What is/are the problem/s with logical positivism?

[sillysmile]

I'm aware that the school of thought lost scientific favour, but I don't really understand what was wrong with it in order for it to have to be replaced.
Does one of the problems have something to do with making the assumption that if some pattern is consistently the same, that it will always stay the same?

Thank you :)

[owleye]

I'm aware that the school of thought lost scientific favour, but I don't really understand what was wrong with it in order for it to have to be replaced.
Does one of the problems have something to do with making the assumption that if some pattern is consistently the same, that it will always stay the same?

Thank you :)

I'm not in strong touch with the movement's life cycle though I've read some of Michael Friedman's "Revisiting Logical Positivism". First, I'm not sure Logical Positivism or Logical Empiricism has been replaced (though I suppose there's something to be said about the success of Quine-Duhem). Indeed, even within the scientific community, apart from its instrumentalist outlook, there's a sense in which inference to the best explanation seems to be at work in scientific enquiry. Consider how the process of science is described in the text of an AP Biology course I'm reading (Cambell/Reece 6th edition): skipping over the first two processes (Observing and Questioning), Hypothesizing becomes the key element around which science takes action, developing a logic (the so-called hypothetical/deductive -- "if ... then" logic), from which experiments are carried out based on the predictions of the hypotheses to see whether there is any merit to them -- the 'deductive' part. Much more can be said here but the gist of it is that there is in a sense of a logical sequence going on within the scientific method.

From a philosophical perspective, about which there are others on this board who will be able to say more and correct what I say, notably Lomax and xcthulhu, I understand that the movement reached its peak with A.J Ayers, in his doctrine of verificationism, from which it subsequently suffered because there's a logical flaw it introduced, respecting the inability to verify itself. I don't know whether it has been resolved, but it may have motivated others to move toward a more pragmatic/holistic approach to the connection between theory and observation. (I think the failure of Carnap's attempt to define the logic may also have contributed to its lack of support.)

Despite this, Friedman in his revisitation, took up the challenge of logical positivism by taking a look at the motivation behind these positivists. With the fall of Kant's a priori stance regarding the forms of intuition of space and time due to a number of factors about which there's a huge amount written already on this subject, Relativity Theory brought about a significant puzzle respecting our knowledge of space and time, as it might exist apart from what are now considered contingent qualities of consciousness, respecting the ability of humans to be able to observe (or measure) reality.

Among the thinkers seeking to solve this problem (my memory fails me and I don't have the book in front of me) Herman Weyl developed a mathematical approach which allowed us to draw inferences about it in a self-scaling way, by drawing on the ability to observe reality with some accuracy at a local level and extend it outward to a global understanding through instrumentation designed in accordance with that mathematics. A number of other approaches seeking to unravel how such knowledge is acquired, which would require me to do a little research. (Note that Friedman's critique centers around the problem that logical empiricists had because they wished to develop a direct connection between theory and reality by eliminating the 'observational' aspects of observation -- reacting in this way to the downfall of Kantian intuition.)

Since these early days (in and around the first part of the twentieth century), there's been a revolution of sorts (this is the way I see it anyway) that was introduced with the information-theoretic approaches to knowledge acquisition (see, for example, Fred Dretske). It is in this context that I believe logical positivism ought to be revisited, wherein the relationship between the theoretical and the observed ought to take shape. The idea here is that we don't observe phenomena so much as we gather information from the world, process it and subsequently generate a phenomenal world based on that information, one which more or less faithfully represents it. The fidelity represented is due to evolutionary forces having favored it, which of course means it now becomes a discipline within biology. Though there's work yet to be done, I think it favors an epistemology that is more down to earth, one which should be fruitful in finding the logic that must be there, because whatever it is that is the scientific method must be something that is in touch with reality in order for it have succeeded.

James

[wuliheron]

In a word, logical positivism is its own worst problem. A.J. Ayers, a major proponent of logical positivism, said "I suppose the most important [defect]...was that nearly all of it was false." It didn't just loose scientific favor, its not even popular among philosophers anymore because it has so many problems if it were a horse we'd take it out and shoot it to put it out of its misery.

