Pinker on Palin's "nucular"

« previous post | next post »

In an op-ed in Saturday's New York Times, Steve Pinker tries to explain or extenuate some of Sarah Palin's linguistic derelictions, real and alleged. Among other things, he says that Palin shouldn't be taxed for saying "nucular," which is

 …not a sign of ignorance. This reversal of vowel-like consonants (nuk-l’-yer —> nuk-y’-ler) is common in the world’s languages, and is no more illiterate than pronouncing “iron” the way most Americans do, as “eye-yern” instead of “eye-ren.”

I agree with Pinker's overall conclusion that Palin shouldn't be on the hook for this one, but I think both of the claims here are wrong. It's not a phonetic process, and if it isn't exactly a sign of ignorance, it's the legacy of it. 

First off, if this were a case of a phonetically motivated sound change, like the one that turns Februrary into "febyuary," it would be a more complicated and mysterious one than common-or-garden variety metathesis or dissimilation (which is another route that people have gone in trying to explain this one phonetically.) You'd have to explain not just how /y/ and /l/ came to be transposed, but why the new version should wind up not, as Pinker gives it (in Times-reader-friendly phonetics) as "nuk-y’-ler" but rather as "nuk-yu-ler." Where could that that  /u/ have come from? And why aren't people tempted to go around saying "likular" (as in "The first outcome is likular than the second") which would be the natural result when likelier underwent the same process that nuclear did. 

In fact nuclear isn't a hard word to say, the way February is — not even three times fast. But when the word first entered the popular language it was a relatively obscure one. So it isn't surprising that some people would have reanalyzed it on the model of more familiar items like molecular and particular. In short, thisis  an analogical reformation, not the result of a phonetic process. (I dealt with this all some years ago in my book Going Nucular; you can find the title essay here.)

What about Pinker's second point, that Palin's pronunciation of the word is "not a sign of ignorance"? Well, not of her ignorance, anyway. It's fair to assume that "nucular" was the dominant pronunciation in the ambience she grew up in, as it was for Bill Clinton, and that she acquired it "naturally." But at its inception, the "nucular" pronunciation was the result of ignorance, or at least of unfamiliarity with the item, which is why it tends to be more frequent in the varieties used by less-well-educated speakers (or maybe I should say it's less frequent in the varieties used by literate ones). 

That doesn't mean that speakers who pick up the "nucular" pronunciation from family, friends, or teachers can be accused of ignorance themselves — they weren't the ones who came up with the reanalysis that motivated the pronunciation. But it does explain why such speakers might want to correct their pronunciation once they're made aware of it — not just because the "nuclear" variant happens to be used by better educated speakers, but because it conforms more closely to the word's orthography, and because this is, in its nature, a word that belongs to literate discourse. 

Palin has to be aware that many people consider her pronunciation nonstandard, and she (or her handlers) seems to have made some effort at correction, which is presumably why she pronounced the word as "new clear" when reading off the teleprompter in her convention speech. Since then, though, it's been "nucular" all the way, which may be part of the "let Palin be Palin" strategy. 

George Bush, on the other hand, can't be exculpated for saying "nucular." After all, it isn't likely that that version was frequently heard at Andover, Yale, or the Kennebunkport dinner table. In his mouth, it's what I've described as a "faux-bubba" pronunciation. (As I noted in the "Going Nucular" piece, this disaffectation* is not uncommon among Pentagon and DOD types when they're referring to weapons, though not when they're referring to families or medicine.) And for some reason — I feel this strongly, but am not up to justifying it right now — deliberately down-shifting to a misanalyzed pronuncation of nuclear is a lot more culpable as linguistic slumming goes than merely dropping a g now and again.

And what of Palin's g-dropping, which Pinker also refuses — too quickly, I think — to condemn? I'll get to that in another post.    


*An attested word, it turns out — who knew?



  1. Marinus said,

    October 5, 2008 @ 3:54 am

    I take it Feburary is a typo for February? That had me confused for a little while.

  2. Kevin Iga said,

    October 5, 2008 @ 4:02 am

    What's wrong with a variant pronunciation, even when not caused by a general phonological change? We have two well-accepted pronunciations of "either": [i: ∂r] and [ai ∂r]. The second derives from interference from German when the house of Hanover took over the English throne. One might go around saying that the second one is therefore "wrong", and that we should prefer the first. But do we?

    Similarly with aluminum (US) and aluminium (UK). One might try to make a case in these non-phonological variations that one is more "correct" because it occurred earlier, or because it is more logically derived etymologically, but do we? And if we did, would we therefore have a better English language?

    As far as communication is concerned, as far as I can tell, everyone understands both nucular and nuclear quite readily, and the only people who are really even noticing the variation are a few crusaders.

    Nunberg says, "But it does explain why such speakers might want to correct their pronunciation once they're made aware of it — not just because the "nuclear" variant happens to be used by better educated speakers, but because it conforms more closely to the word's orthography, and because this is, in its nature, a word that belongs to literate discourse."

    The first argument says that the speech of the educated has more prestige, and therefore, one should adopt it. I would say that it depends on the context. In Hawaii, where I grew up, standard American English was the prestige dialect when applying for an office job, but Hawaiian Pidgin English had advantages when running for political office, in showing that you could relate to the common person. Why always side with the academics? It's not always in your interest to do so.

    As to orthography, that's subject to change as well, as the "aluminum/aluminium" case demonstrates. We sometimes hear complaints that orthography should be altered to fit the current pronunciation rules, so why should we try to fit pronunciation to orthography? Do you really pronounce "sherbet" as it is spelled? "Wednesday"? or in the case given in the original post, "February"?

