Prologue Magazine

Race, Nationality, and Reality, Part 2

INS Administration of Racial Provisions in U.S. Immigration and Nationality Law Since 1898

Summer 2002, Vol. 34, No. 2

By Marian L. Smith


Whether to apply a "color test" or to determine whiteness by reference to an ethnological table of the families and races of mankind was the basic question put to naturalization officials for years thereafter. Court officers, attorneys, and private citizens regularly requested a ruling from the Department of Commerce and Labor on the question. But, as Naturalization Division Chief Richard K. Campbell admitted in 1907, such a ruling would not "be considered authoritative if one were adopted," because the ultimate power to decide the issue lay with the courts.12

At the same time, Campbell understood the 1906 naturalization act to have given his agency a clear mandate to guide naturalization courts toward uniformity. When the various U.S. attorneys who represented the government in naturalization cases asked Campbell how they should proceed without a definitive court ruling, he began to offer the opinion of his office regarding racial eligibility to naturalization and to suggest a course of action for the attorneys. While Campbell's writings indicate that he personally considered "off-color races not only unfit but as well undesirable additions to the body politic," his strategy consisted of efforts to force courts into making decisions one way or the other. He began in August 1908 by advising Andrew Balliet, assistant U.S. attorney in Seattle, to direct clerks of court to refuse to accept declarations of intention or file petitions for naturalization on behalf of "Hindoos," or East Indians. In doing so, Campbell hoped to push the applicants toward a higher court that might issue a precedent decision.13

Beyond his effort to coordinate the activities of U.S. attorneys, and through them influence clerks of courts, Campbell could offer little in the way of practical instruction. He received numerous queries from clerks like J. W. Porter, clerk of the circuit court, Champaign County, Illinois, who asked if the division could supply a "list of countries from which, and from which only, natives may apply for naturalization?"14 Campbell replied that there was no list, nor would such be of any "practical value." Rather, a list of countries, representing nationalities, "would be very apt to mislead you" since §2169 does "not apply to countries or to natives of any country, solely, as white persons may come from any country to the United States and be eligible to citizenship."15 Like the courts, the Naturalization Division could offer no clear guidance on Clerk Porter's practical question.

Copies of Richard Campbell's earlier letter to U.S. Attorney Balliet, which implied that clerks of court were not to accept naturalization applications by any person racially ineligible to citizenship, eventually forced Campbell and the division into retreat. In the next year, 1909, New York immigrant advocate Justin S. Kirrah launched a campaign on behalf of Syrians, Turks, East Indians, and others deemed non-white by Campbell. In a complaint to the secretary of commerce and labor, Kirrah accused the division chief of "personal malice" toward certain ethnic groups and of interfering with work of the courts and "abuse of his office and usurpation of functions appertaining to Federal officials other than himself."16 At the same time, the Board of Delegates on Civil Rights of the Union of American Hebrew Congregations also complained to the secretary about the department's denial of naturalization to Turks, Syrians, Armenians, Palestinians, and Jews, pointing out that if the ruling were upheld, "it would, if living, exclude David and Isaih [sic] and even Jesus of Nazareth himself."17

Secretary of Commerce and Labor Charles Nagel soon admonished the chief of naturalization for pretending to decide for the courts whether a person was racially qualified to apply for naturalization, advising the division to "be concerned more with the present reputation than with the original nativity of the applicant." Campbell responded by mailing a new letter of instructions to each chief naturalization examiner (representatives of the division in naturalization hearings). He admitted that racial eligibility was a question of law "and is not a question which this Department or any officer thereof is authorized to decide." On the other hand, it remained the examiners' duty to advise the court of each and every petitioner's race.18 Though the 1906 Naturalization Act put the Naturalization Division in charge of "all matters" related to naturalization, this 1909 episode demonstrated the limited power held by the administrative agency at that time. The division and the courts actually shared, or enjoyed divided, authority over various aspects of the naturalization process. And as Campbell later remarked, "[d]ual authority and responsibility never have been and never will be productive of anything but confusion."19

