¹ The idea of legitimate authority as being based on dependent reasons can be found in Raz, Joseph, ‘Authority, Law and Morality’, Ethics in the Public Domain (Oxford: Clarendon Press, 1996), 210–237Google Scholar.

² I will not dwell upon individual coping mechanisms in this paper, but since the individuals help bring about what they did not wish the group to do, they often resort to ignoring their normative attitudes toward the aim or engage in self-deception concerning what they approve or disapprove of. On the concept of ignoring, see Spelman, Elizabeth V., ‘Managing Ignorance’, Race and Epistemologies of Ignorance, ed. Tuana, Nancy and Sullivan, Shannon (New York: SUNY Press, 2007), 119–131Google Scholar.

³ John Searle's view of status functions as the basis of institutional reality would suggest that status functions (with the associated desire-independent reasons for action) existed for the party members only, while the rest were acting out of fear (based on desire-dependent reasons) and not because of the recognition of the validity of the status-functions. However, it could be the case that there were no desire-independent reasons based on the validity of status-functions for the party members (or the majority of them) either. Then there was no reductively defined group agency and the corresponding institutional reality, and what constituted the existence of the state would remain a puzzle. Moreover, the explanation based on desire-dependent reasons Searle gives works for the case of occupiers using direct force, and not pretending the populace loves them. (Making the Social World: The Structure of Human Civilization (Oxford: Oxford University Press, 2010), 164Google Scholar.) Moreover, it works well with shorter-term regimes and with rather simple sets of orders, mostly prohibitions, like curfews. It does not work for a society lasting for more than half a century and based on massive propaganda and pretense. Engaging in elaborate rituals, even if only out of fear, still constitutes the validity of the rules at the societal level.

Searle does give a good description of what I would call the condition of legitimacy: ‘In order to be rationally binding on an agent, desire-independent reasons for action contained in institutional facts must be, explicitly or implicitly, created as such by that agent’ (Ibid., 131).

⁴ French, Peter, ‘The corporation as a moral person’, American Philosophical Quarterly 16: 3 (1979), 207–215Google Scholar.

⁵ Ludwig, Kirk, ‘The Ontology of Collective Action’, From Individual to Collective Intentionality ed. Chant, S., Hindriks, F. and Preyer, G. (Oxford: Oxford University Press, in press, expected publication 2013)Google Scholar.

⁶ Kirk Ludwig, personal communication.

⁷ I would like to repeat that the group members are performing a joint action, just not the one that reductively explains how the regime exists. That a joint action is performed prompts a typical response from those who want to defend reduction. For example, the response goes, political prisoners may labor at a mine in a joint action that can be described as ‘working together to avoid punishment’. While ‘avoiding punishment together’ allows individuals to share a meaningful purpose, it still does not fall under an official description satisfactory for explaining the continuity of the official group agent, say ‘supplying the state with the raw materials it needs to increase its military potential’, since none of the prisoners endorses the goal. There is an interesting contrast with the demonstrators here: the joint intention to avoid punishment may not materialize from the corresponding individual intentions in the demonstrators' case. The intentions of fellow-prisoners to avoid punishment in performing forced labor are generally clear to others because their status is relatively unambiguous: they are deprived of freedom by the open use of force and not merely its threat, and there is no public pretense that they are willingly advancing the state's goals. As the prisoners, the demonstrators fear to reveal that their individual goal in participating is to avoid punishment, but the demonstrators face a greater uncertainty about the intentions of others. Thus it is much less likely that the demonstrators' joint intention can be that of avoiding punishment together.

⁸ It should be noted that in some cases of permanent minorities, like an ethnic minority that desires to secede, the violation of the necessary condition may also occur, but in this case it is a valuable tool in finding how the bounds of membership affect legitimacy. I discuss this in the last section of the paper.

⁹ Raimo Tuomela's view of sociality can be construed as having a similar emphasis on the relation of individual and group intentions in his notion of the we-intention. He argues that the group norms and their collective acceptance are conceptual conditions for a participant's we-intention and the reason for her to perform her part (and performing it). Thus, when individuals act in the we-mode, they make group reasons their reasons by constructing them in the mode of acceptance (The Philosophy of Sociality: The Shared Point of View (Oxford: Oxford University Press, 2007), 92Google Scholar). I consider his view in the next section.

¹⁰ Tuomela thinks that a group, ontologically, consists of its members functioning as group members with collective commitment to the group ethos (The Philosophy of Sociality, 124); groups are not agents over and above group members (Ibid., 4); groups are agents or persons only in a metaphorical sense (Ibid., 145).

¹¹ Tuomela, The Philosophy of Sociality, 182.

¹² Tuomela, ‘Cooperation and the We-Perspective’, Rationality and Commitment, ed. Peter, Fabienne and Schmid, Hans Bernard (Oxford: Oxford University Press, 2007), 227–257 at 244Google Scholar.

¹³ Tuomela's account of group intention also demands that a collective's non-operative members tacitly accept the commitment of its operative members because all of them exist in the ‘we-mode’ that ascribes intentions to the group (‘Joint Intention, We-mode and I-mode’, Shared Intentions and Collective Responsibility; Midwest Studies in Philosophy XXX, ed. French, Peter A. and Wettstein, Howard K. (Boston, MA & Oxford, UK: Blackwell Publishing, 2006), 35–58 at 47Google Scholar).

¹⁴ Bratman, Michael E. ‘Shared Intention’, Ethics 104: 1 (Oct. 1993), 97–113, at 106CrossRefGoogle Scholar.

¹⁵ Similar concerns apply to Margaret Gilbert's theory. Gilbert's strongest statement of what group agency involves is that members form a plural subject when they form a joint commitment to believe a certain proposition as a body. The formation of a joint commitment occurs when each member expresses their readiness to be so committed with others under conditions of common knowledge. But in an oppressive society members can behave as if forming a joint commitment to a societal goal while merely going through the motions of ‘forming a commitment’. See: Gilbert, Margaret, On Social Facts (Princeton, New Jersey: Princeton University Press, 1989)Google Scholar.

¹⁶ French, Peter, Collective and Corporate Responsibility (New York: Columbia University Press, 1984), 13Google Scholar.

¹⁷ French, ‘The corporation as a moral person’, 207.

¹⁸ Ibid., 211.

¹⁹ Ibid., 214.

²⁰ Philip Pettit persuasively argues that groups are social integrates or institutional persons whose rules for group rationality and integrity cannot be straightforwardly correlated with majority rule. Thus, group agents have an ontological status independent of their individual members. The members act jointly to set up certain common goals and goal-setting procedures for later occasions (Pettit, Philip, ‘Joint Actions and Group Agents’, Philosophy of the Social Sciences 36: 1 (March 2006), 18–39, at 34CrossRefGoogle Scholar). They act jointly to set up a body of judgments for rationally guiding action in support of those goals, and a procedure for making such judgments in the future. They act jointly to delegate tasks to individuals in pursuit of the group goals (Ibid., 33).

²¹ The account of illegitimacy the hybrid account puts forward has various practical implications. One of them is the capacity to assess whether a group fails to qualify for a moral group right. A group right concerns a group's entitlement to a collective good, which we can define as any good that serves the interests of individuals as members of a group – e.g. language or culture. In oppressive environments, officially promoted collective goods may not represent the true interests of group members. A strategy my account would employ for the determination of whether a group's claim to a right has merit is to investigate whether, in joint actions aimed at the enjoyment of the collective good to which the group claims a right, reductive group intentions align with the institutional expression of group intentions. If they do not, the group doesn't qualify for a moral group right to the collective good the enjoyment of which the joint actions are supposed to realize.