Are there limits to the U.S. model?

Exporting U.S. concepts of race: are there limits to the U.S. model?

Virginia R. Dominguez

In the fall of 1993, I found a gem at the Bishop Museum Archives in Honolulu. There, in the middle of the Kalanianole Collection, was a priceless letter to Hawaiian King Kalakaua, dated January 26, 1885, from Dr. E. Arning, an English-speaking Swedish physician brought in by the king’s government to help evaluate and improve public health. “Sire,” Arning wrote, “in compliance with your Majesty’s request on behalf of the Board of Geneology as to ranging [ranking?] the Polynesian, more especially the Hawaiian race under the headings of Huxley’s classification of races based on hair, colour, and form of cranium I beg to submit the following.” In four handwritten pages, Arning went on to assess what “is generally accepted,” presumably among Euro-American scientists; what additional measurement of skulls he undertook himself in order to provide the King with a more informed answer; and what his conclusions were, based on cranium, skin, and hair (Kalanianole Collection Letters).

According to Arning, it was generally accepted that “the Polynesians are either mesocephalic or brachycephalic in opposition to the dolichocephalic Papuan race inhabiting the New Hebrides, the Viji group, and the New Britannia Archipelago, and other islands of the South Pacific” but also that “it has been generally noted, even by the first observers, at a time when the race was purer and not under the influence of the recent influx of foreign races, that a great variety exists among the Hawaiians… regarding the colour of the skin and the colour and form of the hair.” Though he included a copy of Huxley’s classification as an addendum to his letter, Arning concluded “that it is not feasible to appoint an exact place for the Hawaiians in Huxley’s classification.”

A century later, this language looks out of date, and the obsession with cranial measurements even seems quaint (see Marks, 1994; “Group Urges Cessation of Race-Based Statistics,” 1997). But what makes this such a find is neither the taxonomic obsession of late-nineteenth-century European scientists (for example, Wells, 1883; Winchell, 1881) nor evidence that they had a hard time placing the indigenous people of Hawai’i in their classification schemes (see Davis, 1979) but, rather, the fact that King Kalakaua himself had sought the expertise of a visiting European scientist regarding the racial classification of Hawaiians.

I have thought a great deal about this letter and different ways to interpret it, and have come to the conclusion that it is an example of a kind of data we must seek and utilize more often–that is, data about how U.S. concepts of race travel and what one can see when they are made to travel outside a very specific geohistorical world that one can almost never see when they operate as doxa.

There is in Arning’s letter to Kalakaua much more than readily meets the eye–and much that highlights just how much is taken for granted in late-twentieth-century talk of race in the United States (for example, Office of Management and Budget, 1994, 1995, and 1997; Spencer, 1997). Without Kalakaua’s own letter (or written record of a verbal request to Arning), I cannot, of course, be sure of his intentions, rationale, or understanding. But exploring the options, in light of what we do know about public discourse in the Kingdom of Hawai’i and the United States at the time, is still suggestive.

One possible interpretation of Kalakaua’s request is that he understood the Anglo-American concept of race but could not find Hawaiians on any of the competing scientific or popular taxonomies that circulated at the time–and sought clarification. A second plausible interpretation, however, is that he was aware that Europeans and Euro-Americans classified people by something called race, but that it was not a classification he used himself and, hence, needed “an informant” to explain it. A third interpretation might be that competing discourses on where Hawaiians fit circulated among European and Euro-American residents in the kingdom–and in the print media in the United Kingdom and United States–and that Kalakaua had simply sought an authoritative answer that might settle the question. I think it is even worthwhile to contemplate a fourth option: that Kalakaua understood enough about the privileging of whiteness and denigrating of blackness in late-nineteenth-century Anglo-American discourse on race to seek scientific backing for the claim that Hawaiians were closer to whites than to blacks.

One thing is clear. King Kalakaua’s query was of a totally different order than the superficially related queries about race that circulated in the United States at the time and those that circulate in the United States even now in the late-twentieth century for, unlike the United States in either the late-nineteenth or the late-twentieth centuries, Kalakaua’s Hawai’i did not classify people in terms of “race.” Kalakaua was curious–and I believe unclear–about the entire system of classification in much the same way a traditional cultural anthropologist might be with regard to a culturally foreign taxonomy (of plants or diseases, for example)–with one exception. Kalakaua’s curiosity was neither purely intellectual nor purely scientific. By the 1880s, Hawaiian exposure to U.S. racializing practices had grown considerably and extended to Kalakaua himself.

So through Arning’s letter to Kalakaua we have a chance (1) to focus on a historical moment in which the soon-to-be-subordinated population is aware of “race” but not yet subjected by it, and (2) to contrast life with and without “race” before the thinkability of race sets in. Even in all the recent, welcomed publishing flurry on the social construction of whiteness and blackness and the sociohistorical shaping of racial categories (for example, Davis, 1991; Frankenberg, 1993 and 1997; Omi and Winant, 1986; Spencer, 1997; Stoler, 1997), there are usually at best only hints of the possible–but very real–unthinkability of “race.”

On the Relative Foreignness of “Race”

Consider the Hawaiian case. I posed the possibility that Kalakaua understood the Anglo-American notion of race but could not figure out where to put Hawaiians, but I believe there is little evidence to support that interpretation. The first census of the population of the Hawaiian Islands to use racial terminology and to demand classification by race came after Kalakaua–in fact, after the demise of the independent Kingdom of Hawai’i in the 1890s and immediately upon annexation by the U.S. government under President McKinley in 1898 (see Schmitt, 1977).

The Kingdom of Hawai’i was clearly not an egalitarian state or society before annexation, but its terms of distinction and modes of differentiation had very little overlap with Anglo-American racial ones (see Ii, 1963; Kamakau, 1976, 1991, and 1992; Kame’eleihiwa, 1992; Sahlins, 1992). No institutional practices promoted social, reproductive, or civic exclusivity on anything resembling racial terms before the American period. There was no known ban on marriages, except for prohibition of a marriage if the foreigner was shown to have a wife in his place of origin. It was easy for foreigners to become citizens. Unlike the extensive differentiating and disempowering laws put in place throughout the nineteenth century in numerous parts of the U.S. mainland, no parallel laws–customary or legislated–seem to have existed in the Kingdom of Hawai’i.

