¹ Some studies do descriptively identify norms as important but do not pursue the analytical relationship between norms and sanctions. A notable exception is Kim Richard Nossal who argues that sanctions are sometimes used punitively to express moral disapproval; see ‘International Sanctions as International Punishment’, International Organization, 43 (1989), pp. 301–22CrossRefGoogle Scholar. However, the act of punishment, rather than the target's response, becomes the analytical focus. Sanctions are thus successful by definition.

² Realist analyses generally adopt materialist and statist assumptions. As Kim Richard Nossal rightly points out, the ‘generic’ theory of sanctions derives primaril y from the perspective of hegemonic states, which is an additional reflection of the predominance of the realist theoretical perspective; see Rain Dancing: Sanctions in Canadian and Australian Foreign Policy (Toronto, 1994)Google Scholar.

³ For the broadest empirical study of sanctions see Hufbauer, Gary Clyde, Schott, Jeffrey J. and Elliott, Kimberly Ann, Economic Sanctions Reconsidered: History and Current Policy, 2nd edn (Washington, DC, 1990)Google Scholar. In addition, the pariah state literature starts from the assumption that concerted international sanctions are possible and thus seeks to demonstrate their implications for the target state. See in particular Geldenhuys, Deon, Isolated States: A Comparative Analysis (Cambridge, 1990)Google Scholar.

⁴ Although he focuses on states’ adoption of sanctions in Economic Statecraft (Princeton, 1985)Google Scholar, David A. Baldwin explicitly excludes international consensus and domestic factors. For discussions of why sanctions analysis needs to recognize inter-state interaction, rather than relying on single-state explanations, see Stephanie Lenway, Ann, ‘Between War and Commerce: Economic Sanctions as a Tool of Statecraft’, International Organization, 42 (1988), pp. 397–426CrossRefGoogle Scholar, and Tsebelis, George, ‘Are Sanctions Effective?’, Journal of Conflict Resolution, 34 (1990), pp. 3–28CrossRefGoogle Scholar.

⁵ Noteworthy exceptions are: Jentleson, Bruce W., Pipeline Politics: The Complex Political Economy of East-West Energy Trade (Ithaca, NY, 1986)Google Scholar; Mastanduno, Michael, Economic Containment: CoCom and the Politics of East-West Trade (Ithaca, NY, 1992)Google Scholar; and Martin, Lisa L., Coercive Cooperation: Explaining Multilateral Economic Sanctions (Princeton, 1992)Google Scholar. However, none of these works analyzes normative consensus as a factor motivating the adoption of sanctions. These limitations are symptomatic of the theories being tested. For details of the more general debate over the role of norms in international relations theories, cf. Kratochwil, Freidrich and Ruggie, John G., ‘International Organization: A State of the Art on the Art of the State’, International Organization, 44 (1986), pp. 753–75CrossRefGoogle Scholar; Keohane, Robert O., ‘International Institutions: Two Approaches’, International Studies Quarterly, 32 (1988), pp. 379–96CrossRefGoogle Scholar; and Klotz, Audie, Norms in International Relations: The Struggle against Apartheid (Ithaca, NY, 1995)Google Scholar.

⁶ See, for example, Knorr, Klaus, The Power of Nations (New York, 1975)Google Scholar; Doxey, Margaret P., Economic Sanctions and International Enforcement (New York, 1980)CrossRefGoogle Scholar; Daoudi, M. S. and Dajani, M. S., Economic Sanctions: Ideals and Experience (London, 1983)Google Scholar; Kaempfer, William H. and Lowenberg, Anton D., International Economic Sanctions: A Public Choice Perspective (Boulder, 1992)Google Scholar. These ‘goods’ need not be solely economic commodities, however; arms as well as sporting matches could be translated into such a utility calculus.

⁷ Rhodesia is the classic example. Johan Galtung makes an even more pessimistic claim, that sanctions can be counterproductive by generating new elites whose interests lie in maintaining international isolation. See ‘On the Effects of International Economic Sanctions’, World Politics, 19 (1967), pp. 378–16CrossRefGoogle Scholar. Minter, William and Schmidt, Elizabeth offer a more optimistic assessment in ‘When Sanctions Worked: The Case of Rhodesia Re-examined’, African Affairs, 87 (1988), pp. 207–37CrossRefGoogle Scholar. Baldwin also offers a lengthy catalogue of the analytical problems of conventional studies which, he argues, underestimate the effectiveness of sanctions (Economic Statecraft, p. 205). See also Jentleson's discussion of the difficulties of measuring success, in Pipeline Politics, pp. 31–4. Geldenhuys’ list of indices in Isolated States attempt s to contribute a more balanced and consistent evaluation of different types of sanctions.

