I'm growing to hate linguistics

Mundane & Pointless Stuff I Must Share: The Off Topic Forum

Moderator: Moderators

Post Reply
Catharz
Knight-Baron
Posts: 893
Joined: Fri Mar 07, 2008 7:54 pm

I'm growing to hate linguistics

Post by Catharz »

I'm taking a class in introductory linguistics, and much to my horror I've discovered that a lot of linguistics is just bullshit.

Steven Pinker ("The Language Instinct") setiously uses the argument "But he's wrong, wrong, wrong" to help 'disprove' linguistic relativism. Noam Chompsky is a fucking genius, but he's also a conceited prick.

Aside from personal attacks, the real reason I'm posting this is the following:

In formal logic, one of the common constructions is the "implication," following the notation p -> q == 'if p then q.'
An interesting thing about implications is that if p -> q is true, then 'not p' -> q is also true. This allows for vacuous proofs, which are damn' useful.

Anyway, I'm sure many of you know about symbolic logic, so I'll get to my point. In syntactics, lingusts deal with something called "phrase structure rules." These rules are things like 'if you combine an adjective and a noun phrase, you get a noun phrase.' They're written like "NP -> Adj NP."

In other words, symbolic logic. But, in their infinite wisdom, the linguists flipped the arrows! Not only does this make no sense (the rules are 'productive' in the opposite direction of the arrows), but it's god damn fucking confusing to someone who is used to logic!

*deep breath*
Ok, I think I'm done ranting.
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by Username17 »

Catharz wrote:
In formal logic, one of the common constructions is the "implication," following the notation p -> q == 'if p then q.'
An interesting thing about implications is that if p -> q is true, then 'not p' -> q is also true. This allows for vacuous proofs, which are damn' useful.


That's not true.

P -> q is true if p*q, it's true if ~p*q, it's true if ~p*~q, and it's false if p*~q.

~p -> q is true if ~p*q, it's true if p*~q, it's true if p*q, and it's false if ~p*~q.

The two statements are ture and false under different circumstances. While both statements are true if both p and q are true, the statements are in no way the same and cannot be used to prove each other.

---

As to linguistic relativism, it is definitely true that speakers of some types of Chinese perform basic mathematical functions with a different part of the brain than do speakers of some types of Russian - leading to an objective proof of some of the conceits of the Sapir-Whorf hypothesis. However, those who would argue that our labels for languages have some sort of objective meaning is a hillarious and indefensible notion.

Just the existence of "Chinese" - a "language" that has multiple non-intelligible grammatical structures is pretty much the elephant in the room as far as the idea that our language labels are anything more than political contrivances.

There is a real reason that we describe Chinese as a language. But it has more to do with the extent of the Han Dynasty than it does to the actual experience of people speaking and listening to one another. Sometimes we talk about "Serbocroatian" as a language and sometimes we talk about "Serbian" and "Croatian" - really the only difference is whether we want to accent the similarities or the differences of the peoples who live in former Yugoslav republics.

-Username17
User3
Prince
Posts: 3974
Joined: Thu Jan 01, 1970 12:00 am

Re: I'm growing to hate linguistics

Post by User3 »

I think he meant that the conditional is true if and only if the contrapositive is true.

In other words (well, symbols):

p -> q <=> ~q -> ~p
Catharz
Knight-Baron
Posts: 893
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by Catharz »

FrankTrollman at [unixtime wrote:1175105476[/unixtime]]
That's not true.

I was being stupid. What I meant was that if P is false, the implication is still true. Which you already said.

FrankTrollman at [unixtime wrote:1175105476[/unixtime]]As to linguistic relativism, it is definitely true that speakers of some types of Chinese perform basic mathematical functions with a different part of the brain than do speakers of some types of Russian - leading to an objective proof of some of the conceits of the Sapir-Whorf hypothesis. However, those who would argue that our labels for languages have some sort of objective meaning is a hillarious and indefensible notion.

It's cool how Mandarin acts almost like a "functional" language in its isolating morphology. Also makes it easier to learn.

Oh, one of the intersting things I learned by reading Whorf: He never actually articulated the theory attributed to him and Sapir. He implies it, but the "Hypothesis" is in many ways just a straw man.

FrankTrollman at [unixtime wrote:1175105476[/unixtime]]Just the existence of "Chinese" - a "language" that has multiple non-intelligible grammatical structures is pretty much the elephant in the room as far as the idea that our language labels are anything more than political contrivances.

There is a real reason that we describe Chinese as a language. But it has more to do with the extent of the Han Dynasty than it does to the actual experience of people speaking and listening to one another.

China might be considered special based on how much they invest in 'One China' propaganda.

