"Animals aren't rational."

Como pode aparecer

The strongest version

Strongest version (steelman). In the philosophical tradition, the argument comes from Aristotle, Thomas Aquinas, Descartes and Kant; today it appears, for example, in the contractualism of Peter Carruthers. Charitably reconstructed, it goes like this:

1. P1Only rational beings (capable of abstract reasoning, deliberation and self-reflection) have full moral status, that is, interests that create direct duties for us.
2. P2Non-human animals aren't rational in this sense.
3. C1So non-human animals don't have full moral status.
4. P3If a being lacks full moral status, it's permissible to use it for important human ends, provided there's no gratuitous cruelty.
5. C2So it's permissible to use animals for food, clothing and research.

There's also a moderate version: rationality wouldn't be a condition for having some moral status, but would give human interests much greater weight. Both versions are dealt with in this chapter.

The argument is valid: if the premises were true, the conclusions would follow. The problem lies in its soundness: P1, P2 and P3 are false or lack sufficient justification.

Full answer

1. a

   Reconstruction and diagnosis

   As we've seen, the argument is formally valid. So it has to be assessed premise by premise. Three points give way: the normative criterion (P1), the empirical claim (P2) and the bridge between status and use (P3). Before them comes an earlier problem: the word "rational" is ambiguous, and the ambiguity creates a dilemma.

   The dilemma of ambiguity. "Rational" can have two senses:

   • Strong sense: abstract reasoning, propositional language, the ability to reflect on one's own reasons.
   • Weak sense: learning, making inferences, solving new problems and adjusting behavior to goals.

   Rational in what sense?

   • Strong: abstract reasoningP2 becomes true, but P1 excludes babies and people with profound intellectual disabilities
   • Weak: learning, inferring, solving problemsP1 now includes those humans, but P2 becomes false for many animals

   There's no sense of "rational" that makes P1 and P2 true at the same time without excluding humans whom the speaker protects. Those who use the argument usually switch between the two senses without noticing: the strong one to exclude animals, the weak one to include all humans.

2. b

   Where the argument breaks

   1. The criterion is irrelevant to what's at stake (P1).

   A characteristic only justifies different treatment if it relates to the interest affected. Requiring political understanding in order to vote can be debated; requiring political understanding in order not to be beaten is absurd. Rationality is relevant to some interests, such as political autonomy and moral responsibility, but not to the interest in not feeling pain, fear or deprivation. Jeremy Bentham summed this up in 1789: "The question is not, Can they reason? nor, Can they talk? but, Can they suffer?"

   From here comes the sentientist foundation. Having interests means being able to be affected, for better or worse, from the point of view of whoever lives the experience. A stone has no interests; a pig does. That's why sentience is a necessary and sufficient condition for moral inclusion, and the principle of equal consideration of interests (Singer) requires that similar pains count similarly. The criterion is also the most parsimonious: a single trait explains why we protect adults, babies, people with dementia and, already today, dogs and cats.

   2. Species overlap (P1 against humans themselves).

   The argument is old: it appears in Porphyry, in the third century, and in Bentham. Today it's called the argument from species overlap (Horta, 2014), a name replacing the older "argument from marginal cases". In the form of modus tollens:

   1. Q1. If rationality is necessary for full moral status, then humans without rationality don't have full moral status.
   2. Q2. Humans without rationality (newborns, people with profound intellectual disabilities, people with advanced dementia) have full moral status.
   3. C. So rationality isn't necessary for full moral status.

   Q2 is accepted by the speaker. The argument doesn't propose downgrading these humans: it shows the criterion is wrong and that the moral community should be widened, not narrowed (Naconecy). The usual escape routes have their own chapters: the "species norm" (A11) judges individuals by what others have; "potential" (A12) doesn't cover those who will never have rationality; "special relationships" (A13) make protection depend on someone caring.

   3. The empirical premise is a caricature (P2).

   As early as 1871, Darwin held that the mental difference between humans and other animals is one of degree, not kind. Research in animal cognition has confirmed this gradual picture (see the table in c). An honest caveat is needed here: citing animal intelligence helps undo the false "rational versus irrational" divide, but using it as the foundation is a trap. It would mean accepting the speaker's criterion and concluding that chickens or fish, supposedly less intelligent, matter less. The argument against exploitation doesn't depend on animals being intelligent; it depends on their being sentient.

