Kuhn, historical philosophy of science, and case-based reasoning

Configurations

Volumn 6, Issue 1, 1998, Pages 51-85

  Nickles, Thomas ^a  

^a UNIVERSITY OF NEVADA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0003027137     PISSN: 10631801     EISSN: 10806520     Source Type: Journal    
DOI: 10.1353/con.1998.0008     Document Type: Review

References (112)

1. Thomas S. Kuhn, The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1962), p. 1.
2. See Thomas S. Kuhn, "Postscript-1969" to the second edition of Structure (Chicago: University of Chicago Press, 1970), pp. 174-210;
3. Thomas S. Kuhn, idem, "Logic of Discovery or Psychology of Research?" and "Reflections on My Critics," in Criticism and the Growth of Knowledge, ed. Imre Lakatos and Alan Musgrave (Cambridge: Cambridge University Press, 1970), pp. 1-23 and 231-278;
4. Thomas S. Kuhn, idem, "Second Thoughts on Paradigms," in The Structure of Scientific Theories, ed. Frederick Suppe (Urbana: University of Illinois Press, 1974), pp. 459-482.
5. "Logic of Discovery" and "Second Thoughts" are reprinted in Thomas S. Kuhn, The Essential Tension: Selected Studies in Scientific Tradition and Change (Chicago: University of Chicago Press, 1977), chaps. 11 and 12.
6. A nearly complete bibliography of Kuhn's works may be found in Paul Hoyningen-Huene, Reconstructing Scientific Revolutions: Thomas S. Kuhn's Philosophy of Science (Chicago: University of Chicago Press, 1993).
7. For Stephen Toulmin, Kuhn's work represents the removal of the final obstacle to the historicization of everything. For Alasdair MacIntyre, the Import of Kuhn's work is that "scientific reason turns out to be subordinate to, and intelligible only in terms of, historical reason" and is therefore humanized ("Epistemological Crises, Dramatic Narrative, and the Philosophy of Science," Monist 60 [1977]: 453-471 [quote on p. 464];
8. reprinted in Paradigms and Revolutions: Applications and Appraisals of Thomas Kuhn's Philosophy of Science, ed. Gary Gutting [Notre Dame: University of Notre Dame Press, 1980], pp. 54-74).
9. Charles Taylor speaks of the victory of "universal hermeneutics" in "Understanding in Human Science," Review of Metaphysics 34 (1980): 25-38, at p. 26.
10. But compare the later Kuhn: Thomas S. Kuhn, "The Halt and the Blind: Philosophy and History of Science," British Journal for the Philosophy of Science 31 (1980): 181-192;
11. Thomas S. Kuhn, idem, "Commensurability, Comparability, Communicability," in PSA 1982, ed. Peter Asquith and Thomas Nickles (East Lansing: Philosophy of Science Association, 1983), pp. 669-688 and 712-716;
12. Thomas S. Kuhn, idem. "The Trouble with the Historical Philosophy of Science," Robert and Maurine Rothschild Distinguished Lecture, Harvard University, 19 November 1991.
13. At least in his published work. Nancy Nersessian first made me realize the extent to which cognitive psychology drops out of Kuhn's original, interdisciplinary project, aside from a bit of linguistic analysis. For her own conception of how history and cognitive science might be integrated, see Nancy Nersessian, "Opening the Black Box: Cognitive Science and History of Science," Osiris 10 (1995): 194-211.
14. In "Kuhn and the Cognitive Revolution" (in this issue) she develops Kuhn's theory of cognition in terms of Eleanor Rosch's prototype theory. In sect. VI, I shall briefly take up the extension of this into schema theory.
15. Thomas S. Kuhn, Black-Body Theory and the Quantum Discontinuity: 1894-1912 (Oxford: Oxford University Press, 1978), is an internalist history of the emergence of early quantum theory from its background in classical statistical mechanics.
16. As quoted by William James, "The Will to Believe," in The Will to Believe and Other Essays in Popular Philosophy (New York: Longmans, Green, 1897), § ii, (widely reprinted).
17. William James, The Will to Believe and Other Essays in Popular Philosophy, Ibid., § x.
18. Karl Popper, Conjectures and Refutations (New York: Basic Books, 1962).
19. Robert Merton, "Science and the Social Order" (1938),
20. reprinted in Merton, The Sociology of Science (Chicago: University of Chicago Press, 1973), chap. 12.
21. The latter, paradoxical statement is the title of a popular book by Wayne Dyer, M.D.: You'll See It When You Believe It (New York: William Morrow, 1989).
