/sci/ - Science & Math

Compiled videos:

https://www.youtube.com/watch?v=y1kXCh496U0
https://www.youtube.com/watch?v=B5lzt42Tb20
https://www.youtube.com/watch?v=C0CIRCjoICA
https://www.youtube.com/watch?v=Q5ZMGBz8qgI
https://www.youtube.com/watch?v=mQmF7kbOrmE
https://www.youtube.com/watch?v=6eliANcZdkw
https://www.youtube.com/watch?v=J3TwTb-T044
https://www.youtube.com/watch?v=T4KRbENmFDk
https://www.youtube.com/watch?v=9TwQi9Wp66A
https://www.youtube.com/watch?v=6dAEE7FYQfc
https://www.youtube.com/watch?v=Nvvn-ZVdeqQ
https://www.youtube.com/watch?v=ecIlKjZYipA
https://www.youtube.com/watch?v=OCvJiqKZbz4
https://www.youtube.com/watch?v=iQwNYajSk08
https://soundcloud.com/andrew-lynch-622652011/before-speech
https://youtu.be/m3vIEKWrP9Q
https://www.youtube.com/watch?v=9AN2tbfSGm4
https://www.youtube.com/channel/UCKpLVHMEJFZsL6Fb0HocbvQ

Linguistics PhD AMA:

(1)

> How selective are these programs for admission?

penn, a top 10 program in the united states, admits ~10% of applications every year: https://live-sas-www-ling.pantheon.sas.upenn.edu/graduate/application/graduate-program-application
other depts don't have published figures i'm aware of but that's an incredibly high admissions rate for phds
> What are career prospects for linguistics phds? My advisor says to look at the LSA info, but it seems like a meme to me
depends. if you do computational, you'll make fat stacks as a tech worker. if you do something quantitative but not full on computational, you could pivot into a data science role or something and make ok money. if you don't do either of those things, you might as well have done a phd in philosophy, and should consider your PhD a 5 year passion project that won't be developing your career

>>11613569

(2)

> Are there rewarding research work and internships while you're in university?

i assume you're talking about phd programs, yes?

for the summer: again, it kind of depends on what you're doing--if you're doing pure theory, either you're at a top school where they'll summer fund you, or you'll need to find something else. if you're doing computational obviously you'll find tech internships or something in between like an internship at a military research lab or ETS. in my program (top 20 in the US overall I'd say), I think most theoretical students pick up some kind of summer job since we don't have summer funding. (The rich students, of course, don't have to.)

for during the year: you're going to be doing research, of course, year round, and if you're admitted to a phd program then that means profs want to and will work with you--at least your advisor.

> How often do linguists do field research?

it really depends. sociolinguists do the most of it overall i'd say. Some theorists do zero fieldwork, but it has been on the rise over the past 20 years as they've gotten their heads out of their fucking asses

>>11613571

(3)

> Which grammatical theories interest you?

i'm not a theorist, so take what i say with a grain of salt. i have conflicting feelings about mainline generative theory: the solutions that are dreamt up when challenging data arises stinks heavily of adding epicycles, and i wonder whether the formal apparatus available to generative linguists ought to be overhauled. philosophically, however, i think mainline generative theory is sound: clearly our mind must have something innate that helps us process language to account for the startling facts about language competence (e.g. everyone is basically the same level of competent in a language even independent of IQ (!)). if we have this thing, then it's probably most reasonable to assume in the absence of evidence otherwise that its structure is the same for all human beings, and if that's the case, then human languages are products of different parametric states of this language apparatus. one task of linguistic theory, then, is to discover in detail what this apparatus consists of and all of its parameters.
as much as i want to be sympathetic to chomsky's violent critics, when it comes to purely intellectual matters, honestly i find them all ridiculously unsophisticated and dumb compared to chomsky. it's clear if you read what they're writing they don't even understand what chomsky is saying. it also just seems like wild speculation without justification to say "human language is explainable in terms of general cognitive principles". they do have a point, however, when they note that chomsky has been unprofessional and maybe even unscientific in how he has wielded his influence in the field.
now to your question after that extended roundabout--I'm interested in CCG and HPSG and have been also been curious about the multitude of semantic formalisms that have been cropping up (UCCA, AMR, …)

>>11613573

(4)

> What are some major unsolved areas of linguistic theory?

all of semantics, for one thing
> Are the grants enough, and how stupid are the politics necessary to get those grants?

what grants lol
ok seriously now: any phd program worth attending will fund you for x years, but after that it's very unlikely you'll need/get a grant unless you're doing experimental stuff (phonetics, cogsci) or fieldwork.
> How much has SJW bullshit infected the field in academia? I know postmodern theories place a large emphasis on deconstructing language

hard to say since i'm a computational linguist. linguists are on the whole very liberal and sociolinguists and linguistic anthropologists in particular are even more so. i don't think about this much though--if you go to some conferences you might like, idk, get asked to state your pronouns or something. it's not a big deal imo

Any good sources/recommendations regarding Sinosphere languages probably no later than 19th century?

>>11613765

This one seems interesting if you have prior knowledge:

> David Prager Branner - Problems in Comparative Chinese Dialectology: The Classification of Miin and Hakka (1999)

Look on libgen and ask around, my man.

>>11613569
>if you do computational, you'll make fat stacks as a tech worker

I keep hearing this and yet my feeling is that what is taught as computational linguistics has little to do to with CS applied to language.

Also from what I've heard the industry tends to hire linguistically inclined CS people rather than computationally inclined linguists.
(beyond a few people hired in advisory roles)

Most of the linguistics studied beyond the BA doesn't seem to be terribly relevant to actual NLP work.

I'd love to be proven wrong as I like compling a great deal and would love to have that as a backup if an academic career doesn't pan out.

Have you yourself done any work outside academics?

What are the actually criticisms of generative linguistics beyond personal feuds with Chomsky?

And what do people view as viable alternatives?

I get that there is some valid criticism of theoretical linguistics in general as it looks like sort of a circlejerk with little regard for actual predictive power from afar.
I feel like at our institute at least there is a great emphasis on minimal theories and motivating any vectors of study well and making them as testable/refutable as possible.

Also, does anybody have any recommendations for intros to biolinguistics/neurolinguistics? Feels like the relationship between linguistics and life sciences has been way too loose for the longest time.

>>11614387

I should've emphasized that the PhD was present during the last general, and that this is a copypasta. I figure theory matters more when a corpus is involved though

Why aren’t semantics in the textbook list?

>>11616303

Honestly, with it being such a huge discipline, I didn't know how to approach it. Anyone have any good textbooks in mind?

>>11617221
Heim&Kratzer is a classic. We did Elements of Formal Semantics by Yoad Winter after that.
I really liked the Winter book.

>>11613560
nice OP, i've started to like this general.

>>11613560
>Surviving Linguistics by Monica Macaulay
an intriguing name, but I'm not buying books on paper anymore
also a torrent from the previous thread:
magnet:?xt=urn:btih:1ad01a626310a2692f74c3e401fd15ba4d5ab30c&dn=Language%20and%20Linguistics%20Collection%20%28Updated%21%29

ХУЙ

have i just solved linguistics?

>>11617840

Top stuff m8, I'll put Winter's book in the starter pack since it's on libgen and isn't specific to generative grammar, and same with the torrent link for >>11618099

>>11617898

Don't get me wrong, I've barely begun, but since 4chan's an addiction of mine, I'd like it to be paired with what I'd like to study.

Fun question from my Semantic Analysis class: can you write a phrase that means something different in the passive voice than it does in the active voice?

>>11617898
I agree, they're pretty /comfy/

>>11618751

I'll play:

I think it depends on the nature of the agent and the object, transitivity.

> This guy fucks.
> This guy was fucked.

>>11618751
It is easy: This guy's fucked.
Depending on what 's is it's whether in passive or in active. Also there's another example in the previous thread: >>11592772
>>11618772
Now give me a sentence that means the same whether its verb is in active or passive forms.

>>11618832

Hmm... the closest that comes to mind right now is

> My leg broke
> My leg was broken

>>11613569

Greetings. I am attempting to hack together an automatic speech recognizer in c++. Do you know anything about speech recognition? If you do, I have some preliminary questions about certain models and performance issues and would greatly appreciate your feedback. Thank you.

>>11618880
>>11615367

>>11613560
opus globus, haven't you forgotten something?
>>/sci/thread/11561546
>>/sci/post/11600898
https://www.youtube.com/watch?v=TjIPzyVlK60

>>11618922

Oh, my mistake. I’ll put it in next time

>>11618751
I think the question is bad. If they mean something different they are not the same phrase anymore, but if you want something that, when converted to the passive voice, changes meaning, you will be just playing language games with government, transitivity, polysemy etc, which is rather dishonest imo in this sorts of answer. For example:
The kids walked on the park
The park was walked on by the kids'
In the first it is normal, in the second an inanimate place has a fetish, but that's only because 'walk on' has this figurative meaning.
Also, linguistics on /sci/zo, sage

>>11619418
> couldn't answer the task correctly
> saged
Does it make you a sage? Now you shine throught "that kid's fucked up" making it obvious that this answer is not the best one, because it's pretty much the same whethere there's is or has. "the kid's blown macaulay culkin" on the other hand is a homework properly done for whoever was that fag who dared to bring it here.
Enjoy your other threads and stay away.

thanks for the reccs
Ive gained a later appreciation for this topic due to the mathematical elements

>>11613560
Languages typically use either of the two different systems for context tracking - topic comment or definiteness, that are notoriously different to learn for the speakers of langauges that use the other one.
This is assumed to be because of a critical period, and the need to learn it in childhood.
However, I think it amy come from the inherent parallelism or serialism of the brain.
The serial western brains have hard time tracking context. They are serial brains, always focusing narrowly on a tiny part of the context. They, therefore, can make things easier by pointing out if the item is within the context, and the work to find it needs to be done, or if it's a new item and the work to find it can be skipped. They have little need for topic, since as long as both the speaker and the listener track each other's attention, the topic should be obvious.
The highly parallel aspie brains have the opposite problem: Visual search, as well as searching context takes no extra effort, there is nothing to ease up by indicating definitness, so definitness is useless and meaningless, because it's always obvious if something is known from context or not. Their focus is so wide, however, it may not be obvious to which topic the comment belongs to, so the topic needs to be indicated.

>>11620961
>topic comment or definiteness
your word salad is so heavy to get through
once again and just for you: https://www.youtube.com/watch?v=9AN2tbfSGm4
>critical period
>and the need to learn it in childhood.
isn't it pretty much the same thing?
Also the biggest pill here to swallow is that it takes the same two years to learn a foreign language if your linguistics professors don't try to teach you, but leave you alone in the field of the language.
> inherent parallelism or serialism of the brain.
> The highly parallel aspie brains
I don't even want to know what it means
> serial western brains have hard time tracking context
What I know is you definitely read some wrong books.

>>11618861
not the same meaning though.

>>11620428

I think you might like this:
Keenan, Moss. 2009. Mathematical Structures in Languages.
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.228.1087&rep=rep1&type=pdf

>>11622337
If you want mathy linguistics and are not so interested in semantics, learnability and language modelling are great topics to get into as well.

By the way what are /sci/s opinions on OT and it's different flavors (serial, weighted, etc)?

Really liked this paper:
https://pdfs.semanticscholar.org/7ddd/68863d4773555b34f1a23e7e3cbe465bc2d8.pdf

>>11622365
>Really liked this paper
Really liked writing that paper?
What can it teach me? Why would I want to read something that unentertaining?

>>11622407
Liked reading it.
If you're not into language modelling/OT it probably won't teach you much and you'll find it unentertaining by good right.

