User talk:TheseusDeuteros

Feel free to leave a message here!

TheseusDeuteros (talk) 19:16, July 28, 2013 (UTC)

Ánagin

 * Hey, I'll give my review here if you don't mind :D

So far, I've edited your phonology table a bit, merging the top cells (and relocating /ɦ/ to nonsibilant fricatives from plosives as I believe that was just a random mistake). I'll have to comment that, so far, your phonology is something that just doesn't look that possible to me.

I mean, the phonology is very unstable and statistically really hard to come by, if not impossible. There's just too many gaps in it and a very noticeable lack of balance.

First, [d̪] and [d] are barely - if not not-at-all - distinguishable, just like [ʋ] and [v]. You lack [p], while having a typologically unlikely heterorganic affricate, [ps], and you have [tʃ] but no [ʃ], and [ʒ] but no [d͡ʒ]. You basically have /tʃ ʒ/ when it would be much more believable to have /ʃ ʒ/ or /tʃ dʒ/. You have distinctive labialisation on only one sound - which isn't how suprasegmentals really work - that is, /g/ vs. /gʷ/ - I'd expect at least a /kʷ/ (I don't know of any language that had /gʷ/ but not a /kʷ/ - even Latin had /kʷ gʷ/, and English has them in the form of /kw gw/); you also have labialisation without an assorted /w/, which I don't think is impossible, but can't cite any example like that. You also have /ʒ s/ but no /z/.

Now, second, concerning the vowels: you don't need to use /ä/ - that is just plain Wikipedia being stupid. In general linguistics, /a/ is the central sound and /æ/ is the front (thus, /ɑ/ is their back counterpart) You also have /y/ but lack /i/ - this is, I believe, impossible. I mean, Attic Greek had /i y/ but no /u/, but /u y/ is just something rather hard to have. Languages like to economise on features - if one front vowel is rounded, it'll have an unrounded variant, and there usually will be several more front unrounded vowels. I've made your table neater (deleted unused cells) and rewrote the terminology so that it uses "High > Low" instead of "Close > Open" as that particular type of terminology is another Wikipedia stupidity example: most linguists use "High > Low" in works and it's a generally accepted thing to do.

The phonological constraints and phonotactics are actually quite good, though you're lacking a syllable structure.

Well, that's it - if you need anything explained, I'm not that far away :D 09:16, August 12, 2013 (UTC) ~) The Elector, Darkness Immaculate

Hi! Well, I'll just start from the top...

That with /ɦ/ must've been a stupid mistake indeed. But for some reason the table messed up a couple of times, so that might've also been the reason for its weird placement :)

Oh, if it looks quite impossible, then let's see what we can do about it!

First off,  [d̪] and [d] aren't actually two separate sounds in Ánagin; they're allophones. Only in a very limited amount of situations, /d/ is pronounced [d̪]. Does that mean I should remove it from the table and only make a note about it, or something else...?

Then, you say [ʋ] and [v] are barely distinguishable, but I'm not sure if I agree here (unless Wikipedia has made another mistake). Namely, in Dutch, my native language, [ʋ] is the sound represented by 'w', as in "woord" (meaning 'word'), whereas [v] is the 'true v' just like in English. For me, [ʋ] is much closer to [w], I think, than to [v].

Now moving on to [ps]! This is an oddity in Ánagin I can explain :) The ancient languages that preceded Ánagin (most notably Proto-Tágwerian) did have the sound [p]. It was a very common sound, actually. They also had [ps]. First, two separate letters were used, but a separate letter representing [ps] started to emerge later. It wasn't used on all islands surrounding Tánagwer, but it became more common over time. Over the ages, though, Ánagin started to develop and [p] changed to [b]. After a couple of centuries, [p] had disappeared completely. However, pronouncing [bs] is rather difficult! Therefore, [ps] still exists. It's a very rare sound, though. I believe I won't create more than twenty words that contain it.

As for the siblant fricatives and the affricates, I'll explain as much as I can explain. I might have to change some things indeed, though. [tʃ] is just a separate sound in Ánagin and [ʃ] indeed doesn't exist. Is that really that weird? And yes, Ánagin does have [ʒ], but it has a story similar to the explanation of [ps], as it's very uncommon too. It occurs when 'r' and 's' follow each other. Just like in Polish 'rz', 'rs' is pronounced [ʒ]. In earlier times, the people of Tánagwer did pronounce it like [rs], later it became [rz] and finally [ʒ]. However, does this make any sense at all?

