Taking the History of Science Really Seriously

By Scott A. Kleiner

This book is intended to offer a historical, conceptual and methodological analysis of selected iconic shifts (revolutions) in scientific belief and practice. Part I is a reexamination of The Copernican Revolution in the light of philosophical thinking about the history of science that has emerged in the last 60 years since its publication in 1959 (Kuhn I). The philosophy and historiography drawn from Part I is then sought and compared with that evident in the history of evolutionary inquiries (Part II) and genetics (Part III). Since turning my focus to the Philosophy of Biology in the early 1970's, I have sought studies in its history with the objective of articulating a theory of change and discovery with the guidance of epistemic values realized in inquiries (questioning strategies) in the physical sciences (Kleiner, 1968, 1970).
Particularly evident in the three disciplines here studied are the trade-offs between those willing to sacrifice realism (ontic comprehensiveness) for mathematical fruitfulness (instrumentalism). I was surprised to learn from Rosenberg's Instrumental Biology that there was instrumentalism in biology, because wasn't it obvious that Darwinian histories (Kitcher, 1984) had to be fully realistic to be causally efficacious? in explaining phylogeny? But, of course, R.A. Fisher's fruitful program articulated in articles (1915-1921) and a book in 1930 was instrumental biology. As with Galileo, Kepler and Newton, the pursuit of the goal of enhancing the realism of mathematical models was a driver of discovery as the Copernican Revolution unfolded (Kleiner, 197?), but also as the Darwinian-Mendelian revolutions unfolded. I thus advocate a heuristic and qualified realism, not the unrealizable 'literal' realism brought up in recent philosophical debates (Laudan, 1986)
Newton settles for an ontology that gives progressively epistemically accurate models only up to a point, the point at which empirical resolution is technically limited and, as we now know, causal determinativeness declines into chaos (Randall, 2014). Mendelism and evolutionary theory has a similar escalation of realism and indeterminism from linkage, adding chromosomes to the ontology (late 19th century cytology), to karyotypes and sub-karyotypes, to subpopulations emerging within genetic populations, to cells, developing biological individuals, reproductive and symbiotic groups, genes, their sequences, enzymes and theirs, regulators, enhancers and repressors, major transitions (Michod, 1999), and comparative developmental studies (Raff, 1996).
Mendelian populations were initially studied with single-locus models and were, like the Laplacian Ideal, fully deterministic in their etiology (at least in Fisher's ideal infinite populations). Introducing linkage in computer models of the Mendelian reproductive cycle gives rise to chaos, where small variations of allelic fitnesses produce diverging evolutionary trajectories. The same outcome occurs with the proliferation of small Mendelian isolates, where sampling error generates sampling error and disparity among these isolates.

• Language: English
• Publisher: BookBaby
• Release date: Oct 15, 2020
• ISBN: 9781098313975

Book preview

History of Science Really Seriously

© 2020 by Scott A. Kleiner

All rights reserved. This book or any portion thereof may not be reproduced or used in any manner whatsoever without the express written permission of the publisher except for the use of brief quotations in a book review.

ISBN (Print): 978-1-09831-396-8

ISBN (eBook): 978-1-09831-397-5

Part I

A Tale of Three Kuhns

Contents

Preface: An Autobiographical Note

Chapter I.1:Kuhn I: The Copernican Revolution (1959)

Chapter I.2:Questions and Inquiry

Chapter I.3:What is a scientific revolution?