22. On Whig history, see Herbert Butterfield, The Whig Interpretation of History (London: Bell, 1931);
23. Thomas Nickles, "Good Science as Bad History: From Order of Knowing to Order of Being," in The Social Dimensions of Science, ed. Ernan McMullin (Notre Dame: University of Notre Dame Press, 1992), pp. 85-129;
24. Thomas Nickles, idem, "History of Science and Philosophy of Science," Osiris 10 (1995): 139-163.
25. In Thomas S. Kuhn, "Revisiting Planck," Historical Studies in the Physical Sciences 14 (1984): 231-252, Kuhn (p. 246) prefers the term ethnocentric to Whig. I agree, but retain the more commonly used word.
26. See Peirce's classic essay, "The Fixation of Belief" (1877), in Collected Papers of Charles Sanders Peirce, ed. Charles Hartshome and Paul Weiss (Cambridge, Mass.: Harvard University Press, 1934), vol. 5, pp. 358-387 (widely reprinted).
27. Joseph Rouse, "Science as Practice: Two Readings of Thomas Kuhn," chap. 2 of Rouse, Knowledge and Power: Toward a Political Philosophy of Science (Ithaca, N.Y.: Cornell University Press, 1987).
28. In Engaging Science: How to Understand Its Practices Philosophically (Ithaca, N.Y.: Cornell University Press, 1996), Rouse contends that it is practices that limit and shape the domain of intelligibility, both as a field of action and as a space of prepositional representation.
29. For a vigorous critique of "practice" accounts, see Stephen Turner, The Social Theory of Practice: Tradition, Tacit Knowledge, and Presuppositions (Chicago: University of Chicago Press, 1994).
30. James, "Will to Believe" (above, n. 6), § ii.
31. Thomas S. Kuhn, The Copernican Revolution (Cambridge, Mass.: Harvard University Press, 1957).
32. For James on "spectator's" knowledge, see Max Fisch, Classic American Philosophers (New York: Appleton-Century-Crofts, 1951), p. 28.
33. Although he does not mention James, see Stephen Toulmin, "Death of the Spectator," a chapter in Toulmin, The Return to Cosmology (Berkeley: University of California Press, 1982), pp. 237-254.
34. See Thomas S. Kuhn, "Mathematical versus Experimental Traditions in the Development of Physical Science," Journal of Interdisciplinary History 7 (1976): 1-31;
35. reprinted in Essential Tension (above, n. 2), chap. 3.
36. Kuhn, "Postscript" (above, n. 2).
37. Steve Fuller, "Being There with Thomas Kuhn: A Parable for Postmodern Times," History and Theory 31 (1992): 241-275;
38. Steve Fuller, idem, "Teaching Thomas Kuhn to Teach the Cold War Vision of Science," Contention 4 (1994): 81-106. A "fuller" account of these matters is forthcoming in his book on Kuhn, the manuscript of which I have not seen. I am more sympathetic to historical and to "deep cultural" accounts of science than is Fuller, who rejects both.
39. These are my words, not Kuhn's, although he explicitly rejected, as ridiculous, the Fopperlans' call for perpetual scientific revolution. See especially the exchanges among Kuhn and Popperians of various stripes, including Lakatos, in Lakatos and Musgrave, Criticism (above, n. 2).
40. Kuhn joined his historian colleagues in dismissing history as written from the point of view of great scientists. This was a courageous move on their part (albeit professionally self-interested), for until Kuhn's generation, most historians of science had been practicing scientists, often eminent ones; scarcely anyone else was technically competent. Furthermore, the move disconnected working scientists from their own history in two ways: (a) by rewriting the job description of historians as studying the history of a period in its own terms, ideally in complete ignorance of modern science (completely reversing the old task of explaining the historical emergence of presentday science, which requires up-to-date scientific knowledge); and (b) by denying that scientists are the best authorities on their own previous work. Historians claim to understand the work and aims of X and of X's predecessors better than X does. This professional move by historians became more strained when internalist history yielded to social history and sociology of science, some of whose practitioners advocate technical ignorance even in the study of contemporary scientific work.
41. Compare the question (raised in conversation by Thomas Sederqvist) whether priests and ministers should be exposed to the real history of Jesus and the Gospels.
42. See below; the articles cited in note 11; and Thomas Nickles, "A Multi-Pass Conception of Scientific Inquiry," Danish Yearbook of Philosophy, ed. S. A. Pedersen and T. Soderqvist (in press).
