Bell, Magendie, and the proposals to restrict the use of animals in neurobehavioral research.

The discovery by Magendie of the sensory and motor functions of the dorsal and ventral roots of the spinal nerves provides an illuminating case study of the scientific and ethical considerations that arise when one contemplates restricting neurobehavioral research on animals because of the suffering it causes them. Such restrictions reduce the number of worthless experiments only at the cost of reducing the number of worthwhile experiments—experiments that shed new light on the sources of behavior and provide the knowledge that enables us to alleviate human suffering. Therefore, one should urge the abandonment of animal research in part or in toto only if one believes that the moral value attached to the avoidance of animal suffering is greater than the moral value attached to the enrichment of human understanding and the alleviation of human suffering. A bill called the "Research Modernization Act" is now before Congress, where it is picking up influential support. The bill would ban most surgical experiments using live animals, on the theory that the same knowledge may usually be gained by computer simulations, experiments on bacteria, and so on (see Broad, 1980). The bill would establish a review committee that would allow at most one experiment of a given type to be done on live animals. The proponents of this legislation claim that the law is a moral imperative and that it would not cause serious harm to research in the life sciences. I wish to argue that this bill would devastate behavioral neurobiology and that it is an affront to moral sensibility. Behavioral neurobiology tries to establish the manner in which the nervous system mediates behavioral phenomena. It does so by studying the behavioral consequences of one or more of the following procedures: (a) destruction of a part of the nervous system, (b) stimulation of a part, and (c) administration of drugs that alter neural functioning. These three techniques are as old as the discipline. A recent addition is (d) the recording of Vol. 36, No. 4, 357-360 Copyright 1981 by the American Psychological Association, Inc. 0003-066X/81/360<I-0357$00.75 electrical activity. All four procedures cause the animal at least some temporary distress. In the past they have frequently caused intense pain, and they occasionally do so now. Also, they often impair the animal's proper functioning, sometimes transiently, sometimes permanently. From the beginning, this enterprise has provoked moral censure, to which the experimentalists have often reacted defensively. The terms of this debate have changed hardly at all in 200 years. Consider the following passage, written shortly after 1800 (Le Gallois, 1813,'pp. 19-21): Before I close this introduction, I wish in some degree to exculpate the physiologists who make experiments upon living animals, from the reproaches of cruelty, so frequently uttered against them. I do not pretend wholly to just i fy them, I would only remark, that the most part of those who utter these reproaches may be deserving of the same. For example, do they not go, or have they never gone a hunting? How can the sportsman, who for his own pleasure mutilates so many animals, and often in so cruel a manner, be more humane then the physiologist who is forced to make them perish for his instruction? Whether the rights we assume over those animals be lawful or not, it is certain that few people scruple to destroy, in a variety of ways, such of those animals as cause them the least inconvenience, though ever so trifling; and that we only feed the most part of those that surround us, to sacrifice them to our wants. I can scarcely comprehend that we should be wrong in killing them for our instruction, when we think we are right in destroying them for our food. I own that it would be barbarous to make animals suffer in vain, if the object of the experiment could be obtained without it. But it is impossible. Experiments upon living animals are one of the greatest lights of physiology. The This is a slightly revised version of a paper presented to the Symposium on the Ethics of Animal Experimentation at the meeting of the American Psychological Association, Toronto, August-September 1978. Requests for reprints should be sent to C. R. Gallistel, Department of Psychology, University of Pennsylvania, 3813-15 Walnut Street T3, Philadelphia, Pennsylvania 19104. AMERICAN PSYCHOLOGIST • APRIL 1981 • 357 difference between the dead and the living animal is infinite. If the ablest mechanician is unable to discover all the effect of a machine after having seen it work, how could the most learned anatomist devise, by the study only of the organs, the effect of a machine as prodigiously complicated as the body of an animal. To find out its secrets, it is not enough to observe the simultaneous exercise of all the functions in the animal, while in health; it is above all important to study the effect