There is little evidence of “race-like” notions in Hawaiian society prior to the arrival of Captain Cook at the tail end of the eighteenth century. And though there is evidence that Hawaiians became aware of Anglo-American uses of the term race and of its connection to slavery in the U.S. South by the 1850s and 1860s (as evident in the Hawaiian-language newspapers of the period), all the other evidence suggests that the idea of race failed to take hold institutionally or conceptually in nineteenth-century Hawaiian society. We do know that there was intense social ranking among Hawaiians throughout this period and that they used bilineal genealogical reckoning to determine everyone’s social rank (see Kamakau, 1991). But these alone are not evidence of “racial thinking.” Body talk appears to be limited. References to “Polynesians with reddish hair” and “red-eyed outcasts” do show an invocation of phenotypic traits as named markers of two of the many named categories. And yet, with the exception of reference to the “height at which each ali’i stood,” there is no other reference to the body, no systematic invocation of phenotypic features or even ancestors’ phenotypes, and no reference whatsoever to skin color in any of the key documents written by nineteenth-century Hawaiian historians (Ii, 1963; Kamakau 1976, 1991, and 1992; Malo, 1951; Thrum, 1917).

References to skin color do appear in the Hawaiian language newspapers as early as the 1850s–as does the specific term haole–but it is revealing to note how they do so. First of all, while Hawaiian language newspapers date back to 1834, all but two of them–up to 1861–were missionary publications and all newspapers, bulletins, and newsletters (till 1861) had Anglo-named editors. The vast majority of the writing in the first three decades of newspaper life in Hawaiian was done by Anglo-American missionaries and their Anglo-American affiliates (see Mookini, 1974). Secondly, a great deal of the invocation of skin color or the use of the term haole appears reactive, that is, commenting on the relevance of skin color in the United States, its connection to slavery elsewhere, haole references to Hawaiians’ “brown skin,” and so on. As Eleanor Nordyke writes,

White persons became known as haole, from the Hawaiian

words ha or “breath” and ‘ole “without,” a term originally

used by Hawaiians for persons who could not speak the

Hawaiian language and did not understand the native culture.

Haole did not indicate skin color in its early usage–the

term was applied in reference to a stranger. In the early

nineteenth century the Hawaiian words ha’ole ‘ele ‘ele

were used in speaking about Black foreigners. Haole was

applied to foreign residents from America, the British

Empire, France, and Russia (1989, p. 43).

It is interesting to note that Kalakaua’s Hawai’i chose to adopt a different European taxonomy by the mid- to late-nineteenth century–that of nationhood and nationality–but not that of “race.” Nation and nationality became increasingly invoked in governmental practices and public discussions towards the end of the nineteenth century. The Hawaiian censuses of 1872 and 1884 invoked and classified people by age, sex, and “nationality,” and listed under nationality were: Natives, Half-castes, Hawaiian-born foreigners, Americans, British, Germans, French, Portuguese, Norwegians, Chinese, Japanese, Polynesians, and Other Nationalities. The point implies an openness to, rather than a blanket rejection of, foreign taxonomies–and fits in with what we know about other features of public life and monarchical demeanor at the time. As the portraits hanging on the walls of the Iolani Palace in Honolulu, the palace itself, and their interests in foreign relations all imply, mid- to late-nineteenth century Hawaiian monarchs were arguably Europhilic. So the nonadoption of a racial taxonomy may not have been accidental. What we do know for sure is that “race” and “color” remained elusive as principles of classification and modes of reference at least into the 1880s.

Everything changed the minute the United States annexed Hawai’i in 1898. The very next census–of 1900—blatantly classified the population by “color.” The Table of General Nativity counted all those “native born” and those “foreign born.” It also divided the population into: Total White, Native White, Native White of native parentage, Native White of foreign parentage, Foreign White, Chinese, Japanese, Negro. The table titled “Native Born Population of Hawaii” (1900) divided the population into categories and named them as follows: Hawaiian (29,787), Part Hawaiian (7,843), Caucasians (7,283), Portuguese (9,163), Negroes (178), South Sea Islanders (60), Japanese (4881), and Chinese (4021) (see 1904 or see 1905 Thrum’s Hawaiian Annual, p. 19). While the invocation of “nationalities” did not immediately disappear, it is the invocation–and classification–in explicit racial terms that then prevailed at least into the 1960s.

The Apparent Necessity of “Race”

The change is striking and immediate, and revealing of both Hawaiian life without “race” and U.S. life with “race.” Whatever it was that puzzled Kalakaua was obviously no obstacle for the U.S. federal bureaucracy. The difficulties Kalakaua might have had with the taxonomy of race were at some level shared by the U.S. federal bureaucracy, but instead of rejecting the taxonomy in favor of something else–like the Hawaiian monarchy appears to have done–they kept finding a way to make the racial taxonomy work.

This was true of U.S. census efforts nationwide just as much as U.S. census efforts in previously “raceless” Hawai’i. Their insistence on a racial taxonomy is relentless (see also Anderson, 1988; Dominguez, 1986; Lee, 1993). What may be more surprising is the evidence of constant struggles over the categories themselves and the U.S. census office’s record of categorical flip-flopping over at least 150 years. All U.S. censuses from the beginning in 1790 have asked about race or color, and nearly all have debated what to do with nonwhites (including the forthcoming census of 2000) (see Iowa City Press-Citizen, October 1, 1997; U.S. Department of Commerce, Bureau of the Census, 1990; Wright, 1900).

It is interesting to contrast the U.S. mainland’s taxonomic practices with Hawai’i’s in the years just prior to and during Kalakaua’s reign. From at least 1850 on, U.S. census officials were explicit in their instructions to marshals but, looked at in succession, they come across as obsessed and puzzled at the same time.

For example, instructions to census marshals and assistant marshals in 1850 included the required question about “color.” Under heading 6, entitled “Color,” the instructions stated that “in all cases where the person is white, [the marshal is to] leave the space blank; in all cases where the person is black, [he is to] insert the letter B; if mulatto, [he is to] insert M.” “It is very desirable,” it went on, “that these particulars be carefully regarded” (Wright, 1900, p. 152). The color concern extended to slaves as well. Schedule 2 for “Slave Inhabitants” asked for the name of the slaveholder and any previous slaveholder, the slave’s age, sex, physically defective status (as in the case of free persons), and color. Here, too, instructions were to be very specific: “Under heading 5, entitled `Color,’ insert, in all cases, when the slave is black, the letter B; when he or she is mulatto, insert M. The color of all slaves should be noted” (Wright, 1900, p. 153). Despite a tendency toward binarism that I have noted before in White By Definition (1986), this was a ternary moment in the U.S. racial taxonomy.

Instructions to assistant marshals participating in the 1870 census-the first after the abolition of slavery and the end of the U.S. Civil War–responded to key social and legal changes but maintained the key obsession with “race” or “color.” “Columns 4, 5, and 6,” it specified, “must, in every case, be filled with age, sex, or color of the person enumerated. No return will be accepted when these spaces are left blank” (Wright, 1900, p. 157). And on the same page, a modified message about “color” appeared:

It must not be assumed that, where nothing is written in this

column, “White” is to be understood. The column is always

to be filled. Be particularly careful in reporting the class

Mulatto. The word is here generic, and includes quadroons,

octoroons, and all persons having any perceptible trace of

African blood. Important scientific results depend upon

the correct determination of this class in schedules 1 and 2

(Wright, 1900, p. 157).