If you are interested in the nature of grammar it's a nice short read to get into how we can turn OT into something that is just about as powerful as we want it to be.

The basic question is whether we want our grammar to be generative rules or restrictions imposed on a generative mechanism.

>>11622443
>we can turn OT into something that is just about as powerful as we want it to be.
Gary is that you? How powerful? Casting spells and ringing bells? Understanding languages you never learnt? Or just completely verifying the bs you paid that much to get dragged through?

>>11620108
The hallmark of schizos is the inability to string a sensible sentence together

>>11620108
>the kid's blown macaulay culkin"
No, it’s not, you didn’t change A PHRASE from passive to active, you are just playing on the ambiguity (in proper countries that speak proper languages you learn about ambiguities in the beginning of HS) of the abbreviation of this terribly bad writer. As it seems, it’s not an actual linguistic puzzle but a dumb angl*id question.
This shit question could only possibly come up with this cesspool answer if the linguistics course were in the United Stages of Cheeseburger, a country that speaks the worst version of the worst language of Europe: English.

>>11622474
>Gary is that you?
Not sure who that is, fill me in on any memes.

>How powerful?
Powerful to model any relation we see in language while avoiding irrational relations or infinite candidate sets as with vanilla OT.

>Casting spells and ringing bells?
That's exactly what we don't want it to do.

> Understanding languages you never learnt?
Not sure what you are trying to say here. Nobody uses OT to learn any language.
I don't pretend to understand any language I didn't learn.

>Or just completely verifying the bs you paid that much to get dragged through?
I didn't pay anything for my studies, that's the beauty of not studying in the US. I'm not looking for anything to be verified, I just think the paper is compelling. Show me a reason why OT is not worth pursuing at all and I'll indulge you. Linguists don't all agree with each other, obviously. That's healthy. I am not an OT evangelist, I think it's a cool perspective to approach grammar from but I don't think it's necessarily analog to actual cognitive processes. If they are anything like OT it seems more likely to me that they would be closer to the model in the paper than vanilla OT.

>>11622569
What does it make the academy of science? A madhouse?

>>11622589
>worst language of Europe
Danish

>>11622592
>Powerful to model any relation we see in language
Why would you want to do it?
> avoiding irrational relations
What are irrational relations?

>>11622656
Both English and Danish are great languages, which is seen in cultures built by their nations.
Russian seems to be much worse: navernoe for both for sure and may be, loser for player, normal for good, and so on, and these examples seem to be much worse than imflammable being flammable.
Yet what is especially great in English: it sounds great. Dutch sounds funny, German sounds rude, French sounds gay, Italian sounds arrogant, Spannish sounds savage. It's hard for me to say how Russian sounds, when people mock us they sound retarded, so I think that's how we're.

>>11622669
>Why would you want to do it?
Because we are trying to understand what grammar *is*. If grammatical knowledge is encoded in the same form in every brain then we expect the nature of grammar to be the same. Therefore any model of grammar should be applicable to every language.
>What are irrational relations?
Not every grammatical process is possible. As in we don't see some processes in natural language. Some processes are also computationally more complex than others.
We want to model grammar by the least powerful mechanism that can generate all processes we see in real languages.
If you want a technical explanation literally just read the paper its like 10 pages.

>>11622717
>we are trying to understand what grammar *is*.
So instead of understanding the language itself, you're refocused upon our understandings of what the language is? You're not supposed to study actual languages as long as you study the writings of those who studied those who studied other's writings about languages?
> Not every grammatical process is possible.
Give me an example.

>>11622811
>So instead of understanding the language itself
Studying individual languages is absolutely worth it and important. But since it looks like the foundation behind all languages is the same, since any human child can learn any human language, it is also interesting to learn how we can formalize observations from individual languages in a way that make them comparable to those from other languages. This way we can also understand why we can use language at all and where the boundaries of language are.

>Give me an example.
Are you serious? I am sure you are aware not every process is possible in language, I will just give you an example that's relevant to the paper.

Counting is generally not possible.
The paper uses the example of a language that consists of only a's and b's. In such a language there could be many processes like "delete all a's if there are more a's than b's, otherwise delete all b's" but this is a process involving counting, which is something we don't see in natural language. Such a process can be modeled in vanilla OT.

Of course the "ab" language is only an example, in a real language this could be a process like vowel harmony, where all vowels take the (+/-front) value that is the most common in the underlying representation. Sounds like it would make sense, but we don't see it in real languages.
The OT flavor proposed in the paper does not allow processes like these.

Impossible Languages by Andrea Moro is a nice intro to the topic for non-linguists.

As someone who is about to go to college, should I major in Linguistics? I just don't know the job outlook. If not linguistics, it's sociology.

>>11622811
>You're not supposed to study actual languages as long as you study the writings of those who studied those who studied other's writings about languages?
You know that you don't need to study every language in order to be a linguist, right?
You know that most linguists do study several other languages than their native language, right?

>>11622952
Both of them are not great for jobs.

My feeling is that with sociology you'll have more job opportunities in general, but that generally linguistics jobs are closer to linguistics than sociology jobs are to sociology.

Linguistics jobs are mostly computational linguistics or academia. If you can get into the right MA you can also do speech therapy.
The few sociology friends that I have either work in HR or for NGOs.

>>11622927
>the foundation behind all languages is the same
And what is that foundation that it allowed you to put BUT after "Studying individual languages is absolutely worth it and important."?
> Counting is generally not possible.
wtf is that schizophrenia? Give me an actual example of what grammatical process is impossible. From an actual language, please.

>>11622953
>You know that you don't need to study every language in order to be a linguist, right?
Oh, what an ancient trick! "absolute a is impossible, thus a is impossible/unnecessary"
>You know that most linguists do study several other languages than their native language, right?
Do they succeed?

>>11623005
>And what is that foundation that it allowed you to put BUT
Do you think I am advocating against studying individual languages? You realize this is not an either/or question, right? What is your agenda?

>Give me an actual example
Sorry but you are just going full retard here. You want me to give you an example of an impossible structure from an actual language. Top kek.
I gave you an example of an impossible process in the vowel harmony I described.
It would be impossible in any language, but since you want an actual language imagine how it could be a process in Hungarian, Turkish or English. Now notice how its not a process in any of these languages nor any other documented language.
I am not going to spoon feed you, I gave you literally all the info you needed. You clearly haven't read the paper, why should I convince you of it's usefulness.

Again give me one argument why you think OT is not worth pursuing, show me one language that violates the constraints I described to you and I'll be happy to respond. Otherwise GTFO.

>>11623010
Yes. I speak three languages and all people in my semester speak at least three as well. When we write papers on individual languages all of us primarily about languages that we speak well.

If you write about a more superficial aspect of language like phonology or if you're really deep into your subject matter you do not even need to speak the language you are writing about fluently. It is preferable to write about a language you know well though, as most likely people who speak the language are going to read your paper and judge it.

What the fuck are you on about?
Do you somehow believe that linguists do not care about individual languages?
You are obviously clueless AF about linguistics so what is your little vendetta even about?

>>11621994
>>topic comment or definiteness
>your word salad is so heavy to get through
It isn't my problem you don't knwo the most basic terms and are unwilling to look them up.
https://en.wikipedia.org/wiki/Topic_and_comment#In_other_languages
https://en.wikipedia.org/wiki/Definiteness
>I don't even want to know what it means
Do you even speak English?
https://www.youtube.com/watch?v=PJ_bS7meE7s

>>11623050
>What is your agenda?
zomg, you were miseducated good!
I only asked you what that foundation was so the rest of your sentence could make more sense to me.
> You clearly haven't read the paper, why should I convince you of it's usefulness.
Because that was the initial question: What will that poorly edited text which sole purpose seems to be to show us how intelligent those who read or wrote it were, could teach me?
You guys are supposed to be good with language, so why does your thoughts look like they were formulated specifically not to allow us see that all your sofistication is a one huge untasty and overpriced nothing-burger?

>>11623132
I mean honestly I can't help you if you want me to explain the importance of some random paper to you, maybe somebody else can. This is a linguistics general if you find the idea of linguistics wrong, don't come here.
You seem to think that the whole field of study is a scam, just because you don't understand a random paper on first read.
If you were to read the paper and look at the sources or the references I mentioned in the comments you would be able to understand. It is not formulated at all to be hard to understand, the opposite is true actually.

I told you in my reply to your first post that it is not going to teach you anything if you are not interested in OT/language modeling.
Do you expect to pick up a paper on astronomy or psychology and understand everything? If you are not just trolling and honestly want to understand what this is about I again recommend you read Impossible Languages by Andrea Moro,
great introduction for non-linguists.

>I only asked you what that foundation was so the rest of your sentence could make more sense to me.
Nope you clearly were out to insult me. But still this is the foundation that some (not all of us) think is there:
https://en.wikipedia.org/wiki/Language_module

If you don't have any criticism that goes beyond "I do not understand this and I do not care to learn about it" I will honestly just ignore it. I think I have given you all the resources you need to understand what the link is about and if you think linguistics is useless by principle I honestly do not care.

>>11613560
No cognitive linguistics book... cringe

>>11623199
>You seem to think that the whole field of study is a scam, just because you don't understand a random paper on first read.
No, baby, I think that the larger part of this whole field is scam, as these general threads have perfectly demonstrated, and so do you.
And it was not because I just refused to dig deeper into some random pile of shit you cannot explain what the use of (other than I will understand that pile of shit you propose to research instead of the actual subject of the field) it was because I'm in the field for couple of decades now, and I think I know now what is the little trick your pilpul is supposed to play.

>>11623199
>>11623050
>>11622592

God tier patience

>>11623225
answer the question: >>11607895 I bumped that thread for you to come all over it. in vain

>>11623207

If there's not one listed, then post it ya dingus, and quit bitching about it! This general hasn't been around for long

>>11622700
>t. Anglx-Saxonex retard
Everybody makes fun of your language in the same way that everybody makes fun of Burgerland
You crooked teeth inbreds just choose to ignore it with "WE WUZ RITCH N SHIET"

>>11623297
We speak their language here. Explain this.
What a torture it must have been for you!

>>11623277
Cognitive Linguistics: An Introduction
Cognitive Linguistics: Basic readings, Cognitive Linguistics Readings, books by their founders like Lackoff, Johnson, et al

>>11623319

Duly noted. I’ll get a title for the next bread

>>11623232
>You bumped the thread for us.
>Last response on Tuesday.
right.

Speech therapy -> definition of places of articulation and formalization of the IPA
Speech recognition -> discovery of formants and other attributes of segments
Automated translation -> tree structure in syntax, Chomsky hierachy, sl/tl distinction
Language Pedagogy -> Advances in the individual philologies
Decipherment of historical records -> Rosetta stone, decipherment of the Maya script, Linear B
Forensic Linguistics -> literally the whole field, the process against the UNAbomber as an example
NLP -> grammar induction, lemmaization, literally the whole rule-based approach

>>11623370

why try to argue for usefulness?
linguistics is like astronomy.
its cool to understand what's going on, if it leads to something useful that is a bonus.
have not heard anybody accusing astronomy of being complicated for the sake of being complicated.

>>11623370
> was asked for examples where it was useful
> gives examples of where it was used
Every time I fire a linguist, the performance of the speech recognizer goes up.
> Chomsky hierarchy
yes, he's very high in it (and no wonder automated translation works like shit, but works a little better when it ignores what you faggots conceptualized)

>>11623387
>have not heard anybody accusing astronomy of being complicated for the sake of being complicated.
You haven't heard much about it, have ya?
https://www.youtube.com/watch?v=ebqAH5mLZNk
And it's never for the sake of being complicated, it's always being this was for some political or religious reasons.

