Zompist Bboard Again

This doesn't quite answer the question, but...

etymonline wrote:ics in the names of sciences or disciplines (acoustics, aerobics, economics, etc.), a 16c. revival of the classical custom of using the neuter plural of adjectives with Greek -ikos "pertaining to" (see -ic) to mean "matters relevant to" and also as the titles of treatises about them. Subject matters that acquired their English names before c. 1500, however, tend to be singular in form (arithmetic, logic, magic, music, rhetoric). The grammatical number of words in -ics (mathematics is/mathematics are) is a confused question.

That is, words like physics, acoustics, politics etc. were plural in Greek, and this was consciously imitated in English. Using "politicks" with the plural is pretty common-- indeed, you can still find "politics" used this way.

Here's one example, from 1700, of "mathematicks" used as a plural.

My walls are becoming structurally unsound from the sheer number of forehead-shaped dents, and I am beginning to suspect that I am the butt of an elaborate practical joke.

Let's take this one step at a time.

Grammatical suffixes persevere after a word is abbreviated or initialized. You can say "I DM'ed him," with the past tense suffix being preserved. If you abbreviated "I direct messaged him" as "I DM him," with the tense suffix being chopped off along with the rest of the word, it would be incorrect, at least if your intention was to present a sentence in the past tense.

We understand this, yes? Please raise your hand if I've lost you so far.

Now derivational endings, on the other hand, are usually lost when a word is abbreviated. "Cubic," for example, is often abbreviated "cu." I can abbreviate "governor" as "gov," but I cannot abbreviate "governors" as "gov." The derivational ending -or gets chopped off with no problem, but if I chop off the grammatical ending, it changes the meaning of the sentence, because readers expect the grammatical information to be there. In short, the derivational suffix is an inherent part of the lexical word, while the grammatical suffix is there to explain the word's role in a phrase or clause. There are obviously some exceptions to this, for example "died" is sometimes abbreviated as "d." with the grammatical ending removed, but these tend to be very context-specific, in set phrases or constructions where there is almost guaranteed to be no ambiguity. If I want to shorten the word "linguistics," I can say "ling." I am not required to include the -ic. However, if I wanted to shorten something like "I linguisticed the hell out of that class," I would have to say "ling'ed" or something similar. Likewise, derivational endings are in some rare cases preserved on abbreviations, but it is limited to cases where the ending is highly concpicuous, like "I'd like to DM-ize this conversation," where -ize is an unexpected ending not predictable from context.

So we understand the pattern, yes? Grammatical sufixes remain when we shorten words, because they do not just belong to that word. Derivational suffixes, at least in cases with no ambiguity, are lost when we shorten words.

Please, for the love of Black Baby Jesus, tell me you understand. Got it? Thank Christ.

Now how do we apply this pattern to the words "math" and "maths?" The suffix -s in this case is a derivational suffix, with no historical connection to the plural suffix -s. This is admittedly very stupid, and the English tendency to reduce all suffixes to -s is something that even an amateur conlanger would avoid, but coincidences like this are what we have to work with. American English dutifully treats this derivational suffix as a derivational suffix: just like the -ic, the -s is lost when the word is abbreviated: mathematics > math.

Still with me?

Now let's look at "maths." The -s suffix could not survive the shortening if it were consistently understood by every generation of British speakers to be a derivational suffix. After all, British speakers immediately know that "lings" and "mathic" are wrong. We can see from thousands of examples that British speakers are perfectly capable of applying the logic of preserving grammatical but not derivational suffixes to abbreviations. So where does "maths" come from?

Go ahead and write down your top three answers, and I'll tell you if you got it correct.

...

Did you write down "because at some point in the last three hundred years it was reanalyzed as a grammatical suffix?" That's correct! That is literally the only explanation for how this could occur, unless you think every British speaker is having a momentary stroke every time they say this one specific word.

Now some of you have your hands up because you're about to ask me about number concord on verbs and pronouns. Put your filthy hands down. Nobody is talking about number concord, past or present. I'm talking to you about calculus and you're asking me a question about parking validation. Stop it. Stop it immediately; it's not relevant. English could switch to using "xe" to describe every noun in the dictionary, and it would not prevent us from having reanalyzing a suffix at some point in the past. They are two different things.

Since my Swedish example went over your head, I will give you a Romanian one. The word for "arm" comes from a Latin neuter. As the neuter gender started to disappear in Vulgar Latin, pre-Romanian speakers reanalyzed the neuter plural -a as a feminine ending, and started using feminine adjective forms and the like, but only in the plural. Thus, we have masculine concord and feminine concord on the same word, so is it masculine or feminine? Neither, it's neuter. The reanalysis of suffixes and the reorientation of ajective/pronoun concord are two separate things. This is further born out by the fact that pre-Romanian speakers reanalyzed -a as a feminine ending for plural "arms," even though the prototypical feminine ending is not -a in the plural, but the singular. The pre-Romanians did not feel the need to apply singular feminine concord to plural arms. That would just be silly. Morphology and concord are not the same. The behavior of one does not negate the behavior of the other.

