Signal failure on central line

This book, we are told, has been translated into English from the Hebrew. It shows. I had not previously known that birds bark, that wolves hunt gazelles in deserts or that tigers menace bulls, whatever "bulls" might mean. This is regrettable, for the authors have something important to say. They offer a unifying theory that explains previously baffling aspects of animal signalling and try to hold up a mirror in which ordinary human behaviour takes on a surprising new significance.

But do they? I thought that the "stotting" of antelopes - their leaping with four legs in the air when attacked by predators - had long ago been recognised as a signal to the predator that the animal stotting was too strong and agile to be caught and that the predator would be wiser to single out the weaker of the herd, but these authors appear to believe that this is their original hypothesis. I would have thought, too, that more scientific data and less subjective conjecture would be merited to explain why male cichlid fish make nest depressions in the sand that are constantly washed away, and that they do so merely to impress females of their superior masculine determination.

The quintessence of The Handicap Principle is that by finding food and avoiding predators in spite of having to drag around such an extravagant and cumbrous train of feathers, the peacock hopes to convince the peahen that he is the high-quality mate she is seeking to father her chicks.

"Most of us have heard", the authors say, "of the 'abandoned kitten' who climbs to the tree-top and cries for help all night. Kindly neighbours try to help it assuming that it has lost its mother. We believe the kitten is trying to blackmail its mother into continuing to nurse it. Babbler fledglings I cry loudly when hungry. An observer senses intuitively that they cry in order to be fed, and it was a long time before we realised what was wrong with this explanation. After all, the cries pose a real danger. They can be heard over a great distance - hundreds of yards. They reveal to predators the fledglings' location - and at that age the latter can barely fly, or even hop. The common explanation is that the cries tell the parents where the fledglings are and that they are hungry. But the cries are often loudest when the parents are nearby and know just where to find the fledglings. ... We believe that the cries are actually meant to be heard by predators. The fledglings say, as it were, 'cat, cat, come and get me. I am here and I don't care who knows it until my parents feed me'." The fledglings are forcing their parents to feed them by endangering themselves; like the deceitful kitten, dangerous blackmail again.

In both cases, the authors assume, it is in the interest of the young to blackmail their parents into feeding and caring for them since the risk of attracting a predator is less than that of being abandoned, but they neglect to account for the possibility of the predator getting there first. Here The Handicap Principle is difficult to accept, and with other examples in the book it is not supported by sufficient evidence. In many cases, with no evidence.

The authors extend this argument even to the cries of human babies, where there are no predators and little chance of their being abandoned by indifferent parents. Here, though, the motivations are interpreted as the adults' desire to avoid the loss of prestige that continual baby crying would attract, "making a bad impression on his or her partner, neighbours and friends". I can hardly think of a more specious argument!

Human females are unique among mammals in the excessive amount of blood and associated matter they discharge in the menstrual flow. This prodigal waste of vital tissue, we are told, informs a potential mate of her ability to produce good and healthy offspring. Perhaps this was true in the Stone Age, though why only in humans, and what purpose does it serve today when ladies do not, as it were, wear their menstrual conditions on their sleeves? The Zahavis offer no explanation.

The authors believe that natural selection encompasses two different and often opposing processes. One kind of selection favours straightforward efficiency, and it works in all areas except signalling - such as the stotting antelopes or the wailing kittens. This selection makes gestures other than signals more effective and less costly; "utilitarian selection" the authors suggest calling it. The other kind of selection by which signals evolve, they write, results in costly features and traits that look like "waste". It is precisely this costliness, the signallers' investment with signals, that make signals reliable. They suggest calling this process "signal selection".

As the Zahavis see it, most of what Darwin defined as "sexual selection" is better understood to be "signal selection". Signal selection differs from sexual selection in that it includes all signals - not only those that affect potential mates and sexual rivals, but also signals sent to all other rivals, partners, enemies or anyone else. At the same time, we are asked to believe, signal selection excludes features that improve actual fighting ability which are selected for straightforward efficiency.

The authors describe many similar examples of puzzling animal behaviour and explain it with the "handicap principle". The publishers claim that the authors have discovered "a missing piece of Darwin's puzzle", which many find difficult to accept, though in their introduction the authors claim acceptance. Good science is looking for data that may refute and not only confirm a hypothesis. This has not been done with the rigour that the authors and publishers would have us believe.

Harry Miller is a fellow of the Zoological Society.