Hostname: page-component-cd9895bd7-p9bg8 Total loading time: 0 Render date: 2024-12-25T17:53:38.728Z Has data issue: false hasContentIssue false
  • English
  • Français

Modelling complex conflicts: formalism or expertise?

Published online by Cambridge University Press:  26 October 2009

Peter G. Bennett
Get access Rights & Permissions [Opens in a new window]

Extract

The use of game theory to study international conflict has seen peaks and troughs of enthusiasm, high hopes alternating with disillusionment. A revival in recent years is evidenced by the wide interest aroused by non-technical, accessible applications—notably Axelrod’s work on cooperation. Meanwhile, the increasing use of computers to analyze many types of decision lends credence to the notion that with this new-found analytical power, one can move beyond the simplified models previously on offer. Indeed, several pieces of game-based software are now commercially available. The enthusiastic might claim that we stand on the threshold of an age in which no important decision involving conflict should be made without the backing of powerful, on-line analysis. The even more enthusiastic (of whom more below) are prone to claim that the threshold has already been crossed. However, there remain those for whom such claims cut very little ice, and there Is little consensus amongst theorists as to what sort of practical help can be offered to decision-makers.

Type
Research Article
Information
Review of International Studies , Volume 17 , Issue 4 , October 1991 , pp. 349 - 364
Copyright
Copyright © British International Studies Association 1991

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Axelrod, R., The Evolution of Cooperation (New York, 1984)Google Scholar. See also, for example Nicholson, M.: Formal Theories in International Relations (Cambridge, 1989)Google Scholar; Zagare, F. C., (ed.): International Interactions, 15 (1990)Google ScholarSpecial issue on modelling international conflict.

2 As will be seen below, this illustrates a more general point about the helpfulness (or otherwise) of game models in practice.

3 Howard, N., Paradoxes of Rationality (Massachusetts, 1971).Google Scholar

4 Howard, N., ‘The Usefulness of Metagame Analysis’, J. Opi Res. Soc, 37, (1986) 430–432Google Scholar. Howard, N., ‘The Present an d Future of Metagame Analysis’, Euro. Jnl of O. R. 32, (1987) 1–25.Google Scholar

5 Fraser, N. M. and Hipel, K. W., Conflict Analysis: Models and Resolutions (Elsevier, 1984).Google Scholar

6 Illustrating a similar point, the ‘hawks’ in Ex Comm pressed hard for the plan of disarming the Turkish missiles, then launching an air strike followed by an invasion. See Sorensen, T., The Kennedy Legacy (New York, 1969).Google Scholar

7 Brams, S., Superpower Games (New Haven, 1985).CrossRefGoogle Scholar

8 See e.g. Kreps, D. M. and Wilson, R., ‘Sequential Equilibria’, Econometrica, 50 (1982), pp. 863–94.Google Scholar

9 Thomas, L. C., ‘Using Game Theory and its Extensions to Model Conflict’, in Bennett, P. G. (ed.), Analysing Conflict and its Resolution (Oxford, 1987).Google Scholar

10 For some differing evaluations, see, for example Janis, I., Victims of Groupthink (Boston, 1972)Google Scholar; Bernstein, B. J., ‘We Almost Went to War’, Bulletin of the Atomic Scientists, 32 (1976), pp. 13–21Google Scholar; Berresford, A. and Dando, M. R., ‘Operational Research for Strategic Decision-Making: the Role of World-View’, J. Opl Res. Soc, 29 (1978), pp. 137–146.Google Scholar

11 Bennett, P. G., ‘Toward a Theory of Hypergames’, Omega, 5 (1977), pp. 749–751Google Scholar; Bennett, P. G., ‘Hypergames: Developing a Theory of Conflict’, Futures, 12 (1980), pp. 489–507.Google Scholar

12 Bennett, P. G., ‘Beyond Game Theory-Where?’ in Bennett, (ed.), Analysing ConflictGoogle Scholar. This paper used the hypothetical example of leaders who saw world events in terms of‘actors’ defined by their religion, and the mutual difficulties in understanding that would arise in dealings with those used to thinking in terms of national or class-based actors. Recent events suggest that the case is not so hypothetical.

13 Harsanyi, J. C., ‘Games with Incomplete Information Played by “Bayesian” Players’, Management Science, 14 (1967), pp. 159–82.Google Scholar

14 ‘If you hadn’t acted, you would have been impeached’, Robert Kennedy is quoted as telling J. F. Kennedy. Abel, E., Missile Crisis (Philadelphia, 1968)Google Scholar. See also, inter alia, Allison, G. T., Essence of Decision: Explaining the Cuban Missile Crisis (Boston, 1971)Google Scholar.

