Linguistic Miscellany Thread

[Qwynegold]

Consider this sentence: I know the boy whose father died, and he told me he had a secret. My question is, who had a secret?

[KathTheDragon]

My initial interpretation was the boy had a secret, but it is structurally ambiguous.

[bradrn]

My initial interpretation was the boy had a secret, but it is structurally ambiguous.

I agree.

[Qwynegold]

Hmm, I thought so. I had another tussle with the relative clauses in a conlang, and I was trying to figure out what to do when there is something that can't be relativized. (I know that sentence had more to do with anaphora than relative clauses, but anyway...) What should I write in a grammar? Should I just write "this and this and this thing can't be relativized", without offering any suggestions on what to do instead?

[bradrn]

Hmm, I thought so. I had another tussle with the relative clauses in a conlang, and I was trying to figure out what to do when there is something that can't be relativized. (I know that sentence had more to do with anaphora than relative clauses, but anyway...) What should I write in a grammar? Should I just write "this and this and this thing can't be relativized", without offering any suggestions on what to do instead?

I’d phrase it something like this: ‘All elements of the clause may be relativised, except X and Y and Z. If necessary, these may be relativised indirectly, through use of Feature1 (Section a.b), Feature2 (Section c.d), or Feature3 (Section g.h).’

[Qwynegold]

Allright, thanks! I think I have figured out now how to do this. Now I just need to write it down in my grammar... I'll post my solution to the above sentence, without any explanations, even though no one asked, just so I can feel like I've accomplished something.

1SG.M-NOM know DEF.SG.M boy₁-ACC about father₂-NOM of.M DEF.SG.M he₁ PST die and.INCL DEF.SG.M he₂-NOM PST tell 1SG.M-ACC about secret-NOM PST EXIST with.COM DEF.SG.M he₂

(or in more easy to understand notation: I know the boy₁ about father₂ of the he₁ died and the he₂ told me about secret was with the he₂)

[Kuchigakatai]

People who pronounce "spider" with [ɐɪ] (as in "fight, fire" as opposed to [aɪ] as in "find, file") are basically saying spite-er. Clearly many think spiders are full of spite.
~shower thoughts~

Consider this sentence: I know the boy whose father died, and he told me he had a secret. My question is, who had a secret?

I've noticed that pretty much all ancient Latin distinguishes this kind of thing, using is or hic for the last one mentioned (the latter = the father) and ille for the previous one (the former = the boy). I wonder if this was also true of natural speech, or if it was more of a mannered thing of the written language. So much writing consistently respects it though...

[Hominid]

People who pronounce "spider" with [ɐɪ] (as in "fight, fire" as opposed to [aɪ] as in "find, file") are basically saying spite-er. Clearly many think spiders are full of spite.
~shower thoughts~

I always pronounce that vowel as [ɐɪ] (or whatever) before intervocalic /d/, unless it's morpheme-final, I believe. So, the word "sider" (which I have never used, but apparently means "one who takes a side") is not a homophone of "cider" for me.

It gets weird when there's a <t> though. I'm pretty sure I pronounce "United States" with [aɪ] and "united" (the verb) with [ɐɪ], which makes no sense given the above rule, but that could be a weird artifact of how I learned those words when I was very young. I haven't paid attention to how other people with Canadian raising say those words.

Also, I definitely have [aɪ] in "fire".

[bradrn]

People who pronounce "spider" with [ɐɪ] (as in "fight, fire" as opposed to [aɪ] as in "find, file") are basically saying spite-er. Clearly many think spiders are full of spite.

People have different vowels in fight and find? It’s the same vowel for me.

EDIT: Actually, now that I think about it, there is a length difference, but I believe that’s just the usual pre-voiced-consonant vowel lengthening rule.

[Travis B.]

People who pronounce "spider" with [ɐɪ] (as in "fight, fire" as opposed to [aɪ] as in "find, file") are basically saying spite-er. Clearly many think spiders are full of spite.

I pronounce spider or cider not with [əe̯], as with fight, or fire before fortis obstruent) but with [əːe̯] (as with fire otherwise). Note that rider (and the nonexistent word sider) have [aːe̯] for me. Also note that latter and ladder are not homophones for me, as the former has [ɛ] while the latter has [ɛː] (yes, this is a true minimal pair for vowel length for me).

[anteallach]

People who pronounce "spider" with [ɐɪ] (as in "fight, fire" as opposed to [aɪ] as in "find, file") are basically saying spite-er. Clearly many think spiders are full of spite.

People have different vowels in fight and find? It’s the same vowel for me.

EDIT: Actually, now that I think about it, there is a length difference, but I believe that’s just the usual pre-voiced-consonant vowel lengthening rule.

It's basically the "Canadian raising" pattern (which is nowhere near restricted to Canada): the shorter allophone before fortis consonants has a raised starting point. Some other accents monophthongise the long allophone but not the short one.

