Conlangery Podcast (Latest Ep: #94: Face and Politeness)

[clawgrip]

Japanese also has this weird thing where asking a question with a negatively conjugated verb means a suggestion. So 行かないか – literally "go.NEG INT" means "shall we go?" or "do you wanna go?". Dunno how widespread this might be. And か also means or, so it fits with that particular tendency.

This is a standard function of politeness and is quite widespread. While in English we express politeness generally by either obscuring our true question, "I was wondering if you knew anyone who could help," or making the question (and thus the answer) hypothetical, "Would you like to go?" Japanese expresses politeness by making it as easy as possible for the addressee to reject the proposition. The speaker will make it easy for addressee to confirm the opposite of what they actually want confirmed by asking negative questions, e.g. "You likely don't know anyone who can help, right?" "Are you not going?" The addressee can theoretically confirm these statements with "hai" (though not in practice, since we all recognize politeness strategies for what they are).

These politeness patterns carry over even into informal speech, e.g. ケイタイ知らない？Keitai shiranai? portable.(phone) know.NEG? "Do you know where my phone is?" is a normal way of asking questions. This is just like politeness patterns carrying over into informal speech in English, e.g. "Any idea where my phone is?" uses the "any idea" polite obfuscation phrase even in an informal situation.

[finlay]

Widespread crosslinguistically? In other languages?

[clawgrip]

Widespread crosslinguistically? In other languages?

Ah, by widespread I meant in Japanese. Maybe I misunderstood you there.

[finlay]

Yeah, I wasn't very clear. I was wondering if you would be able to find that in other languages or not.

[Ollock]

Conlangery #46: Conlanging for Conworlds

[finlay]

I'm SO glad that Japanese is a language that likes loanwords... It'd be so much harder if they tried to use calques or invented words for everything (they have these, as you alluded, for older words like 電車 for train or 電話 for phone (electric-car and electric-talk) but not newer ones like コンピュータ (compyuuta) - although wiktionary gives an alternative for computer made from kanji, which I reckon hasn't caught on in quite the same way...)

[Rodlox]

I was reading Billions & Billions by Carl Sagan, and one chapter opens with this line from Euripides' Ion:

Is this, then, true or mere vain fantasy?

Is the "then" a discourse particle?

[Gojera]

Yeah, I wasn't very clear. I was wondering if you would be able to find that in other languages or not.

I've always thought it was very analogous to some forms of suggestions in English. "Why don't we go to Qdoba this time?"

[Ollock]

Conlangery #47: Isolating and Analytic Languages

[finlay]

You were discussing Japanese earlier; I'm fairly sure conjoined NPs use と to join them and only have the case particle at the end, which is a fairly good case for them being particles or postpositions rather than suffixes.

So,
バナナを食べる - banana o taberu - eat a banana
リンゴを食べる - ringo o taberu - eat an apple
リンゴとバナナを食べる - ringo to banana o taberu - eat an apple and a banana

There's evidence from prosody that they're at least phonologically part of the previous word, so you might want to call them a clitic. This is why some people hyphenate them when transcribing. (ringo-to banana-o taberu)

[Ollock]

You were discussing Japanese earlier; I'm fairly sure conjoined NPs use と to join them and only have the case particle at the end, which is a fairly good case for them being particles or postpositions rather than suffixes.

So,
バナナを食べる - banana o taberu - eat a banana
リンゴを食べる - ringo o taberu - eat an apple
リンゴとバナナを食べる - ringo to banana o taberu - eat an apple and a banana

There's evidence from prosody that they're at least phonologically part of the previous word, so you might want to call them a clitic. This is why some people hyphenate them when transcribing. (ringo-to banana-o taberu)

That's what I was thinking was the case, but I avoided answering on the podcast because I wanted to get just this kind of clarification. Thanks.

[Gojera]

Yep. Japanese has been borrowing heavily from English recently, but previously it used to borrow more from other European languages, German especially. So パン pan "bread" is from a Romance language, while アルバイト arubaito "part-time work" is from German Arbeit. I was looking up the Japanese words for climbing ropes and pitons: they're from German, ハーケンとサイル.

Y'all have talked about Japanese quite a bit the last few sessions; it has some fairly weird features. If y'all can find a good grammar, you might think about doing it as a featured natlang.

