Font Size: 
Indicator V-5 Multilingualism
NOTE TO READERS: Please include the following reference when citing data from this page: "American Academy of Arts and Sciences, Humanities Indicators,".
Back to Section V-A

Updated (4/30/12) with American Community Survey data for 2010 and General Social Survey data for 2006.

Because Americans’ multilingualism has implications not only for the nation’s ability to compete in a global marketplace but also for its capacity to develop and execute effective foreign policy, the extent to which Americans gain fluency in multiple languages is an important question. Such multilingualism is the focus of this indicator, which also looks at where proficiency in languages other than English is acquired. (For data concerning trends in foreign-language course-taking, see Indicator I-7, Language Course Enrollment in High Schools, and Indicator II-7, Postsecondary Course-Taking in Languages Other than English (OTE).)

In January of 2006, President George W. Bush launched the National Security Language Initiative, which was designed to “dramatically increase the number of Americans learning critical need foreign languages such as Arabic, Chinese [Mandarin], Russian, Hindi, Persian, and others through new and expanded programs from kindergarten through university and into the workforce.” 1 For all the present concern about what is perceived as a national foreign-language deficit, however, existing data on multilingualism are of limited use in gauging the true extent of the country’s achieved fluency in multiple languages. This is true for several reasons. First, such data are based on self-report—currently no system objectively measures and registers individuals’ multilingual capabilities. Second, the national trend data covering the greatest length of time, those collected by the U.S. Census Bureau, reflect a concern with immigrants’ ability to acquire English-language skills. Thus these data do not capture those individuals who have gained their proficiency in non-English language(s) via formal instruction; nor do they account for those who may have learned a non-English language in their childhood homes (and still speak it fluently) but who do not use that language in their own homes as adults. Moreover, Census Bureau data do not measure the extent of individuals’ proficiency in their non-English “home” language. Finally, data collected by the bureau and other organizations on this topic are structured to measure Americans’ proficiency in just one language other than English and thus do not reveal how many people have facility in three or more languages. Although these data cannot measure the full extent of the nation’s multilingualism, they are the best available and are presented below.

Figure V-5a displays relevant Census Bureau data. While a language question has appeared on almost every decennial census since 1870, only since 1980 have respondents been asked not only if they speak a language other than English but also how proficient they are in English. Between 1980 and 2010, the percentage of Americans age 18 or over who were multilingual, defined in this case as individuals who (1) report speaking a non-English language at home and (2) characterize themselves as speaking English “well” or “very well,” rose from 9.2% to 14.9%.

Figure V-5a, Full Size
Supporting Data Supporting Data

In 2010, the majority (55%) of multilingual Americans age 18 or older spoke Spanish or Spanish Creole in addition to English (Figure V-5b). Just under a quarter spoke another Indo-European language, and 16% spoke an Asian or Pacific Island language. Included among the remaining 5% of multilingual Americans whose second language is labeled “Other” in Figure V-5b were those individuals who spoke indigenous languages of North, Central, and South America; Semitic languages (including Arabic); and languages of Africa.

Figure V-5b, Full Size
Supporting Data Supporting Data

Data from the 2006 General Social Survey (GSS), administered by NORC at the University of Chicago, provide a somewhat different estimate of the extent of multilingualism in the United States. The focus of the GSS was respondents’ proficiency in a non-English language, and thus the survey did not include questions about their proficiency in English. An advantage of GSS data, however, is that they capture the two key groups missed by the Census Bureau: those individuals who learned a non-English language outside the home and also those who learned the language at home as children but who now, while still fluent in the non-English language, speak only English in their own homes. The GSS data also reveal where those Americans who speak at least one language in addition to English developed their proficiency in the non-English language.

The GSS indicates that approximately 15.7% of those proficient enough in English to complete the survey in that language described themselves as speaking another language “well” or “very well” (Figure V-5c). Only 2.6% of English-speaking respondents spoke another language proficiently and had acquired that proficiency in school. Most speakers of English who described themselves as proficient in another language had learned that language at home.

Figure V-5c, Full Size
Supporting Data Supporting Data

1 U.S. Department of State, “National Security Language Initiative,” Fact Sheet (January 5, 2006), The other languages deemed “critical” are Azeri, Bengali, Gujarati, Japanese, Korean, Marathi, Pashto, Punjabi, Tajik, Turkish, Urdu, and Uzbek.
Back to Top

Skip Navigation Links.  

View figures and graphics: