Balog: a strictly monovalent conlang

Imralu · Post by **Imralu** » Sat Jun 26, 2021 11:07 am

Hello. So, I'm on-again-off-again working on a language that has some pretty weird features and I'm really curious if the strict monovalency thing/lack of syntactic transitivity has been anadewed. (Anadew is a transitive verb now, OK?) I haven't done a thread for one of my languages for years and years, and I basically just want to talk about a couple of features, but for anyone interested, there's a loose description of the (stupid) setting and not terribly-exciting phonology in here:

More: show

Monovalency:
The equivalent of transitivity is formed by juxtaposition of clauses. Most sentences follow a somewhat ergative pattern, with many verbs with a rather passive meaning being used (be.killed, be.consumed, be.built etc.), however instead of an ergative case to introduce the agent as an argument of that verb, the agent is the subject of a separate clause clause with rather instrumental meaning that tells how the action was done.

Tsag uvvaŋ, mol essanok.

tsag: use.end.of.held.spear

u=: DEF(a)=

vaŋ: hunt

|: |

mol: be.killed.for.food

e=: DEF(d)=

sanok: be.pig

The hunter killed the pig (for food) (by stabbing it with a spear).

More: show

Kqal uvvaŋ, magaz iddauz.

kqal: chop.with.axe

u=: DEF(a)=

vaŋ: hunt

|: |

magaz: fall.over

i=: DEF(e)=

dauz: be.tree

The hunter cut the tree down (with an axe).

More: show

B'uvvaŋ, magaz iddauz wettsik.

b: use.hand

u=: DEF(a)=

vaŋ: hunt

|: |

magaz: fall.over

i=: DEF(e)=

dauz: be.tree

we=: ATTR=

tsik: be.small

The hunter pushed/pulled the small tree over (by hand).

More: show

Wež oož, b'uvvaŋ, magaz iddauz.

wež: use.eye

o=: DEF(c)=

ž: 1S

|: |

b: use.hand

u=: DEF(a)=

vaŋ: hunt

|: |

magaz: fall.over

i=: DEF(e)=

dauz: be.tree

I saw the hunter topple the tree by hand.

More: show

Nounlessness:

Another feature is one that's common to virtually all of my conlangs but AFAIK not conclusively demonstrated in any natlang: a lack of a lexical distinction between nouns and verbs. (Salishan languages come close but some content words apparently need to be marked for some syntactic positions that other content words don't need to be marked for. Lojban doesn't have a lexical distinction between verbs and nouns though.) For convenience sake, I call the content words "verbs" simply because they can be unmarked in predicates but are marked as subjects by one of a set of clitics that attach to the front of the subject phrase. (These are the things glossed as "DEF(x)=" here. The letter in brackets will be explained below under Rank.) Each verb is also its own agent noun, but because the relationship is consistent, calling it zero-derivation would essentially just needlessly double the size of the lexicon. Alternatively, you could regard all the content words as nouns and regard the predicates as containing a zero copula, which is indicated by the absence of the subject marker, but that's getting a bit convoluted. It's the same thing, just a less parsimonious description and calling them verbs is simpler than going to Lojban route and calling them something like brivla.

This means that you can swap the elements in a clause and change topic and focus without altering any semantic information.

Magaz iddauz.

magaz: fall.over

i=: DEF(e)=

dauz: be.tree

The tree fell over.

More: show

Dauz immagaz.

dauz: be.tree

i=: DEF(e)=

magaz: fall.over

be.tree DEF=fall.over
"What fell over was a tree."

More: show

Functions associated with nominal or verbal morphology are not restricted to either the subject or the predicate of a clause, nor to use in conjunction with a subset of the category of verbs based on semantic criteria. For example, TAM markers, when present, are not restricted to predicates or to use with semantically "verby" verbs. The structure of a predicate phrase and a subject phrase differs only in the presence or absence of a subject marker, which provides information about definiteness, specificity and rank.

Rank:
A rank is assigned to every subject of a clause in Balog. Rank is used to differentiate between multiple referents and the assignment of rank to a referent is not fixed but is, rather, a flexible assignment that varies according to the context and the other referents in the discourse. Rank applies not only to third person referents, but also to first and second persons. Within a discourse, as far as possible, each rank level should have no more than one referent. When there are more than five referents in a discourse, some doubling up is necessary, however this requires repetition of lexical verbs within the subjects, just as, in English, a story about two men tends to necessitate more use of their names because the referent meant by "he" or "him" is not always clear.

The five ranks are indicated by the vowels in the following order: u, a, o, e, i. As clitic prefixes, these are glossed as DEF(a), DEF(b), DEF(c), DEF(d) and DEF(e) respectively. Clitic prefixes cause gemination of the following consonant, except when attached to a verb that consists only of a single consonant, in which case, the vowel is lengthened instead. For example, an A-ranked Balog would be referred to as ubbalog, but would be addressed in the second person as uul. The number-neutral third person pronouns are, respectively, uu, aa, oo, ee, ii.

The main metric, according to which rank is assigned is: Who would win in a physical fight. Physically larger, stronger, more combat-capable individuals are always assigned higher rank in discourse than less combat-capable individuals. Inanimate objects that are incapable of effecting any consequences are thus generally assigned E-rank (marked by the vowel i-) although of course, if I say that I was injured by a burning tree that fell on me, the the tree would be ranked higher than me and the fire would be ranked higher than the tree. If it was me who started the fire, when I refer to my lighting of the fire, I would rank myself higher than the fire and then reassign ranks when talking about the consequences. E.g. something like:

use.hand o=I, come.into.being e=fire, location i=base of tree
(I started a fire at the base of the tree.)

consume a=fire, be.burnt o=tree, fall.over oo, be.injured i=I
(The fire burnt the tree and caused it to fall over and injure me.)

Without the need for the reassignment of ranks, the fire could have simply been referred to again as ee and the tree, or at least its base, as ii, but because, at that point in the story, "the tables turned", everything needed to be reassigned, necessitating the reiteration of the lexical verbs "(be.)fire" and "(be.)tree". In the second sentence, the tree was mentioned again, but within the framework of that rank assignment, the rank C (o) was unambiguously assigned to it alone, meaning it could be referred to again using oo. Until the next reassignment of rank, the tree can now be referred to with oo.

Also, note that I marked myself with i= (rank E) in the second sentence. This was not necessary because rank D (e=) was available. However, by marking myself with i=, I have also conveyed that I am pathetic (and as ineffectual as an inanimate object can be, far less effectual than the fire and the tree), a way of acknowledging that I was stupid for getting injured by a situation that I caused.

The metric of physical effectuality doesn't always apply in instances where there is a clear imbalance of political power or formal rank. If you could clearly beat someone in a fight but they could end you because of money or connections or they are higher (formally) ranked than you, you would assign them a higher rank than you. Within the hierarchy of the militaristic Balog empire, higher (formally) ranking Balog will always be referred to with higher (linguistic) ranks and non-Balog are always assigned lower ranks than Balog, but same-(formally-)ranked individuals will assign rank based on perceived physical combat ability. Where two individuals in a discourse do not agree to their rankings in each other's speech, the matter may be settled physically.

When assigning rank to a sole individual without others (yet) in the discourse a rough guess is made as to the rank the referent is likely to be assigned in the context of others. Thus, a judgement is made of someone any time they are referred to, even in the absence of other referents to be compared with. Generally, without context, the A-rank (u=) is assigned to indicate political, finantial or social power. Someone who just looks like a badass would default to the B-rank (a=). An average-Joe would default to the C-rank (o=) and a weakling to the D-rank (e=). The E-rank (i=) is by default, occupied by inanimates, but it is also frequently used as a way of talking down to people perceived as especially lowly. The context of the speaker also plays a role here. A Balog would always refer to a non-Balog by default with e= or i= as, by Balog standards, all non-Balog are essentially physically ineffectual. A non-Balog speaking Balog to another non-Balog would (at least when within earshot of Balog) always refer to a Balog with u= by default.

Inspiration for rank:

More: show

I got tired of the usual (real or grammatical) gender-based distinctions and, at the moment, I'm writing a story about a gay couple and really getting annoyed with how often I have to use their names to disambiguate. I thought it would be cool to explore a system that would give them different pronouns. I was having a whinge about this with a close friend of mine, who then joked that it would be cool to have different pronouns for doms and subs and I filed that away as an idea.

I also felt inspired by the assignment of physical spaces to referents in sign languages. In a sign language, it never matters if you're talking about two third-persons of the same gender or not - you just give them a unique location and can keep referring to them. The choice of location for each referent in sign languages is sometimes arbitrary, but it's often influenced by the referents' actual or habitual locations (e.g. pointing at someone or at their chair in their absence, through the wall towards their office or situating someone's partner or friend as closer to them in the signing space than to others.) I thought about having a completely arbitrary system that is just set at random for a discourse, but it seems likely that, in an oral language where these would not be locations but different sounds, certain assignments would easily pick up connotations.

And for an orc language, I then thought having it basically based on who could kick whose arse, felt right and fun and horrible. Like, seriously, if this were a real language and I had to learn it, I'd be like "OMFG, this is such bullshit!", but it's perfect for a horrible conworld. If I use this lang much here in threads, don't read too much into my choice of rank for various people. It's virtually impossible to use this awful system over the internet where you can't literally size people up. From memory, I have already ranked Jal higher than me and Bradrn lower than me, based simply on the fact that I remember that Jal is very tall and I am taller than average, so probably taller than Bradrn, but ... that obviously is unclear and also, well, at the moment, I'm plagued by pretty bad back pain, so I could probably be destroyed in a physical fight by most people. I just have to pick one of the vowels so ... yeah, please nobody take it personally.

Propositional:
In addition to the subject clitics u=, a=, o=, e=, i=, there is also the complementiser clitic ä=, which I gloss as DEF(p). P stands for for propositional as C for complementiser was already taken, the contrast between DEF(c) and DEF.C might get confusing, and simply using C on its own is not an option because this also changes to show definiteness and specificity. It also can't be glossed as a complementiser when used in the third person pronominal form ää as that simply translates to "it", referring to a mentioned proposition rather than a single tangible referent.

Here is an example of the difference between the definite, specific complementiser ä= (DEF(p)=) and the indefinite, non-specific complementiser ähä= (NSPC(p)=).

M'oož, nam ämmelets eel.

m: feel.emotionally

o=: DEF(c)=

ž: 1S

|: |

nam: be.liked

ä=: DEF(p)=

melets: be.happy

e=: DEF(d)

l: 2S

I'm glad you're happy.
I like that you're happy.

More: show

M'oož, nam ähämmelets eel.

m: feel.emotionally

o=: DEF(c)=

ž: 1S

|: |

nam: be.liked

ähä=: NSPC(p)=

melets: be.happy

e=: DEF(d)

l: 2S

I like the idea of you being happy.

More: show

____________________________________________________________________________________________
*Sorry about all the awkward uses of the word "status" as a verb here. I initially called this "rank" rather than status, but because there is also formal rank in the militaristic sense, I decided to change it to status, but it doesn't have the nice range of uses that "rank" has. I also keep finding myself writing rank when I mean status, so I just did a search and replace and made a mess of it and now I think I'll just call status "rank" and call actual rank "formal rank" or whatever. Urgh.
____________________________________________________________________________________________
EDIT: I fixed all mention of status to rank and changed formatting of examples and added extra syntactic sentence breakdowns hidden in "more" tags to show the parts of sentences.

Qwynegold · Post by **Qwynegold** » Sun Jun 27, 2021 1:44 am

Hmm, you seem to have had similar inspiration as Xwtek in his orclang.

I was wondering about everything marked as definite. Are there any indefinites besides that clitic ähä? (I didn't quite understand that part of the grammar btw.)

bradrn · Post by **bradrn** » Sun Jun 27, 2021 3:34 am

I don’t have much to say about this, apart from saying I can see nothing obviously wrong with this system. (Mind you, that’s probably high praise when coming from me. I can always find things wrong.)

One comment: I’d imagine you’d probably get various combinations of monovalent verbs grammaticalising into causatives, applicatives etc. In that case much of the ‘syntax’ of this language would be devoted to specifying which combinations of verbs have which idiomatic meaning. (Such things may or may not be categorisable as serial verb constructions — there would be similarities, but the label is really meant for verbs which share their arguments.)

Imralu wrote: ↑Sat Jun 26, 2021 11:07 am The main metric, according to which status is assigned is: Who would win in a physical fight. Physically larger, stronger, more combat-capable individuals are always assigned higher status in discourse than less combat-capable individuals.

What happens if people disagree on status? I can imagine people getting into linguistic conflicts where each attempts to assign the highest status to themself. (Or the reverse: they attempt to be humble by assigning highest status to the other person.)

It's virtually impossible to use this awful system over the internet where you can't literally size people up. From memory, I have already ranked Jal higher than me and Bradrn lower than me, based simply on the fact that I remember that Jal is very tall and I am taller than average, so probably taller than Bradrn, but ... that obviously is unclear and also, well, at the moment, I'm plagued by pretty bad back pain, so I could probably be destroyed in a physical fight by most people. I just have to pick one of the vowels so ... yeah, please nobody take it personally.

