One thing that annoys me in reading Durrell is the invariable insertion of u after q even when the q represents Arabic qāf. This is not unique to him—it exists in many books written about the Middle East before, say, the ’60s—but he’s particularly thoroughgoing about it; for instance, the OED entry for qasida ‘an Arabic or Persian panegyric or elegiac poem or ode’ has a bunch of citations, ranging from 1819 to 1971, but only one with the qu- spelling:

1958 L. DURRELL Balthazar iv. 82 He was delighted to hear some music and listened with emotion to the wild quasidas that the old man sang.

Another example from Balthazar (on the penultimate page, p. 242 of my Dutton edition): “I was terribly upset when Balthazar told me that he had fallen down those stairs at the central Quism and killed himself” (qism ‘part, section’ being an Egyptian term for a police station). I simply don’t understand the rationale. If you want to provide a folksy anglicized version, why not use k? The vast majority of your readers won’t know the difference between Arabic qāf and kāf, and wouldn’t be able to pronounce the qāf correctly anyway, so why not write “kasida” and “kism”? If you want to be scientific and use the q, why on earth toss in that pointless u, which adds only the certainty of mispronunciation?

While I’m at it, there’s an interesting word in a description of the Alexandria harbor in the first chapter of Clea (yes, I’ve reached the last book of the Quartet; the quote is from p. 34 of the Dutton edition): “Framed by the coloured domes there lay feluccas and lateen-rig giassas, wine-caiques, schooners, and brigantines of every shape and size, from all over the Levant.” When I couldn’t find giassa in the OED or Webster’s Third International, I started to worry, but the internet came through again—the Project Gutenberg text of R. Talbot Kelly’s Peeps at Many Lands: Egypt (1916) contains this very illuminating sentence: “These native boats are of several kinds, from the small ‘felucca,’ or open boat used for ferry or pleasure purposes, to the large ‘giassa,’ or cargo boat of the river.” But it will surprise no one to learn that I’m still curious about the word itself. It looks Italian, but it’s not in the Vocabolario Etimologico della Lingua Italiana or my own dictionaries; it looks like it should be pronounced /jasa/, but Egyptian Arabic does not use the /j/ sound. Any information will be much appreciated.

Update (Jan. 2022). See now the comment below by ktschwarz, who found this OED entry, unrevised since 1933:

gaiassa, n.
Pronunciation: /ɡʌɪˈasə/
Forms: Also 1800s caiash.
Etymology: < Arabic ḳayyāsa.

A high-stemmed vessel with lateen sails used on the Nile for carrying freight.


  1. At least he doesn’t change a following letter into “u”, like all those people who write “Al Queda”.

  2. marie-lucie says

    I have seen the name of the Northern Canadian town of Iqaluit written incorrectly Iqualuit.
    It seems that people who are not used to seeing a q without a following u automatically add a u to q (except at the very end, as in Iraq: Iraqi seems to be holding up, at least in print, because of the influence of Iraq).
    The oral counterpart of this written tendency is the pronunciation of u-less q as kw – a prime example being the Australian airline Qantas which apparently everyone (starting with the airline’s publicity spots) pronounces quantas, as if the word had something to do with quantity.
    So whether Durrell wrote qaf or quaf would have been immaterial to the readers, except those nitpickers who prefer their foreign words to be linguistically correct.

  3. Oh, Marie-Lucie, you are too optimistic. “Iqualuit” gets 40,000 Google hits. “Iraqui” is the right way to spell the adjective in Spanish, but even when restricted to English-language pages, Google still finds 282,000.

  4. Besides, it may not be so much Durrell as his publishers: yet another manifestation of the Idiot Copy Editor God, may his noodliness ever decrease.

  5. The word shows up spelled ‘ghiassa’ in Gilbert Parker’s The weavers. Perhaps this spelling indicates a غ ghayn in an Arabic original. Also, does not ق qāf sometimes get Romanized to ‘g’/’gh’, as in the surname of the Libyan leader? (I don’t know Arabic, but I thought the second spelling might tip off a more knowledgeable reader.)