Once again we note intense concern with the specification of “color,” but note that this time “Schedule 1–Inhabitants” provided enumerators with five “Color” categories to choose from: “White, W; Black, B; Mulatto, Mu; Chinese, C; Indian, I.”

If Kalakaua had been exposed to any of these forms, he might indeed have wondered where Hawaiians fit in. Hawaiians were of little concern on the U.S. mainland, even though the introduction of a fivefold classification on the official U.S. census suggests early efforts to “rein in” non-European and non-African people. The racial taxonomy continued its primary emphasis on the European and African diasporas and on granting categorial status to their mixed offspring. Instructions to these assistant marshals interestingly appear to have recognized more than just three categories of mixed European and African offspring–at least in public and casual usage–but hint at the need to deny them official federal standing. By this point, it is obvious that the taxonomy was under strain and an effort was being made to harness its social unruliness. But the result was as before: more detailed instructions and better discipline, not abandonment.

The 1880 census continues the theme of continuity and change. Instructions to enumerators in 1880 repeated the explicit requirement from the 1870 census to include information on the age, sex, and color of every individual counted. They also repeated the 1870 description of “Color” and color categories practically verbatim (Wright, 1900, p. 171). A direct message to enumerators about the main goals of the census was included under “duties of enumerators.” “It is the prime object of the enumeration,” they were told, “to obtain the name, and the requisite particulars as to personal description, of every person in the United States, of whatever age, sex, color, race, or condition, with this single exception, viz: that `Indians not taxed’ shall be omitted from the enumeration” (Wright, 1900, p. 168).

Schedule I for “the Indian Division” continued the questions about marital status, occupation, age, and sex, but under “personal description” made four of the eight items questions about degree of “Indianness.” Two of these were social indicators–years of wearing “citizen’s dress” and years, if any, on reservations. But two explicitly used racial language and imply familial relations:

Column 10: If this person is of full-blood of this tribe, enter

“/.” For mixture with another tribe, enter name of latter.

For mixture with white, enter “W;” with black, “B;” with

mulatto, “Mu.”

Column 11: If this is a white person adopted into the

tribe, enter “W. A.;” if a negro or mulatto, enter “B. A.” (U.S.

Department of Commerce, Bureau of the Census, 1990).

By 1890, immediately before the United States took over Hawai’i, Cuba, Puerto Rico, the Philippines, Guam, and Samoa, instructions to census takers followed a pattern already set–of compulsory racialization and simultaneously of a struggle to figure out what to do with nonwhites. Color, sex, and age–columns 4, 5, and 6–appeared on the main census form immediately after name, veteran status, and relationship to head of family. As usual, no elaboration was offered regarding “sex,” but specificity was asked for with regard to age, even though date of birth was not included. Under column 4 (“color”), eight categories were given–up from five in the previous census.

Census takers were instructed to “Write white, black, mulatto, quadroon, octoroon, Chinese, Japanese, or Indian, according to the color or race of the person enumerated” (Wright, 1900, p. 187). Words of caution followed:

Be particularly careful to distinguish between blacks, mulattoes,

quadroons, and octoroons. The word “black” should

be used to describe those persons who have three-fourths or

more black blood; “mulatto,” those persons who have from

three-eighths to five-eighths black blood; “quadroon,” those

persons who have one-fourth black blood; and “octoroon,”

those persons who have one-eighth or any trace of black

blood (Wright, 1900, p. 187).

The federal census had apparently caved in to a social and legal discourse that wanted a certain kind of genealogical specificity to a greater degree than at any other point in U.S. history.

No prior U.S. federal census had coded quadroons and octoroons separately, and none would in the succeeding century, but it is important to see this as part of the context within which Kalakaua sought an answer from Arning about the racial classification of Hawaiians. What, one might ask, is this race thing that keeps changing yet remains so compulsory? Multiple racial categories appear to have been recognized officially in 1890–indeed insisted upon–by the U.S. federal government. Yet even here there was no obvious place for Hawaiians. With increasing economic and political pressure threatening the survival of the independent Kingdom of Hawai’i, Kalakaua was probably not the only Hawaiian to wonder where Hawaiians would fit in if taken over socially and/or politically by the United States.

With no designation of a separate Hawaiian “race” on these official forms, I wonder if there was not talk (that I have not been privy to) of the possibility that Hawaiians might be classified as a race of some kind of mulattoes, quadroons, or octoroons–or some mixed offspring of Chinese and black. There were scientific debates about the racial origins of the Polynesian and, more specifically, Hawaiian people that European and European–diaspora scientists participated in (see Davis, 1979; Smith, 1918). In only one case that I have been able to document did the scientist–Abraham Fornander, a Swedish-born resident of Hawai’i married to a member of the Hawaiian ruling class–argue that the Hawaiians were members of the Aryan race (see Fornander, 1909, and 1916-1920). All other taxonomic conclusions made Hawaiians “colored” within the U.S. racial taxonomy. And since “colored” at the turn of the twentieth century definitely meant inferior, even if “free,” it is hard to imagine the English-speaking, well-travelled Kalakaua not aware of the assigned inferiority of “colored” people in the United States. It is, of course, harder to know whether he secretly hoped that Arning would find a way to declare Hawaiians white (or Aryan).

At the turn of the twentieth century, the U.S. federal bureaucracy remained highly “color” conscious, and not just in the segregated South. By 1900, two years after the annexation of Hawai’i and the U.S. defeat of Spain in 1898, the census no longer requires great genealogical specificity when it codes for “color,” but it continues to ask about “color or race” right after it asks for name and relationship to the “head of family.” Whereas the 1890 census coded eight color categories, the 1900 census provides only five options: W for white; B for “black (negro or of negro descent)”; Ch for Chinese; Jp for Japanese, and In for Indian “as the case may be” (see U.S. Department of Commerce, Bureau of the Census, 1990).

Note that uncertainty, flip-flopping, and insistence all seem to have coexisted, but that “Color or race” remained a must, no matter how unsure they were about how many races (other than white) there were and which distinctions needed to be made or altered. The 1910 instructions to enumerators reintroduced a distinction between black and mulatto, and for the first time offered “Other” as an option. Six labeled categories were given, besides “Other”: W for white, B for black, Mu for mulatto, Ch for Chinese; Jp for Japanese, In for Indian. “For census purposes,” it said explicitly, “the term `black’ (B) includes all persons who are evidently full-blooded negroes, while the term `mulatto’ (Mu) includes all other persons having some proportion or perceptible trace of negro blood” (U.S. Department of Commerce, Bureau of the Census, 1990).

