Olympic Games Competition:

Structural Correlates

of National Success*

DONALD W. BALL

Univerisity of Victoria, British Columbia, Canada

Olympic Games Competition: Structural Correlates of National Success

THERE ARE innumerable sports, games, and contests throughout the modern world, but pre-eminently at the international level the games are the Olympic Games; Baron de Coubertin's turn-of-the-century resurrection of the ancient Greek spectacles. Although supposedly - and as reiterated by Inter- national Olympic Committee officials (cf. Brundage, 1956: 35) - but competi- tions between individuals, the games are in fact, as is inter-nation sport in general (cf. Heinila, 1966; McIntosh, 1963: 89-93, 188-203; Morton, 1963: esp. 65-103), an arena of intense between-polities contesting. Overall team performances become sources of national pride and shame as countries vie with one another for the prestige associated with high-ranking total performances. If war can be, as it has, likened to a game (Avedon and Sutton-Smith, 1971: 271-303; Liddell Hart, 1971), the Olympic Games may be likened to a qua- drennial war, with medals and team points and international reputations rather than territory as the victor's spoils. In the following the concern will be with some of the possible social structural correlates, national indicators, of these Olympic Game outcomes.

Perspective

Each Olympics brings with it new reports of heightened international con- flicts and disputes (not to mention intranational ones, e.g. Edwards, 1960; Kieran and Daley, 1969: 410, 431, 433-435, 444): accusations of illegal * An earlier version of this paper was prepared for the session on the Sociology of Sport,

American Sociological Association, Denver, September, 1971. Support for this analysis has been provided by a Faculty Research Grant (08 518) from the University of Victoria.

187

subsidies and financial aid, training methods, and the like; solicitations of political asylum; and since 1952, alternative methods for scoring team perform- ances (Kieran and Daley, 1969: 318, 382-383; Chester, 1968: 155) - a practice officially barred by international rules at the insistence of the modern Games' founder, de Coubertin (Morton, 1963: 93) in the 1890's.

The examples of such politically implicated "incidents" are legion. Among the better known: 1936: The "Nazi Olympics," Hitler's attempt to exploit the Games as a

propaganda showcase (Mandell, 1971; Chester, 1968: 82-91; Kieran and Daley, 1969: 153-181).

1.952 : Against U.S. opposition, the Soviet Union enters a team in the Summer Games (Chester, 1968: 104; Morton, 1963: 34-35, 79-91); the West and the IOC fear communist-inspired demonstrations; a two-Chinas jurisdictional dispute (Kieran and Daley, 1969: 219-227, 232ff, 265- 266, 231-232, 277). The Games end with P claiming the "world superiority" of the Soviet athletes, and asserting that they had won more medals than any other country (Kieran and Daley, 1969: 265-266; Morton, 1963: 85-86; the actual count was U.S.A. 41 Golds to U.S.S.R.'s 23, and U.S.A. 75 total medals to the U.S.S.R.'s 68).

1956: The Suez invasion and the Hungarian uprising are reflected in the Games (Chester, 1968: 119-121 ), e.g. some Arab nation teams withdraw in protest of the Suez actions (Kieran and Daley, 1969: 280-281).

These examples could easily be multiplied, including some involving actual physical violence, e.g. the free-for-all between the U.S.S.R. and Hungarian water polo teams in 1956 (Morton, 1963: 87-88; Kieran and Daley, 1969: 316-317), as well as those featuring intricate political maneuvers such as characterized the Tokyo Games in 1964 (pages 375-377); the Games in which the IOC was finally successful in their efforts to ban the tabulation of unofficial, point-based national standings (pages 318, 383-384; Chester, 1968: 155). Albeit unsystematically and without social-scientific rigor, Atkinson's only- somewhat-fictional account of the 1968 Olympics, dramatically captures the essence and spirit of this international conflict in his novel, The Games (1968: esp. 5, 24, 123, 333-334, 342, 380-381, 383-384).

Although the Games are ostensibly competitions between individuals (or teams in particular sports), this is belied by the attention given to the perform- ance of their nationals by governments and quasi-governmental agencies as well as the press and the public (Morton, 1963; McIntosh, 1963; Heinila, 1966, and the citations therein). Whether financed via public appeal or state subsidy or support, polities are both overtly and covertly concerned with the outcomes of the Olympics - as well as "the game within the games" which surrounds the jockeying for site-selection and support.

