For a long time in the history of Western philosophy it was assumed that nonhuman animals are incapable of rationality. Aristotle observes in his zoological writings that many animals appear to exhibit rational capacities, yet in his psychological and ethical writings he excludes nonhuman animals from community with humans on the grounds that animals are incapable of phronesis, the notion of practical wisdom that Aristotle considers crucial for active citizenship. The Stoic philosophers formalize Aristotle’s position in the psychological and political writings into a cosmic principle according to which animals were created expressly to satisfy human needs and lack the cognitive capacities necessary to occupy the cosmopolitan standpoint from which rational beings contemplate the divine logos alongside the gods.1 Saint Augustine follows this Stoic line of thinking and maintains that even though animals have sufficient subjective states of awareness to feel pain, this pain has no moral significance inasmuch as animals do not participate in koinoia or community with human beings. René Descartes refuses to acknowledge that animals are sentient and argues for the legitimacy of experimenting on animals, as well as killing and eating them, on the grounds that animals are mere machines with no inner states of awareness whatsoever. Kant returns to the Aristotelian-Stoic-Augustinian view according to which animals do have subjective states of awareness, but he argues that, because animals lack the capacity for rational choice, they are neither subjects nor objects of direct moral concern.
Already in antiquity there were proponents of the view that animals are rational or at least that animals have complex inner states of awareness in virtue of which they should be considered direct beneficiaries of moral concern. The most conspicuous animal advocates in ancient philosophy are Plutarch and Porphyry.2 Plutarch wrote three texts on animals in which he criticizes the Stoic position by offering a wide array of examples of animal thought and ingenuity and in which he deplores the human practice of meat eating. Porphyry argues in On Abstinence from Killing Animals that there is no need to kill and eat animals (indeed, that the practice interferes with the process of soul purification), that the practice of animal sacrifice is an offense against the gods (for whom animals are beloved), that we owe duties of justice to animals, and that animals exhibit at least imperfect rationality. But many of the anecdotes offered by these ancient thinkers bear the traces of an unmistakable anthropomorphism that ultimately detracts from the credibility of their arguments. One such anecdote is offered by Plutarch (1995 [ad 70]):
Cleanthes [a Stoic philosopher], even though he declared that animals are not endowed with reason, says that he witnessed the following spectacle: some ants came to a strange anthill carrying a dead ant. Other ants then emerged from the hill and seemed, as it were, to hold converse with the first party and then went back again. This happened two or three times until at last they brought up a grub to serve as the dead ant’s ransom, whereupon the first party picked up the grub, handed over the corpse, and departed.
(p. 369)
Another such anecdote concerns Chrysippus’s dog. Sextus Empiricus (1996 [ca. ad 190]), who, like Plutarch, is at pains to criticize the Stoic position concerning animals, offers the following account and interpretation of this anecdote:
According to Chrysippus, who was certainly no friend of non-rational animals, the dog even shares in the celebrated Dialectic. In fact, this author says that the dog uses repeated applications of the fifth-undemonstrated argument-schema, when, arriving at a juncture of three paths, after sniffing at the two down which the quarry did not go, he rushes off on the third one without stopping to sniff. For, says this ancient authority, the dog in effect reasons as follows: the animal either went this way or that way or the other; he did not go this way and he did not go that; therefore, he went the other.
(p. 98)
Lest we suppose that this tendency to anthropomorphize is a relic of antiquated prejudices that we wise postmoderns have long since outgrown, it is worth bearing in mind that this tendency manifests itself even in contemporary ethology. One illustrative example is the work of Donald Griffin, a prominent cognitive ethologist and the codiscoverer of echolocation in bats. In an effort to counter the influence of behavioral ethology, according to which animal behavior can be adequately understood without any reference to subjective inner states of awareness, Griffin maintains that a wide variety of animals engage in conceptual abstraction. He suggests, for example, that “it seems reasonable to suppose that when pigeons are working hard in Skinner boxes to solve these challenging problems [of making fine discriminations between different types of objects], they are thinking something like: ‘Pecking that thing gets me food’” (Griffin 1992:131). On Griffin’s view, even invertebrates such as honeybees “may think in terms of concepts” (p. 139).
