Unnamed isolating(ish) ergative proto-language

[Abun]

Hey everybody, I’m a new conlanger (sort of, I’ve dabbled in it for some 10-odd years but never consistently enough) who has just joined this forum (thanks to the admin btw), hoping to participate in discussions as well as get (and give) some feedback

My most recent project is still unnamed as of now. It is supposed to become a proto-language for more fleshed-out languages later. I have so far only decided on the very basics of the culture speaking it, a plain-dwelling human-like people, so I have so far avoided creating vocabulary which is most likely connected to religious concepts (including such things as “fire, sun, live” ect.). Because of this (and because I’m more of a grammarian to begin with xD) the lexicon so far only consists of roughly 100 words or so.

The phonology I purposefully kept rather simple. It is not prototypical SAE (there is both aspiration and voicing contrast in the stops as well as initial [ŋ] and nasal+stop clusters) but it’s not too outlandish to the European ear:

Nasals: /mnŋ/ <mnŋ>
Plosives: /p pʰ b bʰ t tʰ d dʰ k kʰ g gʰ/ <p ph b bh t th d dh k kh g gh>
Liquids: /rl/ <rl>
Fricatives: /s h/ <s h>
Vowels: /a aː e eː i iː o oː u uː/ <a ā e ē i ī o ō u ū>
Phonotactics: (s/N) (P) (s/L) V ((N) (P)/L) (s)

sPs clusters are illegal. Also NP clusters have to be in the same POA. I haven’t spent much thought on allophony seeing as it’s a proto-language anyways. Some consonant clusters will of course cause assimilations to be more easy to pronounce, but for the proto-language I kept it in the non-assimilated form and will work out different patterns for daughter languages later. I do know /s/ will be voiced to [z] between voiced phonemes but that’s all for the moment.

As far as grammar is concerned, I started out intending it to be agglutinating. But I guess I overdid it a little with trying to give each of the affixes an etymology, so I ended up realizing I might just as well go another step back and have those affixes be real lexical morphemes. Since almost all my inflectional affixes got “lexicalized” in this process and I don’t have much in terms of derivation yet, the rest is nearly completely isolating with monosyllabic roots. So I’ll just post what I have in terms of syntax so far (only the very basics). Happy to read your feedback

1. Basic Syntax
The most basic syntactical order is SVO but due to the ergativic nature of xxx, this sentence structure in the narrowest sense can only be found in transitive sentences. Since xxx is a largely isolating language, the cases are not marked with affixes but are inferred via the word order. This means that the component in absolutive (the patient (P)) always comes after the verb, while the component in ergative (the agent (A)) always comes before it. So the sentence order is better described as “AVP”.

For transitive sentences this results in SVO order:

• (1) Tākh hapikar ŋgīn ī.
  father VOL-see-FIN mother PROG
  The father is looking at the mother.

• (2) Nākh khū haphraksār gemep tho.
  man that.one VOL-die-CAUS-FIN gemep PERF
  That man has killed a gemep. (gemep being a kind of large antelope-like creature used as cattle)

Note that the verbal root has to be augmented with the suffix -ar to mark it as a finite form.

For most intransitive verbs, the agent is absent and therefore the order is VP:

• (3) Ghēntar lat.
  flow-FIN water
  The water flows.

• (4) Loŋar pirs ī.
  sleep-FIN child PROG
  The child is sleeping.

• (5) Phōtar ndap.
  grow-FIN tree
  The tree grows.

This structure is also used for stative verbs (see below) and nouns which can be verbalized and treated like an intransitive verbal root. So “X is Y” is expressed as “X Ys” (or in the correct order: “Ys X”):

• (6) Thontar bōn.
  cold-FIN wind
  The wind is cold.

• (7) Ntūar ŋkhūn.
  old-FIN woman
  The woman is old.

• (8) Guntūar ŋgin.
  elder-FIN mother
  Mother is an elder.

However, there are some intransitive verbs (most importantly verbs of motion) where the only argument is indeed not the patient but actively performing the action. In these cases the argument is percieved as an agent and therefore put in front of the verb:

• (9) Golok nī hasālar tho.
  NOMZ-hunt big.group VOL-leave-FIN PERF
  A large group of hunters have left.