Duncan (1962) has proposed a model of social order, originally formulated for communities and societies, which seems equally useful as an interpretive

188

scheme for international social order. Essentially, Duncan argues that social order arises in, is sustained by, and is altered through the symbolic exchanges of hierarchically ranked superiors, equals, and inferiors. In other words, a central constituent of order, disorder, and change is a kind of "status-contesting," as members of the system attempt to maintain and/or alter their rank via the strategic presentation of symbols. And a symbolic resource potentially available to members of the international order is overall team Olympic Game perform- ance. Thus, the concern of nation-states to be, in the language of gamesman- ship, "one-up," thereby making their others "one-down" (Potter, 1948; Ball, forthcoming, and the references by Potter cited therein). Further, given the popularity of sport in general, and the Olympics in particular, this makes such performances readily meaningful to relevant constituencies, both domestic and foreign. The Olympics provide a useful arena for the production and distribu- tion of the symbols of successful international status-contesting. They become a stage where is played the drama of the politics of symbolic conflict (for a basically domestic account of symbols and politics, see Edelman, 1964).

It would be a mistake, then, to dismiss the Olympics as "mere games," for although the extent is unknown - if ever knowable - they may serve as sur- rogates for other, possibly more destructive types of international conflict (Natan, 1958), where not only gains but losses too, might be of much greater magnitude. As long as the Olympic Games exist as a potential alternative and symbolic mode of between-nations conflict, they are worthy of serious scientific consideration.*

Such consideration is, of course, contrary to the "spirit of the games" even if consistent with a sociological concern with sport and politics. Thus, Avery Brundage has programmatically reiterated the Olympic sentiment that

The Games are not, and must not become, a contest between nations which would be entirely contrary to the spirit of the Olympic Movement and would surely lead to disaster. For this reason there is no official score of nations and tables of points are really misinformation because they are entirely inaccurate. To be correct they would have to be weighted since it is certainly unfair to give the winner of the marathon or decathlon, a winning gymnast, pistol shooter or yachtsman, and a winning football soccer or basket- ball team the same score. Moreover, the factor of population should be considered...

...Neither the Olympic Games nor any sport contest can be said to indicate the superiority of one political system over another, of one country over another... The IOC resents attempts to use the Games as a political instrument or to pit one country against another. We trust that you will do everything in your power to discourage the publication of scoring tables, which are quite worthless... (Brundage, 1956: 35).

Problem Noble as the sentiment may be, that the claim by IOC officials that the

Games are not competitions between nations but between individuals, appears patently false is obvious to any who have followed the Olympics (regardless of

* The writing of this predates, but is not superceeded by the terrorist attack at the 1972 Games in Munich.

189

the particular media employed). Brundage himself, in calling attention to "the factor of population" suggests otherwise. Thus, albeit paradoxically, both the official Games ideology and Brundage's own qualification provide a point of departure by suggesting a testable proposition.

In the null version the proposition is this: indicators of national structural variables, e.g. demographic, economic, and political, will bear no significant relationships to total national Olympic Game scores. Thus, unless there are relationships between the structure of nation-state polities and the overall team performance of their Olympic athletes, we may take the International Olympic Committee officials at their word regarding the purity and apolitical nature of the Games; that is, that they involve individual competition, not international rivalries and competition. The null is the official position of the IOC; its rejec- tion is consistent with the perspective advanced in the citations and paragraphs above, as well as a more general view predicting systematic outcomes of Olympic contesting.

As caveats, two possible "exceptions to the rule" of the proposition (in its null form) might be predicted: one naive, the other somewhat more socio- logically sophisticated. The first, the naive one, as noted by Brundage, would relate to sheer population size. This would be a simple prediction that national Olympic success and population size would be directly related: i.e. that ceteris paribus, the larger the population, ergo the larger the pool of potential winners in any event from which to select team-members (cf. Brundage, 1956). The second exception which might be expected would have to do with national economic status: the "richer" the country, the better the team performance vis-a-vis other national teams; this due to nutritional opportunities, availability of facilities and coaching, et cetera, and a host of wealth-related variables too numerous to catalogue (McIntosh, 1963: 134-160). In short, relations obtain- ing in contradiction to the null should be based upon resources alone, i.e. the greater the resources, the more successful the team - but essentially because the presence of these resources, human and economic, increases the probability of having successful individuals ( qua team-members). Big and rich countries should be more successful than little and poor nations just because they are big and rich.

