Physics and Philosophy

I wrote, "Mr. Bob Mac is a good case in point. His whole orientation was towards invalidating certain aspects of Objectivism. He had made up his mind that the philosophy was wrong, and by God, nobody was going to convince him otherwise." Cal replied:

How do you know? We could as well say that you have made up your mind that Objectivism is right and that nobody is going to convince you otherwise. This is cheap psychologizing and in fact an ad hominem argument: you "know" that he has "made up his mind", suggesting that he has done so in advance, without really considering the arguments pro and con. You don't consider the possibility that he has considered all the arguments and has come to the conclusion that they're wrong, no he disagrees with you and therefore he must be biased.

Cal, go back and read the exchange between us, and you'll see what I'm talking about. He was not simply making the argument that you're making. He was being completely unresponsive and disingenuous. Don't just look at his conclusion; look at the manner in which he defended it - especially his dismissal of the Objectivist axioms as question begging.

I also wrote, "So, if I understand you, you're claiming that Joseph's argument begs the question - that the author of An Introduction to Logic, and a lecturer in philosophy at the University of Oxford committed an obvious petitio. Well, I'm sure that the Oxford University Press would be glad to know about that, so that they can issue a retraction and expunge this embarrassing philosophical faux pas from their now tarnished reputation." You replied,

This is a perfect example of an argument from authority: because he's written an introduction to logic and is lecturer in philosophy his argument must be right. Don't you think there are also many professors in philosophy who think that Rand's philosophy is bunk?

Of course, and I should not have written that, because it does sound like an argument from authority. I was simply being sarcastic, because I was so fed up with Bob's evasiveness. Of course, you don't base an argument on appeals to authority. I would be the last person ever to suggest that.

But remember also that Bob was arguing that the Objectivist axioms of existence, consciousness and identity are themselves question begging, which is absurd, as these axioms are the basis of logic and of any logical fallacies, such as question begging. To say that the axioms are question begging would divest the concept "question begging" of any meaning. It's a classic example of the fallacy of the stolen concept - of accepting and using a concept in the very act of denying the concept(s) on which it logically and genetically depends.

In fact Bob is right. If you look at Joseph's argument: If a thing is to have any determinate nature and character at all, there must be uniformity of action in different things of that character, or of the same thing on different like occasions. His supposition is that a thing has a determinate nature, and then he tells us that it under the same circumstances can only produce the same effect. Well, that is of course a classic example of begging the question: you suppose that its nature is deterministic and that supposedly "proves" that it does behave deterministically.

By "determinate" nature Joseph doesn't mean a deterministic nature; he means a specific nature. In other words, he means that if a thing is to possess a specific character or identity, then there must be uniformity of action in different things of that character, or of the same thing on different like occasions.

And the argument of the Law of Identity is of course also fallacious: A thing, to be at all, must be something, and can only be what it is. This does not imply that it must behave deterministically, however, that is an extra assumption that is not contained in the definition of the Law of Identity.

I see what your saying, Cal. I do see your argument, even though I don't agree with it, but if Bob was making that argument, it certainly wasn't obvious. He was arguing that to base the law of causality on the law of identity is question begging, viz.,

You assert that the law of identity demands, infers, or deductively requires causality. This is 100% identical to the begging the question fallacy.

Of course, if the law of causality isn't based on the law of identity, then it has no basis whatsoever and can no longer be considered a valid principle of scientific induction. Don't assume that because Bob happens to agree with you that he is doing so for the same reasons - that he is making the same argument; he isn't. In any case, I would say that causal necessity does imply deterministic behavior, and that such behavior is therefore not an assumption but an implication of the law of identity. You continue,

This is a big flaw in the Objectivist reasoning and no argument from authority or argument ad hominem can change that.

I was not making an argument from authority; I was not saying that Bob was wrong, simply because he disagreed with Joseph; if I gave that impression, I certainly didn't intend to. Nor was I saying that Bob was wrong because he was biased, which is what an argument ad hominem would entail. But he was very definitely biased, which is something that even he admitted in one of his posts. If a man concedes his own anti-Objectivist bias - if he states it explicitly - and then argues like it for all the world, then I think it's rational for me to conclude that he is biased.

