The Shakespeare Conference: SHK 24.0176 Monday, 15 April 2013
Date: April 12, 2013 1:53:27 PM EDT
Subject: Elizabethan Pronunciation Question
The main evidence would come from rhymes. There are 11 instances of rhymes in the canon with forms of bow (head): bow rhymes with allow, now, vow; bowed with proud and vowed. All these would have had a ‘schwa + u’ diphthong (that is, the first element was more centralized than it is today). The only exception is bowing with growing (Tem 4.1.112).
There are 6 instances of rhymes with bow (weapon), all with the same /o:/ long vowel: with doe, go, know, woe. There are no rhymes with a word containing the diphthong.
So everything depends on that bowing/growing rhyme. Looking at it from the grow point of view, grow usually rhymes (33 times) with other words in /o:/, such as know and show. But there are two other instances where it rhymes with a word containing the diphthong - brow (VA 141), allowing (WT 4.1.16).
So, the bulk of the rhyming evidence suggests that the two phonemes were usually kept apart, but the phonetic difference between them was evidently not so great to prevent them being made to rhyme upon occasion. There are a few other word-pairs which show the same sort of thing, such as town/known and glow/brow.
I don’t think there’s a strong phonological case for pronouncing bow (weapon) with a diphthong, therefore, but I wouldn’t rule it out if there were a plausible semantic motivation (such as a possible pun).