Conlang thread

Have any of you tried learning a constructed language? I remember being autistic enough to try lojban at uni, I actually had a friend who did the same and we would have a circlejerk.

Remember when we had /lang/ and some random asshole complained about it and admin kike deleted it?

No because I'm not retarded.

Two dweebs in my year learnt klingon though. Very surprisignly one of them dropped out.

My autistic special interest wants to reconstruct Proto Anglo-Frisian

>>4 That's not completely a conlang though. Historic languages have value.

>>1
Lojban is not perfect, an AI(neural network) would design a better language without
any preconceived notions: syntax-optimized easy-as-esperanto as-precise-as-lojban as-dense-as-ithkuil,etc.
Maybe even add some viral ideas(making words sounds reflect semantic values more precisely) to insert utility loanwords into English
to allow it naturally replace complex slang.
English is quite ripe for neologistic hostile takeover, i'd say such a
language can be subverted(like Newspeak) and even 'upgraded' with
a sufficiently smart AI trained to devise vocabulary with a basis of more
effective language(not Snow Crash fan, but biologically we're monkeys trying to devise vocalizations of complex mental machinery and monkeys are not pinnacle of technology - we're not introspecting our brain but evolving means to do so..an AI brain mapper would allow precise semantic 1:1 map to concept-word dictionary foundation to create a hyper-precise language referencing the exact brain mental process/network activated by the word)

>>6
Don't give mentifex any ideas

>>7
I don't give out any concrete ideas, just hints at what design
i'd like to see in future: software-defined languages,
logical brain-computer interfaces that map exactly without 'interpretation'
and language evolution guided by AI models not hobby linguists
spending 10 hours to think of a precise meaning of 'set'.
Instead of programming AI natural language processing,
upgrading the language to a level of precision inherent in programming languages structures where AST of text
can be modeled as an exact graph with 0 ambiguous content(no garden path sentence, no alternate meaning, context-free structure).
Such language will be superior within human-to-human communications
too by eliminating dependencies on context and reducing semantic space of meanings to exact concept references.

>>5
Where do you draw the line between reconstructed and constructed languages? Do they fall in the conlang category? Some proto languages have historical atestations, but 99% of the language is made up.

>>9
Depends on how much you can verify how accurate a historic language is. I guess the cut is where it stopps being useful to interpret the world.

The only time I use old languages is to understand new words (eg "define word" in google then it gives you the cool tree of origin) or read old texts. If there exists text and the we know somewhat accurately what the symbols and words mean the lang probably is useful.

An interesting example would be "In the beginning was the Word", the first line of the bible. Doesn't make much sense in English but if you know that the pre-translated word was logos not lexis and know what it meant in Greek at the time it makes sense. When I first heard that a new world opened.

No idea if my understanding is correct but I found out that old civilisations had less separation from word and thought. The Egyptians didn't write down the bad parts of their methology because they believed it had power. So if you're interested in that you've to search 3rd party sources. Or the Greeks thought of logos as kinda the flow of the world. I'd have to invest more energy to understand, I think the concept isn't really around anymore so one has to describe it instead of translate. Of course, nowadays, depending on the ideology, people interpret logos differently.

>>6,8 So you have to construct an accurate model of a human brain before the AI can make a better language? At this point communication through soundwaves isn't neccesary anymore is it?

>>11
Humans need sound-based languages, it will be necessary until everyone is connected electronically(if ever).

can some russian or czech or something please tell me about the validity of interslav? If I learn interslav, can I read russian books? Will ukrainian qts understand me?

>>12 If we model the brain (if ever) we surely are able to send it signals directly as well.

>>14
..and how would you send something deep into the brain?

>>15 If you've already modeled the brain you know which impulses to send.

>>10
Sorry homes, you have reconstructed and classical languages mixed up. Ancient Greek isn't reconstructed. Ancient Greek was written down 2000 something years ago, while reconstructed languages have no written record, except in a few cases like Proto-Norse where we can say that there are gravestones and swords and whatnot with a few grammatically incomplete but nonetheless existent inscriptions. The most famous and the most useful reconstructed language is Proto-Indo-European, which is said to have been spoken before writing existed, therefore the entire thing is conjectural, and its grammar and lexicon have been revised a few times over the centuries.