As for Objectivism's claiming that the law of identity implies determinism, don't forget that Objectivism advocates (non-deterministic) free will, so it doesn't make that claim. Granted, some Objectivists claim that in the sub-atomic realm, determinism must obtain, but I don't think that this is an official Objectivist position. It is my position, however, based on Joseph's argument. What Joseph is saying is that, under the same conditions, a thing could only act differently if it were different, because its action is a function of its identity - of what it is. If we deny this, we're saying that the nature or identity of a thing has nothing to do with its behavior, but in that case, its action would be completely unrestricted, because if its nature doesn't restrict it, then nothing else could. Remember, a thing is nothing but its attributes, including its action, so if its attributes aren't the same, then on what grounds can we say it's the same thing? How can it be the same thing under the same conditions, yet have different attributes? If it is truly the same thing, then it must have the same attributes, including the same action. That's the argument. Whatever is wrong with it (assuming that there is something wrong with it), I don't think it begs the question. We could express it formally as follows:

The same thing must possess the same attributes under the same conditions.
Action is an attribute.
Therefore, the same thing must perform the same action under the same conditions.

If you want to attack this argument, you will have to challenge either the truth of the premises or the validity of the conclusion. If you're interested in continuing this discussion, the ball is in your court. The conclusion is unquestionably valid given the premises, so if the conclusion is false, then one or both of the premises must be false. But I don't see how they could be. To say that the same thing need not possess the same attributes under the same conditions - that it need not take the same action - is to say that we have no basis for predictions of any kind. It is to say that just because something behaved a certain way in the past under a given set of conditions is no reason to believe that it will behave that way in the future.

- Bill

Post 101

Sunday, May 21, 2006 - 9:29pm

Excellent reasoning, Bill.

Ed

Post 102

Tuesday, May 23, 2006 - 12:28pm

Bill:

By "determinate" nature Joseph doesn't mean a deterministic nature; he means a specific nature. In other words, he means that if a thing is to possess a specific character or identity, then there must be uniformity of action in different things of that character, or of the same thing on different like occasions.

This formulation is of course rather vague: what exactly is "uniformity of action"? I think the only meaningful interpretation is that a thing under exactly similar circumstances (a condition that Joseph doesn't mention) will always act in exactly the same way - and that is the same as saying that it behaves deterministically. And if that is not what he means (which seems very unlikely to me), then he cannot conclude that it will behave deterministically. So no matter how you look at it: his conclusion is already implied in his assumption, which is merely a different way of saying the same thing. BTW, I was struck by the similarity of his argument and that of Rand's argument. Before I knew that he lived a century ago, I thought that he was heavily influenced by Rand. So I wonder if it is the other way around: did Rand perhaps base some of her ideas on his book? Or to paraphrase her own words: did she plagiarize his ideas? (Not that I think that it would be plagiarizing, but I think she shouldn't get away with that formulation either.)

Of course, if the law of causality isn't based on the law of identity, then it has no basis whatsoever and can no longer be considered a valid principle of scientific induction.

The law of causality is not based on the law of identity, it is based on empirical evidence. We observe that (at least in the macroscopic world) things under similar circumstances will always behave in the same way. We also observe that this is not the case for subatomic particles, but that doesn't mean that scientific induction no longer works. That the behavior of such particles contains a random element doesn't imply that their behavior is completely unpredictable, the success of QM is a direct refutation of that. While we can't exactly predict the behavior of a single particle, we can predict the statistical behavior of an ensemble of those particles, and those predictions have an astounding accuracy.

As for Objectivism's claiming that the law of identity implies determinism, don't forget that Objectivism advocates (non-deterministic) free will, so it doesn't make that claim. Granted, some Objectivists claim that in the sub-atomic realm, determinism must obtain, but I don't think that this is an official Objectivist position.

Well, I'm not sure what the official Objectivist position is, but from all that I've read from Rand and Peikoff I gather that their position is that the world is deterministic, except for living beings (or is it only humans? I'm not sure about that), which is of course a contradiction, as living beings are also part of the physical world. Saying that it is the character of living beings (or humans) that they are not deterministic is an arbitrary assertion. The usual argument is that this is based on introspection, which is of course also a form of empirical evidence and not a corollary of the law of identity, as this doesn't tell us anything about when things behave deterministically (inanimate matter) or not (living beings). For the record: my view is just the other way around: on the most fundamental level the world is not deterministic, but at higher levels the statistical behavior of the subatomic world results in a highly deterministic macroscopic world. That is also true for the human brain; although its working components are small according to our everyday standards, they're still big (and warm) enough to make the effects of the random behavior of the subatomic realm negligible, so I'm in fact a hardcore compatibilist.