Yeah, dunno about jal but this sounds entirely correct for me

I’m not strong at all. (I had to transport a table downstairs today and could barely carry it even with help.)

As for the internet more generally, but might it become based on prestige? Highest status is the admin, then the moderators, then the users in decreasing order of experience.

Imralu · Post by **Imralu** » Sun Jun 27, 2021 5:04 am

Qwynegold wrote: ↑Sun Jun 27, 2021 1:44 amHmm, you seem to have had similar inspiration as Xwtek in his orclang.

Hi! Hmm, not sure what you mean as I had a look at his Rkou and I honestly can't see any similarities other than the fact that they are both spoken by orcs. Or is that all you meant?

Qwynegold wrote: ↑Sun Jun 27, 2021 1:44 amI was wondering about everything marked as definite. Are there any indefinites besides that clitic ähä? (I didn't quite understand that part of the grammar btw.)

Ah, yes. I just didn't really get into that. Here's the full table of all of the subject markers. Completely regular, so perhaps a bit boring to some, but I usually don't aim for a huge amount of naturalism in my conlangs and many of them are conlangs in-world too.

Code: Select all

Definiteness:   [DEF___][INDEF_________]
Specificity:    [SPEC__________][NSPC__]
Glossed as:     DEF     SPEC    NSPC
       Rank A:  u=      uqu=    uhu=
       Rank B:  a=      aqa=    aha=
       Rank C:  o=      oqo=    oho=
       Rank D:  e=      eqe=    ehe=
       Rank E:  i=      iqi=    ihi=
Propositional:  ä=      äqä=    ähä=

Here are a couple of examples showing specific and non-specific indefinites:

Wež oož, b'uquvvaŋ, magaz iqiddauz.

wež: use.eye

o=: DEF(c)=

ž: 1S

|: |

b: use.hand

uqu=: SPEC(a)=

vaŋ: hunt

|: |

magaz: fall.over

iqi=: SPEC(e)=

dauz: be.tree

I saw a hunter topple a tree by hand.

The specific indefinite subject markers indicate that the speaker has a specific hunter and a specific tree in mind but, unlike the definite, doesn't expect the listener to know which ones. Once they've been introduced and assigned a rank like this, however, they can then be referred to again with the definite u= and i= respectively, and since the rank makes it clear what's what, the verbs vaŋ and dauz don't need to appear again, meaning you are only likely to need the definite pronominal forms uu (he = the hunter) and ii (it = the tree) from here on, unless a reassignment of ranks occurs.

Kq'ähäwwež oož, b'uhuvvaŋ, magaz ihiddauz.

kq: NEG

ähä=: NSPC(p)=

wež: use.eye

o=: DEF(c)=

ž: 1S

|: |

b: use.hand

u=: DEF(a)=

vaŋ: hunt

|: |

magaz: fall.over

i=: DEF(e)=

dauz: be.tree

I didn't see any hunter topple any tree by hand.
Lit.: It is not that I saw any hunter topple any tree by hand.

In this sentence, the negative forces the use of the non-specific rather than the specific, similarly to how "any" often appears in negative sentences when "some" might appear in the positive equivalent. The non-specific is also often used in sentences about wanting, looking for etc. Unlike English, a distinction is always made between

I'm looking for a book. (I have a particular book in mind.) → iqi=
I'm looking for a book. (I don't have a particular book in mind.) → ihi=

With iqi=, the sentence wouldn't necessarily mean "any old book will do", as "any" often at least implies in English. It doesn't mean that there are no further conditions, just that there is no specific entity or entities intended by the designation.

In first and second persons, such as oož (I) in the examples above, logically, only the definite would really be used as the speaker and the listener both obviously know which individual is referred to. Perhaps, idiomatically, the specific indefinite oqoož could be used to acknowledge that the listener isn't expected to know the speaker (although they will generally know which individual is speaking: "you don't know me/you haven't met me before" is quite difference from "you don't know which person is currently speaking to you"), or it could also be used when speaking anonymously (anonymous letter, Catholic-style confession booth kind of thing, speaking menacingly to someone in the dark, etc.), but that would be quite marked.

Perhaps the non-specific could be used in first and second person forms kind of as the impersonal pronoun "one". I suppose the main contender for this role would be uhuuŋ (NSPC(a)=1P.INCL) indicating that a non-specific "any of us" is intended, with the highest rank used to leave it the most open ended. (The highest rank within a group is used to refer to the whole group, thus when referring to all people, rank A has to be used. When referring to all people other than the emperor, for example, the rank B would be used to leave room for the emperor in rank A).

The negation of the whole sentence above is performed by subordinating it with ähä-. Thus, at the top level of the sentence, we have the verb "kq" "be not" and the subject "ähä=[SUBORDINATED CLAUSES]". I'm not sure I can really make my explanation of the propositional subject markers clearer. The main thing I wanted to show was that different shades of meaning can be achieved with definiteness/specificity distinctions on the complementiser (propositional subject marker) itself. Maybe I can explain it better here.

Melets eel.

melets: be.happy

e=: DEF(d)

l: 2S

You're happy.
(Independent clause making an assertion)

ämmelets eel

ä=: DEF(p)=

melets: be.happy

e=: DEF(d)

l: 2S

that you're happy
(Subordinated clause used as the subject of another clause: there is an imbedded assertion in here that you are, in fact, happy (or were or will be etc., as there are no TAM markers here) and the listener could dispute that)

äqämmelets eel

äqä=: SPEC(p)=

melets: be.happy

e=: DEF(d)

l: 2S

a specific instance of you being happy
(Subordinated clause used as the subject of another clause: there is no assertion imbedded in here other than the assertion that there has been or will be at least one instance in which you are happy. The listener could only really dispute it if their entire existence never contains a single moment of happiness.)

ähämmelets eel

ähä=: NSPC(p)=

melets: be.happy

e=: DEF(d)

l: 2S

the general idea of you being happy
(Subordinated clause used as the subject of another clause: there is no assertion in here at all - this cannot be contradicted as it is only talking about the idea, which could be imaginary.)

In my example sentences in the last post, ämmelets eel and ähämmelets eel were given as the subject of a clause containing the verb nam meaning "appeal (to sb)", "be liked (by sb)". The "liker" with nam is in a clause with the verb m "feel emotionally". For example, you could say

M'oož, nam eel.

m: feel.emotionally

o=: DEF(c)=

ž: 1S

|: |

nam: be.liked

e=: DEF(d)

l: 2S

I like you.

There are two clauses there. The second one simply asserts that you are liked by someone. The first indicates who the liker is by asserting that I feel something emotionally. Because of the nature of juxtaposition, a true pedant could say that you could potentially interpret the sentence as meaning that I feel something completely unspecified, not necessarily affection, and, unrelatedly, that you are liked by someone completely unspecified who is not necessarily me. However, the pragmatic rules of Balog mean that these clauses, when spoken with an intonation contour that indicates that they belong to the same sentence, are understood together and that it is clear that what I feel is affection and that the person who likes you is me. In any case, the subject eel, which equates to the object in the English translation (but to the subject in a Spanish translation, btw., "Me gustas.") can be replaced by a clause introduced with ä=, äqä=, ähä=.

For the sake of extra clarity, I've been writing commas between clauses when possible (even though it makes it feel very stilted), but it isn't really possible here because the subordinate clause introduced by ä(qä/hä)= begins within the word.

Imralu · Post by **Imralu** » Sun Jun 27, 2021 6:07 am

bradrn wrote: ↑Sun Jun 27, 2021 3:34 am I don’t have much to say about this, apart from saying I can see nothing obviously wrong with this system. (Mind you, that’s probably high praise when coming from me. I can always find things wrong.)

Ooh, thanks!

bradrn wrote: ↑Sun Jun 27, 2021 3:34 amOne comment: I’d imagine you’d probably get various combinations of monovalent verbs grammaticalising into causatives, applicatives etc. In that case much of the ‘syntax’ of this language would be devoted to specifying which combinations of verbs have which idiomatic meaning. (Such things may or may not be categorisable as serial verb constructions — there would be similarities, but the label is really meant for verbs which share their arguments.)

Yes! That's exactly what I've had in mind. I'm finding it a bit hard to find a way to mark the structure of phrases like this in my dictionary in a way that is unequivocal and easy to use.

I feel like the verb b "use.hand" might get a bit overused and become used simply to introduce agents with emphasis, a bit like "I did it myself" in English, like, by my hand. There is also the pretty generic, insipid g "(be).cause", "be.reason" and that could be used to obscure how something was done or could perhaps indicate deliberate volition as opposed to accidental actions. Perhaps it could have a different original meaning that still arises from time to time and it merely got weakened to a kind of neutral cause. (Hypothetically, if g simply took over the whole system and got used all the time, it could simply be described as an ergative prefix, but as it currently stands, there is essentially an open class of ergative markers that convey all kinds of additional information.)

I can imagine a verb like "stumble" being used to indicate any accidental action, like "I stumbled, the bird flew away" = "I accidentally released the bird." Potentially, that could weaken the verb "stumble" so much that to literally talk about stumbling you'd have to say something like "I use foot, I stumble". I've also thought of things like euphemisms for killing someone in a blind rage (because typecasting orcs is fun: #JustOrcyThings) like "I got lost, he died", but I have already divided up the semantic space of dying and being killed into some very specific stuff , including veb "be.killed.in.anger", "be.killed.for.revenge" ... of course, that could be used with "I got lost", but it's not exactly very euphemistic. Perhaps they wouldn't often bother with a euphemism here though, but potentially "I got lost, he died in battle" could work. Anyway, yeah, there's a huge potential for a lot of odd and interesting pairings of clauses that could create some pretty impenetrable idiomatic meanings.

bradrn wrote: ↑Sun Jun 27, 2021 3:34 amWhat happens if people disagree on status? I can imagine people getting into linguistic conflicts where each attempts to assign the highest status to themself. (Or the reverse: they attempt to be humble by assigning highest status to the other person.)

As I said before, it is settled physically, however I imagine that it is seldom settled by an all-out brawl but by codified contests such as arm-wrestling or actual wrestling (with rules) or something more culturally interesting. There's a Korean game whose name I don't know (it was always initiated with me by a fast double clap and then "assuming the pose"), but which I'm incidentally quite good at, that involves planting your feet next to each other and trying to knock your opponent off balance by only touching their hands. Whoever has to move their feet loses and besides directly pushing on your opponents hands, there is a lot of bluffing, dodging their hands, unexpectedly resisting or bending with their push which can cause them to lose balance. (Does anyone know the name?) That's not what I have in mind exactly, but that genre of things. In any case, whoever loses has to accept and adopt the speech-ranking ... if they don't, then it may come to a brawl.

bradrn wrote: ↑Sun Jun 27, 2021 3:34 amYeah, dunno about jal but this sounds entirely correct for me I’m not strong at all. (I had to transport a table downstairs today and could barely carry it even with help.)

Well, there have been times lately when I haven't even been able to stand vertically and have to just stand at some fucking weird angle, obviously using a whole bunch of muscles in a very unstable and unsustainable way just to keep me somewhere close to vertical. Like, literally anyone could probably drop me to the floor with a light push ... unless adrenalin in case of conflict would make me suddenly overcome the stupid stuff my torso is doing.

That leads to another problem in written communication though. It's more permanent than a simple speech act. If I write something to someone while clearly able to beat them in a fight and then they read it some time later after I have an accident (or the lockdown-induced effects of inactivity on my stupid scoliotic (that can be the adjective of scoliosis, right?) torso), they might be like "Ha, so you think you're a tough guy, huh?" I don't think the average Balog writes very much, but I'd have to come up with a cultural solution for this: either it's just accepted as "You were stronger than me at the time" or, lololololol, "NOW I'M STRONGER THAN YOU SO YOU HAVE TO EDIT YOUR WRITING! ALL OF IT!" I think I'll go with the former although the latter is pretty funny.

bradrn wrote: ↑Sun Jun 27, 2021 3:34 amAs for the internet more generally, but might it become based on prestige? Highest status is the admin, then the moderators, then the users in decreasing order of experience.

Hmm, yes, except in most cases you can't see who's been around longer. Before the move of the ZBB in 2018, it would have been possible here, but now it's all been obscured. It looks like we all joined in 2018, but I joined the original site in 2006 (in May, maybe, from memory??) and I mostly can't remember who's been around longer than me and who hasn't. Like, I joined the new site a few days before Qwynegold, but I can't remember if Qwynegold is an oldbie who's been around longer than me. I don't remember with you either. People have also changed user names and because I haven't been around very consistently, a lot of names simply look familiar and that's all I can say about them.

One thing about the internet though: in nearly all cases, things can't be settled physically and there is no actual danger of ranking someone lower than you when they could actually kick your arse, so I imagine, if this language were used on something like the internet, there would just be a lot of people arrogantly ranking themselves higher than everyone else and either this would result in flame-wars or it would just be ignored and used to form judgements about who is arrogant and who isn't. Potentially, there would be a politeness standard where second person is placed above first person, always, and third persons could be placed even higher or lower depending on what you want to say about them. Maybe placing 2nd person above 1st person would even become a rule of politeness on some forums. I don't know. I think something like that would change the language. When it's spoken mostly face to face, this system seems like it could work well (although horribly, lol), but in fairly anonymous, written communication, I think a change would come about.