The labialisation of 'g', makes a separate letter for [gʷ]. And yes, it is distinctive, but only in a few words. Maybe I'll add [kʷ] too, though, if that'd make more sense. But [gʷ] is a separate letter in the alphabet... So would I then also have to create a separate letter for [kʷ]? Or would it also be possible to have [gʷ] as a separate letter, but [kʷ] as 'k + w', or [k + ʋ]?

Lastly, for the consonants, yes, Ánagin doesn't have [z], but does have [s] and [ʒ] in a few words. Is that impossible?

When it comes to the vowels, I indeed might have to change some illogical things. Let's see...

I've got rid of [ä] by now, also the note about it.

The main problem indeed is the lack of [i]. Is it impossible too, if I tell you that [y] (represented by 'ý' in Ánagin's orthography) is hardly used, only in roughly twenty-five words? Namely, I'd like Ánagin not to have [i], since it doesn't fit with the 'sound' I want to give it.

Anyways, that's all I can say. Oh, and thanks as for the phonological constraints! That was a lot of work! And yes, the syllable structure is missing, but I'm working on that. I was thinking of (C)(C)V(C)(C).

EDIT - By the way, would you be able to give me your e-mail adress, perhaps? Because it's rather inconvenient talking through Wikia's talk pages.

Ánagin again
1) You should probably find a way to distinguish phones and phonemes in the table. I do that by only including phonemes in my tables; you can describe allophones elsewhere, such as another section. You could also include both allophones and phonemes in the table, but put (brackets) around the allophones

2) As to [ʋ] vs. [v], they might be distinguished here and there, but Dutch /ʋ/ is closer to [β] than, say, my /ʋ/, which is closer to [v]. I mean, it's more distinguishable than [d̪ d], but the distinction also often has some other elements

3) Beware, though, as if it's a rare sound, it'll likely just fall back to [p]

4) You know, Polish  isn't actually [ʒ], or even voiced most of the time. It's more of a [z̠], and it assimilates to voicing to previous consonants - = [ts̠ɨ]. It didn't come out as a result of [r+z], as is etymologically traced to  (probably something like [trɪjɛ] from Proto-Slavic. If it came from [ɾz], I'd expect it to become [r] (trilled) rather than [ʒ]

5) I'd say you should first do the phonology, then do the orthography. You should focus on a balanced phonology, sure, but if you do irregularities, they should be natural. I mean, in Kti, I have a consonant system of /t d k ʔ s z ʃ ʒ x m n r/, but it's relatively natural; the instability surfaces in the peripheral dialects that solve the older system in different way (such as Coastal-Eastern /t d k ŋ ʔ s z ð ʂ ʐ x m n r/ ) as Kti is just one result of language change. Phonologies shouldn't be designed according to orthographies, it's the orthographies that should be designed according to phonologies.

X) You also have <á> in the language's name, but no explanation, and you don't mention it at all after that.

6) The frequency of [y] doesn't really matter to its presence - I mean, if it's so infrequently used, it'll just probably shift to [i]. Hell, it'll probably shift to [i] anyhow as it doesn't have anything to contrast the [+rounded] feature with, and that [y] is harder to pronounce than [i]. You know, even in a system that actively distinguishes between [i ɛ] and [y œ], they can, and do, merge into a simple [i ɛ] set. This is what happened to Bavarian German (Vögel > Vegln (birds)- ignore the added <-n> as it's analogy - demonstrates the unrounding of <ö> to )

7) Be careful about (C)(C)V(C)(C) as that means you could, in your case, have words such as =  = <[CC-V-CC]-[CC-V-CC]>. You'll probably need to tune it to your liking, such as limiting it to, say, "CCVCCVCC", that is, to not allow more than two consecutive consonants in the middle of words or something.

8) I'm at "xxxxxxxxxxxxxxxxxxxxxxxxxxx" :D

13:22, August 12, 2013 (UTC) ~) The Elector, Darkness Immaculate

'''I'll reply via e-mail, okay? And thank you very much again, as well!'''