David Harriman's Book

[Robert Campbell]

Here is how Harriman sets the stage for his opus:

As our knowledge of the physical world has advanced, our understanding of knowledge itself has lagged behind. I witnessed this gap between physics and epistemology during my college years at the University of California, Berkeley. In my physics lab course, I learned how to determine the atomic structure of crystals by means of x-ray diffraction and how to identify subatomic particles by analyzing bubble-chamber photography. In my philosophy of science course, on the other hand, I was taught by a world-renowned professor (Paul Feyerabend) that there is no such thing as scientific method and that physicists have no better claim to knowledge than do voodoo priests. (Chapter 1, p. 5)

Robert Campbell

[Robert Campbell]

In the quotation from page 5, do you think Harriman is identifying the alternatives fairly?

There are lots of college physics courses where students learn how to use X-ray diffraction.

How many college philosophy of science courses are taught from the standpoint of the late Paul Feyerabend?

Robert Campbell

[Xray]

The book makes some overblown claims, and Harriman's objections to modern physics are enough to discredit him personally with serious and philosophically sophisticated scientists, but the book is well worth reading as food for thought.

On the Logical Leap thread, you wrote about Harriman's arguments:

He does say, following Rand's entity/attribute/relation ontology, that space is a relationship, and that relations cannot be bent, only entities can. It is a puerile argument based upon his lack of understanding of the difference in senses between space in the sense of geometers and the modern physicists' concept spacetime. Again, we run into the problem of translating phenomena into the context of a scale where conventionally human-scale notions fail to apply.

For those who, like Xray, don't understand how ludicrous Bohr's claim is, allow me to illustrate.

I go to the office of a private detective and tell him that I want him to locate a Blictri.

"Okay." he replies, "What is a Blictri?"

"A Blictri has no identity, no determinate characteristics. It is nothing in particular."

He asks, "So how will I know a Blictri if I find one?"

"You won't. Have you been listening to me? It has no identity.

"So are there any signs, tracks, or whatever that I can look for?"

"Nope. I don't know what a Blictri is -- no one does; there is no way to identify one -- so I cannot tell you what to look for."

The private detective, failing to appreciate my profound understanding of Blictris, throws me out of his office and tells me never to come back.

Ghs

This merely illustrates once again the problem of conventional human-scale notions not being applicable.

Ba'al has commented on this many times, for example here:

Wave and Particle are mathematical abstractions. Any extended body is not a particle. The Wave and Particle abstractions make good analogs in a mathematical description of the physical world. What is -really- Down Below is neither a wave nor a particle, but for certain situations the particle description fits well, for other situations the wave description fits well.

Edited September 13, 2010 by Xray

[George H. Smith]

In the quotation from page 5, do you think Harriman is identifying the alternatives fairly?

There are lots of college physics courses where students learn how to use X-ray diffraction.

How many college philosophy of science courses are taught from the late Paul Feyerabend's standpoint?

Robert Campbell

Feyerabend is regarded as something of a quack by many philosophers of science. His methodological anarchism is an extreme case, not something in the mainstream.

Harriman overstates the extent of scientific skepticism by philosophers of science. It is the business of philosophers to raise questions and to explore alternatives, but these should not be confused with conclusions or definitive statements.

At one point Harriman even quotes from and criticizes Morris Cohen and Ernest Nagel (from An Introduction to Logic and Scientific Method), calling them philosophers "who traffic in the arbitrary deny [sic] the existence of real knowledge" P. 72). This is nonsense Their book is anything but skeptical about knowledge and science. Cohen in particular, who also wrote the excellent work Reason and Nature: An Essay on the Meaning of Scientific Method, would have agreed with Harriman on a number of significant issues. It just so happens that Morris and Nagel disagreed with Harriman on some specific points. When I first scanned the endnotes of Harriman's book and saw the reference to Cohen and Nagel, I assumed he had used them to support one of his own points.

