MIGRATION EMERGENCIES AND HUMAN RIGHTS IN HAITI
Paper prepared for the Conference on Regional Responses to Forced Migration
in Central America and the Caribbean
September 30-October 1, 1997
'Haiti' and 'forced migration' have been synonyms for much of this decade. The story of the 1991 coup that launched a thousand paltry rafts, the successful US effort to intercept the refugee armadas at sea and return their passengers first to Haiti and later to the camps at Guantánamo Bay, the abrupt end of the refugee outflow once Aristide was returned to Haiti on the wings of the US and UN forces in 1994, is by now well-known. I do not intend to revisit it in great detail in this paper, but I will touch upon its implications for any future refugee emergencies involving Haiti. I would like to focus most of my remarks on Haitian migration pressures that have been evident since 1994 which involve migration patterns that both predate and were reshaped by the 1991-94 refugee crisis. These more enduring migration patterns contribute to political instability in Haiti (and other Caribbean states) and exacerbate the impact of refugee crises when they do occur. Because of their complexity, they also tend to be avoided in discussions about responses to migration emergencies.
The 1991-94 Haitian Refugee Crisis
The US and the Crisis
The 1991-94 Haitian refugee crisis was the culmination of political tensions that had been building in Haiti for at least 20 years around the cruel dictatorship of Papa Doc Duvalier, his son and the series of military governments which followed them. Haitian refugees had actually begun arriving on the US mainland in 1972, but their efforts to flee the Duvalier dictatorship only led to arrest, jail, the denial of asylum and swift expulsion as successive US governments refused to recognize the repression in Haiti and branded them economic migrants. US policy shifted in 1980 when the Carter Administration found itself confronting simultaneous influxes of both Haitian and Cuban refugees. Unable to overtly treat the two refugee populations differently, Carter created a new immigration classification: 18,000 Haitians and 125,000 Cubans became neither refugees, nor asylees, but simply 'entrants,' whose fate was to be decided at a later date by legislation. The Reagan Administration shifted US policy once again, establishing the interdiction-at-sea program in September 1981, and promising that Haitians able to escape the Coast Guard boats would face lengthy detention in federal prisons and INS centers. According to the US immigration service, 433 boats were intercepted and 25,551 Haitians returned to Port-au-Prince under the interdiction program from 1981 to 1991. Only 28 persons were permitted to enter the US to pursue refugee claims. 1
Baby Doc Duvalier was forced out of Haiti in 1986 and four military governments preceded the elections that brought Aristide to the presidency in February 1991. After Aristide's election, refugee flows dropped dramatically: fewer than 1,200 took to the seas in 1990, one-third of the number of refugees intercepted in each of the years 1987, 1988 and 1989. 2
The surprise coup in September 1991 opened the refugee floodgates. Within six months of the coup the US Coast Guard had intercepted more than 38,000 Haitians at sea; 10,747 were eventually allowed to pursue asylum claims in the US following screening by immigration officials on board ships or at the US naval base at Guantánamo Bay. An estimated 10% of the population of Port-au-Prince and Haiti's other large cities fled into the mountains, generating an internally displaced population of perhaps 300,000. 3 A further 30,000 crossed into the Dominican Republic.
Initial US reluctance to repatriate Haitians plucked from the sea withered in the face of the refusals of other Latin American states to share the refugee burden. A temporary legal rebuff to the Bush Administration's repatriation plan led to the establishment of camps at Guantánamo Bay and the establishment of an in-country refugee processing program at the US embassy in Port-au-Prince. When Guantánamo filled up with 12,000 refugees, President Bush ordered the summary return of all Haitians picked up at sea. He had been given the liberty to do so by a US Supreme Court ruling that the Refugee Convention did not apply on the high seas.
President Clinton adopted his predecessor's policies shortly after taking office in January 1993, despite having severely criticized this same policy in the 1992 presidential campaign. Almost all of the Guantánamo refugees were permitted to enter the US to pursue asylum claims while those picked up at sea were returned to Haiti despite the horrid human rights situation in the country. Under intense political pressure, the Clinton Administration finally agreed to initiate on-ship refugee processing in June 1994, but the single naval vessel designated for the screenings was quickly overwhelmed. With the news prompting an outflow of more than 10,000 refugees in the 10-day period following the establishment of the shipboard screening process, the Clinton Administration recognized that even a re-opened Guantánamo Bay would prove incommensurate to the task. The Administration sought to obtain a network of safe haven zones for Haitians in the Caribbean; when Caribbean states refused to resettle the refugees, Clinton sent them to Guantánamo. With his options for controlling the refugee flood severely limited, the president committed the US to quickly restore democratic government to Haiti. The exodus of Haitian refugees slowed and finally stopped with the September 1994 invasion.
