IN THE UNITED STATES VIRGIN ISLANDS*
(Revised February, 1999)
*An earlier version of this paper was presented at the International Conference on "Pluralism in the Late Twentieth Century," 7-9 December 1989, ISER UWI, St. Augustine, Trinidad and Tobago.
Klaus de Albuquerque and Jerome L. McElroy have had extensive teaching, research and policy experience in the U.S. Virgin Islands. Their academic research and consulting throughout the Eastern Caribbean has focused on small-scale agriculture, inter-island migration, the implications of political status change, sustainable tourism strategies and, more recently, longitudinal patterns of serious crime in the region.
United States Virgin Islands society is a complex of diverse racial and ethnic groups: black islanders, West Indian immigrants, Puerto Ricans in-migrant, Frenc immigrants from St. Barthelemy, and transplanted mainland "continentals." Recent U.S. Censuses of the Population and Housing reveal these groups exhibit distinct socoio-economic and residential profiles. Compared to whites, blacks and Hispanics have lower income, educational and occupational status, less home ownership but larger families, and higher unemployment and poverty rates. These groups also engage in differing recreational and associational activities. This evidence suggests a heterogeneous society (Smith, 1965) with a high degree of social and cultural pluralism.
Following Kuper (1969a) we contend that pluralism is a condition of most complex societies. A society is pluralistic when it has, at the minimum, two groups (eidetic or scalar) with differing institutional practices in a system of relatively permanent and regular relationship with each other. It follows, therefore, that pluralism refers quite simply "to a diversity of groups which sustain social relationships with one another" (Kuper, 1969b:465). If these groups share the same set of basic or "compulsory" institutions, but can be distinguished by their membership in differing "alternative" and "exclusive" institutions, then the ensuing society is often defined as a heterogeneous society (Smith, 1965). If, however, a "society" consists of national cultural segments which have minimal contact with each other and are differentially incorporated into the polity, then a plural society is said to exist (Smith, 1969)1.
We have argued elsewhere (de Albuquerque and McElroy, 1985) that the United States Virgin Islands (USVI) can be best characterized as a heterogeneous society with a high degree of social and cultural pluralism (in terms of group diversity and the differential incorporation of these groups into the polity and the economy). What saves the USVI from being labeled a plural society is the fact that the various racial/cultural segments participate in the same basic institutions but are distinguished by their differing "alternative" (e.g., occupations) and "exclusive" (e.g., sports clubs) institutional affiliations (see Despres, 1968:11). In broad terms, the USVI experience embraces four constitutive and interacting elements: (1) a variety of racial, ethnic and nationality groups with differential access to wealth, status and power; (2) a highly complex system of social stratification based on race/color and class, the two often being coincident; (3) a historical shift in the racial tradition from a more "flexible racial tradition" to a more "rigid racial tradition," the latter being associated with the United States' acquisition of the islands (de Albuquerque and McElroy, 1985); (4) a persistent immigrant tradition after 1960, with two types of immigrants, one white and superordinate, and the other black and subordinate, to the native population. Kuper (1969b:46) has argued that of the various types of pluralism, the racial and cultural pluralism exhibited by a society like the USVI, is the most "enduring," "comprehensive" and "historically unique."
This paper attempts to document the correlates of pluralism in the USVI. It draws evidence primarily from the 1980 and 1990 U.S. Censuses of Population and attempts to demonstrate that the various racial, ethnic, and nationality groups have differing family, household, occupational, labor force, income, and educational characteristics, are often residentially segregated, and engage in widely disparate and exclusive recreational and associational activities. These groups, for all intents and purposes, constitute separate communities and the relationship between them is most often colored by mutual distrust, misunderstanding, and hostility.
The Virgin Islands were so named by Columbus during his second voyage in 1493. Three of the islands that now make up the United States Virgin Islands (USVI) were colonized progressively and eventually controlled by Denmark: St. Thomas in 1692, St. John in 1717, and St. Croix in 1733. The economy of the Danish West Indies (St. Thomas, St. John and St. Croix) shifted with the fortunes of sugar and commerce and began a steady deterioration after the abolition of slavery in 1848. Between 1835 and 1917, the population declined by forty percent as a result of persistent emigration and natural decrease. To counteract the effects of emancipation and emigration on the local labor supply, the Danish government set up an immigration bureau in 1855 to recruit laborers from the nearby British, French and Dutch islands.2 At roughly the same time Bretons (French) began to arrive in St. Thomas from the island of St. Barthelemy. This complemented earlier minor migrations from Saba and St. Maarten (Crane, 1971; Lewis, 1972). By the time of the United States purchase of the islands in 1917, almost one fifth of the total population had been born in other West Indian islands.
The transfer of rulership brought with it an attendant change in the pattern of race relations. Whereas Danish Virgin Islands society was a highly stratified pigmentocracy where race/color relations were carefully controlled by an intricate system of etiquette, the new American Naval administration, which governed between 1917 and 1930, was overtly racist and quite unprepared to deal with the highly complex constellation of race-color-class patterns they encountered. Virgin Islanders of all colors and classes were treated rather crudely and were pointedly excluded from social intercourse with Naval personnel. The racial situation did not significantly improve under a succession of civilian governors appointed in Washington, and it was only in the 1960s with the emergence of the civil rights movement on the United States mainland, stronger local government, and the appointment of local governors, that the American presence lost some of its harsher colonial manifestations.
All the while, sporadic waves of immigrants from nearby islands had been attracted to the Territory (technically the USVI political status is that of Unincorporated Territory) by periodic booms and lax immigration restrictions. For example, the U.S. repeal of the prohibition on alcohol measurably increased the demand for workers in the St. Croix sugar fields, and cane cutters were contracted annually from the Leeward Islands, particularly Antigua. During World War II many Eastern Caribbean workers were recruited to build a submarine base on St. Thomas and to replace local laborers who had volunteered or been conscripted for military service.
The most extensive immigration, however, began after 1960 as a result of a new diversification policy of aggressive tourism promotion and export industrialization. During the 1960s, real per capita income grew ten percent per year fueled by the boom in hotel construction, related services, and light and heavy manufacturing (McElroy, 1978). To meet the demands of this rapidly expanding economy, unprecedented numbers of Eastern Caribbean workers were recruited under a non-immigrant worker program. Most, subsequently had their statuses adjusted, and stayed on to build a new life and raise their children in the USVI (de Albuquerque and McElroy, 1982).
