EDGAR W. SCHNEIDER
American English is an Inner Circle variety (Kachru 1985) and one of two major “reference accents” of global English; as such, it has been a relevant but not a prominent topic in the field of World Englishes, which is more concerned with Outer Circle and Expanding Circle varieties. However, viewing it in this perspective definitely makes sense, given that centuries ago American English began as the first of Britain’s colonial (and later postcolonial) offspring, and it went through the same process of linguistic and cultural appropriation that has shaped other postcolonial varieties and has been described in the “dynamic model of Postcolonial Englishes” (Schneider 2003, 2007) – it is also a product of the colonial expansion of the British Empire in much the same way as the Englishes of, say, Australia, New Zealand, and South Africa. In comparison with these and other world Englishes, a longer time depth in association with sociopolitical developments is responsible for its character as a more influential and “stable” variety which by now has completed the entire developmental cycle of emerging varieties.1 On the other hand, a history of in‐migration has contributed to a blurring of the distinction between L1 and L2 varieties and the importance of effects of language contact not that much different from Outer Circle and other Englishes. More than others Mufwene (1996, 2001) has emphasized the fact that (white) American English has been shaped by language contact and essentially the same processes as African‐American English and other “disenfranchised Englishes” (2001: 106).
The distinctive nature and the varieties of English in North America are a product of the continent’s settlement history, with individual accents and dialects having resulted from unique mixtures of settlers from different regions of the British Isles and elsewhere and their ways of speaking.2
As is well known, the first English‐speaking permanent settlers founded the South Atlantic colonies (beginning with Jamestown, Virginia, in 1607) and New England (where the Mayflower landed the Pilgrim Fathers in 1620). Many of them were Puritans and came as religious dissenters, not because of poverty; their region of origin was primarily southern England. For generations these colonies maintained relatively strong political and cultural ties with their mother country, which is why the accents of New England and the South share relatively prominent linguistic features with southern British English, and to some extent with one another. Examples include the nonrealization of a postvocalic /r/, which in conservative New England and Southern accents is not pronounced in words like car, card, four, and fourth; the retention of /j/ in tune or new, or the “Boston a” in half and rather.3 From the original bridgeheads via urban hearths like Boston, MA, Richmond, VA, and then Charleston, SC, such accents took root in these regions, in accordance with Mufwene’s Founder Principle (Mufwene 2001). Eastern New England has continued this tradition largely to the present day: with important cultural centers and economic prosperity through trade, whaling and later early industrialization those who had established themselves there saw little reason to leave, so linguistically and culturally the region is somewhat different from the rest of the US. Similarly, a conservative and aristocratic plantation culture with a distinctive accent and culture established itself in the coastal South and expanded along the South Atlantic plains into Georgia. The downside of this culture was the infamous institution of slavery, with Africans having been forcibly brought to the region as early as in the late seventeenth century and, in large numbers, throughout the eighteenth and early nineteenth centuries.
Later waves of immigrants in the seventeenth century came through mid‐Atlantic ports, where the Quakers had established themselves in Pennsylvania, and their religious tolerance made the location attractive for many newcomers. Unlike the early wave, a majority of them came from northern and western England, Scotland, and also Ireland, and they tended to be of less affluent origins. Hence, very broadly it can be stated that a mixture of the working‐class speech from these regions constituted the basis of colonial mid‐Atlantic American speech, which later, after the colonial period, became the basis for the mainstream, inland‐northern and western type of American English.
When eighteenth‐century immigrants found the best lands along the coast taken, and hostile Indians and the earlier presence of the French prevented straight westward movement, settlements spread with a strong southwestern bend into the Great Valley of the Appalachian mountains. Many of these settlers were so‐called Ulster Scots, labeled Scotch‐Irish in the US, who found the landscape, climate and economic possibilities in the mountains familiar and favorable and thus rooted their culture and language features there (with linguistic traces like “positive anymore” to be still observed in the region today; see Montgomery 2006).