Please understand this, I am begging you, the building cannot take any more it's going to collapse.

zompist wrote: ↑Sun May 01, 2022 10:05 pm Here's one example, from 1700, of "mathematicks" used as a plural.

Curiously, in only two instances where accidence reveals it to be plural does it occur without the article. There is also one case where it must be singular, and that lacks an article, and there is one further case where there is no article but the verb but does not reveal its number. As a grammatical subject, it overwhelmingly occurs with the article and is construed as plural. This three hundred year old usage is alien to my grammar in this respect.

I find it somewhat funny, that despite treating seemingly endless singular nouns as plural (eg. the government, countries, companies, bands, teams, etc.), the one noun that would actually make some sense being plural, at least by its looks, math(s), is treated singular in British English.

zompist wrote: ↑Sun May 01, 2022 10:05 pm This doesn't quite answer the question, but...

etymonline wrote:ics in the names of sciences or disciplines (acoustics, aerobics, economics, etc.), a 16c. revival of the classical custom of using the neuter plural of adjectives with Greek -ikos "pertaining to" (see -ic) to mean "matters relevant to" and also as the titles of treatises about them. Subject matters that acquired their English names before c. 1500, however, tend to be singular in form (arithmetic, logic, magic, music, rhetoric). The grammatical number of words in -ics (mathematics is/mathematics are) is a confused question.

That is, words like physics, acoustics, politics etc. were plural in Greek, and this was consciously imitated in English. Using "politicks" with the plural is pretty common-- indeed, you can still find "politics" used this way.

Here's one example, from 1700, of "mathematicks" used as a plural.

Ah, cool.

I feel that etymonline entry could do with a bit more precision though. Consulting the LSJ, I see Ancient Greek did use a neuter plural for 'politics' (τὰ πολιτικά), but all three of 'arithmetic, music and rhetoric' were feminine singular, like the word for 'mathematics'. 'Logic' is attested in the substantivized feminine singular form of the word, the neuter plural, and also the neuter singular at least by Zeno (who is mentioned to list the three branches of philosophy as τὸ φυσικόν, τὸ ἠθικόν, τὸ λογικόν)... 'Acoustics' seems unattested in the corpus the LSJ considered, though the faculty or sense of hearing is in the neuter singular. For 'economics' it seems the word is attested in the masculine singular and the neuter plural as book titles, but more often in the feminine singular:

LSJ wrote:ὁ οἰ. title of treatise on the duties of domestic life, by Xenophon; and τὰ οἰ. title of treatise on public finance, ascribed to Aristotle, cf. X. Cyr. 8.1.14: ἡ -κή (sc. τέχνη) domestic economy, husbandry, Pl. Plt. 259c, X. Mem. 3.4.11, etc.;

The etymonline comment seems to suggest the ancients had a custom to use the neuter plural but if anything the feminine singular was more common.

Moose-tache wrote: ↑Sun May 01, 2022 10:36 pm...

Ah, I get it now. Linguistics > ling is a good analogy for mathematics > math. The emotionality wasn't all that needed though, I'd say...

Moose-tache wrote: ↑Sun May 01, 2022 10:36 pm Please understand this, I am begging you, the building cannot take any more it's going to collapse.

Put your hard hat on; there's more coming.

You missed out a further complication with the simple Romanian neuters. Plural ouă of ou 'egg' is the only inherited word that keeps the feminine-like singular ending. Others use a feminine plural form instead, such as the plural of braț 'arm' being brațe. Italian behaves the same.

mocha wrote: ↑Sun May 01, 2022 10:53 pm I find it somewhat funny, that despite treating seemingly endless singular nouns as plural (eg. the government, countries, companies, bands, teams, etc.), the one noun that would actually make some sense being plural, at least by its looks, math(s), is treated singular in British English.

There's another one - meths for methylated spirit(s). In this sense, meths too is usually construed as singular, though one can find it in the plural. This too seems to be a Commonwealth word. Now, one can construct a story whereby both mathematics and maths were construed as plural, accounting for the 's' in the abbreviation, but mathematics has largely become singular, and maths has followed suit. The word meths may have the same history; methylated spirits is also often used as a singular, perhaps more commonly than as a plural.

zompist wrote: ↑Sat Apr 30, 2022 5:41 pm You'll find both because there is no fixed rule.

So what about a name that has final /s/ only in hearing, but not orthographically? For example, would one write Condoleezza Rice's or Rice'?

Raholeun wrote: ↑Tue May 03, 2022 1:59 am

zompist wrote: ↑Sat Apr 30, 2022 5:41 pm You'll find both because there is no fixed rule.

So what about a name that has final /s/ only in hearing, but not orthographically? For example, would one write Condoleezza Rice's or Rice'?

I could only write (and say) "Rice's", but let's see if anyone disagrees!

zompist wrote: ↑Tue May 03, 2022 2:08 am

Raholeun wrote: ↑Tue May 03, 2022 1:59 am

zompist wrote: ↑Sat Apr 30, 2022 5:41 pm You'll find both because there is no fixed rule.

So what about a name that has final /s/ only in hearing, but not orthographically? For example, would one write Condoleezza Rice's or Rice'?

