anti-Semetic

On the spelling of this word with the letter E (rather than I) and its pronunciation with [ɛ] (rather than [ɪ]), both recently noticeable in my country because of the great increase in anti-Semitic rhetoric and actions here: worth systematic investigation.

A few examples, with minimal commentary, beginning with Amy Schumer on Kanye West, in this YouTube video:


(#1) In the video, she pronounces the word twice, both times with [ɛ]; in the cover image above, it’s spelled with the letter E, but everywhere in the text the word gets its standard spelling, with the letter I

Next, a visual from a Daily Mail (UK) story on the Goyim Defense League, “Hate group called Goyim Defense League is blamed for flooding six states with thousands of anti-Semitic flyers that claim COVID response is masterminded by Jewish leaders including heads of CDC, Pfizer and BlackRock”


(#2) The word is spelled with the letter I throughout the story

Then an editorial cartoon, by Chip Bok for the Tampa Bay Times 11/8/22:


(#3) Here we have only text, no speech; but it’s surely significant that the paper’s reporting on anti-Semitism has the word spelled with the letter I throughout

Then a WBNS-TV (Columbus OH) news report (YouTube here) that displays the header:

Racist, anti-semetic phrases found spray painted in Ohio State resident hall

though of course the tape has [ɪ] throughout, and all the other text has the letter I.

Finally, in much the same vein, a website entry with the header:

Ye’s recent Twitter anti-Semetic escapade ought to be condemned in full

while the body of the posting begins:

Last week, rapper Kanye West posted anti-Semitic comments on his social media platforms. The statements posted on his accounts are vile, and there is no excuse for Kanye West’s behavior in promoting these fringe conspiracy theories. NBC News reported that because of his disgusting anti-Semitic post, Twitter and Instagram temporarily froze his comments.

and continues with the standard spelling throughout.

What I do. Now, one of my functions is to notice and record stuff — I’m famous for the little notebook and pen in my shirt pocket — to accumulate examples, to characterize the phenomena as best I can and point them out to others (who might be able to study them systematically, using resources and tools not available to me), and (insofar as I can) to say why they’re of interest.

Occasionally, I’m able to provide answers. But I generate far more questions than I could possibly try to answer on my own. So mostly I’m a source of research topics, not results. As at the top of this posting:

On the spelling of this word with the letter E (rather than I) and its pronunciation with [ɛ] (rather than [ɪ]), both recently noticeable in my country because of the great increase in anti-Semitic rhetoric and actions here: worth systematic investigation.

At this point, I searched to see if Ben Zimmer had looked at the phenomenon — it’s the sort of topic that catches his interest, and he’s superb at untangling things for a general audience, smoothly deploying his background knowledge, resources, and tools for clarification. (Meanwhile, behind the screen, the lexicographers and sociolinguists roll up their sleeves to embark on the hard task of collecting, analyzing, and interpreting data.)

Ben had also noticed anti-Semetic lately (in speech specifically) but had not in fact written on it. He wondered if the Recency Illusion might be at play, and suggested the possibility of some influence from cosmetic, hermetic, and the like.

Ah, I had very carefully framed what I’d said to him

In the past few weeks, I’ve seen a fair amount of anti-Semetic as a spelling for anti-Semitic and pronunciations of the word with accented [ɛ], separately and together.

That is, I said I’d recently noticed the usages, not that they were recent; as to the actual age of the usages, I have no idea.

But then Ben begins to speculate on the source of the usages, suggesting words in –etic [ɛtɪk] as models. Another possibility is a speech error analysis, in which the e [ɛ] of Semitic perseveres into the following syllable. And more complex accounts of the details: say, some speakers have [ɛ] as a variant of [ɪ] in this one word, for whatever reason, and that is then is sometimes carried over into the spelling. And probably more.

Further complications: these motivating considerations are not mutually exclusive, and might sometimes reinforce one another. And then it’s quite likely that what’s going for one speaker is different from what’s going on for another. And finally, as always, once the variants are in the mix, they can propagate from one speaker to another, without regard to their original motivation.

