New Conlanger Requesting Guidance on Proto-Lang

[BarkMiner]

Intro/Mission Statement

Hi all, newbie looking for help with a conlang. This conlang is for an RPG setting, and is meant to be an ancient language that will have a heavy influence on the world languages that follow. I’m currently referring to it as proto-ogrish (as it is originally spoken by a particular tribe of Ogres), but it needs an internal name at some point. I’m essentially just wanting some feedback at this early stage as to whether I’m on the right track given my goals so I can adjust early if need be. Goals for this project are:

• The language should be reasonably naturalistic (if someone who knows their stuff were to see it it shouldn’t be completely embarrassing)

• The language should express some concepts relevant to the speakers more easily than English or French (the languages I am familiar with), and other concepts with more difficulty

• The resulting language family should similarly contain languages that are conceptually distinct from one another, but have enough cognates to provide clues as to their relatedness

• At this stage I would like to invest minimum effort, as the root language and languages that follow after it will only feature minimally in the materials I currently plan on writing, however…

• It should be a solid enough foundation that I can expand on it in the future.

Early Phonology
Consonant inventory: p,b,t̪,d̪,k,g,x,h,m,n,l,w
Vowel inventory: i,i:,ə,ə:,ɑ,ɑ:

Initially these may only be combined by pairs with a consonant followed by a vowel, but only some pairs are legal at this stage. Any consonant that may be followed by a short vowel may be followed by its long equivalent. Any consonant may be followed by ɑ. [p,b] may be followed by i. [x,h,m,n,l,w] may be followed by ə. [t̪,d̪,k,g] may be followed by [i,ə].

Early Grammar
Sentence role is marked by suffixes that come at the end of the related clause.

-kɑ is suffixed onto the final word of the clause containing the subject and a present-tense verb.

-hɑ is suffixed onto the final word of a clause containing a past tense verb (at this initial stage we only have past perfect).

-gɑ is suffixed onto the final word of a clause containing the direct object.

Nouns are also marked with case suffixes:

Singular/Familiar: Unmarked
Plural: -bɑ
Greater Plural: -bɑbɑ (greater plural often means all of something, but when contrasted with a regular plural can mean more than the previously referenced plural)
Sub-singular: -mɑ (sub-singular refers to either a partial version of the noun or none of it)
Unfamiliar: -pɑ (this is used when the noun is unknown to either the speaker or the listener)

The unfamiliar suffix always comes before a quantity suffix. The clause always ends in kɑ or gɑ, which is always placed at the end of any other case markings on the final noun.

At this stage in the language nouns are used as adjectives. With the subject marker placed at the end of the noun clause rather than the main noun it can be difficult to distinguish what a verb is referring to. This is complicated by the language having no clear word order rules. Thus verbs are conjugated to agree with their subject in person and plurality (but not familiarity).

1st person singular: unmarked
1st person plural: -[name]bɑ (so someone named hibi would end a verb referencing ‘us’ with ‘-hibibɑ’)
1st person greater plural: -[name]bɑbɑ
1st person subsingular: -[name]mɑ
2nd person singular: -t̪əgi
2nd person plural: t̪əgibɑ
2nd person greater plural: -t̪əgibɑbɑ
2nd person subsingular: -t̪əgimɑ
3rd person singular (animate): -t̪əgɑ
3rd person plural (animate): -t̪əgɑbɑ
3rd person greater plural (animate): -t̪əgɑbɑbɑ
3rd person subsingular (animate): -t̪əgɑmɑ
3rd person singular (inanimate): -t̪əxə
3rd person plural (inanimate): -t̪əxəbɑ
3rd person greater plural (inanimate): -t̪əxəbɑbɑ
3rd person subsingular (inanimate): -t̪əxəmɑ

Though there is no required word order for a sentence, there are word-order conventions. Sentences generally end with either the subject or the verb, as both end in -kɑ which also marks the end of the sentence, so it can be easier to arrange things this way. In the present tense the word order defaults to SOV (subject, object, verb), but in other tenses where the verb phrase does not end in -kɑ the default switches to VOS.

Phrases default to head-initial. Generally the default patterns are broken for the sake of drama, suspense, puns, or comedy in a sentence. Initially this meant that surprising adjectives were more likely to come before a noun, but quickly this came to add a sense of irony to expected adjectives when they come before a noun.

Pronouns are rarely used, and imply unfamiliarity with the subject. Generally nouns and proper nouns are simply repeated throughout a conversation. One would simply use their own name rather than use a pronoun, or simply not include a subject as the speaker is assumed to be the subject when none is supplied. Using a pronoun to refer to someone you know rather than their name can be seen as rude or imply insult. The third person animate pronoun can be used to refer to the self occasionally to express distance from the self (the exact meaning of this can be contextual but usually it implies shame). The main pronouns follow:

2nd person singular: t̪əgi
2nd person plural: t̪əgibɑ
3rd person singular (animate): t̪əgə
3rd person plural (animate): t̪əgəbɑ (when quantity is known it is more common to use 3rd person singular followed by a number, though if this is the subject the verb will still agree with the plural)
3rd person singular (inanimate): t̪ixə
3rd person plural (inanimate): t̪ixəbɑ

Basic Examples of Early Language
Where word order is relevant:

t̪ixəxəwə = warrior
wəhə = mighty

t̪ixəxəwə wəhəkɑ (“warrior mighty” is likely to imply that the warrior actually is mighty)
wəhə t̪ixəxəwəkɑ (“mighty warrior” is more likely to be mocking the warrior and imply the opposite)

Intransitive example:

kit̪ɑ = to fly
kigə = bird

kit̪ɑkɑ = I fly
kigəkɑ kit̪ɑt̪əgɑkɑ = The bird (familiar) flies

Transitive example:

bə = to find (this will later become a main copula)

t̪ixəxəwə wəhəkɑ kigəgɑ bət̪əgɑkɑ = the warrior mighty (familiar) finds the bird (familiar)
t̪ixəxəwə wəhəkɑ kigəpɑgɑ bət̪əgɑkɑ = the warrior mighty (familiar) finds a bird (unfamiliar)
bət̪əgɑhɑ kigəpɑgɑ t̪ixəxəwə wəhəkɑ = found was a bird (unfamiliar) by the warrior mighty (familiar)

I hope this is enough to give a basic idea of how the initial language would be spoken. Let me know if I should provide some more examples to make things clearer at this point.

Early Evolution
In rapid speech, a lot of adjustments end up being made. Repeated pairs of consonants and vowels are often simplified by removing the first vowel and lengthening the second. For example the greater plural suffix bɑbɑ -> bbɑ: (the b is pronounced twice). This occurs everywhere where the same consonant-vowel pair is repeated, but the vowel loss also extends to extremely common morphemes consisting of two consonant-vowel pairs, particularly where both vowels are the same. The ɑ is dropped from -kɑ, -gɑ, and -hɑ suffixes.

xəwə, a morpheme which means ogre, mud, or person, is used frequently in compounding and simplifies to xwə. xw becomes understood as a single consonant by native speakers and like x and w may be followed with either ɑ or ə. In repeated consonant-vowel pairs of x or w, xw is used rather than xx or ww when vowel reduction occurs (e.g. xəxə -> xwə:).

For many common morphemes the extended vowel following the vowel loss is also eventually lost.

The final stop kk is developed to clarify the ending of an entire idea. This is often used at the end of long sentences, ideas, or simply to indicate that a speaker has finished speaking.