Raholeun · Post by **Raholeun** » Mon Jun 28, 2021 1:41 am

I am sort of fascinated after reading all this. Some brief comments:

- Your text examples only seem to contain a limited subset all of Balog phonemes. Some listed consonant sounds are not represented in the examples. Is that intentional, or do <tshq, dž> etc only occur in a certain register of the language?
- In the last post you note that there is room for some idiomatic meaning arising. I really like that idea and it might open some pathways into historical derivation. That got me thinking; rank is deferred to the prowess or strength of a person or object. But I could also imagine that the word itself always gets ranked a certain way, for instance because the word was loaned from the effeminate elves or some subservient Orc race. Then rank gets lexicalized, adding a layer of complexity.

Fun stuff.

Ares Land · Post by **Ares Land** » Mon Jun 28, 2021 8:07 am

I don't have any particular questions right now, but I like all of this a lot, especially the 'monovalence' and specific/nonspecific parts!

bradrn · Post by **bradrn** » Mon Jun 28, 2021 8:44 am

Imralu wrote: ↑Sun Jun 27, 2021 6:07 am
bradrn wrote: ↑Sun Jun 27, 2021 3:34 amOne comment: I’d imagine you’d probably get various combinations of monovalent verbs grammaticalising into causatives, applicatives etc. In that case much of the ‘syntax’ of this language would be devoted to specifying which combinations of verbs have which idiomatic meaning. (Such things may or may not be categorisable as serial verb constructions — there would be similarities, but the label is really meant for verbs which share their arguments.)
Yes! That's exactly what I've had in mind. I'm finding it a bit hard to find a way to mark the structure of phrases like this in my dictionary in a way that is unequivocal and easy to use.

I know that the Kalam dictionary simply treats idioms as subheadings under the main word, e.g.:

Pawley & Bulmer wrote: TAN- [tá·n], v. Rise above a surface; thus:
1. Climb, go up, ascend. Compare jak-. Mon kogm tawi, tan amnoŋ! Step on the knot (in the tree) and climb up! Wog agiabin, maŋlaŋ at tani, yp yneb-tek ayab, kosyam ptk opin. I was burning off the garden when the flames rose up and almost burnt me, and I ran away in fear.
2. (of hair, feathers, trees, etc.) Grow. Compare jak-, yob g-. (G) Wask matp ksen ap tanub ogok tbl, ognap wask wos ogok tbl, dad apl akl,… Young, newly grown wask trees are cut down, sometimes mature trees are cut, and (the outer bark) scraped). Jbl ognap gu nep sŋok tani mdek. Some eugenia trees grow there in the flatter parts (of the forest).

[…]

tan ap tan ap yap g-, climb up and down. (G) Mab nb okok tan aptan apyap gl npal. They climb up and down the trees all around. (G) B ogok … kmn cb klek nŋl tan aptan apyap gl pakl tek... The men find the animal’s dung and climb up following its habitual path and kill it.
ap tan-, = atan-, ascend or rise suddenly, fly up, as sparks from a fire. Guñgolp ap tanb. Sparks flew up.
ap tan jak-, rise to the top, reach the top, as water filling a container. Tin ñg ap tan jakp. The tin has filled with water.
at tan-, = atan-, rise up, rise to the top. Kob agiey, skum at tani kotp yu gp. They are burning off grass, and the rising smoke has obscured the house.
d ap tan-, bring something upwards, move something up.
d ap tan d ap yap-, bring something up and down, move back and forth.

But in your case this might be more difficult, since your idiomatic phrases won’t necessarily be contiguous. Perhaps you could mark them as ‘verb1 … verb2’ or something?

I feel like the verb d "use.hand" might get a bit overused and become used simply to introduce agents with emphasis, a bit like "I did it myself" in English, like, by my hand. There is also the pretty generic, insipid g "(be).cause", "be.reason" and that could be used to obscure how something was done or could perhaps indicate deliberate volition as opposed to accidental actions. Perhaps it could have a different original meaning that still arises from time to time and it merely got weakened to a kind of neutral cause. (Hypothetically, if g simply took over the whole system and got used all the time, it could simply be described as an ergative prefix, but as it currently stands, there is essentially an open class of ergative markers that convey all kinds of additional information.)

I can imagine a verb like "stumble" being used to indicate any accidental action, like "I stumbled, the bird flew away" = "I accidentally released the bird." Potentially, that could weaken the verb "stumble" so much that to literally talk about stumbling you'd have to say something like "I use foot, I stumble".

All this sounds reasonable. Also, I’d be interested in hearing more about the open class of ergative markers.

I've also thought of things like euphemisms for killing someone in a blind rage (because typecasting orcs is fun: #JustOrcyThings) like "I got lost, he died", but I have already divided up the semantic space of dying and being killed into some very specific stuff , including veb "be.killed.in.anger", "be.killed.for.revenge" ... of course, that could be used with "I got lost", but it's not exactly very euphemistic. Perhaps they wouldn't often bother with a euphemism here though, but potentially "I got lost, he died in battle" could work. Anyway, yeah, there's a huge potential for a lot of odd and interesting pairings of clauses that could create some pretty impenetrable idiomatic meanings.

Note also that languages tend to innovate new words for culturally-specific concepts. In this case, the behaviour you describe reminds me of Austronesian amok, the word in this case coming from Malay. If your people go into a murderous rage often enough, they’ll probably innovate a word for the concept as well.

bradrn wrote: ↑Sun Jun 27, 2021 3:34 amYeah, dunno about jal but this sounds entirely correct for me I’m not strong at all. (I had to transport a table downstairs today and could barely carry it even with help.)
Well, there have been times lately when I haven't even been able to stand vertically and have to just stand at some fucking weird angle, obviously using a whole bunch of muscles in a very unstable and unsustainable way just to keep me somewhere close to vertical. Like, literally anyone could probably drop me to the floor with a light push ... unless adrenalin in case of conflict would make me suddenly overcome the stupid stuff my torso is doing.

Oh, that sounds awful. I hope you recover soon.

That leads to another problem in written communication though. It's more permanent than a simple speech act. If I write something to someone while clearly able to beat them in a fight and then they read it some time later after I have an accident (or the lockdown-induced effects of inactivity on my stupid scoliotic (that can be the adjective of scoliosis, right?) torso), they might be like "Ha, so you think you're a tough guy, huh?" I don't think the average Balog writes very much, but I'd have to come up with a cultural solution for this: either it's just accepted as "You were stronger than me at the time" or, lololololol, "NOW I'M STRONGER THAN YOU SO YOU HAVE TO EDIT YOUR WRITING! ALL OF IT!" I think I'll go with the former although the latter is pretty funny.

I actually think it would end up as both, depending on absolute status. If Random Goon 1 wins a fight with Random Goon 2, said goon probably isn’t going to demand all references to him be rewritten (for practical reasons if nothing else). If the Great Orc King gets killed by the Next Great Orc King (because realistically, how else would orcs choose their king?), there’ll probably be trouble for anyone who doesn’t immediately rewrite their documents to praise the Next Great Orc King.

bradrn wrote: ↑Sun Jun 27, 2021 3:34 amAs for the internet more generally, but might it become based on prestige? Highest status is the admin, then the moderators, then the users in decreasing order of experience.
Hmm, yes, except in most cases you can't see who's been around longer. Before the move of the ZBB in 2018, it would have been possible here, but now it's all been obscured. It looks like we all joined in 2018, but I joined the original site in 2006 (in May, maybe, from memory??) and I mostly can't remember who's been around longer than me and who hasn't. Like, I joined the new site a few days before Qwynegold, but I can't remember if Qwynegold is an oldbie who's been around longer than me. I don't remember with you either. People have also changed user names and because I haven't been around very consistently, a lot of names simply look familiar and that's all I can say about them.

Yeah, same with me. I was minimally active on the old board (though a long-time lurker), so I wouldn’t know much about this either. (My involvement on the old board was mostly restricted to me promoting my SCA. Actually, now that I think about it, most of my most recent posts have been promoting my new SCA. Funny how things go full circle.)

One thing about the internet though: in nearly all cases, things can't be settled physically and there is no actual danger of ranking someone lower than you when they could actually kick your arse, so I imagine, if this language were used on something like the internet, there would just be a lot of people arrogantly ranking themselves higher than everyone else and either this would result in flame-wars or it would just be ignored and used to form judgements about who is arrogant and who isn't. Potentially, there would be a politeness standard where second person is placed above first person, always, and third persons could be placed even higher or lower depending on what you want to say about them. Maybe placing 2nd person above 1st person would even become a rule of politeness on some forums. I don't know. I think something like that would change the language. When it's spoken mostly face to face, this system seems like it could work well (although horribly, lol), but in fairly anonymous, written communication, I think a change would come about.

Hmm… I suppose it could end up like this? I personally think that most people would intuitively rank people based on perceived subjective status within the community, with a handful of arrogant types ranking themselves highest and getting disliked for it.

Ares Land wrote: ↑Mon Jun 28, 2021 8:07 am I don't have any particular questions right now, but I like all of this a lot, especially the 'monovalence' and specific/nonspecific parts!

Specificity seems to be a trend right now. First my Hlʉ̂, then your a-Uttes, and now Balog have all had it as an important grammatical component. Perhaps us conlangers are finally paying more attention to semantics!

Imralu · Post by **Imralu** » Tue Jun 29, 2021 2:01 pm

Raholeun wrote: ↑Mon Jun 28, 2021 1:41 am I am sort of fascinated after reading all this. Some brief comments:

Thank you!

Raholeun wrote: ↑Mon Jun 28, 2021 1:41 am- Your text examples only seem to contain a limited subset all of Balog phonemes. Some listed consonant sounds are not represented in the examples. Is that intentional, or do <tshq, dž> etc only occur in a certain register of the language?

No, that's not intentional at all. I have only shown about 15 lexemes so far, about 20 including the subject markers, but also, it's quite natural that some sounds are more common than others. I don't think any language has a pretty much equal use of all the phonemes. Also, they can be distributed quite strangely as well. For example, /ð/ in English is quite common in use, being in many of the most common grammatical words (the, this, that, these, those, they, them, there, with, whether, either, neither) and many kinship terms (mother, father, brother) and then pretty sparsely distributed throughout the rest of the lexicon (... bother ... dither ... weather ...). I think I'm just picking words to have the right feel for me. I'm sure some of the consonants will get used quite rarely.

Raholeun wrote: ↑Mon Jun 28, 2021 1:41 am- In the last post you note that there is room for some idiomatic meaning arising. I really like that idea and it might open some pathways into historical derivation. That got me thinking; rank is deferred to the prowess or strength of a person or object. But I could also imagine that the word itself always gets ranked a certain way, for instance because the word was loaned from the effeminate elves or some subservient Orc race. Then rank gets lexicalized, adding a layer of complexity.

Yeah, I thought about that too, but I can only really imagine that happening with the lowliest of low concepts and even then, if you're talking about two very low concepts, there will always be the pressure to rank one higher and there is also always the possibility of denigrating someone by placing them lower in this ranking than some disgusting, low concept. With high-ranking people, they could almost get lexicalised with u=, for example, the emperor ... but then, if talking about the deities of the Balog religion with the emperor in the same discourse, it could be seen as a slight to the deities to rank the lowly human emperor as equal with them. I can't really imagine much pressure for anything to become fixed completely as there would always be situations where giving a different rank would be tempting within a particular discourse. I think the rank-system itself will have to stay flexible, but there is a lot of room for bizarre conventions to arise

I just imagined what might happen if someone is talking loudly and mentions the emperor in a low rank within earshot of someone high-ranking who might report it to the emperor. That's speechcrime and everyone would be trying to distance themselves from it if they saw who had overheard it, so I just imagined that people would pretent to literally not understand it.

Ares Land wrote: ↑Mon Jun 28, 2021 8:07 am I don't have any particular questions right now, but I like all of this a lot, especially the 'monovalence' and specific/nonspecific parts!

Thank you!

bradrn wrote: ↑Mon Jun 28, 2021 8:44 amI know that the Kalam dictionary simply treats idioms as subheadings under the main word, [...]

But in your case this might be more difficult, since your idiomatic phrases won’t necessarily be contiguous. Perhaps you could mark them as ‘verb1 … verb2’ or something?

Thanks for the link! Yeah, the main problem is that I usually don't write nicely formatted dictionaries but instead write what are (hopefully) cleverly formatted dictionaries in Excel (or rather the Open Office equivalent) so that I can search for things easily, and usually my system gets a bit weird with stuff like this. I think using numbers is a good way to do it. Thanks for the tip!

bradrn wrote: ↑Mon Jun 28, 2021 8:44 am
All this sounds reasonable. Also, I’d be interested in hearing more about the open class of ergative markers.

OK, I'll paste from my still pretty unorganised dictionary. This is not an exhaustive list and I've randomly added entries in there for the words for the body parts themselves (regularly derived from the instrumentals with the suffix -ed ... but I think I need more variety there and irregularity, suppletive forms etc.). Some of the English translations are a bit iffy ... like "use knowledge" for "know", but I'm also adding other translations a lot of the time to indicate what it may actually end up meaning in a sentence. More will be added including more combining body part and manner, such as "use hand violently", "use hand lovingly", "use hand skillfully" etc.

