English questions

[bradrn]

On the other hand, the mere fact that different levels of interruptability exist should suggest that by itself, ‘interruptability’ alone isn’t useful as a criterion. People have made this point more strongly with phonology: the phonological hierarchy is typically postulated to be something like ‘foot < word < phrase’, but in fact there seem to be more (or possibly less) levels than that, which don’t even necessarily line up with each other, so using them as a criterion is cross-linguistically a bit pointless. (See e.g. https://doi.org/10.1111/lnc3.12364 — incidentally another great article on wordhood!)

Haspelmath is deeply into cross-linguistic comparisons, so that bothers him a lot more than it bothers me.

Oh, that article isn’t Haspelmath’s. And I’m quite deeply into cross-linguistic comparison, but it doesn’t bother me either that (say) the phonological hierarchy doesn’t have exactly three strictly delineated levels — it just means that that theory is straightforwardly wrong, and that it’s meaningless to talk about a ‘phonological level’ unless you’re talking about a specific language.

Also, I don’t really understand what kinds of ‘taking fuzzy categories into account’ you’d like to see here. He’s looked at ten different criteria, and demonstrated that each is neither sufficient nor necessary to define the kinds of things which people have called ‘words’. If those criteria are themselves fuzzy or ill-defined, that just reinforces his point.

Fuzzy doesn't just mean vague.

I know what it means. (Remember, I’ve read your Lexipedia!)

The thing is, "make very precise definitions with 100% binary values, then use those for definitions" is precisely not dealing with continua. For the most part, he's showing that binary definitions don't work well. (There's a page where he literally shows these as pluses and minuses.) Again, Chomskyans get hung up on those pluses and minuses, but they're probably wrong about that.

Yes, exactly: he’s taking lots of examples of binary definitions, and showing that they just produce nonsense. The pluses and minuses aren’t his, though — they’re his summary of what binary tests other people have used.

I don't know what an approach using continua would look like, though if anyone wants to use it as a dissertation topic, I'd read that. I expect you'd get a lot of correlations rather than ironclad laws, and that you'd get a lot of clumping that doesn't turn out to be universal.

I agree with this — and it would be a very interesting research topic indeed! I’m less optimistic than you about the clumping, though (or ‘clustering’ as Haspelmath calls it). This was in part the point of the ‘pluses and minuses’: if there were any clumping, we might expect to see correlations there, but I can’t see much of anything.

[Travis B.]

Does anyone else have a partially non-rhotic pronunciation of order? I often have [ˈɔːɾʁ̩ˤ(ː)] for this word, and have heard others here with the same pronunciation, and actually find it hard to give a fully rhotic pronunciation of the word in quick speech. (Note that the vowel in the first syllable is not THOUGHT but NORTH/FORCE ─ THOUGHT is more open than that.)

[Raphael]

When talking about types of (computer) files, do you usually put a dot in front of them? That is, do you usually talk about docx files or .docx files?

[Ryusenshi]

Does anyone else have a partially non-rhotic pronunciation of order? I often have [ˈɔːɾʁ̩ˤ(ː)] for this word, and have heard others here with the same pronunciation, and actually find it hard to give a fully rhotic pronunciation of the word in quick speech. (Note that the vowel in the first syllable is not THOUGHT but NORTH/FORCE ─ THOUGHT is more open than that.)

Not very surprising: the phenomenon is well-documented in words like quarter, or even thermometer (even though there are two syllables in-between).

When talking about types of (computer) files, do you usually put a dot in front of them? That is, do you usually talk about docx files or .docx files?

It depends. No for PDF, MP3, WAV, JPEG, docx. Yes for mono-letters like .o or .c. A big grey area in-between.

[vlad]

Does anyone else have a partially non-rhotic pronunciation of order? I often have [ˈɔːɾʁ̩ˤ(ː)] for this word, and have heard others here with the same pronunciation, and actually find it hard to give a fully rhotic pronunciation of the word in quick speech. (Note that the vowel in the first syllable is not THOUGHT but NORTH/FORCE ─ THOUGHT is more open than that.)

https://home.csulb.edu/~nhall2/dissimilation_paper.pdf

Swadesh (1947:142) claims that dissimilatory deletion of /r/ after stressed /o/ regularly leaves behind a vowel that has no more than a marginal presence in such contexts otherwise:

[...] Colloquial variants with r lost by dissimilation provide additional illustrations of o before a consonant other than r: kónər, kwódər, kòpəréɪʃn beside kórnər, kwórdər or kwórtər, kórpəréɪʃn (corner, quarter, corporation). [NB: Swadesh is distinguishing between o, ou and ɔ.]

