What Can Philosophy and the Cognitive Sciences Give Each Other?

Sushchin Mikhail Aleksandrovich ORCID: 0000-0002-8805-6716 PhD in Philosophy Senior Researcher, Institute of Scientific Information for Social Sciences of the Russian Academy of Sciences 117418, Russia, Moscow, Nakhimovsky Prospekt, 51/21 sushchin@bk.ru Other publications by this author

Introduction

In most cases, philosophy is considered as one of the six disciplines included in the complex of modern cognitive sciences (apparently referring primarily to part of the research in the fields of philosophy of consciousness, epistemology and philosophy of science, affecting the issues of cognitive sciences). In addition to philosophy, cognitive psychology, cognitive neuroscience, the field of research and development of artificial intelligence, cognitive linguistics and cognitive anthropology traditionally belong to the complex of cognitive disciplines.

It is widely known that the philosophers introduced (x. Putnam, J. Fodor, D. Dennett, E. Clark, J. Hochvi, etc.) a significant contribution to the development of some of the most important theoretical concepts and concepts of modern cognitive sciences. Meanwhile, there is currently no clear understanding of the place and role of philosophy as part of cognitive disciplines. At the same time, the question of the role and place of philosophical research in the complex of modern cognitive disciplines is raised and discussed quite rarely. In this work, as a starting point, we will take the positions set out in the articles of three well-known modern philosophers: D. Dennett, P. Tagard and V.A. Lectorsky. After considering the points of view of Dennett, Tagard and Lectorsky, the article will present my own understanding of possible ways of interaction between philosophy and specific cognitive disciplines.

Three concepts of interaction between philosophy and cognitive disciplines

Let's start from the point of view outlined in Dennett's work. As he notes, philosophers are able to make a significant contribution to cognitive research if they start from the dominant constructive attitude in science and are aimed primarily at formulating testable hypotheses ^[1], as well as clarifying the relationship between scientific concepts and concepts like free will, moral responsibility, etc. According to Dennett, there are a number of reasons why the scientific community is skeptical about what philosophers are doing. One of these reasons is the attempts of professional philosophers to prove or refute on the basis of "first principles" anything that relates to the conduct of empirical sciences. (Dennett ironically calls this style attempts to prove on the basis of first principles that bumblebees cannot fly ("bumblebee deductions"), recognizing that he himself carried out such "deductions", deliberately rejecting the idea of "grandmother neurons", "grandmother neurons" ^{[1, p. 232]}.)

According to Dennett, philosophers can contribute to progress in cognitive sciences in several ways. So, they can recognize promising, but vaguely expressed ideas and make them clearer for further research – for example, to help express a dubious idea in a form suitable for refutation. Many scientists, Dennett notes, confidently believe that they are able to do this work themselves, without the help of philosophers. However, in practice, they tend to quickly become aware of the real complexity of the problems they are dealing with and appreciate the advice of professional philosophers. In addition, having abandoned the classical philosophical techniques of refutation based on counterexamples and reductio ad absurdum, philosophers, according to Dennett, are also able to contribute to progress by helping scientists develop experiments. As an example, Denneta refers to his own ideas, which stimulated experimental work on problems about false beliefs, blindness due to inattention, perceptual filling (filling-in), etc. Finally, philosophers, according to Dennett, are able to contribute to the development of cognitive research by discussing issues of free will, moral responsibility, the existence of suffering, etc.

P. Tagard offers a different vision of the interaction of philosophy and special cognitive disciplines ^[2]. In his opinion, philosophy significantly influences cognitive sciences in the way of clarifying general and normative issues. Among the general questions Tagard considers questions about the nature of scientific theories and explanations, the role of computer modeling, the relationship of various cognitive disciplines to each other. He refers to normative issues as questions about the Bayesian nature of human thinking, about the establishment of norms themselves, etc.