Also, you've hinted before that you might do one of your (the hosts') conlangs as a featured conlang. I suspect that you're also a little reluctant to do that sort of thing, to toot your own horn, so to speak. But I think it would be a really good idea. We've heard you talk about them in dribs and drabs over 40+ episodes, and it's helpful, I think to know where you're coming from. So, I think you should feature Aeruyo or one of William Annis's languages, or something. I also always thought Inyauk had a really nice aesthetic, or maybe you could get Bianca on to talk about her cat or house elf language

[clawgrip]

You were discussing Japanese earlier; I'm fairly sure conjoined NPs use と to join them and only have the case particle at the end, which is a fairly good case for them being particles or postpositions rather than suffixes.

So,
バナナを食べる - banana o taberu - eat a banana
リンゴを食べる - ringo o taberu - eat an apple
リンゴとバナナを食べる - ringo to banana o taberu - eat an apple and a banana

There's evidence from prosody that they're at least phonologically part of the previous word, so you might want to call them a clitic. This is why some people hyphenate them when transcribing. (ringo-to banana-o taberu)

I'm not sure if a native speaker perspective really matters or not, but in my experience, most Japanese people seem to perceive these to be a part of the previous word, i.e. ringoto, bananawo. In my experience, Japanese people tend to write them as such in Romanized transcriptions. Here is an example of text written by a Japanese speaker: "gengowa ookuno hito to kaiwa dekiru saisyono syudan nanode motto ookuno hitokara ironna tisiki wo manabitaidesu. tada sonokuninokotobawo hanaseru dakedeha jyuubun dewa naito omounode hibi benkyou no mainitidesu!!" Interestingly, this person has chosen to write "to" independently after a noun (hito to), but has connected it to the verb (naito). Also, wo is variable in this text: tsisiki wo but sonokuninokotobawo.
I'm not saying this person's blog definitively shows us how we should treat the words, only that it is an interesting look into one native speaker's opinion of how Japanese is divided up.

[finlay]

You were discussing Japanese earlier; I'm fairly sure conjoined NPs use と to join them and only have the case particle at the end, which is a fairly good case for them being particles or postpositions rather than suffixes.

So,
バナナを食べる - banana o taberu - eat a banana
リンゴを食べる - ringo o taberu - eat an apple
リンゴとバナナを食べる - ringo to banana o taberu - eat an apple and a banana

There's evidence from prosody that they're at least phonologically part of the previous word, so you might want to call them a clitic. This is why some people hyphenate them when transcribing. (ringo-to banana-o taberu)

I'm not sure if a native speaker perspective really matters or not, but in my experience, most Japanese people seem to perceive these to be a part of the previous word, i.e. ringoto, bananawo. In my experience, Japanese people tend to write them as such in Romanized transcriptions. Here is an example of text written by a Japanese speaker: "gengowa ookuno hito to kaiwa dekiru saisyono syudan nanode motto ookuno hitokara ironna tisiki wo manabitaidesu. tada sonokuninokotobawo hanaseru dakedeha jyuubun dewa naito omounode hibi benkyou no mainitidesu!!" Interestingly, this person has chosen to write "to" independently after a noun (hito to), but has connected it to the verb (naito). Also, wo is variable in this text: tsisiki wo but sonokuninokotobawo.
I'm not saying this person's blog definitively shows us how we should treat the words, only that it is an interesting look into one native speaker's opinion of how Japanese is divided up.

Native speaker's judgements basically count for shit in theoretical linguistics (you could argue that this is a bad way of approaching it, but this ain't the time), but they're certainly interesting. The very fact that there's variability in when this person puts a space before certain particles is telling, too; and I would put this more down to the fact that Japanese doesn't have spaces in the native orthography, so it's essentially arbitrary where we put them in transcribed Japanese. Japanese people when transcribing tend to keep in the orthographical quirks of kana spelling a lot more than English people; hence here you see 'si' and 'sya' rather than shi and sha, and I think here you've got ha instead of wa and wo instead of o. I've seen people spelling 新宿 as Shinjyuku on many occasions, whereas I write it Shinjuku, because writing it as j rather than z (the kunreisiki transcription should be Sinzyuku) precludes the need for a y (and also because this is 'official' insofar as it's the one that the train companies use). Of course, a lot of the conventions of Hepburn romanisation are based on English phonology rather than Japanese, meaning that there are kind of extra rules to learn compared with writing it in kana.