Chapter I.4:Kuhn II: The Structure of Scientific Revolutions

Chapter I.5:The Copernican Revolution

Appendices

Bibliography

Preface

An Autobiographical Note

In an intermediate graduate level course in philosophy of science taught by Dudley Shapere in the Fall of 1963, I was introduced to the recently published Structure of Scientific Revolutions by Thomas S. Kuhn. In the previous year I had taken two quarters of Robert Palter’s history of science, which covered ancient Greek astronomy, Copernicus, Galileo, Kepler, Newton, Huygens, Faraday and Maxwell, among others. In the history class, I was introduced to Kuhn’s Copernican Revolution (Kuhn I; 1959) from which I have taught ever since. I was thus well prepared for what I considered a breath of fresh air, after being immersed in Ernest Nagel’s iconic Structure of Science, early volumes of the Minnesota Studies in the Philosophy of Science and other sources for the ‘received view’ of the philosophy of science, then and currently known as Logical Empiricism. Shapere’s critical review of the Structure came out in the Philosophical Review in 1964, from which I hesitated in becoming a Kuhn (Kuhn II) disciple. However, I have been gratified to see valuable elements of Kuhn II, particularly the appreciation of now obsolete scientific theories, re-emerge in the writings of Mary Hesse, Imre Lakatos, Dudley Shapere, Philip Kitcher, Ernan McMullen, Lindley Darden, Larry Laudan and others.

The recent festschrift celebrating 50 years since the first edition of the Structure (Daston, L. et. al., 2014) focuses entirely on Kuhn II with no mention of Kuhn I (from The Copernican Revolution, 1959) or Kuhn III (Rationality, Objectivity, and Scientific Reason, a chapter in The Essential Tension, 1977). Errol Morris’ The Ashtray (2016), though dramatically readable, gives an equally one-sided view of Kuhn’s philosophy of science. In the studies that follow I shall argue that this rendition of scientific realism will not stand up to philosophical or historical scrutiny.

Since Kuhn II is the best known of his work (witness the near universal usage of paradigm) I shall begin here with an exposition of outstanding points in Kuhn II before discussing important differences between it and Kuhn I and Kuhn III. I will then continue with an analysis and epistemic appraisal of the principal episodes documented in Kuhn I, which covers a historical process sustained for nearly 1400 years from the homocentric sphere ‘paradigm’ of the fourth century B.C.E to the reception of Newton’s Principia Philosophica Naturalis in the late 17th century C.E.

In the introduction to his Structure Kuhn claims that theories now regarded as falsified or obsolete, e.g. geocentric planetary astronomy, have many of the features that we value in more recent and contemporary scientific research programs. These include a core of ontological assumptions and prescribed research activities, and a sustained effort toward generating and interpreting observations of the subject matter at hand, and consensus among empathetic and cooperating members of a linguistic community. Though the celestial sphere is no longer part of post Cartesian cosmology, it remains an illusion generated by the earth’s rotation. Unlike a fantasy or socially- constructed legend, it can be valued as a frame of reference for observation and a guide to terrestrial navigation. Though planetary spheres, epicycles, eccentrics and equants were banished in the Keplerian revolution, the geometry and ray optics by which their su-perimposed rotations could represent observed planetary trajectories remain in the optical toolbox for much more recent scientific endeavors.

The novice student of the history and philosophy of science should be reminded repeatedly that obsolete scientific theories are not unscientific because of their apparent errors or metaphysical content but can be epistemically legitimate products of research practices under historically imposed technical and epistemic constraints. The phenomenologists’ strategy of bracketing out all conceptual constructs could be applied in instructing students not to view the past history of science through the bias of the contemporary scientific cannon. However, this historiographical heuristic must be used with care so not to endorse the epistemic, conceptual and ontological relativism implicit in viewing paradigms as isolated but authoritative episodes in a chronology of science.

The history of science has vastly enriched my appreciation of the history of philosophy, particularly from the writings of Plato, Aristotle, medieval Neo-Aristotelians, Keplerian Neo-Platonism, Galileo’s Aristotelianism, Bacon, Whewell and Mill, Newton and Locke, Hume and Kant, Mach and Einstein, and much else. My appreciation of the history of science has also been enriched more recently by Imre Lakatos’ ‘Methodology of Scientific Research Programmes’ (1970), Dudley Shapere’s Galileo (1974, David Hull’s Philosophy of Biological Science (1974), his Science as Process (1986), Michael Ruse’s The Darwinian Revolution (1975), William Wimsatt’s theory of heuristics (2007), Larry Laudan’s Science and Values, Lindley Darden’s Scientific Change (1999), Philip Kitcher’s The Advancement of Science (1993), and work on the logic of questions highlighted by Jaakko Hintikka.