Not surprisingly, the population schedule developed by the Census Bureau to count Indians now required specification of the “proportions of Indian and other blood,” immediately after asking for tribal affiliation and parents’ tribal affiliation. Three options were given: Indian, White, Negro. Enumerators were told to count fractions, unless “the Indian is a full-blood.” “For Indians of mixed blood all three columns should be filled, and the sum, in each case, should equal 1, as 1/2, 0, 1/2; 3/4, 1/4, 0; 3/4, 1/8, 1/8; etc.” (U.S. Department of Commerce, Bureau of the Census, 1900). Lack of trust or genealogical suspicion was also evident. “Wherever possible,” enumerators were told, “the statement that an Indian is of full blood should be verified by inquiry of the older men of the tribe, as an Indian is sometimes of mixed blood without knowing it” (ibid.).

From the vantage point of the Bureau of the Census and other administrative offices in both Washington and Honolulu, Hawai’i was then just one more place to adapt, recategorize, and incorporate within the racial taxonomy. Over the previous decade the official census taxonomies had shown their ability to expand and adjust to changing public, political, social, legal, and scientific positions. The fact that Hawaiians did not employ a racial taxonomy themselves seems to have mattered little.

For the first years after annexation, government officials resorted to a kind of terminological duality and simultaneity that reflected the new apparent “need” for racial thinking but also the ill fit between the white-black mainland obsession and the much more acknowledged pluralism of society in Hawai’i even after the U.S. takeover. “Census Returns for the Hawaiian Islands, 1910” (Thrum’s Hawaiian Annual, 1911, p. 19) compared the numbers of people of different races (its official term) showing up in the 1900 and 1910 censuses. Listed this time under “Race” were: “Hawaiians, Part-Hawaiians, Portuguese, Spanish, Porto Ricans, Other Caucasians, Chinese, Japanese, and All Others.” The category labeled “Hawaiian” showed a decrease of 3,688 while that labeled “Part-Hawaiian” showed an increase of 4,637. The “Chinese” population was also in decline (down 4,064), while the “Japanese” population posted a huge increase and appeared to be 40 percent of the total population of the Hawaiian Islands. Those distinctions were maintained and indeed focused on. Portuguese, Spanish, and Porto Ricans appear to have been treated as types of Caucasians along with “Other Caucasians,” but the noted social relevance was as Portuguese, Spanish, and Porto Rican.

“A Report of the Superintendent of Public Instruction,” dated 1920 (see Thrum’s Hawaiian Annual, 1921, p. 18), included a table labeled “Nationality of Pupils,” but at the head of each column of named and counted “groups” it read “Races” once again. This time the named “races” were: Hawaiian (3,293); Part-Hawaiian (4,100); Anglo-Saxon (1,033); Scandinavian (34); Portuguese (5,304); Filipinos (941); Spanish (379); Chinese (3,721); Japanese (17,541); Porto Rican (1,068); Korean (508); Other Foreigners (373). Note the absence of “Negro,” “Black,” or even “colored” in these lists. Nationwide, Hawai’i was being romanticized through postcards, music, dance, and a burgeoning tourist industry intent on selling Hawai’i as both foreign and safe for (white) Americans, that is, as a place not constituted by racial tension, without a history of slavery and, hence, without negroes. If there were no Jim Crow laws, it was presumably because they weren’t needed.

This was, after all, a deeply segregationist period of life in at least a third of the United States at the time, a period of massive waves of working class European immigration (some more welcome than others), and a period of Orientalist backlash against the importation since the mid-1880s of large numbers of Chinese-speaking and Japanese-speaking workers. The 1920 census reflected a great deal of this. It listed sex, color or race, age at last birthday, and single, married, widowed, or divorced [status] as the four items sought under the category of personal description. Home ownership, rental, or mortgage status moved ahead of personal description, and immigration history, naturalization status, place of birth, and mother tongue followed immediately. Note the date: over two decades since the opening of Ellis Island, with millions of southern and eastern European immigrants pouring into the United States, not to mention inflow across the Pacific and the Rio Grande, and on the eve of the first official, across-the-board restrictions on immigration into the United States. Immigration and naturalization loomed larger than ever, but still not larger than “color or race.”

By 1930, the federal census had expanded its section on “Home data,” and “personal description” now consisted of sex, color or race, age at last birthday, marital condition, and age at first marriage. More significantly, there was an expansion in the number of racial types to be counted–now nine–and specific, detailed instructions for reporting race were given to enumerators, beyond any developed prior to 1930. Unlike nineteenth-century censuses that had insisted on a distinction between black (or negro) and mulatto (or even quadroon and octoroon), the 1930 census was essentially adding names of other populations it was now fully racializing as distinct and identifiable.

Paralleling the proliferation of “racial types” was, in fact, a conceptual consolidation of both whiteness and blackness, neither of which have been allowed to be officially splintered since the 1910 census. Black and mulatto disappeared into the presumably homogeneous category of Negro. White remained unsplintered, despite the huge waves of European immigration that characterized the first three decades of the twentieth century, and the derogatory verbal and visual representations that accompanied many of these flows.

Mexican, Filipino, Hindu, and Korean appeared for the first time on the Federal census. The census coded W for White, Neg for Negro, Mex for Mexican, In for Indian, Ch for Chinese, Jp for Japanese, Fil for Filipino, Hin for Hindu, and Kor for Korean. For “other races,” it said on the actual schedule, “spell out in full.” According to the 1990 historical compilation of census forms, procedures, and instructions on which I have drawn extensively,

a person of mixed White and Negro blood was to be

returned as Negro, no matter how small the percentage of

Negro blood; someone part Indian and part Negro also was

to be listed as Negro unless the Indian blood predominated

and the person was generally accepted as an Indian in the

community. A person of mixed White and Indian blood was

to be returned as an Indian, except where the percentage

of Indian blood was very small or where he or she was

regarded as White in the community (U.S. Department of

Commerce, Bureau of the Census, 1990, p. 60).

In an effort to count Mexicans, enumerators were instructed to code as Mexican “all persons born in Mexico, or having parents born in Mexico, who were not definitely White, Negro, Indian, Chinese, or Japanese” (ibid.).

The bottom line was also clearly stated. The presumed purity and sameness of “the white race” was to be preserved statistically, even, or perhaps especially, when there was ample evidence of sexual and familial mixing in various regions, states, and territories. The most extreme obsessive mechanisms for presumably preserving “the purity of the white race” in Louisiana actually date from this period (see Dominguez, 1986). Historically, it is the same period of time during which increasingly vocal and insistent clamor for Aryan purity became official as Hitler and his Nazi party took over Germany. The historical simultaneity is both noteworthy and underacknowledged in the context of the United States. “Any mixture of White and some other race,” in the 1930 U.S. census, “was to be reported according to the race of the parent who was not White; mixtures of colored races were to be listed according to the father’s race, except Negro-Indian (discussed above)” (ibid.).

Limits to the U.S. Model?

