[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/lit/ - Literature


View post   

File: 57 KB, 750x742, 1591653090606.jpg [View same] [iqdb] [saucenao] [google]
17861783 No.17861783 [Reply] [Original]

Hello Anons.
I recently heard that other languages have tenses similar to how we have 'past, future and present' tense but for how confident they are in the information they are speaking about. I have no clue if this is true but it feels like it'd solve 99.99% of the issues we currently have in the west.
As someone who tries to be very exact with how confident I am on information I wish english had this.

ITT: We make up endings to words to somehow convey confident levels to information similar to other languages.

>> No.17861811

>>17861783
jt;dr

>> No.17861832

>>17861811
;c

>> No.17861839

>>17861783
>...of the issues we currently have in the west
Examples, any of?

>> No.17861842

>>17861783
I haven't heard of any language with such a feature, it certainly can't be a European language. Do you remember which language it was?
>it feels like it'd solve 99.99% of the issues we currently have in the west
I can say with extreme confidence that it wouldn't.
(Note how your sentence already clearly communicates your confidence in the statements - "I have no clue if this is true", "it feels like".)

>> No.17861860

>>17861839
I think we are at the most divided we've ever been.
Why? Information and the transmission of it.
It's very difficult to communicate the level of confidence in data without doubling or tripling the amount of words.

In regards to issues, it feels like with everything political everyone is so certain. I've started equating it to reddit, where everyone says they know for certain and obviously x is the case. Idk all of our political issues we seem unable to find the truth to many of the problems we run upon, so much so that it feels like there's two realities in regards to these.

>>17861842
I believe I heard it was an Asian language? Maybe Chinese?

>> No.17861880

>>17861860
Ah, yes, Chinese, the cluster of languages in a totalitarian country inhabited by insect-people, sounds like a great solution to the problems of our culture.

>> No.17861887

>>17861783
English has 12 verbal tenses. Spanish has 16. Are you referring to that sort of thing?

>> No.17861893

>>17861880
At least they all seem to agree on what the facts are.

>>17861887
What are the 4 extra they have?

>> No.17861907

>>17861893
>At least they all seem to agree on what the facts are.
With the occasional help of tear gas and rubber bullets, yes.

>> No.17861919

>>17861860
>american problems

>> No.17861921

>>17861783
English has other means of communicating degrees of certainty, you're just have a bad grasp of the language and don't know how to do it effectively. Read more, write more.

>> No.17861929

>>17861921
Give examples whilst extending your sentences as little as possible.

>> No.17862032 [DELETED] 

That's not tense. It's having a grammatical requirement of putting something into preset categories of a specific piece of information. Some languages also have 'source'. e.g. witnessed vs non-witnessed vs common knowledge. You can do all this in English and any other language but these concepts are referring to specific inflections or grammaticalised things that strictly categorise. Communicating in a language without this usually includes more informatio, isn't grammatically required, or highly variable and irregular (impossible to say here's how and when 100% in a systematic prescription of the language).

Grammatical number and person also applies, many languages don't have either. That means a normal noun isn't 3rd person because there is no true, distilled person like that. The information is subsumed by other information or completely unnoticed/irrelevant. The practical result of this example is that 'referring to yourself in the 3rd person' is not a thing, not weird, for most words.

>>17861929
He went home.
He might've went home.

You can do it simply with an auxiliary verb. Not really the same because a true example would have that information distilled and grammaticalised.

>> No.17862061

>>17861783
You’re talking about grammaticalised evidentiality marking. It’s fairly common cross linguistically and if applied to English would really make very little difference.

>> No.17862096

>>17861783
It wouldn't solve all your problems but English first language speakers seem incapable of using the correct mood (esp. subjunctive vs indicative), remembering the right participle, or understanding a negative interrogative apart from blatant sarcasm.

>> No.17862102

>>17861783

Sounds like Imparfait in French.

>> No.17862154

>>17862102
Doesn’t that just correspond to past progressive? Also >>17861783
how do you feel about how I communicated my uncertainty in the preceding sentence?

>> No.17862213

>>17862154
You had to extend your statement significantly.

>He went home.
>He might've went home.

Almost as many letters as the rest of the statement to describe your uncertainty.

>> No.17862223
File: 45 KB, 1143x544, not grammaticalized.png [View same] [iqdb] [saucenao] [google]
17862223

>>17861929
These in the pic easily include more information/more distinctions than most languages with evidentiality marking as they don't usually have as many categories. Any language can do it including those with evidentiality marking. Just as we add ungrammaticalised temporal information in addition to tense-aspect. Another example would be adding prepositions to dative nouns. Though the dative case can be used alone and every noun must be in one case or another, it doesn't give you much so prepositions and more are used. The information it gives you is just 'distinction' from words in other cases.

>>17862213
Letters don't mean anything in English orthography. 'Might've' is one, short syllable.

>> No.17862237

>>17862213
>might have went
>might have went

>> No.17862238

>>17862223
It seems inefficient. Do you not agree that if you were to sit down and design a new version of English you could make this more efficient?