    As to whether this word "in its nature… belongs to literate discourse", I'm not sure I know what is meant. All of the above examples can be used in writing, and are. Conversely, "nuclear" is a a word that appears in everyday conversation, from concerns about world affairs to discussing occupations of relatives or friends in the nuclear industry. "Literate" is not synonymous with "educated", or "formal". But even if these are meant, the word "nuclear" has been used by uneducated people, in roughly the same contexts as for educated people (not counting physicists and oncologists and other such specialists): fears of terrible modern weapons. And as for formal/informal distinctions, "nuclear" has hung around the informal register long enough to have a shortened form, "nuke", that is widely recognized, and this form has generated a verb form "to nuke" and a noun form. It is informal enough to have certain widely-recognized figurative uses: in microwaving food, in a massive attack in a video game, or in loud demonstrations of anger, to name a few.

    Nunberg also claims that "nuclear" is not difficult to say, unlike "February". Actually, this is a matter of what you're used to saying. On an objective level, both involve consonant clusters of similar quality: cl versus br. If anything, "br" has the advantage of both consonants being voiced. The best case I can make against "February" is that the point of articulation similarity in "bru" makes it hard to distinguish from "bu" for the hearer. Both "nuclear" and "February" are difficult for some non-native speakers and in both cases, "nucular" and "Febyuary" are much easier, because of the lack of the consonant cluster. Not that we should judge English by its non-native speakers, but I want to point out that ease of pronunciation is in the mouth of the beholder.

  3. Peter said,

    October 5, 2008 @ 4:03 am

    If there are distinct differences in (say) the average levels of formal education or of other social characteristics of people who pronounce the word "nuclear" one way versus another, then one can readily imagine people making (witting or unwitting) decisions to pronounce the word a certain way in order to identify with, or to be identified with, one social group or the other. Former Austalian Prime Minister Bob Hawke, who had been both a Rhodes Scholar and a trades union official, was well-known for his ability to switch his accent and vocabulary depending on the audience he was addressing; it is not clear that this behaviour was ever deliberate or witting.

    It is precisely BECAUSE the "nucular" pronunciaton is not used by people with formal education that George W Bush uses it (whether wittingly or not). I think your criticism of him unfair, since it privileges the standard pronunciation of the word by the formally-educated over other pronunciations. By what right is one pronunciation privileged over another? Is this not a form of prescriptivism?

  4. Peter Metcalfe said,

    October 5, 2008 @ 4:34 am

    She did pronounce nuclear as "new-clear" at the RNC convention, primarily because the teleprompter spelled the word (as is standard practice) as "new-clear". So I think she (or at least some people within her circle) consider new-clear as the formal pronunciation as opposed to the more folksy nucular.

  5. Andy J said,

    October 5, 2008 @ 5:05 am

    Is there any evidence as to how 'nucular' speakers then pronounce the word nucleus? I would have thought this discontinuity might well cause some speakers to question their nucular pronunciation. Or perhaps the two words are seen as utterly separate in that, apart from in scientific circles, nuclear in the popular weapons sense is a concept devoid of scientific meaning.

  6. Jack Collins said,

    October 5, 2008 @ 5:29 am

    Lewis and Short record that the earliest spelling of the Latin nucleus, "kernel" was nuculeus, a diminutive of nux, "nut." The related nucula, small nut, was also attested as a female name.

    A quick search of the OED finds dozens of words ending in -cular, in particular, the adjectival forms of nouns ending in -cle or -cule, deriving from the Latin diminutive suffix -culus. There are NO words (aside from compounds of the adjective clear) ending in -clear except nuclear and compounds thereof.

    I think this is a case of sound-change by analogy, pure and simple. You can't even call it an eggcorn or folk-etymology, since the -cleus in nucleus really is a variant of the diminutive suffix -culus.

    [(myl) See "'Nucular' solecism traced to 200 B.C.", 1/2/2004, where the definitive finger of blame is pointed at Plautus. More seriously, this does seem to be a case of circular (circlear?) analogy, where the end is near the beginning — by accident. But it's merely near the beginning, because the original form is NOT nuculus but rather nuculeus — and (except when analogy is involved) etymology ain't horseshoes.]

  7. Tristan McLeay said,

    October 5, 2008 @ 6:33 am

    Kevin Iga, “ February ” and the “ nucular ” pronunciation of “ nuclear ” both still have a consonant cluster. The English orthography masks this, by spelling /juː/ and /jə/ as a long u, but the /kj/ and /bj/ still definitely exist.

    Geoff Nunberg, you ask “ Where could that that /u/ (in the second syllable of ‘ nucular ’) have come from ? ”. English C/j/ clusters are almost exclusively followed by either a /ə/ or a /uː/, and if it's the former the orthography clearly indicates that it comes from an /uː/ originally. Furthermore, there is (at least for me) a lot of overlap between an /l/ and a /u/. A small amount of coarticulation between the schwa and the following lateral, along with some subconcious knowledge of English sound patterns, could I think easily cause people to interpret the [kjəl] sequence as standing for /kjuːl/, and then to use this.

  8. Faldone said,

    October 5, 2008 @ 7:43 am

    Febyuary from February seems to me to be more of a tendency to drop Rs after bilabials or labio-dentals. Other examples are infastructure and speak. The glide in Febyuary is probably from some other force.

  9. Peter F said,

    October 5, 2008 @ 8:12 am

    If you remember on of the Couric-Palin interviews, Palin wanted to use the word 'caricature' but was unsure how to pronounce it. She simply uttered 'cari-', paused, and Couric helped her out, not by correcting her, but by providing a synonym, 'mock'.

