Cats have a theory of the mind


What the dog thinks about the cat

There are animals that think definitely, namely the members of our species. People think all the time. You grasp thoughts, ponder them, combine them into sequences of thoughts, thereby forming new thoughts, sometimes expressing them in word and writing and much more. People think when they build houses, when they eat or drink. But people don't just think while they act, they think primarily in order to act. Thoughts are the basis for people to perform certain actions in opposition to others.

But what about other animal species? When a raven builds his nest, does he think he's building a nest? Or if a bee shows its conspecifics with a bee dance where to find nectar, does it know that it is giving directions? Can a chimpanzee be surprised, taken aback, and have a growing doubt? Can we even grasp the thinking of animals? If yes how? If not, why not?

Those are tough questions. The answer to this depends on various factors: on the one hand, of course, on what it means to think in the first place, and on the other hand, on how certain animal behavior is to be interpreted. To get a better grip on these confusing questions, the following example can help. It comes from the American philosopher Norman Malcolm (2005, p. 86) and has often been used both in philosophical discussions and in the feature pages:

Let's assume that our dog is chasing the neighbor's cat. It races at full steam towards an oak, but suddenly swings away at the last moment and disappears on a nearby maple. The dog does not see this maneuver and, when it arrives at the oak, stands on its hind legs, scratches the trunk with its paws as if to climb up, and barks excitedly up at the branches. We who watch the episode from the window say: "He thinks the cat climbed this oak tree."

It is neither incomprehensible nor inappropriate to say that the dog thinks the cat climbed the oak tree. With this we can explain the dog's behavior well. The dog is, so to speak, on the wrong track: he mistakenly thinks that the cat is hiding on the oak. And because he thinks this, he barks up the oak.

We have ascribed a certain thought to the dog. This also gave us better access to the question of whether animals think. Thinking seems to presuppose something like thoughts of a certain content. So the question that we must answer first is: Do animals have thoughts? The example suggests a positive answer. The dog has a thought. What he thinks explains what he does.

But caution is advised here. We explain the dog's behavior to ourselves by ascribing a certain thought to it. But why do we ascribe this thought to him? On the one hand, of course, because of his behavior. On the other hand, because this explanation makes his behavior appear rational.

The first part of the answer is a bit unsatisfactory. We explain what the dog does through his thoughts, and we infer his thoughts through what he does. Aren't we going around in circles? Is the behavior explicit enough to reveal a particular thought? We haven't done anything to rule out alternative explanations of the behavior. Perhaps the dog is simply following a genetic program unknown to us.

Unfortunately, the second part of the answer is also somewhat unsatisfactory. The dog's behavior may seem rational. But is the process that produced this behavior rational? Is the dog acting for reasons? In short: Perhaps we have only given the dog reasons that we would have in his place; maybe we just gave it our inner voice. Then we anthropomorphized the dog's behavior, i.e. we treated him mentally as we would mentally treat a member of our species. However, we have not shown that the dog has thoughts, but only that we can describe its behavior as if it were controlled by thoughts. Or to speak with Gary Larson: Even if the dog Ginger no longer eats the rubbish after being insulted by his master, he does not do so because he would have understood his master's words. How could he?

Conditions for thoughts

Because of these difficulties, many philosophers prefer to address the question "Can animals have thoughts?" What does a being have to be able to do so that it can have thoughts at all? What are the conditions for having thoughts?

- content. When we say that someone can think, we usually mean that they are able to think something. Obviously we never have thoughts just like that, but when we think we always think something. It is therefore a characteristic of thoughts to have a content. The content of the thought that we ascribed to our dog was that the cat is on the tree.

- terms. Thoughts are usually composed of several components. The main components of the dog thought are: "cat" and "tree". These are terms. They are another characteristic of thoughts. As Kant says, thinking is always done in terms. So if the dog actually has the thought that the cat is on the tree, then it must evidently have the terms “cat” and “tree” in some way. But what are concepts? Concepts can be described as classification principles: They are something general to which a particular, individual object is subsumed. Subsumption takes place on the basis of criteria that the object shares with others who also fall under this term. These criteria must of course be known in some way. Because classifying things is not simply a matter of being disposed or a simple causal reaction to objects with the same properties. Rather, to have a concept means to judge that certain things fall under it.

