Volume 76 1967 > Volume 76, No. 4 > How many Hawaiians?, by Robert C. Schmitt, p 467 - 476
HOW MANY HAWAIIANS?
Time and again, Hawaii statisticians are asked: How many residents of the Islands have Hawaiian blood? How many of these persons are “pure” Hawaiians?
Quoting official census statistics no longer seems to provide an adequate answer. A query on race has appeared in every official census conducted in Hawaii, from the first all-island count in 1849 to the most recent in 1960. Many authorities have questioned the accuracy of these data, however, particularly in more recent enumerations. It has long been suspected that some part Hawaiians tend to forget their non-Hawaiian ancestry in replying to census queries and “pass” as pure Hawaiians. There is also a strong possibility, largely disregarded in earlier discussions, that other part Hawaiians fail to mention their Hawaiian blood and thus merge statistically with a non-Hawaiian group.
There is little question that the number of pure Hawaiians has sharply declined during the 189-year period since the first white contact, while the part Hawaiian population has increased at a rapid rate. Most modern historians and sociologists agree with Romanzo Adams that approximately 300,000 Polynesians lived in the Hawaiian Islands when Captain Cook arrived in 1778. 1 War, infanticide, disease, reduced fertility, and out-migration contributed to their early decline, with syphilis and recurring epidemics as perhaps the most important factors. Hawaiian depopulation has continued at an almost constant rate since the 18th century, averaging 1.8 percent a year between 1778 and 1823, 2.1 percent annually from 1823 to 1853, 1.8 percent from 1853 to 1878, 1.7 percent between 1878 - 468 and 1910, 2.0 percent from 1910 to 1940, 1.6 percent in 1940-1950, and 1.5 percent in 1950-1960. The part Hawaiians meanwhile recorded steady gains, with annual growth rates of 5.1 percent in 1853-1878, 4.2 percent in 1878-1910, 4.7 percent in 1910-1940, 4.0 percent in 1940-1950, and 2.2 percent in 1950-1960. 2 By April 1, 1960, according to alternate census tabulations, there were either 10,502 or 11,294 pure Hawaiians and either 91,597 or 91,109 part Hawaiians in the State. 3 Totals for selected years are reported in table 1. 4
TABLE 1: HAWAIIANS, PART HAWAIIANS, AND NON-HAWAIIANS, FOR HAWAII: 1778 TO 1960
Source: Estimates for 1778 and 1823 by Romanzo Adams and official census data for 1853-1960, cited in Robert C. Schmitt, Demographic Statistics of Hawaii, 1778-1965 (Honolulu: University of Hawaii Press, 1968).
The Hawaiian decline and part Hawaiian rise were relative as well as absolute. Americans, Europeans and Asians migrated to Hawaii in great numbers during the 19th and 20th centuries and soon overtook the native population. Persons of non-Hawaiian blood, less than 3 percent of the total population in 1853, passed the 50-percent mark around 1887 and reached an all-time high of 86.2 percent in 1930. Pure Hawaiians, who accounted for 100 percent of the population of the Island chain on contact, fell to 49.7 percent in the 1884 census and 1.7 percent by 1960. Part Hawaiians constituted 1.3 percent of the - 469 population in 1853, reached a record high of 14.8 percent of the total in 1950, and subsided to 14.5 percent a decade later.
This, roughly, is the official record. How accurate are these statistics?
Not very, according to recent evidence. This evaluation stems from independent survey findings, a comparison of cohort statistics from successive censuses, birth and death tabulations by the Hawaii Department of Health, anthropological field checks, blood-typing studies, and records of the Department of Hawaiian Home Lands and Kamehameha Schools.