This book is intended to offer a historical, conceptual and methodological analysis of notable shifts (revolutions!?) in scientific belief and practice. Part I is a reexamination of The Copernican Revolution in the light of philosophical thinking about the history of science that has emerged in the last 60 years since its publication in 1959. The philosophy and historiography drawn from Part I is then sought and compared with that evident in the history of evolutionary inquiries (Part II) and genetics (Part III). Since turning my focus to the Philosophy of Biology in the early 1970’s, I have sought studies in its history with the objective of articulating a theory of change and discovery with the guidance of some epistemic values realized in the physical sciences.

Particularly evident in the three disciplines here studied are the trade-offs between those willing to sacrifice realism (ontic comprehensiveness) for mathematical fruitfulness (instrumentalism). I was surprised to learn from Rosenberg’s Instrumental Biology that there was instrumentalism in biology, because wasn’t it obvious that Darwinian histories (Kitcher, 1984) had to be fully realistic to be causally efficacious in explaining phylogeny? But, of course, R.A. Fisher’s fruitful program articulated in articles and a book in 1930 was instrumental biology. As with Galileo, Kepler and Newton, the pursuit of the goal of enhancing the realism of mathematical models was a driver of discovery as the Copernican Revolution unfolded, but also as the Darwinian-Mendelian revolutions unfolded. I thus advocate a heuristic and qualified realism, not the unrealizable ‘literal’ realism brought up in recent philosophical debates.

Newton has to settle for an ontology that gives progressively epistemically accurate models only up to a point, the point at which empirical resolution is technically limited and, as we now know causal determinativeness declines into chaos (Randall, 2014). Mendelism and evolutionary theory has a similar, escalation of realism from linkage, adding chromosomes to the ontology (late 19th century cytology), to karyotypes and sub-karyotypes, to subpopulations emerging within genetic populations, to cells, developing biological individuals, reproductive and symbiotic groups, genes, their sequences, enzymes and theirs, regulators, enhancers and repressors, and much else.

Mendelian populations were initially studied with single-locus models and were, like the Laplacian Ideal, fully deterministic in their etiology (at least in Fisher’s ideal infinite populations). Introducing linkage in computer models of the Mendelian reproductive cycle gives rise to chaos, where small variations of allelic fitnesses produce diverging evolutionary trajectories. The same outcome occurs with the proliferation of small Mendelian isolates, where sampling error generates disparity among these isolates.

What is the book about? Rationality and Objectivity in the history of physical (planetary astronomy and classical physics) and biological sciences (evolution and evolutionary genetics).

The neglected story of interest begins in Kuhn’s Copernican Revolution (1959) which features a history with epistemically and semantically connected sub-episodes. These are not apparent in the Structure (1962, 1970), whose episodes are scattered among several fields and historical periods. Internal epistemic values favored in the Structure raise the specter of epistemic relativism and social conventionalism, a reductionistic and confused philosophy of science. Transcen*dent and multiple epistemic values, as suggested in the ‘Rationality, Objectivity and Scientific Change’ (1977) (but also by Craver, 2002) open the possibility of paradigm change justified by indisputable success in one dimension of valuation, and trade-offs and compromises among several. The case for Copernicanism is inconclusive in 1550 with instrumentalism compromising realism, but a form of realism prevails and progresses in Kepler and Newton. R.A.. Fisher’s Mendelian populations are fictions, but the single locus models formulated therein are highly successful in solving problems faci*ng Darwin’s Natural Selection. The organism, the interactor, was a fictional aggregate (mosaic) of Mendelian characters that gradually became more realistic and coherent with the addition of linkage, karyotypic dynamics, developmental demographics, developmental ecology and genetics.

The three phases of Kuhn’s philosophical career are the inspiration for this work, but it is not about Kuhn, but rather about scientific change, its progression, progress and rationalization. It is a theory of scientific discovery: Discoveries are, as Kuhn says in his 1959, answers to interesting questions. But what is a question? What makes a question interesting? ‘Inquiry’ appears in the title of many works in the philosophy of science, but I have found little sustained inquiry about inquiry.