Griffin’s motivation is to arrive at more edified conceptions of the subjective lives of animals than either the Western philosophical tradition or behavioral ethology allow. Griffin would have us believe that animals, all the way down to at least some invertebrates, engage in conscious thought that is conceptual and predicatively structured. To say that nonhuman animals are capable of conceptual abstraction can mean many things; indeed, a survey of the literature on concepts shows that there is nothing even remotely like a consensus on the question what a concept is. In my own work I follow the definition offered by Colin Allen and Marc Hauser (1991): “To have a concept of X where the specification of X is not exhausted by a perceptual characterization, it is not enough just to have the ability to discriminate X’s from non-X’s. One must have a representation of X that abstracts away from the perceptual features that enable one to identify X’s” (p. 227). On this view, conceptual ability is more than and something essentially different than discriminatory ability; it is the ability to form representational content that is separate from any particular perceptual experience to which that representational content pertains. The ability to distinguish, say, black from white does not require any conceptual ability. To recognize that black and white are both colors does require conceptual ability, and, in attributing conceptual ability to animals, Griffin is attributing to them precisely this kind of sophisticated mental operation. Moreover, to the extent that intentional states such as beliefs and desires are predicatively structured and employ concepts, as when I state or recognize that something is the case, Griffin, in attributing conceptual ability and thought to animals, is implicitly attributing intentional capacity to them. Thus Griffin would presumably say that the piping plover, in performing its broken wing display to lure predators away from its young, has a more or less specific set of beliefs: that the predator is endeavoring to catch the plover’s young, that the predator will be fooled by the broken wing display into thinking that the plover will be easier to catch than the plover’s young, etc. Intentionality in this sense involves a variety of sophisticated mental capacities, including the ability to recognize that one’s own mind is one mind among many minds, that other minds operate in essentially the same ways as my mind, that these operations involve phenomena such as beliefs and desires, etc. And to the extent that beliefs and desires are by their very nature predicatively structured—a belief is a belief that something is the case, and a desire is a desire that some state of affairs obtain—the attribution of intentionality to nonhuman animals is the attribution of all these abilities to them.3
To adopt such a view of many if not all nonhuman animals is to seek as far as possible to erode what was traditionally taken to be a difference in kind between humans and nonhuman animals and to represent that difference as nothing more than one in degree. Griffin and others who take this tack on behalf of animals appeal to physiological similarity and evolutionary continuity between humans and animals as a basis for concluding that animals, like humans, possess conceptual and in some cases predicative abilities. But even Griffin acknowledges the claim of the tradition, that the human possession of language constitutes an important difference between humans and animals (Griffin 1992:4). Thus he finds himself caught in a peculiar contradiction: He wants to attribute to many animals the capacity for conceptual thought and acts of predication and he wants to acknowledge that there is something distinctive about human language. But if the distinctiveness of human language does not consist in, or at least include, conceptual and predicative abilities, in what does it consist? It seems to me that thinkers such as Griffin, in their zeal to assert that animals have rich inner lives, go too far from one extreme to the other in attributing to animals capacities that animals simply do not seem to possess. Specifically, it seems implausible to attribute predicative intentionality and conceptual ability to any but perhaps the most sophisticated nonhuman animals, such as higher primates, dolphins, and some birds.
I say “perhaps” because the results of much of the observation that has been performed on animals that lie closest to the human-animal cognitive dividing line are highly ambiguous. Researchers such as Griffin are forced by the very nature of their vocation to reject Thomas Nagel’s (1974) challenge to cognitive ethology, a challenge according to which we can never know what it is like to be an animal of another species inasmuch as differences in perceptual apparatus make for fundamentally different sorts of subjective encounters with the world. Griffin (1992:237–238) argues that while we may not be able to arrive at absolutely precise accounts of the inner lives of other animal species, we can nonetheless learn a great deal about the mental lives of animals by analogy to our own experience. It is widely accepted in the community of cognitive ethologists that analogy is an indispensable tool in the endeavor to understand the subjective experience of animals. My colleague at Bucknell, the ethologist Douglas Candland (1993), has noted our strong tendency to conceive of the mental lives of animals in terms of our own intentionality, our capacity to form beliefs and desires: “When my dog barks at me, do I not attribute to the dog some purpose, a purpose reflected in my reflecting about the contents of the dog’s mind? In this way, the mind of my dog is inextricably bound to my mind, for I have no other way to create its mind other than by applying the categories and concepts of my own” (p. 369). But this by itself is no guarantee that what we infer by means of analogy is an accurate representation of the inner states of other animal species; and it certainly constitutes no proof that animals employ the same cognitive apparatus as we do. Even Candland urges us to “deny the arrogance of thinking that we are objective and [to] devote our attention to examining our own categories, and thereby the power and weakness of our own natures” (p. 369).
