The principal changes (with the vowels shown in IPA) are roughly as follows. However, exceptions occur, the transitions were not always complete, and there were sometimes accompanying changes in orthography:
This means that the vowel in the English word date was in Middle English pronounced [aː] (similar to modern dart); the vowel in feet was [eː] (similar to modern fate); the vowel in wipe was [iː] (similar to modern weep); the vowel in boot was [oː] (similar to modern boat); and the vowel in house was [uː] (similar to modern whose).
The effects of the shift were not entirely uniform, and differences in degree of vowel shifting can sometimes be detected in regional dialects both in written and spoken English, for example in the speech of much of Scotland.
Not all words underwent certain phases of the Great Vowel Shift. ea in particular did not take the step to [iː] in several words, such as great, break, steak, swear and bear. Other examples are father, which failed to become [ɛː], and broad, which failed to become [oː].
Shortening of long vowels at various stages produced further complications. ea is again a good example, shortening commonly before coronal consonants such as d and th, thus: dead, head, threat, wealth etc. (This known as the bred-bread merger.) oo was shortened from [uː] to [ʊ] in many cases before k, d and less commonly t, thus book, foot, good etc. Some cases occurred before the change of [ʊ] to [ʌ]: blood, flood. Similar, yet older shortening occurred for some instances of ou: country, could.
Note that some loanwords such as soufflé and Umlaut have retained a spelling from their origin language which may seem similar to the previous examples, but since they were not a part of English at the time of the Great Vowel Shift, they are not actual exceptions to the shift.
The surprising speed and the exact cause of the shift are continuing mysteries in linguistics and cultural history, but some theories attach the cause to the mass immigration to the south east of England after the Black Death, where the difference in accents led to certain groups modifying their speech to allow for a standard pronunciation of vowel sounds. The different dialects and the rise of a standardised middle class in London led to changes in pronunciation, which continued to spread out from that city.
The sudden social mobility after the Black Death may have caused the shift, with people from lower levels in society moving to higher levels (the pandemic also hit the aristocracy). Another explanation highlights the language of the ruling class; the medieval aristocracy had spoken French, but by the early fifteenth century, they were using English. This may have caused a change to the "prestige accent" of English, either by making pronunciation more French in style, or by changing it in some other way, perhaps by hypercorrection to something thought to be "more English" (England was at war with France for much of this period). Another influence may have been the great political and social upheavals of the fifteenth century, which were largely contemporaneous with the Great Vowel Shift.
Because English spelling was becoming standardised in the 15th and 16th centuries, the Great Vowel Shift is responsible for many of the peculiarities of English spelling. Spellings that made sense according to Middle English pronunciation were retained in Modern English because of the adoption and use of the printing press, which was invented by Gutenberg in Germany around 1440.
German and Dutch also experienced sound changes resembling the first stage of the Great Vowel Shift: long i changed to [ai] and in Dutch to [ɛi] (as in Eis and ijs, 'ice'), and long u to [au] and the Dutch [œy] (via [yː]) (as in Haus and huis, 'house'). For this reason the German "ei" is pronounced closer to [ai] and in Dutch "ij" to [ɛi]; however, otherwise, those languages kept their spellings far more consistent.
The comparisons between English and the Continental Germanic languages are interesting for a variety of reasons. It is hardly surprising, given the huge differences between the structures of Old English and Old High German vowel phonology, that the "shifting" of long vowels would differ in detail as much as they resemble one another. Thus, there is no indication that non-high long vowels in English (i.e., all vowels not ) did anything but just move up in tongue-body position (there is no hint, for example, of the diphthongal features of Modern bee, bay, bone in any of the orthoepic [pronunciation] manuals of the 17th and 18th centuries). In German, original *[oː] is now [uː], as Proto-Germanic *fōtuz "foot" > German Fuß, but the process was totally different (as well as much earlier than the English developments): already in the very earliest Old High German texts (9th cent.) the vowel in question is consistently written -uo-. That is, it had "broken" into a nucleus with a centering glide. This complex nucleus "smoothed" as the term has it in Middle High German, becoming the [uː] of Modern German. The [oː] of Modern German has a variety of sources, the oldest of which is Proto-Germanic *aw, which smoothed before /t d r x/ (so rot 'red', Ohr 'ear', Floh 'flea', etc.) Elsewhere the sound was written -ou- in OHG. Similarly original *ay became [eː] before /r x w/, remaining what was written -ei- elsewhere.
Then, once again in the 15th or 16th centuries (about the time that uo was becoming [uː]), the long high vowels diphthongised much as in English, except that there were three of them: [uː] > [au], [iː] > [ai], and [yː] > [øy] > [oi]. In some German dialects original /ou, ei/ remain distinct from these new diphthongs, but in standard German they fell together as /au/ and /ai/, the latter somewhat eccentrically written -ei- as a rule, a holdover of the days when /ei/ was the only such diphthong.
Ideally, the term shift should be reserved for a sequence of interconnected changes, such as the two Germanic consonant shifts rather than used as a practical synonym of "sound change". The diphthongisation of the long high vocoids of Middle High German might qualify as a "shift"; the breaking and raising of *ō (and the very rare *ē²) do not really qualify as "shifts", still less their smoothing into modern [uː] and [iː].