The Response of Caribbean States
While the US response to the refugee crisis was the most visible in the region, few other states responded in more generous manner. The Dominican Republic, whose government had not-so-quietly supported the Haitian military and permitted supplies (particularly fuel) to seep though the international embargo during the coup years, sought to stem the flow of what ended up as a tide of 30,000 refugees in the months after the coup. 4 After the failure of the Governors Island Agreement to return Aristide to Haiti in October 1993, another wave of military repression in Haiti sent 2,500 to 3,000 additional Haitians into the Dominican Republic. In response, the Dominican military bolstered the police and army presence on the border and reportedly forced many Haitians back to Haiti, sometimes handing fleeing refugees over to the Haitian military. 5 The asylum process was highly restrictive. Under procedures established by the Dominican government in conjunction with UNHCR, refugees were required to apply to UNHCR for asylum. If UNHCR determined that an applicant met the requirements for political refugee status it notified an intergovernmental commission, the Comisión Nacional para los Refugiados (CONARE), which had the responsibility for granting refugee and asylum status. Out of a pool of perhaps 30,000 refugees fleeing the anti-Aristide coup, only 2,800 Haitians were able to seek UNHCR recognition. While approximately 1,300 were recognized under UNHCR's mandate, CONARE granted refugee status to only 35 applicants.
Efforts to secure refuge in other Caribbean states had very mixed results. Jamaica placed its refugees in an isolated camp and offered minimal refugee processing. The government of The Bahamas was actively hostile toward the refugees who landed on its shores, threatening (in some cases succeeding) to send refugees back to Cedras' Haiti, and minimizing opportunities for the processing of refugee claims. Other countries in the region refused to accept Haitian refugees for settlement, despite intense pressure from the US. This group included Venezuela, whose president had sent the plane that took Aristide out of Haiti and to refuge in Caracas.
Aftermath of the Crisis
The end of the 1991-94 Haitian refugee crisis did not end serious concerns about Haitian migration flows. The most obvious concern has been the possibility that political and economic instability might generate a new outflow of migrants, perhaps set off by an abrupt increase in political violence. Such was the case in November 1995 when the US Coast Guard intercepted more than 1,100 Haitians at sea fleeing a rise in political violence and uncertainty prior to the December presidential election won by Rene Preval. 6
Just as seriously, the US response left the Refugee Convention in tatters in the Caribbean, its fundamental proscription of refoulement savagely violated, its contribution to the anti-immigration, anti-refugee mood sweeping North America and Europe clear and wide-ranging. US, regional and international attention turned to rebuilding Haiti's economy and political institutions. International donors have financed projects to disband the military and build a new national police force, reform the judicial structures, support the development of a democratic, rights-respecting civic culture, and rebuild the shattered economy. A US/UN military and police presence have provided security against political violence and a large Coast Guard fleet still guards the sea lanes against rafters.
But three years after the US/UN intervention, the county remains mired in a seemingly-endless political crisis and the economy is moribund. The UN civilian and military missions are scheduled to pull out at the end of November, although it seems likely that the US and UN will arrange for a small military and police presence to remain to avoid any kind of destabilizing event that might generate a new refugee outflow.
While the US and UN have invested hundreds of millions of dollars in Haiti, nothing has changed on the refugee front. No regional mechanisms have been developed to deal with future crises emanating from Haiti, Cuba or elsewhere. The US response would probably mirror its reaction to the 1991-94 situation: raise a wall of ships around Haiti and repatriate refugees or reopen Guantánamo Bay as a temporary safe haven. It is also likely that states in the region will follow the lead of the US and seek to close their borders and quickly repatriate any Haitians who do make it ashore. While this view applies with particular force to the Dominican Republic, it is most likely true of Jamaica, The Bahamas, the Turks and Caicos, Martinique, Guadeloupe, St. Martin and French Guiana, all of whom have deported or expressed interest in repatriating a large number of Haitians residents since 1994.