If we deny this, we're saying that the nature or identity of a thing has nothing to do with its behavior, but in that case, its action would be completely unrestricted, because if its nature doesn't restrict it, then nothing else could.

That is a false dichotomy: the fact that the behavior of some things contains a random element does not imply that the identity of that thing has nothing to do with its behavior and that it's behavior is "completely unrestricted", nothing could be farther from the truth, so this is a straw man (or should I say straw particle?).

Post 103

Tuesday, May 23, 2006 - 4:24pm

I am beginning to think that perhaps QM does mean that certain areas of the subatomic world are not, in the traditional sense, deterministic. I also think that it also provides evidence that consciousness is non-deterministic from a scale - the higher the level of consciousness, the lower the level of determinism. What I find odd is why people who say the universe is non-deterministic to the extent of QM often then argue for determinism in consciousness? Why is that?

Post 104

Tuesday, May 23, 2006 - 5:39pm

Kurt, one answer is that it was all pre-determined by a super-consciousness (one that doesn't even need a material body), in order to watch us wriggle and writh in agonizing contradictory postulation -- only to then cling to our immaterial (abstract) math formulas as if they give the comfort of a warm blanket in the cold, dark night of the cosmos.

But that's just one answer. And, I caution you to take even that answer with a grain of salt (as I might have been pre-determined to write those words, in that way, at that time, yada-yada).

Ed <------------ [thinks he's self-conscious]

Post 105

Tuesday, May 23, 2006 - 10:57pm

I wrote, "By 'determinate' nature Joseph doesn't mean a deterministic nature; he means a specific nature. In other words, he means that if a thing is to possess a specific character or identity, then there must be uniformity of action in different things of that character, or of the same thing on different like occasions." Cal replied,

This formulation is of course rather vague: what exactly is "uniformity of action"? I think the only meaningful interpretation is that a thing under exactly similar circumstances (a condition that Joseph doesn't mention)...
In so many words, he does.
...will always act in exactly the same way - and that is the same as saying that it behaves deterministically. And if that is not what he means (which seems very unlikely to me), then he cannot conclude that it will behave deterministically. So no matter how you look at it: his conclusion is already implied in his assumption, which is merely a different way of saying the same thing.

What he's arguing is that the law of causality is implied by the law of identity, not that it is the law of identity. If I argue that the law of identity implies that the same thing must have the same attributes under the same conditions, I am drawing out the implications of the law of identity; I am not simply restating the law itself.

I wrote, "Of course, if the law of causality isn't based on the law of identity, then it has no basis whatsoever and can no longer be considered a valid principle of scientific induction." Cal replied,

The law of causality is not based on the law of identity, it is based on empirical evidence. We observe that (at least in the macroscopic world) things under similar circumstances will always behave in the same way.

But if you don't base causal inference on the law of identity, then how do you interpret the empirical evidence and apply it to similar situations in the future? More important, if you don't think that the law of causality is based on the law of identity, then how do you answer Hume? Hume argued that all we can say empirically is that things have behaved a certain way in the past, not that they will necessarily behave that way in the future. According to Hume, all we observe empirically are spatial conjunction and temporal continuity; we don't observe any necessary connection, any production of the effect from antecedent causes.

The problem of how to answer Hume - especially if one is an empiricist - is known in philosophy as "the problem of induction." Joseph's answer, in which he argues that the law of causation is implied by the law of identity, appears in a chapter entitled "Presuppositions of Induction." Induction is a process of reasoning from the particular to the general - of extending one's empirical observations beyond the range of the past and present into the future. In order to do that, one needs the law of identity - one needs the principle that like actions will occur under like conditions. Without that principle, one cannot generalize - one cannot reason inductively. But to reason inductively is to embrace the principle of causal necessity. If one cannot say that a certain kind of action must, by the nature of the acting entity, occur under the same conditions in the future, then one cannot engage in a process of inductive generalization.

