Classical Merthic (Attempt at Semi-Speedlanging)

[cybrxkhan]

Inspired by roninboddhisatva’s speedlanging, I’m taking a break from Aidisese, and settled on a semi-speedlang project, Classical Merthic. I’ll try to post a bit of something new about it every few days and so, and hopefully within a couple of weeks I’ll have a feasible conlang. Any comments, advice, thoughts, and suggestions are welcome!

Merthic is supposedly to superficially sound like Irish/Gaelic, but its morphology is a bit similar to the Semitic languages (it will be triconsonantal), and syntactically it is similar to the East Asian languages (i.e. Chinese and Japanese). It’ll be SVO with a very small amount of inflections.


HISTORY AND OVERVIEW

Classical Merthic, or Merthic, is the ancestral language of the languages of the entire Merthic family, dominating parts of the two largest continents of Arteran, Kanis and Itholia. Merthic, with its unique script, was the lingua franca of most of western Kanis and northwestern Itholia for more than a thousand years, and continued to be used in formal religious and administrative situations for even longer, even in nations whose peoples had no genetic or linguistic connection to the original speakers of Merthic.


PHONOLOGY

Below is the phonology of Merthic; it is extremely similar to Irish, with a few differences.

Note: Like in Irish, velarized (broad) and palatialized (slender) consonants contrast with each other. Most of the velarized/broad consonants have a palatialized/slender equivalent, and vice versa. I might make this contrast something grammatical, but I’m not sure at this point.

Consonants:
Stops: /pʲ pˠ bʲ bˠ tʲ tˠ dʲ dˠ c ɟ k g qʲ q/ <p ph b bh t th d dh c ch g gh x xh>
Fricatives: /fʲ fˠ vˠ s ʃ ɣ h/ <f fh v vh s sh gc h>
Nasals: /mʲ mˠ nʲ nˠ/ <m mh n nn>
Approximants: /w j ɾʲ ɾˠ lj lˠ/ <w cg r rr l ll>

Note: /j/ is the ‘slender’ equivalent of ‘broad’ /ɣ/; /w/ is the ‘broad’ equivalent of ‘slender’ /vj/ and vice-versa.

Vowels:
/a a: e e: i i: o o: u u:/ <a á e é i í o ó u ú>

Diphthongs:
/ea io iu ia ua oa/ <ea io iu ia ua oa>

Syllable-Initial Clusters:
Non-Approximant + Approximant
/s ʃ/ + /mʲ mˠ nʲ nˠ/

Syllable-Final Clusters:
Approximant + Non-Approximant
/mʲ mˠ nʲ nˠ/ + /s ʃ/

Note that every cluster has an ‘opposite’ – i.e., every initial cluster has a final cluster equivalent, and vice versa. They are considered to be the same consonant in the sense that they can be interchangeable depending on their position in a syllable (i.e., if /sʲmʲ/, an initial cluster, becomes a final cluster for whatever morphological or grammatical reason, then it changes to /mʲsʲ/).

Additionally, clusters must agree in broadness and slenderness. Thus, /fˠɾˠ/ is not a possible cluster.


Phonotactics:

Syllable structure is (C)V(C), where C is any consonant or consonant cluster and V is any vowel or diphthong.

Stress falls on the first syllable of a word, except if that first syllable does not contain an initial consonant; in that case, the stress then falls on the next syllable.



SAMPLE SENTENCES

These are obviously lacking some of the grammatical features that I'll make later on, but they're good to get a feel of Merthic:

Willem otuimgce odhúhath clinnad.
[’wi.lˠemʲ o.’tʲuimʲ.ɣe o.’dˠu:.hatˠ ’kʲlʲi.nˠadʲ]
ruler warring contemplate love.
[The] ruler who wars contemplates about love.

Otúmagc ohútlar.
[o.’tʲu:.mʲaɣ o.’hu:.tʲlʲaɾʲ]
war be.extremely.evil
War is [an] unforgivable evil.

