"The creator of the controversial field of cognitive ethology, Donald R. Griffin has spent more than three decades researching animal cognition. In a completely revised and updated edition of his classic, Animal Minds: Beyond Cognition to Consciousness, Griffin . . . discusses his own and others' research findings including those of his critics. . . . Griffin's book will enlighten, delight and even ruffle some feathers."
"This is the book anyone interested in animal intelligence has been waiting for. Destined to become a classic, it is a brilliant synthesis of our current knowledge about the mental life of other species from the scientist who has contributed more to this subject than anyone else."
"It is difficult for us to imagine effectively dealing with the external world without our ability of internal representation. Yet, it has almost universally been assumed that all other animals are optimally adapted while totally lacking any such ability. Is that assumption a fallacy, or does it encapsulate our scientific inadequacy? Donald R. Griffin is a pioneering scholar who does not cede the world of some animal's inner representation as an impossible barrier. Instead, he presents it as an exciting objective scientific frontier."
An excerpt from
In Favor of Animal Consciousness
A hungry chimpanzee walking through his native rain forest comes upon a large Panda oleosa nut lying on the ground under one of the widely scattered Panda trees. He knows that these nuts are much too hard to open with his hands or teeth and that although he can use pieces of wood or relatively soft rocks to batter open the more abundant Coula edulis nuts, these tough Panda nuts can only be cracked by pounding them with a very hard piece of rock. Very few stones are available in the rain forest, but he walks 80 meters straight to another tree where several days ago he had cracked open a Panda nut with a large chunk of granite. He carries this rock back to the nut he has just found, places it in a crotch between two buttress roots, and cracks it open with a few well-aimed blows. (The loud noises of chimpanzees cracking nuts with rocks had led early European explorers to suspect that some unknown native tribe was forging metal tools in the depths of the rain forest.)
In a city park in Japan, a hungry green-backed heron picks up a twig, breaks it into small pieces, and carries one of these to the edge of a pond, where she drops it into the water. At first it drifts away, but she picks it up and brings it back. She watches the floating twig intently until small minnows swim up to it, and she then seizes one by a rapid thrusting grab with her long, sharp bill. Another green-backed heron from the same colony carries bits of material to a branch extending out over the pond and tosses the bait into the water below. When minnows approach this bait, he flies down and seizes one on the wing.
Must we reject, or repress, any suggestion that the chimpanzees or the herons think consciously about the tasty food they manage to obtain by these coordinated actions? Many animals adapt their behavior to the challenges they face either under natural conditions or in laboratory experiments. This has persuaded many scientists that some sort of cognition must be required to orchestrate such versatile behavior. For example, in other parts of Africa chimpanzees select suitable branches from which they break off twigs to produce a slender probe, which they carry some distance to poke it into a termite nest and eat the termites clinging to it as it is withdrawn. Apes have also learned to use artificial communication systems to ask for objects and activities they want and to answer simple questions about pictures of familiar things. Vervet monkeys employ different alarm calls to inform their companions about particular types of predator.
Such ingenuity is not limited to primates. Lionesses sometimes cooperate in surrounding prey or drive prey toward a companion waiting in a concealed position. Captive beaver have modified their customary patterns of lodge- and dam-building behavior by piling material around a vertical pole at the top of which was located food that they could not otherwise reach. They are also very ingenious at plugging water leaks, sometimes cutting pieces of wood to fit a particular hole through which water is escaping. Under natural conditions, in late winter some beaver cut holes in the dams they have previously constructed, causing the water level to drop, which allows them to swim about under the ice without holding their breath.
Nor is appropriate adaptation of complex behavior to changing circumstances a mammalian monopoly. Bowerbirds construct and decorate bowers that help them attract females for mating. Plovers carry out injury-simulating distraction displays that lead predators away from their eggs or young, and they adjust these displays according to the intruder's behavior. A parrot uses imitations of spoken English words to ask for things he wants to play with and to answer simple questions such as whether two objects are the same or different, or whether they differ in shape or color. Even certain insects, specifically the honeybees, employ symbolic gestures to communicate the direction and distance their sisters must fly to reach food or other things that are important to the colony.