Unless the courts issued enough decisions to determine the eligibility of every possible "race" that might apply for citizenship, or until the Supreme Court might issue broad guidance on the definition of white persons, federal officials continued to live amid uncertainty. Those at the Department of Interior's General Land Office were frequently embarrassed by the fact that they accepted homestead applications from aliens racially ineligible to citizenship, then had difficulty issuing final patents when the homesteader was unable to naturalize. Often, judges frustrated with contradictory federal rules would admit racially ineligible homesteaders to citizenship as a protest to conflicting federal practices.20

Other absurd cases continued to arise, like the case of Carmen Garcia, a Mexican teacher at the Government Indian School in Riverside, California. While employed by one arm of the U.S. government to teach Americanization on the reservation, Garcia was found racially ineligible to naturalize by another arm of that government because her ancestry was three-quarters Papago Indian.21 In correspondence with U.S. attorneys, Campbell sought cases in which an objection to the naturalization of a racially ineligible applicant might make its way to the Supreme Court. Attorneys, however, were understandably reluctant to defend such objections without evidence, and there was no settled opinion on what constituted evidence of race. The courts increasingly turned to science for an answer to the question, and the state of early twentieth-century American ethnology was peculiarly qualified to confuse the issue further.

When trying to advise a local judge about the racial eligibility of "Hindoos," or East Asians, the assistant U.S. attorney in Seattle consulted a scientific library in an effort to learn whether or not East Indians were white. He learned that ethnologists considered East Asians to be "from Aryan stock," specifically "Caucasic" or "Caucasian." He then advised the court to accept a Hindu's application, since it seemed "from the best source of information obtainable by this office on that subject that they are included in the term 'Caucasian.' If that is true they have a right to make their declaration." Expressing some doubt that his action was appropriate, the attorney wrote the Division of Naturalization for Campbell's opinion.22

Campbell frequently voiced his disregard for "ethnologists, of whom the authors of the statutes had as little knowledge as they did of this rather speculative science." Campbell said the division based its determination that Hindus were not white persons on previous court decisions holding that Indians, Chinese, Japanese, Hawaiians before annexation, and natives of British Burma were all nonwhite Further, "prior to the Civil War negroes, of whatever shade, were ineligible." Arguments that Hindus were Caucasians were irrelevant, for it was "immaterial whether the writer is ethnically an Aryan or a Mongolian. His claim in that respect is probably correct. The fact is that he is not a white person." To determine who was and was not white, Campbell was happy to leave the issue to the courts, whose decisions would be based not on science, but on "common understanding":

The law refers to persons, and confines the right to become naturalized to those who are white. It does not seem to me that the introduction of expert testimony would result otherwise than it does where testimony of this character is used to establish any doubtful point, the question of sanity, for example. Without being able to define a white person, the average man in the street understands distinctly what it means, and would find no difficulty in assigning to the yellow race a Turk, or Syrian, with as much ease as he would bestow that designation on a Chinaman, or Korean.

The result desired is one that can be reached definitely, not by any general rule, but by a slow process of elimination.

Campbell's approach now agreed with that taken by the lower courts, which considered each race or ethnic group on a case-by-case basis. The accumulation of court decisions over time, however, did not necessarily bring clarity. Campbell observed in 1916 that the language of §2169 "has been fruitful of much honest diversity of construction by the public and by the courts . . . in the absence of legislative attempt to remove its obscurity."23

The Supreme Court first attempted to construe the meaning of "white persons" in 1922, in the case of Ozawa v. The United States.24 In its decision, the court denied naturalization to Takao Ozawa, a Japanese born in Japan, on the grounds that he was not a white person. The written decision referred to white persons as those "popularly known as the Caucasian race." To many, this seemed a nod to the ethnological argument rejected by Campbell, and consequently "gave rise to a difficult situation in regard to those races who were not white of skin, but who were classified as Caucasian by ethnologists because of a remote common ancestry with admittedly white persons."25

The Supreme Court soon clarified its reasoning, and rejected scientific ethnological classification as a determiner of race, in its 1923 United States v. Thind decision. Bhagat Singh Thind was "a high caste Hindu of full Indian blood" born in Punjab, whom the court denied eligibility to naturalization. The written opinion explained that the words "white person" were to be understood as those found "in common speech, to be interpreted in accordance with the understanding of the common man, synonymous with the word 'Caucasian' only as that word is popularly understood."26 The Thind decision became the touchstone of judicial policy toward racial eligibility for another two decades.

