(Conlangs) Q&A Thread - Quick questions go here [2010-2020]

[Creyeditor]

Okay, if insertion/epenthesis and deletion are sandhi processes and you do not consider zero allophones, I understand how something can be sandhi without being allophony. If the number of segments is changed, it can be sandhi, but it can't be allophony.

[...]Sandhi - as I understand it - is when a phoneme undergoes transformation to a different phoneme altogether due to its environment at a word or morpheme boundary. [...]

Wait, wait. If a phone alternates with another phone under some condition, doesn't that make them allophones of a single phoneme? How can one phoneme undergo transformation to another phoneme?

The verb stem dviṣ combined with the 2SG suffix -si becomes dvekṣi (the vowel change is a regular unrelated process), whereby the disallowed cluster ṣs has been resolved by turning it into sth wildly different.

Could you not say that /ṣ/ has an allophone [k] before /s/ and /s/ has an allophone [ṣ] after /ṣ/?

All of k ṣ s are full phonemes in their own right, so this is clearly not an example of allophony (at least not in the strictest sense of the word).

This is actually not an argument, I think, because there are chain shifts that are considered allophony (e.g. /a/>[e],/e/> under some condition) and neutralizations that are considered allophony (e.g. German final devoicing where /d/ > [t] and /t/>[t]).

One of the first things I taught myself to say in Sanskrit when I was a silly teenager was I am the most beautiful. I made a (probably incorrect) superlative of śobhanah → śobhanatamah (most radiant ?) and then added the 1SG form of the verb 'to be' āsmi. Now sandhi operates here, too. The combination ah + ā becomes o and the whole thing becomes śobhanatamosmi. Again o is its own phoneme in Sanskrit. This is not allophony.

Okay, I agree that this is not allophony, but I would say that this is because it is fusion i.e. the number of segments is changed. /o/>[o] existing as a separate phoneme is not a good reason IMO.

[hr][/hr]
Now, maybe this has to do with a slightly different understanding of allophony. In my understanding it does not really rely on native speaker analysis, but instead on complementary distribution and alternation.
If one segment changes into another segment under certain phonological conditions, they are allophones.
If two sounds are in complementary distribution, i.e. they only occur under different, non-overlapping conditions, they are allophones only if they are also sufficiently similar phonetically.

To illustrate why this does not depend on other existing phonemes, let's say we have a language with intervocalic voicing and word final devoicing.
[at] but no *[ad]
[ada] but no *[ata]
and [at]+[a] is pronounced [ada]
and [ada] is clipped to [at] under a certain condition.
also [a]+[da] is [ada] and [a]+[ta] is also [ada]
Notice that they still contrast word initially. [ta] means something different from [da].

This means we have the following phoneme allophone pairs. Note how we have voiced and voiceless phonemes that each map to voiced and voiceless allophones. The existence of one phoneme does thus not prevent other phonemes from having the same allophone. This is what I would call neutralization.
/t/ [d]
/t/ [t]
/d/ [d]
/d/ [t]

I hope this has been somewhat understandable.

[DesEsseintes]

Wait, wait. If a phone alternates with another phone under some condition, doesn't that make them allophones of a single phoneme? How can one phoneme undergo transformation to another phoneme?

I should have worded that better. It should probably be sth like "when a phoneme assumes the quality of another phoneme in that language's inventory."

Could you not say that /ṣ/ has an allophone [k] before /s/ and /s/ has an allophone [ṣ] after /ṣ/?

You could, but I don't think it helps to lump it all under allophony. I see this as a morphophonological process at a different level than saying sth like "/s/ is dental before dental stops and alveolar before alveolar stops" (assuming there isn't a dental/alveolar distinction in the language).

Also, is that really what you think is going on here? That /ʂ/ is being realised [k] and /s/ as [ʂ]? I personally don't think it is.

This is actually not an argument, I think, because there are chain shifts that are considered allophony (e.g. /a/>[e],/e/> under some condition) and neutralizations that are considered allophony (e.g. German final devoicing where /d/ > [t] and /t/>[t]).