By the 1920s, not surprisingly, there was much more “race talk” in Hawai’i, but at least some of it seems to have toyed with the U.S. racial taxonomy. A terrific example of this is the rise of a public and academic obsession with racial mixing. Racial mixing becomes something to be noted, counted, monitored, theorized, and, in certain liberal quarters, praised. In an era of great institutionalized segregation in at least a third of the United States and a deepening of obsessions with racial purity–in the United States as well as in Nazi Germany–the racialization inherent in the monitoring of intermarriage in Hawai’i came both from self-labeling liberal quarters trying to combat “racism” and from far more conservative mainland-oriented quarters concerned with maintaining racial lines for a host of reasons. Since Jim Crow laws were never adopted in Hawai’i, even during this period, Hawai’i’s relative non-Americanness was a matter of great appeal and simultaneously of great concern.

Amid these processes of production and consumption of “Hawaii,” the territorial government of Hawai’i collected statistics on intermarriage. Invariably these were always and only conceptualized as marriage in which the bride and the bridegroom were of different races. The practice became so ingrained that it continued well into the statehood era, that is, after 1959. From 1947 through 1968, Thrum’s Hawaiian Annual/All About Hawaii included a special table on intermarriage (across racial lines). A near obsession with mixtures was increasingly evident in numerous government offices and agencies, most explicitly with reference to the difficulties it posed (and still poses) to all attempts to count the population of Hawai’i. Both in 1967 and 1968, All About Hawaii published side by side statistical tables on the “Population of Hawaii by Race: 1853 to 1960” (attributed to the Department of Planning and Economic Development) and the “Ethnic Stock of the Population of Oahu: April 1, 1964-March 31, 1966.”

It is actually in the seemingly deracialized–and seemingly ethnicized–table that the intermarriage obsession is most evident. The actual categories listed were: “All Races,” “Unmixed,” “Other Caucasian,” “Mixed, Two Strains,” “Mixed, More Than Two Strains,” and “Race Not Reported.” Similarly, the 1970 U.S. census was heavily used by All About Hawaii in its ninety-first edition (from 1974). In its section entitled “Characteristics of the Population” (part of an eighteen-page segment called “The People of Hawaii”), the population again was classified and counted by age, sex, and race. A note, however, said that “persons of mixed race [are] included with unmixed groups” [though] “for separate data on persons of mixed and unmixed race” the reader is referred to Table 9 (All About Hawaii, 1974, pp. 81-98).

Eleanor Nordyke’s discussion of “ethnicity” in Hawai’i details the coding specifications articulated in 1980 by the State of Hawai’i’s Department of Health, adapting 1978 criteria developed in cooperation with the National Center for Health Statistics (Nordyke, 1989, pp. 106-7). How to handle Hawai’i’s “mixtures” was the operational question. Accordingly,

a. If a mixture is of Hawaiian and any other race, code as Part Hawaiian.

b. If a mixture is of White races (White, Caucasian, Puerto Rican, Portuguese, Cuban, or Mexican), code to the first race listed.

c. If a mixture is of White (White, Caucasian, Puerto Rican, Portuguese, Cuban, or Mexican) and any other race, code to the non-White race.

d. If a mixture is of non-White races (except Hawaiian and Part Hawaiian), code to the first race listed.

In addition, Nordyke added,

The State Health Department determines a child’s race from the parents’ ethnic group following [the] 1978 procedures:

a. If both parents are of the same race, child’s race is parents’ race.

b. If either parent is of unknown race, child’s race is that of the parent with the known race. c. If either parent is Hawaiian or Part Hawaiian, child’s race is Part Hawaiian.

d. If either parent is Black, child’s race is Black (except Hawaiian and Part Hawaiian).

e. If parents’ races are White but not the same, (Caucasian, Puerto Rican, Portuguese, Cuban, or Mexican), the child’s race is that of the father’s race.

f. If one parent is White and the other parent is non-White, child’s race is that of the non-White parent.

g. If both parents are non-white but not the same race, child’s race is that of the father.

While Nordyke quoted these contemporary coding specifications in order to show just how “complex” the statistical problems are in the study of demography in Hawai’i (1989, p. 104), I am much more interested in them because of what they reveal about the depth, or entrenchment, of racial thinking in institutionalized procedures for maintaining categorization and counting in terms of race. Obviously a great deal of marriage and parenting in Hawai’i goes on now–as it apparently has for far longer–without much evidence of people caring about the lines of demarcation introduced by mainland Americans and institutionally transported to Hawai’i in the hundred years of Americanization (see Okamura, 1981). They make the institutional commitment to a racial taxonomy much more poignant, for they show that it goes on even in the face of personal and familial actions that constantly complicate the classification system beyond belief and arguably call the usefulness of their resulting numbers much into question.

Limits to Doxic Change

But how such social relations, alignments, and stakes are framed in authorized and public circles is something else. Sociologists, alongside demographers, participated in a continuing narrative of race as early as the mid-1920s (and possibly earlier, as suggested by Porteus, 1962). Romanzo Adams’ pioneering sociological work, The Peoples of Hawaii (1933), prepared for an international conference in Honolulu that took place in 1925, explicitly racialized the population of Hawai’i (see Lind, 1980, p. ix). His protege, Andrew Lind, developed his career around the analysis of race relations, intermarriage, and the processes “by which the many peoples of Hawaii were becoming one people” (ibid.).

In Lind’s career, we see perhaps more clearly than in any other single bit of data both the depth of the racialization that came in with the American era and the forms of resistance that accompanied it (Lind, 1955, 1980). As faculty advisor to the student-run journal Social Process in Hawaii, founded in 1935, he actively helped shape both the research projects his students undertook and the articles and research reports they published. A hopeful, antidiscriminatory vision is evident in his own selections and is echoed in that of so many of his students’ contributions. A disciple of University of Chicago sociologist Robert E. Park, Lind strove to critique mainlanders who argued that the rate of intermarriage in Hawai’i was so great that it was retarding the process by which Hawai’i could be decolonized. Lind did not disagree with the emphasis on race, though he rejected the imposition of U.S. mainland classifications and exclusions. In fact, race remained at least until the late 1970s the primary axis of Lind’s work. Drawing directly from Robert E. Park early in his career, he openly hoped that racial intermarriages would happen.

As if in cahoots with Robert E. Park, the U.S. federal censuses from midcentury on appear simultaneously liberalizing and insistently racializing, with a continued penchant for the naturalness and unfragmentability of “whites.” The 1950 census dropped the term color in favor of the sole term race, and provided six labeled categories in addition to “Other” (for white, Neg for negro, Ind for American Indian, Jap for Japanese, Ch for Chinese, and Fil for Filipino). Blood fractioning of the “Indians” counted continued in full force. Four boxes were included on the schedule: for full, one-half, one-quarter, and less “Indian blood.” The 1960, 1970, 1980, and 1990 censuses have all (1) continued the practice of making the “color or race” question one of the five, six, or seven questions asked for all persons (even since the advent of probability sampling), and (2) experimented with the terms used and the number and names of categories coded. They have all, however, also maintained a question formulated racially–in which “White” is one option–and coded as a race question.

