![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
In the recent literature on concepts, two extreme positions concerning animal minds are predominant: the one that animals possess neither concepts nor beliefs, and the one that some animals possess concepts as well as beliefs. A characteristic feature of this controversy is the lack of consensus on the criteria for possessing a concept or having a belief. Addressing this deficit, we propose a new theory of concepts which takes recent case studies of complex animal behavior into account. The main aim of the paper is to present an epistemic theory of concepts and to defend a detailed theory of criteria for having concepts. The distinction between nonconceptual, conceptual, and propositional representations is inherent to this theory. Accordingly, it can be reasonably argued that some animals, e.g., grey parrots and apes, operate on conceptual representations.
Notes
Notes
[1] Peacocke (1992) labels this the “principle of dependence.” It is a general methodological principle that is motivated by an analogous view on meaning which relies on understanding: “As a theory of meaning should be a theory of understanding, so a theory of concepts should be a theory of concept possession” (p. 5).
[2] From our point of view, it is inadequate to assume that a thermostat has concepts because it would be completely superfluous for the explanation of its behavior. It can be explained entirely by simple mechanisms including representations of the external world but without presupposing any concepts that characterize the content of these representations. To suppose concepts in the case of thermostats is just as inadequate and superfluous as it would be to try to explain the fact that a table remains in the middle of the room by assuming that it has an intention to do so.
[3] Since the representational content is often context-dependent we implicitly assume that representational contents are defined by functional roles in the relevant contexts. The question of the context-dependency of representational contents will not be discussed in this paper.
[4] The fact that the content is causally relevant seems obvious because the bird does not only have representations of Monarch butterflies but also of other kinds of butterflies.
[5] By referring to Pylyshyn's account of plasticity, we do not subscribe to his general account of cognition as symbol manipulation.
[6] According to personal communication with Julia Fischer, one of the leading experimenters dealing with Rico, there was only one test in which they started to deal with categories. This unpublished test gave evidence that Rico was principally able to learn the categorization of balls as a class even if the dog seems not to rely on this more complex ability when he identifies single objects.
[7] According to Davidson, one cannot have a concept without having a belief, since concepts come only as parts of beliefs.
[8] “Meeting conditions (i)–(iii) above may provide good grounds for attributing concepts to animals, even though these conditions need to be neither necessary nor sufficient for concept possession” (Allen, 1999, p. 37).
[9] Although improvements in the accuracy of the alarm calls during the development of vervets have been observed, it is an open question whether these are caused by a capacity to detect their own discrimination errors.
[10] It is possible to test different levels of discrimination competences without using acoustic labels. In the first part of a test, monkeys were shown a pair of objects (e.g., two identical cups). Then they were shown two pairs of objects, one identical with the first pair and a different pair (e.g., two identical balls). By giving positive or negative feedback, the monkeys can be easily trained to always discriminate the identical pair of objects but also to always discriminate the different pair of objects. In a more advanced test, they were shown first either a pair of two identical objects representing the sameness relation (two identical cups) or a pair of two different objects representing the difference relation (a ball and a shoe). Then they were shown two pairs of novel objects, e.g., two identical puppets and a second pair consisting of a puppet and a radio. The monkey then had to choose which pair of objects is also representing the sameness relation (or the difference relation) as presented in the first case. This kind of strategy was also used to illustrate that monkeys can be trained to distinguish food and nonfood on the conceptual level because they have to evaluate a banana and an apple as instantiating sameness (belonging to the same category) but banana and cup as instantiating difference (belonging to different categories). These kinds of tests nicely allow us to distinguish perceptual sameness/difference from conceptual sameness/difference (Bovet & Vauclair, 2001). In an analogous line of nonlinguistic tests, it can in principle be checked which discrimination abilities a cognitive system has.
[11] The wasp—not being disturbed—“drags the cricket in the burrow, lays her eggs alongside, closes the burrow, and flies away, never to return. In due course, the eggs hatch and the wasp grubs feed off the paralysed cricket” (Wooldridge, 1963, p. 82).
[12] In recent work it is questioned whether the description of the behavior of the Sphex is adequately evaluated as a rigid behavioral program (Keijzer, 2001). But even if that example cannot play its role here the general idea is clear enough.
[13] One can also characterize these interrelations by inferential relations, e.g., what is green is not red. But of course a cognitive system having perception-based concepts need not be able to construct explicitly such propositional inferences as indicated above. These inferential relations are implicit in the representation of properties as belonging to the same dimension. Concerning the distinction of implicit and explicit representations see Dienes and Perner (1999).
[14] According to this definition, the classification adequately expressed by the demonstrative utterance ‘This is a ball’ is only a conceptual but not a propositional representation. This seems implausible at first glance because the philosophy of language defines propositional contents as contents of that-clauses. Looking at the mental representation, this is no longer implausible. The classification of an object as a ball relies—in the basic case—on a sensory representation of the object and the conceptual representation of the property being a ball. The sensory representation is not already a conceptual one. To put it in other words: A demonstrative identification is not per se a conceptual one. Therefore, the utterance ‘This is a ball’ is an adequate expression of a conceptual representation (but not a propositional representation) in such a basic case of mental fixation of the reference.
[15] The classifications may only be available in different situations but, nevertheless, they are in principle available (Newen & Vogeley, 2003, p. 534).
[16] We only mention that we additionally distinguish metarepresentations, i.e., a representational form that includes an intentional system, an attitude and a propositional content. Metarepresentations have the logical structure that is necessary for an explicit representation of propositional attitudes. Although there is a debate on whether or not animals have metarepresentations, there is no conclusive evidence available at the moment. It is only clear that normal four-year-old children are able to use metarepresentations. In this paper we are not contributing to this debate concerning theory of mind (instead, see Newen & Vogeley, 2003).
[17] Here our evaluation differs from that of Pepperberg (1999), who claims that the capacity of answering the questions ‘What's same?’ and ‘What's different?’ implies that Alex has both concepts (p. 64).
[18] Kanzi is able to understand complex unusual and untrained sentences like ‘Can you put the ball on the pine needles?’ His understanding was tested by presenting him many things including a ball and pine needles and Kanzi was able to do what he was expected to do (Savage-Rumbaugh et al., 1998, pp. 58–62).
[19] We suggest that a basic natural language competence is defined by using stimulus-independent representations that can be systematically combined such that compositionality, productivity and systematicity are fulfilled.
[20] Despite important differences, Barsalou (1999), Fodor (1998), and Prinz (2002) offer theories of concepts which essentially rely on nomic mind–world relations. These theories therefore have the disadvantage mentioned above.
[21] The concept EXPLOSION as a type of event is equivalent to the concept BEING AN EXPLOSION used to classify situations which can be treated as complex objects.