    It looks that in many of these mistakes, ignorance looks like the most plausible explanation.

  10. Chris Johnson said,

    October 5, 2008 @ 9:16 am

    I consistently pronounce 'nucleus' as [nukliəs], and I imagine most, if not all, other 'nucular' speakers do too. I don't think I've ever heard "nuculus".

    I use the two pronunciations for 'nuclear' in more or less free variation. I never notice which form a speaker uses unless I'm specifically listening for it. I think I am more likely to use 'nucular' when saying the word 'thermonuclear' than when saying 'nuclear' itself. I have no idea whether my usage varies by domain or not. It doesn't seem like I had any trouble treating 'nucular' as an adjectival form of 'nucleus'.

    I think the 'nucular' pronunciation has hopped over social and class boundaries more than people realize. It's not a very common word. I think it may be spread more by mass media than by the family dinner table. I had an upper middle class upbringing that valued education, and I even wanted to be a physicist in 5th grade. Nevertheless, I think 'nucular' was my usual form growing up. (I've shifted towards [nukliɚ] over the years, but my girlfriend still catches me sometimes.) Bush could very well be 'slumming' linguistically, but I'm sure there are more than a few kids from elite families that picked up 'nucular' from the culture at large.

    (I've known for a long time that 'nucular' was a less prestigious pronunciation, but in the past few weeks I've been shocked at just how stigmatized this feature of my speech is – especially by people 'on my side' politically. As a speaker of a pretty standard variety of American English, this is a new and unpleasant position for me.)

  11. Alan Gunn said,

    October 5, 2008 @ 9:42 am

    Military (mis)pronunciations could probably be a whole topic. I remember being surprised years ago that President Carter, who served on nuclear submarines in the navy, said "nucular." A few years ago I learned from my son, an army officer, that most people in the army call IEDs "improved," rather than "improvised," explosive devices. I haven't heard the "improved" version on TV, though, so perhaps the public-relations people get special instructions.

  12. Mary Kuhner said,

    October 5, 2008 @ 10:18 am

    I'm a long-term West Coast US academic from an academic family, and I also suspect that I use both pronunciations in free variation. I hear the two as exactly the same word–that is, I can hear that the pronunciations are different but it doesn't seem to affect my parsing at all. I wouldn't notice, I think, if a speaker alternated pronunciations sentence to sentence.

    I'm having trouble producing spontaneous pronunciations now that my attention has been drawn to the subject, but I think nucleus has the same two pronunciations for me. I tend to encounter both of these words in the context of cell biology, not physics or military affairs.

    When we critique student presentations my colleagues are fairly picky about speaking register, but I've never heard this word corrected in either direction. I don't think it's seen as an issue in my particular speech community.

  13. Andrew said,

    October 5, 2008 @ 10:31 am

    I am struck by the distinction you draw between 'nuk-y'-ler' and 'nuk-yu-ler'. I had always supposed that the spelling 'nucular' represented something like 'nuk-y'-ler'; I have heard people (in the UK) saying 'nuk-y'-ler', whereas I am not aware of hearing people actually sounding the u. (I would pronounce 'particular' as something like 'partic-y'-ler', so I do not find it surprising that this spelling is used to represent this pronunciation.) If Palin and others actually do sound the (invisible) u, I think that makes it more surprising than it would otherwise be.

  14. dd said,

    October 5, 2008 @ 11:20 am

    The "op-ed" link doesn't seem to go anywhere…

  15. Adam Buchwald said,

    October 5, 2008 @ 11:29 am

    Re: whence the /u/, it comes from the grammar (of course).

    As another poster above mentioned, there's a pretty strong phonotactic restriction in English where "y" (or /j/ in IPA) following tautosyllabic consonants must be followed by /u/. So, for example, we have words like cute (/kjut/) but not */kjit/, */kjat/, */kjot/. This has been argued by many to be a piece of evidence indicating that /ju/ is actually a diphthong with an onglide (as opposed to other English diphthongs). While it is certainly the case that the /u/ part of this vowel gets reduced (as in most pronunciations of "popular"), it still is 'underlyingly' /ju/.

    [As a side note, there is a strong sound change sweeping the nation in which the entire /ju/ diphthong is being reduced to schwa, heard with increasing commonality in words like accurate (/&k@r@t/). Pay attention: someone you know says this (and you might even say it too).]

    Re: "phonetically-driven", the change certainly seems to be driven by sound structure considerations, far more than the "iron" example.

    It seems that the "wrong" pronunciation of nuclear is much less phonologically marked than the "right" one; the hiatus has been removed between syllables 2 and 3, syllable 3 now has an onset, and depending on your analysis of /ju/, one could argue that the consonant cluster at the beginning of syllable 2 has been removed (if we analyze /ju/ as a diphthong, syllable 2 begins with a singleton /k/) as well. So one could argue that people who say nucular merely have their markedness constraints working overtime.

    In this sense, it differs from iron in which the "wrong" pronunciation has 2 onset-less syllables (resulting in hiatus) rather than a second syllable which is CVC as in "eye-ron." So that example may be a red herring, but it doesn't negate the differences in phonological markedness between nuclear and nucular.

  16. Kevin Iga said,

    October 5, 2008 @ 11:43 am

    Tristan McLeay,

    Correction accepted. But if anything, it strengthens my point.

    About foreign speakers: I was thinking of Japanese, which does not typically allow consonant clusters, but does allow Cja, Cju, and Cjo, (and has a syllabic n that might be interpreted as forming a consonant cluster with the next phoneme if that is a consonant).