The 1803 Louisiana purchase, followed by the Lewis and Clark expedition, ultimately opened the inland and western parts of the continent for westward expansion and the continuous spread of the region settled by British and European immigrants. A deplorable consequence of this process was the cruel fate of the Native American population, who were continuously driven out of their homelands, decimated, and relocated forcefully. The Great Lakes Area and the Upper Mississippi region were settled predominantly by people from the inland northern parts of the original colonies, from western New England and upstate New York. Throughout much of the nineteenth century new lands further west were being taken, a process advanced by historical events like the building of the transcontinental railroad, the California gold rush of 1848–1849, or the admission of Texas to the Union in 1845.
Linguistically, the opening up of the Midwest and West can be characterized as a continuous and increasing process of mixing and blending of people with different regional origins and of the accents they brought with them. Dialect contact resulted in koinéization, the emergence of a middle‐of‐the‐road variety in which extreme dialectal forms (which, being used by only a minority, were communicatively inefficient) tended to be rubbed off, so American English has frequently been perceived as surprisingly homogeneous – a view which, however, may also be challenged. It is true that along the east coast, dialect differences between the various regions are strongest, and the further west we move the less conspicuous speech differences become. On the other hand, scholarship has shown and speakers know that even in the West there are significantly different regional and local speechways.
Schneider’s (2007: 251–308) extensive survey of the history of American English in the light of the “dynamic model” suggests that the “foundation phase” began with the earliest settlements and was followed by the extended “exonormative stabilization” period when the earliest colonies were established and the influential early colonial cities founded (after c. 1670, the foundation year of Charleston, SC). The third phase of nativization, with the variety beginning to distinctly go its own ways, was triggered by the independence movement in the mid‐1770s, and gave way to “endonormative stabilization” with the end of the independence wars and the beginnings of the westward expansion, lasting throughout the nineteenth century. After 1898, the year of the Spanish‐American War, when the US entered the world scene and began to grow into its superpower role, the country’s stability allowed the onset of phase 5, “differentiation,” with its recent emphasis on distinct regional and ethnic dialects as group identity markers.
The early American settlers were faced with radically new experiences and objects, and to meet the need to designate these they either borrowed or coined new words. By the eighteenth century such “Americanisms” abounded, and lexicographers, most notably the patriotic Noah Webster, began to record and emphasize the lexical distinctiveness of American English – it is interesting to see that this “linguistic declaration of independence” followed the political separation of the United States from her British mother country. Webster’s influence, in his famous “blue‐backed speller” (The American Spelling Book, first published in 1783), of which during the nineteenth century 100 million copies were sold, and then in his monumental 1828 American Dictionary of the English Language contributed substantially to an awareness and the solidification of such lexical differences, and so for a long time the search for and documentation of Americanisms remained an essential component of the scholarly study of American English. Two mid‐twentieth‐century scholarly dictionaries epitomize these activities: Craigie and Hulbert (1938–1944) document the American vocabulary, understood broadly as things American including British survivals associated with American culture, in the philological fashion of the OED, while Mathews (1951) narrowed his definition of Americanisms to words of American origin only.
Dialect words have been the second major object of American lexicography. The American Dialect Society, founded in 1899, pursued the explicit goal of supporting the compilation of an American Dialect Dictionary equivalent to Joseph Wright’s English work, and the realization of that goal was seen in the second half of last century. Directed originally by the late Frederic G. Cassidy, the monumental Dictionary of American Regional English (DARE) project, based upon both a reading program along OED lines and a 50‐state lexical dialect survey, now provides systematic coverage of words and expressions which are not in general use in the US, in that they are restricted to certain regions or ethnic groups (Cassidy, Hall, von Schneidemesser et al. 1985–2013). Another landmark of regional lexicography, focusing on the conservative southeastern mountain dialect region, is Montgomery and Hall (2004).
Building upon earlier European dialect atlas models, in the late 1920s an initiative was launched to systematically collect data for a projected “Linguistic Atlas of the United States and Canada,” to be directed by Hans Kurath. Because of the vastness of the region and the limitation of resources this project has never materialized as such but was broken down into a series of smaller, regional Linguistic Atlas projects. Methodologically, trained interviewers selected representative informants from regionally scattered localities and recorded their responses to a predetermined questionnaire of several hundred phonological, lexical, and morphological questions in fine phonetic notation, so that in the end millions of individual responses were put together as maps or lists (see Atwood 1963). By the end of the 1930s Kurath finished and published the Linguistic Atlas of New England (LANE), the model project for many to follow, and organized field work along the entire east coast for the Linguistic Atlas of the Middle and South Atlantic States (LAMSAS), a project whose data have been computerized and are still being analyzed by means of sophisticated statistical methods (Kretzschmar & Schneider 1996 and Kretzschmar 2009, as well as recent work by Kretzschmar and others). A series of similar projects followed, to cover almost the entire continent (Kretzschmar 2005). The most recent, and in many ways most modern (using audio recording and computerization technology from the outset) addition is the Linguistic Atlas of the Gulf States (LAGS; Pederson 1986–1991), which details the South, the most distinctive dialect region of the US.
