The diversity of the Latinx population and the concentration of national origin groups in different parts of the country also point to the diversity of varieties of Spanish spoken within the US. The make-up of local Latinx populations shapes the Spanish spoken in different areas of the country, such that the Spanish one hears in El Paso, Texas is more likely to sound like Mexican varieties, whereas the Spanish heard in Washington Heights, New York is more likely to sound Dominican (see Chapter 10 for a discussion of some of these linguistic characteristics). Still, this is not to say that the Spanish spoken in these areas is just like what you might hear in Mexico or the Dominican Republic; not only do all language varieties change and develop in different ways over time, but contact with English, and with other varieties of Spanish, also has an impact on how Spanish is spoken in the US (see Chapter 10). Further, we want to stress that ‘Mexican Spanish’ and ‘Dominican Spanish’ are not monolithic, homogenous entities. Instead, there is significant geographic and social variation within nations and national origin groups (in Chapter 4 we discuss the ideologies linking language to nation).
Statistics on Language Ability and Use
In our overview of how many people speak Spanish at home we relied on official statistics from the Census Bureau. Before going further in our discussion of the patterns of language knowledge and use, let’s stop to examine where those statistics come from and how they are produced. In particular, we’ll describe the Census Bureau’s question about language and the kinds of language data that the Census Bureau does and doesn’t collect, and we’ll explain how the Census Bureau’s language question limits our knowledge about Spanish and Spanish-speakers in the US (Leeman, 2004, 2018c). In Chapter 5 we will explain and critically analyze the Census Bureau’s statistics on ethnoracial identity.
Since 1980, the Census Bureau has produced statistics about language using a three-part question which is asked of persons aged five and older (see Figure 2.2). The first part of the question asks whether the person speaks a language other than English at home. If the answer is yes, the next part of the question provides a write-in box to identify the non-English language, and the third part asks how well the person speaks English, with four response options: Very Well, Well, Not Well and Not at All. This question does not appear on the census itself (which is conducted every ten years), but rather on the Census Bureau’s American Community Survey (ACS). The ACS is an annual sample-based survey that collects a wide array of social, economic and other data from US households. The language data are used in decisions about where voting materials in Spanish and other minority languages are needed as well as in the implementation of other language policies (see Chapter 8).
Figure 2.2 American Community Survey language question (2015)
There are several aspects of this language question, and the resulting statistics, that are worth noting. First, the ACS asks only about language use, not language knowledge, and only about home use. Thus, if a person knows another language but doesn’t speak it at home, this isn’t recorded by the Census Bureau. For this reason, the actual number of people who know Spanish is certainly higher than the 41 million reported by the Census Bureau, since many people who know Spanish don’t speak it at home (Leeman, 2004). Secondly, the question only provides one write-in box for a non-English language. As a result, there is no way of knowing whether respondents speak more than one non-English language at home. The proportion of immigrants from Latin America, and especially Central America, who speak indigenous languages (such as K’iche’, Mixtec and Nahuatl) is growing (Bazo Vienrich, 2018). Some such individuals also speak Spanish, but others don’t. However, because the ACS only provides space for one non-English language, multilingual individuals who answer that they speak an indigenous language (such as Aymara, K’iche’, Mixtec, Nahuatl or Quechua) are omitted from counts of people who speak Spanish. On the other hand, if they answer that they speak Spanish, their use of indigenous languages goes unrecorded.
The third and fourth points we want to highlight are related to the English-speaking ability question. Rather than an objective assessment, the question relies on the respondents’ subjective judgment about how well the person speaks English. Obviously, this raises some concern about the reliability and validity of the results, since people differ in the criteria they use, and notions about ‘good’ and ‘bad’ ways of speaking might impact responses (Leeman, 2015). On the other hand, it clearly wouldn’t be feasible or desirable to give everyone a language exam! The fourth aspect of the ACS language question that we want to mention is that it doesn’t ask about ability in the non-English language. Thus, while the question allows for classification people based on four levels of English-speaking ability, there is no way to know about their ability in Spanish (or any other non-English language). Some people who speak Spanish at home are completely fluent, whereas others have more limited ability, but there is no way to tell them apart in the ACS data (Leeman, 2004, 2018c).
A final limitation of the ACS is that, because it is based on a sample (rather than on the entire population, like the census), it isn’t possible to calculate statistics for small areas. Thus, we can only use ACS data to compare the percentage of people who speak Spanish in different states or in large cities, but not in less populated counties, smaller cities or neighborhoods.
The reason that the Census Bureau’s language questions are the way they are is related both to language policies and to language ideologies (Leeman, 2004, 2018c). As we discuss in Chapter 8, there are a few US language policies that require services in minority languages but these are designed for people with limited speaking ability in English (Gilman, 2011). In some cases (e.g. the Voting Rights Act), the requirement depends on how many such people live in a given area, and ACS statistics are used to make that determination. In contrast, there aren’t any policies that require statistics about ability in non-English languages (Leeman, 2018c). Further, dominant language ideologies don’t see minority languages as a particularly interesting social characteristic (Leeman, 2004), something we will delve into further in Chapter 4. For this reason, while the ACS provides invaluable statistics on home language use and English-speaking ability, corresponding statistics about knowledge of Spanish must come from other, non-governmental sources.
The English-speaking Ability of Spanish-speakers and Latinxs
One thing that the ACS statistics do allow us to determine is the English-speaking ability of people who report speaking Spanish at home, and thus to evaluate the oft-repeated claim that Spanish-speaking immigrants and their offspring don’t learn English. The supposed failure to learn English, which is often framed as a refusal or unwillingness to do so, is a common trope on social media and in online comments, letters to the editor and everyday discourse, one that is often taken for granted or presented as fact without any evidence. For example, in a recent appearance on Meet the Press, former NBC news anchor Tom Brokaw answered a question about the high levels of support among residents of Wyoming and South Dakota for building a wall on the US–Mexico border as follows:
And a lot of this, we don’t want to talk about. But the fact is, on the Republican side, a lot of people see the rise of an extraordinary, important, new constituent in American politics, Hispanics, who will come here and all be Democrats. Also, I hear, when I push people a little harder, ‘Well, I don’t know whether I want brown grandbabies.’ I mean, that’s also a part of it. It’s the intermarriage that is going on and the cultures that are conflicting with each other. I also happen to believe that the Hispanics should work harder at assimilation. That’s one of the things I’ve been saying for a long time. You know, they ought not to be just codified [sic] in their communities but make sure that all their kids are learning to speak English, and that they feel comfortable in the communities. And that’s going to take outreach on both sides, frankly. (NBC News, 2019)
There is a lot to unpack in Brokaw’s statement, including his acknowledgement that support for a border wall is rooted