Wittgenstein was the first big hint they had a serious problem. You could compare him to a modern Socrates who pointed out the contradictions in how people use words and language and the endless language games philosophers play. Nothing speaks quite as loudly as success and while the logical positivists promised a scientific and philosophical revolution the later contextualists following in Wittgenstein's footsteps actually produced one. Its too soon to say exactly where the chips will fall, but its nowhere near logical positivism.

Personally I can't wait. The neurologists, behaviorists, linguistic annalists, and others are rapidly making progress towards a real scientific theory of language that will finally put all the B.S. in science and philosophy to rest once and for all.

[sillysmile]

Thank you both :) I'll get back to this when I have a bit more time to absorb it and reply.

[lucaspa]

I'm aware that the school of thought lost scientific favour, but I don't really understand what was wrong with it in order for it to have to be replaced.
Does one of the problems have something to do with making the assumption that if some pattern is consistently the same, that it will always stay the same?

Thank you :)

That's part of the problem. What you described is the Problem of Induction. Induction is a form of logical thinking that says: if a1=b,a 2=b, a3=b, .... ax=b, then all a's = b. If the sun rose the day before yesterday, yesterday, and today, then the sun will rise tomorrow. It seems intuitively true. However, David Hume showed that "proving" something by saying it has always been that way in the past therefore it will always be that way in the future is circular logic. You are presuming the very thing you are trying the prove.

The heart of Positivism was the Verification Principle. This principle said that only statements that can be "verified" were "meaningful" and, therefore, true. One problem with this is that the Verification Principle, itself, cannot be verified! Oops.

The other problem was that "verification" depends on induction, and thus runs into the Problem of Induction I described above. Even simply statements like "the atomic number of gold is 79" cannot be verified because we cannot, and have not, looked at every atom of gold in the universe!

So positivism bit the dust. It survives among atheists because it offers the only "scientific" means of disproving, or at least dismissing, God. Which, quite frankly, is what the Logical Positivists had in mind. Since statements about God are not verifiable, according to Positivism, they are not meaningful and therefore, not true.

[lucaspa]

First, I'm not sure Logical Positivism or Logical Empiricism has been replaced (though I suppose there's something to be said about the success of Quine-Duhem). Indeed, even within the scientific community, apart from its instrumentalist outlook, there's a sense in which inference to the best explanation seems to be at work in scientific enquiry.

Logical Positivism was replaced by Popper's Falsificationism as a means of doing science. Falsificationism did not work as a means of telling science from not science, but does work quite well as how science is done.

LP was never "inference to the best explanation". That is more Whewell's Consilience of Inductions.

Consider how the process of science is described in the text of an AP Biology course I'm reading (Cambell/Reece 6th edition): skipping over the first two processes (Observing and Questioning), Hypothesizing becomes the key element around which science takes action, developing a logic (the so-called hypothetical/deductive -- "if ... then" logic), from which experiments are carried out based on the predictions of the hypotheses to see whether there is any merit to them -- the 'deductive' part. Much more can be said here but the gist of it is that there is in a sense of a logical sequence going on within the scientific method.

Notice the DEDUCTIVE part of "hypothetical/deductive". What you have described is Popper's view of science. Positivism depended on induction, not deduction.

Inductive logic stays around because it just seems to work so well. Despite Hume's conclusively showing that it shouldn't work and can't be relied upon as "proof". Every time we draw a line through several points on a graph, we are using induction. Because we have some points and they are on the line, we induce that all the points are on the line. And it seems to work.

Several years ago some authors tried to combine the deductive logic of the hypothetico-deductive method and induction. The basic idea was that scientists started with hypotheses. Once they failed to falsify the hypotheses and could put them together in a "model" (biochemical model) in a review paper, then any new data was inductive and could be used to strengthen the model that way. I have the paper in PDF format if anyone is interested.

[owleye]

There's a need to correct the authors of the quotes in your post as they are wrongly attributed, but as they seem to be addressed to something I wrote, I'll respond to them. I think that Popper has had more influence on scientists in consideration of how they view science, than has Kuhn, but in my mind, scientific advance occurs because its theories are considered better (leaving this term undefined for now), not because they haven't yet been falsified. (I'll forgo my reasons here for emphasizing 'comparison'.) I will concede, however, that a theory's not yet having been falsified does have the virtue of it be instrumentally valuable.