Noun and verb clauses now may also end with x, which indicates uncertainty in the clause. What this uncertainty means depends a bit on context (examples below).

Ending a verb clause in a d now indicates imperfect past. Ending a verb clause in t indicates either an imperative or the imperfect future. Ending a verb clause in tk now indicates an imperfect imperative or imperfect future, and kt indicates habitual present.

Updated noun cases:

Plural: -bɑ
Greater Plural: -bbɑ (greater plural often means all of something, but when contrasted with a regular plural can mean more than the previously referenced plural)
Sub-singular: -mɑ (sub-singular refers to either a partial version of the noun or none of it)
Unfamiliar: -pɑ (this is used when the noun is unknown to either the speaker or the listener)
Unfamiliar plural: -pbɑ:
Unfamiliar sub singular: -pmɑ:
Unfamiliar greater plural: -pbɑ:bɑ

Updated agreement conjugations:

1st person singular: unmarked
1st person plural: -[name]bɑ
1st person greater plural: -[name]bbɑ
1st person subsingular: -[name]mɑ
2nd person singular: -t̪gi
2nd person plural: t̪gi:bɑ
2nd person greater plural: -t̪gibbɑ:
2nd person subsingular: -t̪gi:mɑ
3rd person singular (animate): -t̪gɑ:
3rd person plural (animate): -t̪gɑ:bɑ
3rd person greater plural (animate): -t̪gɑ:bbɑ
3rd person subsingular (animate): -t̪gɑ:mɑ
3rd person singular (inanimate): -t̪xə
3rd person plural (inanimate): -t̪xə:bɑ
3rd person greater plural (inanimate): -t̪xə:bbɑ
3rd person subsingular (inanimate): -t̪xə:mɑ

Updated Pronouns:

2nd person singular: t̪gi
2nd person plural: t̪gi:bɑ
3rd person singular (animate): t̪gə
3rd person plural (animate): t̪gə:bɑ
3rd person singular (inanimate): t̪xə
3rd person plural (inanimate): t̪xə:bɑ

Basic Examples of Evolved Language
Riffing off of one of our previous example sentences:

t̪ixəxəwə wəhəkɑ kigəpɑgɑ bət̪əgɑkɑ = the warrior mighty (familiar) finds a bird (unfamiliar)

Is now pronounced

t̪ixəxwə wəhək kigəpɑg bət̪əgɑk

t̪ixəxwə wəhəx kigəpɑg bət̪əgɑgk = the mighty warrior (familiar) a bird (unfamiliar) found? Did the mighty warrior find a bird? Note that subject and object are now more ambiguous when there is a question about the subject, though the direct object signifier ‘g’ provides a clue.

t̪ixəxwə wəhəx kigəpɑx bət̪əgɑxk = mighty warrior (familiar) bird found? Difficult to translate but indicates that the speaker knows something happened involving a mighty warrior and a bird, one of them finding the other, and is uncertain about any of the details. Perhaps: What happened with the mighty warrior and a bird finding one or the other?

Conclusion and Questions

I could go into more examples but if you've made it through all the above I already appreciate you greatly and respect your time. I figure if more examples are needed you probably know better what to ask for than I know what to include. First and foremost before I evolve this any further, does this seem like a reasonable start? Am I missing anything major?

Also, I'm working on the writing system which has a sort of complex evolution. The initial pictographs are written by ogres on cave walls (both painted and carved) but it's adapted into a phonetic alphabet by the goblins they trade with on the basis of ogrish pictograms. This alphabet will at first encode consonant-vowel pairs as well as stop consonants, though it will neglect/group things that don't appear as distinct to the goblins as it does the ogres. From there it will evolve separately and become the basis for the written languages throughout the entire region with significant variation by the 'present' the rpg is set in. The initial goblin versions of the script are woven into or dyed into fabric, and the ogres use this convention for communicating with the goblins at first but when they adopt the phonetic alphabet more fully they also carve/paint it on stone. Curious for thoughts/pointers as to how this history might effect what types of symbols are used for letters. I have some very basic ideas, as well as some notes on how the writing is going to influence the further evolution of the language, but would like a sanity check from some more experienced folks before I go any further.

Once again thank you so so much if you made it this far, looking forward to any advice.

[VaptuantaDoi]

Early Phonology
Consonant inventory: p,b,t̪,d̪,k,g,x,h,m,n,l,w
Vowel inventory: i,i:,ə,ə:,ɑ,ɑ:

This looks mostly fine, except for a few points:
- If you've only got one series of coronal stops, you don't really need to transcribe them as /t̪ d̪/ rather than just /t d/; just mention that they're dental rather than alveolar in the phonology section
- Having /x h/ as the only fricatives is unnatural; I'd expect at least /s/ to balance it out (although having said that, I'm sure there's some natlang counterexamples; apparently Abipon is claimed to have just /x ħ/. Even so for this inventory it makes more sense to add /s/, or instead drop one of /x h/.)
- Your vowels are a bit weird, but not obscene. I'd expect them to behave like a vertical system, with realisations like [i~ɨ e~ə~o a~ɑ].

Initially these may only be combined by pairs with a consonant followed by a vowel, but only some pairs are legal at this stage. Any consonant that may be followed by a short vowel may be followed by its long equivalent. Any consonant may be followed by ɑ. [p,b] may be followed by i. [x,h,m,n,l,w] may be followed by ə. [t̪,d̪,k,g] may be followed by [i,ə].

This seems like a weird set of restrictions to me, although you could probably justify it somehow.

Early Grammar
Sentence role is marked by suffixes that come at the end of the related clause.

-kɑ is suffixed onto the final word of the clause containing the subject and a present-tense verb.

-hɑ is suffixed onto the final word of a clause containing a past tense verb (at this initial stage we only have past perfect).

-gɑ is suffixed onto the final word of a clause containing the direct object.

I'm not sure I totally get what you mean with these. -kɑ and -hɑ look like clause-final tense clitics, but -gɑ is like a valency marker. I'd have to see how they're used in more detail to give more feedback.

Nouns are also marked with case suffixes:

Singular/Familiar: Unmarked
Plural: -bɑ
Greater Plural: -bɑbɑ (greater plural often means all of something, but when contrasted with a regular plural can mean more than the previously referenced plural)
Sub-singular: -mɑ (sub-singular refers to either a partial version of the noun or none of it)
Unfamiliar: -pɑ (this is used when the noun is unknown to either the speaker or the listener)

The unfamiliar suffix always comes before a quantity suffix. The clause always ends in kɑ or gɑ, which is always placed at the end of any other case markings on the final noun.

These aren't case suffixes, they're number suffixes. I've never seen anything like the 'sub-singular' be morphologically marked but I guess it's not impossible.

1st person plural: -[name]bɑ (so someone named hibi would end a verb referencing ‘us’ with ‘-hibibɑ’)

That's very unusual and seems pretty cumbersome to me. And also not really naturalistic - in technical terms, it looks like agent noun incorporation, which even highly synthetic languages tend to avoid for various reasons.

In the present tense the word order defaults to SOV (subject, object, verb), but in other tenses where the verb phrase does not end in -kɑ the default switches to VOS.

