About
Content
Store
Forum

Rebirth of Reason
War
People
Archives
Objectivism

Post to this threadMark all messages in this thread as readMark all messages in this thread as unreadBack one pagePage 0Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9Forward one pageLast Page


Post 180

Friday, July 2, 2004 - 2:09amSanction this postReply
Bookmark
Link
Edit
What, exactly do you mean by universals and can you give some examples of universals and explain why they are universals, according to your understanding of the term, rather than just concepts (of existents or attributes) or just attributes?

To contrast with that, are there concepts or attributes you would not classify as universals, and can you give examples of those, if there are any?

End of Question
 
I have another request. Since you post and this question really relate to Ed's, "Intentional Conceptualism," to be fair to him, would you mind answering my question on that thread. It would be helpful if you cut my question and pasted it into you answer for reference there, that's why I marked it. This is only a suggestion, however.

Thanks!

Regi
Um... later perhaps.  I'm too lazy right now. 


Post 181

Friday, July 2, 2004 - 10:43amSanction this postReply
Bookmark
Link
Edit
Joe,

Beginning at the end:

Since I have no idea what you're looking for, I don't know if that answered your question.
 
You did a good job anyway. I only wanted to know exactly where you thought context was relevant to concepts and you told me exactly that.

Are you referring to the referents of a concepts?  If so, I don't see how a wider knowledge would change the referents.
 
Yes, that's exactly right. Context cannot change the meaning of the concept. I hear Objectivists say things all the time that sound like they think context affects the meaning, or that new knowledge expands the meaning of a concept. While we might discover new things (actual existents) that we did not formerly know were referents of a concept, the concept does not change. It always meant those referents, even if we didn't know they existed.

My only other comment is about this: ... when you form the concept, you go through a process of differentiation.  You observe the referents in a wider context of what you know, and recognize that they are similar to each other, and different from other things.
 
I do not disagree here, except that the, "difference from other things," is seldom recognized explicitly, nor is it required. Things could not be recognized at all if they were not different from other things. To recognize them implies they are different, else they would never have been noticed in the first place. The problem with recognizing a things difference from other things explicitly is that every thing and every class of things is different from everything else. To identify explicit differences it would be necessary to specify a thing's differences from every other kind of existent there is, or, at least the ones known.

To identify man as the, "rational animal," it is only necessary to specify what qualities a human being has, (all of the qualities of an animal plus the quality of rationality). It is not necessary to specify that man is different from the other animals explicitly, because it is implied by the fact he has a quality no other animal does, and animality is sufficient to differentiate man form all other things, e.g. plants, and the non-living.

Thanks again, Joe,

Regi
 




Post 182

Friday, July 2, 2004 - 8:11pmSanction this postReply
Bookmark
Link
Edit
Rodney, I beg your pardon for my misattribution. I’m not sure that I follow your point one, but in (2) you say: “…the objects may have any of a range of different lengths. That is the universal aspect of length that is mind-independent …which is recognized (as an objective fact) in the mental separation. It does not imply that length as such exists as some sort of entity in itself. Length is expressed only in the being of some physical thing.”

To my mind, the above expresses Aristotelian universals, or moderate realism. As I understand it, moderate realism is the view that universals have real existence in objects, but cannot be separated from those objects, so it’s true that length “as such” does not exists an entity in itself; it exists as an attribute “in the being of some physical thing.”  For a moderate realist, specific lengths also count as universals, since it’s claimed they can be repeated in numerous objects.

Rodney: “As a general point, I do not think it is fruitful to try to grasp AR’s theory by juggling past theories, which may be full of confusion and error …Simply observe reality first-hand…How did I first grasp the concept of length, and does it imply there is some metaphysical “essence of length”?”

Rand placed her theory of concepts within the context of the problem of universals, so it’s probably useful to investigate the actual problem and the previous solutions she was trying to correct.

I can’t remember how I first grasped the concept of length, but I would agree with Rand that abstract ‘length’ is derived from the consideration of the specific lengths of individual objects, and that the term ‘length’ refers to a quality of similarity or resemblance in this attribute between individual objects. In that case, while there is no universal quality in the object that is shared with like objects, the general term can be grounded in real-world objects, although not with the sort of precision implied by measurement omission.

Brendan 


Post 183

Sunday, July 4, 2004 - 10:41amSanction this postReply
Bookmark
Link
Edit
"[T]he previous solutions she was trying to correct": This phrase indicates the mistaken critical-rationalism premise. Rand did not build her theory by critiquing past theories; she basically looked at reality, and using logic first-hand (as opposed to testing the logic of others' theories), sought the facts. And all philosophers should do the same, rather than primarily evaluating preexisting ideas (which can easily slip into a primacy of consciousness premise, and I think often does).