The Supreme Court's reasoning in the Thind case had significance beyond determining that Hindus were racially ineligible to naturalize. While trying to divine the original intent of Congress in using the term "white persons," the court referred to Congress's more recent creation of the "Asiatic barred zone"27 defined within the Immigration Act of 1917. In the immigration law, Congress designated a geographic area from which immigration was prohibited, and the court concluded "it is not likely that Congress would be willing to accept as citizens a class of persons whom it rejects as immigrants."28 The court did not argue that Congress in 1917 was attempting to define the term "white persons" when drawing the zone, only that in doing so Congress again demonstrated that "common understanding" of nonwhite, non-African regions of the world. Furthermore, the Thind decision suggested that additional races from outside the barred zone could still be found ineligible to naturalization.

Yet the reference to immigration law and its barred zone seemed to offer an end to the confusion of naturalization officials, judges, and naturalization attorneys across the nation. The Immigration Act of 1917 described the barred zone with latitudinal and longitudinal precision. If one took Thind literally, one need only consult a map of the zone to determine the eligibility of any naturalization applicant.29 Commissioner of Naturalization Raymond F. Crist, who replaced Richard Campbell in 1923, called the Thind decision "a source of gratification to administrative officers" who longed for a settlement of the legal question by the Supreme Court, regardless of whether the court decision upheld or overturned previous administrative decisions. Supreme Court attention to disputed provisions of the nationality law were especially important, he noted, as the Constitution mandated uniformity in application of naturalization laws. Crist seemed to think the Thind decision would "clarify the atmosphere of doubt hitherto existing in the interpretation of the statutes."30

A Detroit judge soon demonstrated that questions of racial eligibility would continue to be decided on "common understanding," as recommended in Thind, rather than the Asiatic barred zone. Following the Thind decision, a number of suits commenced to cancel citizenship earlier granted to other Hindus. Facing proceedings to cancel his 1921 naturalization, John Mohammed Ali told the U.S. District Court in Detroit in 1925 that though born in India, he was not East Indian or Hindu. Rather, he was properly Arabian, for his ancestors originated in Arabia. The court replied that Ali's ancient ancestry was not at issue. Ali had dark skin and fit all other criteria that had disqualified Thind from naturalization. The court, however, did not definitely hold that persons of the Arabian race were ineligible. The decision in Ali's case, by canceling his citizenship but not deciding the eligibility of Arabs, only served to renew controversy over the eligibility of those peoples who bordered on the barred zone.31

Afghanistan was only partially included within the barred zone, but in 1928 a California judge also dismissed an Afghan's petition for naturalization on the basis of the Thind decision. In re Feroz Din, the court denied naturalization to an Afghan because he was neither white nor of African nativity or descent. It was not necessary to consult scientific evidence, the judge noted in his brief decision, for "[w]hat ethnologists, anthropologists, and other so-called scientists may speculate and conjecture in respect to races and origins may interest the curious and convince the credulous, but is of no moment in arriving at the intent of Congress in the statute aforesaid."32

Any remaining hope that immigration law's barred zone would provide a rational system for determining racial eligibility was dashed by the Immigration Act of 1924.33 In a marvelous example of legal cross-referencing, the 1924 act denied eligibility to immigrate to those denied eligibility to naturalize under U.S. nationality law— the very nationality law recently interpreted by reference to immigration law! Put simply, any person ineligible to naturalize was now ineligible to immigrate. By essentially incorporating §2169 into immigration law, Congress offered both the Immigration Service and the Department of State's Foreign Service the opportunity to wrestle with the question of who was, and was not, racially eligible to naturalize.