Otherwise, if the Game's establishment's claims are correct, relationships should be few and unsystematic; if they are wrong a pattern of structural relationships should obtain between nations and their overall team scores in the Olympic Games. Contrary to the proposition then, conservatively it might be hypothesized that there are systematic relationships between variables indexical of the structure of national social, economic, and political arrange- ments and total national Olympic Game scores.

Method

The nation-state polities fielding Olympic teams are many, as are the indicators and sources of information concerning their structural variation. The

190

problem at hand dictates recourse to comparative, cross-national analysis (Marsh, 1967; Etzioni and DuBow, 1970). A major, codified source of such structural data on the nations of the world is A Cross-Polity Survey, by Arthur S. Banks and Robert B. Textor, which compiles information on almost 60 structural variables for all of the world's "independent polities... as of April 1, 1963" (Banks and Textor, 1963: v).

The indicators employed by Banks and Textor are diverse, and include measures of population size, composition, and social characteristics; economic indices; and political structure, styles, and behaviors. These measures are then dichotomized (see Chart 2) and run, each one against all others. The result is a computer printout of all 2 X 2 tables each equal to or less than .10 in terms of significance level as determined by the Fisher's Exact Test (for a more complete description of their methods, e.g. coding, dichotomizing, see Banks and Textor, 1963: 1-53; and for the sources and codes themselves, pages 54- 117). Consistent with their format, this analysis begins by running dichotomized national scores for the 1964 Summer Games against the data coded in the Cross-Polity Survey (however, the .10 level of acceptance/rejection is ignored). The 1964 Olympics were chosen for analysis since they were the Games closest to the Banks and Textor cut-off date of April, 1963, and would thus minimize social change induced errors.

Scoring the Games

Traditionally there have been two competing methods used in calculating national team scores, the North American and the European. The former assigns ten points for a first place, and five through one for succeeding places ( 10-5-4-3-2-1 ) . This method, which gives greater weight to first places, has been favored by U.S. Games-watchers. The European scheme (7-5-4-3-2-1) has been endorsed by the Soviet Union (Menke, 1969: 738; Morton, 1963: 86-87). In the Games themselves, however, only three places are officially recognized through the awarding of medals : first = gold, second = silver, and third = bronze. In this analysis a scoring method based upon awarded medals only is followed. Team scores are calculated on a basis of summating the following: gold medal = 3 points, silver medal = 2 points, and a bronze medal = 1 point (see Chart 1).

Since some of the Banks and Textor variables were originally coded on more than two attributes, e.g. high, medium, and low, it is necessary to collapse them, as did Banks and Textor themselves, into dichotomies to make them congruent with the Cross-Polity format. Chart 2 indicates these relabelled di- chotomous variables constructed from the Banks and Textor codes (these desig- nations make cutting points apparent to anyone consulting the originals (pages 54-117).

The results of the extension of the Banks and Textor data are shown in Table 1, where 55 dichotomous comparisons of team performance and national indicators are presented. Since some of these measures bear but dubious inde-

191

Chart 1

Nations above and below median on 1964 summer Olympic game Scores (scores in parentheses) a

a This is a sample (N = 36) of an original universe of 41 nations, the scores of which are available in the press, and summarized in the World Almanac for 1965, pages 182-183. Scoring is for medal winning only: Gold = 3, Silver = 2, and Bronze = 1; this differs from some systems which score for up to the first six places. Units in the universe but excluded from the sample are (a) Bahamas and Kenya, information for which, i.e. national indicators, are not available in Banks and Textor (1963), (b) Germany and Korea, divided into two nations except in Olympic competition where a single team was entered, and (c) Great Britain, a sub-unit of the United Kingdom. Scores, for informational purposes, for these units were: Bahamas = 3, Germany = 87, Great Britain = 37, Kenya = 1, and Korea = 5. For the sample the range of scores above the median is 188-10, and below the median it is 8-1.

b The host nation: Typically hosts are more successful, at least in part because of their ability to enter larger than usual teams at relatively low financial expenditure.

