Why do we use logic?

[chompchompchomsky]

Logic, in an Aristotelian sense, that structures things in a progressive gradual understanding moving in one direction. Algorithmic logic, too, is pervasive in our creations (Computers, and their derivatives). And yet our minds don't do that! Our minds use a function of conceptual blending, whereby we arrive at conclusions by blending together unrelated concepts to form new, meaningful things: But we don't make anything that can do that, and we don't communicate in that way? Why do we translate our cognitive form of reasoning into a completely different model? Where is the usefulness in that?

[Agent 008]

Neural networks.

/thread

[chompchompchomsky]

Neural networks.

/thread

cop out.

[chompchompchomsky]

No ideas? Nobody?

[Reality Apologist]

What can you possibly mean by 'where's the usefulness?' The computer you're typing on right now, the electricity powering it, the cell phone in your pocket, and virtually every other piece of modern technology is fundamentally based in a mathematically represented science, which is based in the laws of logic. We use it precisely because it helps us do things like this--it's a very powerful way of representing the world, even if it doesn't capture every interesting aspect of it.

[a334jv2df]

Logic, in an Aristotelian sense, that structures things in a progressive gradual understanding moving in one direction. Algorithmic logic, too, is pervasive in our creations (Computers, and their derivatives). And yet our minds don't do that! Our minds use a function of conceptual blending, whereby we arrive at conclusions by blending together unrelated concepts to form new, meaningful things: But we don't make anything that can do that, and we don't communicate in that way? Why do we translate our cognitive form of reasoning into a completely different model? Where is the usefulness in that?

What's 1234+1*1?

/thread

[Gantz Graf]

I figured this would be obvious, logic probably exists due to the symbolic nature of our cognition. If you're constantly blending symbols and coming to new conclusions, you'd have to have some sort of internal 'bullshit detector' to stay grounded and to make sure you weren't just jumping at shadows.

[Haiti's Space Agency]

What's 1234+1*1?

/thread

one thousand two hundred and thirty-five.

[Reality Apologist]

one thousand two hundred and thirty-five.

That seems like a logical conclusion.

[chompchompchomsky]

What can you possibly mean by 'where's the usefulness?' The computer you're typing on right now, the electricity powering it, the cell phone in your pocket, and virtually every other piece of modern technology is fundamentally based in a mathematically represented science, which is based in the laws of logic. We use it precisely because it helps us do things like this--it's a very powerful way of representing the world, even if it doesn't capture every interesting aspect of it.

Yet it is a completely foreign way of representing the world, and it doesn't capture any of the "interesting aspects" of algorithmic thinking because it is an entirely different, complete system. What I'm asking is why.

[chompchompchomsky]

I figured this would be obvious, logic probably exists due to the symbolic nature of our cognition. If you're constantly blending symbols and coming to new conclusions, you'd have to have some sort of internal 'bullshit detector' to stay grounded and to make sure you weren't just jumping at shadows.

OBVIOUS????? Yes! Indeed you do have an internal bullshit detector, but you don't have an external one, you've translated your cognitive style entirely. again, why?

[Gantz Graf]

OBVIOUS????? Yes! Indeed you do have an internal bullshit detector, but you don't have an external one, you've translated your cognitive style entirely. again, why?

Legibility/joy in writing/ease of interpretation/all of the above?

[chompchompchomsky]

It is only legible, joyful and easy because you're accustomed to the translation. But it would be just so without any translation, wouldn't it?

[Gantz Graf]

It is only legible, joyful and easy because you're accustomed to the translation. But it would be just so without any translation, wouldn't it?

Is it possible for the synthesis of internal symbols to be conveyed in their natural/pure form?

Is their any other option?

I'm content with imperfect translations if the only alternative is an even more incomplete translation.

[Syphilis]

Logic, in an Aristotelian sense, that structures things in a progressive gradual understanding moving in one direction. Algorithmic logic, too, is pervasive in our creations (Computers, and their derivatives). And yet our minds don't do that! Our minds use a function of conceptual blending, whereby we arrive at conclusions by blending together unrelated concepts to form new, meaningful things: But we don't make anything that can do that, and we don't communicate in that way? Why do we translate our cognitive form of reasoning into a completely different model? Where is the usefulness in that?