In 1960, the question was: “Is this person–White, Negro, American Indian, Japanese, Chinese, Filipino, Hawaiian, Part Hawaiian, Aleut, Eskimo, (etc.)?” In 1970, it read “COLOR OR RACE Fill one circle: White, Negro or Black, Indian (Amer.) Print tribe, Japanese, Chinese, Filipino, Hawaiian, Korean, Other–Print race.” In 1980, the schedule once again asked “simply”: “Is this person–White, Black or Negro, Japanese, Chinese, Filipino, Korean, Vietnamese, Indian (Amer.) Print tribe, Asian Indian, Hawaiian, Guamanian, Samoan, Eskimo, Aleut, Other–specify?” And in 1990: “Fill ONE circle for the race that the person considers himself/herself to be.” Listed were: “White, Black or Negro, Indian (Amer.) [and it requested the name of the enrolled or principal tribe], Eskimo, Aleut, and Asian or Pacific Islander (API);” this last choice was subdivided into “Chinese, Filipino, Hawaiian, Korean, Vietnamese, Japanese, Asian Indian, Samoan, Guamanian, and Other API.” As in earlier censuses throughout the century, “Other” remained a possible identification.

This multiplication of official categories is often misconstrued, however–and so are four accompanying bureaucratic moves: (1) the move toward self-reporting and away from classification by phenotypic observation; (2) the move toward the separate and separable enumeration of people of Latin American ancestry in the United States; (3) the move toward far more detailed specification of people largely entering the United States across the Pacific, rather than across the Atlantic; and (4) the addition, in the long form given out to a sample of the population, of a question about the respondent’s “ancestry or ethnic origin.”

Self-reporting sounds like self-determination but is severely confined by the requirements that only one answer be given and that the answers, except is a small minority of cases, fit into the categories given. The only exception is about to happen in the forthcoming census of the year 2000, when “multiracials” will be given the option of marking more than one box (Iowa City Press-Citizen, October 30, 1997; see also OMB, 1997; Spencer, 1997). But that concession accentuates the larger point: that there has been absolutely no room in these classification procedures for the type of answer large numbers of residents of contemporary Hawai’i typically give when asked “what” they are–for example, Irish-Chinese-Hawaiian-Portuguese, or Hawaiian-Filipino-English-Puerto Rican-German?

Many people are also not aware of procedural limitations to self-determination within the bureaucracy. The 1960 census, for example–the first to use self-enumeration and the mails extensively-still instructed census workers reaching those not responding by mail (1) to consider “Southern European and Near Eastern nationalities” as “White,” (2) to classify “Asian Indians” as “Other” and write in “Hindu,” and (3) to classify “Puerto Ricans, Mexicans, or other persons of Latin descent” as “White,” “unless they were definitely Negro, Indian, or some other race,” presumably in the eye of the census worker (U.S. Department of Commerce, Bureau of the Census, 1990). Similarly, the 1970 census, reputedly designed for self-identification on the part of the respondent, allowed the enumerator “to fill in blanks by observation when this was possible” (U.S. Department of Commerce, Bureau of the Census, 1990, p. 83). Door-to-door census takers were instructed to assume that a respondent’s relatives living in the same housing unit “were also of the same race unless the census taker learned otherwise” (ibid.). They were also given a long list of possible write-in entries in the enumerator’s manual and instructed on how these were to be treated–actually, I would argue, on how they were to be translated.

Chicano, LaRaza, Mexican American, Moslem, or Brown were accordingly to be changed to “White.” Brown (Negro) was to be classified as “Negro” or “Black.” A separate question, asked of a sample of households, inquired into the “person’s origin or descent” but limited the answers to those claiming, or being identified as having, ancestors from a Spanish-speaking country: “Mexican, Puerto Rican, Cuban, Central or South American, Other Spanish,” or “No, none of these.” Even the 1980 census, which no longer allowed census workers themselves to enter race by observation, contained relevant instructions (see also Rodriguez and Guzman, 1992).

When, for example, a single response was not forthcoming from the respondent because of a claimed racial mixture, “the mother’s race” was to be reported by the official enumerator. “If this was not satisfactory,” instructions were to enter “the first racial group given” (U.S. Department of Commerce, Bureau of the Census, 1990, p. 89). Unlike instructions issued in 1970, instructions in 1980 were to allow “Brown” or “Mexicano” to be entered as “Other” (unless one of the listed categories was chosen). If a person was reportedly unable or unwilling to select a single group in the separate question about Spanish origin, “and only part two was Spanish (as in `Irish-Cuban’),” the “No, not Spanish/Hispanic” circle was to be filled.

In contrast to all of these singular and detailed specifications and restrictions in the clearly racial question, census workers were instructed to allow multiple answers to the new–and not compulsory–question, “What is this person’s ancestry?” Clearly experimental, the official census schedule added: “If uncertain about how to report ancestry, see instruction guide.” Below the line to be filled in was a presumably helpful list of answers to serve as examples: “Afro-Amer., English, French, German, Honduran, Hungarian, Irish, Italian, Jamaican, Korean, Lebanese, Mexican, Nigerian, Polish, Ukrainian, Venezuelan, etc.” That this was the closest thing to incorporating “ethnic talk” into the federal census was even clearer in the 1990 forms, which posed the question, “What is this person’s ancestry or ethnic origin?” and gave as sample answers this time: “German, Italian, Afro-Amer., Croatian, Cape Verdean, Dominican, Ecuadoran, Haitian, Cajun, French Canadian,Jamaican, Korean, Lebanese, Mexican, Nigerian, Irish, Polish, Slovak, Taiwanese, Thai, Ukrainian, etc.”

The novelty presented by this question in the past two censuses actually highlights—despite appearances–the doxic nature of race on the U.S. mainland and the unthinkability of U.S. life without “race.” The optional and experimental nature of this question of ancestry accentuates the compulsory nature of the racial questions. The former is optional and novel; the latter, compulsory and as old as the first U.S. census in 1790. If ancestry issues in general were fundamental to American nationhood–or more precisely had become fundamental in the wake of the ethnic and multicultural movements of the last twenty to twenty-five years–that question would have become compulsory for all respondents, and would have obviated the need for the ever-present race question. Likewise, the proliferation of Asian-Pacific Islander categories is not to be mistaken as a general question about country of origin or claims to ancestry. While it counters the extreme Orientalizing rhetoric of earlier decades by which huge numbers of individuals in the United States were lumped together under one or two “Asian/Oriental” categories, all continue to be juxtaposed to “White” and “Black” (or “Negro”) and neither “White” nor “Black” are categorically allowed to be fragmented.