FrankTrollman at [unixtime wrote:1175105476[/unixtime]] Sometimes we talk about "Serbocroatian" as a language and sometimes we talk about "Serbian" and "Croatian" - really the only difference is whether we want to accent the similarities or the differences of the peoples who live in former Yugoslav republics.

-Username17

That and Serbian is written in Cyrillic, right?
User avatar
josephbt
Knight
Posts: 325
Joined: Fri Mar 07, 2008 7:54 pm
Location: Zagreb, Cro

Re: I'm growing to hate linguistics

Post by josephbt »

Yeah, you really shouldn't use "Serbo-Croatian" anywhere near a Serbian or a Croat. People here still have ethnic problems, and saying that it is the same language could get you really hurt. I'm not kidding.
Here, in Croatia, there is an institute whos only purpose is to establish the differences between Serbian and Croatian language. The fact that it is harder for someone from NW Croatia to understand someone from SE Croatia, than it is to understand someone from anywhere in Serbia, doesn't give them any problems.

wrote:Claimer: I'm sane. I don't care if you call it Serbo-Croatian or Serbian or Croatian.

engi

Blood for the Blood God!
User avatar
tzor
Prince
Posts: 4266
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by tzor »

I admit that I am on occasion dense so I'm going to have to do the logic on this the long way. From my computer background I was always under the impression that "if X then Y" (or X -> Y in your notation) was represented in logic as NOT(X) OR Y or in the more familiar notation !X+Y

Remember that in logic statements MUST BE TRUE.
If Y is true then the statement is always true.
If X is false then the statement is always true.

"If I have blue hair then Frank is correct" is true because I do not have blue hair.

Now what does (Not X) imply? Using logic that would become X+Y and there is no way that !X+Y implies X+Y at least not that I can think of ... but let's do the math

(!X+Y)->(X+Y)
= !(!X+Y)+X+Y
... (As I got confused with all the logic math I just threw up a spreadsheet and looked at the answers)
= !(!X & !Y)

So in other words (X->Y)->(!X->Y) is simply not true when both X and Y are simply not true.

Does that make sense? Because I tend to find logic so confusing.
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by Username17 »

You don't negate portions of statements. You compare whole statements and determine whether the statements are true or false.

If you have multiple statements, and those statements contradict each other you don't have a valid argument - your premises are simply wrong.

So if you have:

X and Y
!X

Then you have false premises.

-Username17
Neeek
Knight-Baron
Posts: 652
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by Neeek »

tzor:

You are using very bad notation, for starters, so bad that I'm having trouble understanding wtf you mean.

+ is, if used, usually "and" (is only true if both sides are true)
v is usually "or" (is true if either side is true)
-> is implication (if/then, is true if both parts are true or the first part is false)
<-> is equivalence (true if both sides are the same)
! or ~(or several other things) is negation (true if the statement is false)

From there, there are a fairly large number of rules that have to be followed to manipulate these symbols.

For example:
(!X v Y) -> (X v Y) can be rewritten as:
!(!X v Y) v (X v Y) which can be rewritten as:
(X + !Y) v (X v Y) which is to say the premise is true when X is true or Y is true, because (X + !Y) is only true when X is true and Y is false, and the "or"s make the whole statement true whenever that part is true.
technomancer
Journeyman
Posts: 140
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by technomancer »

In computer logic, OR is +, AND is *. This makes sense if you look at it the way computers do -- false is 0 and true is anything else (usually 1).

A * B is AND, because if one of the two is 0, then the whole statement is 0 (False)

A + B is OR, because if at least one of the two is 1, then the whole statement resolves to 1 or more (True)
Catharz
Knight-Baron
Posts: 893
Joined: Fri Mar 07, 2008 7:54 pm

Re: I'm growing to hate linguistics

Post by Catharz »

Huh, from discrete math I'm accustomed to the notation of ¬ = NOT, ^ = AND, v = OR, &#8853; = XOR, -> = implication, and <-> = equivalence.

Oddly, we never really went over the equivlence of * and + to OR and AND, although we did use binary matricies to represent logical operations. My thoughts on computer logic have been tainted by Lisp, where 0 is true and the only false is Null (= (), #f).

Anyway...
So in other words (X->Y)->(!X->Y) is simply not true when both X and Y are simply not true.

In philisophical logic Frank is probably right. If it's representing a system, it's false (that is,. it crashes) exactly when you said it is.

If X is false any Y is true, the statement is true (T)->(T).
If X is true and Y is true, (T)->(T).
If X is true and Y is false, (F)->(T).
If X is false and Y is false, (T)->(F), which is false.

The whole damn' thing is vacuous because you can ignore everything on the left side of the conditional. It's equivalent to !X->Y.

At least I think that's how it works...
Post Reply