So a systematic investigation of the phenomena will be a truly hard task. And a fair amount of the data will be inscrutable: we won’t be able to say what’s going on in this specific instance on this specific occasion — only what the larger-scale effects look like.

I’ve already gone further in this matter that I usually do; I’m perfectly comfortable reporting on what I’ve seen, and leaving the real work for others. Don’t go telling me that if I reported it, I’m responsible for analyzing it — because I’m a Famous Linguist, I get this every so often — that will only make me truculent, and angry, and despairing (because I have great trouble finding time in my life for almost anything these days).

9 Responses to “anti-Semetic”

  1. Lise Menn Says:

    I’ve been watching the [ɛ] pronunciation, too, especially on national network news, often considered to exemplify a General American standard.

    • arnold zwicky Says:

      You (implicitly) raise an interesting question here. Let’s grant that the [ɛ] variant is particular to this one lexical item; there’s still the question of how it’s distributed in the population — is it associated with big sociocultural variables (region, racioethnicity, age, sex, class, rurality, etc.)? or is it more of an individual matter? or mostly a matter of membership in local sociocultural communities? In particular, is it associated with the factors that go into General American (of which region is one)?

  2. Robert Coren Says:

    I’m aware of some regional accents (as widely divergent as western PA and Scotland) in which /ɛ/ and /ɪ/* are barely distinguishable, if at all.

    * That’s supposed to be the small capital I of IPA, but WordPress apparently recognizes it as a letter and insists on translating it into the local sans-serif font.

    • arnold zwicky Says:

      I have to admit to being baffled here, at least initially. Let’s restrict this to American English (as in my original posting). The only reported across-the-board merger of /ɛ/ and /ɪ/ in North America is in Newfoundland, which apparently got the feature from Irish. Western Pennsylvania English has been much studied, and none of the sources on it report such a merger. For many speakers, WPE does participate in the PIN/PEN merger (characteristic of the American South, but also found in Northern Midlands speakers, from western Pa. through Iowa or so), but that’s only before nasals.

      I suspect you’re hearing a lowered allophone of /ɪ/ from some WPE speakers; /ɪ/ and /ɛ/ are still distinct for these speakers, and they have no trouble hearing the difference between, say, PIT and PET, BIT and BET — but for your perceptions, the two vowel productions both “sound like” variants of /ɛ/.

      • arnold zwicky Says:

        Oh, yes: Amy Schumer does not seem to be a speaker with generally lowered /ɪ/, at least to my ears; for her, the phenomenon seems to be a fact about this one lexical item.

      • Robert Coren Says:

        Back in the early 1960s I attended a summer camp one of whose owners and a good many of the staff and campers were from Pittsburgh, and I remember that at least some of them pronounced “milk” in a way that I would have spelled “melk”. (I think we’re both familiar with the phenomenon of people who can make distinctions in their own idiolect that are unclear or nonexistent for speakers of other idiolects, like the marry/merry/Mary triplet.)

        I would be interested to know how you persuade WordPress to reproduce the short-i IPA symbol.

      • arnold zwicky Says:

        To Robert on MILK in Pittsburgh: this is a word-specific lowering of /ɪ/ before the consonant [l], widespread in MILK, attested also for PILLOW for some speakers, but not, so far as I can tell, for HILL, KILL, WILT, STILT, BILK, etc.

        As for the symbols in WordPress, your small-cap I shows upon my screen just as you intend, not as a dotless lower-case I. I have no idea why. WordPress has been behaving very erratically and bizarrely for me recently, to the point where it’s very hard for me to prepare my postings at all.

      • Robert Coren Says:

        Oh, so maybe it’s just my display that’s doing weird things: what I see is a (serif-less) capital I.

  3. Anon Says:

    Another factor here: when you’ve only heard “semitic” as part of the word “antisemitism,” you’ve only heard the schwa sound, since the “i” is shortened. So you don’t know if the original vowel in “semitic” is /ɛ/ or /ɪ/.

Leave a Reply to Lise MennCancel reply


Discover more from Arnold Zwicky's Blog

Subscribe now to keep reading and get access to the full archive.

Continue reading