[watwut]

No, there does not have to be advantage nor clear answer to question. A trait that produces no advantage can remain. A trait that produces disadvantages can remain too - just less often.

Improbable things happen in practice. Just less often.

[h0l0cube]

> No, there does not have to be advantage nor clear answer to question. A trait that produces no advantage can remain.

What you're talking about is a spandrel, and eusociality is not that. I'd recommend reading the article more closely. Much of it covers the evolution argument with depth and nuance – or at least more depth than I've seen in the HN comments section.

https://en.wikipedia.org/wiki/Spandrel_(biology)

[watwut]

I am not talking about either.

[schneems]

To paraphrase my understanding of your position: a suicidal trait picked up by an entire species is a totally neutral trait that somehow neither provides advantage nor disadvantage and still hasn’t been lost and the reason that it provides neither positive nor negative benefit to that species is unknowable.

[watwut]

It is suicidal only in very limited rare circumstances. It also kills a single otherwise short living individual unable to reproduce. The individual belongs to an otherwise large colony.

So yes, the trait has negligible impact on survival of the species and it is entirely plausible for them to survive despite the trait being super slight disadvantage.

Also, they did not "picked it". They are insects, they made no decision.

[imtringued]

It is only suicidal in specific contexts and even then it is technically not suicide because it is the attacker that kills the bee most of the time.