These sources differ widely in the degree of care taken in their preparation. The Kamehameha Schools and Department of Hawaiian Home Lands (both of which limit their services to persons with Hawaiian blood) go to considerable effort to obtain accurate information on the racial backgrounds of persons using their facilities. Anthropological and genetic studies are similarly painstaking. A somewhat greater range of error is possible in the household sample and birth and death statistics published by the State Department of Health; although the survey personnel and registration clerks of the Department are well-trained and experienced, the nature of their work precludes in-depth probing of racial ancestry. Least accurate of all available sources is the decennial census, with its hastily-trained, inexperienced enumerators and superficial coverage of race.
Direct evidence regarding the accuracy of census statistics is provided by the Hawaii Health Surveillance Program survey, conducted on the Island of Oahu by the Hawaii State Department of Health. A probability sample of 17,837 persons, interviewed by carefully trained public health nurses and professional survey staff using a detailed schedule, provided data for the two-year period ended March 31, 1966. This survey indicated an average of 5,643 Hawaiians and 98,524 part Hawaiians (out of an estimated non-institutional, non-barracks population of 565,433) living on Oahu in 1964–1966. 5 The 1960 Census, in contrast, reported either 6,647 or 7,420 Hawaiians and either 68,871 or 68,395 part Hawaiians on the Island. 6 Although some of the difference between Health survey and census findings can be attributed to sampling variation, a later survey date, and omission of persons in institutions and military barracks from the Health study, part is obviously related to such matters as schedule design, interviewer training, and editing procedures. Of the two enumerations, the Health survey is in all likelihood the more accurate. Neither count, of course, is free of errors caused by unrecognised or unadmitted racial mixture.
Misreporting has inflated Hawaiian population totals for many years. Romanzo Adams, for example, estimated that 9,780 persons classified as pure Hawaiians in the 1930 U.S. Census were actually part Hawaiians. As a consequence, the number of full-blooded Hawaiians in the Islands in 1930 was actually 12,856 instead of 22,636, and the number of part - 470 Hawaiians was 38,004 rather than 28,224. 7 No allowance was made for part Hawaiians incorrectly classified as non-Hawaiians. Adams wrote:
For more than a hundred years there has been a passing over of part-Hawaiians into the Hawaiian group. That is, part-Hawaiians, especially the darker complexioned ones, frequently are ignorant of their possession of non-Hawaiian blood or they think that their little non-Hawaiian blood is of no practical importance and so they claim to be full-blooded Hawaiians. It is estimated that, as a consequence of such passing and of further unrecognised intermixture, about 43 percent of the so-called Hawaiians of today have a little of the blood of people who came after 1778. In making this estimate the historic situation covering over a century and a half has been considered. 8
An unpublished study by W. A. Lessa suggests that Adams's estimate, if anything, greatly overstated the number of full-blooded Hawaiians surviving in 1930. Lessa wrote:
Regarding Hawaiians, the very thorough research I did in 1930-1932 turned up only about 1,700 Hawaiians, all others not being “pure” by the strict criteria I applied, whereby I rejected any subject with the slightest known admixture. The present figure ought to be perhaps well under 1,000, considering the small size of Hawaiian families and the marked tendency to intermarry with other racial groups. My data are very close to being as accurate as can be . . . 9
Additional evidence is provided by the State Department of Hawaiian Home Lands. This agency serves families in which at least one member has 50 percent or more Hawaiian blood. In April 1965 a Department official estimated that only about 100 of the 1,700 families on land administered by the Department had any pure Hawaiian members. 10
A study in Hawaiian population genetics by Morton, Chung and Mi lends further support to these conclusions. These authors analysed 172,448 live birth certificates and 6,879 fetal death certificates for the years from 1948 through 1958. Noting “considerable evidence that pure Polynesians lack [type] B, and acquire it only through admixture,” they also studied records of the Honolulu blood bank. They concluded that “if the present frequency of B in Hawaiians was attained by a constant rate of admixture, starting from a pure Polynesian population with no B genes, then the Hawaiians born before 1810 were substantially pure Polynesian. . . . On the blood type evidence, ‘Hawaiians’ have 8.5% - 471 Caucasian admixture and 13.7% Chinese admixture, while ‘Caucasian-Hawaiians’ have 8.4% Chinese admixture and ‘Chinese-Hawaiians’ have 14.6% Caucasian admixture.” 11 Morton and his co-authors thus agree with Adams's conclusion that unreported intermixture dates from the early post-contact period. A similar stand was recently taken by McArthur after careful study of historical materials. 12
Vital statistics compiled by the State Department of Health reinforce the impression that full-blooded Hawaiians may be rarer than is generally recognised. As indicated in table 2, five-year totals on the number of pure Hawaiian births have dropped from 1,342 in 1940-1944 to 427 in 1960-1964. Hawaiian births as a percent of the Hawaiian and part Hawaiian total fell from 9.1 in 1940-1944 to 1.7 in 1960-1964. Even these figures are misleading, since the pure Hawaiian counts include illegitimate babies born to Hawaiian mothers and unknown fathers. Such births accounted for 43.3 percent of the pure Hawaiian total in the most recent five-year period. If illegitimate babies are omitted (since most are probably part Hawaiian), the quinquennial totals for full-blooded Hawaiians have declined from 1,126 to only 242 during this twenty-five year period.