This book makes extensive use of primary sources in the history of science, particularly extensive quotations from Copernicus, Galileo, Kepler, Newton, Lyell and Darwin, but also some research papers in evolutionary genetics available online from jstor.org. Biographical sketches help in establishing a historical setting for iconic research programs in the history of a discipline, but I offer only occasional sketches where I deem them relevant, not whole biographies. Scientists have different personalities, some content to follow prescribed ‘normal heuristics’, the orthodoxy, others wishing to inquire outside the bounds of orthodoxy; so, personal histories of prominent participants are relevant to illuminating the ‘revolutionary research programs’, that, by my contention, mark the important turning points in the history of science.

Darwin’s ‘use and disuse’ theory, his theory of the evolution of the eye, of eusocial organisms and his naturalistic taxonomy have not received the attention they are due. Dobzhansky’s (and some of his students’) research program is hardly an extension of orthodox ‘New Synthesis’ ideology (Cf. Michod, 1981), but embraces the promise of karyotypic evolution. Late in his career Dobzhansky considers experiments on the mobility of individual pseudoobscura flies as possible ecological grounds for dividing Fisher‘s populations into the finite and divergent peripherals of Wright and Mayr.

This book reflects fifty years of teaching The Copernican Revolution, which I consider to be an accessible introduction for advanced undergraduates and graduates at all levels to the history and philosophy of science. As did Kuhn, I offer it as an exemplar of how to ‘bracket out’ the biases of recent science and consider empathetically past paradigms, with all their seemingly crazy metaphysics and limited empirical access, as credible programs of research.

Chapter I.1

Kuhn I: The Copernican Revolution (1959)

A historiography of science should embody a criterion for demarcating science from non-science through a historical timeline. The timeline represents a history of one or more converging or diverging scientific disciplines. The timeline contains a succession of episodes and various cognitive, ontological and methodological commitments. One such historiography, subscribed to more by scientists than historians, chronicles and describes the most recently successful scientific enterprises but dismisses all or most seemingly contrary predecessors as unscientific, called ‘Whiggism’ (Butterfield, 1931), or, more descriptively, chauvinism of the present.

In the introductory essay to the Structure (1962) Kuhn notes his surprise at finding that Aristotle’s natural philosophy of the Cosmos, Earth and Planets meets most if not all requirements for scientific status. Immediate observations of these objects and their movements are cited in arguments for a two-sphere cosmos and for planetary motions. Empirical constructs such as methods for recording solar locations and motions, for identifying and re-identifying planets and stars, for setting up terrestrial and celestial coordinate systems are an essential part of the 4th century research program. An epistemic objective offered in Plato’s Republic prescribed ‘saving’ seemingly chaotic appearances (wanderings) by mathematical constructs analogous to the Pythagorean theory of harmonies.

Nevertheless, Kuhn’s sampling of the history of science is episodic and scattered among several scientific disciplines, including electricity, chemistry, relativity and classical mechanics, as well as planetary astronomy. The overall picture of the history of science offered in the Structure can be compared to punctuated equilibrium in evolutionary theory. The history of a scientific discipline is punctuated with a succession of wholly distinct scientific paradigms that endure for various bounded time intervals and are subject to empirical challenges only within their local conceptual and deontic (valuational) environment. Another historiography with an evolutionary analogy would embody Darwinian gradual and cumulative progression, tree-like branching of disciplines with some reticulation (crossing of branches), with some lineages going extinct and others conserved and supplemented, absorbed or ramified by successors. These two historiographies are not necessarily dichotomously distinct: punctuated equilibrium allows species to be linked by a succession of descendants and founders which may not be evident in the fossil record. Darwinian gradualism allows episodic and lasting innovations, founders of clades, as well as partial and wholesale deletions. In sum, the historical progression of science can be sketched as a ramifying tree with some reticulate branches. (Cf. Kuhn, 1962, XIII)

