pronunciation: words ending with an unstressed /ɚ/

Nino83

Senior Member
Italian
Hello everyone.

Is there any rule, in words ending with an unstressed /ɚ/, in order to know if the stressed syllable is open or closed (i.e if its vowel is long or short)?
Words formed by affixes are excluded (words like make > maker or rob > robber, for example, because in these cases the suffix doesn't influence the vowel of the derived word, or suffixes like -ator, in which the preceding vowel is part of the suffix and is always long).
The problem arises when there is a single consonant in the spelling (in cases like rubber it's clear that the stressed consonant is closed, because of the double "bb" in the spelling).

colour, cover => /ʌ/, short
favour => /eɪ/, long
 
Last edited:
  • JulianStuart

    Senior Member
    English (UK then US)
    Rule? In English pronunciation? :D
    I don’t think the existence of the schwa at the end affects the preceding vowel - it’s the single or double consonant that would be the closest to a rule. There is a word coaler which (for me) ends with the same schwa as colour and has the long vowel because it has one l, but then there’s collar which has the short vowel (indicated by the double l).
     

    GreenWhiteBlue

    Senior Member
    USA - English
    I hope you understand that you are speaking of a minority accent in English. For most native English speakers, words such "maker", or "rubber", or "robber" do not "end with an unstressed /ɚ/."
     

    sumelic

    Senior Member
    English - California
    In general, there is no particularly good rule for the pronunciation of stressed vowels in morphologically simple disyllabic words that have penult stress and end in the "VCVC" spelling pattern (where the first "C" represents a single consonant letter representing a single consonant sound, so not "x"). This is one of the most unpredictable spelling patterns of English.

    In the specific case where the vowel letter is "u", we can say that the pronunciation is expected to be "long u" (/juː/, or after certain consonants or clusters, /uː/), because "short u" usually only occurs in orthographically closed syllables. Examples of words pronounced with "long u" in this position: spelled with -o(u)r, rumo(u)r, tumo(u)r, stupor, furor; spelled with -er, tuber, super. But the word "sugar" has an exceptional pronunciation with the "put/book" vowel.

    In other cases, there is variance, sometimes even for one and the same word.

    A brief overview of words spelled with "VCor" or "VCour": Words with a "long a" like favo(u)r include vapo(u)r, savo(u)r, labo(u)r. But valo(u)r is pronounced with a "short a" sound, as in "trap" ("val-" also has an irregularly short pronunciation in the related word valiant). The words clamo(u)r and squalor are in present-day English generally pronounced with "short" vowel sounds ("trap" in the first, and "lot" or "cloth" in the second), but there seems to be some evidence that pronunciations with "long a" existed historically.

    The words liquor, vigo(u)r, rigo(u)r are pronounced with "short i", except the OED notes that "rigor" (always so spelt) in the phrase "rigor mortis" (or as an abbreviation for it) may be pronounced with a "long i" sound, in accordance with the general rule of lengthening penult stressed vowels in open syllables in the traditional English pronunciation of Latin (e.g. in words like "apex", "vitae", etc).

    The words dolo(ur) and odo(u)r are pronounced with "long o"; the word hono(u)r is pronounced with "short o", and the word color, as you note, has a particularly irregular sound-spelling correspondence due to the phenomenon where the letter "o" represents a "short u" sound in certain words taken from French.

    The word tremor seems to usually be pronounced with "short e", but the Oxford English Dictionary makes note of a variant pronunciation with "long e".

    Morphologically simple words ending in <VCer> show variation as well. "Water" is of course highly irregular, as is the example "cover" that you mentioned. There are words with "long e" in this position (like "fever") and "short e" (like "leper", "clever"), "long i" (like "viper") and "short i" (like "river"), "long o" (like "clover") and "short o" (like "proper"). I guess I can't think of any word spelled with <aCer> and pronounced with the "trap" vowel, but this might just be an accidental gap. Of course, words spelled with intervocalic "v" are expected to be somewhat unpredictable due to the traditional avoidance of spellings with doubled "v", but as these examples show, there is also some amount of unpredictability when the single medial consonant is something else.
     
    Last edited:

    dojibear

    Senior Member
    English - Northeast US
    For most native English speakers, words such "maker", or "rubber", or "robber" do not "end with an unstressed /ɚ/."
    I think they do, for most AE speakers. In the AE parts of the WR dictionary, "robber" is written with this IPA sequences: /ˈrɒbəʳ/. The IPA symbols /əʳ/ and /ɚ/ are two IPA spellings for the same sound. And WR shows the AE for "rubber" as /ˈrʌbɚ/.

    Both spellings are called "R-colered vowels" (a wikipedia article title): the combination of a vowel and English R is often pronounced (in AE) as a single vowel sound, which is different than the sound of that vowel without the R.

