Automata, Receptacles, and Selves

After rejecting Carruthers' conflation of levels of consciousness as implausible and conceptually muddled, and Carruthers' claim that nonhumans are automata as undermined by evolutionary and ethological considerations, we develop a general criticism of contemporary philosophical approaches which, though recognizing nonhuman consciousness, still see animals as mere receptacles of experiences. This is, we argue, due to the fact that, while in the case of humans we grant a self something that has not only a descriptive but also a prescriptive side, requiring at least non-interference in the case of nonhumans we focus only on the descriptive aspects. Consequently, we treat humans as equals whatever their capacities, but we order nonhumans in a hierarchy based on their cognitive level. We conclude that such double standards are not only inconsistent but also self-serving. 1. Carruthers On Awareness Central to Carruthers' argument in the target article (Carruthers, 1998) is a distinction "between two different sorts of subjectivity -between worldly-subjectivity and mentalstate-subjectivity" (section 3). First-order representational (FOR) theories of consciousness may be able to account for the first sort of subjectivity, but since they cannot deal adequately with the latter they must be rejected in favor of higher-order theories. Mental-state subjectivity is the kind that really matters, it is the sort of 'subjectivity of experience' that is the real 'hard problem' of consciousness. This distinction is defective in at least two ways: it obscures and misplaces the actual 'hard problems' -epistemological, metaphysical, and moral -of consciousness, and it conflates a number of distinctions and differences in cognitive functioning which do not in fact coincide. Sigmund is having a kidney stone attack. He is not aware that this is what is going on, but he is very much aware of the pain and pressure in his abdomen. He makes his way to the hospital emergency room. The examining physician asks him where he hurts and Sigmund locates the pain as well as he can. Then the physician asks him to describe the pain as throbbing, piercing, etc. Only at this point does Sigmund attend to the pain as a mental state. His pain certainly had phenomenal properties, but it was not those properties which concerned Sigmund, but rather the highly aversive state of his body. Now that the question has been asked Sigmund attends to the pain qua pain and tries to describe its phenomenology. Perhaps, had Sigmund been of a more reflective bent, he might have thought about his mental state, his experiences, earlier. But he didn't. He was absorbed by the pain. According to Carruthers, the important 'subjectivity of experience' is not the sort of (mere) worldly/bodily-subjectivity of Sigmund writhing in the car on the way to the hospital, but the mental-state-subjectivity elicited for the first time in this episode by the physician's question. (Perhaps, if only she hadn't asked, his suffering wouldn't really be suffering.) But this is just wrong. The 'hard problem' of consciousness isn't this second-order reflective experience of experience. Sigmund's experience of his experience looks a lot like his coming to have beliefs about his mental states. We do indeed, many of us, often come to have such introspective beliefs. As theorists we can say lots of things about beliefs, in terms of dispositions to natural language utterance or other behavior, or tokenings of mentalese, or something, and beliefs about our own states can be discussed in the same terms. But the sheer sensuousness of experience, the painfulness of the pain, the smell and taste and color of the orange, reduce most theorists to inarticulate ostension. ("You know, the way it seems.") This is really the hard problem, about which most of us are almost completely clueless. This is why it is so tempting to give up and declare consciousness illusory or unstudiable or mysterious. Carruthers rightly rejects 'mysterian' claims about consciousness. But his own theory escapes mystery by dismissing our most intense conscious experiences as illusory. Carruthers seems to be claiming that since bats (let us assume) lack the sort of complexity needed to reflect upon their experience, since they can't think about what it is to be a bat, there is no such thing as the subjective character of chiropteric experience, nothing it is like to be a bat. But this is as silly as concluding from the fact that bats are (presumably) ignorant of the fact that they have livers that there are no such things as bat livers. Moreover, Carruthers' distinction is much too simple, for there are many degrees of complexity of worldly/bodily-subjectivity and of mental-state-subjectivity, and that distinction just does not correspond with lacking/having a theory of mind, or