Actually, our minds do act logically. Just because you can't see or understand the logic, it doesn't mean it's not there.

[DarkMage35]

When you try to logically explain the basis for logic, things can get weird. We do what we must because we can.

See: Zen and the Art of Motorcycle Maintenance, Taoism, Nihilism

[Sanchez_Machine]

It makes me look smarter than I actually am. :D

[Giggles The Panda]

When you try to logically explain the basis for logic, things can get weird. We do what we must because we can.

See: Zen and the Art of Motorcycle Maintenance, Taoism, Nihilism

+1 for the peasant with the Portal reference.

[Ambient]

I am still quite confused as to what logic is exactly; and whether it is synonamous with a principle like rationality.

I suppose we use logic because very often, it is of benefit to use in a material sense; and in a spiritual sense provides secuirity within ourselves.

In other words, we like to know whats going on!

Anyone wanna clear up the logic/rationality thing?

P.S The philosophy of pythageris may be of interest to you; hes definitly one of my favourite mathmaticians ( who trivially also had a semi- cultic/religous following back in the day which I thought was funny, because it was essentially a math's cult).

[Uriah Heep]

Logic is how we make sense of things to ensure proper communication and understanding.

[Ambient]

Logic is how we make sense of things to ensure proper communication and understanding.

How is this different from say, a mythological yarn?

Isnt the purpose of such a thing also clarity or "to make sense of things"; and that this quality will follow through in our linguistics and literature.

[Uriah Heep]

How is this different from say, a mythological yarn?

Isnt the purpose of such a thing also clarity or "to make sense of things"; and that this quality will follow through in our linguistics and literature.

Well unless you are talking about mythological yarn it works, perhaps you are confusing logic with using words in proper context?

I dont know, i'm just trying, i'm no english proffesor

[Agent 008]

The real question, is our brain equivalent to a Turing machine?

In other words, can our brain fundamentally be computed (simulated) on any computer?

[Ambient]

Well unless you are talking about mythological yarn it works, perhaps you are confusing logic with using words in proper context?

I dont know, i'm just trying, i'm no english proffesor

Nah im just confused now :p

I was trying to structure my reply the same way you structured your post.

So you talked of logic as "making sense" which would follow through into communication and understanding.

So I said.

Isnt mythology something which people and cultures use for "making sense" which would follow through into communication and understanding.

To you agree with the similarity I have drawn?

[Ambient]

The real question, is our brain equivalent to a Turing machine?

In other words, can our brain fundamentally be computed (simulated) on any computer?

Elements/Incompletley yes.

Houlistically/Completley no.

Might depend on what you mean by "fundamentally".

Just a guestimate though; but I wonder if a computer can conceptualise things like God or divinity and such.

The human mind is not really an imitator or a simulator; more of an emulator. Im not sure a robotic or digital agent is capable emulating human agency.

Thats what cyborg's are for :)

[Σnigma]

The human mind is not really an imitator or a simulator; more of an emulator. Im not sure a robotic or digital agent is capable emulating human agency.

They couldn't. The fundamental basis of the computer is the proof; if x = y, then such and such happens. To try and go from linear logic to human rationalization is beyond our ability to recreate, because frankly, we don't even know how our own brains work. How can we recreate something we don't even know?

[Reality Apologist]

Logic is a set of principles for reasoning formally (that is, symbolically--in a way that is content-neutral) about propositions. It is a set of rules that tell you how to (and how not to) infer the truth of propositions from other true propositions.

Modus pones is a classic example of first order logic:

P --> Q
P
---------
Q

If P, then Q.
P is true.
Therefore, Q must be true.

Logic is useful, in part, because the variables 'P' and 'Q' can stand for any proposition whatever:

If Jim is a nert, then Jim is a skert.
Jim is a nert.
Therefore, Jim is a skert.