Dhiheth uhulmta venac himlat.
[’dˠi.hetˠ ’u.hulʲmʲ.tʲa ’vʲe.nʲakʲ ’hi.mʲlʲatʲ]
ruler weakly.attested invent writing
[The] weakly attested philosopher invented writing.



FEATURES

Here are some basic ideas about the grammar of Merthic, which I'll work on the next few rounds. Everything is subject to change, of course:

- SVO, head-first, and prepositional
- Tri-consonantal morphology, akin to the Semitic languages
- Nominative-Accusative
- Zero Copula
- Small pool of adpositions
- Measure Words
- Serial Verb Construction
- Heavy use of compounding to make new and different words
- Minimal affixation in verbs and nouns
- Large number of grammatical voices
- Honorific speech that is grammaticalized, like in Japanese
- Many different honorifics depending on social status and gender
- Heavy use of particles to indicate aspect and mood
- Evidentiality system with 2 or 3 differentiations, which will be optional



Next round I'll post about the triconsonantal morphology, which will be a bit fun, since it's easy to slide in random references with this kind of morphology.

[Mr. Z]

I can't say much about the phonology, because I simply don't know much about phonology. Though it seems you have many, many consonants that are foreign to me.

The sample sentences look nice; you certainly don't use too many words to form sentences. I like the features of the language. They look good and useful.

Good luck!

[Ulrike Meinhof]

The combination Irish/Chinese/Semitic sounds cool. Looking forward to seeing more of it.

[cybrxkhan]

Thank you for the kind words. Anyhow, here's the triconsonantal morphology:

MORPHOLOGY

Alright, so this round I’ll be going into the triconsonantal morphology a bit. Most of you probably know about this concept like the back of your hand, so I don’t think there’s any need to explain how the basics of it works in excruciating detail.

Anyhow, basic Merthic morphology is similar to that of Arabic, as most of its roots and basic words are triconsonantal, meaning that they are determined by a unique set of three consonants, which, depending on the pattern they follow with vowels, can mean different things. Below are a few of the possible patterns (periods are used to separate syllables):

Ca.CáC = transitive verb
Ci.CeC = person
CeaC.Cá = adjective describing a state
o.CuiC.Ce = strong adjective describing a state
aC.CioC = verb describing a state or intransitive verb


Merthic uses these patterns from the following tri-consonantal roots (as you can tell, the roots are all references to some random things, some more obvious than others):

Cl-Nn-D = love
Clannád = to love
Clinned = lover
Cleanndá = affectionate
Ocluinnde = loving
Alcnniod = to be in love*

H-Tl-R = evil
Hatlár = to commit evil towards, to wrong
Hitler = villain
Healtrá = evil*
Ohuitlre = evil beyond redemption
Ahtlior = to be evil

W-Ll-M = power; rule; authority
Wallám = to dominate, to control, to rule over
Willem = ruler; a superior; leader
Weallmá = powerful; having power/authority
Owuillme = extremely powerful; dominating; dominant
Awlliom = to be powerful

Dh-H-Th = thought; mind
Dhaháth = to think about
Dhiheth = thinker; scholar; philosopher
Dheahthá = intelligent; pensive
Odhuihthe = contemplative; meditative
Adhhioth = to think

H-ML-T = write, record
Hamlát = to write/record about
Himlet = writer; author; scribe
Healmtá = written; attested; recorded*
Ohuilmte = well-attested*
Ahmliot = to write

Br-T-N = human, person
Bratán = to give birth to
Briten = person; human
Breatná = human (adj.)
Obruitne = N/A
Arbtion = to be human*

*Notice that the consonants in the cluster switch because of a change from a syllable initial to a syllable final position.


Sample Sentences (these, again, are not including other grammatical features I’ll put in later on):

Willem ocluinnde bratán briten dheahthá.
ruler loving give.birth human intelligent.
[The] loving ruler gave birth [to an] intelligent human.

Himlet odhuihthe dhaháth hamlát hitler.
scribe contemplative think write villain
[The] contemplative scribe thinks about and writes about [a] villain.

Eire cleanndá clannád wallám múir dhiheth.
3SG.FEM affectionate love dominate many scholar.
She who is affectionate loves to dominate/control [the] scholar a lot.