Complete reversal of basic constituent order based on tense is weird. Languages don't generally have rules which reverse more than two segments, whatever they may be. I'd suggest changing one of these in such a way that the order of two constituents is common between them - e.g. have SOV vs. VSO. (And bear in mind that VOS is a rared default order; OSV and OVS are even rarer)

2nd person singular: t̪əgi
2nd person plural: t̪əgibɑ
3rd person singular (animate): t̪əgə
3rd person plural (animate): t̪əgəbɑ (when quantity is known it is more common to use 3rd person singular followed by a number, though if this is the subject the verb will still agree with the plural)
3rd person singular (inanimate): t̪ixə
3rd person plural (inanimate): t̪ixəbɑ

It's odd that they all begin with /t/, but there's some natlang precedent. Not having first person pronouns at all is unnaturalistic, even if they're rarely used.

I hope this is enough to give a basic idea of how the initial language would be spoken. Let me know if I should provide some more examples to make things clearer at this point.

They couldn't hurt.

Early Evolution
In rapid speech, a lot of adjustments end up being made.

[...]

Your sound changes are ok, although they're a bit irregular. Sound changes are generally phonologically conditioned; say, the final vowel of a polsyllabic word is dropped, rather than just in bound morphemes.

Conclusion and Questions

I could go into more examples but if you've made it through all the above I already appreciate you greatly and respect your time. I figure if more examples are needed you probably know better what to ask for than I know what to include. First and foremost before I evolve this any further, does this seem like a reasonable start?

It's definitely distinctive for a beginner's conlang, which is a good thing! Some of your terminology could be clearer, and it's unnaturalistic in parts, but it's got potential to be really interesting. If you haven't read it already, there's a pretty good beginner's guide online by Mark Rosenfelder which might help clarify some of your ideas, and it also touches on making naturalistic sound changes.

(I don't have enough expertise to comment on your writing system ideas, but I'm sure there's someone here who does and will)

[BarkMiner]

- Having /x h/ as the only fricatives is unnatural; I'd expect at least /s/ to balance it out (although having said that, I'm sure there's some natlang counterexamples; apparently Abipon is claimed to have just /x ħ/. Even so for this inventory it makes more sense to add /s/, or instead drop one of /x h/.)
- Your vowels are a bit weird, but not obscene. I'd expect them to behave like a vertical system, with realisations like [i~ɨ e~ə~o a~ɑ].

Can you help me to understand why this is, to avoid future mistakes like it in the future?

This seems like a weird set of restrictions to me, although you could probably justify it somehow.

Is it unusual to restrict what consonants and vowels can appear together at all, or is the way I have done it what is strange?

These aren't case suffixes, they're number suffixes. I've never seen anything like the 'sub-singular' be morphologically marked but I guess it's not impossible.

Afaik there's no precedent for this but it seemed like a relatively harmless place to stray off the beaten path and somehow felt right.

1st person plural: -[name]bɑ (so someone named hibi would end a verb referencing ‘us’ with ‘-hibibɑ’)

That's very unusual and seems pretty cumbersome to me. And also not really naturalistic - in technical terms, it looks like agent noun incorporation, which even highly synthetic languages tend to avoid for various reasons.

Not having first person pronouns at all is unnaturalistic, even if they're rarely used.

Got it. With the proper noun-pronoun thing the idea was basically to subvert fantasy expectations from the viewer's point of view. I really really hate the dumb orcs/ogres/whatever trope that's characterized with weird lazy dialect like these characters referring to themselves in the third person. I wanted to turn that on its head and have that speech quirk turn out that that kind of self-reference was based on a completely different set of linguistic assumptions from their native tongue rather than stupidity. But I may have gotten a bit overzealous trying to bake that into the language, I'll figure out a first person pronoun to be uncommonly used and maybe incorporate that into verb conjugation.

I'll mess with pronouns generally a bit and try to better understand where they might come from as you're right in general they're sort of arbitrary and artificial feeling at the moment.

Complete reversal of basic constituent order based on tense is weird. Languages don't generally have rules which reverse more than two segments, whatever they may be. I'd suggest changing one of these in such a way that the order of two constituents is common between them - e.g. have SOV vs. VSO. (And bear in mind that VOS is a rared default order; OSV and OVS are even rarer)

This will require some thought to resolve. Initially I was going to have no default word order at all, which I know is sort of rare, but as I was toying with it I realized that it's more convenient to use certain word orders and a default order would likely naturally come about even if it isn't grammatically required. For convenience I'll stick mostly to SOV for now until I have a better idea what I'm doing.

Your sound changes are ok, although they're a bit irregular. Sound changes are generally phonologically conditioned; say, the final vowel of a polsyllabic word is dropped, rather than just in bound morphemes.

I think I follow but I'm not certain -- are you basically saying that such change rules are likely to occur at the word level rather than the morpheme level?

It's definitely distinctive for a beginner's conlang, which is a good thing! Some of your terminology could be clearer, and it's unnaturalistic in parts, but it's got potential to be really interesting. If you haven't read it already, there's a pretty good beginner's guide online by Mark Rosenfelder which might help clarify some of your ideas, and it also touches on making naturalistic sound changes.

Thanks so much for all the help, will definitely check out the article here shortly.

Since you asked, some further examples (avoiding too many things that are likely to change in the near future on the basis of your feedback).

Di = To See
Tɑ: = The Sky

Tɑ:g dik = I see the sky
Təgik tɑ:g ditəgik = You (unfamiliar) see the sky
Tɑ:mɑg dik = I don't see all of the sky -- Probably "I can't see the sky", but might vary based on context.
Təgik tɑ:g ditəgixk = Do you (unfamiliar) see the sky?
Təgik tɑ:x ditəgik = Is it the sky that you (unfamiliar) see?

I'll keep working with this, as it stands I see where the pronouns and agreement conjugations are a little awkward though maybe sound change over time (more properly implemented) will add a more natural feel. I'm also terrified to dive into more complex compound sentences, though I think I have the starting groundwork for it there's a lot of grammatical stuff to look into I'm a bit intimidated by.

[Salmoneus]

Hi all, newbie looking for help with a conlang. This conlang is for an RPG setting, and is meant to be an ancient language that will have a heavy influence on the world languages that follow. I’m currently referring to it as proto-ogrish (as it is originally spoken by a particular tribe of Ogres), but it needs an internal name at some point. I’m essentially just wanting some feedback at this early stage as to whether I’m on the right track given my goals so I can adjust early if need be. Goals for this project are:

• The language should be reasonably naturalistic (if someone who knows their stuff were to see it it shouldn’t be completely embarrassing)

• The language should express some concepts relevant to the speakers more easily than English or French (the languages I am familiar with), and other concepts with more difficulty

• The resulting language family should similarly contain languages that are conceptually distinct from one another, but have enough cognates to provide clues as to their relatedness

• At this stage I would like to invest minimum effort, as the root language and languages that follow after it will only feature minimally in the materials I currently plan on writing, however…

• It should be a solid enough foundation that I can expand on it in the future.

Hi.
OK, that sounds reasonable. I would express some caution on the 'should expess some concepts relevant to the speakers more easily than English or French' aspect, as this is hard to really make work in practice. Grammatically, some languages make some sorts of propositions easier to express than others, but in too broad and content-neutral a way to really match cultural concerns; lexically, of course, a language will have more words for more culturally-significant things, but that's fairly peripheral to the actual structure of the language.

I'd also say that "at this stage I would like to invest minimum effort" is an extremely familiar sentiment around these parts... usually followed by years of effort! [everyone's big project always seems to be 'the thing I thought I'd spend very little time on before moving on to what I thought was the big project'...]

