Damn it, you can't just say that; there is math involved! On the one hand, you have the evolutionary cost to yourself: if you die, then you can't have any more children. On the other hand, you have the evolutionary benefit to your close kin: they have access to more resources with you gone. This benefit is weighted by how closely these other people are related to you. A rule of thumb is that you break even if you sacrifice your life to save the life of two siblings, four nephews/nieces, or eight cousins.
Do you really expect that a get-depressed-and-die instinct would reliably result in at least that much benefit to someone's close kin? If not, then you don't really expect such an instinct to evolve by kin selection.
>Do you really expect that a get-depressed-and-die instinct would reliably result in at least that much benefit to someone's close kin? If not, then you don't really expect such an instinct to evolve by kin selection.
Consider the case of cancer. Cells have builtin functions for commiting suicide when something goes awry, and also do not breed excessively normally. However since knocking those inhibitions out is advantageuos in the short term, cancer arises again and again. The mutations seems to confer an advantage, for the first generation, the second generation... quite a few generations actually. Then suddenly, as the cancer cells reach critical mass, and the parasite/parasitee ratio grows to large, it becomes a huge liability. The cancer cells die. Every single one. The end.
evolutionary cost to yourself: if you die, then you can't have any more children.
If we suppose that people only consider suicide in bad situations, where, say, their expected reproductive capability is maybe 1/8th that of their surviving relatives, then the evolutionary cost is 1/8th of what you assume. In this case, you would only need to save one cousin, or to have 1/4 of a chance of saving a sibling, or 1/2 a chance of saving a niece/nephew. Or maybe not "saving" a niece, but making it practical for your siblings to deliberately have another niece. ... The scenario remains questionable, and the 1/8 is of course made up, but it does seem possible.
If there exists such a thing as group selection. This is not entirely clear and many biologists remain unconvinced about that theory (Richard Dawkins most notably).