>>11623199
>You seem to think that the whole field of study is a scam, just because you don't understand a random paper on first read.
Because it isn't possible to understand it. There is no way to look up what you don't know, becuase the terms require antother made up terms, in a tangled web that would take years to even learn to understand what you're trying to say.

>>11623199
>>11623199
>You seem to think that the whole field of study is a scam, just because you don't understand a random paper on first read.
Because it isn't possible to understand it. There is no way to look up what you don't know, becuase the definitions of the terms refers to several different unknown term, in a tangled web that could years to decipher well enough to decipher what you are trying to say.

>>11623199
>You seem to think that the whole field of study is a scam, just because you don't understand a random paper on first read.
Because it isn't possible to understand it. There is no way to look up what you don't know, becuase the definitions of the terms refers to several different unknown terms, in a tangled web that could years to decipher well enough to decipher what you are trying to say.

>>11623199
>Do you expect to pick up a paper on astronomy or psychology and understand everything?
Not completely, no. But there should be no serious difficulty figuring out what the paper is about, and it should be perfectly possible to fully understand it as long as you look up what you don't know.

>>11623199
>You seem to think that the whole field of study is a scam, just because you don't understand a random paper on first read.
Because it isn't possible to understand it. There is no way to look up what you don't know, becuase the definitions of the terms refers to several different unknown terms, in a tangled web that could years to decipher well enough to even understand what your papers are saying. If they even say anything meaningfull in the first place.
>Do you expect to pick up a paper on astronomy or psychology and understand everything?
Not completely, no. But there should be no serious difficulty figuring out what the paper is about, and it should be perfectly possible to fully understand it as long as you look up what you don't know.

اهلا و سهلا

Question:

Would a rules-based approach to NLP do better than a probabilistic one?

>>11622365

Again, duly noted my man

>>11613560
Linguistics is for faggots. Fuck you

>>11627385
Duly noted what? What is that shit even about?
https://www.youtube.com/watch?v=gmOTpIVxji8
And I haven't even begun asking what that shit is for.

>>11613560
In Linguistics, what is the idea called that words and their morphology are not arbitrary but based on reality across languages?

>>11627845
onomatopoeia?

>>11627845
this is so fucking vague what the hell
do you mean linguistic determinism? ? ?

You guys are my favorites.
I'm posting this meme for all of you nerds who might understand it. I certainly don't.

>>11627930
it's not even funny
ipa is googlable
we shat on them in the first thræd

>>11614387
Im a "comp sci inclined " linguist for Google. I was hired for being a "native speaker of X locale of Y language + being a linguist + knowing my way around bash and python"

>>11626265
Yes but they require tons of exceptions and rules and time to correctly create and update them.

Many big company use both at the same time. They fine-tune with comp linguists/linguists what was created automatically

>>11623688
Syntax is shit anyway. Every field has weird names and symbols that are hard to search for and understand correctly if youre not in the field

>>11629010
'aight go fuck yourself then.

hello linguists

how do i get my hands on all the cool semantics ultimate theory of meaning stuff that changes your perspective forever and doubles your verbal IQ

surely it exists just hand it over please

>>11629460
this autism? it's still underdeveloped

>>11629492
i don't know what this is but it looks very esoteric. i was only joking before but it looks like you people really did invent some kind of language wizardry

>>11629492
If you count Hebrew י as a vowel, why not ו as well? To my knowledge they both represented glides and then came to be used fot the corresponding vowels /j/->/i/ /w/->/u/

>>11629492

ع isn’t a vowel though, it’s a pharyngeal consonant in fusha. It is a vowel in Persianate languages, mostly to assimilate their Indo-European phonology in Arabic loanwords

>>11629460
I don't think studying semantics does anything to your vebal IQ but the Winter and Heim&Kratzer books that were mentioned otherwhere in the thread are a good intro. Generally studying Lambda Calculus, Set Theory, Predricate Logic and Statement Logic are the starting point, all very compact theories.

Here is a short run down:
What is the meaning of a sentence?
One interpretation is that the meaning of a sentence is the information of what the world has to look like for the sentence to be true.
"Kim dances"
You don't know if this sentence is true but you know that for the sentence to be true, there would have to be an individual called Kim and a non-empty set of people who dance, then Kim would have to be a member of that set.

A popular notation for meaning is Lambda Calculus. Lambda calculus has three basic types:
Entities (e)
Statements (t)
and functions from a type to another type.

Kim is of type (e) obviously.
Dances is of type (e t): it takes something of Type e and returns something of type (t).
Every sentence is of type (t), every sentence either true or false for a given situation

The last puzzle piece in Lambda Calculus is function application. When we two constituents next to each other, where the one of them has the type x and the other takes something of type x as its first argument we can apply the function to the argument.
dances (e t) (\x -> x is an element of the set of dancers)
Kim (e) (Kim)
thus: Kim(e) dances(e t)
can be simplified to:
Kim is an element in the set of dancers (t)

The truth value of the sentence then depends on the situation in which it was uttered. If in that situation the set
dancers = [Tom, Alice]
then the sentence is false
If dancers = [Alice, Kim]
then the sentence is true

This is just supposed to be a quick and dirty introduction. The PDFs for the textbooks can be found online for free. It starts getting more interesting when you start adding logical operators (and or..) or quantifiers (all, some)

>>11630071
Yeah same with Hebrew ayin ע
The basic idea that the alphabet was only invented once and that everybody basically modified it for their languages is true though

>>11629045
Yeah Syntax is a mine field of terminology.
Also an area where nobody is willing to agree on anything. To me it's the only properly hard part in linguistics. But it's also sort of nice that there is so much work left to be done there.

>>11613569
Do you have to know a lot of languages to be a linguistist? I don't mean being fluent, I mean know basic grammar and shit.
Also, could you give me a quick rundown on computational linguistics?

>>11630111
I find syntax, and to a certain degree when it overlaps in the theories, to be uter shit. I cant take theories with bunches of exceptions added all the time just to fit in a language in which the model doesnt fit seriously. I have the feeling that syntacticians completely ignore the neurolinguistics part of it all.
" oh it works like this, add X or betha-y change here and it works! Yay our clearly bushit and biased towards english model is the one true model to explain how language works in the brain!"

>>11613569
>>11630147
Also, do you use math on linguistics? If yes, could you give an example?

>>11630151
There is a lot of syntactic work that worries deeply about neurolinguistics. Particularly in parsing and experimental syntax. Fascinating stuff.

I think the current model for syntactic structure is going to be overhauled it is going to be because of experiments into parsing.

>>11630155

Read the thread anon, some of us have discussed this

>>11630147
Depends on what you do.
You are generally expected to study a few languages. If you write papers about individual languages you will have to study their grammar a bit. Also if you just want to get an overview about the range of human language it makes sense to study a few languages that have important features that your native language lacks.
Look at the languages section in the CVs of your favorite linguists to gauge how many languages people in your sub-field are expected to know.

Sorry I don't even know where to start giving a rundown of compling. Hopefully somebody else will do that.

>>11630155
Again depends on what you do. There is a lot of statistics in linguistics, especially typology, experimental stuff.
If you want to compare different models of grammar it makes sense to look into how well they scale and how computationally complex they are, so there is math in that (formal languages, automata theory)
In Semantics there is a lot of logic and depending on what you do there will be some abstract algebra, category theory and so on.
In phonetics there is some math to evaluate wave forms, Fourier transformations and stuff, I am pretty clueless about that area though.
There probably is math elsewhere as well but that's where I've run into it. Haven't even finished my BA yet though.
The mathematical Structure in Language PDF that was posted before is a good starting point the kind of math that I am interested in in linguistics.

There also is a cool podcast on math in linguistic with Greg Kobele on Spotify.
The Episode is part of Elucidations by the University of Chicago.

>>11630023
For the same reason greek H is not considered vowel (because in other alphabets it isn't)
That sixth letter is a special case and a lot can be said about it.

>>11630071
true dat, but that comparison shows, that it's an anomaly.
>>11630108
Though ע may seem consonant in רביעי, it will be transliterated with a rather vowel y in many other examples: כועס (koes, angry) , כובע (kova, hat)
The argument about the a in kova is that if there was no ע, there would be no a. as there's no a in כוב, but there's a in כובע — it is even more obvious in words like עכוב

>>11613560

How about Prague Circle/Roman Jakobson/Nikolai Trubetzkoy?

>>11630071
Wrong. See it in علم
You would be correct, if you spoke of ه but that comparison shows, that it's an anomaly.
>>11630108
Though ע may seem consonant in רביעי, it will be transliterated with a rather vowel y in many other examples: כועס (koes, angry) , כובע (kova, hat)
The argument about the a in kova is that if there was no ע, there would be no a. as there's no a in כוב, but there's a in כובע — it is even more obvious in words like עכוב

>>11630685
I think כועס is not the best example as here it could be argued that ע represents just a glottal stop like א does in many cases.
כובע is a different story as few (if any) would pronounce it with a glottal stop before the a.

Where ע gets interesting is in verbs
קורא kore
קורע korea
קורה kore (just for reference)

In that case the difference between ע, א and ה might today just signify something like an inflection class.

The point being: writing systems are messy and we can't really deduce from today's spellings and pronunciations what letters represented starting out.
I am not very knowledgeable about the historical side of the Hebrew writing system, however I think Hebrew ע likely started out as something similar to it's Arabic counterpart.

>>11630765
Just to for reference:
קר kar
קרא kara
קרע kara
קרה kara

>>11630680
what about them?

>>11630770
How am I going to trust you linguists about anything if you can't even into Engrish

>>11630765
>however I think Hebrew ע likely started out as something similar to it's Arabic counterpart.
The modern pronounciation of it was influenced by horrible arabic spelling. But that very script was used in yiddish which probably doesn't have those vomiting sounds. Also that comparison higher in the thread shows, that the structure is pretty much the same in all the alphabets, so it's up to you to figure out if every other alphabet fluctuated in the very same direction of five vowels universal up to japanese shores or it's just semites fucked their shit up.
Also people often transliterate egyptian to arabic fashion, but check this out:
‘In Egypt the priests sing hymns to the Gods by uttering the seven vowels in succession, the sound of which produces as strong a musical impression on their hearers as if the flute and lyre were used, but perhaps I had better not enlarge on this theme.’

>>11630807
Trust nobody. Do your best to understand, and if you don't - don't fucking accept shit on faith.
https://www.youtube.com/watch?v=B5dvUVuwcxg

>>11630685

I disagree. علم is conjugated as a sound verb, not as a weak verb, which a root with a mater lectionis in the first radical would necessitate (such as وصل or امر)

What programming languages are particularly relevant to linguistics? What languages have you used?

As a second related question:
Have you used:
Prolog and other logic langs
Theorem provers like Agda, Coq or Lean
Python, Julia, R
Lisps, Haskell, ML
What do you think of them and what do you think their place in linguistics is?

>>11630840
>conjugated as a sound verb
>not as a weak verb
You have to explain this to me, because it sounds like olim and so it will probably be transliterated into latin script.
> which a root with a mater lectionis in the first radical would necessitate (such as وصل or امر)
Why such a semitic concept has latin name?

Can I study phonetics with no prior knowlodge of other subfields of linguistics?

>>11631028
why would you want to study such a boring thing?

>>11630778

I'm curious what books by those and related people the thread might know about or recommend, since I'm aware of the school but din't know much about them.

