Baumeister et al. argue that groups can be better than the sum of their parts, and they present in their target article a theory of what makes that possible. The theory posits two stages in successful group formation: a phase emphasizing group identity and a second phase that stresses the differentiation of the individuals, particularly with respect to their work-related capacities to contribute to the group. The authors hold that although dissent may be beneficial, “background” diversity, such as race, age, and gender diversity, may harm the first phase in a way that affects the final functioning of the group. Background diversity is thus distinct from the work-related diversity (“specialized function, skills, training, expertise, etc.” [sect. 2.4, para. 9]).
A peculiar feature of the authors' claim is that there can be a harmful “excessive diversity” of ethnicity, gender, race, or age in a group, and that diversity can get to be too much of a good thing (sect 3, para. 11). The oddness of the feature is that if one were to act in light of this view, in many instances one would be in a legally questionable position. According to Title VII of the 1964 US Civil Rights Act, hiring or firing on the basis of gender, race, or age is illegal in a company of 15 or more employees. Title IX of the Education Amendments Act also places clear prohibitions on many exclusions based on background diversity.
Many theorists have argued that today appealing straightforwardly to a need to exclude members who would increase ethnic or background diversity is widely thought wrong (Brownstein & Saul Reference Brownstein, Saul, Brownstein and Saul2015; Dovidio Reference Dovidio2010; Holroyd Reference Holroyd2012). One problem with that mode of thinking is that we find that largely unconscious, implicit biases or convictions about unsuitability may lurk behind otherwise sincere statements that there were no suitable women or people of color to include, for example. Consequently, groups deciding policies on abortion, family leave, student loans, health-care resources, welfare for unwed mothers, and so on, may not include any members of the groups most affected by the policies.
The first and most obvious question that argument raises concerns the quality of the outcomes of such group deliberations when the group does not include minorities. I am not suggesting that there are somehow purely biological or innate differences between groups of white men and groups that are more diverse with respect to gender or race. Rather, the idea is that experience grounds a great deal of our knowledge, and members of such diverse groups can have very diverse access to knowledge.
Such knowledge may not be easily shared. People occupying distinctive social positions, particularly subordinate ones, may have knowledge that is generated by their social position. Thus, in her Lettres de Mlle Aïssé à Madame C of 1728, Madame Cornuel famously opined, “No man is a hero to his valet” (Ratcliffe Reference Ratcliffe2011, p. 111). And although that sort of knowledge is at least sometimes shareable (though perhaps not always [Harding Reference Harding2004]), recent discussion of microaggressions makes it very clear that members of a dominant group may have a very hard time understanding it. Members of the majority group may see in the complaints of minorities about continual daily harassment a victim culture leading to restrictions on free speech (Schmidt Reference Schmidt2015). Equally problematic, members of a majority group may have a great deal of difficulty retaining knowledge that comes from a minority perspective and giving it a place in their viewpoint (Burgess et al. Reference Burgess, Ryn, Dovidio and Saha2007).
A second problem arises with the efficacy of the decisions homogeneous groups reach. In cases where the decisions are supposed to affect and even shape the behavior of the gender and racial minorities who are not participating, will that happen if such people have no say in the process? This question really asks for facts that I do not have, though I note that, for example, when police, prosecutors, and grand juries can be seen to be in some sense one group, outsiders may strongly object to the process if they disagree with the result. Riots argue against the efficacy of the group's decisions.
A third question concerns morality rather than legality. Though we may find it convenient to restrict or reject inclusion based on gender, race, or age, doing so generally gets bad press when the discrimination is uncovered. It is seen as unfair.
Finally, I want to note that conflicts between diversity and group identity may be solved or at least mitigated. There has been a great deal of recent work on how to do so, and there has been definite progress (Brownstein & Saul Reference Brownstein, Saul, Brownstein and Saul2015; Dovidio Reference Dovidio2006; Gaertner & Dovidio Reference Gaertner and Dovidio2000; Jones et al. Reference Jones, Dovidio and Vietze2014; Kang et al. Reference Kang, Gray and Dovidio2014). Outcomes of diverse groups may indeed excel those of more homogeneous groups.