---------------------------------------

Early Phonology
Consonant inventory: p,b,t̪,d̪,k,g,x,h,m,n,l,w
Vowel inventory: i,i:,ə,ə:,ɑ,ɑ:

OK. Seems reasonable. Three notes:
- I can't read it, but I think that's a dental mark under /t/ and /d/? That's fine, but bear in mind you don't actually need to include that - the general principle is that you only need to include information that disambiguates from something else (eg if you have both dental and alveolar stops, then you need to mark one pair as explicitly dental). Otherwise it's normal to just use the simplest valid representation - in this case /t/ can be used for either alveolars or dentals. Having said that, I've no actual problem with you including the dental marks here rather than saying in words 'the coronal stops are alveolar'. It just may be easier for you not to have to type the marks every single time!

- it's extremely unusual for a human language to have multiple fricatives and no /s/. I also think it may be unusual to have /l/ without /r/ (if a language only has one of these two, it's common for the phoneme to be realised in either way depending on context). However, since this language is spoken by ogres, I have no problems with this, maybe ogre tongues/mouths are just a bit different from ours. I note in fact that they don't have any non-dorsal fricatives or approximants other than /l/, which is sort of a stop anyway (the centre of the tongue touches but the edges don't), and this seems a plausible limitation for a species that perhaps has a larger and less agile tongue than humans do.

- The vowel inventory is a little unusual; usually the three vowels would be /i a u/, or occasionally (mostly in North America) /i a o/ or /e a o/. It's odd not to have a fully back and rounded vowel. But, it probably does happen somewhere in North America, and in any case, these aren't humans, so OK. The one slightly odd thing to me is that you do have /w/, so you do have lip rounding, so it's odd that they don't have any rounded vowels. But not impossible. However, if it were me I'd make "no lip rounding for ogres" a thing and replace /w/ with... whatever the symbol for a velar, uvular or pharyngeal approximant is.

Regarding symbols, again, you can just call your low vowel /a/, you don't have to use the specific awkward symbol for a back low vowel every time. Since in this language the two soudns don't contrast (and I'd be extremely surprised if your /A/ (that's another way of writing your vowel, in an encoding system called 'x-sampa', which is easier that copy-pasting IPA) isn't sometimes realised as [a], particularly given there are only three vowels - fewer phonemes usually means more allophones).

Which actually brings me to an important question: do you understand yet the words phone, phoneme and allophone? If not, you should learn about them, since they make things a lot easier.

Short version: the sounds that commonly appear in a language are called 'phones'. However, many individual phones are essentially meaningless, because they are equivalent to other phones. Sometimes one speaker may say a word using one phone, while another speaker may say the same word using a different phone. So in describing languages we instead talk about 'phonemes', which are meaningful phonetic units: two things are only phonemes if replacing one with the other changes the meaning of the word, or makes the word meaningless. So in standard English, there is no phonemic distinction between alveolar and dental /t/ - there is only /t/. Pronouncing 'pat' with a dental /t/ may make you sound French, but it doesn't change what word you're saying. On the other hand, in some Irish dialects of English, this IS a phonemic distinction: [pat̪] is an entirely different word from [pat] in those dialects (the former is what we spell 'path' - standard interdental fricatives become dental stops in this dialect).

Similarly, when you break up a language into phonemes, you find that the 'same' phoneme is actually pronounced differently in different phonetic contexts. So, in standard English, /t/ is pronounced differently in the word 'tack' from how it is pronounced in the word 'stack': the former is aspirated, and the latter is unaspirated. In Mandarin, this difference would be phonemic and hugely important, but in English it is not phonemic: if you pronounce 'stack' with an aspirated stop, everyone will still understand you, and may not even notice that you sound a little foreign. This alternative pronunciations of the same phoneme in different contexts are called allophones. Which allophones a phoneme has depends on the language. As a result, there is no 'correct' representation of a phoneme itself - phonemes are abstract, not concrete. Phonemes are the abstract entities represented in actual speech by phones, while two phones that represent the same phoneme are allophones. Conventionally, phonemes are represented in transcriptions by a symbol that evokes the most common phone that instantiatiates that phoneme, but this isn't actually necessary - there's a famous grammar of Marshallese that spells its vowels with little pictures (teacup, football, can't remember the other two). It does this because the phonemes have so many different allophones, with none being more 'fundamental' than any others, that the author felt that randomly picking one to represent the phoneme would be misleading.

[also note: the same phone can be an allophone of more than one phoneme! For instance, the phone [x] may be an allophone of the phoneme /h/ in initial position, but an allophone of /k/ in intervocalic position...]

Initially these may only be combined by pairs with a consonant followed by a vowel, but only some pairs are legal at this stage. Any consonant that may be followed by a short vowel may be followed by its long equivalent. Any consonant may be followed by ɑ. [p,b] may be followed by i. [x,h,m,n,l,w] may be followed by ə. [t̪,d̪,k,g] may be followed by [i,ə].

And here those concepts are relevant to a basic rule of transcription: phonemes go /inside slashes/, while phones go within [square brackets], and graphemes (written representations) go within <angle brackets>. In this case, you're looking for slashes, not brackets, because you're talking about phonemes.

I'd also suggest that this is where you need the concept 'syllable' (a subdivision of a word consisting of a 'nucleus' (usually a vowel) and usually some coda or onset (consonants)). And ideally 'phonotactics' (rules about which phonemes can go next to which phonemes, and which allophones appear in those locations). In this case, the phonotactics of your language only allow simple, CV syllables; and additionally, labial stops may not be followed by schwa, while dorsals, nasals and /l/ may not be followed by /i/.

This isn't impossible, but I do wonder why these rules exist. Saying dorsals can't be followed by /i/ makes sense, as they're at opposite ends of the mouth - presumably at some point in the past /ki/ became /k@/ (that's x-sampa for schwa), or maybe /ti/ or something. But I don't know why you can't have nasals followed by /i/. In particular, why you can have /pi/ and /bi/ but not /mi/. [in theory it's also weird to not allow /li/ as a random rule, but in practice /l/ is often secondarily velarised/uvularised/pharyngealised, so it would fit with the other dorsals in that regard despite being primarily coronal itself]. Similarly, why is /p@/ not allowed!?

-------------------------------------------

Early Grammar
Sentence role is marked by suffixes that come at the end of the related clause.

-kɑ is suffixed onto the final word of the clause containing the subject and a present-tense verb.

-hɑ is suffixed onto the final word of a clause containing a past tense verb (at this initial stage we only have past perfect).

-gɑ is suffixed onto the final word of a clause containing the direct object.

This, I'm afraid, I don't understand. Perhaps some examples?
I'm not sure you're using the word 'clause' correctly. I thought at first you meant 'phrase', but I don't think you can mean that either.

Generally, you're going to have a lot of clauses - perhaps most clauses - containing subjects, verbs, AND objects.

I would suggest beginning by thinking about basic word order in simple clauses. [EDIT: I see you say later that word order is free. That adds some complications, but in any case you might want to lead with that rather than bury it!]

Nouns are also marked with case suffixes:

Singular/Familiar: Unmarked
Plural: -bɑ
Greater Plural: -bɑbɑ (greater plural often means all of something, but when contrasted with a regular plural can mean more than the previously referenced plural)
Sub-singular: -mɑ (sub-singular refers to either a partial version of the noun or none of it)
Unfamiliar: -pɑ (this is used when the noun is unknown to either the speaker or the listener)

The unfamiliar suffix always comes before a quantity suffix. The clause always ends in kɑ or gɑ, which is always placed at the end of any other case markings on the final noun.