That's just as valid as something like:

If I drank the whisky, then I will be intoxicated.
I drank the whisky.
Therefore, I am intoxicated.


The power of logic lies (in part) in this ability to reason about all sorts of propositions without respecting the content of the proposition: being able to reason about the world in a totally syntactic way is a very powerful resource.

[Silverfuck]

It terrifies us that we are fallible, that our knowledge and understanding of the world is fallible. Logic is a tool we use to quiet that fear and convince ourselves that we can understand.

[Reality Apologist]

They couldn't. The fundamental basis of the computer is the proof; if x = y, then such and such happens. To try and go from linear logic to human rationalization is beyond our ability to recreate, because frankly, we don't even know how our own brains work. How can we recreate something we don't even know?

You're certainly right that digital computers won't be able to duplicate consciousness: I think the Chinese Room (http://en.wikipedia.org/wiki/Chinese_room) shows that pretty convincingly. That's not to say, though, that we won't be able to simulate the activity of the brain with a powerful digital computer. We might get something that is behaviorally indistinguishable from consciousness, even if the mechanism behind that behavior is interestingly different. This is what the Blue Brain (http://en.wikipedia.org/wiki/Blue_Brain_Project) project is aiming at--by exactly modeling the behavior of individual neurons, the Blue Brain people hope to get something that behaves in every way as if it were conscious.

[chompchompchomsky]

Thank you, Apologist, for posting a formal definition of logic. (that is what I meant by it). And it's pretty easy to see that our minds don't do that. And no, Syphyllis, Our minds do not use logic, they use conceptual blending, and it is not just some extraordinarily complicated version of logic: it is a completely different system.

[Uriah Heep]

Nah im just confused now :p

I was trying to structure my reply the same way you structured your post.



So I said.



To you agree with the similarity I have drawn?

Ahh yeah haha, Dont worry I couldnt get any deeper, but I do see your point.

[Mor3BL7]

Logic, in an Aristotelian sense, that structures things in a progressive gradual understanding moving in one direction. Algorithmic logic, too, is pervasive in our creations (Computers, and their derivatives). And yet our minds don't do that! Our minds use a function of conceptual blending, whereby we arrive at conclusions by blending together unrelated concepts to form new, meaningful things: But we don't make anything that can do that, and we don't communicate in that way? Why do we translate our cognitive form of reasoning into a completely different model? Where is the usefulness in that?

tl/dr

[Reality Apologist]

Thank you, Apologist, for posting a formal definition of logic. (that is what I meant by it). And it's pretty easy to see that our minds don't do that. And no, Syphyllis, Our minds do not use logic, they use conceptual blending, and it is not just some extraordinarily complicated version of logic: it is a completely different system.

You're absolutely right that our minds don't work like that--philosopher/cognitive scientist Andy Clark sums up our cognitive abilities as being "good at frisbee, bad at logic." That is, we're quite good (better than anything we've seen so far) at identifying all kinds of patterns and predicting how those patterns will proceed, but we're bad at careful, formal, syntactic reasoning. That's why we use logic as a system though--we're pretty good at following simple rules, and the structure of logic lets us transform complex cognitive tasks (like reasoning about the consequences of a large number of related propositions) into a set of simpler rule-application tasks. We're not good (in other words) at thinking logically innately, but that's why we've invented a tool to help us do it: formal logic.

[Agent 008]

They couldn't. The fundamental basis of the computer is the proof; if x = y, then such and such happens.

Formally reasoning what can and cannot be computed is a big branch of Computer Science.

http://en.wikipedia.org/wiki/Computability_theory_%28computer_science%29

[Lux Aeterna]

Language is linear, our minds are not.

Devise a method of transmitting simultaneous streams of non-linear information between humans. Until then, we will have to stick with logic...

[Ambient]

Language is linear, our minds are not.

Devise a method of transmitting simultaneous streams of non-linear information between humans. Until then, we will have to stick with logic...

Put exctacy in the water systems :p

At least thats what my war&peace class thought would be a decent means of achieving negative peace :p

Maybe a universal auxillery language would do the job lux?