⁸ Hufbauer et al. measure effectiveness in terms of the target state's response, considering both ‘the extent to which the policy outcome sought by the sender country was in fact achieved, and the contribution made by the sanctions’, but they acknowledge the highly subjective nature of their evaluations (Economic Sanctions Reconsidered, p. 41).

⁹ Drawing on the international political economy literature leads to a distinction between coercive costs and market costs. Threats to survival can be made through military means or economic sanctions. However, sanctions can also inflict economic costs on the target which are generated through manipulation of market mechanisms. These costs I characterize below as incentives that induce actors to change, rather than threats which coerce change.

¹⁰ Most notable was South Africa's ability to evade the oil boycott. Continuing debates over the general economic costs of sanctions are complicated by both a recession in the latter half of the 1980s and the role of non-governmental sanctions. Generally analysts recognize that financial restrictions, including governmental restrictions on further loans, international bankers’ refusal to grant credit, and corporations’ changing risk assessments, contributed to South Africa's economic woes. For details, see the report of the Commonwealth Committee of Foreign Ministers on Southern Africa, Banking on Apartheid: The Financial Sanctions Report (London, 1989)Google Scholar. Thus, in the South African case, governmental sanctions operated within and reinforced normal financial market mechanisms.

¹¹ For a discussion of the difficulties of measuring norms see Goertz, Gary and Diehl, Paul F., ‘Toward a Theory of International Norms : Some Conceptual and Measurement Issues’, Journal of Conflict Resolution, 36 (1992), pp. 634–64CrossRefGoogle Scholar. For an alternative view of socialization which focuses primarily on elite beliefs, see Ikenberry, G. John and Kupchan, Charles A., ‘Socialization and Hegemonic Power’, International Organization, 44 (1990), pp. 283–315CrossRefGoogle Scholar.

¹² For an historical overview of the rise of a norm of racial equality see Vincent, R. J., ‘Racial Equality’, in Bull, Hedley and Watson, Adam (eds.), The Expansion of International Society (Oxford, 1984), pp. 239–54Google Scholar. For the most comprehensive summary of sanctions against South Africa, see Geldenhuys, Isolated States.

¹³ For an analysis of domestic unrest see Marx, Anthony W., Lessons of Struggle: South African Internal Opposition, 1960–1990 (New York, 1992).Google Scholar

¹⁴ The African perspective differed in its historical grounding. Although equating apartheid with European Nazism, African objections to apartheid arose from experiences with European colonization and discrimination. Eastern bloc states supported a norm of racial equality within a framework of economic justice. For an historical overview of the processes which created a dominant international order based on Western European ideals, see Bull and Watson (eds.), Expansion of International Society.

¹⁵ For an overview of the early multilateral debates see Bissel, Richard E., Apartheid and International Organizations (Boulder, 1977).Google Scholar For details of the UN debates and resolutions, see Yearbook of the United Nations (New York, annual).

¹⁶ For overviews of Commonwealth debates over apartheid see Darwin, John, Britain and Decolonization (London, 1988)CrossRefGoogle Scholar; Doxey, Margaret, The Commonwealth Secretariat and the Contemporary Commonwealth (London, 1989)CrossRefGoogle Scholar; Mansergh, Nicholas, The Commonwealth Experience, Vol. 2: From British to Multiracial Commonwealth (Toronto, 1982)CrossRefGoogle Scholar; Miller, J. D. B., Survey of Commonwealth Affairs: Problems of Expansion and Attrition 1953–1969 (London, 1974)Google Scholar. For details of official declarations see Commonwealth Secretariat, The Commonwealth at the Summit: Communiques of the Commonwealth Heads of Government Meetings, 1944–1986 (London, 1987)Google Scholar. The discussions during Summit meetings are officially secret.

¹⁷ Japan, which has become South Africa's largest trading partner in recent years, receives attention as the biggest hole in sanctions implementation. Japan's adoption of diplomatic, cultural, computer, nuclear, gold, iron, steel and airline bans in 1986 appeared to be the result of (or preventive concern over) pressure from the US. Because it has been a follower rather than leader in apartheid debates, and because its ties to South Africa are solely economic rather than military and cultural, Japan is not considered in detail here.

¹⁸ For a sympathetic historical overview of US policy toward South Africa, see Coker, Christopher, The United Stales and South Africa, 1968–1985: Constructive Engagement and its Critics (Durham, NC, 1986)Google Scholar; for a more critical one, see Minter, William, King Solomon's Mines Revisited: Western Interests and the Burdened History of Southern Africa (New York, 1986)Google Scholar.

¹⁹ On race politics in Britain, see Fryer, Peter, Slaying Power: The History of Black People in Britain (London, 1984)Google Scholar; and Goulbourne, Harry (ed.), Black Politics in Britain (Aldershot, 1990).Google Scholar

²⁰ For an overview of British policy toward South Africa, see Barber, James, The Uneasy Relationship: Britain and South Africa (London, 1983)Google Scholar; and Berridge, Geoff, Economic Power in Anglo-South African Diplomacy (London, 1981)CrossRefGoogle Scholar. On foreign-policy-making in the British parliamentary system more generally, see Barber, James, Who Makes British Foreign Policy? (Milton Keynes, 1976)Google Scholar.