Over six decades after Park, Lind, and Adams first articulated their hopeful but deeply racialized visions of intermarriage, I sit here wondering how much has and has not changed, and what King Kalakaua would make of it now. The hopefulness expressed by Lind and Adams–indeed the fervent, even edenic, desire to create a model racial paradise in Hawai’i–comes across in the 1990s as simultaneously utopian, naive, patronizing, and “racist.” But the paradox is that the United States may not have moved all that far, because the continuing commitment to race as doxa leads well-intentioned bureaucrats, politicians, scientists, and activists to promote change that amounts to little more than the spinning of wheels.

Kalakaua’s Question–in 1998

The Kalakaua-Arning exchange could be read quite narrowly with late-twentieth-century U.S. eyes and that, I am arguing, would be a major mistake. Such a reading would no doubt entail treating it as evidence that a small subordinate population wants cultural recognition by insisting on clear taxonomic recognition over which it seeks some form of agency. But this narrow reading would be as problematic as I find its late-twentieth-century counterpart to be. The issue was not then and is not now that “some minority populations now want the classification system to include separate categories for them,” or that the United States might today be “a multiracial, multiethnic society, with increasing heterogeneity among the minority populations” or that “a person’s self-identification may change over time or in different circumstances, and society’s characterizations of racial and ethnic identity also change over time” (Edmonston et al., 1996, p. 2).

These quotes capture the gist of the 1996 report issued by the National Research Council’s Committee on National Statistics on “the federal standards for racial and ethnic classification” and the July 1997 report issued by the Interagency Committee for the Review of the Racial and Ethnic Standards advising the Office of Management and Budget about possible changes to OMB’s “Statistical Policy Directive No. 15,” originally issued in 1977. In both cases, the texts reveal a desire, or at least a willingness, to make concessions to “minority” populations–sometimes explicitly called minority populations and sometimes called population groups, but almost invariably categorized as nonwhite. They do not, however, reveal any internalization of the larger point that “race” itself is a taxonomic system that has neither existed throughout time nor across space, and that major implications follow from it.

The message suggested by the Kalakaua-Arning exchange is a powerful one–that it is possible and productive to track the ways in which the notion of race is (or has been) exported, the ways it is (or has been) consumed, and the ways it becomes–but need not be—doxa. In contrast, contemporary “revisionist” racial discourse (especially in its most public textual forms) in the United States largely assumes that there is a growing diversity in the United States due to recent patterns of immigration and an increase in “interracial marriages” that require an adjustment in the number of racial categories officially recognized and tabulated on U.S. censuses, household surveys, administrative forms, and medical research. The former suggests the possibility of social organization without the taxonomy of race; the latter suggests the unthinkability of governmental administration without the constant monitoring and disciplining of the taxonomy of race.

Late 1990s texts cannot apparently escape racial thinking, even when they include some rhetoric about the social or historical construction of race as a taxonomy. And the examples abound. “The principal objective of the review,” the 1997 Interagency Committee’s report still states, “is to enhance the accuracy of the demographic information collected by the Federal Government” (Office of Management and Budget, Federal Register, 1997, p. 36874) and to enable historical continuity in the data “on race and ethnicity that have provided 20 years of information for a variety of purposes” (ibid.). A similar message is interspersed throughout Spotlight on Heterogeneity, the National Research Council’s 1996 report on the federal standards for racial and ethnic classification. It says that “the desires of small groups to have separate statistical breakdowns for all government data can conflict with the need for reliable statistics” (Edmonston et al., 1996, p. 2) and that “the need for continuity in categories for historical analysis runs counter to the equally laudable goal of adapting to the changing demographic composition of the U.S. population” (p. 2).

Even in its assertion that it is “critically important.., to test current categories,” the issues are all about adjustments within the taxonomy of race, and mostly about adjustments within what it continues to see quite unquestionably as the category of nonwhites. Five topics are highlighted for special research: “alternative ways of collecting racial and ethnic data” (always conceptualized, of course, as racial and ethnic data); “field testing of the wording and order of questions in surveys” (to see what impact they have on individual patterns of racial self-identification); “self-identification labels and categories” (to see just how different “subjective” answers are from observers’ identifications); “differences between self-identification and observer identification” (to see how “reliable” racial self-identification is and what procedural adjustments might be necessary to compensate for “deviance”); and “racial and ethnic classification of immigrants” (to decide how best to fit them into the U.S. racial taxonomy) (Edmonston et al., 1996, p. 3). The political maneuvering entailed in all of this is, of course, tremendous and part of the multiple levels of politics that shape census taking (see also Alonso and Starr, 1987; Anderson, 1988).

But I offer a different emphasis–one of possibility–in the materials I have presented and juxtaposed here. My account of the introduction of the concept of race into Hawai’i is just as much a story of a society without “race” as it is a story of how it becomes a society with “race,” whose concept it is, how it weaves its way into the society, and how it adapts and is adapted as it travels under particular geopolitical circumstances. And my account of the history of race as a federal taxonomy in the United States is an account of adjustments, rather than of substantive change. While the former ultimately becomes a story of incorporation and racialization, it does not begin that way; and while the latter appears to be a story of change and historical accuracy, it has quite a different look from afar.

References

Adams, Romanzo, The Peoples of Hawaii (Honolulu: The Institute of Pacific Relations, 1933).

Adams, Romanzo, Interracial Marriage in Hawaii (New York: Macmillan, 1937).

Alonso, William, and Starr, Paul, eds., The Politics of Numbers (New York: Russell Sage Foundation, 1987).

Anderson, Margo J., The American Census: A Social History (New Haven: Yale University Press, 1988).

“Census Won’t Have Multiracial Category,” Iowa City Press-Citizen, October 30, 1997, p. 7A.

Davis, Eleanor Harmon, Abraham Fornander: A Biography (Honolulu: The University Press of Hawaii, 1979).

Davis, F. James, Who Is Black? One Nation’s Definition (University Park: Penn State University Press, 1991).

Dominguez, Virginia R., White By Definition: Social Classification in Creole Louisiana (New Brunswick: Rutgers University Press, 1986).

Edmonston, Barry, Goldstein, Joshua, and Lott, Juanita Tamayo, eds., Spotlight on Heterogeneity: The Federal Standards for Racial and Ethnic Classification, summary of a workshop, Committee on National Statistics,

National Research Council (Washington, D.C.: National Academy Press, 1996).

Fornander, Abraham, An Account of the Polynesian Race, Its Origin and Migrations, and the Ancient History of the Hawaiian People to the Times of Kamehameha I, 3 vols. in one, compiled by John F. G. Stokes (Rutland, Vt.: Charles E. Tuttle Co., 1969[1878-1885]).

