Theories, Family Resemblances and Analogy

Part of the Synthese Library book series (SYLI, volume 197)


The related problems of metaphor and analogical reasoning have recently become of interest within three separate disciplines: philosophy of science, philosophy of language, and Artificial Intelligence. Classically the study of analogy has been concerned with analogical meaning (related to metaphor), and with analogical argument, which was seen in standard logic texts up to the 19th century as a very poor relation of inductive and hypothetical reasoning.1 These two aspects began to come together in the 19th century with explicit discussion of both meaning and argument from the point of view of scientific models.2 Subsequently positivist philosophy of science added “analogy” to its consideration of inductive argument as an important ingredient of induction, and post-positivist philosophy began to consider analogical meaning in models as crucial for the understanding of theoretical concepts and the problem of meaning variance between theories.3


Concept Formation Linguistic Term Analogical Reasoning Successful Prediction Family Resemblance 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    For example J. S. Mill, A System of Logic, London, 1843, Bk. III, Ch. 20 and W. S. Jevons, Principles of Science, Macmillan, London, 1874, Ch. 28.Google Scholar
  2. 2.
    See W. Thomson, Papers on Electrostatics and Magnetism, 2nd ed. London, 1884, p. 29f;Google Scholar
  3. 2a.
    W. J. M. Rankine, Miscellaneous Scientific Papers, London, 1881, p. 209;Google Scholar
  4. 2b.
    J. C. Maxwell, Scientific Papers, Cambridge University Press, 1890, I, p. 155;Google Scholar
  5. 2c.
    and M. Hesse, The Structure of Scientific Inference, Macmillan, London, 1974, Ch. 11.Google Scholar
  6. 3.
    Hesse, ibid., specially Chs. 2 and 9.Google Scholar
  7. 4.
    Among the many recent books and collections of papers devoted to the philosophy of metaphor are D. E. Cooper, Metaphor, Blackwell, Oxford, 1986;Google Scholar
  8. 4a.
    M. Johnson (ed.), Philosophical Perspectives on Metaphor, University of Minneapolis Press, 1981;Google Scholar
  9. 4b.
    G. Lakoff and M. Johnson, Metaphors We Live By, Chicago University Press, 1980;Google Scholar
  10. 4c.
    D. Miall (ed.), Metaphor: Problems and Perspectives, Harvester, Brighton, Sussex, 1982;Google Scholar
  11. 4d.
    A. Ortony (ed.), Metaphor and Thought, Cambridge University Press, 1979;Google Scholar
  12. 4e.
    P. Ricoeur, The Rule of Metaphor (trans. R. Czerny), Routledge and Kegan Paul, London, 1978;Google Scholar
  13. 4f.
    J. F. Ross, Portraying Analogy, Cambridge University Press, 1882;Google Scholar
  14. 4g.
    and S. Sachs (ed.), On Metaphor, Chicago University Press, 1979. There is also an extended discussion in H.-G. Gadamer, Truth and Method (English trans.) Sheed and Ward, London, 1975, p. 387ff.Google Scholar
  15. 5.
    I have discussed this issue in “The cognitive claims of metaphor”, in Metaphor and Religion, Theolinguistics 2 (ed. J. P. van Noppen), Brussels, 1984, and “Texts without types and lumps without laws”, New Literary History, 17, 1985–86, 32.Google Scholar
  16. 6.
    See M. Arbib and M. Hesse, The Construction of Reality, Cambridge University Press, 1986, Chs. 2 and 3.CrossRefGoogle Scholar
  17. 7.
    See D. Rothbart, “Analogical information processing within scientific metaphors”, and P. Thagard, “Dimensions of analogy”, this volume.Google Scholar
  18. 8.
    For a discussion of this example, see M. Hesse, The Structure of Scientific Inference, Ch. 11.Google Scholar
  19. 9.
    M. Black, Models and Metaphor, Cornell University Press, Ithaca, N.Y., 1962, Chs. 3 and 13.Google Scholar
  20. 10.
    P. H. A. Sneath and R. R. Sokal introduce their theory of biological taxonomy by referring to polythetic classes and family resemblances, and give references to methods in information retrieval which are based on word clusters according to similar principles (Numerical Taxonomy, Freeman, San Francisco, 1973, pp. 21, 448). See also G. Dunn and B. S. Everitt, An Introduction to Mathematical Taxonomy, Cambridge University Press, 1982;Google Scholar
  21. 10a.
    N. Jardine and R. Sibson, Mathematical Taxonomy, Wiley, London, 1971;Google Scholar
  22. 10b.
    and K. Sparck Jones, “Some thoughts on classification for retrieval”, J. of Documentation, 26, 1970, 89; and “Clumps, Theory of”, in Encyclopedia of Library and Information Science (eds. Kent and Lancour), Marcel Dekker, New York, Vol. 5, 1971, 208.CrossRefGoogle Scholar
  23. 11.