TABLE 2: LIVE BIRTHS, FOR HAWAIIANS AND PART HAWAIIANS: 1940 TO 1964
Source: Hawaii Department of Health, annual reports and records.
Comparison of census data with birth statistics provides further evidence. The 1960 U.S. Census reported 26.6 percent more Hawaiians under ten years of age than appeared in statistics on live births for the preceding decade; yet the Census showed 21.7 percent fewer part Hawaiians. When the 1960 Census count on persons 10 to 19 is compared with 1950 Census data on persons under 10, there appears to have been a - 472 5.2 percent increase in Hawaiians and a 17.0 percent decrease in part Hawaiians. When allowance is made for normal mortality and out-migration, these comparisons (presented in greater detail in Table 3) strongly indicate an overrepresentation of pure Hawaiians and under-representation of part Hawaiians in the Census.
TABLE 3: SELECTED GROUPS OF HAWAIIANS AND PART HAWAIIANS: 1940 TO 1960
Source: Hawaii Department of Health, annual reports for 1940-1959 and records. U.S. Bureau of the Census, U.S. Census of Population: 1950, Bulletin P-C52, tables 29 and 30, and U.S. Census of Population: 1960, Final Report PC(2)-1C, table 61. Kamehameha Schools, records.
The proportion of Hawaiian births differs widely by source, varying inversely with the degree of care likely to have been taken in compiling the data. Statistics developed by the Kamehameha Schools (which follow a policy of restricting enrolment, except for faculty children, to Hawaiians and part Hawaiians) provide an example. A careful check of enrolment records as of November 1, 1959, showed that only seventeen (or 1.0 percent) of the 1,749 boys and girls in the school were unmixed. 13 Birth statistics for the corresponding age group—that is, Hawaiian and part - 473 Hawaiian children born 1942 to 1954—report 5.3 percent as full-blooded. Among Hawaiian and part Hawaiian children five to seventeen years old enumerated in the 1960 Census, 6.5 percent were listed as unmixed. Additional information appears in Table 3. Although these differences may reflect differential mortality, migration and schooling patterns, it seems more likely that they stem chiefly from interviewer procedures and response verification.
Part Hawaiians are becoming progressively more fractional. In 1965, for example, more than half of the 4,732 Hawaiian and part Hawaiian babies born had one part Hawaiian and one non-Hawaiian parent (see Table 4). The composite or average blood mixture of students at the Kamehameha Schools dropped from 50 percent in 1954 to 36 percent in 1964, and the proportion with at least half-Hawaiian ancestry declined from 60.9 to 34.2 percent of the student body during the same ten-year span (see Table 5). The schools reported forty-one pure Hawaiian students
TABLE 4: LIVE BIRTHS OCCURRING IN HAWAII, BY RACE OF FATHER AND MOTHER: 1965
Source: Annual Report, Department of Health, State of Hawaii, Statistical Supplement, 1965, p. 11.