There is an unfortunate tendency among orthodox O'ists, including Harriman, to emphasize their disagreements with other philosophers rather than their points of agreement, even when the latter are more fundamental. There is also a tendency to overstate the significance of disagreements.

I have typically taken the opposite approach. Readers of ATCAG may recall that I favorably mention philosophers other than Rand in my discussion of the contextual nature of knowledge before I mention Rand at all. That was a deliberate decision on my part.

Ghs

[Robert Campbell]

George,

I'm a member of the Philosophy of Science Association, albeit not a very active one.

I've neither encountered defenders of Feyerabend's views at the PSA events I've been to, nor seen defenses of his views in the pages of Philosophy of Science.

But the news out of Irvine, CA, is that philosophers of science are, almost to the last man and woman, a wretched pack of subjectivists and nihilists.

Is it possible that the Orthos are not merely exaggerating the extent and signficance of their disagreements with other philosophers?

Could they be trying to discourage their students from finding out what other philosophers of science have actually said?

Robert Campbell

[George H. Smith]

For those who, like Xray, don't understand how ludicrous Bohr's claim is, allow me to illustrate.

I go to the office of a private detective and tell him that I want him to locate a Blictri.

"Okay." he replies, "What is a Blictri?"

"A Blictri has no identity, no determinate characteristics. It is nothing in particular."

He asks, "So how will I know a Blictri if I find one?"

"You won't. Have you been listening to me? It has no identity.

"So are there any signs, tracks, or whatever that I can look for?"

"Nope. I don't know what a Blictri is -- no one does; there is no way to identify one -- so I cannot tell you what to look for."

The private detective, failing to appreciate my profound understanding of Blictris, throws me out of his office and tells me never to come back.

Ghs

This merely illustrates once again the problem of conventional human-scale notions not being applicable.

Your comment merely illustrates your astonishing inability to grasp even the most fundamental of philosophical issues. Your excuse-for-all-occasions would apply to a physicist who claimed that the subatomic world is populated by tiny reindeer and a Santa who rides about in a sleigh. Your critical faculties, such as they are, are reduced to a quivering blob whenever you encounter some nonsensical statement -- literally nonsensical, in Bohr's case -- by some physicist. Bohr's farts had as much meaning as his statement about identity. I'm sure you would have been impressed by them as well.

Ghs

[sjw]

Could they be trying to discourage their students from finding out what other philosophers of science have actually said?

Robert Campbell

Conspiracy theorist! Conspiracy theorist!

They certainly have the motive. The more acolytes, the more money. I actually had an Objectivist offer that to me as an explanation for why they were doing some absurd thing. He said you can't expect them to do something that would shut off their revenue stream. So the theory *he*, as an ARI-supporter, offered to me for an explanation of folly was that the people running ARI are just greedy charlatans not interested in truth but only interested in maximizing revenue.

Shayne

[Robert Campbell]

Here's Harriman on the contextual certainty of first-level generalizations. Bolding has been applied by me.

Since first-level generalizations are the basis of all higher-level generalizations, they cannot be threatened or undermined by the latter. Like sense perception itself, they are impregnable to overthrow by any future discovery. This does not mean that the first-level generalizer is omniscient. On the contrary, it means that knowledge is contextual, and, therefore, that—on any level of generalization, from first to last—certainty does not require omniscience.

Since generalizations are always reached in and through a specific cognitive context, their proper statement necessarily includes that context. To quote from Objectivism: The Philosophy of Ayn Rand:

**Harriman quotes two paragraphs from Peikoff's Chapter 5, pp. 171-172**

A child learns, for example, that pushing a ball makes it roll. Later he discovers that this does not happen if the ball reaches a certain weight, or if it is glued to the floor, or if it is made of iron and sitting on top of a strong magnet. None of this overthrows the initial first-level generalization. On the contrary, the latter is necessary for anyone to be able to consider subsequent qualifications. One cannot reach or validate "Pushing moves a ball only under X conditions" until one has first grasped the elementary fact that "pushing moves a ball."