pendence to one another, it is not completely legitimate to accept or reject particular comparisons of one finding versus another on the basis of significance levels. It is, however, useful to look at the overall pattern of significant findings; further this avoids the arbitrary setting of levels by the writer, while leaving readers free to make their own decisions (although in the text, "significant" will refer to values equal to or less than .05). Of the 55 tests, the cumulative percentages of those reaching the various (one tailed) levels - which under conditions of perfect independence would be the expected proportions - are:

P value < . 01 < . 05 <,15 <.20 <.25 :5.30 :5.40 < . 50 > . 50

Cum. % 9% 38% 47% 53% 62% 71% 75% 76% 84% 91% 100% (N/55) 5 21 26 29 34 39 41 42 46 50 55

192

Chart 2

Names of attributes of indicators

Table Top Row Bottom Row No.

2 High (300,000 square miles or more) Low (less than 300,000 square miles) 3 High (17,000,000 persons or more) Low (less than 17,000,000 persons) 4 High ( 100/square mile or more) Low (less than 100/square mile) 5 High (2% per year or more) Low (less than 2% per year) 6 High (20% of pop. in cities of 20,000 or Low (less than 20% in cities of 20,000 or

more; 12.5% in cities of 100,000 more; less than 12.5% in cities of population or more) 100,000 population or more)

7 High (34% or more) Low (less than 34%) 8 High ($ 5,000,000,000 or more) Low (less than $ 5,000,000,000) 9 High ($ 600 and above) Low (less than $ 600)

10 High (U.N. assessment at least 0.25% of Low (less than 0.25% of total) total)

11 I High (near self-sastaining growth rate or Low (not currently near being self-sus- better) taining)

12 High (50% or more) Low (less than 50%) 13 Complete (at least intermittently) Absent (either internally or externally) 14 High (100 or more) Low (less than 100) 15 Substantially Christian Christianity not important 16 Homogeneous Heterogeneous 17 High (90% or more either caucasoid, Low (less than 90% in any of three

mongoloid, or negroid) categories) 18 High (85% or more have same native Low (less than 85% has same native

tongue) tongue) 19 Through 1913 Since 1914 20 Historically (or significantly so and not No (or partially through colonial domina-

a former colony) tion) 21 1 See Table 1 See Table 1 22 European (or derived, e.g., through colo- Non-European (autochthonous, develop-

nization) ed and non-developed tutelage) 23 Advanced Transitional 24 Doctrinal and developmental Conventional and traditional 25 High mobilization of resources Limited or non-mobilization of resources 26 Constitutional and authoritarian Totalitarian 27 High (stable since World War I) Low (limited or unstable since WWI or

before) 28 Polyarchy (at least limited) Non-polyarchic (or pseudo-polyarchic) 29 Competitive Non-competitive 30 Autonomous (at least partial) Low autonomy 31 High (re: integration, homogeneity, fac- Low (re: integration and homogeneity;

tionalism, etc.) high factionalism or opposition) 32 Extreme (sectional groups present and Low (sectional groups, if present, are less

active) active) 33 Significantly present Less significant to absent 34 Significantly present Less significant to absent 35 Significantly present Less significant to absent 36 Frequent or occasional Infrequent or rare 37 Significant Negligible

193

Chart 2 (continued)

Table Top Row Bottom Row No.

38 Significant Limited 39 Significant Limited or negligible 40 Significant Limited or negligible 41 One-party (or one dominant) More than one party 42 Territorial, regional, or ethnic Aggregative or class-based 43 High stability Low stability or non-stable 44 Pronounced or moderate Negligible or absent 45 Elitist Moderate or non-elitist 46 Pronounced to moderate Negligible 47 Federalistic forms Unitarian forms 48 Significant Limited or negligible 49 Presidential Parliamentary, monarchial, or commu-

nist 50 Effective Ineffective 51 Unicameral Bicameral 52 Dominant or strong Weak 53 Modern Semi-modern or transitional 54 Interventive or supportive Neutral 55 Politically significant Not significant 56 Civil or common (or combined) Other (e.g. Muslim, Scandanavian, Com-

munist, mixed) 57 Member (or quasi-member) Non-member 58 High (wants: needs ratio of satisfaction) Low (wants: needs ratio of satisfaction) 59 High rate of change Low rate of change 60 High (low stability) Low (high stability) 61 1 Traditional and transitional Modern

