Objective Knowledge: An Evolutionary Approach

One of the notions which pervade the whole book, "verisimilitude", had been defined by Popper in a seemingly unobjectionable way in the 1960s, and verisimilitude was thought by Popperians (including Popper) to be an accessible and legitimate aim of science, given that truth was seen as an important but very elusive target. Popper even tells us here (chapter 8) that with his novel definition he has rehabilitated the notion of "verisimilitude" just as Tarski had rehabilitated the notion of truth. This turned out to be a vain hope. Popper's definition of verisimilitude was shown to be completely wrong - in that two FALSE theories could not be compared with respect to their verisimilitude in Popper's sense -, and moreover, since the 1970s all the work which has been done on this topic seems to support the conclusion that verisimilitude is neither a clear nor a useful a notion. Yet Popper had maintained that "we cannot do without this idea". The consequences of this failure for Popper's account of scientific knowledge, and for this book in particular, should therefore be evident for everyone. Moreover, the negative results concerning verisimilitude were discovered after the first edition of this book had been published. The make-shift amendments in the second edition are hardly enough to improve matters.

The conception of knowledge as a Darwinian process is a nice idea, but it is rather vague and also too emphatic and one sided: knowledge also has its "Lamarckian" aspects. The story about the amoeba and Einstein (Einstein is not ESSENTIALLY more intelligent than the ameba) is funny. The production of correct answers cannot, it seems, be reduced to the sheer overproduction of hypotheses and the elimination of incorrect ones. The process of HOW some hypotheses are designed from initial data is also important - a logic of discovery, that is. Popper is not interested in this, despite the title of his classic book on scientific method - the reason being...that any process of discovery is not DEDUCTIVELY VALID!

Another curious feature of the "objective knowledge" which Popper describes is that it resides in a platonic heaven of "statements in themselves": it is a knowledge "without a knowing subject" (sic), although, curiously enough, it is somehow dependent (if I understood this platonic myth correctly) on what we humans do.

The chapter on "The aim of Science" contains a point which was made by Popper in 1949. Newton's theory does not entail Kepler's third law nor Galileo's law of falling bodies: it is actually incompatible with them. The incompatibility with Galileo's law was perhaps more well known before Popper wrote this essay than the incompatibility with Kepler's law. But the lesson which Popper derives from this, namely, that inductivism is refuted, is certainly spurious.

The chapter on clouds (inderterministic systems) and clocks (deterministic systems) is suggestive in the poetic wording and the stories, but does not add much to the debate of determinism-indeterminism. Popper believes that all systems are clouds, although some more clocklike than others. Here is an argument: the determinist thesis implies that a deaf physicist would have been be able to write Mozart's compositions just by knowing Mozart's physical state at a certain time and predicting what he would write in the pentagram; but this is absurd. Therefore determinism is wrong.

The chapter on Evolution and the tree of knowledge is all wrong. Popper's views on the (un)scientific character of evolutionary theory were shown to be wrong by scientists and philosophers alike. This time, Popper says that the only thing Darwin did was to show that evolutionary explanations "can exist", that is, "are not logically impossible" (!), and that no Darwinist has ever provided evolutionary explanations of anything at all. Later Popper admitted that his views on Darwinism were sheer mistakes, but even so the later reformulations of his views were found to be also terribly misleading and confused. What is even more curious, Popper objects to the usual definition of fitness in terms of reproduction rates on the grounds that it does not take into account that such rates might be due not to fitness but to fecundity; but his amended statistical definition of comparative fitness (A is more fit than B if its survival rate is greater and its fecundity rate is less or equal) has all the vices of every attempt to DEFINE fitness in terms of survival rates: it renders evolutionary explanations circular (A survived because fit, and A is fit because it survived).

The "Logic" part of Popper's "A realist view of Logic, Physics and History" (ch. 8) is extremely odd. He defends classical logic on sheer PRAGMATIC grounds (its utility as a canon of critical procedure), but he does not answer the question of whether there is any CORRECT logic amongst the many logics, which is the WHOLE question of "realism" about logic.