    Wikipedia notes that sometimes the R-coloring only affects the end of the vowel, so you have two vowel sounds. This is represented by using two symbols such as [ɑɚ] or [ɑɹ], rather than the single symbol [ɑ˞].

    Is there any rule, in words ending with an unstressed /ɚ/, in order to know if the stressed syllable is open or closed (i.e if its vowel is long or short)?
    The only rule like this (that I know about) is the "silent E" spelling rule, where words with a long vowel are spelled with a silent E added on the end. But in that case the E is not an unstressed syllable: it only exists in the spelling.

    (cross-posted)
     

    Nino83

    Senior Member
    Italian
    For most native English speakers, words such "maker", or "rubber", or "robber" do not "end with an unstressed /ɚ/."
    Hi, GreenWhiteBlue. As dojibear said in AmE it's [ɚ] (a rhotic schwa, r-coloured schwa) or [ɹ̩] (syllabic "r") while in BrE it's a simple schwa [ə] (where Peter and pita have the same pronunciation).

    Thank you everyone for your answers. :thumbsup:
     

    Nino83

    Senior Member
    Italian
    Because in words with two syllables, normally, stressed closed syllables and, consequently, short vowels are indicated by double consonants or groups of consonants (except Cl and Cr, like table and April, where the syllable is open and the vowel long) but this doesn't happen in words ending with unstressed /ər/.
     

    Nino83

    Senior Member
    Italian
    Good examples!
    Let's say that this happens when a disyllabic word ends with a syllabic consonant (often sonorants, r, n, maybe also l).
     

    TheCrociato91

    Senior Member
    Italian - Northern Italy
    As dojibear said in AmE it's [ɚ] (a rhotic schwa, r-coloured schwa) or [ɹ̩] (syllabic "r") while in BrE it's a simple schwa [ə] (where Peter and pita have the same pronunciation).
    What I think GreenWhiteBlue meant is that not all American English variants are rhotic, therefore not all AmE speakers pronounce the syllabic r. That being said, I do believe American English is predominantly rhotic nowadays.
     

    dojibear

    Senior Member
    English - Northeast US
    Wikipedia's article on "Rhoticity in English" says:

    the historical /r/ is not pronounced except before vowels in "non-rhotic varieties", including most of the dialects of modern England, Wales, Australia, New Zealand, South Africa, and, variably, some parts of the eastern (particularly northeastern) coastal United States.

    When I lived near Boston, we talked about "Boston accents" and "Harvard accents", which I think were non-rhotic. And of course the "Boston Brahmin" accent (snooty elite posh and pretentious) imitates UK speech, so is non-rhotic. I heard non-rhotic local accents on Cape Cod and in coastal Maine. Those are the only ones I know. Since Boston is on the ocean, all of those are "coastal" accents.

    In northern (inland, not coastal) New Jersey we not only pronounced written Rs but we added an R sound to some words, such as "wash" ("warsh") and "Washington" ("Warshington"). Is there a fancy name for that too?
     

    sumelic

    Senior Member
    English - California
    Good examples!
    Let's say that this happens when a disyllabic word ends with a syllabic consonant (often sonorants, r, n, maybe also l).
    Unfortunately, I don't think the phonological structure of the word really helps you predict the pronunciation of a vowel letter in this context very well. I am more familar with spelling-to-sound correspondences than the reverse, but there are certainly a fair number of words with the -VCVC spelling pattern and a short vowel in the first syllable, even in cases where the final sound is a sonorant. Of course, many of these are words with medial v, like travel, ever, sever, novel, that might not count as I mentioned earlier, but there are also things like camel, rebel, carol, beryl, pedal, petal, model, vigil; tenor, scholar, manor; ebon, lemon, linen, wagon, felon, heron. The double-consonant spelling is more regular, and does occur in many disyllabic words ending in resonants (like lesson, cellar, adder, grammar, matter), but despite that, it is not used for a fair number of words like this with short vowels in the first syllable. I think it's a similar situation for words ending in non-resonant sounds with the VCVC spelling pattern, actually (although these are less common): we have short vowels in gamut, planet, profit, epoch, syrup, polyp.

    There are some fairly reliable rules for specific affixes: e.g. disyllabic adjectives ending in VCal or VCar, from Latin -alis/-aris (like final, polar) are almost all pronounced with long vowels in the first syllable (the only exception I know of is "moral", and in many American English accents you don't have to worry about that because of the merger of "short o" and "long o" before /r/).

    As you mentioned, the agent or instrument suffix -er, and the comparative suffix -er, also regularly trigger consonant doubling, so words with these affixes spelled with VCer usually have long vowels.
     

    Nino83

    Senior Member
    Italian
    Thank you very much, sumelic!
    I'll ask why there is this ambiguity, if there is any historical reason, in these words in the EHL forum.
     
    < Previous | Next >
    Top