10.9 Sociolinguistic correlations: Ethnicity

Like gender, ethnicity is socioculturally and sociolinguistically complex. Language and ethnicity are intricately linked and often co-constitutive. That is to say, each is often circularly defined; divisions between languages are often defined with reference to divisions between cohesive cultural groups that use those languages and ethnic groups are often defined with respect to the language that the group uses (e.g., think about how Czech and Slovak are mutually intelligible but understood as distinct languages, spoken by distinct ethnic groups). Around the world, people tend to live in close proximity to other members of their ethnic group. This is true both in places where that ethnic group is indigenous or in contexts of colonialism and diasporic migrations. This means that people often – but certainly not always – have social networks that are ethnically homogeneous. The linguistic consequence of this is that, because we tend to use language in the same way as the people we interact with most, ethnolects of many languages have emerged. Ethnolects are varieties of an ambient (standardized) community language used by a minoritized ethnic group. That’s not to say that the ambient standard doesn’t also have ethnoracial associations though! In North America for example, while people tend to assume that the ambient standards of Canadian English, Quebecois French, American English, and ASL are ethnically-neutral, they are ideologically associated with whiteness and European settlers.

I don’t have the space to dive into the complex intersections of language, ethnicity, race, and prestige but what I want to do is demonstrate the importance of linguistic variation with respect to ethnolects. The Lumbee Tribe of North Carolina is an Indigenous group in the United States and in fact, with 45 000 members, the Lumbee Tribe is the largest Indigenous group that lives east of the Mississippi River. The majority of Lumbee people live in Robeson County, North Carolina, a multiethnic area: 40% of residents are Lumbee, 35% are Anglo-American, and 25% are African American (Wolfram, Daugherty, Cullinan 2014).  These three groups, though living in close proximity, each live mostly selectively-segregated within the county. The Lumbees’ political situation is thorny; they have state-recognized Indigenous status in North Carolina but are not federally recognized with formal tribal status by the US Government. For almost a century and a half, the Lumbees have been unsuccessfully petitioning for full federal recognition. A major roadblock for the Lumbees’ petition has been their language history. Knowing one’s ancestral language is a key piece in demonstrating descent and is key to federal recognition. However, the Lumbee people’s ancestral language was taken from them very early on in the settler-colonial history of North America (they were documented as speaking English as early as 1730!). In an effort to combat this mitigating factor, a group of sociolinguists have been documenting the uniqueness and time depth of the Lumbee English ethnolect to help provide evidence of Lumbee Tribe’s culturally distinctive language that will satisfy the settler-colonial state.

Many of the lexical, phonological, and morphosyntactic characteristics of Lumbee English are shared by their Anglo-American and African American neighbours, or by nearby Appalachian English or North Carolina’s Outer Banks English. However, Lumbee English is composed of a unique mix. For example, the /aɪ/ phoneme in Lumbee English is raised and backed to [ɑ̝ɪ], something shared with Outer Banks English but none of the others; while Lumbee English’s ‘for to’ complementizer (e.g., I want for to get it) is shared with Appalachian English, but no others; and, Finite be (e.g., she bes there) is only shared with their Anglo-American neighbours. There are also a few features that are completely unique and point to the Lumbees’ long history of use of English. One of these is perfective be. Until the mid-seventeenth century or so, standard English exhibited a categorical alteration between be and have as markers of perfect aspect depending on the main verb of the sentence. Eventually, in most varieties of English, perfective be was lost, but in German, a language closely related to English, the alternation still happens, as in (6).

In Lumbee English, like in German, many main verbs retain perfective be both in present and past tenses as in the examples in (7) from Wolfram (1996: 9) and Dannenberg (1999: 67).

While perfective be in the present tense is shared with a small handful of other isolated English-speaking communities (none near Robeson County), perfective be in the past tense appears to be completely unique to Lumbee English. The presence of perfective be in Lumbee English suggests that the language has been in use within the community for a very long time: at least as long ago as when perfective be was more generally common among English speakers.

Lastly, several linguistic variables in Lumbee English demonstrate further evidence of its uniqueness. Both consonant cluster reduction and was/were levelling are variables shared by Lumbee English and their African American and Anglo-American neighbours. But for both variables, Lumbee English exhibits a unique pattern of conditioning factors.

Variable consonant cluster reduction involves variation between complex codas and reduced codas. So a word like disk may variably be realized as [dɪsk] or [dɪs] and a word like grilled might be realized as [ɡɹɪld] or [ɡɹɪl]. There are two important linguistic conditioning factors that correlate with this variation. First is the morphological complexity of the word: is the word monomorphemic like disk and mist or bimorphemic like grill-ed and miss-ed? Second is the following segment: is the next sound a vowel, a consonant, or a pause? The three varieties spoken in Robeson County differ in terms of how these factors interact and which variant is favoured in specific contexts. For example, Lumbee English patterns with African American English and differs from Anglo-American English in one way: in both Lumbee English and African American English the reduced variant is less likely to occur in monomorphemic words that appear before vowels like in rest [ɹɛst] easy. At the same time, Lumbee English patterns with Anglo-American English and differs from African American English in another way: in Lumbee English and Anglo-American English, the reduced variant is more likely in bimorphemic words that come before pauses like she’s blessed [blɛs] (Torbet 2001: 381).

Like with consonant cluster reduction, was/were levelling, the variable realization of were as was, as in we were/was and they weren’t/wasn’t, is shared by everyone in Robeson County. However, only in Lumbee English does the polarity of the sentence constrain the variation (such that affirmative sentences favour levelling to was and negative sentences disfavour levelling) (Wolfram and Sellers 1999: 103).

The evidence for the uniqueness and long history of the Lumbee ethnolect of English is strong. If a unique ancestral language is a requirement for purposes of federal recognition by the US Government, it seems that Lumbee English should qualify as such. Unfortunately, although in November 2020 the Lumbee Recognition Act of 2019 was passed in the House of Representatives it failed to pass in the Senate. However, in April 2021, a bipartisan group of lawmakers from North Carolina introduced new bills to try again.

References