Hitler ohuiltre cann arbtion.
Villain great.evil NEG be.human.
[The] villain who is evil beyond redemption is not human.


Besides that I’ve made some minor adjustments to the phonology. I hope to get the script up next round, but it’s being an asshole because I’m trying to make a logographic system that features the triconsonantal morphology.

[Nortaneous]

I was going to say something, but then I read Hitler ohuiltre cann arbtion. as Hitler o'hitler can abortion. so I'm going to bed and I'll do it in the morning.

Also, in one of the example sentences in your first post, you have "uhulmta". How is that /lm/ cluster allowed in your phonotactics? I can't see a way to get CCC clusters where the middle C isn't one of /s ʃ/.

[cybrxkhan]

I was going to say something, but then I read Hitler ohuiltre cann arbtion. as Hitler o'hitler can abortion. so I'm going to bed and I'll do it in the morning.

Also, in one of the example sentences in your first post, you have "uhulmta". How is that /lm/ cluster allowed in your phonotactics? I can't see a way to get CCC clusters where the middle C isn't one of /s ʃ/.

The syllables would be

u . hulm . ta

So I don't see a problem, since it isn't a CCC cluster per se... Regardless, CC clusters are fine in both the syllable onset and the syllable coda.

However I did have some issues with definitions of phonotactics and clusters previously... so would my above explanation be inadequate or something?

[finlay]

Hitler is your word for villain?

lol

also this bit is why nortaneous is saying that:

Syllable-Initial Clusters:
Non-Approximant + Approximant
/s ʃ/ + /mj mˠ nj nˠ/

Syllable-Final Clusters:
Approximant + Non-Approximant
/mj mˠ nj nˠ/ + /s ʃ/

You haven't allowed for /lm/ in the coda. Edit: oh, you wrote "approximant + non-approximant"... fair enough i guess. I'd deregularise it a bit by arbitrarily disallowing certain clusters - I would especially disallow clusters of a broad and slender consonant together. Make a full list of what's allowed, essentially.

Also palatalisation (Irish slender consonants) is indicated by ʲ in the IPA, not j. And are you trying to make it look anything like Irish? because you haven't copied the system of knowing whether it's slender or broad depending on what vowel it's next to.

[Bristel]

Also palatalisation (Irish slender consonants) is indicated by ʲ in the IPA, not j. And are you trying to make it look anything like Irish? because you haven't copied the system of knowing whether it's slender or broad depending on what vowel it's next to.

Yes, I think that if you integrate this feature, it would scream "IRISH" more than it does now... although the words look very Irish anyway.

Clannad looks like a reference to the anime?

[cybrxkhan]

Hitler is your word for villain?

lol

Some of the references are not so subtle.

also this bit is why nortaneous is saying that:

Syllable-Initial Clusters:
Non-Approximant + Approximant
/s ʃ/ + /mj mˠ nj nˠ/

Syllable-Final Clusters:
Approximant + Non-Approximant
/mj mˠ nj nˠ/ + /s ʃ/

You haven't allowed for /lm/ in the coda. Edit: oh, you wrote "approximant + non-approximant"... fair enough i guess. I'd deregularise it a bit by arbitrarily disallowing certain clusters - I would especially disallow clusters of a broad and slender consonant together. Make a full list of what's allowed, essentially.

Ah, broad + slender... Yeah, I was thinking about only allowing clusters to be of the same type, and originally had that in my notes, but I took it out as I was unsure about that. I'll take that advice.

Also palatalisation (Irish slender consonants) is indicated by ʲ in the IPA, not j. And are you trying to make it look anything like Irish? because you haven't copied the system of knowing whether it's slender or broad depending on what vowel it's next to.

Actually I am aware of how to indicate palatalization in IPA, but I think what happened was when I copied this from my word document onto here, the subscripting was loss. Goddamn formatting issues! I've fixed that, anyways.

