In anarchist discourse, a group-entity is usually distinguished from an individual hominid, or animal groups from a single living being of any sexual species. All group-entities are assumed to have certain characteristics in common, most of them sociopathic or destructive to the interests of individuals.
Groups are seen often as akin to asexual beings, reproducing by copying (e.g. cloning) rather than sharing, and uninterested in interaction other than consumption (often consuming each other in misnamed 'mergers'). Group relations are described as limited to 'eat-or-be-eaten', in contrast to individuals who (do also eat but) can engage in more complex types of interaction with the more complex cognition that sexual reproduction and social living require.
Stricter anarchists generally deny the assertion that private corporations, for instance, are fundamentally different from trade unions or political parties or religious institutions or even non-governmental organizations. All such entities are seen as self-interested and interested only in their own propagation - their relation to individuals is predatory, parasitic, and only rarely symbiotic. Even if those individuals perceive those groups as serving their own interests, the anarchist argues, they are actually accepting a proxy for those interests.
In particular the distinction is important for sexual interests, which are the most obvious interests that group-entities cannot have, having no sexual feelings, and able to deal in sex only as a commodity not as a direct organismic interest, e.g. a dating game show which is motivated to indulge sexual interest but only to sell advertising, thus distorting the mating process to serve commerce. A response to this argument is that groups also do not feel hunger, but can act as a means for individuals to satisfy their hunger - this being the main motive for the corporation and collective economic effort, back to earliest systems of irrigation.
Bob Black argued that it was not feelings or intentions or urges but the actual model of collaboration that distincted the group-entity. The sexual individual animal can conceive (pun intended) of seduction, dance, flirtation, and other means of cooperation that do not involve one entity consuming and destroying the identity of the other. Group-entities have no such skills nor capacities - they interact with each other in a way remniscent of predator-prey relations, where the predator will consume and destroy the identity of the prey, which no longer reproduces but contributes to the predator's energy. Corporate mergers, for instance, are often presented as relationships of equals cooperating, but in fact usually one executive group or the other will be drastically reduced, the organizational structure and ethics of one of the two being obliterated. Corporations do not get each other pregnant, nor do they nurture child corporations - joint ventures being a notable exception.
Some anarchists employ arguments remniscent of sociobiology and relate the behavior of other animals to those of humans - including that of animal groups. They may rely on observations from ecology and biology. For example, in a fish school, the fish have certain habits that make the school a cohesive group-entity, and which they perceive protect them from predators. But it may also be true that the school itself acts to attract predators, prevent creative or evasive actions, and acts with some collective intelligence to cut out its weaker members - none of which necessarily benefits the individual fish obeying 'school rules'. Much of anarchist discourse consists of comparing human behavior to that of other pack or social animals, and focuses on the 'collective stupidity' such habits imply - also known in psychology as groupthink. John Zerzan considers even ideas of "number" to reflect such a groupthink.
Some argue that humans are the stupidest of hominids when massed socially, no better than domestic animals bred for slavery, and have thus lost much of their capacity to solve ecological or social problems with any creativity. War, fundamentalist religion and pathological consumption are often seen as mere symptoms of this. These phenomena are sometimes presented as arguments against civilization itself as a process. The theory of eco-anarchism focuses on reducing reliance on group-entities specifically to reactivate creativity, and strengthen inter-generational ties as an alternative to same-generation peer-groups (seen as prone to peer pressure) or 'school'. Daniel Quinn's book Ishmael presents a positive vision of tribalism and hominid personhood (the book being structured as a conversation between man and ape) is an influential work of this sort.
An interesting question is whether a family which exists primarily or only to provide economic, infrastructural, instructional and emotional support to specific individuals who are related, is a group-entity distinct from feelings or methods of individual sexual animals. Families unlike the more economic or political entities focused on the outside world, do seem sometimes to merge or alter themselves so as to relate effectively in ways analogous to individuals' - indeed, a traditional family is formed by exactly such a 'dance' between two such individuals. Positive notions of tribe would seem to flow from kin or clan relationships, as studied in cultural anthropology and seemingly universal in not only human but also Great Ape societies.
Lewis Thomas noted in Lives of a Cell, 1975, that people deprived of traditional life-ways could often behave in ways remniscent of the sociopathic and predatory ways that nation-states interact, e.g. in zero-sum games. The Iks, a people he used as example, were not longer able to hunt and pursue their forest lifestyle, and when forced to become farmers were reduced to playing cruel practical jokes on each other, revelling in each others' errors and misfortunes, and squabbling in ways that reminded Thomas of the United Nations, i.e. not very united at all, and not reflective of a real 'society'.
Accordingly, nationalism and the interaction between nation-states (or the political factions or parties that control them might be the most appropriate or least controversial application of the theory of group-entities. Its application below that level, even to bioregional democracy, is questionable to the degree that the group or the entity does have some biological identity, e.g. an ecoregion, a species, a food chain, the carbon cycle of the atmosphere, the water cycle of a watershed. Such bases for organization would seem, like that of the kin, clan, or family, to align biological and group interests, and therefore overcome the objections that group and individual entities have nothing in common.
Adapted from Wikipedia article, "Group entity" http://en.wikipedia.org/wiki/Group_entity under the GNU Free Documentation License.