AR Philosophy >
Morality of AR >
Speciesism - Index
It's a dog's life
04 June 2005
Ian Duncan is professor of applied ethology and chair of animal welfare at the department of animal and poultry science, University of Guelph, Ontario, Canada
A PAINTING tells a thousand tales. Pieter Bruegel's hunting dogs, for example, look as dejected as their owners in his painting Hunters in the Snow. But it tells us something really important: that from the Renaissance on, art and literature seem to have been way ahead of science in granting animals - or at any rate, mammals - some acceptance of sentience. The secular intellectual world of Leonardo da Vinci, Michel de Montaigne, Erasmus, Shakespeare and Francis Bacon took animal sentience for granted.
Philosophers were the big nay-sayers. There is a clear line of argument for non-sentience running from Aristotle to St Thomas Aquinas and René Descartes to Immanuel Kant. They tended to draw a thick dividing line between non-human animals and humans, allowing the latter to use the former for whatever purpose they liked. Descartes is usually singled out for special blame for introducing the idea of animals as "automata".
The Enlightenment brought fresh philosophical challenges over the question of animal sentience. Scottish philosopher David Hume, in particular, wrote: "Is it not experience, which renders a dog apprehensive of pain, when you menace him or lift up the whip to beat him?" And in the wake of the studies of Charles Darwin less than a century later, feelings came to be viewed as "adaptations" to pressures of natural selection.
About 120 years ago, then, scientists and non-scientists alike began to accept some degree of animal sentience. But through much of the 20th century this acceptance seems to have gone into reverse, largely because of behaviourism, which eschewed any study of animal feelings.
One of the leading exponents of this theory, B. F. Skinner, laid it out clearly as late as 1975: "We seem to have a kind of inside information about our behaviour, we have feelings about it. And what a diversion they have proved to be. Feelings have proved to be one of the most fascinating attractions along the path of dalliance." In other words, studying feelings was meaningless. Skinner and others with similar viewpoints exerted a powerful influence on the way that European ethologists were thinking at that time.
The pattern was eventually broken by Donald Griffin with his key paper presented at the International Ethology Conference in Parma, Italy, in 1975, and followed by his book, The Question of Animal Awareness (Rockefeller, 1976), in which he made a powerful case for animal feelings.
But if animals had feelings, where did that leave animal welfare? Around the early 1980s, Marian Stamp Dawkins and I concluded that it was impossible to give animal welfare a precise scientific definition. The best we could manage was a broad, working description, much of which centred on feelings.
There is now a gradual acceptance that feelings govern welfare, and therefore feelings should be measured when assessing welfare. So animal welfare science has been largely concerned with developing methods by which we can "ask" animals what they feel about the conditions under which we keep them and the procedures we subject them to.
The biggest obvious problem with feelings is that they are subjective, and therefore not available for direct investigation. But we don't need to know exactly what animals are feeling: an indication of how positive or negative a feeling is is a good start. We can investigate this by using indirect methods, such as by looking at preference testing, motivational testing and understanding animal communication.
Preference testing was pioneered by Stamp Dawkins and Barry Hughes, both of whom were working with poultry. In their tests animals were given a choice over certain aspects of their environment, and it was assumed by the researchers that they would choose in the best interests of their own welfare. There have now been many studies of preference strength, and they tell us a lot about fear, for example, by seeing how far an animal will work to avoid it.
In one experiment a hen had to learn to avoid a scary inflatable balloon in its box. It was shown a warning light 20 seconds before the balloon was inflated, so the hen would have time to move. Some of the hens took a long time to learn how to avoid the balloon - but they all got there in the end. Interestingly, some of the hens gave alarm calls during the experiment, but the number of alarm calls decreased as the birds learned what was happening and became less afraid.
These techniques are now coming of age - which is good news for scientists and animals alike. But there are three areas where we will have to concentrate our future efforts. The first is understanding the role that pleasure plays in welfare; the second is deciding just where on the phylogenetic scale sentience begins; and the third is working out where in development sentience begins. Watch this space!