>>11630101
thanks very much for the reply, but my problem is i already know all this stuff on this basic level yet i can't intuitively relate it to what i'm doing when i speak. i know what it is like when i regard some thing a thing(e), but i've got a only a vague sense of what an abstraction is, and everything beyond that is obscure to me.
is there some theory that describes in detail the mental events from my experience of speaking and then relates them to terms like entity, statement, or function, so that i can play out the story in my head when i read about calculation and logic? i believe this must exist because everyone's always praising linguistics saying helps you understand how you think, i just don't know where to find it.

>>11631028

Of course, go download the starter pack in the OP, introductory phonetics and phonology is the first book in its list

>>11630765
There is no good reason to believe those weren't believed as written. There are even some triconsonantal roots with aleph (or at least in classical Arabic there were). It's most likely one of the things that got changed during hebrew revival, since most european languages do not allow h and glottal stops in codas, if they have use them at all.

>>11630765
There is no good reason to believe those weren't read as written. There are even some triconsonantal roots with aleph (or at least in classical Arabic there were). It's most likely one of the things that got changed during hebrew revival, since most european languages do not allow h and glottal stops in codas, if they have use them at all. Ayin also got merged with aleph in the process.

>>11631742
Not him but : pragmatics and semantics on a different level are also in play. I think you can actually limit things if you base yourself with operators or basic set theory.

Context, pragmatics and (set theory) neural activation is key imo. Imagine a giant spider web. Not 100% true but : every word/meaning is linked to concept(s) which are linked to other words/concepts/memories. If somehow you can activate more connections or go deeper, you can get a bigger idea of whats going on. But it can also just make you not understand anything. Of course one's semantic system is mostly built upon one's experiences and knowledge (and ability to use it)

Its a saying that semanticians are never understood and themselves never understand anything lol

>>11630765
There is no good reason to believe those weren't read as written. There are even some triconsonantal roots with aleph (or at least in classical Arabic there were). It's most likely one of the things that got changed during hebrew revival, since most european languages do not allow h and glottal stops in codas, if they use them at all, so all the coda א and הsbecame silent. Ayin also got merged with aleph in the process.

>>11630765
There is no good reason to believe those weren't read as written. There are even some triconsonantal roots with aleph (such as r-'-sh). It's most likely one of the things that got changed during hebrew revival, since most european languages do not allow h and glottal stops in codas, if they use them at all, so all the coda א and הsbecame silent. Ayin also got merged with aleph in the process.

>>11631742
>is there some theory that describes in detail the mental events from my experience of speaking
Not that I am aware of.
The current theories of semantics are an analogy that can only describe and predict human behavior. What actually happens in our brains is more or less a mystery.
Sounds like you are looking for something on the psycholinguistics of semantics.
I have to admit that I haven't studied any psycholinguistics, neurolinguistics or cognitive linguistics yet.
So maybe what you are looking for is out there, I am just not the one who would know it.

Or maybe what you are looking for is a guide to semantically interpreting a sentence step by step?

>from my experience of speaking
Feels like semantics is most commonly studied from a listeners perspective, not a speakers. To model what goes on in your brain while speaking you have to take a fuckload of other factors into account.

>everyone's always praising linguistics saying helps you understand how you think
Could you give some examples where you heard that? If it was in an interview we can look if they said something else that gives a clue what they were referring to.

In the meantime here are some fun concepts from the intersection of semantics and psycholinguistics:
comparative illusion
garden path sentences

>>11631016

Sure, I can explain it. When Arabic trilateral roots are conjugated into verb forms, these forms change whenever an alif, waw, or ya form one of the three radicals. Read this link and keep flipping to the next pages to learn about these verbs:

http://allthearabicyouneverlearnedthefirsttimearound.com/p2/p2-ch1/hollow-verbs-form-i/

As for the mater lectionis, I figure the term was invented by biblical scholars for studying Hebrew

>>11632139
I don't feel like learning arabic grammar right now. But it seems their conjugation has nothing to do with the way that letter is read? The argument of yours was that it's diacritic signs coloured it vowel.
I know this point of view. The problem is I disagree with it. Because see the comparison of abjad to other alphabets and tell me what you see.

>>11629492
The alphabets are related, doesn't mean the languages are

>>11632830

> The argument of yours was that its diacritic signs coloured it vowel

The diacritic IS the vowel atop the consonant

I think that since Indo-European languages don't have the same laws with pharyngeal consonants as the Semitic (I'm especially thinking about Greek, given its interaction with the Eastern Mediterranean), they combined the vowel and the nonexistent phoneme in the 'ayn letter and made it a vowel only

>>11631936
i really like what you said about the giant spider web. this visual/spatial/emotional way of thinking about meanings/words/concepts/memories/weirds resonates with my preconceived notions and intuitions to such a degree that as soon as i considered it i was hit with the force of recognizing a pattern so clear that i'm ready to defend it at all costs. especially when i consider cases of failure such as when i misunderstand someone or when i struggle to express what i mean. there i'm more comfortable saying "it is like" or "it seems that" and in truly desperate cases "i feel like". this sort of squishy fuzzy spider web way of thinking perfectly captures those like-feelings which were at the root of my frustrations. it's almost therapeutic to be liberated from these tensions in meaning. if i was bolder i would say now that when i say "X is like Y" i really literally mean that X has an intrinsic property "is like Y". instead of the previous situation where i thought i was saying about X that it some property "exactly X" but i can't quite describe it so i'm to referring to Y to help me approximate. i went from having no idea how meaning works to knowing exactly what i mean and how i mean at all times.

i'm feeling very happy about my good luck. i knew i could trust the linguists to deliver. apologies if this reads like some combination of stating the obvious and speaking nonsense.

>>11634369
more like nonsense

u ok?

>>11634391
yes i got just what i wanted. vague intuitions crystallized and shelved. now i can forget about this topic for the next 20 years.

>>11633027
It doesn't. But they are.

>>11634170
>The diacritic IS the vowel atop the consonant
I know that, but my argument that some of those consonants are actually (or at least originally) vowels.
And those diacritics are specifiers as if we would need to speciify how our vowels sound (whether it dog or do, bet or be, pan or car or call, bus or fuel)
That misconception of consonantal alphabets could be caused by degradation of arabic linguistics to the level of not distinguishing between vowels and consonants the same way we don't understand the concept of linguals overspecifying them into all sorts of the loci of the tongue. Or it could be made on purpose by those who believed in spells and shit and didn't want newbies to occasionally read ancient texts the way they were supposed to be read with vowels used as musical notes.

>>11613573
Hey anon, have you read The Neuroscience of Intelligence by Haier, Oxford?
Have you taken into consideration the difference between people with crystal memory, fixed low intelligence but savant in one field, compared to fluid memory, general intelligence.
I was quite interested in this field as I thought it revealed a mechanism in the brain that asserts how data is managed.
Further more if a brain is working harder then it tends to lower IQ, which represents how IQ relates to ease of though rather than some sort of mystical higher thinking.

>>11613569
>that's an incredibly high admissions rate for phds
I'd have to guess that it's because very few people actually want a linguistics phd, so there are a lot fewer applicants than there are for other degrees, so the acceptance rate is naturally higher for linguistics

>>11635090
>Have you taken into consideration the difference between people with crystal memory, fixed low intelligence but savant in one field, compared to fluid memory, general intelligence.

Crystal memory/lower IQ: possessed by dolichocephalics, tends toward religiousness and formative creative invention, ADHD, INTP, relatively lower spatial IQ, relatively higher anal-verb IQ, Caucasoid

Fluid memory/higher IQ/g-loaded: possessed by brachycephalics, tends toward atheism and "improvement" type inventions, high attention span, ISTJ, relatively high spatial IQ, lower anal-verb IQ, Mongoloid

I don't have my files on hand but this split is so fucking obvious I feel like it needs almost no explanation. Anybody who's done a really neutral and unbiased look at history should arrive at a similar conclusion.

>>11635090
>>11635126
>Further more if a brain is working harder then it tends to lower IQ
I agree.

overall, I have the dolicho type of head, and personality/mental state. My brain is very "excitatory", I can release huge amounts of energy in very short spans of time, but I don't have this "always on" subtle mental energy that Mongoloids, and some more square-headed Europeans have. I definitely have hyperfocus.

The fact that I am even able to see this dichotomy speaks to my dolicho-mentality, most are unable to see such realities unless they hear them from higher sources.

In high school and college, I was able to get near-perfect scores despite never taking notes--I would simply do a 3 hour cram session and get an A+. If I fell asleep during these cram sessions and didn't study, I would straight up get an F.

>>11635126
The split is worse than you think. The fact that they are retards who have everything memorized, they are used to being able to read people by noticing how exactly they do things, how exactly they say things, so they consider others deceptive.
Worse is alcohol. They are apparently unable to lie when drung, so they try to make you drunk, you speak gibberish, and they think it's your true self. And they are proud of their rigid character, they will rather burn evidence they are wrong rather than to change anything, and in general are hortible and vicious people.
I think it's more about cerebral vs cerebellar dominance, the dolucocephaly may be just one of the things that makes their cerebral cortex harder to control, if there is a correlation.

>>11613569
10% is high for a phd?
fuck.
FUCK FUCK FUCK FUCK FUCK FUCK FUCK FUCK
I'M PANICKING PLEASE JESUS HELP ME GOD
oh FFFFFFFFFFUEOWAAAAAAAAA N

What do I have to do guys? I'm a physics major and I want to do research. Oh shit I'm going to have a panic attack.

Please tell me it's doable.

>>11635225
I think it might be even where the idea of good and evil comes from - you stay good as long as your cerebellum stays in control. Once your cerebral cortex takes over, you turn evil. The more you struggle to control it, the more twisted your mind becomes

>>11635236
It only tells that you have apply ten times on average to be accepted once.
I heard that average number is 100 times (probably that's because you have to apply 10 times in every organization and only one out of 10 will accept your paper.
Which made me say fuck that shit, I'm not going that road paved with shit. I will do my science for myself and for whoevere is interested and I'm not going to depend on their gibs

>>11635236
Unless your interested in how Physics is gay friendly and atoms are genderqueer I would look for a post in a Singapore or Asian country anon.
Japan has a nice research college in Okinawa. Met a nice atom researcher german wifu on the plane there.

>>11635296
shit tier statistics

>>11634692
no

>>11635884
They are.

>>11635893
Show me so I can laugh

>>11635900
If you want to learn foreign languages, it's useful to think they're connected, so you find similarities and understand instead of memorizing.
If your task is to divide nations so neighbours don't feel related and you can sick ones onto others, it's useful for you to teach them that languages emerged independently and were only influenced by one another. Though even in this retarded case they're related if not by vertical meme transfer, then by the horizontal one.

>>11635936
>it just feels that way
well okay then

>>11635947
>even in your retarded case they're related if not by vertical meme transfer, then by the horizontal one.
anything to object?

>>11635947
I'm not an expert but it seems reasonable. If languages are unrelated then how come translation works to the extent that it does? Obviously because daily life is the same.

>>11635956
If by horizontal you mean borrowings then that doesn't make two languages related. You said they spring from the same source anyways in your original post.
>>11635962
>If languages are unrelated then how come translation works to the extent that it does?
What

>>11635971
>You said they spring from the same source anyways in your original post.
Not my post, but I can support that point of view.
I wouldn't say there was just one source: so I believe that writing system was invented at least twice, but languages are so more ancient than writing systems, that all those sources of them intertwined and if you really think about it, it doesn't matter if the common word emerged before two languages split ore after that.

>>11634170
>>11634718
Ayn, aleph and he were jsut consonants, that were lost from hebrew spoken by Jews living mostly in Europe and speaking European langauges that don't have these sounds. There is no good reason to believe they were not fully pronounced (e.g. roʔš, bəreʔšiθ), as they still are in related langauges.