[chompchompchomsky]

You're absolutely right that our minds don't work like that--philosopher/cognitive scientist Andy Clark sums up our cognitive abilities as being "good at frisbee, bad at logic." That is, we're quite good (better than anything we've seen so far) at identifying all kinds of patterns and predicting how those patterns will proceed, but we're bad at careful, formal, syntactic reasoning. That's why we use logic as a system though--we're pretty good at following simple rules, and the structure of logic lets us transform complex cognitive tasks (like reasoning about the consequences of a large number of related propositions) into a set of simpler rule-application tasks. We're not good (in other words) at thinking logically innately, but that's why we've invented a tool to help us do it: formal logic.

^This. I think that's the best suggestion yet.

[Rust]

Reality Apologist pretty much covered it. We are biological beings, and our mind is the result of thousands of years of evolutionary forces which have favored cognitive tools geared towards success in their environment. Thus, we use things like heuristics (http://en.wikipedia.org/wiki/Heuristic)which are not logical (they pretty much do not follow the rules of inference by definition) but are still nonetheless successful strategies more often than not.

We use formal logic because it aids us to reach conclusions that our innate heuristics (or other cognitive processes ) can't or have trouble reaching. This means we can develop things like mathematics, which in turn helps is develop a multitude of natural sciences, responsible for a myriad of different beneficial things.

[justanotherweirdo]

Devise a method of transmitting simultaneous streams of non-linear information between humans. Until then, we will have to stick with logic...

music?



shmusic?
http://images.starpulse.com/Photos/Previews/Walk-Hard-movie-31.jpg

[Ambient]

music?



shmusic?
http://images.starpulse.com/Photos/Previews/Walk-Hard-movie-31.jpg

Music and exctasy water?

[The_Wizard]

Music and exctasy water?

Music, exctasy water, and delicious food.

[AmazingBadass]

No real perceptual meaning besides justifying ourselves as humans. I'm going to search for a short paragraph. Hold.

People react, they have impact, and they own their own lifestyle. Sometimes this is preferable to the ones who are lethargic, without movement, and follow between others lives. I have been processed in such a way that unbearable meaning to compromise has nothing to give except reason. Reason is something that must be lived without if the ability to adapt must be met. I think once someone has given in to the vertigo of life, they cannot bear meaning to an actual existence of one another. The impacts of life’s past make the future more prevalent, yet the present further obsolete in a sense of renowned ownership. I take pride in past mistakes of my current world, as many people ought to. I think my Love is injudicious for the way I’ve allowed it to take control over circumstances indifferent to my positional prejudice. The pathetic fall for the already taken and the strong fall for the taken that cannot be attained except for through forgotten wisdom of a selfless mind. Most are such, because that’s how we were built as a race, as a population of a fake individualist society.

[chompchompchomsky]

Your grasp of grammar, syntax and vocabulary are breathtaking. Absolutely breathtaking.

[The Jitterskull]

Logic is just fun. It breaks down when you start getting high up but it makes things here so easy at times...


<3 logic

[Mr.Happy]

That's not to say, though, that we won't be able to simulate the activity of the brain with a powerful digital computer. We might get something that is behaviorally indistinguishable from consciousness, even if the mechanism behind that behavior is interestingly different. This is what the Blue Brain (http://en.wikipedia.org/wiki/Blue_Brain_Project) project is aiming at--by exactly modeling the behavior of individual neurons, the Blue Brain people hope to get something that behaves in every way as if it were conscious.

Say we design a computer, modelled exactly on the human neural structure. It has sensory inputs much the same as ours, including both the elementary-school five senses and the other, lesser-known ones that we nevertheless possess. It is able to learn and form (or activate) new connections in the same way as a human does, reacting to its external environment in accordance with how it has been taught to act. It is fully Turing-compliant, capable of responding indistinguishably from a human being in response to any scenario you care to throw at it; its responses are emergent, not pre-programmed, although they are of course based on prior experience as ours are. It is capable of choosing what to output, or what not to output; in other words, it is capable of making deductions without necessarily communicating them with the outside world, a distinct set of internal and external processes. I think this is a reasonable goal for those working in the field of artificial intelligence.

