Phonological history of English high back vowels

Most dialects of modern English have two high back vowels: the near-close near-back rounded vowel /ʊ/ found in words like foot, and the close back rounded vowel /uː/ (realized as central [ʉː] in many dialects) found in words like goose. This article discusses the history of these vowels in various dialects of English, focusing in particular on phonemic splits and mergers involving these sounds.

Historical development

In the Old English vowel system, there was a pair of short and long high back vowels, /u/ and /uː/, both written ⟨u⟩ (the longer vowel is often distinguished as ⟨ū⟩ in modern editions of Old English texts). There was also a pair of back vowels of mid-height, /o/ and /oː/, written ⟨o⟩ (the longer one often ⟨ō⟩ in modern editions).

The same four vowels existed in the Middle English system. The short vowels were still spelt ⟨u⟩ and ⟨o⟩, but long /uː/ came to be spelt as ⟨ou⟩, and /oː/ as ⟨oo⟩. Generally the Middle English vowels descended from the corresponding Old English ones, although there were certain alternative developments – see English historical vowel correspondences.

Due to Middle English open syllable lengthening, short /o/ was mostly lengthened to /ɔː/ (an opener back vowel) in open syllables; this development can be seen in words like nose. In the Great Vowel Shift, ME long /oː/ ended up being raised to /uː/, in words like moon; ME long /uː/ was diphthongized, becoming the present-day /aʊ/, as in mouse; and the /ɔː/ of nose was raised and later diphthongized, leading to present-day /oʊ ~ əʊ/.

At some point, short /u/ developed into a lax, near-close near-back rounded vowel, /ʊ/, as found in words like put. (Similarly, short /i/ has become /ɪ/.) According to Roger Lass, this laxing occurred in the 17th century, but other linguists have suggested that it may have taken place much earlier.^[1] The short /o/ remaining in words like lot has also been lowered, and unrounded in some accents (see low back vowels).

Shortening of /uː/ to /ʊ/

In a handful of words, including some very common ones, the vowel /uː/ was shortened to /ʊ/. In a few of these words, notably blood and flood, this shortening happened early enough that the resulting /ʊ/ underwent the "foot–strut split" (see next section) and are now pronounced with /ʌ/. Other words that underwent shortening later consistently have /ʊ/, such as good, book, and wool. Still other words, such as roof, hoof, and root are in the process of the shift today, with some speakers preferring /uː/ and others preferring /ʊ/ in such words. For some speakers in Northern England, words ending in -ook such as book, cook still have the long /uː/ vowel.

Foot–strut split

The vowel in the word sun

The foot–strut split is the split of Middle English short /u/ into two distinct phonemes /ʊ/ (as in foot) and /ʌ/ (as in strut). The split occurs in most varieties of English; the most notable exceptions being those of Northern England and the English Midlands.^[2]

The origin of the split is the unrounding of /ʊ/ in Early Modern English, resulting in the phoneme /ʌ/. In general (though with some exceptions), unrounding to /ʌ/ did not occur if /ʊ/ was preceded by a labial consonant (e.g., /p/, /f/, /b/) and followed by /l/, /ʃ/, or /tʃ/, leaving the modern /ʊ/. Because of the inconsistency of the split, the words put and putt became a minimal pair, distinguished as /pʊt/ and /pʌt/. The first clear description of the split dates from 1644.^[3]

In non-splitting accents, cut and put rhyme, putt and put are homophonous as /pʊt/, and pudding and budding rhyme. However luck and look are not necessarily homophones; many accents in the area concerned have look as /luːk/, with the vowel of goose.

The absence of this split is a less common feature of educated Northern English speech than the absence of the trap–bath split. The absence of the foot–strut split is sometimes stigmatized,^[4] and speakers of non-splitting accents may try to introduce it into their speech, sometimes resulting in hypercorrections such as pronouncing butcher /'bʌtʃə/.^[5]

The name "foot–strut split" refers to the lexical sets introduced by Wells (1982), and identifies the vowel phonemes in the words. From a historical point of view, this name is inappropriate because the word foot did not have short /ʊ/ at the time the split happened, but underwent shortening at a later time.

