LaFollette (2011) on Animal Experimentation

 Hugh LaFollette, “Animal Experimentation in Biomedical Research.” Chapter 29, pages 796-825, in The Oxford Handbook of Animal Ethics, Tom L. Beauchamp and R. G. Frey, editors, Oxford University Press, 2011. 

• The Common View: animals matter, but we (humans) can use them for our purposes when the benefit to us is significant.

• Lenient Views (animals barely matter) and Demanding Views (animals matter a lot) also exist – and for these, the potential benefits to humans are essentially irrelevant: the Lenient view allows animal exploitation for trivial human benefits, and the Demanding View forbids animal experimentation even for large human benefits.

• The Historical View was an extreme form of the Lenient View, one in which animals were essentially things, like rocks.

• Almost everyone now believes that it is wrong to cause an animal unnecessary pain, and that animal lives are valuable to them.

• For many people, these various starting points lead to a consideration of balancing costs imposed on animals with benefits to humans -- but some animal advocates suggest that such a balancing is unethical.

• The Argument Supporting Experimenting on Animals: (1) the research works, we can generalize from the anymals to humans; (2) animal experiments have been essential historically for medical advances; and (3) scientific methodology (controlled experiments and the need for “intact systems,” not just a hunk of cells) bolsters claims (1) and (2).

• There's an information bias: we hear about the seeming successes but not about the failures, the blind alleys we have been led down by inappropriate animal models.

• The similarity problem is the question of whether the animals condition is sufficiently like the human condition being examined. Given similarity, the inference problem is whether we can draw reliable conclusions about humans from the animal experiments. [These problems to some extent parallel internal and external validity concerns.] Much animal experimentation fails similarity or inference.

• When animal experimentation is presented as a choice something like choosing between the life or your baby and the life of your dog, the dog will not win. But the actual choice we are faced with is not between your baby and your dog – it is the institution of animal experimentation that is in the balance. 

• The harm we do to animals in experiments is present and real; the claimed benefits are down the road and uncertain. Maybe researchers have special obligations to those future humans down the road – but they also have special obligations to the captive animals in their labs.

• Would we use the “balancing test” in considering whether to experiment on nonconsenting humans? Isn’t this inconsistency a devastating counter to the “pro” case for animal experimentation?

• Speaking of inconsistency, there is the basic one of needing anymals to be similar to humans to scientifically justify experimenting on anymals, but not similar to humans to morally justify experimenting on anymals. [Peter Singer, for one, noted this paradox in 1973.]

• The fact that animal experiments might have value to humans is not in itself dispositive. More telling is the question of whether those benefits are procurable in ways that do not involve animal experiments.