Recon languages are not spoken for communication purposes, of course, their sole value is that they provide explanations for how daughter languages evolved and related languages diverged. But it's all just word games basically, probably the most complicated form of word game, as it's all made up and hypothetical but the rules are internally consistent and are able to make accurate predictions about modern languages.

>>17 That's completely retarded holy shit. Increadibly innacurate pseudo science. Please tell me this isn't a mainstream thing.

When the next trendy idea comes around it's absolutely going to be revised again

>>18
Are you retarded? PIE is the foundation of modern comparative linguistics and arguably modern linguistics as well as the resources that were used to codify it were used to make just the innovations that led to universal grammar. I highly doubt you know anything enough about this subject to refute it. But as I have said, PIE is an accurate model that accurately explains consonant and vowel shifts and morphological features across multiple generations and branches of Indo European languages. Tell me your retarded proposal that accomplishes as much.

>>19
Are you retarded? sambo!

>>19
Of course it neatly fills all the gaps. It'a artifical and made to rationalize the guesswork.

Universal grammar is another pseudo science thing. If it had any basis in reality natural lang processing software would use it. But it's all neural networks. Almost like the haughty theories are hot air and don't walk the walk.

>>13
I'm intrigued. How would I learn it?

>>21
Wow, it's almost like a computer and the human brain are two different things. The nature of reality now is determined by how well a computer can simulate it? Simulate love, motherfucker.

>>23

Simulate love

1.pheromones
2.visual models
3.vibration(e.g. pulse)
4.heat simulation
5.touch and tissue simulation
6.ai memory(NN adjusts to your expectations)
7.behavior simulation(NN trained on females)
8.Electromagnetic simulation of human organs(and possibly chakras)

Love is heart chakra radiation, but i don't know the exact frequencies
or how its going to be replicated, though i'm sure it will be replicated given sufficient advancement.

>>23 Grammar isn't real. It's a model made after the language already evolved. This is evidenced by grammar based approaches failing. nice strawman fag

>>24
Transhumanist garbage. You will never fuck a computer.

>>26
The grammar asserted by neurolinguistics is not the grammar you are taught in school. It is not prescriptive or even verbal, it is formalized logic and is computational. Prescriptive and descriptive grammars are models, the actual cognition involved in processing and creating thoughts using strict syntactical rules that you are not even conscious of while doing is not a model, your own post demonstrates that it is real. It is in the category of things that cannot be said to not exist because to even make the argument assumes its existence: you cannot form a thought without grammar, you cannot communicate without grammar; all thoughts are inherently grammatical, no matter what model you use to describe that thought process. Even your attempts at making ungrammatical sentences will follow unconscious syntactical rules.

It's not even the case that neural networks don't use construction grammar or formal grammar, where did you even get this assumption? It's so obvious you don't know what you are talking about. You are just reacting from your gut to topics you know nothing about. If you are the same person I was responding to initially before this retarded argument began, it's obvious you know nothing, you thought Greek was a proto language and you thought John 1 was the beginning of the Bible, and if you're not that person, why did you even step in with this aside? The conversation is about conlangs, and recon langs fall into this category, your ignorant opinion on linguistics doesn't have anything to do with that.

>>27

You will never fuck a computer.

Has there been any object or animal that not been fucked by man?
Rule34 applies anywhere. In particular there is little to stop a man fucking
his own computer if he deems the act attractive - the limit just isn't there.
A VR simulation that is close to a real thing would make 99.95% of man
choose the more superior option that skips dealing with screeching, manipulative femoid monkeys.

>>27
you will never fuck a waifu
https://old.reddit.com/r/WaifuSingularity/

>>27

grammar

So a pattern of neurons firing is considered grammar? I'm smart enough to recognize psychoanalysis as an inaccurate model but you seem to believe it to be logos.

ad hominem

Didn't add anything to the conversation.

>>30
You're not smart enough to contribute to the conversation in any way.

>>31
You're not smart enough to contribute to the conversation in any way.

>>32
You're smart enough to contribute to the conversation in a meaningful way.

>>33
You're smart enough to contribute to the conversation in a meaningful way.

Esperanto > Lojban