43. For an excellent discussion of Kuhn's situatedness problem, see Hoyningen-Huene. Reconstructing Scientific Revolutions (above, n. 2). chap. 3.
44. See also the title essay of Stanley Fish, Is There a Text in This Class? The Authority of Interpretive Communities (Cambridge, Mass.: Harvard University Press, 1980), pp. 303-321;
45. Stanley Fish, idem, "Being Interdisciplinary Is So Very Hard to Do," Profession 89 (MLA): 15-22. Fish contends that cross-disciplinary discourse always faces problems of Whiggish appropriation, compounded by incommensurability. It is a consequence of recent literary criticism, with its attention to reader response, that the very idea of "a reading" of a work is Whiggish, reflecting the situation of the reader.
46. The locus classicus for the idea of "strong reading" is Harold Bloom, The Anxiety of Influence (Oxford: Oxford University Press, 1973).
47. Nickles, "Good Science" (above, n. 11), p. 98.
48. See Thomas S. Kuhn and John Heilbron, "The Genesis of the Bohr Atom," Historical Studies in the Physical Sciences 1 (1969): 211-290. The study of citation networks (citation and cocitation analysis) is a useful way to track changes over time in the status of frequently cited works.
49. For a description and references, see Marc De Mey, The Cognitive Paradigm (Dordrecht: Reidel, 1982), pp. 124 ff.
50. But see MacIntyre, "Epistemological Crises" (above, n. 3),
51. Herbert Butterfield, The Englishman and His History (Cambridge: Cambridge University Press, 1944), on the positive role of Whig redescription, as healing cultural tears and retaining a place in the successful narrative for the earlier work. Redescription is not entirely negative.
52. This is not the place to pursue the interesting, reflexive question for historians themselves, who are solving their own current research, career, and life problems in writing history. Is there anything more deliciously ironic than the historians' own story of the long struggle and eventual triumph of anti-Whiggish history?
53. An extended example (which I have discussed in articles cited above, n. 11) is Kuhn's Black-Body Theory (above, n. 5). In 1905 and after, Hendrik Antoon Lorentz and Paul Ehrenfest misinterpreted Max Planck's 1900 work as a solution that Planck never proposed (and later repudiated) to a problem ("the ultraviolet catastrophe") that Planck had not entertained.
54. This Whig-anti-Whig (and anti-anti-Whig) counterpoint partly explains differences between philosophers of science and (other) science studies practitioners. Many historians, sociologists, and anthropologists of science complain that philosophers are "pro science" as well as being Whiggish progressives. While most science studies practitioners assume the methodological stance of distanced outsiders and attempt to make science strange, philosophers tend to consider themselves quasi-insiders and are more interested in the claims the sciences make about the universe than in the detailed processes of producing those claims. For them the cosmos, not the scientific community, is the reality most worth studying. In Social Epistemology (Bloomington: Indiana University Press, 1988), Steve Fuller rejects both these positions for a normative-critical stance toward science.
55. Kuhn, "Postscript" (above, n. 2), p. 191.
56. Although only published in 1974, in the Suppe volume (above, n. 2), "Second Thoughts" dates from the 1969 conference.
57. Hoyningen-Huene, Reconstructing Scientific Revolutions (above, n. 2), p. 90.
58. See Allen Newell and Herbert A. Simon, Human Problem Solving (Englewood Cliffs, N.J.: Prentice-Hall, 1972).
59. Quoted from R. Duda and E. Shoitliffe, "Expert Systems Research," Science 220 (1983): 261-268.
60. For more on expert systems, see, e.g., Patrick Winston, Artificial Intelligence, 3rd ed. (Reading, Mass.: Addison-Wesley, 1992);
61. James Martin and Steven Oxman, Building Expert Systems: A Tutorial (Englewood Cliffs, N.J.: Prentice-Hall, 1988);
62. F. Hayes-Roth, "Expert Systems," in Encyclopedia of Artificial Intelligence, 2nd ed., ed. Stuart Shapiro (New York: Wiley, 1992), pp. 477-489.
63. The central thesis of Hubert and Stuart Dreyfus, Mind Over Machine (New York: Free Press, 1986), is that reducing skilled, intuitive expertise to a system of rules necessarily dumbs it down.
64. In sum, RBR does not provide a workable solution to the Meno problem: the problem of how inquiry (learning) is possible, raised by Plato in his dialogue Meno. By contrast, CBR attempts to solve the Meno problem by rhetorical (analogy- or similarity-based) pattern matching.