Hellenic and Hellenistic Planetary Astronomy: The Paradigm of Paradigms

On p.90 in the Structure of Scientific Revolutions (first edition, 1962) Kuhn describes paradigms as punctuating history with wholly distinct variations of several components, ontology, method and empirical access. These components are aptly illustrated in the homocentric sphere cosmology of Plato and Aristotle. It is evident, from Kuhn’s description of his ‘Aristotle experience’, that the homocentric universe was the exemplar for Kuhn’s theory of paradigms in the Structure. The overall positive heuristic (Lakatos, 1970) for planetary astronomy was formulated by Plato: replace the whole ‘wandering’ planets appearing to immediate experience with Pythagorean mathematical constructs that demonstrate the planets’ hidden regularities, i.e. ‘save’ the appearances. Plato’s pedagogy of three disciplines, mathematics, music and astronomy embodies the belief that mathematics is the way to the perception of the Good, the highest order of timeless form embodying precision, functional coherence and universality (Timaeus, Republic).

Paradigms contain, or exemplify for research programs) these components:

A world view or ontology: The two-sphere universe defines a ‘space’ within which celestial motions occur, bounded by the celestial sphere and the spherical and concentric earth. Celestial spheres undergo timeless uniform rotation from the moon’s orbit beyond. According to Aristotle, the four Empedoclean terrestrial ‘elements’ have distinctive natural motions of gravity and levity in radial earth-centered directions only. They are thus substantively distinct from a fifth element composing celestial objects, such as the moon and the sphere carrying it. The lunar dichotomy distinguishes two domains of motion and substance: (i) The celestial contains uniformly rotating geocentric spheres and is timeless (eternal in motion and substance). (ii) The terrestrial, which is subject to a multiplicity of episodes of generation and corruption, forced as well as natural motions, tripartite ‘souls’ (activities: vegetative, animative and cognitive), substantial transmutation and chance encounters.

Geometrical optics holds in both celestial domains: Rectilinear rays, perspective theory, image and shadow projection, is central to the empirical arguments for the two-sphere universe as they appear from Aristotle (his theory of the rainbow in De Caelo) through Ptolemy and beyond. (M.A. Smith, 2014)

Overall research objectives exemplify characteristic epistemic values natural and forced motions, final, material and efficient causes added to Plato’s geometrical and numerical ideals. They are codified in problems and expressed in questions addressed to the community. Observational ‘appearances’ should be ‘saved’ with geometrical constructions using homocentric spheres with axes adjusted post hoc at various angles and at various rotational speeds. These objectives are in accord with Plato’s program of reducing celestial wanderings to kinematic (geometry of motion) regularity.

In Aristotle’s cosmology terrestrial motions are to be explained by natural life cycles of kinds (animals and plants) and radial natural locomotion for the four Empedoclean elements. Horizontal trajectories are ‘forced’, caused and sustained, by external applied forces. Water and Earth can be forced upward against natural gravity, as Fire and Air can be constrained downward against their natural levity. The terrestrial objects we encounter are mixtures of various proportions of these elements, and their transmutations explain various geological and meteorologyical phenomena, such as volcanism, rain, springs, etc.

Preferred research problems include differences between theoretical expectations and observed planetary orbits (empirical anomalies). Plato’s motion of the ‘same’ (diurnal rotation) versus various motions of the ‘other’ (eastward motions unique to each planet) are an initially formulated constraint on solutions to these problems (Republic). Kuhn calls these puzzles, and are problems, whose possible solutions are specified by the paradigmatic ontology and research objectives. Puzzle solutions are constrained by a negative heuristic, which protects core beliefs, and can be prioritized by positive heuristics that guide further inquiry in the paradigm (Lakatos, 1970, Wimsatt, 1980). The formulation and pursuit of solutions to well defined puzzles are what Kuhn calls normal science.