    More generally, it is conceivable that someone might make the argument that consonant clusters with glides are "easier" than consonant clusters with liquids. I'm not convinced this is a universal, rather than particular to English speakers, or particular to me. But I suppose a frequency table for various languages' use of such consonant clusters might be helpful (and if the existence of one implies the existence of the other). Anyone know of such evidence? In that case, "kli -> kjul" might be viewed as part of a phonological process, albeit applied very sporadically in English.

    About contrasts with "nucleus": The original post alluded to contrast with "nuclear" in phrases such as "nuclear family", which is never pronounced "nucular family". This lends more credence to the idea that for many speakers, "nuclear" in "nuclear weapons" is not a morphologically derived form of "nucleus" but has a separate lexical entry.

  17. Bob Ladd said,

    October 5, 2008 @ 12:08 pm

    Whether nucular originated as a metathesis or as an unclassifiable analogically-influenced change to a new scientific word (like aluminum – aluminium [thanks, Kevin Iga] and polyethylene – polythene and probably others), the fact remains that it is apparently used by a lot of speakers that the people at Language Log Plaza would have to recognise as "standard" (cf. the comments by Chris Johnson and Mary Kuhner). Seems to me this puts Geoff Nunberg in the unenviable position of deploying a prescriptivist shibboleth when it suits his political purposes. I also find the idea of deliberate "faux-bubba" style unconvincing – everybody uses different registers talking in different contexts (cf. Peter's comment about former Australian Prime Minister Bob Hawke), and everybody engages in what Erving Goffman called "presentation of self". Few people are both cynical enough and talented enough to achieve significant and convincing modifications to their overall presentation-of-self on demand. So I doubt that either Palin's or Bush's manner of speaking is calculated; on the contrary, I think it's about what you would expect of people who, for different reasons, never really mastered careful formal style and now find themselves in positions where a command of that style would be useful.

  18. Dan T. said,

    October 5, 2008 @ 12:09 pm

    I can't help mispronouncing the medication name Metoprolol (a blood pressure medication) as "Metropolol", shifting the "r" in analogy to "Metropolis".

  19. Robert Coren said,

    October 5, 2008 @ 1:04 pm

    I have a vague and distant memory (which I hope somebody else of A Certain Age can confirm or refute) that President Eisenhower habitually said "nucular", which would have given that variant something of a running start.

    Personally, I found Palin's "nucular" a lot less bothersome than her "eye-rack". Now, that's iggerant.

  20. James said,

    October 5, 2008 @ 1:20 pm

    Bob Ladd,

    Seems to me this puts Geoff Nunberg in the unenviable position of deploying a prescriptivist shibboleth when it suits his political purposes.

    That's a weird thing to say. Geoff was completely explicit about this, writing "I agree with Pinker's overall conclusion that Palin shouldn't be on the hook for this one," and at the end explains that even if her pronunciation derives from an incorrect re-analysis, it isn't Palin's re-analysis so the ignorance is not hers.

    So where is the anti-Palin prescriptivism?

  21. Arnold Zwicky said,

    October 5, 2008 @ 1:35 pm

    Robert Coren asks about Eisenhower and nucular: yes indeed. Geoff Nunberg mentioned this in his 2002 "Going Nucular" Fresh Air piece archived here and then in his 2004 book with this title.

  22. Bob Ladd said,

    October 5, 2008 @ 1:59 pm

    @James: I was referring to Geoff's overall take on nuclear/nucular, including the idea that we should be in the business of "exculpating" (or not) George Bush for his pronunciation of the word, and the idea that, having been made aware that "better educated speakers" say nuclear, Palin (or her "handlers") should at least consider mending her ways. He presumably wouldn't say things like this about split infinitives or between you and I, because those are well-established shibboleths on which anti-prescriptivist linguists (myself included, of course) pretty much have a party line. But for some reason he doesn't see the party line as covering nuclear/nucular – here he allows his own linguistic prejudices (which I also share) to get the better of him. The style of his arguments about nuclear/nucular strikes me as awfully close to the style of argument deployed by prescriptivists about their own favourite shibboleths, including the Argument from Etymology and the Argument from Authority.
    @Robert Coren: As "somebody else of A Certain Age", I'm pretty sure you're right about Eisenhower. I seem to recall my mother – a dyed-in-the-wool prescriptivist – complaining about this when I was young and Eisenhower was president.

    GN Look, what I'm trying to say is that what makes GWB's "nucular" pronunciation culpable is that it's phony, and not simply that it's incorrect. You wouldn't want a horror of prescriptivism to leave you incapable of applying words like "phony," "affected," "pretentious," "cloying," and so forth to certain choices of variants.

  23. Amy Vaughan said,

    October 5, 2008 @ 2:04 pm

    You say not to give people who say nucular a hard time for being 'ignorant' because it's not their fault, but rather the fault of someone else's ignorance – but I have a hard time seeing why, exactly, reanalysis is the result of ignorance. Reanalysis seems to be as natural (though frequently more conscious) and motivated as formal-phonetic sound change. I'm not actually convinced that this isn't a case of FPSC myself, but I'm more concerned by your tacking up reanalysis to apparently being so dumb one doesn't know how to pronounce something. There are plenty of other cases of language change via reanalysis that don't get chalked up to being ignorant – are you sure that you're not being overly affected by your own cultural and political biases here?

