Linguistic Miscellany Thread

[malloc]

I just stumbled on this amazing piece of Mongolian calligraphy this morning:

[Linguoboy]

- Learnèd irregular plurals (crisis - crises, cherub - cherubim/cherubs, one ninja - three ninja/ninjas). For a lot of these there's a struggle between the learnèd plural and the regular plural, with different levels of acceptance for either (matrix - matrices and millennium - millennia are very accepted, campus - campi and stadium - stadia mostly occur as rare jokes).

One of these things is not like another. Ninja is not a "learnèd plural" in the same way as matrices or campi. It's the absence of a distinct plural form. As you say, this pattern exists already in English and sometimes foreign borrowings are assimilated to it rather than taking the usual -s plural.

There's also a subclass of pseudo-learnèd plurals that are analogical extensions of true learnèd forms. Generally, this involves the extension of Latin second declension -i to third- or fourth-declension nouns, e.g. octopi (vs. octopodes), peni (vs. penes), excursi (vs. excursus), etc.

[Ryan of Tinellb]

matrix - matrices

I had mathematics lecturers refer to two indices, one indice /ˈɪndəsi/; and two minima, one minima. I think I've also heard the same for nuclei.

- Learnèd irregular plurals (crisis - crises, cherub - cherubim/cherubs, one ninja - three ninja/ninjas). For a lot of these there's a struggle between the learnèd plural and the regular plural, with different levels of acceptance for either (matrix - matrices and millennium - millennia are very accepted, campus - campi and stadium - stadia mostly occur as rare jokes).

One of these things is not like another. Ninja is not a "learnèd plural" in the same way as matrices or campi. It's the absence of a distinct plural form. As you say, this pattern exists already in English and sometimes foreign borrowings are assimilated to it rather than taking the usual -s plural.

I disagree, Linguoboy. It takes a bit of "learning" to know that Japanese doesn't have plural marking (or that the plural is marked by a null morpheme ).

The converse startled me, when my 日本人の日本語の先生, i.e.: my Japanese language teacher who is himself Japanese, referred to 'kanjis'.

[Linguoboy]

I disagree, Linguoboy. It takes a bit of "learning" to know that Japanese doesn't have plural marking (or that the plural is marked by a null morpheme ).

I still think there's a difference between learning that a plural distinction exists and memorising the actual forms (which are often quite irregular from the POV of a speaker of English).

Cf. Italian plurals like biscotti and spaghetti. I think pretty much the entire populace is "learned" enough to know that Italian nouns have mandatory singular and plural forms. However, unless they've specifically studied Italian, they don't know what the actual forms are and so they treat these nouns as invariable, either as count nouns (e.g. one biscotti, two biscotti) or mass nouns (e.g. one piece of spaghetti).

[Ryan of Tinellb]

I can agree to disagree.

Cf. Italian plurals like biscotti and spaghetti. I think pretty much the entire populace is "learned" enough to know that Italian nouns have mandatory singular and plural forms. However, unless they've specifically studied Italian, they don't know what the actual forms are and so they treat these nouns as invariable, either as count nouns (e.g. one biscotti, two biscotti) or mass nouns (e.g. one piece of spaghetti).

My Italian plurals are like the half-learned Latin plurals mentioned earlier, eg: octopi. I picked up the rule (singular) a, o → (plural) e, i in primary school LOTE, but I didn't study the language long enough to learn the exceptions to that rule.

spaghetto, biscotto, graffito follow the rule. Can anyone give me an example of an Italian word in English that didn't originally follow it?

[Ryan of Tinellb]

My internal lexicon has 'Cf:' as "compare wif [sic with]" instead of "confer".

[Pabappa]

My internal lexicon has 'Cf:' as "compare wif [sic with]" instead of "confer".

Ah, I didnt know that. I just learned "cf. = compare" from an etymological dictionary long ago and yet never bothered (that i remember) to look up the etymology of "cf" itself. Ive probably spelled it out as "c.f." a few times.

The food thing .... to me, *spaghetto is just silly. It sounds like a parody of Gepetto. sgraffito is an art term, but i havent seen it without the s-. I think food terms like that have their own rules .... I cant think of even one case where we've borrowed the singular & plural from Italian and use them reliably for the singular and plural in English. imagine .... macarono, provolona, fettucina, linguino, maybe extended to genericized proper names like *toblerona, etc . the anomalous pronunc of "bologna" might have to do with it being analogized to the rule of most Italian foods ending in -i even though we didnt change the spelling. and my parents say "kielbasi" instead of "kielbasa" .... i think it goes even beyond Italian. Food terms really have their own rules.

