According to Wiktionary, the English word 'tooth' can be pronounced as /tʊθ/ (as opposed to its regular pronunciation in RP of /tuːθ/) in certain areas of Wales and the British Midlands.

Is there any information about where this shift comes from? As a native British English speaker, I can attest to having heard this be used, but this variant seemed highly out of place. Furthermore, I was unable to find any other similar words which had undergone the same shift, e.g. booth (which resembles 'tooth' in both their Modern English and respective Middle English forms) and sooth (which matches 'tooth' in Modern and Old English forms). Neither of these two examples have similar shifts listed as variants.

Is there any literature on when/how this shift came about? I am particularly interested in finding out:

  • At what point in time the shift was first introduced/became commonplace;
  • Whether this (or similar) shifts are also present in other lesser-known dialects, which simply are not listed on Wiktionary;
  • Possible reasons for the shift (e.g. as part of a larger local vowel shift more generally unrounding close-back vowels, which just happened to affect 'tooth' as well);
  • Why other words, including those with similar origins to 'tooth' (see examples above) have not experienced the same change;
  • Whether other areas/dialects also had this difference at one point, but have now moved to the 'standard' pronunciation (if so, why?);
  • Whether the similarity in the variation found in both Wales and the Midlands is a mere coincidence, or there is some reason for the resemblance.

Thanks in advance for any information/literature on the topic!

1 Answer 1


I don't know anything about the change in pronunciation of this particular word, so this is just a partial answer.

The more general sound change this is a part of is shortening of /uː/ (from Middle English /oː/) to /ʊ/. This is a sporadic sound change that occurred in RP as well (in other words) and is one of the major sources of /ʊ/ in Modern English. For example, it occurs in foot, soot, good and almost all -ook words.

This sound change shows up in a number of words, but not in tooth, in the 1644 text The English primrose, by Richard Hodges, which is written with diacritics indicating the pronunciation of the words. Hodges writes hood and good as<hꝏ̯d>and <gꝏ̯d>, with a vowel that he describes as a short version of the vowel in words like tooth (which Hodges writes as <tꝏth>). Hodges also has short <ꝏ̯> in <hꝏ̯p>, unlike in modern RP. Hodges is cited in The Cambridge History of the English Language, by Roger Lass (2000), which is cited by the Wiktionary article "Phonological history of English close back vowels: FOOT–STRUT split". Based on the way Hodges and Wallis (another source on early pronunciation) describe the sound of the shortened vowel in words like good, Lass says that it was most likely a short vowel with the quality [u] at this point in time, rather than [ʊ] (Lass 90).

Even though Hodges doesn't provide direct evidence for the pronunciation of tooth with a shortened vowel existing at this time, the presence of the shortening sound change in other words makes it seem plausible that tooth might have had a shortened vowel for people other than Hodges in 1644. However, it's hard to estimate how recently the sound change might have applied, since there isn't anything I know of that would prevent it from occurring more recently.

Sporadic shortening in Early Modern English is also attested in the RP pronunciations of some words that had Middle English /ɛː/, which regularly developed to modern RP /iː/ (as in heath, beneath, mead, wheat), but in some words such as death, breath, dead, thread, threat developed instead to short /ɛ/.

  • Thanks for this answer! Clearly, it's very popular, and you've indeed raised some points about the general shortening shift which I hadn't really considered. I've given you an upvote for now, but unless I get another answer in the near future addressing more completely the locational variation (which, as you yourself point out, your answer doesn't really get to the bottom of), I'll also mark this as answer. Commented Mar 29, 2021 at 22:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.