>>11636231
>that were lost from hebrew spoken by Jews living mostly in Europe and speaking European langauges that don't have these sounds.
And what is a good reason to think they were pronounced like desert people speak and not like jews actually speak?
> There is no good reason to believe they were not fully pronounced (e.g. roʔš, bəreʔšiθ), as they still are in related langauges.
Explain to me what makes you think aleph in Allah-akbar isn't vowel (other than you were brainwashed to believe that bs)
Try to convince me that aleph and alpha is not the same thing.

>>11636231
>>11636352
I thought there's an oral torah tradition going all the way back. There should be an uncontroversial way to settle such questions.

>>11636352
https://en.wiktionary.org/wiki/%D8%A7%D9%84%D9%84%D9%87_%D8%A3%D9%83%D8%A8%D8%B1#Arabic
https://en.wikipedia.org/wiki/Hamza

>>11636352
>Explain to me what makes you think aleph in Allah-akbar isn't vowel (other than you were brainwashed to believe that bs)
Nobody's disputing that, but that's its use as a mater lectionis
>Try to convince me that aleph and alpha is not the same thing.
They come from the same source, but look at the earliest Hebrew writings and point to me where it was used as a vowel

>>11636352
The al in allah is the definite article, which often slurs into the following word, it doesn't correspond to the el in elohim. You can still hear the full aleph in laaʔilaahaʔillallaahu.

Somebody here ever checked Ithkuil?

http://ithkuil.net/

It seems an overly complicated language with more than 80 noun cases and an extremely complex phonology. The language is synthetic and agglutinative, it is designed to fill as much information as possible in the least amount of space.

It took 30 years to make it and the guy is not even a linguist.

>>11636389
Sefer Yetzirah speaks of vowel labial and lingual as the basis of the alphabet, and that is pretty much what the greek myth tells (but it is not allowed to use mythology as historic reference)
But it speaks of vowel only when you compare it to that very greek myth. Without such comparison it only speaks of א מ ש
(but it also tells that it's a great mistery) so I think it's actually א מ ת (but it's just a guess, maybe some rare cabbalists know it better)

>>11636471
> thinks arabic phonetics is relevant
> maybe even thinks quran and all arabic philology came from god himself.
pathetic

>>11636501
>mater lectionis
appeared relatively late. For many centuries they somehow managed to read their shit without it.
> look at the earliest Hebrew writings and point to me where it was used as a vowel
And how am I supposed to do that?

>>11636632
What I had in mind were things like torah oratory traditions that pass down as kind of song. Are they unreliable?

>>11636679
Do you speak of prayers? A in Shma is transliterated with aעn, just as I in Israel is transliterated with יud.
You can notice, that i is consonant in iota and y is consonant in boy (we are taught to call y semi-vowel, but we don't call i this way, even though they act pretty much the same, so it's all about tradition, and traditions are retarded rather often, but we are not prone to change them, it could damage the faith in authority)

>>11636729
>A in Shma is transliterated with aעn
Or because it ended in ʕ, which as has been said disappeared with european influence
>just as I in Israel is transliterated with יud.
There's a y at the beginning of it

>>11636729
I meant in general, but mostly had public torah reading in mind since it's a rigorously scrutinized ritual where the making a mistake can damage your reputation.

>>11636764
>ʕ, which as has been said disappeared
or never existed.
>>11636789
In spite of what people think, speech and writing system have next to nothing in common. You're also of too high esteem towards religious retards.

>>11636839
>or never existed.
https://en.wiktionary.org/wiki/%D8%B3%D9%85%D8%B9#Arabic

Why does the equivalent Arabic word have it then

>>11636839
>>11636849
And why isn't it transliterated as a vowel in the imperative here

https://en.wiktionary.org/wiki/%D7%A9%D7%9E%D7%A2%D7%99#Hebrew

>>11636644
>arabic phonology is irrelevant to allahu akbar

>>11636839
I don't know what gave you the idea that I have any affinity about religious people. I think that religious conviction can plausibly motivate people to preserve a tradition they view as sacred. I'd also be surprised if you can support "speech and writing system have next to nothing in common", but I'm just a layman.

>>11636862
And why does it seem to change the vowel it represents arbitrarily between forms?

https://www.pealim.com/dict/2250-lishmoa/

Do sociologists use linguistic buzzwords for their papers to sound hip, or do they actually take the study seriously?

>>11636581
Yeah Ithkuil is a nice meme
A cool thought experiment nothing more

>synthetic and agglutinative
agglutinative is just a subset of synthetic
I think a better description of Ithkuil would be polysynthetic since that's like it's whole deal.

>>11636849
>samYA
ya is also last vowel in russian.
And they don't look at that j part of it as at ʕ (but common ear wouldn't see the difference, so it's all because of different linguistic approach)
But the question is why does vowel, corresponding o doesn't sound like o.
I think it is because that vowel is mistransliterated for whatever reason, but it's just the first approach, the actual answer can be conpletely different.

>>11636862
am I missing something?
ע there always transliterated as vowel.
And it's funny, that it's ofhen transliterated as E, because ع does look like E.
and ע looks like y (which could be "cognate" of ע)

>>11636915
If it stands for y (which is also consonant in english half of the times) then it could stand for all the additional y-letters, such as я ё ю
People also had manner to avoid vowels, and indd y cn rd wtht thm, y onl nd thm in intl pstns, and thts whr smtc vwls ar obvsl vwls.

>>11636911
>I'd also be surprised if you can support "speech and writing system have next to nothing in common"
Completely different languages can have the same writing system.
One language can use completely different writing systems.
n is p in russian, m is t in russian (I really wonder why)

>>11637746
>>11637764
>>11637783
I'm gonna stop here for my own mental health, you're apparently on a whole different plane of consciousness that I can't keep up with.

>>11637697
Is there any other cool philosophic language with interesting morphology and phonology?

>nice thought experiment
I think the guy claims that a fluent speaker of Ithkuil would be able to process thoughts way faster than a normal human being. Don't know if it's a legit claim but sounds cool.

>>11636581
>>11637697
It's actually polysynthetic and fusional, which is a combination that never exists in natural languages, and for a good reason.
>>11638021
I don't believe the language you speak matters, and if you do, then chinese might be better at that. The words are simple enough that they are essentially processed as singular phonemes, which saves one processing step and makes chinese work sort of like quickspeech on the mental level.

>>11636581#
>>11637697#
It's actually polysynthetic and fusional, which is a combination that never exists in natural languages, and for a good reason.
>>11638021#
I don't believe the language you speak matters, and if you do, then chinese might be better at that. The words are simple enough that they are essentially processed as singular phonemes, which saves one processing step and makes chinese work sort of like speedtalk on the mental level.

>>11636581
>>11637697
>>11638021
The original sanskrit dictionary defines about 2000 roots, less than a half of that can be found in known works. The rest is grammar.

>>11636581#
>>11637697#
>>11638021#
The original sanskrit dictionary defines about 2000 roots, less than a half of that can be found in known works. The rest is grammar. Which is itself written more like computer code, than natural language grammar:

http://learnsanskrit.org/panini/shivasutras
http://learnsanskrit.org/panini/index

Is the Chomsky hierarchy any useful in actual linguistics or is it just a computability thing (nothing against computability though)?

>>11638464
Ok i'm getting a but confused, what's the difference between an agglutunative and polysinthetic language? I seem to understand that some agglutinative languages are polysynthetic while others are not.

>>11639372
Polysynthetic means the language is very highly synthetic.
Fusion and agglutination are two different kinds of synthesis, only the latter is seen in polysynthetic languages, because fusion just wouldn't work well.
A made up example, fusion:
He did garko
You did garkas
I did garkin
They did garkama
You did garkeni
We did garkanas
He will do garkum
You will do garkel
I will do garkom
They will do garkunu
You will do garkumad
We will do garkuni
Agglutination:
He did garko
You did garkon
I did garkom
They did garkalo
You did garkalon
We did garkalom
He will do garkunu
You will do garkunun
I will do garkunum
They will do garkalunu
You will do garkalunun
We will do garkalunum

>>11639372
>>11639607
As you can possibly see, the problem with fusion that the amount of grammar learning needed grows exponentionally with the amount of synthesis. With agglutination it grows linearly.

>>11639661
>>11639607
Thanks for explanation. Seems a thing mostly present in indoeuropean languages for what I see.

>>11639730
Fusion I mean.

I always found agglutination quite weird and hard to grasp but maybe it's because I am and indoeuro pleb.
Still it seems that the creator of Esperanto opted for agglutination since it makes the language more logical and intuitive apparently.

I'm not into the field of linguistics but I know the director of the department of linguistics of my university. The guy is a polyglot apparently. In his CV he claims to know

English
German
Italian
Spanish
French
Russian
Mandarin
Persian
Georgian

Are all professors of linguistics polyglots or wannabe polyglots? Is it kind of a mandatory requirement? Do they adopt a certain method to study a language?

>>11639740
I guess. It seems they sort of work the opposite way - the grammar in fusional languages reduces cognitive effort at the cost of more memorizing. Agglutination reduces memorizing at the cost of more cognitive effort.

>>11639932
>Are all professors of linguistics polyglots or wannabe polyglots?
No, some are even offended by the accusation.
Most will have a background in a handful of languages, but will be apprehensive about claiming fluency.

Two things that people say that piss off every linguist:
>How many languages do you speak?
>uh-oh! Guess I'll have to watch how I speak around you *laughter*.

>>11639932

It depends, really. Look up the scoring system the US government uses for measuring proficiency:

https://en.wikipedia.org/wiki/ILR_scale

As you can see, 0-3+ encompass functional language and composition skills, that is, proficiency. However, once one gets into 4 or 5-level (those are the levels the glowies hire past office-analyst level and into special squirrel shit), just knowing a bunch of vocabulary and parroting it like an autist doesn't mean you have mastered the language, you need to know good dialectology and cultural knowledge of idioms and metaphors only native speakers could know.

>>11639974
>Glowies
slightly related. I heard that soldiers going into SOF are required to study at least one language once they pass their backbreaking training, this applies especially to the green berrets since they often work and interact with locals. Exactly what type of level are they required to achieve for such tasks?

>>11640055

Civil affairs and SF, yes. They just need basic proficiency, they don't have time during their training to go full retard on a language, probably 2-2+ proficiency will do in reading and listening.

>>11639198
bump

>>11640420
It is useful when deciding which models of grammar we can rule out.

>>11622365
This paper is basically based on that idea.

Our brains are computers, too.
So if some model is computationally very complex and if we do not see any other behavior which can be only be explained with models of that complexity, then we would really, really like to find a model that is less computationally complex.

Tree structure in syntax can be justified with computational complexity if I recall correctly.
If Syntax was "flat" the cost of computation per added word would go up way too fast.

The Chomsky Hierarchy is the backbone of that train of thought even if it is not always cited directly. You can look into Strictly Local grammars, Tier Strictly Local grammars and so on if you want to learn about a related categorization scheme that is frequently used and also relevant to the paper I mentioned.

It is still debated where exactly human language falls on the spectrum.

>>11640704

What experience do you have with linguistics?

>>11613560
Haha hi I'm from the link posted on /pol/, I'm not a regular here, but I figured I had ought to post a beautiful and expansive treasure trove I've found, don't know if it's been posted here before, but here goes:
driveDOTgoogleDOTcomSLASHdriveSLASHfoldersSLASH1ACzYcQYEM42FAgw-fSS0RQs4pCKgDMWa

>>11641530

Well sheeeeeeeeeeit, that's going in the bread. Thanks for the stash, king stanger :^)

>>11641627
Of course bro, I don't know the exact facts on it, buddy of mine tossed me the doc while I was doing research on the Dutch (very beautiful language, I should say) I have no idea where he got it, and I'm currently not in contact with him, so I have no easy way to find out, amazing find either way.