65. E.g., the PI program of Keith Holyoak and Paul Thagard has a little bit of everything. It is a production system (based on "if-then" production rules) that can fire in parallel and that incorporates some CBR (but without indexing), some schema theory, and some analogical reasoning. For a description of this and Thagard's ECHO program, which provides a vaguely Kuhnian account of scientific revolutions, see Paul Thagard, Computational Philosophy of Science (Cambridge, Mass.: MIT Press, 1988);
66. Paul Thagard, idem, Conceptual Revolutions (Princeton: Princeton University Press, 1992).
67. For the cognitive turn in psychology, applied to the analysis of science, see De Mey, Cognitive Paradigm (above, n. 26).
68. For a sampling of more recent work, see Ronald Giere, ed., Cognitive Models of Science (Minneapolis: University of Minnesota Press, 1992).
69. The basic reference is Janet Kolodner, Case-Based Reasoning (San Francisco: Morgan Kaufmann, 1993).
70. Some older references are Christopher Riesbeck and Roger Schank, Inside Case-Based Reasoning (Hillsdale, N.J.: Erlbaum, 1989);
71. DARPA, Proceedings, Case-Based Reasoning Workshop (San Mateo, Calif.: Morgan Kaufmann, 1989);
72. Menachem Jona and Janet Kolodner, "Reasoning, Case-Based," in Encyclopedia of Artificial Intelligence (above, n. 35), pp. 1265-1279.
73. Kuhn, "Postscript" (above, n. 2), p. 190. This paragraph suggests that there need be no sharp breaks between normal and revolutionary CB research, but I cannot pursue this anti-Kuhnian theme here.
74. See Berndt Brehmer, "In One Word: Not from Experience," in Judgment and Decision Making, ed. Hal Arkes and Kenneth Hammond (Cambridge: Cambridge University Press, 1986), pp. 705-719.
75. Paul Feyerabend, "Realism and the Historicity of Knowledge," Journal of Philosophy 86 (1989): 393-406.
76. "The Essential Tension" is the title essay of Kuhn's collection (above, n. 2). This tension has since been documented by cocitation studies, which show, among other things, how works (some of them old standards) cited in scientific articles become reconceptualized through the lens of the newer work.
77. The case-study approach was Kuhn's own introduction to the history of science. The first several years of his teaching were in the course at Harvard designed by President James B. Conant and chemist Leonard K. Nash, centered on the two-volume text, Harvard Case Histories in Experimental Science (Cambridge, Mass.: Harvard University Press, 1948), edited by Conant and Nash.
78. In "Being There" (above, n. 19), p. 263n75, Fuller says the course was modeled on case-based teaching in law and business schools, with the intention that students elicit general principles from the cases.
79. See Stephen Toulmin, Human Understanding (Princeton: Princeton University Press, 1972), chap. 1; and see pp. 148 ff for the idea of "genealogies of problems." Toulmin is another philosophical source of CBR models. He has long pointed to case law (versus statute law) as a model for scientific thinking without rules.
80. See Albert Jonsen and Stephen Toulmin, The Abuse of Casuistry: A History of Moral Reasoning (Berkeley: University of California Press, 1988).
81. I am also indebted to Dudley Shapere, Reason and the Search for Knowledge (Dordrecht: Reidel, 1984). Shapere has always emphasized the importance of studying cases.
82. For various approaches to schema theory, see William Brewer and G. V. Nakamura, "The Nature and Functions of Schemas," in Handbook of Social Cognition, vol. 1, ed. R. S. Wyer, Jr., and T. K. Srull (Hillsdale, N.J.: Erlbaum, 1984), pp. 119-160;
83. David Rumelhart, "Schemata and the Cognitive System," in Handbook of Social Cognition, ibid., pp. 161-199;
84. Michael Arbib, "Schema Theory," in Encyclopedia of Artificial Intelligence (above, n. 35), pp. 1427-1443;
85. Michael Arbib and Mary Hesse, The Construction of Reality (Cambridge: Cambridge University Press, 1986);
86. chaps. 6 and 14 of David Rumelhart, James McClelland, and the PDP Research Group, Parallel Distributed Processing, 2 vols. (Cambridge, Mass.: MIT Press, 1986);
87. Sandra Marshall, Schemas in Problem Solving (Cambridge: Cambridge University Press, 1995).
88. See also Eleanor Rosch on prototype theory; e.g., Eleanor Rosch and B. B. Lloyd, eds., Cognition and Categorization (Hillsdale, N.J.: Erlbaum, 1978).