During the 1970s the USVI experienced three export-induced recessions and a dramatic decline in construction activity. Tourism problems were exacerbated by crowding in the main towns on heavy cruise ship days, public utility breakdowns, and rising crime. Unemployment reached double-digit levels and fiscal deficits met by U.S. Congressional bailouts became routine. These changes, which resulted in tighter immigration controls, severely restricted the inflow of Eastern Caribbean Labor. At the same time, however, there was considerable in-migration from the United States mainland. During the 1980's, the economy regained momentum and population growth spurted in some years to above 3 percent per annum. The growth was due primarily to the influx of white mainlanders, and to a lesser extent Puerto Ricans, on the heels of a renewed tourism/finance mini-boom, sparked in part by political instability in other resort destinations and economic downturns in Puerto Rico, Texas, Louisiana, and Florida. In the 1990s, population growth declined significantly with many Eastern Caribbean immigrants returning as retirees to their home islands and many long-term residents returning to the continental U.S. The 1990s, have also seen an increase in the number of migrants from the Dominican Republic and Haiti. The economy of the USVI has taken a downturn in the 1990s following the destruction caused by two major hurricanes--Hugo in 1989 and Marilyn in 1995. The latter wrought extensive destruction on St. Thomas and St. John, while the former devastated St. Croix. Both hurricanes severely damaged the tourism plant, affected many businesses, and resulted in significant out-migration to the United States.
Hurricane Hugo also postponed a referendum on the future political status of the USVI. Chafing under the lack of equality (Virgin Islanders, although U.S. citizens, cannot vote in presidential elections and do not have voting representation in the U.S. Congress), the Virgin Island Legislature established a new Commission on Status and Federal Relations in 1988 (there had been several earlier commissions in the 1980s) and began the task of political education on various political status options. But the electorate, consisting of a very large minority of persons born outside of the USVI, was largely uninterested in the status question. When a referendum was finally held in 1993, voter turnout was poor (27.5%) and the majority who voted (80%) opted for the status quo (remaining an Unincorporated Territory), rejecting the uncertainty of incorporation, or integration through statehood, or going it alone (independence) (Krigger, 1994)
Governor Schneider and several other USVI politicians began pushing in 1996 for a new Constitutional Convention to draft a constitution (subject to approval by the U.S. Congress), one feature of which would address the USVI-Federal relationship. A constitution would also define who is a Virgin Islander, a definition that is bound to be highly controversial given the number of migrant groups resident in the USVI. An attempt to define "Virgin Islander" was quietly written into a bill (1996) allowing casino gambling on St. Croix, and, when it became public knowledge, the definition raised considerable ire and opposition (the strongest reaction coming from naturalized citizens born elsewhere in the Eastern Caribbean) because it linked "Virgin Islander" to having one ancestor born in the Virgin Islands prior to 1917.
The question of native Virgin Islanders versus non-natives simmers in the background and resurfaces periodically, as it did in the casino bill, and over the New Visitor Center the U.S. National Park Service was building on St. John. Complaints regarding the latter revolved around the lack of consultation with the local community, and the question whether all the various groups had an equal right to speak for St. John--native St. Johnians, long-term Eastern Caribbean migrants and white continentals, part-year residents, and recently arrived in-migrants from the U.S. mainland. In November 1998, Governor Schneider running for a second term was defeated by former Commissioner of Education, Charles W. Turnbull. The new Turnbull administration's major concern revolves around reviving the moribund economy, and in particular, reducing the recurring budget deficits.
Data were obtained from the U.S. Censuses of Population and Housing for the USVI. We relied heavily on the 1980 and 1990 censuses. We should caution, however, that the 1990 census was concluded less than seven months after hurricane Hugo, when St. Croix was still in the throes of the recovery process and many residents were off-island having lost both their houses and their jobs. For this reason we report both 1980 and 1990 data, and where we have doubts about the quality of the 1990 figures, report only the 1980 data.
DIVERSITY OF GROUPS
In two previous articles (de Albuquerque and McElroy, 1985, 1989), we documented the diversity of racial, ethnic and nationality groups in the USVI, and noted some of their respective demographic and socio-economic characteristics.
According to Table 1, between 1917 and 1970, the most significant change in the racial composition of the population has been the increasing importance of whites, stemming originally from the American presence following the purchase of the islands from Denmark (1917), and in the 1960s and 1970s, from the rising in-migration of mainland entrepreneurs, professionals, construction, and service workers, attracted by expanding insular economic opportunities. After 1970, the white component of the population declined from 18.2 percent to 14.8 percent in 1980 and 13.5 percent in 1990.
Several prominent observers of USVI society claim that the white population is larger than the census estimates, pointing to its transiency--many whites are part-year residents or live on boats. The latter population of "yachties" has been historically undercounted in the censuses.
(Table 1 About Here)
Krigger (1988) maintains that this population was undercounted in 1980. Undercounting of white residents also occurred in 1990 because many whites had left temporarily after the onslaught of hurricane Hugo.
The most significant increase in the white population has occurred on the island of St. John. In 1917, only four whites were enumerated on St. John. By 1960, whites constituted 10.5 percent of the population, and by 1980 their proportion had more than doubled to 26.3 percent. The 1990 census count placed the white population of St. John at 34.5 per cent (Table 1). Not only are whites on St. John very noticeable, but their presence is often overwhelming, particularly when one factors in an average daily population of several hundred white tourists. Thus the sense black St. Johnians have of being an embattled minority is very real.
Another noticeable trend is the declining significance of the black population through 1960 (Table 1) resulting from the increasing importance of both white and "mixed and other" (mainly Puerto Rican) minorities, and the rise in the black proportion, thereafter, reflects the heavy immigration from the Eastern Caribbean. Since 1980, we estimate that the black proportion of the population has declined significantly on St. John, (from 72.6 % to 64.7 %), has risen slightly on St. Thomas, but has declined by seven percentage points on St. Croix primarily because of the increase in the "mixed and other" populations (mostly Hispanic).