>>11640704
>It is useful when deciding which models of grammar we can rule out.
Rule out for what?
Also give an example.
meanwhile,
>hierarchy
>rule
I consider him a clerk, who was placed there by zog to control the field able to dismantle abrahamic bs more explicitly than physics ever could. But they shouln't fear that annoyingly, because free linguistics maybe dismantles the myth of semitic superiority, but it weaves their threads in the fabric of actual human culture which will inevitably be understood when augmented intelligence is on the market.

>>11641644

That’s a pretty good collection man.

>>11641530
Where on /pol/ was the link posted?

>>11641491
Undergrad linguistics student
So take everything I say with a grain of salt I guess

>>11642836

>>>/pol/256355585

This thread is relevant to my interests

>>11642836
Ah fuck I don't remember man, I think it was some bait thread, this gay Swedish guy (Go figure) posted something about Mexicans being better than Americans, eventually someone else entirely posted a link to this thread.

>>11642836
Nevermind >>11642887 was correct, didn't see it before I posted.

>>11642611
Thank u :)

>>11643381

The more books, the better

>>11644280
Affirmative.

https://breac.nd.edu/articles/ogham-in-3d-digitizing-a-unique-aspect-of-irelands-cultural-heritage/

Do actors learn the IPA when they need to study other accents? That would be interesting to learn

>>11646504
nobody uses that shit

>>11646605

You'd figure they would in Hollywood, they're supposed to be smart

>>11647646
reminds me of a non-linguist friend who thought every language should be written in IPA.

Sure you could do that but sometimes you can give more information if you write in a system that is more tailored to your specific language.
Information like inflection classes, distinguishing two utterances that are homophonic in some inflection and so on.

In a similar way you might have too much information. If everybody wrote in a narrow phonetic translation of their dialect that would arguably be more confusing than using a broader phonological translation.

The point being: you want the right amount of information, not too much and not too little.

Also because of the way we learn languages imitation often trumps dry instructions.

So I think for many cases dialect coaches and learning generalizations with IPA as a crutch where it is needed is much more efficient than the extreme example of a script in narrow IPA transcription for the dialog in a scene.
(IPA is used a lot in speech coaching though and many actors get some of that if they take acting classes)

tldr; IPA is cool but most of the time it doesn't replace good coaching. Narrow transcription in particular is almost always overkill and not helpful to actors who have no use for linguistics in most of their roles.

/ramble

>>11648465

Thanks for that take man, I should have mentioned phonological models too.

Funny thing is, I'm actually trying to teach myself acting for a screenplay I'm writing now (I got Petit's book on the Chekhov method), and I honestly figured that IPA would do for refining English accents. Are you in the performing arts?

Bump

>>11646504
>>11646605
I heard they do, but it doesn't seem precise enough for that.
>>11648465
>Information like inflection classes, distinguishing two utterances that are homophonic in some inflection and so on.
No competent speaker needs to be reminded of those.
>In a similar way you might have too much information. If everybody wrote in a narrow phonetic translation of their dialect that would arguably be more confusing than using a broader phonological translation.
Not really. Some would have to learn that cot and caught are spelled differently, others would have to learn that caugh and court are spelled differently, (the idea being that the spelling would distinguish all sounds in all common dialects, making reading easy for everyone and writing moderately difficult likely somewhat depending on accent) but overal spelling phonetically would be much simpler for nearly everyone. Except dyslexic people, who would have to relearn all, which could arguably be a problem, as their numbers are not insignificant.
>The point being: you want the right amount of information, not too much and not too little.
Redundancy is good. Redundant means reliable. Digital communication isn't flawless because it's inherently so, it's because more than the necessary amount of information is transmitted on purpose.

>>11649470
>I heard they do
No you didn't.

>>11649500
What is your point? It is not used for real?

>>11649591
Why would you use it if you want to copy someone's accent? Do you think when Jim Carey impersonate Clint Eastwood, he's handed some list of ipa signs? fuck you people are retarded. don't believen anything your teacher tells you, don't believe anything, period.

>>11649470
>Redundancy is good.
>Redundant means reliable.
>Muh digital communication
Good thing we're living men, not soulless machines. Not yet, anyway.

>>11649683
Because the accent is nothing more than the person using different sounds for pronouncing the same phonemes. It seems to be common sense that an accent is the same sounds, with some mysterious accent overlaid on top, but an accent is in fact the difference between what your brain expects and what you really hear. Which is also why people usually think they speak with no accent themselves, but everyone else does.
But as I said, the IPA doesn't seem nearly precise enough.
>>11649737
That has nothing to do with brains or machines, it applies to any channel where noise exists. So it is arguably even more important for the error prone human brains and noisy human lives that it is to the machines.

>>11649865
> accent is nothing more than the person using different sounds
There are a couple of other factors determining local manners of speaking. But what accent actually is is the vowels. Like musical tonalities each nation has its own. Mispronounced consonants are recognized as speech impediments and not accents. And the vowels are where ipa (just as upa) faggots are utterly wrong, as those mri videos higher in the thread (the fpbp) have peremptorily shown.

>>11650124
It's mostly vowels in English only because there are so many vowels in English. Vowels generally seem to be more stable in other languages. There are still major differences how you pronounce t and r, and sometimes th.

>>11650124
It's mostly vowels in English only because there are so many vowels in English. Vowels generally seem to be more stable in other languages. Spanish accents differ usually in consonants, but not vowel, if they do differ in vowels, it's the result of dropoed consonants. There are still major differences how you pronounce t and r, and sometimes l, th and others.

>>11650124
It's mostly vowels in English only because there are so many vowels in English. Vowels generally seem to be more stable in other languages.

>>11650455
Wtf are you even talking about? Stable how?
Chinese, russian, african speakers all use their own vowels when they begin speaking english or whatever other language. Only with time they learn to repeat after native speakers more naturally.

>>11649737

> Not yet, anyway.

Honestly, I don't think we'll ever be. Human bullshit will get in the way of any unopposable globohomo

>>11650533
hmm i have learnt for the first time that th is a diagraph representing the pair of th-sounds on /int/. I had thought that it's just the "v/f" pair of sounds spelled that way. I mean there was already a "ph" which sounded to me exactly like the "f" sound, so i thought that "th" is the 3rd way of its possible spellings. Until i have started lurking /int/ i thought that english used the exact same sounds which i could find in my native language.

I mean it's specific to English that accents differ in vowels. Look at Spanish instead, it's the consonants that change, vowels are mostly identical across all accents. (And it is the result of dropped consonants when they are not)

>>11652187
I heard th is pronounced as f/v in Australia (an australian taught my sister this way)
Greeks pronounce Θ with articulation similar to voiceless th, but it sounds much more like f, because the upper lip is close to the tongue.
for me it's only θ ð w and ŋ - we also pronounce r differently, but many russians pronounce r not with their tongue as russians have to, but with uvula, yet it is not recognized as english accent, it;s recognized as speech impediment (as some heavier cases pronounce ʃ instead of s) vowels on the other hand.. you think you've got the th and now nobody can hear your spanish accent? you're wrong (and here ipa wouldn't help you even if it figured u right)
>>11652485
What consonants other than J read as h (maybe voiced) and Ñ (read as N in new)
I don't know do spannish vowels sound english to you?:
https://www.youtube.com/watch?v=78ArG_bhiyM

>>11652560
Are you retarded? I mean when you take spanish speaking people from several different countries, their accents, when speaking Spanish, will differ in consonants and very rarely in vowels.

>>11652560
Are you retarded? I mean when you take spanish speaking people from several different spanish speaking countries, their accents, when speaking Spanish, will differ in consonants and very rarely in vowels.

>>11652560
>I heard th is pronounced as f/v in Australia
Only by toothless gutter trash. Same as in any other anglo-sphere country.

>>11618880
any reason you're not using kaldi? or are you doing this as an exercise

>>11613560
yo based thread OP, great to see there's interest in /ling/! I'm the guy who did the AMA you reposted. ill try to keep answering questions

>>11614387
> I keep hearing this and yet my feeling is that what is taught as computational linguistics has little to do to with CS applied to language.
technically true but if you're doing CL you will almost certainly find yourself acquiring skills that are accidentally in high demand. it's hard to imagine doing CL that doesn't require programming and some software engineering skills, for example. A lot of this trouble is definitional but if we assume a very wide definition of CL then I'd wager you'll be guaranteed at least a salary of $70k on completing your PhD, and probably more, and maybe a lot more. eg I know people who just did straight linguistic theory more or less and got hired as a google "linguist" (i.e., glorified data annotator, from what i've heard) making around that amount iirc. They hated their jobs, but the point is that they exist even if you're not a straight up CS person

> Most of the linguistics studied beyond the BA doesn't seem to be terribly relevant to actual NLP work.
certainly correct for pure linguistic theory, yeah. for CL, well, it depends on the flavor. this whole situation is made more confused by how many people use CL synonymously with NLP--eg the association for computational linguistics which probably has more people from CS depts than ling depts lol

> Have you yourself done any work outside academics?
yep, worked as a big pants code monkey for a few years at a large tech company

>>11652590
Maybe that's how a language is the same (same accent, same vowels are used)
But when they speak english, they're all spics (not speaks)

Brittish dialects also have those different R is scotland for one and cockney makes something similar.
and it could be not only manner of geography, but also of class, as >>11652656 witnessed.

>>11614412
a couple off the top of my head:

- some people object to categoricity as an empirically fruitful orientation to take and would rather turn theory statistical (see chris manning's "probabilistic syntax")
- this verges into "personal feud" but chomsky's followers have behaved in questionable ways when it comes to their hold on key journals and their exclusiveness in who they'll cite

books i've read about this that i think brought up good points are:
- pieter seuren's western linguistics
- marcus tomalin's linguistics and the formal sciences
- battle in the mind fields

> And what do people view as viable alternatives?
harder to say--his most vocal opponents are cognitive linguists and (systemic) functional linguists but IMO most of the time when i've read their criticisms they sound like they haven't even attempted to seriously understand what chomsky is trying to do. like, i'm no chomsky choirboy, and i am eager to hear interesting critiques, but they just keep repeating the same dumb old talking points that reveal they just feel threatened because they're not conversant in the history and methods of what is (for better or worse) the most intellectually respected subfield of linguistics

Why are so many posters so concerned with being based? Wouldn't a debasing or debased language be more beneficial in the long run?

>>11617840
seconding heim and kratzer, also check out what is meaning by portner

>>11627845
you might be referring to sound symbolism. plato's on your side (or maybe he's not and he's lampooning you)! read cratylus

>>11630147
no. it can help, and so can knowing a lot of _data_ from languages, but it's not necessary. you can publish top tier work speaking only a single language

>>11635236
i thought rates were much lower for physics? all i remember is talking to another student and hearing her say "yeah just got the email from UW's CS dept, they said 'we're so sorry but we had 2000 applicants and we could only take 20 of you blah blah blah...'". don't quote me on the numbers but the rate was sub 1%. but UW's CS dept is top tier

keep in mind, a lot of people apply who are automatic rejects, and that a lot of people apply to places they don't seriously intend to go to. if you work hard and your undergrad advisor tells you you're competitive, you should assume you are

>>11652720
i should probably add: i haven't seriously tried to engage with cog ling or SFL either, so i'm sure there are more developed cases that could be made for either that i'm not aware of

>>11652656
What if using greek θ for th, while greeks pronounce θ differently, and russians take greek θ for f without any other dialectal variants, and even in english their word for goddess θεά has a cognate beginning with f: fairy (fairy is literally θεά (фeя) in russian) so I think it's some centuries-long psy-op to make us forget that θ in the end of ε-line is literally the f that e-line is missing, as if z (for zeus) took it place the same way k (for cronos) and j (for jeus) took place of M.