[Travis B.]

Anyone else have laxing of vowels before /l/ in common contractions, such as we'll, you'll, he'll, and she'll? (For me we'll and wool, he'll and hill, and she'll and shill are homophones outside careful speech.)

[zompist]

Anyone else have laxing of vowels before /l/ in common contractions, such as we'll, you'll, he'll, and she'll? (For me we'll and wool, he'll and hill, and she'll and shill are homophones outside careful speech.)

Yes, except that for me "we'll" is homophonous with "will"— "wool" has a back vowel (same as "you'll" in fact).

[Space60]

Unlike other season words, the word "autumn" can't take the definite article for me. I'd say "in the spring", "in the summer", "in the fall" and "in the winter", but not *"in the autumn". I'd say "in autumn" if I were using "autumn" instead of "fall". "Fall" is the word I use most of the time to refer to the season.

[missals]

Anyone else have laxing of vowels before /l/ in common contractions, such as we'll, you'll, he'll, and she'll? (For me we'll and wool, he'll and hill, and she'll and shill are homophones outside careful speech.)

Likewise - in fact, I would say that my variety has completely lost the full-vowel or stressed forms of contractions, and replaced them in all contexts with forms resulting from the analogical re-stressing of the unstressed contractions.

So I have:

we'll unstressed [wɪl], stressed [wɪl] - NEVER [ˈwi.əl]
he'll unstressed [hɪl], stressed [hɪl] - NEVER [ˈhi.əl]
you'll unstressed [jəl], stressed [jəl] - NEVER [ju.əl]
she'll unstressed [ʃɪl], stressed [ʃɪl] - NEVER [ˈʃi.əl]
I’ll unstressed [æl], stressed [æl] - NEVER [ˈaɪ̯.əl]
you’re unstressed [jɚ], stressed [jɚ] or [jɔɹ] - NEVER [ˈju.ɚ]
aren't unstressed - [ɑɹnt], stressed [ɑɹnt] - NEVER [ˈɑɹənt]

So, even if I was delivering a formal speech that included the phrase "He'll be the one...", I couldn't say [ˈhi.əl] - my only choices would be [hɪl], or the separate words he will, no in-between.

Probably relatedly, my pronunciation of our and your is always [ɑɹ] and [jɚ] (or [jɔɹ]) - never [aʊ.ɚ] and [ju.ɚ].

[Whimemsz]

Anyone else have laxing of vowels before /l/ in common contractions, such as we'll, you'll, he'll, and she'll? (For me we'll and wool, he'll and hill, and she'll and shill are homophones outside careful speech.)

Likewise - in fact, I would say that my variety has completely lost the full-vowel or stressed forms of contractions, and replaced them in all contexts with forms resulting from the analogical re-stressing of the unstressed contractions.

So I have:

we'll unstressed [wɪl], stressed [wɪl] - NEVER [ˈwi.əl]
he'll unstressed [hɪl], stressed [hɪl] - NEVER [ˈhi.əl]
you'll unstressed [jəl], stressed [jəl] - NEVER [ju.əl]
she'll unstressed [ʃɪl], stressed [ʃɪl] - NEVER [ˈʃi.əl]
I’ll unstressed [æl], stressed [æl] - NEVER [ˈaɪ̯.əl]
you’re unstressed [jɚ], stressed [jɚ] or [jɔɹ] - NEVER [ˈju.ɚ]
aren't unstressed - [ɑɹnt], stressed [ɑɹnt] - NEVER [ˈɑɹənt]

So, even if I was delivering a formal speech that included the phrase "He'll be the one...", I couldn't say [ˈhi.əl] - my only choices would be [hɪl], or the separate words he will, no in-between.

Probably relatedly, my pronunciation of our and your is always [ɑɹ] and [jɚ] (or [jɔɹ]) - never [aʊ.ɚ] and [ju.ɚ].

This is similar for me, except: (1) my unstressed/neutral forms for "we'll" and "I'll" are /wl̩/~/wʊl/ [wɵʟ] (like Travis, homophonous with "wool") and /ɑl/ [ɑʟ] (same as my normal stressed "I'll" -- I don't have /æ/ or [æ] in this), (2) "she'll" varies between /ʃɪl/ [ʃɪɫ] and /ʃl̩/~/ʃʊl/ [ʃɵʟ], and (3) I can have "decontracted" forms of "I'll", "you're", "aren't", etc. in emphatic/clarifying contexts ([ˈaɪ̯.ɵʟ], [ˈɑɹɨ̞nt] ...) though in some of these cases I'm much more likely to use a fully decontracted form like "I will"

[Zaarin]

- Learnèd irregular plurals (crisis - crises, cherub - cherubim/cherubs, one ninja - three ninja/ninjas). For a lot of these there's a struggle between the learnèd plural and the regular plural, with different levels of acceptance for either (matrix - matrices and millennium - millennia are very accepted, campus - campi and stadium - stadia mostly occur as rare jokes).

