Monday, 18 June 2012


Harry Campbell writes,

Who better than you to consult about my life-long puzzlement over the pronunciation of Ulysses. I've always stressed the second syllable, I’m not sure why, but initial stress seems to be the norm. What are the rights and wrongs of this, from a classicist’s point of view?

With last Saturday being Bloomsday, the airwaves were full of discussions of James Joyce’s novel Ulysses. And everyone I heard on BBC Radio Four called it ˈjuːlɪsiːz.

Not me. I’ve always called it juˈlɪsiːz, like the eponymous hero of the Odyssey, Greek Odysseus əˈdɪsjuːs and Latin Ulysses or (less usually) Ulixes (apparently via Sicilian Greek Οὐλίξης). In accordance with the Latin stress rule (stress the penultimate if it is a heavy syllable), all three forms of the name are traditionally stressed on the penultimate. The double s of Odysseus and Ulysses, like the x of Ulixes, makes the syllable heavy.

In the 12th edition of EPD (1963), the last to be edited by Jones himself, the pronunciation of Ulysses is given as
    juˈlɪsiːz (rarely ˈjuːlɪsiːz)
(I’ve modernized the phonetic notation.)

By the 14th edition (ed. Gimson and Ramsaran, 1977) the word ‘rarely’ has been removed, but priority is still given to juˈlɪsiːz. It was only when Peter Roach took over as editor that the priority was changed, placing the antepenultimate-stressed version first. Judging by what I heard on the radio, Peter was right to make this change.

Perhaps I ought to do the same for LPD. Or at least conduct a preference poll.

The Merriam-Webster Collegiate and other dictionaries suggest that Americans still retain the traditional penultimate stressing. Perhaps it’s only the Brits (and only some of them) who have abandoned it.

The influence of my schooldays in the Classical Sixth is still strong. I don’t think I’m going to change my own pronunciation.


  1. The Tom Lehrer song only works with the traditional stressing:

    And the head coach wants no sissies,
    So he reads to us from something called 'Ulysses'.

  2. Many of my pronunciations are influenced by my classical education — but not names I was familiar with before I took up Latin and Greek.
    The version with initial stress was the only form I was aware of back in the 1950's.

  3. Personally I've always put the stress on the first syllable and have rarely heard it on the second, which goes to show how different people's experiences of a word can be!

    I wonder what influence Ulysses S. Grant may have had on changing the pronunciation. Obviously I've no idea how he preferred his name to be said, but it's hard to think of other ways that it could get into regular public discourse as it's not the popular of names!

    1. "... not the *most* popular of names", that should say!

    2. No influence at all, I should say. I have not studied either mythology or history past the secondary level, but I have never heard any pronunciation except /juˈlɪsiz/, either. It was originally Hiram Ulysses Grant's middle name, but due to an error he was registered at the U.S. Military Academy as "Ulysses S. Grant" (his mother's maiden name was Simpson) and he retained this form of his name for the rest of his life. His nickname at the Academy was "Sam", because of the initials "U.S." being read as "Uncle Sam".

      I would guess that /ˈjuːlɪsiːz/ is a barbarism that someone started either out of ignorance or a desire to set a new fashion, and that caught on. None the worse for that, of course: the same might be said of the modern pronunciation of July, which was /ˈdʒuːli/ until about two centuries ago.

    3. Yes. In fact, one of its spelling variants is Hyrum. It's a Bible name, and so follows the rules for Anglo-Latin pronunciation despite not being of Latin origin.

  4. Since I read the title of today’s blog this morning I’ve had the theme song to the classic ’80s cartoon Ulysses 31 playing on loop in my head. In that song the name is repeatedly pronounced with initial stress, and as far as I remember it’s pronounced that way in the cartoon itself as well. The massive popularity of Ulysses 31 (a science-fantasy reimagining of the Odyssey) can’t have done much to halt the advance of the initial-stressed pronunciation.

    This change clearly constitutes an exception to the rules for pronunciation of Latin loanwords, but aren’t there other exceptions that have always been around? Romulus, for example, is always pronounced 'rɒmjələs, rather than 'rɒmələs as I’d have expected. That’s before the confusion caused by the inconsistent switch to continental values for Latin vowels discussed previously in this blog. Surely there are others?

