¹ These illustrations are taken from Amnesty International Report 1985, describing human rights conditions in 1984.

² Paine, Thomas, Common Sense (Harmondsworth, England: Penguin Books, 1976), 65.Google Scholar

³ Quoted in Huntington, Samuel, Political Order in Changing Societies (New Haven: Yal University Press, 1968), 41.Google Scholar

⁴ Ibid., 52.

⁵ Ibid., 41.

⁶ Chomsky, and Herman, , The Political Economy of Human Rights: The Washington Connection and Third World Fascism (Boston: South End Press, 1979), 8.Google Scholar

⁷ See Huntington, Samuel, “Will More Countries Become Democratic?” Political Science Quarterly 99 (Summer 1984), 193–218CrossRefGoogle Scholar. Wiarda and Kline contrast the British and Iberian influences as follows:

The Spanish and Portuguese colonies were founded on a set of institutions that were absolutist, authoritarian, hierarchical, Catholic, feudal or semifeudal, two-class, corporatist, patrimonialist, orthodox, and scholastic to their core. By contrast, the British colonies … derived from a set of institutions and practices that were fundamentally different.…”

Wiarda, Howard J. and Kline, Harvey F., eds., Latin American Politics and Development (Boston: Houghton Mifflin, 1979), 21–22.Google Scholar

⁸ Huntington (fn. 7), 206.

⁹ Kirkpatrick, , “Dictatorships and Double Standards,” Commentary 68 (November 1979), 34–45.Google Scholar

¹⁰ Howard, and Donnelly, , “Human Dignity, Human Rights, and Political Regimes,” American Political Science Review 80 (September 1986), 801–18.CrossRefGoogle Scholar

¹¹ Kirkpatrick (fn. 9), 44. We should note that the use of regime type in an analysis of human rights violations risks tautology because concern for human rights is an element of regime definition. In our research, however, classifications of regime types are independent of our data on human rights.

¹² Howard and Donnelly (fn. 10), 802.

¹³ See the discussion in Carleton, David and Stohl, Michael, “The Foreign Policy of Human Rights: Rhetoric and Reality from Jimmy Carter to Ronald Reagan,” Human Rights Quarterly 7 (May 1985), 205–29CrossRefGoogle Scholar. Our discussion of the methodology draws upon their imaginative research and follows their lead.

¹⁴ See the Foreign Assistance Act of 1961 as amended, in Legislation on Foreign Relations Through 1985, Committee on Foreign Relations (Washington, DC: G.P.O., 1986), 86Google Scholar, for a more recent statement of this section, in which the phrase “causing the disappearance of persons by the abduction and clandestine detention of those persons” has been added after the passage on “prolonged detention without charges.”

¹⁵ Ibid., 125.

¹⁶ See Harry M. Scoble and Laurie S. Wiseberg, “Problems of Comparative Research on Human Rights,” in Nanda, Ved R., Scarritt, James R., and Sheppard, George W. Jr., Global Human Rights: Public Policies, Comparative Measures, and NGO Strategies (Boulder, CO: West view Press, 1981), 162–64Google Scholar, for a discussion of the Freedom House and State Department reports.

¹⁷ The rank-ordering of a country into one of the five categories for the two scales was difficult. While identifying the most and least serious offenders on either dimension was a relatively straightforward procedure, the task was considerably more challenging for the majority of countries in our study. In order to produce meaningful measures for analysis, the scaling of each country involved an iterative process of categorization. First, using the information on each country from our coding sheet, both of us separately rank-ordered all countries on both scales. Next, we compared our rankings, did an initial intercoder reliability check, and tried to make explicit the calculations that we went through to arrive at a particular rank order for a country. To resolve our differences, we agreed upon the following quantitative distinctions: “rarely” referred to countries in which one to ten cases of political prisoners or incidents of torture or killing were reported by Amnesty International; “sometimes” referred to countries with 20 to 90 cases; “often” to 100 or above, and “very often” to 1000 or more. Because our coding sheet has more than one entry to capture information on political prisoners and their treatment, we also agreed that if a given country had a number of entries on these dimensions, the numerical threshold required for an assigned scale would be lower. Finally, we made adjustments in our rank-ordering for several countries to reflect these decisions. Namibia, while given its own country entry in the Report, was not included in our analysis because of the general lack of other data.

Although these procedural efforts cannot ultimately remove the necessarily qualitative judgment for the two scales, we are generally quite confident that our effort helped us to construct a meaningful and empirically useful measure.

Finally, in a decision that may well be controversial, we departed from Amnesty International's definition by not counting judicial executions or capital punishment on our tortureand-killing scale. For the sake of consistency, we also excluded judicial amputations—as found, for example, in some Moslem countries—from our analysis.

¹⁸ The comparison between our aggregate human-rights scale score (the sum of the ranks on the political prisoner and torture scales) for 1984 and Carleton and Stohl's rank-ordering of countries for 1983 is based upon their assessment from Amnesty International for that year (fn. 13, 227–28). The range of our scale scores was 0 to 8; it was 1 to 5 for Carleton and Stohl. In all, comparable information between the two datasets was available for 46 countries.

