One of the really interesting ethical questions about suicide (among other actions) is, what counts as the choice of the real person - especially if a person making a decision is shown to have conflicting desires and motivations?
Most of the time when we make decisions, we have conflicting motivations - such is the nature of a decision. Cognitive science research suggests that, most of the time, we don't even know why we make a particular decision. We may feel that we are carefully weighing the pros and cons of action and inaction based on carefully considered criteria, but most of the time that is not how our brain apparatus actually works. (In fact, some suggest that decisions based on instinct are usually better than decisions made based on more rational criteria.)
Given this background of conflict, should we sometimes forcibly prevent a decision maker from acting until he is certain?
With many decisions and consequent actions - whether to eat a cheeseburger, whether to go to work, whether to get a divorce - people have many conflicting motivations, conscious and unconscious. There's some value to waiting to make a decision until one is sure - which could include a friend forcibly preventing someone from making a decision until the decision maker is sure - especially since many decisions, like suicide, are irrevocable. But there's also a cost to waiting to be certain (e.g. time spent being hungry until you eat the cheeseburger or being miserable until you get a divorce/commit suicide).
Taking an action is a decision between two options - acting and not acting. Both have consequences. Forcing someone to not act is making a decision for him and imposing the costs of that decision upon him without his consent.
I think a waiting period for serious decisions (like California's 48-hour waiting period for marriage) - which amounts to forcing someone to take more time to think about a decision (and imposing the costs of that time on him) - is acceptable if the costs of the waiting period were found to be, on average, smaller than the costs of poor decisions without the waiting period. But this would have to be from a perspective of maximizing happiness, rather than promoting liberty. Waiting periods are paternalistic - I'm not a hardcore libertarian, so I think that can be okay at times where it's not very intrusive. Ultimately, though, I think not only the decision of whether to act, but the decision of how long to wait before acting, should rest with the actor. And the costs (in terms of suffering) of being forced not to commit suicide are substantial - the longer the delay, the higher the potential costs.
The weirder question, which I've been struggling with, is what to do about people whose desire to commit suicide changes over time? If I sign something at age 18 that says I want to be forcibly prevented from committing suicide if I ever try it, should that be enforced when I'm 80 and want to die peacefully? If I want to die at 18, should it make a difference that I might change my mind later? I don't have much of a framework from which to answer that one.
Chip's suggestion - that we go with the "one that's speaking, whenever" - is attractive in its simplicity, humanity, and apparent respect for liberty. But if our society followed it strictly, it would prevent us from ever increasing our happiness by binding ourselves. The whole idea of a contract (from sales to employment to marriage) is to increase our overall happiness by binding the actions of our future selves. Similarly, if I had a fairly happy life but very occasionally went into a despairing funk and wanted to die, I might think I'd be better off if I could prevent myself from committing suicide during that period. (Just as I, in my real incarnation, would feel myself better off if I could prevent my future addled self from docilely swallowing the activated charcoal if a future suicide attempt proved unsuccessful.) Are present and future benefits and costs allowed to weigh against each other?
Do I owe anything to my future self - since, in a sense, it is me? Can I take anything from my future self - again, since it is me - by either imposing suffering by not committing suicide, or removing its "chance at life" by committing suicide? Ethically, do I stand in relation to my future self as toward my present self, or as toward a totally different person?
In response to an email from reader Elizabeth, who also pointed me to the Wilson book - thanks!