Say such a machine is constructed, and it works perfectly. How exactly is that machine only a simulation of consciousness? What makes it something that behaves as if it were conscious in every way, rather than something which is actually fully conscious?

~

As for the subject of the thread, it's hardly surprising that the human brain, which has been shaped by a few billion years of evolution (of course, for most of that time it wasn't a human brain or anything remotely resembling it, but it has a genetic ancestry), isn't as inherently logical as something designed for the sole purpose of functioning logically. The only forces guiding evolution are reproductive success and survival success; the goals of blind, stumbling life are far different from the goals of a method designed to make correct deductions and inferences about everything. Rust said a rather similar thing, and was quite correct.

[Agent 008]

As for the subject of the thread, it's hardly surprising that the human brain, which has been shaped by a few billion years of evolution (of course, for most of that time it wasn't a human brain or anything remotely resembling it, but it has a genetic ancestry), isn't as inherently logical as something designed for the sole purpose of functioning logically. The only forces guiding evolution are reproductive success and survival success; the goals of blind, stumbling life are far different from the goals of a method designed to make correct deductions and inferences about everything. Rust said a rather similar thing, and was quite correct.

But luckily, software evolves much faster than hardware.

We have "installed" logic on our brains (some of us haven't), and it is working very smoothly on our hardware when invoked.

[Mr.Happy]

^ Indeed.

Since the demands of our society are moving much faster than the evolution of our brains (which are barely evolving now, if at all), since there's a limit to how effectively we'll be able to 'install' logic on our brains and since there will naturally be an ever-increasing demand for intelligence and logical thought amongst some strata of society (military and science mainly), I reckon it won't be more than 50 years before we're literally installing capacity-boosting hardware in our brains.

Imagine downloading the latest neural update and suddenly being able to think more quickly and clearly. Awesome.

[PirateJoe]

Moving to inhuman condition

[Reality Apologist]

Say we design a computer, modelled exactly on the human neural structure. It has sensory inputs much the same as ours, including both the elementary-school five senses and the other, lesser-known ones that we nevertheless possess. It is able to learn and form (or activate) new connections in the same way as a human does, reacting to its external environment in accordance with how it has been taught to act. It is fully Turing-compliant, capable of responding indistinguishably from a human being in response to any scenario you care to throw at it; its responses are emergent, not pre-programmed, although they are of course based on prior experience as ours are. It is capable of choosing what to output, or what not to output; in other words, it is capable of making deductions without necessarily communicating them with the outside world, a distinct set of internal and external processes. I think this is a reasonable goal for those working in the field of artificial intelligence.

Say such a machine is constructed, and it works perfectly. How exactly is that machine only a simulation of consciousness? What makes it something that behaves as if it were conscious in every way, rather than something which is actually fully conscious?


Good. There are, I think, a few distinctions that are worth making here that might inform the discussion a bit. First (and most importantly), I think it's important to distinguish between cognition and consciousness--that is, between the sorts of activities that any system engages in when it uses representations of the world in order to find food, make predictions, and generally get around and the sort of thing that systems like our brains do in order to instantiate those representations in a particular way. Let me say a little more about this.

Cognition--information processing--is widely (and, I think, rightly) acknowledged to be a set of functional attributes: that is, the act of exploiting representations of the world in order to better function in the local environment is one which can be realized in a very large number of ways. We do it when, for example, we use our "mental maps" of to navigate familiar cities; programmers make robots that can do it when they build something like Stanford's Stanley. (http://en.wikipedia.org/wiki/Stanley_(vehicle)) It would be unfairly chauvinistic of us to suppose that only cognition being implemented in our particular kind of wetware counts as "legitimate" cognition--that is (once again), cognition is a behavioral or functional set of abilities, and anything that behaves in the right sorts of ways counts as a cognitive system.