As for knowing the slenderness and broadness depending on vowel, I was actually not aware of that... Although frankly to be honest I don't think I'll put it in. Obviously I don't feel like copying Irish outright, since I'm just going for something like the theme park version, so as long as it superficially looks like Irish, that's good enough for me. However, I think it's a feature I can use in some of the daughter languages of Merthic. Do you have any good resources explaining that? Because for some reason the few resources I used didn't really explain that part too well or at all.

Clannad looks like a reference to the anime?

Actually it was a reference to the band. No, just kidding, you're right dead-on. It's a shameless reference. Especially considering the fact that it's bad Irish on the part of whoever thought of the name. Well, most of the triconsonantal roots will be shameless references anyways, so yeah.

[Bristel]

Are there going to be bi-consonantal roots as well? Or quad?

[Risla]

hahaha, the current root for "be evil" in South Eresian is /ʔiˈtɬeɾ/ <hitlér->. Maybe I should change it.

[cybrxkhan]

Are there going to be bi-consonantal roots as well? Or quad?

Almost certainly. I'll probably use them mainly for foreign loanwords and onomatopoeic stuff, but there'll probably be some more normal ones, in order to keep consistent with the general atmosphere of Arabic morphology. Likewise due to sound changes some of the triconsonantal roots might simplify (or complicate themselves).

hahaha, the current root for "be evil" in South Eresian is /ʔiˈtɬeɾ/ <hitlér->. Maybe I should change it.

Oh no, I love having these kinds of little Easter eggs in my conlangs.

[finlay]

Also palatalisation (Irish slender consonants) is indicated by ʲ in the IPA, not j. And are you trying to make it look anything like Irish? because you haven't copied the system of knowing whether it's slender or broad depending on what vowel it's next to.

Actually I am aware of how to indicate palatalization in IPA, but I think what happened was when I copied this from my word document onto here, the subscripting was loss. Goddamn formatting issues! I've fixed that, anyways.

As for knowing the slenderness and broadness depending on vowel, I was actually not aware of that... Although frankly to be honest I don't think I'll put it in. Obviously I don't feel like copying Irish outright, since I'm just going for something like the theme park version, so as long as it superficially looks like Irish, that's good enough for me. However, I think it's a feature I can use in some of the daughter languages of Merthic. Do you have any good resources explaining that? Because for some reason the few resources I used didn't really explain that part too well or at all.

This applies to Scottish Gaelic as well, incidentally: basically, the consonant is slender if it is next to an orthographical <i> or <e> and broad if it is next to an orthographical <a>, <o> or <u>. So the vowels aren't always what they seem, and it can be quite difficult to tell what they're meant to be if you're unfamiliar with it. However, an accented vowel is always long, and AFAIK the vowel sound usually follows this.

You therefore get things like the imaginary words <tí> /tʃi:/ (slender because it's next to <i>) and <toí> /ti:/ (broad because it's next to <o>).

The consonants can also be "aspirated" if you add an h next to them, which is a second level of complexity: essentially, some consonant graphemes represent four phonemes. Aspiration here doesn't mean the same thing as it does in linguistics, by the way - for example, bh and mh represent /v/, or /w/ if they're slender I think, and th/sh represent /h/.

The example from Scottish Gaelic (which probably applies to Irish too) is the name Seumas, which is /ʃeməs/ (slender s next to e, broad m and broad s), but turns to Sheumais in the vocative, with an aspirated first S and slender second S next to i. This is /heməʃ/, from whence we get the anglicised Hamish. I think the spelling Seamus is the Irish way of spelling it, incidentally, although I'm not sure about that.

Anyway, that's a rough guide to it. You're already most of the way there by knowing what palatalisation is – it's just the vowels that are the difficulty with Gaelic spelling because you might have two or three orthographic vowels to compensate for palatalisation or velarisation of one of the surrounding consonants, and it's not always obvious how to reconcile them into a single vowel sound unless you know the language.