with lacking/having a notion of self. By speaking of 'seeming or appearance' Carruthers blurs the distinctions among a range of states, including, inter alia, the manifestation of objects in one's experience, itself ranging from the simplest awareness to the most stirring epiphany, and the consciousness of semblance and of a (possibly deceptive) appearance distinguishable from reality. Some of these states do indeed presuppose a very complex subject. Carruthers' equivocation obliterates the boundaries between the simpler and the more complicated states. The end result is to render simple consciousness conceptually homeless. But Carruthers' conflation of levels of consciousness may well point in the opposite direction. Even though consciousness does not logically entail self-awareness, "it may be ... that what we know about evolutionary pressures (as well as animal behavior and physiology) suggests that actually existing conscious animals are probably self-aware" (DeGrazia, 1996, p.175). For self-awareness is not an all-or-nothing capacity, but comes in degrees. It is curious that Carruthers speculates on the evolutionary advantages of a 'theory-of-mind' faculty and completely forgets the evolutionary advantages of the more basic level of, e.g., bodily self-awareness. We know of no vertebrate that fails to distinguish its own body parts from objects in its environment. Finally, given his all-or-nothing view of consciousness, which excludes any gradual steps, it is hard to imagine how in the first place beings who were fully unconscious could abruptly evolve "a capacity for HOTs [Higher Order Thoughts] because of the role such thoughts play in predicting and explaining, and hence in manipulating and directing, the behaviors of others" and suddenly start to think and reason about the beliefs, desires, intentions, and experiences of others. Surprisingly, Carruthers' distinction between worldly/ bodily-subjectivity and mentalstate-subjectivity turns out to have little or no moral significance even for him. In note 14 (1998) he writes: "My present view is that it is first-order (non-phenomenal) disappointments and frustrations of desire which are the most basic objects of sympathy and (possible) moral concern" (p. 18).<1> So nonhumans can be disappointed, can have their desires frustrated, and we can appropriately respond with sympathy, but these disappointments and frustrations are non-phenomenal. This is just incoherent. A stone rolling down a slope that pitches up against a tree is not frustrated. A dog running down a slope that is blocked by a fence may be frustrated. Disappointments and frustrations are possible only for the sentient. Disappointments are conscious states. 'Non-phenomenal disappointment' is muddle or (at best) metaphor. 2. Nonhuman Consciousness In his Treatise of Human Nature, David Hume claims that no truth appears to him more evident, than that nonhuman animals ('beasts') are endowed with thought and reason as well as human beings, and adds that the arguments are in this case so obvious, that they "never escape the most stupid and ignorant" (Hume, 1739/1888). More recently, American philosopher John Searle, speaking of consciousness and epistemology, wrote in a similar vein: I do not infer that my dog is conscious, any more than, when I came into this room, I inferred that the people present are conscious. I simply respond to them as is appropriate to respond to conscious beings. I just treat them as conscious beings and that is that. If somebody says, 'Yes, but aren't you ignoring the possibility that other people might be unconscious zombies, and the dog might be, as Descartes thought, a cleverly constructed machine, and that the chairs and tables might, for all you know, be conscious? Aren't you simply ignoring these possibilities?', the answer is: Yes. I am simply ignoring all of these possibilities. They are out of the question. I do not take any of them seriously. (Searle, 1998, pp. 4950). To this, one could add a Strawsonian point. Our ascription of conscious states to ourselves conceptually requires, as Strawson argued (1959), that those same states be ascribable to others. We learn what 'hurt', and 'happy' mean, and who we are, in interaction with others who treat us as beings that can be hurt or happy, as centers of experience, and who themselves are centers of experience that can be hurt or happy. Experimental work has confirmed Strawson's view. Developmental psychology shows that there is a kind of symmetry in the information we have about ourselves and others: as our 'mental' concepts develop or are triggered, they automatically apply both to own internal states and to the behavior of others (Gopnik and Meltzoff, 1994). We become selves as we come to recognize selves. And our partners in this are, often, not just Mommy, but Kitty. But Carruthers is not only