Interpreting shifts in the "mixed and other" population is less straight forward because of the ambiguities that underlay the old system of race/color classification. Declines recorded between 1917 and 1930 partially derive from changing definitions of race imposed on local census enumerators by the U.S. Bureau of the Census. Similarly, abrupt declines between 1960 and 1980 were primarily due to a change-over from the traditional enumerator designation system to self-identification. In addition, in 1980 and 1990 the category of "mixed or other" was reduced to a single category of "other", reflecting the American inability to perceive of race in other than black and white categories. The striking increase in the "mixed and other" population of St. Croix between 1930 and 1960 was due to in-migration of Puerto Ricans to St. Croix from the off-shore islands of Vieques and Culebra (Table 1). In 1980, Hispanics constituted 14.2 percent of the USVI population, with St. Croix having the largest proportion (22.2%). Racially the Hispanic population in 1980 was broken down as follows: 57.6 percent black, 16.3 white, and 26.1 "other" (U.S. Bureau of the Census, 1984b). By 1990, the Hispanic proportion of the USVI population had only increased marginally to 14.8 percent (this figure differs from the 14.5% derived from block data). Racially, the largest proportion of Hispanics (57.6%) reported themselves as "other" (mixed), 34.3 percent identified themselves as "black," and only 8.1 percent as white (U.S. Bureau of the Census, 1993b). These shifts represent a significant change in racial self-identification between 1980 and 1990, with Hispanics preferring to refer to themselves as "mixed" or "brown" in 1990, rather than
"black," most likely to distinguish themselves from native black Virgin Islanders and Eastern Caribbean immigrants. The 50 percent decline in the number of Hispanics reporting themselves "white" reflects the growing Hispanic Crucian identity, and widespread negative sentiments in the USVI towards whites.
Defining discrete ethnic categories in the USVI is a difficult exercise because (1) ethnicity is often circumscribed by nativity, color, jural status and socio-economic marginality; (2) groups have multiple identities from which to choose, that is, ethnicity is often situational; and (3) there are numerous instances where ethnic self definitions and definitions by others are at variance.
Table 2 shows the nativity composition of the population for various census years beginning in 1917 and dramatically underscores the importance of immigration and in-migration (from Puerto Rico and the continental United States) in shaping the USVI population. Of particular note is the substantial decline in the native born component of the population, from 76.2 percent in 1917 to 47.2 percent in 1980. The native born proportion of the population rose to 49.0 percent by 1990, and currently we estimate that native born Virgin Islanders are a majority for the first time since 1970. Although not a discrete ethnic group, native Virgin Islanders often act as if they constitute a separate group. Caught between their lucrative American connection and their Eastern Caribbean heritage, Virgin Islanders frequently exhibit a kind of cultural marginality. Unwilling to fully embrace white or black America, or their Leeward Island cousins, from whom they cannot be meaningfully separated socio-culturally, they have shown an inordinate preoccupation with trying to uncover and authenticate a separate Virgin Islands culture.
In an earlier paper (de Albuquerque and McElroy, 1985), we noted that nativity, color, family name, and length of residence, play a very important part in the sociological definition of a Virgin Islander. Recent evidence suggests that this definition is becoming more flexible (the inclusion of the USVI born children of Eastern Caribbean immigrants) but not sufficiently elastic to also include the USVI born children of white and black mainlanders.
Fluctuations in the population born elsewhere in the Eastern Caribbean (Table 2) parallel the economic fortunes of the USVI economy. As noted earlier, massive immigration during the 1960s was in response to major transformations in the economy: expanded infrastructure, heavy industrial construction, and burgeoning labor-intensive tourism. The pace of immigration from the Eastern Caribbean slowed down in the 1970s because the economy contracted and immigration restrictions were consequently enforced (de Albuquerque and McElroy, 1982).
In 1960, the largest single group of non-native born Leeward Islanders were from the neighboring British Virgin Islands (Table 2), a testament to the relative ease of movement within the Virgin Islands archipelago. However, after 1960, St. Kitts, Nevis and Anguilla and Antigua/Barbuda became the most important suppliers of labor to the USVI, followed by Dominica, St. Lucia and Trinidad and Tobago.
(Table 3 About Here)
Data on nativity by island of residence suggest some island specific patterns of migration. For example, persons from Aruba, Bonaire, Curacao, and Trinidad and Tobago, with oil refining experience, were recruited/gravitated to St. Croix because of the Hess Oil refinery, one of the largest in the world. Additionally, persons from the Windward Islands of St. Lucia, St. Vincent and Grenada were more likely to be attracted to St. Croix because of its agricultural tradition and new industrial development, while Leeward Islanders from St. Kitts, Nevis, Anguilla, the British Virgin Islands, and St. Maarten found employment in the hotels and other tourist related enterprises on St. Thomas. Antiguans, because of their long term association with St. Croix (through earlier recruitment as cane cutters), apparently showed a preference for St. Croix as a destination. Data for 1990, although not complete, show that the patterns remain. The majority of Anguillians, Kittitians/Nevisians, British Virgin Islanders and Dominicans (from Dominica) continue to live on St. Thomas, while Trinidadians and St. Lucians dominate on St. Croix (U.S. Bureau of the Census, 1993c).
We argued earlier that Eastern Caribbean immigrants could not be meaningfully separated from native Virgin Islanders by a juridical dichotomy (native/immigrant) which ignores that both groups are part of a larger socio-cultural community. Yet, Eastern Caribbean immigrants have been treated and portrayed in the USVI as if they constituted a separate ethnic group. Admittedly, there are some linguistic and other differences that tend to demarcate St. Lucians, Dominicans, Antilleans (especially from the ABC islands), and the descendants of immigrants from St. Bartholomew, as separate groups, but Kittitians, Nevisians, Antiguans and Anguillians can only be distinguished from Virgin Islanders on the basis of nativity/nationality. However, all Eastern Caribbean immigrants have been historically lumped together, and had associated with them, the derogatory term "garot," suggesting that they were poor, unwashed and uneducated, a common stereotype of immigrant groups.
The strong sense of "home island" manifested by Eastern Caribbean migrants reinforced the perception on the part of Virgin Islanders that they were "not part of us." Likewise, by being denied full participation in USVI society in the 1960s, Eastern Caribbean migrants were pushed to maintain strong home island identities by organizing voluntary associations (sport clubs, benevolent societies, improvement associations, etc.) and by periodic visits home. Even the eventual full incorporation as U.S. citizens of many of these immigrants did not result in any significant denouement of home island attachment and identity. In fact, as many East Caribbean islands achieved political independence and began the process of economic restructuring, segments of the Eastern Caribbean community began expressing an active interest in returning home. While return migration and circulation have always been a feature of inter-island migration in the Eastern Caribbean (McElroy and de Albuquerque, 1988), the volume of returnees from the USVI has risen significantly in the late 1980s and the 1990s, as many of the original migrants reached retirement or near-retirement age. In addition, overcrowding, widespread drug use, crime, and an undercurrent of violence in the USVI have seriously affected their perceived quality of life in their adopted home.