Preferred empirical procedures: Empirical rules and heuristics, instruments, reliable informational processes are drawn from geometrical optics and perspectival construction, and from Aristotelian mechanics. In the Structure Kuhn embraces Gestalt perception, a holistic but ambiguous cognitive response to sensory inputs. Gestalt perceptions are legitimately employed for identifying and reidentifying and naming stars and galaxies, as the constellation Ursa Major, has always been used for identifying and re-identifying Polaris. Such methods remain embedded in the names we give to celestial objects, e.g. Alpha Centaurus.

Theoretical problem-solving protocols (Cf. Lakatos’ heuristics, 1970): Resolve an anomaly by adding a concentric sphere or post hoc adjustment of inclination and rotation for a match with quantitative and qualitative features of observed planetary ‘wanderings’. The imperative to conserve embedded uniformly rotating spheres concentric with the earth, is a negative heuristic prioritizing the retention of core cosmological assumptions.

Comment: Heuristics are guides or strategies that facilitate but do not guarantee the achievement of a desired goal. Possible paths to a goal, the location of a desired object, can easily exceed limitations of human activity, so we use selective constraints or biases to narrow these possibilities to some that we, with our current powers, can implement. A bias can contain information as where the object can be found or how it might appear in various circumstances. Biases should be useful in placing the goal within our powers of search, but they can also misdirect us. The object can be mistakenly conceived or nowhere near where we think it is. Certain questions contain such biases, those that explicitly describe (constrain) possible answers, without fully divulging what we want to achieve. Paradigms or research programs prescribe such constraints for a scientific research tradition: The five geocentric planets must have their orbits represented as a superposition of uniform circular motions. Heuristics are distinct from rules in that rules allow no violations: The compound circles give way to heliocentric ellipses. The quintessential planet, the Moon, becomes a terrestrial projectile with an escape velocity. Heuristics also have limited domains of application within a discipline but can also be violated or suspended (Cf. Feyerabend’s Against Method, 1974; Kleiner, 1979). Heuristics are subject to ‘pragmatic’ justification, that is, evaluation in terms of epistemic values promised and mitigated in the light of other such values (Kuhn III, Wimsatt, 1980). Lakatos’ negative heuristic advises against (does not forbid) challenging the core of a paradigm, in this case the homocentric sphere ontology with accompanying sublunary and superlunary elements and behaviors. Again, the advice is presumptive and can be suspended for specific pragmatic advantages. Thus, the presumption that all celestial motions be referred to a common center is suspended in the addition of eccentrics, equants and minor (adjusting) epicycles to the theoretical resources. The cost of this exception is the explanatory coherence of Aristotle’s program and the benefit is indefinite power of improving accuracy in saving the appearances. Positive heuristics chart a way forward in a research program and can advise the suspension or abandonment of the constraints provided by the negative.

A language for communicating within a scientific community propositions about observable and unobservable occurrences and including paradigm specific (English) terms such as ‘space’, ‘planet’, ‘Earth’, gravity’, ‘levity’ and ‘motion’ (Feyerabend, 1962, Hanson,1960). Meanings are determined by shared and heuristically protected belief (Lakatos’ negative heuristic, 1970; C.I. Lewis’ pragmatic a priori) that implicitly define concepts (see Newton’s ‘Definitions’ below), and a terminology of designators, the latter of which can vary among local communities. The same designator can designate different concepts in different paradigms, so respective proponents can ‘talk through one another’ by equivocating terms (Kuhn, 1962, Chapter III). Aristotelian ‘gravity’ refers to the center of the universe, an empty but dynamically privileged point in space, but Galilean ‘levity’ designates Archimedean buoyancy (de Motu), a product of the ‘weight’ of the enclosing medium with no essential reference to a privileged point in space. Newtonian ‘gravity’ is an interaction between two or more pieces of matter in centerless infinite space. The Aristotelian reference to the center of the universe is impossible in Newton’s infinite homogeneous and anisotropic space. Aristotle cannot refer to Newton’s corpuscles without embedding himself in Newton’s conceptual system (see below).