These efforts at massive repatriation reveal a migration dynamic that poses serious political and economic problems for Haiti. It also exposes these migration populations to serious human rights abuses as a result of their tenuous immigration status. It is to this dynamic that I turn in the remainder of this paper.
Haitian Migration in the Caribbean
The 1991-94 refugee crisis exacerbated longstanding tensions in various Caribbean states related to historic Haitian migration flows. Haitian migrants in the Caribbean are part of a complex pattern of economically- and politically-induced population movements that have long characterized the region. Laborers recruited formally or secretly to work temporarily harvesting seasonal crops end up forming permanent communities which are augmented by later waves of migrants which expand into other economic sectors (usually agri-business, construction, domestic service, tourism). These communities tend to be geographically, culturally and linguistically isolated from the host society and very poor, often lacking access to basic services. Many migrants are given temporary working visas which must be annually renewed by the employer. The rest work without proper visa documentation. Children of the migrants are often denied proper residency permits or citizenship in states with a jus soli nationality principle. Attending school and obtaining needed social and health services can be difficult. With the risk of deportation hovering in the background, employers are able to exploit the immigrants. 7
With uncertain immigration status, the immigrants cannot complain or press for better living and working conditions. They are generally avoided by the host populations and subjected to discrimination when contact is made. When economic cycles turn downward, immigrant groups are blamed for taking jobs away from nationals and for requiring too much educational, health and social service spending (when those services are available). Elections are won by candidates promising to deport all illegal immigrants, and a number of governments are able to deport large numbers of Haitian residents, including the second- and third-generation descendants of Haitian immigrants who have become fully culturized and who know little of Haiti, have few, if any, relatives there, and do not speak Creole (or French).
Aristide's return in November 1994 and the subsequent re-establishment of democratic institutions in Haiti prompted a number of Caribbean states to initiate programs for the repatriation of the Haitian populations within their borders. The United States returned 20,000 from Guantánamo between September 1994 and January 1995; The Bahamas repatriated approximately 8,000 Haitians from February to December 1995, and is seeking to extend the repatriation accord with Haiti; the Dominican Republic deported 20,000 in the first three months of this year; the Turks and Caicos is in the process of repatriating 3,000 this year; smaller numbers have been deported from other island states; and Martinique has expressed interest in a repatriation program like that of the Turks and Caicos. Repatriated Haitians have returned to find a government unable to provide the resources or administrative structure for resettlement and an economy which has yet to revive enough to offer employment opportunities. Many of them attempt to return to the state from which they were deported.
Haitians in the Dominican Republic
Haiti and the Dominican Republic share an island and a long and antagonistic history dating form the colonial period. Racial, language and cultural differences have melded with historical memory (the 22-year Haitian occupation for Dominicans, Trujillo's 1937 slaughter of at least 10,000 cane cutters for Haitians) and political demagoguery to leave the Haitian populations in the Dominican Republic vulnerable to serious discrimination, abuse and economic exploitation. The seasonal migration of low-wage Haitian cane cutters, which began at the turn of the century, gradually left a large permanent population in Dominican agricultural zones and major cities. That migration seems to have been supplemented in recent years by a much broader flow seeking work in a broad range of Dominican economic sectors, including agribusinesses like rice and coffee, construction, domestic service, tourism and textile factories in free-trade zones. 8 The population has grown to an estimated 500,000 residents today.
This population has often been used politically within Dominican domestic politics and in bilateral relations with Haiti. Three months prior to the September 1991 military coup which drove Aristide from Port-au-Prince, President Balaguer responded to the international criticism of Dominican treatment of Haitians in the sugar industry by ordering a massive deportation of Haitian cane cutters. In the ensuing chaos, an estimated 30,000 men, women and children flooded across the border. The deportations were effectively reversed by the coup, which sent approximately 30,000 Haitians fleeing into the Dominican Republic.
Periodic mass expulsions of Haitian residents has been the rule in Haiti-Dominican history. Trujillo's massacre of 10,000-20,000 Haitians in the border area in 1937, a wave of deportations in 1981 (perhaps more than 4,000) following press reports of the extensive use of Haitian labor in agriculture, Balaguer's decree in 1991, a military-directed campaign deporting several thousand Haitians prior to Aristide's aborted return attempt in October 1993, a massive deportation of 20,000 Haitian residents between January and March of this year, and a new deportation campaign launched during the first week of September 1997 are the major episodes of this long-running and bitterly tragic drama between the two states.