Rather than taking a previous theory as a benchmark, and describing AR's as a "moderate" variety of it, one must recognize that reality is in no sense "moderate," that facts abide by the principle of the excluded middle. Reality is "extreme," and Rand's theory is "extreme" Objectivism. If there are elements of truth in some "realist" ideas, fine, say so and let it go at that; but continually adding refinements and qualifications to an idea that at best is muddled and at worst is mistaken, or combining multiple instances of such ideas in hopes of "approaching" truth, only compounds the confusion and is unscientific.


Post 184

Sunday, July 4, 2004 - 5:07pmSanction this postReply
Bookmark
Link
Edit
Rodney writes:
>This phrase indicates the mistaken critical-rationalism premise.

Which is...?

>Rand did not build her theory by critiquing past theories; she basically looked at reality, and using logic first-hand (as opposed to testing the logic of others' theories), sought the facts.

So, you are saying as she "looked at reality" and used logic, her theories are beyond criticism? (In other words, we must *take it or leave it*?)

- Daniel

Sanction: 9, No Sanction: 0
Sanction: 9, No Sanction: 0
Post 185

Monday, March 13, 2006 - 1:15pmSanction this postReply
Bookmark
Link
Edit
You can see the glaring mistake that Daniel Barnes is making here. He is actually deliniating a pretty good critique of deductive reason. I consider deductive reasoning to be one tool out of many in an overall process that will and should inevitably rely more on induction.

If deductive reasoning is relied on exclusively, or more plausibly if there is some PRETENSE that it must be relied on alone or some felt need that it ought to be the only tool then I would say (though many here would probably disagree) that Daniels criticisms are valid.

This is the gigantic mistake that Daniel is making:

Instead of calling the thing he is criticising by a name that would make him aware of his blunder. Instead of calling this thing 'Deductive Reasoning' he's calling it 'Essentialism' and blaming the whole thing on Aristotle.

"Karl Popper compared Aristotelian arguments to eternally sharpening your pencil and never actually getting around to writing anything."

Anyone whose read Aristotle would know that this is not so. Aristotle is a fellow who gets right to the meat of it and has written on an astounding array of subjects.

But anyone who has tried to read Popper would know that it was HE and not Aristotle that was mucking about with long-winded deductive 'windbaggery' (to use a Barnesian word).

Now here's the irony. It is Popper who tried to deep-six induction. It is HIS influence that therefore is likely to bog philosophy down in deductive overuse.

And the beauty of this is that Barnes has been able to try this on by being sloppy about defining his terms.

Post 186

Tuesday, December 29, 2009 - 5:04amSanction this postReply
Bookmark
Link
Edit
This is an oldie, but I believe it to contain an error: I think the layman form of the Bayes equation would read thusly:

[The probability of A, given B] =


[The probability of B, given A] x [The probability of A]
_________________________________________

[The probability of B]
Ed


Post 187

Tuesday, December 29, 2009 - 6:37amSanction this postReply
Bookmark
Link
Edit
Ed, I tried really, really hard, to follow this thread, and I read through 10 as you told Mr. Boese to do. It's not that I don't understand "probabilities"; I don't understand how they get to [The probability of A, given B]. A=A is not a probability that is affected by the existence of B. In one part of the thread it was mentioned that someone goes to the store for milk, and it isn't there. The possibility that the store might be out [B] does not alter the fact that milk has been purchased there before and will be again.

A=A doesn't mean the milk will be there. A=A means "Milk is sold in grocery stores." If it is true that a particular grocery doesn't sell milk, then A=A means "That particular store doesn't sell milk."

Change milk to "coffee". I walked into a convenience store last month in a low-income neighborhood. They did not sell what I was looking for--a cup of coffee. Not even a mechanical machine to make it cup by instant cup. While that still seems odd to me, the truth is that A=A in the case of that particular store means "That store does not sell cups of hot coffee."

Where does "probability" lie in such a proposition? Why would the Bayesians try to turn a known particular into a probibility?


Post 188

Tuesday, December 29, 2009 - 6:56amSanction this postReply
Bookmark
Link
Edit
Mr. Clark, some examples of using Bayes' Theorem are here.

Note that the examples given involve at least two contingencies. The milk and coffee scenarios you give in post 187 involve only one contingency.


Post 189

Tuesday, December 29, 2009 - 10:25amSanction this postReply
Bookmark
Link
Edit
Curtis,

Let me break it down for you, using Daniel. The research question would be this:

Given the incidence of Daniel appearing to lie (as in the other thread when questioned about relative frequencies), what is the probability of Daniel being a "troll"?

