Hostname: page-component-cd9895bd7-8ctnn Total loading time: 0 Render date: 2024-12-27T06:06:29.507Z Has data issue: false hasContentIssue false
  • English
  • Français

An Attempt to Simulate the Outbreak of World War I

Published online by Cambridge University Press:  01 August 2014

Charles F. Hermann  and
Margaret G. Hermann
Charles F. Hermann
Affiliation:
Princeton University
Margaret G. Hermann
Affiliation:
Educational Testing Service
Get access Rights & Permissions [Opens in a new window]

Extract

Political games and simulations are models or representations of particular political systems and their associated processes. They are techniques for reproducing in a simplified form selected aspects of one system, A, in some independent system, A′. Games and simulations have a dynamic quality produced by the complex interaction of properties in the model. This feature enables them to generate states of the system that differ radically from those present originally. The kinds of transformations that may occur between the initial and final states of a simulation or game are difficult to represent by other means, despite a diversity in modeling procedures ranging from verbal descriptions to differential equations. Because of their apparent applicability to many problems of politics, as well as their novelty, games and simulations have been developed in a variety of areas in political science. They have been used in research, instruction, and policy formation. Although the application of these techniques has been increasing, systematic evaluation of their performance is only now beginning. This essay reports one type of evaluation.

The researchers sought to structure a simulation of international politics so it would reproduce features of the political crisis that preceded the beginning of the First World War. Two separate trials or runs of the simulation were performed as a pilot project. With two runs, the data are sufficient only to illustrate what might be done in an expanded research program.

Type
Research Article
Information
American Political Science Review , Volume 61 , Issue 2 , June 1967 , pp. 400 - 416
Copyright
Copyright © American Political Science Association 1967

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.)

Footnotes

1

This research was conducted under Contract N123(60530)25875A from Project Michelson, U. S. Naval Ordnance Test Station, China Lake, California. An earlier report on this project was distributed by the contractor as Studies in Deterrence X: Validation Studies of the Inter-Nation Simulation, NOTS Technical Paper 3351, December, 1963. The authors wish to acknowledge their indebtedness to Harold Guetzkow, principal investigator and mentor; Thomas W. Milburn, director of Project Michelson; and Robert C. North and his colleagues at the Stanford Studies in Conflict and Integration who generously shared their document collection and data analysis on the outbreak of World War I. The Center of International Studies at Princeton University supported the first author during the preparation of the present article.

References

2 In addition to the studies cited elsewhere in this paper, the variety is suggested by the following illustrations: Benson, Oliver, “A Simple Diplomatic Game,” in Rosenau, James A. (ed.), International Politics and Foreign Policy (New York: Free Press, 1961), 504–511Google Scholar; Davison, William P., “A Public Opinion Game,” Public Opinion Quarterly, 25 (1961), 210–220CrossRefGoogle Scholar; Abelson, Robert P. and Bernstein, Alex, “A Computer Simulation Model of Community Referendum Controversies,” Public Opinion Quarterly, 27 (1963), 93–122CrossRefGoogle Scholar; Bloomfield, Lincoln P. and Whaley, Barton, “The Political-Military Exercise: A Progress Report,” Orbis, 8 (1965), 854–870Google Scholar; Scott, Andrew M. with Lucas, William A. and Lucas, Trudi M., Simulation and National Development (New York: Wiley, 1966)Google Scholar; Singer, J. David and Hinomoto, Hirohide, “Inspecting for Weapons Production: A Modest Computer Simulation,” Journal of Peace Research (1965), 18–38CrossRefGoogle Scholar; Robinson, James A., Anderson, Lee F., Hermann, Margaret G., and Snyder, Richard C., “Teaching with Inter-Nation Simulation and Case Studies,” American Political Science Review, 60 (1966), 53–65CrossRefGoogle Scholar.

