First detailed conlang- notes/suggestions?

[scorpryan]

[pre]This is my first real attempt at creating a conlang, as well as my first post here; be gentle (but by all means constructive). I will give some basic information below- let me know what you think, and what needs to be added/altered/removed or plain doesn't make sense. Oh, also-- any tips on typing IPA here? I mostly copied and pasted, and some of the symbols look awkward (the dent-alveolar 't' below is not lined up correctly with the diacritic...)?

Morphosyntactic alignment: ergative-accusative, split ergativity
Typology: agglutinative
Word order: free, SOV preferred

Primary stress is word-initial.
Secondary stress falls on every third syllable.


Phonology
[pre]
Orthograph IPA Notes
b /b/
p /p/ unaspirated
k /k/ unaspirated
g /g/
d /d/
t / t̪ / like in Finnish
f /f/
v /v/
h /ħ/ voiceless pharyngeal fricative; very breathy and forceful. always pronounced.
j /d͡ʒ/
m /m/
n /n/
ng /ɳ/ /ɳg/ medially, /ɳ/ finally
l /l/ alveolar, not dental
r /r/ trilled
s /s/
ş /ʃ/
z /z/
ç /θ/ as in English thin
č /t͡ʃ/
a /a/ as in Spanish mal
e /e/ as in Spanish que
i /i/ as in Spanish si
o /o/ as in Spanish queso
u /u/ as in Spanish subir
ı /ɯ/ see 'further notes'
ă /ə/
ô /ɔ/
w /w/
y /j/
' /ʔ/
[/pre]

Further notes
* Consonants are never aspirated.
* Consonants are almost never devoiced or sonorized, no matter what they come into contact with. There are only two exceptions; the perfective preverb 'g-' will become 'k-' before a verb root that begins with a 'g', and the habitual preverb 'č-'
becomes 'j-' if the verb root begins with a 'č' already. Otherwise, consonant voicing is never changed to aid pronunciation. A final 'b' is always a /b/, for example.
* /r/ is heavily trilled, like in Spanish 'rr' or anywhere in Finnish.

* a macron ◌̄ means that primary stress shifts to that syllable and the vowel in question is lengthened slightly.

* ı is basically the same as in Turkish (close back unrounded). That being said, it is sometimes realized as /ɪ/ in certain dialects, especially in rapid speech.


Noun Morphology

Plural marker: in-

Case
Absolutive-: -Ø
Ergative: -ktu
Accusative: -t, ıt
Lative/Dative: -mer
Ablative: -rol
Benefactive: -sta
Locative: -el
Genitive: -i
Caritive: -van
Abessive: -şa, -aş
Instructional: -r, ır
Genitive II: ◌̄ri

Possessive Suffixes
1st: -ni
2nd: -ya
3rd: -sa
1st pl: -min
2nd pl: -yain
3rd pl: -sain

Cltics
-'kô urgency
-'kis diminutive
-'lă question
-'lu negative question "isn't it? is it"
-'ta exclamatory

Allowable noun stem endings: a, e, i, o, u, ô, ă, k, t, l, m, z, d, f, ç, s

Particles (that I know of so far):
ş vocative, honorific
h emphasis
fa' ein used in negative commands
fa', fu negative

* adpositions are used in many cases.
* the second genitive has the meaning of "of the ___(s)". Whereas the genitive is used strictly for possession, Genitive II functions more like an adjective, and is used to describe nationality, origin, affiliation, or even likeness, and has a more a "poetic" register. The dotted circle with macron represents the final vowel of the noun, which will receive the primary stress. If the noun stem has no final vowel, one will be added. so ilya (ocean) > ilyāri (of the oceans)

Verbal paradigms

Transitive series

Code: Select all

Preverb       +       Root      +      Subj      +      Obj      +      Tense                 +      Mood
Ø imperfective                       -ık                 -măt               -Ø (nonfuture)           -s  conditional
g perfective                          -eş                 -şăt                -i future                    -kui  past subjunctive
č habitual                            -wa                -Ø                                                  -va  subjunctive
ā passive voice                      -om               -tkim                                              -if   optative
                                          -ya                -yăt
                                          -wan             -Ø

The subject and object affixes above are listed in order (1st, 2nd, 3rd, 1st pl...) and obviously not all can be combined with each other. Also, after -eş, -tma is used instead of -măt, so the YOU + ME combo is -eştma.
The 3rd person object is intentionally null. In the transitive conjugation, 3rd person objects are implied. So lahwa means "he/she/it reads it."
Note that this is a template of all existing affixes for the transitive series conjugations, and not all of them are possible in all combinations. Instead, refer to list of tenses.