I take one potentially crucial weakness in this connection to be the tendency to assume that, because of our evolutionary continuity with other species, nonhuman animals must possess the same sophisticated capacities for linguistic intentionality and concept formation that we possess. I see nothing wrong with granting that animal species that lie near the human-nonhuman divide are capable of some conceptual abstraction and at least the rudiments of linguistic communication; one positive lesson of evolutionary theory is that humans do lie on some sort of continuum with other animal species and that the historical supposition that we, unlike all other animals, are created in God’s image is the expression of a wish to be unique and to be superior to all other earthly beings. But some differences in degree are so significant as to constitute differences in kind; and I believe that the possession of conceptual ability and predicative intentionality, both of which are needed to transform communication into language, distinguish humans from all but perhaps the most sophisticated nonhuman animals. Even in the face of some extraordinary ethological discoveries, it should still be apparent to us that there are some differences in kind between humans and nonhuman animals. In particular—and I accept Nagel’s proposition that we cannot ultimately be certain about this sort of thing—nonhuman animals do not appear to be capable of contemplating the distant future or the remote past. To do things that are somehow related to the distant future or remote past is not to be able to contemplate past or future moments; I believe that animals that do exhibit some sort of relation to the remote past or future may well do so in a fundamentally different way than humans do. Specifically, I believe that all but perhaps the most sophisticated animals relate to past and future nonconceptually, and this mode of relating to past and future significantly restricts the capacity of animals to engage in practices such as long-range planning. Bear in mind here that practices such as storing nuts for the winter can be explained without any reference to conceptual abstraction and thinking about the future as such.4 I also believe it is in virtue of the lack of conceptual ability in most animals that it does not make sense to attribute anything like moral agency to animals. For activities such as long-range planning and moral choice involve the capacity to represent oneself to oneself as a self among other selves, to take an explicit inventory of one’s needs and desires as well as the needs and desires of other agents, and to represent and relate to one another specific parts of space and moments of time that extend far beyond what is immediately perceivable. Such capacities involve formal, predicatively structured intentionality, which is the ability to relate logical subjects and predicates to one another in ways that represent the world; this activity, in turn, requires conceptual ability, inasmuch as the logical subjects and predicates, as well as the possible relations between them, are themselves conceptual units. The manipulation of these conceptual units enables us to grasp reality as such, which is to say that we represent states of affairs and contingencies explicitly and in relation to one another. My best sense is that any being capable of operations such as the consideration of counterfactuals must be capable of forming concepts and intentional states, and I find it hard to believe that many if any nonhuman animals are capable of contemplating counterfactual conditions such as “what if it were to rain next Tuesday?” I further consider it more than coincidental that language is predicatively structured, i.e., I proceed on the assumption that all and only those beings capable of intentionality are capable of language, and vice versa: All and only those beings capable of language are capable of intentionality. Why? Because, as the Stoic philosophers observed, human beings but not animals appear to be able to contemplate isolated moments and events, even those far removed from the present, as such, which is to say that humans appear to be unique among living beings in possessing the capacity to consider a moment or an object in the abstract.5
Of course, there is no knowing this with any degree of certainty. The importance of Nagel’s challenge to the science of ethology consists in its recognition that our judgments and inferences about the mental states of nonhuman animals must ultimately remain hypothetical, which is to say that science cannot have the last word about the nature of animal cognition. Griffin himself proceeds on the basis of an open acknowledgment that he is employing a colossal and ultimately unprovable hypothesis when he attributes conceptual ability to animals. Thus pragmatic criteria become indispensable, if not ultimately dispositive, in addressing questions pertaining to animal cognition. In particular we have to ask ourselves whether, on the basis of what we experience in ourselves and what we experience in animals, it seems prudent to attribute to animals the sophisticated abilities associated with intentionality, in particular the capacity to contemplate individual units of meaning “as such.” The capacity to contemplate things as such is perhaps the most sophisticated form of abstract thinking, and on my view it is in all likelihood possessed exclusively by linguistic, which is to say human, beings. An excellent example of the as such, and one of great relevance in reflecting on the moral status of animals, is the capacity to contemplate rights and duties. Human beings are capable of taking on duties in virtue of their ability to contemplate what a right and a duty are in the abstract and in turn to reflect on specific rights and duties. Thus it makes sense to say that human beings, at least those of a certain degree of cognitive and moral maturity, possess rights and duties. And while I believe it makes perfect sense to say that sentient nonhuman animals have rights (such as the right not to be killed and eaten by human beings), I consider it neither sensible nor prudent to say that animals have duties toward other sentient beings. To be able to take on a duty is to be able to contemplate that duty, which presupposes, among other things, the ability to contemplate what a duty in general is. A very apparent if ultimately unprovable difference between humans and nonhuman animals is that humans appear to possess the kind of cognitive abilities in virtue of which it makes sense to hold humans morally and legally responsible, whereas animals precisely do not. This is why, on the view that I have developed in my work on the moral status of animals, I argue that it would make no sense whatsoever, for example, to say that lions have duties not to harm gazelles, even though it makes perfect sense to say that human beings have duties not to harm nonhuman animals. It makes sense, even if you do not ultimately agree with it. But it would not comparably make sense to attribute moral responsibility to nonhuman animals, because they give no indication of being able to step back from their bodily desires and evaluate them in the ways in which human beings seem quite clearly to be capable of doing.
