Words like tomorrow, sorrow, arrow, follow, borough contain /o/, as in the diphthong /oʊ/, which was /wə(n)/ in Middle English which was weakened from Old English /x/ or /ɣ/ + some sort of vowel. There was no vowel /o/ there in the first place. For example: morrow < ME morwen (no /o/) < OE morgen (no /o/); follow < ME folwen (no /o/) < OE folgian (no /o/).

Has there been a theory on when and why it was inserted there? Also, why does North American English have /oʊ/ consistently regardless of whether the original Old English consonant was /x/ or /ɣ/, while British English has /ə/ for Old English /x/ (as in borough < ME borwe < OE burh) and /əʊ/ (as in bow < ME bowe < OE boga; however arrow in both NAE and BE < ME arwe < OE earh)?

  • 1
    I cannot understand your first paragraph. What do you mean inserted? Do you mean evolved?
    – Lambie
    Commented Jun 22, 2018 at 17:58
  • @Lambie There was no /o/ there in the first place. Look at the Middle English and Old English forms. Notice there was only a sequence of /l/ or /r/ + /w/ in Middle English and + /x/ or /ɣ/ in Old English. Therefore, the existence of /o/ should be "insertion". All I could hypothesize right now is some vague sort of inherent physical quality of Middle English /w/ perhaps coupled with the loss of Middle English /ə(n)/ that makes an insertion of /o/ inevitable, to make words, like, easier to pronounce or something. Commented Jun 22, 2018 at 18:02
  • 1
    Ok, now at least I can understand it. But, how do you know how ME morwen was pronounced? You say no /o/ but you don't say how it was pronounced....
    – Lambie
    Commented Jun 22, 2018 at 18:13
  • @Lambie More or less as it was spelt. English spelling and pronunciation were much more consistent back then. Anyway, if you know something about these former forms of English, you'll know what I'm talking about, so there's really no need for long and unnecessary explanation. Commented Jun 22, 2018 at 18:44
  • 1
    IMO "tomorrow" is a poor choice of example-word for this question's title -- I thought from the title that the question was about the prefix "to-". Perhaps use "arrow" instead, as it has only one o?
    – Rosie F
    Commented Jan 13, 2019 at 9:15

2 Answers 2


Old English [ɣ] became [w] in Middle English in certain environments when surrounded by voiced segments. Thus /sorɣe/-> /sorwǝ/. This in turn was subject to epenthesis and produced variants like soruwe, meaduwe for Mod Eng "sorrow" "meadow." The final /ǝ/ was subject to deletion like other /ǝ/ in later Middle English. The epenthetic vowel represented by o was retained because a word-final cluster *rw was impossible. The contemporary standard pronunciation with /o/ is at least in part a spelling-pronunciation. In the early 17th century, pronunciations like /windǝ/ and /arǝ/ for "window" and "arrow" were not so stylistically marked, and are described by Ben Jonson as normal pronunciation in reference to o which "in the last Syllabes before w...frequently looseth: as in willow, billow." I.e. /bɪlǝ wɪlǝ/. By the end of the eighteenth century, John Walker was describing it as substandard: “The vulgar shorten this sound and pronounce the o obscurely, and sometimes as if followed by r, as winder and feller; but this is almost too despicable for notice. Good speakers preserve the diphthong in this situation”

If it hadn't been for the 18th century zest for spelling-pronunciations, we might well think of "Sarah" and "Arrow" as being perfect rhymes today. The 18th and early 19th centuries were a period of great social change, social mobility and thus social insecurity in urban areas. The obsession with propriety and gentility led to obsession with propriety in language. At the same time, mass-literacy was spreading on a scale unprecedented in English history. An environment like that tends to encourage language change, and this one in particular was a breeding-ground for new pronunciations derived from spelling. Newly-literate people who spoke various non-standard Englishes, when trying to "better themselves" linguistically, often had no better guide to correct pronunciation than the traditional orthography. And our orthography is a nasty one for learners.

  • Having grown up in Derbyshire I would find the sentence "'Ah shot an arra aht 't winda." meaning "I shot an arrow out of the window" not only comprehensible but quite familiar. I know these aren't exactly the same pronunciations as your example but they are very close.
    – BoldBen
    Commented Jan 13, 2019 at 7:42

Epenthesis to avoid word-final C + /w/ or C + /j/

English words can't end in a consonant followed by /w/, so presumably the /o(ʊ)/ developed either as an epenthetic vowel between a consonant and word-final /w/, or as a vocalization of the /w/. The modern English spelling suggests the "epenthetic vowel" analysis.

Similarly, in place of a consonant followed by word-final /j/ we see consonant + /i/ in modern English, e.g. belly < OE bælġ < PG "*balgiz" according to Wiktionary.

Variation between /əʊ/ and /ə/ probably isn't based on the identity of the Old English consonant

I'm not convinced by your suggestion that the pronunciation of "borough" as /ˈbʌrə/ in British English can be explained in terms of it coming from Old English "burh". As Alex Foreman's answer says, /ə/ instead of /əʊ/ can occur in an unstressed final syllable for many other words (including ones like meadow that didn't use to have /x/), not only for "borough". It's true that "borough" is spelled with "ough" instead of "ow", but I don't know whether that's related to the pronunciation.

Also, the pronunciation of the monosyllable bow as /bəʊ/ is not really relevant, because in stressed syllables, /ə/ is not possible, or at least not expected, in most accents of English. (The /əʊ/ in a Standard Southern British English pronunciation of bow is usually analyzed as a single diphthongal vowel phoneme, not as a sequence of /ə/ + /ʊ/.)

I think that Old English words with "g", "h", or "u" or "w" could correspond to words pronounced with final unstressed /əʊ/ in modern English.

  • /əʊ/ from words with g: I've read that in word-final position, Old English devoiced /ɣ/ to [x] (but in writing, either "g" or "h" could be used). The "h" in "burh" actually seems to be a devoiced /ɣ/: it could be spelled "burg", and [ɣ] shows up in other parts of its declension like the genitive plural burga.

  • /əʊ/ from words with h: The OED says the word "farrow" (cognate to Anglo-Saxon fearh, from PG *farhaz) is pronounced as /ˈfarəʊ/ in British English.

  • /əʊ/ from words with u or w: callow corresponds to OE calu (which had w in inflected forms).

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.