Blagojevich: my bleeping phonological error

« previous post | next post »

At one point toward the end of my post (here) on the Blagojevich affair, I made a bad linguistic mistake. And on a cute and moderately interesting point. In brief (I have little time this morning), for unknown reasons but in a fairly well studied way, expletive insertion inside words in English works prosodically only when there is a weak stress somewhere before the insertion point and a strong stress immediately after it: I'm not going to KALama-fuckin'-ZOO!. It doesn't work when there is only an unstressed syllable before (so ?I'm not going to Chi-fuckin'-CAgo is nowhere near as good, because Chi- is too light), and it is hopeless when the stress (say, because it is on the first syllable) has to precede the insertion (*I'm not going to ABi-fuckin'-lene). The remaining details will quite probably be explained here by one of the Language Log phonologists (you could even read quite a bit about the details in a paper by Arnold Zwicky and me, item no. 124 here). Here's my mistake: I don't know Serbian, and I get my news mainly from print. And at the point when I wrote, I thought the name was Blago-JE-vich. It isn't. It's Bla-GO-jevich. The readers who have objected in comments on Ben's latest post are correct: I put the insultingly inserted expletive into his name in the wrong place.

*BlaGOfuckinjevich is completely impossible, for me as well as the commenters who objected. As properly pronounced, this name has the wrong prosodic shape to work well with an inserted expletive (there isn't enough stress on the syllable before -go: the Bla- is basically stressless). I knew that what I wrote was not great; but given the correct stress position, I made a clear error, by my own judgment, and I would have abandoned the attempt to insert if I had realized where the correct stress went. I didn't find out how the name was correctly pronounced before I posted. This was careless. I would probably be excoriated by judgmental readers if I opened comments, so I hope my mouse doesn't slip and hit the "Allow Comments" button by mistake… Oops!


  1. Paul said,

    December 16, 2008 @ 5:22 am

    Is this morphophophuckinology?

  2. jva said,

    December 16, 2008 @ 6:06 am

    So this means that Obama is in worse position than Bush was?

  3. Karen said,

    December 16, 2008 @ 6:43 am

    Worse, or better, I'd say – depending on how often he wants his named deformed like this!

  4. Graham said,

    December 16, 2008 @ 6:46 am

    I always love a good use of expletive infixation, and for my money what Geoff had originally is all good. The stress on syllables is usually sufficient enough to determine where the expletive should be correctly placed, but the more complex structure of unbelievable allows for either:

    un-fucking-believable or unbe-fucking-lievable,

    the latter being my personal favourite despite there being two strong syllable stresses at the beginning (un and be). Syllable stress is all good for the majority of cases, but sometimes the morpheme boundaries must come in to play to achieve that exacting dramatic effect.

  5. Karen said,

    December 16, 2008 @ 6:46 am

    ps – you don't need to know Serbian. All you need to know is how the governor pronounces his name, which very easily may not be how his ancestors did. That said, though I'm not a Serb speaker (Russian yes, Serb no), I can't think of a Serb name with the stress on the -je-.

  6. Faldone said,

    December 16, 2008 @ 7:29 am

    I can see Geoff's ignorance of the pronunciation of the governor's name; it probably wasn't saturating the newscasts in Scotland the way it was in the USA and one of the commonly heard comments early in the news cycle was a comment on the difficulty in pronouncing Blagojevich's name. As for the insertion point of infuckingfixes there's another rule, sometimes in competition with the rule of prosody. The infix should go between morphemes. Since Blagojevich doesn't have any morphemes recognizable to the average US English speaker that rule has no effect for us and the infix must follow the prosody rule. I suppose the -be- in unbelievable is, technically a morpheme, but it doesn't have the weight of the un-, so it isn't as common to hear unbefuckinglievable.

    Note: Just a quick, unvetted Google fight; unfuckingtbelievable gets 44.5 Khits, unbefuckinglievable gets 3.7K, so the latter is not non-existent.

    Oh, and about infixing Bush's name, it would have been perfectly normal to have said "Doublefuckingyou'. Two birds and all that.

  7. Robert M Maier said,

    December 16, 2008 @ 8:24 am

    On a tangent, this reminds me of a comment of Seleskovich: she pointed out that the pronunciation of her name could be 'translated' from the original Serbian (SE-leskovich), using different stress patterns, into French (selesko-VICH), Russian (seles-KO-vich), and English (se-LES-kovich).
    Which takes the rule of speaker's own pronunciation out into the uncertain grounds of linguistic relativity again, I suppose.

  8. Shay said,

    December 16, 2008 @ 9:16 am

    It's good to know that "Blagoje-fuckin'-vitch" still works quite well.