First, let's calculate the "prior" probability of lying [B]. This would be the probability of anyone lying, regardless of context. In the absence of other data, folks lie, let's say, about 10% of the time -- so the prior (or "background") probability of lying is 0.1. Now, let's say that the "prior" probability of being an internet troll [A] is 20%, or 0.2. And let's say that trolls lie 90% of the time (0.9). Now we can get the equation:

[the probability of someone being a troll, given that they have appeared to lie; (A, given B)] =

[the probability of them lying, given that they are a troll] x [the prior probability of them being a troll]
_______________________________________________

[the prior probability of them lying]

Now, let's punch in the numbers. The probability of Daniel being a troll, given he appeared to lie is ...

[0.9] x [0.2]
_______

[0.1]

So, the probability of Daniel being a troll is 1.8, or 180%.

Oh crap, something seems wrong with my math here. It's weird to have a probability be more than 100%. Will someone please take a few minutes of time and double-check my work?

:-/

Ed

(Edited by Ed Thompson on 12/29, 10:33am)


Sanction: 4, No Sanction: 0
Sanction: 4, No Sanction: 0
Post 190

Tuesday, December 29, 2009 - 10:50amSanction this postReply
Bookmark
Link
Edit
Ed,

Your understanding of the Bayesian formula is, shall we say, incorrect.

Let's try rephrasing the numbers a bit.

20%, or 200 out of 1000 forum posters are trolls.

90%, or 180 out of 200, forum posters who are trolls will lie.

10%, or 80 out of 800, forum posters who are not trolls will lie.

If there are 1000 posters, what fraction of posters who lie will actually be trolls?


The answer: The total number of posters who lie is 180+80=260. The fraction of those posters who lie who are trolls is 180/260 = 0.69 = 69%.


Post 191

Tuesday, December 29, 2009 - 10:58amSanction this postReply
Bookmark
Link
Edit
And just for fun, let's try a variation of the above - let's say that only 2% of internet posters are trolls instead of 20%. Given those numbers , can you say what fraction of posters who lie are trolls?


Post 192

Tuesday, December 29, 2009 - 11:07amSanction this postReply
Bookmark
Link
Edit
Ed T.,
You omitted part of P(B) = P(B|A)P(A) + P(B|A')P(A') here. A' = non-troll
Showing more clearly how Daniel followed Bayes' Theorem:
(0.9*0.2) / (0.9*0.2 + 0.1*0.8) = 0.18 / (0.18 + 0.08) = 0.18 / 0.26

(Edited by Merlin Jetton on 12/29, 11:16am)


Post 193

Tuesday, December 29, 2009 - 12:06pmSanction this postReply
Bookmark
Link
Edit
Thank you, Ed and Merlin, for trying. Maybe after I read it several times.

And as for you, Mr. Boese, the fact that you didn't quit while ahead but corrected Ed about "trolls", of which he accused you of being, doesn't say much for your character. You should have gone back under your bridge.


Post 194

Tuesday, December 29, 2009 - 2:50pmSanction this postReply
Bookmark
Link
Edit
Curtis,

I've just started a thread in the Banter board that you may wish to contribute to.

As for this thread, Ed asked for someone to help correct his work. I knew the correct way of doing it, and so I offered him the best help I could think of. Your post seems to indicate that you would prefer that I did not help him at all, and left Ed's misunderstandings uncorrected - that is, for me to not contribute even useful information to this board's threads - that is, for the board to be poorer by my not making even positive contributions at all, let alone make posts asking about Objectivism and trying to learn more about it. If this is truly how you feel, then you should be asking the moderators of this board to ask me to stop posting, or to put me on moderation, and/or to ban me outright.

But, in the meantime, I am curious - is my above explanation of Bayesian numbers done in such a way that you can answer my question, in which the percentage of trolls is changed from from 20% to 2%?


Post 195

Tuesday, December 29, 2009 - 9:52pmSanction this postReply
Bookmark
Link
Edit
Daniel,

My understanding of the Bayesian formula is, shall we say, improved.

And just for fun, let's try a variation of the above - let's say that only 2% of internet posters are trolls instead of 20%. Given those numbers , can you say what fraction of posters who lie are trolls?
Okay. Let's try rephrasing the numbers a bit.

2%, or 20 out of 1000 forum posters are trolls.

90%, or 18 out of 20, forum posters who are trolls will lie.

10%, or 98 out of 980, forum posters who are not trolls will lie.

If there are 1000 posters, what fraction of posters who lie will actually be trolls?

The answer: The total number of posters who lie is 18+98=116. The fraction of those posters who lie who are trolls is 18/116 = 0.16 = 16%. Which makes it highly unlikely that even liers are trolls.

