Congress Vote on the Gulf War

Base on Krus, D. J., & Webb, J. M. (1993) Quantification of Santayana's cultural schism theory. Psychological Reports, 72, 319-325.

Abstract. The January 12th, 1991 U.S. Congress vote, giving President Bush powers to initiate military operations against Iraq, was analyzed with respect to personal and religious background of the voting senators and representatives. The results of the analysis showed that aside of the party affiliation, the outcome of the vote was significantly influenced by religious affiliations of the members of Congress. These empirical findings are interpreted within the theoretical context of Santayana's hypothesis that the Catholic-Protestant schism is one of the determinants of the niveau of our society.

In contrasting Catholics and Protestants, the last of the grand old generation of American philosophers, George Santayana, makes incisive observations pertaining to the ideological undercurrents within Western civilization frequently classified at the cultural level as Judeo-Christian and Graeco-Roman traditions and expressed at the religious level by Catholics and Protestants. These observations scattered throughout his voluminous work are well summarized by Durant (1938, p.542) who notes that

"Christianity was at first a combination of Greek theology with Jewish morality; it was an unstable combination, in which one or the other element would eventually yield; in Catholicism the Greek element triumphed, in Protestantism the stern Hebraic moral code. The one had a Renaissance, the other a Reformation"

Protestantism is convinced of the importance of success and prosperity; it abominates what is disreputable; contemplation seems to it idleness, solitude selfishness, and poverty a sort of dishonorable punishment. It is constrained and punctilious in righteousness; it regards a married and industrious life as typically godly, and there is sacredness to it, as of a vacant Sabbath, in the unoccupied higher spaces which such an existence leaves for the soul. It is sentimental; its benevolence is optimistic and aims at raising men to a conventional well being. Protestantism was therefore attached from the first to the Old Testament, in which Hebrew fervor appears in its worldly form.

Protestantism, being the dominant religion of the United States, sets the tenor of our country. The combination of competence and fervor, as a major characteristic of American civilization, makes for one of the most resolute and militaristic nations of the contemporary world. The Catholic-Protestant schism, an underlying cause of the Thirty Years War (Wright, 1965), seems also to be relevant in tracing the origins of the 1991 Middle East War, or Desert Storm, as it was termed.

The decisive factor in the inception of Desert Storm was the January 12th, 1991 Senate and House vote, giving President Bush war powers. At that time, there was a major division of opinions among the members of Congress. Representative Solarz (D, New York) headed the war party and Representative Gonzales (D, Texas) attempted to initiate impeachment proceedings against President Bush. The resulting vote was close, especially in the Senate, where the pro-war motion carried by only 5 votes.

On the surface, the January 12th vote of Congress appears to be strictly a party matter. Republicans voted for the war, while the Democrats opposed it. The final outcome, however, was determined by the cross-over votes of Democrats giving their support for the war. Inquiry into the personal and religious backgrounds of the voting members of the 102nd Congress suggests that this cross-over vote was related to by the age-old Catholic-Protestant schism. According to Santayana's observations, this hypothesis seems likely. Protestants, closely tied to the Old Testament, traditionally showed an intense preoccupation with the Middle East. The present investigation attempted to test this hypothesis and subjects these concerns to empirical scrutiny.

The result of the Senate and House roll calls, as reported in the New York Times (1991a, 1991b) and the Arizona Republic (1991a, 1991b), served as the data in this investigation. Each vote was coded as 1 if cast for the war and as 0 if opposing the war. Religious affiliation of senators and representatives (see Pullen, 1991) was classified as Catholic (134), Methodist (75), Episcopalian (56), Baptist (54), Presbyterian (49), Jewish (41), Lutheran (21), Latter Day Saints (11), and Greek and Eastern Orthodox (7) categories. Members of the Unitarian (10), Church of Christ (8), Congregationalist (5), Christian Scientists (3), Disciples of Christ (3), Apostolic Christians (1), Dutch Reformed (1), Evangelical Covenant (1), Pan-African Christians (1), and Seventh Day Adventists (1) denominations were coded as belonging to the Protestant category together with congressional members who identified themselves only as "Protestant" and one Congress member who described himself as "Christian." The numbers in parentheses, following each religious denomination represented in Congress, give the number of members in each category. Also included in the analysis was information about gender of the members of Congress, their party affiliation, and age.

