al 2017 ; Pleasant and Barclay 2018 ; Sylwester et al, 2013 ). In some societies, cooperators are almost or equally as likely to be punished as defectors (Gächter et al 2010 ; Gächter and Herrmann 2011 ; Herrmann et al 2008 ).…”

“…In most social dilemmas, interacting with a cheat means that partners incur losses and experience disadvantageous inequity (Raihani and McAuliffe 2012 ), meaning that these motives cannot easily be disentangled. Studies which have attempted to isolate the proximate basis of punishment decisions in the laboratory found that punishment decisions are at least (or sometimes more) sensitive to inequity than to losses and suggest that punishment might often be motivated by disliking being worse off than others, rather than disliking being cheated (Bone and Raihani 2015 ; Carlsmith et al 2002 ; Dawes et al 2007 ; Gächter et al 2017 ; Houser and Xiao 2010 ; Johnson et al 2009 ; Masclet and Villeval 2008 ; Paál and Bereczkei 2015 ; Raihani and McAuliffe 2012 ).…”

“…Abbink and Sadrieh 2009;Dawes et al 2007;Paál and Bereczkei 2015;Raihani and McAuliffe 2012;Wu et al 2009) or towards those who cooperated more than the punisher (Anderson andCinyabuguma et al 2006;de Melo and Piaggi 2015;Falk et al 2005;Goette et al 2012;Herrmann et al 2008;Irwin and Horne 2013;Pfattheicher et al 2017;Pleasant and Barclay 2018; …”

Machiavellian intelligence hypothesis revisited: What evolved cognitive and social skills may underlie human manipulation.

“…HM people, however, have to pay the costs: In the long run and/or in a tightly integrated group, they are easily detected and likely to be rejected from most relationships (Wilson, Near, & Miller, 1998). LMs, on the other hand, outperform HMs in group situations that require coordinated action and where the chance of punishment is high (Paal & Bereczkei, 2015). At the same time, they frequently lose competition in face-to-face and emotionally loaded situations because they cannot detach themselves from peronal and value implications, and emotional involvement might distract them from rational way of thinking (Geis, Weinheimer, & Berger, 1970).…”

Decoupling altruistic punishment: immune punishers learn to not cooperate, but still punish hypocritically.

“…An interpretation reinforced by our finding that many punishers, 35%, were inconsistent in their use of punishment across all five rounds, choosing to punish both socially and anti-socially at different times (Table 2). These results suggest that cooperation and punishment are not linked traits, as often assumed, and that immune individuals were punishing out of a sense of hypocrisy, competitiveness, or confusion 22,[51][52][53] . Individuals may have thought they could somehow gain from the punishment, either directly, or indirectly, via a chain of interactions.…”