These are only a few of the more striking examples of versatile behavior on the part of animals that will be discussed in the following pages. Although these are not routine everyday occurrences, the fact that animals are capable of such versatility has led to a subtle shift on the part of some scientists concerned with animal behavior. Rather than insisting that animals do not think at all, many scientists now believe that they sometimes experience at least simple thoughts, although these thoughts are probably different from any of ours. For example, Terrace (1987, 135) closed a discussion of "thoughts without words" as follows: "Now that there are strong grounds to dispute Descartes' contention that animals lack the ability to think, we have to ask just how animals do think." Because so many cognitive processes are now believed to occur in animal brains, it is more and more difficult to cling to the conviction that this cognition is never accompanied by conscious thoughts.
Conscious thinking may well be a core function of central nervous systems. For conscious animals enjoy the advantage of being able to think about alternative actions and select behavior they believe will get them what they want or help them avoid what they dislike or fear. Of course, human consciousness is astronomically more complex and versatile than any conceivable animal thinking, but the basic question addressed in this book is whether the difference is qualitative and absolute or whether animals are conscious even though the content of their consciousness is undoubtedly limited and very likely quite different from ours. There is of course no reason to suppose that any animal is always conscious of everything it is doing, for we are entirely unaware of many complex activities of our bodies. Consciousness may occur only rarely in some species and not at all in others, and even animals that are sometimes aware of events that are important in their lives may be incapable of understanding many other facts and relationships. But the capability of conscious awareness under some conditions may well be so essential that it is the sine qua non of animal life, even for the smallest and simplest animals that have any central nervous system at all. When the whole system is small, this core function may therefore be a larger fraction of the whole.
The fact that we are consciously aware of only a small fraction of what goes on in our brains has led many scientists to conclude that consciousness is an epiphenomenon or trivial by-product of neural functioning, as discussed by Harnad (1982). But the component of central nervous system activity of which we are consciously aware is of special significance, because it is what makes life real and important to us, as discussed in detail by Siewert (1998). Insofar as other species are conscious, the same importance may well be manifest. Animals may carry out much of their behavior quite unconsciously. Many may never be conscious at all. But insofar as they are conscious, this is an important attribute.
Although nonconscious information processing could in theory produce the same end result as conscious thinking, as emphasized by Shettleworth (1998) and others, it seems likely that conscious thinking and emotional feeling about current, past, and anticipated events is the best way to cope with some of the more critical challenges faced by animals in their natural lives. As pointed out by the philosopher Karl Popper (1978), what he termed "mental powers" are very effective in coping with novel and unpredictable challenges. This is especially true of many animals under natural conditions, where mistakes are often fatal. The effectiveness of conscious thinking and guiding behavioral choices on the basis of emotional feelings about what is liked or disliked may well be so great that this core function is one of the most important activities of which central nervous systems are capable.
The nature of animal minds was a major subject of investigation until it was repressed by behaviorism, as discussed in chapter 2. Darwin, Romanes, Lloyd Morgan, von Uexkull, and many other scientists of the nineteenth and early twentieth centuries were deeply interested in animal mentality. This history has been thoroughly reviewed by Schultz (1975), Wasserman (1981), Boakes (1984), Dewsbury (1984), and R. J. Richards (1987), and especially cogently by Burghardt (1985a, 1985b). What is new is the accumulated results of a century of active and successful investigation of animal behavior. These discoveries have now provided a wealth of data about the complexities and versatility of animal behavior under natural conditions, and what they can learn to do in the laboratory. We can therefore return to the investigation of animal minds with far better and more extensive evidence than what was available to nineteenth-century biologists.