In several instances, the religious affiliation of Congress members was not listed. In those cases we were able to obtain this information from the Americans United for Separation of Church and State (Silver Springs, MD). Representatives listing party affiliation as "Independent," listing their religious affiliation as "none," or representatives who wished not to reveal their religious affiliation were excluded from the analysis. Only 5.4% members of Congress were excluded from the analysis on these grounds. Obtaining religious affiliation of senators and representatives was also hampered by the fact that one representative whose religious affiliation was not listed in the Pullen's handbook (Silvio Conte; R, MA) died shortly after the January 12th vote. After a series of inquiries, however, we were able to secure this information and include his vote in the analysis. In addition, voting results in the New York Times were not reported accurately. When compared with the voting results reported by the Arizona Republic, a discrepancy was noticed in the case of Representative Lantos (D, CA). The New York Times listed him as voting against the war and the Arizona Republic listed him as voting for the war. A call to Mr. Lantos' office resolved this controversy.

Results

As an initial step in validating Santayana's hypothesis, the votes of Catholics either for or against the war were contrasted with all other votes. Seventy-nine Catholics voted against the war, 55 voted for the war. The votes of other members of Congress were divided 234 for the war and 135 against the war. Excluded from the analysis were members of Congress who did not cast their votes (2 among the 435 members of the House and 1 among the 100 members of the Senate), and 29 members of Congress (5.4%) with religious affiliation not given. The cross tabulation of these frequencies returned a chi square of 20.124 (p < .00001) with directions of the observed differences supporting Santayana's hypothesis. A scale analysis of the votes of members of Congress on the war issue was performed on recorded categories of religious denominations, as shown in Figure 1, describing a continuum of increased likelihood to vote for the war.

FIGURE 1. Religious Affiliation of Members of the 102nd U.S. Congress and the
likelihood of their pro-war vote

The frequencies of votes opposing the war and the pro-war votes were in each category as follows: Catholic (79/55), Lutheran (9/12), Baptist (23/31), Protestant (21/36), Methodist (27/48), Episcopalian (17/38), and Presbyterian (12/37). The Catholic denomination was selected as a pivotal category, combined with each Protestant denomination shown in Figure 1, and correlated with the vote on the war issue, serving as a criterion. The obtained correlations for each Protestant denomination, listed in the order of increasing magnitudes, were Lutheran (r=.111), Baptist (r=.149), Protestant (r=.203), Methodist (r=.220), Episcopalian (r=.255) and Presbyterian (r=.305). These correlations can be interpreted as the indicators of convergent validity for Santayana's hypothesis. Evidence for the discriminant validity of Santayana's hypothesis comes from the observation that the Jewish vote did not show any relationship to this postulated continuum. The Jewish vote was split evenly with 21 Jewish members of Congress voting against the war and 20 voting for the war. A similar split (4 for the war, 3 against the war) was observed with respect to the Greek and Easter Orthodox category. However, given the relatively small number of voting members in this and the Latter Day Saints category, we refrained from using their classifications. The coefficients of correlations for all variables included in the study are shown in Table 1.

TABLE 1. Correlations between votes cast on the 1991 war with Iraq
and indices on the Members of the 102nd U.S. Congress.

	2	3	4	5
1. The War Issue Vote	*-.675*	*.237*	.067	-.014
2. Party Affiliation		*-.147*	-.036	-.006
3. Religious Affiliation			.086	*.158*
4. Gender				.025
5. Age

Contrasting Catholics with other members of Congress with respect to the vote on the war issue, the phi coefficient of correlation between these binary categories was -.200. The magnitude of this relationship increased to .237 when the postulated Catholic - Lutheran - Baptist - Protestant - Methodist - Episcopalian - Presbyterian 7 point scale was used in lieu of the binary Catholic - Protestant scale used in the course of the pilot study, as shown in the Table 1. This observed increase lends further support to Santayana's hypothesis.

