The question of whether animals possess minds — and if so, what kind of minds they possess — has occupied philosophers since antiquity, but it is only in the past half-century that the discussion has acquired the empirical grounding necessary to make it something more than an exercise in speculative metaphysics. The transformation has been driven not by any single breakthrough but by the patient accumulation of evidence from ethology, comparative psychology, and neuroscience, evidence that has progressively narrowed the conceptual space within which the old Cartesian denial of animal consciousness can be maintained. Descartes's characterisation of animals as elaborate automata — beings whose behaviour, however superficially purposive, is entirely explicable in mechanical terms — was always morally convenient; it is now, in addition, scientifically untenable.
The difficulty, however, lies not in establishing that animals are conscious — a proposition that the Cambridge Declaration on Consciousness of 2012 affirmed with unusual institutional authority — but in determining what their consciousness is like. Here the philosophical obstacles are formidable, and they derive not from any deficiency in our observational methods but from a structural limitation inherent in the very enterprise of understanding a subjectivity radically unlike one's own. The philosopher Thomas Nagel framed the problem with celebrated precision in his essay on the experience of being a bat: even if we knew everything there was to know about the neurophysiology of echolocation, we would still not know what it is like to navigate the world through sonar, because the qualitative character of that experience — what Nagel called its 'what-it-is-likeness' — is accessible only from within the perspective of the creature that has it.
Nagel's argument is sometimes taken to establish the impossibility of cross-species understanding, but this is to draw a stronger conclusion than the premises warrant. What the argument actually demonstrates is that complete empathic identification with another species is impossible — that we can never fully inhabit the experiential world of a bat or an octopus or a dog. But this limitation, significant as it is, does not preclude all understanding; it merely constrains its character. We can observe, measure, and interpret the behaviour of other species; we can identify functional analogues of our own emotional and cognitive states; we can construct models that, while necessarily incomplete, capture something genuine about the inner lives of non-human animals. The knowledge we obtain is partial, perspectival, and provisional — but the same, if we are candid, is true of the knowledge we have of other human minds.
The most philosophically interesting recent work in this field has concerned the emotional lives of animals, and in particular the question of whether non-human species experience anything that merits the designation of grief. The evidence, drawn principally from observations of elephants, cetaceans, and certain corvids, is suggestive though necessarily inconclusive. Elephants have been observed returning repeatedly to the remains of deceased companions, touching the bones with their trunks in behaviour that is difficult to characterise as anything other than ritualistic. Orcas have been documented carrying the bodies of dead calves for days or weeks, long past the point at which any practical function — thermoregulation, protection from predators — could plausibly be attributed to the behaviour. Crows have been seen gathering around dead members of their species in what researchers have tentatively described as 'funerary' assemblies.
The temptation, when confronted with such evidence, is to oscillate between two equally unsatisfactory poles: the anthropomorphic, which projects human emotional categories onto non-human behaviour without adequate justification, and the mechanistic, which denies the relevance of emotional categories altogether and insists upon explanations couched in the arid terminology of stimulus and response. The most intellectually honest position, I would suggest, lies between these extremes — in the recognition that animal grief, if it exists, is neither identical to human grief nor reducible to reflexive behaviour, but is instead something for which our existing conceptual vocabulary may simply be inadequate. The challenge is not merely to answer the question but to develop a language in which the question can be properly asked.