Let's see how this one goes...

[mira]

Aight, time for another attempt at a conlang. Something new this time. This thread won't be organised at first and I'll just throw notes and ideas down here to begin with. Once I get something concrete, I'll make a more nicely formatted post on it and add it to a contents thingy below or something. If you're on the CBB Discord then you'll have seen the writing system already (and also know that I made it before the phonology but shhhhhh).

Feel free to post thoughts and suggestions and such, I'll try and respond to all of them (no Janko, I do not have any numbers for you).

This one'll be a priori, probably agglutinative with some hints of isolating. OSV order (because it's great), hexadecimal number system, and some other stuff. Main goal is to have some form of lexicon for once as I've always got stuck at that point in the past. I've even made a pretty dictionary template that I'll be using as I build up the lexicon. I'm going to try and grow the lexicon alongside the grammar, so I can put in examples as I create the grammar, rather than putting it in as an afterthought.

Contents
*tumbleweed* (give me a minute)

[Davush]

The lexicon has always been the downfall of my conlangs (I've recently reached the grand total of around 300 words in Qutrussan which is a first) so I'm with you on that. Nice templates always help as well. Also looking forward to seeing the writing system!

[mira]

Phonology things first, and I'll do the bit on the writing system next. I tried to be fairly symmetrical with this, but throw in the odd extra sound to make the phonology a bit bigger and more interesting.

Code: Select all

/ p b t d k g   / < p b t d k  g    >
/  m   n   ŋ    / <  m   n   ng     >
/ f v s z ʃ ʒ h / < f v s z sh zh h >
/  w   ɾ   j    / <  w   r   j      >
/      t͡s t͡ʃ    / <      ts  tc     >
/ a e i o ɯ     / < a e i o u       >

Nothing too fancy. Rough symmetry with a front/mid/back set of plosives, nasals, fricatives, and approximants, throw in a /h/ and a couple of affricates, then a 5-vowel system.

Phonotactics is where I'm trying to make it a bit more interesting.

• (C)V(N/V) syllable structure
  • Any consonant at the start of the syllable. Optional at the start of a word, compulsory within a word.
  • Put a vowel in the middle
  • Optional nasal/vowel at the end (but not both). The nasal at the end is represented only with an <n>, but will change between the nasals depending on the following consonant (same/similar place of articulation). Defaults to /n/ when word final. Stays as /n/ before <tc> though.
• If the last syllable in a word starts with a nasal or a fricative and is paired with the same vowel as the previous syllable, then the final vowel isn't vocalised, so <kana> is just /kan/, and <taishi> is just /taiʃ/.
• Available diphthongs in syllables are: <ai au ei eu> which become /aɪ aɯ eɪ eɯ/.
• <wu uw ji ij> are all invalid.

[mira]

The lexicon has always been the downfall of my conlangs (I've recently reached the grand total of around 300 words in Qutrussan which is a first) so I'm with you on that. Nice templates always help as well. Also looking forward to seeing the writing system!

300?! My record is about 28, with a close second place of 3. I'm just beginning the writing system post now, so you won't be waiting too long for that.

[Davush]

The lexicon has always been the downfall of my conlangs (I've recently reached the grand total of around 300 words in Qutrussan which is a first) so I'm with you on that. Nice templates always help as well. Also looking forward to seeing the writing system!

300?! My record is about 28, with a close second place of 3. I'm just beginning the writing system post now, so you won't be waiting too long for that.

My average is probably closer to 10 throughout the hundreds (!) of conlangs I've started and stopped, so I'm sure you can beat your record there.

[mira]

Writing system time! I'm pretty proud of this one. It's a syllabary, looks concerningly Japanese, but not too similar. It was made by first looking at some other scripts and picking out the more prominent shapes (most of the shapes are from katakana, hence the similar look), then assigning a shape to a different sound, then experimenting with different ways of combining the shapes to get something that looked coherent and presentable. I then made a font out of it as you can see in the chart below:

The font is done using opentype ligatures, so I can type in "pa" and get the whole symbol out as one thing which makes it incredibly easy to type. The hyphen-y character is the general nasal at the end of a syllable as mentioned under phonotactics. Word boundaries are represented with a little dot between the syllables. The font is monospaced so the characters will fit neatly to a grid.

The system is intended to be written in columns left-to-right, but I'll be doing it all in rows for the purposes of compactness, and the same thing is being done in the dictionary template.

[Pabappa]

I like the script. The glyphs look like they've been slowly simplified over time from ligatures of previously existing consonant plus vowel signs. This is *unlike* Japanese, though in Japanese the glyphs seem to have evolved towards a convergent shape anyway.

One minor question about the phonology though, and I'm not sure about this, but ...