I would not consider a chain shift in vowel sounds allophony, but rather a morphophonological process, or if it happens systematically within morphemes, some kind of vowel harmony, which I don't consider allophony (but perhaps I'm wrong). There are exceptions I can think of though, like /i u/ lowering to [e o] before uvulars in kalallisut. By the way, I did not say that mergers weren't allowed under allophony. Final devoicing and a resultant merger should be considered allophony as the sounds do not change from coming into contact with another morpheme.

Your d/t example has no instances of your phonemes coming into contact with other morphemes so I cannot say whether there is any sandhi at work. But no, I didn't really understand the point you were trying to make with that example.

[Creyeditor]

Could you not say that /ṣ/ has an allophone [k] before /s/ and /s/ has an allophone [ṣ] after /ṣ/?

You could, but I don't think it helps to lump it all under allophony. I see this as a morphophonological process at a different level than saying sth like "/s/ is dental before dental stops and alveolar before alveolar stops" (assuming there isn't a dental/alveolar distinction in the language).

Oh, so the Sandhi processes you gave only happened with certin classes of morphemes? So they were morpheme specific phonological processes?

This is actually not an argument, I think, because there are chain shifts that are considered allophony (e.g. /a/>[e],/e/> under some condition) and neutralizations that are considered allophony (e.g. German final devoicing where /d/ > [t] and /t/>[t]).

I would not consider a chain shift in vowel sounds allophony, but rather a morphophonological process, or if it happens systematically within morphemes, some kind of vowel harmony, which I don't consider allophony (but perhaps I'm wrong). There are exceptions I can think of though, like /i u/ lowering to [e o] before uvulars in kalallisut. By the way, I did not say that mergers weren't allowed under allophony. Final devoicing and a resultant merger should be considered allophony as the sounds do not change from coming into contact with another morpheme.

Yes I agree that vowel harmony is really a difficult case.

Your d/t example has no instances of your phonemes coming into contact with other morphemes so I cannot say whether there is any sandhi at work. But no, I didn't really understand the point you were trying to make with that example.

The plus signs were meant to indicate morpheme boundaries. Sorry, I did not mention it. The point was that the difinition of allophony allows for phonemes 'to transform to other phonemes' or maybe their main allophones, but this could still be considered allophony. But it probably still misses the point because the phonological processes happen both inside a morpheme and at morpheme boundaries in this example.

[Salmoneus]

Whether sandhi is allophony is the wrong question - it's a category error. They're two different types of things, and the terms arise in different levels of analysis.

Allophony is a phonological term, a phonological phenomenon. Sandhi is a morphological term, a morphological process.

The process of sandhi will often employ a form of allophony (likewise, "words" employ phonemes but that doesn't mean that words are a subset of phonemes). But where it involves extreme alternations between phonemes that are not close to one another, it makes little sense to call it allophony. Does English have a phoneme /n*/, which only appears in one word ("an"), and which has two allophones, one a nasal and one a null?

[Creyeditor]

Whether sandhi is allophony is the wrong question - it's a category error. They're two different types of things, and the terms arise in different levels of analysis.

Allophony is a phonological term, a phonological phenomenon. Sandhi is a morphological term, a morphological process.

Okay, that is an interesting take on the question. Now, I have a follow up question. In tonal phonology, sandhi is often between phonological words (which are the domains for other morphophonological processes). Would you say than that sandhi is a mophological phenomenon that is only dependent on the adjacent word?

The process of sandhi will often employ a form of allophony (likewise, "words" employ phonemes but that doesn't mean that words are a subset of phonemes). But where it involves extreme alternations between phonemes that are not close to one another, it makes little sense to call it allophony. Does English have a phoneme /n*/, which only appears in one word ("an"), and which has two allophones, one a nasal and one a null?

At least for some Austronesian languages something similar is often assumed, a phoneme (or morphophoneme) /N/, that is only present in the prefix meN- (or whatever form you have in your language) and that triggers and undergoes allophony that is different from all other nasals. Some of them (in some languages) are null.