Neither Haiti nor the Dominican Republic have ever expressed a serious willingness to tackle the migration conflict. Migration was somewhat controlled during the Duvalier years when the dictator (and later his son) sold their citizens for work in the Dominican cane fields, but the post-coup government has had little interest in and capacity to control migration outflows to Haiti's neighbor. A positive step forward in 1996 -- the establishment of a bilateral commission to address issues of mutual concern, including migration -- has been derailed by the massive deportations at the beginning of this year. Outrage inside Haiti provoked the Preval government to strongly condemn the Dominican policy, generating tensions which have frozen further discussions of the issue. A protocol agreed to by the two governments governing the process for repatriations has not been followed. The Dominican government, however, has recently expressed a greater sensitivity to international and domestic criticism on this issue and has modestly improved the way in which the deportations are carried out. Suggestions by governmental officials that an international organization supervise the repatriations to assure that international norms are complied with have not yet born fruit.
Other Vulnerable Populations: The Bahamas and Turks and Caicos
Similar inchoate repatriation exercises are occurring elsewhere in the Caribbean. Jamaica used the damage wrought by a hurricane to return almost of its Haitian population in 19--. Haiti signed a two-page agreement with The Bahamas in January 1995 for the return of 800 Haitians and Bahamians of Haitian descent each month over a 12-month period. Commitments to review the immigration status of each individual selected for deportation and provide legal residency for those who were long-time residents or had been born in The Bahamas were not fulfilled. Similarly, Haiti has entered into an agreement with the Turks and Caicos to repatriate 3,000 of its 10,000 Haitian residents; the Turks and Caicos agreed to grant legal residence to 1,000 Haitians. These numbers were pulled out of a political hat and have no necessary relation to the immigration claims individual members of the Haitian population may have in the islands.
Responses to the Repatriations: Haiti and the International Community
The problem for Haiti is that the state has no capacity to absorb or reintegrate returnees. The economy is, at the moment, moribund. Thus, jobs for thousands of recent arrivals will be almost impossible to find. Housing, education and job training funds are scarce. The International Office for Migration (IOM) is operating a migrant reintegration program whose effectiveness is limited. The Haitian National Office for Migration (NOM) is surviving on a grant from the European Union for emergency assistance for returnees and lacks the technical capacity or manpower to develop a Haitian migration policy or provide much direct help to returnees. It must turn to IOM for technical assistance, basic staff training and the implementation of assistance programs for the returnees. In the welter of pressing domestic economic and political crises, migration is an issue that no one in the Haitian government wishes to address. And, finally, even if it wished to tackle migration problems, the Haitian government lacks the capacity to do so -- it has no data on migration patterns, cannot track the movements of its citizens (most of whom lack any kind of official identity document), and the new Haitian National Police has a tiny border patrol and coast guard which cannot control the flow of border or sea traffic.
The problem for the migrants who are deported is they usually have no way to demonstrate their legal (or moral) right to reside in the host country -- repatriations (particularly in the Dominican Republic) are usually carried out randomly and collectively, often separate families, and often return long-time residents to a country where they no longer have family or friends and children to a land whose language they may not speak.
International governmental organizations have not been of much assistance -- migration issues of this nature do not appear to fall within anyone's mandate. UNHCR limits itself to convention refugees and displaced persons. The World Bank and the Inter-American Development Bank tend to ignore the migration effects of economic development strategies they finance, even though the transition to an open, market-based economy is expected to generate migration outflows in the short term. 9 The European Union has targeted funds for joint Haiti-Dominican projects in the border area through Lomé IV, but it is not clear that those funds will be employed to address migration concerns. Although the Organization of American States has played an important role in resolving the 1991-94 Haiti crisis (and the Inter-American Commission on Human Rights has launched a study of migrant labor in the hemisphere), broader Caribbean migration issues are not a focus of its present work either. CARICOM might be a candidate to take an interest in these issues now that Haiti has become a member, but migration has not been high on its agenda in the past.