²¹ Unlike the Commonwealth where Britain was the only opponent of sanctions, the EC was more divided. For example, Denmark strongly advocated sanctions, Germany joined Britain in opposing sanctions, and French policy varied. For details on the European debates and Britain's role, see Holland, Martin, The European Community and South Africa (London, 1988)Google Scholar; and Thatcher, Margaret, The Downing Street Years (New York, 1993)Google Scholar.

²² For an overview, see Heriber t Adam and Moodley, Kogila, The Opening of the Apartheid Mind: Optionsfor the New South Africa (Berkeley, 1993), pp. 52–8Google Scholar.

²³ See Barber, James and Barratt, John, South Africa's Foreign Policy: The Search for Status and Security, 1945–1988 (Cambridge, 1990)Google Scholar for the foreign policy response to sanctions. For the strategic capabilities argument see Price, Robert M., The Apartheid State in Crisis: Political Transformation in South Africa 1975–1990 (New York, 1991), pp. 220–48Google Scholar.

²⁴ For example, see Becker, Charles M., ‘Economic Sanctions Against South Africa’, World Politics, 39 (1987), pp. 147–73CrossRefGoogle Scholar. Although more explicit than most in stating this political connection, Kaempfer and Lowenberg rely on general ‘economic hardship’, their primary economic consequence of sanctions, to generate ‘political pressure’ through interest groups within the target state; see International Economic Sanctions, p. 11. While generating much of their model from the South African case-and arguing for the importance of political consequences of economic sanctions-they do not ofler substantial empirical support for their analysis; also see their ‘Applying Economic Sanctions: Toward a Public Choice Framework’, in Odell, John S. and Willett, Thomas D. (eds.), International Trade Policies (Ann Arbor, 1990), esp. p. 187Google Scholar.

²⁵ That interest group analyses underestimate political institutions is a particularly problematic limitation given that groups within South Africa are fighting specifically over possible alternative institutional arrangements for a post-apartheid South Africa. On the relationship between groups, institutions and power in South Africa see Horowitz, Donald L., A Democratic South Africa? Constitutional Engineering in a Divided Society (Berkeley, 1991)Google Scholar. On the power of the South African state in relation to the economy and domestic resistance, see Greenberg, Stanley B., Legitimating the Illegitimate: State, Markets and Resistance in South Africa (Berkeley, 1987)Google Scholar, as well as his critique of the liberal view that economic change necessarily produces democratic reforms, in ‘Economic Growth and Political Change: the South African Case’, Journal of Modern African Studies, 19 (1981), pp. 667–704CrossRefGoogle Scholar. Furthermore, even though transnational and exiled anti-apartheid activists played a crucial role in generating global sanctions, international influence on domestic actors has been underestimated; notable exceptions include Marx, Lessons of Struggle, and Price, Apartheid State in Crisis. For an extended discussion of the relationship between international norms and domestic politics, see Klotz, Norms in International Relations, esp. ch. 2.

²⁶ Wording and emphasis were marginally different across agreements but represent a surprising degree of consistency given the difficulty of international agreement in general. Some actors called for more stringent requirements, such as more concrete evidence of a transition to majority rule (such as one-person one-vote elections); the measures listed here represent the more conservative interpretation of reform.

²⁷ In other words, South Africa is a ‘hard case* for an explanation stressing international influence. In addition, public acknowledgement of the importance of sanctions is unlikely to be found since it was illegal to support sanctions.

²⁸ New York Times, 3 May 1991, A l l.

²⁹ Controversy persisted, however, about whether the South African government had released all political prisoners; opposition movements claimed many remained incarcerated for political crimes, a position which the government refused to accept. This point became particularly noteworthy in US debates over the repeal of sanctions, with President Bush, never an advocate of sanctions, supporting the South African government's interpretation. Despite objections by the ANC as well as domestic US supporters of sanctions, such as TransAfrica and the Congressional Black Caucus, Bush supported a State Department determination that all political prisoners had been released when he signed an executive order lifting US sanctions (New York Times, 11 July 1991, A10).

³⁰ New York Times, 3 May 1991, A11.

³¹ As in the US debates over who were really political prisoners and who were ‘ordinary’ criminals, debate flourished over what was ‘irreversible’ commitment to political reforms. Mandela and the ANC argued for the maintenance of international sanctions pressures until an election based upon one-person-one-vote. Their position, however, did not sway predominantly conservative Western leaders.

