The very first sentence in the essay is:
“NAZI” is a short word. It has only two syllables, like “rac-ist” or “kill-er.”I love it when people do things like this, because it lets me be a linguistics nazi by pointing out where someone has mistaken the folk model of language for the scientific model. This is not the way linguists would divide these words into syllables. The correct way would be "ra-cist" and "ki-ller." Or, more properly, using phonetic symbols from the International Phonetic Alphabet (the little line ˈ is placed in front the stressed syllable):
[ˈre.səst] [ˈkɪ.lər]Notice that we can use a period, rather than a hyphen, to separate syllables. Notice also that there's only one "l" in the phonetic rendition of "killer"; that's because there's only one "l" sound in that word, and in linguistics we don't put two sounds in there unless they really belong in there.
Now, I don't expect Keret to write like a linguist in The New York Times. But, then again, maybe I should. This kind of thing perpetuates misconceptions about language that I have to correct every semester in my classes.
UPDATE (Jan 19). A friend reminds me that Keret's division of "rac-ist" and "kill-er" does reflect the morphological divisions in these words. So, "racist" is race + -ist, and "killer" is kill + -er. These are derivational suffixes, as opposed to inflectional suffixes like the -s in cats and dogs.
Perhaps taking divisions between morphemes to reflect syllable divisions is further evidence of the lack of phonological awareness among many otherwise educated people, especially English speakers.