The choice of designators is not always arbitrary but is sometimes conventional for a variety of community utilities: Descriptive designators (‘white breasted nuthatch’ or Newton’s vis insita—innate force) aid amateur ornithologists in the field or embody Newton’s essentialism regarding matter. But descriptive names have a downside: Mayr pointed out that different species or geneses can look alike in the field, as with downy and hairy woodpeckers, and yet be phylogenically distant from one another. When Newton realized he needed a clear distinction between ‘active’ and ‘passive’ forces he chose ‘vis inertia’. On the other hand, non-descriptive numerical designators in different circumstances are convenient for tracing migration paths or for military administration.

Comment: Kuhn II embeds a philosophy of scientific language in which meanings are determined by core paradigmatic beliefs. Aristotelian ‘gravity’ cannot be translated into Newtonian terms without generating contradictions. (Feyerabend, 1962). (Cf. Quine, 1953 on synonymy: a and b are synonymous just if a can be substituted for b, and conversely, in all contexts salve veritatis.). Shapere, in his review of the Structure (1964), points out that holistic semantics applied globally within a paradigm begs the question of absolute discontinuity between paradigmatic languages. However, these semantic incommensurables are local, limited to a few terms, gravity, space, motion, earth, planet. They exclude light, shadow, body, tower, stone, iron, terrestrial location, and some natural kinds shared by those within and outside the Aristotelian tradition. (Cf. Galileo, Dialogues, 1632, below.)

Paradigms are ‘defined’ by examples of successful research, exemplars, (Structure 2nd edition, Postscript 1970) as embodied in textbooks and other media for cultural transmission of the paradigm. Kuhn also mentions there are ‘formal’ components of paradigms, e.g. spherical geometry, trigonometry and mathematical representations of location, velocity and acceleration, oscillations, waves, frequencies and amplitudes. In the holistic semantics of Kuhn II these formal concepts cannot be shared among multiple paradigms, an implausible stretch for those focused on interparadigmatic continuity.

Comment: The paradigm concept is complex and multifaceted, not ambiguous (Cf. M. Masterman in Lakatos and Musgrave 1970). Lloyd (1983) held that ‘theories’ are ‘families’ of realizable models like Kuhn’s exemplars, but models of what if ‘syntactical’ paradigmatic postulates are to be dismissed? Kuhn’s exemplars, like Lloyd’s models, must exemplify something even if they share only non-transitive Wittgenstinian familial similarities. Lakatos’ distinction between the ‘core’ and ‘periphery’ of research ‘programmes’ plausibly addresses this problem. The content of peripheral applications is variable under a constraining core. But the core is preserved by a heuristic, which by Wimsatt’s (1980) conception can be variable within pragmatic values. The core also embodies ‘ideals of natural order’ (Toulmin, 1989), i.e. fictional null processes such as Newton’s First Law and Fisher’s Mendelian population (Cartwright, 1983; see Below).

A framework for interpreting history: Whig historiography interprets the history of science as a linear progression to a currently accepted paradigm, e.g. from Democritus to Dalton leaving Aristotle outside of ‘proper’ science. This misrepresentation of history could be considered a pedagogical and rhetorical fiction for persuading newcomers, students, the public and other outsiders, to join a given scientific community.

Beyond the core Aristotelian ontology there is also a periphery of variant content, Democratean infinite space, Pythagorean 10-body planetary system circulating a central fire, Aristarchus’ heliocentric system, proposals of terrestrial rotation, and Venus and Mercury as solar satellites. Orthodox Aristotelians are likely to regard these items as speculative at best, though some are exploited much later in revolutionary developments (Copernicus, Descartes, Tycho Brahe), so they endure historically as peripheral slack around the paradigmatic core.

