The region defined as the Southern United States by the US Census Bureau comprises a quarter of the country’s geographical area and contains a third of its population. It would be unreasonable to expect uniformity of speech over such a large area, and one can certainly find a great deal of variation within the region. On the other hand, attempts to draw distinct boundaries between one subregion and another are fraught with difficulties. Differences in race, class, education, and changes over time, bring in complexities that confound attempts to define accents in a purely geographic way. There are, however, generalizations that can be made about all southern accents, and at least one major subdivision between southern accent types, that will serve to locate actors as they begin to create the specific speech characterization appropriate to their character. In this handout we will build a picture of one major variety: the Non-Rhotic Southern accent. While rhoticity is certainly a salient feature in distinguishing between types of southern accent, the term “non-rhotic” is used here as a convenient label for a constellation of features, not limited to a lack of rhoticity, that make up a reasonably distinct variety of southern accent. In our description of this accent we do not make the claim that any feature is unique to this accent and there are, indeed, many features shared with other varieties. You may have noticed that I’ve left out a major cultural feature in my description of this accent, and that is race. There’s a lot to discuss in that one word, and we’ll be having that discussion, but for now we can say that the Non-Rhotic accents spoken by Black and White people in the South share many features. Starting here is intentional, but it’s just a start.
A good starting point for feeling the oral posture of this accent is to relax the jaw. Certainly there are numerous Southerners with tight jaws, but this first step will allow space for step two, the retraction of the tongue root, without introducing an unreasonable amount of tension.
It can also be a useful tool to begin with a strong posture of rhoticity – which will certainly involve tongue root retraction but will also recruit other muscles of articulation – and to gradually reduce that rhoticity until it is felt, but produces no definite sound of rhoticity.
Many speakers also hold slight tension in the buccinators muscles, and in the outer ring of the orbicularis oris.
Southern speech is commonly stereotyped as slower than other varieties of English, but this is not borne out by the studies that have set about to measure this feature. So what creates this impression? First there tends to be more variety in vowel length, making some vowels seem drawn out. The term “drawl” which we’ll discuss below refers to this increased length, as well as a tendency for lengthened vowels to change in shape as they proceed. In addition, Southern speech also tends to change pitch more slowly, making gently curving pitch contours. The range of pitch change is generally greater than in other American accents, but there aren’t as many abrupt changes as we might hear in Received Pronunciation (RP). All this taken together leads to an impression of unhurried, languorous flow.
There is also a tendency to shift primary stress to the first syllable of some words, such as: cement, police, hotel, pecan, July, Detroit, umbrella, tv, Thanksgiving, insurance. While this is something of a stereotype, and by no means universal, it can still be heard.
A Magnetic Center
The oral posture has a noticeable effect on STRUT and FOOT, drawing them closer together around a point on the vowel chart between the two.
STRUT → ə̝ ɘ̠
This phoneme is generally realized in more central position in American accents, and this tendency is particularly strong in Southern accents. A southern pronunciation of “blood” might be indistinguishable from an R.P. speaker’s “blurred”.
FOOT → ʊ̜̈ ɘ̹
A centralizing and unrounding of this sound brings FOOT and STRUT into a near merger. This is different from the merger in some UK accents, where the merged FOOT/STRUT sound remains rounded . It’s interesting to note that the historical movement of these words has followed this same centralizing course. Words like “food” “foot” and “flood” began in English with the same close, back, rounded vowel. The differences in the surrounding consonants “pulled” some of these words toward a more central and unrounded state. In the American South this drift has gone a bit farther.
Diphthongization, or breaking of monophthongs
English vowels vary considerably in duration depending on their context. For example, most of us will spend more time with the vowel of “beam” than we will with “beat”. We also vary length according to the stress we place on a word. What is distinctive in Southern accents is the way vowels change in shape throughout that duration. This is called a “drawl” and though this word can carry some negative connotation, it has a certain accuracy in that it describes the way the sound is drawn out. More than duration, though, it is the shifting of the vowel, turning a monophthong into a diphthong or triphthong, that produces the feature that we hear as particularly Southern. The same factors that influence length will affect the prominence of this feature.
Ongliding of Long Vowels
This is similar to a tendency in Cockney of initiating FLEECE and GOOSE from a lax starting point. The subtle differences in oral posture between the two accents can be distinctly heard here.
