¹ Wittgenstein, L., Philosophical Investigations, (Oxford: Blackwell, 1976), translated by Anscombe, G. E. M., sections 65–70, 31–33.Google Scholar

² Mandelbaum, M., ‘Family Resemblances and Generalization Concerning the Arts’, American Philosophical Quarterly, 2, No. 3 (07 1965), 221.Google Scholar

³ For a selection of such counter-examples see, Midgley, M., ‘The Game Game’ Philosophy, 49, No. 189 (07 1974), 236CrossRefGoogle Scholar; and Simon, M. A., ‘When is a Resemblance a Family Resemblance?’, Mind, 78, No. 311 (07 1969), 409.Google Scholar

⁴ Caillois, R., Man, Play, and Games (London: Thames and Hudson, 1962), 14–26Google Scholar. I have altered Caillois' names for these categories. In the original they are: agon, alea, mimicry, and ilinx. See p. 36.

⁵ Caillois, , Man, Play, and Games, 8–9.Google Scholar

⁶ Lewis, D., Convention, (Cambridge Mass.: Harvard University Press, 1969), 16–24.Google Scholar

⁷ Quoted in Kennedy, Michael's introduction to, The Autobiography of Charles Hallé (London: Elek Books, 1972), 10.Google Scholar

⁸ Wittgenstein's original example in section 66 of the Investigations is Kampfspiele. I have been unable to find this term in a German/English dictionary, but the Duden Deutsches Universal Wörterbuch A-Z, (Mannheim: Duden Verlag, 1989)Google Scholar gives ‘Direct physical contact in a ball game which is taking place between two teams’, e.g. in Football, Handball, Rugby, Hockey etc. For example, ‘By ever more aggressive contact the player was finally beaten in an exciting Kampfspiel.’ (I am grateful to Stephen Nesom for this translation). There does not seem to be a direct English equivalent for this term—perhaps ‘tussle’ or ‘tackle’ come closest—and it certainly does not use the word ‘game’; this is no doubt the reason why Anscombe substituted ‘Olympic games’ in her translation. The difficulty for translation here is important. The noun Spiel is a transformation of the verb Spielen and therefore often occurs in contexts which are not correctly rendered by ‘game’ e.g., ‘Schauspiel’ (theatre play). As Hacker and Baker write: ‘The activities characterized as Spiele are a wider and more varied set than those called “games”. “Spiele” is used as an internal accusative of “Spielen”, so any activity which can be characterized as playing… is called “ein Spiel”. This does not affect Wittgenstein's point.’ (Baker, G. P. and Hacker, P. M. S., An Analytical Commentary on Wittgenstein's Philosophical Investigations (Oxford: Blackwell, 1983), 132Google Scholar.) The wider extension of ‘Spiel’ may not affect the point Wittgenstein is making, but it may mean that this is true of German but not of English. See Vendler, Z., ‘Linguistics and the a priori’, Philosophy and Linguistics, Lyas, C. (ed.) (London: Macmillan, 1971) 261Google Scholar for more discussion of these issues.

⁹ Searle, J., Speech Acts (Cambridge University Press, 1969), 33–35.CrossRefGoogle Scholar

¹⁰ This point is made in Khatchadourian, H., ‘Common Names and “Family Resemblances”’, in Pitcher, (ed.), Wittgenstein: The Philosophical Investigations (New York: Doubleday, 1966), 205–217.CrossRefGoogle Scholar

¹¹ I have more to say about art in ‘The Definition of “Art”’, The Philosophical Quarterly 41, No. 164 (07 1991), 271–286Google Scholar; and ‘Why “Art” Doesn't have Two Senses’, The British Journal of Aesthetics 31, No. 3 (07 1991), 214–221.Google Scholar

¹² Manser, A., ‘Games and Family Resemblances’, Philosophy, 42, No. 161 (07 1967) 216–217.CrossRefGoogle Scholar

¹³ The example is discussed in Manser, , ‘Games and Family Resemblances’.Google Scholar

¹⁴ I would like to thank J. M. Bell and the late Flint Schier for help with earlier versions of this paper.