With respect to deduction, I think of it as emphasizing the agreement with theory. With respect to induction, I think of it as emphasizing the negation of it being falsified. As such, it is merely taking a different perspective on the empirical (contingent) aspect that science has to deal with.

James

[lucaspa]

I think that Popper has had more influence on scientists in consideration of how they view science, than has Kuhn, but in my mind, scientific advance occurs because its theories are considered better (leaving this term undefined for now), not because they haven't yet been falsified

The reason they are considered "better" is that all the competitors HAVE been falsified :). Instead, the theory has suvived attempts to falsify it. Repeatedly.

With respect to deduction, I think of it as emphasizing the agreement with theory. With respect to induction, I think of it as emphasizing the negation of it being falsified.

???

Deductive logic in symbolic form: If H, then C. Not C, then not H. In words it is "if the hypothesis is true, then the consequences are true. If the consequences are not true, then the hypothesis is not true."

Inductive logic is arguing from the specific to the general. As I posted, it's general form is:
a1 = b, a2 =b, a3 =b, .... an = b, therefore all a's = b. The classic example is: swan 1 is white, swan 2 is white, swan3 is white, therefore all swans are white.

I suppose your thinking that deduction is agreement with theory bears some relation to deduction, but you miss the falsification of not C.

What Popper noted was that, if you look only for evidence "for" ANY theory, you will find it. No matter how wrong the theory is, you can always find some evidence "for" it. So the evidence that really counts is the evidence AGAINST it. That's the falsification. This is how Popper actually put it:

"1. It is easy to obtain confirmations, or verifications, for nearly every theory -- if we look for confirmations.
2. Confirmations should count only if they are the result of risky predictions.
3. Every 'good' scientific theory is a prohibition: it forbids certain things to happen. The more a theory forbids, the better it is.
4. Every genuine test of a theory is an attempt to falsify it, or to refute it.
5. Confirming evidence should not count *except when it is the result of a genuine test of the theory:* and this means that it can be presented as a serious but unsuccessful attempt to falsify the theory." [emphasis Popper's]
Summary : The criterion of the scientific status of a theory is its falsifiability, or refutability, or testability, not its confirmation.
Also Popper wrote: "I thought that scientific theories were not the digest of observations, but that they were inventions -- conjectures boldly put forward for trial, to be eliminated if they clashed with observations, with observations which were rarely accidental but as a rule undertaken with the definite intention of testing a theory by obtaining, if possible, a decisive refutation." Karl Popper, Conjectures and Refutations, 1963 p 38.

"the experimental results may square with the hypothesis, or they may be inconsistent with it. ... but no matter how often the hypothesis is confirmed -- no matter how many apples fall downward instead of upwards --the hypothesis embodying the Newtonian gravitational scheme cannot be said to have been *proved to be true*. Any hypothesis is still sub judice and may conceivably be supplanted by a different hypothesis later on. ...To my mind the great strength of Karl Popper's conception of the scientific process is that it is realistic -- it gives a pretty fair picture of what actually goes on in real-life laboratories." "The Threat and the Glory", by P.B. Medawar (Nobel Prize winner in medicine), HarperCollins, New York, 1990 (original publication 1959). pp 96-101.

[owleye]

The reason they are considered "better" is that all the competitors HAVE been falsified :). Instead, the theory has suvived attempts to falsify it. Repeatedly.

I left out my arguments intentionally as I didn't wish to get into a discussion with you on the topic of the scientific method, so I won't be pursuing your points, relevant as I think they are to them. With respect to logical positivism, my points relating to the idea that it might not be entirely dead were two-fold. One was based on it being revisited by Michael Friedman, the other on the basis that logical positivism/logical empiricism might yet make a comeback because there was a semblance of logic within the text of an AP biology course I'm currently reviewing.

I took your critique to be only toward the latter, as I don't think I made my point clear enough that my audience could figure out what was being said.