Also, we could take the native names of both languages: Gaeilge for Irish, which is /ge:lʲɟə/ according to Wikipedia - so we have a broad g followed by a slender lg cluster. The purpose of the "a" vowel is to show that the first g is broad. For Scottish Gaelic, it's Gàidhlig, which is /ka:lʲikʲ/ (they also devoice /g/, incidentally): broad <g>; accented <à> to show that it's long; <i> to show that the next consonant is slender; <dh> is aspirated d, which is /ɣ/, but elided in slender position or something; slender <l>; <i> shows that the next and previous consonants are slender, and indicates /i/; and finally a slender <g>.

[cybrxkhan]

This applies to Scottish Gaelic as well, incidentally: basically, the consonant is slender if it is next to an orthographical <i> or <e> and broad if it is next to an orthographical <a>, <o> or <u>. So the vowels aren't always what they seem, and it can be quite difficult to tell what they're meant to be if you're unfamiliar with it. However, an accented vowel is always long, and AFAIK the vowel sound usually follows this.

You therefore get things like the imaginary words <tí> /tʃi:/ (slender because it's next to <i>) and <toí> /ti:/ (broad because it's next to <o>).

The consonants can also be "aspirated" if you add an h next to them, which is a second level of complexity: essentially, some consonant graphemes represent four phonemes. Aspiration here doesn't mean the same thing as it does in linguistics, by the way - for example, bh and mh represent /v/, or /w/ if they're slender I think, and th/sh represent /h/.

The example from Scottish Gaelic (which probably applies to Irish too) is the name Seumas, which is /ʃeməs/ (slender s next to e, broad m and broad s), but turns to Sheumais in the vocative, with an aspirated first S and slender second S next to i. This is /heməʃ/, from whence we get the anglicised Hamish. I think the spelling Seamus is the Irish way of spelling it, incidentally, although I'm not sure about that.

Anyway, that's a rough guide to it. You're already most of the way there by knowing what palatalisation is – it's just the vowels that are the difficulty with Gaelic spelling because you might have two or three orthographic vowels to compensate for palatalisation or velarisation of one of the surrounding consonants, and it's not always obvious how to reconcile them into a single vowel sound unless you know the language.

Also, we could take the native names of both languages: Gaeilge for Irish, which is /ge:lʲɟə/ according to Wikipedia - so we have a broad g followed by a slender lg cluster. The purpose of the "a" vowel is to show that the first g is broad. For Scottish Gaelic, it's Gàidhlig, which is /ka:lʲikʲ/ (they also devoice /g/, incidentally): broad <g>; accented <à> to show that it's long; <i> to show that the next consonant is slender; <dh> is aspirated d, which is /ɣ/, but elided in slender position or something; slender <l>; <i> shows that the next and previous consonants are slender, and indicates /i/; and finally a slender <g>.

Man, I feel bad that you typed that all up. Anyhow, thanks for all of that, it was pretty informative. I think some of that I can use for the allophony and dialectal sound changes in Merthic, but otherwise I think I'm keeping most of what I have so far, or at least the general structure; this information will still come in handy when I try making the daughter languages though, so thank you very, very much!

[finlay]

no problem - I find I can type for ages and only realise afterwards that I've made a wall of text.

I also do this thing where I consider posts to be more like drafts, especially here where it doesn't tell others that you've edited it unless someone's posted after you. So I often end up editing it when I think of a new paragraph or five that could fit well into the existing post.

[cybrxkhan]

MERTHIC SCRIPT

Originally I wanted Merthic's script to be similar to Chinese, but gave up on the idea. Basically, the Merthic script now is a kind of a Hangul-esque abjad in structure and its looks are vaguely similar to the runic scripts. Each symbol represents a specific consonant, and the way the symbols of a word are group discern whether it is monoconsonantal, biconsonantal, triconsonantal, or quadconsonantal. I'm not sure whether it's good or not, or whether I should change it a bit, so I'd like to see if you guys have any comments or suggestions.

Anyhow, below is a visual explaining structure of a Merthic character:




And here are a couple of the example sentences written in the script. The first line is the formal script proper, the second line is the 'cursive' form of the script, the third line is as if it were to be written with Latin characters, and the fourth line is the Romanization.