As Tagard notes, any advanced field of scientific knowledge inevitably faces questions about the nature of knowledge and reality. For example, advanced research in the field of theoretical and experimental physics deals with questions about the nature of space and time, as well as a number of methodological problems. As an advanced discipline, cognitive science, according to Tagard, faces the following general questions: what is the nature of explanation and what is the nature of theory? How to evaluate alternative theories? How do different special cognitive disciplines, such as cognitive psychology and cognitive neuroscience, relate to each other? Is psychology reducible to neuroscience? What is the role of computer modeling in cognitive research? These questions, according to Tagard, ultimately affect the limits of the development of cognitive science, and ignoring them usually leads to the repetition of philosophical concepts of the nature of scientific knowledge, which turned out to be unfruitful in the past.

In addition to general issues, philosophy is able to benefit cognitive sciences in the way of discussing normative issues. It is customary to attribute a descriptive character to fundamental science, but applied science also has a normative dimension in terms of what it tries to do to improve a person's life. Philosophy is a discipline within cognitive science with the richest experience in discussing normative issues. An example of a normative question is the question of establishing norms of inference and reasoning. There is no consensus among professional logicians as to which standards should be considered appropriate. Some philosophers believed that pure intuition could act as a basis for establishing standards of inference, but disagreement among logicians suggests the need to find a more complex method. According to Tagard, ignoring philosophy is equivalent to implicit and unprofessional philosophizing. In addition, according to Tagard, philosophers can benefit cognitive science by defending its key concepts of computation and representation.

In a broader context, the interaction of philosophy and cognitive sciences is considered by V.A. Lektorsky ^[3]. The lecture examines three global scenarios of interaction between philosophy and specific cognitive sciences. The first scenario, which gained popularity at the dawn of the so-called analytical philosophy thanks to the works of L. Wittgenstein, G. Ryle, R. Carnap, etc., suggests that philosophy should ignore the specific results of cognitive research. Instead, philosophy should concentrate on identifying the necessary dependencies in the structure of consciousness and cognition with the help of some special non–empirical methods - first of all, of course, logical analysis of language, so beloved by adherents of analytical philosophy. However, this kind of installation has encountered significant difficulties. As Lectorsky writes, "The reasoning of an analytical philosopher may look quite strict within the assumptions from which he proceeds.... The question, however, is which intuitions should be considered relevant for solving the corresponding problem and which logical tools should be used as a means. <...> In fact, different analytical philosophers do not explicate the content of the corresponding intuitions, which they claim, but build explanatory models, as a rule, incompatible with each other (at the same time, ideas about possible rules of logical argumentation also change). Therefore, there is no agreement on what should be considered a good result of the analysis" ^{[3, p. 13-14]}.

The second scenario, which is in many ways the opposite of the first and arose as a reaction to it, assumes the naturalization of philosophy. In general, this idea was expressed by the American philosopher and logician W. Quine in the late 1960s. According to Quine, philosophy should abandon claims to normativism, in fact, reducing its efforts to generalizing research in psychology and the brain sciences. Quine called this program of cognition research "naturalized epistemology." Since then, a number of projects have been developed in an anti-normative spirit, including the idea of neurophilosophy by Paul and Patricia Churchland, calling for the replacement of the concepts of "folk psychology" used in everyday life (concepts of thought, desire, intention, etc.) with the language of some hypothetical future progressive neuroscience.

At the same time, as Lectorsky points out, there is another possible way for philosophy and specific cognitive sciences to interact when they enter into a dialogue in which both their mutual enrichment and mutual criticism take place. According to Lectorsky, philosophy has always "tried to critically assess the everyday world and go beyond the accepted cultural stereotypes, creating new ways of intellectual and practical activity. Building a picture of the world, she always tried not only to separate it from the world of everyday ideas, but also to find transitions between them: both in understanding reality and in relation to what a person should do" ^{[3, p. 15]}. So, for example, George Berkeley critically evaluated the foundations of Newton's mechanics, which had just appeared at that time, and its key concepts of absolute time and absolute space, and did not just describe how concepts arise from a set of sensations. In this regard, Lektorsky notes, it is important to note that it is possible to understand the naturalization of epistemology in such a way that it does not merge with cognitive research, but interacts with them and does not lose its normative character. Thus, Lektorsky believes, epistemology can critically analyze the prerequisites of cognitive sciences and enrich itself due to the results obtained in them.

