Recent evidence suggests that the ability of animals to understand the subjective states of other animals depends upon the type of animal in question. Gordon Gallup, Jr. posits that the cognitive mechanism which provides an animal with the ability to understand the subjective state of other animals is the concept of self because: "knowledge of mental states in others presupposes knowledge of mental states in oneself." (2). This "knowledge of self", which Gallup also calls empathy, then allows animals to make inferences about the knowledge of others. Research by Gallup and Povinelli indicates that the only animals with a self-concept are chimpanzees, orangutans, and humans based on an experimental method called the mirror test. During this test, animals are placed in front of a mirror and their reactions are recorded. After a significant period of time in front of the mirror, the animal begins making faces at itself. In order to test self-concept, Gallup anaesthetized each animal (after repeated presentations in front of the mirror) Experimenters then put marks on each animals eyebrows and ear, and put the animal back in front of the mirror to see if it noticed the marks. The assumption is that animals who notice the marks will have a concept of self because they attribute the marks to themselves, not to some other animal in the mirror. Animals who do not notice the marks will not have a concept of self, because, as Gallup speculates: "Maybe the reason most species cannot process mirrored information about themselves stems from an inability to conceive of themselves. Correctly inferring the identity of the reflection presupposes an identity on the part of the organism making that inference."(2)
Yet does the existence of self concept imply that it is possible for animals to understand the subjective state of other animals? Povinelli does not deny the possibility that the reason why these chimpanzees perform on the mirror test is because they have a concept of self. What he does contest, however, is whether this self concept can then be used to support the argument that these animals have what Gallup terms empathy and what Povinelli terms "psychological understanding."(2) Povinelli uses the simple act of seeing in order to illustrate his point. His laboratory successfully showed that chimpanzees "can reason about the visual perspective of others."(2) This result was found in experiments in which humans made eye contact with chimpanzees and then looked over the chimpanzees shoulder, causing the chimpanzee to look over his shoulder to see if someone else was there. Povinelli then created what he terms a "seeing/not seeing" paradigm in which there are two experimenters which adopt different postures: one experimenter is facing towards the animal and the other experimenter has their back to the animal, or places their hands over their eyes. It was hypothesized that if the chimpanzee has a concept of self, he would gesture only to the experimenter who could see him. However, findings indicated that the ability to see the experimenter made a difference. Povinelli interprets the results of the mirror test to indicate a concept of self in chimpanzees in which the animal recognizes that their behavior and the behavior of the animal in the mirror are the same. This recognition, however, can not necessarily then be attributed to the ability of the animal to infer psychological states about itself and others. Povinelli's explains the reason why certain primates might have this ability (and others do not ) through evolutionary theory. According to this theory, certain primates had to develop a higher level recognition process as a result of their life style of swinging through trees, which required higher levels of recognition and navigation of their behavior and movements. Gorillas, for example, are not included in the higher recognition category supposedly because they adapted to life on land and no longer had a use for higher level recognition. Be careful with using higher level recognition and self concept interchangeably.
Others do not share the view that, even with a knowledge of self, it is possible to understand the subjective state of another animal, human or otherwise. Thomas Nagel would argue that no, it is not possible for humans or even other chimpanzees to understand the subjective state of another chimpanzee because every state is subjective, and there is no current valid measure of the objective state. In other words, before one considers the problem of subjective and objective states, one must have a way to objectify a subjective state. Nagel illustrates the fundamental problem that phenomenological theory addresses in his article "What is it like to be a Bat" when he states: "But no matter how the form may vary, the fact that an organism has conscious experience at all means, basically, that there is something it is like to be that organism-something it is like for the organism. We may call this the subjective character of experience." (3). Both Nagel and Andrew Jordan (4).argue that we may be able to fabricate human cognition, but we will never be able to fabricate human experience. Therefore, Nagel proposes to find "an objective phenomenology not dependent on empathy or imagination."(3)
One method which might be considered an effort toward such objectification (from Nagel's point of view) is through the use of drug discrimination studies. This field of pharmacology dedicates itself to the ability to determine or discriminate subjective states of animals through a drug discrimination paradigm. The basis of this theory relies upon Pavlovian Conditioning and the concept of stimulus generalization(5). The classic example of Pavlovian Conditioning is a paradigm in which a dog is presented with food paired with a bell or tone. After repeated paired presentations of tone paired with food, the tone alone comes to illicit salivating. Stimulus generalization occurs when a higher and lower tone also produce higher and lower salivation levels respectively. Thus a different tone may illicit the same physiological response to a stimulus, which then causes similar behavior. Jaffe found that cocaine has similar subjective effects to amphetamine, thus it is difficult for subjects to discriminate between the two drugs and subjects will generalize the effects of cocaine to amphetamine or amphetamine to cocaine to the extent that the behavioral result is similar.( (6).) In addition, experimenters also postulate that the amphetamine-like subjective are the cause of the addictive quality of cocaine. Gauvin, et al (7). conducted a study in which different combinations of over the counter stimulants such as caffeine, ephedrine, phenylpropanol-amine, and saline were tested for generalization to (the subjective state of)cocaine. Results indicated that as long as the mixture combination included ephedrine, rats generalized the effects of the drug to cocaine. However, no single element of a mixture generalized to the effects of cocaine. The purpose of this study was to find a possible combination of over the counter stimulants which would produce the same subjective experience of cocaine but not have the addictive property of self administration, thus decreasing cocaine craving during withdrawal.
One potential problem with the ideology behind these studies, however, is that these experiments study the subjective experience of an altered state (the state produced by the drug) as opposed to the baseline subjective state of an individual. Comparative psychologists like Povenelli or Gallup might also consider these findings problematic because they maintain that in order to compare animal subjective states to human subjective states is necessary to have a concept of self. If these researchers continue to use rats, which are not part of the group established by the comparative psychologists as having a concept of self, then studies about the subjective experience of these rats are not generalizable to human behavior.
In conclusion, the debate about animal consciousness and the subjective state of animals depends upon the interpretation of the evidence. Clearly there are those who question whether certain types of evidence can and do accurately represent a subjective state. In addition, the question of how productive it might be to pose such a question and conduct research in this area also depends upon interpretation. Povenelli argues that research with animals that do not have what he considers a concept of self(i.e. every animal but humans) may not be the most productive way to go about this problem because animals do not have the same cognitive ability as humans and therefore any research conducted on non-human animals could not be generalized to human behavior. Nagel develops this point further when he argues that it is not possible for any animal to experience what it is like to be another animal. On the other hand, drug discrimination research provides useful information about the nature of altered subjective states (in animals and humans) and this information can be used to study abuse, addiction and withdrawal of drugs.
3) Nagel, Thomas. (1974) What is it like to be a bat? Philosophical Review, 83:435-450.
6) Jaffe, J. Drug Addiction and Drug Abuse. In: Gilman, A.G.; Goodman, L.S.; Rall, T.W.; Murad, F., eds. The Pharmacological Basis of Therapeutics . New York: Macmillan; 1985:532-581.
7) Gauvin, et al. (1993) The discriminative stimulus properties of legal, over-the-counter stimulants administered singly and in binary and ternary combinations. Psychopharmacology, 100:309-319.