We find Smaldino's argument lucid and compelling. His clear delineation of group-level traits (GLTs) pinpoints the sorts of social complexity that cannot be explained by selection of individuals or kin and highlights the importance of coordination, in addition to cooperation, as a fundamental determinant in cultural evolution. We think the examination of GLTs should be taken further and that by doing so the concept may help catalyze advances in fields that draw on evolutionary theory.
Smaldino's call for a “between-levels perspective” (sect. 7, para. 2) parallels one for a mesoeconomic approach (Dopfer Reference Dopfer and Antonelli2011; Reference Dopfer2012) from evolutionary economics. Mesoeconomics is often considered as a placeholder for economic studies that do not fit neatly in either micro or macro categories (Dopfer Reference Dopfer and Antonelli2011). However, if economics can grow to include Smaldino's culturally transmissible GLTs subject to natural selection, we feel that mesoeconomics could “gain an independent and genuine place in the architecture of economics” (Dopfer Reference Dopfer and Antonelli2011).
However, first, GLTs may need to be redefined. Smaldino argues that GLTs are the phenotypic effect of social organization, rather than that organization itself. We agree that group-level social organization does have “phenotypic” effects (e.g., the music of a rock band) and that those phenotypic effects are subject to natural selection. However, if we call the phenotypic effect of social organization the “trait,” how then do we talk about different social structures with similar phenotypic effects, or the multiple phenotypic effects of a single type of social organization? In biology, the concept of pleiotropy refers to the condition in which one gene generates multiple phenotypic effects. Such a thing is surely also possible with social organization. This issue makes the “phenotypic effect” definition of GLT awkward to apply. For example, the music played by the rock band may be recorded and played back. We would all agree that the performance is the phenotypic effect of the rock group's organization, but if we use Smaldino's definition, then the recording of the performance would also be considered a GLT. But certainly, the recording is not a group-level organizational trait, but the output of one. Defining GLTs as the details of social organization itself solves a number of issues with the concept.
This redefinition has the benefit of simplifying how we conceptualize institutional selection. If a GLT is an organizational structure, then we can see that group-level phenotypic effects can be the result of individual-level traits, emergent GLTs, aggregate group behaviors, or any combination thereof. Discussions of group selection tend to focus on aggregate group-level effects exerting influence on the prevalence of cooperative individual-level traits within a population. GLTs are behaviors that cannot be enacted by a single individual and are built on interdependencies between individuals (e.g., zone defense, a queue, square dance). For this reason, a GLT can only exist when organization matters. Organization is only likely to matter in the context of preexisting behavioral heterogeneity and/or culturally transmitted differentiated social roles. Smaldino's general insight is that models of cultural evolution stand to be advanced substantially by the addition of these types of heterogeneity. We also believe that such a consideration will necessarily help the study of the development, or ontogeny, of social structures.
In examining the social ontogeny of structures, coordination and cooperation leap out as central dimensions. Smaldino suggests that both are necessary components of GLTs, but he stops short of defining them as hard requirements. Both cooperation and coordination influence the development of a social structure and bear on Smaldino's aggregate versus emergent distinction.
Some social structures may begin their development with a set of differentiated actors and then come to develop cooperation over time. An archetypal example of such a structure is the market economy in which individuals possess differentiated resources and skills to trade. In this example, each person has many deficits and one surplus, so most random pairings can lead to mutually beneficial exchange with only the most limited amount of cooperative action. Thus, the social structure of the market is an important boundary sitting between aggregate group behaviors such as flocking, on one hand, and Smaldino's emergent, cooperative, role-differentiated structures, on the other.
Other social structures may start with a cooperative set of individuals and develop role differentiation over time. As an example, a cooperative group happens upon a novel challenging scenario, requiring their combined effort, such as hunting a new and dangerous type of prey. Here, high levels of preexisting cooperation facilitate the development of differentiated and interlocking roles, helping the group become successful. A less cooperative group may not have had the same success.
These two archetypal institutions are further differentiated by their ongoing ontogenetic paths. In the market example, coordination can be selfish and mutually beneficial. The mutual benefit and low requirements for cooperative investment tend to make such distributed social structures resilient. Coordination based on preexisting differences is easy to develop and takes no work to maintain. Moreover, high levels of cooperation do not necessarily follow. The market example contrasts with the cooperative group scenario in which high levels of cooperation potentiate high levels of integrated and coordinated action. It is no surprise that groups that are both more coordinated and more cooperative can accomplish more complex and demanding feats. NASA would not have been able to put humans on the moon with coordination alone; cooperation and central planning are required. But the ontogenetic order of cooperation and coordination in the development of a social structure, and the social preconditions that determine that order, may determine which type of structure emerges. This begs the question: When and how do market-like organizations become more highly cooperative and tightly integrated and vice versa?
Can a coordinated, weakly cooperative organization become a more coordinated, highly cooperative one? Do economic integrations easily give rise to political and social integration? In 2013, with the recent challenges faced by the European Union, this question is not idle speculation. The scale and scope of our political-economic challenges highlight the urgent need for this research to be extended to socioeconomic theory and policy.
