Another triconsonantal conlang

[Herr Dunkel]

But
what
You confuse me. I shall take my leave

[Plusquamperfekt]

But
what
You confuse me. I shall take my leave

OK again, DePaw asked me why I have <š> instead of <z> and then I wrote him that I do not want to have a voicing contrast in my root system, since I would get mixed clusters with voiceless and voiced consonants, which are not only a little difficult to pronounce, but it would be very likely that there would be an assimilation resulting in many homophones.

For example, if I had /pt/, /pd/, /bt/ and /bd/, many people would pronounce all of them [pt].

[Herr Dunkel]

Ah I misunderstood.
Still, you could without an issue do like 600 verbs.

[Astraios]

Could I just point out that Modern Hebrew is a triconsontal language which distinguishes all four of /pt/ /pd/ /bt/ /bd/.

[Plusquamperfekt]

Could I just point out that Modern Hebrew is a triconsontal language which distinguishes all four of /pt/ /pd/ /bt/ /bd/.

Yes, but you forget zet my natiff languache hess final devoicing. For me it's difficult to pronounce them differently

[Astraios]

So don't have final clusters :p

[KathTheDragon]

Btw, triconsonantal languages don't do what you're proposing. Just, no. This looks, at best, one of those scientificy artificial auxlangs pretending to be a triconsonantal language.

I suggest you read this thread for starters.

[Salmoneus]

And I suggest that EVERYBODY CALM THE FUCK DOWN.

OK, there are clearly some issues with plusquam's idea here, and it's fair enough to point them out in a civil way.
But the purpose of the forum is for people to share their work, NOT for everybody to try to prove how badass and smart they are by competing to shout people down the loudest.

[Linguist Wannabe]

Caveat: I know virtually nothing about the Afro-Asiatic language family other than the basics that everyone (with an interest in linguistics) knows about. So much so that I have never even embarrassed myself trying to make a triconsonantal root language.

If you don't want to have clusters of mixed voicing, couldn't you just have a rule that inserts a non-phonemic schwa [ə] between two clusters of mixed voicing. I seem to recall that some Semitic language does that (not sure if it's Arabic or Hebrew or what).

You could also use phonemic aspiration.

Also, even 720 verb roots might be enough. I think some (polysynthetic) North American languages only have about 800 verb roots. You will have your work cut out for you developing enough morphology to make a fully expressive language from such a small number of roots though. Maybe you should do some reading on polysynthetic languages.

Also, maybe this is repeating what other people have said, but you might want to think about how the language has evolved. Mecislau's post here viewtopic.php?f=10&t=9392&start=25 (that begins with "Alright, part one of the origin of Semitic triconsonantal roots... ") would be a good starting point.

[Yng]

He said specifically that he doesn't want this language to be naturalistic or have diachronics, so that thread's probs not that relevant.

[Hallow XIII]

That is all well and good but you should look at his first post where it said that this conlang is not intended to be naturalistic.

But it also reaffirms what I said about the ZBB not being a good place for this sort of thing. The fact of the matter is, it is basically only good for feedback on that exact parameter, and, furthermore, people here almost instinctively react to this sort of language with the assumption that the maker knows nothing whatsoever.

[Linguist Wannabe]

That is all well and good but you should look at his first post where it said that this conlang is not intended to be naturalistic.

But it also reaffirms what I said about the ZBB not being a good place for this sort of thing. The fact of the matter is, it is basically only good for feedback on that exact parameter, and, furthermore, people here almost instinctively react to this sort of language with the assumption that the maker knows nothing whatsoever.

I didn't mean to imply "that the maker knows nothing whatsoever". I just gave a link to something that I think I would be very interested to read if I was into making Semitic-style triconsonantal root conlangs.

[Plusquamperfekt]

Hey people, thank you very much for the link. I've just read what Hallow wrote about the types of feedback in this forum. For this reason I've come to the conclusion that I should have explained why I am trying to create a conlang about which even I say that it is not very naturalistic and not very aesthetic either, since the phonology is rather uninteresting.

I would like to try out how efficient a language can be, i.e. how much can be expressed with very little text. I know that there have been already other conlangs with this objective, but I would like to test to which degree translations in a conlang made my myself would be shorter than original versions in English for example.

[Linguist Wannabe]

I would like to try out how efficient a language can be, i.e. how much can be expressed with very little text. I know that there have been already other conlangs with this objective, but I would like to test to which degree translations in a conlang made my myself would be shorter than original versions in English for example.

If you want to do that, are you allowed to tweak the phonology? (e.g. you could fit more "information" in a shorter "space" if you made any one of the following phonemic: Vowel length, consonant gemination, consonant palatalisation or velarisation, tone, more voicing / phonation / aspiration contrasts etc.)

Also, have you thought about making your language isolating and / or pro-drop? Dropping things such as pronouns and time markers if they are obvious from the context is one way to be able to make shorter texts possible in a conlang.