  9. Orange said,

    December 16, 2008 @ 9:23 am

    Rod is short for Milorad, which would seem to lend itself to infixation better than the last name does. Too bad he doesn't use his full name—though really, I suppose it's a blessing. Milofuckin'rad Blagojevich has far too many syllables.

  10. Eyebrows McGee said,

    December 16, 2008 @ 9:35 am

    It's okay, Geoff, we Chicagoans have long training with difficult-to-sight-read (in English) Polish names, so Blagojevich came relatively naturally to us. We sort-of expect the rest of the US — not to mention the rest of the world — to struggle with our local politicians when they're in the national or international press. (One of the more amusing bits of the presidential election, to me, was watching the press gradually come to a consensus on how "Obama" is pronounced … and in certain foreign languages, it just clearly isn't. The "a" vowels seemed surprisingly difficult for some BBC reporters to produce.)

    When I went to Duke for grad school, at orientation they were teasing us and saying we'd know we were real Dukies when we could pronounce and spell "Krzyzewski" without cheating. I was like, "Shuh-SHEF-ski? How's that hard?" Only slowly did it dawn on me that none of them were from Chicago or other nearby places where Polish sausage is easily purchased and Krzyzewski is a totally logical spelling. :)

    On a different note, I remember (somewhat vaguely; I was 17 or so) when Blagojevich first ran for the US House in 1996. His biggest radio commercial was a bunch of little kids attempting to pronounce "Blagojevich" and failing with adorable cuteness ("I just can't say it!") with a benevolent adult narrator periodically saying, "Blagojevich" correctly and making short statements about his policy, and by the way, did you notice his name is hard to pronounce?, because his name was new on the scene. It was quite successful at gaining him pretty wide name recognition, and overcoming the difficulty of running a newly-arrived "Blagojevich" against an incumbent "Flanagan."

    In Illinois he's generally "Blago" in casual written conversation (as on the internet), as "Blagojevich" doesn't type real smoothly; in conversation he's typically "that IDIOT." ("Which idiot?" "Our governor." "Oh, THAT idiot.")

  11. Eyebrows McGee said,

    December 16, 2008 @ 9:36 am

    "and in certain foreign languages, it just clearly isn't. The "a" vowels seemed surprisingly difficult for some BBC reporters to produce"

    These, incidentally, are two separate thoughts which needed a more clear separation. But they're pretty amusing run together. :)

  12. Nick Z said,

    December 16, 2008 @ 9:50 am

    Because SBE and American English /a/ often have different phonetic realisations it can be difficult for SBE speakers accurately to reproduce American pronunciation; and sometimes we even assign it to the wrong phoneme – e.g. /o/, /a:/ for /a/ etc. So in proper names, which do not have a corresponding SBE pronunciation, it's not surprising there's some variation.

  13. Eyebrows McGee said,

    December 16, 2008 @ 10:37 am

    @Nick, oh, I know — I can only make one short-a sound, and it is the Chicago sound, and that sound does not occur in the rest of the US, let alone overseas. When I lived in Raleigh-Durham I was totally unable to properly pronounce "Raleigh," to the endless amusement of my friends. So I don't begrudge anybody else their As. Some of the BBC commentators (and other newscasters around the world) settled into a pronunciation for "Obama" that sounded basically like what I'd think Obama wound sound like in a SBE accent (or whatever other accent was on the table), and I certainly expect American names to have unique pronunciations in foreign accents and foreign languages. What was amusing to me was when other commentators would be wandering all over the place with it, even within a single story, because of those troublesome As. As if they were trying to negotiate a settlement between what it wanted to sound like in their head, what it sounded like in American English, and what it sounded like in the majority pronunciation of their local language/accent.

  14. Lazar said,

    December 16, 2008 @ 10:47 am

    @Eyebrows McGee: Just out of curiosity, do the BBC commentators in question pronounce Obama with the vowel sound of "cat" or with the vowel sound of "spa"?

    [The latter. –GKP]

  15. mollymooly said,

    December 16, 2008 @ 10:50 am

    AFAIK, the first a in "Obama" has the PALM vowel. For most Americans, this is the same as the LOT vowel. For most Britons, it's not, but Britons still use the PALM vowel and not the LOT vowel. However, there are many "foreign" words where Britons use the TRAP vowel and Americans use the PATH vowel; Wikipedia lists:
    annato, BangladeshA2, Caracas, chiantiA2, Galapagos, GdańskA2, grappaA2, gulagA2, HanoiA2, JanA2 (male name, e.g. Jan Palach), KantA2, kebab, Las (placenames, e.g. Las Vegas), Mafia, mishmashA2, MombasaA2, Natasha, Nissan, Pablo, pasta, PicassoA2, ralentando, SanA2 (names outside USA; e.g. San Juan), SlovakA2, Sri LankaA2, Vivaldi, wigwamA2, YasserA2

  16. Andrej Bjelakovic said,

    December 16, 2008 @ 10:55 am

    Here in Serbia, the name is usually pronounced with the stress on the first syllable, although I can easily imagine there being many families with the name Blagojević who pronounce it with the stress on the second syllable.
    However, as someone already pointed out, it's pretty much irrelevant how Blagojevich's ancestors pronounced it; what's matters is how he and majority of his countrymen pronounce his name, and that is most definitely /bləˡgɔɪəvɪtʃ/.