   4. The bridge between status and use doesn't hold (P3).

   Even in the moderate version, where rationality would give human interests more weight, the argument doesn't reach its conclusion. First, that extra weight would have to be enormous to outweigh intense, prolonged suffering, plus death, in exchange for taste and convenience, since there are alternatives (J11). Second, the "important human ends" of P3 are, in practice, mostly preferences, not needs. Third, in the Kantian version, duties toward animals are only indirect: cruelty would be wrong because it hardens human character. It follows that torturing a dog in secret, with no effect on anyone, wouldn't be wrong in itself, a conclusion almost no one accepts.

3. c

   Data and sources

   The table separates solid findings from fragile ones. A book that accuses the other side of using selective data can't do the same.

   Animal	What was observed	Study	Reliability
   Mammals, birds, fish and many invertebrates	Strong scientific support for conscious experience in mammals and birds; a "realistic possibility" in all vertebrates and in cephalopods, decapods and insects	New York Declaration on Animal Consciousness, 19 Apr. 2024	High as a synthesis of experts (it's a declaration, not a study)
   New Caledonian crow	Bent a straight wire into a hook to retrieve food from a tube	Weir, Chappell and Kacelnik, 2002	High; later studies show bending twigs is part of the species' natural repertoire, which reduces the "invention" aspect but not the flexibility
   Cichlid fish (Astatotilapia burtoni)	Inferred the hierarchy among other fish just by watching fights (transitive inference)	Grosenick, Clement and Fernald, 2007	Moderate
   Cleaner wrasse (Labroides dimidiatus)	Reacted to a mark on its own body seen only in a mirror	Kohda et al., 2019	Debated: the behavior was observed, but whether it indicates self-awareness is contested
   Sheep	Recognized dozens of other sheep's faces for more than a year	Kendrick et al., 2001	Moderate
   Domestic chicken	Review of studies: object permanence, self-control, simple numerical concepts	Marino, 2017	Moderate (literature review)
   Pig	Used a mirror to find hidden food	Broom, Sena and Moynihan, 2009	Low: not replicated by Gieling et al., 2014. Don't use as proof

   The overall picture is one of continuity, with capacities unevenly distributed among species. None of these data, however, is needed for the moral conclusion: the evidence of sentience, which is far more robust than the evidence of complex reasoning, is enough.

4. d

   The strongest counter-reply: contractualism

   The philosophically strongest version comes from contractualism, defended on this subject by Peter Carruthers (1992). Morality would be the set of rules that rational agents would accept in order to live together. Only those who can understand, accept and follow rules are party to the contract. Animals would be left out and would have, at most, indirect protection, because cruelty reveals bad character. Non-rational humans would be included for two reasons: the contractors care about them, and excluding them would create social insecurity, since anyone could come to lose their rationality. A close variant, Carl Cohen's, grants rights through membership of a species of moral agents and is dealt with in chapter A11.

5. e

   Why the counter-reply also fails

   It confuses who makes the rules with whom the rules protect.

   Moral agents are those who can have duties; moral patients are those who can be harmed. Babies are patients without being agents, and no one concludes that their protection is therefore a favor.

   It makes the protection of vulnerable humans contingent.

   If a baby is protected only because the contractors care about them, then torturing an orphan nobody cares about wouldn't be wrong in itself. The theory fails precisely the test of our firmest moral judgments.

   The contract's own device points the other way.

   Behind Rawls's veil of ignorance, rules are chosen without knowing what position one will occupy. Mark Rowlands (1997) observed that if we also didn't know whether we'd be human or some other sentient being, we'd choose rules protecting all sentient beings. Excluding animals only appears when the contract is designed by those who already have power, and power isn't justice (A2).

   Even if accepted, it doesn't justify the real practice.

   Contractualism itself condemns cruelty; large-scale farming and slaughter systems, with confinement and mutilation, would hardly pass even that minimal filter.

Reasoning errors

Biases

Extrapolations

The universalization test applies the principle "whoever isn't rational may be used" to other cases. If the person rejects the consequences, they must abandon or reformulate the principle.

Contradictions

What the argument gets right

Connections

A1 usually appears right after a claim of superiority and, once answered, opens the way for attempts to save the species boundary by other routes.

References