However, there's also this thing about spider and some other words. Travis apparently has a diphthong in this word which is long (as expected before /d/) but with the raised starting point (normally only found before fortis consonants). I'm not sure what the conditioning for this is, or whether it's actually a marginal phonemic split.

Also some people apparently have a raised starting point before /r/.

[Pabappa]

It's just sporadic .... I dont think you can really write a decisive rule. It's like most splits, I think .... sometimes it happens, sometimes it doesnt. Like bath/trap for example. Anyway, for me "fire" and "wire" dont rhyme, .... "fire" has raising and "wire" does not. It makes no difference which sense of "fire" is meant ... "hire" and "fire" dont rhyme either.

Regarding the United States, I suspect i have both pronunciations depending on context. Way back in elementary school when we did the pledge of allegiance, i think it was a proper [aɪ] diphthong, not raised ... but maybe thats just because we were drawing out the pronunciation. Low vowels seem to be better for that .... I remember giving a presentation once where i said "uuuuuuuuuuu-NIIIIIIIII-ted STATES!" with a vowel so low i'd call it either IPA [ ɔ ] or IPA [ɒ].

[Travis B.]

Wire and fire both have [əe̯]~[əːe̯] ([əːe̯] when disyllabic) for me, as do lyre and hire, but liar and higher have [aːe̯]. Note that wired, fired, and hired all have [əːe̯].

[Nortaneous]

I have raising before monomorphemic [-ɾəɹ] but not before [-əɹ] (or [-gəɹ]) - so spider and cider have [ʌj], but rider, fire, tiger have [aj] ~ [a].

It would be possible to, with a sufficiently deep phonetic analysis, posit a rule of d > t / V_V.

[Travis B.]

I have raising before monomorphemic [-ɾəɹ] but not before [-əɹ] (or [-gəɹ]) - so spider and cider have [ʌj], but rider, fire, tiger have [aj] ~ [a].

It would be possible to, with a sufficiently deep phonetic analysis, posit a rule of d > t / V_V.

That doesn't work for me since fire and tiger also have raising for me.

[Qwynegold]

Consider this sentence: I know the boy whose father died, and he told me he had a secret. My question is, who had a secret?

I've noticed that pretty much all ancient Latin distinguishes this kind of thing, using is or hic for the last one mentioned (the latter = the father) and ille for the previous one (the former = the boy). I wonder if this was also true of natural speech, or if it was more of a mannered thing of the written language. So much writing consistently respects it though...

Huh, that's interesting. This conlang is inspired by Latin, but I didn't know about that.

[Kuchigakatai]

A list of all ISO 639-3 (2019) languages by code and their corresponding name on Wikipedia. Two mirrors:
- https://controlc.com/ca7d884a (ControlC.com, will expire in a month)
- https://pastebin.com/B42yGTwp (Pastebin.com, erroneously detected as having "offensive" content, maybe due to the Anal language, etc.)

For most of them, unless they already have the world "language", you can just add "language" after and that'll get you to the correct Wikipedia page. The list contains 8163 entries, although including a few entries like "[zxx] No linguistic content". It was derived from this Wikipedia list.

[Curlyjimsam]

Hmm, I thought so. I had another tussle with the relative clauses in a conlang, and I was trying to figure out what to do when there is something that can't be relativized. (I know that sentence had more to do with anaphora than relative clauses, but anyway...) What should I write in a grammar? Should I just write "this and this and this thing can't be relativized", without offering any suggestions on what to do instead?

Things like passive and applicative are commonly used by languages to turn things that can't be relativised into things that can. E.g. if you can only relativise subjects, the passive lets you turn objects into subjects so they can be relativised too.

[Qwynegold]

A list of all ISO 639-3 (2019) languages by code and their corresponding name on Wikipedia. Two mirrors:
- https://controlc.com/ca7d884a (ControlC.com, will expire in a month)
- https://pastebin.com/B42yGTwp (Pastebin.com, erroneously detected as having "offensive" content, maybe due to the Anal language, etc.)

For most of them, unless they already have the world "language", you can just add "language" after and that'll get you to the correct Wikipedia page. The list contains 8163 entries, although including a few entries like "[zxx] No linguistic content". It was derived from this Wikipedia list.

I wish the Finnish and Swedish versions of Wikipedia had articles about as many languages as the English one has. There are so many languages and dialects that just don't have any name in Swedish or Finnish.

[Qwynegold]

Things like passive and applicative are commonly used by languages to turn things that can't be relativised into things that can. E.g. if you can only relativise subjects, the passive lets you turn objects into subjects so they can be relativised too.

Yeah, I'm already using the passive voice for some things. But I'm not going to add any applicative into this language when I've gotten this far. :/