The optative mood is not to be confused with the imperative, which serve different purposes.

When called for, the indirect object affixes may REPLACE the direct object affixes used above, but they are never used together in one word. Note that the I.O endings do not include number, only person (number will be implied by context):
-mer 1st indirect object
-yer 2nd indirect object
-(z)er 3rd indirect object

Officially, there is no past tense. Instead, the perfective aspect is used to convey actions that happened in the past (in absence of future-tense marker). Technically, there is no marking for present tense either; the subject endings are merely added onto the imperfective stem, which is "present tense" by default until the future marker -i is added. *****I am having trouble and would be open to critique here... I am fine with using the Present Perfect to convey the past tense, etc... but what about the past imperfect? It seems like there is a huge gap in meaning if I do not have one. My idea (shown below) was to have "was ___ing/used to ___" be represented by using the habitual aspect (which is present tense by default, "people sing") plus adverbs, etc. that indicate a past-tense time frame. So someone could use the habitual non-future form of the verb with other words that indicate the action occurred at some point in the past. Is that plausible...?****


Present tense (imperfective nonfuture) root+subj+obj tulwa "he's singing"
Present perfect (perfective nonfuture) g+root+subj+obj gtulwa "he sang/he has sung"
Future root+subj+(obj)+i tulwa-i "he will sing"
Habitual č+root+subj+obj čtulwa "he sings (in general)" OR "he was singing" depending on context
Future Perfect g+root+subj+(obj)+i gtulwa-i "he will have sung" (possible but rare)
Conditional root+subj+(obj)+(e)s tulwas "he would sing" gtulwas "he would have sung"
Subjunctive, two arguments root+subj+(obj)+va tulwava "that he sings, that he would sing"
Past Subjunctive root+subj+(obj)+kui gfizwayer tuleşkui "he told you to sing"
Optative root+subj+(obj)+if tulwaif "may you sing"
Passive ā+root+subj+ă ātulwă "it is sung"
Aorist root+ă tulwă "sing, sung"
1st Infinitive root+i tuli "I like to sing"
2nd Infinitive root+ava tulava "I want to sing" (used in subjunctive clauses where there is only one subject)
3rd Infinitive root+asse tulasse "He came in singing"
4th Infinitive root+uman tuluman "You must come so that you can/in order to sing."

Intransitive series

I am not going to go into as much detail or put every template of every tense here (they are mostly the same), but here are the intransitive subject suffixes:

1st -eb
2nd -iş
3rd -u
1pl -im
2pl -ya
3pl -waz


Here are some random examples from the lexicon so far (which is at about 250-300 words):


kară- to love
fată- to want
yalimă- forever, always
julma- treasure, beloved
hsed- far
ihwenna- almost, about to, nearly
ilya- ocean
lotos- sleepy
mekyad- clean, pure
judyar- rare
nyelva- sweet
ruz- mouth
melek- servant
yotemă- bed
nennă- to come
dragavi- barbaric, foolish, unrefined
hsani- wind
mertă- to thank
lalu- song
pagwi- small, short
yară- to see, watch
maharet- ancient, venerable
silimen- together
julă- to keep
tefit- night
odwa- judge
tengă- to do
orya- story, tale
kizme- pain, longing
evek- widow
orek- storyteller, poet
dyară- to travel
ôkya- fruit
jecwed- many, several
hajă- to save, protect
hajik- (n.) guard
sadă- to heal
herek- woman
nimo- spouse
jotă- to bring, carry
una- sibling
ikisin- hence, therefore, thus
umă- to eat
hwen- fast
twelıç- fair, gentle, holy
kalolmă- teacher
geç- soldier
şemit- land, country

I feel like I should stop now. Please share your thoughts.[/pre]


***I am sorry for the screwed up formatting... I really tried. I have no idea how to create tables or to make the [pre] preformatting tag work... if someone can help me out I will go back and make this more read-able. Sorry!

