Humans are animals but a very distinct and unique kind of animal. Our anatomical differences include bipedal gait and enormous brains. But we are notably different also, especially in our individual and social behaviors and in the products of those behaviors. With the advent of humankind, biological evolution transcended itself and ushered in cultural evolution, a more rapid and effective mode of evolution than the biological mode. Products of cultural evolution include science and technology; complex social and political institutions; religious and ethical traditions; language, literature, and art; and electronic communication.

In this chapter, I explore ethics and ethical behavior as a model case to illuminate the interplay between biology and culture. I propose that our exalted intelligence – a product of biological evolution – predisposes us to form ethical judgments, that is, to evaluate actions as either good or evil. I further argue that the moral codes that guide our ethical behavior transcend biology in that they are not biologically determined; rather, they are products of human history, including social and religious traditions.

HUMAN ORIGINS

Mankind is a biological species that has evolved from species that were not human. Our closest biological relatives are the great apes and, among them, the chimpanzees and bonobos, who are more closely related to us than they are to the gorillas, and much more than they are to the orangutans.

The concept of pathology has a built-in normative character. An individual (or a state) is pathological if and only if it is not normal, that is, if it fails to be as an individual (or a state) of that sort ought to be. In other words, a pathological individual (or state) is an individual (or state) that does not conform to the standards to which all the individuals (or states) falling under its very sortal concept must conform.

When applied to mental disorders, this view entails that the concept of psychopathology has a normative character. A human being is psychopathic if and only if he fails to have the mental capacities that are normal for humans, that is, the mental capacities that humans ought to have. Human specific mental capacities include emotional responses to the environment, as well as perceptual and cognitive abilities. Let us call this the “received view of psychopathology.”

The received view has two main implications concerning psychiatry. The first implication is that psychiatry is ethically relevant, in two respects. First, the results of psychopathology may help to define what is normal for humans, and thus psychiatry may help to determine ethical norms. Second, the psychiatric treatment of mental disorders rests on various assumptions concerning what is normal for humans, and thus psychiatry has profound ethical bearings.

The second implication of the received view of psychopathology is what we could call the “universal treatment thesis.”

INTRODUCTION

Crucial pieces of the mechanism of protein synthesis were discovered by biochemists and molecular biologists in the 1950s and 1960s. At the outset, the approaches of these fields were very different, focusing on different components and finding different aspects of the mechanism from different ends. By about 1965, the results from the different approaches were integrated. The scientific work leading to this integration reveals general strategies for discovering mechanisms.

This instance of interfield integration, like many other discovery episodes in the biological sciences, crucially involves discovering a mechanism. Focusing centrally on mechanisms provides new ways of thinking about discovery, interfield integration, and reasoning strategies for scientific change. Philosophers of science have separately analyzed scientific discovery, interfield relations, mechanisms, and reasoning strategies. This chapter brings together these disparate topics. A unified approach yields reasoning strategies in discovering mechanisms that integrate results from different fields.

Many philosophers of science (e.g., Popper 1965) have been skeptical about finding methods for reasoning in discovery. Even those who have had much to say about scientific change (e.g., Kuhn 1962; Laudan 1977; Kitcher 1993) have not even discussed reasoning strategies for discovering new paradigms, traditions, or practices. Nonetheless, a few philosophers have worked on discovery (e.g., Nickles, ed. 1980a, 1980b; Meheus and Nickles, eds., 1999). Reasoning in discovery is a more tractable problem if discovery is viewed as an extended process of construction, evaluation, and revision (Darden 1991).

INTRODUCTION

Anomalies often challenge biological generalizations. Among the generalizations of widest scope in biology are natural selection and the central dogma of molecular biology. Because natural selection often produces novel variants, it might be expected that wide-scope generalizations would be infrequent (other than the theory of natural selection itself). Furthermore, biological regularities are evolutionarily contingent (Beatty 1995); whatever regularity has evolved can evolve away. Nonetheless, the central dogma of molecular biology is one of the most general findings in all of biology. It provides a schema for protein synthesis that involves unidirectional information flow from nucleic acids to proteins but not back. If information does not flow back into the genetic material, then there seems to be no mechanism for the inheritance of adaptive acquired characters. Although adaptive characters might be acquired during the life of one organism, their inheritance requires a change in the genetic material passed to the next generation. Because NeoLamarckian mechanisms require such inheritance, the lack of backwards flow of information strengthens the case for NeoDarwinian natural selection as the account of adaptations against any version of NeoLamarckism. The contemporary NeoDarwinian theory of natural selection claims that mutations arise spontaneously; that is, they are produced independently of their fitness in a given environment.

Anomalies challenging generalizations of such wide scope get attention. In a paper in 1988, John Cairns and colleagues claimed to find a new class of mutations in bacteria, called “directed mutations” and later “adaptive mutations.”