The gloss and translation of the two sentences in the above examples are:

Eire cleanndá clannád wallám múir dhiheth.
3SG.FEM affectionate love dominate many scholar.
She who is affectionate loves to dominate/control [the] scholar a lot.

Willem ocluinnde bratán briten dheahthá.
ruler loving give.birth human intelligent.
[The] loving ruler gave birth [to an] intelligent human.

[cybrxkhan]

SERIAL VERB CONSTRUCTIONS

This is one of the more interesting parts of Merthic, in my opinion, so any comments and thoughts are very much welcome. Merthic allows for serial verb construction, like in many East Asian languages. Merthic’s serial verb construction allows for the expression of mood, aspect, and even negation, which are otherwise absent.

NEGATION

One of Merthic’s most distinguishing traits is the lack of grammatical negation. There is no way to technically negate a sentence in Merthic – no inflection, no affix, no particle, no adverbial phrasing, etc. (although almost all the daughter languages did acquire negation from other languages). In order to ‘negate’ a sentence, you either supply an alternative verb (i.e., “I am not strong” --> “I lack strength”), or use serial verb constructions. I’ll make examples in English, first, to explain things (note that in Merthic there are verbs which express copula + adjective, i.e. "to be Irish" would be a verb on its own, independent of "to be" and "Irish"):

You don't love me! --> You be.averse love me. ("You are averse to loving me!")
Scottish people do not speak funny. --> Scottish people lack speak funny. ("Scottish people lack funny speech.")
The king cannot raise an army. --> The king be.unable raise army. ("The king is unable to raise an army.")
I cannot be stupid. --> I prevent be.stupid. ("I prevent myself from being stupid.")


Two example sentences in Merthic:

Pharoh satáp acglioch.
house prevent be.high
[The] house cannot be tall.

Willem franás clannád eire.
ruler lack love 3SG.FEM
[The] ruler does not love her.

Personally I don't know if this kind of thing - a lack of formal negation - is attested in any natlang, but I don't think this is too outlandish a feature, and it allows for - hopefully - more expression on part of the speaker.



MOODS

Some moods are also usually indicated by serial verb construction. Although there are quite a number of possibilities with serial verb construction, I’ll highlight some of the more common ones:

Subjunctive

The subjunctive is indicated by a verb construction using the verb “tanáh”, which means “to want” or “to desire”. For instance:

Gealig tanáh aptiot.
1SG desire eat
I want to eat.


Imperative

There are a variety of verbs that can be used to indicate an imperative mood, and each vary in terms of politeness. “Badáms”, meaning “to obey”, “to capitulate”, is a formal term used by superiors to inferiors; “Manách”, meaning “to appease” or “to satisfy”, is a neutral, casual term used by equals; and “Sacrát”, meaning “to concede”, “to accept”, is an informal term used by inferiors to superiors, or amongst those who are intimate. There are many more that can be used, but these were the most common.

Thus, for example:

Dhavéah manách patát.
2SG appease eat
Please eat.

Caigh badáms vanác pharoh.
2SG obey build house.
Build a house.


Conditional/Potential

Like the imperative, a number of verbs can be used to indicate a conditional mood. For instance, “adfriorr”, meaning “to be possible”, and “albnios”, meaning “to be hypothetical” or “to be theoretical”, is commonly used. Note that in this case the conditional mood also acts as the potential mood. Anyhow, an example sentence:

Axanagc albnios clannád eire.
3SG.MASC be.hypothetical love 3SG.FEM
He would love her./He could love her./Possibly, he loves her./If he loves her




ASPECTS

Again, serial verb constructions can indicate the equivalent of grammatical aspects in other languages.

Perfect

The perfect aspect is indicated by a construction of “lamáth”, meaning “to finish/complete”, and another verb or verbs; it can also be used to indicate that the subject has finished the action. For instance:

Hitler lamáth patát briten.
villain finish eat person
[The] villain had eaten people./The villain finished eating people.

Progressive

The progressive is indicated by a construction of “fanát”, meaning “to continue”, and another verb or verbs.

Hitler fanát patát briten.
villain continue eat person
The villain is eating people.