[Keenir]

Anyway, I guess what I was asking was, if anyone had any thoughts on what one could do to spice things up a bit and make it look less like it was kit-bashed

whats wrong with it being kit-based?

Nothing, inherently. I was just looking for some suggestions on how to mess with it a bit to help distance it from the source some more than I was already able.

well, maybe start with doing what you mentioned, and then see how it looks (I bet it won't look as kid-based as you think)...and then maybe reverse the direction its written in (either horizontally or vertically)

maybe add a rule like "in the script, every word must end in a vowel, even if the word doesn't."
(or, for added oddness, have the script end each word in a vowel only if the spoken word doesn't; if the spoken word does end in a vowel, the script's word cannot)

[Taurenzine]

So I have a question about articles: In english, articles used with plural nouns are a little strange; the article "the" works fine, but not the article "a". Some examples are that saying "The Apples" works but saying "An Apples" doesn't. You have to say "Some Apples" instead. This is a silly question because I think I already know the answer, but both of these articles could theoretically work without needing a system like this, right? If you just used the articles and attached it to a noun, singular or plural, it could work, right?

[qwed117]

So I have a question about articles: In english, articles used with plural nouns are a little strange; the article "the" works fine, but not the article "a". Some examples are that saying "The Apples" works but saying "An Apples" doesn't. You have to say "Some Apples" instead. This is a silly question because I think I already know the answer, but both of these articles could theoretically work without needing a system like this, right? If you just used the articles and attached it to a noun, singular or plural, it could work, right?

Sure. The thing in english is a number concord system, because some is the plural of a, to some degree. In a system with number concord, you could see something like "An apples"

[KaiTheHomoSapien]

Confusingly enough, "some" can be a singular indefinite as well: "some woman called this afternoon..."

But yes, there's no reason why you can't have a dedicated plural indefinite article, or an indefinite article that works with both singular and plural nouns.

In English, the plural alone seems to almost be inherently indefinite ("I bought apples at the store" is indefinite. "Some" wouldn't really change the sense here), which may explain why indefinite plural articles are rare or there is no dedicated form other than a specialized use of "some".

[Sumelic]

So I have a question about articles: In english, articles used with plural nouns are a little strange; the article "the" works fine, but not the article "a". Some examples are that saying "The Apples" works but saying "An Apples" doesn't. You have to say "Some Apples" instead. This is a silly question because I think I already know the answer, but both of these articles could theoretically work without needing a system like this, right? If you just used the articles and attached it to a noun, singular or plural, it could work, right?

Theoretically, it is possible to have articles, affixes or clitics that indicate degree of definiteness, but not number. I can't think of any specific example right now of a language where the main/only indefinite article is number-invariant, but probably there are some.

However, there are specific common paths of development for articles that have contributed to the situation in English. In a great many languages (including English) that have indefinite articles, they developed from the numeral "one". So in languages with this path of development, there is often a natural connection between an indefinite article and singular grammatical number.

[Sumelic]

Hmm. Actually, maybe I shouldn't have said that "probably there are some" in my previous post. At least, "Determination" from 1969 (by Edith A. Moravcsik) doesn't seem to mention any: https://archive.org/stream/ERIC_ED10571 ... te+article

[svld]

Anyway, I guess what I was asking was, if anyone had any thoughts on what one could do to spice things up a bit and make it look less like it was kit-bashed and also end up resulting in characters that wouldn't immediately be mistaken for just... "wrong Tangut."

Stroke shapes of Tangut seems to be the same as chinese calligraphy, maybe you can mess with it's stroke shapes?

[Omzinesý]

So I have a question about articles: In english, articles used with plural nouns are a little strange; the article "the" works fine, but not the article "a". Some examples are that saying "The Apples" works but saying "An Apples" doesn't. You have to say "Some Apples" instead. This is a silly question because I think I already know the answer, but both of these articles could theoretically work without needing a system like this, right? If you just used the articles and attached it to a noun, singular or plural, it could work, right?