theoretically unsatisfying. He is also empirically uninformed. Many of the "concepts which it would be implausible to ascribe to most species of animal" (1998, section 6) are in fact reasonably ascribed to animals of many species. Pigeons can grasp concepts, as it has been repeatedly demonstrated in Skinner box experiments. They can learn to distinguish, e.g., scenes with or without a body of water, or pictures containing a particular person from others with no people or different individuals (Griffin, 1992). Moreover, starving and thirsty pigeons when pecking keys for food use brief pecks with high force, and when pecking for water employ gentler pecks with prolonged contact between beak and key. It seems quite reasonable to infer that they are thinking about eating or drinking (Griffin, 1992). Unfortunately, experiments designed to verify the presence of these cognitive abilities pay no attention to what the foodor water-craving attitude may mean phenomenally for the pigeons. The experimenters demonstrate the conceptual achievement but ignore its phenomenal import. Carruthers ignores the demonstration. Consider, next, the capacity to entertain the thought, [that is dangerous] about a particular perceptually-presented animal. It is difficult to understand how, without it, it would be possible to conceive of most prey-predator relationships, where the dangerous/nondangerous distinction is obviously more vital than when theorizing about consciousness. When minnows encounter pike, the minnows tend to come together into a compact school, and then a series of individual inspections begins, during which the minnows seek to find out how likely the predator is to attack (Griffin, 1992). Gazelles, like minnows, seem to evaluate the probability that a predator will attack. Even when lions are plainly visible, Thompson's Gazelles do not automatically flee; they go on grazing, but the members of the group look around at different times in what is called 'predator vigilance', and as soon as a gazelle sees something unusual or suspicious, he becomes alert and gives an alarm snort, or 'quiff' (Griffin, 1992). As for the "standard story from the primatology and 'theory-of-mind' literatures" (Carruthers 1998, p. 217) and predicting and explaining the behaviors of others, there is an entire body of literature about the nonhuman apes.<2> After reviewing the debate, Gomez concludes that apes are persons who do not describe themselves as persons, but "may act and feel as persons in the most essential sense of the word, which I take to be the ability to recognize others and themselves as individual subjects capable of feeling and behaving intersubjectively" (Gomez, 1998, p. 61). One of the best indicators of a 'theory of mind' is tactical deception, which is widely present in the great apes. In discussing it, Savage-Rumbaugh even mentions episodes suggesting an attribution of a 'theory of mind' to others. Consider: at Gombe, an adult male was alone in a feeding area when a box was opened electronically, revealing the presence of bananas. A second chimp arrived, and the first one quickly closed the box and ambled off nonchalantly. He then waited until the intruder departed and then quickly opened the box to retrieve the bananas. But the other chimp had simply hidden, and appeared once again, thus deceiving the deceiver (Savage-Rumbaugh, 1994). In general, social animals must constantly deal with relational, and therefore predictive and manipulatory, problems. With a clear reference to the role of the mirror-test in singling out self-awareness, Frans de Waal so describes the 'social mirror': A macaque or baboon can hardly function without knowing the social position of each group mate, the kinship network, which individuals are likely to side with each other in a fight, the possible reactions of others to particular actions, and so on...Understanding one's surrounding equals understanding oneself. (de Waal, 1996, pp. 68-9). So, many nonhumans are 'natural psychologists', and not only for their own species. For example, rhesus mothers tend to hold their infants together with the infants of females who outrank them, in the attempt to establish connections with higher-ups in the hierarchy (de Waal, 1996). Inter-species cooperation is widespread. Along the shores of Australia and West Africa, native people net-fishing for mullet developed a mutually beneficial relationship with bottlenose dolphins. They set their nets and beat on the water's surface, and dolphins swim toward shore in response to this sound and herd schools of mullet toward the nets. Some mullet are caught by the dolphins and others are entrapped in the nets: thus, both the fishermen and the dolphins benefit (Herzing & White, 1998). Finally, think of the perhaps elementary but genuine theory of mind demonstrated by any kitten or puppy inviting another animal (human or nonhuman) to play. Such simple but important facts as cats keeping their claws retracted when batting at a human hand show that many animals effectively categorize the things with which they deal as sentient or not (without, of course, having an articulable concept of sentience). Young children obviously make and understand ascriptions of mental states and of the subjective aspects of reality ('Yummy') much before the age at which they (according to Carruthers) have the concept of seeming. 