Chapter I.2:

Questions and Inquiry

One aspect of the inattention of recent philosophy to the discovery process has been its focus upon propositions, semantic structures, rules, models and concepts to the exclusion of questions and heuristics. Propositions can be used to affirm, deny, suggest and record laws, theories, facts and experimental protocols. Scientific belief systems consist of the propositions identified by a given community of researchers, and often a scientific discipline is identified with such a system. Questions, on the other hand, when asked, specify, suggest, recommend, prioritize demand or endorse epistemic aspirations and research agendas instrumental to the generation of knowledge. They can be put forth to recommend that a research community’s attention be focused on some item, some missing element in a causal chain, some phenomenon unexplained, a mismatch between observed motion and saving constructs, an entity whose effects have been ignored. If one were to regard science as a constellation of human activities or a ‘practice’ (Kitcher, 1993), questions should be considered as important as beliefs in a philosophical theory of science. Considering questions in philosophy of science is needed for a complete account of scientific activity because desires, in this case epistemic desires or desires to know, are as important as beliefs in guiding scientific research (Kleiner, 1992).

Much philosophical effort has been devoted to normative principles for entertaining, accepting or rejecting propositions in science, but little has been said about like principles for the appraisal of questions. It has been suggested that the logic of scientific discovery might fruitfully be construed as the logic of questions (Hintikka, 1981, Sintonen, 1990). Questions (or the statement of problems) convey beliefs (with varying specificity and certainty) about items of presumed ignorance that should be remedied in (Plato, Meno, Newton Queries, Bromberger on knowing our ignorance; Peirce, Dewey on doubt; Shapere on specific doubt), and among which various priorities should be attached. However, for this suggestion to be developed into a philosophically illuminating position, a ‘logic’ (or, better, a heuristic) of questions suitable for the practice of science must be at least demonstrated if not fully elaborated.

There is a body of literature devoted to ‘erotetic logic’ or the logic of questions (e.g. Prior and Prior, 1955, Aqvist, 1975, Belknap and Steele, 1976, Hintikka, 1976, Kleiner, 1992). The focus of this literature is principally on these topics: (i) formally expressing various types of question, (ii) the link between a question’s presupposition and the validity or ‘askability’ of the question, (iii) relations between questions and various kinds of answer, and (iv) relations between questions and component questions or sub-questions.

Formal erotetic logic provides links between belief and question that carry us a considerable distance in our efforts to understand scientific practice. A logically necessary condition for asking a question is that its presupposition be believed to be true. We may ask What is the molecular mechanism for cell differentiation? only if we believe There is a molecular mechanism for cell differentiation. In science these beliefs and their conceptual prerequisites are provided by the Kuhnian paradigm (Kleiner, 1970, 1993), about which more will be demonstrated below. The question-answer relation is important for providing an understanding of how informed inquiry is possible without the possession of the information sought (i.e. answers). This is the problem of inquiry set forth in Plato’s Meno, and it is addressed by showing how questions can contain incomplete information about a subject, but sufficient information to inform epistemic desiderata, where this information is implicit in but not constitutive of the desired answers.

The same relation can explain how experimental inquiry is ‘theory laden’ without begging the very questions addressed to nature by the scientist. One might ask What is the mass of the Sun?, a question that presupposes that the Sun has mass. This presupposition carries all of the conceptual baggage of Newtonian celestial mechanics. The answer to the question, "The mass of the Sun is M", where M is some large multiple of Earth Mass Units (EMU’s), is not divulged by the question. This question would not be available to someone uninformed by Newton’s conceptual innovations or would be rejected by someone who believed that the Sun is massless or that mass-values of an object like the Sun are somehow not unique.

The question-answer relation is also informative about serendipitous scientific discovery which is broadly defined as occurring when a question Q generates an answer to a different question Q’ (Kantorovich and Ne’eman 1989). A surprise answer can defeat the question Q that was instrumental in producing the answer by falsifying Q’s presupposition. In searching for ‘centers of creation’ (from Lyell’s Principles of Geology) Darwin collects mockingbird specimens from several Galapagos Islands. Once convinced that the specimens from the different islands are distinct species, Darwin embraces the thesis that these distinct species descend from common ancestors, a thesis that he believes is incompatible with the existence of centers of creation. The ‘informed minds’ that serendipitous discoveries illuminate are minds that are aware of or can come up with the alternative or novel question Q’, which will have presuppositions, conceptual and