¹⁹ For the purposes of the statistical analyses, we collapsed categories one and two on both the political prisoner and torture scales. The income data are from World Development Report, 1986 (New York: Oxford University Press, 1986)Google Scholar. Where there were reporting gaps, which was most often the case with second-world countries, we used The World Almanac and Book of Facts, 1986 (New York: Newspaper Enterprise Association, Inc., 1985)Google Scholar. In the case of Brunei, we used The CIA Factbook (Washington, DC: C.I.A., 1986).Google Scholar

²⁰ Although our human rights measures are not per capita measures, there is a small positive association between the size of population and the frequency of violations on both human rights measures: the larger the population, the larger the potential pool of victims, and the larger the reports of violations. Nevertheless, when we controlled for size of population (small, medium, or large) using an analysis of variance technique and a hierarchical log linear technique (controls that we use throughout these analyses) the association between income and human rights violations remained. See Norusis, Marija J., SPSS^x Advanced Statistics Guide (New York: McGraw-Hill, 1985), 195–254Google Scholar, 297–325. The population estimates for 1984 are from The World Almanac (fn. 19). For Albania, Gambia, U.S.S.R., and Zaire, population estimates were available only for 1983.

²¹ The I.M.F. lists the following as “industrial countries”: United States, Canada, Australia, Japan, New Zealand, Austria, Belgium, Luxembourg, Denmark, Finland, France, Germany, Iceland, Italy, Netherlands, Norway, Spain, Sweden, Switzerland, United Kingdom. I.M.F., , Direction of Trade Statistics Yearbook 1986 (Washington, DC: International Monetary Fund, 1986)Google Scholar. The classification of nations into first-, and second-, and third-world nations was based upon Bertsch, Gary K., Clark, Robert P., and Wood, David M., Comparing Political Systems: Power and Policy in Three Worlds (New York: Wiley, 1986), xiii–xvGoogle Scholar, and their categorization of 170 nations for 1983. The first-world nations excluded from our dataset for this analysis were: Canada, Greece, Ireland, Israel, Italy, Japan, Spain, Switzerland, United Kingdom, and the United States. The second-world nations excluded were: Albania, Bulgaria, People's Republic of China, Cuba, Czechoslovakia, East Germany, Hungary, Kampuchea, North Korea, Laos, Poland, Romania, Soviet Union, Vietnam, and Yugoslavia. The rest of the nations in our dataset were classified as third-world countries for this analysis, including five nations (Barbados, Cyprus, Djibouti, Grenada, and the Seychelles) that were classified as “mixed systems, status uncertain, or otherwise unclassified” by Bertsch et al. Because of missing data for some nations, the total number of nations in the trade and investment analyses was 90 and 85, respectively.

²² The investment measure is the sum of “Total Other Official Flows plus Private Sector Net” from the member countries of the OECD's Development Assistance Committee (which corresponds to the I.M.F.'s list of industrial countries except Luxembourg, Iceland, and Spain). OECD, Geographical Distribution of Financial Flows to Developing Countries (Paris: OECD, 1986).Google Scholar

²³ See Firebaugh, Glenn and Gibbs, Jack P., “User's Guide to Ratio Variables,” American Sociological Review 50 (October 1985), 713–22CrossRefGoogle Scholar, and Uslaner, Eric, “The Pitfalls of Per Capita,” American Journal of Political Science 20 (February 1976), 125–33CrossRefGoogle Scholar, for a methodological discussion of the use of ratio variables.

²⁴ The information on colonial powers is from Nielsen, Waldemar A., The Great Powers and Africa (New York: Praeger, 1969)Google Scholar, inside front cover; The World Almanac (fn. 19); and Banks, Arthur S., ed., Political Handbook of the World, 1986 (Binghamton, NY: CSA Publications, 1986).Google Scholar

²⁵ Data for years of independence are from Appendix B in Russett, Bruce and Starr, Harvey, World Politics: The Menu for Choice (San Francisco: W. H. Freeman, 1981 and 1985)Google Scholar, originaand 2d eds., 577–83 and 589–96.

²⁶ To identify authoritarian regimes, we used the regime typology presented in Shively, W. Phillips, Power and Choice (New York: Random House, 1986), 248–49Google Scholar. From this typology, the ian category. Like most such efforts, the regime typology from Shively has some questionable classifications. For example, Uruguay, South Korea, and Honduras were listed under “Democracy,” with the qualification “heavy military involvement.” We placed those countries in the authoritarian category. Finally, to identify totalitarian regimes, we used the classification of Marxist nations by Bertsch et al. (fn. 21).

²⁷ South Africa was also classified as a democracy in the Shively listing, and the following were termed countries “difficult to classify": Philippines, El Salvador, Haiti, Nepal, Chad, and Iran. When these seven countries are added to the authoritarian category, the number of authoritarian regimes in the two high-torture dimensions increases from 52% to 61%, but this does not significantly change the results.

²⁸ Howard and Donnelly (fn. 10), 814.

²⁹ See Anderson, Charles W., “System and Strategy in Comparative Policy Analysis,” in Gwyn, William B. and Edwards, George C., eds., Perspectives on Public Policy Making (New Orleans: Tulane University Press, 1975)Google Scholar, for a theoretical discussion of policy conditions.