89. Alternatively, we might relate exemplars to the body of psychological and philosophical literature on exemplar-based accounts of concepts. See, e.g., E. E. Smith and D. L. Medin, Categories and Concepts (Cambridge, Mass.: Harvard University Press, 1981);
90. Scott Allen and Lee Brooks, "Specializing the Operation of an Explicit Rule," Journal of Experimental Psychology 120 (1991): 3-19;
91. Muhammad Ali Khalidi, "Two Concepts of Concept," Mind and Language 10 (1995): 402-422. Smith and Medin develop an exemplar-based account of concepts, which, however, is implemented as a feature-based account. Allen and Brooks study the effect of practice on the speed and accuracy of applying categorization rules; they conclude that, in favorable environments, performance undergoes a transition from being rule-based (even automated or compiled rules) to being case-based.
92. See Arbib, "Schema Theory" (above, n. 46), p. 1427.
93. Hoyningen-Huene, Reconstructing Scientific Revolutions (above, n. 2), chap. 4, distinguishes exemplars as accepted local problem solutions from exemplars taken more globally as models for teaching and research.
94. For habits of mind, see Howard Margolis, Patterns, Thinking, and Cognition (Chicago: University of Chicago Press, 1987);
95. Howard Margolis, idem, Paradigms and Barriers (Chicago: University of Chicago Press, 1993).
96. Ulric Neisser, Cognition and Reality (San Francisco: Freeman, 1976), p. 22.
97. See also Barry Barnes, Thomas Kuhn and Social Science (New York: Columbia University Press, 1982), chap. 2, on training and finitism.
98. Rumelhart, "Schemata and the Cognitive System" (above, n. 46), p. 182.
99. The directness gives these cognitive processes an immediate, intuitive flow. We do not consciously have the sense of having to derive or calculate. The world just presents itself in this fashion, which makes these (learned) responses seem almost self-evident. On acquired intuition, see Dreyfus and Dreyfus, Mind Over Machine (above, n. 36);
100. Margolis, Patterns (above, n. 49), who develops an explicitly Kuhnian theory of cognition, based on pattern recognition, without computational detail. Kuhn was sympathetic to Margolis's treatment.
101. "Humanist thought throve on example," notes John Lyons, in Exemplum: The Rhetoric of Example in Early Modern France and Italy (Princeton: Princeton University Press, 1989), p. 12.
102. See also Jonsen and Toulmin, Abuse of Casuistry (above, n. 45). For recent AI treatments of analogy and other Kuhnian themes, see Thagard's two books (above, n. 38).
103. Taking a sociological rather than a mentalistic approach, Andy Pickering's work highlights the value of direct modeling in scientific research, starting with "Exemplars and Analogies: A Comment on Crane's Study of Kuhnian Paradigms in High Energy Physics," Social Studies of Science 10 (1980): 497-502, 507-508. Sociologists and social historians express serious misgivings about individual-cognitivist approaches to science studies.
104. See Steve Fuller, "The Philosophy of Science since Kuhn: Readings on the Revolution That Has Yet to Come," Choice 27 (1989): 595-601.
105. We can view the CBR movement as yet another attempt to capture the subtleties of rhetoric in a more precise manner. Admittedly, the actual implementation of CB systems involves rules - e.g., indexing rules, and retrieval and modification algorithms. However, at the gross level described above, CBR remains notably different from RBR.
106. See Ronald Giere, Explaining Science: A Cognitive Approach (Chicago: University of Chicago Press, 1988);
107. Ronald Giere, idem, "The Cognitive Structure of Scientific Theories," Philosophy of Science 61 (1994): 276-296, for a recent, very liberal treatment of the semantic conception of theories from a broadly Kuhnlan perspective.
108. Nancy Cartwright, in How the Laws of Physics Lie (Oxford: Oxford University Press, 1983) and later works, has gone beyond the semantic view, in roughly this direction.
109. I do not claim that such a methodology is sufficient for normal science. See Rouse, Engaging Science (above, n. 13), for a deflationary account of science along culture studies lines, with politics of science displacing epistemology as the central concern.
110. Consult the resulting volume, Scrutinizing Science: Empirical Studies of Scientific Change, ed. Arthur Donovan, Larry Laudan, and Rachel Laudan (Dordrecht: Kluwer, 1988).
111. For a response, see Thomas Nickles, "Remarks on the Use of History as Evidence," Synthese 69 (1986): 253-266.
112. The reference is to chap. 5 of Structure.