The chapter on Tarski (ch.9) is also mainly incorrect. His discussion of the problem of truth bearers in note 1 is completely muddled and rash. For instance, he says that he employs "sentence" as a synonym of "interpreted sentence OR PROPOSITION" (!). The interpretation of Tarski's theory as a theory of correspondence with FACTS is entirely arbitrary. Tarski nowhere talks about facts, but Popper speaks EVERYWHERE about them, even of "supposed" facts, of "real" facts, of "the world of facts" and what not. He also says that "Tarski's theory" allows us to define REALITY as "that with which true sentences correspond". Reality would in turn be "the set of real facts". It is needless to say that these grotesque fancies are not to be found nor suggested in Tarski's careful and precise work on truth.

There is almost nothing to be learnt from this book, and much to become confused about.


Good overview of 20th century philosophy of science   December 7, 2002
Dr. Lee D. Carlson (Baltimore, Maryland USA)
32 out of 39 found this review helpful

In a recent article on the relation between natural philosophy and quantum chromodynamics (the physical theory of the strong nuclear interaction), Frank Wilcek, a well-recognized researcher in elementary particle physics, included the following entertaining passage:

A man walks into a bar, takes a seat on the next-to-last stool, and spends the evening chatting up the empty stool next to him, being charming and flirtatious, as if there were a beautiful women in that empty seat. The next night, same story. And the next night, same story again. Finally the bartender can't take it any more. She asks, "Why do you keep talking to that empty stool as if there were a beautiful woman in it?".

The man answers, "I am a philosopher. Hume taught us that it's logically possible that a beautiful woman will suddenly materialize on that stool, and no one has ever refuted him. If one does appear, then obviously I'll seem very clever indeed, and I'll have the inside track with her."

"That's ridiculous", says the bartender, who happens to be a physicist. "Plenty of very attractive women come to this bar all the time. You're reasonably presentable, and extremely articulate; if you applied your charm on one of them, you might succeed".

"I thought about trying that," he replies, "but I couldn't prove it would work."

I included this passage in this review not to ridicule the work of David Hume but to emphasize that his philosophy of science is in no way troubling. The author of this book though spent most of his professional life attempting to refute the views of Hume and then justify the practice of science "objectively". In the first few paragraphs of this book, the author sounds bitter about the lack of recognition for his work on "the problem of induction", which he felt Hume had shown to have devastating consequences on the "truth" of science. The search for an objective, rational "foundation" of science has occupied the time of this author and many others, who hold to the idea that scientific knowledge needs such a foundation and the Humean challenge must be answered. To those readers who agree with the author in this regard, this book would be of interest. To those who do not, this book could possibly be read as an exercise in mental gymnastics. There are some places in the book where issues are raised that are important in fields such as artificial intelligence, but as a whole the book is typical of 20th century philosophy of science: it holds as axiomatic that scientific knowledge needs an underlying foundation.

Since I personally do not believe the David Hume has to be answered at all, a review of the author's arguments against Hume would be misplaced. Having read Hume's works in detail, and having walked away from them puzzled as to why they are considered so "formidable" or "devastating", my interest in this book was purely subjective: that of gaining insight as to why many philosophers of science are so deeply troubled by Hume's philosophy and other science skeptics. Finishing the book still left my questions unanswered in this regard, and judging by a perusal of the literature on the philosophy of science, Humean skepticism is still considered the "thing to answer". Scientific truth is still held in doubt to a large degree, and debates on it in the social and political realm usually take place in the context of religion or why creationism should be taught in the public schools.

But science needs no foundation. The game of philosophy should now be what consequences science has for philosophy. What theories of truth, of ethics, of knowledge, are possible for philosophy because of science? If this book were rewritten to reflect this attitude, its content would be very different, possibly more elaborate in its views. The avenues that science opens up in ethics, epistemology, and ontology are rich in information theory, mathematics, logic, and many other areas. Scientific and technological advances are exploding at an unprecedented rate, and no Humean challenge or backlash can stop it.....thankfully.