In any case, I took you as one who has studied the faults of logical positivism and its history and were painting a picture of "Hypothetical Deduction" as it was used in the text. You came to understand this concept was an inductive one, not deductive, which in my mind meant only that it shouldn't have been called by that name. My interpretation of it, which I thought was charitable, was that Popper's idea explained science's logic by turning deduction on its head, so to speak. Science's task is not to deduce anything, rather to put some hypothesis to the test of falsification. While I grant that this is one of the main features of the scientific method (and is why theorists are not only required to put up testable hypotheses, but that the testing is to be performed by independent testers, and indeed, the results need to be verified and repeated independently). However, this doesn't account for the acceptance of a theory, which is what I'd been trying to get at. No matter, logical positivism's intention is to put science on firm ground by establishing its logic. That they have failed to accomplish this (on the basis that "verificationism" can be verified and other reasons), might not indicate that it is dead, but only awaiting another good idea.

With respect to deduction, I think of it as emphasizing the agreement with theory. With respect to induction, I think of it as emphasizing the negation of it being falsified.

???

Deductive logic in symbolic form: If H, then C. Not C, then not H. In words it is "if the hypothesis is true, then the consequences are true. If the consequences are not true, then the hypothesis is not true."

Inductive logic is arguing from the specific to the general. As I posted, it's general form is:
a1 = b, a2 =b, a3 =b, .... an = b, therefore all a's = b. The classic example is: swan 1 is white, swan 2 is white, swan3 is white, therefore all swans are white.

You don't have to lecture me on deductive and inductive logic. I'm sufficiently familiar with them that I understand their weakness and strengths. Science obviously makes use of both forms of argument in their work. However, as far as I'm aware, logical positivism, despite the prowess in logic of these positivists (I'm thinking here of Carnap), the success of science cannot be explained in accordance with any logic they were able to come up with. I make use of 'emphasis' because this is how I understand the role of logic in justification of whatever knowledge claim or position one happens to have about the world. I'm not thinking of it as a mathematical or logical enquiry.

Getting back to the topic, I'm no longer sure you were even addressing the topic. I've granted the significance of Popper on instrumentalist grounds. However, I'm less sure of his role in altering how science conceives of itself. I'll further grant his role as a logician, respecting how science no longer is able to prove its theories in accordance with the two sorts of logic you've mentioned. But, as I see it, science already recognized this as a result of having so long been under the misapprehension that Newton was the last word in scientific theory, especially in consideration of how Einstein's theory supplanted it. Such a thing will tend to awaken them to the possibility of their being wrong. It was Popper who came along and firmed all that up for them. It's no wonder that scientists should fall in line, despite that it takes away one of their main arguments for the authority of science. However, with the "Structure of the Scientific Method", Kuhn demonstrated how advances in science (Feyerabend would object here) actually occurred, taking into consideration the psychology and sociology of science. I can appreciate that scientists themselves don't wish to be put under a microscope, so to speak, but his observations can't be so easily dismissed. In any case, as you can tell, I tend to favor the method of science along the lines of Kuhn over Popper, though of course, they had different objectives. Kuhn's analysis is more descriptive, and less epistemic, whereas Popper's is the reverse. Despite this, I think Kuhn's point deserves respect when considering its constructive aspects. (I suppose I've gone off the track that I was trying to avoid at the start. If the topic itself was the scientific method, I would go into this further. For now, though, I'll leave it at that.)

James

[Lomax]

Hi Lucaspa,

The reason they are considered "better" is that all the competitors HAVE been falsified :). Instead, the theory has suvived attempts to falsify it. Repeatedly.

I think Owleye's right on this one. It's not strictly accurate to say that falsificationism replaced verificationism, since they were first published within months of each other, and Popper even conceded to Ayer that the falsification principle itself is best justified on weak-verificationist grounds.

Deductive logic in symbolic form: If H, then C. Not C, then not H. In words it is "if the hypothesis is true, then the consequences are true. If the consequences are not true, then the hypothesis is not true."

Owleye mentioned the Duhem-Quine thesis a while back, and this is why Kuhn has had more practical influence than Popper: any consequences can be made to fit with any theory, by means of introducing or eliminating auxiliary theories. The problem becomes even worse for those of us who buy into Two Dogmas. Falsificationism and verificationism are still both very important tools, but neither of them can do the job without some kind of pragmatist adhesive, I think.

It's good to see some discussion of Popper on the forum though; we don't get enough of it, and I'm as guilty as anybody else.

Lomax