6 ideas
Morgan's Canon says an animal's behavior should never be explained by a 'higher' psychological faculty if a 'lower' one will do. The book argues this builds a bias against animal minds into the method: it assumes a ranked ladder of faculties and treats one error type, over-attribution, as worse than under-attribution. Evidence, not a default preference for simpler explanations, should decide between hypotheses.
Anthropomorphism is wrongly attributing human traits to animals. Anthropectomy is wrongly denying animals traits they actually share with humans. Both are errors, so a science that guards only against the first will systematically underestimate animal minds.
Clever Hans, a horse in early 1900s Berlin, seemed able to count and do arithmetic by tapping his hoof. Oskar Pfungst showed that Hans stopped tapping when he picked up tiny, unintended cues in the questioner's posture, and failed when the questioner didn't know the answer. The case shows that apparent cognition can come from subtle social reading, and it became the model for blind experimental controls.
Everyday understanding of others is often modeled as attributing beliefs and desires in order to predict behavior. The book argues people also predict and explain others through traits, situational norms, stereotypes and sensitivity to emotional states, without representing beliefs at all. Asking whether an animal 'has theory of mind' is therefore too coarse, since animals may be socially competent by these other routes.
In most tests, an animal that tracks what another can see could succeed simply by tracking the observable cues that a mental state would be inferred from, such as line of gaze. Behavioral experiments therefore struggle to tell genuine mental-state attribution apart from sophisticated behavior-reading.
Animals such as chimpanzees and cetaceans show socially maintained patterns: behavior specific to a group, conformity, punishment or protest when expectations are violated, and caring for others. The book argues these can count as proto-norms or moral practices even without explicit rules or language. On this view, morality comes in degrees and grows out of shared social expectation rather than reasoned principles.