A program for the simulation of rational social normative trust, predictive `trust,' and predictive reliance between agents will be introduced. It offers a tool for social scientists or a trust component for multi-agent simulations/multi-agent systems, which need to include trust between agents to guide the decisions about the course of action. It is based on an analysis of rational social normative trust (RSNTR) (revised version of M. Tuomela 2002), which is presented and briefly argued. For collective agents, belief conditions for (...) collective agency should be added. For the various forms of trust agents must have (at least) subjectively rational reasons to believe that the conditions of the trust account are fulfilled. A list of such reasons (of varied weights), e.g., given by empirical research, can manually be built into a parameter file or be generated by a calling program in a fixed format. From this list of reasons the program randomly generates a belief base for the agents of the artificial society. Reasons can be chained together so that one set of reasons satisfies several belief conditions. The program checks if the conditions are fulfilled for the artificial agents' social normative trust/predictive `trust'/`predictive reliance' in another agent that he will perform an action X. Each outcome is logged to a result file. In conclusion we discuss various aspects of the application of a trust component of the suggested kind in empirical research, social simulation, and multi-agent systems. (shrink)

In this paper we will study two central social notions, acting as a group member and collective commitment. Our study of the first of these notions is – as far as we know – the first systematic work on the topic. Acting as a group member is a central notion that obviously must be understood when speaking of the “we-perspective”, group life, and of social life more generally. Thus, not only philosophy of sociality, philosophy of social science, political and moral (...) philosophy but also the various social sciences need this notion and should benefit from our analyses and arguments.Collective commitment is the other “we-perspective” notion studied in our paper. We have argued for its importance as representing a kind of social glue needed for group members when thinking and acting as a group. In contrast to some other studies (e.g. Gilbert, 1989, 2000, Castelfranchi, 1995) our most elementary notion of collective commitment is not intrinsically normative but is only instrumentally “normative” and intention-relative. Thus our treatment covers more ground than the previous accounts do. (shrink)

In this paper, an account of rational social normative trust (RSNTR) and a context for rational trust (Y) will be offered and briefly argued. The account concerns a person’s trust in another person that he will perform a specific action. Rational social normative trust is conceived as the trustor’s accepting attitude vis-à-vis his dependence on the trustee. This is an attitude that the trustor acquires non-intentionally, because of his belief, due to their relationship of mutual respect, that he is entitled (...) to expect of the trustee, on social or (quasi-) moral normative grounds, that the trustee will intentionally gratify him by his action, and because of his belief that he will indeed do so at least in part in deference to his rights. The “trust context Y” involves the conditions for a situation where the trustor can rationally consider whether intentional gratification may be expected. Rational trust is distinguished from the more general technical concept of rational reliance by the presence of context Y and the trustor’s expectation of the trustee’s intentional gratification of the trustor. The trustor’s social normative expectation of the trustee’s intentional “acting with goodwill” towards him is the central belief of the trustor involved in rational social normative trust. This belief does not require the trustor’s belief of the trustee’s genuine attitude of goodwill towards the trustor. The account distinguishes social normative trust from predictive “trust” by the trustor’s socially grounded normative, as opposed to only predictive, expectation of the trustee’s intentional gratification of the trustor. Normative trust, and also predictive “trust,” are more than an evaluation of a person’s trustworthiness. In addition to expecting of the trustee and/or expecting that the trustee will intentionally gratify him, the trustor feels comfortable about being dependent on the trustee, and has an accepting attitude regarding his dependent position. Both in normative and predictive trust, the trustor may decide to depend, or to refrain from depending, on the trustee for an action. In the case of normative trust, the trustor genuinely trusts the person he decides to depend on, but in the case of predictive “trust,” his “trusting” is comparable to relying on some features of the trustee or the situation. “Deciding to trust” is to make a bet on someone – to act as if one trusted. The account of rational trust (RSNTR) will be applied to a case where the trustee is a collective agent. Criteria for collective agency is then added. Collective agency is discussed, mainly, in the light of Raimo Tuomela’s work. When the trustor is a collective, the criteria for collective agency should be satisfied for the trustor as well. (shrink)