    [GN] Maybe the problem here is that "ignorance" has both normative and factual meanings. When you run into an unfamiliar name and reanalyze or misremember it on the model of a more familiar one — say, recalling the name of Gen. McKiernan as "McClellan" — that's an error born of ignorance in the factual sense of the term, inasmuch as someone more familiar with the name probably would have got it right. That doesn't mean it's culpable or a sign of being dumb, but it doesn't mean either that it isn't a mistake or that you wouldn't try to correct yourself after it was pointed out to you, rather than saying, "Oh, well that was a perfectly natural form of linguistic change." Presumably a lot of folk etymologies (all of them?) have this character. But I think it's delusional to imagine that a descriptivist commitment requires us to banish words like "mistake" and "error" from our lexicon, particularly when those notions play such an important role in speaker's own behavior.

  24. dave allen said,

    October 5, 2008 @ 3:05 pm

    how can anyone from a nation which pronounces aluminium as aluminum take offence at "nuculer"

  25. Andrew said,

    October 5, 2008 @ 4:04 pm

    Dave Allen: well, they also spell aluminium as aluminum, so they can't be accused of simple mispronunciation in that case. (I'm not suggesting they should be accused of it in the case of 'nucular'. But they can be.)

    Bob Ladd: I think one difference between this and other prescriptivist shibboleths is its relative newness. People have been using split infintives for hundreds of years. The word 'nuclear' has been in widespread use for no more than sixty years, and when it was first introduced presumably it was pronouced 'nuclear'. It seems rather surprising that a variant form should develop and become a regular part of the language in such a short time (very short if Eisenhower was using it). So it's to be expected that some people, when first exposed to it, will see it, not as a legitimate variant form, but as a mistake.

  26. language hat said,

    October 5, 2008 @ 4:29 pm

    I agree with Kevin Iga and Chris Johnson, and with Bob Ladd's "Seems to me this puts Geoff Nunberg in the unenviable position of deploying a prescriptivist shibboleth when it suits his political purposes." For the entire length of GW Bush's presidency I have been arguing with people who mocked him for the "nucular" pronunciation, which seems to me entirely unexceptionable, and one result of the politically based mockery (or, in Geoff's case, gentle and nuanced but unmistakable disapprobation) is reactions like Chris's: "I've been shocked at just how stigmatized this feature of my speech is – especially by people 'on my side' politically. As a speaker of a pretty standard variety of American English, this is a new and unpleasant position for me."

    I really wish people would concentrate on the policies of politicians and (in Robert A. Hall's resonant phrase) leave their language alone.

  27. Bob Ladd said,

    October 5, 2008 @ 4:51 pm

    @Dave Allen: How can anyone from a nation which pronounces polyethylene as polythene take offence at "aluminum"?
    But seriously, I agree with the general drift of the comments from Kevin Iga and Amy Vaughan: there is nothing surprising about new technical vocabulary getting reanalysed and incorporated into the language in its reanalysed shape. In the case of polyethylene – polythene and aluminum – aluminium these have settled down as American/British differences, but in the case of nuclear – nucular we instead have a new shibboleth in the making. The only basis I can see for the different treatment is that nuclear is morphologically related to nucleus in a fairly transparent way, whereas for English speakers first encountering aluminum or polyethylene the morphological structure of all those syllables must have been considerably less obvious. So it was easier for alternative pronunciations to get established without attracting the attention of prescriptivists.
    @ Andrew: polyethelyne – polythene can only be a decade or two older than nuclear – nucular, so I don't think it's the recency of the development that is at issue.

  28. Bob Ladd said,

    October 5, 2008 @ 5:53 pm

    Quick reply to GN's responses: I completely agree that we can and should still make judgements about style, and the McKiernan/McClellan example is a clear case of a mistake that most speakers would actually correct. But the issue is (1) whether GWB's nucular really is phony (i.e "faux-bubba"), and (2) whether McKiernan/McClellan is a reasonable analogue for nuclear/nucular. On (1): you may be right, but some of the comments here suggest you may not be. Maybe one of GWB's Andover or Yale classmates is a Language Log reader and will remember clearly that he used to say nuclear; otherwise I don't believe you can be sure that he's putting it on. On (2): I find the analogy unconvincing. Polythene is just as clearly an etymological/derivational "mistake" as nucular, but I don't think anyone would suggest that British speakers should be expected to switch to polyethylene after the misanalysis is pointed out to them. Like it or not, polythene – polyethylene and nucular – nuclear are established variants; McKiernan/McClellan is a different kind of error.

  29. Mark F. said,

    October 5, 2008 @ 8:23 pm

    Jimmy Carter came up, as a user of the "nucular" pronunciation, but I've always thought his version was a little idiosyncratic. He seemed to elide out the second syllable almost entirely, so that it sounded a little like "nuke lah". This is different from the common pattern of distinctly pronouncing "nucyular". Am I remembering right? If so, is this just "nucular" followed by a phonetic transformation? If it's an example of the sound change Adam Buchwald was talking about, it'd be a pretty early one.

    I tried to find audio on the web of Carter saying the word "nuclear", however he pronounced it, but came up empty.

  30. Robert S. Porter said,

    October 5, 2008 @ 9:24 pm

    Perhaps this is just the historian in me, but isn't this issue highly dependent upon the history of the term. I understand Geoff's point that it was originally an error or misunderstanding and in general there is nothing prescriptivist about pointing out an error. But this 'error' goes back 5 or more decades and people have been saying it this way throughout. With the name example, I agree. We shouldn't simply allow McKiernan as "McClellan", but with nuclear it is not a recent change or a proper name. Nuclear clearly is an example of linguistic evolution, even if the impetus to change was fairly immediate and ignorant.

  31. dr pepper said,

    October 5, 2008 @ 10:06 pm

    Are there people who drop the first r in "infrastructure"? I've never heard it myself.

  32. Robert S. Porter said,

    October 5, 2008 @ 11:17 pm

    I'd say it's probably fairly common to drop the first r. I probably do it myself some of the time.