Based upon lexical data from LANE and LAMSAS, Kurath (1949) postulated the now classic regional division of American dialects into three main dialect regions (North – Midland – South), with several subregions and the general proviso that the distinction is likely to get weaker or disappear the further west one moves. Atwood (1953) and Kurath and McDavid (1961) found this division confirmed on the basis of morphological and phonological data, respectively. Using lexical data from DARE, Carver (1987) was the first and only author so far to challenge this threefold division, arguing instead for a binary distinction into North and South only. However, on closer investigation the differences between both areal classifications are minor, essentially a matter of categorization and conceptualization: Kurath had observed “North Midland” and “South Midland” subdivisions which in Carver’s book resurface as “Lower North” and “Upper South,” respectively. Essentially, it seems clear that in terms of regional dialects American English shows two core areas, the North and the South, and a broad transition band in between.
William Labov’s classic study of New York City pronunciation (1966) and other work from that period (Labov 1972) founded a new subdiscipline of linguistics, the systematic study of sociolinguistic variation and change. Like dialectologists before him (who had already sampled speakers from different social strata), Labov’s goal was to study the down‐to‐earth intricacies of real‐life speech, but he was more interested in the social dimension of speech variability and in the theoretical modeling of why languages vary and how this affects language change (Labov 1994–2010). He developed new methods and concepts to reach these goals: the tape‐recorded “sociolinguistic interview,” with free conversation meant to stimulate informants to converse freely and without much effect of the “observer’s paradox,” in which then the realizations (“variants”) of predetermined variables are looked for and interpreted, using quantifying methodology. Typically, the frequency of certain variants is correlated with dimensions like social class, gender, age, and also style. Adopting and developing this methodology, sociolinguists such as Labov, Walt Wolfram, Guy Bailey, and many others have since investigated numerous communities across the US, usually interpreting a limited number of variables in the light of specific hypotheses of language variation and change.4
Labov and his followers detected and investigated a vigorously ongoing sound change, the “Northern Cities Shift,” broadly to be characterized as a clockwise rotation of the short (checked) vowels, which is far advanced among young speakers in many inland‐northern urban areas (Labov 1994: 177–201). They carried out a new and large‐scale dialect survey project of the entire US known as the “Telsur” (“telephone survey”) project with the aim of documenting regional sound systems and sound changes on a broad, national basis. The result of this is the phonological Atlas of North American English (Labov, Ash, & Boberg 2006), a multimedia product which thoroughly analyzes and exemplifies an immense number of audio data from across the US. Condensing this wealth of information into a new regional division of American English, Labov basically confirms Kurath’s three main areas (with the South expanding more widely into the Midlands than previously assumed) and adds a fourth one, the West. He finds that while the North, the South, and the West have fairly homogeneous vowel systems and patterns of change, the Midland is characterized by extreme diversity, a residual region where individual cities have developed dialect patterns of their own (cf. Murray & Simon 2006).
Typically American English is seen as against British English, and distinguishing features on the levels of phonology, lexis, orthography, and grammar tend to be juxtaposed in list form in textbooks. For example, American versus British choices are reported to include the lexical items gas (vs. petrol), fall (vs. autumn), railroad (vs. railway), etc.; the pronunciations /æ/ (vs. /ɑː/) in dance, grass, or can’t, unrounded /ɑ/ (vs. /ɒ/) in lot or dollar, and postvocalic /‐r/ in car, card, and so on; on the grammatical level, have (vs. have got) for possession, will (vs. shall) for first‐person future reference, and a more liberal use of the past (for the present perfect) tense; and spellings like theater, honor, recognize, and plow (vs. theatre, honour, recognise, plough). Much of this requires qualification and a more careful phrasing, however: not infrequently “American” words or pronunciations exist in Britain as well but are constrained to the status of regional dialect forms, stylistically marked choices, or slightly different usage conditions.5 American innovations are being adopted in British speech as well.