On the ways of mutual influence of philosophy and cognitive sciences

So, we have considered three points of view on the interaction of philosophy and specific cognitive disciplines. Starting in a general sense from the position expressed in the work of the Lecturer, I believe that philosophy and cognitive sciences can enter into a productive dialogue in which their mutual enrichment can take place. The result of this dialogue may be the strengthening or weakening of certain theories or concepts, as well as theoretical progress.

First, it should be said about the possible influence of philosophy on cognitive sciences. Following Tagard, I believe that philosophy can have the greatest impact on the development of cognitive sciences by clarifying a number of issues related to the conduct of the philosophy of science. The famous philosopher of science, B. van Fraassen, argues that any concept of the philosophy of science should be able to answer questions about the nature of scientific theories and what they do ^{[4, p. 7]}. In my opinion, the most important methodological problem for cognitive sciences at the moment is the problem of theoretical progress. It can be formulated as follows. So, on the one hand, in the cognitive sciences (in a broad sense), incredible progress has been noted in the field of improving practices. This applies both to direct tools for studying the brain and cognition (tools for brain research, eye trackers, etc.), and to data processing tools (computers, and their hardware, and the corresponding software). A whole chasm separates modern tools for research of cognitive processes and the equipment that was used at the dawn of modern cognitive sciences – generations of devices replaced each other. This is an indisputable fact. On the other hand, is it possible to talk about “comparable” theoretical progress in cognitive sciences? (Of course, I'm not saying that strict comparisons can be made between these areas. The question of to what extent they are commensurate remains open.) In a general sense, a negative answer can be given here, and the reasons for the extremely slow and insignificant progress of theoretical ideas in the cognitive sciences are precisely the problem indicated.

Closely related to the problem of theoretical progress is the problem of choosing a leading attitude in theoretical, methodological and explanatory contexts. Should such an attitude be pluralistic or monistic? In recent years, a number of authors have discussed the possibility of finding the so-called "grand unification theory" in cognitive sciences, which is sufficiently similar to the image of the paradigm from the first edition of The Structure of Scientific Revolutions ^[5]. Generally speaking, the value of unification is often emphasized in the cognitive literature. Here the question arises: do cognitive sciences need a kind of monopolist paradigm? Should cognitive scientists, as P. Feyerabend ironically wrote at the time, commenting on the attempts of some social scientists to improve their disciplines based on a specific understanding of Kuhn, limit criticism, reduce the number of comprehensive theories to a single theory and create a normal science (i.e., produce theoretical unification) that accepts this theory as a paradigm ^{[6, p. 198]}? In my recent work, I defended the point of view that a pluralistic strategy is preferable for cognitive sciences in both theoretical and methodological contexts ^[7]. Moreover, according to D. Shapir, the explicit expression of discontent, discussions about the foundations of theories, the proliferation of different points of view have always been more or less present in the history of science ^{[8, p. 42]}. Theoretical and methodological pluralism is designed to ensure progress in cognitive research through mutual criticism and preservation of alternatives. However, great care is required when it comes to the explanatory context. Despite the numerous evidences of heterogeneity of cognitive systems (the same human brain), the question of the preferred explanatory attitude should not be solved exclusively in a theoretical way – empirical research should have the last word here.