I notice your script lacks glyps for /ji/ and /wu/. Is this because the language cannot have /ji/ and /wu/ phonetically? if so you may want to reconsider the contrast between /ti/ vs /tsi/ vs /tši/, /si/ vs /ši/, etc ... as I believe that Japanese lost its /ji/~/i/ contrast by the same process that merged down the contrast between /ti/ and /tši/. In other words, if your /i/ is phonetically [ji] in isolation, it is likely that it will also be [ji] when it follows a consonant, and that there cannot be a contrast between a consonant and its palatalized equivalent before a phonetic ["i"]. (I'd say the same would apply to /u/~/wu/ if your phonology had labialized consonants. Jp used to, but they merged into their plain counterparts. )

[mira]

I notice your script lacks glyps for /ji/ and /wu/. Is this because the language cannot have /ji/ and /wu/ phonetically? if so you may want to reconsider the contrast between /ti/ vs /tsi/ vs /tši/, /si/ vs /ši/, etc ... as I believe that Japanese lost its /ji/~/i/ contrast by the same process that merged down the contrast between /ti/ and /tši/. In other words, if your /i/ is phonetically [ji] in isolation, it is likely that it will also be [ji] when it follows a consonant, and that there cannot be a contrast between a consonant and its palatalized equivalent before a phonetic ["i"]. (I'd say the same would apply to /u/~/wu/ if your phonology had labialized consonants. Jp used to, but they merged into their plain counterparts. )

Yes, <wu uw ji ij> are all invalid as per the phonology post a bit further down. The language isn't supposed to be directly or indirectly based on Japanese, it's just that the writing system looks a fair bit like Katakana. The grammar and lexicon shouldn't end up being noticeably Japanese-y at all. /i/ is in isolation anyway.

[Pabappa]

Yes, <wu uw ji ij> are all invalid as per the phonology post a bit further down. The language isn't supposed to be directly or indirectly based on Japanese, it's just that the writing system looks a fair bit like Katakana. The grammar and lexicon shouldn't end up being noticeably Japanese-y at all. /i/ is ["i"] in isolation anyway.

OK thanks. You can use slashes for those sequences .... the angle brackets indicate the transcription, but not every language (even conlangs) uses a phonemic orthography.

On the idea of /i/~/ji/ being merged but not /ti~tsi~tši/.... my impression is that languages with vowel inventories of the type /a, ja, u, ju, i/ have the missing member because the /i/ behaves as if it were /ji/, including palatalizing all preceding consonants. You may not hear the /j/ because the glide is assimilatedi nto the vowel but it behaves just the same as any other /j/.

There seem to be plenty of Australian Aboriginal languages that merge /ji/ into /i/ without the palatalization of consonants, though it always seems to be spelled as if it were /ji/ without /i/ rather than /i/ without /ji/. I dont know what the names are actually pronounced like.

Did this language have a shift of /ji/ > /i/, and /wu/ > /u/, at some point in its history, or did those combinations never arise?

[Davush]

I think /i/ without /ji/ is very common and well attested, after all /j/ is generally just a non-syllabic /i/ so acoustically it's not an easy distinction to maintain.

Palatalization after /i j/ may or may not occur, it's not a given, and there's no reason to say it must happen in OTheB's language either. /ti tsi tʃi/ distinction isn't unheard of.

If a language has /i j/ but no /ji/, there's no reason to analyse all /i/ as actually underlying /ji/. Palatalization can occur because of plain /i/ as well (since /i j/ are more or less equivalent anyway, one just behaves more like a consonant).

@OTheB: Nice script, I see what you mean about it being slightly Japanese. Can I ask what program you use to make it?

[mira]

@OTheB: Nice script, I see what you mean about it being slightly Japanese. Can I ask what program you use to make it?

I made it with Inkscape and FontForge. The glyph editor in FF isn't very nice to use, but you can copy and paste stuff in directly from inkscape. I made each symbol by drawing the path for it, stroking the path with a really thick line, then converting the outline of that to a whole path. I then just copied it into FF, arranged all the glyphs I needed, wrote the ligature rules in a text editor, imported that opentype features file, and exported. It's not too hard once you figure it out, and I can definitely recommend using programmed ligatures to make your script much easier to type.

[mira]

Did this language have a shift of /ji/ > /i/, and /wu/ > /u/, at some point in its history, or did those combinations never arise?

Absolutely no idea. I'm not making the language with any history in mind. It's completely a priori, and it'll be more on the personal/engineered language side than the realistic side. Not allowing /ji/ and so on is just a design choice as the sound is very hard to pronounce distinctly from /i/ and I don't really like the sound to begin with, ergo it isn't in the language. I imagine stuff will start to be much more coherent as I work on more of the language.

[k1234567890y]

I somehow want to help with the lexicon...;-;