Do animals have morals? Darwin thought they did, and that by studying them we can see how human morality has evolved. He pointed out that their capacity for love and sympathy, particularly between parents and offspring, and the powerful set of instincts that bind animal societies together, can lead to behaviour that in humans we would unhesitatingly call moral. He instanced recorded cases of wild birds — crows and pelicans — that have gone blind but have been fed and kept alive by other members of their group. Marc Bekoff and Jessica Pierce are keen to extend Darwin’s claims, drawing on modern research. Bekoff specialises in animal behaviour and Pierce in philosophy, so they are equipped to approach the problem from both sides, and their hope is that they may improve the way animals are treated by showing they are more like humans than is generally supposed.

Their book is well meaning but not well written. It is pitilessly repetitive and frequently offers belief as a substitute for argument. Their examples of moral animal behaviour often seem questionable. They relate, for example, that a female elephant in Natal once rescued a group of captive antelopes by undoing the latches on the gate of their enclosure with her trunk. No other details are given, so, as with other of their anecdotes about animals in the wild, what happened and why remain purely speculative. An instance of fellow feeling among laboratory animals that they cite took place in a group of diana monkeys that had been trained to insert a token into a slot to obtain food. The oldest female in the group could not manage this and her mate, after watching her unsuccessful attempts, picked up the tokens she had dropped, put them in the slot, and allowed her to have the food that came out. This looks more convincing than the liberationist elephant, but it does not make much advance on Darwin’s observation that animals help the handicapped in obedience to their instinct for group solidarity.

One of the book’s strongest sections draws on Bekoff’s interest in the rules that govern play among young wolves, coyotes and domestic dogs. It is in play, he argues, that individual animals learn co-operation, trust and forgiveness, and develop a sense of right and wrong. Play requires a mutually agreed standard of fairness, and if one animal violates it by being too aggressive, play will stop until the offender has indicated by gestures that it apologises, and has been forgiven. Pups that repeatedly offend are driven out of the group. It is not clear how Bekoff and Pierce explain the inability of these delinquent pups to learn social skills. But they note that work on baby rhesus monkeys indicates that if they are separated from their mothers at birth, or otherwise starved of early nurturing, they are likely to grow up socially incompetent, and perhaps it is the same with coyote pups.

From a human angle, that might make them less culpable, and so less immoral. Their education is to blame. But for Bekoff and Pierce that is probably no excuse. Morality, they insist, is “species relative”: that is, rhesus-monkey morality, human morality and coyote morality are all different. So that whereas human morality entails free will and moral choice, rhesus-monkey and coyote morality need not. “Even behaviour that is conditioned or instinctual can count as moral.” Animal morality shows itself in action. It consists in obeying social rules. On the Bekoff-Pierce model, it would be impossible for a moral rhesus monkey to behave immorally, or for an immoral rhesus monkey to pretend to be moral for the sake of convenience, whereas human beings manage these things all the time.

Knowing what goes on inside an animal, and whether what goes on includes moral choice and a belief in free will, are obviously impossible, which is why Bekoff and Pierce do not make them obligatory requirements for moral animals. The problem of empathy, which they define as “feeling with” another creature and sharing its emotion, is the obstacle their book keeps crashing into. They describe several experiments that are designed to show that animals feel empathy. Rats undergo an increase in heart rate and blood pressure when they are made to watch other rats being decapitated. When mice are injected with acetic acid, which causes a severe burning sensation, they writhe in pain; and other mice, watching them, become more sensitive to pain themselves, and writhe more violently if they are injected. More sophisticated experiments aim to demonstrate that animal empathy can express itself as benevolence. In 1959, a researcher at Brown University trained rats to press a lever in order to get food. He then constructed what Bekoff and Pierce describe as a “torture chamber” in an adjoining cage, where the bottom of the cage was an electric grid on which a second rat’s delicate pink paws were placed. When the first rat pressed his food lever, a surge of electricity ran through the grid, giving the other rat an electric shock. It was found that rats would not press the food lever if they could see a fellow rat receiving a shock.

Besides being morally disgusting, this experiment does not prove that the first rat experienced fellow feeling or benevolence. It may simply have been alarmed and distressed by the sight, smell and noise of a tortured rat nearby, and have avoided causing a repeat performance out of pure self-concern. More to the point, the three experiments do little to encourage Bekoff and Pierce’s pious hope that people will be kinder to animals if they realise they are moral beings. Their arguments become hopelessly tangled as they circle round the impossibility of getting inside an animal’s being. It is true, they admit, that we shall never be able to “enter into the subjective experience of another”. Yet, astonishingly, they go on to claim that “this is not an impediment to understanding how animals feel and think”. We can “safely infer” that their experience of pain, joy, envy, compassion and love is “probably very much akin” to ours, and that they have cognitive and emotional lives “every bit as rich as our own”.

This wishful thinking, masquerading as science, takes no account of language, or of the huge difference it has made to human brain development. Inadvertently, Bekoff and Pierce slip into writing about animals as if they had human language. “Ought” and “should”, they claim, “play an important role in their social interactions”. On the contrary, they play no role at all, because animals do not use words, and are not tangled in the vast web of concepts that words bring. In the end, the Bekoff-Pierce approach claims too little for animals, not too much. Likening them to people is defamatory, because they are innocent and we are not. It also diminishes them, because by comparison with humans they look handicapped. What is needed is the imagination to see that they are utterly beyond our knowledge — that language is the impenetrable veil that will always prevent us from knowing what it would be like to be them, and to experience things in all their blazing uniqueness without language getting in the way.

Wild Justice The Moral Lives of Animals by Marc Bekoff and Jessica Pierce

Chicago UP £18 pp204

Times Recommends

• The 100 Best Books of the Decade
• The Times picks the 50 best paperbacks of 2009
• The conversation: James Ellroy