Similarly, Newton's laws are not contradicted by Einstein's discovery of relativity theory (see Chapter 4). On the contrary, Newton's science remains absolute within Newton's context, and that science alone is what makes possible the later expansion of this context, when men discover operative factors unknown in Newton's day.

The knowledge possessed by a rational inducer is always limited, but it is nevertheless real. Because it is limited, it is open to future qualifications. Because it is real, however, the qualifications have no negative significance; they are purely positive, a step forward in the cognitive enterprise, not a step backward crushing their own roots. (Chapter 1, pp. 19-20)

[sjw]

Since first-level generalizations are the basis of all higher-level generalizations, they cannot be threatened or undermined by the latter. Like sense perception itself, they are impregnable to overthrow by any future discovery. This does not mean that the first-level generalizer is omniscient. On the contrary, it means that knowledge is contextual, and, therefore, that—on any level of generalization, from first to last—certainty does not require omniscience.

Speaking of conspiracy theories, why do you suppose Harriman wants you to be as a little child, accepting what appears at face value, not questioning earlier generalizations? Religions have the same epistemological standards. But they're more honest, they call it "faith."

Shayne

[George H. Smith]

Ba'al has commented on this many times, for example here:

Wave and Particle are mathematical abstractions. Any extended body is not a particle. The Wave and Particle abstractions make good analogs in a mathematical description of the physical world. What is -really- Down Below is neither a wave nor a particle, but for certain situations the particle description fits well, for other situations the wave description fits well.

This strikes me as very plausible, but even abstractions have identity. They have specific characteristics qua abstractions that make them this particular abstraction rather than some other.

As for what exists Down Below -- whatever it is, it will be what it is. It will be something in particular. It will have identity.

Ghs

[anthony]

Shayne,

I find those combined quotations by Harriman and Peikoff to be entirely reasonable. Is this rationalization to you?

I'm genuinely interested to know.

Tony

[George H. Smith]

Here's Harriman on the contextual certainty of first-level generalizations. Bolding has been applied by me.

Since first-level generalizations are the basis of all higher-level generalizations, they cannot be threatened or undermined by the latter. Like sense perception itself, they are impregnable to overthrow by any future discovery. This does not mean that the first-level generalizer is omniscient. On the contrary, it means that knowledge is contextual, and, therefore, that—on any level of generalization, from first to last—certainty does not require omniscience.

Since generalizations are always reached in and through a specific cognitive context, their proper statement necessarily includes that context. To quote from Objectivism: The Philosophy of Ayn Rand:

**Harriman quotes two paragraphs from Peikoff's Chapter 5, pp. 171-172**

A child learns, for example, that pushing a ball makes it roll. Later he discovers that this does not happen if the ball reaches a certain weight, or if it is glued to the floor, or if it is made of iron and sitting on top of a strong magnet. None of this overthrows the initial first-level generalization. On the contrary, the latter is necessary for anyone to be able to consider subsequent qualifications. One cannot reach or validate "Pushing moves a ball only under X conditions" until one has first grasped the elementary fact that "pushing moves a ball."

Similarly, Newton's laws are not contradicted by Einstein's discovery of relativity theory (see Chapter 4). On the contrary, Newton's science remains absolute within Newton's context, and that science alone is what makes possible the later expansion of this context, when men discover operative factors unknown in Newton's day.

The knowledge possessed by a rational inducer is always limited, but it is nevertheless real. Because it is limited, it is open to future qualifications. Because it is real, however, the qualifications have no negative significance; they are purely positive, a step forward in the cognitive enterprise, not a step backward crushing their own roots. (Chapter 1, pp. 19-20)

The highlighted portions of this passage are merely a restatement of Peikoff's version of Rand's contextualism that I criticized here.

[sjw]

Shayne,

I find those combined quotations by Harriman and Peikoff to be entirely reasonable. Is this rationalization to you?

