[MD] Step two

[ARLO JAMES BENSINGER JR]

[Jan-Anders]
Transposed to human and animal organisations this means that groups of different members playing certain roles are making better results than individuals.

[Arlo]
I think its important to stress that 'individuals' whose behavior is mediated by a social-semiotic system (e.g. "language") are still acting "socially" even if there are, at that precise moment, no other humans co-present. In other words, "social activity" is not measured by counting the number of biological agents present at any given time, but by looking at the nature of the activity in question. 

As Siouxsie sings in Israel: Even though we're all alone. We are never on our own. When we're singing, singing.

In the example above, it is the semiotically-mediated nature of the activity of the 'different members' that evidences social activity. This also starts to allow us to zoom our focus in at the fractal boundary between biological and social patterns. As you suggest (and I agree) certain animal (non-human) groups do evidence (what I'd call) proto- or rudimentary social behavior, precisely because we see proto-semiosis in certain animal activity. Certainly, we do not see a level of semiotic complexity among animal activity that we do see within human activity. But I do believe this is a difference of 'phenotype' not of 'genotype'.