I will take it for granted that behavior and consciousness (human and nonhuman) result entirely from events that occur in their central nervous systems. In other words, I will proceed on the basis of emergent materialism as analyzed by Bunge (1980, 6), Bunge and Ardilla (1987, 27), and Mahner and Bunge (1997, 205-12), and assume that subjective consciousness is an activity of central nervous systems, which are of course part of the physical universe. Just what sort of neural activity leads to consciousness remains a challenging mystery, as I will discuss in detail in chapter 8. But there is no need to call upon immaterial, vitalistic, or supernatural processes to explain how some fraction of human or animal brain activity results in conscious, subjective thoughts and feelings.
No one seriously denies that we experience conscious thoughts and subjective feelings, even though we cannot describe them with complete accuracy and therefore no one else can experience them exactly as we do. The question under consideration in this book is the extent to which nonhuman animals also experience something of the same general nature as the subjective feelings and conscious thoughts that we know at first hand. The content of an animal's conscious experience may be quite different from any human experience. It may ordinarily be limited to what the animal perceives at the moment about its immediate situation, but sometimes its awareness probably includes memories of past perceptions or anticipations of future events. An animal's understanding may be accurate or misleading, and the content of its thoughts may be simple or complex. A conscious animal must often experience some feeling about whatever engages its attention. Furthermore, any thinking animal is likely to guide its behavior at least partly on the basis of the content of its thoughts, however simple or limited these may be.
In their Dictionary of Ethology Immelmann and Beer (1989) define animal consciousness as "immediate awareness of things, events, and relations," but they hasten to add the conventional behavioristic claim that "statements about the nature of this awareness in comparison to that of humans are dismissed by tough-minded scientists as idle speculation." But speculation is where scientific investigation often begins, and I hope to stimulate new and enterprising inquiries that will significantly reduce our current ignorance. Consciousness is not a neat homogeneous entity; there are obviously many kinds and degrees of consciousness. Many scientists (for example, Hauser 2000, xiii) feel that terms such as consciousness are too vague and slippery to be useful in scientific investigation, and they are very reluctant even to begin talking about the possibility of nonhuman consciousness without first settling on clear-cut definitions. On the other hand, Francis Crick (1994, 20) is not so easily inhibited:
Everyone has a rough idea of what is meant by consciousness. It is better to avoid a precise definition of consciousness because of the dangers of premature definitions. Until the problem is understood much better, any attempt at a formal definition is likely to be either misleading or overly restrictive or both. If this seems like cheating, try defining for me the word gene. So much is now known about genes that any simple definition is likely to be inadequate. How much more difficult, then, to define a biological term when rather little is known about it.
An unnecessary confusion has arisen concerning the meanings attached to terms such as awareness, emotion, mind, and conscious. In ordinary usage they denote conscious mental states, but many scientists justify their aversion to these terms and the concepts they designate by arguing that they cannot be defined with the precision necessary for scientific analysis. This has led to a sort of semantic piracy committed by defining mental states in essentially behavioristic terms. Such terms as positive or negative emotion replace like or fear when a living or even a nonliving system is more likely to respond in one way than another, ruling out implicitly by choice of terms subjective experiences such as liking or fearing. Crist (1996, 1998, 1999) has lucidly analyzed the pervasive effect of this "mechanomorphic" terminology. Thinking is often redefined explicitly or implicitly as information processing. Calling it cognition has become a popular way to study animal mentality without recognizing that at least some of the cognition is probably accompanied, and influenced, by conscious subjective experiences.
The Content of Animal Consciousness
Animal thoughts and emotions presumably concern matters of immediate importance to the animals themselves, rather than kinds of conscious thinking that are primarily relevant to human affairs. Consciousness is not an all-or-nothing attribute. It varies widely within our species, and it would be remarkable if the content of every animal's consciousness were identical. Conscious thinking and strong emotional feelings can ordinarily deal with only one or a very few things at a time. Large, complex animals must also be able to organize and retain information about innumerable perceptions and potential actions of which only one or a very few can be the focus of conscious awareness at any one moment.
