Sentience

Sentience is the capacity for subjective experience — for there to be something it is like to be the creature in question. Paradigmatically, it is the capacity to feel: pain, pleasure, fear, comfort, distress, relief. Sentience is the hinge on which modern animal ethics turns, because it is the feature that makes a being’s treatment matter to the being itself.

This article covers what sentience is (and is not), how contemporary science tries to detect it, the two declarations that reshaped the field, the precautionary framework that now guides policy, and the expanding evidence in fish, cephalopods, and insects.

Consciousness, sentience, sapience

The terms are often run together; they should not be. Consciousness is the broader philosophical category — any state in which there is phenomenal experience at all. Sentience is the ethically-loaded subset: the capacity to have experiences with valence, states that feel good or bad to their bearer. Sapience refers to higher-order cognitive capacities — abstract reasoning, language, self-reflective thought — and is neither necessary nor sufficient for sentience. An infant is sentient without being sapient; a sophisticated chess engine is sapient-adjacent without being sentient at all.

The ethical work is done by sentience, not sapience. A being that cannot suffer has no stake in how it is treated; a being that can, does — regardless of whether it can compose a sonnet.

Nociception is not pain

A common confusion: nociception is the neural detection of tissue damage, the reflex arc that pulls a hand from a flame. Many animals (and some plants, in a looser sense) exhibit nociception. Pain is the conscious, affective experience that typically accompanies nociception in sentient beings — the awfulness of the burn, not merely the withdrawal reflex.

Lynne Sneddon’s influential framework asks whether an animal displays behaviours that go beyond reflex: does it attend to the wound, trade off pain-avoidance against other goals, show altered motivation for hours or days, respond to analgesics in human-comparable ways, and possess the neural architecture to support affective processing (Sneddon, 2015)? A ratcheting up of these criteria is what separates “detects damage” from “feels pain”.

The Cambridge Declaration (2012)

In July 2012, a group of prominent neuroscientists — including Philip Low, David Edelman, and Christof Koch — signed the Cambridge Declaration on Consciousness in the presence of Stephen Hawking. Its central claim: the neurological substrates that generate conscious experience in humans are not unique to humans. Mammals, birds, and “many other creatures, including octopuses” possess the relevant substrates (Low et al., 2012).

The declaration did not invent the evidence; it consolidated two decades of comparative neuroscience into a public, signed statement that shifted the burden of proof. After 2012, the default scientific posture moved from “prove these animals are conscious” toward “explain why they would not be”.

The New York Declaration (2024)

Twelve years later, a second declaration went further. The New York Declaration on Animal Consciousness, signed in April 2024 by Kristin Andrews, Jonathan Birch, and dozens of researchers from comparative cognition, neuroscience, and philosophy, asserts two points (Andrews, Birch et al., 2024):

1. There is strong scientific support for attributions of conscious experience to other mammals and birds.
2. There is a realistic possibility of conscious experience in all vertebrates (including reptiles, amphibians, and fish) and in many invertebrates — at minimum cephalopod molluscs, decapod crustaceans, and insects.

The declaration’s second point is the more consequential. “Realistic possibility” is not “proven”; it is the threshold at which ignoring the possibility becomes morally reckless.

Birch’s precautionary framework

Philosopher Jonathan Birch led the 2021 LSE review, commissioned by the UK Department for Environment, Food and Rural Affairs (DEFRA), of sentience in cephalopod molluscs and decapod crustaceans (Birch, Burn, Schnell, Browning & Crump, 2021). The review set out eight criteria, grouped into neural architecture, pain-like behaviour, and responses to analgesia and noxious stimuli, and rated each taxon against them. Octopuses scored “very strong” evidence; crabs and lobsters scored “strong” or “substantial” evidence on most criteria.

The review fed directly into the UK Animal Welfare (Sentience) Act 2022, which added cephalopod molluscs and decapod crustaceans to the list of animals legally recognised as sentient.

Birch’s later book The Edge of Sentience (2024) generalises the approach into a precautionary framework. The core move: where there is a realistic possibility of sentience, the expected moral cost of treating the being as if it were insentient is too high to ignore. Policy should apply proportionate protections keyed to the strength of evidence, not binary inclusion-or-exclusion.

Fish pain: the Sneddon programme, and Key’s dissent

Lynne Sneddon’s programme at the University of Liverpool has produced the most systematic behavioural and neural evidence for pain in fish. Rainbow trout injected with noxious chemicals in the lip show rocking, rubbing, suspended feeding, altered respiration, and attentional deficits — and these changes are reversed by morphine (Sneddon, 2015, and earlier work from 2003 onward). Zebrafish will pay a cost (leaving a preferred environment) for access to analgesic-laced water when injured.

Not everyone is persuaded. Neuroscientist Brian Key’s Why fish do not feel pain (2016) argues that fish lack the cortical structures he takes to be necessary for conscious pain. The ensuing exchange in Animal Sentience is a useful reference for how such disagreements are adjudicated. The majority view, reflected in the NY Declaration, is that Key’s neocortex-centric criterion is too restrictive — birds lack a neocortex and plainly have rich cognition — and that functional equivalence in fish forebrain structures (pallium, telencephalon) is a plausible substrate.

Octopuses and other cephalopods

Cephalopods are the poster organisms for the convergent evolution of minds. Their last common ancestor with vertebrates was a flatworm-like creature more than 500 million years ago; whatever cognition they have, they built from scratch, with a nervous system distributed two-thirds into their arms. And yet: tool use in Amphioctopus marginatus, individual recognition of human handlers, REM-like sleep states with colour changes, and the behavioural and neural signatures of nociceptive sensitisation and protective wound-tending (Birch et al., 2021). The EU has required anaesthesia for cephalopod research since 2013.

Insect sentience

The most contested frontier. Lars Chittka’s The Mind of a Bee (2022) synthesises decades of work on bumblebee cognition: bees display numerical cognition, tool-use-like behaviours, play (rolling wooden balls for no food reward), metacognition-like opting-out of difficult trials, and pessimism-like judgement biases after aversive events. None of this proves sentience, but each is a behavioural signature that, in mammals, is taken as evidence for it.

Crump, Birch and colleagues’ 2022 review in Animal Welfare applied a framework analogous to the decapod review to insects, finding substantial evidence in some orders (particularly adult Diptera and Hymenoptera) and weaker evidence in others (Crump et al., 2022; see also the NY Declaration’s explicit mention of insects). Given that insects are farmed at scales approaching the trillion, the stakes of the question are not merely academic.

Why sentience is the moral pivot

Ethics is not, at bottom, about intelligence, language, or relatedness to humans. It is about whose experiences count. If a being can have experiences that matter to it — that it prefers to avoid or seeks out — then those preferences are a claim on how it may be treated. To ignore the claim because the being belongs to a different species is to smuggle a morally irrelevant property into the centre of the argument.

The empirical work on sentience does not, by itself, settle the ethical question. But it removes the easiest escape route — the insistence that only humans are “really” conscious. That route closed, for mammals and birds, in 2012; it closed, provisionally but seriously, for fish, cephalopods, and at least some insects, in 2024. The rest is a question of what we do about it.

See also

• Ethics — the trunk article on the moral case for veganism.
• Speciesism — the error sentience exposes.
• Animals — the beings the evidence is about.
• Argument from marginal cases — the structural reason sapience cannot do the work sentience does.