A horse named Hans; the search for a 'talking animal'

UNESCO Courier, Feb, 1988 by James Serpell

It took the scientific world many decades to recover from this ignominious piece of self-deception and, when news of talking animals resurfaced again in the early 1960s, it involved not horses, but dolphins. It had long been recognized that marine mammals, such as dolphins, were intelligent and highly trainable creatures which could be taught to perform astonishing tricks for public entertainment. But in 1961, a book entitled Man and Dolphin was published containing far more exciting observations.

The author, John C. Lilly, was a neurologist who had been conducting experiments on the brains of live dolphins, focusing particularly on their vocal behaviour. Despite early setbacks (many of his experimental animals died under anaesthetic), Lilly succeeded in training dolphins to vocalize on command, and he noticed that some of his animals spontaneously produced passable imitations of human speech. This finding, coupled with the fact that dolphins possess exceptionally large and complex brains, inspired Lilly to make the optimistic prediction that "within the next decade or two" humans would establish two-way communication with another species, probably a marine mammal.

Lilly's book attracted considerable media attention and stimulated a number of futuristic novels and films. More surprisingly, perhaps, his arguments were so convincing that research on human-dolphin communication began to attract considerable government funding. Alas, despite these efforts, Lilly's talking dolphin never materialized. Within a decade it was apparent that the spectre of Clever Hans had simply reappeared in an unfamiliar disguise, and the entire programme of research was discontinued.

Evidently undaunted, the scientific community once again shifted its attention, this time to focus on anthropoid apes. As our nearest living relatives, apes have long been the objects of special fascination, and the idea of teaching them to speak was far from new. The earliest experiments in this direction, conducted by various people between 1896 and the 1940s, were, however, largely unsuccessful. In 1947, for example, a young American psychologist and his wife adopted a month-old infant chimpanzee called Vicki, and brought her up as their "daughter". Vicki was reared as a child for six years until her death, and subjected to an intensive programme of language training. Despite this, she learned only four words--papa, mama, cup and up--which she whispered rather than spoke. It was concluded that apes were incapable of acquiring human language, if only because they lack the necessary vocal apparatus to pronounce words effectively.

In spite of these early negative results, talking apes hit the headlines again in the 1970s when Allen and Beatrice Gardner at the University of Nevada began teaching a young chimpanzee called Washoe American Sign Language (a system of hand signals devised for the deaf). After four years of intensive training, the Gardners claimed that Washoe had acquired a vocabulary of 132 different signs which she could combine together into apparently meaningful "sentences". Owing to the particularly close relationship which existed between Washoe and her trainers, however, it was difficult to rule out the so-called "Clever Hans effect".