University of Nevada, Reno

Basque Center

   Issue 61
   Issue 60
   Issue 59
   Issue 58
   Issue 57
   Issue 56
   Issue 55
   Issue 54
   Issue 53
   Issue 52
   Issue 51
   Issue 50
   Issue 49
   Issue 48
   Issue 47
      Tree Trunks
      AAA Session
      Goian Bego
      BSP on CNN
  Issue 46

Basque Studies Program Newsletter · Issue 47, 1993

Re-counting Basques
by William A. Douglass

The 1980 U.S. census marked the first time that Americans were allowed to specify their ethnic identity(ies). For students of immigration and ethnicity this was a watershed event. Those of us with an interest in Basques were particularly elated since the approach in previous censuses, which emphasized “nationality” rather than ethnic identity, glossed most Basques as either French or Spanish nationals. A Basque from Havana was Cuban; one from Manila was a Filipino, etc. In short, the census was a highly flawed document for analyzing Basque immigration into the United States (in some, but not all, cases Basques could be inferred; for example, “Spanish” nationals in Idaho could be assumed to be Bizkaian Basques, however, those in California or New York could not).

The Basque Studies Program played a privileged role in the modification of the 1980 census schedule. When contacted by the Bureau of the Census for advice, we argued successfully that an internal distinction should be made between French and Spanish Basques at the very least. We would have preferred even greater refinement (e.g. Bizkaians, Navarrese, Souletins, etc.) but such was impossible given that each new category complicated enormously the Bureau’s task. As it stands, for normal reporting purposes (i.e. publications by the Bureau of the Census), “Basques” is one of only three Western European categories that is further subdivided (Cypriots and Portuguese being the other two). Thus, for Basque specialists the U.S. census has gone from being an inept exercise to becoming a highly significant one. In 1990 the Bureau of the Census conducted its second survey of the American population employing ethnic identity distinctions. What can be learned from a Basque perspective by comparing the two censuses?

First, it should be noted that, as in 1980, respondents could specify more than one ethnic identity. Unlike the 1980 exercise, however, when only one in five respondents was asked to do so, the 1990 census requested that all Americans answer the question (and more than 90 percent complied). Obviously, the 1990 results are much more reliable than the 1980 ones, which projected population totals from the sample. In 1980 the smaller the Basque population of a particular state the greater was the probability of error. In 1990 this was much less of a factor.

Table One details the 1990 totals of French, Spanish and undifferentiated or generic Basques in the United States. It should be noted that reference is to ethnic identity and not birthplace. Given the highly circumscribed immigration of Basques into the country over the past several decades, a majority of the respondents were born in the United States. The figures in parentheses in Table One are the totals reported in the 1980 census. Table Two presents the eleven states with 800 or more Basques in the 1990 census. Again the results for 1980 are in parentheses.


Table 2: States with 800 or More Basque Residents


Number of Basques, 1990 & (1980)

  California 19,122 (15,530)
  Idaho 5,587 (4,332)
  Nevada 4,840 (3,378)
  Oregon 2,257 (2,253)
  Washington 1,770 (1,134)
  Utah 1,422 (873)
  Arizona 1,316 (1,100)
  New York 1,300 (1,426)
  Texas 1,248 (887)
  Florida 1,189 (859)
  Colorado 955 (955)

Several comments may be made. First, in both censuses California predominates with about one-third of the nation’s total Basque population. In both counts California led second-place Idaho by nearly a four-to-one margin. In both censuses Basques were present in all fifty states. In the 1980 one Delaware, with 21 persons, reported the smallest Basque population of any state, whereas by 1990 Vermont’s two Basques ranked as the fewest. It is quite evident that sampling bias and coding error contaminated some of the 1980 results, particularly in the states with small Basque populations. For example, it is just as hard to explain why West Virginia had 106 Basques in 1980 as it is to understand why they had declined to nine persons over the intervening decade.