Similarly, Canepari (2005:89) claims that when /r/ is dropped from /or/ ‘words do not become ambiguous, because the vowel timbre alone is distinctive’. He describes the vowel that is left as back rounded lower-mid, higher than [ɔ] but lower than [o], and his illustrations of vowel quality do not show this vowel occuring in American English except as a result of dissimilation. Aside from the slight difference in transcriptions (which is probably just a matter of transcription systems), Canepari’s and Swadesh’s descriptions are consistent with one another.

Swadesh and Canepari’s descriptions mean that dissimilation creates a new kind of vowel contrast. Before most consonants, General American has a contrast between oʊ/ɔ (odor [ˈoʊɾər], audit [ˈɔɾɪt]), but no contrast between oʊ/ɔ/o. The phone [o] occurs only as an allophone of /oʊ/ or /ɔ/ before /r/, as in order [ˈorɾər].However, when /r/ is deleted through dissimilation, this allophone remains, creating words like order [ˈoɾər]. This in effect creates a oʊ/ɔ/o contrast which does not exist except when the [o] derives from dissimilation.

[Travis B.]

I noticed a little oddity in the English here today. Normally there is a LOT split in most NAE varieties where tomorrow, sorrow, borrow, and sorry end up identical to START while Florida, forest horror, and most other words with LOT before /r/ end up identical to NORTH/FORCE. The big exception to this is old-school East Coast dialects which have the same vowel as START in Florida and like. My own dialect superficially varies from the former because sorry ends up with NORTH/FORCE rather than with START.

However, I just noticed that this is not true in my dialect, because LOT in the first set of words, sorry aside, ends up as [aʁˤ] whereas START has either [ɑʁˤ] or [ʌʁˤ] depending on fortisness/lenisness and a marginal phonemic split of its own. It should also be noted that this [aʁˤ] is not productive ─ no words other than the original tomorrow, sorrow, borrow, etc. set have this, and this is not productive at all. I find trying to force a word other than these, such as a loanword (e.g. hara kiri) or simply making up a word, to use this pronunciation feels very unnatural, and only START feels acceptable to me in these cases.

[bradrn]

I don't know what an approach using continua would look like, though if anyone wants to use it as a dissertation topic, I'd read that. I expect you'd get a lot of correlations rather than ironclad laws, and that you'd get a lot of clumping that doesn't turn out to be universal.

I agree with this — and it would be a very interesting research topic indeed! I’m less optimistic than you about the clumping, though (or ‘clustering’ as Haspelmath calls it). This was in part the point of the ‘pluses and minuses’: if there were any clumping, we might expect to see correlations there, but I can’t see much of anything.

I actually found some very interesting presentations on this yesterday (Tallman et al. 2018, Tallman et al. 2019, Tallman et al. 2019). They survey 17 different American languages, write down a linear template for each one, and then see what kinds of constituents each wordhood test isolates. For a few languages (e.g. Naso and Cupʼik) many tests strongly converge on a single ‘word’ unit, but for most it’s much more ambiguous — and Hup has no evidence for a ‘word’ level whatsoever. (A fact which should be quite apparent from Epps’s grammar — she puts spaces between things, but it’s obvious that the grammar doesn’t really care about whether two things form one ‘word’ or not.) The last presentation I cited has a section at the end on defining ‘words’ as a fuzzy category, which you might find particularly interesting.

[zompist]

I actually found some very interesting presentations on this yesterday (Tallman et al. 2018, Tallman et al. 2019, Tallman et al. 2019). They survey 17 different American languages, write down a linear template for each one, and then see what kinds of constituents each wordhood test isolates. For a few languages (e.g. Naso and Cupʼik) many tests strongly converge on a single ‘word’ unit, but for most it’s much more ambiguous — and Hup has no evidence for a ‘word’ level whatsoever. (A fact which should be quite apparent from Epps’s grammar — she puts spaces between things, but it’s obvious that the grammar doesn’t really care about whether two things form one ‘word’ or not.)

Definitely interesting! I'm not clear (probably because of the Powerpoint format) where the list of slots comes from and whether it's supposed to be cross-linguistic.

The results would seem to be similarly frustrating to the words-always-exist school and the words-never-exist school.

Also, nice to see Anthony Woodbury still working. If anyone has read Geoffrey Pullum's essay "The Great Eskimo Vocabulary Hoax", Woodbury is the scholar of Eskimo he cited.

The last presentation I cited has a section at the end on defining ‘words’ as a fuzzy category, which you might find particularly interesting.

Again, not enough detail in the slides to understand what he's talking about. However, when I say "fuzzy" I don't mean that things have to be real-valued. A significant set of discrete values will do— e.g. Tallman's slots. (I don't get the impression that Tallman thinks that these slots are the last word and that finer distinctions are ruled out.)