Stabilizing the Migration Flows: Ruminations in Lieu of Recommendations
Stabilizing these kinds of migration flows is a rather daunting task, given the necessity of cooperation across regional and international organizations and governments. What, for example, would be necessary in the Haiti-Dominican Republic context (a promising situation to the extent both countries share a discreet geographic space) to begin to address the migration conflicts? First, one would have to recognize a number of rather discouraging facts: Haiti's presently destitute economy, degraded environment and feeble government capacity 10; the economic differentials between the two countries and well-developed Haitian migration networks that will continue to facilitate migration to the Dominican Republic; economic and political transition difficulties in the Dominican Republic, itself a sender of large numbers of migrants to the US via Puerto Rico; the political sensitivity of the migration issue on the island; the lack of technical resources and funds for migration matters.
Second, it seems clear that any approach must resolve the legal status of the Haitian populations in the Dominican Republic. A plausible solution would be to grant permanent residency to long-term residents, citizenship to those born in the Dominican Republic (as stipulated by the Dominican constitution) and work permits to shorter-term residents with jobs. All remaining non-documented Haitian migrants would be subject to deportation procedures agreed to between both governments and supervised by an independent local or international organization. The two states might then govern their labor arrangements with a migrant labor accord supervised by an independent organization. 11 In turn, the two states -- but particularly Haiti -- would have to gain greater control over the border and restrict to the extent possible unauthorized crossings.
Third, development organizations and donors will have to include migration stabilization as a short-term goal of their work in both countries. Specialists insist that economic growth will eventually resolve poverty-linked migration outflows, but Haiti (as well as the Dominican Republic) faces years -- perhaps decades -- before growth could, under the best of circumstances, reach that level. Therefore shorter-term options must be considered. This might mean that the international financial institutions incorporate migration avoidance into their development programs for Haiti (and its neighbor), targeting out-migration regions for both rapid-impact and small scale sustainable economic projects. This might mean European Union support for major infrastructure investments (irrigation, hydroelectric) and free trade zones in the border region. It might mean technical and financial assistance to both governments for designing and implementing modern immigration and border control systems. 12
Fourth, one or more international and/or regional organizations must take the lead to put migration on the agenda in the Caribbean. Haiti and the Dominican Republic have great difficulty (for historical and political reasons) discussing migration in a policy context. The two states have not been able to agree even to study the issue (migration flows and migrant characteristics; a census of the Haitian population in the Dominican Republic) or hold joint workshops. They need a broader regional umbrella structure to both make serious discussions politically possible and to receive the technical assistance and financing for migration management programs. While the medium- and long-term migration challenges are daunting, and the prospects for significant improvements on these issues not promising, it is not clear that short-term efforts to identify vulnerable populations, assist states with migrant legalization and humane repatriation where appropriate, and focus development work on migration stabilization may not generate small, but important improvements.
At stake with these improvements is the ability of tens of thousands of Haitian men, women and children to escape lives lived under intense discrimination and critical poverty, the prospects for Haiti as a democratic state, and the possibility for reducing tensions between Haiti and its neighbors. Haiti's recovery will eventually depend upon its ability to compete in the international marketplace. Economic recovery will therefore mean challenging its Caribbean neighbors in the quest for a fair share of international capital, technology, markets and consumer and tourism dollars. Competitive challenges generate political friction and can lead to retaliatory expulsions of expatriate Haitian populations, always a threat to Haitian political, as well as economic, stability. The way beyond this stability dilemma lies in interlocking formal regional discussions (and perhaps agreements) concerning immigration, refugees, trade and economic development. Those discussions and agreements, however, will only come about as the result of regional initiatives that are not yet on the horizon.
1 Norman L. Zucker and Naomi Zucker, Desperate Crossings: Seeking Refuge in America (New York: ME Sharpe, 1996), p. 78.
3 Open Society Institute Special Report, A Proposal to Establish a Temporary Refugee Scheme in the Caribbean Region for Refugee and Migration Emergencies (December 1995), p. 3. Approximately 68,000 Haitians were picked up at sea between the coup and the date of Aristide's return.
4 US Committee for Refugees, 1995 World Refugee Survey, p. 177.
6 William Branigin, 'Surge of Haitian Boat Refugees is Largest Since Aristide's Return', Washington Post, November 29, 1995, p. A28.