Defining French and Puerto Rican Virgin Islanders as separate ethnic groups is even more problematic because they have a history of long residence on the islands and have been creolized for the most part. Nevertheless, there do exist real cultural differences, largely linguistic and religious, and in the case of Puerto Ricans, also culinary. Both groups have long been negatively stereotyped by black and brown Virgin Islanders. The French were portrayed as unintelligent, incestuous, and socially inferior (de Albuquerque and McElroy, 1985) while Puerto Ricans were characterized as dirty, quarrelsome, clannish money grubbers with "white superiority" beliefs (Senior, 1947:18). Some of these stereotypes have been laid to rest, but ethnicity for both groups remains clearly situational. French Virgin Islanders celebrate "Bastille Day" and Puerto Ricans several Puerto Rican holidays and religious feasts. Indeed, both groups are proud of their heritage and frequently remind other Virgin Islanders of it. But when there is competition for scarce resources, they are quick to claim a native "I born here, too" identity.
Mainlanders from the United States, commonly referred to as "continentals," constitute yet another group. Earlier (1950-1970) white continental in-migrants were usually wealthy retirees with a fondness for tropical living, refugees from the financial world, and a few Bohemians. Since 1970, continental in-migrants have been mostly businessmen, professionals, technicians, craftsmen, construction and service workers, bent on recreating the type of society they left behind. So successful have they been that much of the USVI that tourists come into contact with has been transformed into a Key West/California/Texas version of the Caribbean, replete with Jimmy Buffet style bands, Texas barbecues, Maine lobster nights, Sunday hoedowns, and Monday night American football at the local watering holes. The grafting of this somewhat alien tourist-oriented service culture onto the USVI fabric has produced considerable social distance between white continentals and black and brown islanders. Whites live in the wealthier residential neighborhoods or on boats. They socialize with each other, patronize white establishments, hire whites in preference to others, and send their children to mostly white private schools. Just as Furnivall (1948) noted in his classic characterization of Netherlands India and Burma as plural societies, whites in the USVI meet their fellow islanders only in the marketplace but do not combine. The result is a society that is fractured, or to use Furnivall's term, has no "common social will."
In addition to these five major ethnic groups, there are several smaller enclaves in the USVI mosaic. Black continentals constitute a somewhat special case. Set apart from Virgin Islanders by linguistic and other cultural differences, they were often lumped together with white continentals but now have found greater acceptance particularly among non-white professional circles. Since the mid-1970s, Palestinians and East Indians have become increasingly visible in the USVI. The former have come by way of Puerto Rico and New York and within a short time have progressed from itinerant fruit vendors to owners of gas stations and supermarkets on St. Croix and novelty/department stores on St. Thomas. Palestinians have become an important element in the growing Muslim community in the USVI. East Indians (from India via Jamaica, Curacao, and the Canary Islands) have cornered the market on the sale of jewelry, cameras, electronics, and tourist novelty items. They have successfully penetrated the premiere Caribbean duty-free shopping area in Charlotte Amalie, St. Thomas. The India Association in St. Thomas periodically shows Indian films and runs a special weekly radio program of Indian music.
CORRELATES OF RACE/COLOR, ETHNICITY AND NATIONAL ORIGIN
Table 4 presents data on selected socio-economic characteristics by race and ethnicity (Hispanic). When compared to whites, blacks and Hispanics fare rather poorly and tend to manifest those characteristics associated with low socio-economic status -- they have higher unemployment rates, larger households and families, more single parent households, are less well educated, have lower labor force participation rates, tend to be concentrated in blue collar and service occupations, and are more likely to live in poverty. Clearly the association of race/color and privilege, or lack thereof, is a very noticeable feature of USVI society.
(Table 4 About Here)
In 1980, blacks and Hispanics had significantly larger households, were twice to three times as
likely to have these households headed by females, and had lower home ownership rates (Table 4). In point of fact, a sizeable portion of the black and Hispanic population of the USVI is congregated in public housing projects.3 Their family characteristics also differ from whites in the following ways: a greater percentage of families with children, fewer married couple families, and a larger proportion of single parent families. In 1990, the percentage of black and Hispanic female headed households increased, paralleling U.S. national trends. However, household and family size declined overall. The 1990 census data also show blacks, whites, and Hispanics owning their own homes at roughly similar rates. The increases in black and Hispanic home ownership rates are due to an aggressive local government policy of building affordable low and middle income housing.
Black and Hispanic socio-economic marginality also manifests itself in educational attainment. In 1980, only 39 to 40 percent of blacks and Hispanics were high school graduates as opposed to 85.2 percent of whites. The figures for college graduates were even more disparate; 7 percent of all blacks aged 25 years and above had graduated from a four-year college versus 39.5 percent for whites. By 1990, black and Hispanic educational attainment had shown some improvement, but still lagged far behind whites.
In 1979, white median household income was roughly twice that of black and Hispanic households, $17,281, $9,908, and $9,655, respectively. An alarming proportion of black (37.1%) and Hispanic (39.1%) persons were living below the poverty level in 1979 (Table 4). When poverty status was broken down by age, the figures reveal the stark reality of the USVI economic miracle. Fifty-seven percent of the black and 62 percent of the Hispanic elderly were living in poverty in 1979. Poverty rates for 1989 are more encouraging since they show an overall decline in the poverty level for all groups and significant declines for those persons 65 years and over. Much of this decline is due to the impact of U.S. Federal programs to alleviate poverty, particularly among the elderly. By 1989, the two to one income differential between whites ($35,689) and Hispanics ($17,466) still remained, while blacks ($22,050) had closed the gap somewhat, but still lagged behind whites significantly. With a low wage-high cost of living economy1 dominated by whites, and almost entirely geared to serving tourists, the prospects of narrowing income inequality between racial/ethnic groups are remote. Gini coefficients computed on data on income distribution show a heightening of income inequality between 1940 and 1980 (de Albuquerque, 1989).