The < > helps me search for things by the beginning or end. Not always useful, but surprisingly often. It will definitely be useful if I want to rhyme stuff, lol. Also, it tells me if I've already created a word because the > automatically gets suggested.

Also, none of these words are actually marked for number. For example wež means "use eyes" but can apply to one or more people. The derived wežed "be eyes", refers to a set or set of eyes. Wežad refers to an individual eye or individual eyes. There are also singulative and plurative suffixes to make it clear: =wežedan "pair of eyes", =wežedim "pairs of eyes", =wežadan "eye", =wežadim "(individual) eyes"

Oh, and the references to horns ... that's the alient symbiont ... they essentially end up looking like horns on a host.

Code: Select all

WORD            MORPHOLOGY      POS     ENGLISH
<dwar>		-dwar-		V	<use top of head> <headbutt>
<zaŋ>		-zaŋ-		V	<use hair> <flick with hair>
<moš>		-moš-		V	<use forehead> <headbutt>
<gož>		-gož-		V	<use mind> <think>
<žel>		-žel-		V	<use knowledge> <know>
<gwež>		-gwež-		V	<use mind's eye> <imagine seeing>
<gok>		-gok-		V	<use mind's ear> <imagine hearing>
<mev>		-mev-		V	<use mind's body> <imagine feeling>
<m>		-m-		V	<use emotion> <use heart> <feel>
<wež>		-wež-		V	<use eyes (pair)> <see>
<wežak>		-wež-ak-	V	<use eye (individual)> <see with one eye> <see out of the corner of one's eye>
<wežed>		-wež-ed-	V	<be eyes (pair)> <be a pair of eyes> <be pairs of eyes>
<wežad>		-wež-ak-ed-	V	<be eye (individual)>
<wežuž>		-wež-už-	V	<use eyes (pair) with glasses>
<wežakuž>	-wež-ak-už-	V	<use eye (individual) with a monocle>
<wežužed>	-wež-už-ed-	V	<be glasses (pair)> <be eye glasses>
<wežakužed>	-wež-ak-už-ed-	V	<be monocle>
<kel>		-kel-		V	<use horn>
<keleb>		-kel-eb-	V	<be horned one>
<keled>		-kel-ed-	V	<be horn>
<keluv>		-kel-uv-	V	<use both horns>
<kelak>		-kel-ak-	V	<use one horn>
<keluved>	-kel-uv-ed	V	<be horn (pair)>
<kelad>		-kab-		V	<be horn (pair)>
<hif>		-hif-		V	<use nose> <smell>
<hifed>		-hif-ed-	V	<be nose>
<zok>		-zok-		V	<use ears> <hear>
<zoked>		-zok-ed-	V	<be ears>
<zokak>		-zok-ak-	V	<use one ear> <hear with one ear> <overhear>
<zokad>		-zok-ak-ed-	V	<be one ear>
<mul>		-mul-		V	<use mouth> <eat> <talk> <speak>
<muled>		-mul-ed-	V	<be a mouth>
<kan>		-kan-		V	<use teeth> <bite>
<kaned>		-kan-ed-	V	<be teeth>
<kanad>		-kan-ak-ed	V	<be one tooth>
<lob>		-lob-		V	<use tongue> <lick>
<lobed>		-lob-ed-	V	<be tongue>
<sak>		-sak-		V	<use lips> <kiss> <suck>
<saked>		-sak-ed-	V	<be lips>
<sakad>		-sak-ak-ed-	V	<be one lip>
<dev>		-dev-		V	<use beard>
<deved>		-dev-ed-	V	<be beard>
<devad>		-dev-ad-	V	<be beard hair>
<nyag>		-nyag-		V	<use neck>
<bul>		-bul-		V	<use shoulder> <shoulder>
<baž>		-baž-		V	<use upper arm> <use biceps>
<kab>		-kab-		V	<use elbow> <elbow> <nudge>
<dab>		-dab-		V	<use lower arm> <block with lower arm>
<veŋ>		-veŋ-		V	<use wrist> <twist> <masturbate>
<b>		-b-		V	<use hand>
<bak>		-b-ak-		V	<use hand (individual)>
<bad>		-b-ak-ed-	V	<be hand (individual)>
<buv>		-b-uv-		V	<use hands (pair)>
<buved>		-b-uv-ed-	V	<be hands (pair)>
<taŋ>		-taŋ-		V	<use open hand> <slap>
<buw>		-duw-		V	<use fist> <punch>
<guŋ>		-guŋ-		V	<use one finger>
<guwiŋ>		-guwiŋ-		V	<use fingers>
<gawaŋ>		-gawaŋ-		V	<use thumb>
<buž>		-d-už-		V	<use gloved hand>
<mok>		-mok-		V	<use chest>
<maž>		-maž-		V	<use breast> <use pectoral muscle>
<neŋ>		-neŋ-		V	<use nipple> <breastfeed>
<bwel>		-bwel-		V	<use back>
<gum>		-gum-		V	<use belly>
<lig>		-lig-		V	<use genitals>
<mun>		-mun-		V	<use vagina>
<žub>		-žub-		V	<use penis>
<zeb>		-zeb-		V	<use testicles>
<fug>		-fug-		V	<use anus> <defecate>
<bam>		-bam-		V	<use buttock>
<ŋal>		-ŋal-		V	<use leg> <walk> <run>
<ŋaled>		-ŋal-ed-	V	<be leg>
<ŋalad>		-ŋal-ak-ed-	V	<be leg (individual)>
<ŋalak>		-ŋal-ak-	V	<use leg (individual)>
<val>		-val-		V	<use thigh>
<gaž>		-gaž-		V	<use knee>
<wod>		-wog-		V	<use calf>
<ŋim>		-ŋim-		V	<use shin>
<mes>		-mes-		V	<use ankle>
<duw>		-duw-		V	<use foot> <kick> <nudge with foot>
<d>		-d-		V	<use sole of foot> <tread> <step on> <crush underfoot> <walk>
<žeb>		-žeb-		V	<use toe>
<muğ>		-muğ-		V	<use body>
<wuŋ>		-wuŋ-		V	<use body weight>
<has>		-has-		V	<use skin>
<žum>		-žum-		V	<use muscle>	
<tav>		-tav-		V	<use stick> <hit with stick> <dig with stick>
<gat>		-gat-		V	<use end of thrown spear> <throw spear>
<tsag>		-tsag-		V	<use end of held spear> <stab with spear>
<žits>		-žits-		V	<use shaft of spear> <hit with spear>
<gev>		-gev-		V	<use held rock>
<tug>		-tug-		V	<use large projectile> <throw large object>
<teg>		-teg-		V	<use small projectile> <throw small object> <shoot>
<his>		-his-		V	<use arrow> <shoot arrow>
<kkal>		-kkal-		V	<use axe> <chop with axe>
<gag>		-gag-		V	<use club> <club>

bradrn wrote: ↑Mon Jun 28, 2021 8:44 amNote also that languages tend to innovate new words for culturally-specific concepts. In this case, the behaviour you describe reminds me of Austronesian amok, the word in this case coming from Malay. If your people go into a murderous rage often enough, they’ll probably innovate a word for the concept as well.

Yeah, but it may just be that innovative. In US English, they generally just use the phrase mass shooting. It's funny that "amok" isn't used more often in English. I guess it's been made too bland by overuse ("The kids are running amuck again!"), but in German, they use Amoklauf for things like that.

But yeah, I thought about a word for losing control and snapping, like "get lost" could easily begin to be understood on its own. I thought about the English idiom "see red" and how something like that could then also encode the assumed consequences.

Oh, that sounds awful. I hope you recover soon.

Thanks! Off-topic rant:

More: show

I actually think it would end up as both, depending on absolute status. If Random Goon 1 wins a fight with Random Goon 2, said goon probably isn’t going to demand all references to him be rewritten (for practical reasons if nothing else). If the Great Orc King gets killed by the Next Great Orc King (because realistically, how else would orcs choose their king?), there’ll probably be trouble for anyone who doesn’t immediately rewrite their documents to praise the Next Great Orc King.

Yeah, you're right, and there would probably be emperors who wouldn't care and nutty ones who put people to death for ever having written of them in a low rank without even giving the opportunity to edit. "How dare you not have foreseen the position I would one day hold. Death for you!"

Yeah, same with me. I was minimally active on the old board (though a long-time lurker), so I wouldn’t know much about this either. (My involvement on the old board was mostly restricted to me promoting my SCA. Actually, now that I think about it, most of my most recent posts have been promoting my new SCA. Funny how things go full circle.)

I definitely know your name. Often, things like names just get stored in my head with no accompanying information. Several years ago, I was walking across a bridge and a string of words entered my head and I had no idea from where and it ran around in my head for a while ... it turned out to be someone's username from here, something like "So Haleza Grizeh", but my brain just filed that under "miscellaneous words" without attachment to anything else. Bradrn is in there too, but I have no idea how long it's been there.

Hmm… I suppose it could end up like this? I personally think that most people would intuitively rank people based on perceived subjective status within the community, with a handful of arrogant types ranking themselves highest and getting disliked for it.

But not everything on the internet is in a community. You can just ask questions into the ether and have them answered on some sites with no requirement of even having a log-in, just a username that you pick at that moment.

Ares Land wrote: ↑Mon Jun 28, 2021 8:07 am I don't have any particular questions right now, but I like all of this a lot, especially the 'monovalence' and specific/nonspecific parts!
Specificity seems to be a trend right now. First my Hlʉ̂, then your a-Uttes, and now Balog have all had it as an important grammatical component. Perhaps us conlangers are finally paying more attention to semantics!

Hmm, isn't it more pragmatics than semantics. The book, a (specific) book and a(ny) book are not really semantically different, only pragmatically as it depends on my own knowledge and my expectations of your knowledge. In any case, it's not the first time I've had this in a language. One I started working on back in 2013 also had this three-way DEF, SPEC, NSPC distinction but I never considered it interesting enough to post about it. I remember agonizing over which marking to give question words because they kind of fit into the quadrant that I haven't covered: you know which one, but I don't, and I think that might have been all I mentioned about it anywhere. I also got annoyed that the Leipzig Glossing Rules seem to completely ignore specificity (which is why I added abbreviations for it in my signature) and often it's even been ignored by linguists describing languages with a specificity distinction. I've seen it described in really roundabout ways like "the definite article is used for all specific referents, regardless of whether the listener is expected to know the identity of the referent or not", ... um, that's not a definite article. That's a specific article! In fact, I think the only time I've seen it referenced in a grammar is when there is at least a three-way distinction, although even then, often it's described in very funny ways.

malloc · Post by **malloc** » Tue Jun 29, 2021 8:57 pm

That looks quite interesting, especially the monovalent concept. Is there any particular reason or inspiration behind this idea? I also like the idea of eliminating nouns. My own project has focused on that, although everyone said eliminating the noun-verb distinction was impossible when I presented my ideas.

Raholeun · Post by **Raholeun** » Wed Jun 30, 2021 2:51 am

Imralu wrote: ↑Tue Jun 29, 2021 2:01 pm
Raholeun wrote: ↑Mon Jun 28, 2021 1:41 am- In the last post you note that there is room for some idiomatic meaning arising. I really like that idea and it might open some pathways into historical derivation. That got me thinking; rank is deferred to the prowess or strength of a person or object. But I could also imagine that the word itself always gets ranked a certain way, for instance because the word was loaned from the effeminate elves or some subservient Orc race. Then rank gets lexicalized, adding a layer of complexity.
Yeah, I thought about that too, but I can only really imagine that happening with the lowliest of low concepts and even then, if you're talking about two very low concepts, there will always be the pressure to rank one higher and there is also always the possibility of denigrating someone by placing them lower in this ranking than some disgusting, low concept. With high-ranking people, they could almost get lexicalised with u=, for example, the emperor ... but then, if talking about the deities of the Balog religion with the emperor in the same discourse, it could be seen as a slight to the deities to rank the lowly human emperor as equal with them. I can't really imagine much pressure for anything to become fixed completely as there would always be situations where giving a different rank would be tempting within a particular discourse. I think the rank-system itself will have to stay flexible, but there is a lot of room for bizarre conventions to arise

Take, for instance, the sentence "Love will tear us apart" (song). Tearing apart is not something that is done delicately, so it would be implied that "love" strongly outranks the plural pronoun. However, all Orcs know love is something from the realm of cowards, weaklings and invertabrates and altogether not very orcish. Would "love" then be assigned a high rank like the sentence implies, or would the speaker simply refuse to rank such a concept higher than "us"?

Imralu · Post by **Imralu** » Wed Jun 30, 2021 2:33 pm

malloc wrote: ↑Tue Jun 29, 2021 8:57 pmThat looks quite interesting, especially the monovalent concept. Is there any particular reason or inspiration behind this idea?