I'm genuinely interested to know.

Tony

I think their theory is silly. I think a theory of knowledge based in what children do and think is silly. See Spot Run. See Spot Roll Ball. See Spot Form a Gen For Life.

It is not silly to talk about what children do in the context of child development, but if adults root their certainty of the world around what they learned as children then they're just as dogmatic as any brainwashed religious nut.

Shayne

[sjw]

The highlighted portions of this passage are merely a restatement of Peikoff's version of Rand's contextualism that I criticized here.

Wow George, that brilliantly sums up about 80% of my own view on this, with the other 20% taking exception to your characterization of knowledge concerning the Ptolemaic model. I think there is an important distinction you miss: the issue of whether or not the ancients were morally justified (were they honest) vs. epistemologically justified.

The issue is that epistemology is a kind of technology, and we must hold our most advanced conception of it at a given time in mind when judging history. We have to be able to say "if they had followed this *method*, they would have come to the right answer: they didn't know at that time what the right model was because they didn't have enough evidence to make a determination." We have to constantly revise our "epistemology technology" in accordance to errors, such that we can avoid similar errors in the future. Ideally we want an epistemology that would if properly executed have avoided all past scientific errors without actually having to know in advance the discovery that revealed the errors. Thus we have the highest-tech epistemology possible to us for future discoveries.

Shayne

Edited September 13, 2010 by sjw

[Ellen Stuttle]

Here's Harriman on the contextual certainty of first-level generalizations. Bolding has been applied by me.

[....]

Similarly, Newton's laws are not contradicted by Einstein's discovery of relativity theory (see Chapter 4). On the contrary, Newton's science remains absolute within Newton's context, and that science alone is what makes possible the later expansion of this context, when men discover operative factors unknown in Newton's day.

[....]

The highlighted portions of this passage are merely a restatement of Peikoff's version of Rand's contextualism that I criticized here.

Harriman says in his Preface that the material in Chapter 1 is taken "nearly verbatim" from Peikoff's lectures. But shouldn't the physicist of the team edit the physics for clarity? His not doing so bodes problems ahead.

"Newton's laws are not contradicted by Einstein's discovery of relativity theory," the text says. The "laws" sounds as if Newton's laws of motion are meant. However, the contradiction between Newton and general relativity pertains to the theoretic framework of the inverse-square Newtonian law of gravity. The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

So how is Harriman going to get Newton's science to come out contextually "absolute"?

Ellen

[George H. Smith]

Harriman says in his Preface that the material in Chapter 1 is taken "nearly verbatim" from Peikoff's lectures. But shouldn't the physicist of the team edit the physics for clarity? His not doing so bodes problems ahead.

"Newton's laws are not contradicted by Einstein's discovery of relativity theory," the text says. The "laws" sounds as if Newton's laws of motion are meant. However, the contradiction between Newton and general relativity pertains to the theoretic framework of the inverse-square Newtonian law of gravity. The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

So how is Harriman going to get Newton's science to come out contextually "absolute"?

Ellen

A post I wrote yesterday on another thread is relevant to this issue. See here . The following is based on my analysis in that post.

If I were a Peikovian, here is how I would respond to your post, in general terms.

Newton's laws were justified within the context of knowledge available at the time; they were therefore true. Relativity theory is similarly justified, so it is true as well. Truths do not conflict. There are degrees of truth.

So far as I can recall, Peikoff never expressly defends the notion of degrees of truth, but this notion is an important aspect of the traditional coherence theory of truth. And it has been pointed out previously that Peikoff, despite his technical adherence to the correspondence theory of truth, says a number of things (e.g., about complete integration) that are characteristic of the coherence theory.

An idea can be more or less integrated, and since there can be degrees of integration (i.e., coherence), there can also be degrees of truth, according to the coherence theory. But I should note that Peikoff's use of both the correspondence and coherence theories is not necessarily inconsistent. A philosopher can claim that correspondence is the meaning of "truth," while also maintaining that coherence is the primary criterion, or test, of truth.