I'll just come back to lack of spaces... in Japanese words are quite often separated by which system they're written in; it's usually fairly clear that something in katakana is different from something in kanji. I guess hiragana are a bit different because they're sort of more often used for suffixes, especially on the verbs, but I reckon you can tell from the example above that at least the writing system considers the suffixes "separate" from the katakana words here.

[clawgrip]

Yeah, I know that trusting a native speaker with no training to describe their language is asking for trouble. I just find it interesting that in my experience, Japanese people as a general rule tend to consider the particles to be a part of the word they follow. My wife seemed certain of it. But even if it is a part of the word, that doesn't really mean much. Latin has the serial que that is considered part of the word (in that it is not separated by a space), yet it is still a clitic that applies to the whole phrase.

On the subject of orthography, I can live with tyu and sya and so on, but jyo and cyo for example really bother me because they're bastard hybrids of Nihonshiki and Hepburn and don't really benefit anyone.

[Gojera]

I've also heard native speakers say that before, that they didn't know that the particles were separate from words until they went to school, they had thought they were part of the word. I think it's a common perception among Japanese, but we're always taught they're short words called particles.

Some of them are purely case-marking, but the particles include not only these but a variety of other grammatical and discourse words that are clearly not. They mark individual words, whole NPs, or clauses, and sometimes these uses overlap in difficult-to-pin-down ways.

が ga is used to mark a noun that is a subject or direct object of a verb, or, following a verb, a conjunctive particle meaning "but". の no is a genitive marker or it nominalizes a verb.

で de, like に ni, is usually used to mark the location, instrument, or manner of a noun, but can be interpreted as a form of copula, and も is a conjunctive particle meaning "also". But then you get words like どこも dokomo "everywhere, nowhere", どこでも dokodemo "anywhere", and どこにでも dokonidemo "to anywhere". Whereas でも demo after a verb means "but" and ので node after a verb means "because".

[clawgrip]

I guess you're talking about how some particles can occur in sequence, while others are prohibited. は, を, and が must be the last particle in the sequence, but anything else is fair game.

Even weirder is that in Old Japanese, が and の used to be variants of the same thing, though I'm not knowledgeable enough about Old Japanese to say exactly where they were similar and where they differed. But that's why you can find が in various place names where の would make more sense, e.g. 自由が丘 Jiyūgaoka, 関ヶ原 Sekigahara, 茅ヶ崎 Chigasaki (incidentally ヶ is an abbreviation of 箇, something about which I asked and subsequently stumped multiple Japanese people and was forced to research on my own).

[Ollock]

Conlangery #48: Designing a Sound System

[Aurora Rossa]

Regarding the Nahuatl vowel system, I have often heard that languages generally prefer front vowels over back vowels. The back vowels apparently sound more similar to each other on an acoustic level than the corresponding front vowels and people thus find them somewhat harder to distinguish. That leads some languages to merge back vowels while allowing more distinctions among front vowels, so systems like /a e i o/ or /a e E i o u/ show up a lot.

Something about Volapük has always reminded me somehow of steampunk or perhaps dystopian fiction. It seems like something you would see on the signage of some fictional Eastern European country from the early twentieth century.

[patiku]

Where's you hear that? Was someone talking about it down at the club?

_{i said it in #dts a while back}

[Aurora Rossa]

Where's you hear that? Was someone talking about it down at the club?

_{i said it in #dts a while back}

Hmm, really? I must have forgotten that you said it, but gotten the impression stuck in my brain unconsciously.

[finlay]

Yeah, I know that trusting a native speaker with no training to describe their language is asking for trouble. I just find it interesting that in my experience, Japanese people as a general rule tend to consider the particles to be a part of the word they follow. My wife seemed certain of it. But even if it is a part of the word, that doesn't really mean much. Latin has the serial que that is considered part of the word (in that it is not separated by a space), yet it is still a clitic that applies to the whole phrase.

On the subject of orthography, I can live with tyu and sya and so on, but jyo and cyo for example really bother me because they're bastard hybrids of Nihonshiki and Hepburn and don't really benefit anyone.

To follow up on this a little bit, I was talking to a Japanese guy yesterday who said that を is pronounced [wo] or [uo], differently from お, and that おう and オー are different vowels too: [ou] and [o:]. I've always been taught that each pair is pronounced identically, and this is the way they're represented in most romanizations (the first two are both o and the second two are both ō in Hepburn). We were mainly speaking in English to each other, and my Japanese isn't that great, so I didn't really get the chance to test it out by hearing him speaking naturally, but is this actually the case or is it just a misconception based on the kana orthography?