Tables 5 to 7 provide a closer look at the relationship between race and ethnicity and income by controlling for education, occupation, and industry classification. A quick perusal of Table 5 indicates that much of the variation in income between whites, blacks, and Hispanics in the USVI disappears when education level is controlled. Whites with one to four years of high school have slightly higher incomes than blacks and Hispanics, but this advantage disappears when it comes to college. A weighted regression of race (a dummy variable was constructed) and education on income, performed on the data, confirms that education is the most significant (.001 level) predictor of income. However, when we control for occupation and industry, it is apparent that white income in all occupations and industry
classifications is higher. One could argue that higher incomes for whites in the same occupation or industry are a result of their higher educational attainment, but this argument loses some of its cogency when one looks at occupations with minimal educational requirements or where practical experience is valued more. A commonly held perception in the USVI is that whites command higher incomes because they are a favored group and not because of their higher levels of education. Indeed, blacks and Hispanics often insist that higher white educational attainment is incidental because whites are most often employed in service related occupations which do not require a college education or specific technical skills.
(Table 7 About Here)
The 1980 census of population (U.S. Review of the Census, 1985) also reports data on income by place of birth. Persons born on the mainland United States (mostly white) had the highest median income (US$ 9,872) in 1979, followed by those born in Jamaica (mostly professionals) at US$ 9,556, and those born in Asia (US$9,500). Persons born in the USVI had a median income of US$ 6,534. Of those born elsewhere in the Eastern Caribbean, median incomes were as follows: Trinidad and Tobago ,US$ 8,231, Aruba, Bonaire and Curacao, US$ 7,700, Barbados, US$ 7,563, Anguilla, US$ 6,422, Dominica, US$ 6,402, Antigua, US$ 6,365, and St. Kitts/Nevis, US$ 6,293.
Income differentials among the West Indian immigrant population are largely due to occupation, with Trinidadians and ABC Islanders working at more lucrative occupations in the oil refining industry in 1980, while largely unskilled Leeward Islanders were confined to service related occupations in tourism and private households.
Occupational and Industry Distribution
One of the structural dimensions of pluralism is the prevalence of racial and/or ethnic stratification in the occupational distribution. The USVI variant of job segregation is reflected in Table 4, which contains data on employment by occupation and industry classification. The data show that the white minority is over-represented in higher-paying white collar occupations while the black majority and Hispanic minority are over-represented in lower-paying blue collar and service occupations. The dominance of whites in trade and commerce (wholesale and retail trade, finance, insurance and real estate) and the dependence of blacks and Hispanics on the public sector is also evident in Table 4. In 1990, 31.6 percent of black, 13.6 percent of white, and 26.4 percent of Hispanic workers were employed by either the U.S. Federal or the USVI local government.
There are also some noticeable linkages between ethnicity/nativity and occupation and industry classification. In examining 1980 census data, de Albuquerque and McElroy (1989) found that Puerto Ricans tend to be over-represented in manufacturing (precision production, fabrication, machine operation) and construction, and under-represented in managerial and professional occupations. Native-born Virgin Islanders on the other hand are over-represented in U.S. Federal and local government jobs (Table 4 shows that in 1980, 38.2% of black workers were in government jobs), and do much better than Eastern Caribbean immigrants when it comes to managerial and professional occupations. Of the Eastern Caribbean immigrants, those born in the British Virgin Islands seem to have found more opportunities in managerial and professional occupations, possibly because of their longer duration of residence and their closer ties (familial and cultural) to U.S. Virgin Islanders. Anguillians and St. Lucians are over-represented in the construction industry (there are quite a number of Anguillian building contractors on St. Thomas), Trinidadians in manufacturing (the Hess Oil connection), and Antiguans, Kittitians and Nevisians in the services sector (as domestics, hotel workers, etc.) (U.S. Bureau of the Census, 1985, and 1993b).
The preponderance of Eastern Caribbean immigrants in the less desirable lower-paying occupations is a legacy of their earlier tenuous status as non-immigrant laborers and of the ruthless exploitation they experienced in the 1960s and early 1970s. Several reports written during this period document that Eastern Caribbean migrants were "in dire need of manpower and related services" (USVI Department of Labor, 1969:69) and institutionally neglected in terms of health, welfare, and educational services (SERD, 1968:52). In response to these reports, the United States Federal Government introduced a series of legal reforms, including allowing "Alien" children access to USVI public schools.5 While these reforms improved conditions, they could not significantly alter the occupational and social status of Eastern Caribbean migrants in the USVI. Support for this group gradually developed through the local government's "Alien Emphasis Program" and through voluntary associations such as the "Alien Interest Movement." By the mid-1970s, with various U.S. Federal programs in place, and with the number of immigrants exceeding non-immigrants, Eastern Caribbean migrants began to make inroads in all areas of USVI society. In addition to some notable improvements in occupational status, the political clout of a growing population of naturalized citizens of Eastern Caribbean origin was recognized by local politicians. Widespread integration also took place among the working class as sexual unions and marriages between Virgin Islanders and Eastern Caribbean migrants began to occur with some frequency.
In two earlier articles, we noted the spatial dimensions of race/color and ethnicity in the USVI as it manifested itself in noticeable residential segregation (de Albuquerque and McElroy, 1985, 1989). Table 8 attempts to capture this segregation by marshalling data on race and ethnicity by island and block unit. Methodologically speaking, attempting to capture residential segregation at the sub-district level is imprecise because some census block units (e.g. in St. John) are quite large, and the segregation that exists on the ground in the form of enclaves and gilded ghettoes, usually occurs over smaller geographical areas. However, some of the broader contours of residential segregation are
discernible. For example, in St. Thomas whites dominate Water Island, and those parts of the East End, North Side, and West End with beach/cliff frontage and expansive water views. On the other hand, Eastern Caribbean migrants are concentrated in the town of Charlotte Amalie and its environs.