[spanick]

Cool stuff! You're very thorough. I'm sure others will have plenty to say but I just have a couple comments.

First, I'd like to know more about the stress system. Normally, when I think of predictable stress systems, I think of feet composed of to syllables i.e. (σσ) = a foot and stress is either iambic (σσ́) or troachic (σ́σ) but I can't (off the top of my head) think of a reason secondary stress would be every third syllable. Other things to think about with stress: What is the directionality of the stress? Is there extrametricality? Are there degenerate feet? Are unfooted syllables allowed? Are feet quantity sensitive?

Second, for <ng> it looks like you used the IPA symbol for the palatal nasal/ɳ/ instead of the velar nasal/ŋ/.

EDIT: if you did intend the palatal nasal, then it seems an out of place phoneme and the combination /ɳg/ particularly odd.

[Chengjiang]

ng /ɳ/ /ɳg/ medially, /ɳ/ finally

Is this sound intended to be a velar nasal as in sing? If so, you've got the wrong symbol. <ŋ> is used for the velar nasal. <ɳ> represents a retroflex nasal, a sound made with the tip of the tongue against the palate found in Indo-Aryan, Dravidian, and Pama-Nyungan languages.

Second, for <ng> it looks like you used the IPA symbol for the palatal nasal/ɳ/ instead of the velar nasal/ŋ/.

That's the retroflex nasal. The palatal nasal is <ɲ>.

* a macron ◌̄ means that primary stress shifts to that syllable and the vowel in question is lengthened slightly.

I'd recommend using an acute accent to indicate stress rather than a macron. It has a tradition of use to mark stress in many languages while the macron doesn't, and it also comes with ready-made combinations with <ă> and <ô> thanks to Vietnamese.

Are you planning to have the stress diacritic (whatever it is) not replace the dot on <i>, to keep it distinct from stressed <ı>? I think there are versions of <i> with those diacritics that retain the dot, but I'm not sure.

Also, you haven't defined any phonotactics for this language, so we don't have a clear idea of what types of syllable are allowed. I mean, I don't see any clear cases of syllables ending in two or more consonants or beginning with three or more, but that's just based on the lexicon sample you gave.

Plural marker: in-

Does this assimilate at all to a following consonant? Are there allomorphs such as /im/?

Case
Absolutive-: -Ø
Ergative: -ktu
Accusative: -t, ıt
Lative/Dative: -mer
Ablative: -rol
Benefactive: -sta
Locative: -el
Genitive: -i
Caritive: -van
Abessive: -şa, -aş
Instructional: -r, ır
Genitive II: ◌̄ri

Possessive Suffixes
1st: -ni
2nd: -ya
3rd: -sa
1st pl: -min
2nd pl: -yain
3rd pl: -sain

Cltics
-'kô urgency
-'kis diminutive
-'lă question
-'lu negative question "isn't it? is it"
-'ta exclamatory

What order do these suffixes follow if more than one is found on a word? Also, where are the clitics found? Clause-finally?

I'm wondering about allomorphs for a lot of these, especially since you don't describe phonotactics for this language. Most languages with grammatical affixes have alternate forms for these affixes depending on the phonetic environment, such as <-es> for English <-s> after sibilants.

Particles (that I know of so far):
ş vocative, honorific
h emphasis
fa' ein used in negative commands
fa', fu negative

Where do these particles go? You need to tell us their positions relative to other words.

* adpositions are used in many cases.

That's awfully vague. When and how are adpositions used?

The optative mood is not to be confused with the imperative, which serve different purposes.

These purposes being what? I can guess based on their names, but that would be a guess and nothing more. If you're going to say that two things are used differently, you need to say what the difference is.

When called for, the indirect object affixes may REPLACE the direct object affixes used above, but they are never used together in one word.

How are ditransitive verbs conjugated, then? How does this system handle situations like "I gave it to you"?