    1. Romulus ˈrɒmjələs, ˈrɒmjʊləs follows an entirely regular pattern: cf stimulus, modulus, calculus, cumulus, tumulus, convolvulus.

  5. You're right, I hadn't noticed that. It seems that in open syllables u is always long, hence the j in these words.

    1. Not long in Latin! These all have Latin -ŭlus. Pretty much all Latin u, long and short, get a preceding /j/ in English.

    2. Latin vowel length is discarded once it is used to determine the position of the stress; after that, English spelling rules are applied. So ulla, as in the phrase lectio sine ulla delectatione 'reading without any enjoyment', is not yodified thanks to the ll.

  6. My own pronunciation (and perhaps that of many of my generation) has been heavily influenced by the 1980s cartoon (; "U-ly-SEE-ee-EE-ee-EEZ".
    On a closer listening, I notice that the spoken preface tends towards stressing the penult, but the theme song places stress on the first and last syllable...

  7. I've always had first syllable stress in "Ulysses". Antepenult stress is pattern of most classical proper names ending in /i:z/ (Socrates, Demosthenes, Pericles, Androcles, Diogenes, Aristophenes, etc. -- exceptions include Achilles and Archimedes).

    In addition, the pattern of vowel lengths long-short-long, with stress on the second syllable, strikes me as somewhat unusual in English. I can't immediately think of any (other) examples than "u-LY-sses", although I'm sure there are some.

    1. Isn't the initial vowel of Ulysses short: [jʊˈlɪsiz]?
      And the penult of Archimedes has long [iː], right?
      At any rate, I find the pattern of Ulysses and Achilles unusual too. It only makes sense to me if the ending is reinterpreted as a 'short'/unstressed [i] plus 'grammatical' [z]. Compare, say, crevalles are fish, HP antes up billions, the finale's tonight, karate's evolved, a machete's handle, etc.
      That [z] is 'added on' in some sense in Ulysses and Achilles is supported by its absence in the adjectival forms ulyssean and achillean.
      That the es ending is somehow reduced is also suggested by the fact that North Americans can freely flap [t] in Achates [əˈkɑɾiz] or Penates [pəˈnɑɾiz], just as they do in diabetes, as well as in karate's, machetes, committees, etc.
      Elsewhere, a related odd pattern seems 'forced' by related words, for example, putrescent : putrescine [ˌpjuːˈtɹɛˌsiːn]; rubella : rubellite [ˌɹuːˈbɛˌlaɪt] (also [ˈɹuːbəˌlaɪt]); vibrissa : vibrissae [ˌvaɪˈbɹɪˌsaɪ] ~ [ˌvaɪˈbɹɪˌsiː].
      A few more thoughts:
      - The following examples are similar to Ulysses/Achilles, depending on how [st] is syllabified: Orestes, Agonistes, Cerastes.
      - Both [jʊˈlɪsiːz] ~ [ˈjuːlɨˌsiːz] occur in Canada,
      - Another Greek name that admits either antepenultimate or penultimate stress is Philoctetes [fɪˈlɑktəˌtiːz] ~ [ˌfɪlɑkˈtiːtiːz].
      - Odysseus is [ˌoʊˈdɪˌʃuːs] for some.

  8. Oddly enough the OALD gives /ˈjuːlɪsiːz/ as the American pronunciation; both options are given as possible in British English.

  9. Regarding classical names, I've only today found out that the name of the king Ixion is pronounced ɪkˈsaɪən. I thought it was ˈɪksiən.

  10. Oxford Dictionary of English has only n
    under Ulysses.

  11. I can't say I recall ever hearing it with a stress on the first syllable in the US. I had never even considered doing that. My anecdotal experience doesn't mean much but if I had to guess I would assume that Merriam-Webster is accurate in their assessment.

  12. "By the 14th edition (ed. Gimson and Ramsaran, 1977) the word ‘rarely’ has been removed, but priority is still given to juˈlɪsiːz."

    For what it's worth, already in the 13th edition (ed. A. C. Gimson, 1967) the entry ''Ulysses'' reads like that (at least to judge from the 1974 reprint "with corrections").