• (10) Gatad hokhlant pikī haphātar tho.
  1s lake look-AP VOL-go-FIN PERF
  I went to the lake. (Not happy with the way the allative postposition is formed so that’s likely to be revised later)

Of the three main arguments, only the verb is obligatory; the other other two can be omitted. If only the patient is omitted, it is implied to be the current topic of speech:

• (11) Tākh hapikar ī.
  father VOL-see-FIN PROG
  The father is looking [at him/her/it].

• (12) Nākh khū haphraksār tho.
  man that.one VOL-die-CAUS-FIN gemep PERF
  That man has killed [him/her/it].

If the agent of a transitive verb is omitted, the meaning can be compared to passive clauses in English. Depending on the context, it can be implied that the current topic of speech is the agent but this need not be the case:

• (13) Hapikar ŋgīn ī.
  VOL-see-FIN mother PROG
  The mother is being looked at (by someone unknown or the current topic).

• (14) Haphraksār gemep tho.
  VOL-die-CAUS-FIN gemep PERF
  The gemep was killed (by someone unknown or the current topic).

If agent and the patient are both omitted, the current topic of speech is understood to be the patient while the agent is not specified:

• (15) Hapikar ī.
  VOL-see-FIN PROG
  [She] is being looked at.

• (16) Haphraksār tho.
  VOL-die-CAUS-FIN PERF
  [It] was killed.

2. Verbal phrases
2.1 Dynamic and stative predicates
xxx distinguishes dynamic and stative predicates. The former describe actions and other dynamic processes, the latter conditions and states. The question of which type a particular predicate belongs to is decided by what kind of word is used as its head.

All stative verbs are intransitive and the large majority would be translated to English as adjectives. However there are a few stative verbs which would be translated as transitive verbs although they are intransitive in xxx. Ŋim ‘to know’ for example describes a state instead of a process, and indeed it is an intransitive stative verb in xxx which would be more closely rendered as ‘to be known’. “x knows y” is translated to xxx as “y is known to x”. Another important class of stative predicates is nominal roots used as verbs.

Dynamic predicates on the other hand always have a verbal head, which can be either transitive (e.g. lok “to hunt”) or intransitive (nas “to eat”).
The differentiation between the two is highly important for which aspects a verb can take (the perfective aspect for example is incompatible with stative verbs).

2.2 Aspects
Verbs in xxx do not conjugate by tense, person or numerus. However there are four different aspects: neutral, perfective, progressive and mutative. These are indicated with particles which most commonly occur at the end of the sentence but are also used as subordinating particles as well (see below). The function of the different aspects will in the following be seperately discussed in detail.

2.2.1 The gnomic aspect
The gnomic aspect is used to express general truths. It was probably originally indicated by the sentence final de, but it seems that even before the split of the known descendant languages this sentence final had fallen out of use and the gnomic aspect was marked by the absence of a sentence final (the particle is still used in relative clauses however, cf. 3.1):

• (17) Thontar lat.
  cold-FIN water
  Rain is cold.

The gnomic aspect is also used to express habitual actions, i.e. actions which are repeated regularly:

• (18) Gukhū hanasar ron.
  3s VOL-eat-FIN meat
  He eats meat.

This sentence does not mean that he eats meat without ever stopping but that he regularly eats meat (in a modern sense, the sentence could be translated as “he is not a vegetarian”).

2.2.2 The perfective aspect
The perfective aspect is marked by the particle tho. It stresses the notion of the action being a complete process with a beginning and and end. For example:

• (19) Gukhū hanasar ron tho.
  3s VOL-eat-FIN meat PERF
  He has eaten meat.

Unlike in the the English perfect tense, the completion of the action need not have an effect on the present. For example, the above sentence in English implies some influence of the meat-eating on the present (maybe "he" got sick because the meat was spoiled). The xxx sentence on the other hand merely expresses that the meat-eating is done at the time that is spoken about (this point of time may also be in the past or future).

Note also that because the perfective aspect considers the verb as a process with a beginning and an end, it can never be used with stative predicates.

2.2.3 The progressive aspect
The progressive aspect is expressed with the particle ī. It expresses that the action is in progress at the moment we are talking about. It is therefore for the most part equivalent to the English progressive forms although the xxx progressive aspect (like all aspects) does not differentiate whether the action happens in the past, present or future.

• (20) Gukhū hanasar ron ī.
  3s VOL-eat-FIN meat PROG
  He is eating meat.

The progressive aspect is also used with stative verbs to express states that are true at the moment but not generally (in that case one would use the gnomic aspect):

• (21) Thontar lat ī.
  cold-FIN water PROG
  The water is cold (this water right now, not generally).