A useful occasion for reflection on the question of sophisticated conceptual abilities in animals is provided by Irene Pepperberg’s work with Alex the parrot. A close reading of Pepperberg’s work shows an acknowledgment on her part that we cannot ultimately know the cognitive mechanisms that were at work when Alex the parrot performed his remarkable cognitive tasks. For example, Pepperberg documented Alex’s ability to answer questions about the number and color of objects present on a table; he was able to do this with up to six objects. Pepperberg (2005:197) notes that Alex’s accuracy on this task was “comparable to that of chimpanzees and very young children.” Given that “‘number sense’ requires handling abstract concepts—representations and relations,” Pepperberg concluded that “Alex [used] and comprehend[ed], in appropriate situations, abstract utterances at a representational level” and that he may “have shown a numerical competence not unlike that of humans” (p. 204). But, at the same time, Pepperberg acknowledges that Alex might simply have been making some kind of elaborate associations between present sense impressions and past experience, or he might have been “subitizing,” which means that he may simply have been instantaneously recognizing a pattern on the table the way one might recognize a particular die or domino without actually counting the dots.
I do not doubt Pepperberg’s conclusion that Alex did as well on this task as chimpanzees or young children. What I do doubt is that an experimental setup of this kind can offer conclusive evidence that, as Pepperberg urges, Alex comprehended “abstract utterances at a representational level.” At the same time, Pepperberg does offer one observation that might lend credibility to the claim that Alex was capable of abstract thought. In the experiments with colored objects, Alex spontaneously and appropriately employed the word none. In a previous study, he “had been taught to use ‘none’ to indicate absence of information in one situation and, without training, transferred its use to another when specifically queried” (p. 202). In the study in question, Alex spontaneously used the term none appropriately, without Pepperberg having solicited it as a possible answer. Pepperberg concluded from this that Alex employed a notion that is “abstract and relies on violation of an expectation of presence” (p. 203). Pepperberg notes that “even young children and some apes have some difficulty with ordinal use of zero” (p. 203), and hence that it would be hasty to conclude that Alex possessed an ordinal comprehension of zero. Nonetheless, Pepperberg urges the conclusion that Alex possessed a sophisticated capacity for conceptual abstraction; she acknowledges but resists the possibility that Alex was engaging in some form of association that is in no way dependent upon the capacity for abstraction.
In the case of cognitively highly sophisticated animals such as Alex, I don’t see why we need to exclude categorically the possibility that they employ conceptual abstraction and intentionality. But the attribution of such capacities to other animals strikes me as highly implausible for reasons that I have touched upon already in connection with the notions of right and duty. These reasons are elaborated by Donald Davidson (1985) in his reflections on the dynamics of belief formation. Davidson presents a holistic conception of belief according to which any being capable of having one belief must be capable of having a whole array of beliefs and a grasp of the difference between truth and falsity; to have these capacities, a being must be capable of linguistic intentionality. “One belief demands many beliefs, and beliefs demand other basic attitudes such as intentions, desires and, if I am right, the gift of tongues. This does not mean that there are not borderline cases. Nevertheless, the intrinsically holistic character of the propositional attitudes makes the distinction between having any and having none dramatic” (p. 473).6 Propositional attitudes or intentional states, like language, are predicatively structured and involve the use of concepts. Any being that possesses any of these abilities possesses them all. To suppose that a being could be capable of intentionality and concepts without being capable of language would be to suppose that language consists simply in the attachment of words to fully formed, predicatively structured thoughts—it would be to misunderstand completely the nature of language. Beings that are capable of concepts and intentionality, which is to say linguistic beings, are not simply able to relate to a complex array of objects and conditions; they are able to represent these objects and conditions as such, which is to say that they are able to relate to them as explicit objects of contemplation. Wittgenstein makes a good point when he observes that the dog cannot believe that his companion human will return the day after tomorrow. But he makes a mistake when he supposes that the dog is capable of believing “that his master is at the door” right now (Wittgenstein 1953:174). The dog may well have a mental state that is equivalent to the state of belief in a human being, but the dog’s mental state, I believe, is not an intentional state with a propositional attitude and conceptual content.