Thus, it is necessary to look into dialects: American English is anything but homogeneous – the notion encompasses not only a rich array of regional forms and some social variation but also, and increasingly so, ethnic varieties shaped by effects of language contact and differential degrees of integration of generations of immigrants into the American mainstream culture.6
Regional dialect differences primarily depend upon different pronunciation patterns and lexical choices. Obviously, the spread of individual forms varies from the strictly local to elements which set off larger dialect regions from adjacent ones. Linguistic atlas data and publications and many other sources provide ample illustration of such variants; for reasons of space I restrict myself to pointing out some of the best‐known characteristics of three large regions. The inland northern region, extending westward from western New England into the Great Lakes area, comes closest to an “unmarked” accent globally perceived as “typically American.” The Midlands are essentially a transition region with a small number of features of their own and an increasing number of northern or southern features the further one progresses in the respective direction.
New England pronunciation is most strongly characterized by the lack of a postvocalic /r/ and by a low [a] in words like bath, glass, or aunt (known popularly as the “Boston a”). Conservative dialects from the eastern part of the region maintain a distinction between the vowels in Mary ([e:]), merry ([ɛ]), and marry ([æ]). Lexical items characteristic of the region include pail ‘bucket’, darning needle ‘dragonfly’, angleworm ‘earthworm’, grinder ‘submarine sandwich’, and rotary ‘traffic circle.’
Southern English, the topic of much recent research (e.g. Nagle & Sanders 2003), is clearly the most distinctive of all American dialects, also a product of a strong regional identity. Well‐known features include the so‐called “Southern drawl” (a lengthening and breaking tendency of vowels, as in [ɪə] in bit)7, lack of rhoticity (now recessive), the monophthongization of /aɪ/ (e.g. time [ta:m]; generally before voiced consonants and in free position, with regional and social restrictions before voiceless consonants), homophony of mid and high front vowels before nasals (known as pin/pen‐merger), the second‐person plural pronoun y’all, double modals like might could, an inceptive future fixin’ to, and words such as light bread, pulley bone ‘wishbone’, mosquito hawk ‘dragonfly’, granny woman ‘midwife’, or jackleg ‘unprofessional, dishonest’. It is interesting to see that some traditional features of Southern English are now being given up while new regional shibboleths are emerging. Bailey (1997) claimed that Southern English originated as late as the post‐Reconstruction period after the loss of the Civil War, as a deliberate means of strengthening Southerners' regional identity against outside political dominance – a hypothesis which was recently challenged, or at least modified, by Montgomery, Ellis, and Cooper (2014).
English as spoken in the West lacks salient characteristics but is regarded as prestigious nationwide. The low back vowels of lot and thought are merged, and high back vowels as in goose or foot are frequently fronted. Younger California speakers tend to lower their lax front vowels (so six sounds like sex, sex like sax, and sax like socks; cf. Gordon 2004). Regional words include borrowings like canyon or corral and coinages like parking strip or chippie ‘woman considered to have loose morals’ (Carver 1987).
Numerous sociolinguistic studies from many locations, urban and rural, have yielded insights into some principles governing speech variability and have identified a few robust distributional tendencies. Obviously, the familiar pyramid‐shape of dialectal variation applies: the higher a speaker’s social status, and the more formal a speech situation, the less likely dialectal forms are, and vice versa. Women have widely been found to be leading in linguistic changes, that is, to adopt and spread linguistic innovations more rapidly than males. While the use of regional words carries no stigma and certain traces of regional accents are acceptable also among upper‐class speakers (consider recent US presidents from the South), nonstandard grammatical phenomena (like multiple negation, the use of ain’t or preverbal done, nonconcord copula forms, or nonstandard relativization) are socially stigmatized but hardly regionally diagnostic.