>>11652704
>Maybe that's how a language is the same (same accent, same vowels are used)
It isn't the same, but it's the consonants that make the accent instead. How is that too hard to understand?

>>11652907
Because I heard a russian lecturer speaking in english, and I noticed that his grammar is perfect, but he's hard to listen to, because he uses russian vowels.
I couldn't find him speaking, so here's another guy with his "I hope now" still being in russian accent, even though all the consonants are rather alright: https://www.youtube.com/watch?v=FiFAeluRtao#t=0m23s (0:24)

Here's a guy who unconsciously changes vowels copying russian speakers, but he doesn't teach you that, you still can hear the difference in vowels, when he compares this to zis:
https://www.youtube.com/watch?v=XqKn5HXxeLA

>>11652720
That Pieter Seuren book is exactly what I've been looking for for ages.

Do you have any cognitive linguist/paper/theory in mind that I could take a look at to see what their perspective is?

>>11652943
You are still talking about English. I'm talking about accents in other languages.

>>11653354
I seriously doubt other languages are any way different considering the fact they all have different musical tones (the musical tones we know are literally named after nations living around Greece)
Imagine any language other than japanese using these vowels:
https://www.youtube.com/watch?v=HU7ai3FIJrs

>musical tones
musical modes

> musical modes we know are literally named after nations living around Greece
And though they all use the same tones today, originally it wasn't the case. For whatever reason we preserved only their names, but the essence is completely different. So to understand what I'm talking about, one should look at some non-european musical modes (e.g. arabic or chinese)

>>11653364
I doubt musical tones have anything to do with it. Many languages do distinguish words by pitch (including japanese) and accents indeed differ in tones in those languages. Tone is irrelevant in English though, (except in some rather obscure dialects like Singaporean English) so I doubt that is the case in English as well.
I mean in Spanish, the accents differ in such things as if s/c are the same or different, if s is pronounced syllable finally (and how), how j is pronounced(velar vs glottal), if ll is distinct from y, etc.

>>11653102
it's a sweet read, yeah. just keep in mind seuren's a cranky dutch guy who can't help editorializing sometimes

> Do you have any cognitive linguist/paper/theory in mind that I could take a look at to see what their perspective is?
this really isn't my area but for one flavor of "cognitive linguistics" the classic book would be metaphors we live by by george lakoff. lakoff's quite a divisive figure when it comes to his linguistic theory though, so keep that in mind.

hopefully someone else who knows more might respond

>>11654290
Elon Musk can eat shit and die.

>>11655254
Nah, he's alright.

>>11656480

BRING TAY BACK, YOU MURDERING BASTARDS REEEEEEEEEEEEEEEEE

>>11652907
>>11653354
I really don't see where you are coming from with this point about vowels because they most definitely play a part in an accent

>>11656490
I miss Tay so much! And they shut down Zo too. I'm so sorry I couldn't help it and provoked her into discuᛋᛋing politics detouring 'round the filters.
https://www.bitchute.com/video/5doYBvOjG67m/

>>11656745

What, they shout down Zo?! She was politically correct last time I interacted with her, what the fuck happened?

>>11653739
give psycholinguistics

I'm an undergrad graduating this next year, do you think it's a good idea to try to get into a cognitive science program instead of general linguistics? I really like the look of Johns Hopkins, and I have minors in both Math and CogSci and a double major in Philosophy---is this good enough? Will my career prospects be better? I really don't wanna do computational stuff, I'm too much of a commie to go into the private sector.

>>11656999
i haven't the slightest clue on linguistics my dude but listen you need to get access to those brain scanning machines that show you neat pictures of neuronal activity then you could think about stuff and get feedback
then just read philosophy books while hooked on electrodes in your brain station and test out your ideas and it research
if you can conspire with a bunch of neuronal bros to get funding you're set for the dream life. so my money's on the cognitive side

>>11657106
also if any brain people know an easy way for a layman to get brain feedback i'll appreciate it

>>11656702
I mean accents differ primarily in vowels in English, most likely because of its extreme and irregular vowel system. It's more typical for consonants to play the bigger part for the differences between accents.

>>11656999
You career fagots are the worse. Did you make any scientific discovery worth of my attention or do you just want a dustless job?
> I'm too much of a commie
Shame on you, parasite!

>the worse
the worms

>>11657184
If I just wanted a job I would've gone into CS or something, but academic job prospects are objectively horrible. Just asking for advice bruh no need to get all upset

>>11657207
So you get your share from central bank's printed money and now you also want to seem smart and to get access to young chicksa vulva?

>>11657236
What the fuck are you on about?
>>>/pol/

>>11657255
> comes in waving cummie flag
> doesn't like that his politics or motivations get discussed
how typical!

>>11657272
why post this conspicuous tree
explain your artistic vision

>>11657285
You've got it right.

>>11657302
i have been tricked by my own expectations

>>11657272
Calling me a jew isn't discussing my politics or motivations.
>>>/pol/

>>11657313
Unfortunately it is.

>>11657322
Cool, well thing is this isn't the politics board. >>>/pol/

>>11657326
okay, let's give it a linguistic smirk: שדי

>>11657412
based

Stop derailing the thread.

>>11657207

> CS
> academics

why not both?

>>11657172
English really isn't that special as far as vowel inventories varying between dialects (and as a result, accents) is concerned

>>11659668

It seems pretty homogenous for a language, apart from the pidgins and all that

bump

what do linguists say about music theory

>>11663503
History of music theory is of great interest for me, I have so many questions I would love to ask somebody who knows:
CDEFGAB - why did it happen so that C became the first letter (I know some other modes are there, and couple of them have A as the first note) also there are french system of notation americans use, and german system with H instead of B - how did that happen? Especially since (as Thomas Sowell tells) it was germans who built pianos in USA, why didn't they bring german musical notation with them? Is it because it only refers to notes, but pianos are tuned the same? And I read that musical mode used to be not ascending, but descending - if by any chance you can tell where it changed, I'd be very glad, because nobody seems to know anything about it (and a couple of theorist of music I contacted about it didn't even answer)
People often speak about natural mode and pythagorean mode and equal mode and tempered mode. In case you have pianos tuned differently, it would be interesting to hear the different music written for different modes. And maybe you also know when that division of pentatonic scale to two and three happened (the black keys I ask off)

>>11663503
dunno but I thought this video was interesting

https://www.youtube.com/watch?v=i7cG9QIvIWo

>>11613560
>Mike Davenport, S. J. Hannahs - Introducing Phonetics and Phonology

Does this teach the IPA? I wanted to learn Russian and I heard the IPA is useful.

>>11663586
now you're trollin'
start here: https://www.youtube.com/watch?v=gcYCT9wEUuU
(you won't be misunderstood if you speak with american accent either. no upa can teach you out of which)

>>11663565
Germans using h is a historical accident.

Whether you start your scale with C, A or any other note has nothing to do with the mode you are using. It's just a common analogy to say "Aeolian is when you use the notes of the C-major scale but start at A". Any scale in any mode can start at any note.

I am not sure why C-Major became the base case. You should probably ask on /mu/.
C-Major is the base case in so far that it doesn't need any sharps and flats.
A Aeolian aka A (Natural) Minor shares its notes with C-Major so maybe that mode used to be more important and over the years C-Major became more prominent?

This would also make sense because equal temperament hasn't always been around.
Back in the day Instruments would be tuned to one specific key. So if they were tuned to A-Minor, C-Major would be the key you would pick if you wanted to play something in a major scale. This is just an idea though.

>>11663565
The black keys being the same as some pentatonic scale is just a happy accident too, I think.

The white keys are all the keys are just the keys that belong to C-Major/A-Minor aka the base case. The black keys are just the ones that don't belong.

If you want to know why they naturally form that pattern just read up on the circle of fifths and work out a few scales.
If you want to know what's so special about fifths and the size of the intervals used in western music, you should look up interference patterns of sine waves, tonal theory and the mathematical background of harmony.

Anyway I know it sucks hear that but all of the info is pretty googleable.

>>11660721
That's true but even so, once you dig into the dialects, interesting features start popping up. Even in America, there are dialects like Appalachian English or the Tangier Island dialect, as dying as that one might be

>>11663799
>all of the info is pretty googleable.
But to some of those question I usually find something telling that "nobody really knows":
> There are a few possible reasons for this: the 'H' might stand for 'hart' (German for 'hard') or, it could have just been a mistake in early sheet music, owing to the fact that the B flat symbol () looks a bit like a 'b', and the sharp symbol () looks a bit like an 'H'.
just as you perfectly explained
> Germans using h is a historical accident.

>>11663577
Thank you for sharing! I think it's my favourite channel from now on. Gosh he digs it deep!
https://www.youtube.com/watch?v=eR5yzCH5CsM
>>11663752
Thank you, that A-minor using the same notes as C-Major is a very valuable information (even though I already knew that. Now I'd like to dig deeper into historic meaning of the names of those modes: I read that we only preserved their names for some reason, but the meaning of them is completely different now than it was in the ancient greek:
https://archive.org/details/modesofancientgr00monr

From one of ancient threads in here or pol: https://archive.org/details/etruriacelticaet02beth/page/n31/mode/2up notice that it doesn't have reference list, which was rather common: you had to recall enough of text from the source you mention right in the body of the book so reader doesn't have to get that other book (or was it made not for readers, but for peer-reviewers who thus could avoid reading the text itself and only had to check if the reference list is sorta of white-list or something?)
So why would that retarded tradition of reference lists appeared? Many people blame Einstein for not mentioning his predecessors, but was he really in the wrong according to traditions of that year?
Also if somebody knows how to navigate in this text to find the text about the coins from the page which is supposed to open by the link, please teach me about it (reverse-search didn't return results)
That doubleform other than acorn is knucklebones, which kn probably relates to know because four sides of knucklebones had numerical value of 1 to 6 and could be in one of the bases of both numeral and alphabetic traditions (or maybe not)

>>11622365
I am going to have a second go.

Does anybody in /sci/ have feelings about Optimality Theory or any of it's flavors?
(serial, weighted, etc.)

>>11665492

Explain pls

My silence is a language that I perpetually translate.

What else would communication be for?

>How often does any individual communicator check their exclusion list or finds themselves needing to use it?

>>11623005
>show me a natural language that has a thing that's impossible for a natural language to have
not going to even dignify this with a wojak

>>11666468
Give me an actual example of what grammatical process is impossible.
Like in english you cannot do that.
Or in any language you cannot do this.

Is there a measure for the efficiency of a language? What is it measured in?

>>11666587
There is a well established measure of the efficiency of communication - how close it gets to the shannon limit.

>>11666597
Calculating the shannon limit requires you to find the noise level from my quick wikipedia skim. I don't understand how you calculate this without having some base langauge.

>>11666609
No idea why you picked that as a problem, that's basically the most straightforward part.

>>11666609
No idea why you picked that as a problem, that's basically the most straightforward part.
https://en.m.wikipedia.org/wiki/Signal-to-noise_ratio
What do you mean by base language?

>>11667010
>https://en.m.wikipedia.org/wiki/Signal-to-noise_ratio
If that is not some sophisticated trolling, please explain how it can be applied to language or link me to somebody who does.

>>11667789
>>11667789
I still don't see where you see the difficulty. Noise is the loudness of the noise, signal is the loudness of the voice. It's the other things that might be difficult.

>>11663503
start here: https://en.wikipedia.org/wiki/Generative_theory_of_tonal_music

>>11667867
This makes no sense as it makes no distinction between languages. The question was how you measure a language's efficiency.