>> No.17862256

>>17861783
>I wish english had this.
I'm esl but isn't english nuanced enough, having may, might, should, could... ?

>> No.17862299

>>17862256
Not about it being nuanced it's about it being inefficient.
Source: I talk like a retard whenever trying to communicate anything political to attempt to provide exactly how I obtained and how confident I am on every piece of information. Doing this extends statements significantly. All I'm saying is can't we improve this?

>Bob looks hungry.
>Bob seems hungry.
>Bob would be hungry.
>Bob must be hungry by now.

vs something like

>Bob l'hungry.
>Bob s'hungry.
>Bob wb'hungry.
>Bob mb'hungry by now.

idfk I'm just thinking and based on what I heard about other languages.
ITT: We discuss problems in English and suggest improvements.

>> No.17862329

>>17862213
My man I don't use pictograms I use phonetics. These letters have a sound and the language its own cadence and emphasis. I need to ask you; WHY this is necessary? I can say "I am certain Paul left an hour ago." Or I am not sure if Paul left an hour ago." Or "Paul left an hour ago" or "Paul was here until an hour ago." Or "I think Paul left." Or "I bet Paul left." OR "Paul was here, but I think he left."
There. you have just conveyed certainty in the past tense.

Why do you need a hot and cold value to "I think Paul left." You could say "I am 20 percent certain Paul left and hour ago." But then your a dweeb and no on likes you.
You want to make another stupid rule for this language and have it be like "I thinkum Paul left an Hour ago." But then people think your retarded.

>> No.17862346

>>17862238
Languages are not designed. And constructed languages are 99% of the time splicing things from natural language, not designed from the ground up in any way.

How is it more efficient having grammaticalised evidentiality where there is only two distinctions (as most languages).

Firstly, you can't say both, it has to be one or the other. This is also why tense-aspect often doesn't really align to anything tangible.
Secondly, with so few distinctions, as is almost always the case with something grammaticalised (it is a process of information simplification and regularisation after all), you'll often want to specify further which means you will have doubled information needlessly. Having to say the same thing twice doesn't sound very efficient or precise. Precision comes from optionality. Grammaticalised things tend to become more abstract and intangible because of being obligatory. Not solid, precise information.

>>17862299
Ok dude you're an utter retard. Please read the image (article here https://en.wikipedia.org/wiki/Evidentiality)) and study linguistics before making a fool of yourself again. You don't even know what you're trying to change/add.

If you want to be clear in political discourse then do so, it has nothing to do with English or grammaticalised evidentiality.

>> No.17862391

>>17862346
Also, this might whet your appetite: http://www.ithkuil.net/

>> No.17862552

>>17862299
>>Bob looks hungry.
>>Bob seems hungry.
These have the exact same number of syllables as the original sentence "Bob is hungry". The third one has one more syllable, the fourth you couldn't even make up an unpronounceable contraction that would encompass all the new syllables, so in effect it also has the difference of only one syllable more than your proposed model.
In short, your ideas are pointless and dumb through and through.

>> No.17862983
File: 277 KB, 497x480, E07737F8-C6E8-4DDF-9F38-B261A9D792BC.png [View same] [iqdb] [saucenao] [google]
17862983

>>17861783
In Turkish there are two past tense suffixes, one for first-hand knowledge and one for uncertain knowledge. Only example I know of, if you want google "miş" suffix and see for yourself

>> No.17863007

>>17862983
Sounds useful anon.
Do Turkish people seem to be in better alignment politically then us in the west? Or atleast seem to know the facts?

>> No.17863014

>>17862983
came here to say this. its the same as adding supposedly or allegedly to what you are saying tho

>> No.17863023

>>17863014
Do you think people in english utilize this as they should as opposed to just making concrete statements?

>> No.17863087

>>17863023
in a way adding "miş" makes it more concrete. because adding "supposedly" introduces a bit of skepticism while adding "miş", depending on the context might imply "thats what they say at least" or "its definitely true but i just now heard it/heard it from someone else". we'll still use miş if an earthquake happens for example.

>> No.17863282

>>17861783
99% of the stuff you'll read online about language affecting culture is conjecture by people who have no idea what they're talking about. sapir-whorf was discredited a long time ago, and if there's any really meaningful link between language and some sort of cultural expression, it's the latter forming the former, not the other way around.

tl;dr having some confidence particle in english wouldn't really change shit

>> No.17863306

>>17863282
How do you explain tribes that don’t understand perceptions of time? Or the evolution of the word blue? Or any of the other trite things you can point out about a given language that effects a speakers perception of the world. If there is no linguistic context to individualized language then how do you explain the entire study of propaganda and the field of marketing and copy? It’s like you think words bare no meaning when that’s the only thing they actually do.

>> No.17863400

>>17863306
I don't wanna be an asshole so if this comes off that way lmk.