3. Consciousness and Selfhood Unlike Carruthers, most contemporary moral philosophers take animal consciousness for granted. Moreover, they think that consciousness matters, and that whenever consciousness be it human or nonhuman is present duties of beneficence, or golden rule arguments, or appropriate consequentialist weightings, come into play. But although they avoid a position as eccentric as that of Carruthers, most modern moral thinkers still contrive somehow to discount nonhuman subjectivity and its significance. Animal consciousness counts, surely, but for much less than human consciousness, even human consciousness of a type and level surely attained by many nonhumans. How can this be? As the human/nonhuman dichotomy itself shows, the ground tends to be species membership. While the appeal to species membership can be straightforward, we will not deal with this approach here. Many authors have argued conclusively that species membership is just as morally arbitrary a basis for assigning moral status as race or sex. We are concerned not with overt speciesism, but with covert forms tacitly embodied in various approaches to the moral status of nonhuman animals. There is an implicit recourse to double standards when treating similar aspects in humans and in nonhumans. As far as consciousness is concerned, this causes a difference in focus. When dealing with human consciousness, some things are deemed to be relevant. When, on the other hand, it comes to nonhuman consciousness, other aspects come to the forefront. Recently, it has been argued that taking nonhumans seriously means taking their minds seriously.<3> It is difficult not to agree with this emphasis, especially in view of a position such as Carruthers'. However, even when nonhuman consciousness is recognized, it appears in a somewhat distorted way. We would therefore say that taking nonhumans seriously also means taking their subjectivity seriously. As Thomas Nagel has stressed in a famous essay, what makes the mind important is subjectivity what it feels like to have a certain point of view on the world. Nagel considers two peculiarities in subjectivity: the peculiarity of being a specific being's subjectivity, with all the attendant questions of how particular senses and physiological characteristics can affect one's point of view; and the peculiarity of subjectivity as a fact which resists objectivization and, therefore, standard scientific examination. Both of these aspects, as we have argued, are fully in play in the worldly/bodily-subjectivity Carruthers wants to discount. The former aspect, being a specific being's subjectivity, is, so to speak, the descriptive aspect of the self. To be myself means to have particular characteristics and capacities, a specific point of view on the world, and also a particular psychological history (or narrative). Those of us with both reflective mental-state-subjectivity and linguistic skill can describe, more or less adequately, how things seem to us and how things are going for us. Other similarly equipped can understand how we are similar and how we are different. Many other animals, human and nonhuman, lack our descriptive powers, and some may lack mental-state-subjectivity or lack the richness of such subjectivity that we enjoy. But for each sentient being there is a way things seem to her and a way things are going for her. Our understanding of the subjectivity of a nonhuman animal is generally even more tentative and more risky than our understanding of the subjectivity of another human. But unless blinded by rage or dogma or the grip of a theory, we know that they have lives, lives different from ours. Different lives, but lives entitled to equal consideration. More and more, contemporary egalitarianism discounts differences as 'irrelevancies'. That an individual is female or male, old or young, black or white, intelligent or stupid, is not deemed to matter ethically. In particular, human rights theorists stress that differences of cognitive capacities should not play a role in determining the moral status of human beings. Both everyday morality and international declarations warrant even greater basic moral protection to those members of our species who, because of different sorts of limitations or impairments are, so to speak, descriptively deficient. What is the rationale