You have
Access
Baumeister et al. argue that groups can be better than the sum of their parts, and they present in their target article a theory of what makes that possible. The theory posits two stages in successful group formation: a phase emphasizing group identity and a second phase that stresses the differentiation of the individuals, particularly with respect to their work-related capacities to contribute to the group. The authors hold that although dissent may be beneficial, “background” diversity, such as race, age, and gender diversity, may harm the first phase in a way that affects the final functioning of the group. Background diversity is thus distinct from the work-related diversity (“specialized function, skills, training, expertise, etc.” [sect. 2.4, para. 9]).
A peculiar feature of the authors' claim is that there can be a harmful “excessive diversity” of ethnicity, gender, race, or age in a group, and that diversity can get to be too much of a good thing (sect 3, para. 11). The oddness of the feature is that if one were to act in light of this view, in many instances one would be in a legally questionable position. According to Title VII of the 1964 US Civil Rights Act, hiring or firing on the basis of gender, race, or age is illegal in a company of 15 or more employees. Title IX of the Education Amendments Act also places clear prohibitions on many exclusions based on background diversity.
Many theorists have argued that today appealing straightforwardly to a need to exclude members who would increase ethnic or background diversity is widely thought wrong (Brownstein & Saul Reference Brownstein, Saul, Brownstein and Saul2015; Dovidio Reference Dovidio2010; Holroyd Reference Holroyd2012). One problem with that mode of thinking is that we find that largely unconscious, implicit biases or convictions about unsuitability may lurk behind otherwise sincere statements that there were no suitable women or people of color to include, for example. Consequently, groups deciding policies on abortion, family leave, student loans, health-care resources, welfare for unwed mothers, and so on, may not include any members of the groups most affected by the policies.
The first and most obvious question that argument raises concerns the quality of the outcomes of such group deliberations when the group does not include minorities. I am not suggesting that there are somehow purely biological or innate differences between groups of white men and groups that are more diverse with respect to gender or race. Rather, the idea is that experience grounds a great deal of our knowledge, and members of such diverse groups can have very diverse access to knowledge.
Such knowledge may not be easily shared. People occupying distinctive social positions, particularly subordinate ones, may have knowledge that is generated by their social position. Thus, in her Lettres de Mlle Aïssé à Madame C of 1728, Madame Cornuel famously opined, “No man is a hero to his valet” (Ratcliffe Reference Ratcliffe2011, p. 111). And although that sort of knowledge is at least sometimes shareable (though perhaps not always [Harding Reference Harding2004]), recent discussion of microaggressions makes it very clear that members of a dominant group may have a very hard time understanding it. Members of the majority group may see in the complaints of minorities about continual daily harassment a victim culture leading to restrictions on free speech (Schmidt Reference Schmidt2015). Equally problematic, members of a majority group may have a great deal of difficulty retaining knowledge that comes from a minority perspective and giving it a place in their viewpoint (Burgess et al. Reference Burgess, Ryn, Dovidio and Saha2007).
A second problem arises with the efficacy of the decisions homogeneous groups reach. In cases where the decisions are supposed to affect and even shape the behavior of the gender and racial minorities who are not participating, will that happen if such people have no say in the process? This question really asks for facts that I do not have, though I note that, for example, when police, prosecutors, and grand juries can be seen to be in some sense one group, outsiders may strongly object to the process if they disagree with the result. Riots argue against the efficacy of the group's decisions.
A third question concerns morality rather than legality. Though we may find it convenient to restrict or reject inclusion based on gender, race, or age, doing so generally gets bad press when the discrimination is uncovered. It is seen as unfair.
Finally, I want to note that conflicts between diversity and group identity may be solved or at least mitigated. There has been a great deal of recent work on how to do so, and there has been definite progress (Brownstein & Saul Reference Brownstein, Saul, Brownstein and Saul2015; Dovidio Reference Dovidio2006; Gaertner & Dovidio Reference Gaertner and Dovidio2000; Jones et al. Reference Jones, Dovidio and Vietze2014; Kang et al. Reference Kang, Gray and Dovidio2014). Outcomes of diverse groups may indeed excel those of more homogeneous groups.