Thanks! I can't remember exactly what my inspiration was except for a couple of main ideas: some languages tend to gravitate towards only having one fully expressed argument in a clause, tending to avoid establishing two new entities. I also wanted to play around with ergativity and I was also really keen to try out something with a lot of marking of instrumentality, but it felt like it should be associated with the expression of the agent rather than the patient.

I also like the idea of eliminating nouns. My own project has focused on that, although everyone said eliminating the noun-verb distinction was impossible when I presented my ideas.

Lol. It's not impossible. Lojban doesn't have a noun-verb distinction and that's not what makes the language hard to use. Virtually none of my conlangs have had a lexical NV distinction for the last 15 or so years and that's never a bit that causes problems.

There are, however, a lot of ways to not get rid of the lexical noun-verb distinction but claim you have and I've seen it claimed before for languages where all the content words can appear in all the syntactic positions where content words can appear, but with unpredictable changes in meaning. For example, something like this is similar to what I've seen before:

Code: Select all

How NOT to do it:

word  | noun meaning | verb meaning | adj meaning
------|--------------|--------------|------------
blort | food         | eat          | edible
------|--------------|--------------|------------
spram | size         | big          | big one
------|--------------|--------------|------------
blops | speech       | speak        | spoken
------|--------------|--------------|------------
mlam  | fish         | fish         | fishy
------|--------------|--------------|------------
mlamo | fisherman    | be fisherman | fisherman-like
------|--------------|--------------|------------
drub  | comb         | combed       | combed
------|--------------|--------------|------------
glug  | beauty       | beautify     | beautiful

You could indeed say that you can't use syntactic means to say that a word is only one part of speech because there are no gaps in the table, each word can occupy the same slots as the others, but with changes in meaning like this, all we're looking at is zero derivation. Each meaning associated with the part of speech of a word is essentially a related but separate lexical entry. ("Finger means what as a verb?!") Then you'd also have to have a syntactical difference in sentences to indicate the difference between "It eats" and "It is food" or "He fishes" and "He is a fish". It then becomes very easy to have syntactic test to distinguish what part of speech something is. For example, mlam¹ meaning "fish" (the animal) would likely need a copula to appear in the predicate (is a fish) whereas mlam² meaning "fish" (the activity) could not be used with a copula but would appear in the predicate with whatever verb markings are normal. Thus, you would easily be able to demonstrate that mlam¹ and mlam² belong to different parts of speech and one of these parts of speech would very conveniently be called nouns and the other verbs.

I rarely post a great deal about this aspect of my conlangs, but the few times I have and someone has told me I cannot possibly have no lexical, it's turned out that they've essentially assumed that I'm just kidding myself with this zero derivation thing. I think it's not something you can just claim and get away with without arguing because a lot of people have seen people kidding themselves with zero-derivation or that kind of "there-is-stonying-beneath..." kind of thing that Zompist was warning about in the first LKC.

Raholeun wrote: ↑Wed Jun 30, 2021 2:51 amTake, for instance, the sentence "Love will tear us apart" (song). Tearing apart is not something that is done delicately, so it would be implied that "love" strongly outranks the plural pronoun. However, all Orcs know love is something from the realm of cowards, weaklings and invertabrates and altogether not very orcish. Would "love" then be assigned a high rank like the sentence implies, or would the speaker simply refuse to rank such a concept higher than "us"?

Yes, I will have to have a think about how situations like that are handled. In any case, I wouldn't say that "love" would lexicalise to always have a particular rank, just because it may always be ranked below human referents (or at least Balog referents), because it can always be put in a sentence with an even more pathetic, invertebrate concept like ... compassion [Orcish mocking laughter].

I'm also not sure I'll typecast them quite to that extent. Perhaps the balog will view love as weakness, vulnerability, but the fact that it does have the capacity to make you weak and vulnerable (e.g. your enemies can get at you by killing your lover) means it is deserving of its place above the people being torn apart by it (or perhaps rather that they are so pathetic, they should be ranked below it).

In any case, the purpose of the ranks is to keep referents separate pronominally and avoid rementioning the lexical verbs again and again, and that purpose is defeated if there are continual reassignments of rank as every new sentence contains a different power balance. What I mean is that "love" wouldn't necessarily get ranked above the people just because it's tearing them apart in that particular sentence. You could do it, to emphasise its power over them or to emphasise a flip of power, from something they controlled to something that controls them, but if you have previously mentioned love, say in the E rank and the people in the C rank, you could also just continue the previous assignment of ranks.

I have a feeling that describing all the ins and outs of when to use what rank when discussing whom, whom and what together could probably fill up a whole book, in-world ... like, a handy ettiquete guide for non-Balog to avoid putting their foot in it (and potentially getting killed) when talking to or near Balog. That seems a bit ambitious for me, so I guess I'll keep thinking about it and make stuff up on the fly a lot and hopefully keep it fairly consistent.

bradrn · Post by **bradrn** » Thu Jul 01, 2021 10:25 am

Imralu wrote: ↑Tue Jun 29, 2021 2:01 pm
bradrn wrote: ↑Mon Jun 28, 2021 8:44 amI know that the Kalam dictionary simply treats idioms as subheadings under the main word, [...]

But in your case this might be more difficult, since your idiomatic phrases won’t necessarily be contiguous. Perhaps you could mark them as ‘verb1 … verb2’ or something?
Thanks for the link! Yeah, the main problem is that I usually don't write nicely formatted dictionaries but instead write what are (hopefully) cleverly formatted dictionaries in Excel (or rather the Open Office equivalent) so that I can search for things easily, and usually my system gets a bit weird with stuff like this. I think using numbers is a good way to do it. Thanks for the tip!

Yes, I find that Excel etc. are pretty bad at this sort of thing. The grid format is really not well suited for things like this, which are best communicated through long-form explanations and examples. Possibly you might like SIL Toolbox — I use it for my dictionaries, and the main view is somewhat Excellish.

bradrn wrote: ↑Mon Jun 28, 2021 8:44 am
All this sounds reasonable. Also, I’d be interested in hearing more about the open class of ergative markers.
OK, I'll paste from my still pretty unorganised dictionary. This is not an exhaustive list and I've randomly added entries in there for the words for the body parts themselves (regularly derived from the instrumentals with the suffix -ed ... but I think I need more variety there and irregularity, suppletive forms etc.). Some of the English translations are a bit iffy ... like "use knowledge" for "know", but I'm also adding other translations a lot of the time to indicate what it may actually end up meaning in a sentence. More will be added including more combining body part and manner, such as "use hand violently", "use hand lovingly", "use hand skillfully" etc.

Thanks for the info! But I suppose I was more interested in knowing: how can there be ergative languages in a monovalent conlang, when the ergative marks the subject of bivalent verbs?

bradrn wrote: ↑Mon Jun 28, 2021 8:44 amNote also that languages tend to innovate new words for culturally-specific concepts. In this case, the behaviour you describe reminds me of Austronesian amok, the word in this case coming from Malay. If your people go into a murderous rage often enough, they’ll probably innovate a word for the concept as well.
Yeah, but it may just be that innovative. In US English, they generally just use the phrase mass shooting. It's funny that "amok" isn't used more often in English. I guess it's been made too bland by overuse ("The kids are running amuck again!"), but in German, they use Amoklauf for things like that.

But yeah, I thought about a word for losing control and snapping, like "get lost" could easily begin to be understood on its own. I thought about the English idiom "see red" and how something like that could then also encode the assumed consequences.

‘Word’ was probably the wrong term to use here — I should have said ‘conventionalised expression’ or something more like that. Your example of ‘mass shooting’ is a good one: that’s something which happens a lot in the US, so Americans now have a conventionalised expression for such an event. Contrast this to my idiolect: in Australia we have no mass shootings, and so to me that expression sounds no more conventional than any other.

Hmm… I suppose it could end up like this? I personally think that most people would intuitively rank people based on perceived subjective status within the community, with a handful of arrogant types ranking themselves highest and getting disliked for it.
But not everything on the internet is in a community. You can just ask questions into the ether and have them answered on some sites with no requirement of even having a log-in, just a username that you pick at that moment.

Now that’s a point I hadn’t considered. In this case, I’m not sure what would end up happening.

Ares Land wrote: ↑Mon Jun 28, 2021 8:07 am I don't have any particular questions right now, but I like all of this a lot, especially the 'monovalence' and specific/nonspecific parts!
Specificity seems to be a trend right now. First my Hlʉ̂, then your a-Uttes, and now Balog have all had it as an important grammatical component. Perhaps us conlangers are finally paying more attention to semantics!
Hmm, isn't it more pragmatics than semantics. The book, a (specific) book and a(ny) book are not really semantically different, only pragmatically as it depends on my own knowledge and my expectations of your knowledge. In any case, it's not the first time I've had this in a language. One I started working on back in 2013 also had this three-way DEF, SPEC, NSPC distinction but I never considered it interesting enough to post about it. I remember agonizing over which marking to give question words because they kind of fit into the quadrant that I haven't covered: you know which one, but I don't, and I think that might have been all I mentioned about it anywhere. I also got annoyed that the Leipzig Glossing Rules seem to completely ignore specificity (which is why I added abbreviations for it in my signature) and often it's even been ignored by linguists describing languages with a specificity distinction. I've seen it described in really roundabout ways like "the definite article is used for all specific referents, regardless of whether the listener is expected to know the identity of the referent or not", ... um, that's not a definite article. That's a specific article! In fact, I think the only time I've seen it referenced in a grammar is when there is at least a three-way distinction, although even then, often it's described in very funny ways.

Possibly it’s more pragmatic than semantic. This isn’t really an area I know much about. I do know that question words would be non-specific though.

malloc wrote: ↑Tue Jun 29, 2021 8:57 pm I also like the idea of eliminating nouns.

This is an idea I find interesting. It seems like natlangs regularly try to get rid of their verb class, by making it very small (or even closed) and sending the remainder over to a more nouny class, but I haven’t yet found one with a small or closed noun class, with most ‘nouns’ pushed into a more verby class. I’m convinced this reflects some basic cognitive property of the human brain — we prefer to think about things rather than actions, or something like that.

Imralu wrote: ↑Wed Jun 30, 2021 2:33 pm
I also like the idea of eliminating nouns. My own project has focused on that, although everyone said eliminating the noun-verb distinction was impossible when I presented my ideas.
Lol. It's not impossible. Lojban doesn't have a noun-verb distinction and that's not what makes the language hard to use. Virtually none of my conlangs have had a lexical NV distinction for the last 15 or so years and that's never a bit that causes problems.

There are, however, a lot of ways to not get rid of the lexical noun-verb distinction but claim you have and I've seen it claimed before for languages where all the content words can appear in all the syntactic positions where content words can appear, but with unpredictable changes in meaning. … You could indeed say that you can't use syntactic means to say that a word is only one part of speech because there are no gaps in the table, each word can occupy the same slots as the others, but with changes in meaning like this, all we're looking at is zero derivation. … Then you'd also have to have a syntactical difference in sentences to indicate the difference between "It eats" and "It is food" or "He fishes" and "He is a fish". It then becomes very easy to have syntactic test to distinguish what part of speech something is.

If you haven’t seen it already, I can highly recommend François’s paper on word classes in Hiw. (It’s probably one of the most enlightening papers I’ve ever read; I’ve already recommended it once on the ZBB.) For me it really solidified this notion of zero-derivation vs one word class, and in particular I really like the way François approaches syntactic tests as a way to distinguish word classes. (Basically, each word class can occur in a different set of syntactic environments. Even if, say, nouns and verbs can regularly appear in exactly the same environments, they can be told apart if there are other environments where only one or the other occurs. To truly unify nouns and verbs, you would then need to make them completely indistinguishable.)

Vardelm · Post by **Vardelm** » Thu Jul 01, 2021 4:16 pm

bradrn wrote: ↑Thu Jul 01, 2021 10:25 am
malloc wrote: ↑Tue Jun 29, 2021 8:57 pm I also like the idea of eliminating nouns.
This is an idea I find interesting. It seems like natlangs regularly try to get rid of their verb class, by making it very small (or even closed) and sending the remainder over to a more nouny class, but I haven’t yet found one with a small or closed noun class, with most ‘nouns’ pushed into a more verby class. I’m convinced this reflects some basic cognitive property of the human brain — we prefer to think about things rather than actions, or something like that.

Do the Salishan languages not count?

Imralu · Post by **Imralu** » Thu Jul 01, 2021 4:32 pm

bradrn wrote: ↑Thu Jul 01, 2021 10:25 amYes, I find that Excel etc. are pretty bad at this sort of thing. The grid format is really not well suited for things like this, which are best communicated through long-form explanations and examples. Possibly you might like SIL Toolbox — I use it for my dictionaries, and the main view is somewhat Excellish.

Thanks, yeah, I really got turned off SIL things a long time ago. Not only did I have really unexpected formatting irregularities throughout my file (some entries had different font sizes or types for no apparent reason), a file got corrupted one time and only recent back-up I had saved was somehow also corrupted.

Thanks for the info! But I suppose I was more interested in knowing: how can there be ergative languages in a monovalent conlang, when the ergative marks the subject of bivalent verbs?

Ah, yes. Well, I said there is essentially an open class of ergative markers. For example, if all that was ever used was g "cause", you could just analyse that as an ergative marker rather than a verb and analyse this sentence as one single transitive sentence with ergative alignment. .