A good example of the problem appeared in early 1930, in India, on the basis of a rumor that a United States court had declared Parsees eligible to citizenship.34 Immediately, U.S. Consul at Bombay Wilbur Keblinger was deluged with Parsee immigrant visa applications. "Inasmuch as practically all of the Parsees in the world live in the consular district," he wrote the secretary of state, "it would be appreciated if the Department could inform this Consulate whether or not such a decision has been handed down."35 Though guidance provided to the State Department by the assistant commissioner of naturalization assured U.S. consuls that Parsees remained ineligible to citizenship because they were native to the Asiatic barred zone, a U.S. consul in London continued to be troubled by the case of a Parsee actor who wished to go to Hollywood. The applicant claimed to be Parsee, but not native to the barred zone.36

The level of confusion overseas matched that at home, where racial exclusions to citizenship did not always appear uniform. "Does race or color mean nothing?" asked a resident of Portland, Oregon, who read in his local paper that two Japanese and one Chinese were naturalized by a U.S. court. His newspaper did not explain the three were U.S.-born women who had previously lost their U.S. citizenship by marriage to aliens. Two of the three were Asian Americans. One of the so-called Japanese women was actually of Scandinavian descent and had acquired Japanese nationality and "race" by marriage. All three women, because they were U.S.-born, were allowed to naturalize despite their "race" under special provisions of a law amended by Congress in 1931. Congress again waived racial requirements for citizenship in 1935 when allowing for the naturalization of racially ineligible World War I veterans.37

The confusion of race with nationality became more obvious and problematic for immigration and naturalization officials after a 1933 - 1934 reorganization of the executive branch recombined the Immigration Bureau and the Naturalization Bureau into one agency, the Immigration and Naturalization Service. Separate since 1913, the two bureaucracies became one devoted to enforcing both immigration and nationality law. That the two used different racial classification systems for procedural and statistical purposes initiated some reconsideration and revision of internal, administrative guidance on the question. As recently as 1930, the Department of Labor concluded that no change could be made to the designation of races and peoples on certain immigration and naturalization forms because the data was required "by law."38 But by the mid-1930s the INS began, at times, to use its administrative discretion to alter its classification of race.

Change began with amendment of the traditional List of Races or Peoples devised by the Immigration Service on Ellis Island in 1898. Internal instructions issued September 11, 1936, announced two changes to the list and four additions. The two revisions illustrated a continuing mixture of race (as color) and nationality on the list. "African (black)" changed to "Negro," no doubt simplifying the classification of blacks from the Caribbean or Central and South America. Previously, blacks from Cuba or the West Indies, for example, were designated as "African" though typically not native to Africa. Another change combined the statistical codes for "Italian (north)" and "Italian (south)."39 In 1898, the northern and southern Italians were classified separately because the list depended heavily on language to identify differing races or peoples. By 1936, international politics held greater sway.

The same change from an ethnic to a political definition of "peoples" can be seen in the four 1936 additions to the list: Albanian, Estonian, Latvian, and Filipino. The Immigration Service had classified Latvians as Lithuanians for nearly four decades. But in 1935 the Latvian consul general began a campaign to convince the commissioner of immigration and naturalization to separate the two groups. His argument was simple: "'Lithuanian' is, of course, not a race, nor are Latvians Lithuanians, nor Lithuanians Latvians."

The North American Manx Association made a similar plea in June 1937, pointing out that the "native race or people of the Isle of Man" were as distinct from the English as were the Irish, Scottish, or Welsh, and deserved similar recognition on the List of Races or Peoples. INS Circular No. 152, of August 12, 1937, announced the addition of "Manx" to the list. And protests from the Mexican government "that Mexicans did not belong to the 'colored' races" caused INS to issue new guidance emphasizing that Mexicans were considered white for immigration and naturalization purposes. A similar response to protests from Brazil in 1942 had INS revising all references to "Spanish American" to read "Latin American."40

By far the most pressing, and embarrassing, item on the List of Races or Peoples in the late 1930s was the term "Hebrew." The American Jewish Committee protested the classification of Hebrew as a race as early as 1930, warning that such "inquisition" into religion by the government was "improper and susceptible of unfortunate abuse." At that time, the solicitor for the Department of Labor wrote a long memorandum on the legal requirement for including race— and Hebrew as a race— on both immigration and naturalization forms. The department found the American Jewish Committee's complaint groundless and rejected their request.41 In the following years, as Nazi persecution of Jews in Europe increased, dissatisfaction with the presence of Hebrew on the list widened and deepened.