Recognizing that an animal's consciousness may be quite different from any human thoughts and feelings makes the problem of identifying and analyzing it more difficult. We may, however, tend to exaggerate this difficulty, because many basic concerns are likely to be very similar for most animals that have any conscious experiences at all. Hunger and desire for food, fear of dangers, affection, and hatred may well have much in common for any animal that experiences them. Once we give up the implicit assumption that any conscious experiences of other animals must be a subset of human experiences, we are faced with the difficulty of determining, or indeed even clearly imagining, what such nonhuman experiences are actually like to the animals who experience them. For instance, the range of odors that various animals can distinguish, and to which they react in different ways, suggests variations in subjective emotional feelings that odors may elicit. If a particular species of animal is capable of some type of conscious experience that our brains cannot generate, how can we ever hope to learn what it is? Although total and perfect understanding seem at first to be beyond our reach, enterprising investigation can probably achieve significant if incomplete understanding.
Sensory and perceptual capabilities are an obvious category in which interspecies differences are probably large and important and in which we can anticipate a beginning of such investigation. As discussed in chapter 2, the philosopher Thomas Nagel (1974) aroused great interest by raising the question "What is it like to be a bat?" He chose this example because bats' reliance on echolocation, detecting obstacles and capturing rapidly moving prey by hearing echoes of one's own sounds, seems at first glance so remote from any human experience as to be truly beyond our ken. But this specific example may not be as telling as it seems on first hearing. For blind people also carry out a form of echolocation, detecting objects by hearing echoes of sounds they make. To be sure, human echolocation is severely limited compared to that of bats and dolphins with respect to the types of targets that can be detected and discriminated from each other, and especially with respect to the ability to distinguish relevant echoes from competing sounds. But the fact remains that despite enormous differences in resolution and practical usefulness many blind people, and also well-practiced blindfolded subjects, do make good use of echolocation, as reviewed by Griffin (1958) and Rice (1967a, 1967b).
The case of human echolocation has an intriguing and perhaps significant further ramification. Many blind people succeed quite well at echolocation but do not realize that they are detecting objects by their sense of hearing. Yet if they are prevented from making any sounds, if their hearing is blocked, or if they are subjected to masking noise, they lose almost completely their previous ability to detect obstacles and avoid collisions with them, as demonstrated by the meticulous experiments of Supa, Cotzin, and Dallenbach (1944). But this specialized type of auditory discrimination is not wholly divorced from consciousness. Many other skillful users of echolocation are well aware that they detect obstacles by hearing some difference in the sound of their voices, footsteps, cane taps, or other sounds they have found to be helpful in finding their way. Echolocation is of course only one case in which sensory or perceptual input channels to the central nervous system differ from those with which we are familiar. But it is an instructive example of how critical scientific investigation can lead to understanding of behavioral and subjective experiential phenomena that at one time appeared mysterious and inexplicable.
It is important to distinguish between perceptual and reflective consciousness. The former, called "primary consciousness" by Farthing (1992), Lloyd (1989), and others, includes all sorts of awareness, whereas the latter is a subset of conscious experiences in which the content is conscious experience itself. Reflective consciousness is thinking, or experiencing feelings, about thoughts or feelings themselves, and it is often held to include self-awareness and to be limited to our species. I will return later in this chapter and in chapter 14 to some of the important ramifications of this distinction. But most of the evidence reviewed in this book suggests perceptual rather than reflective consciousness, and to avoid tedious repetition of perceptual I will use the term consciousness to mean this rather than implying that in particular cases the animal is capable of both categories of conscious thinking.
Recognizing our ignorance is a necessary first step toward reducing it. The customary view of animals as always living in a state comparable to that of human sleepwalkers is a sort of negative dogmatism. Because we know far too little to judge with any confidence when animals are or are not conscious, the question of animal consciousness is an open one, awaiting adequate scientific illumination. There is of course no reason to suppose that other animals are capable of the enormous variety of thinking that our species has developed, largely through the use of our magnificent languageespecially written language, which allows the dissemination and preservation of knowledge far beyond what can be achieved by direct communication and individual memories. The principal difference between human and animal consciousness is probably in their content.