Goož magaz iddauz.

g-: ERG-

o=: DEF(c)=

ž: 1S

magaz: fall.over

i=: DEF(e)=

dauz: tree

I felled the tree.

However, I could also say Boož magaz iddauz, "I felled the tree by hand". Perhaps you could call b- a marker of the "manual ergative" → "manergative" case. There's essentially no limit though:

Doož magaz iddauz. "I kicked the tree over." → oož is marked in the "pedergative" case
Tšqanagoož magaz iddauz. "I sneezed the tree over." → oož is marked in the sternutergative case! (Not a single hit on Google. Disappointing!)

My point was basically that verb phrases like these basically fill the role of the ergative case while also providing additional information. G, b d and tšqanag are all verbs, not merely types of ergative case markers and as such, the things they mark are additional clauses. (D'oož, for example, is an entire valid sentence on its own basically just meaning "I do something with my hands", "I use my hands".) As verbs, they can also be embedded within a subject phrase, such as ottšqanag "the (C-ranked) one who sneezes", and the order of consistutents in the clause tšqanag oož "I sneeze" could also be reversed to create ž'ottšqanag "the one who sneezes is me", just as magaz iddauz "the tree fell over" can be reversed to dauz immagaz "the thing that fell over was a tree". Tšqanag ottšqanag. Dauz iddauz. Magaz immagaz. (The sneezer sneezes. The tree is a tree. The faller-over-er falls over."

‘Word’ was probably the wrong term to use here — I should have said ‘conventionalised expression’ or something more like that. Your example of ‘mass shooting’ is a good one: that’s something which happens a lot in the US, so Americans now have a conventionalised expression for such an event. Contrast this to my idiolect: in Australia we have no mass shootings, and so to me that expression sounds no more conventional than any other.

Yeah. I'm also Australian, but I would say "mass shooting" is part of my idiolect as a loan from US English ... although I suppose just about everything could count as a loan in an idiolect.

Hmm… I suppose it could end up like this? I personally think that most people would intuitively rank people based on perceived subjective status within the community, with a handful of arrogant types ranking themselves highest and getting disliked for it.
But not everything on the internet is in a community. You can just ask questions into the ether and have them answered on some sites with no requirement of even having a log-in, just a username that you pick at that moment.

Now that’s a point I hadn’t considered. In this case, I’m not sure what would end up happening.[/quote] Language evolution I guess, or the development of specific cultural norms. For example, on the internet in German, most people use informal address, du/ihr etc., on forums where they would be likely to use Sie if they spoke to strangers in real life.

Possibly it’s more pragmatic than semantic. This isn’t really an area I know much about. I do know that question words would be non-specific though.

Well, there are actually many possible justifications for using definite/indefinite, specific/non-specific marking as far as I can see it. For example, you could justify the definite because the listener is assumed to know the identity

Code: Select all

                               | Listener assumed to  | Listener not assumed   |
                               | know which one       | to know which one      |
-------------------------------|----------------------|------------------------|
       Speaker knows which one:| definite specific    | indefinite specific    |
-------------------------------|----------------------|------------------------|
Speaker doesn't know which one:| definite nonspecific | indefinite nonspecific |
-------------------------------|----------------------|------------------------|

Because I only have three marked categories ...

definite specific: V=
indefinite specific: VqV=
indefinite nonspecific: VhV=

..., the remaining square (definite, nonspecific) has to share the marker of one of the others. It depends which one you think is the most salient feature I guess: if the question as to whether the listener is assumed to know the identity (definiteness) is the first criterion you look at, then it should clearly be definite. If the speaker's own knowledge is the main criterion, then the indefinite nonspecific. I think I will generally go with the definite, thus specificity is not marked in the definite, only the indefinite, meaning I have a marking system like this

Code: Select all

                               | Listener assumed to  | Listener not assumed   |
                               | know which one       | to know which one      |
-------------------------------|----------------------|------------------------|
       Speaker knows which one:|                      | VqV= (SPEC=)           |
-------------------------------| V= (DEF=)            |------------------------|
Speaker doesn't know which one:|                      | VhV= (NSPC=)           |
-------------------------------|----------------------|------------------------|

Because the referents of question words are generally unknown to the speaker and assumed to be known to the listener, they logically fall into that definite-nonspecific quadrant and so I think I'll be marking them as definite in this language rather than as non-specific. Perhaps the non-specific could be used in some rhetorical questions to convey a "you don't know the answer to this, nobody does".

Maybe I'll instead come up with a marking for that fourth quadrant and use it more or less as the interrogative thing and that would mean that I could use a verb like "be person" with it for "who" instead of having a specific "be who". Perhaps it should also be external to the rank system because it is likely to be quite hard to pick a rank for an unknown entity.

This is an idea I find interesting. It seems like natlangs regularly try to get rid of their verb class, by making it very small (or even closed) and sending the remainder over to a more nouny class, but I haven’t yet found one with a small or closed noun class, with most ‘nouns’ pushed into a more verby class. I’m convinced this reflects some basic cognitive property of the human brain — we prefer to think about things rather than actions, or something like that.

I'm curious what natlangs you're referring to. I know Japanese can more or less be said to have a closed verb class, but it's still not that small. There are also a lot of languages that make frequent use of ... I forget what they're called ... verbs like etmek in Turkish, and I guess that's mostly what Japanese does, but I've never seen one of those languages that comes close to getting rid of the distinction.

I've kind of had the impression it's more the opposite in natlangs that get close to it, such as in Lushootseed where you could regard ʔux̌ʷ as a verb meaning "go" and sbiaw as a verb meaning "be coyote" and both can be nominalised by putting them in essentially what can be regarded as a relative clause:

ti: SPEC

sbiaw: be.coyote

= (the) one that is a coyote → the/a coyote

ti: SPEC

ʔux̌ʷ: go

= (the) one that goes → the/a "goer"

This is probably a massive oversimplification of Lushootseed grammar which looks terrifyingly intimidating in most example sentences, and apparently some verbs do need to be marked in ways nouns don't when used as arguments ... although ʔux̌ʷ doesn't need to be ... meaning it belongs to the lexical class of nouns, I guess? In any case, this is basically the way Balog works as well as my old conlang Iliaqu. In both Balog and Iliaqu, predicates consist of [verb phrase] and subjects (and in the case of Iliaqu, also objects, other arguments and adjuncts) consist of [nominaliser] (+ [verb phrase]). Actually, in Iliaqu, there is also a way to verbalise nominalisers, meaning you could have a verb phrase consisting of an embedded noun phrase with an imbedded verb within and so on. Shenanigans. In practice, there is rarely a need to go so many levels deep. Haven't devised that for Balog yet and maybe it won't need it. (In Iliaqu, it was good not only for bringing definiteness and specificity into predicates, but also the meanings of cases and that won't be the ... case in Balog because there are only subjects.) In any case, the big open class of words in Iliaqu and Balog all belong to one class and because they're unmarked in predicates and marked by something which can be thought of as the head of a relative clause when appearing in subjects, they're most conveniently thought of as verbs.

In the majority of my other languages, the open class of word has been more easily thought of as nouns as they're unmarked as subjects and marked by a predicate marker (which can also be thought of as an invariant copula) in predicates, e.g. "I fell the tree" is, in one of my languages:

na: 1S

i: COP

zyendu: feller

ya: GEN

vyada: tree

I fell a tree.
I'm a feller of a tree.

You could regard the phrase i zyendu as verb meaning "fell" and regard ya as an accusative marker, but you can flip the sentence around the copula to get this.

zyendu: feller

ya: GEN

vyada: tree

i: COP

na: 1S

The one who fells a tree is me.

Zyendu, vyada and na can occur in all of the same syntactical positions. You could drop "ya" as well; zyendu vyada is syntactically more like "tree feller" than "feller of tree", but in most instances, essentially the same.

The funny way that this causes accusative-like roles to coalesce with genitive-like roles also means that passive essentially coalesces with "have" because, for example, people who are killed have killers, things that are eaten have eaters etc. I have specified that into finer shades of meaning. I use the noun ne as basically an "inadvertent-haver". One who or that which "has" through no action of their own, as opposed to, for example, za which is a "holder", one who has by holding, whether literally or metaphorically, with more volition than ne.

vyada: tree

i: COP

ne: inadvertent.haver

(ya): (GEN)

zyendu: feller

The tree (through no doing of its own) has a feller.
The tree is felled.

na: 1S

i: COP

ne: inadvertent.haver

(ya): (GEN)

hugo: brother

I have a brother.
I am "brothered"

If you haven’t seen it already, I can highly recommend François’s paper on word classes in Hiw. (It’s probably one of the most enlightening papers I’ve ever read; I’ve already recommended it once on the ZBB.) For me it really solidified this notion of zero-derivation vs one word class, and in particular I really like the way François approaches syntactic tests as a way to distinguish word classes. (Basically, each word class can occur in a different set of syntactic environments. Even if, say, nouns and verbs can regularly appear in exactly the same environments, they can be told apart if there are other environments where only one or the other occurs. To truly unify nouns and verbs, you would then need to make them completely indistinguishable.)

Thank you, thank you! I have a feeling I may have read this a while back and then lost and forgotten it as it seems familiar. I think I'm going to go for a night-walk in the forest instead of reading it now though.

bradrn · Post by **bradrn** » Fri Jul 02, 2021 12:49 am

Imralu wrote: ↑Thu Jul 01, 2021 4:32 pm
bradrn wrote: ↑Thu Jul 01, 2021 10:25 amYes, I find that Excel etc. are pretty bad at this sort of thing. The grid format is really not well suited for things like this, which are best communicated through long-form explanations and examples. Possibly you might like SIL Toolbox — I use it for my dictionaries, and the main view is somewhat Excellish.
Thanks, yeah, I really got turned off SIL things a long time ago. Not only did I have really unexpected formatting irregularities throughout my file (some entries had different font sizes or types for no apparent reason), a file got corrupted one time and only recent back-up I had saved was somehow also corrupted.

Huh, was that with SIL Toolbox? Because I’ve never had a problem with that one. (And it stores its data in plain text files, so I’m not sure how it would be possible for those to get corrupted.)

Thanks for the info! But I suppose I was more interested in knowing: how can there be ergative languages in a monovalent conlang, when the ergative marks the subject of bivalent verbs?
Ah, yes. Well, I said there is essentially an open class of ergative markers. For example, if all that was ever used was g "cause", you could just analyse that as an ergative marker rather than a verb and analyse this sentence as one single transitive sentence with ergative alignment. .

Goož magaz iddauz.

g-
ERG-

o=
DEF(c)=
ž
1S

magaz
fall.over

i=
DEF(e)=
dauz
tree

I felled the tree.

However, I could also say Doož magaz iddauz, "I felled the tree by hand". Perhaps you could call d- a marker of the "manual ergative" → "manergative" case. There's essentially no limit though:

Boož magaz iddauz. "I kicked the tree over." → oož is marked in the "pedergative" case
Tšqanagoož magaz iddauz. "I sneezed the tree over." → oož is marked in the sternutergative case! (Not a single hit on Google. Disappointing!)

My point was basically that verb phrases like these basically fill the role of the ergative case while also providing additional information. G, d b and tšqanag are all verbs, not merely types of ergative case markers and as such, the things they mark are additional clauses. (D'oož, for example, is an entire valid sentence on its own basically just meaning "I do something with my hands", "I use my hands".) As verbs, they can also be embedded within a subject phrase, such as ottšqanag "the (C-ranked) one who sneezes", and the order of consistutents in the clause tšqanag oož "I sneeze" could also be reversed to create ž'ottšqanag "the one who sneezes is me", just as magaz iddauz "the tree fell over" can be reversed to dauz immagaz "the thing that fell over was a tree". Tšqanag ottšqanag. Dauz iddauz. Magaz immagaz. (The sneezer sneezes. The tree is a tree. The faller-over-er falls over."

OK, so are you saying that these aren’t actually ergative markers, they’re just verbs which often occur in places where you would expect an ergative to go? In that case, I would avoid calling them ‘ergative’. That term has been misused enough already.

‘Word’ was probably the wrong term to use here — I should have said ‘conventionalised expression’ or something more like that. Your example of ‘mass shooting’ is a good one: that’s something which happens a lot in the US, so Americans now have a conventionalised expression for such an event. Contrast this to my idiolect: in Australia we have no mass shootings, and so to me that expression sounds no more conventional than any other.
Yeah. I'm also Australian, but I would say "mass shooting" is part of my idiolect as a loan from US English ... although I suppose just about everything could count as a loan in an idiolect.

Now that it’s not 1am here and I can think properly, let me be more clear: I recognise the term ‘mass shooting’ as a conventionalised expression for a certain event, and I might even use it myself if telling someone else about this horrible thing I saw in the newspaper (which itself uses the term when reporting on goings-on in the US), but if—God forbid—one were to happen somewhere in Sydney, I don’t think it would be the first term to spring to mind, and I’d probably consider it a special case of a ‘tragedy’.

Now that’s a point I hadn’t considered. In this case, I’m not sure what would end up happening.
Language evolution I guess, or the development of specific cultural norms. For example, on the internet in German, most people use informal address, du/ihr etc., on forums where they would be likely to use Sie if they spoke to strangers in real life.