2.2.4 The mutative aspect
The mutative aspect is marked by the particle ŋen and indicates a change in situation. It expresses the notion that whatever the verb describes is a new development:

• (22) Gukhū hanasar ron ŋen.
  3s VOL-eat-FIN meat MUT
  He is eating meat now (he wasn’t before).

The use of the mutative aspect in some contexts overlaps slightly with the perfective. Compare for example:

• (23) Gatad ghēt pikī haphātar tho.
  1s lake look-AP VOL-go-FIN PERF
  I went to the lake.

• (24) Gatad ghēt pikī haphātar ŋen.
  1s lake look-AP VOL-go-FIN MUT
  I have gone to the lake.

The first sentence stresses that I went to the lake at all. The second sentence on the other hand focuses that I was somewhere else before and then went to the lake instead.

Also, the mutative is frequently used with stative predicates (in contrast to the perfective which can only be used with dynamic ones):

• (25) Thontar lat ŋen.
  cold-FIN water MUT
  The water has become cold.

• (26) Gatad sin ŋimar ŋen.
  1s for known-FIN MUT
  It is known to me now. I know it now.

2.3 The volitional prefix ha-
Many verbs can be augmented with the volitional prefix ha-. This prefix expresses that the action is performed intentionally and actively. Conversely, if a verb lacks this prefix, the action is understood to be involuntary. Therefore, the logical subject of a volitional verb always has to appear in front of it. Notice the difference between the following sentences:

• (27) Loŋar guntū ŋen.
  sleep-FIN elder MUT
  The elder has fallen asleep (without intending to).

• (28) Guntū haloŋar ŋen.
  elder VOL-sleep-FIN MUT
  The elder has fallen asleep (intentionally).

This does not mean however that a non-volitional verb always has its logical subject appearing after it. If the verb is transitive (i.e. has a patient), its subject is still an agent and therefore appears in front of the verb:

• (29) Golok haphraksār gemep tho.
  hunter VOL-kill-FIN gemep PERF
  The hunter killed the gemep (intentionally).

• (30) Golok phraksār gemep tho.
  hunter kill-FIN gemep PERF
  The hunter killed the gemep (accidentally).

2.4 Negation
Verbal phrases are negated with the negation verb mut which is placed in front of the verb. The negation verb takes any conjugation endings the main verb might have had:

• (31) Phōtar ndap.
  grow-FIN tree
  The tree grows.
  
  (32) Mutar phōt ndap.
  not-FIN grow tree
  The tree doesn’t grow.

The plain negation verb negates facts. If the logical subject is sentient though, the use of the plain negation verb may indicate that the subject tried to perform the action but wasn’t successful. To express that the subject didn’t perform the action on purpose, one augments the negation verb with the volitional prefix ha-. In this case the main verb cannot take the volitional prefix as well:

• (33) Golok haphraksār gemep tho.
  hunter VOL-die-CAUS-FIN gemep PERF
  The hunter killed the gemep.
  
  (34) Golok mutar haphraksa gemep tho.
  hunter not-FIN VOL-die-CAUS gemep PERF
  The hunter didn’t kill the gemep (he tried to but didn’t succeed).
  (35) Golok hamutar phraksa gemep tho.
  hunter VOL-not-FIN die-CAUS gemep PERF
  The hunter didn’t kill the gemep (he didn’t want to).

3. Nominal phrases
Complex nominal phrases are comprised of a noun and one or several attributes which modify it. The attribute can consist of a verbal (adjectival) clause (relative clause), a plain noun or verb (adjective), a possessor or a numeral. Since all the others can be viewed as special cases of the verbal clauses, we will start from the latter.

3.1 Relative clauses
In order to transform sentence into a relative clause, one simply places the entire sentence in front of the noun one wants it to modify. The aspect particle here fulfills a subordinating function besides just marking the aspect. Inside the clause, the modified noun is replaced with the relative pronoun es:

• (36) Nākh khū halokar gemep ī.
  man that.one VOL-hunt-FIN gemep PROG
  That man is hunting gemep.
  
  (37) es halokar gemep ī nākh
  ES VOL-hunt-FIN gemep PROG man
  the man who is hunting gemep

• (38) Loŋar pirs ŋen.
  sleep-FIN child MUT
  The child has fallen asleep.
  