Approaches to Animal Awareness
In spite of all the reasons advanced to justify a neglect of animal consciousness, a few scientists have recognized that we cannot adequately appreciate animals without understanding their subjective experiences. For example, the psychologist Walker (1983) edged closer to doing so than most psychologists when he reviewed extensive experimental evidence concerning animal learning and problem-solving in a book titled Animal Thought. He summarized his conclusions as follows: "Some kind of mental activity is being attributed to the animals: that is, there is considered to be some internal sifting and selection of information rather than simply the release of responses by a certain set of environmental conditions. Knowledge of goals, knowledge of space, and knowledge of actions that may lead to goals seem to be independent, but can be fitted together by animals when the need arises" (81). "Our organ of thought may be superior, and we may play it better, but it is surely vain to believe that other possessors of similar instruments leave them quite untouched" (388). Although Walker tip-toed cautiously around the question of nonhuman consciousness, he did push at the behavioristic fence.
The comparative psychologist Roitblat (in Roitblat and Meyer 1995, 3) advocates "a perspective on animal behavior . . . that views animals not as passive reflex devices, but as active information processors, seeking information in their environment, encoding it, and using it for their benefit in flexible and intelligent ways." But he holds back from explicit recognition that seeking information and using it might sometimes be conscious processes. On the other hand, Burghardt (1997) has recently recommended that scientific ethology be expanded to include the private experiences of animals. He proposes that understanding the subjective experiences of animals be recognized as a fifth basic objective in addition to the four that Tinbergen advocated (proximal mechanisms of causation and control, ontogeny, evolutionary history, and survival value).
Because private experiences include subjective feelings as well as cognition, Burghardt (1997) feels that "cognitive ethology" is not a sufficiently inclusive term. When I first suggested this term (Griffin 1976, 102) it seemed appropriate to emphasize that animal behavior could not be adequately understood without considering animal cognition. But over the past quarter-century we have come to appreciate more fully that in most cases subjective emotional feelings are at the heart of human mental experiences, and very likely those of animals as well, as discussed by M. Dawkins (1993) and Damasio (1999). To the extent that animals make conscious choices about their actions, this will usually entail deciding to do what they perceive as helpful in achieving a desired goal, and the desire often seems to be intense, whether the animal wants something or strives to avoid it. Subjective feelings and emotions are thus an important part of cognitive ethology. Limiting it to nonconscious information processing, or to thoughts devoid of emotional importance to the animal itself, is an obsolete relic of behaviorism.
An increasing part of the renaissance of scientific concern with consciousness is renewed attention to emotional feelings as well as "factual" thinking about objects and events. For example, Panksepp (1998), Damasio (1994, 1998, 1999), and LeDoux (1996) have reviewed recent analyses of the neural correlates of emotion. As usual, most of the scientific attention has been directed at human emotions. But Cabanac (1999, 176) has boldly proposed an objective and evolutionary approach based on "emotional fever." Because gentle handling produces an increase in body temperature and heart rate in rats, mice, and lizards but not in frogs or goldfish, he suggests that "the first elements of mental experience emerged between the amphibians and reptiles." This is a sweeping conclusion to base on a few simple physiological measurements, but the search for physiological correlates of mental experiences must begin somewhere.
It not at all clear, however, whether increase in heart rate or the set point for body temperature is a reliable correlate of emotional experience. And the conclusion that emotional feelings are totally absent in all amphibians and fishes seems rather premature. There is a tendency to take some well-studied reflex, such as insect-catching by frogs, as representative of all behavior of members of an entire group of animals. Sensory and neural mechanisms of insect-catching by frogs have been analyzed in detail, and because frogs will also catch small inedible moving objects this behavior has come to be seen as an inflexible reflex. But the courtship behavior of frogs or nest-building fishes involves more versatile behavior than catching flies.