In the 1980 census there was clearly a coding problem which introduced an error with respect to French Basques (someone else was being counted as such in at least one state sample), since the most egregious or counterintuitive discrepancy in the results was in the Nebraska tally. In that state 2,707 persons were reported as French Basque in 1980, whereas no one was by 1990. According to the 1980 census, Nebraska’s 2754 Basques represented the fourth largest concentration in the United States, yet none of us working in Basque immigration history had ever met one! The state’s figure of 45 Nebraska Basques in 1990 is much more reasonable.

The Nebraska correction should be kept in mind since, in 1980, it represented an overcount of U.S. Basques by a factor of over 6 percent. If this error is deducted from the 1980 totals the projected population in that year is slightly over 40,000 Basques in the United States. Therefore, the increment in the interim represents a rise on the order of about 16 percent rather than the approximately ten percent increase reported between the two censuses. Where is the growth concentrated and how might we interpret it?

We find that California, Idaho, Nevada and Oregon are, in descending order, the four states with the largest concentrations of Basques in both censuses (once the Nebraska totals are purged from the 1980 one). The first three have all experienced considerable and comparable growth rates between the two censuses. Indeed, their combined total increase of 6,317 alone accounts for most of the increase in the Basque-American community. At the same time, the growth rate is somewhat smaller in the southern Idaho-southeastern Oregon region of Bizkaian settlement than in Nevada, probably a reflection of the latter’s overall rapid growth rate compared with the former’s more modest one.

Washington state’s Basque population increased dramatically as well, probably reflecting the Seattle effect as a magnet of migration from throughout the Pacific Northwest. Utah moved from eleventh to sixth position, likely due to the similar magnetic draw of the greater Salt Lake City area as a regional center. There seems to be a Sun-belt phenomenon discernible in the increase in the Basque populations of Arizona, Texas and Florida, whereas Colorado’s total remained perfectly static and New York’s actually declined somewhat.

Over the past decade there has been little movement of Basques into the United States. Indeed, it is likely that the scant immigration was largely offset by the departures of Basques returning to Europe to live. Therefore the overall growth of the Basque population of the United States is largely attributable to natural increase among Basque-Americans, coupled with their disposition to continue to self-identify as Basques. We might also infer at least the possibility that Basques in the United States manifest a greater propensity for endogamy, or marriage within the ethnic group, than most other Americans of European descent. These conclusions derive from the fact that of those persons reporting Basque ancestry 78.9 percent made it their first designation (recall that it was possible to list up to three ethnic identities, including simply “American”). Only Greeks and Cypriots manifested a greater propensity to retain their primary ethnic designation.

There is, however, an interesting internal trend discernible within the Basque-American population away from either “French Basque” or “Spanish Basque” identity in favor of the simply “Basque” one. Thus, in 1980 11,919, or 28 percent of the listed total of 43,140 Basques opted for French Basque, 8,534, or 20 percent, for Spanish Basque and 22,686, or 52 percent, for Basque as their primary identity. In 1990 the percentages shift to 12, 16, and a dramatic 72 percent respectively.

We might speculate that this is one sign of the “ageing” of the Basque-American community in which children of Basque parents are more prone to see themselves in “Basque” terms rather than “Spanish Basque” or “Bizkaian” ones, for example. It is even possible that we are witnessing a cumulative effect of the proliferation of Basque social clubs and creation of NABO in the 1970s and 1980s. The clubs and NABO tend to emphasize Basqueness rather than Old World Basque regional distinctions.

In conclusion, the ethnic ancestry data derived from the last two U.S. censuses represent a qualitative improvement over the former national origins approach. For students of ethnicity its importance can scarcely be overstated. However, it is also critical to recognize its limitations. The outcome provides us with only a first approximation, one which lacks either context or nuance. Two persons responding that they are Basque may be invoking entirely different levels of awareness and personal commitment. A man born in rural Bizkaia with recent American citizenship obviously means something quite different by “Basque” than does a third-generation American with one Basque grandparent. Thus, the census data represent a good point of departure for students of immigration and ethnicity, but should never be confused with any kind of final statement regarding the status and nature of Basque ethnicity in the United States.


Copyright © 2000 the Center for Basque Studies, University of Nevada, Reno. All rights reserved. Updated 20 September 2000. E-mail: