Possible classifications of English accents

In what follows, a few possible classifications of English accents are sketched out. As the first set of chapters (Chapters 2–6), that is, the chapters introducing the major international varieties of English, are either exclusively or primarily concerned with pronunciation, we only consider pronunciation-based relationships here. This means that differences of grammar, and subtypes of English created by such differences, are ignored.

However, it is true for all (linguistic) classifications that they are usually based on connections and groupings that result from a historical chain of events: a modification to the linguistic system (e.g., a sound change) which starts in a certain geographical or social sphere is unable to affect the whole of the rest of the population, its spread is incomplete, and therefore a division or gap is brought about between two speech communities. This is illustrated by several examples below.

1.2.3 POSSIBLECLASSIFICATIONSOF ENGLISHACCENTS

The following figure (inspired by Figure 1.1 in Trudgill – Hannah 1982:

5) shows pronunciation-based relationships among the major English-speaking geographical areas. It uses eleven factors, that is, eleven pro-nunciation features which only characterize a certain subset of the accents spoken at those localities.¹ For example, line a marks the deletion of non-prevocalic [r] (called R-dropping) in England, Wales*, South Africa, Aus-tralia and New Zealand. (See below and Section 2.1 for more detail.) The BATH-broadening line (line b) separates accents which pronounce "broad"

[] (rather than "flat" []) in a group of words containing bath, can't, dance, last, etc., frequently referred to as the BATH-words (see Section 2.1). The tapping/flapping line (line e) indicates which standard varieties are characterized by the voicing (i.e., tapping or flapping) of intervocalic [t], i.e., which are the so-called tapping dialects/accents, in which certain [t]'s turn into the so-called tap or flap ([]), pronounced with a single quick tapping movement of the tongue (like the r in Hungarian perec 'pretzel'). (As a matter of fact, tapping is found in most non-standard accents, but it has found its way into the standard in Canada and the US only – cf. Section 4.2.)

a: R-dropping, b: BATH-broadening c: LOT Unrounding, e: tapping/flapping d: COT=CAUGHT

f: syllabic [r] in BIRD

g: PULL=POOL, h: WHICH≠WITCH i: TRAP=DRESS

j: front [a] in BATH k: THREW≠THROUGH

1 Keep in mind that the countries included in the figure are large enough to have internal variation – the figure only represents the majority accents. E.g., as we will see presently, the R-dropping line (a) could in fact cut both England and the US into two.

1.2.3 POSSIBLECLASSIFICATIONSOF ENGLISHACCENTS

Note that BATH-broadening, R-dropping, tapping, and the like are sound changes, innovations which arose in a particular segment of the English-speaking community, spread to its other segments but did not spread to all of them – thus dialectal differences have been created. The sound changes and phenomena mentioned in the figure will be treated in more detail in later chapters.

On the whole, the figure reveals that it is possible to distinguish between two main types of English accent: an "English" type (English English, Welsh English, South African English, Australian English, New Zealand English) and an "American" type (US English, Canadian English), with Scottish and Irish English located somewhere between the two, perhaps forming a separate category.

Note, however, that national boundaries do not necessarily coincide with the borderlines between speech communities, as has been mentioned in footnote 1 above. Even within one country, different segments of the population may exhibit different behaviours. To illustrate this, let us consider the issue of R-dropping in more detail. As we have seen above, in certain accents of English (parts of England, Wales*, South Africa, Australia and New Zealand) the sound [r] is only found before a vowel in pronunciation (e.g., in the word red) – all other orthographic* (= spelt / written) r's are deleted, i.e., dropped (e.g., in care, hard). Such accents are called non-rhotic*, while accents with all orthographic* r's pronounced are called rhotic*.

The chart below shows rhoticity in the accents of English. It is based on Harris' system (1994: Chapter 5), which distinguishes four subtypes (dubbed type A, B, C, and D – see column 2) and Table 2.2 in Trudgill – Hannah (1982: 15). It also includes the traditional terminology of the rhotic/non-rhotic* bifurcation of accents (column 1). Column 3 lists the major accents which need be distinguished, and columns 4 to 8 specify the phonological environments relevant to the discussion: a "yes" means a pronounced [r] in the given position, a "no" means R-dropping. The dash ( – ) stands for the [r] in question, the closing square bracket ( ] ) indicates a morpheme boundary, the vertical line ( | ) indicates a pause, "Á" is a stressed vowel,

"A" is an unstressed vowel, "V" is short for (any) vowel, and "C" is short for consonant. Finally, the braces ( { ... } ) enclose objects in a disjunctive (either/or) relation. Columns 7 and 8 distinguish between two cross-morpheme cases, one is called Linking-R (cross-morpheme-final [r] pronounced)

1.2.3 POSSIBLECLASSIFICATIONSOF ENGLISHACCENTS

and Intrusive-R (non-historical [r] inserted). (For more explanation and more examples, see the next chapter.)

1 2 3 4 5 6 7 8

B RP yes yes no yes no/variable

2

USEng yes yes variable yes variable

5 AusEng yes yes no yes yes

6 NZEng yes yes no yes yes

rhotic A ScotEng yes yes yes yes no

10 IrEng yes yes yes yes no

11 CanEng yes yes yes yes no

12 Mid-West

USEng yes yes yes yes no

(Abbreviations of the accents: RP = Received Pronunciation, SouthEngEng = Southern English English, USEng = US English, AusEng = Australian English, NZEng = New Zealand English, SAfEng = South African English, ScotEng = Scottish English, IrEng = Irish English, CanEng = Canadian English)

There are several conclusions to be drawn from the chart. First, the traditional rhotic/non-rhotic* distinction is not fine-tuned enough: notice that the non-rhotic label actually subsumes three categories. Second, certain geographical areas, e.g, England or the US, exhibit such a degree of internal variation that they need to be broken down into sub-classes or sub-regions.

In the case of England, RP is distinguished from Non-RP Southern English English (and in fact, a "Non-RP South-west" category may as well be

1.2.3 POSSIBLECLASSIFICATIONSOF ENGLISHACCENTS

squeezed into type A – see the next chapter). In the US, the situation is so complex that there are as many as four distinct sub-regions (East New England US English, New York City US English, South East US English, Mid-West US English – this final category represents GA). In addition, within New York City extensive internal variation has been recorded (basically, social class variation – see Labov* 2006 and Section 4.1).

In Chapters 2–6, a whole lot of pronunciation features will be introduced, all of which separate accent types in English in a similar fashion.