Stadia is the standard plural of the measurement, but that would very rarely come up in anything but academic historical discourse, so it still qualifies under "learnèd plurals."

I just stumbled on this amazing piece of Mongolian calligraphy this morning:

That is gorgeous. Syriac and its derived scripts are my favorites.

[Raholeun]

Question: are there attested natlangs that have a sound phonemically that is only found in a certain word class, or that is restricted to derivational morphology as opposed to root lexemes? If so, what sounds what that be?

[Pabappa]

Question: are there attested natlangs that have a sound phonemically that is only found in a certain word class, or that is restricted to derivational morphology as opposed to root lexemes? If so, what sounds what that be?

yes there are a few examples .... some Uralic languages have voiced stops only in gradations of words and in loans ... that is, theyre bound by grammar and cannot occur in independent roots. Finnish is an example of this. english /ð/ comes close to being bound to function words, and a few people could perhaps argue that words like "feather" are just showing a voiced allophone of /θ/. but thats an argument all its own. there's also the voiced pronunciation of "thank" gaining ground so i'd say english does not have a phoneme that is bound in such a way.

i think Middle Japanese may have had closed syllables only in loans from Chinese. im not clear on the chronology of the changes.

Ive heard that the Arabic pronunciation of "Allah" has a phoneme unique to that word, /lˁ/, but I think the pronunciation with the plain /l/ is acceptable in some contexts.

Some Russian speakers have /ɣ/ only in religious terms, due to influence from Church Slavonic.

[Whimemsz]

.

[Ares Land]

Apparently it turns out that there is a correlation between the gender gap and speaking a language with grammatical gender:

http://documents.worldbank.org/curated/ ... PS8464.pdf

I'm curious about what you think of it. I was thinking the Sapir-Whorf hypothesis was pretty much dead myself...

[zompist]

(I'm not trying to pick on you today, Ars! You just happened to post things that are interesting to talk about.)

I mostly skimmed the article, so I can't say how good their statistics are. What I didn't see, though, was any consideration of historical (economic) change, or detailed power structures, which I think would decimate their arguments.

Most interesting was their depiction of gendered languages in the world, which I reproduce for convenience.



First thing to notice: English is evidently assigned as "non-gendered". Which is maybe OK as a rough linguistic observation, but it's going to have a huge effect on their conclusion-- English is the native language of people in many countries, mostly wealthy ones. Doesn't English have a robust gender system in pronouns and kinship terms? That one data point is going to have an enormous impact.

Or: Peru, Bolivia, and Paraguay are shown as lighter in color (fewer gendered languages). That's presumably because Quechua, Aymara, and Guarani are non-gendered. Only-- the political culture of these countries is entirely dominated by the Spanish speakers.

Or: 100 years ago, the linguistic data would be almost the same, but the economics entirely different. Most notably, China and India would be starving poor instead of relatively developed. And though speaking the same languages, people were in general much more male-dominated. 1000 years ago, the economics would be even more different: the most prosperous areas of the world would be the Arab caliphate (gendered) and China (non-gendered). The massive changes in that time are largely due to one thing: Europe's enormous worldwide surge. Maybe there's an argument that it's due to gendered language instead, but I don't even see the authors considering the change.

Finally, I'd note my usual objection to this sort of Whorfianism: do they really really believe that Turkey, China, and Indonesia are better for women than France, Germany, and Norway?

[Ares Land]

Yes, it's intriguing enough but not entirely convincing.
It would be interesting to compare attitudes, all things being equal with similar culture and economic development. Like Britain and France, or Norway and Finland.

I was a little bothered by that bit in the intro as well:

For example, Chen (2013) demonstrates that speakers of languages that demarcate
the future as separate from the present (e.g. English) save less than those whose languages
make no such distinction (e.g. German).

[Moose-tache]

I was a little bothered by that bit in the intro as well:

For example, Chen (2013) demonstrates that speakers of languages that demarcate
the future as separate from the present (e.g. English) save less than those whose languages
make no such distinction (e.g. German).

will vs wollen. Clearly one is a tense and the other is a modal. Try to keep up! /s