The ethologist Marian Dawkins (1993, 1998) has contributed significantly to the recognition of the central importance of emotions in cognitive ethology. She has developed procedures by which an animal's preferences can be evaluated by allowing it to choose between different environments. More recently she has critically analyzed evidence for animal consciousness and the difficulties of determining whether animals have conscious experiences. Her basic conclusion is that at least mammals and birds are probably conscious at times but that there are many pitfalls that must be carefully avoided in scientific attempts to determine the existence and content of animal consciousness. She sums up the balance of evidence as follows: "Our near-certainty about (human) shared experiences is based, amongst other things, on a mixture of the complexity of their behavior, their ability to 'think' intelligently and on their being able to demonstrate to us that they have a point of view in which what happens to them matters to them. We now know that these three attributescomplexity, thinking and minding about the worldare also present in other species. The conclusion that they, too, are consciously aware is therefore compelling. The balance of evidence (using Occam's razor to cut us down to the simplest hypothesis) is that they are and it seems positively unscientific to deny it" (1993, 177).
Three contributors to a recent symposium volume (Dol et al. 1997) have argued that many animals have conscious experiences of some sort, although other contributors disagree, as will be discussed in chapter 2. Van der Steen (1997) considers consciousness a heterogeneous "umbrella concept" but concludes that although the subject is a difficult one it can be studied scientifically and that some animals are probably conscious at times. Meijsing (1997, 57) reviews the diverse views of scientists and philosophers and concludes: "From an evolutionary point of view, as soon as there is locomotion there is perceptual awareness and as soon as there is perceptual awareness there is self-awareness (meaning awareness of an animal's own body)." A somewhat similar view has been developed by Sheets-Johnstone (1998). And Wemelsfelder (1997, 79) argues that subjective experience is not "hidden" but is expressed in behavior: "Attention is not a by-product of the ability to process information, it forms the very condition for that ability, enabling the animal to evaluate and apply acquired information in flexible and adaptive manner." These views, however, fail to recognize that much complex human behavior takes place without any awareness of what our bodies are doing, so that additional evidence is needed to distinguish between complex and adaptive behavior and behavior that indicates consciousness. Recognizing that we cannot be certain which animals are conscious, R. Bradshaw (1998, 108) and others recommend that when issues of animal welfare are concerned it is best to "assume animals do have consciousness in case they do; if they do not it does not matter."
The taboo against considering subjective experiences of nonhuman animals has become such a serious impediment to scientific investigation that it is time to lay it aside and begin the difficult task of investigating the subjective experiences of nonhuman animals, as recently advocated by Burghardt (1997). To accomplish this we will have to overcome the effective indoctrinationoften accomplished by nonverbal signals of disapprovalthat has inhibited students and young scientists from venturing into this forbidden territory and subjected those who do so to criticism and ridicule, as exemplified by the strong opinion of Boakes (1992) quoted in chapter 2. One result is that students of animal behavior are inhibited from reporting versatile behavior that suggests conscious thinking, and scientific journals sometimes refuse to publish data or interpretations that support the inference of animal consciousness, as described by Searle (1990a, 1990b), Whiten and Byrne (1988), and Heinrich (1995, 1999).
Evidence Suggesting Animal Consciousness
There are several types of scientific evidence that provide promising insights into what life is like for various animals. One category of evidence is the versatility with which many animals adjust their behavior appropriately when confronted with novel challenges. Animals encounter so many unpredictable challenges under natural conditions that it would be very difficult if not impossible for any combination of genetic instructions and individual experience to specify in advance the entire set of actions that are appropriate. But thinking about alternative actions and selecting one believed to be best is an efficient way to cope with unexpected dangers and opportunities. In theory such versatility might result from nonconscious information processing in the brain. But conscious thinking may well be the most efficient way for a central nervous system to weigh different possibilities and evaluate their relative advantages.
A second major category of promising evidence about animal thoughts and feelings is their communicative behavior, which will be discussed in detail in chapters 9-12. And a third type of evidence is available from neuropsychology. For what little is known about the neural correlates of conscious, as opposed to nonconscious, thinking does not suggest that there is anything uniquely human about the basic neural structures and functions that give rise to human consciousness. Chapter 8 will analyze this evidence in some detail.