In St. John, whites are concentrated in Central, Coral Bay, and East End subdistricts. Since land is scarce and housing very expensive, Eastern Caribbean migrants have been confined almost entirely to the town of Cruz Bay and its environs (de Albuquerque and McElroy, 1989). Census subdistricts, which are larger on St. Croix, are even more inadequate when it comes to capturing the residential segregation that exists on the ground, with the exception of the East End which is very heavily white (Table 8). There are some noticeable pockets/enclaves of ethnic/nationality groups. Blacks (native and Eastern Caribbean immigrants) dominate the two towns of Christiansted and Frederiksted, and North West and South West subdistricts. Hispanics are disproportionately to be found in Christiansted subdistrict and North Central and South Central. Antiguans congregate in North Central and Sion Farm, Dominicans in North West and Sion Farm, and Kittitians and Nevisians in North and South West, Sion Farm, and North and South Central (de Albuquerque and McElroy, 1989).
Table 9 attempts to look at residential segregation by focussing on smaller census geographical units--namely the block numbering areas (BNA). Unfortunately, trying to uncover segregation through BNAs affords slight improvement and all we can note are those BNAs where non-Hispanic blacks, non-Hispanic whites, and Hispanics are either over-represented or under-represented. Nevertheless, Table 9 0clearly lends itself to more specific interpretations of the spatial distribution of poverty and affluence as it relates to race and ethnicity. For example, certain BNAs having water frontage, or high elevations with expansive sea views, are associated with expensive dwellings, high household incomes, and a disproportionate number of whites, while low lying areas with no sea views often correspond to black and Hispanic lower middle income housing developments. The BNAs also allow us to pick out another aspect of residential segregation--namely between public and private housing. As noted earlier, a very significant portion of native Virgin Island blacks and Hispanics are congregated in public housing projects on both St. Thomas and St. Croix. For example, BNA 9703 on St. Croix, which encompasses a large public housing project, had the largest concentration of non-Hispanic black poverty (53.6% of households) and Hispanic poverty (58.0% of households) in 1990.
Despite their earlier tenuous status and bottom position in the USVI social hierarchy, Eastern Caribbean migrants have fared much better in terms of poverty and housing quality. Originally as non-immigrant workers, cut off from many local and Federal programs and benefits, Eastern Caribbean immigrants worked hard, often at several jobs, and saved at relatively high rates. Once they were permitted to adjust their immigration status, they began to move to lower middle income housing developments. In 1990, this group was to be found scattered widely--from urban neighborhoods, to suburban lower-middle and middle income developments, all the way to homes in exclusive locations. There are, however, still noticeable pockets of immigrant settlements reflecting island origins. For example, in St. Croix, Trinidadians are congregated in BNA 9705 and 9706, St. Lucians in 9709 and 9711, Antiguans in 9707 to 9714. Similarly, in St. Thomas, Kittitians and Nevisians are concentrated in BNAs 9610 to 9614 (U.S. Bureau of the Census, 1993c).
Recreational and Associational Patterns
Earlier it was indicated that the various racial and ethnic groups in the USVI can also be distinguished in terms of their recreational and associational patterns. The different groups occupy separate social worlds and through their associational and recreational activities reaffirm their respective identities and reinforce the heterogeneous nature of USVI society.
At the informal level, socializing often takes place within the group; white "yachties" socialize with other white "yachties"; black stateside (mainland) professionals seek out each other; Eastern Caribbean immigrants patronize establishments of fellow immigrants; Puerto Ricans frequent their own restaurants; and black Virgin Islanders, as an embattled minority, find comfort in each other's company. However, stratification in the USVI is also class-based, and classes occasionally cut across ethnic and racial boundaries. Blacks are often thrust together with Puerto Ricans in public housing projects; the fraternity among professional groups (e.g. lawyers) results in some after-hours socializing; and, at the level of the USVI elite, informal relationships across racial/ethnic lines are fairly common. However, in the main, groups in the USVI tend to observe racial and ethnic boundaries in many areas of social life.
Whites are to be found in the numerous white bars and establishments. They join only certain beach clubs and dominate the three yacht clubs. Not to be outdone, native Virgin Islanders have organized various associations (e.g., Virgin Islands 2000) to promote their interests and to closely monitor developments on their island home. Eastern Caribbean migrants, because of their early tenuous status, were forced to create the kinds of adaptive voluntary organizations commonly associated with immigrant groups. Thus, the world Eastern Caribbean migrants have created is replete with sports clubs (usually cricket clubs), benevolent associations, friendly societies, and rotating credit unions, each representing a particular island. There is the Nevis Benevolent Association, the Anguilla Improvement Association, to name two. During the St. Thomas Carnival, many of these associations put up their own booths (food and drink stalls) and cater to their members and visiting friends from "home."
One of the structural dimensions of pluralism is the differential incorporation of groups into a polity. Furnivall (1948) and Boeke's (1953) classic plural society describes a colonial polity where the subject population is excluded from the political process and the entire system of structured inequality is maintained by colonial control of the instruments of violence. USVI history in this century details the struggle for full incorporation into the American political system, beginning with U.S. citizenship in 1927, the elected Governors Act (1960), an elected representative (non-voting) to the U.S. Congress (1968), and several attempts in the 1970s to gain voter approval for a Local Constitution (see Boyer, 1984). The 1996/97 discussions on a new Constitutional Convention reflected perennial concerns about the second class citizenship conferred by the present status of "Unincorporated Territory." As an unincorporated territory, the Virgin Islanders do not have voting representation in the U.S. Congress and cannot (those who are U.S. citizens) vote for the President of the U.S. In addition, certain non-fundamental provisions of the U.S. Constitution do not extend to the USVI.
What is of central interest to this paper is the way in which the various groups are incorporated into the polity. The significant but declining population of Eastern Caribbean immigrants who have not sought U.S. citizenship are naturally excluded from the political process (there were 17,464 non-citizens of a total population of 101,809 enumerated in 1990). Currently, almost half of Eastern Caribbean immigrants are naturalized U.S. citizens, and they have demonstrated their potential political clout by supporting members of their community in their bids for election to the unicameral legislature. One of the most popular vote getters in the 1986 and 1988 senatorial elections was a person who was born in the British Virgin Islands but is often now considered a Virgin Islander.
In the 1990s, Eastern Caribbean immigrants have been exercising greater voting clout and in 1996 elected Dominican-born David Jones and St. Kitts born Roosevelt David to the 22nd Legislature. In the 1998 elections, five out of 15 senators elected were born in the Eastern Caribbean. These results underscore the importance of an umbrella political organization (The Association of Caribbean Organizations) which mobilized the Eastern Caribbean vote. It also reflects the rejection of nativistic attempts to legitimize a very restrictive definition of "Virgin Islander."