  33. Skullturf Q. Beavispants said,

    October 5, 2008 @ 11:53 pm

    My mother, generally a very articulate woman, pronounces "literature" as something like "lit-a-tcher", with the first r completely elided.

    I agree that more evidence is probably needed before we can decide whether GWB's "nucular" is an affectation (e.g., as Bob Ladd says, witnesses from his younger years saying he used to say "new-clear".)

    And I share the surprise of Andrew above that such a widespread variant as "nucular" can arise so prevalently in such a short time.

  34. blahedo said,

    October 6, 2008 @ 12:05 am

    From the OP: "which is why it tends to be more frequent in the varieties used by less-well-educated speakers (or maybe I should say it's less frequent in the varieties used by literate ones)."

    I am both well-educated and literate, and I have the two pronunciations in more or less free variation, as do many of my well-educated, literate friends. (As an aside, the two pronunciations I have are /'nu.kj@.lR/ and /'nu.klI.jR/ or /'nu.kl@.jR/—note that both have a reduced vowel on the middle syllable, making the swap of /j/ and /l/ both easier and more plausible.) As such, I don't accept Geoff's premise that

    "it isn't likely that that version was frequently heard at Andover, Yale, or the Kennebunkport dinner table."

    —at least, not without evidence. I have caught more than one educated, literate person bitching about W's "nucular" and then using the same pronunciation themselves; and so my presumption will be that this is another instance, albeit possibly more subtle, of "I never split infinitives" or "I never end a sentence with a preposition" or "I never use the passive voice": a prescriptive shibboleth that the prescriber "knows" they use correctly. Some do, some don't, but I'm gonna want some hard evidence here before I believe this distinction is as categorical as Geoff and others are making it out to be. Has anyone got any numbers on this?

  35. Joe said,

    October 6, 2008 @ 1:14 am

    This one I don't hold against her, as much as I honestly see her as another Miss Teen South Carolina (the "some people they don't have maps" girl for those who don't remember). Of course, I'm originally from the Midwest and I originally used that pronunciation because everyone I knew used it.

    It wasn't until my college physics professor gave me crap about it that I changed my pronunciation, but I don't think I was "ignorant" given that I was in an honors course at the time. Honestly, I could probably teach the average person a thing or two about the fundamental forces & particles if I wanted to.

    But it has morphed into a peeve, so it seems unwise to use it no matter how condescendingly you could lecture someone about QCD given the excuse.

  36. Breffni said,

    October 6, 2008 @ 2:57 am

    Bob Ladd:

    In the case of polyethylene – polythene and aluminum – aluminium these have settled down as American/British differences, but in the case of nuclear – nucular we instead have a new shibboleth in the making.

    I'm not sure whether you mean nuclear-nucular is a British/American shibboleth in the making, but I've heard two documentaries recently in which the British narrator (in voiceover, therefore presumably a professional actor – same narrator in both) used "nucular" throughout. Can't remember the titles of the documentaries, unfortunately.

    Dave Allen: "aluminum" was "the name given by its discoverer, Sir H. Davy c 1812 (for which he had first of all used ALUMIUM)… Both alumium and aluminum lived for some time." – OED. Not that that trumps "aluminium", but it's probably worth bearing in mind.

  37. Andrej Bjelakovic said,

    October 6, 2008 @ 5:48 am

    I find it odd that Pinker used the word iron as an example. Because, as far as I know, in all varieties of English (and not just by most Americans) it is pronounced as /'aɪə(r)n/ and never as /'aɪrən/.
    It seems to me that a word like jewelry would serve better as an illustration – outside US /'ʤu:əlri/ is the variant usually used, whereas if I'm not mistaken most Americans say /'ʤu:ləri/.

  38. Andy J said,

    October 6, 2008 @ 6:33 am

    @Andrej B. The more common spelling in British English is jewellery, which points towards the /'ʤu:ləri/ prounciation, whereas the jewelry spelling would accord with the /'ʤu:əlri/ pronunciation. That is not to say that those who pronounce it in a particular way necessarily use the corresponding spelling, of course.

  39. Andy J said,

    October 6, 2008 @ 7:01 am

    Polyethylene/Polythene. I have a half-memory tucked away in the back of my mind (a very cluttered place), for which I have so far been unable to find any supporting evidence, that polythene was a trade mark of the British company Imperial Chemical Industries (ICI) who first developed polyethylene commercially. This was the reason why the spelling 'polythene' was adopted in the UK. The company certainly trade-marked Perspex, a word which is still used generically.

  40. Andrew said,

    October 6, 2008 @ 9:20 am

    Bob Ladd: I don't think polyethylene/polythene is a fair comparison for nuclear/nucular. First, because it's a British/American distinction, so people on one side of the divide might well not know that the other form exists, or, if they do, that it relates to the same stuff. Second, because they are spelt differently, so no one can accuse the other side of simple mispronunciation. Third, because it's hard to see how this differentiation could arise by accident; it seems likely to be the result of a decision made by someone (as, I believe, aluminium/aluminum is), so even before I saw Andy J's comment above I was assuming something like that had happened. On the other hand I find it hard to believe that someone decided to introduce a new pronunciation for 'nuclear' (though you never can tell; if anyone has evidence that this happened it would be interesting to hear); it seems more likely that the first person to say 'nucular' was making a mistake and would have accepted correction.

    I agree that now, when the pronunciation has been in use for a least fifty years and is the accepted pronunciation in some communities, it is silly to go on calling it a mistake. But given that it began as a mistake, and this quite probably happened within living memory, I don't find it surprising that people react to it in a different way from longer-established linguistic variations.

