[This is a guest post by Steven Pinker of Harvard University. —GKP]
I agree with Geoff Nunberg that the nucular pronunciation is not the result of a phonetic process that applies across the board in these dialects. It's a lexical phenomenon, though one with a phonetic motivation, and I didn't distinguish the two in my Times Op-Ed piece. In this regard I think it's related to Febuary, jewlery, iurn, purty, and Kirsten (from Christine). I also agree that there is an analogical attraction to words like binoculars, particular, circular, vascular, and muscular, but it is one that may have prevailed because the weak perception of the order of the adjacent sonorants in nuclear failed to resist the tug.
But I don't agree with other aspects of the analysis.
I don't hear a yu in the middle syllable of nucular, but rather a schwa; I suspect the perception of the u is influenced by the spelling. Nor is it clear that this is a faux-bubba disaffectation (both great words, by the way), or a sign of ignorance. Jimmy Carter, during his 1980 debate with Reagan, boasted that he was a nucular engineer, and both Eisenhower and Mondale have been credited with the pronunciation as well. Last Friday I spoke to the Strategic Studies Group at the Naval War College in Newport, RI, and heard the pronunciation from two of the senior analysts there. According to Merriam-Webster, "Though disapproved of by many, pronunciations ending in [kyələr] have been found in widespread use among educated speakers including scientists, lawyers, professors, congressmen, United States cabinet members, and at least two United States presidents and one vice president. While most common in the United States, these pronunciations have also been heard from British and Canadian speakers."
I also doubt that nucular represents conscious linguistic slumming by Bush or Palin. People generally end up with the accents of their late childhood and early adolescent peers, so Midland and Houston were the formative influences on Bush's accent, rather than Kennebunkport and Andover. (It seems unlikely that Bush is faking his entire Texan accent.) And I wonder whether nucular is enough of a bubba shibboleth to grant a politician more points than he or she loses among the mainstream pundits.
Since my op-ed came out I've received many emails with other analyses. One suggested that nucular involves a back-formation incorporating nuke, but the timing is wrong: nucular dates back to the 1950s, but nuke was still unfamiliar in the mid-1970s. (In 1976, a Harvard graduate-school friend of mine saw the bumper sticker "No Nukes" and thought that it was a racist slogan!) Another correspondent suggested that nucular betrays an ignorance of the connection to nucleus. But do we know whether these speakers also pronounce nucleus as nuculus? And do we know whether a nuclear-sayer could give a clearer explanation of the relation between an atomic nucleus and nuclear power than a nucular-sayer? We all use technical terms with pronunciations that obscure their original composition, such as helicopter (helico + pter, "spiral wing"), fax (from fac + simile "make similar" transmission), modem ("modulator-demodulator"), and so on.
I think we can all agree that there's a master's thesis in here for someone: we lack good data on the regional, class, and age distribution of the two pronunciations, and the linguistic and psychological factors that they correlate with.
— Steven Pinker