The Hard Problem of Gravity

[lupus_in_fabula]

Because without anyone to understand the computation, how can it mean anything in particular?

A computer runs a program to add two numbers. Can we in any sense say that it is actually adding two numbers in the absence of someone to give it that interpretation?

A boulder runs down a hill and ends up resting next to another boulder. A person observing this might interpret it as 1+1 = 2. Is computation taking place? He might be ignoring another boulder ten yards away. There might be a couple of other smaller rocks that he's ignoring. Are all the computations with different results taking place at the same time?

We can interpret almost anything as computation. Does that mean that something is going on independently of our interpretation?

Yet that which is referred to could continue unaffected, regardless if there is a particular understanding, description or labeling of it.

I would actually say many things are going on independently of our interpretation of them (read: formal understanding & labeling), that might also include underlying processes that makes interpreting them as such possible in the first place.

If I understand RD correctly, the last humans' understanding, meaning and interpretation is irrelevant in this case (they are simply particular kinds, human kinds). They are only happening because of certain computations in a particular complex structured system in the first place. They are also happening in a different system from that of the computer. Hence when the last human dies, it would simply mean that certain computation ends in one system but other kind of computation still continues in the other.

 

[westprog]

Illusions exist, but not as the things we perhaps thought they were. "A sensation that actually exists" may give too much credit.

As we retread out steps... I don't think it's possible for an illusion to be illusory. It has to be a real illusion. The word "illusion" implies something there to be deceived.

 

[westprog]

Yet that which is referred to could continue unaffected, regardless if there is a particular understanding, description or labeling of it.

I would actually say many things are going on independently of our interpretation of them (read: formal understanding & labeling), that might also include underlying processes that makes interpreting them as such possible in the first place.

If I understand RD correctly, the last humans' understanding, meaning and interpretation is irrelevant in this case (they are simply particular kinds, human kinds). They are only happening because of certain computations in a particular complex structured system in the first place. They are also happening in a different system from that of the computer. Hence when the last human dies, it would simply mean that certain computation ends in one system but other kind of computation still continues in the other.

It might be the case that there's a process going on when a computer carries out a computation, independent of human understanding - but it's a physical process, and it only has physical meaning. A payroll program, for example, wouldn't be a payroll program if there were no humans being paid. It would be a pattern of electrons. I don't see why such a pattern would have any more significance than any other way that electrons organise themselves. There's no detectable field generated from computation. No forces are exerted. The detectable effect is on the human being looking at the printout.

 

[PixyMisa]

A huge part of the problem that I see is that we bandy these words about -- experience, feeling -- as though we all know what they mean. Sure, we have a vague sense and know how we use them, but that vague sense is a problem since it gives us the 'feeling' that we have a handle on the situation when we don't. What drove people nuts about Socrates was his constant harping about 'what do you mean by x?' -- because it became obvious that they didn't really know.

That's the great thing about online forums - much less hemlock.

 

[cyborg]

There's no detectable field generated from computation. No forces are exerted. The detectable effect is on the human being looking at the printout.

The detectable effect being what?

I don't see why such a pattern would have any more significance than any other way that electrons organise themselves

It doesn't - meaning is synthetic.

Saying, "a payroll program is meaningless to the machine," is no surprise. It's as meaningless to a goathearder too.

Meaning is synthetic - it is never innate.

 

[westprog]

The detectable effect being what?

An alteration in behaviour. Which we could, if we wanted to, put down entirely to the workings of known physical laws.

In which case, the only detectable effect would be in our own experience - or "private behaviour", as Mercutio prefers.

It doesn't - meaning is synthetic.

Saying, "a payroll program is meaningless to the machine," is no surprise. It's as meaningless to a goathearder too.

Meaning is synthetic - it is never innate.

And it's physically meaningless, as far as I can tell. There's no such thing as a "meaningful computation" outside of someone to interpret it.

 

[PixyMisa]

Because without anyone to understand the computation, how can it mean anything in particular?

The thermostat heats, and the thermostat takes heat away.

A computer runs a program to add two numbers. Can we in any sense say that it is actually adding two numbers in the absence of someone to give it that interpretation?

Only if the answer is correct.

We can interpret almost anything as computation. Does that mean that something is going on independently of our interpretation?