None of these things are case suffixes. Possibly -ka and -ga might be case suffixes, depending on what you mean exactly. But these things aren't. Most of these things are just 'numbers'. [and 'sub-singular' int he sense of 'a partial version' is just called 'partitive'. Partitives don't normally cover negatives, but it doesn't seem impossible]. Cases refer to the role of the noun in the sentence (subject, object, etc).

On the 'unfamiliar' - do you mean that the item is completely unknown, as in they've never heard anyone mention it before, or that it's unknown in the context of the sentence, as in they don't know which item it's currently referring to?

In English that's a distinction between "I saw some cat I'd never seen before" and "I saw a cat (I'd actually seen it before but I haven't mentioned it yet so you don't know which one I'm talking about yet)".

The latter is what we call 'indefinite', and it's not that weird to have definiteness pattern with number. The former is... I don't know what that's called. It's not as common in human languages. But it doesn't strik me as inherently implausible.

At this stage in the language nouns are used as adjectives.

I'm not sure what you mean by this. Nouns are nouns and adjectives are adjectives. Do adjectives look exactly like nouns? If so, how do you tell the difference?

With the subject marker placed at the end of the noun clause rather than the main noun it can be difficult to distinguish what a verb is referring to. This is complicated by the language having no clear word order rules. Thus verbs are conjugated to agree with their subject in person and plurality (but not familiarity).

1st person singular: unmarked
1st person plural: -[name]bɑ (so someone named hibi would end a verb referencing ‘us’ with ‘-hibibɑ’)
1st person greater plural: -[name]bɑbɑ
1st person subsingular: -[name]mɑ
2nd person singular: -t̪əgi
2nd person plural: t̪əgibɑ
2nd person greater plural: -t̪əgibɑbɑ
2nd person subsingular: -t̪əgimɑ
3rd person singular (animate): -t̪əgɑ
3rd person plural (animate): -t̪əgɑbɑ
3rd person greater plural (animate): -t̪əgɑbɑbɑ
3rd person subsingular (animate): -t̪əgɑmɑ
3rd person singular (inanimate): -t̪əxə
3rd person plural (inanimate): -t̪əxəbɑ
3rd person greater plural (inanimate): -t̪əxəbɑbɑ
3rd person subsingular (inanimate): -t̪əxəmɑ

Though there is no required word order for a sentence, there are word-order conventions. Sentences generally end with either the subject or the verb, as both end in -kɑ which also marks the end of the sentence, so it can be easier to arrange things this way. In the present tense the word order defaults to SOV (subject, object, verb), but in other tenses where the verb phrase does not end in -kɑ the default switches to VOS.

Phrases default to head-initial. Generally the default patterns are broken for the sake of drama, suspense, puns, or comedy in a sentence. Initially this meant that surprising adjectives were more likely to come before a noun, but quickly this came to add a sense of irony to expected adjectives when they come before a noun.

Pronouns are rarely used, and imply unfamiliarity with the subject. Generally nouns and proper nouns are simply repeated throughout a conversation. One would simply use their own name rather than use a pronoun, or simply not include a subject as the speaker is assumed to be the subject when none is supplied. Using a pronoun to refer to someone you know rather than their name can be seen as rude or imply insult. The third person animate pronoun can be used to refer to the self occasionally to express distance from the self (the exact meaning of this can be contextual but usually it implies shame). The main pronouns follow:

2nd person singular: t̪əgi
2nd person plural: t̪əgibɑ
3rd person singular (animate): t̪əgə
3rd person plural (animate): t̪əgəbɑ (when quantity is known it is more common to use 3rd person singular followed by a number, though if this is the subject the verb will still agree with the plural)
3rd person singular (inanimate): t̪ixə
3rd person plural (inanimate): t̪ixəbɑ

Basic Examples of Early Language
Where word order is relevant:

t̪ixəxəwə = warrior
wəhə = mighty

t̪ixəxəwə wəhəkɑ (“warrior mighty” is likely to imply that the warrior actually is mighty)
wəhə t̪ixəxəwəkɑ (“mighty warrior” is more likely to be mocking the warrior and imply the opposite)

As a general note, it's rare to have a really explicit marker of sarcasm like this, because the whole point of sarcasm is mimicking non-sarcastic speech as directly as possible.

Intransitive example:

kit̪ɑ = to fly
kigə = bird

kit̪ɑkɑ = I fly
kigəkɑ kit̪ɑt̪əgɑkɑ = The bird (familiar) flies

Transitive example:

bə = to find (this will later become a main copula)

t̪ixəxəwə wəhəkɑ kigəgɑ bət̪əgɑkɑ = the warrior mighty (familiar) finds the bird (familiar)
t̪ixəxəwə wəhəkɑ kigəpɑgɑ bət̪əgɑkɑ = the warrior mighty (familiar) finds a bird (unfamiliar)
bət̪əgɑhɑ kigəpɑgɑ t̪ixəxəwə wəhəkɑ = found was a bird (unfamiliar) by the warrior mighty (familiar)

I hope this is enough to give a basic idea of how the initial language would be spoken. Let me know if I should provide some more examples to make things clearer at this point.

Early Evolution
In rapid speech, a lot of adjustments end up being made.

A point here: language change is inevitable and (almost) universal. It's not just a matter of people speaking 'rapidly' (people mostly speak at the same speed, unless they're intentionally speaking slowly for some reason).

Repeated pairs of consonants and vowels are often simplified by removing the first vowel and lengthening the second. For example the greater plural suffix bɑbɑ -> bbɑ: (the b is pronounced twice).

What do you mean by 'pronounced twice'?

The ɑ is dropped from -kɑ, -gɑ, and kɑ suffixes. This occurs everywhere where the same consonant-vowel pair is repeated, but the vowel loss also extends to extremely common morphemes consisting of two consonant-vowel pairs, particularly where both vowels are the same.

Just a reminder: a "consonant-vowel pair" can just be called a "syllable"

xəwə, a morpheme which means ogre, mud, or person, is used frequently in compounding and simplifies to xwə. xw becomes understood as a single consonant by native speakers and like x and w may be followed with either ɑ or ə. In repeated consonant-vowel pairs of x or w, xw is used rather than xx or ww when vowel reduction occurs (e.g. xəxə -> xwə:).

For many common morphemes the extended vowel following the vowel loss is also eventually lost.

The final stop kk is developed to clarify the ending of an entire idea.

This doesn't happen in real language: things aren't just 'developed' to 'clarify' things. They evolve naturally from things that already exist.

This is often used at the end of long sentences, ideas, or simply to indicate that a speaker has finished speaking.

This is called a 'tag' morpheme, innit.

Noun and verb clauses now may also end with x, which indicates uncertainty in the clause.

Why?

What this uncertainty means depends a bit on context (examples below).

Ending a verb clause in a d now indicates imperfect past. Ending a verb clause in t indicates either an imperative or the imperfect future. Ending a verb clause in tk now indicates an imperfect imperative or imperfect future, and kt indicates habitual present.

This all makes it sound like someone just sat aroun and invented things to say!

But also: the whole point of having a proto-language is not having to just invent things in the daughter languages. I'm not sure why you're presenting a daughter language here where features DON'T appear to have evolved from the proto-language. I mean, why not jsut make thsi daughter language the proto-language? [if all subsequent languages evolve from this daughter language then this IS the proto-language, technically!]