In other words, if trolls aren't common in the first place (if only 1 in 50 online are trolls), then I won't think that you are a troll -- regardless of all of your other actions. The beauty of Bayesian thinking is that you get to omit so much relevant data (rather than integrate it like you have to do in Objectivism). Like utilitarianism, you just simply pop out certain numbers like I did, above (utilitarians use numbers called "utils"), claim that they are good numbers which were "hard-earned" -- and then you get to take over a whole country with Utopian social engineering!

Weeeeeeeeeeeeeeeeee!

I read your link and it talks about the very error I committed. It said that somewhere around half of doctors -- when presented with percentages (rather than relative frequencies) -- got it wrong. So, the upside is that I have a 50% chance of being as smart as a doctor. Which sure has me relieved. I am especially comforted by the fact that I can prove this now with ... [yep, you guessed it!] ... Bayesian reasoning!

Woooohoooooooooo!

Troll or not, Daniel, I'll be the first to admit that I have learned from you. For starters, you've been a great help to me in understanding how far you can go wrong with Bayesian formulizationing.

Ed

(Edited by Ed Thompson on 12/29, 10:07pm)


Post 196

Tuesday, December 29, 2009 - 10:07pmSanction this postReply
Bookmark
Link
Edit
Ed,


> The answer: The total number of posters who lie is
> 18+98=116. The fraction of those posters who lie who are
> trolls is 18/116 = 0.16 = 16%.

And we have a winner! Feel free to accept a gold star. :)


> My understanding of the Bayesian formula is, shall we
> say, improved.

> Troll or not, Daniel, you've been a great help to me in
> understanding how far you can go wrong with Bayesian
> formulizationing.

I'm glad I was able to help. :)

I was specifically trying to offer a way to think about Bayesian logic that completely avoided the standard mathematical formula. For a lot of people, that sort of formula is a 'black box', effectively a piece of magic that they don't understand... but if understanding of the principles behind the formula can be gained, then the formula itself is almost an afterthought.

Plus, as Eliezer put it, "Do not walk to the truth, but dance. On each and every step of that dance your foot comes down in exactly the right spot. Each piece of evidence shifts your beliefs by exactly the right amount, neither more nor less. What is exactly the right amount? To calculate this you must study probability theory. Even if you cannot do the math, knowing that the math exists tells you that the dance step is precise and has no room in it for your whims."


Post 197

Tuesday, December 29, 2009 - 10:11pmSanction this postReply
Bookmark
Link
Edit
Daniel,

You posted too soon (my last edit was finished just after your post got posted). It may behoove you to double-check to see if you missed anything crucial.

Ed


Post 198

Tuesday, December 29, 2009 - 10:14pmSanction this postReply
Bookmark
Link
Edit
Ed,

(responding to what you added in your edit to post #195,)

> In other words, if trolls aren't common in the first
> place (if only 1 in 50 online are trolls), then I won't
> think that you are a troll -- regardless of all of your
> other actions. The beauty of Bayesian thinking is that
> you get to omit so much relevant data (rather than
> integrate it like you have to do in Objectivism). Like
> utilitarianism, you just simply pop out certain numbers
> like I did, above (utilitarians use numbers called
> "utils"), claim that they are good numbers which were
> "hard-earned" -- and then you get to take over a whole
> country with Utopian social engineering!

Actually, this is based on a somewhat naive view of Bayesianism... which becomes much more useful, and more powerful, when applied to /multiple/ pieces of evidence; so that, for example, if someone demonstrates multiple individual pieces of evidence of trollery, then, while no individual piece of evidence would be enough to significantly increase the odds that they are a troll, when combined, each one adds a certain amount to the percentages, eventually tipping the odds.

This is, in fact, precisely how Bayesian spam filters operate, to determine whether or not a piece of email is spam or not.


(Also, I could be wrong, but I think the word you are trying to refer to is "utilons", not "utils".)


Post 199

Tuesday, December 29, 2009 - 11:05pmSanction this postReply
Bookmark
Link
Edit
Daniel,

I have been hard at work in my philosophical arm-chair, and I have pain-stakingly come up with a useful new scale using a soon-to-be-commonplace standard.

It's called the "U-troll." If someone online performs some actions frequently enough, then they earn a new "U-troll." If the frequency of "U-trolls" that they earn breaks 50%, then they can be decisively categorized as a "UTrollitarian." A variant of the "UTrollitarian" is when someone persistently but arbitrarily bids up the price of products auctioned online (in order to establish the greater good, as always). This variant is then called an "E-BAYesian."

In cases where both behaviors are present, you get a "UTrolliBayesian" -- someone who not only takes up your time, but also your money.

Ed

(Edited by Ed Thompson on 12/29, 11:05pm)


Post to this threadBack one pagePage 0Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9Forward one pageLast Page


User ID Password or create a free account.