    E. Rosch, “Cognitive reference points”, Cognitive Psychology, 7, 1975, 532;CrossRefGoogle Scholar
  24. 11a.
    E. Rosch, “Cognitive representations of semantic categories”, J.Experimental Psychology, General, 104, 1975, 192;CrossRefGoogle Scholar
  25. 11a.
    E. Rosch, “Principles of categorization” in Cognition and Categorization (ed. E. Rosch and B. B. Lloyd), Wiley, New York, 1978, p. 27;Google Scholar
  26. 11b.
    E. Rosch and C. B. Mervis, “Family resemblances: studies in the internal structure of categories”, Cognitive Psychology, 7, 1975, 573;CrossRefGoogle Scholar
  27. 11c.
    and C. B. Mervis and E. Rosch, “Categorization of natural objects”, Ann. Rev. Psychology, 32,1981, 89.CrossRefGoogle Scholar
  28. 12.
    Relations of cue properties can be used to define the similarities and differences between objects. It is important to notice that “similarity” need not be a symmetrical relation. For example, Mervis and Rosch (op. cit. p. 94) note that Mexico is perceived as more similar to the United States than the United States is to Mexico. This asymmetric relation can be represented by a similarity function S(ab) which depends on the proportion between the number of a properties that also belong to b to the total number of a properties judged relevant to the comparison. Then S(ab) is generally not equivalent to S(ba). Cooper (op. cit. p. 186) is surely mistaken in arguing that since similarity is a symmetric relation, similes cannot be types of metaphor, and metaphors cannot be based on similarities.Google Scholar
  29. 13.
    M. Turner, “Categories and analogies”, and M. Johnson, “Some constraints on embodied analogical understanding”, this volume.Google Scholar
  30. 14.
    Dictionaries and thesauruses contain much of the information needed to construct such a nearness distribution of linguistic terms, see Sparck Jones and Jackson, op. cit. Google Scholar
  31. 15.
    Rosch, “Cognitive representations of semantic categories”, 226, and K. Dahlgren, “The cognitive structure of social categories”, Cognitive Science, 9, 1985, 379.CrossRefGoogle Scholar
  32. 16.
    Hesse, “Texts without types and lumps without laws”.Google Scholar
  33. 17.
    Cooper (op. cit. p. 279) sees it as an objection to the kind of theory adopted here that it cannot be the case that “unknown to the [speakers] themselves, they are employed full-time in speaking metaphorically”. His argument is that they must have a theory of correct and consistent speech that guarantees that non-metaphoric talk is the norm. But (i) a theory of metaphoric connections does not have to be (consciously) “known to speakers” any more than the rules of grammar do, and (ii) on the theory presented here, a “theory of correctness” of metaphoric talk, though itself in principle metaphoric (because all talk is), would be at the relatively “literal” end of the scale of metaphoricalness, as a contribution to applied logic.Google Scholar
  34. 18.
    See my discussion of “grue” in The Structure of Scientific Inference, Ch. 3.Google Scholar
  35. 19.
    Ibid., Ch. 5.Google Scholar
  36. 20.
    What P. Suppes calls the “combinatorial jungle” (“Concept formation and Bayesian decisions”, in Aspects of Inductive Logic (ed. J. Hintikka and P. Suppes), North-Holland, Amsterdam, 1966, p. 21).Google Scholar
  37. 20a.
    See also R. Carnap, The Logical Foundations of Probability, Routledge, Chicago, 1962, p. 124.Google Scholar
  38. 21.
    B. de Finetti, “Foresight: its logical laws, its subjective sources”, in Studies in Subjective Probability (ed. H. E. Kyburg and H. E. Smokler), Wiley, New York, 1964, p. 120.Google Scholar
  39. 22.
    See my Models and Analogies in Science, Sheed and Ward, London, 1963 and The Structure of Scientific Inference, Ch. 11.Google Scholar
  40. 23.
    Ibid., Ch. 9.Google Scholar
  41. 24.
    Alternatives are C. Glymour’s “bootstrapping” theory, in Theory and Evidence, Princeton University Press, 1980, and I. Niiniluoto, “Analogy, transitivity, and the confirmation of theories”, in Applications of Inductive Logic (ed. L. J. Cohen and M. Hesse), Clarendon, Oxford, 1980, p. 218.Google Scholar
  42. 25.
    Rosch (op. cit. 1978, p. 28) makes it clear that her theory is not intended as a theory of the development of categories in children or adults, but it may surely provide suggestions for such a theory.Google Scholar
  43. 26.
    Cooper (op. cit. p. 257f) has the best recent discussion, but in the end rejects the primacy of metaphor.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1988

Authors and Affiliations

  1. 1.Department of History and Philosophy of ScienceCambridge UniversityUK

Personalised recommendations