Updated noun cases:

Plural: -bɑ
Greater Plural: -bbɑ (greater plural often means all of something, but when contrasted with a regular plural can mean more than the previously referenced plural)
Sub-singular: -mɑ (sub-singular refers to either a partial version of the noun or none of it)
Unfamiliar: -pɑ (this is used when the noun is unknown to either the speaker or the listener)
Unfamiliar plural: -pbɑ:
Unfamiliar sub singular: -pmɑ:
Unfamiliar greater plural: -pbɑ:bɑ

Updated agreement conjugations:

1st person singular: unmarked
1st person plural: -[name]bɑ
1st person greater plural: -[name]bbɑ
1st person subsingular: -[name]mɑ
2nd person singular: -t̪gi
2nd person plural: t̪gi:bɑ
2nd person greater plural: -t̪gibbɑ:
2nd person subsingular: -t̪gi:mɑ
3rd person singular (animate): -t̪gɑ:
3rd person plural (animate): -t̪gɑ:bɑ
3rd person greater plural (animate): -t̪gɑ:bbɑ
3rd person subsingular (animate): -t̪gɑ:mɑ
3rd person singular (inanimate): -t̪xə
3rd person plural (inanimate): -t̪xə:bɑ
3rd person greater plural (inanimate): -t̪xə:bbɑ
3rd person subsingular (inanimate): -t̪xə:mɑ

Updated Pronouns:

2nd person singular: t̪gi
2nd person plural: t̪gi:bɑ
3rd person singular (animate): t̪gə
3rd person plural (animate): t̪gə:bɑ
3rd person singular (inanimate): t̪xə
3rd person plural (inanimate): t̪xə:bɑ

Basic Examples of Evolved Language
Riffing off of one of our previous example sentences:

t̪ixəxəwə wəhəkɑ kigəpɑgɑ bət̪əgɑkɑ = the warrior mighty (familiar) finds a bird (unfamiliar)

Is now pronounced

t̪ixəxwə wəhək kigəpɑg bət̪əgɑk

t̪ixəxwə wəhəx kigəpɑg bət̪əgɑgk = the mighty warrior (familiar) a bird (unfamiliar) found? Did the mighty warrior find a bird? Note that subject and object are now more ambiguous when there is a question about the subject, though the direct object signifier ‘g’ provides a clue.

t̪ixəxwə wəhəx kigəpɑx bət̪əgɑxk = mighty warrior (familiar) bird found? Difficult to translate but indicates that the speaker knows something happened involving a mighty warrior and a bird, one of them finding the other, and is uncertain about any of the details. Perhaps: What happened with the mighty warrior and a bird finding one or the other?

I'm afraid I don't understand these, but maybe I'm just too sleepy right now.

incidentally, a good way to make your examples more transparent is using interlinear glossing - translating each morpheme individually, as it were.

Conclusion and Questions

I could go into more examples but if you've made it through all the above I already appreciate you greatly and respect your time. I figure if more examples are needed you probably know better what to ask for than I know what to include. First and foremost before I evolve this any further, does this seem like a reasonable start? Am I missing anything major?

Well, obviously this is only a rough sketch, so lots of major things are missing, inevitably (eg: relative clauses). Having said that, I'd suggest focusing on improving clarity and use of linguistic conventions in describing what you have. I mostly understand you, but not entirely.

Also, I'm working on the writing system which has a sort of complex evolution. The initial pictographs are written by ogres on cave walls (both painted and carved) but it's adapted into a phonetic alphabet by the goblins they trade with on the basis of ogrish pictograms. This alphabet will at first encode consonant-vowel pairs as well as stop consonants

Just fwiw, a script that encodes syllables, rather than phonemes, is a syllabary, not an alphabet.


Anyway, this is a good start, and you've clearly put some thought into it, and I hope I haven't discouraged you at all.

[Salmoneus]

OK, so, reassuringly, I see I agree with everything VaptuantaDoi said. Except that I would distinguish between things that are unnatural for humans and things that might be unnatural in any language, and because this is explicitly a language for a non-human species I'm OK with some things they pick up on.

I definitely think that the lack of /s/, for instance, could be perfectly fine for non-humans: different tongues will work differently. The grammatical stuff (lack of 1st person pronouns, tense changing word order) is going to be more controversial. [although I actually think the 'tense changing word order' thing is actually naturalistic even for humans, in theory... it just requires a complicated backstory!].


Philosophically, I think the lack of 1st person pronouns is defensible. Essentially, what they're doing is having each speaker have their own unique 1st person pronoun, and that doesn't seem implausible to me: proper names of speakers are one of the most fundamental parts of language. [Bottlenose dolphins introduce themselves to one another by saying their own name when they meet another dolphin, for instance. Fun fact: dolphins have both forenames and surnames, and their surnames are matrilineal. That is, each dolphin's name is a part of their mother's name plus a novel element of their own. I'm not sure whether the matrilineal names are family names though (passed down through generations) or just matronymics (new in each generation).]

Importantly, while these pronouns LOOK like normal names, they aren't. Because if Bob says he did something, Bob incorporates his 'name' as a pronominal affix, whereas if Bob says some other guy named Bob did something, the name isn't incorporated. This makes the 1st person reference unambiguous and seems to fulfill all the functions of a first person pronoun. Likewise, even if you've never met Bob you're not confused when Bob said Bob did something, because the incorporation of the name as an affix both tells you that this guy did it AND tells you that this guy is called Bob. I actually think that works very well! Even though it is, as Vaptu said, absolutely NOT something that humans have ever or almost certainly would ever do in a human language.

----------

Just to answer a couple of your questions (hopefullly not stepping on Vaptu's toes...)

- it's not uncommon to restrict what consonants can be adjacent to what other consonants. These restrictions arise either because certain combinations are changed, or because certain phonemes can only arise in certain contexts in the first place. For instance, I think it's perfectly unremarkable to restrict dorsal consonants (velars and back) from preceding /i/, because these wo sounds are at opposite ends of the mouth, and there's therefore a natural tendency to have one drag the other to the middle. Most likely, all instances of /ki/ have simply become /k@/. That's fine. It's just that some of your restrictions don't have such obvious physical motivations. None of them look "wow, that woudl never happen!", they just look "huh... why on earth is that?"

[I'd also bear in mind: with a very small inventory having too many phonotactic restrictions can reduce your number of allowed syllables rather dangerously, leading to either longer words or a lot of homophones!]

- WHY is it bizarre for human languages to have /x/ and /h/ but no /s/? Fundamentally, for two reasons. Firstly, because when you have few sounds in a language you usually end up with soudns that sound very different from one another; /s/ is one of the most distinctive sounds possible, so is very common, whereas /h/ and /x/ sound extremely similar to one another (huge numbers of languages, including Old English, have [h] and [x] as allophones). And also quite similar to /k/ and /g/ and /w/. So it's weird to, as it were, fill up this already murky soundspace, while leaving the most distinctive part of soundspace empty. HOWEVER: when transcribing non-human languages, our symbols are approximations. Maybe some feature of ogre throats not evident from human IPA makes these sounds very acoustically distinctive to them.

