Great Vowel Shift — part 2

Aug 2, 2010 by

In the last posting in July, I explained the basics of the Great Vowel Shift, a historical process that is responsible for a lot of the “mess” in the English spelling system, as far as vowels are concerned. As you may recall, the Great Vowel Shift applied only to long vowels, bringing them one step higher on the “vowel ladder” and turning the high vowels into diphthongs (in fact, it appears that mostly likely the change affecting high vowels happened first and once the high vowel spots were open, the rest of the vowels started raising too).

Importantly, however, the Great Vowel Shift did not affect short vowels. In another round of complication, some previously long vowels of Middle English were shortened before the Great Vowel Shift could apply to them. These vowels did not undergo the shift, creating yet another discrepancy between spelling (which reflects Middle English pronunciation) and modern pronunciation. Here are a few examples.

In the process of the so-called bred-bread-merger, the long front mid-low vowel (a lower-case epsilon in IPA) was shortened before coronal consonants. These are consonants pronounced with the tip of the tongue — corona in Latin — touching the roof of the mouth anywhere between the upper teeth and the hard palate. The coronal consonants which were the environment for the bred-bread-merger are /t/ and /d/. Thus, in dean we now pronounce [i:] (as a result of the Great Vowel Shift), but dead is pronounced [ded], not [di:d] (cf. deed). Simply put, dead escaped the Great Vowel Shift because its vowel was shortened in the bred-bread-merger so it was not long so the Great Vowel Shift didn’t apply. Other examples of the bred-bread-merger include head (vs. heed), threat and of course bread.

Another vowel that was shortened before the Great Vowel Shift “monster” could get to it is oo, which was shortened from [u:] to [ʊ] in many cases before /k/, /d/ and less commonly /t/. As a result of have book ([bʊk] not [bawk]), foot ([fʊt] not [fawt]) and good ([gʊd] not [gawd]). In another twist, this shortening process in the abovementioned words happened after the change of [ʊ] to [ʌ], while in other words [u:] to [ʊ] happened before [ʊ] to [ʌ]. How do we know? Well, that’s why good doesn’t rhyme with blood or flood!

Similarly, some instances of ou shortened from [u:] to [ʊ] and then changed to [ʌ], as in country, which explains why this word is pronounced the way it is rather than [ku:ntri], [kawntri] or [kʊntri].

Although the Great Vowel Shift applied only to long vowels, some short vowels changed around the same time. In addition to the already-mentioned [ʊ] to [ʌ] (except in northern English dialects, where they still say [bʊt] for but or [lʊv] for love), the short [a] was changed to the fronter vowel [æ], as in cat.

While the long and short vowels were changing, the diphthongs were not left in peace either. The Middle English diphthong [aj] as in day changed into [ej] and merged with the former long vowel [a:]. Hence, May and mane have the same vowel. Furthermore, the Middle English diphthong [aw] as in cause changed into the “open-o” (back mid-low rounded vowel) thus merging with the short-o of Middle English. As a result, cause and cost have the same vowel.

All these processes of vowel change in Renaissance English resulted in some interesting alternations in today’s English. For example, the vowel in the Old English ce:p-te shortened to kep-te so it didn’t raise in the Great Vowel Shift, while the vowel in the Old English ce:-pan remained long and raised to /i/. Hence today we have the present-past alternation keep ~ kept. Another example involves the “open-o” vowel in the Old English word for ‘nose’ lengthening in Middle English, so as a result of the Great Vowel Shift it was raised to the “closed-o” (back mid-high rounded vowel). But the vowel in nosþyrl (which derives from nosu + þyrel literally, ‘nose+hole’) stayed short. Hence, today we have nose ~ nostril. Similarly, the first vowel of Old English ste-lan was lengthened in Middle English to ste:le and now we have steal ~ stealth.

Why did these changes happen (in the way they did)? That’s the question for the next posting.

Like this post? Please pass it on:

Previous Post
| Next Post

Related Posts

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below:


  • Špule


    I find your article really helpful, but I cannot understand one thing. You say that "another vowel that was shortened before the Great Vowel Shift 'monster' could get to it is oo, which was shortened from [u:] to [ʊ] in many cases before /k/, /d/ and less commonly /t/. As a result of have book ([bʊk] not [bawk]), foot ([fʊt] not [fawt]) and good ([gʊd] not [gawd])." But in your first article on the GVS, you wrote that oo was pronounced like [o:], so it was shifted to [u:] (the chart also shows that). Thus, the quoted paragraph seems to me like a mistake.

    Also, I'd like to hear your opinion on words like "muse" or "tune", where /u/ is pronounced like [ju:], not [au]. Is this also a result of the GVS?

  • Asya Pereltsvaig

    @Špule: Thank you for your insightful comment and for catching the discrepancy. What I failed to point out in my posting is that some of these words currently spelled with "oo" used to be spelled with "ou" (pronounced [u:]) and hence would have been subject to the u:-to-aw change, had the u:-to-ʊ change not happened, as described in my posting. This is what happened to room (formerly spelled roum). For other "oo" words the theory is that the [o:] changed to [u:] (e.g., [mo:n] to [mu:n], moon) BEFORE the the [u:] to [aw] change applied. If the shortening of [u:] to [ʊ] didn't happen in these words when it did, we would have had [bawk] 'book' etc. Note that this analysis is compatible with the push-chain theory of the GVS, but not with the pull-chain theory. Which is the correct approach is still up for the grabs.

  • Špule

    @Asya Pereltsvaig: Wow, thanks, this looks even more complicated than I thought… One more thing: if moon was originally spelled with oo ([mo:n] and consequently [mu:n]) and it didn't shorten to [mʊn] (unlike some words ending in /k/, /d/, /t/), why don't we have [mawn] now?

  • Asya Pereltsvaig

    @Špule: It is all in the timing of the changes. Which in most cases we have to infer from the observed results. May not be the most theoretically satisfactory approach, but short of finding tape recordings of Middle English speakers, that's all we can do ;)