Jeremy Trombley has a neat post listing three criticism of anthropocentrism: boundary, agential and perspectival.
The boundary criticism holds that anthropocentrism is wrong because it makes arbitrary distinctions between beings that merit ethical consideration and beings who do not.
The agential criticism is that anthropocentric philosophies fail to recognize nonhumans as active participants in the constitution of the world (including the human world).
The perspectival criticism is that anthropocentric philosophies fail to understand the world from nonhuman “points of view”. Since this is liable to be impossible, Trombley suggests, we must “content our selves with boundary or agential non-anthropocentrisms, but these will always fall short of the true non-anthropocentric ethics that we need.”
This typology is arguably incomplete. I’d add two related complementary forms to the mix – transcendental anthropocentrism (or transcendental humanism) and instrumentalism.
Transcendental anthropocentrism (TA) is really just another word for what Meillassoux refers to as “correlationism”. It is the claim that the nature of value, agency, or reality consists in the way these are disclosed or represented – where disclosure or representation is invested with forms derived from human phenomenology or human language.
Instrumentalists claim that only humans or persons are authentic actors while the “agency” of nonhumans like animals or tools is a human construction or projection. For example, if one believes that language use is the sine qua non of intentionality and agency, then humans (and notional non-human language users like Mr Spock) are the the only agents and representers.
Instrumentalism short circuits the agential and perspectival critiques. If instrumentalism is true we properly speak of nonhuman agency or representing in quotation marks. Something similar is implied by transcendental humanism. If a world is only disclosed to entities with culture then animals and tools clearly don’t have access to a world and thus no perspective on it.
The best reason for thinking that instrumentalism is false is arguably Darwinian. Our representational capacities are evolved from the representational capacities of nonhumans lacking human linguistic prowess. Thus, however remarkable our propositional understanding is, it could not have arisen by saltation. Some form of representation – hence misrepresentation – must have preceded it (e.g. map-like or array-like representations). Or as Patricia Churchland puts it in Brainwise: “Even if human language is unique in its complexity and representational power, general considerations from evolutionary biology imply that a theory postulating the absence of any continuity between linguistic representations and nonlinguistic representations would need skeptical scrutiny” (283).
However, even if instrumentalism is false, it does not follow that we are wrong to make ethical boundary distinctions between humans and nonhumans. Transcendental humanism might be false, even though only a very restricted subset of nonhumans has a perspective. Similarly, it seems likely that only a fairly restricted set of nonhumans have any agency to speak of. Finally, capacities for agency and representation may be widely distributed but hugely variable in nature and scope.