In the aggregate these data suggest that the null form of the proposition be rejected; i.e. that which proposes that a relationship between national structural indicators and teams success does not exist. On the whole, these data argue persuasively otherwise, e.g. with almost four of ten tests reaching the .05 level. Although the level of measurement is admittedly crude and independence dubious, the rejection of the null of no relationship seems completely warranted given the magnitude of the findings. Thus, the exploration of the positive relationships involving the structural correlates of national Olympic Games success appears worthy of continued investigation. Keeping in mind the prob- lems of the data, it still seems worthwhile to examine some of the sub-sets and/or particular findings as shown in Table 1, even if only descriptively.

An overview of the 2 X 2 data

The following will, because of their number, but briefly describe relation-

194

Table 1

Summary of 2 x 2 analyses of national indicators and Olympic scores

195

Table 1 (continued)

a Numbers 2-57 are equivalent to the "Raw Characteristic" variable numbers in A Cross- Polity Survey by A. S. Banks and R. B. Textor (Cambridge: The M.I.T. Press), 1963, described in pages 54-117.

b Cutting points of the dichotomy were established by visual inspection, using two criteria: (a) the attempt to minimize marginal variation, and (b) to avoid violating logic and/or theory, e.g., a coding of none versus some or much. These criteria are not always independ- ently satisfiable.

c For names of attributes of national indicators (labels of rows), see Chart 2. d This measure reaches unity with a zero value in any cell.

ships between the 55 various national indicators drawn from Banks and Textor and 1964 Summer Games team performance.

Demography and Ecology Neither size in area (2),* population (3), nor population density (4) are

significantly related to team success; thus providing disconfirmation of the "naive caveat" mentioned above. Population growth rate (5), however, is both significantly and negatively related to national team performance.

In terms of population composition, high scores are associated with ur- banization (6, 7), a high rate of literacy (12), linguistic (18) and racial homo-

* Numbers in parentheses refer to the "raw characteristic" variable numbers used by Banks and Textor, and also appearing in the first column of Table 1. Because of space limita- tions, the reader is urged to consult the more complete description of the variables available in their work (pages 54-117).

196

geneity (17), but not religious homogeneity (16), although countries charac- terized by western as contrasted to non-western faiths (15) are the more success- ful in the Games. In sum, westernized relatively homogeneous polities, with low population growth-rates and high rates of literacy, are those for which Games success may be predicted, where additionally there is a high rate (14) of newspaper circulation.

Economy As expected, economic prosperity is strongly related to team performance.

A high gross national product, both nationally (8) and on a per capita basis (9), international financial status (10), i.e. size of assessment for United Nations support, and a developed, self-sustaining economic order (11) all are positively and significantly associated with national Olympic success. Thus, the second "exception to the rule" is confirmed, i.e. successful nations in Olympic competi- tion are those which can afford it.

The Polity As the title would suggest, the Banks and Textor data weighs heavily to-

ward variables characterizing the political system. Polities successful in Olym- pic competition are those which date their independence prior to World War I (19), i.e. the older established (read western) nations of the world.

In terms of political structure (20-32), high performance is related to: being historically western (20), being modernized via European influence (22) and in an advanced stage of this process (23) ; having an explicit doctrinal ideology (24) leading to a high degree of political mobilization of resources, human and otherwise (25); a totalitarian rather than a constitutional authority (26), but one which is stably entrenched (27). Further, neither degree nor form of re- presentation (28, 29), nor freedom of group opposition (30) are significantly related to national team performance, though all three are negative in direc- tion of association. Finally, high political enculturation (31), i.e. little or no extreme opposition, factionalism, and so on, along with a lack of sectionalism (32) are both related to Olympic success. Structurally, high performance polities are modernized and westernized, with extremely centralized govern- ments and highly integrated members.

As regards political competition (33-43) : the articulation of the demands of interest groups (33) such as trade unions, industrial organizations; institutional groups (34) e.g. within the governmental bureaucracy; non-associational groups (35), for instance ethnic and religious blocs; and anomic groups (36), those lacking organization or structure over time; and political parties (37) - all but the first are negatively related to team scores, only the non-associational groups at a significant level. Similarly, the presence of aggregated constit- uencies shows little association with Olympic performance, whether the referent is political parties (38), the executive (39), or the legislature (40). Generally

197

then, both the success and size of various interest-group constituencies shows little relationship to Games scores, what there is more often being negative in direction.