  6. John C., tred carefully. I think that LH might be protective of the Idiot Copy Editor Gods. (That was my thought too, though).
    The “caique” is cognate with “kayak” via Turkish languages. At my URL.

  7. /kw/ for Qantas doesn’t seem entirely unreasonable, considering that the Q in the acronym stands for ‘Queensland’.

  8. Perhaps there’s some French influence? Incidentally, “Clea” is, for my money, the best book of the four, and more than makes up for “Mountolive” which is the least good.

  9. Remember my “queueing” story, LH?

  10. it may not be so much Durrell as his publishers
    Yes, that occurred to me, and I should have been more explicit about it. As for the Idiot Copy Editor Gods, no worries—I may be a copy editor, but I’m well aware of the number of rule-bound idiots who fill that position.
    The word shows up spelled ‘ghiassa’
    Excellent! And there’s a root gh-(w)-S ‘to plunge, submerge; to practise pearl-fishery’ that it might be a derivative of (it has a derivative
    ghawāSa ‘submarine’). Many thanks!

  11. This page has illustrations of various vessels, including (in the penultimate row, on the left) “Ghiassa o markab de Egipto”; markab is simply an Arabic word for ‘boat.’

  12. Isn’t the g pronunciation for qaf a trait of the Gulf? I was always curious why Qaddafi something was transliterated with a G; surely his name is not pronounced in that manner in Libya.

  13. marie-lucie says

    [g] for [q] :
    I think that interpreting the q sound as “hard g” is quite common – perhaps more than interpreting it as a k-sound. This may be because the uvular q, being pronounced farther back in the mouth, has a duller resonance than the velar k and is in that respect closer to the voiced velar g? I have also encountered this substitution in old transcriptions of some Native American words, as well as in attempts by untrained persons to pronounce the sound.
    ACW: I can believe that qu for q is prevalent on Google, but I was thinking more of the media.
    Conrad: I too liked Clea best – Mountolive seemed written by a different person, as it was much more conventional in structure and style than the other three books. Could it be that it was written first, kept under wraps by the author, and the other three written later in a less conventional manner, building on characters and incidents from Mountolive?

  14. I’m afraid I have no idea about the compositional history behind the Quartet, although I agree with your assessments. D himself conceived (as far as I recall) the four books corresponding to the four dimensions–and Mountolive alone represents ‘time’, so perhaps this is a post factum justification of its difference to the other books. Perhaps a stray Durrellian here can enlighten us. I find the ending of Clea quite poignant and beautiful–the mechanical hand is such a remarkable image.

  15. I just realised that I might have spoilt something for LH. Apologies for that…

  16. disgusted of alexandria says

    I thought LH was REreading it? You may however have spoiled it for me. Normally I’m about 80% through a book before I realise it’s the same one someone had told me the ending of years earlier. I’ll get back to you in 2020.

  17. The story with the Mummer is that he signed his name to a letter (in English, written to some American schoolkids in 1986) as “Moammar El-Gadhafi.” So like it or not, that’s his name in Latin letters, just as Benjamin Netanyahu is not “Binyamin” nor “Son-Of-Right-Hand”.
    As for the Quartet, did Queneau necessarily write the “Narrative” version of the silly story about the young man, the bus, and the button first of all?

  18. marie-lucie says

    As for the Quartet, did Queneau necessarily write the “Narrative” version of the silly story about the young man, the bus, and the button first of all?
    Perhaps not, but the Quartet is not four rewritings of a story which is totally devoid of interest, simply done as an exercise in stylistic virtuosity. Even though there are different incidents and points of view in the other three books, they are obviously written by the same author. Mountolive just does not have the same feel as the others, and it also could stand on its own as a single novel, while the others are interconnected with each other and one needs to read them all in order to understand what really happened. This is why I feel that Mountolive could have been written first, before the author had developed his own style and voice. Of course I have no proof, and I am not even a great fan of Durrell.