3 The simulation runs were conducted in the summer of 1961 at Northwestern University. The exploratory nature of these runs led the authors to question whether the pilot study should be published. The supply of the original Navy report, however, is now exhausted. Moreover, no more complete set of historical runs has been conducted to date. Because a number of other published materials have discussed these pilot runs, it seems appropriate to make a fuller description of the World War I simulation more widely available. In doing so, the authors wish to caution that the work is primarily an examination of a means of evaluating simulations rather than a direct validation of the Inter-Nation Simulation. For examples of how this pilot project has been discussed elsewhere, see Herzog, Arthur, The War-Peace Establishment (New York: Harper, 1963), esp. 183–184Google Scholar; Verba, Sidney, “Simulation, Reality, and Theory in International Relations,” World Politics, 16 (1964), esp. 507–515CrossRefGoogle Scholar; Robinson, James A. and Snyder, Richard C., “Decision-Making in International Politics,” in Kelman, Herbert C. (ed.), International Behavior (New York: Holt, Rinehart and Winston, 1965), 445, 512Google Scholar; and Singer, J. David, “Data-Making in International Relations,” Behavioral Science, 10 (1965), p. 77CrossRefGoogle ScholarPubMed.

4 Hermann, Charles F., “Validation Problems in Games and Simulations with Special Reference to Models of International Politics,” Behavorial Science, in pressGoogle Scholar.

5 For a discussion of the distinctions made be tween games and simulations, see Hermann, Charles F., “Games and Simulations of Political Processes,” International Encyclopedia of the Social Sciences, in pressGoogle Scholar.

6 Guetzkow, Harold, Alger, Chadwick F., Brody, Richard A., Noel, Robert C., and Snyder, Richard C., Simulation in International Relations (Englewood Cliffs, N. J.: Prentice-Hall, 1963)Google Scholar.

7 The international history, the statements of domestic and foreign policy, and the personality sketches appear as appendices in Technical Paper 3351, op. cit. A complete set of the diplomatic messages in the form in which they were used in the simulation is on file with the contracting agency and with the International Relations Program, Northwestern University. For values assigned the basic parameters, see Table 1

8 Koch, Howard E. with the staff of the Stanford Studies in Conflict and Integration, Documentary Chronology of Events Preceding the Outbreak of the First World War: 28 June-6 August, 1914 (Stanford Univeraity, mimeographed, 1959)Google Scholar.

9 Some illustration of how the criteria were employed is appropriate. The selection of Berthe-lot provides a good example of the application of the first two criteria. Although Berthelot did not have the legitimate authority to make binding decisions for his government, he nevertheless extensively influenced French foreign policy during July of 1914. With the president, premier (who was also the foreign minister), and political director of the foreign ministry on a mission to Russia in July, Berthelot was placed in charge of the foreign ministry. He became the chief advisor during this time to Bienvenu-Martin, acting Premier and Foreign Minister. Bienvenu-Martin, officially the Minister of Justice, was a novice at foreign affairs, and relied heavily on Berthelot. For further evidence on the role of Berthelot, see Challener, Richard D., “The French Foreign Office: The Era of Philippe Berthelot,” in Craig, Gordon A. and Gilbert, Felix (eds.), The Diplomats: 1919–1939 (Princeton, N. J.: Princeton University Press, 1953), 49–85Google Scholar. The third criterion was important in the choice of Conrad. Far less autobiographical and biographical material is available on the Austro-Hungarian political leaders than on those from the other countries. Most information concerns Emperor Franz Joseph, a logical selection in addition to Berchtold for the simulation. However, because Franz Joseph was quite old and recovering from a serious illness, he was not as influential on the decisions as other officials. Among the key figures, Conrad had the most available material.

10 No inter-coder reliability was performed in the content analysis of the personality traits. For this reason, as well as because of the very limited sample of materials that could be examined for each figure, the selected traits must be considered only as tentative approximations. A list of the sources used in the content analysis appears in Technical Paper 3351, op. cit.

11 Sources for the personality categories were Cattell, Raymond B., The Sixteen Personality Factor Questionnaire, rev. ed. (Champaign, Ill.: IPAT, 1957)Google Scholar; Edwards, Allen L., Edwards Personal Preference Schedule (New York: Psychological Corp., 1953)Google Scholar; Gough, Harrison G., California Psychological Inventory (Palo Alto, Calif.: Consulting Psychologists Press, 1956)Google Scholar.

12 Harrison G. Gough, ibid.

13 Osgood, Charles E., Suci, George J., and Tannenbaum, Percy H., The Measurement of Meaning (Urbana, Ill.: University of Illinois Press, 1957)Google Scholar.

14 Personal correspondence from Harrison G. Gough, dated July 25, 1961.

15 For a description of this earlier simulation research, see Brody, Richard A., “Some Systemic Effects of the Spread of Nuclear Weapons Technology: A Study through Simulation of a Multi-Nuclear Future,” Journal of Conflict Resolution, 7 (1963), 663–753CrossRefGoogle Scholar.