You have
Access
We find Smaldino's argument lucid and compelling. His clear delineation of group-level traits (GLTs) pinpoints the sorts of social complexity that cannot be explained by selection of individuals or kin and highlights the importance of coordination, in addition to cooperation, as a fundamental determinant in cultural evolution. We think the examination of GLTs should be taken further and that by doing so the concept may help catalyze advances in fields that draw on evolutionary theory.
Smaldino's call for a “between-levels perspective” (sect. 7, para. 2) parallels one for a mesoeconomic approach (Dopfer Reference Dopfer and Antonelli2011; Reference Dopfer2012) from evolutionary economics. Mesoeconomics is often considered as a placeholder for economic studies that do not fit neatly in either micro or macro categories (Dopfer Reference Dopfer and Antonelli2011). However, if economics can grow to include Smaldino's culturally transmissible GLTs subject to natural selection, we feel that mesoeconomics could “gain an independent and genuine place in the architecture of economics” (Dopfer Reference Dopfer and Antonelli2011).
However, first, GLTs may need to be redefined. Smaldino argues that GLTs are the phenotypic effect of social organization, rather than that organization itself. We agree that group-level social organization does have “phenotypic” effects (e.g., the music of a rock band) and that those phenotypic effects are subject to natural selection. However, if we call the phenotypic effect of social organization the “trait,” how then do we talk about different social structures with similar phenotypic effects, or the multiple phenotypic effects of a single type of social organization? In biology, the concept of pleiotropy refers to the condition in which one gene generates multiple phenotypic effects. Such a thing is surely also possible with social organization. This issue makes the “phenotypic effect” definition of GLT awkward to apply. For example, the music played by the rock band may be recorded and played back. We would all agree that the performance is the phenotypic effect of the rock group's organization, but if we use Smaldino's definition, then the recording of the performance would also be considered a GLT. But certainly, the recording is not a group-level organizational trait, but the output of one. Defining GLTs as the details of social organization itself solves a number of issues with the concept.
This redefinition has the benefit of simplifying how we conceptualize institutional selection. If a GLT is an organizational structure, then we can see that group-level phenotypic effects can be the result of individual-level traits, emergent GLTs, aggregate group behaviors, or any combination thereof. Discussions of group selection tend to focus on aggregate group-level effects exerting influence on the prevalence of cooperative individual-level traits within a population. GLTs are behaviors that cannot be enacted by a single individual and are built on interdependencies between individuals (e.g., zone defense, a queue, square dance). For this reason, a GLT can only exist when organization matters. Organization is only likely to matter in the context of preexisting behavioral heterogeneity and/or culturally transmitted differentiated social roles. Smaldino's general insight is that models of cultural evolution stand to be advanced substantially by the addition of these types of heterogeneity. We also believe that such a consideration will necessarily help the study of the development, or ontogeny, of social structures.
In examining the social ontogeny of structures, coordination and cooperation leap out as central dimensions. Smaldino suggests that both are necessary components of GLTs, but he stops short of defining them as hard requirements. Both cooperation and coordination influence the development of a social structure and bear on Smaldino's aggregate versus emergent distinction.
Some social structures may begin their development with a set of differentiated actors and then come to develop cooperation over time. An archetypal example of such a structure is the market economy in which individuals possess differentiated resources and skills to trade. In this example, each person has many deficits and one surplus, so most random pairings can lead to mutually beneficial exchange with only the most limited amount of cooperative action. Thus, the social structure of the market is an important boundary sitting between aggregate group behaviors such as flocking, on one hand, and Smaldino's emergent, cooperative, role-differentiated structures, on the other.
Other social structures may start with a cooperative set of individuals and develop role differentiation over time. As an example, a cooperative group happens upon a novel challenging scenario, requiring their combined effort, such as hunting a new and dangerous type of prey. Here, high levels of preexisting cooperation facilitate the development of differentiated and interlocking roles, helping the group become successful. A less cooperative group may not have had the same success.
These two archetypal institutions are further differentiated by their ongoing ontogenetic paths. In the market example, coordination can be selfish and mutually beneficial. The mutual benefit and low requirements for cooperative investment tend to make such distributed social structures resilient. Coordination based on preexisting differences is easy to develop and takes no work to maintain. Moreover, high levels of cooperation do not necessarily follow. The market example contrasts with the cooperative group scenario in which high levels of cooperation potentiate high levels of integrated and coordinated action. It is no surprise that groups that are both more coordinated and more cooperative can accomplish more complex and demanding feats. NASA would not have been able to put humans on the moon with coordination alone; cooperation and central planning are required. But the ontogenetic order of cooperation and coordination in the development of a social structure, and the social preconditions that determine that order, may determine which type of structure emerges. This begs the question: When and how do market-like organizations become more highly cooperative and tightly integrated and vice versa?
Can a coordinated, weakly cooperative organization become a more coordinated, highly cooperative one? Do economic integrations easily give rise to political and social integration? In 2013, with the recent challenges faced by the European Union, this question is not idle speculation. The scale and scope of our political-economic challenges highlight the urgent need for this research to be extended to socioeconomic theory and policy.