Three measures are devoted specifically to party systems (41-43). Quan- titative aspects (41), e.g. one party, one-dominant, two, et cetera, and party stability (43) show little association with team success. Having a Communist- dominated party system (42) on the other hand, indicates a strong positive relationship with Olympic success.

Another dimension of the polity is the position of leaders. Personalismo, the tendency for partisans to follow a leader for non-ideological reasons (44) is negatively but not significantly related to Game scores, as is leadership char- isma (46) ; while elitist leader recruitment (45) is both positive and significant in its association.

Lastly, the question of political power forms and distributions and national team success: in terms of vertical power distribution (47), federalism is related to low scores, unitary forms to high ones; horizontal distributions show no relationship at all to team performance. A weak executive as contrasted to legislatures, parliaments, or Communist systems (49) as well as Communist Bloc membership (57) are significantly associated with successful Olympic team scores. Finally, taken by themselves, neither current status of the legis- lature (50), executive (52), nor the character of the legislature (51) or the bureaucracy (53) show particularly strong relationships with team scores. Neither do the possibilities of military involvement in the polity (54), nor that of the police (55) or the character of the legal system (56).

Summary

Overall, the successful nation-state in Olympic competition would appear, in a kind of visual factor analysis, to be: stable and homogeneous in population, literate, modern and western, with little institutionalized domestic political competition, economically prosperous, characterized by a strong central government staffed by an elite, and probably (<.05) a member of the Com- munist Bloc. Such is the summary portrait which emerges.

It should be remembered that (1) these data are located in time; i.e. 1963 for national indicators, summer of 1964 for Olympic team scores; and (2) very wasteful in terms of information thrown away, i.e. by dichotomizing variables in many cases susceptible to more than nominal and/or ordinal measurement. Not only Games scores, but several national indicators are capable of higher, more sensitive levels of measurement, which sacrifice less information.

Discussion

The frequency, directions, and magnitudes of the Banks and Textor derived data strongly argue for the utility of using nations as units of analysis in the study of sport, politics, and Olympic Game competition; the disclaimers of

198

the IOC to the contrary. If the Games may be surrogate for other forms of inter-nation conflict, then the outcomes of such contests are important data for the understanding of, or prediction about world order and disorder.

The findings herein, although suggesting the validity of the problem, also implicitly exert pressure toward a better quality of data and measurement. Many of the dichotomized indicators employed here and in Banks and Textor studies are based upon continuous variables of interval and ratio levels. Having demonstrated the utility of the approach herein, such data can, in the future, be employed in more sensitive correlational analyses.

For example, an important and frequently employed measure in cross- societal research is Marsh's "Index of Societal Differentiation" (Marsh, 1967: 31-37, 329-374). For modern nation-states this measure is based upon a combination of "( 1 ) percentage of males in... non-agricultural occupations, and (2) gross energy consumption in megawatt-hours per capita for one year" (page 332). It is conceptualized as "the number of structurally distinct and functionally specialized units in a society" (page 31). The range of the index is from zero to a potential of infinity, with the United States, at 109.4, the current high reported by Marsh. For the 1964 Olympic Game nations reported upon here, the low is Nigeria's 16.2, with a median of 41.35.

When the two medians, Marsh's index and team scores, are employed to dichotomously array the participating nations, the distribution is high-high and low-low = 12 per cell, and high-low and low-high = 6 per cell. The Fisher's Exact Probability is less than .05 and Yule's Q = .60. Such a strategy, how- ever, throws away valuable information.

Since both Marsh's index and the Game score index have potential zero points, use of a Pearsonian correlation seems justified (also see Labovitz, 1970). The coefficient for these two variables is a relatively robust +.42 (r2 = .17), indicating a reasonably strong relationship between societal differentiation and team scores. Such a finding is implicit in the 2 X 2 dervied summary portrait above.

Taken as a whole, these findings suggest that the key to understanding national Olympic team scores is resources: possession and mobilization ability. Overall, Game success is related to the possession of resources, both human and economic, and the centralized forms of political decision-making and authority which maximize their allocation. Such is the price of national Olympic success.