Also, this is the same process that led to the creation of the PRICE and MOUTH diphthongs during the Great Vowel Shift.
FLEECE → ɘ̆i
The vowel is first formed in a lax position, perhaps [ɘ] or [ɨ] and glides toward the tenser [i] as the vowel progresses. In word-final positions, or when followed by a voiced consonant, the effect is more noticeable. This feature also varies according to grammatical stress.
GOOSE → ɘ̆u
In the same way, this vowel is first formed in a lax position, arriving at[u] from a more central beginning point.
Who did you use to fool around with in school?
Southern Breaking in short (or lax) vowels
Some vowels in English only occur in checked positions. That is, they never occur in a word-final postion. They also tend to be shorter in length and for this reason they are often called “short” vowels. These are the vowels in KIT /ɪ/, DRESS /ɛ/, TRAP /æ/, LOT /ɑ/, STRUT /ʌ/, and FOOT /ʊ/. The first three are susceptible in Southern speech to a process called “Southern breaking” in which they “break” into two parts, becoming diphthongs, or, in some cases separate syllables with an intervening consonant.
In stressed monosyllables followed by a voiced consonant, particularly a bilabial, breaking can be quite prominent.
DRESS → ɛə̆ ɛjɪ ɛɪ̆
A similar breaking occurs with this short vowel, when followed by /d/, /m/, /l/ or /ʤ/. In other cases, when followed by /ɡ/, /ʃ/, /ʒ/, or /ʒ/, the effect can be so strong as to shift the vowel all the way to the diphthong /eɪ̆/.
TRAP (BATH) → æə̆ æɪ̆ æjɛ
In all American accents, the TRAP and BATH lexical sets are both merged to /æ/. Breaking occurs here under similar conditions as with the other short vowels. In some environments a more palatal offglide can produce a diphthong similar to PRICE.
LOT ≠CLOTH→ lɑt klɒʊ̞̆̆θ
While RP and some American accents merge these two sets, American Southern makes a distinction. It might be useful to think of this as a difference in length, which can lead to Southern Breaking. In this scheme, LOT is always short, while CLOTH is potentially lengthened, and therefore “drawled.”
Other Breaking of Long Vowels
THOUGHT → ɒʊ̞̆̆
As in many American accents, THOUGHT merges with CLOTH. In the South, these can be realized close to [ɔ] but as the phonetic context and sentence stress increases the length of the vowel, a stronger breaking occurs. In all cases, the vowel grows rounder as it progresses.
Contrary to tendency to “break” vowels as described above, some diphthongs are “smoothed” to monophthongs.
PRICE→ pɹaɪ̆s PRIZE/PRY → pɹaˑz pɹaː
This distinctive Southern sound is often realized with a weak second element, or as a pure monopthong. This effect can vary based on the surrounding consonant environment, with the monophthongal form occurring before voiced consonants or in word final positions, and a diphthongal form before unvoiced consonants
MOUTH → aʊ̜̈̆̆
The second element of the diphthong can be extremely unstressed and unrounded. There may also be nasalization in the first element.
Other diphthong changes
GOAT → ɤ ʊ̆
For some Southern speakers, this sound is strongly centralized and unrounded in both its elements. For this non-rhotic accent, the adjustment is quite small.
This subtle effect may simply flow from the tongue retraction characteristic of the oral posture. The first element is lowered and centralized.
The first element of this diphthong can be unrounded, and the second element shortened.
Speakers throughout the south might drop the post-vocalic r. The degree and manner of rhoticity varies by region, age, education, and race. The variety we are studying here is one of the “non-rhotic” forms. Unlike most non -rhotic dialects of English, this dialect resists elision. That is to say that in a phrase like far away where a speaker of a non-rhotic accent of say, New York City, would ordinarily use a consonant /r/ to link far with away, pronouncing the phrase [fɑɹəweɪ̆], the speaker of this accent would not, pronouncing the phrase [fɑ.əweɪ̆]. I understand this as form of rhoticity in which an articular action (bunching) is retained, serving a gestural purpose in the flow of speech, but with very little acoustic result. This lack of r-coloring and r elision can be applied to every instance of an r except for the beginnings of words or those instances where an awkward pronunciation might result.
START → ɑə̆ ɑ̹˞ ɑʌ̆
This diphthong needn’t travel far in tongue position to produce its shadow of rhoticity. Even in its oldest, least rhotic form, there is some diphthongal shift. There may also be some lip corner advancement.