And secondly, because some soudns are just plain easier than others to say, and those sounds tend to get used more. For humans, the easiest sounds are coronals - sounds made with the front of the tongue near the front of the mouth. This is basically because human foretongues are very nimble. Dorsal sounds - made with the body of the tongue - are much more difficult for us, because the back of our tongue isn't as nimble. It's surprisingly nimble, and there are languages with LOTS of dorsal sounds. But in general languages tend to have more coronal sounds than dorsal ones and virtually none have more dorsals than coronals (if you don't coutn secondary articulations). As a result, it's a general rule that if you only have one of... well, almost anything... it's likely to be coronal. If you have one nasal, it's /n/; if you have one lateral, it's /l/, and so on. If you have one fricative, it's /s/.

But again, it's possible that ogre tongues just work differently. Maybe they find coronal sounds difficult and dorsal sounds easy!


[tangent: I've also been toying with ogres as an intelligent 'first race in history', fwiw. In my case it's probably subliminal D&D influence!]

[BarkMiner]

Hi.
OK, that sounds reasonable. I would express some caution on the 'should expess some concepts relevant to the speakers more easily than English or French' aspect, as this is hard to really make work in practice. Grammatically, some languages make some sorts of propositions easier to express than others, but in too broad and content-neutral a way to really match cultural concerns; lexically, of course, a language will have more words for more culturally-significant things, but that's fairly peripheral to the actual structure of the language.

I'd also say that "at this stage I would like to invest minimum effort" is an extremely familiar sentiment around these parts... usually followed by years of effort! [everyone's big project always seems to be 'the thing I thought I'd spend very little time on before moving on to what I thought was the big project'...]

Thanks for the heads up! And yeah, you're totally right, this part of the project is already waaaay bigger than I meant it to be and it feels like I'm barely getting started.

I can't read it, but I think that's a dental mark under /t/ and /d/? That's fine, but bear in mind you don't actually need to include that - the general principle is that you only need to include information that disambiguates from something else (eg if you have both dental and alveolar stops, then you need to mark one pair as explicitly dental). Otherwise it's normal to just use the simplest valid representation - in this case /t/ can be used for either alveolars or dentals. Having said that, I've no actual problem with you including the dental marks here rather than saying in words 'the coronal stops are alveolar'. It just may be easier for you not to have to type the marks every single time!

Yup, dental marker, and I'll keep that in mind for the future.

- it's extremely unusual for a human language to have multiple fricatives and no /s/. I also think it may be unusual to have /l/ without /r/ (if a language only has one of these two, it's common for the phoneme to be realised in either way depending on context). However, since this language is spoken by ogres, I have no problems with this, maybe ogre tongues/mouths are just a bit different from ours. I note in fact that they don't have any non-dorsal fricatives or approximants other than /l/, which is sort of a stop anyway (the centre of the tongue touches but the edges don't), and this seems a plausible limitation for a species that perhaps has a larger and less agile tongue than humans do.

- The vowel inventory is a little unusual; usually the three vowels would be /i a u/, or occasionally (mostly in North America) /i a o/ or /e a o/. It's odd not to have a fully back and rounded vowel. But, it probably does happen somewhere in North America, and in any case, these aren't humans, so OK. The one slightly odd thing to me is that you do have /w/, so you do have lip rounding, so it's odd that they don't have any rounded vowels. But not impossible. However, if it were me I'd make "no lip rounding for ogres" a thing and replace /w/ with... whatever the symbol for a velar, uvular or pharyngeal approximant is.

These explanations are awesome! Adds a layer of depth that aligns aesthetically with what I was going for and fixes some of the jank. Just taking these suggestions wholesale (with maybe "ʁ̞" replacing the "w").

Regarding symbols, again, you can just call your low vowel /a/, you don't have to use the specific awkward symbol for a back low vowel every time. Since in this language the two soudns don't contrast (and I'd be extremely surprised if your /A/ (that's another way of writing your vowel, in an encoding system called 'x-sampa', which is easier that copy-pasting IPA) isn't sometimes realised as [a], particularly given there are only three vowels - fewer phonemes usually means more allophones).

Good to know. Is there a default for encoding "i" in English since we can be so weird about that sound?

Which actually brings me to an important question: do you understand yet the words phone, phoneme and allophone? If not, you should learn about them, since they make things a lot easier.

I did not fully understand this and the rundown was super helpful, thanks!

This isn't impossible, but I do wonder why these rules exist. Saying dorsals can't be followed by /i/ makes sense, as they're at opposite ends of the mouth - presumably at some point in the past /ki/ became /k@/ (that's x-sampa for schwa), or maybe /ti/ or something. But I don't know why you can't have nasals followed by /i/. In particular, why you can have /pi/ and /bi/ but not /mi/. [in theory it's also weird to not allow /li/ as a random rule, but in practice /l/ is often secondarily velarised/uvularised/pharyngealised, so it would fit with the other dorsals in that regard despite being primarily coronal itself]. Similarly, why is /p@/ not allowed!?

Honestly there's no good reason, it was just kind of how things ended up looking after I coined initial words. This can change.

On the 'unfamiliar' - do you mean that the item is completely unknown, as in they've never heard anyone mention it before, or that it's unknown in the context of the sentence, as in they don't know which item it's currently referring to?

So this is part of a broader project to subvert fantasy tropes. I mentioned this in a reply earlier in the thread but here's the full detail on this aspect of the language:

I really really dislike how in popular fantasy orcs, ogres, goblins, what have you are characterized as "stupid" using lazy dialect where they for no apparent reason seem to be unable to comprehend pronouns, refer to themselves in the 3rd person, etc... The goal here was to bake an avoidance of pronouns into their mother languages in a way that makes those linguistic quirks make sense, but if you bother to try to understand it instead of stupidity you find a rich set of cultural norms that differ from your own.

The distinction between the familiar and the unfamiliar was supposed to be a part of that conceptualization, where heavy linguistic distinction is made around whether the speaker/the listener are familiar with who/what they're talking about. When to use familiar vs. unfamiliar is a bit contextual in how I'm thinking about it now. When referring to an inanimate object it might be closer to a definite vs. indefinite article, while when referring to a person it might be closer to formal vs. informal (tu/vous in French, but applicable also to the 3rd person).

At this stage in the language nouns are used as adjectives.

I'm not sure what you mean by this. Nouns are nouns and adjectives are adjectives. Do adjectives look exactly like nouns? If so, how do you tell the difference?

This is probably me just being a noob at conlanging -- I had understood that adjectives tend to develop from other aspects of a sentence at first?

What do you mean by 'pronounced twice'?

Also probably a dumb thing to note. In English we expect a sequence of two identical consonants to generally not voice the second one (e.g. generally is not pronounced with two ls). So I was just trying to clarify that when I write "bb" I mean "bb".

The final stop kk is developed to clarify the ending of an entire idea.

This doesn't happen in real language: things aren't just 'developed' to 'clarify' things. They evolve naturally from things that already exist.

The idea was that this was a reduplication of -k. So -k is already used to indicate a stop, and eventually to clarify speakers start using -kk to indicate a more final stop. Does that make sense or still weird?

This all makes it sound like someone just sat aroun and invented things to say!

But also: the whole point of having a proto-language is not having to just invent things in the daughter languages. I'm not sure why you're presenting a daughter language here where features DON'T appear to have evolved from the proto-language. I mean, why not jsut make thsi daughter language the proto-language? [if all subsequent languages evolve from this daughter language then this IS the proto-language, technically!]

My initial approach was to start with something super simple and basic and to iterate on it to get it to a point where it feels as though it's been spoken for some amount of time. The current version isn't really a daughter language that will be presented anywhere, but I'm trying to incorporate it into a history. Makes sense that a lot of these things should be bumped back up to the beginning or cut though as they seem to stem from nowhere.