Stages of the Foot–Strut split, as described by Wells (1982), p. 199

	mood goose tooth	good foot book	blood flood brother	cut dull fun	put full sugar
Middle English	oː	oː	oː	u	u
Great Vowel Shift	uː	uː	uː	u	u
Early Shortening	uː	uː	u	u	u
Quality Adjustment	uː	uː	ʊ	ʊ	ʊ
Foot–Strut Split	uː	uː	ɤ	ɤ	ʊ
Later Shortening	uː	ʊ	ɤ	ɤ	ʊ
Quality Adjustment	uː	ʊ	ʌ	ʌ	ʊ
RP Output	uː	ʊ	ʌ	ʌ	ʊ

In modern standard varieties of English, e.g. Received Pronunciation (RP) and General American (GA), the spelling is a reasonably good guide to whether a word is in the FOOT or STRUT lexical sets. The spellings o and u nearly always indicate the STRUT set (common exceptions are wolf, woman, pull, bull, full, push, bush, cushion, puss, put, pudding and butcher), while the spellings oo and ould usually indicate the FOOT set (common exceptions are blood and flood). The spellings of some words changed in accordance with this pattern: e.g. wull became wool and wud became wood. In some recent loan words such as Muslim both pronunciations are found.

Development of the sequence /juː/

Earlier Middle English distinguished the close front rounded vowel /y/ (occurring in loanwords from Anglo-Norman like duke) and the diphthongs /iu/ (occurring in words like new), /eu/ (occurring in words like few)^[6] and /ɛu/ (occurring in words like dew).

By the time of Late Middle English, /y/, /eu/, and /iu/ had merged as /ɪu/. In Early Modern English, /ɛu/ merged into /ɪu/ as well.

This /ɪu/ has remained as such in some Welsh, northern English, and American accents, in which threw /θrɪu/ is distinct from through /θruː/. In the majority of accents, however, the falling diphthong /ɪu/ turned into a rising diphthong, which became the sequence /juː/. This change had taken place in London by the end of the 17th century. Depending on the preceding consonant and on the dialect, this either remained as /juː/ or developed into /uː/ by the process of yod-dropping.^[7] Hence the present-day standard English pronunciations of duke /d(j)uːk/, new /n(j)uː/, few /fjuː/ and rude /ruːd/.

Foot–goose merger

The foot–goose merger is a phenomenon that occurs in Scottish English, Ulster varieties of Hiberno-English, Malaysian English and Singaporean English,^[8] where the vowels /ʊ/ and /uː/ are merged. As a result, pairs like look/Luke are homophones and good/food and foot/boot rhyme. The quality of the merged vowel is usually [ʉ] or [y] in Scottish English and [u] in Singaporean English.^[9] The use of the same vowel in "foot" and "goose" in these dialects is not due to phonemic merger, but the appliance of different languages' vowel system to the English lexical incidence.^[10] The full–fool merger is a conditioned merger of the same two vowels before /l/, making pairs like pull/pool and full/fool homophones.

Other changes

In Geordie, the goose vowel splits into what may be considered two distinct phonemes:

• /uː/, in morphologically closed syllables (as in bruise /bruːz/);
• /ɵʊ/, in morphologically open syllables – at the end of a word (as in brew /brɵʊ/), but also at the end of a morpheme (as in brews /brɵʊz/).^[11]

Compare the similar development of the high front fleece vowel.

The change of /uː.ɪ/ to [ʊɪ] is a process that occurs in many varieties of British English where bisyllabic /uː.ɪ/ becomes the diphthong [ʊɪ] in certain words. As a result, "ruin" is pronounced as monosyllabic [ˈɹʊɪn] and "fluid" is pronounced [ˈflʊɪd].^[12]

See also

• Phonological history of the English language
• Phonological history of English vowels
• Phonological history of English consonants
• Yod-dropping

References

1. Stockwell, R., Minkova, D. (2002), Interpreting the Old and Middle English close vowels, Language Sciences 24(3–4), pp. 447–457.
2. Wells, ibid., pp. 132, 196–99; 351–53
3. Lass, Roger (2000). "Phonology and Morphology". In Lass, Roger. The Cambridge History of the English Language iii: 1476-1776. Cambridge, UK: Cambridge University Press. pp. 88–90. ISBN 0-521-26476-6.
4. Wells, ibid., p. 354
5. Trudgill, P., Sociolinguistic Patterns in British English, E. Arnold 1978, p. 86.
6. http://www.courses.fas.harvard.edu/~chaucer/pronunciation/, http://facweb.furman.edu/~wrogers/phonemes/phone/me/mvowel.htm
7. Wells (1982), p. 206.
8. HKE_unit3.pdf
9. Wells, John C. (1982). Accents of English. Cambridge: Cambridge University Press. pp. 400–2, 438–39. ISBN 0-521-22919-7. (vol. 1). ISBN 0-521-24224-X (vol. 2)., ISBN 0-521-24225-8 (vol. 3).
10. Macafee 2004: 74
11. Watt, Dominic; Allen, William (2003), "Tyneside English", Journal of the International Phonetic Association, 33 (2): 267–271, doi:10.1017/S0025100303001397
12. Wells (1982), p. 240.