The Hopi tribe "having a different conception of time" was essentially baseless conjecture by Whorf, and other examples like the Russian distinction between sinij and goluboj (sorry for transliteration russanons) aren't really meaningful enough to make any sort of claim about their implications (Winawer's 2007 study basically had such a small distinction as to render is entirely meaningless). Most modern linguists are also very skeptical on grandiose claims about Pirahã being some miraculous linguistic exception as well, fwiw.

What OP (and Sapir-Whorfists in general) is claiming is that linguistic features have a direct and major effect on speakers' cultural perceptions of things. All the research we've run either categorically refutes claims in this vein, or is contradictory and inconclusive on them. Besides, if there are major intercultural/interlinguistic differences between cultures/languages, it's far more likely that a cultural difference resulted in a linguistic change, and not the other way around, like Whorfianism suggests.

Basically, to get back to the context of this thread, English having some confidence marker would not have a significant effect on our culture, and it's 99% likely that we just denote exactly what OP is talking about in some different way than the Turks do.

>> No.17863402

>>17861783
>but for how confident they are in the information they are speaking about

Youre delusional if you think this wouldn't be weaponized in rhetoric by corrupt politicians

>> No.17863412

>>17863007
AHAHAHA what are you doing?

>> No.17863418

>>17863402
Oh God, and advertising. Anon, this is a genuinely terrible idea.

>> No.17863421

>>17863400
Thanks for the reply. Sorry if my post was aggressive.

>> No.17863425

>>17863402
>>17863418
It's evidently already done in English

>> No.17863431

>>17863425
Seemingly obvious, forsooth.

>> No.17863434

>>17861783
>how we have 'past, future and present' tense but for how confident they are in the information they are speaking about. I have no clue if this is true
Yeah. E.g. Turkish:
> 1. OP'un annesi götten vurduruyordu.
> 2. OP'un annesi götten vurduruyormuş.

In the first sentence I have evidence that OP's mom is getting fucked in the ass. I know this because either I fucked her myself, or I've seen her getting fucked with my own eyes or on some video.

But in the second sentence, although the sentence is formally in past tense it might either mean that a) I've been told that OP's mom gets fucked in the ass (and this still happens to this day) or b) I've been told that OP's mom used to get fucked in the ass in the past.

>> No.17864136

>>17861783
It’s just a syntactic sugar for I heard that.
An example in Turkish,
gitti: he went.
gitmis: I heard that he went.
gitmismis: I heard that they say that he went (which I highly doubt).
gitmismismis: he didn’t actually go.

Also:
gitmisti/gittiydi: He had gone.

See there is so much more information density but it doesn’t actually convey a meaning that is not available in some other languages. The meaning is there.

>> No.17864407
File: 157 KB, 633x758, 1601253397106.jpg [View same] [iqdb] [saucenao] [google]
17864407

>>17863007

>> No.17864659

>I recently heard that other languages have tenses

>> No.17864734
File: 39 KB, 480x480, 56256342_122774958894853_2337040554913944696_n.jpg [View same] [iqdb] [saucenao] [google]
17864734

>>17862391
>ithkuil
are you into ithkuil anon? i took a break from learning 2011 because of the new language
im surprised no one mentioned the name of the grammatical category at the start of the thread. ig there's not many linguistics autists here...
desu i dont think it would change much though because in english you can always use phrases to mark evidentials, like "i heard from someone that-". it's just more concise
>>17862346
>Precision comes from optionality. Grammaticalised things tend to become more abstract and intangible because of being obligatory. Not solid, precise information
interesting, though i know with ithkuil at least a lot of the categories if unmarked are just assumed to be in their default sense. for instance, the default evidential is the confirmative, so i will always assume that is someone's meaning and i can really confirm what they are saying through direct observation unless they say otherwise

>> No.17864758

>>17863007
people will be stupid regardless of language. a sexy precise language only matters to people trying to discuss things seriously

>> No.17864773

>>17863434
I can confirm she does, yeah

>> No.17865347

>>17864758
and it's called maths.

>> No.17865674

>>17863400
Winawer's study showed there's a significant advantage in non verbal tasks, and beyond that the studies into blue cones show that Russian speakers are physiologically different in their ability to detect variant shades of blue. Pooh-poohing an anon on the evolution of the word blue when it is the kind of evolution you can see within an individual is going a bit far anon: it's like not understanding the fruit fly evolution experiment in middle school.

>> No.17866649

>>17861842
It's called evidential marking. Quechua has it, as an example off the top of my head; it's pretty common in New World languages, actually, I think.

>>17861887
English has two verbal tenses, Spanish has three. The rest are various combinations of tense, aspect, and mood.

>> No.17867767

>>17861907
As if they dont use tear gas and rubber bullets in the west.

>> No.17867802

>>17861842
>I haven't heard of any language with such a feature, it certainly can't be a European language
I don't know if it's exactly what OP is referring to, but what is called the subjunctive mood is quite common in Indo-European languages. It's found in English too and is often expressed with auxiliary verbs.