  (39) loŋar es ŋen pirs
  sleep-FIN ES MUT child
  the child which has fallen asleep

If the modified noun is the patient or only argument of the verb, the relative pronoun can be omitted:

• (40) loŋar ŋen pirs
  sleep-FIN MUT child
  the child who has fallen asleep

• (41) gatad hanasar tho ron
  1s VOL-eat-FIN PERF meat
  the meat which I ate

It can also be omitted if the modified noun is the agent and the patient is explicitly stated:

• (42) Haphraksār gemep tho nākh
  VOL-die-CAUS-FIN gemep PERF man
  the man who killed the gemep

In all other cases, the use of the relative pronoun is obligatory.

3.2 Possession
TD

3.2 Numerals
TD

4 Adverbial clauses
TD

5 Questions
TD

[2+3 clusivity]

Great first showing; welcome to the board.

If agent and the patient are both omitted, the current topic of speech is understood to be the patient while the agent is not specified:(15) Hapikar ī.VOL-see-FIN PROG[She] is being looked at.(16) Haphraksār tho.VOL-die-CAUS-FIN PERF[It] was killed.

Cool.

2.2.4 The mutative aspect The mutative aspect is marked by the particle ŋen and indicates a change in situation. It expresses the notion that whatever the verb describes is a new development:(22) Gukhū hanasar ron ŋen.3s VOL-eat-FIN meat MUTHe is eating meat now (he wasn’t before).The use of the mutative aspect in some contexts overlaps slightly with the perfective. Compare for example:(23) Gatad ghēt pikī haphātar tho.1s lake look-AP VOL-go-FIN PERFI went to the lake.(24) Gatad ghēt pikī haphātar ŋen.1s lake look-AP VOL-go-FIN MUTI have gone to the lake.The first sentence stresses that I went to the lake at all. The second sentence on the other hand focuses that I was somewhere else before and then went to the lake instead.

This seems to me to be a fused recent past + perfective, perhaps why you are getting an overlap. CF a lot of European languages where tense and aspect are difficult to separate.

[Abun]

Great first showing; welcome to the board.

Thanks

This seems to me to be a fused recent past + perfective, perhaps why you are getting an overlap. CF a lot of European languages where tense and aspect are difficult to separate.

Recent past + perfective yes, but relative to the point of reference, not the time of speaking, so the sentences could just as well refer to past or future developments. However it is probably closer to the use of perfect tenses in English (not French though if I'm corect) than a true perfective. Actually the inspiration comes from the use of sentence-final le (了) in Mandarin which is used in this way and differs in meaning from the homonymic verbal suffix which indicates perfective aspect. There is of course some overlap if the sentence happens to end in a verb (also the particles have the same etymology and indeed the two aspects aren't differentiated in at least some of the non-standardized Chinese languages).
In fact, I was at first thinking about scrapping the distinction but when I thought about it more carefully, I realized that they have quite different nuances and would be applied in different contexts. If I'm telling a story for example, the perfective would be used to establish a setting: "Yesterday I went to the lake and you won't believe what I saw there!" The "mutative" (not happy about the name but I couldn't come up with anything better, yet xD) on the other hand marks a turn: "I got up in the morning, ate some breakfast and then I went to the lake to have a swim."
In any case, I'm sure to have some of the descendants merge those two

[Curlyjimsam]

Point of interest: there are, as far as I know, no real-world languages which demonstrate ergativity purely in terms of word order without case or agreement marking. This doesn't necessarily mean you can't do this in your language - I do a similar thing to you in my main conlang - but some conlangers like to be as realistic as possible, so if you're one of them, this is something you might like to be aware of.

[Abun]

Point of interest: there are, as far as I know, no real-world languages which demonstrate ergativity purely in terms of word order without case or agreement marking. This doesn't necessarily mean you can't do this in your language - I do a similar thing to you in my main conlang - but some conlangers like to be as realistic as possible, so if you're one of them, this is something you might like to be aware of.

Yes, I'm aware that this is at least highly unusual in natlangs, but it doesn't bother me much to be honest. It wasn't planned like that from the beginning (as mentioned I started out planning it as an agglutinative language with case markers), and at least in one of the two main descendant families it is going to develop towards the agglutinative direction and quite probably become inflecting even later. I have another branch in planning which is going to go the opposite direction and lose/lexicalize at least most of the small affix-system present in this language but I'm not sure yet whether the sentence order is going to stay the way it is. More likely I'm going to change it into a more "normal" nominative-accusative SVO system later (unless I get the feeling I have to set it apart from Chinese which I'm already heavily stealing from xD)