I am having trouble and would be open to critique here... I am fine with using the Present Perfect to convey the past tense, etc... but what about the past imperfect? It seems like there is a huge gap in meaning if I do not have one. My idea (shown below) was to have "was ___ing/used to ___" be represented by using the habitual aspect (which is present tense by default, "people sing") plus adverbs, etc. that indicate a past-tense time frame. So someone could use the habitual non-future form of the verb with other words that indicate the action occurred at some point in the past. Is that plausible...?

You could do that. You could also just use imperfective marking, leaving its past versus present tense status up to context.

jecwed- many, several

You don't define a value for simple <c>. How is this word pronounced?

EDIT: You say that this language has split ergativity. What conditions the split? It's not actually clear from this description.

[spanick]

Second, for <ng> it looks like you used the IPA symbol for the palatal nasal/ɳ/ instead of the velar nasal/ŋ/.

That's the retroflex nasal. The palatal nasal is <ɲ>.

Whoops, yeah that's what I meant. My bad. Thanks!

[scorpryan]

stress: I was meaning to imitate the stress pattern of Hungarian, so what I meant to say was not "every third syllable" but every ODD numbered syllable.
Does that make more sense?

Yes, I meant the velar nasal. I will go back and edit it.

I really prefer the macron instead of the acute accent, just visually speaking, because the acute reminds me so much of Romance languages, and I wanted
to stay far away from imitating them or ending up with a language that looks Tolkien-eque, but you make a logical point. I suppose I should change that.

The stress/lengthening diacritic will never be necessary on ı. Whenever ı occurs, it is always unstressed, unless it occurs word-initially. I don't intend for there to ever be an occasion where
stress is counterintuitively switched to a syllable containing ı (which is what the stress diacritic is for;
to indicate unusual stress patterns contrary to the word-initial rule).

I am still working on the phonotactics of this language. I do have a running list of phonotactic constraints somewhere that I need to dig up. As far as allowable
syllable types, I need to sit down and work it out. I am still learning about linguistics as I go, so i am not sure how I should mark it... I can say that the rest of my lexicon does not exhibit any obvious departures from the sample I posted, so I guess I could say this language has:

CV
CCV
VC

I don't think coda clusters will be allowed, but certain 2-consonant onset clusters are allowed (<st>, <kt>, <hl>, <hs>, <hw>, <hr>, <kn> are the only ones I can think of right now, I need to make a complete list).

In- does not assimilate. It is always in-.


Noun declension looks like this:

Plural marker-Stem-Case-Possessive-Clitic.

It always follows this order, there is no metathesis. So, yes, clitics are found clause-finally.


A few things to note: if a noun ends in ô or ă it is most likely going to be 3 syllables. that final vowel is deleted/replaced in most cases, making the word function as consonant-final throughout the case declension. a noun never ends in a diphthong.

Ergative
the ergative ending <-ktu> causes elision of stem-final consonant
final consonants. geret-gerektu

Accusative
the accusative ending <-t> will be <-ıt> after
stem final consonants (coda clusters not allowed).

Lative/Dative
no change after <a e i o u k t l z d f ç s>, just
add <-mer>.

<ô ă> are deleted. if the consonant before them is
<m>, it is also deleted.

<m> is deleted. it is not doubled.

Ablative
<a e i o u k t l m z d f ç s> just add <-rol>
<ô ă> are deleted then <-rol> is added.


Benefactive
the benefactive is <sta> after <a e i o u k t l
f>. pile-pilesta

<m z d ç s> are deleted when the benefactive <sta>
is added. navid-navista

<ă ô> are deleted, and then the remaining consonant
follows the above rules. kalolmă-kalolsta.

Locative
<a> -ael
<e> -el
<i> -iel
<o> -oel
<u> -uel
<ô ă> delete and replace with -el
<k t l m z d f ç s> simply add -el.

Genitive
simply add <-i> to the stem. if the word ends in <i> already.... I haven't decided? I think it should simply remain the same, and thus -i nouns are
indistinguishable from their genitive constructions... but perhaps I could make a stress shift to the final <i> using the diacritic, or make it doubled (ii) ... the doubling seems unrealistic though, since
doubling doesn't occur anywhere else in the language.

Caritive
simply add <-van>... ô, ă deleted.