Anyway, this is a good start, and you've clearly put some thought into it, and I hope I haven't discouraged you at all.

Nope, not discouraged one bit! I was feeling a little lost and this feedback has been extremely helpful.

[BarkMiner]

[tangent: I've also been toying with ogres as an intelligent 'first race in history', fwiw. In my case it's probably subliminal D&D influence!]

Once again thank you thank you for all the feedback, the further notes are also really helpful to clarify. And yeah in my case it's explicit D&D influence. I've always really loved the concept of Ogre mages and had a personal love of Orcs. Very broadly this project is a sort of post-apocalyptic fantasy ttrpg setting centered around both having a lot of that comfy classic fantasy feel but subverting or exploring in more detail a lot of the tropes of the genre that are often sort of copy-pasted. I usually shy away from writing really trope-y stuff but I've been pretty inspired on this one and feel like it's coming together as both a very unique setting and one that will feel familiar and cozy to fans of LotR or older editions of D&D, as well as breathing some life into some tropes and ideas that have gone a bit stale.

Further aside -- The notes on what might make sense for ogres but not humans are super helpful, and will be great for understanding how loan words and the like are going to work as this progresses.

[VaptuantaDoi]

- Having /x h/ as the only fricatives is unnatural; I'd expect at least /s/ to balance it out (although having said that, I'm sure there's some natlang counterexamples; apparently Abipon is claimed to have just /x ħ/. Even so for this inventory it makes more sense to add /s/, or instead drop one of /x h/.)
- Your vowels are a bit weird, but not obscene. I'd expect them to behave like a vertical system, with realisations like [i~ɨ e~ə~o a~ɑ].

Can you help me to understand why this is, to avoid future mistakes like it in the future?

I'll do my best, but bear in mind phonology is a complicated field, and there's an exception to every rule. It's something you kinda have to just get a feel for.

With regards the fricative inventory, it's generally true that the phonemes in an inventory are spread as far apart in the phonological space as possible, because this enables maximal distinctiveness, so it's easier to tell minimal pairs apart. Having just /x h/ as fricatives goes against this; they're both quite far back in the mouth and they're acoustically similar; to put it simply they don't sound very different from each other. This means that speakers might struggle to distinguish them, which could lead to either a merger of the two phonemes, or a divergence, say fronting (x > ç) or voicing (x > ɣ).
Having two fricatives in the back of the mouth space also leaves a big gap for everything between bilabial and palatal fricatives. This kind of gap tends to lead other phonemes to migrate into this space, either by allophony or sound change. This is especially relevant for gaps where phonemes are easy to pronounce (as /s/ is) and easy to distinguish (as /s/ also is). Speakers are naturally inclined to use this avenue that enable easier production and reception of speech.
Having said all this, languages sometimes just don't do this for whatever reason. Most Australian aboriginal languages don't have any fricatives at all, and no-one really knows why, but still it's been stable for millenia as far as anyone can tell. So basically don't feel obliged to add /s/, especially if you can justify it biologically - sibilants require a fair bit of tongue flexibility, so you can just say that ogres' tongues are a bit thicker than humans', and you're entirely justified. You could also have [​s​] as an allophone; it's very common to have it as an allophone of /t/ before /i/, and this could be phonemicised later.

Moving onto the vowels, there's a similar story with the vowel space. The reason /i u e o a/ is such a common vowel system is that it makes very efficient use of the vowel space; you can have quite a bit of variability without confusion. However, your vowel system has one great thing going for it - it has two maximally distinct vowels, /i/ and /ɑ/; you can hardly get two more distinct vowels at all. Your third vowel is also somewhere in between, which means it doesn't encroach too much on the peripheral vowels (vowels at the outside edges of the vowel chart).
There's another factor at play in vowel systems, which is why I mentioned a vertical vowel system. Putting it simply, vowel systems plotted onto the vowel chart usually fall into one of three shapes (I won't go into three dimensions cos it's not relevant here):

1) triangular

Code: Select all

i     u
 e   o
   a

(e.g. Latin, Spanish, Hawai'ian, tons of others)

2) quadrangular

Code: Select all

i     u
ɛ     ɑ

(e.g. Akkadian, Malagasy, Cayapa)

3) vertical

Code: Select all

   ɨ
   ə
   ä

(e.g. the Sepik languages, some Northwest Caucasian languages)

Your inventory is closest to a vertical system, so it makes sense that it would behave like one. The thing about vertical systems is that they don't seem to fill out the vowel space very well, because they inevitably have lots of holes in them. However, the make do by treating all the vowels as basically just features of high; e.g. /ɨ/ is just treated as [+high], or in other words, /ɨ/ is just a generic symbol for "any high vowel", which is probably realised as some or all of [i ~ y ~ ɨ ~ u]. This allophony can be either in free variation, or conditioned. Free variation means that speakers will produce any of say [i ɨ u] for a given instance of /ɨ/, and then they might change it at some other point; this is the case in for example the aboriginal Australian language Arrernte. The alternative is that they are conditioned by surrounding consonants; for example, the allophone [​i​] may occur next to palatal consonants, and [​​u​] next to labial consonants; this is the case in Ubykh.
Your vowel system is slightly different, in that it's skewed diagonally downwards. That's still pretty naturalistic - high vowels tend to fill the front part of the vowel space more than the back; there's a lot more languages without /u/ than there are languages without /i/. Just expect that your vowels will expand to fill the space available to them; there's a lot of unused space in the back rounded area, so /ə/ will likely have an allophone [o] or [ʊ]. Likewise the mid and low front area is underpopulated, so I'd also expect [e] to crop up somewhere, and maybe [æ] as well. The North American language Wichita has a pretty similar vowel inventory to your conlang - /i ɛ ɑ/ - and these phonemes vary between [i ~ ɪ ~ e], [ɛ ~ æ] and [ɒ ~ a], which means that the vowel space is pretty well covered. Basically, your vowel inventory is fine, but when you develop it further consider expanding or shifting the vowel inventory to make it more normal. If you're interested in learning more about vowel systems, there's a pretty good thread on the old version of the other board which goes through this stuff in a lot more detail.

This seems like a weird set of restrictions to me, although you could probably justify it somehow.

Is it unusual to restrict what consonants and vowels can appear together at all, or is the way I have done it what is strange?

The latter. There is one way I can envision your system developing, which I'll elucidate on:

The ancestor of your conlang has the following inventory:
/p b t d k ɡ x h m n l j w/
/ə(ː) a(ː)/

Syllable structure is CCV, where the only permissible onset clusters are /pj bj tj dj kj ɡj/. The vowels /ə a/ have the following allophones:

/ə/ > [i, ə, u]
/a/ > [a, ɑ]

The allophone [​i​] is most common next to the palatal glide /j/, so /pjə/ is pronounced [pji]. Over time, the glide /j/ is lost, which makes [​i​] phonemic, but only following /p b t d k ɡ/.

This scenario explains why /i/ only occurs after plosives, but not why /ə/ can't occur following /p b/. Maybe /pə bə/ > [pɔ bɔ] > /pɑ bɑ/ or something. If I were you, I'd just drop these as rules, and retain them as just general tendencies.

[BarkMiner]

I'll do my best, but bear in mind phonology is a complicated field, and there's an exception to every rule. It's something you kinda have to just get a feel for.

...

Once again thank you so so much for the detailed explanations. Taking this and a read of some of the materials you linked into another draft!