That is interesting. Do you have any idea why this is? (And I’m assuming it’s not because Germans look down on random internet strangers…)

Possibly it’s more pragmatic than semantic. This isn’t really an area I know much about. I do know that question words would be non-specific though.
Well, there are actually many possible justifications for using definite/indefinite, specific/non-specific marking as far as I can see it. For example, you could justify the definite because the listener is assumed to know the identity
Code: Select all
                               | Listener assumed to  | Listener not assumed   |
                               | know which one       | to know which one      |
-------------------------------|----------------------|------------------------|
       Speaker knows which one:| definite specific    | indefinite specific    |
-------------------------------|----------------------|------------------------|
Speaker doesn't know which one:| definite nonspecific | indefinite nonspecific |
-------------------------------|----------------------|------------------------|
Because I only have three marked categories ...

definite specific: V=

indefinite specific: VqV=

indefinite nonspecific: VhV=
..., the remaining square (definite, nonspecific) has to share the marker of one of the others.

My understanding is that no natlang has such a system, precisely because it leaves a gap. The usual system seems to be definite specific/indefinite specific/non-specific, with the non-specific article unmarked for definiteness.

(This admittedly is based on a sample of… um, no languages, because I haven’t investigated this topic much. But I’m pretty sure I read somewhere that this is the usual division. Possibly in a grammar of Hausa? Anyway, I’d be interested in hearing any counterexamples if you have any.)

Maybe I'll instead come up with a marking for that fourth quadrant and use it more or less as the interrogative thing and that would mean that I could use a verb like "be person" with it for "who" instead of having a specific "be who". Perhaps it should also be external to the rank system because it is likely to be quite hard to pick a rank for an unknown entity.

I’d be interested to see if you can get this to work. I imagine that something like this would have the definite nonspecific article fusing with the verb to form a single interrogative series, precisely because such a combination would only be used as an interrogative.

This is an idea I find interesting. It seems like natlangs regularly try to get rid of their verb class, by making it very small (or even closed) and sending the remainder over to a more nouny class, but I haven’t yet found one with a small or closed noun class, with most ‘nouns’ pushed into a more verby class. I’m convinced this reflects some basic cognitive property of the human brain — we prefer to think about things rather than actions, or something like that.
I'm curious what natlangs you're referring to. I know Japanese can more or less be said to have a closed verb class, but it's still not that small. There are also a lot of languages that make frequent use of ... I forget what they're called ... verbs like etmek in Turkish, and I guess that's mostly what Japanese does, but I've never seen one of those languages that comes close to getting rid of the distinction.

Closed-verb-class languages are a favourite of mine. They’re mostly attested from New Guinea, particularly Trans–New Guinea: Kalam is my favourite with ~200 verbs, but also Kobon, Oksapmin, Eipo and the entire Chimbu–Wahgi branch have closed verb classes. Outside TNG, the Yam language Komnzo also has a closed verb class; the Skou language Wutung has only 32 monomorphemic verb roots, though interestingly these are not a closed class as they include a Tok Pisin loanword. They’re also very common in northern Australia: Jingulu can be said to have only three verbs, though that depends on your definition of ‘verb’. The West African languages Eʋegbe and Kanuri also have closed verb classes, as do the Barbacoan language Tsafiki and the Sino–Tibetan language Khaling. And then as you say there’s Japanese, though I’m not sure how much that counts.

These languages tend to compensate by making heavy use of light verb constructions: that is, they’ll say something like ‘I had a walk’ for ‘I walked’, turning what would be a verb in other languages into a noun. (Or a coverb, this being a noun which is restricted to light verb constructions: quite a few of those languages I listed above have open coverb classes. This strategy is particularly prominent in northern Australia.) Kalam is unusual in that it prefers to use SVCs:

Bin: woman

ak: the

ñapanŋaŋ: child

anup: him

sop: soap

ak: the

wik: rub

d: touch

ap: come

tan: descend

d: touch

ap: come

yap: ascend

g-e-b.: do-PRS.PROG-3s.

The woman is soaping her child.

I've kind of had the impression it's more the opposite in natlangs that get close to it, such as in Lushootseed where you could regard ʔux̌ʷ as a verb meaning "go" and sbiaw as a verb meaning "be coyote" and both can be nominalised by putting them in essentially what can be regarded as a relative clause:

ti
SPEC

sbiaw
be.coyote

= (the) one that is a coyote → the/a coyote

ti
SPEC

ʔux̌ʷ
go

= (the) one that goes → the/a "goer"

This is probably a massive oversimplification of Lushootseed grammar which looks terrifyingly intimidating in most example sentences, and apparently some verbs do need to be marked in ways nouns don't when used as arguments ... although ʔux̌ʷ doesn't need to be ... meaning it belongs to the lexical class of nouns, I guess?

Dixon says the difference is more subtle: ‘van Eijk and Hess (1986) point out that when a verb is head of an NP it cannot take possessive affixes, whereas a noun can’. Sutton (2004:199) says that ‘Nouns contrast with verbs in that they may take the possessive affixes and are
not (or not usually) inflected for the progressive aspect, while verbs (with a few exceptions) must be nominalized to take the possessives, and they are (with a few exceptions) inflected for the progressive aspect … Noun roots differ also from verb roots in their usual shapes’, though this definition seems somewhat too exception-ridden for my taste. (I suppose you could say that these exceptions are multicategorial words which are members of both the verb and the noun class.)

Salishan languages are however quite different to what I was talking about. I was referring to languages which have a clear difference between verbs and nouns, but have a substantially reduced verb class; however Salishan languages are quite different in that they have both verbs and nouns, but obsc

The funny way that this causes accusative-like roles to coalesce with genitive-like roles also means that passive essentially coalesces with "have" because, for example, people who are killed have killers, things that are eaten have eaters etc.

This is not so unexpected: English uses ‘have’ for the perfect, which is of course extremely similar to the passive as a category. (IIRC Quite a few Indo–Iranian languages have become ergative in the past tense by unifying their passive with their perfect.)

Imralu · Post by **Imralu** » Fri Jul 02, 2021 3:38 pm

bradrn wrote: ↑Fri Jul 02, 2021 12:49 am
Imralu wrote: ↑Thu Jul 01, 2021 4:32 pm
bradrn wrote: ↑Thu Jul 01, 2021 10:25 amYes, I find that Excel etc. are pretty bad at this sort of thing. The grid format is really not well suited for things like this, which are best communicated through long-form explanations and examples. Possibly you might like SIL Toolbox — I use it for my dictionaries, and the main view is somewhat Excellish.
Thanks, yeah, I really got turned off SIL things a long time ago. Not only did I have really unexpected formatting irregularities throughout my file (some entries had different font sizes or types for no apparent reason), a file got corrupted one time and only recent back-up I had saved was somehow also corrupted.
Huh, was that with SIL Toolbox? Because I’ve never had a problem with that one. (And it stores its data in plain text files, so I’m not sure how it would be possible for those to get corrupted.)

Oh, no. It was in something else by SIL. I remember that I tried to open the corrupted files in Notepad, hoping that the information was stored as text, but it was just that wall of symbols you get if you open any random file type in Notepad. OK, I think I'll check it out. My Excel / Notepad systems have been working very well for me. I just have to think carefully and then be consistent with formatting. They don't come out very aesthetically presentable though, but they're just for me. If they were more aesthetic, then I'd probably enjoy using them more though.

OK, so are you saying that these aren’t actually ergative markers, they’re just verbs which often occur in places where you would expect an ergative to go? In that case, I would avoid calling them ‘ergative’. That term has been misused enough already.

Yeah, I think we're talking past each other a bit here. I'm definitely not calling them ergative markers. I've just compared them with ergative markers.

Now that it’s not 1am here and I can think properly, let me be more clear: I recognise the term ‘mass shooting’ as a conventionalised expression for a certain event, and I might even use it myself if telling someone else about this horrible thing I saw in the newspaper (which itself uses the term when reporting on goings-on in the US), but if—God forbid—one were to happen somewhere in Sydney, I don’t think it would be the first term to spring to mind, and I’d probably consider it a special case of a ‘tragedy’.

I'm sure if it happened regularly in Australia, the media would come up with some bizarre phrase for it like "coward gunning".

That is interesting. Do you have any idea why this is? (And I’m assuming it’s not because Germans look down on random internet strangers…)

Hmm, it's not about looking down on people. Like, if you call a worker in a bank "du", it doesn't feel like you're looking down on them, more like you're just being uncomfortably familiar. The nametags of supermarket cashiers literally say stuff like "Fr. Schmidt". The formality feels really weird to me and also counterproductive because when you hear the staff talking to each other, whoops, now you know their full name. (Don't get me started on apartments not having numbers and just having your name on the buzzer, door and letterbox. This place is like a paradise for stalkers.) I can't really give much of an answer for why "Sie" is not so common on the internet other than ... it just doesn't feel the same as that formal, distanced language you use with strangers in real life. Young people talking to other strangers about the same age use "du" as well ... in Apple stores here, they have a policy of using "du" with everyone.

My understanding is that no natlang has such a system, precisely because it leaves a gap. The usual system seems to be definite specific/indefinite specific/non-specific, with the non-specific article unmarked for definiteness.

I can only really imagine it being likely in a natlang where definiteness and specificity are marked independently of each other and even then, it still seems unlikely. The only time I can imagine it making any sense is for question words and maybe also those situations where someone has mentioned someone in a vague way so you don't know their identity (like "Does this person have a name?")

Tongan marks specificity with the choice of article:

ha tangata = a (nonspecific) person
(h)e tangata = a (specific) person

And marks definiteness with final stress:

(h)e tangatá = the person

I don't think ?ha tangatá is possible at all, but if it were my conlang, I'd be tempted to use ha with final stress to mean "which ... ?"

By the way, I was reading a grammar of Pacific Englishes (from memory, focusing on research from Samoa, Solomon Islands and Fiji) and the conclusion about article use was that "the" is more of a specific article than a definite article in these varieties of English.

I’d be interested to see if you can get this to work. I imagine that something like this would have the definite nonspecific article fusing with the verb to form a single interrogative series, precisely because such a combination would only be used as an interrogative.

Yeah, the challenge I can see would be that, if this marking is all I'm using to form question words, as I don't currently have a way to embed a noun phrase within a verb phrase (as I do in Iliaqu), so I couldn't really use these within the predicate (unless I do that). So, to ask "who are you?", "you" would have to be the predicate and "who" the subject of the clause. Reversing it so that "you" is the subject and "person" is the predicate, the only place the "which"-marking could be, as a nominaliser, would be on "you", so that would only allow "which you is a person". (That sounds like a very bizarre but cutting criticism.

) If I instead simply have a verb for "who" or a means of deriving question words within a verb phrase (not relying on the nominaliser), then it's easy and they can be in whichever clause element makes sense. (Potentially gravitating towards the predicate as that is more focusy and the subject is more topicky, although I haven't been super consistent with that.)

Closed-verb-class languages are a favourite of mine. They’re mostly attested from New Guinea, particularly Trans–New Guinea: Kalam is my favourite with ~200 verbs, but also Kobon, Oksapmin, Eipo and the entire Chimbu–Wahgi branch have closed verb classes. Outside TNG, the Yam language Komnzo also has a closed verb class; the Skou language Wutung has only 32 monomorphemic verb roots, though interestingly these are not a closed class as they include a Tok Pisin loanword. They’re also very common in northern Australia: Jingulu can be said to have only three verbs, though that depends on your definition of ‘verb’. The West African languages Eʋegbe and Kanuri also have closed verb classes, as do the Barbacoan language Tsafiki and the Sino–Tibetan language Khaling. And then as you say there’s Japanese, though I’m not sure how much that counts.

These languages tend to compensate by making heavy use of light verb constructions: that is, they’ll say something like ‘I had a walk’ for ‘I walked’, turning what would be a verb in other languages into a noun. (Or a coverb, this being a noun which is restricted to light verb constructions: quite a few of those languages I listed above have open coverb classes. This strategy is particularly prominent in northern Australia.) Kalam is unusual in that it prefers to use SVCs:

Yes, light verbs! That's the term I had forgotten. As for Skou verbs, pronouns are a basically closed class and yet English still loaned they-them-their from Old Norse. I think all closed classes are at most "closed-ish". I've also seen prepositions described as a closed class, but there are quite a few loan prepositions in English. Prepositions seem less closed than pronouns.

Bin
woman
ak
the
ñapanŋaŋ
child
anup
him
sop
soap
ak
the
wik
rub
d
touch
ap
come
tan
descend
d
touch
ap
come
yap
ascend
g-e-b.
do-PRS.PROG-3s.

The woman is soaping her child.

Wow! "Touch come descend touch come ascend" is pretty cool. It reminds me of my compound word for "(one who) rub(s)" in Zyange: touchₐ-moveₐ.

Dixon says the difference is more subtle: ‘van Eijk and Hess (1986) point out that when a verb is head of an NP it cannot take possessive affixes, whereas a noun can’. Sutton (2004:199) says that ‘Nouns contrast with verbs in that they may take the possessive affixes and are
not (or not usually) inflected for the progressive aspect, while verbs (with a few exceptions) must be nominalized to take the possessives

So, I can't say "my goer" with the same structure as I could say "my coyote"?