  17. Jonathan Lundell said,

    December 16, 2008 @ 11:28 am

    WRT the Tribune's "bluh-GOY-uh-vich", it's easy to see how Bla-GO-jevich can be heard that way, and it makes Bla-GO-fucking-jevich difficult. The most natural sounding variation, to me, is "[Rod|Governor] fucking Blagojevitch" (where [a|b] is "a or b or neither").

  18. jfruh said,

    December 16, 2008 @ 12:08 pm

    I'm sort of interested in the phenomenon of not knowing how names are pronounced generally. I work at home and get most of my news from print and the Internet, like Geoff, whereas my wife commutes by car and is an NPR junkie. This has given rise to some amusing confusiong while we discuss the news (my attempt to pronounce Saddam Hussein's home town, Tikrit, with short "i"s prompted especially vigorous eye-rolling).

    But it's true that there are whole classes of names that I recognize by sight but are so long and/or gnarly that I've never really try to sound them out and generally just gloss over with my eyes. I think Ahmadinejad had been president of Iran for a year before I even knew how many syllables were in his name.

  19. Arnold Lambtally said,

    December 16, 2008 @ 12:14 pm

    I have often heard Commonwealth speakers pronounce Obama to rhyme with Alabama, at least to my ears.

  20. Coby Lubliner said,

    December 16, 2008 @ 12:28 pm

    I would have thought that Milošević would have given people an idea of how four-syllable surnames ending in -ević (-evich) are stressed in Serbofuckincroat.

  21. Daz said,

    December 16, 2008 @ 7:10 pm

    Fwiw, all the pronunciations I've heard on radio and TV have been bluh-GAW-yuh-vitch.

    (Of course, GAW-yuh sounds like "Goya". And the j in the spelling seems to be pronounced like a y.)

  22. Ran Ari-Gur said,

    December 16, 2008 @ 7:20 pm

    Can it go the other way? Like, can "Chi-fuckin'-cago" occur, but force "chi" to be pronounced with secondary stress (and presumably a long "i" sound /aI/)? Or is "Chi-fuckin'-cago" only possible for speakers who would pronounce the "chi" that way even in plain old "Chicago"? (And likewise, can "Abi-fuckin'-lene" exist, but force the "lene" to be pronounced with primary stress?)

  23. Lazar said,

    December 16, 2008 @ 8:07 pm

    @Daz: No, it's definitely [bləˈgɔɪ.ə.vɪtʃ]. The thing is that in American English, "Gaw-ya" doesn't sound like "Goya" – compare the difference between the predominant pronunciation of "lawyer", [ˈlɔɪ.ɚ] "loy-er", and the alternative, [ˈlɒ:.jɚ] "law-yer", which sounds quite different and is relegated to the status of a marked regional pronunciation (see ). From my perspective as a native AmEng speaker (and a member of the solid majority that prefers [ˈlɔɪ.ɚ]), it would sound unnatural for me to use [bləˈgɒ:.jə.vɪtʃ] because such a [V.j…] sequence would have no precedent at all in my dialect.

    Or for some cruder evidence, "bla goy" gets 190 Google hits and "bla goya" gets 174, whereas "bla gaw" only gets 1 (unrelated to the Governor) and "bla gaw ya" and "bla gaw ye" get none.

  24. Boris Blagojević said,

    December 16, 2008 @ 9:21 pm

    @Ran Ari-Gur:
    That's what I did when I wrote 'bla-fucking-gojevich'. I put a secondary stress on 'bla', and pronounced it with /ɑː/, and it sound OK to me. But I'm not a native speaker, of course.
    The lengthening of the schwa comes naturally to me, since [ɑ] is closer to the original [a].

  25. dr pepper said,

    December 17, 2008 @ 3:14 am

    British speakers could just say Blaggerovich.

    Also: "path vowel" and "trap vowel"? I've never heard anyone pronounce them differently in normal speech. I've only heard "path" pronounced "pauth" as a deliberate affectation.