I could only write (and say) "Rice's", but let's see if anyone disagrees!

I also would only write or say "Rice's" myself.

Travis B. wrote: ↑Tue May 03, 2022 10:47 am

zompist wrote: ↑Tue May 03, 2022 2:08 am

Raholeun wrote: ↑Tue May 03, 2022 1:59 am
So what about a name that has final /s/ only in hearing, but not orthographically? For example, would one write Condoleezza Rice's or Rice'?

I could only write (and say) "Rice's", but let's see if anyone disagrees!

I also would only write or say "Rice's" myself.

Same here. *Rice' looks like a non-native error.

One thing to remember is that English does not treat final /s/ after a vowel as "plural-like"; only /z/ can be "plural-like" in this position. However, it does count as sibilant, so one gets /ˈraɪsəz/ for "Rice's". Final /s/ only counts as "plural-like" when it comes after a fortis obstruent in English.

Travis B. wrote: ↑Tue May 03, 2022 1:43 pm One thing to remember is that English does not treat final /s/ after a vowel as "plural-like"; only /z/ can be "plural-like" in this position. However, it does count as sibilant, so one gets /ˈraɪsəz/ for "Rice's". Final /s/ only counts as "plural-like" when it comes after a fortis obstruent in English.

However, dice's spots seems abnormally rare compared to dice spots.

Richard W wrote: ↑Tue May 03, 2022 4:35 pm

Travis B. wrote: ↑Tue May 03, 2022 1:43 pm One thing to remember is that English does not treat final /s/ after a vowel as "plural-like"; only /z/ can be "plural-like" in this position. However, it does count as sibilant, so one gets /ˈraɪsəz/ for "Rice's". Final /s/ only counts as "plural-like" when it comes after a fortis obstruent in English.

However, dice's spots seems abnormally rare compared to dice spots.

But I would interpret dice spots as being a compound rather than a genitive construction.

Travis B. wrote: ↑Tue May 03, 2022 4:46 pm

Richard W wrote: ↑Tue May 03, 2022 4:35 pm However, dice's spots seems abnormally rare compared to dice spots.

But I would interpret dice spots as being a compound rather than a genitive construction.

That's why we've only got frequency to go on. I think dice may be a noun being used attributively; do you think it's a univerbation?

Richard W wrote: ↑Tue May 03, 2022 4:59 pm

Travis B. wrote: ↑Tue May 03, 2022 4:46 pm

Richard W wrote: ↑Tue May 03, 2022 4:35 pm However, dice's spots seems abnormally rare compared to dice spots.

But I would interpret dice spots as being a compound rather than a genitive construction.

That's why we've only got frequency to go on. I think dice may be a noun being used attributively; do you think it's a univerbation?

Compounds that could be interpreted as genitive constructions, e.g. rabbit hole, are pretty common in English, and in most of such cases could not be interpreted as a genitive of a plural where the genitive and plural are not marked separately.

Travis B. wrote: ↑Tue May 03, 2022 5:07 pm Compounds that could be interpreted as genitive constructions, e.g. rabbit hole, are pretty common in English, and in most of such cases could not be interpreted as a genitive of a plural where the genitive and plural are not marked separately.

That doesn't mean that such phrases are compounds. One test is whether they are written without a space. Another is whether they bear meanings of their own, which is how rabbit hole makes it into Wiktionary.

I think this sort of modification by juxtaposition in English, as in "a forum discussion, a rabbit house", is commonly called "composition" (and its results "compounds") though? The likes of "rabbit house" vs. "rabbit hole" seem well worth distinguishing but I at least am not aware of good terminology...

Travis B. wrote: ↑Tue May 03, 2022 5:07 pm Compounds that could be interpreted as genitive constructions, e.g. rabbit hole, are pretty common in English, and in most of such cases could not be interpreted as a genitive of a plural where the genitive and plural are not marked separately.

Richard W wrote: ↑Tue May 03, 2022 5:14 pm That doesn't mean that such phrases are compounds. One test is whether they are written without a space. Another is whether they bear meanings of their own, which is how rabbit hole makes it into Wiktionary.

Kuchigakatai wrote: ↑Tue May 03, 2022 7:48 pm I think this sort of modification by juxtaposition in English, as in "a forum discussion, a rabbit house", is commonly called "composition" (and its results "compounds") though? The likes of "rabbit house" vs. "rabbit hole" seem well worth distinguishing but I at least am not aware of good terminology...

‘Noun compound’ is a formal term: these examples are all noun compounds, resulting from the operation of noun compounding (or composition, if you prefer). They can be interpreted as genitives because both possessive constructions and noun compounds are forms of nominal modification with very general semantics — in most cases they are more or less interchangeable, except for situations involving actual possession. (Indeed, in languages which do not make heavy use of noun compounds, possessive constructions tend to be used as a replacement.) ‘Rabbit hole’ is a noun compound which has lexicalised, having gained an idiomatic meaning.

Oh yes, "compounding", not "composition". I might be getting too influenced by the Spanish translation.

No one expects the Spanish translation!