All this brings us to another important question for the philosophy of cognitive sciences – the question of the nature of large cognitive theoretical trends (or groups of individual theories), such as classical computational cognitivism, connectionism, embodied cognition (or, in another classification, "4E" cognition, 4E Cognition ^[9]) and predictive processing. How should these directions be understood? As paradigms in the manner of T. Kuhn ^[5], research programs in the spirit of I. Lakatos ^[10], or as research traditions in the sense of L. Laudan ^[11]? Or, perhaps, it is worth abandoning the universalism inherent in the theories of Kuhn, Lakatos and Laudan, and opt for the concept of B. Von Eckardt's research conceptual framework specially developed for understanding cognitive trends ^[12]? A series of my recent publications has been devoted to this issue ^{[13, 14]}. I argued that none of the mentioned concepts, primarily due to a number of descriptive problems, can be taken as such as a basis for understanding cognitive theoretical trends. Instead, these directions are proposed to be understood as homogeneous or heterogeneous in structure theoretical complexes designed to promote the proliferation of scientific theories to ensure progress in relation to a number of so-called advantages of a good theory (predictive success, the ability to give unintended explanations of known facts, empirical adequacy, etc.). In my opinion, complexes of cognitive theories are designed to promote theoretical progress in cognitive sciences.

Summarizing what has been said, we note that other important questions of the philosophy of cognitive sciences include questions about the nature of individual cognitive theories (what is the status of the theory, what are its structure and functions, how does the theory differ from the model, etc.), what type of scientific explanation is best suited for cognitive disciplines, as well as questions of attitude different cognitive disciplines are related to each other (in particular, whether some cognitive disciplines are reducible to others or not).

Thus, the last of the issues mentioned above has a fairly long history of discussions. At the beginning of the development of cognitive sciences, such disciplines as artificial intelligence, linguistics, philosophy and psychology played a leading role. In that era, cognitive disciplines were strongly influenced by the so-called computer metaphor and the concept of functionalism, which questioned the relevance of neuroscience for cognition research. However, great progress in the creation of new brain research tools and the development of connectionism in the field of artificial intelligence has led to a significant strengthening of the role of neuroscience as part of a complex of cognitive disciplines. The question began to be raised about the possibility of reducing such a higher-level discipline as psychology to neuroscience. Perhaps the apotheosis of modern reductionism was the work of the Churchland spouses already mentioned. My other work was devoted to this issue, in which I argued (having considered the main arguments of both Patricia Churchland ^[15] and other modern supporters of reductionism) that at the moment there are no prerequisites for reducing psychology to neuroscience ^[16].

In addition to the issues of philosophy of science briefly discussed above, philosophy is directly related to the discussion of the foundations of cognitive sciences and their key concepts of representation and computation. These concepts have served as the foundation for modern cognitive sciences since their emergence in the 1950s. Strictly speaking, cognition was largely understood as a calculation based on mental representations. However, the crisis of classical computational cognitivism, criticism from authors such as J. Gibson, opened the way for alternative projects in which an attempt was made to abandon these ideas when explaining cognitive processes. It seems that philosophy can play one of the main roles in discussions about the representational and computational nature of cognitive processes. So, in my other previous work, I tried to bring together the most important arguments of opponents of the representational understanding of cognitive processes and phenomena (the problem of the homunculus, the hypothesis of the world as its own model, the possibility of non-representational concepts of cognition). Having considered each of these arguments in detail, I came to the conclusion that the concept of representation retains its significance and leading role for cognitive research ^[17]. It is too early to talk about serious competition from non-representational theories of cognition.

At the same time, of course, the very concepts of computation and representation raise many questions. For example, if representation is a substitute, then are colors, smell, and taste represented? Or, how do the concepts of representations, computing and information processing relate? After all, the concepts of computing and information processing are also central concepts of cognitive sciences and are currently used by many cognitive scientists as synonyms. However, if we talk about the classical concepts of A. Turing and K. Shannon, which gave rise to the relevant literature and to which it is customary to refer, it immediately turns out that computing and information processing are by no means the same thing. The question of the relationship between the concepts of computing and information processing was discussed in detail in the works of G. Piccinini and his co-authors ^[18].