Whites, despite their minority status, were able to disproportionately influence the outcome of senatorial elections in the 1980s (Leary and de Albuquerque, 1989). The 17th Legislature (1987-1988) was headed by a white native born Virgin Islander of Danish ancestry, and included another native born white, a French Virgin Islander, and two other whites born on the U.S. mainland. This disproportionate representation (whites constituted 14 percent of the population, yet a third of the 15 member Legislature was white) did not go unnoticed or unremarked. The influence of whites in the legislature, however, has been declining since 1994. In 1998 (23rd Legislature), only one white, a French Virgin Islander, and former President of the 22nd Legislature, was elected to the senate.
The ability of the small community of French Virgin Islanders to consistently elect one or two representatives to the legislature in the 1980s was a testament to the growing importance of white voters during the decade. In a study of the 1986 elections, Leary and de Albuquerque (1989) suggest the disproportionate influence of white voters on elections was due to their higher turnout rate and the possibility that white voters only voted for a few candidates (presumably white) rather than the full slate (7) of candidates (an analysis of the returns from white and black dominated precincts showed that the average voter in white dominated precincts voted for far fewer candidates than the average voter in black dominated precincts). This bullet voting (maximizing your vote by voting for one candidate rather than a full slate of seven candidates for St. Thomas and St. Croix, respectively) has long been associated with
French Virgin Islanders, and it appeared to be a practice of the entire white community. In the 1990s, there were proportionately fewer white voters because of white flight in response to high levels of crime and two devastating hurricanes.
The large community of Puerto Ricans on St. Croix were for a long time unable to realize their potential voting strength, but since the advent of populist Hispanic politicians (Hector Citron, Liliana Bellardo), the community has become a political force to contend with. The full political incorporation of Puerto Ricans into the USVI polity was finally achieved with the election of Vieques-born Juan Luis as Governor in 1978 (he was returned to a second term in 1982).
The increasing political clout of Hispanics and Eastern Caribbean immigrants, has been disturbing to native-born Virgin Islanders, who long enjoyed supremacy in the political arena. Being native-born black still provides a political head start, but aspirants to office now have to contend with a new kind of racial/ethnic politics in the USVI. Candidates who ignore the necessity of a broad-based appeal are often reminded of the unsuccessful, though credible, bid of Adelbert Bryan for Governor in 1986.6 Since campaigns are becoming increasingly more expensive, dependence for financing on the white business community is increasing, and along with it the character of USVI politics is changing, the nativism of Senators Adelbert Bryan and Celestino White notwithstanding. Governor Schneider was elected in 1994 on a broad-based coalition of conservative native Virgin Islanders, Eastern Caribbean immigrants, continental whites, and the overwhelming support of the business community (white, East Indian, and Arab). It would appear that this same broad-based coalition, plus Hispanics, were responsible for Charles Turnbull's success in November, 1998.
We have argued in this paper that the USVI can be characterized as having a high degree of racial and ethnic pluralism. While not quite a plural society a la Furnivall (1948) or M.A. Smith (1965;1969), the USVI remains, nevertheless, a largely segmented society. Since race, ethnicity and national origin are the major lines of cleavage, they tend to structure all social relationships, creating a fragile equilibrium that is sustained by the necessity of economic intercourse but continuously threatened below the surface by mutual suspicion, antagonism, prejudice and racism.
Such an uneasy social equilibrium, thinly held together by the impersonal bonds of the marketplace, has both short-run and long-run consequences. It was in part the frayed community fabric and the absence of a widely shared history and cultural tradition (in addition to the severity of the destruction) that resulted in the outbreak of looting and violence in the wake of hurricane Hugo in September of 1989. Such a disaster, laid bare some of the long-standing cleavages festering throughout USVI society. It is this same frayed fabric that may prevent the successful resolution of the most critical issues facing the Territory in the near future. These include the issue of a future local constitution, greater local participation in the tourist economy and the future direction of this economy, the control of the business sector by minority communities (whites, East Indians, and Arabs), and the persistence of an underclass of black and Hispanic youth with little chance for social mobility.
In the first instance, the absence of any strong social consensus means that progress in writing a constitution acceptable to voters will be difficult. The major stumbling block here will be the question, "Who is a Virgin Islander?" In the second case, the conduct of tourism policy in the next decade will largely determine whether the contemporary high-density mass market tourism economy can prove viable for the long-term future. If current high-import and mass merchandising directions continue then USVI tourism is in trouble. We have argued elsewhere (de Albuquerque and McElroy, 1989) that such a tourism style is unsustainable as currently practiced in the region, since it inevitably produces negative effects (1) on the fragile ecology and amenity base of the economy, (2) on the fragile stability of an already stressed social system, and (3) ultimately on the competitive attractiveness of the destination. Unless all segments of the community can join together to develop a more socially adaptive and environmentally compatible tourism style, which seems unlikely at this stage of USVI tourism development, and unless there is a concerted effort to distribute tourism's benefits more equitably among the various groups, the long-term sustainability of USVI society is at risk. The latter addresses the third issue of greater local control and participation in the tourist industry. This will be hard to achieve given the nature of USVI tourism--a tourism plant controlled my multinational hotel chains. Likewise, the greater involvement of blacks and Hispanics in business, will be difficult to bring about, but the USVI government can target loans and other business incentives to these groups. The various youth and vocational training programs have failed to address the problems of a fairly significant number of economically disenfranchised black and Hispanic youth for whom crime has become a way of life. Unless they can be better integrated into the mainstream, crime threatens to destabilize not only the tourist-based economy but also the delicate social fabric.
1 For an extensive discussion of the concepts of pluralism and the plural society, see de Albuquerque, 1977.
2 According to Miller and Boyer (1980:Note 8), "as many as 1,700 laborers had arrived (in St. Croix) by 1864 from Barbados and St. Eustatius alone."
3 The USVI has the notoriety of having the highest percentage of persons housed in public housing projects under the American flag.
4 Census figures on poverty in the USVI are underestimated because they are based on a national poverty level. Cost of living in the USVI is 20-30% higher than on the U.S. mainland.
5 The USVI Department of Education ruling barring children of non-immigrant workers from entering the public schools was successfully challenged in Federal Court (Hosier v. Evans, 1970) resulting in the enactment of public law 91-225.