7 See, e.g., National Coalition for Haitian Rights, Beyond the Bateyes: Haitian Immigrants in the Dominican Republic (May 1996); Wilfredo Lozano, ed., La cuestion haitiana en Santo Domingo: Migracion internacional, desarrollo y relaciones inter-estatales entre Haiti y Republica Dominicana (Santo Domingo: FLASCO: Centro Norte-Sur de la Universidad de Miami, 1993).
8 Ruben Silie, 'Politica migratoria y relaciones interestatales Republica Dominicana y Haiti,' paper presented at the conference Republica Dominicana y Haiti: Hacia el 2000, Santo Domingo, Dominican Republic, 1996, pp. 9-12.
9 Josh DeWind and David Kingly argued in 1986 that World Bank and USAID development strategies promoting a shift away from domestic agriculture toward export crops and assembly export production only intensified the impoverishment of the rural population and increased migration from Haiti to the U.S. and other Caribbean states. Josh DeWind and David Kinley, Aiding Migration: The Impact of International Development Assistance on Haiti (New York: Immigration Research Program, Center for the Social Sciences, Columbia University, 1986). In a recent paper, Philip Martin and J. Edward Taylor explain that "[f]reer trade may increase imports before exports rise, producing a currency crisis, devaluation, recession, and emigration, as in Mexico in 1995. Multinationals tend to use more imported components, so that breaking up local monopolies and attracting direct foreign investment can increase imports, the use of capital-intensive production techniques, and exports, without increasing the number of jobs. Finally, aid in the form of infrastructure can have the perverse effect of stimulating emigration, as when better roads meant to help farmers to market their crops also permit cheap imported food to reach even the countryside, destroying jobs and stimulating emigration." Managing Migration: The Role of Economic Policies (Paper prepared for the Migration Policy in Global Perspective conference at the New School for Social Research, September 8, 1995, revision dated December 21, 1995), p. 31. Regardless, Martin and Taylor argue that liberal economic polices can be effective in the longer run.
10 This collection of problems is crucial for migration issues. As William Wood notes, 'Haiti's alarming levels of land degradation, exacerbated by decades of misguided agrarian policies, is intertwined with desperate living conditions in rural areas and is a major factor in rural to urban migration and even emigration from Haiti's shores. Haiti's people have endured successive generations of intensifying poverty, soil erosion, deforestation, and shrinking average farm sizes, leaving many peasant families unable to meet bare subsistence levels. Since the 1800s, Haitians have tried to escape poverty by seeking work on sugar cane plantations in neighboring countries, often under very harsh conditions. Recent labor immigration restrictions, continued rapid population growth in Haiti, declining crop yields, and very high unemployment rates are intensifying pressure in Haiti's already destitute rural economy and heavily degraded landscape.' William B. Wood, 'Ecomigration: Linkages Between Environmental Change and Migration,' Migration Policy in Global Perspective Series, Working Paper No. 3 (The International Center for Migration, Ethnicity and Citizenship, New School for Social Research, 1996), p. 13.
11 The International Organization for Migration is assisting Costa Rica and Nicaragua with the implementation of a bilateral migrant worker accord, acting as a neutral intermediary in the recruitment, contracting, monitoring and repatriation of the seasonal workers. See Convenio de Mano de Obra Migrante entre el Gobierno de Costa Rica y el Gobierno de la Repdblica de Nicaragua para Regular el Ingreso y Permanencia de Trabajadores Migrantes No Residentes, dated January 6, 1993.
12 More importantly, if economic policies are to work, the US must permit the new economic policies adopted by emigration states to function by opening its markets to free trade. US development initiatives such as the Caribbean Basin Initiative, for instance, usually exclude most of the goods in which Caribbean countries have a comparative advantage and only permit imports of goods assembled in export zones, a practice that does not foster linkages with producer or other sectors of the economy crucial to fostering a sustainable export development strategy (although it does provide jobs). Sugar producers in the US have a powerful lobby and have been able not only to increase tariffs for Caribbean sugar but also to continue to import labor from the Caribbean to cut Florida sugar. The US tariffs appear to have been high enough to be responsible for the loss of substantial jobs in Caribbean sugar producing nations, ironically contributing to US-bound emigration from these very same countries. Robert C. Smith, 'Current Dilemmas and Future Prospects of the Inter-American Migration System,' Migration Policy in Global Perspective Series, Working Paper No. 5 (The International Center for Migration, Ethnicity and Citizenship, New School for Social Research, 1997), pp. 31-32.