[bradrn]

I actually found some very interesting presentations on this yesterday (Tallman et al. 2018, Tallman et al. 2019, Tallman et al. 2019). They survey 17 different American languages, write down a linear template for each one, and then see what kinds of constituents each wordhood test isolates. For a few languages (e.g. Naso and Cupʼik) many tests strongly converge on a single ‘word’ unit, but for most it’s much more ambiguous — and Hup has no evidence for a ‘word’ level whatsoever. (A fact which should be quite apparent from Epps’s grammar — she puts spaces between things, but it’s obvious that the grammar doesn’t really care about whether two things form one ‘word’ or not.)

Definitely interesting! I'm not clear (probably because of the Powerpoint format) where the list of slots comes from and whether it's supposed to be cross-linguistic.

The impression I get is that the slots are not cross-linguistic — this analysis was purely language-internal. For each language, I think they just wrote down the linear structure of the sentence. It certainly is difficult to decipher, though.

The results would seem to be similarly frustrating to the words-always-exist school and the words-never-exist school.

Agreed. On my part, it’s forced me to revise my opinions closer to ‘words can exist, but only language-internally’.

I don't get the impression that Tallman thinks that these slots are the last word and that finer distinctions are ruled out.)

I do get that impression, because 1 slot contains 1 morpheme. At least grammatically, it makes no sense to have a word which is smaller than a morpheme.

[Raphael]

"I don't care what you or anyone else say"

or

"I don't care what you or anyone else says"?

[bradrn]

"I don't care what you or anyone else say"

or

"I don't care what you or anyone else says"?

Either, I think.

[Estav]

"I don't care what you or anyone else say"

or

"I don't care what you or anyone else says"?

I don't think my grammar actually has a specific solution for these kinds of situations. As a conscious decision, I would go with agreement with the nearer element ("what you or anyone else says"), which sounds a bit better to me and I think may be the normatively preferred method; but it doesn't subjectively feel to me like this is "the correct" way of putting it, just the more acceptable sounding wording of the two problematic alternatives.

[Raphael]

Thank you!

[Richard W]

"I don't care what you or anyone else say"

or

"I don't care what you or anyone else says"?

Although agreement with the nearer alternative is a popular choice, I thing there's a tendency for 'you or anyone else' to be treated as a plural subject. A clear example of this, I think, would be I'd be upset if you or anyone else hurt themselves.' The object can also turn up as 'himself' or 'themself'. I think 'herself' would be aggressive unless it was overwhelmingly likely that 'anyone else' would be female.

[zompist]

"I don't care what you or anyone else say" or "I don't care what you or anyone else says"?

You might be interested in linguist Arnold Zwicky's discussion.

In short: it's kind of not covered by English grammar. Zwicky cites two possible rules (agree with the nearer NP; agree with the higher-plurality NP) and suggests that even so, some sentences are just hard to make work, e.g. "Neither you nor I [is, am, are] responsible."

(FWIW I'd go with "says" in your example, but I'd be strongly tempted to reword to avoid the problem.)

[elgis]

When is it acceptable to use a plural verb after an interrogative who?

1. Who are they?
2. Who make these toys? (as in "Who are the people that make these toys?")
3. Who bring their own food?
4. Who have been to the island?

[Estav]

When is it acceptable to use a plural verb after an interrogative who?

1. Who are they?
2. Who make these toys? (as in "Who are the people that make these toys?")
3. Who bring their own food?
4. Who have been to the island?

As a rule, "who" as an interrogative pronoun triggers singular agreement on a verb. This doesn't affect "Who are they?" because the subject of that sentence is "they", not "who". I think it's not completely impossible for a plural verb to be used when "who" is the subject, but definitely uncommon, and I can't think of any common type of sentence where it would tend to be used. Maybe it might be preferred with some kinds of predicates that are unambiguously plural, like "Who are in disagreement?" or "Who are united?"

[Raphael]

Thank you again, everyone!

(FWIW I'd go with "says" in your example, but I'd be strongly tempted to reword to avoid the problem.)

I could, of course, replace "I don't care what you or anyone else says" with "I don't care what you say, or what anyone else says". Problem is, that would mean adding more words. And I've already been told to make changes towards fewer, rather than more, words where possible.

[zompist]

I could, of course, replace "I don't care what you or anyone else says" with "I don't care what you say, or what anyone else says". Problem is, that would mean adding more words. And I've already been told to make changes towards fewer, rather than more, words where possible.

Probably "I don't care what anyone says" will work just fine.

[Travis B.]

Does anyone vary whether metal and medal are homophones based on register? For me they both have short vowels in everyday speech, but when I emphasize medal I may pronounce it with a long vowel.