of the derangement, or the cessation of such or such a function. It is in determining by this analysis what the function of such or such an organ is, as well as its relation with the other functions, that the art of experiments upon living animals consists. But to be able to do it with some degree of precision, it is indispensably necessary to multiply the victims, on account of the variety of circumstances and accidents which may render their result uncertain or inconclusive. I should be tempted to say of physiological experiments, what has been said of charities: perdenda sunt multa; ut semel ponas bene. SENECA. [Translation: Many are a waste, that one may come out well.) The passage just quoted seems to me to contain most of the basic facts and positions in the debate between behavioral neurobiologists and antivivisectionists. Let me first summarize what I take to be matters of fact: 1. Experimental surgery causes pain and distress to animals. 2. Researchers are well aware of this pain. Since the discovery of ether in 1847, they have used anesthetics to reduce or prevent the pain, wherever such reduction or prevention does not affect the conclusions that can be drawn from the experiment. 3. There is no way to establish the relation between the nervous system and behavior without some experimental surgery. 4. Most experiments conducted by behavioral neurobiologists, like scientific experiments in general, may be seen in retrospect to have been a waste of time, in the sense that they did not prove anything or yield any new insight. 5. There is no way of discriminating in advance the waste-of-time experiments from the illuminating ones with anything approaching certainty. Such judgments are necessarily made under conditions of high uncertainty. As shown by the theory of signal detection, a necessary consequence of this uncertainty is that any attempt to reduce the number of neurobehavioral experiments by prior evaluation of their possible significance will necessarily give rise to many "false negatives," without eliminating "false positives." That is, prior restraints on neurobehavioral experiments will lead to rejection of experiments whose results would in fact have been important and allowance of experiments whose results will prove unimportant. This will be true no matter how stringent and cumbersome the a priori evaluation (see Figure 1). These five statements must be taken as facts. Any attempt to advance a proor antivivisectionist position by denying one or another of these statements evades the ethical question by denying the very circumstances that give it force. The force of these circumstances can best be appreciated by the study of specific historical cases. One case that should be analyzed at length by anyone contemplating restricting neurobehavioral experiments is the discovery that the dorsal and ventral roots of the spinal cord are sensory and motor, respectively. In 1822 Francois Magendie discovered that in young puppies the dorsal and ventral roots of the peripheral nerves come together outside the spinal column, so that they can be separately severed with relative ease. Magendie had been wondering for some time what would be the effect of cutting one or another root on the behavior of the limb or body segment served by the nerve. In the other animals he was familiar with, the roots fused before exiting from the spine. They could only be cut individually after breaking open the spine, which, in the days before anesthesia, was all but impossible to do without damaging the spinal cord. Soon after discovering the favorable anatomical disposition of the roots in young puppies, Magendie began exposing the spines of 6-8-week-old puppies and cutting either the dorsal or the ventral roots of one or more nerves. After several such experiments he was able to publish his famous three-page communication in which he concluded that the dorsal roots carried sensory signals while the ventral roots carried motor signals (Magendie, 1822). Magendie's experiments place the ethical problems posed by neurobehavioral research in sharp relief for the following reasons: (a) The results were of the utmost importance, (b) The animals used were puppies and the pain of the necessary surgical procedure was both intense and unalleviated by anesthetics, whose discovery lay 25 years in the future, (c) Other very similar experiments had been conducted by some of the leading neuroscientists of the day—most notably the English anatomist Charles Bell—without yielding the decisive, all-important insight, (d) The experiments, because they rapidly became well-known and because they were sometimes performed in public, incurred widespread moral censure and helped fuel the antivivisection movement in 19th century England. Let me elaborate on these points. First, as regards the significance of the results, I can do no 358 « APRIL 1981 • AMERICAN PSYCHOLOGIST -Q '5 E 'S Rejected proposals Proposals that would ' prove worthless ", .Proposals that would