  19. I’ve been googling around and can find no indication that Mountolive was written first; in Conversations (pp. 26-27), he says “Justine was held up by bombs, but she took about four months—really a year, because the whole middle period I dropped in order to deal with the Cyprus job [Bitter Lemons]. I wrote Balthazar in six weeks in Sommières, I wrote Mountolive in two months in Sommières, and finished Clea in about seven weeks in all.” It seems much more likely to me that Mountolive is different because of the “objective” role it plays in the Quartet (which I seem to recall reading came to him as a whole in a flash in Alexandria).

  20. Kenneth Rexroth admired the first two books a lot but speculated that the next two were written under the pressure of a publishing contract.
    I liked Clea least. To me it neither brought anything new as good as what came before, nor did it sum things up satisfactorily or bring things to a conclusion. The quartet just tailed off.

  21. David Marjanović says

    My brother treats qu as totally inseparable. He goes so far as to write the mathematical variables as p and qu. I think this attitude is quite common among Standard Average European speakers.
    Jabal al-Lughat says the sound shift from [q] to [g] is a Bedouin thing, which probably means it’s not so much restricted geographically as being a difference between countryside/desert and city.

  22. Hurray for approximate search functions! Searching the OED for giassa returns (among other things) this entry, unrevised since 1933:

    gaiassa, n.
    Pronunciation: /ɡʌɪˈasə/
    Forms: Also 1800s caiash.
    Etymology: < Arabic ḳayyāsa.
    A high-stemmed vessel with lateen sails used on the Nile for carrying freight.

    (The ḳ must be a romanization of qāf.) The same spelling is found in Merriam-Webster Unabridged, with a definition that appears cribbed from the OED: “a Nile cargo boat with high stem and lateen rig”. Not in any other English dictionaries, but mentioned in the 1911 Encyclopædia Britannica under “boat”. Curiously, the only Wikipedia to have an entry for it is German.

    Finally, a Google image search with this spelling finds several pictures, such as this print for sale by the National Maritime Museum in London. The Look and Learn Historical Picture Archive is very helpful in explaining not just what it looks like, but why: “They have low hulls to enable them to dock easily even where there is a shallow draught, and a sharply-rising prow to give impetus when running on to sand.”

    The “caique” is cognate with “kayak” via Turkish languages.

    For the benefit of random passers-by: see this followup for the definitive debunking.

  23. Very nice indeed! Thanks much for that; I’ll add an update to the post.

  24. I should add that I haven’t mellowed over the years — I still deeply resent “the invariable insertion of u after q even when the q represents Arabic qāf.”

  25. “Iqualuit” gets 40,000 Google hits.

    Not really.

    First, when reporting a Google search we should say whether or not an item is enlosed in quotes. The quotes “precisify”. These searches return different results:


    Second, that number of hits at the top is an estimate. A wild over-estimate, and practically meaningless for comparisons of spellings. To compare prevalences for Iqualuit and Iqaluit, do use quotes:


    Then look through successive pages of results till there are no more hits. Shouldn’t take long. If offered a search that includes omitted hits, do it: and again go through the successive pages. The number arrived at after all this is a far more reliable basis for comparing against Iqaluit, once that has been put through the same process. And it’s all case insensitive by the way. That’s how it used to work. Is it still? Hard to know with Google.

    Better, of course (at least with formally published material):


    There are three forms on that chart.

    “Iraqui” is the right way to spell the adjective in Spanish, …

    (Read “iraquí”, with an acute.) That spelling of the adjective is far more undertandable than “Iraqui” in English. If English sought to render the word “natively” it might grossly come out as “Iraki”, “Irakee”, Iraquee”, or indeed “Iraqui”. But in Spanish, if “Iraq” is chosen as the base then only “iraquí” (normally lower case) and perhaps “iraqui” would be possible as native representations. Cf “yanqui”. In fact Spanish Wikipedia shows both “Irak” and “Iraq” as the base; the article goes on to use “Irak” and “iraquí”, consistently. Compare Italian, again using its Wikipedia: “Irak” is possible, “Iraq” is used in the article, and “iracheno” is the adjective.

    …, but even when restricted to English-language pages, Google still finds 282,000.

    Again, no.

Speak Your Mind