behind this? Usually, what the authors in question claim is that equal interests are owed equal consideration, whatever the other (descriptive) characteristics of the individual. And this, because each individual values her interests. Interests bring in an evaluative aspect of the self which adds to the descriptive one. But the root goes deeper than interests. Why would the self see the satisfaction of its interests as good and value it, if it did not value itself? Others' satisfactions may be considered too, but nothing has the immediate, indisputable value that the satisfaction of our interests has for us insofar as those interests pertain to us. As it has been noticed, even if we prove to be good for nothing, we believe, nevertheless, that we are still entitled to life, to liberty, to the pursuit of happiness... Call this the phenomenological proof for the existence of intrinsic value. The question How do we know that intrinsic value exists? is similar to the question How do we know that consciousness exists? We experience both consciousness and intrinsic value introspectively and irrefutably. (Callicott, 1995). The self has thus an evaluative aspect, which is what underlies intra-human egalitarianism. We can, however, go one step further. Something more can be said of the non-descriptive side of the self something that might explain what it is that really resists objectification, and why a protective fence around it is morally called for. Richard Hare advances a suggestion. The 'I', he says, is not wholly a descriptive word but in part prescriptive, and in identifying ourselves with another individual, we identify with her prescriptions (Hare, 1981). In fact, Hare does not say 'another individual'. What he says is 'another person' and he actually seems to have in mind a specific, intellectualized kind of 'I'. Curiously, however, the first example of a prescription he gives is the preference not to be made to suffer (by the dentist). A second relevant passage in the same book is worth quoting fully: [T]he short answer to the problem about the meaning of statements about other people's states of mind is that terms like 'I' and 'you' have no descriptive content in the strict sense; that is to say, if you and I just changed places, the world would be no different in its universal properties. So, the meaning of the predicate in 'You are in pain' is exactly the same as in 'I am in pain'; there is no difficulty in knowing what it means, and this sameness of meaning is indeed the reason why it is possible to teach the use of the predicate. (Hare, 1981, p. 123). If we go back to Nagel's distinction, we may say that, if to be myself is something that can be described to others, and they can understand that we are different, to be myself is something that cannot be described and yet others can understand that, in this, we are the same.<4> As in the case of the predicate 'in pain', there is no difficulty in knowing what it means to be a self, and we could say that "this sameness of meaning is indeed the reason why it is possible to teach the use" of the subject. Alan Gewirth (1978), who developed a reconstruction of morality quite different from Hare's, stemming as it does not from the utilitarian but from the rights strand of moral theory, offers the best explanation of the partial prescriptivity of the self. The agent, he claims, pursues goals through her action. If she pursues them, she must see them as good (evaluation). But if she sees her goals as good, she must implicitly or explicitly command that others should not interfere with her attempt to reach those goals (prescription). Thus, the self starts with a fundamental negative prescription: do not interfere with my action, do not prevent me from pursuing what I deem good either by depriving me of freedom, or by undermining my abilities through subtractions from my welfare, or by killing me. It might be said that, once again, what is involved is a specific, intellectualized kind of self, which is able explicitly to state its prescriptions. But there is no cogent reason for claiming this. Merely by reflecting on what it is to be a self, the intellectualized self can realize that the same profound, radical prescriptive aspect is present in other conscious beings, regardless of whether it is articulated or not. And in fact, human rights doctrines clearly show how reflective beings, just because they are reflective, can see that other beings have simpler selves which are equally important to them. This is, we believe, the deep reason why, in contemporary egalitarianism, descriptive elements including biological, social and cognitive aspects are seen as irrelevancies. What is recognized is that they should morally give way to the prescriptive element embodied in the fundamental request for non-interference (more mundanely, non-exploitation).