I trust my wife. And no, this isn't the set-up to some awful joke. There may be plenty of those to come--but this is not that.
I trust my wife and yet there are lots of things that I don't trust her to do. It's easy to give you examples and none of them spell disaster for my marriage.
For instance, I trust my wife, but I don't trust her to save the last glass of champagne. I trust my wife, but I don't trust her to save me the last chocolate biscuit. I trust my wife, but there are many things I don't trust her to run a five minute mile.
What's going on here? The distinction is well known in the philosophy of trust--it's one between generalised trust ('x trusts y') and relational trust ('x trusts y to carry out some action, A'). And, crucially, these two notions of trust--generalised trust and relational trust--are not contradictories. I don't contradict myself by saying that 'I trust my wife', and that 'I don't trust my wife to save me the last of the biscuits'.
So far, so good. We had the appearance of a tension and we've now managed to smooth it over. We've managed to avoid contradiction.
What gets philosophically interesting in this space is two-fold and will lead us in the direction of infants. Don't worry, we're getting there.
First, can we give an analysis of what it means to say that 'x trusts y to carry out some action, A')? That is, can we use other concepts to understand what's said here? It would be good if we could. Philosophical projects often look to understand some concepts in terms of other, more basic concepts, and given the trust seems to have a moral component (if someone breaks your trust, it seems like they've done something morally wrong--a bit like breaking a promise) it would be good if the analysis could explain where that moral component comes from.
Second, which of generalised trust ('x trusts y') and relational trust ('x trusts y to carry out some action, A') is more fundamental? As in, can we understand generalised trust in terms of relational trust? Or do we need to go the other way around? Or, in fact, are the two concepts just completely different?
Different philosophers take different stances on this. For instance, Hawley thinks we can understand generalised trust in terms of relational trust. So, the idea might be something like: generally trusting someone is a matter of trusting them to do most things or the most important things. So, when I speak of 'trusting my wife' what I'm saying is that, for most actions (or most actions I care about) I trust her to perform them. There is something pretty plausible about that.
In contrast, there are others--and I think that this is probably the majority view at this stage--who think that this gets the story wrong and that generalised trust cannot be understood in that way. Paul Faulkner has a really nice paper on this. He gives five reasons for thinking that generalised trust is basic, the last of which we'll focus on here.
Faulkner's notes that:
Indeed, I think it's a part of what it is to trust to be able to reflect upon and experience the moral dimension of that trust. To think otherwise is just to mistake the nature of trust--at least, so I claim here.
If that's right, then that undercuts one of Faulkner's arguments for thinking that generalised trust is basic. Perhaps more pressingly, it also cuts to the hear of a key constraint on what a theory of trust should deliver. Put simply: don't worry about whether a baby trusts you.