[I]t would be absurd for us to hope that we can know more of any object than belongs to the possible experience of it or lay claim to the least knowledge of how anything not assumed to be an object of possible experience is determined according to the constitution that it has in itself.

* * *

It would be… a still greater absurdity if we conceded no things in themselves or declared our experience to be the only possible mode of knowing things….

[Kant, Prolegomena to Any Future Metaphysics]

A certain line of skepticism about normative epistemology has become more or less standard in contemporary philosophy of science. It runs like this.

Scientific method is a matter of choosing among theories on the basis of evidence.

But in “conceptual revolutions”, meaning, truth, and even what counts as observable can all be theory-relative.

The role of pictures and visual modes of presentation of data in science is a topic of increasing interest to workers in artificial intelligence, the psychology of problem solving, and increasing numbers of scientists in all fields who must deal with problems of how to represent large quantities of complex multidimensional data in an intelligible fashion. The use of pictures is marvelously illustrated by but not limited to the biological sciences, so I will use examples from elsewhere as appropriate. With the development of our visual technology—television, videotape, and the computer, the uses (and misuses) of visualization has properly become a matter not only of theoretical but also of practical concern. I will start with a practical story because it has multiple morals, both practical and theoretical—most of which lead elsewhere than I wish to go here.

Herbert Simon’s work on scientific discovery deserves serious attention by philosophers of science for several reasons. First, Simon was an early advocate of rational scientific discovery, contra Popper and logical empiricist philosophers of science (Simon 1966). This proposal spurred on investigation of scientific discovery in philosophy of science, as philosophers used and developed Simon’s notions of “problem solving” and “heuristics” in attempts to provide rational accounts of scientific discovery (See Nickles 1980a, Wimsatt 1980). Second, Simon promoted and developed many of the crucial techniques and methods used in cognitive science. One is the use of computers to model internal psychological processes, a technique central to his account of scientific discovery. Another is protocol analysis, the use of the verbal reports of experimental subjects in psychology to construct accounts of their psychological processes. Protocol analysis is given a detailed formulation by Simon (Simon and Ericsson 1984), and is modified for use in the study of scientific cognition in the paper on Krebs (Kulkarni and Simon 1988).

The obvious metaphysical differences between Newton and Leibniz concerning space, time, and motion reflect less obvious differences concerning the relation between geometry and physics, expressed in the questions: what are the invariant quantities of classical mechanics, and what sort of geometrical frame of reference is required to represent those quantities? Leibniz thought that the fundamental physical quantity was “living force” (mv2), of which every body was supposed to have a definite amount; this notion violates the classical principle of relativity, since it makes a physical distinction between uniform velocity and absolute rest. But Leibniz did not try to represent this physical quantity in a spatio-temporal reference frame, assuming, instead, that all such frames are equivalent so long as they agree on the relative motions (changes in the mutual Euclidean distances) among bodies.

A recent set of experiments by Cairns et al. (1988) purport to show that some mutations in a strain of E. coli bacteria are directed in the sense that they are induced, somehow, in an environment that is suited for them. If this claim is true it would limit the validity of the neo-Darwinian assumption that evolution proceeds by “random variation and natural selection” which requires that there be no correlation between the genesis of a particular mutation and the fitness of the corresponding phenotype in the environment in which the mutation occurs. Instead, it would argue for the occurrence of some “neo-Lamarckian” processes in evolution in the sense that the genesis of some mutations would have a positive correlation with the fitness of the resultant phenotype in that environment. Thus, one of the most fundamental tenets of the neo-Darwinian interpretation of evolution would be brought into question.

We have yet to fully understand the mariner or the measure to which logical empiricism emerged as a conventionalist response to both traditional Kantian and empiricist epistemology and to the apparent triumphs of “conventionalist stratagems” (in Popper’s aspersive locution) in the foundations of science. By “conventionalism”, however, is here understood a broader sense than customary, an extrapolation of views on the foundations of geometry and physics (associated in the first instance with Poincaré“) to an encompassing epistemological consideration of the development and validity of scientific concepts generally. In this new construal, the concepts of science are neither derivable from sense experience, nor are they transcendentally valid a priori conditions of its possibility. Rather they are “free creations of the human mind” whose provenance is “logically arbitrary”, as Poincare’ and (subsequently) Einstein put it.

