"A Japanese" or "a Chinese" were common in English a few decades ago. This has declined very significantly. Interestingly, people often use these as self-descri

"A Japanese" or "a Chinese" were common in English a few decades ago. This has declined very significantly. Interestingly, people often use these as self-descriptions (I think especially non-native speakers). There is a similar phenomenon with "a Jew", which was once common and where now most speakers would use "a Jewish person" (except, to some extent, for Jews using it as a self-description, although "I am Jewish" is more common than "I am a Jew", but their denotation is also different because the former foregrounds religious identification and the latter foregrounds ancestry or ethnic identification). Apart from the "person-centered language" phenomenon that other people have mentioned, I think there's the closely related euphemism treadmill https://en.wikipedia.org/wiki/Euphemism#Euphemism_treadmill https://en.wikipedia.org/wiki/Euphemism#Euphemism_treadmill where even a completely neutral term may acquire a negative connotation because of its widespread use in a negative sense, or by people intending to give it a negative connotation. Or people may simply fear that a neutral term has acquired a negative connotation by recalling that it was widely used in the past. As other commenters have pointed out, there are still demonyms which are common to use in the singular and their idiomaticity has tended to change (mostly in the direction of stopping). Although I think there's an interesting phenomenon where new countries' demonyms feel plausible to me in the singular ("an East Timorese", "a South Sudanese", "a North Macedonian", "a Kosovan"/"a Kosovar"). It's hard to prove but I suspect that the relative novelty or unfamiliarity of these terms acts to alleviate speakers' internalized fear that they might inadvertently be using the same terminology that was previously used by xenophobes! But I'm not totally confident in that. It would be interesting to gather more data by surveying a lot of native speakers about a lot of demonyms and trying to quantify the possible influences of (1) familiarity, (2) phonology, (3) history of ethnic conflict, ... Another one that feels kind of parallel is the arbitrariness in whether you use a definite article in the name of a country. For example, "the Sudan" and "the Ukraine" used to be very common in English but it's not unusual to see the country names without the article now. On the other hand, "the United Kingdom", "the Netherlands", and "the Maldives" are still always "the". This is different in other languages, where for example "Switzerland" is never "the" in English, but is always "the" in German ("die Schweiz"); "Japan" is never "the" in English, but is always "the" in Portuguese ("o Japão"), as is Brazil ("o Brasil"). It's challenging, though maybe not impossible, to give a clear account of the origins of all of these distinctions... Edit: in Arabic, too, some countries are "the" and some aren't, but it also doesn't correspond with English practice.