  26. Lane said,

    December 17, 2008 @ 10:17 am

    I was at an event with Peter Bergen, the terrorism analyst for CNN and an Oxford-educated Englishman, who confused me several times by referring to a supposedly famous terrorist leader, Abdullah Razzam. It took me a couple of minutes, wondering why I had never heard of this person, to realize he was doing what Brits and Kennedys do: replacing the glottal stop Americans use to separate "Abdullah" and "Azzam" with an R-sound… But he did not do the British "pat" vowel for Arabic names, I don't think. He seems to have half-Americanized himself, just enough to confuse me.

    Also, there's a paper out there–Mark Liberman sent it to me once–on the PATH versus PALM vowels in American speaker perception when used in foreign names. As most people would probably guess, Americans find the "Irahk" pronunciation more prestigious than the "Irack" one. Which is why it has always struck me as odd that BBC Brits use the "pat" vowel in Vivaldi and pasta. I always wondered if it was anti-snobbery, trying intentionally not to sound like snooty foreigners.

  27. Lazar said,

    December 17, 2008 @ 12:39 pm

    @dr pepper: Huh? "Trap" and "path" use different vowel phonemes for about half or more of the population of England, as well as pretty much all Australians and New Zealanders. It's called the trap-bath split, and it's most definitely not an affectation. It involved a raising of /{/ in various contexts, most notably before voiceless fricatives, so that in standard British English, "path" uses /A:/ (the same vowel as "car" or "palm"), and "trap" uses /{/ (the same vowel as "cat"). Nobody in the entire English speaking world pronounces it as "pauth".

  28. DMV said,

    December 17, 2008 @ 3:11 pm

    Hmmm. I hope G.K.P. didn't take my "objection" in the comments to the other post as "judgmental." I got that impression from this post. I was just genuinely curious and thought I could be missing something.

    I have no background in linguistics whatsoever. Therefore, for my part, I come to Language Log a blank slate, not to make nitpicky judgments about the posters.

    But I could be in a minority. :)

  29. Jarek Weckwerth said,

    December 17, 2008 @ 3:16 pm

    @Lane: "Which is why it has always struck me as odd that BBC Brits use the "pat" vowel in Vivaldi and pasta. I always wondered if it was anti-snobbery, trying intentionally not to sound like snooty foreigners."

    John Wells once wrote about this on his blog (you need to scroll down to 13 Nov 2006), and I think the paper you mention also touches upon this, but I can't seem to find it on my lappie at the moment, so I can't confirm — time to do some tidying up on the disk.

    Supposedly it's about the quality of TRAP in UK English. It's more open (more like IPA [a] or Spanish or Italian /a/) than AmE TRAP. As a result, the Brits tend to feel that it corresponds better than PALM to /a/ in other languages. AmE TRAP is not sufficiently [a]-like, apparently (especially for people from the Northern Cities…), so Americans default on the next-best alternative, i.e. PALM.

  30. Lazar said,

    December 17, 2008 @ 5:16 pm

    @Jarek Weckwerth: I think that quality-based explanation might be even more compelling for Candian English, which has a strong preference for /{/ in foreign words and which tends to use a more open/central [a] for that phoneme; but I suppose it might make sense with British English, because from what I've gathered, not only are [a]-like pronunciations predominant in the North, but they're also becoming quite common in the South as well, although in the past RP used a much closer value for that phoneme.

    Another consideration that strikes me as interesting is phonemic distribution: British English /A:/ has expanded to include historical instances of /A:r/, and it's acquired many instances of historical /{/ in the trap-bath split, so perhaps you could say on balance that /A:/ is big enough, and poor old /{/ could use a few more instances. Likewise, in Canadian English, /A:/ has expanded to include most of the instances of historical /Q/ and /O:/, giving it a very wide distribution, so again it seems like /{/ should receive the foreign words. In General American, on the other hand, /A:/ has only acquired historical /Q/, and /{/ hasn't lost its trap-bath split cases, so perhaps it makes more sense to assign them to /A:/. I dunno, am I overthinking things?

  31. sleepnothavingness said,

    December 17, 2008 @ 8:18 pm

    So am I to understand that tfuckingmesis would be OK but dystfuckingmesis not so much?

  32. Merri said,

    December 19, 2008 @ 9:27 am

    Said pattern (placing the expletive behind the first accent and just before the second) seems to find its justification into articulation rather than English language. In French, there are accents on the first and last syllables of such long names, and in trying to create the French equivalent, 'blagojemerdovitch' sounded wonderfully smooth, while other insertions sounded strange.

  33. Derrick said,

    December 26, 2008 @ 3:33 pm

    A funny cartoon on the subject:

RSS feed for comments on this post