Schlick has been accused of a number of philosophical sins over the years, most notably his rather casual, and frequent, traversing of the borders between language, experience, and reality. While we allow our scientists the freedom to roam creatively throughout the peripheral regions of Epistemology and Metaphysics, we are not so tolerant of our philosophers. We know that Schlick gave up the physics laboratory for the philosopher’s armchair, and we expect him to stick to a particular position.

Schlick’s colleagues in the Vienna Circle were not totally blameless in this regard, either, and it may be that part of the charm of their work is precisely their inability to stick to the positions that solidified over 2500 years of western philosophy.

Philosophers now treat the relationship between Classical Mendelian Genetics and molecular biology as a paradigm of nonreduction and this example is playing an increasingly prominent role in debates about the reducibility of theories ranging from macrosocial science to folk psychology. Patricia Churchland (1986), for example, draws an analogy between the alleged elimination of the “causal mainstay” of classical genetics and her view that today’s psychological theory will be eliminated by neuroscience. Patricia Kitcher takes an autonomous rather than eliminativist view of the reported nonreduction in genetics and reasons that psychology will retain a similar autonomy from lower level sciences (1980 and 1982). Although Churchland and Kitcher offer different interpretations of the apparent failure of molecular biology to reduce classical genetics, they agree that this failure will help illuminate theoretical relations between psychology and lower level sciences. The appearance of the Mendelian example along side the usual ones from physics and chemistry marks a turning point in philosophy of science.

Konrad Lorenz does not view behaviors as innate; he does not even regard differences among behaviors (of different species) as innate. Rather, he construes information (about the environment to which the behavior is adapted) as the innate component of (some) behavior. His noted deprivation experiments are intended to withhold environmental sources of that information from the organism: should the organism nevertheless exhibit behavior evidencing possession of such information, then that information must be innate. Lorenz interprets this conclusion to mean that the information is transmitted to the organism genetically, by way of a sort of blueprint in the genome.

This paper is a summary of my doctoral dissertation on philosophical interpretations of Einstein’s special and general theories of relativity, submitted to the Dept. for History of Science, Univ. of Hamburg, in 1989, which was recently published in the Series Science Networks at Birkäuser.2 After a brief overview of its content I will focus on a discussion of the method employed to analyse philosophical interpretations of a physical theory.

I take Newton’s arguments to inverse square centripetal forces from Kepler’s harmonic and areal laws to be classic deductions from phenomena. I argue that the theorems backing up these inferences establish systematic dependencies that make the phenomena carry the objective information that the propositions inferred from them hold. A review of the data supporting Kepler’s laws indicates that these phenomena are Whewellian colligations—generalizations corresponding to the selection of a best fitting curve for an open-ended body of data. I argue that the information theoretic features of Newton’s corrections of the Keplerian phenomena to account for perturbations introduced by universal gravitation show that these corrections do not undercut the inferences from the Keplerian phenomena. Finally, I suggest that all of Newton’s impressive applications of Universal gravitation to account for motion phenomena show an attempt to deliver explanations that share these salient features of his classic deductions from phenomena.

Selection operates at many levels. Some of the most obvious cases are organismic, such as changes in coloration under the influence of predation (cf. Kettlewell 1973; also Endler 1986). It also operates at other levels. Meiotic drive involves selection for a gene, independently of its effect on the organism. At a higher level, there may also be selection for patterns of colony growth in social insects, again under the influence of predation (cf. Wilson 1971). The appropriate level of selection is a matter of the causal patterns exhibited, or the target of selection. It can also be understood as a question of the appropriate level at which to seek explanatory laws.

Robert Brandon (1982, 1990) first clearly distinguished the question of the level of selection from debates over the units of selection. In doing so, he argued that the phenotype is commonly the target of selection, whatever the unit of selection might be.

There is, at present, controversy surrounding the role of the mathematical models which typify the more theoretical portions of ecology and evolutionary biology. Within these sciences there has been controversy about the “testability” of these models, both in terms of the ability of the model to make precise enough claims about the world, and in terms of our ability to determine the values of theoretical parameters. There has been concern, particularly in ecology, about the lack of realism characteristic of most models as well. These are, for the most part, issues involving the nature of theoretical explanation. Much of the current debate about null hypothesis testing, the unfalsifiability of theory, and the role of theory is symptomatic, it would seem, of the fact that these sciences have not settled on what to expect from their most mathematical theories. If there is one emerging theme, it is the need, again especially in ecology, for pluralistic theoretical treatments - which suggests that they may have been expecting too much.