The "Hard Problem"
The lack of definitive evidence revealing just what neural processes produce consciousness has led Chalmers (1996) to designate the question of how brains produce subjective awareness as the "hard problem." He and others claim that it is such a difficult problem that normal scientific investigation is unable, in principle, to solve it, and that consciousness must be something basically distinct from the rest of the physical universe. But this view, fortunately, has not seriously interfered with a striking renaissance in the scientific investigation of consciousness. In the 1990s numerous neuroscientists, psychologists, philosophers, and others have taken up active investigation and discussion of consciousness. Even distinguished molecular biologists such as Edelman (1989), Edelman and Tononi (2000), and Crick (1994) have joined the quest. One of the most inclusive of several international conferences devoted to consciousness and related subjects led to two massive volume edited by Hameroff et al. (1996, 1998). General reviews of this reawakened concern with consciousness have been published by Crook (1980, 1983, 1987, 1988), Baars (1988, 1997), Chalmers (1996), Flanagan (1992), and Searle (1992, 1998).
Although much of our behavior takes place without any awareness, and this includes most of our physiological functions and the details of such fairly complex actions as coordinated locomotion, the small fraction of which we are aware is certainly important. It is a completely reasonable and significant question whether members of other species experience anything, although the content of their conscious experiences is likely to be quite different from ours. Perhaps we can never discover precisely what the content of nonhuman experiences are, because scientific understanding is seldom complete and perfect. But it seems probable that we can gradually reduce our current ignorance about this significant aspect of life.
The Comparative Analysis of Consciousness
Despite the renaissance of scientific and philosophical interest in consciousness, one of the most significant and promising approaches to the general question has been largely neglected. This is what biologists call the comparative method: analyzing an important function in a variety of species in which it occurs, sometimes in simpler forms, in which it can be studied more effectively without the many interacting complications that obscure its basic properties in the more complicated animals. Crick and Koch (1998), leaders in the renewal of scientific studies of consciousness, take it for granted that monkeys are conscious. But they prefer to defer investigating nonhuman consciousness because they claim that "when one clearly understands, both in detail and in principle, what consciousness involves in humans, then will be the time to consider the problem of consciousness in much simpler animals" (97).
Restricting scientific investigation to the most complex of all known brains may be unwise, however, for insofar as consciousness can be identified and analyzed in a variety of animals, certain species might turn out to be especially suitable for investigating its basic attributes. Obvious analogies are the use of fruit flies for investigations of genetics, squid giant axons for analyzing the biophysics of nerve conduction, laboratory rats and pigeons for studies of learning, and Aplesia for detailed analysis of the cellular and molecular basis of learning. It would have been unwise for the early investigators of genetics or learning to limit their research to primates, and the same may be true for contemporary and future studies of consciousness.
Perceptual and Reflective Consciousness
The psychologist Natsoulas (1978) emphasized a major distinction that is often overlooked. One widespread and important meaning is what he designates as Consciousness 3, following the Oxford English Dictionary: "the state or faculty of being mentally conscious or aware of anything." This Natsoulas calls "our most basic concept of consciousness, for it is implicated in all the other senses. One's being conscious, whatever more it might mean, must include one's being aware of something" (910). Another important meaning is what Natsoulas, again following the OED, calls Consciousness 4, which he defines as "the recognition by the thinking subject of his own acts or affections. . . . One exemplifies Consciousness 4 by being aware of, or by being in a position to be aware of, one's own perception, thought, or other occurrent mental episode" (911). The other shades of meaning analyzed by Natsoulas (1983, 1985, 1986, 1988) are less important for our purposes, but these two impinge directly on the issues that will be discussed in this book.
Natsoulas' Consciousness 3 is similar to conscious perception, although its content may entail memories, anticipations, or imagining nonexistent objects or events, as well as thinking about immediate sensory input. An animal may think consciously about something, as opposed to being influenced by it or reacting to it without any conscious awareness of its existence or effects. It is convenient to call this perceptual consciousness. Consciousness 4, as defined by Natsoulas, entails a conscious awareness that one is thinking or feeling in a certain way. This is conveniently called reflective consciousness, meaning that one is aware of one's own thoughts as well as the objects or activities about which one is thinking. It is a form of introspection, thinking about one's thoughts, but with the addition of being able to think about the thoughts of others. The distinction between perceptual and reflective consciousness is important for the sometimes confused (and almost always confusing) debate among scientists about animal consciousness. Many scientists use the unqualified term conscious to mean reflective consciousness and imply that perceptual consciousness is not consciousness at all. On the other hand, the existence and distribution of relatively simple perceptual consciousness is important in its own right, and when we understand it better we will be in a much better position to investigate whether any animals are also capable of reflective consciousness.