INTRODUCTION

The synthetic theory of evolution is a multilevel theory that serves to synthesize knowledge from fields at different levels of organization. It provides a solution to the problem of the origin of species. Biologists attempted to solve this problem for years, during which time the key fields emerged and developed to the point that the synthesis was possible. Prior to the synthesis, debate occurred about what components were necessary to solve the problem and alternative theories were proposed. Had any of the prior theories been correct, then the fields that exist within evolutionary studies would have been different. Which fields exist is a contingent fact about the nature of the world and the way we study it. Which fields are synthesized to solve certain problems is contingent on the nature of the solution and the stage of development of various fields at the time the solution is proposed.

Relations among levels of organization, fields of study, and the problem of the origin of species are the subjects of this chapter. The focus is on the synthetic theory of evolution as proposed by Theodosius Dobzhansky in 1937, the multiple levels within it, the knowledge from different fields that it synthesized, and its various predecessors. The final section discusses the concept of a synthetic theory and contrasts that with previously studied interfield theories.

THE SYNTHETIC THEORY

Theodosius Dobzhansky published the seminal book outlining the evolutionary synthesis in 1937: Genetics and the Origin of Species.

INTRODUCTION

In many fields of science what is taken to be a satisfactory explanation requires providing a description of a mechanism. So it is not surprising that much of the practice of science can be understood in terms of the discovery and description of mechanisms. Our goal is to sketch a mechanistic approach for analyzing neurobiology and molecular biology that is grounded in the details of scientific practice, an approach that may well apply to other scientific fields.

Mechanisms have been invoked many times and places in philosophy and science. A key word search on “mechanism” for 1992–1997 in titles and abstracts of Nature (including its subsidiary journals, such as Nature Genetics) found 597 hits. A search in the Philosophers' Index for the same period found 205 hits. Yet, in our view, there is no adequate analysis of what mechanisms are and how they work in science.

We begin (Section 1.2) with a dualistic analysis of the concept of mechanism in terms of both the entities and activities that compose them. Section 1.3 argues for the ontic adequacy of this dualistic approach and indicates some of its implications for analyses of functions, causality, and laws. Section 1.4 uses the example of the mechanism of neuronal depolarization to demonstrate the adequacy of the mechanism definition. Section 1.5 characterizes the descriptions of mechanisms by elaborating such aspects as hierarchies, bottom-out activities, mechanism schemata, and sketches.

Philosophers of science have had relatively little to say about theory construction. Theories were treated by Popper (1965) and the logical empiricists (e.g., Hempel 1966) as if they arose all at once by a creative leap of the imagination of a scientist, a process whose study was viewed as the province of the psychologist. Only after the creative leap, they agreed, were the philosopher's logical tools useful to evaluate the theory so produced. Even more historical accounts concerned with scientific change, such as Kuhn's (1970), did not discuss the way paradigms or the theories within them were constructed, except that they somehow arose in response to anomalies of their predecessors. Lakatos (1970), who proposed criteria for evaluating progressive research programs, did not discuss how the scientist constructs the program originally, although later additions, he claimed, resulted in some way from the “positive heuristic.” Even Laudan (1977), who made much of the commonplace observation that science is a problem-solving activity, focused on the use of solved or unsolved problems to evaluate theories and research traditions; he did not provide an analysis of how a scientist goes about solving a problem. Thus, most of twentieth-century philosophy of science, from the logical empiricists to the most recent work, has been within the context of justification, not the context of discovery.

Dichotomizing science into these mutually exclusive contexts and concentrating on justification to the exclusion of discovery distorts the ongoing process that characterizes science.

INTRODUCTION

This chapter discusses representation of scientific theories and reasoning in theory change. A scientific theory may be represented by a set of concrete exemplary problem solutions. Or, alternatively, a theory may be depicted in an abstract pattern, which when its variables are filled with constants becomes a particular explanation. The exemplars and abstractions may be depicted diagrammatically, as they are in the cases from Mendelian and molecular genetics to be discussed. One way that a theory grows is by adding new types of exemplars to its explanatory repertoire. Model anomalies show the need for a new exemplar; they turn out to be examples of a typical, normal pattern that had not been included in the previous stage of theory development. A special-case anomaly indicates the need for a new exemplar or abstraction, but it has a small scope of applicability. Thus, our subjects here are exemplars, abstractions, diagrammatic representations, and anomalies and the roles they play in the representation of explanatory theories and in the change of such theories. Examples will be taken from Mendelian and molecular genetics.

EXEMPLARS, ABSTRACTIONS, AND DIAGRAMS

Thomas Kuhn (1970, 1974) discussed the importance of “exemplars,” which he characterized as concrete problem solutions in which a formalism is applied and given empirical grounding. Kuhn said that exemplars are taught by the use of problems in textbooks.