Recent research has tended to emphasize the indexicality of linguistic, notably sociophonetic, choices, that is, the fact that by speaking in a certain way speakers actively express their identities and their social affiliations (Eckert 2000)
Immigration has continued to shape the linguistic landscape of the US, and many ethnic varieties are products of language contact, frequently involving language shift on the side of a minority group from an erstwhile ethnic language to the dominant one, English – modifying the latter in this process. The best‐known case in point, African‐American English (AAE), is discussed in Chapter 18 of this volume (see also Lanehart, Green, & Bloomquist 2015). Ethnic variation also raises a number of questions concerning conflicting identities, varying discourse conventions, or intercultural communication (Fought 2006).
Relatively little attention has been devoted to the English of Native Americans, which varies from speech with no discernible “accent” to contact varieties (cf. Leap 1993; Rowicka 2005). Distinctive features seem to lie less in transfer of phonology or grammar (possibly with the exception of some special patterns of tense use) than on the pragmatic level (expressions of respect and politeness, discourse organization, etc.). Lumbee English in North Carolina has been shown to feature distinctive vocabulary (e.g. ellick ‘coffee’, sorry in the world ‘badly’) and grammar (finite be, as in She bes there, and I’m for I’ve, as in I’m been there) (Wolfram, Dannenberg, Knick, & Oxendine 2002).
Demographic changes and migration effects give special prominence to Hispanic varieties of English. Some work has been done on Puerto Ricans in New York City and very little on Cuban immigrants in Miami, while the “Chicano English” of descendants of Mexican immigrants is fairly well researched (Fought 2003; Santa Ana & Bayley 2004). Characteristic features include some aspects of pronunciation (e.g. strongly monophthongal vowels) and several prosodic phenomena (e.g. a different system of vowel reduction and distinctive intonation contours).
Cajun English is spoken in Louisiana, predominantly by younger speakers who, two generations after the language shift from French to English, sense a loss of their cultural heritage and have fueled a “Cajun Renaissance.” Features include high rates of final consonant deletion (not only in clusters), the monophthongization of diphthongs, lack of aspiration in word‐initial stops, and “heavy nasalization,” also of consonants (Dubois & Horvath 2004).
Further linguistic research would also be required concerning the linguistic integration of Asian immigrant groups. Except for some work on Vietnamese English, hardly anything has been done in that area.
Due to the relatively strong degree of mixing, mutual accommodation, and koinéization that occurred during the colonial period and even more strongly in the phase of westward expansion, American English has traditionally been perceived as relatively homogeneous, at least in comparison with British dialects. Based on limited factual evidence, Krapp (1925) coined and the phonetician Kenyon disseminated the notion of “General American,” which became popular during the 1930s and can still be found cited in some sources today, to refer to a putatively homogeneous normative type of American English (in practice, it probably meant accents not distinctively New England or Southern). However, dialect geographers like Kurath, Atwood, and others strongly opposed this notion, arguing that there is no nationally uniform standard accent of American English and that on closer investigation American dialects show a great deal of phonetic, lexical and grammatical variability. This assessment is based on the voluminous atlas evidence and has been confirmed by works like Frazer (1993), which shows how much variability there exists even in the “Heartland,” a region where speakers believe that they “have no accent.”
Thus, in line with phase 5 as postulated by Schneider (2003, 2007) in the emergence of postcolonial Englishes, American English has transcended the stage of emphasizing homogeneity and proceeded to increasing diversification, both regional and social. In other words, not only culturally but also linguistically the traditional “melting pot” metaphor, assuming that immigrants have been assimilated to join a mainstream culture, is now giving way, if only gradually, to a “salad bowl” conceptualization, in which individual groups remain recognizable through the retention of ethnolinguistic characteristics. This becomes all the more apparent considering the “divergence hypothesis” of African‐American English (AAE) (e.g. Bailey & Maynor 1989) and comparable dissociating trends affecting other ethnic, regional, and social varieties. It is noteworthy that varieties as diverse as AAE, Chicano English, Cajun English, Southern English, and the “brogue” of Ocracoke, NC (Wolfram & Schilling‐Estes 1997) have all been stated to be products of recent strengthening processes of locally or ethnically based group identities.