Abessive
vowel final stems take <-şa>.
consonant final stems take <-aş>.
ô, ă deleted.

Instructional
vowel final stems take <-r>
consonant final stems take <-ır>


Genitive II
vowel-final stems take a stress diacritic on the
final vowel, then add <-ri>.

consonant-final stems take <-āri>

ô, ă deleted.


...I think that pretty much covers any possible interaction. Tell me if you catch something. The possessive suffixes begin with nasals/glides/sibilants
that make them easy to attach/pronounce in combination, and the clitics are always preceded immediately by a glottal stop. I think these are the only necessary
allomorphs. There is no vowel harmony.

Location of particles: the vocative ş and emphatic h can go in any position. they are free-floating, but most often occur at the head of a clause, provided
they don't run into a duplicate letter that might hide them (like another h at the beginning of a word).

fă' ein comes right before the verb in its 2nd person imperfect non-future form to indicate the negative imperative. Actually I haven't decided but I think
that fa' ein will just be the general negative marker that goes before a verb of any kind, and the imperative will simply be identical to regular ol' past tense.

fă' goes before a noun or adjective, as in "not thirsty."

fu is "no" as in "no, I'm not thirsty."

Optative and imperative: imperative is a command made directly to a 2nd person using (presumably?) the same form as the imperfective nonfuture (present tense),
whereas the optative allows the speaker to politely express a wish without urging for its completion, and this wish can apply to any person. Compare "die!" with
"may you die."

Ok, on the ditransitive verbs issue, I am still working out the bugs, but like I said above the subject suffixes carry intransitive or transitive meanings. So if you
use the transitive suffixes, a 3rd person direct object is ALWAYS implied. So anytime you have a sentence like "I gave it to you", where the direct object is a 3rd person, you will not require two endings.

But what about 3-way conjugations where the object is not a 3rd person? "Take us from him"? To solve the ambiguity I could either say that I DO allow both endings to
occur, or I could introduce a series of detached pronouns. Still working on them, but I do have a few direct and indirect object pronouns in mind, which I had
intended to be used strictly for clarification or in instances where you would simply answer a question like "to whom?" with "to him". The accusative ones could be something like emet, eşet, et or eret (?), etc. and the indirect could be emer, eşer, etc.

I guess the logical thing would be to do the first, but the pronunciation of such words kind of... annoys me. I find it unpleasing aesthetically. Then again, it also annoys me to sully the "agglutinative"-ness
of my agglutinative language by introduce separate pronouns I already have a system that allows for pronouns to be attached directly to the verb. What do you suggest?

Oops. I meant <jekwed>, merely an error caused by the fact that the original orthography had both <c> and <k>.

Oh, and I'm not going to lie- on the adpositions question, I have no answer for you yet (ha, ha). I only have about 2 or 3 adpositions in the lexicon so far.
I will say that some may be prepositions and others postpositions, and indicate semantic value appropriate to that case, for example the locative case has a crap-ton of adpositions to indicate meanings like "under" or "next to" ... I'm still working it out.

P.S: again, sorry for the awkward spacing. I typed my response in Notepad first, oops. Last time- I swear.

I also just realized I forgot to post allowable diphthongs. I have a list laying around, I will go back and add that in a bit.

P.P.S: All this time, I've been writing fa' ein. Should be fă' ein.

[spanick]

stress: I was meaning to imitate the stress pattern of Hungarian, so what I meant to say was not "every third syllable" but every ODD numbered syllable.
Does that make more sense?

Definitely, Hungarian is basically trochaic. When you encounter a word with an odd number of syllables, does the final, unfooted syllable also take a secondary stress? For instance, your word <tulava>

[scorpryan]

stress: I was meaning to imitate the stress pattern of Hungarian, so what I meant to say was not "every third syllable" but every ODD numbered syllable.
Does that make more sense?

Definitely, Hungarian is basically trochaic. When you encounter a word with an odd number of syllables, does the final, unfooted syllable also take a secondary stress? For instance, your word <tulava>

No, I should have mentioned that as well. Three syllable words will only have initial stress. For example, <hazeşmer> <(you) help me> / 'haz eʃ mer / but hazeşmeri /'haz eʃ 'mer i /

Which I think is the same as Hungarian? Not sure.