- objectivity. Perhaps the decisive condition of thoughts is their claim to objectivity or their semantic evaluability. Thoughts must be true or false. One who has a thought needs to be able to distinguish it from what it is referring to. So if our dog is to be able to think, then he must be able to wrongly believe that a certain circumstance is present. Only then is it even possible to find out whether his behavior is not just a causal reaction to certain circumstances, but rather is motivated in a rational way.

- functionality. When you have a thought, you usually don't have it just like that. Thoughts are not idle in their carrier, but always fulfill certain functions. A thought usually does not appear alone, but is connected to other thoughts in a systematic way. Furthermore, a thought can be a reason for behavior. Finally, it can in turn be considered and considered. Thoughts are therefore not free-floating entities, but rather form a systematic network in their carrier that fulfills certain functions.

The question now arises as to which conditions a being must meet in order to be able to have functional, semantically evaluable states with a content and a conceptual character. Some philosophers believe that one has to have a language for this. Others, on the other hand, do not consider language skills to be necessary in order to be able to have thoughts (cf. Stephan 2004). We turn to the first group first.

Thoughts with language

There are two basic versions here. One version, especially put forward by Jerry Fodor, considers it obvious that beings such as higher animals who do not have a natural language can also think. For this reason there has to be a language “before” language, a language of the spirit. The problem with this argument in our context, however, is that it already presupposes that animals can think. Because of this, we don't want to look at it in detail here.

The second version assumes that only speakers of a natural language can have thoughts. One of the main advocates of this position is the American philosopher Donald Davidson. (1986; 2005). His starting point is the question of how the content of thoughts can be determined in general. Let's go back to our dog example. How can you determine that the dog does indeed have the thought “The cat is on the tree”? Davidson says that in order to ascribe this thought to the dog, we must ensure that by any description of the tree the dog believes that the cat has climbed it. This means that the dog's thought must have a content or a meaning that can be interpreted (by us). Now, according to Davidson, the meaning of a term like “tree” is determined by two factors: First, it consists in the relationship of this term to the object to which it refers, in this case trees; on the other hand, the meaning of a concept is determined by the inferential relationships in which it stands to other concepts and thoughts. This means that if you have the term "tree", then, according to Davidson, you have to have many general tree ideas at the same time, such as that trees are mostly green in summer and lose their leaves in winter. So if a being has a concept or a thought, then only because it already has several concepts or thoughts that are inferentially related to it. This combination of semantic and epistemic holism forces us, according to Davidson, to attribute thoughts to someone only if they have a network of thoughts that is broadly similar to ours. In the case of animals, however, Davidson said, there is no basis for such an assumption.

In addition, animals cannot fulfill the objectivity claim of thoughts. As noted, this requires the ability to make mistakes. Now a being can only make mistakes if it is also able to act correctly at the moment. That is, mistakes can only be made by someone who can distinguish between appearance and reality. So making mistakes implies that one is able to recognize just this. And for this one must be able to form a (correct) thought about a (recognized as wrong) thought. So you have to be able to form meta-thoughts. And the assumption of such an ability in animals Davidson considers simply to be overwhelming for them.

Thoughts without language

In the recent past, numerous philosophers have attempted to develop a concept of thinking without language (Allen & Bekoff 1997; Bermúdez 2003; Dretske 1988; Millikan 1989; Proust 1997; Sterelny 2003). Their goal is to "naturalize" the mind; H. to assign it (and with it the thought) a place in the natural world that corresponds to the findings of the cognitive sciences and biology. Naturally, these authors turn first to the explanation of thoughts of simple living beings and not to the complex mind of man. But since humans, like all living beings, are a product of evolution, access to the spirit of simple living beings will also tell us something about the human spirit.