Did you use a computer to post that question, or did you just interpret it into existence?

 

[rocketdodger]

That is something to which I cannot agree. No computation can take place without a key to interpret it. Such a key must always be in the possession of an observer, who must be, AFAWK, a human being.

That's also the flaw with the universe simulated with a row of stones idea. Only the person laying out the stones knows what the simulation means. The stones aren't in possession of the key, without which it is meaningless.

So you are saying the behavior of every system in the universe would be identical if there were no humans to observe the difference?

What is this, some kind of joke? You are joking, right?

 

[rocketdodger]

Did you use a computer to post that question, or did you just interpret it into existence?

That, right there, is all that needs to be said.

 

[rocketdodger]

It might be the case that there's a process going on when a computer carries out a computation, independent of human understanding - but it's a physical process, and it only has physical meaning. A payroll program, for example, wouldn't be a payroll program if there were no humans being paid. It would be a pattern of electrons. I don't see why such a pattern would have any more significance than any other way that electrons organise themselves. There's no detectable field generated from computation. No forces are exerted. The detectable effect is on the human being looking at the printout.

You don't see any difference between the behavior of a rock and the behavior of a bacterium?

Interesting...

 

[rocketdodger]

And it's physically meaningless, as far as I can tell. There's no such thing as a "meaningful computation" outside of someone to interpret it.

So what?

Do you think this somehow advances your argument?

 

[yy2bggggs]

An alternate view that still fits the evidence (although it does not fit with our cultural view) is that a certain amount of time without eating will elicit a number of behaviors, both public and private. The private behaviors, since one's parents (and others) cannot see them, are not contingently responded to as the public behaviors are. ...

And how does this work for every animal but a human? How does it work for non-social animals?

 

[cyborg]

An alteration in behaviour.

Right. Which applies just as well, if not more consistently, to the systems dealing with the payroll data in the first place.

And it's physically meaningless, as far as I can tell. There's no such thing as a "meaningful computation" outside of someone to interpret it.

There is no innate meaning to anything - good.

Meaning is just a word to describe the fact that something is emotionally significant to you.

"Someone" interpretting it is redundant. What it requires to be "meaningful" is that something "cares" about it and reacts based on it.

 

[Mercutio]

And how does this work for every animal but a human? How does it work for non-social animals?

They, of course, still have the benefit of the environment's selective effect on their genes (well, their ancestors' genes), and the selective effect of the environment on their behavior (reinforcement and punishment). Social animals would have imitative learning, stimulus enhancement, and teaching. Non-social animals would not.

Social animals display behavior we infer as the ability to recognize others (individually, and as members of their group or outsiders); non-social animals tend not to. Dennett's "Kinds of Minds" (and for the record, I do not agree with Dennett on everything; there are some important points on which we differ. This is just for the sake of anyone trying to suss my position based on Dennett: Don't.) is one analysis of the "minds" one could infer from the behavior of various different animals. (One story in "Kinds of Minds" looks at whether awareness of pain can be inferred; it notes instances of monkeys biting off another's testicles, with so little reaction that one could conclude that he must not have felt [much, at least] pain. Cephalopods are assumed to be able to feel pain (and have, at least in some countries, the same legal protection against torture that vertebrates have; as with veal, you cannot cook one alive no matter how much you want to, whereas you certainly can do so with lobster), whereas no other closely-related mollusks are assumed to be able to. We do all of this inferring, as we do with other people, through behavior.

Many animals can be taught to discriminate color, for instance. Monkeys can be taught to respond to both conscious and non-conscious stimuli (think blindsight) in elegant experiments reported by Weiskrantz, among others (in "Consciousness Lost and Found"). The trick is, it is impossible to say whether they already discriminate, and have to be taught simply what to do when they see a particular color, or whether we actually have to teach the color recognition in the first place. Either way, we have a learning curve. (and in adult animals, of course, there may have been other opportunities to learn color, unless learning history is strictly controlled. I am unfamiliar with this literature, so I can't say one way or another whether such studies have been done.)

So, pretty much the same as in humans, with the exception of the verbal behavior, it would appear.

 

[Mercutio]

As we retread out steps... I don't think it's possible for an illusion to be illusory. It has to be a real illusion. The word "illusion" implies something there to be deceived.