³² For example, the European Community lifted its sanctions in stages in response to de Klerk's reforms. In December 1990, it lifted its voluntary ban on investments in South Africa, but members delayed lifting the arms embargo as part of the UN embargo until after the repeal of the legal pillars of apartheid (New York Times, 5 February 1991, A3); the EC lifted the rest of its economic restrictions on 15 April 1991 in response to de Klerk's continuing reform efforts (New York Times, 16 April 1991, Al). On 11 July 1991, the day after the International Olympic Committee lifted its ban. against South African sporting representation, the US lifted its sanctions, declaring that de Klerk's movements to abolish apartheid were ‘irreversible’ (New York Times, 11 July 1991, Al). In contrast, Britain lifted its sanctions almost immediately after de Klerk's initial reform efforts-specifically Mandela's release-bringing Thatcher under criticism from diverse sources such as the opposition Labour Party as well as fellow European Community members (The Independent (London), 13 and 24 February 1990)Google Scholar.

³³ New York Times, 18 June 1991, A8.

³⁴ On the divisions within Afrikanerdom prior to de Klerk's ascension, see Leach, Graham, The Afrikaners: Their Last Trek (London, 1989)Google Scholar. In its April-May 1989 survey the Investor Responsibility Research Center attempted to gauge white response, in terms of perception not policy, to international economic sanctions. The report's findings substantiate the general claim that sanctions were taken seriously by whites in South Africa. For details see Hofmeyr, Jan, The Impact of Sanctions on White South Africa: Part II, Whites’ Political Attitudes (Washington, DC, 1990)Google Scholar. For an analysis of the socio-economic bases of divisions within Afrikanerdom prior to the peak of Western sanctions see Lipton, Merle, Capitalism and Apartheid: South Africa, 1910–1984 (Totowa, NJ, 1985)Google Scholar.

³⁵ The Conservatives had vehemently contested the precise wording, but de Klerk's vision prevailed (New York Times, 18 March 1992).

³⁶ Geldenhuys, Deon, ‘The Foreign Factor in South Africa's 1992 Referendum’, Politikon, 19 (1992), pp. 47–51CrossRefGoogle Scholar.

³⁷ Geldenhuys, ‘Foreign Factor’, pp. 56–8.

³⁸ New York Times, 18 March 1992, A8. Although voters on both sides of the issue were concerned about violence, their evaluations of the consequences of the vote varied considerably. Thus, apparently many of the ‘yes’ voters were not optimistic about reform but saw a future under more conservative leadership as an even less appealing alternative (e.g. New York Times, 4 March 1992, A6).

³⁹ New York Times, 20 March 1992.

⁴⁰ The referendum lost in only the most conservative areas of the northern Transvaal, and even there de Klerk received 38% of the votes (New York Times, 26 March 1992).

⁴¹ New York Times, 20 March 1992, A7.

⁴² On the sporting boycott generally see Nixon, Rob, Homelands, Harlem and Hollywood: South African Culture and the World Beyond (New York, 1994), pp. 131–54Google Scholar, and on its role in the referendum in particular, pp. 152–3.

⁴³ A variety of international organizations, most notably the UN and the Organization of African Unity, gave the ANC and the PAC official status as the legitimate representatives of the South African people. UN status in particular granted the liberation movements substantial financial resources. For details, see Yearbook of the United Nations.

⁴⁴ For details on ANC-UDF connections see Marx, Lessons of Struggle, p. 139; for a general discussion of foreign financing see Price, Apartheid State in Crisis, p. 233.

⁴⁵ For more details on the resurgent influence of the ‘Charterist’ reform movement see Marx, Lessons of Struggle.

⁴⁶ New York Times, 22 July 1992.

⁴⁷ See Halliday, Fred, ‘The Gulf War 1990–1991 and the Study of International Relations’, Review of International Studies, 20 (1994), p. 116CrossRefGoogle Scholar.

⁴⁸ See, for example, Sparks, Allister, ‘The Secret Revolution’, The New Yorker, 11 April 1994, esp. pp. 69–72Google Scholar; de Klerk, Willem, F. W. de Klerk: The Man in His Time (Johannesburg, 1991)Google Scholar; Ottoway, David, Chained Together: Mandela, de Klerk, and the Struggle to Remake South Africa (New York, 1993).Google Scholar

⁴⁹ See Klotz, Audie, ‘Race and Nationalism in Zimbabwean Foreign Policy’, The Round Table, 327 (1993), pp. 260–5Google Scholar.

⁵⁰ See Nixon's contrast between Afrikaners and Serbs in Homelands, pp. 233–54.

⁵¹ See Nossal, Rain Dancing, pp. 261–7; and Damrosch, Lori Fisler, ‘The Civilian Impact of Economic Sanctions’, in Damrosch, Lori Fisler (ed.), Enforcing Restraint: Collective Intervention in Internal Conflicts (New York, 1993), pp. 274–315Google Scholar.