TABLE 5: RACIAL COMPOSITION OF STUDENTS AT THE KAMEHAMEHA SCHOOLS: 1954 TO 1964
Source: The Kamehameha Schools, records. Data supplied by Jack Darvill, Assistant to the President, May 4 and 7, 1965.- 474
(3.3 percent of the student body) in March 1954 but only seventeen (0.8 percent) a decade later. The 1964-1966 Hawaii Health Surveillance Program survey cited earlier found that 64,478 of the 98,524 part Hawaiians on Oahu admitted to two or more non-Hawaiian strains. (Among the 34,046 with only one non-Hawaiian strain, 18,101 were Caucasian-Hawaiians and 11,983 were Chinese-Hawaiians.) 14
The growing number of Hawaii residents with one-eighth, one-sixteenth, or even less Hawaiian blood suggests the possibility that many such persons, technically classified as part Hawaiians, may be reporting themselves as non-Hawaiians to census enumerators. Evidence supporting such a notion appears in migration estimates, which show a heavy net out-migration for Hawaiians and part Hawaiians. 15 These estimates, computed by the residual method, may in fact reflect “passing” of part Hawaiians as non-Hawaiians as much as they mirror a true out-migration. A part Hawaiian child, correctly classified on its birth certificate but thought of as non-Hawaiian by the census-taker, would appear as an out-migrant in computations of intercensal components of population change.
In the light of the foregoing comments and data, only the roughest kind of estimate of the “true” number of Hawaiians and part Hawaiians living in the Islands can be ventured. Recent figures on “pure” Hawaiians range from Lessa's informal guess of “well under 1,000” to the 1960 Census total of 11,294. Although no independent Statewide estimates of the part Hawaiian population have appeared in print in recent years, Oahu data from the Health survey suggest that the 1960 Census count of somewhat more than 91,000 for the entire chain was perhaps a quarter too low. All things considered, the weight of evidence points to a current population of around 130,000 persons with Hawaiian blood. Of this total, probably not more than a thousand could accurately claim unmixed ancestry, in the strictest sense of the term.
- 476 Page is blank
1 For a discussion of Hawaiian population trends during the 72-year period between Cook's visit and the first complete census, see Schmitt 1968: Chapter II.
2 These annually compounded rates were computed from table 1.
3 The lower Hawaiian and higher part Hawaiian totals were obtained from a special hand tabulation of 1960 Census data, in which non-response was allocated on a prorata basis; the higher Hawaiian and lower part Hawaiian totals, in contrast, were obtained from a machine tabulation in which non-response was usually assigned on the basis of race reported for the household head. See Hawaii Department of Planning and Economic Development 1963.
4 Seventeen official censuses have been taken in Hawaii during this period. For data from those not cited here, see Adams 1933:8-9, Adams 1937:8, Lind 1955:27, and Schmitt 1968: Chapters III and IV.
5 Hawaii Department of Health 1967.
6 See footnote 3 above.
7 Adams 1937:15.
8 Adams 1937:14.
9 W. A. Lessa, personal communicaton dated November 11, 1964. Dr. Lessa, now a Professor of Anthropology at the University of California, Los Angeles, did his Hawaiian research as part of a larger study under the guidance and jurisdiction of Dr. Harry L. Shapiro of the American Museum of Natural History. The statistical work was completed at the Peabody Museum of Harvard University under Earnest Hooton.
10 James C. Clarke, Planning Coordinator, in discussion with the present writer, April 26, 1965.
11 Morton, Chung and Mi 1967:13, 26, 29, and 127.
12 McArthur 1966.
13 For Kindergarten through the 12th grade, excluding 10 non-Hawaiian faculty children. Data supplied by the Office of the President, The Kamehameha Schools, May 24, 1965.
14 Hawaii Department of Health 1967.
15 Schmitt 1961-1962:19-20.