Somehow, late in the 1930s, the service was able to amend the procedure for recording race on naturalization (as opposed to immigration) records. Vague references to this change imply that traditional difficulties in "obtaining and recording proper information as to race" caused a revision of forms wherein the applicant no longer stated his race, but chose a racial designation from a list of those eligible to citizenship. This revision occurred prior to passage of the Nationality Act of 1940, evidenced by a May 7, 1940, edition of form A-2214, Application for a Certificate of Arrival and Preliminary Form for Petition for Naturalization. A blank calling for race does not appear on the form until page three, and refers the applicant to instructions listing the choices of white, African or African descent, or Filipino. "State to which one of these classifications you belong."42

All INS naturalization forms were revised after passage of the Nationality Act of 1940. Based on recommendations offered by a presidential committee studying naturalization problems since 1933, the act recodified all (and reconciled much) previous U.S. nationality law. Among proposals associated with the committee were some recommending elimination of any racial requirement for naturalization. Unfortunately, §303 of the new law extended racial eligibility to only one new group: descendents of races indigenous to the Western Hemisphere. Section 303 not only required INS to revise naturalization forms again, it raised a question of reinterpretation of the old §2169. Furthermore, it begged the question: Who are races "indigenous to the Western Hemisphere?"

Recodification of nationality law forced the INS to examine all legal language and reevaluate its administration. It also became an opportunity to reinterpret §2169, which was not repealed but was replaced for INS purposes by §303. Though the 1940 law covered only naturalization, the legislative links between racial eligibility and immigration had been established earlier. Thus any change to racial classification in administration of naturalization could affect racial classification in the immigration arena.

It was with this opportunity in mind that Commissioner of Immigration and Naturalization Earl G. Harrison began his effort to remove Hebrew from the immigration List of Races or Peoples. Two months after taking office, Harrison asked Henry Bernard Hazard, an immigration and nationality law expert and director of the service's Research and Education Division, to review the practice of racial classification on INS forms and answer the question of whether "'Hebrew' was a race." After more than thirty years of working to standardize immigration and naturalization, Hazard may have shared Harrison's obvious desire to excise the word "Hebrew" from official forms. In any event, Hazard quickly submitted a long analysis of the question, concluding that Jews or Hebrews could not properly be considered a race, "at least not in the sense in which that term is used in the immigration law."43

More important, Hazard concluded that both "race" and "peoples" were subject to administrative determination under naturalization and immigration law. In his review of immigration's list, Hazard noted that the law required a record of each arriving immigrant's race but "has nothing to say about 'the people' to whom the alien may belong. The enlargement of the classification . . . to include 'peoples' appears to have been made arbitrarily, possibly because of the difficulty in determining just what the term 'race' might imply." Similarly, concerning the Nationality Act of 1940, Hazard concluded that neither the law nor regulations defined the term "race." As a result, whatever system the INS used to classify or supply race on naturalization forms was "a matter resting in administrative discretion." One year later, the term "Hebrew" no longer appeared on immigration forms and papers.44

Another area where the INS increased its use of administrative discretion related to racial eligibility to naturalization. As has been shown, the courts traditionally admitted or denied racially ineligible aliens during court proceedings, either according to or despite INS objections. But developments in administrative law generally began to provide INS officers with opportunities to make decisions outside the courtroom. The cases of Majid Ramsay Sharif (Shariph) and Noshad Khan are illustrative.45

Sharif, an Arab, applied for an immigration visa in 1941 but was denied as an alien racially ineligible to citizenship. Khan, an Afghan resident of the United States since 1926, faced deportation charges in 1941 for illegal entry but applied for discretionary relief as his deportation would be a hardship on his U.S. citizen wife and children. Whether the INS could afford relief to Khan and adjust his immigration status depended on whether an Afghan was racially eligible to naturalize. In both cases, because they were not petitions for naturalization, the questions went not to the courts but to the Board of Immigration Appeals (BIA).46 Administrative law could now determine the question of racial eligibility.