NORTH → ɔə̆ ɔ
Historically, this category of words has been pronounced with the more open /ɔ/. In the South, the merger of this category with FORCE words (below) has been underway throughout the last century. For an older speaker or an earlier period, this distinction would still be appropriate.
FORCE → oə̆ o.ə
Words in this category begin with a higher, more rounded vowel. Because of this tighter starting point, a breaking can also occur when stress leads to lengthening of the vowel, producing two syllables: [o.ə]
CURE → ɔə̆
As in many other accents of English, the older form of ʊə̆ has shifted open to merge with FORCE . In more rhotic Southern accents and in unstressed positions, the merger tends towards NURSE.
NEAR → ɪə̆ ɪ.ə
NEAR and SQUARE seem to be more resistant to rhoticity, and more susceptible to breaking than the centering diphthongs listed above.
SQUARE → ɛə̆ ɛ.ə eə̆
The first element can be raised. If lengthened this phoneme can develop into a triphthong: [eɪ̆ə]. In unstressed syllables, the sound can be reduced and centered: [ɛ ]
NARROW → ˈnæɹə
As with RP and other London accents, NRS can group some SQUARE words with TRAP, making them part of a small group Wells calls TRAP+R.
There’s also a special case of GOAT in several words in that sample . Any word that ends with unstressed GOAT would follow the same pattern, and you can sometimes see it represented in spelling: feller, tater, yeller
Other Rhotic Vowels
In some areas, the bunching can be palatal producing a near merger of NURSE and CHOICE. This is by no means universal, and is perhaps a stigmatized and waning feature. It’s worth experimenting with, though, because it may give insights into the posture.
lettER → ə
Even in accents with rhoticity in other lexical sets, the unstressed nature of this phoneme can lead to a non-rhotic realization here. In unstressed prefixes like -pre- or -pro, metathesis can take place transforming “professor” into “perfesser”.
Lack of elision
As mentioned above, an articulation of some sort can occur in instances of post-vocalic/r/ that may not produce much of an acoustic result, but serves the articulatory role of /r/ as a transition between two vowels. This means that this accent, while “non-rhotic” in a very real sense, does not produce a linking /r/ where other non-rhotic accents would.
The effect of nasals
One of the most recognizable features of Southern accents is the raising of DRESS vowels when preceding a nasal. It is also not uncommon for nasalization to occur in any vowel preceding a nasal consonant. Strong nasalization is, however, more strongly associated with rhotic Southern accents.e similarity of the prefixes pre- and per- leads to confusion.
All accents are subject to assimilation and deletion of sounds in the right conditions. This accent goes further than others in some cases, and in at least one case, retains a distinction lost in R.P and most American accents.
Although potentially problematic for a theatrical accent, there is certainly a tendency towards simplification of complex articulations. This occurs more at the ends of words.
BELL→ bɛʟ͉ or bɛɰ
The phoneme /l/ is often velarized in postvocalic positions. This effect is increased in Southern accents to the point where there is no contact at all between the tongue tip and the alveolar ridge. The effect is somewhat similar to postvocalic /r/ in that the consonant action has been reduced and shifted back in the mouth. This has been called “L Vocalizing” by Wells since the resultant articulation is more like a vowel than a consonant. The essential difference here is that this accent does not include the additional lip corner protrusion which is characteristic of L vocalizing in London accents.
The so called “dropped g” is really the substitution of /n/ for /ŋ/. This is present in relaxed speech in all accents of English, but it occurs in this accent in more stressed environments.
“Glide Cluster Reduction”
J.C. Wells proposed this deeply inaccurate description, and we may be stuck with it. What he intended to describe was a lack of distinction in some accents between WHINE and WINE, or between HUE and YOU. He sees tins as a reduction of a consonant sequence – hw for WHINE and hj for HUE. The problem is that nobody really pronounces these as a sequence. Speakers who make a distinction between these phoneme categories pronounce WHINE as /ʍaɪ̯n/ and WINE as /waɪ̯n/. Similarly, many speakers (myself included) prounounce HUE as /çu/ and YOU as /ju/.
In many accents of English the distinction in pronunciation between words like WHINE and WINE has disappeared. It was retained for longer in the South, though it is in currently on the wane.