EDIT: just realized my example here does not represent your question, because it's not unfooted? But still, the same applies. All 3 syllable words have only primary stress.

[spanick]

Your example works. But I could ask the same for any word with an odd number of syllables. Does a 5 syllable word have two secondary stresses (σ́σ)(σ̀σ)(σ̀) or just one
(σ́σ)(σ̀σ)(σ)? Do you see what I mean?

Just a little clarification, a foot is made of two syllables (σσ) so an unfooted syllable is a syllable which is not part of a foot (basically, one, solitary syllable). A degenerate foot is when that single syllable is able to take stress.

[spanick]

Sorry if I'm going on too much about stress, I know it's relatively minor in the over all picture of your conlang. Obviously, I just really like the topic! haha

[scorpryan]

Yes, I see what you mean... I think. So, I will say that the same can be said for any word with an odd number of syllables. That final, odd-numbered syllable will not be stressed.

Code: Select all

<kalolmvanmin'ta>               KAL-olm-VAN-min-'ta. The final clitic <'ta> is not stressed. 
teacher-CAR-1p.POSS-(emphasis clitic)

...Wait a minute. I don't know if that makes sense, because technically there is a glottal stop before <ta>... can a glottal stop carry stress?

Also should note that the plural prefix in- does not take the primary stress. The stress is still on the root. So <in-melek> <the servants> is still in-MEL-ek, not IN-mel-ek.

[scorpryan]

Not at all, you;'re bringing up valid points that are helping me out. As this is my first time doing this, I am clearly leaving a lot out... sometimes intentionally, because certain aspects make my brain hurt a tiny bit more than excite the linguist in me.

Also, while this is an agglutinative language, the stress is actually really important, because stress can be phonemic in this language (hence the use of the macron/acute accent)

[spanick]

. I don't know if that makes sense, because technically there is a glottal stop before <ta>... can a glottal stop carry stress?

It makes perfect sense. No, a glottal stop doesn't take stress. Consonants can take stress in some contexts but usually more sonorous ones like /m/ /n/ or /l/. Given your given phonotactical constraints, you'd have to say that in <kalolmvanmin'ta> the glottal stop is part of a cluster, in this case <-'ta> since you said there are no coda clusters. But /Ɂt/ doesn't seem to be a cluster you want (judging by your list of allowed clusters) so you'll need to account for that somehow. Maybe elide the glottal stop? Epenthesize a vowel in there to break up the cluster? There are many possibilities.

[Chengjiang]

Also, while this is an agglutinative language, the stress is actually really important, because stress can be phonemic in this language (hence the use of the macron/acute accent)

No reason why it shouldn't be. Agglutinative languages aren't less likely to have contrastive stress than other types of language.

[scorpryan]

. I don't know if that makes sense, because technically there is a glottal stop before <ta>... can a glottal stop carry stress?

It makes perfect sense. No, a glottal stop doesn't take stress. Consonants can take stress in some contexts but usually more sonorous ones like /m/ /n/ or /l/. Given your given phonotactical constraints, you'd have to say that in <kalolmvanmin'ta> the glottal stop is part of a cluster, in this case <-'ta> since you said there are no coda clusters. But /Ɂt/ doesn't seem to be a cluster you want (judging by your list of allowed clusters) so you'll need to account for that somehow. Maybe elide the glottal stop? Epenthesize a vowel in there to break up the cluster? There are many possibilities.

I want the glottal stop before all clitics to remain as it is, no change. So should I just create a list of allowable onset clusters that includes all possible combinations that begin with a glottal stop?

[scorpryan]

Also, while this is an agglutinative language, the stress is actually really important, because stress can be phonemic in this language (hence the use of the macron/acute accent)

No reason why it shouldn't be. Agglutinative languages aren't less likely to have contrastive stress than other types of language.

Oh, I thought they were. But my experience is limited to a select few, and I just assumed; for example, I can't think of any example of (I guess you'd call them minimal pairs?) two words in Finnish that change meaning depending on whether or not you stress one syllable or the other. The same goes for my *limited* knowledge of Hungarian and Turkish.