Ah, but it is assumed to do more than merely "exist". Its function is precisely why we are discussing it; if that is illusory, then it may as well be something else entirely.

 

[westprog]

Ah, but it is assumed to do more than merely "exist". Its function is precisely why we are discussing it; if that is illusory, then it may as well be something else entirely.

It possibly has no specific function. It remains difficult to see exactly what it does.

 

[Third Eye Open]

Watch this, and then tell me that bacteria is not conscious.

 

[yy2bggggs]

They, of course, still have the benefit of the environment's selective effect on their genes (well, their ancestors' genes), and the selective effect of the environment on their behavior (reinforcement and punishment). Social animals would have imitative learning, stimulus enhancement, and teaching. Non-social animals would not.

Both social and non-social animals must eat, and social animals evolved from non-social animals. That social animals have this ability does not mean they use it for everything--and it doesn't make sense that they would.

The word "learn" can technically apply in either case, but I think it's a misnomer. "Learn" and "teach" are very strongly related, and this implies a social behavior. A better word would be one that suggests "learning without a teacher"--something like "to discover".

Social animals display behavior we infer as the ability to recognize others (individually, and as members of their group or outsiders)

I wouldn't include "individually" (not sure, for example, that recognition of individuals is important in an ant colony).

The trick is, it is impossible to say whether they already discriminate, and have to be taught simply what to do when they see a particular color, or whether we actually have to teach the color recognition in the first place.

(ETA: D'oh! Can't believe I missed this! Underlined above demonstration which specifically shows you making the exact association I was talking about regarding teach/learn)

Oh nonsense! If presented with stimuli, where the only significant difference is color, an individual shows behavior that strongly favors particular colors, that individual can be shown to discriminate color. Unless you mean something entirely different from what I'm reading into it--it's not only possible to tell, it's quite easy.

(ETA: Not sure whether or not this counts [cognitive dissonance study in monkeys])

I'll allow you to tell me what you mean by "can learn to discriminate color" to see if we're simply mismatching.

Either way, we have a learning curve.

Or a discovery curve.

 

[Ichneumonwasp]

It possibly has no specific function. It remains difficult to see exactly what it does.

You could view it as a form of meta-attention, allowing more efficient use of directed attention and a playground on which behavioral tendencies are sifted.

 

[Mercutio]

Both social and non-social animals must eat, and social animals evolved from non-social animals. That social animals have this ability does not mean they use it for everything--and it doesn't make sense that they would.

Nor does it have to. Part of my point is that we use our consciousness words to cover a wide range of behaviors that may serve different purposes, have been learned differently, or may even be reflexive.

The word "learn" can technically apply in either case, but I think it's a misnomer. "Learn" and "teach" are very strongly related, and this implies a social behavior. A better word would be one that suggests "learning without a teacher"--something like "to discover".

As a behaviorist, I am perfectly satisfied with learning as "a relatively permanent change in behavior brought about by interaction with the environment". That's why I specifically used the term "teaching" in my first paragraph.

I wouldn't include "individually" (not sure, for example, that recognition of individuals is important in an ant colony).

I was not thinking of ants/termites when I wrote of social animals, so if that is what you were thinking of, my bad. Social (as opposed to solitary) cats, rodents, or deer (I may even have species wrong here--am going from memory) are the critters whose characteristics I was speaking of.

(ETA: D'oh! Can't believe I missed this! Underlined above demonstration which specifically shows you making the exact association I was talking about regarding teach/learn)

I was using my definition consistently, if that's what you mean, making a distinction between learning in general, and the subset of learning where humans are specifically setting up contingencies such that we can call it teaching.

Oh nonsense! If presented with stimuli, where the only significant difference is color, an individual shows behavior that strongly favors particular colors, that individual can be shown to discriminate color. Unless you mean something entirely different from what I'm reading into it--it's not only possible to tell, it's quite easy.

In such experiments (discrimination learning), the "strongly favoring" is operationalized by a particular operant behavior--the favoring is termed stimulus control. And yes, there is a learning curve.

Or a discovery curve.

How would you tell the difference? The standard definition of learning fits the observations here--there is a change in behavior--without the assumption of an ability for which no evidence is seen. Certainly you can claim it is a process of discovery--but the onus would be on you to demonstrate it independently of learning.


(btw, your monkey link does not work)