For the group of naturalists, the concept of mental representation (MR) is central. However, this term is articulated very differently. For the American philosopher Fred Dretske, an MR is a structure of the brain or mind that stands for something else. It coordinates perceptions (input) and behavior (output). An MR is, as it were, a map with which a living being navigates in its environment, because by means of its MRs it maps its physical and social environment and guides its behavior in it. Thoughts, so the thesis, are MRs of a certain kind. A living being that represents things in its environment in a certain way and consequently behaves, thinks (Dretske 1997). For Dretske, MRs are structures in the brain that, due to their history, have the natural function of carrying information about the environment, which in turn control behavior (Dretske 1988). Natural functions either arise through processes of evolution or are acquired through learning processes (Dretske 1998). The ability of a bird's eye to perceive shapes is the result of evolution (and a process of maturation). The bird's eye has one (or more) innate natural functions that are used to orient the bird. If the bird learns (for example through trial and error) to distinguish certain forms as bad-tasting insects from other forms, then the bird acquires an MR of this insect, which causes it to stop snapping at the insect when it next encounters it, but to avoid it. Let us assume that the bird encounters another, extremely tasty insect, which, however, looks almost exactly like the bad-tasting insect. The bird sees the insect, but does not see through the mimicry, flies away and misses a bite. In this case, the MR incorrectly represented the second insect as bad tasting.

Is such an MR a thought? Let us remember the four conditions for thoughts. Does the bird's MR meet these conditions? Well, the bird's MR is about the bad-tasting insect, that's its content. In the example, the bird does not connect its MR with other thoughts and certainly does not think about its MR. But the MR triggers his avoidance behavior. The bird thinks that a bad tasting insect is crawling in front of it, but it doesn't know. He thinks that wrongly. The MR shows something that is not present relative to its function. Theories that explain misrepresentations as malfunctions are teleosemantic theories (Dretske 1988, pp. 64-70; Millikan 1989); Proust 1997, pp. 213-34). Through trial-and-error learning, the bird acquires the ability to distinguish food from non-food, insects from nuts, and bad-tasting insects from tasty. Dretske is familiar with the structure that controls this ability to differentiate. Having a concept is also shown here in the exercise of a certain ability to classify.

It looks as if we have a valid candidate for a thought without language in the natural, acquired MR, which has the function of carrying certain information about the environment and which controls the behavior of a living being. However, it is easy to see that such an MR is of limited range.

If thinking is connecting and reflecting on thoughts, then the bird does not think even when it has thoughts. So it would have to be shown how speechless living beings connect and reflect their MRs. Further reflections on this can be found in José Luis Bermúdez (2003). For an explanation of the behavior it would also have to be shown how thoughts about the environment interact with the instincts, motivations, needs or desires of a living being. The behavior control is presumably only done through the coordination of informing and motivating states.

The function of an MR defines what information it should carry. Let us assume that this actually explains the possibility of error. The question remains, what information does the MR carry, what does it represent? Here again we have the problem of salary specification. Even more fundamentally, a theory about what it means for a living being to represent its environment is missing to demonstrate objectivity. How is an environment created for a living being? When does it not just react to sensory stimuli on its body surface, but rather relate to things and events in its environment? Joëlle Proust (1997) and Kim Sterelny (2003) provide further considerations.

Having a concept means being able to acquire and meet differences. Nevertheless, the criterion is felt to be too weak. The normative side of the term must also be taken into account. Colin Allen (2005) has therefore proposed two further criteria. Not only should an animal systematically distinguish some X from some non-X, it should be able to (a) determine some of its own misjudgments and (b) learn from (a) to better distinguish between X and non-X . Proust (2005), on the other hand, refers to the social dimension of the terms and calls for sanctions for incorrect distinctions by social partners.

So Dretske's approach can be expanded despite its limited scope. The differences between the cognitive abilities of humans and other animals are by no means blurred. Ruth Millikan (1989), for example, points out six striking differences.

The idea that mental representations are the non-linguistic carriers of thoughts is not shared by everyone who thinks it is possible to have thoughts without having a natural language at the same time. There are a number of philosophers who, following the criticism by Ludwig Wittgenstein and Gilbert Ryle, consider the concept of MR to be of little use. The German philosopher Hans-Johann Glock takes a so-called holodoxastic approach to the attribution of thoughts (Glock 2000; 2005). According to this, thoughts are not MRs. Also, as with Davidson, one does not ascribe thoughts on the basis of consent to whole sentences, i.e. holophrastically. According to this approach, the content of thoughts is determined on the basis of behavior, body postures or facial expressions. No behaviorism is to be asserted with this. According to Glock, thoughts are not identical to behavior. However, the content of thoughts - even in animals - can be determined on the basis of behavior.

