Trust Between Prisoners? Families? Ethnic Groups? Genders?

I found the most interesting part of Sanderson’s argument her reliance on group identification. She claims that to get out of a prisoner’s dilemma, one must 1) identify with a group, 2) commit to a joint strategy, and/or 3) discuss options with other prisoners. In the case of the regular prisoner’s dilemma, the two prisoners cannot speak with each other. Sanderson holds that they would still be able to both achieve the minimum years simply by “taking the standpoint of collective deliberation” by imagining what would the other prisoner want. Figuring out the desired collective outcome, Sanderson reminds us, forces prisoners to think in terms of “we” instead of “I”. But when regarding themselves as “members of a single collective agency, the parties are committed to acting only on reason s that are universizable to their membership” (29). This, Sanderson says, “rules out the principle of maximizing expected utility as an acceptable principle of rational choice” for people in the prisoner’s dilemma who happen to have collective agency (29). Practically, the prisoner could identify with any group that has a shared intention; yet she also says, that this is not a requirement for group identification, and “we have good reason to practically identify and cooperate with perfect strangers” (30).

I was with her until this last idea—that perfect strangers could get out of prisoner’s dilemmas by sheer concern for the other human. If I don’t know this other prisoner, and they are also guilty of a heinous crime, why do I have any reason to trust her? Why would our shared intention of getting as little jail time as possible constitute reason enough for me to trust she will also not confess? It seems that within the example of an actual prisoner’s dilemma, I don’t understand how this mechanism would work. Sanderson suggests this idea of trust is important in cases where there isn’t complete knowledge that the other person will comply: “Where such common knowledge is absent, it could still make sense to take others on trust. In this case, one regards oneself as part of an imagined common agency, in the hope that others will join and make it real by cooperating” (32). I just don’t see this being possible in the case of prisoners. I don’t think it’s necessary for prisoners to come together and discuss and come up with a deal, because that still would present the same issue of trust: they were strangers before and both are guilty of breaking the law. Sounds pretty untrustworthy to me. Where does group membership actually begin to play a role, if between prisoners doesn’t work? Families? Ethnic groups? Genders? I’m curious as to where we actually begin to trust others to comply and what the minimum requirement would be for trusting someone else within a prisoner's dilemma without verbal commitment. Maybe this line looks different for different people, and I'd be curious to know where you all would individually draw it.