Again, the above is my attempt to explain the Peikovian approach in a sympathetic manner. I am not saying that I agree with it.

Ghs

[George H. Smith]

The following passage by Brand Blanshard (The Nature of Thought, II, pp. 428-9) expresses an idea that also pops up in Peikoff's writings from time to time.

The test of any conclusion is whether or not it coheres with such system [of knowledge] as we have; but that system itself must be revised, and revised perpetually, under the correction of the immanent end [i.e., completely integrated knowledge] that is working through it.

What do we mean by coherence? We mean far more than consistency. Perfect coherence would mean the necessitation of each part [of knowledge] by each and all of the others.

Ghs

[sjw]

The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

In exactly what respect would you say they are not consistent? If Newtonian physics is a limiting case of relativity, then I would call them consistent. By "inconsistent" do you mean "not identical"? That would be silly.

Shayne

[BaalChatzaf]

The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

In exactly what respect would you say they are not consistent? If Newtonian physics is a limiting case of relativity, then I would call them consistent. By "inconsistent" do you mean "not identical"? That would be silly.

Shayne

Einstein's spacetime manifold is non-euclidean. Newtons space is euclidean. That is for starters.

In addition Newtons laws are globally Galilean invariant whereas Einstein's field equations are locally

Lorentz invariant.

The underlying theories differ markedly in mathematical structure.

Yet it is interesting to note that Newtonian gravity is a limiting case of Einsteinian gravity.

That is a good sign. Newton's theory and Einstein's theory both attempt to describe the same reality so one would expect that there would be a range in which both theories produces nearly the same predictions.

Ba'al Chatzaf

[sjw]

The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

In exactly what respect would you say they are not consistent? If Newtonian physics is a limiting case of relativity, then I would call them consistent. By "inconsistent" do you mean "not identical"? That would be silly.

Shayne

Einstein's spacetime manifold is non-euclidean. Newtons space is euclidean. That is for starters.

In addition Newtons laws are globally Galilean invariant whereas Einstein's field equations are locally

Lorentz invariant.

The underlying theories differ markedly in mathematical structure.

Yet it is interesting to note that Newtonian gravity is a limiting case of Einsteinian gravity.

That is a good sign. Newton's theory and Einstein's theory both attempt to describe the same reality so one would expect that there would be a range in which both theories produces nearly the same predictions.

Ba'al Chatzaf

So is your position that none of these differences constitute or derive from limiting cases? For example, suppose we limit the velocity toward zero, and the gravitational effects toward zero, would not the justifiable mathematical structure have to be Newtonian?

Shayne

[BaalChatzaf]

The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

In exactly what respect would you say they are not consistent? If Newtonian physics is a limiting case of relativity, then I would call them consistent. By "inconsistent" do you mean "not identical"? That would be silly.

Shayne

Einstein's spacetime manifold is non-euclidean. Newtons space is euclidean. That is for starters.

In addition Newtons laws are globally Galilean invariant whereas Einstein's field equations are locally

Lorentz invariant.

The underlying theories differ markedly in mathematical structure.

Yet it is interesting to note that Newtonian gravity is a limiting case of Einsteinian gravity.

That is a good sign. Newton's theory and Einstein's theory both attempt to describe the same reality so one would expect that there would be a range in which both theories produces nearly the same predictions.

Ba'al Chatzaf

So is your position that none of these differences constitute or derive from limiting cases? For example, suppose we limit the velocity toward zero, and the gravitational effects toward zero, would not the justifiable mathematical structure have to be Newtonian?

Shayne

The limiting case is the speed of light. If light traveled at infinite speed (it doesn't) then the world would be Newtonian. The Lorentz transform becomes the Galilean transform as the speed of light approaches infinity (i.e. becomes indefinitely large). If light traveled infinitely fast then we could synchronize clocks at any distance and simultaneity would become absolute instead of relative.

