The role of inductive reasoning in the evolution of conventions

Law and Philosophy

Volumn 17, Issue 4, 1998, Pages 377-410

  Sugden, Robert ^a  

^a UNIVERSITY OF EAST ANGLIA   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0032244195     PISSN: 01675249     EISSN: None     Source Type: Journal    
DOI: 10.2307/3505086     Document Type: Article

References (46)

1. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
2. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
3. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
4. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
5. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
6. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
7. David Hume, A Treatise of Human Nature (Oxford: Clarendon Press, 1978); first published 1740. The modern literature includes, e.g., David Lewis, Convention: A Philosophical Study (Cambridge, Mass.: Harvard University Press, 1969); Edna Ullman-Margalit, The Evolution of Norms (Oxford: Clarendon Press, 1977); Robert Sugden, The Economics of Rights, Cooperation and Welfare (Oxford: Blackwell, 1986); Ken Binmore, Game Theory and the Social Contract, Volume 1: Playing Fair (Cambridge, Mass.: MIT Press, 1993); Peyton Young, "The Evolution of Conventions", Econometrica 61 (1993), pp. 57-84; Brian Skyrms, Evolution of the Social Contract (Cambridge: Cambridge University Press, 1996).
8. One of the few papers in the game-theoretic literature which recognizes this problem is Sanjeev Goyal and Maarten Janssen, "Can We Rationally Learn to Coordinate", Theory and Decision 40 (1996), pp. 29-49. Goyal and Janssen argue, I think correctly, that the standard assumptions of rationality and common knowledge cannot explain how individuals learn to coordinate their behavior in repeated games. They show that some well-known theoretical results which appear to explain rational coordination in fact depend on hidden assumptions about the prior existence of coordinated expectations. The present paper is complementary with Goyal and Janssen's. They show how little can be derived from standard rationality assumptions; I look for an alternative model of human reasoning which can explain coordination.
9. Lewis, supra note 1.
10. Id., at 9.
11. Id., at 8-24.
12. Id., at 97-100.
13. Robert Sugden, "Courtesy and Convention: a Theory of Normative Expectations", paper prepared for conference on Practical Rationality, Rules and Structure, Zentrum für interdisziplinäre Forschung, Bielefeld, March 1998.
14. Robert D. Putnam, with Robert Leonardi and Rafaella Y. Nanetti, Making Democracy Work: Civic Traditions in Modern Italy (Princeton, N.J.: Princeton University Press, 1993).
15. This framework is now fairly standard, but it was not in 1969. Erecting this framework was one of Lewis's greatest achievements. See Lewis, supra note 1.
16. Thomas Schelling, The Strategy of Conflict (Cambridge, Mass.: Harvard University Press, 1960).
17. With colleagues at the University of East Anglia, I have replicated Schelling's studies under more controlled experimental conditions, with large samples and substantial payoffs. Our results confirm people's ability to solve coordination problems. See Judith Mehta, Chris Starmer and Robert Sugden, "The Nature of Salience: an Experimental Investigation of Pure Coordination Game", American Economic Review 84 (1994), pp. 658-673. For our sample, the favoured meeting places in London were Piccadilly Circus and Trafalgar Square.
18. For recent attempts to solve this problem, see, e.g., Michael Bacharach, "Variable Universe Games", in Ken Binmore, Alan Kirman and Piero Tani (eds.), Frontiers of Game Theory (Cambridge, Mass.: MIT Press, 1993); Robert Sugden, "A Theory of Focal Points", Economic Journal 105 (1995), pp. 533-550. Michael Bacharach's theory and mine both have the feature that games are represented in terms of the descriptions that the players themselves use; there is no presumption that those players are playing the game under a common description. These theories differ from Lewis's in not treating salience as a primitive concept, but rather trying to explain it as a result of players' making use of their knowledge about descriptions.
19. For recent attempts to solve this problem, see, e.g., Michael Bacharach, "Variable Universe Games", in Ken Binmore, Alan Kirman and Piero Tani (eds.), Frontiers of Game Theory (Cambridge, Mass.: MIT Press, 1993); Robert Sugden, "A Theory of Focal Points", Economic Journal 105 (1995), pp. 533-550. Michael Bacharach's theory and mine both have the feature that games are represented in terms of the descriptions that the players themselves use; there is no presumption that those players are playing the game under a common description. These theories differ from Lewis's in not treating salience as a primitive concept, but rather trying to explain it as a result of players' making use of their knowledge about descriptions.
20. It is a fascinating puzzle to ask whether these different possibilities really are mutually supporting, or whether they are mutually destructive (as in a proof by contradiction). On the latter view, the logically correct but paradoxical implication is that we have proved by contradiction that reason does not give out (since if it did give out, reason would tell us what to do); so we know that reason tells us what to do, but we don't know what it tells us to do.
21. Nelson Goodman, Fact, Fiction, and Forecast (Cambridge, Mass.: Harvard University Press, 1954).
22. The bravest attempt to solve the problem is probably John C. Harsanyi and Reinhard Selten, A General Theory of Equilibrium Selection in Games (Cambridge, Mass.: MIT Press, 1988). Interestingly, however, Selten does not believe that the analysis of perfectly rational play can tell us much about actual human interaction. He is strongly committed to empirical research into human decision-making, and in this context, is deeply sceptical of rationality assumptions. I suspect that Harsanyi has more faith in the power of human rationality.