In the case of a holodoxastic attribution of thoughts to animals, Glock pleads for an offensive handling of the accusation of anthropomorphism. If Davidson is right and the assumption of common thought networks is the only way to determine the content of foreign thoughts - regardless of whose - then we cannot avoid assuming a network of thoughts that is largely similar to ours in the case of animals either. This assumption has the function of a contrast film, before which deviations in the animal's network of ideas can be detected in the first place.

In determining the content of animal thoughts, however, Glock adheres to the same criteria for thoughts as Davidson. So he too is of the opinion that their content is conceptual. Since terms are principles of classification, the discriminative behavior of thinking animals must be judgments. How can one recognize, however, that a reason has determined this discriminative behavior and not rather a stimulus that was causally followed by a certain reaction? Now on the basis of the third condition for thoughts: their claim to objectivity or their semantic evaluability. We have already seen that a thought must be able to exist even when there is nothing in the world to which it applies. Since in animals their behavior is supposed to be a mark of thoughts, one must be able to show that they are able to act wrongly. According to Glock, intentionality, rule-based behavior and voluntariness of animal behavior are decisive for this. The animal must therefore also be able to act differently.

Now we have seen from Davidson that in order to act wrongly, one must be able to form meta-thoughts. So the question that arises for Glock is: what would be a sign of meta-thinking? As we saw with Davidson, the ability to make mistakes implies the further ability to see just that. For this reason, e.g. corrective behavior can be a sign of meta-thoughts. It requires that the person correcting his behavior must have a (correct) thought about a thought (recognized as wrong). And Glock believes that such behavior can be observed in some animals. There are now studies with pigs that show that they can not only differentiate between pairs of objects according to categorization schemes such as "square" or "round", but even correct their actions if necessary - before they press the button that rewards the correct categorization with feed (Allen 2005).

Regardless of how one evaluates the dispute about MR and naturalization, for the question “do animals think?” It is obviously primarily important that behavior and the associated learning processes and skills are the key to processing and Give an answer.

Conscious experience

So far we have talked about thinking and thoughts, but not about the conscious experience (consciousness) of animals. A very natural question is: how do certain things feel to an animal? When it comes to the question of conscious experience, it is advisable to abandon the title question and concentrate more on sensory perceptions or pain, which are often exemplarily associated with conscious experience. In the philosophy of mind, questions about the content of thoughts and perceptions (intentionality) and questions about consciousness (phenomenality) are almost routinely separated. Behavioral researchers also often separate the question of animal cognition from the question of conscious experience (Shettleworth 1998, p. 5ff.). The reason for this is that one does not know how to prove the presence of conscious experience in speechless living beings (Clayton 2001, p. 285).

Can we even gain access to foreign subjects (Kaeser 2003)? For example, could we know what it's like to be a bat? In a famous essay, the philosopher Thomas Nagel answered the last question in the negative (1984). His argument is: If I'm not a bat-like being, then I don't have a bat's perspective. And if I don't have a bat's perspective, then I can't know what it's like to be a bat. One can object that this argument is based on a pseudo-task: What follows from the impossibility of being able to do something? I can't know what it's like to be a bat because I'm not a bat. But it does not follow that it is somehow for a bat to be a bat. In other words: Why is Nagel so sure of the attribution of conscious experiences to bats? In other words, the really crucial question is: Do certain things even feel to an animal at all?

A particularly pressing topic is the question of how animals experience pain and their ability to suffer. The German Animal Welfare Act (§1) states: "Nobody may inflict pain, suffering or harm on an animal without a reasonable reason." Now there are philosophers who question the experience of pain and the ability to suffer in animals (Dennett 2005) who consider the arguments put forward as inadequate viewing (Harrison 1991) or demanding high cognitive abilities for conscious experience, which most animals cannot fulfill (Carruthers 1989). Even balanced assessments do not come to clear conclusions (Allen 2005).