Or are they talking about "my going", in which case: could I not use the same structure to say "my being a coyote"?

and they are (with a few exceptions) inflected for the progressive aspect …

Could that not just be a function of their aktionsart, similar to how state verbs generally don't get inflected for progressive aspect in English? Most concepts that typically end up as nouns have a more durative, stative aktionsart: a tree is a tree for a long time. A faller-over is generally only a faller-over for a brief time and that's probably why we generally say things fall over than that they are fallers-over (and why such a noun is awkward as fuck in English). Something like "grow" can of course have just as long a duration as "be a tree", but "grow" focuses on a dynamic change over time whereas "be a tree" does not. It's not denying that trees change over time, but it's not specifically referring to that, so that's why "tree" seems like a nouny concept and "grow" seems like a verby concept. Being a tree is a state. Growing is an action. I feel like deciding part of speech based on progressive marking is the kind of thing that might lead you to conclude that "like" and "enjoy" are different parts of speech because "enjoy" is frequently given progressive marking and "like" (if we are prescriptivist at least) is not. It's merely a semantic difference of aktionsart: like is more "statey", enjoy, although very similar, is semantically more "actiony".

This is not so unexpected: English uses ‘have’ for the perfect, which is of course extremely similar to the passive as a category. (IIRC Quite a few Indo–Iranian languages have become ergative in the past tense by unifying their passive with their perfect.)

Yeah, I think it's more just strange because I didn't consciously plan it that way at all. I came up with ne so I can have passives and more or less discovered that it means "have" without that being my intention. I already had two other words for "have" (za "holder" and nyu "owner"), so I was pretty delighted to realise I actually have three. I didn't really find it that weird, linguistically. It makes logical sense in my language, it's just more like ... drawing the beginning of a path and just following it to its natural conclusion and ending up somewhere you didn't expect.

Travis B. · Post by **Travis B.** » Fri Jul 02, 2021 5:05 pm

Imralu wrote: ↑Fri Jul 02, 2021 3:38 pm As for Skou verbs, pronouns are a basically closed class and yet English still loaned they-them-their from Old Norse.

Note also that just about every English dialect has innovated some sort of 2nd pl. pronoun, ranging from you guys, to y'all, to you lot, to youse, and so on.

keenir · Post by **keenir** » Fri Jul 02, 2021 11:45 pm

Imralu wrote: ↑Fri Jul 02, 2021 3:38 pm . Reversing it so that "you" is the subject and "person" is the predicate, the only place the "which"-marking could be, as a nominaliser, would be on "you", so that would only allow "which you is a person". (That sounds like a very bizarre but cutting criticism. )

they've been exposed to Among Us memes, maybe?

(sorry)

It makes logical sense in my language, it's just more like ... drawing the beginning of a path and just following it to its natural conclusion and ending up somewhere you didn't expect.

keep up that good work!

bradrn · Post by **bradrn** » Sat Jul 03, 2021 8:18 am

Imralu wrote: ↑Fri Jul 02, 2021 3:38 pm
bradrn wrote: ↑Fri Jul 02, 2021 12:49 am
Imralu wrote: ↑Thu Jul 01, 2021 4:32 pm Thanks, yeah, I really got turned off SIL things a long time ago. Not only did I have really unexpected formatting irregularities throughout my file (some entries had different font sizes or types for no apparent reason), a file got corrupted one time and only recent back-up I had saved was somehow also corrupted.
Huh, was that with SIL Toolbox? Because I’ve never had a problem with that one. (And it stores its data in plain text files, so I’m not sure how it would be possible for those to get corrupted.)
Oh, no. It was in something else by SIL. I remember that I tried to open the corrupted files in Notepad, hoping that the information was stored as text, but it was just that wall of symbols you get if you open any random file type in Notepad. OK, I think I'll check it out. My Excel / Notepad systems have been working very well for me. I just have to think carefully and then be consistent with formatting. They don't come out very aesthetically presentable though, but they're just for me. If they were more aesthetic, then I'd probably enjoy using them more though.

Well, if it’s working for you, there’s really no need to change. The only reason I suggested it was because MDF (the format Toolbox uses) is far more flexible than Excel spreadsheets.

OK, so are you saying that these aren’t actually ergative markers, they’re just verbs which often occur in places where you would expect an ergative to go? In that case, I would avoid calling them ‘ergative’. That term has been misused enough already.
Yeah, I think we're talking past each other a bit here. I'm definitely not calling them ergative markers. I've just compared them with ergative markers.

Ah, sorry, I must have misunderstood you then.

Now that it’s not 1am here and I can think properly, let me be more clear: I recognise the term ‘mass shooting’ as a conventionalised expression for a certain event, and I might even use it myself if telling someone else about this horrible thing I saw in the newspaper (which itself uses the term when reporting on goings-on in the US), but if—God forbid—one were to happen somewhere in Sydney, I don’t think it would be the first term to spring to mind, and I’d probably consider it a special case of a ‘tragedy’.
I'm sure if it happened regularly in Australia, the media would come up with some bizarre phrase for it like "coward gunning".

Yes indeed, I’m sure they would!

That is interesting. Do you have any idea why this is? (And I’m assuming it’s not because Germans look down on random internet strangers…)
Hmm, it's not about looking down on people. Like, if you call a worker in a bank "du", it doesn't feel like you're looking down on them, more like you're just being uncomfortably familiar. The nametags of supermarket cashiers literally say stuff like "Fr. Schmidt". The formality feels really weird to me and also counterproductive because when you hear the staff talking to each other, whoops, now you know their full name. (Don't get me started on apartments not having numbers and just having your name on the buzzer, door and letterbox. This place is like a paradise for stalkers.) I can't really give much of an answer for why "Sie" is not so common on the internet other than ... it just doesn't feel the same as that formal, distanced language you use with strangers in real life. Young people talking to other strangers about the same age use "du" as well ... in Apple stores here, they have a policy of using "du" with everyone.

So do you think du is expanding its range at the expense of sie?

(More generally, it looks like I need to read up on politeness marking — I think I may have totally misunderstood the concept.)

I’d be interested to see if you can get this to work. I imagine that something like this would have the definite nonspecific article fusing with the verb to form a single interrogative series, precisely because such a combination would only be used as an interrogative.
Yeah, the challenge I can see would be that, if this marking is all I'm using to form question words, as I don't currently have a way to embed a noun phrase within a verb phrase (as I do in Iliaqu), so I couldn't really use these within the predicate (unless I do that). So, to ask "who are you?", "you" would have to be the predicate and "who" the subject of the clause. Reversing it so that "you" is the subject and "person" is the predicate, the only place the "which"-marking could be, as a nominaliser, would be on "you", so that would only allow "which you is a person". (That sounds like a very bizarre but cutting criticism. ) If I instead simply have a verb for "who" or a means of deriving question words within a verb phrase (not relying on the nominaliser), then it's easy and they can be in whichever clause element makes sense. (Potentially gravitating towards the predicate as that is more focusy and the subject is more topicky, although I haven't been super consistent with that.)

I’m not sure I understand your problem. Why would you need to embed a noun phrase within a verb phrase to apply a definite nonspecific article?

(Oh, and I’ve just realised I made a mistake in my previous post: I meant to say that the article would fuse with the noun, not the verb.)

Closed-verb-class languages are a favourite of mine. They’re mostly attested from New Guinea, particularly Trans–New Guinea: Kalam is my favourite with ~200 verbs, but also Kobon, Oksapmin, Eipo and the entire Chimbu–Wahgi branch have closed verb classes. Outside TNG, the Yam language Komnzo also has a closed verb class; the Skou language Wutung has only 32 monomorphemic verb roots, though interestingly these are not a closed class as they include a Tok Pisin loanword. They’re also very common in northern Australia: Jingulu can be said to have only three verbs, though that depends on your definition of ‘verb’. The West African languages Eʋegbe and Kanuri also have closed verb classes, as do the Barbacoan language Tsafiki and the Sino–Tibetan language Khaling. And then as you say there’s Japanese, though I’m not sure how much that counts.

These languages tend to compensate by making heavy use of light verb constructions: that is, they’ll say something like ‘I had a walk’ for ‘I walked’, turning what would be a verb in other languages into a noun. (Or a coverb, this being a noun which is restricted to light verb constructions: quite a few of those languages I listed above have open coverb classes. This strategy is particularly prominent in northern Australia.) Kalam is unusual in that it prefers to use SVCs:
Yes, light verbs! That's the term I had forgotten. As for Skou verbs, pronouns are a basically closed class and yet English still loaned they-them-their from Old Norse. I think all closed classes are at most "closed-ish". I've also seen prepositions described as a closed class, but there are quite a few loan prepositions in English. Prepositions seem less closed than pronouns.

This is a very interesting point — at what level of loaning can you say a class is closed? But I suppose you can answer this by comparison to other languages. Here, Kalam provides a relevant contrast — it has ~130 verbs and none of them are borrowed (though speakers reanalyse SVCs as single verbs occasionally), whereas Wutung has 27 containing one loan. Percentage-wise at least, the Wutung verb class is far more open than the Kalam one. And yes, it’s extremely small, but Papuan languages generally tend to have small verb classes.

Bin
woman
ak
the
ñapanŋaŋ
child
anup
him
sop
soap
ak
the
wik
rub
d
touch
ap
come
tan
descend
d
touch
ap
come
yap
ascend
g-e-b.
do-PRS.PROG-3s.

The woman is soaping her child.
Wow! "Touch come descend touch come ascend" is pretty cool. It reminds me of my compound word for "(one who) rub(s)" in Zyange: touchₐ-moveₐ.

Yeah, Kalam does a lot of that sort of thing. Some other good examples:

bradrn wrote: ↑Mon Aug 24, 2020 10:15 pm gos tmey nŋ- thought bad perceive ‘dislike, hate something’

mnm ag ask ay- talk say avoid stabilize ‘leave or avoid (a topic)’

ap tan ap yap g-p-ay tam come ascend come descend do-HAB-3PL junction ‘crossroads’ (i.e. ‘junction where they come up and go down’; a nominalization)

ñag jw yok- shoot withdraw displace ‘rout (the enemy) in a war’

pwŋy pag yk- impinge disturb open ‘prise something open or free’

d nŋ- touch perceive ‘feel something (by touching)’

kwy ap-INFL nŋ- odour come perceive ‘smell something’

It helps that most Kalam verbs are really vague: tk-, for instance, means ‘cross (a divide), sever, separate, interrupt, cut off, transform, suddenly become (night, day, clear), have a child’, amongst other things.

Imralu wrote:
Dixon says the difference is more subtle: ‘van Eijk and Hess (1986) point out that when a verb is head of an NP it cannot take possessive affixes, whereas a noun can’. Sutton (2004:199) says that ‘Nouns contrast with verbs in that they may take the possessive affixes and are
not (or not usually) inflected for the progressive aspect, while verbs (with a few exceptions) must be nominalized to take the possessives
So, I can't say "my goer" with the same structure as I could say "my coyote"?

Or are they talking about "my going", in which case: could I not use the same structure to say "my being a coyote"?

and they are (with a few exceptions) inflected for the progressive aspect …
Could that not just be a function of their aktionsart, similar to how state verbs generally don't get inflected for progressive aspect in English? Most concepts that typically end up as nouns have a more durative, stative aktionsart: a tree is a tree for a long time. A faller-over is generally only a faller-over for a brief time and that's probably why we generally say things fall over than that they are fallers-over (and why such a noun is awkward as fuck in English). Something like "grow" can of course have just as long a duration as "be a tree", but "grow" focuses on a dynamic change over time whereas "be a tree" does not. It's not denying that trees change over time, but it's not specifically referring to that, so that's why "tree" seems like a nouny concept and "grow" seems like a verby concept. Being a tree is a state. Growing is an action. I feel like deciding part of speech based on progressive marking is the kind of thing that might lead you to conclude that "like" and "enjoy" are different parts of speech because "enjoy" is frequently given progressive marking and "like" (if we are prescriptivist at least) is not. It's merely a semantic difference of aktionsart: like is more "statey", enjoy, although very similar, is semantically more "actiony".

I really wouldn’t know about any of this; Salishan languages aren’t my strong point.

This is not so unexpected: English uses ‘have’ for the perfect, which is of course extremely similar to the passive as a category. (IIRC Quite a few Indo–Iranian languages have become ergative in the past tense by unifying their passive with their perfect.)
Yeah, I think it's more just strange because I didn't consciously plan it that way at all. I came up with ne so I can have passives and more or less discovered that it means "have" without that being my intention. I already had two other words for "have" (za "holder" and nyu "owner"), so I was pretty delighted to realise I actually have three. I didn't really find it that weird, linguistically. It makes logical sense in my language, it's just more like ... drawing the beginning of a path and just following it to its natural conclusion and ending up somewhere you didn't expect.

Yeah, it’s really nice when that happens!

Zompist Bboard Again

Balog: a strictly monovalent conlang

Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang

Re: Balog: a strictly monovalent conlang