Fornander, Abraham, Fornander Collection of Hawaiian Antiquities and Folk-Lore: The Hawaiian Account of Their Islands and Origin of Their Race with the Traditions of Their Migrations, & c. As Gathered from Original Sources, tr., rev., and ill. with notes by Thomas G. Thrum; Bernice P. Bishop Museum Memoirs, nos. 4, 5, and 6 (Honolulu: Bishop Museum, 1916-1920).

Frankenberg, Ruth, White Women, Race Matters: Social Construction of Whiteness (Minneapolis: University of Minnesota Press, 1993).

Frankenberg, Ruth, ed., Displacing Whiteness: Essays in Social and Cultural Criticism (Durham: Duke University Press, 1997).

Garcia, Joseph, “Panel Opposes Change in Census Listing: Hawaiians Linked to Asians, Islanders,” The Honolulu Advertiser, July 9, 1997, p. A1.

“Government Enlists Thousands to Devise Race Labels: Census Bureau Prepares Survey,” Iowa City Press-Citizen, October 1, 1997, p. 7A.

“Group Urges Cessation of Race-Based Statistics,” Chronicle of Higher Education, September 19, 1997, p. A20.

Ii, John Papa, Fragments of Hawaiian History, tr. Mary Kawena Pukui, ed. Dorothy B. Barrere (Honolulu: Bishop Museum Archives, 1963).

Kamakau, Samuel Manaiakalani, Ruling Chiefs of Hawaii, rev. ed. (Honolulu: The Kamehameha Schools Press, 1992).

Kamakau, Samuel Manaiakalani, Tales and Traditions of the People of Old: Na Mo’olelo a ka Po’e Kahiko, tr. by Mary Kawena Pukui from newspaper articles in Ka Nupepa Kuokoa and Ke Au ‘Oko’a from 1868-1870 (Honolulu: Bishop Museum Press, 1991).

Kamakau, Samuel Manaiakalani, The Works of the People of Old: Na Hana aka Po’e Kahiko, tr. by Mary Kawena Pukui from newspaper articles in Ke Au ‘Oko’a from October 14, 1869 to November 3, 1870 (Honolulu: Bishop Museum Press, 1976).

Kame’eleihiwa, Lilikala, Native Land and Foreign Desires/Ko Hawai’i ‘Aina a me Na Koi Pu’umake aka Po’e Haole (Honolulu: Bishop Museum Press, 1992).

Lee, Sharon M., “Racial Classifications in the U.S. Census: 1890-1990,” Ethnic and Racial Studies 16:1 (January 1993): 75-94.

Lind, Andrew, ed., Race Relations in World Perspective (Honolulu: University of Hawaii Press, 1955).

Lind, Andrew, Hawaii’s People, 4th ed. (Honolulu: The University Press of Hawaii, 1980).

Malo, David, Hawaiian Antiquities, tr. Nathaniel Emerson, rev. ed., Bernice P. Bishop Special Publication, no. 2 (Honolulu: Bishop Museum Press, 1951 [1898]).

Marks, Jonathan, Human Biodiversity: Genes, Race and History (New York: Aldine de Gruyter, 1994).

Mookini, Esther K., The Hawaiian Newspapers (Honolulu: Topgallant Publishing Co., 1974).

Nordyke, Eleanor C., The Peopling of Hawai’i, 2d ed. (Honolulu: University of Hawaii Press, 1989).

Office of Management and Budget, “Advance Notice of Proposed Review and Possible Revision of OMB’s Statistical Policy Directive No. 15, Race and Ethnic Standards for Federal Statistics and Administrative Reporting; and Announcement of Public Hearings on Directive No. 15,” Federal Register 59:110 (June 9, 1994): 29831-35.

Office of Management and Budget, “Interim Notice of Review and Possible Revision of OMB’s Statistical Policy Directive No. 15, Race and Ethnic Standards for Federal Statistics and Administrative Reporting: Summary and Analysis of Public Comments and Brief Discussion of Research Agenda,” Federal Register 60:166 (August 28, 1995): 44674-93.

Office of Management and Budget, “Recommendations from the Interagency Committee for the Review of the Racial and Ethnic Standards to the Office of Management and Budget Concerning Changes to the Standards for the Classification of Federal Data on Race and Ethnicity,” Federal Register 62:131 (July 9, 1997): 36874-946.

Okamura, Jonathan, “Situational Ethnicity,” Ethnic and Racial Studies 4 (October 1981): 452-65.

Omi, Michael and Winant, David, Racial Formation in the United States: From the 1960s to the 1980s (New York: Routledge and Kegan Paul, 1986).

Porteus, Stanley D., A Century of Social Thinking in Hawaii (Palo Alto: Pacific Books, 1962).

Rodriguez, Clara, and Cordero-Guzman, H., “Placing Race in Context,” Ethnic and Racial Studies 15:4 (1992): 523-42.

Sahlins, Marshall, Anahulu: The Anthropology of History in the Kingdom of Hawaii, vol. 1 (Chicago and London: University of Chicago Press, 1992).

Schmitt, Robert C., Historical Statistics of Hawaii (Honolulu: The University Press of Hawaii, 1977).

Smith, Percy B., “Review of Fornander Collection of Hawaiian Antiquities and Folk-Lore,” Journal of the Polynesian Society 27 (June 1918): 78-80.

Spencer, John Michael, The New Colored People: The Mixed-Race Movement in America (New York and London: New York University Press, 1997).

Stoler, Ann, “Racial Histories and Their Regimes of Truth,” Political Power and Social Theory 11 (1997): 183-206.

Thrum, Thomas G., “Brief Sketch of the Life and Labors of S. M. Kamakau, Hawaiian Historian,” Hawaiian Historical Society Annual Report 26 (1917): 40-61.

Thrum’s Hawaiian Annual/All About Hawaii (Published in Honolulu since 1875; cited volumes are from 1904, 1905, 1911, and 1921).

U.S. Department of Commerce, Bureau of the Census, 200 Years of U.S. Census Taking: Population and Housing Questions, 1790-1990 (Washington, D.C.: U.S. Government Printing Office, 1990).

Wells, Samuel R., New Physiognomy, or Signs of Character (New York: Fowler and Wells, 1883).

Wilkerson, B. B., “Native Hawaiians/Census Proposal Harmful to Samoans,” in Letters and Commentary, The Honolulu Advertiser, July 22, 1997, p. A7.

Winchell, Alexander, Preadamites; or a Demonstration of the Existence of Men before Adam (Chicago: S.C. Griggs and Co., 1881).

Wright, Carroll D., The History and Growth of the United States Census, Prepared for the Senate Committee on the Census (Washington, D.C.: Government Printing Office, 1900).

COPYRIGHT 1998 New School for Social Research

COPYRIGHT 2000 Gale Group