The Soviet psychologist Lev Vygotsky proposed that human children are incapable of conceptual abstraction, that they relate to objects and situations by means of what Vygotsky called pseudoconcepts. These are mental tools that lack the logical ordering and unity of genuine concepts but nonetheless allow the being employing them to make complex associations between present stimuli and past experiences or future possibilities. The pseudoconcept is “only an associative complex limited to a certain kind of perceptual bond” based on a “concrete, visible likeness” (Vygotsky 1986 [1934]:119).7 For example, when an individual is presented with a triangle and asked to pick out all the triangles in an array of objects, the selection process can be based on a concept or on an association made between the triangles in the array and the image of the triangle with which the individual was first presented. On Vygotsky’s view, a human child who has not yet developed the capacity to grasp the logical relations involved in a conceptual grasp of triangularity must employ a pseudoconcept, the functional equivalent of a genuine concept, in selecting the triangles. This fact is obscured by the ability of adults and children to communicate with one another and by the fact that the adult and the child may make the same selection. Vygotsky argues that “the functional equivalence between complex and concept” has “led to the false assumption that all forms of adult intellectual activity are already present in embryo in the child’s thinking and that no drastic change occurs at the age of puberty” (p. 121). It is at the age of puberty, on Vygotsky’s view, that humans first become capable of genuine conceptual abstraction (p. 98).
Even if Vygotsky’s claim that human children are incapable of conceptual abstraction has become implausible, his notion of the pseudoconcept and his claim that genuine abstraction is inseparable from linguistic ability have important implications for the understanding of animal behavior. Davidson is right to suggest that there is a dramatic difference between linguistic and nonlinguistic beings, and I see no conflict between this suggestion and the Darwinian view that most if not all differences between human and nonhuman animals are differences in degree rather than differences in kind. I think what makes us hesitant to acknowledge basic differences between human and nonhuman animal cognition is the fear that we will fall into the old trap of supposing that we are somehow not animals and that by acknowledging basic differences between human and animal cognition we will condemn animals to the kind of inferior moral status that since antiquity has made us feel entitled to exploit animals to gratify our own desires. But, once we recognize that differences in cognitive ability have no moral significance whatsoever, we should be able to overcome this fear and begin to acknowledge that animals don’t think the way we do and that their ways of relating to the world, while rich and complex, are not conceptual or predicatively structured. In doing so, we will stop trying to recreate animals in our own image and begin to let animal beings be the beings they truly are.
Notes
References
Allen, C., and M. Hauser. 1991. “Concept Attribution in Nonhuman Animals: Theoretical and Methodological Problems in Ascribing Complex Mental Processes.” Philosophy of Science 58:221–40.
Candland, D. K. 1993. Feral Children and Clever Animals: Reflections on Human Nature. New York: Oxford University Press.
Davidson, D. 1985. “Rational Animals.” In E. Lepore and B. McLaughlin, eds., Actions and Events: Perspectives on the Philosophy of Donald Davidson, pp. 473–480. Oxford: Basil Blackwell.
Griffin, D. R. 1992. Animal Minds. Chicago: University of Chicago Press.
Nagel, T. 1974. “What Is It Like to Be a Bat?” Philosophical Review 82:435–456.
Pepperberg, I. M. 2005. “Number Comprehension by a Grey Parrot (Psittacus eritha cus), Including a Zero-like Concept.” Journal of Comparative Psychology 119:197–209.
Plutarch. 1995 [ad 70]. Moralia. Vol. 12. Cambridge: Harvard University Press.
Sextus Empiricus. 1996 [ca. ad 190]. The Skeptic Way: Sextus Empiricus’s “Outlines of Pyrrhonism.” New York: Oxford University Press.
Steiner, G. 2005. Anthropocentrism and Its Discontents: The Moral Status of Animals in the History of Western Philosophy. Pittsburgh: University of Pittsburgh Press.
——. 2008. Animals and the Moral Community: Mental Life, Moral Status, and Kin ship. New York: Columbia University Press.
Vygotsky, L. 1986 [1934]. Thought and Language. Cambridge: MIT Press.
Wittgenstein, L. 1953. Philosophical Investigations. Trans. G. E. M. Anscombe. New York: Macmillan.