9

It seems that in the words Englishman, Frenchman, and Scotsman, the ‑man part is pronounced /mən/ (just like in Roman).

Whereas in snowman, the ‑man part is pronounced /mæn/ (just like in no man).

Why is it that when ‑man is appended to snow‑ to make snowman, the pronunciation of the man part doesn’t change?

Wouldn’t it logically follow for it to be pronounced /ˈsnowmən/ not /ˈsnowˌmæn/?

10
  • Compare mailman, sand-man, businessman, milkman, snow-man, mad-man, ice-man, middleman, ape-man, wise-man with bridesman, workman, marksman, draftsman, fireman, fisherman, freshman, policeman, countryman, showman, sportsman, toolsman, walkman, Welshman, coachman, dairyman.
    – tchrist
    Oct 22, 2017 at 2:19
  • 1
    One explanation is that “new” noun–noun compounds like snow-man and ape-man tend not to lose the stress on the second syllable, while adjective–noun compounds that are nationalities like Cornishman and Welshman tend to do so. But you may be hard-pressed to find an all-encompassing law that governs all cases, considering how many “old” noun–noun compounds of professions like coachman and journeyman have lost the secondary stress. We'll see what people come up with.
    – tchrist
    Oct 22, 2017 at 2:29
  • 1
    I don't think there is a rule that governs the pronunciation. It is probably how the words were derived from Old English or other source language. Oct 22, 2017 at 2:41
  • 1
    @Gary'sStudent: No. For one thing, in OE they didn't call the residents of the country Scotsmen but Scottas, and the country itself was Scotta lande. The word Scotsman itself was not used until the late 14th century. Englishman was Engliscman (pronounced with initial and final stresses). (N.B. the sc in Old English was prounced like our sh.) The OE for Frenchmen would have been frencisce men (or menn) with a stress on men as you would say it today if separated into two different words.
    – Robusto
    Oct 22, 2017 at 3:11
  • 1
    "Snowman" is a noun-noun compound and so has primary-secondary stress, but "Frenchman" is not a compound (though etymologically it might have derived from one) and so has the same stress a morphologically simple word would have.
    – Greg Lee
    Oct 22, 2017 at 5:55

3 Answers 3

4

In a comment to the original question, Greg Lee wrote the following brief answer:

"Snowman" is a noun-noun compound and so has primary-secondary stress, but "Frenchman" is not a compound (though etymologically it might have derived from one) and so has the same stress a morphologically simple word would have.

0

The all-encompassing law that covers all cases is that the pronunciation of the ‘man’ suffix depends on uniqueness. Where uniqueness occurs, the original pronunciation of ‘man’ is preserved. Where uniqueness does not occur, the pronunciation of ‘man’ is reduced as noted.

So, considering specific cases: You have only one milkman, there is only one sand-man, ‘businessman’ is usually cited only in a context where it is a unique reference, you have only one milkman, there is only one snowman (and his name is ‘Frosty’!), and so on, so these all have the original pronunciation of ‘man’, whereas ‘Englishman’, ‘Frenchman’, ‘fisherman’, ‘draftsman’, ‘coachman’ etc. are all decidedly non-unique, and so have a reduced pronunciation of ‘man’.

2
  • 1
    I was going to take issue with your "only one milkman/businessman" assertion, and still disagree. But the pronunciation of many of these profession or role examples is actually rather fluid anyway. Certainly in plenty of the southern English accents with which I'm familiar, the last syllable is often unstressed/schwa.
    – Chris H
    Oct 22, 2017 at 14:49
  • 2
    I utterly fail to see how you have only one milkman, but not only one coachman. They are completely equivalent in their uniqueness. Oct 22, 2017 at 15:49
0

I am finding, by experimentation, that it is a purely physical matter of how my lips are poised on ending the first syllable and how they prepare for the next syllable.

'Snowman' requires an 'oma' in the middle which is comfortable to say. 'Englishman' would need a 'shma' which is not comfortable. So I produce a 'shmi' sound instead.

This is purely subjective to myself, I must point out.

2
  • 1
    Are you certain this is a matter of comfort? If your thesis were true, then we should not be able to distinguish in pronunciation the two contrasting cases of ❶ "When an English man marries a French man” (which is actually gender-specific) versus ❷ "When an Englishman marries a Frenchman” (which is generic and so no longer gender-specific). The first case has no reduction to unstressed schwa in the respective “man” elements—yet the second case does undergo reduction, despite the two sequences being otherwise the same. Doesn't that disprove your comfort hypothesis?
    – tchrist
    Oct 22, 2017 at 15:13
  • 1
    The notion behind this is often called ‘ease of pronunciation’ in linguistics, but it’s a red herring. While there are a limited set of near-universal developments that can be said to be almost universally considered ‘easier’ to pronounce, the vast majority of things people consider ‘easier’ is purely down to familiarity with words pronounced like that—which is wide of the mark and not relevant here. There is no physiological difference in ease of pronunciation between /a/ and /ə/ in this case, regardless of the previous syllable. Oct 22, 2017 at 15:45

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.