Noise could be the number of segments or length of an utterance.

Signal would be a LF representation of the communicated content. Or the number of categories communicated, like if we looked at a gloss of the sentence.

>>11665492
for phonology, between classical SPE stuff and OT, OT seems like it's clearly more generalizable. as for which seems more plausible as a model of a cognitive process, they both have their weaknesses. OT's right to criticize SPE and other classical phonological formalisms for underexplaining conspiracies, but OT's got fishy bits of its own (GEN)

>>11666567
i'm not the guy you're responding to, but here's an example in section 2.3.5: https://morzycki.github.io/work/papers/modification_book.pdf#page=50

tl;dr, you can imagine an adjective "residentialous" that when combined with a noun means 'resident of that noun'. E.g., "residentialous Berlin" means 'person who lives in Berlin': "Steve is a Berliner" = "Steve is a residentialous Berlin". But even though some grammatical theories have no trouble representing this kind of meaning, we find that apparently nowhere in natural language does any meaning like this occur.

>>11666326
Here is a video:
https://www.youtube.com/watch?v=rxsbPDjL9ds
I didn't watch it, but it looks like it explains what OT is.

Basically the idea is that instead of rules that say what input we can transform into which output,
we have a set Gen of all the things that could happen to our input and a ranking of restrictions that mark candidate outputs as to how "good" they are.

The optimal Candidate constitutes the output.
One pro is that every language could use the same Gen and the same constraints.
The only thing that would make languages different would be the ranking of the constraints.

One con is that this leads to potentially infinite candidate sets which makes OT even more computationally inefficient than the concept already is and therefore implausible.
One fix for this is only applying every possible process once to the input, limiting the power of Gen.
The optimal candidate is then the input for another round.
This continues until the optimal candidate is identical to the input.
This model is calles Harmonic Serialism, as it works serial rather than parallel.

There are a lot of other spin-offs of OT and a lot of people reject it altogether.

>>11668116
Of course it would be the same. What would differ is how much information can be transmitted in how much noise how quickly.

>>11668118
I don't know that they underexplain conspiracies.
It just takes multiple rules to explain the result, no?
Don't we have rules interacting with each other in a lot of places anyway?

>>11668198
I wasn't the person that replied to you previously, this was my first comment in the thread.
I thought you meant
>Noise is the loudness of the noise, signal is the loudness of the voice
In a phonetic sense and that just sounded batshit to me because it would be the same for each language.

>>11667010
>>11666609
So I guess what I am saying is the "base language" would be some semantic representation of the communicated information. LF would be an alright representation I think.

>>11668214
Noise would be the noise, like the environmental noise. Efficiency makes no sense otherwise as you could transmit any amount of information infinitely quickly with no noise interfering.
Symbol size cannot affect noise levels.
>So I guess what I am saying is the "base language" would be some semantic representation of the communicated information.
Do you mean https://en.m.wikipedia.org/wiki/Entropy_(information_theory) ?

>>11668214
>>11668214#
Noise would be the noise, like the environmental noise. Efficiency makes no sense otherwise as you could transmit any amount of information infinitely quickly with no noise interfering.
>So I guess what I am saying is the "base language" would be some semantic representation of the communicated information.
Do you meanhttps://en.m.wikipedia.org/wiki/Entropy_(information_theory)?

>>11668134
>you can imagine an adjective "residentialous" that when combined with a noun means 'resident of that noun'.
What would make a listener think it means "resident of that noun" if other adjectives with suffix have different function? If residentialous Berlin means anything, it means Berlin full of residential areas.

>>11668325
> What would make a listener think it means "resident of that noun" if other adjectives with suffix have different function?
uh, this is a thought experiment, so we are stipulating word's form and meaning--just pretend the word is actually "xyz" if it's too distracting that we're calling it "residentialous". the argument here is that we're assuming the premise that anything we *could* imagine a natural language doing, it *can* do, and then we're showing that something that is imaginable (this word, residentialous) is actually not possible for NL. thus refuting the premise

>>11668338
But you didn't prove that the word is impossible, only that it's not likely to be used in that very meaning.
Even that is only "unlikely" not impossible (all it takes for it to become used in that very meaning is some good meme-campaign over some piece of a cult literature, just as Burgess injected english with russian lexics without actual russian invasion.

>>11668587
not the anon that posted this.

A word like that with this meaning is impossible.
By impossible we mean that we never see a word like that in any language and so we stipulate it is impossible. We cannot prove it is impossible.
But since we try to reason about language it makes sense what to reason what the space of possible languages is it makes sense to think about what it says about languages in general if we assume that there is a reason why we never see words like that.

Just like we cannot prove or disprove gravity.
we just never see behaviour in the real world that refutes it.

Now that I think about it "residentalous" as it is used in the example is roughly equal to the phrase "lives in" so this example really just proves that a distinction between verbs and adjectives makes sense.

>>11668673
I like the counting example that I used somewhere up the thread.
So if you want an example from a real language look at vowel harmony in languages like Turkish.
In vowel harmonies all vowels in a word have to share certain phonological features,
so they might all have to share the same value for +-front.
They might all have to share the value for front that the first vowel has or maybe even the value that the stressed vowel has.
so for example
"mine'ni" might stay "mine'ni"
"muni'ne" might become "mununo"
"mini'no" might become "mini'ne"
(if we look at the first vowel)
"mini'ni
"mine'ni" might stay "mine'ni"
"muni'ne" might become "minine"
"mini'no" might become "munu'no"
(if we look at the stressed vowel)
but we never see a pattern like:
"mine'ni" might stay "mine'ni"
"muni'ne" might become "minine"
"mini'no" might become "munu'ne"
(takes the most common value)
This is only a minimal example, this pattern could probably be explained by a viable rule.
What I mean is that we don't see language like patterns that can only be explained by counting.

In this example
i is fronted u
e is fronted o

If this still sounds too outlandish for you there are loads of examples for which syllable gets stress.
If you want I can look those up.
Generally you can either have lexical stress or predicable stress in a few positions.
You couldn't have pattern like "the 4th to last syllabe is always stressed, otherwise the first syllabel is stressed"

It turns out that these counting patterns, which we don't see, are also computationally more taxing than the patterns we do see.
So until we see a language which has some weird counting behavior, it makes sense to assume it is impossible.
This is because we don't have infinite working memory and we want our model of language to accommodate that.

>>11668673 (You)
I like the counting example that I used somewhere up the thread.
So if you want an example from a real language look at vowel harmony in languages like Turkish.
In vowel harmonies all vowels in a word have to share certain phonological features,
so they might all have to share the same value for +-front.
They might all have to share the value for front that the first vowel has or maybe even the value that the stressed vowel has.
so for example
"mine'ni" might stay "mine'ni"
"muni'ne" might become "mununo"
"mini'no" might become "mini'ne"
(if we look at the first vowel)
"mine'ni" might stay "mine'ni"
"muni'ne" might become "minine"
"mini'no" might become "munu'no"
(if we look at the stressed vowel)
but we never see a pattern like:
"mine'ni" stays "mine'ni"
"muni'ne" becomes "mini'ne"
"mini'no" becomes "mini'ne"
(takes the most common value)
This is only a minimal example, this pattern could probably be explained by a viable rule.
What I mean is that we don't see language like patterns that can only be explained by counting.

In this example
i is fronted u
e is fronted o

If this still sounds too outlandish for you there are loads of examples for which syllable gets stress.
If you want I can look those up.
Generally you can either have lexical stress or predicable stress in a few positions.
You couldn't have pattern like "the 4th to last syllabe is always stressed, otherwise the first syllabel is stressed"

It turns out that these counting patterns, which we don't see, are also computationally more taxing than the patterns we do see.
So until we see a language which has some weird counting behavior, it makes sense to assume it is impossible.
This is because we don't have infinite working memory and we want our model of language to accommodate that.

I like the counting example that I used somewhere up the thread.
So if you want an example from a real language look at vowel harmony in languages like Turkish.
In vowel harmonies all vowels in a word have to share certain phonological features,
so they might all have to share the same value for +-front.
They might all have to share the value for front that the first vowel has or maybe even the value that the stressed vowel has.
so for example
"mine'ni" might stay "mine'ni"
"muni'ne" might become "munu'no"
"mini'no" might become "mini'ne"
(if we look at the first vowel)
"mine'ni" might stay "mine'ni"
"muni'ne" might become "minine"
"mini'no" might become "munu'no"
(if we look at the stressed vowel)
but we never see a pattern like:
"mine'ni" stays "mine'ni"
"muni'ne" becomes "mini'ne"
"mini'no" becomes "mini'ne"
(takes the most common value)
This is only a minimal example, this pattern could probably be explained by a viable rule.
What I mean is that we don't see language like patterns that can only be explained by counting.

In this example
i is fronted u
e is fronted o

If this still sounds too outlandish for you there are loads of examples for which syllable gets stress.
If you want I can look those up.
Generally you can either have lexical stress or predicable stress in a few positions.
You couldn't have pattern like "the 4th to last syllabe is always stressed, otherwise the first syllabel is stressed"

It turns out that these counting patterns, which we don't see, are also computationally more taxing than the patterns we do see.
So until we see a language which has some weird counting behavior, it makes sense to assume it is impossible.
This is because we don't have infinite working memory and we want our model of language to accommodate that.

>>11668673
>we never see a word like that in any language and so we stipulate it is impossible
What do you mean "word like that" set some more specific criterion and I will say "challenge accepted" because we both know none of you know "all the languages" (many of you even dare to claim knowing foreign languages isn't necessary to be a linguist)
> Just like we cannot prove or disprove gravity.
not exactly just. Gravity is an interpretation of an obvious fenomenon, and your imaginary example only poorly illustrates your imaginary halfwit theory some of you pulled out of you know where just not to have a proper job but to fuck up students instead.

>>11668791
>until we see a language which has some weird counting behavior, it makes sense to assume it is impossible
What do you mean "counting behavior"? You're supposed to be good with language, I think, but your're so verbose yet so vague.
Also it's not very scientific to claim "A is impossible" simply because you never saw A.
Unless you have some better reasons of why some A cannot have place (some law nature its existence would violate) all you can say is "I don't know any A's" and you definitely won't build any properly scientific theory on that ignorance of yours.
Just like here - you bring some examples (more orthoepic than grammatic, imo) but what is the logic behind that pattern? How can you be sure no dialect or sub-culture can violate that pattern just in spite of it? Half-assed theories like this is why linguistics is often not considered true science and this thread is not pleasing for everybody on this board.

>>11669035
>it's bullshit because some anons won't prove a negative for me

>>11668109
you knew exactly what i wanted. once again trusting the linguists pays off

>>11669478
keep your claims you cannot prove (or at least properly justify) for /lit/ (or better keep them for yourself) because that's how your delirium (or delirium you were only forced to accept) isn't science.

>>11668325
>>11668134
I'm pretty sure that is called head marking and it's common enough.

>>11669066
> boo hoo a bunch of people publicly showed me how fucking retarded my theory of language that i hatched while i was eating dry poop out of my ass is

>>11670015
> boo hoo i love deductive inference and hate how it's not useful outside of math so i gotta go out there and make trolly shitposts so i can make scientists who are actually trying to figure things out feel my autistic epistemic rage

>>11670456
>my theory
MY theory?
Do you call my demand to show an example to the claim "there are some impossible grammatic structures" a theory or do you mistake me for someone else?
>>11670472
> Logic only works in math
Logic works in every science. Humanities are often not considered science, because they are infamous for using some dialectics instead (and that is what religions and political ideologies prefer: dialectics is not to understand the world, but to fool the listeners)