On the model of Newton’s Principia, the great majority of successful new theories in physics have been introduced by deduction from the phenomena arguments. In such arguments an explanatory theory is deduced from one or more of the empirical facts, or lower level empirical generalizations, which it is designed to explain, by the device of adjoining suitable higher-level theoretical constraints on the form of the required theory. Those theoretical constraints leave certain parameters, the precise form of certain functions, and so on, in the new theory, undetermined, except with the help of the lower level empirical premises.

Cytogenetics was born of the confluence of two so-far independent fields: cytological studies and breeding studies, which merged through the identification of genes with chromosomes. In this paper I argue that genes were introduced as functional entities. Functional explanations are presented here as a subclass of inferences to the best explanation and I argue that abductive arguments do not offer conclusive proof for the existence of the entities postulated through them. However, functional explanations usually follow a scheme laid out by Fodor (1968): there is a first phase where hypothetical entities are postulated and individuated through their effects; there is a second phase where the physical structures responsible for these effects are sought. I analyze the development of both phases in the construction of the chromosome theory of Mendelian inheritance. I will argue that first-phase theories are important to set conditions of identification for the functions played by a certain structure in a given containing system.

There is a passage in Hume’s Enquiry concerning Human Understanding that I have always found striking and rather charming. It concerns a metaphysical theory that Hume regards as bizarre; and he offers two philosophical arguments in its confutation. It is the first of these that I have in mind:

Let us assume that some organisms, humans at least and the other higher animals, have internal states and behavioral states that represent things external to themselves. One of the questions that everyone would like answered about these states is: In virtue of what does such a representational state get the specific content that it has? An answer to this question that’s popular just now is: In virtue of its biological function. I believe there is a deep reason why such an answer can’t work. I shall present that reason in this paper.

First, we need to be clear about the notion of biological function that’s being appealed to here. In particular, we need to distinguish two fairly different ways of thinking about function. We could think of something’s function as what it contributes to the present functioning or output of some complex system of which it is a part. Roughly, this is Cummins’ view in (Cummins 1984).

David Hull’s approach to science, which culminated in his important book Science as a Process(1988), represents an unprecedented conjunction of philosophy of science with the results and concepts of a particular science. Hull takes an evolutionary approach to the conceptual development of science, importing much of his explanatory framework from comparative biology, the discipline where his empirical observations of scientists have been made. On the surface, such a cozy relationship between data, theory, and metatheory leads to worries about circular reasoning (Mishler 1989). Nevertheless, I will argue here that Hull’s approach is basically sound, and that the strength of his arguments comes precisely from his recognition of key analogies between the evolution of organisms and the conceptual evolution of scientists; however, certain disanalogies must also be taken into account.

Comparative biology can be divided into two distinct (although interrelated) halves based on differing orientations and types of questions asked.

Renormalization in Quantum Field Theory (QFT) has frequently been regarded, by philosophers as well as by scientists, as an exemplary case of bad methodological behavior. The feeling that renormalization was somehow an illegitimate way to extract results, an ad hoc maneuver without an independent rationale, was (and is) common among physicists and philosophers, who wonder, at the same time, about the unprecedented accuracy of the empirical results achieved by the illegitimate method. Teller (1989) has recently tried to dispel the air of illegitimacy around this technique. His lucid presentation, however, leaves one wondering why any sufficiently well-informed person could ever have thought of renormalization as an ad hoc move.

Part of the reason, I think, is that renormalization — or so the common view goes — came to the rescue of a sadly lingering theory, the pre-1947 Quantum Electrodynamics (QED). A theory with such a bad record of solved empirical and conceptual problems could simply not be correct.

David Hull’s book (1988) provides an evolutionary account of the development of science which pays attention to both the social and conceptual aspects of that process. Unlike most philosophers who only invoke Darwinian metaphors in a casual way, Hull takes the analogy between the biological evolution of species and the growth of scientific knowledge quite seriously and by providing abstract definitions of terms such as replicator, interactor and lineage, he makes it possible for us to see clearly the structural similarities between the two historical processes.

Other symposiasts will comment on how tight that analogy really is. I must remark in passing that I have never understood the intense interest which evolutionary epistemologists take in this comparison. Surely our major job is to understand how science works, perhaps by using evolutionary theory as a fallible heuristic, but nothing seems to hinge on the extent of the formal analogy.