A well-known and often used argument assumes similarities between humans and certain animal species. (It is very clearly articulated in Robinson 1996). Some animal species are anatomically and physiologically built in a similar way to humans. When it comes to experiencing pain, six criteria are particularly important: people have (i) pain receptors, (ii) a central nervous system and (iii) a connection between the two: injuries cause events in the central nervous system and these in turn cause conscious pain experiences. (iv) These experiences can be alleviated with analgesics. (v) There are also substances produced by the body to relieve pain (so-called endogenous opioids). (vi) Injuries trigger a specific pain behavior, we generally try to escape the damaging stimulus, try to get rid of the pain, and care for the injury. Now, if an animal species meets all of these criteria, it is likely that a similar type of injury is causing similar conscious pain experiences. This argument is an empirically based analogy argument. Such an argument can only make the matter in question plausible, it cannot prove it. In addition, it leaves the pain experience of such animal species indefinitely in individual cases, which do not meet all of the six criteria mentioned.

Although, as I said, it is customary in the philosophy of mind to separate questions about intentionality from thoughts and about phenomenal consciousness, in the last 15 years a position has been established that regards phenomenality as a special form of intentionality, the so-called representationalism or intentionalism (Dretske 1998). This position allows a very natural theoretical perspective on the consciousness of other species. Michael Tye (1996) argues as follows: If animals can be ascribed a special kind of sensory representations, then one can also ascribe conscious experience to them; Tye calls these representations PANIC states. The acronym does not stand for completely frightened animals, but for: available (poised), abstract (abstract), non-conceptual (nonconceptual), intentional (intentional) content. For example, honeybees' sensory representations of flower colors refer to this very color. That is their intentional content. In order to perceive flower colors, according to Tye, bees do not need any terms for these colors (just as we do not). The salary is non-conceptual. Bees are capable of learning about their ability to differentiate between flower colors. The intentional content is therefore available for further cognitive processes (learning). If now PANIC states are identical with conscious experience, then honeybees have conscious experiences. Tye's theory is elegant because it allows conscious experience to be tied to certain behaviors. And it is stronger than the analogy argument, because it is based on a metaphysical thesis about consciousness.

The theoretical relevance of the Frag

What is the significance of the question of whether animals can think? On the one hand, there is a moral relevance to this question. One can argue that only an interest in one's own future guarantees a right to life. Then only those animals would have a right to life that think into the future and can design a future (Krebs 2003). But what is the theoretical relevance of the question? Two answers come to mind here.

Empirical sciences such as developmental psychology, anthropology and ethology increasingly regard non-linguistic living beings (small children, early hominids, animals) as thinking beings. It seems as if these sciences demand an elaborate concept of non-linguistic thinking that is not primarily formed in analogy to language (Bermúdez 2003). In particular, cognitive ethology, i. i. research into information processing, thought processes, beliefs and consciousness in non-human animals requires such fundamental reflections (Allen & Bekoff 1997).

Now, of course, empirical investigations into the possibility of attributing thoughts to animals cannot be conducted independently of a knowledge of what is actually to be investigated. In other words, in order to ask whether thoughts can be ascribed to animals, it must first be clarified what thoughts actually are. And in this way one inevitably ends up in the field of theoretical philosophy - and in its entire field. Because questions about the spirit of animals take up topics from epistemology, the philosophy of language, the theory of action as well as the philosophy of science. As we have seen, it is important to clarify, among other things, what terms are to be understood as components of thoughts and how their content is determined. This is followed by the question of whether a language is required to dispose of terms.

At least it seems certain that animals do not have a language with a syntax. Since animal behavior, their facial expressions and their postures represent the object of investigation for cognitive ethologists and animal psychologists, one must also go into the field of action theory. Here, among other things, the conditions under which behavior can be viewed as an action - and thus as rationally motivated - are examined, among other things. When can you ascribe intentions to a being and what exactly do you ascribe to it when you do it?

Empirical sciences are obviously dependent on philosophical work here. However, one should not think that philosophers would merely act as “assistants” in the natural sciences. This role would be a surprise, especially for them. No, answering the question of whether animals can think also has a philosophical intrinsic value. Because animals, it can be said, represent a kind of test case for theories of the mind. They can be used to demonstrate the scope of such theories, which phenomena they are able to explain and where their limits are. It is important, however, that philosophers cannot make such decisions from their armchairs alone. For this you are absolutely dependent on the research results of the empirical sciences. The answer to the question about the spirit of animals can rightly be described as an interdisciplinary project in which empirical scientists and philosophers can benefit equally from each other's working methods.