In Sharif's case involving the eligibility of an Arab, the board, like the courts, relied on the Thind decision. Unlike the courts, the BIA was persuaded by a brief for the U.S. government in the Thind case that argued that "whiteness," for lack of a better term, is associated with Western civilization, and Western civilization includes "so much of the Near East as contributed to, and was assimiliable with, the development of Western Civilization of Greece and Rome." Having recalled the cultural link between the ancient and modern western worlds, the board concluded "that it was not intended, either in 1790 at the time of the first enactment of the governing statute or certainly in 1940 at the time of its last enactment, that Arabians be excluded from the group of 'white persons'."47 Unless one is prepared to believe immigration officials were naturally more benign that Supreme Court justices, the Sharif case demonstrates a changed "common understanding" in 1941 from that which persuaded the court in 1923. Upon revisiting the Thind decision, the BIA came to an opposite conclusion.

The board exhibited equal flexibility in the case of Noshad Khan, which had to overcome the 1928 In re Feroz Din decision that Afghans were not white persons. To do so, the BIA questioned whether Americans had any "defined popular or common understanding" of Afghans given there were only an estimated two hundred Afghans then living in the United States. Without any popular guide, the board turned to ethnologists who provided evidence that "an Afghan is the exact prototype of the Persian." Since Persians had always been considered racially eligible to naturalize, Afghans would be as well. Because it departed so far from case law, the BIA forwarded its Khan decision for approval by the attorney general, who in turn requested the opinion of the INS. After the INS legal office supported the Khan decision, the attorney general approved it on May 26, 1945.48 Afghans had thus been found eligible to naturalize and to immigrate despite the opinion of, and without the participation of, the courts.

INS changes to the classification of race and administration of racial provisions in immigration and nationality law reflected changes in American thinking or "common understanding." During the Seventy-eighth Congress, in 1942 and 1943, eleven different bills were introduced proposing elimination of racial barriers to naturalization. While most of the bills only aimed to remove the bars to Filipino, Korean, or East Indian naturalization, they indicated a willingness or desire on the part of the American people to liberalize a restrictive policy maintained since 1924.49 It should not be forgotten that changes to naturalization at home during World War II coincided with administrative naturalization of soldiers serving in the U.S. Armed Forces overseas. Since the First World War, Congress maintained legal provisions facilitating the naturalization of U.S. soldiers despite racial ineligibility. During World War II, the practical and propaganda value of naturalization ceremonies held in Europe, North Africa, and the Pacific, in which representatives of all corners of the earth came together to pledge themselves to American ideals, fostered the image of the United States as the defender of democracy worldwide. But they also stood in uncomfortable contrast to continuing racial exclusions to citizenship on the home front.

Unlike American attitudes toward race, INS statistical methods in 1940 remained much the same as they had been at the turn of the century. The ability, willingness, or practicality of changing INS racial classification and coding depended on the service's perception of its legal requirements to record and report racial statistics. That the INS could alter or amend its statistical system became clear in the late 1930s, when additions and changes were made to the List of Races or Peoples. Further change became possible after the realization, stated in Henry B. Hazard's 1942 memorandum to Commissioner Harrison, that "race" and "people" were not defined by law but by administrative practice.

Race, Nationality, and Reality Part 1
Race, Nationality, and Reality, Part 3



Marian L. Smith is the senior historian for the U.S. Immigration and Naturalization Service, Washington, D.C. She writes and speaks about the history of the agency.



Articles published in Prologue do not necessarily represent the views of NARA or of any other agency of the United States Government.