16 Geographical features are not incorporated in the Inter-Nation Simulation and no explicit geographical statements were included in the masked historical communiques. Therefore, the participants' maps provided data about the changing conception of relationships among nations. The distance between allied and hostile nations as well as the relative size of the nations represented in the maps were analyzed. Although the measuring device is worthy of further exploration, the results from these pilot runs proved to be quite ambiguous and are not reported here. They are included in Technical Paper 3351, op. cit.

17 The chronology of hostilities during this critical period in 1914 was as follows: July 28—Austria-Hungary declares war on Serbia; July 29—Russia orders and then cancels general mobilization; July 30—Russia again orders general mobilization, France alerts troops along German border; July 31—Austria-Hungary begins general mobilization, Germany issues ultimatum demanding Russia stop mobilizing within 12 hours or Germany will mobilize; August 1—France and Germany start general mobilization, Germany declares war on Russia. After August 1, formal declarations of war follow in quick succession from the other major European states.

18 It is to be understood that the references to nations here are to the student decision makers in the simulated nations.

19 See, for example, Albertini, Luigi, The Origins of the War of 1914, edited and translated by Massy, Isabella M. (London: Oxford University Press, 1953), Vol. 2, p. 514Google Scholar; and Fay, Sidney B., The Origins of the World War (New York: Macmillan, 1930), Vol. 2, p. 556Google Scholar.

20 Luigi Albertini, ibid., Vol. 2.

21 Dina A. Zinnes, Robert C. North, and Howard E. Koch, “Capabilities, Threat and the Outbreak of War,” in Rosenau, op. cit., p. 470.

22 Ibid., p. 472.

23 The inter-coder reliability for the simulation content analysis of hostility was .82 and for capability .83. For a description of the statistics used in this paper (the sign test, rank-order correlation, and Mann-Whitney U test) see Siegel, Sidney, Nonparametric Statistics for the Behavioral Sciences (New York: McGraw-Hill, 1956)Google Scholar.

24 Zinnes, Dina A., “A Comparison of Hostile Behavior of Decision-Makers in Simulate and Historical Data,” World Politics, 18 (1966), p. 477CrossRefGoogle Scholar. This article is one of the few efforts to explore simulation validity by what we have described as the hypothesis approach. Statistical tests are conducted on hypotheses using data from World War I and from another series of Inter-Nation Simulation runs which made no attempt to replicate those historical events.

25 The 1914 data for this analysis were obtained from Zinnes, ibid. She performed a similar statistical test using Serbia as well as the five nations represented in the simulation. It is interesting that with the addition of Serbia, the result is not significant.

26 In retrospect, this questionnaire item was not totally satisfactory. Estimating the difference between situations rated “exactly alike” and those rated “somewhat alike” is difficult. Because most participants checked several historical situations, it is not clear how much more the simulation resembled the 1914 crisis than, for example, Hitler's ultimatum to Poland. Furthermore, participants were not asked to state when they became aware of the apparent similarity between the simulation and a past event. Some evidence indicates that a messenger suggested the parallel to World War I to several individuals near the end of the M-run. An unstructured question, which provided no list of historical events, might have reduced the number of references to World War I. The alternative situations in the questionnaire item were selected from events mentioned by high school students who pretested the material for the input phase.

27 The messengers in both runs also completed this questionnaire item. Six out of nine messengers reported a similarity to World War I. It may be that the overall view of events provided by this role increased their awareness of the similarities.

28 For further discussion of the level of specificity issue, see Guetzkow, Harold, “Simulation in International Relations,” in Proceedings of the IBM Scientific Computing Symposium on Simulation Models and Gaming, 12, 1964, Thomas J. Watson Research Center, Yorktown Heights, New York, esp. pp. 264–267Google Scholar.

29 Holland, Edward P. with Gillespie, Robert W., Experiments on a Simulated Underdeveloped Economy (Cambridge, Mass.: MIT Press, 1963), 207–208Google Scholar.

30 Pool, Ithiel, Abelson, Robert P., and Popkin, Samuel L., Candidates, Issues, and Strategies (Cambridge, Mass.: MIT Press, 1965) rev. ed., p. 64Google Scholar.

31 Sidney Verba, op. cit., 511, 513.

32 For a survey of various validity approaches and a discussion of their assets and liabilities, see Charles F. Hermann, Behavioral Science, op. cit.

Submit a response

Comments

No Comments have been published for this article.