  41. Steve said,

    October 6, 2008 @ 10:56 am

    In Britain one not uncommonly comes across the word 'burglar' pronounced 'burgular'. Unlike 'nucular' this is definitely considered a non-standard (or, many would say, 'uneducated') usage, and I cannot imagine any candidate for high political office using it, but all the same it is definitely out there. This might lend a little support to the suggestion that some speakers find the consonantal cluster k or g/l slightly problematic to pronounce, and therefore insert a 'u'.

    The differences between 'aluminum' and 'aluminium' and 'polyethylene' 'polythene' are surely a red herring here, since they result from different spellings and punctuations dating back to the original coining of the words.

  42. J said,

    October 6, 2008 @ 1:56 pm

    While it is certainly open to debate whether or not GWB specifically pronounces nuclear "nucular," I think it's not unfair to conclude it's of a piece with his quite certainly self-consciously developed Texas-speak. I haven't the will to look up the primary sources on this right now, but having been raised by a patrician family with ties to New England as much (if not more than) Texas, GWB certainly didn't simply come by his Texas affect by accident. If the accounts by those claiming to be knowledgeable are to be believed, he (some time ago) consciously adopted the "Big Texan" persona, in part because he related to and enjoyed it, and in part because it helped him win constituencies for first himself, and then his father, in Texas and beyond. Certainly, he sounds little like his father, mother, or siblings, who have acquired less Texan-ness despite sharing a similar amount of Texas background with GWB.

    I don't know if "nucular" is standard in Texan, but like I said, it would be of a piece with GWB's "I'm Like You, a Regular Joe Sixpack Kinda Guy" persona, which, while certainly isn't feigned, could certainly be called a persona he, at one point, consciously chose to adopt. Not to mention his association with Karl Rove, who loved intricate machinations of perceptions and nuance of political pandering, could lead someone to the not-unfair conclusion that, while not necessarily a conscious choice each time he says it, his use of "nucular" is part of a larger package of mannerisms he has intentionally acquired and deployed, to present a (false) image of being a non-elite, average educated, normal, straightforward average American — who just happens to be part of an incredibly established political family. If one can't call this, on the whole, false and phony, then we may as well consider Madonna to be British at this point. So while Bob Ladd may be right that there's scant evidence that the Prez is deploying nucular strategerically, I think the idea that his folksiness and apparent disfluency is, at least in part, a conscious play for faux-bubbery.

  43. Debbie said,

    October 6, 2008 @ 3:48 pm

    Here's the NY Times link again:

  44. Peter said,

    October 6, 2008 @ 5:12 pm

    Mention of other examples reminds me of "film", of which a variant pronunciation is sometimes "fillum". Presumably these two variants have been around longer than has the word "nuclear".

    (There's on old Australian joke that film director Philip Adams was known as "Philp" to those people who said "fillum".)

  45. Gordon Sabaduquia said,

    October 6, 2008 @ 7:12 pm

    Regarding Dwight Eisenhower's pronunciation of "nuclear" — he pronouced the word [nuk-l’-yer], the standard pronunciation, in his speech December 8, 1953. "Atoms for Peace" before the General Assembly of the United Nations. See the Internet archive of presidential speeches about 8 minutes 29 seconds into the speech I am of a "certain age" and cannot recall Eisenhower, Kennedy, Johnson, Nixon, pronouncing nuclear as /nu kyu ler/. That Carter did so is also documented in his public addresses using that word. Reagan and the George Bush, sr. pronounced nuclear [nuk-l’-yer].

  46. Ellen K. said,

    October 6, 2008 @ 8:59 pm

    @ Andrej Bjelakovic. Right here in the middle of the U.S. I pronounce jewelry with two syllables, /'ʤu:lri/.

  47. Ole Stig Andersen said,

    October 7, 2008 @ 12:27 am

    In Danish and several other non-English languages the Bush pronunciation of nuclear does not sound like nucular but like "new killer". Jokes about the intrinsic connection between his strange pronunciation and his mass-murderous policies abound.

  48. Linda Sewell said,

    October 7, 2008 @ 2:23 pm

    Interesting topic…I always wondered about LBJ's pronunciation of "Pedernales" ("Per-den-alis") – the Hill Country river located near LBJ's home in Texas. We've asked the locals how they pronounce this name, and responses were pretty-much split. So, when in Rome…
    Personally, I believe that Palin's (mis-)pronunciation of nuclear indicates (to many on the Left, anyway), that she is sympatico with GWB – a link that she would do best to avoid…

  49. Martyn Cornell said,

    October 8, 2008 @ 9:08 am

    Admittedly it's not a word that comes up in everyday conversation much in Britain, but I don't think I've ever heard anyone in the UK say "nucular" – always "newclear".

    And since everyone is familiar with "unclear", I find it hard to see why its rhymning anagram "nuclear", when it arrived in popular discourse, might be analysed as like "particular", to be mispronounced "nucular" – nobody says "I'm afraid your argument is uncular to me".

    If you want a similar shibboleth, though, I'd suggest "advertisement", which is pronounced by some in the UK "advert-EYES-ment", based, presumably, on the pronunciation of the verb advertise, a perfectly acceptable way of saying the word in Dublin, but which would get you laughed at in London.

  50. David Marjanović said,

    October 8, 2008 @ 8:04 pm

    Incidentally, it's no longer polyethylene to chemists, it's polyethene.

  51. Steve Morrison said,

    October 9, 2008 @ 1:34 pm

    Michael Quinion has a good discussion of the aluminum/aluminium matter here:

  52. Rachael said,

    October 13, 2008 @ 8:01 am

    @Kevin Iga:
    "Do you really pronounce "sherbet" as it is spelled?"