We also observe that this is not the case for subatomic particles, but that doesn't mean that scientific induction no longer works. That the behavior of such particles contains a random element doesn't imply that their behavior is completely unpredictable, the success of QM is a direct refutation of that. While we can't exactly predict the behavior of a single particle, we can predict the statistical behavior of an ensemble of those particles, and those predictions have an astounding accuracy.

But, according to Hume, no matter how much predictive success you've had in the past, you still have no basis on which to infer that that success will continue in the future, even if you assume the same entities and the same conditions. All you can say empirically is what has happened; you cannot say what will happen, unless, you bring in another premise. That missing premise that needs to be supplied is the law of identity - the idea that if you have the same thing and the same conditions, then you must have the same action. Hume, of course, does not recognize the law of identity as an answer to his inductive skepticism. But Joseph does.

I wrote, "As for Objectivism's claiming that the law of identity implies determinism, don't forget that Objectivism advocates (non-deterministic) free will, so it doesn't make that claim. Granted, some Objectivists claim that in the sub-atomic realm, determinism must obtain, but I don't think that this is an official Objectivist position." Cal replied,

Well, I'm not sure what the official Objectivist position is, but from all that I've read from Rand and Peikoff I gather that their position is that the world is deterministic, except for living beings (or is it only humans? ...
It's only humans
...I'm not sure about that), which is of course a contradiction, as living beings are also part of the physical world. Saying that it is the character of living beings (or humans) that they are not deterministic is an arbitrary assertion. The usual argument is that this is based on introspection, which is of course also a form of empirical evidence and not a corollary of the law of identity, as this doesn't tell us anything about when things behave deterministically (inanimate matter) or not (living beings). For the record: my view is just the other way around: on the most fundamental level the world is not deterministic, but at higher levels the statistical behavior of the subatomic world results in a highly deterministic macroscopic world. That is also true for the human brain; although its working components are small according to our everyday standards, they're still big (and warm) enough to make the effects of the random behavior of the subatomic realm negligible, so I'm in fact a hardcore compatibilist.

If you're a hardcore compatibilist, how do you defend compatibilism on purely empirical grounds? We don't "observe" people acting the same way under the same conditions. That's a logical presupposition, isn't it? If your approach relies solely on empirical evidence without any reference to the law of identity, how do you justify compatibilism?

I continued, "If we deny this, we're saying that the nature or identity of a thing has nothing to do with its behavior, but in that case, its action would be completely unrestricted, because if its nature doesn't restrict it, then nothing else could." Cal replied,

That is a false dichotomy: the fact that the behavior of some things contains a random element does not imply that the identity of that thing has nothing to do with its behavior and that it's behavior is "completely unrestricted"...
Wait a minute! I thought you denied that the law of identity had anything to do with causality. Are you now saying that a thing's identity does bear a connection to its behavior?
...nothing could be farther from the truth, so this is a straw man (or should I say straw particle?).

Let me see if I understand you. Are you now saying that a thing's identity does have something to do with its behavior - that that identity requires that its behavior be restricted to a certain way of acting but does not require that it act the same way under the same conditions? If so, then I think you have a problem. Suppose that it appears to act in some new, radically unexpected way? Do you attribute this new behavior to its identity or do you attribute it to a hidden variable? And if you attribute it to a hidden variable, on what grounds do you make that attribution?

- Bill

Post 106

Wednesday, May 24, 2006 - 12:01am

Excellent reasoning, Bill.

Ed

Post 107

Wednesday, May 24, 2006 - 6:43am

Cal, I concur with your post #102. Thanks for taking the trouble to compose it.

Rand's principle that the law of cauality is the law of identity applied to action does not logically imply that in a given circumstance a given kind of thing could only do the same single thing on repeated trials. Rather, it implies that in a given circumstance a given kind of thing could not do exactly the same set of things on repeated trials that another kind of thing could do in those circumstances on repeated trials.

By the way, you and others here may find it interesting that in his 1908 book Identity and Reality, Emile Meyerson writes: "The principle of causality is none other than the principle of identity applied to the existence of objects in time". (p.43 in the 1930 English translation)

Meyerson writes also: "Science is not exclusively empirical; it is also the application to nature, in successive phases, of the principle of identity, the essence of our understanding. But from this principle we can draw no precise proposition by deduction; this is why there can be no pure science, contrary to what Kant supposed." (p. 400)
[The main Kant work being contradicted here is Metaphysical Foundations of Natural Science (1786).]