Many behavioral scientists such as Shettleworth (1998) and philosophers such as Lloyd (1989, 186) believe that it is likely that animals may sometimes experience perceptual consciousness but that reflective consciousness is a unique human attribute. The latter would be much more difficult to detect in animals, if it does occur. People can tell when they are thinking about their own thoughts, but it has generally seemed impossible for animals to do so, although animal communication may sometimes serve the same basic function. The very difficulty of detecting whether animals experience reflective consciousness should make us cautious about concluding that it is impossible. Most of the suggestive evidence that will be discussed in this book points toward perceptual rather than reflective consciousness. Those swayed by a visceral feeling that some important level of consciousness must be restricted to our species may cling to reflective consciousness as a bastion still defended by many against the increasing evidence that other animals share to a limited extent many of our mental abilities. The related question of self-awareness will be discussed in chapter 14.
The relation between these two general categories of consciousness can be illustrated by considering a class of intermediate cases, namely, an animal's awareness of its own bodyfor example, the appearance of its feet or the feeling of cold as a winter wind ruffles its fur. This tends to become an intermediate category between perceptual and reflective consciousness, for an animal might be consciously aware not only of some part of its body but also of what that structure was doing. It might not only feel its teeth crunching on food but also realize that it tastes good. Or it might not only feel the ground under its feet but also recognize that it is running in order to escape from a threatening predator. Furthermore, an animal capable of perceptual consciousness must often be aware that a particular companion is eating or fleeing. This means that it is consciously aware of both the action and of who is performing it. These would all be special cases of perceptual consciousness.
This leads to inquiring how likely it is that such an animal would be incapable of thinking that it, itself, was eating or fleeing. If we grant an animal perceptual consciousness of its own actions, the prohibition against conscious awareness of who is eating or fleeing becomes a somewhat strained and artificial restriction. Furthermore, a perceptually conscious animal could scarcely be unaware of its own enjoyment of eating or its fear of the predator from which it is trying desperately to escape. One could argue that perceptually conscious animals are aware of their actions but not of the thoughts and feelings that motivate them. But emotional experiences are often so vivid and intense that it seems unlikely that when an animal is conscious of its actions it could somehow be unaware of its emotions. We might well pause at this point and ask ourselves whether it is really plausible to claim that an animal can consciously experience emotional feelings and simple thoughts but never be aware that it, itself, is having these experiences. Animals must often feel afraid, but are they incapable of thinking about their fear? Are they sometimes conscious of their actions but never of the thoughts and feelings that motivate them?
Consider the case of an animal that barely escapes from the attack of a predator. It was surely frightened at the time, and if it later sees the same predator it presumably remembers both the event and how frightening it was. If so, this memory of its experience of fear would be a case of thinking about one of its emotional experiences. If we grant such an animal the ability to experience perceptual consciousness of a remembered event, how reasonable is it to rule out the possibility of simple reflective awareness of its remembered fear? To continue with this example, suppose that in its previous narrow escape this animal succeeded only because at the last moment it remembered that it could squeeze into a particular cavity. In this emergency it had been perceptually conscious that a certain action, squeezing between the roots of a particular tree, would get it into a safe retreat. Suppose on the following day it sees the same predator in the same area. Our animal would probably remember not only its fear but the tactic that had saved its life the day before. This would be recalling not only the emotional state of fear but also the simple thought of how to escape. If we insist that all animals are incapable of even such simple sorts of reflective consciousness, we are in effect postulating that a perceptual "black hole" encompasses their most intimate and pressing experiences.