(He was perfectly aware that は and へ have two pronunciations, and that はし means bridge with one accent and chopsticks with another, so he's not completely naive. I'm not sure if it's common to know that; one of my receptionists gave it as an example of a homophone in Japanese. Also, she's adamant that 7 in Japanese is spelt しち but pronounced ひち. Perhaps there's a dialect thing going on somewhere...)

[clawgrip]

To follow up on this a little bit, I was talking to a Japanese guy yesterday who said that を is pronounced [wo] or [uo], differently from お, and that おう and オー are different vowels too: [ou] and [o:]. I've always been taught that each pair is pronounced identically, and this is the way they're represented in most romanizations (the first two are both o and the second two are both ō in Hepburn). We were mainly speaking in English to each other, and my Japanese isn't that great, so I didn't really get the chance to test it out by hearing him speaking naturally, but is this actually the case or is it just a misconception based on the kana orthography?

を is usually pronounced /o/ but it can be pronounced /wo/ sometimes when a speaker is trying to be clear. When pronouncing 眞鍋かをり's name, for example, no one says Kawori, they say Kaori, just as they would for anyone whose name is spelled かおり.

おう depends on context. When う is part of a different morpheme, it does not usually form /o:/. 思う is pronounced /o.mo.u/, not /o.mo:/. 丸の内 is pronounced /ma.ru.no.u.ʨi/ not /ma.ru.no:.ʨi/ (However, I am pretty sure that 下総, <Shimousa>, despite /o/ and /u/ being of separate morphemes, is in fact pronounced /ɕi.mo:.sa/ but this is an exception).

If you see おう in something like お父さん (おとうさん) or 東京 (とうきょう) it is pronounced /o.to:.san/, /to:.kjo:/. However, when people are actually spelling it out orally, they may pronounce them based on spelling, i.e. /to.u.kjo.u/, but this is only for clarity.
This is exactly the problem you were discussing with native speakers and their piss-poor knowledge of their own language. This person recognizes that these letters can be pronounced in a certain way and for whatever reason assumes it's always pronounced that way, but you just can't hear it, or some ridiculous notion like that.

(He was perfectly aware that は and へ have two pronunciations, and that はし means bridge with one accent and chopsticks with another, so he's not completely naive. I'm not sure if it's common to know that; one of my receptionists gave it as an example of a homophone in Japanese. Also, she's adamant that 7 in Japanese is spelt しち but pronounced ひち. Perhaps there's a dialect thing going on somewhere...)

This is funny; your acquaintance is actually branding himself as a native Shitamachi Tokyoite (called Edokko) by insisting on this. The Edokko accent has merged /ç/ and /ɕ/ (this means (ひ and し are indistinguishable), but this is not the standard across Japan by any means.

In my experience, Japanese speakers are notoriously bad at recognizing anything to do with pronunciation. Try getting someone to believe that ん actually has about 5 different pronunciations (all allophones); people will not believe you. The excessive variation of this phoneme is also a major headache for Japanese learners and easily identifies many non-native speakers who simply can't get it right.

[finlay]

That maybe wasn't too clear: they're two different people (which is why I italicised "she"). The guy I know is from Shizuoka. I don't know where the woman's from. The only reason I introduced her into the story was that she gave はし as an example of a homophone, whereas he gave it as an example of a minimal pair with an accent. Perhaps she was glossing over the accent thing.

As for ん, you mean like [m n ŋ ɴ ˜j] or something, right, depending on what the next consonant is? (I've noticed that 新大久保 is pronounced [ɕĩːjoːkubo] or something like that)

[clawgrip]

That maybe wasn't too clear: they're two different people (which is why I italicised "she"). The guy I know is from Shizuoka. I don't know where the woman's from. The only reason I introduced her into the story was that she gave はし as an example of a homophone, whereas he gave it as an example of a minimal pair with an accent. Perhaps she was glossing over the accent thing.

As for ん, you mean like [m n ŋ ɴ ˜j] or something, right, depending on what the next consonant is? (I've noticed that 新大久保 is pronounced [ɕĩːjoːkubo] or something like that)

Oh right, I thought I had seen she, but when I glanced back I saw he. Anyway Just mentally change that part of my post.

That's exactly what I mean about ん. [m n ŋ] are simple enough because they're extremely common variations in all languages, but [ɴ] and especially [˜j] give people a lot of trouble. It took me a long time to comfortably pronounce [˜j] without great effort.