[spanick]

I want the glottal stop before all clitics to remain as it is, no change. So should I just create a list of allowable onset clusters that includes all possible combinations that begin with a glottal stop?

You could do that, it's probably better just to define the phonotactics and syllable structure a bit more. For now it suffices to say that /Ɂt/ is an allowable onset cluster. Maybe /ɁC/? I know you're trying to avoid lots of phonotactical rules when morphemes interact, but this is exactly the type of situation when they can be super helpful.

[scorpryan]

How is this for a start on phonotactics? I removed /kn/ as an allowed onset, by the way.

Phonotactics
1.) Nucleus must be a vowel
2.) No complex codas (no final consonant clusters)
3.) Complex onsets restricted to certain clusters, but due to suffixation, any two-consonant cluster may occur medially
4.) Triconsonant clusters are restricted except for the following, which occur as a result of suffixation: /tst/ /kst/ /lst/ /mst/ /nst/ /fst/

Allowed onset clusters:
/hl/ /hr/ /hs/ /hy/ /ks/ /kt/ /st/
/g/ + consonant
/č/ + consonant
/j/ + consonant
/ʔ/ + consonant (only in clitics)

Diphthongs: /ae/ /ai/ /ea/ /ei/ /eo/ /ia/ /ie/ /io/ /iu/ /oa/ /oe/ /oi/ /ua/ /ue/ /ui/
(ao, au, eu, ou, uo NOT allowed)

Syllable structure: ?
CV
CCV
CVC
V
VC
VV

*I have no idea how to determine the syllable structure. I guess it is (C) (C) V (V) (C)? I feel like I have way too many syllable types.

[spanick]

Good stuff. Yeah, (C)(C)V(V)(C) is probably the best way to put it.

Remember too that two consonants medially CVCCV wouldn't be considered a complex coda because that could be thought of as either CV.CCV or CVC.CV (where the period represents a syllable boundary). Same goes for three consonants in a row: CVCCCV would necessarily be broken up as CVC.CCV according to the syllable structure given.

[scorpryan]

Just realized though, that many verb roots end in two consonants (e.g, <mert->)... I could say that because the verb always has to be found with some form, at least the aorist with the schwa sound at the end, that these don't count as words per se and it would be ok... the problem arises when the 3rd person sing or plural or 2nd person plural endings (-wa, -wan, -ya) are added to them, you get a triconsonant cluster <mertwa> which I want to remain intact... so I guess I should also add that triconsonant clusters can take the form of C+C+semivowel? Or, I could say the syllables here are divided as mer-twa... but that is annoying to me because the <t> is a phonemic quality of the verb root <mert-> and the dictionary/aorist form mertă (to thank).

Confuzzled

[spanick]

Or, I could say the syllables here are divided as mer-twa... but that is annoying to me because the <t> is a phonemic quality of the verb root <mert-> and the dictionary/aorist form mertă (to thank).

This is a non-issue. how the syllables are divided is not related to a word's meaning. /mer.twa/ is merely the way it's pronounced/divided up into useful units for talking about.

[Chengjiang]

Allowed onset clusters:
/hl/ /hr/ /hs/ /hy/ /ks/ /kt/ /st/
/g/ + consonant
/č/ + consonant
/j/ + consonant
/ʔ/ + consonant (only in clitics)

I see some other onset clusters in your lexicon sample, e.g. /dr/ and /tw/.

Also, /slashes/ are normally used for a phonemic transcription in IPA, not for an orthographic transcription.

[Pogostick Man]

I really like that you have a morpheme for urgency.

d /d/
t / t̪ / like in Finnish

Is /d/ also dental?

[mèþru]

Slashes for phonemes, brackets for phones and less-than/greater-than for transcriptions.

[scorpryan]

I really like that you have a morpheme for urgency.

d /d/
t / t̪ / like in Finnish

Is /d/ also dental?

Thank you! I had a few more unconventional clitics for random purposes but I threw them out based on aesthetics. Will probably make them free-floating particles.

Oops, yeah. That means I need to put the same diacritic on it as well, right?

The only contrast between them is voicing, they are both dental stops.


mèþru -- oops. got it.

[Vijay]

Oops, yeah. That means I need to put the same diacritic on it as well, right?

Yes.