Ba'al Chatzaf

[Robert Campbell]

Newton's laws were justified within the context of knowledge available at the time; they were therefore true. Relativity theory is similarly justified, so it is true as well. Truths do not conflict. There are degrees of truth.

So far as I can recall, Peikoff never expressly defends the notion of degrees of truth, but this notion is an important aspect of the traditional coherence theory of truth. And it has been pointed out previously that Peikoff, despite his technical adherence to the correspondence theory of truth, says a number of things (e.g., about complete integration) that are characteristic of the coherence theory.

An idea can be more or less integrated, and since there can be degrees of integration (i.e., coherence), there can also be degrees of truth, according to the coherence theory. But I should note that Peikoff's use of both the correspondence and coherence theories is not necessarily inconsistent. A philosopher can claim that correspondence is the meaning of "truth," while also maintaining that coherence is the primary criterion, or test, of truth.

Again, the above is my attempt to explain the Peikovian approach in a sympathetic manner. I am not saying that I agree with it.

George,

I see your reconstruction of the Peikovian position as a reasonable one. It employs the resources available to the Absolute Idealists, who had some influence on Peikoff's thinking.

I think you can get degrees of integration out of Peikoff's exposition, along with the norm that, other things being equal, more integrated is better.

What Peikoff steadfastly refuses to let in is degrees of truth. I can't find them anywhere in his system.

It could be that degrees of truth would run contrary to Peikoff's Parmenidean allegiance, which sets off what is totally integrated against what is totally resistant to integration, what's part of Existence and what's Nothingness, what's True and what's Arbitrary. (I really wonder, at times, how Peikoff can handle mere falsehood in his system.)

I also think that his doctrine of the arbitrary shows how Peikoff's attachment to coherence does, in fact, get in the way of his attachment to the correspondence theory of truth. If "Gödel's proof of his theorem uses trick statements known as Gödel sentences," is context-ful and true when a metamathematician puts it forward, but context-less, productive of cognitive paralysis, arbitrary, and, "therefore," neither neither true nor false, when Peikoff puts it forward, the correspondence theory of truth is not setting the terms.

Robert Campbell

[sjw]

The limiting case is the speed of light. If light traveled at infinite speed (it doesn't) then the world would be Newtonian.

Then I object to your calling the theories "inconsistent." They are perfectly consistent.

Shayne

[Robert Campbell]

Newton's laws were justified within the context of knowledge available at the time; they were therefore true. Relativity theory is similarly justified, so it is true as well.

Here's a different problem with the Peikovian formulation.

How do you define the relevant context?

If "Newton's context" is taken mean the knowledge actually available to Issac Newton, prior to his death in 1727, why would you even be inclined to expect that Newton's laws would apply in circumstances about which Newton possessed no data but later physicists did?

If "Newton's context" means something broader than the knowledge actually available to Newton, where does it leave off and a different context of knowledge begin?

I suppose Peikoff or Harriman could say that "Newton's" context is, roughly, whatever well-educated physicists were in a position to know between 1665 and 1895.

But how could they make this particular demarcation, except in hindsight?

Robert Campbell

[Ellen Stuttle]

The predictions from Newton's inverse-square law come out right in limiting cases, but the theories aren't consistent with each other.

In exactly what respect would you say they are not consistent? If Newtonian physics is a limiting case of relativity, then I would call them consistent. By "inconsistent" do you mean "not identical"? That would be silly.

By"inconsistent" I mean "inconsistent." The theoretical frameworks from which the predictions are made aren't consistent with each other, as Ba'al has explained.

Here's another example. People holding the theory that the motion of heavenly bodies was circular could manage to mathematically describe the observed orbits of the planets by adding epicycle upon epicycle, but the theory isn't consistent with the Newtonian (or the Einsteinian) theory explaining planetary orbits.

The situation is like that in deductive logic where one might be able to arrive at the same deductive conclusion from syllogisms in which the starting premises are inconsistent with each other.

Ellen