The vast majority of Outer Circle world Englishes are products of British colonialism, and traditionally in these countries British English and RP used to be regarded as the linguistic norm and target of education. Only two such varieties are American derived, namely those of the Philippines and of Liberia. Today, however, an increasing impact of American English on practically all varieties of English around the globe can be observed, manifested in American‐influenced lexical choices or also in certain pronunciations.8 So far the evidence for this phenomenon is largely anecdotal, though the process is referred to repeatedly, and an increasing body of evidence is accumulating from various countries.
The reasons for this growing impact of American English are also underresearched, though it is possible to make plausible educated guesses. Clearly it results from the growing exposure to and the great prestige of American English. Prestige is of course associated with people, so this is a consequence of the dominant role the United States plays politically and economically in the global context; a certain ambivalence can be sensed here in many contexts (via the spread of American popular culture, the practice of adopting American ways of speaking is taken up by some who, presumably subconsciously, regard this as fashionable and symbolizing modernity, high status, and an international orientation, but it is resisted by others who fear a loss of local identities and traditions).9 Exposure reflects the global dominance of the American media and music industries, with Hollywood movies being shown and American TV serials being aired (frequently undubbed) on all continents, and it results from the modern facilities for travel and personal contact (tourism, business travel, also student exchange, and, increasingly so, the Internet).
Of course, the impact of American English on other world Englishes varies from one region to another and is difficult to generalize, but some broader statements can be made. Words travel easily, so the majority of new Americanisms used elsewhere are on the lexical level. Words which seem to be spreading widely and rapidly include gas, guy(s), Hi, movie, truck, Santa (Claus), and station wagon, and adolescent slang and fashion terms like man as a form of address, cool meaning “very good,” or the “new quotative” be like to introduce direct speech. To this may be added older words which have been internationalized so strongly that their American origin may no longer be recognized in many communities, like radio (for older British wireless), commute, fan, star, know‐how, break even, or let’s face it (Gordon & Deverson 1998: 112). As to pronunciation, rhoticity and “jod‐deletion” in words like new, tune are widely perceived as “American” and may be adopted for this effect; and for certain words putatively American pronunciations are getting more widespread, including research stressed on the first and primarily on the second syllable, schedule with /sk‐/, lieutenant with /lu:‐/, etc. The spelling center is clearly preferred over centre outside specifically British spheres of influence, and program rather than programme is also used widely, not only in computing contexts. On the level of syntax, hopefully used as a sentence adverbial and patterns like do you have seem to be diffusing from the US. American influence can even modify the meaning of words, as in the case of billion, which now means “a thousand million” rather than “a million million” even in Britain (Peters 2004: 72).
To refer to just a few more exemplary studies: For Australian English, Taylor (1989, 2001: 324–327) reports some examples and quotes reactions, including fairly emotional and hostile ones, to the perceived “American invasion” of Australian English. Similarly, for New Zealand English Gordon and Deverson (1998) document and discuss a wide array of Americanisms on different language levels, and divided reactions to them. Igboanusi (2003) quotes some examples of “an influx of Americanisms into Nigerian English” (603) and refers to other sources attesting aspects of this process in several countries. Trüb (2008) investigates the amount of American English impact on South African English systematically by eliciting phonological, grammatical, and lexical data from both older and younger speakers and finds a clear increase of the amount of American choices in the younger generation. Schneider (2011) documents varying degrees of preferences for American as opposed to British English options on the levels of lexis, phraseology, and grammar in several Asian Englishes. As a typical example from the Expanding Circle, Edwards (2016: 98) observes that young, highly fluent Dutch speakers of English choose American English much more frequently as a model than British English.
Hence, it appears that American English is enjoying covert prestige in many countries and communities where British English is promoted as the “official” target norm, also in education. Certainly this has to be taken with a grain of salt and is likely to be sociolinguistically conditioned (preferred among the young, in informal contexts, and in association with certain topics and domains), but the process seems widespread and robust. It deserves more intensive investigation and systematic documentation.
As the statements in this chapter have shown, American English is anything but homogeneous; rather, the linguistic landscape of North America displays the kaleidoscope of accents, dialects, and linguistic features associated with both national unity and distinct group identities, which characterizes many modern societies. Thus, in a global perspective, it should not be viewed monolithically, as one of two reference varieties as opposed to British English, but rather as a vibrant set of varieties itself, language forms which internally are associated with distinct sociocultural identities and which globally interact with other world Englishes.