Excellent points. The question would be whether there are multiple genus/specific-difference definitions which would pick out all governments, and not pick out any non-governments.

If so, then that’s great. For instance, “risibility” is something only humans have, even though Aristotelians would list it as a proprium, I think (a property which all humans have necessarily) rather than part of the definition of human. So we could have a definition of humans that went:

Genus: Animal
Specific Difference: Risibility

But Aristotle would claim that risibility depends on rationality, so it’s more appropriate to use “rationality” than “risibility” as the specific difference.

I also think you’re right that the genus of government is “human group.” And I don’t think that “government” is a pseudo-concept. Government might be a pseudo-entity (i.e., people may talk about government as being a kind of thing it isn’t, if you can even call it an “it”), but I think you can at least come to a rational understanding of what it means for a person to “be in a government.”

Andrew–
Sounds right to me. Aristotle was amazingly insightful in many areas, and the genus/specific-difference approach is very helpful.

For example, if preschool teachers were taking their class to the bathroom on a field trip, a taxonomy based on gender would determine who goes into which bathroom. This taxonomy wouldn’t discriminate age. On the other hand, when they are lined up to enter the place, there will probably be different prices for adults versus children, but not different prices for males verus females.
Accordingly, one taxonomy might have species: Government genus: human groups wielding authority or power. Other species within the genus: biological familes, religious groups, etc. The family might be human groups. The other genus within the family might be human groups not wielding obvious authority or poer. The species within this genus might be social clubs.
On the other hand, it might not make any sense to group all governments together within the same species at all. I could imagine a genus based on the idea of “groups that we have the ability to opt out of” and another genus based on the idea “groups that we don’t have the ability to opt out of” Modern governments which allowed for citizens to leave (e.g. The U.S) would be species that belong to the former. Ancient governments or ones that don’t allow citizens to leave (E.G. Cuba) would belong to the latter.
Perhaps this was exactly your point: That maybe the category of “government” doesn’t hold any water.

I heard a rather interesting argument against religious inlusivism (and for religious exclusivism/particularism) that took a similiar track.
The author made the claim that “Religion” is an incoherent category. He said it’s a modernistic, reductionistic move to lump Hinduism, Christianity, Budhism, Wicca, etc. into the same group. Given that they don’t even belong to the same category the author claimed it’s nonsense to claim that they are all doing the same sort of thing and leading to the same sort of place.
(Just for the record, I think he’s wrong on some of this: I think you can note enough commanalities between the major world religions to atleast come up with a definition for religion… But it struck me as a really interesting argument.)