    Yes. How do you pronounce it? I've never heard any other variant.

  53. Anonymous Cowherd said,

    October 14, 2008 @ 7:50 pm

    @Rachael: "Sherbert" is an alternative spelling and pronunciation of "sherbet". Some people may deploy the "sherbert" pronunciation even when reading aloud a "sherbet" spelling, or vice versa. It's the same deal with "alumin(i)um". (The difference is that "alumin(i)um" is a UK/US shibboleth, while "sherbe(r)t" sees both pronunciations used within the US, and maybe the UK too, I dunno.)

    "Poly(e)th(yl)ene" is a new one on me. It seems to be a case of UK and US having two completely different words for the same thing, a la "lift/elevator", "bonnet/hood", "lorry/truck", and so on. In this case, the UK and US words happen to sound vaguely similar, that's all. These words are UK/US shibboleths, but their "shibbolethness" is totally unrelated to pronunciation. Americans pronounce "polythene" and "lorry" the same way Brits do; we just don't pronounce them quite so often.

    Then there's "advertisement", "laboratory", and the like, which have only the one correct spelling in both US and UK, but which US and UK pronounce differently. (Martyn's Dublin pronunciation of "advert-EYES-ment" is also the US pronunciation; the London pronunciation he's thinking of is presumably "ad-VERT-issment".) These are UK/US shibboleths based on pronunciation.

    (There are also UK/US shibboleths based on spelling but not pronunciation; "recognized" versus "recognised", for example.)

    Finally, there's the "nuclear/nucular" thing that Geoff was actually talking about. This is a shibboleth, but not one drawn along UK/US lines. This is not a case of two different words for the same thing. This is not a case of two variant spellings for the same word, as with "sherbert"; "nucular" is not now, nor will ever be, an English word. It's merely a (way of representing textually an) alternative pronunciation for the word "nuclear". The "nucular" pronunciation, like other non-standard pronunciations, has a bit of a stigma attached to it, at least in the US, due partly to its non-standardness and partly to its being over-used by people who are otherwise dimwits: guilt by association.

    I'd say the most well-known shibboleth in the "nucular" category — other than "nucular" itself — would be "libary" (for "library"). Geoff mentioned "Febyuary", but not "libary"; someone else mentioned "infastructure". (I'm not sure I've heard anyone say "infastructure"; at first I thought a more reasonable dialect pronunciation would be "infastucture", but that sounds more like a baby-talk pronunciation.) Another possibility would be the pronunciation or lack thereof of the "H" in certain words: "humble/umble", "humor/yumor".

  54. Rachael said,

    October 15, 2008 @ 6:22 am

    Anonymous Cowherd:

    Ah, I see. Thanks.

    As a non-rhotic British speaker I wouldn't make any difference between "sherbet" and "sherbert" anyway.

    I wonder whether the "sherbert" variant arose via non-rhotic speakers inserting an (unpronounced) R into the spelling, and then rhotic speakers pronouncing it?

  55. Aaron Davies said,

    October 17, 2008 @ 1:40 pm

    I imagine there's quite a few Americans whose only knowledge of polythene comes from Abbey Road….

  56. Estella said,

    October 18, 2008 @ 12:00 pm

    I pronounce 'February' as it's spelt, and it's not at all hard to pronounce. If you are talking assimilation, it's actually easier to pronounce that way, because an 'r' is articulated closer to a 'b' and an 'oo' than a 'y' is. But of course the way you learn to say it becomes easier for you because it becomes motor programmed that way in your brain.

    No one would say 'likular' because 'likelier' derives from 'likely' so they already have 'likely' motor programmed into their brain. However, there is no 'nukely' word from which 'nuclear' follows. Similarly, with 'unclear', we already have 'clear' programmed in our brains as the root of the word. But 'nuclear' is a word people learn as a whole word (and probably before 'nucleus' nowadays, with so much emphasis on nuclear weapons) so there is nothing on which to hang it. And there are a lot of adjectives that do end with '-ular' so it is a motor pattern that the mouth often forms into. It makes sense to me as a mispronunciation (although, as a Brit, I never hear it here – probably because we pronounce it 'nyoo-clear' so 'nyoo-kyoo-lar' would be real mouthful!). And then as a mispronunciation, it's become more and more accepted – and that is how pronunciations change over time. And it is able to happen quite easily in the English language because our spelling is so inconsistent anyway that pronunciation often doesn't reflect spelling.

  57. Irene said,

    October 27, 2008 @ 5:16 pm

    To Aaron Davies – I was sure I had never before heard the word polythene, until you reminded me of Polythene Pam!

  58. AcTurner said,

    February 2, 2009 @ 11:24 am

    Pronunciation is a social marker with nucular being a sure sign of the unsophisticated. President Carter, I believe, served on a nucular submarine nontheless. Nucular is a southern-rural corrolary of grits and gravy. Palin's usage begs for the rural end of the continuum.

  59. AcTurner said,

    February 2, 2009 @ 11:27 am

    Who said it? "nuclear physics is neither new nor clear."

  60. MikeyC said,

    July 29, 2009 @ 4:19 am

    Langauge Hat said: I really wish people would concentrate on the policies of politicians and (in Robert A. Hall's resonant phrase) leave their language alone.


    If we had done that in the past, LH, there'd be no Greek comedies, no clowns, etc., no King Lear, Yes Minister, and so on.

  61. Lee said,

    May 19, 2011 @ 5:42 pm

    Marinus: There's a typo in your mention of the typo. It was Februrary.

RSS feed for comments on this post