INTRODUCTION

This chapter discusses abstract characterizations of selection theories. Finding such abstractions is a task in a larger research program, which is based on the assumption that some scientific theories are representative of types of theories that solve types of problems. Natural selection, clonal selection for antibody production, and selective theories of higher brain function are examples of selection type theories. Selection theories solve adaptation problems by specifying a process through which one thing comes to be adapted to another thing.

When Darwin elaborated his theory of natural selection in 1859, he provided a new type of theory for explaining adaptation problems. Others, such as Burnet (1957) in immunology, argued by analogy from Darwinian natural selection for selection processes in other fields. One analysis of analogy is that two analogues share a common abstraction (Genesereth 1980). Thus, an analysis of natural selection and its analogues aids in the development of an abstraction for selection theories. The selection literature in philosophy of biology includes several discussions of natural selection that provide help in this task. The abstraction for natural selection extracted from these discussions will then be used in analyzing clonal selection for antibody production in immunology and selective theories for higher brain function in neurobiology. Before beginning the analysis of selection theories, however, we will briefly discuss the concept of an abstraction.

THE CONCEPT OF AN ABSTRACTION

An abstraction provides a schematic outline that can be filled in or “instantiated” to give an actual theory.

INTRODUCTION

Understanding the growth of scientific knowledge has been a major task in philosophy of science. No successful general model of scientific change has been found; attempts were made by, for example, Kuhn (1970), Toulmin (1972), Lakatos (1970), and Laudan (1977). A different approach is to view science as a problem-solving enterprise. The goal is to find both general and domain-specific heuristics (i.e., reasoning strategies) for problem solving. Such heuristics produce plausible but not infallible results (Nickles 1987; Thagard 1988).

Viewing science as a problem-solving enterprise and scientific reasoning as a special form of problem solving owes much to cognitive science and artificial intelligence (AI) (e.g., Langley et al. 1987). Key issues in AI are representation and reasoning. More specifically, AI studies methods for representing knowledge and methods for manipulating computationally represented knowledge. From the perspective of philosophy of science, the general issues of representation and reasoning become how to represent scientific theories and how to find strategies for reasoning in theory change. Reasoning in theory change is viewed as problem solving, and implementations in AI computer programs provide tools for investigating methods of problem solving. This approach is called “computational philosophy of science” (Thagard 1988).

Huge amounts of data are now available in online databases. The time is now ripe for automating scientific reasoning, first, to form empirical generalizations about patterns in the data (Langley et al. 1987); then, to construct new explanatory theories; and, finally, to improve them over time in the light of anomalies.

INTRODUCTION

This chapter is about discovery in neurobiology; more specifically, it is about the discovery of mechanisms. The search for mechanisms is widespread in contemporary neurobiology, and, understandably, the character of this product shapes the process by which mechanisms are discovered. Analyzing mechanisms and their characteristic organization reveals constraints on their discovery. These constraints reflect, at least in part, what it is to have a plausible description of a mechanism. These constraints also highlight varieties of evidence that both guide and delimit the construction, evaluation, and revision of such plausible descriptions.

The central example in the following discussion is the continuing discovery of the mechanism of spatial memory. Spatial memory, roughly speaking, is the ability to learn to navigate through a novel environment. The mechanism of spatial memory is multilevel, and recently an integrated sketch of the mechanism at each of these levels has started to emerge. Even though this sketch is far from complete at this time, the example offers a glimpse at the kinds of constraints that are delimiting and guiding this gradual and piecemeal discovery process.

This chapter opens with an analysis of mechanisms, discussing their components and their characteristic spatial, temporal, and multilevel organization. Mechanisms are often discovered gradually and piecemeal. The second section introduces conventions for constructing incomplete and abstract descriptions of mechanisms (namely, the mechanism sketch and the mechanism schema) and for describing the constraints under which the gradual and piecemeal discovery of these sketches and schemata proceeds.

INTRODUCTION

Interactions between different areas or branches or fields of science have often been obscured by current emphasis on the relations between different scientific theories. Although some philosophers have indicated that different branches may be related, the actual focus has been on the relations between theories within the branches. For example, Ernest Nagel has discussed the reduction of one branch of science to another (1961, Ch. 11). But the relation that Nagel describes is really nothing more than the derivational reduction of the theory or experimental law of one branch of science to the theory of another branch.

We, in contrast to Nagel, are interested in the interrelations between the areas of science that we call fields. For example, cytology, genetics, and biochemistry are more naturally called fields than theories. Fields may have theories within them, such as the classical theory of the gene in genetics; such theories we call intrafield theories. In addition, and more important for our purposes here, interrelations between fields may be established via interfield theories. For example, the fields of genetics and cytology are related via the chromosome theory of Mendelian heredity. The existence of such interfield theories has been obscured by analyses such as Nagel's that erroneously conflate theories and fields and see interrelations as derivational reductions.

The purpose of this chapter is, first, to draw the distinction between field and intrafield theory, and, then, more importantly, to discuss the generation of heretofore unrecognized interfield theories and their functions in relating two fields.