Olympic sentiment (quoted in Brundage, 1956, above) has recently been reiterated by the President of the 1972 Olympic Games Organizing Committee, i.e. that nations should not

attach too much importance to victories or yield to the erroneous belief that the Games should serve as an instrument for politics or propaganda of any sort or as a criterion for measuring the value of a political system (Herr Willie Daume, 1971).

Unfortunately, neither historical accounts nor sociological data, suggest the validity of Herr Daume's expression of faith. It could be argued that in an age of nuclear stalemate that it is just as well that they don't. Until ways are found

199

to obviate and/or avoid international conflict (see some of the papers in Fried, et al., 1968; also McNeil, 1965), Olympic competition may provide a relatively benign alternative to the several more lethal forms.* This is, of course, not to imply a cathartic function as regards sport and aggression more generally, but rather to hypothesize about the Games in particular. In any case, this is an issue to address in future research.

REFERENCES

ATKINSON, Hugh 1968 The Games. New York: Bantam Books (Research and background by Phillip Knightley).

AVEDON, Elliott M. and Brian SUTTON-SMITH (eds.) 1971 The Study of Games. New York: Wiley.

BALL, Donald W. Forthcoming "Pottermanship: The psychological sociology of S. Potter (and the Yeovil School)." To appear in Humanities as Sociology, edited by Marcello Truzzi, New York: Charles Merrill.

BANKS, Arthur S. and Robert B. TEXTOR 1963 A Cross-Polity Survey. Cambridge: M.I.T. Press.

BRUNDAGE, Avery 1956 President's statement. Bulletin du Comité International Olympique, 55 (February) : 35; quoted in Morton, 1963: 85.

DAUME, Willie 1971 Announcement of the invitation to the United States to participate in the XX Olympiad. Quoted in Drake, 1971: 24.

DRAKE, Dick 1971 "Olympic Games." Track and Field News, 24 (II May): 24-25.

DUNCAN, Hugh D. 1962 Communication and Social Order. Totawa, N.J.: Bedminster Press.

EDELMAN, Murray 1964 The Symbolic Uses of Politics. Urbana: University of Illinois Press.

EDWARDS, Harry 1969 The Revolt of the Black Athlete. New York: The Free Press.

ETZIONI, Amatai and Fredric L. DuBow (eds.) 1970 Comparative Perspectives: Theories and Methods. Boston: Little, Brown.

FRIED, Morton, Marvin HARRIS and Robert MURPHY (eds.) 1968 War: The Anthropology of Armed Conflict and Aggression. Garden City, N.Y.: The Natural History Press.

HEINILÄ, Kalevi 1966 "Notes on inter-group conflicts in international sport." International Review of Sport Sociology, 1: 29-40; Reprinted in the Cross-Cultural Analysis of Sport and Games, edited by Günther Lüschen (1970): 174-182.

KIERAN, John and Arthur DALEY 1969 The Story of the Olympic Games: 776 B.C. to 1968 (7th ed.) Philadelphia:

J. B. Lippincott. LABOVITZ, Sanford

1970 "This assignment of numbers to rank order categories." American Sociological Review, 35 (June): 515-524.

* Research to pursue this line of investigation is currently in progress.

200

LIDDELL, Hart B. H. 1971 History of the Second World War. Toronto: Collier-Macmillan.

LÜSCHEN, Günther (ed.) 1970 The Cross-Cultural Analysis of Sport and Games. Chicago: Stipes Publishing.

MCINTOSH, P. C. 1963 Sport in Society. London: C. A. Watts.

MCNEIL, Elton B. (ed.) 1965 The Nature of Human Conflict. Englewood Cliffs, N.J.: Prentice-Hall.

MANDELL, Richard D. 1971 The Nazi Olympics. New York: Macmillan.

MARSH, Robert M. 1967 Comparative Sociology: A Codification of Cross-Societal Analysis. New York: Harcourt, Brace and World.

MENKE, Frank G. (Revisions edited by Roger TREAT) 1969 The Encyclopedia of Sports (4th rev. ed).. New York: A. S. Barnes.

MORTON, Henry W. 1963 Soviet Sport. New York: Collier Books.

NATAN, Alex 1958 Sport and Society. London: Bowes and Bowes.

POTTER, Stephen 1948 The Theory and Practice of Gamesmanship. New York: Holt.