More fundamentally, the question of how animals think is also used to help people understand themselves. What makes it different from the animal? The human-animal distinction is used to answer the (not only Kantian) question of what humans are. The animal assumes a borderline position. It is used to draw a line. There are many ways to draw and differentiate between humans and animals, but what is the difference that makes these distinctions possible? Man is an animal, but the anthropological question is to what extent man is not an animal. It often coagulates in a "man-is-the-animal-the-X formula". And humans are classically the animal that thinks or that is rational (on the concept of rationality in animals see the articles in Hurley & Nudds 2006). Today this classic distinction seems to be massively questioned, be it as a result of the consideration of scientific research or cultural-scientific reflection (this is of course not a development of the recent past, as a look at history shows. To conclude, we therefore want to present a few selected historical works.

Historical work

Antiquity. The question of the human-animal distinction has been part of the history of philosophy from the very beginning, as Richard Sorabji (1993) shows in his review of the ancient debate. The starting point of the debate is Aristotle, who denied animals a reasonable ability and consequently had to greatly expand the range of sensory perception so that obvious animal abilities such as memory, sensation and intention could be accommodated in it. Aristotle ‘Rejection of animal reason led to a crisis that had a lasting impact on the philosophy of spirit and the moral philosophy of Hellenism.

Early modern age. Even in the early modern period there was an intense discussion about the spirit of animals. This discussion became particularly explosive because, on the one hand, new empirical models for the analysis of animal behavior were designed within the framework of the so-called new science and, on the other hand, the basic cognitive terms were subjected to a radical test. Unlike in antiquity, this discussion was initially not triggered by the denial of a reasonable ability in animals, but on the contrary by Michel de Montaigne's defense of the reason of animals. By reacting to Montaigne and claiming that animals are living machines, René Descartes gave the discussion a sharp outline (Wild, forthcoming). In an extensive study, Gary Steiner (2005) traces the change in anthropocentrism in Western philosophy and the resulting change in the moral attitude towards animals.
A look at the history of the human-animal distinction shows that the question “Do animals think?” Must always be viewed in a larger context. This not only raises questions about the moral and lifeworld relationship between humans and animals, but also theoretical questions about a human self-image. These questions, it seems, always become urgent when epochal changes in the understanding of the world emerge, as the cited works on antiquity and early modern times show. Today these changes are marked by the life sciences. Few thinkers have set our self-image in motion as much as Charles Darwin and the expansion of his theory of evolution. Questions about the self-image of humans and their place in the natural world are still asked about questions of their relationship and their differentiation from animals, questions like "Do animals think?"

Literature on the subject

Introductions, overviews:

(1) The spirit of animals. Philosophical texts on a current debate. Edited by Dominik Perler and Markus Wild. 450 pp., Kt., € 16.—, 2005, Suhrkamp, ​​Frankfurt a. M.

Positions that believe language is necessary for thoughts:

(2) Davidson, Donald: "Thinking and Redening", in: ders .: Truth and Interpretation, 408 pp., Kt., € 16.—, 1986, Suhrkamp, ​​Frankfurt a. M., pp. 224-46.

(3) Davidson, Donald: “Rational Living Beings”, in: (1) pp. 117-31

Positions that represent a concept of thinking without language:

(4) Allen, Colin & Bekoff, Marc: Species of Mind. The Philosophy and Biology of Cognitive Ethology, cloth £ 27.50, pbk. £ 13.95, Bradford Books, 1997, MIT Press, Cambridge (Mass.).

(5) Allen, Colin: “A New Contemplation of Animal Concepts”, in: (1) pp. 191-201.

(6) Allen, Colin: "Animal Pain", in: Noûs 38 (2005), pp. 617-43.

(7) Bermúdez, José Luis: Thinking Without Words, 2003, Oxford University Press (out of print in bookshops)

(8) Dretske, Fred: Explaining Behavior. Reasons in a World of Causes, 180 p., Pbk., £ 11.50, 1988, MIT Press, Cambridge (Mass.).