In addition, it seems that philosophy should play an active role in assessing the ethical consequences of the development of new cognitive technologies and neurotechnologies (in particular, in evaluating technologies such as mind reading and evaluating new applications in the field of artificial intelligence, etc.). So, over the past decade, a number of works on deciphering subjective experience based on data on its neural carriers (correlates). This area of research in neuroscience has become known as mind reading (sometimes, brain reading). Within the framework of this direction, scientists are trying to visualize the internal experiences of subjects on external media. As D.I. Dubrovsky notes in this regard, "With the help of new methods, a number of mental phenomena expressing a person's desires, assessments, thoughts are already being deciphered. However, while these are still separate and relatively simple mental phenomena, it is still far from deciphering mental activity in its integral context of the inner subjective world of the individual. But things are moving quickly towards this, and even now we are facing acute ethical and social issues (Dubrovsky, 2018). In whose hands will the technology of decoding brain codes end up? Who and why will open our inner world? After all, all social self-organization is based on the principle of relative closeness of the inner world of the individual, who selectively opens it to different people at will. Violation of this principle can lead to blatant inequality, the collapse of the entire system of social self-organization" ^{[19, p. 26]}.

In turn, specific cognitive disciplines can shed new light on traditional philosophical problems.:  the problem of consciousness and the brain, the problem of the subject, the problem of free will, etc. So, the results of specific empirical studies are able to give weight to certain philosophical concepts or make them more problematic. (For example, to once again show the problemativeness of Cartesian dualism, etc.) Under the influence of new trends in science and technology, new philosophical trends arise – the already mentioned functionalism in the philosophy of consciousness or neurophilosophy serve as a good example here.

Also, the analysis of the development of cognitive sciences is able to fill the philosophy of cognitive science with concrete material as a special section of the philosophy of science, contributing to the assessment and refinement of traditional concepts of the philosophy of science. Thus, such traditional topics for the philosophy of science as the problem of demarcation (and its significance for the philosophy of science), the homogeneity or heterogeneity of science, the falsifiability of individual scientific theories, etc. are under the scope of the response from the philosophy of cognitive sciences. For example, research in the field of philosophy of cognitive sciences can be an argument for/against the thesis of the separation of science and the approach to the problem of demarcation from the perspective of the idea of family similarities ^{[20, 21]}.

Conclusion

So, in this article an attempt was made to outline the main ways of interaction between philosophy and cognitive sciences. Starting from the point of view of V.A. Lectorsky about the dialogue of philosophy and specific cognitive disciplines, I argued that philosophy is able to make the greatest contribution to the development of cognitive research, firstly, on the way to clarify a number of important issues related to the conduct of philosophy of science (the question of the nature of major theoretical trends and individual theories in cognitive sciences, the question of on the theoretical progress of cognitive sciences, the problem of the relationship of cognitive disciplines to each other, etc.). Secondly, philosophy and philosophical analysis are at the center of discussions about the fundamental concepts for cognitive sciences - the concepts of representation and computation. (In particular, about the necessity of these concepts for cognitive research, their lack of alternatives, etc.) Thirdly, philosophy plays an important role in assessing the ethical consequences of the development of new cognitive technologies and neurotechnologies (such as mind reading technology, etc.). In turn, cognitive sciences can be (one of) the sources of new directions in philosophy (as it was with functionalism or neurophilosophy), they can shed new light on traditional philosophical problems and fill the philosophy of science with concrete material (specialists from other branches of philosophy could probably expand this statement). The result of the dialogue between philosophy and cognitive sciences can be the strengthening or weakening of both philosophical and scientific theories.

Despite the historically significant role of philosophy in the formation and evolution of cognitive sciences, the question of the place of philosophy in the cognitive disciplines is raised and discussed infrequently. Meanwhile, ignoring philosophy where the arsenal of its means can be especially useful to science, most often leads to implicit philosophizing, moreover, very mediocre ^{[2, p. 249]}.