Post 108

Wednesday, May 24, 2006 - 9:22am

I should specify more fully a statement I made in the preceding post.

I wrote that the principle that the law of causality is the law of identity applied to action implies that "in a given circumstance a given kind of thing could not do exactly the same set of things on repeated trials that another kind of thing could do in those circumstances on repeated trials."

Taking that implication on its face, it is too restrictive. Consider a five-pound bag of sugar and a four-pound bag of sugar. These are two different kinds of things in respect of their mass. One bag belongs the class of things that have a mass of five pounds, whereas the other bag belongs to the class of things that have a mass of four pounds.

Imagine we scoot both bags, at the same time, off the terrace of my second-floor apartment. They will both do the same thing. They will both fall to the ground, having the same instantaneous velocity as each other all along their paths, and they will both hit the ground at the same time. [As the case is classical (and regular, i.e., not chaotic), each bag will do the same single thing on repeated trials, but that is no problem.]

So it is certainly not the case that two things belonging to different kinds must do sets of things at least partly different from each other on repeated trials in every given circumstance they can be in. Rather, they must do at least partly different sets of things in some given circumstance or another.

When the five-pound bag and the four-pound bag are placed on the scales of a beam balance, they do different things. That is enough.

Post 109

Wednesday, May 24, 2006 - 10:27am

Kurt:

What I find odd is why people who say the universe is non-deterministic to the extent of QM often then argue for determinism in consciousness? Why is that?

As I already indicated, for physical reasons: the functional building blocks of the brain are essential macroscopic structures, in which the role of quantum indeterminacy is negligible. It wouldn't explain "free will" anyway, as it would change a puppet steered by deterministic processes into a puppet steered by processes with random elements. But you give yourself already a hint in the right direction when you write:

I also think that it also provides evidence that consciousness is non-deterministic from a scale - the higher the level of consciousness, the lower the level of determinism.

A deterministic system is a system in which for a given state at time t1 there is only one possible state at a later time t2. But we should define carefully what we mean by a "state" of the system, as the answer to the question whether a system is deterministic or not will depend on that definition. A state is a description of that system, as complete as possible down to a given level of detail. Take for example a computer: a description at the lowest software level shows a deterministic system (which may be steered by random effects like random key strokes, but which will evolve in itself in a deterministic way). The same is generally true if we describe the computer down to the essential hardware level (like logic gates). Not quite, as glitches may occur, but if everything goes well these are eliminated before they reach higher levels. However, if we describe the computer in terms of individual atoms and electrons, the system is no longer deterministic. This is no contradiction: the behavior of a single electron may show random elements, the aggregate behavior of zillions of electrons is highly predictable, which makes it possible to create an essentially deterministic structure with building blocks that at the deepest level are non-deterministic. At the other end of the scale we see the opposite effect. Take for example a chess computer. We can describe a chess match as a succession of board positions. This is a typical example of a non-deterministic system: for a given position there is (apart from forced moves) in general more than one next position possible (let alone later positions). This will also be true for a good chess computer: for a given position it will not always play the same move; for one given "chess position" there will be several lower level states possible, each with a different outcome for the next positions, only we can't see them, which makes the next move of the computer unpredictable (it would be a rather boring game otherwise!). So we see that we can emulate a non-deterministic system on a deterministic machine.

Something similar happens in the brain: we can in principle describe the system of consciousness as a succession of states of conscious thoughts. Of course this is not so neat as with the description of a computer, the boundaries are rather fuzzy, what is exactly a thought, a feeling, an impulse, etc. Let's assume it's everything we can find by introspection. It's obvious that a system thus described is not deterministic: a description of such a state at time t1 is compatible with the descriptions of many different states at t2, and this is what we call "free will", we can't predict our own thoughts. But this doesn't imply that a description of the system at a lower level can't be deterministic, only this deterministic substrate remains hidden to us. Suppose I think at a certain moment: "I choose to focus!" Where does that thought come from? Is there a thought "I'm going to think 'I choose to focus" and a thought "I'm going to think 'I'm going to think...'" etc. ad infinitum? Of course not, at a certain moment such a thought just "happens". That is to say: it could in principle be explained by a description of lower level processes, which are inaccessible to us, however. Now the simplest assumption is that in a description at the level of the smallest biological structures such processes are deterministic. This is in accordance with our knowledge of physics and chemistry, and there is no contradiction with the indeterminism of conscious thinking at the intentional level. So there is no need for mysterious quantum effects creating "free will", or a lot of hand waving about "new physics" without any indication how that even in principle could create free will.