6 Bryan's campaign was couched in nativistic and nationalist terms. He was portrayed by the media and his opponents as anti-white and anti-Puerto Rican, and he drew the least amount of support from these two groups (Leary and de Albuquerque, 1989).
Boeke, Julius H., 1953. Economics and Economic Policies of Dual Societies. New York: Institute of Pacific Relations.
Boyer, William W., 1984. America's Virgin Islands: A History of Human Rights and Wrongs. Durham, North Carolina: Carolina Academic Press.
Crane, Julia G., 1971. Educated to Emigrate: The Social Organization of Saba. Assen, The Netherlands: Van Gorcum.
de Albuquerque, Klaus, 1977. "Pluralism and the Plural Society: A Re-evaluation." Western Sociological Review, 8:131-44.
__________________, 1989. "Social Implications of Various Political Status Options for the United States Virgin Islands." Report prepared for the Virgin Islands Status Commission. St. Thomas:
Bureau of Public Administration, University of the Virgin islands..
de Albuquerque, Klaus and Jerome McElroy, 1982. "West Indian Migration to the U.S. Virgin Islands." International Migration Review, 16:61-101.
___________________________________, 1985. "Race and Ethnicity in the United States Virgin Islands." Ethnic Groups, 125-153.
___________________________________, 1989. "Correlates of Pluralism in the United States Virgin Islands: Evidence from the 1980 Census." Paper presented at the International Conference on "Pluralism in the Late 20th Century," ISER, St. Augustine, Trinidad and Tobago, December 7-9.
___________________________________, 1992. "Caribbean Small Island Tourism Styles and Sustainable Strategies." Environmental Management. 16(5): 619-32.
Despres, Leo, 1968. "Anthropological Theory, Cultural Pluralism and the Study of Complex Societies." Current Anthropology, 9:3-26.
Furnivall, John S., 1948. Colonial Policy and Practice: A Comparative Study of Burma and Netherlands India. London: Cambridge University Press.
Krigger, M., 1988. "Race and Ethnic Relations in the U.S. Virgin Islands, 1917-1987," in Paul Leary (ed.), Taking Bearings: The United States Virgin Islands, 1912-1987. St. Thomas: Bureau of Public Administration. University of the Virgin Islands, pp. 86-7.
_________, 1994. "The Implications of the 1993 Status Plebiscite in the United States Virgin Islands." Paper presented to the Second National Conference, "Looking to the Future: Relations between
the United States and American Samoa, Guam, the Northern Marianas, Puerto Rico and the
United States Virgin Islands," Washington, D.C., May 24-27.
Kuper, Leo, 1969a. "Plural Societies: Perspectives and Problems," in Leo Kuper and M. G. Smith (eds.), Pluralism in Africa. Los Angeles: University of California Press, pp. 7-26.
________ , 1969b. "Ethnic and Racial Pluralism: Some Aspects of Polarization and Depluralization," in Leo Kuper and M. G. Smith (eds.), Pluralism in Africa. Los Angeles: University of California Press, pp. 459-87.
Leary, Paul M. and Klaus de Albuquerque, 1989. "The Other Side of Paradise: Race and Class in the 1986 Virgin Islands Election." Caribbean Affairs, 2:51-63.
McElroy, Jerome L., 1978. Economic and Social Impacts of the Virgin Islands Coastal Zone Management Program. St. Thomas: Virgin Islands Planning Office.
McElroy, Jerome L. and K. de Albuquerque, 1989. "Tourism Styles and Policy Responses in the Open Economy/Closed Environment Context." Paper presented to the CCA Conference on Economy and the Environment, Barbados, November.
Miller, Mark and William Boyer, 1980. "Foreign Workers in the U.S. Virgin Islands: Lessons for the United States." Paper delivered at the 1980 American Political Science Association Meeting, Washington, D.C., August 28-31.
Senior, Clarence, 1947. "The Puerto Rican Migrant to St. Croix," San Juan, Puerto Rico: Social Science Research Center, University of Puerto Rico, mimeo.
SERD, 1968. Aliens in the U.S. Virgin Islands: Temporary Workers in a Permanent Economy. Silver Springs, MD: SERD, Inc.
Smith, M. G., 1965. The Plural Society in the British West Indies. Los Angeles: University of California Press.
__________, 1969. "Institutional and Political Conditions of Pluralism.," in Leo Kuper and M. G. Smith (eds.), Pluralism in Africa. Los Angeles: University of California Press, pp. 27-65.
U.S. Bureau of the Census, 1918. 1917 Census of Population of the Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing Office.
______________________, 1953. U.S. Census of Population: 1950. Volume II. Characteristics of the Population. Part 51-54, Territories and Possessions. Washington, D.C.: U.S. Government Printing Office.
______________________, 1963. U.S. Census of Population: 1960. Volume I. Characteristics of the Population. Part 55. Virgin Islands. Washington, D.C. : U.S. Government Printing Office.
______________________, 1973. U.S. Census of Population: 1970. PC(1)-B. General Population Characteristics, Virgin Islands. Washington, D.C.: U.S. Government Printing Office.
______________________, 1984a. U.S. Census of Population: 1980. P-B. General Population Characteristics, Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing Office.
______________________, 1984b. U.S. Census of Population: 1980, P-C. General Social and Economic Characteristics, Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing Office.
______________________, 1985. U.S. Census of Population: 1980, P-D. General Population Characteristics, Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing Office.
______________________, 1991. U.S. Census of the Population: 11990, CPH-1-55. Summary
Population and Housing Characteristics, Virgin Islands of the United States. Washinton, D.C.:
U.S. Government Printing Office.
______________________, 1993a. U.S. Census of Population: 1990, C-1-55. General Population Characteristics, Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing Office.
______________________, 1993b. U.S. Census of the Population: 1990, C-2-55. Social and Economic Characteristics, Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing
______________________, 1993c. U.S. Census of Population and Housing: 1990, CPA-3-55. Population and Housing Characteristics for Census Track and Block Numbering Areas, Virgin Islands of the United States. Washington, D.C.: U.S. Government Printing Office.
U.S. Virgin Islands Department of Labor, 1969. Comprehensive Manpower Plan, Fiscal Year 1970. St. Thomas, U.S. Virgin Islands: Virgin Islands Government Printing Office.