Theoretically, it is possible to have articles, affixes or clitics that indicate degree of definiteness, but not number. I can't think of any specific example right now of a language where the main/only indefinite article is number-invariant, but probably there are some.

However, there are specific common paths of development for articles that have contributed to the situation in English. In a great many languages (including English) that have indefinite articles, they developed from the numeral "one". So in languages with this path of development, there is often a natural connection between an indefinite article and singular grammatical number.

But definite articles usually derive from demonstratives like "that" and "those". Old English also had a plural demonstrative þā. But it seems that singular definite article can much easier be generalized to the plural than the indefinite one.

Many language have a definite article and zero for indefiniteness (Welsh, Arabic etc). The zero of course appears both in the singular and plural.

[DesEsseintes]

Does anyone remember a good guide to writing phonotactics and syllable structure in terms of restrictions? I'm sure I've seen one on here - or referred to here - but so far I can't find it.

[Creyeditor]

I do not know of any guide, but if you are interested in an overview on the way people have looked at it in natlangs, you might want to look at ANUs World Phonotactics Database.

Some things that (also) come to mind:

• syllable structure
• word minimality/maximality constraints
• word structure
• foot structure
• allowed coda/onset consonant clusters (generalizations and exceptions)
• allowed consonant clusters across syllable boundaries (generalizations and exceptions)
• morpheme structure constraints
• (how to handle) vowel hiatus

[DesEsseintes]

I do not know of any guide, but if you are interested in an overview on the way people have looked at it in natlangs, you might want to look at ANUs World Phonotactics Database.

Some things that (also) come to mind:

• syllable structure
• word minimality/maximality constraints
• word structure
• foot structure
• allowed coda/onset consonant clusters (generalizations and exceptions)
• allowed consonant clusters across syllable boundaries (generalizations and exceptions)
• morpheme structure constraints
• (how to handle) vowel hiatus

Thanks for the detailed answer, Crey, but I'm thinking of a guide I saw that had all kinds of notation and parsing algorithms of some sort. If I remember correctly, one of the objectives of the guide was to show that if restrictions are defined and ordered well enough, and methods to solve violations encoded, any input could be turned into a permissible word form. Does anyone remember such a thing? (Did I dream it?)

---

And I have another unrelated question.

TLFKAT has the following plain nasal stops: /m n ŋ ŋʷ ɴ ɴʷ/. I'm considering a morphophonological process whereby the sequence /nw/ becomes m, but /ŋw ɴw/ would just become the labialised nasals. Is it unnatural that it's the alveolar/n/ that becomes m? I feel like /ŋw/ is much likelier to do that (but I don't want that cos I want ŋʷ/. Thoughts?

[Clio]

Does anyone remember a good guide to writing phonotactics and syllable structure in terms of restrictions? I'm sure I've seen one on here - or referred to here - but so far I can't find it.

Micamo posted this on the CBB, on the off chance that's what you're thinking of.

[Creyeditor]

That's a really nice model she has

[Sumelic]

And I have another unrelated question.

TLFKAT has the following plain nasal stops: /m n ŋ ŋʷ ɴ ɴʷ/. I'm considering a morphophonological process whereby the sequence /nw/ becomes m, but /ŋw ɴw/ would just become the labialised nasals. Is it unnatural that it's the alveolar/n/ that becomes m? I feel like /ŋw/ is much likelier to do that (but I don't want that cos I want ŋʷ/. Thoughts?

That seems OK. I'm not sure. In many languages, like English, coronals behave like they are less marked than velars. But it's true that velars fairly often assimilate to following labials in a number of languages. One example I started thinking of was Latin, where /dw/ became /b/ word-initially, but then I realized that doesn't really give a clear answer to your question because Latin had already lost word-initial /gw~gʷ/ via converting it to /w/.

I think the fact that it is morpho-phonology probably gives you a bit more latitude, also, so I would say to go for it!

[Taurenzine]

Is there a point in having a pronouns for "it" in my language if I have a demonstrative pronoun that means "the thing"? like, I don't need it right?