Fixed and Free

| 9 Comments

Stress in English is often said to be “fixed and free”, by which is meant that for each word it is fixed, but that there is no fixed position in the word where it must occur, unlike Czech, Finnish or Hungarian, for instance, where it is invariably on the first syllable of a word, or Polish, where it is (with very few exceptions) on the penultimate.

However, the ‘fixed’ part of this statement has to be hedged around with all sorts of caveats. The stress placement on individual words can change over time,and one of the most frequent complaints made by older people about ‘the young’, is that they mispronounce words by putting the stress in the wrong place.

For most of the twentieth century, there was an argument over the placing of the primary stress in controversy: should it be on the first or second syllable? Advocates of initial stress claim that as it comes from the Latin CONTRA and VERSIA, clearly this is the ‘right’ place to put it. Unfortunately, etymology is never a safe argument to use. If we take the series photograph, photography, photographic, photogravure, the primary stress moves progressively later through the word: first, second, third, and fourth syllable (or, starting from the end: antepenultimate, antepenultimate, penultimate, final), and yet they all begin with the Greek-origin prefix photo-.

Stress has been shifting like this for a very long time. Balcony was, until the middle of the 19th century, stressed on the second syllable. Vagary and quandary were shown in dictionaries until well after the 2nd World War as being stressed on the second syllable, although  it is difficult to find even very elderly people who don’t smile if you suggest this pronunciation to them. Decade is another word that has changed: from /ˈdekəd/ in the early 20th century to /ˈdekeɪd/ (my pronunciation) and now more and more often /dəˈkeɪd/

The cases of balcony, vagary and quandary suggest that the Germanic tendency to stress initial syllables is working here, but controversy and decade show an opposite movement – perhaps in these instances treating the first syllable as an unstressed prefix. Integral, /ˈɪntəgrəl/ is also moving to /ɪnˈtegrəl/. Likewise preferable and combatant. Other examples, such as contribute, distribute, may indicate that the Germanic pull is stronger than that of the prefix (initial stress is taking over from the traditional 2nd syllable stress), but if that is so, what are we to make of communal, where second syllable stress is increasing in frequency (the Longman Pronunciation Dictionary gives a ratio of 68% to 32% in favour of initial stress for British speakers, but that was in 1990 – the newest edition repeats the figures, although I suspect, on no evidence apart from my own ears, that it is now nearer 50-50)?

Many nouns are traditionally distinguished from the identically spelt verb by stressing the two differently. For example contrast (initial stress – noun; final stress – verb). But there are also many cases where there is traditionally no stress difference (e.g. practice/practise, where the spelling does not affect the pronunciation; and dispute – 2nd syllable stress in both noun and verb). These patterns are also changing: the dispute type is joining the contrastive stress group (again 68% to 32% preference for the traditional pattern according to LPD), while the contrastive group is losing the contrast by stressing the first syllable in both noun and verb (e.g. import and export).

9 Comments

  1. Across the Water we are as usual more conservative: CONtroversy remains completely fixed here. Decade shows a more interesting story: /ˈdekeɪd/ is dominant with /dəˈkeɪd/ still a fairly uncommon variant, but /ˈdekəd/ is still standard in the sense ‘ten Hail Marys out of a rosary’. I suspect this is also true of Catholic use in Britain and Ireland.

    INTegral is universal in mathematical senses, whether talking of integration or integers, but inTEGral has some currency in other senses. I doubt if it is taking over, though. PREferable is dominant, though preFERable is also heard. On the other hand, comBATant and comMUNal seem to be far more common than the alternatives.

    “Vagary” and “quandary”? Do I detect a hidden Gilbert and Sullivan fan?

  2. “Controversy” is, of course, still stressed on the first syllable on my adopted side of the pond. There is an interesting blog posting on stress differences between the US and UK here: http://www.yek.me.uk/archive18.html#blog177

  3. John – Yes, I have played for rehearsals for many G & S operettas, but more recently than “Iolanthe”, “vaGARy” is the only pronunciation given in the 5th edition of the Concise Oxford, in 1964, and “quanDARy” is given as an alternative there as well. By the 7th edition (1981), “quanDARy” has disappeared altogether, but “vaGARy” is still hanging on as an alternative. I’ve just consulted the vocal score of “Iolanthe”, and note that in the song, which has six rhymes in each verse ending in ‘-ary’, only ‘Andersen’s Library’ has italics, to show that this is an unusual pronunciation.

  4. One of course has to be very aware of the effects of stress shift. For me the lexical stress in the verbs “import” and “export” is definitely on the final syllable, but in a sentence like:

    We must ___ more!

    I am sure i would shift the stress to the first syllable to avoid a stress clash.

    On another matter — I have a vague memory of hearing or being told that the word “July” was at one time regularly pronounced to be a homophone of the name “Julie” and that (or did I dream this?) there is a very early sound-recording of Tennyson, where he uses this pronunciation.

  5. I don’t know if it was just a personal idiosyncrasy, or a consequence of being born in 1875, but my grandmother systematically stressed “sometimes” on the second syllable, reducing the first vowel to a schwa.

  6. John M. – in your example sentence, I don’t think it’s only stress clash that makes you stress the first syllable of import/export, but also possibly that you want to emphasise the direction of trade. This is a feature of English that would be impossible in languages like Italian and Spanish (as John Wells was saying the other day here, under the heading “nuclear-free zones”), where the stress truly is fixed, and shifted contrastive stress is not allowed.

    Scots (and maybe others) often stress “July” on the first syllable. After giving the various spellings over the centuries, OED1 says “The latter form [which one does it mean? GP] was still accented Juˑly as late as Dr Johnson’s time; it is still /ˈdʒuːli/ in Southern Sc.; the modern Eng. pronunciation is abnormal and unexplained”. This was probably written by Murray between 1899 and 1901, when he was working on I, J, K.

  7. This is what the 2008 draft revision of JULY now says:

    The word was usually stressed on the first syllable in the early modern period, as the form July-flower, due to folk etymology (see gamma forms at GILLYFLOWER n.), implies. The orthoepists Peter Levins (1570) and Elisha Coles (late 17th cent.) both include the word among those which have unstressed -y, and Johnson (1755), W. Johnston Pronouncing & Spelling Dict. (1764), and J. Walker Dict. Answering Purposes of Rhyming (1775) all indicate stress on the first syllable (Johnston also marking the y as ‘long’). Both occurrences of the word in Shakespeare are so stressed, as are most metrical examples down to the late 18th cent. (compare quots. 1704 at alpha, 1736 at alpha, 1781 at Compounds 1). Stress on the first syllable still occas. occurs in Scotland.

  8. I am surprised by your report that some people are stressing “decade” on the second syllable. I don’t believe I’ve ever heard that, and wonder if there’s ever a confusion with “decayed”.

    With several of the other examples, it’s the new pronunciation that I’m familiar with and the old one that seems strange, which is probably not a surprise coming from someone born in the late seventies. Just to give one example, “contrast” for me has always been stressed on the first syllable, noun and verb alike.

  9. Adrian – Maybe de’cade hasn’t reached Australia yet, but the LPD British English survey showed a 14% preference for this pronunciation even in 1990.

Leave a Reply

Required fields are marked *.