(9) Dretske, Fred: “The Nature of Thought”, in: Language and Thinking / Language and Thought, ed. by A. Burri, 1997, de Gruyter, Berlin, pp. 288-300.

(10) Dretske, Fred: Die Naturalisierung des Geistes, 185 S., € 24.80, 1999, Mentis, Paderborn.

(11) Millikan, Ruth Garrett: "Biosemantics", in: The Joural of Philosophy 86/6 (1989), pp. 281-97.

(12) Glock, Hans-Johann: "Animals, Thoughts and Concepts", in: Synthesis 123/1 (2000), pp. 35-64.

(13) Glock, Hans-Johann: “Conceptual problems and the problem of concepts”, in: (1), pp. 153-87.

(14) Proust, Joëlle: Comment l'esprit vient aux bêtes, 391 p., € 24.—, 1997, Gallimard, Paris.

(15) Proust, Joëlle: “The intentional animal”, in (1), pp. 223-43.

(16) Stephan, Achim: "Are animals" difficult to understand "?", In: Deutsche Zeitschrift für Philosophie 52 (2004), pp. 569-83.

(17) Sterelny, Kim: Thought in a Hostile World. The Evolution of Human Cognition, 280 p., Cloth £ 60.—, pbk. £ 19.—, 2003, Blackwell, Oxford.

Access to foreign subjects

(18) Kaeser, Eduard: “Access to the alien subject”, in: Philosophia naturalis 40/1 (2003), pp. 1-42

(19) Nagel, Thomas: “What is it like to be a bat?”, In: ders .: About life, soul and death, 1984, Philo, Berlin, pp. 185–99.

Conscious experience

(20) Carruthers, Peter: “Brute Experience”, in: Journal of Philosophy 86 (1989), pp. 258-69.

(21) Clayton, Nicola et. al .: "Declarative and Episodic-like Memory in Animals: Personal Musings of a Scrub Jay", in: The Evolution of Cognition, ed. by C. Heyes & L. Huber, 2000, MIT-Press, Cambridge (Mass.), pp. 273-88.

(22) Daniel Dennett: "Animal consciousness: what is important and why?", In: (1) pp. 389-407.

(23) Harrison, Peter: “Do Animals Feel Pain?”, In: Philosophy 66 (1991), pp. 25-40.

(24) Robinson, William S .: "Some Nonhuman Animals Can Have Pains in a Morally Relevant Sense", in: Biology and Philosophy 12/1 (1996), pp. 51-71.

(25) Shettleworth, Sarah J .: Cognition, Evolution, and Behavior, 704 p., Cloth £ 89.50, pbk. £ 32.50, 1998, Oxford University Press, Oxford.

(26) Tye, Michael: “The Problem of Primitive Forms of Consciousness: Do Bees Have Sensations?”, In: Consciousness and Representation, ed. by F. Esken & D. Heckmann, 1998, Mentis, Paderborn, pp. 91-122.

Historical work

(27) Sorabji, Richard: Animal Minds and Human Morals. The Origins of the Western Debate, 224 p., Cloth £ 40.—, pbk. £ 18.-, 1993, Duckworth, London.

(28) Wild, Markus: The anthropological difference. The spirit of the animals in the early modern times in Montaigne, Descartes and Hume, 350 pp., Hardcover, € 98.—, 2006, De Gruyter, Berlin.

Further literature

(29) Rational Animals? Edited by Susan Hurley and Matthew Nudds. 568 p., Cloth £ 75 .--, pbk. £ 29.95, Oxford, Oxford University Press 2006.

(30) Krebs, Angelika: "Language and Life", in: Describing animals, ed. by A. Brenner, 2003, Harald Fischer, Erlangen, pp. 175-90.

(31) Malcolm, Norman: "Thoughtless Animals", in: (1) pp. 77-94.


Sarah Tietz is doing her doctorate on the topic “Can animals think? A systematic study on the explanation of beliefs ”at the Humboldt University in Berlin.

Markus Wild is a research assistant at the Institute for Philosophy at the Humboldt University in Berlin.