Help as a form of control

Help is not much considered in the literature of analytic social philosophy. According to Tuomela (2000), when A
helps an agent B (1) A contributes to the achievement of B's goal, and (2) B accepts A's contribution to the goal.
We take a rather different tack. Our notion of help is one-sided and triggered by an attempt. It is one-sided because
we can provide our help to someone without her accepting it. She could be unaware of our actions, or she could be
unwilling to receive it. Helping is based on trying because it is agent B (supposedly) trying to do something that
triggers A's action of help. This is something supported for instance by Warneken and Tomasello's experiments
with toddlers (Warneken and Tomasello 2006, 2009).
Help is interesting in its own right, but also because it allows us to reconsider the philosophical underpinnings of the
central notion in this field that is the notion of control. Help is seen here as a kind of weak interpersonal control,
where an agent A's agency guides an agent B's agency.
When possible, we evaluate our framework on chosen scenarios taken from the literature in philosophy and
psychology. The analysis is driven by a formal, logical approach. In particular, we make use of the modal logics of
agency. This assists us in taking sensible philosophical choices, avoiding blatant inconsistencies. Moreover, the
resulting formalism has the potential to serve as a computational engine for implementing concrete societies of
cooperating autonomous agents.