This makes intuitive sense. As you pointed out, if we design a computer program that's capable of directing an artificial body such that the body acts in all ways as if it were a thinking person, then it would be very strange indeed to say that it was a mere "cognitive simulation." As far as cognition is concerned, there's no room to make a distinction between simulation and duplication: just executing the right behaviors in the right sort of way (that is, through the use of representations) is enough to make a system a cognitive system.

We can contrast that, though, with the particular way that systems like our brains solve the cognition problem--that is, with the particular kinds of representation that our brains exploit. Whatever theory of representation we buy into (that is, however we end up cashing out what it means for one object to represent another one), we'll have to admit a distinction between a representation's content and the particular way that content is represented--between (in other words) a representation qua representation and a representation qua physical object (or physical pattern). This distinction is also an intuitive one: we can distinguish between a leather-bound book, a PDF, and a CD containing a voice recording even if all of them contain renditions of Moby Dick; we can make a distinction between facts about the content that they all share (Moby Dick) and facts about the way that content is represented in each token.

In virtue of differences between these tokens, there will be facts that are true of one representation of Moby Dick that won't be true of another. We can talk about the peak frequency of the representation on the CD, the file format of the PDF, and the page weight of the book--all of these are perfectly sensible ways to talk about the individual representations, but it seems clear that they're predicates that can't apply across the tokens: the book has no file type, the PDF has no peak frequency, and the CD has no page weight. Those are (once again) predicates that are idiosyncratic to different modes of representation, and deal with the particular physical system that's instantiating the content of Moby Dick, not that content itself. This all seems rather straightforward.

Now, what's the point? The brain is a representation engine (that's all it means to be a cognitive agent, remember), so we can ask what unique predicates can be ascribed to neural representations in virtue of their physical structure. That is, what sort of thing can we say about a representation of Moby Dick that's instantiated in the structure and activation of neuronal connections in a brain like mine (or yours)? One thing that immediately comes to mind is that neural representations have the property of being like something to some agent: the pattern of structure and activation of my neurons that represents the car parked across the street (for example) as being red has the property of feeling in some particular way to me. There's something that it's like for my brain to represent something as being red (or 15 feet away, or smooth, or similar in chemical composition to a good single-malt Scotch, &c.). That's what we mean by consciousness, and by conscious experience: the set of unique facts about the particular kind of representation that's instantiated in brains that is "like something"--that has a subjective quality.

Viewed in this way, the fact that some system is conscious and another isn't even if both systems are capable of the same cognitive tasks is no more mysterious than the fact that some system might have a file format and some other system might not even though both systems are representing Moby Dick. Qualitative facts about conscious experience are not facts about the content of representations--and thus not facts that will be duplicated purely through the creation of cognitively identical systems--but rather are facts about the particular way that content is represented in biological brains like ours. If we want to duplicate that, we'll need to figure out what it is that our brains are doing that instantiates those representations and duplicate that feature: that's a taller order than merely duplicating the cognitive behavior of the system, though--we can do that with other kinds of representation.

[Chris]

I always thought it was because we weren't theists.

[a giant pterodactyl]

Logic, in an Aristotelian sense, that structures things in a progressive gradual understanding moving in one direction. Algorithmic logic, too, is pervasive in our creations (Computers, and their derivatives). And yet our minds don't do that! Our minds use a function of conceptual blending, whereby we arrive at conclusions by blending together unrelated concepts to form new, meaningful things: But we don't make anything that can do that, and we don't communicate in that way? Why do we translate our cognitive form of reasoning into a completely different model? Where is the usefulness in that?

your logic does not compute. syntax error on line 4

[zuperxtreme]

^ Indeed.

Since the demands of our society are moving much faster than the evolution of our brains (which are barely evolving now, if at all), since there's a limit to how effectively we'll be able to 'install' logic on our brains and since there will naturally be an ever-increasing demand for intelligence and logical thought amongst some strata of society (military and science mainly), I reckon it won't be more than 50 years before we're literally installing capacity-boosting hardware in our brains.

Imagine downloading the latest neural update and suddenly being able to think more quickly and clearly. Awesome.

Does this count?

http://zoklet.net/bbs/showthread.php?t=46366