15 Bernstein, , ‘We Almost Went to War’.Google Scholar

16 Examples include Brams, Superpower Games, Snyder, G. H. and Diesing, P., Conflict Among Nations (Princetown, 1977)Google Scholar; and Zagare, F. C., The Dynamics of Deterrence (Chicago, 1987)Google Scholar. The last has been roundly criticized by Reynolds on this and other grounds: Reynolds, C., Review of International Studies, 15 (1989), pp. 67–74.Google Scholar

17 Bennett, P. G. and Cropper, S. A., ‘Helping People Choose: Conflict and Other Perspectives’, in Belton, V. and O’Keefe, R., (eds.), Recent Developments in OR (London, 1986)Google Scholar; Bennett, P. G., Cropper, S. A. and Huxham, C. S., ‘Modelling Interactive Decisions: The Hypergame Focus’, in Rosenhead, J. (ed.), Rational Analysis for a Problematic World (New York, 1989)Google Scholar; Radford, K. J., Strategic Planning, an Analytical Approach (Reston, 1980).Google Scholar

18 Burton, J., ‘World Society and Human Needs’, in Light, M. and Groom, A. J. R., (eds.), International Relations: a Handbook of Current Theory (London, 1985), p. 50Google Scholar. See also, for example, his ‘Conflict Resolution as a Political System’, Center for Conflict Analysis and Resolution, George Mason University, 1988.Google Scholar

19 Smail, D., Taking Care: an Alternative to Therapy (London, 1987), p. 33.Google Scholar

20 Braybrooke, D., ‘Reconciling Attention to Needs with Attention to Preferences’, 3rd International Conference on Metagame Theory, York University, Toronto, July 1977, p. 1.Google Scholar

21 In all its apparent contradictions, the typical public stance can be paraphrased thus: ‘Even if I wanted t o give way on this, the people wouldn’t stand for it, and anyway there’s no way in which I’d compromise their needs. I’m more than willing to be reasonable: my door is open at any time, but the other side was responsible for the breakdown of talks, and it’s up to them to make the first move … if they’re prepared to come up with something new, we’d be prepared to listen. We realise that the conflict is damaging to all, but there are some points that are just not negotiable …’.

22 Schrodt, P. A., ‘Pattern Matching, Set Prediction and Foreign Policy Analysis’, in Cimbala, S. J. (ed.), Artificial Intelligence and National Security (Lexington, 1987), pp. 89–107.Google Scholar

23 Schrodt, , ‘Pattern Matching’, pp. 90–1Google Scholar. His reported predictions were as follows:

Egypt: Will protest due to violation of sovereignty Can’t [won’t] do much else because of dependence on U.S. aid

Did not help with the interception (U.S. should emphasise that)

This isn’t very good for Mubarak and will result in protests.

Italy: Will not extradite hijackers

There will be additional attacks on Italians

Will release the remainder of PLO [i.e., Abdul-Abbas]

PLO: Will protest interception

Won’t do much else [in direct response] because:

(a) Arafat claims this operation not endorsed by PLO

(b) The hijackers aren’t very nice—they are small potatoes and not too respectable to Palestinians generally.

Won’t try hijacking ships again.

24 Bueno de Mesquita, , The War Trap (New Haven, 1981)Google Scholar.

25 Thus (though Mesquita is using decision-theoretic rather than game-theoretic models)—it would be acceptable to have Kennedy and Khrushchev as the ‘players’ in the Chicken game, but not ‘White House’ and ‘Kremlin’, and certainly not ‘US’ and ‘USSR’. Some further ramifications of this issue are taken up in the following section.

26 Binmore makes the point well: ‘A frequently-heard complaint is that “game theory doesn’t work”. Such complaints reflect the gulf that exists between theoreticians with long-term scientific aims and practical men concerned with resolving immediate questions of policy. … The difficulty … lies in the fact that men of affairs require answers now and in a form they find acceptable. Abstruse answers offered in the indeterminate future are therefore of little help to their advisers, who must cobble together what models they can from whatever material is available’. (‘Why Game Theory “Doesn’t Work”’, in Bennett, , (ed.), Analysing Conflict, pp. 23–24Google Scholar.

27 Since he admits—indeed stresses—the political context of leaders' decisions, Bueno de Mesquita’s argument for concentrating on individuals really translates into a recommendation for the latter approach.

28 Bennett and Cropper, ‘Helping People Choose’; Bennett, Cropper and Huxham, ‘Modelling Interactive Decisions’; Bennett, P. G., ‘Hypergames as an aid to mediation’, in Mitchell, C. and Webb, K. (eds.), New Approaches to International Mediation (Greenwood, 1988).Google Scholar

29 The classic work of this kind is of course Schelling, T. C., The Strategy of Conflict (Harvard, 1960).Google Scholar

30 Thus Schrodt, for example, uses the term ‘formal’ to cover both knowledge-based as well as purely deductive approaches.

31 Schrodt, , ‘Pattern Matching’, p. 91.Google Scholar

32 Bueno de Mesquita, , The War Trap, p. 3.Google Scholar

33 Milburn, T. W., ‘The Management of Crises’, in Hermann, C. (ed.), International Crises (New York, 1972).Google Scholar

34 de Reuck, A. V. S., ‘The Logic of Conflict’, in Banks, M. (ed.), Conflict in World Society (Brighton, 1984).Google Scholar

35 Nicholson, M., ‘Aborted Wars and “Prosaic” Utility Functions’, University of Manchester, 1986.Google Scholar

36 See, for example, Huxham, C. S. and Dando, M. R., ‘Is Rounded-Vision an Adequate Explanation of Strategic Decision-Making Failures?’ Omega, 9 (1981), pp. 371–9.Google Scholar

37 Jervis, R., ‘Realism, Game Theory and Cooperation’ World Politics, 40 (1980), pp. 317–49.Google Scholar

38 Howard, , ‘Usefulness of Metagame Analysis’, and ‘The CONAN Play’, in Rosenhead, (ed.) Rational Analysis.Google Scholar