23. I was among the pioneers of this approach. See Sugden (1986), supra note 1. But I plead not guilty to most of the charges that I now make against the current form of this literature.
24. See, e.g., Tilman Börgers, "On the Relevance of Learning and Evolution to Economic Theory", Economic Journal 106 (1996), pp. 1374-1385; Tilman Börgers and Rajiv Sarin, "Learning through Reinforcement and Replicator Dynamics", draft manuscript, University College London, 1995 (forthcoming in Journal of Economic Theory).
25. See, e.g., Tilman Börgers, "On the Relevance of Learning and Evolution to Economic Theory", Economic Journal 106 (1996), pp. 1374-1385; Tilman Börgers and Rajiv Sarin, "Learning through Reinforcement and Replicator Dynamics", draft manuscript, University College London, 1995 (forthcoming in Journal of Economic Theory).
26. See, e.g., Robert Boyd and Peter Richerson, "Sociobiology, Culture and Economic Theory", Journal of Economic Behavior and Organization 1 (1980), pp. 97-121. They argue that because the market tends to select firms which maximize profits, there can be selection in favour of employee behavior which, although individually sub-optimal, is collectively optimal (for example, cooperative behavior in within-firm Prisoner's Dilemmas).
27. Peter Taylor and Leo Jonker, "Evolutionarily Stable Strategies and Game Dynamics", Mathematical Biosciences 40 (1978), pp. 145-156.
28. Tilman Börgers and Rajiv Sarin, supra note 1.
29. See Glenn Ellison, "Learning, Local Interaction, and Coordination", Econometrica 61 (1993), pp. 1047-1071; Michihiro Kandori, George J. Mailath and Rafael Rob, "Learning, Mutation, and Long Run Equilibria in Games", Econometrica 61 (1993), pp. 29-56; and Young, supra note 1.
30. See Glenn Ellison, "Learning, Local Interaction, and Coordination", Econometrica 61 (1993), pp. 1047-1071; Michihiro Kandori, George J. Mailath and Rafael Rob, "Learning, Mutation, and Long Run Equilibria in Games", Econometrica 61 (1993), pp. 29-56; and Young, supra note 1.
31. See Glenn Ellison, "Learning, Local Interaction, and Coordination", Econometrica 61 (1993), pp. 1047-1071; Michihiro Kandori, George J. Mailath and Rafael Rob, "Learning, Mutation, and Long Run Equilibria in Games", Econometrica 61 (1993), pp. 29-56; and Young, supra note 1.
32. By using the word 'natural', I mean to exclude experiments in which games are explicitly described to participants in terms of strategies and payoffs, as in game theory itself. The argument of this paper is that game theory fails to take account of the fact that in the games of real life, each person chooses actions under his own descriptions of them, and not under those that the game theorist uses. If this is a weakness in game theory, that will not be picked up in such non-naturalistic experiments.
33. Skyrms, supra note 1, at 93 (italics in original).
34. Id., at 102.
35. This is an example of the class of problems discussed by Goyal and Janssen, supra note 2. Goyal and Janssen show that these problems are crippling for the project of creating a theory of equilibrium selection based on ideal Bayesian rationality.
36. There is also a "knife-edge" mixed-strategy equilibrium.
37. This is not a flippant question. At least in my home town, I think I there are slightly different conventions on the roads at weekday rush hours and at weekends: rush hour drivers seem more ready to give way to vehicles entering busy roads from side streets. By the official rules, traffic on the main road always has priority, so when traffic is heavy, the side-street drivers have to depend on the courtesy of the main-road drivers.
38. This concept of a "pared-down" game corresponds with the concept of an existential game, which I analyze in Sugden (1995), supra note 12. The idea is of a game which is described entirely in propositions of the form "There exist(s) ...", in which no labelling words are used (e.g. "There exist two players ...").
39. Goodman, supra note 14.
40. This temptation leads to the familiar idea that induction can be justified if we presuppose the "uniformity of nature". Even in relation to natural science, I prefer Goodman's more conventionalist approach to "projectibility". Id.
41. A Humean sceptic will immediately ask what I mean by "regularities in the world". But I am not trying to solve Hume's problem of induction. What I am looking for is a psychological theory which explains from outside how human beings make inductive inferences. That theory does not have to be grounded in induction. See Hume, supra note 14; and Goodman, supra.
42. A Humean sceptic will immediately ask what I mean by "regularities in the world". But I am not trying to solve Hume's problem of induction. What I am looking for is a psychological theory which explains from outside how human beings make inductive inferences. That theory does not have to be grounded in induction. See Hume, supra note 14; and Goodman, supra.
43. See John H. Holland, K. J. Holyoak, R. E. Nisbett and P. R. Thagard, Induction: Processes of Inference, Learning and Discovery (Cambridge, Mass.: MIT Press, 1986); John H. Holland, "The Rationality of Adaptive Agents", in Kenneth J. Arrow, Enrico Colombatto, Mark Perlman and Christian Schmidt (eds.), The Rational Foundations of Economic Behavior (Basingstoke: Macmillan, 1996).
44. See John H. Holland, K. J. Holyoak, R. E. Nisbett and P. R. Thagard, Induction: Processes of Inference, Learning and Discovery (Cambridge, Mass.: MIT Press, 1986); John H. Holland, "The Rationality of Adaptive Agents", in Kenneth J. Arrow, Enrico Colombatto, Mark Perlman and Christian Schmidt (eds.), The Rational Foundations of Economic Behavior (Basingstoke: Macmillan, 1996).
45. This might be a way of understanding Putnam's important idea that all forms of social cooperation - even choral societies and amateur football clubs - are mutually supporting, and that networks of civic engagement provide "templates" for future collaboration. Putnam, supra note 8, at 174.
46. Schelling, supra note 10, at 97-98.