The correlation between the party membership and the vote on the war issue was -.675. The only other significant correlation between the variables included in the analysis and the war issue was that between the religious affiliation of the members of Congress and the vote for or against the war (r=.237). Entered into multiple regression analysis with the vote on the war issue serving as the dependent variable, these two variables accounted for 48% of the variance in the criterion variable.

The results of the present study also indicate that the religious affiliations of the members of Congress are not independent of party affiliation. Catholics prefer the Democratic Party (r=-.147) and older members of Congress are more likely to be Protestants (r=.158). The preference of Catholics for the Democratic party has been discussed by Reichley (1986) and the association of age with conservatism by Eysenck (1954) and Webster and Steward (1973). These relationships only tangentially pertain to the tested hypothesis and do not contradict it. These were the only significant correlations, not related to the vote on the war issue, among variables included in the analysis.

Another issue we looked at in the present analysis was the relationship between gender of members of Congress and their vote on the war issue. Eighteen female members of Congress voted against the war and 12 voted for the war. However, the correlation between gender and voting on the war issue (r=.067) was not statistically significant.

The hypothesis under scrutiny was also supported by a discriminant analysis which returned a statistically significant canonical discriminant function with a corresponding Wilk's Lambda equal to .944. Testing the hypothesis that Catholics tended to vote for the war and Protestants, ordered on the Lutheran - Presbyterian scale, against the war, the percentage of correctly classified grouped cases was 60.67%. These observations are congruent with other empirical studies of this issue summarized by Russell (1971).

Discussion

The 1991 Middle East War is relevant to the social sciences because, given the relative simplicity of the conflict, it provides a unique opportunity to study factors leading toward the decision to go to war. It seems difficult to imagine another issue more relevant to social science than studying the decision-making processes when one group of people decides that another group can be sent into combat to face possible death. Facets of this decision-making process have been studied by philosophers, historians, and social scientists during the course of recorded human history. Only recently, however, have social scientists been able to apply modern quantitative methods to the study of these topics. Viewed within this context, Santayana's cultural schism theory stands well when validated by the quantitative analysis of the U.S. Congressional vote reported in the present study. It appears that his theory of cultural schism can be extrapolated from the domain of philosophy to quantitative social sciences.

The present analysis also bears relevance for the theoretical studies of religion. Religious doctrines originating in the Middle East contain effective deterrents against individual aggression; however, they lack effective mechanisms to prevent group aggression. The present study further suggests that religion influences the decision-making processes within our government. The founders of our constitution, while formulating the principles of separation of church and state, were aware of what happened in Europe during the period of the religious wars which culminated in the Thirty Years War.* Considering the manifold implications of the present study, Santayana's cultural schism theory is not without relevance for our post cold-war present and for our future.

References

Durant, W. (1938) The story of philosophy. New York: Garden City Publications.

Eysenck, H.J. (1954) The psychology of politics. London: Routledge & Kegan Paul.

Pullen, D. (1991) The U.S. Congress handbook, 102nd Congress, First Session. McLean, Virginia: B. Pullen.

Reichley, A. J. (1986) Religion and the future of American politics. Political Science Quarterly, 101, 23-47.

Roll-call in House on resolution authorizing use of force (1991a, January 13th) The New York Times, Vol. CXL (48,479), pp. 1, 11.

Roll-call in Senate on a gulf assault (1991b, January 13th) The New York Times, Vol. CXL (48,479), pp. 1,11.

Russell, E. W. (1971) Christianity and militarism. Peace Research Review, 4, (3), 1-77.

Webster, A. C., & Steward, R. A. C. (1973) Theological conservatism. In G. D. Wilson (Ed.) The psychology of conservatism. London: Academic Press. (pp. 129-147).

The authors wish to thank two anonymous reviewers for their help with the improvement of manuscript. The authors also wish to thank to Michael C. Dornan for his help with the library search.

*Intensity of that conflict surpassed that of other types of armed confrontations. In Bohemia, for instance, there were whole sections of the country in which nobody was left to bury the dead. The total population of Bohemia decreased in the 17th century from about 3 million to 500,000. These population changes are representative of other areas of Central Europe afflicted by the Thirty Years War.