Post 110

Wednesday, May 24, 2006 - 10:28am

Bill:

But if you don't base causal inference on the law of identity, then how do you interpret the empirical evidence and apply it to similar situations in the future? More important, if you don't think that the law of causality is based on the law of identity, then how do you answer Hume? Hume argued that all we can say empirically is that things have behaved a certain way in the past, not that they will necessarily behave that way in the future. According to Hume, all we observe empirically are spatial conjunction and temporal continuity; we don't observe any necessary connection, any production of the effect from antecedent causes.

If that really is what Hume says, he's missing the point. What we do, especially in science, but also in our more intuitive "folk physics", is to build a model based on one or more hypotheses. Such a model allows us to find regularities in the world around us and to make predictions. We may arrive at these hypotheses by induction and Hume is right when he says that we never can be sure that our conclusion, in case the hypothesis is correct. We can only see how well it works in practice, but even if it seems to work well for a time, this is no guarantee that it always will work. In that case we'll have to discard the model built on that hypothesis and replace it by a model based on a new hypothesis etc. So the model of the flat earth, the Ptolemaic model, phlogiston theory etc. are examples of models that haven't stood the test of time, even if they were once more or less succesful models that seemed adequate to explain the observed phenomena. So there is in fact no problem of induction. Sure, you never can prove that a theory is correct, but so what? You can test it in practice, and if it's a good theory it will produce useful results - the proof of the pudding is in the eating.

If you're a hardcore compatibilist, how do you defend compatibilism on purely empirical grounds? We don't "observe" people acting the same way under the same conditions. That's a logical presupposition, isn't it? If your approach relies solely on empirical evidence without any reference to the law of identity, how do you justify compatibilism?

See my reply to Kurt's post.

Wait a minute! I thought you denied that the law of identity had anything to do with causality. Are you now saying that a thing's identity does bear a connection to its behavior?

Did I ever deny that a thing's identity bears a connection to its behavior? What I do deny is that its behavior can't have random elements, but I never said that its behavior is completely random, somehow you keep trying to saddle me with that ridiculous viewpoint, which is of course a straw man. An electron doesn't suddenly change into a positron or have spin 1 etc.

If so, then I think you have a problem. Suppose that it appears to act in some new, radically unexpected way? Do you attribute this new behavior to its identity or do you attribute it to a hidden variable? And if you attribute it to a hidden variable, on what grounds do you make that attribution?

If something like that happened I'd first check how solid the evidence is for that unexpected behavior (extraordinary claims demand extraordinary evidence!). But if it appeared that there really was such new behavior, I'd see if that could be explained with the current theory and if it turned out that that didn't work, I'd think how I could change the theory or think of a new theory based on different hypotheses, testing those hypotheses etc. (Of course I don't pretend that I personally could do all that really, that would be the task of the scientific community.) Such things have of course many times happened before.

Post 111

Wednesday, May 24, 2006 - 10:28am

Ed:

Excellent reasoning, Bill.

Jesus, Ed, are you Bill's claque or so? Nearly every reply by Bill to one of my posts is followed by your "excellent reasoning, Bill", "Brilliant post, Bill", "splendid refutation, Bill", "Great post, Bill" etc. ad nauseam. I think Bill is a big boy now, who doesn't need your continuous praise. Why don't you write just one laudatory post per year in which you list all of Bill's brilliant replies?

Post 112

Wednesday, May 24, 2006 - 12:39pm

LOL - ye should call yourself Splendens Jealousus....

Post 113

Wednesday, May 24, 2006 - 7:29pm

Excellent point, Cal.

Ed

Post 114

Wednesday, May 24, 2006 - 9:54pm