Customer Reviews

An Introduction to the Philosophy of Science

5 star:		(5)
4 star:		(1)
3 star:		(1)
2 star:		(0)
1 star:		(0)

 

 

This is perhaps the only book in which Carnap is almost invariably informal. It is a remarkably clear introduction to some important topics in the philosophy of science: the nature of scientific laws, probability, scientific measurement, the structure of space, causality and determinism, theoretical concepts and laws, and a last chapter called "beyond determinism". The point where Carnap gets a bit formal, sections 26-28, is boring and absolutely out of date; his approach to analyticity is certainly not the best available. As for the rest, Carnap's conceptions are generally reliable (although it should be observed that his "logical conception" programme for probability was a complete failure). The sections I enjoyed most are those which deal with the structure of space. Carnap is highly proficient there (Carnap's doctoral dissertation was called "Der Raum") and his philosophical observations are always lucid and precise.

Is this book still relevant, despite being a bit old? The answer is
an unqualified YES. Why is this book the best introduction to the
philosophy of science ever written? Because it is the result of a
collaberation between Rudolf Carnap (a philosophical giant) and Martin
Gardner--the celebrated columnest who gave us so many years of
"Mathematical games," during Scientific American's golden years.

Because it was co-written by a professional writer of popular
mathematics, it is probably the only philosophy of science book which
can be read and understood by the interested lay person. But because
it is based on a series of lecture notes from one of the worlds
all-time great philosophers of science, it doesn't "wimp out" on the
technical level. If you read it you will be brought to the forefront
of philosophy of science, at least as understood by the later logical
positivists.

In short, a remarkable collaberation by two men who were at the top of
their game. Thank God for Dover. For ten bucks you can buy a priceless
book.

This is probably the clearest account of the way Science works and why certain decisions are made. Within it is contained the clearest explanation of portions of General Relativity (as a concrete example of why Einstein presented the theory in the manner he did) I've ever read, as well as many other little interesting tidbits. Some parts of the book are a bit dry, but that is probably because this book is a rewriting of college lecture notes given by the author.

There are lots of books on philosophy and the sub category known as the philosophy of science. Science is the primary interest to me and, regards philosophy, I want to know how and why it is that we know what it is we think we know. I.e., epistemology is of interest. In this regards, while other books on philosophy have come and gone or remained on my bookshelf unreferenced, I have clung to Carnap's since purchasing over a decade ago. Praise be to Dover. Perhaps a future printing of these 300 pages might also include a Glossary addendum.

The context of my interest in philosophy arose long ago in terms of Evolution (the theory of common descent); a radio report in the mid-1980s regards Evolution that didn't make sense; and my own prior college education that was rich with science, mathematics, and engineering. While Carnap doesn't write much directly regards Evolution, he does give justification to the science that I know, an empirically based science that confirmed the validity of my prior formal and continuing informal education.

The radio report, the details of which are long forgotten, was that some Evolutionist now thought of Evolution as a "fact." I have since learned that the association of Evolution with fact was largely initiated in a paper by Julian Huxley in the early 1960s. That inappropriate idea has since caught on amongst the idealogs of Evolution. As rich as my higher education had been in physics, chemistry, organic chemistry, and associated laboratory courses, all I knew regards natural history is what I had absorbed in high school biology where Darwin's theory was briefly taught as a theory. Of course we all thought Evolution to be a true theory just as Newton's or some other. My education had led me to believe, and I continue to believe, that scientific theories, even true theories, are theories and that facts are facts. Facts, by definition, are considered to be true. Theories, on the other hand, are never considered to be facts even though they may or may not be so well confirmed so as to be considered to be "true." In science, "fact" is not a synonym for "a truth" or "true", etc. Thus, theories such as Newton's Theory of Motion, Einstein's Theory of Special Relativity, etc., may be well confirmed to the point of being considered "true." But even then it is important, in science, to maintain the integrity of the distinct notion of "fact." Again, in this view, no theory, no matter how well confirmed, is ever a fact. One might think of the analogy to a game of baseball. The game itself is never considered to be the baseball. Similarly, a theory or law of gravity is not equivalent to the observations of a falling apple.

Carnap's book is one of the few philosophy books that I have come across that even addresses, without a whole lot of ambiguity, the notions of facts, laws, and theories in science. Furthermore, while some might think that positivism is dead, if positivism is the label that is attached to Carnap and apparently it is, then my belief is that positivism is and ought to be alive and well. If, as some philosophers seem to argue, the symbolic language of positivism isn't appropriate to modern science then, certainly, update the symbolics while not throwing out the proverbial baby with the proverbial bath water. Mathematics has long been considered to be the universal language of the sciences. Surely, mathematical symbolism and logic is thus preferable to any philosophical symbolism and logic especially if the latter can't be easily transformed across disciplines. Surely, implementing the former ought not be problematical to the latter but rather seen as an enhancement to both science and philosophy.

According to Webster's dictionary the relevant definition of "positivism" is:

"3. a system of philosophy that is based solely on the positive data of sense experience; empiricism; especially, [also P-}, a system of philosophy organized by Auguste Comte, which is based solely on positive, observable, scientific facts and their relations to each other and to natural law; it rejects speculation on or search for ultimate origins."

In other words, "seeing is believing." But don't tell Evolutionists you agree with Ptolemy on this. They might have theoretical problems and their conscious may lead them to need to do a general recall on many of their books. Of course, we need to give adequate consideration to the idea of verification. It is important that our observations are correct observations. Verification of facts is one of the hallmarks that distinguishes rigorous science from, say, eyewitness testimony which can often be erroneous.

In the world of more modern science a caveat regards observations and observables concerns quantum realities in two areas: (1.) At the micro-level of quanta and quantum waves and (2.) Regards quantum entanglements as asscociated with the macro-world. (It wouldn't be a surprise if quantum entanglements were causal regards biological mimicry between, for example, some species of flowering orchids and their insect pollinators.)

One might balk at reading of limits placed on philosophy (and science) by positivism regards the "speculation on or search for ultimate origins." But I have come to see that such a limit does make sense in terms of natural science. Furthermore, such a thought seems to be very much in line with the thoughts of Kurt Godel ([1931], 1992) whose theorem proved that within any system of logic there are limits to what can be proved and disproved within the system. E.g., scientists now believe that the universe (i.e., the empty space between galaxies) is expanding faster than the speed of light. Therefore, can we possibly ever hope to know anything regards the "starry sky" we observe that lies beyond our own galaxy? Where precisely are those stars --- exactly, "now!"? Do those stars even continue to exist? Where does philosophy intersect with theology?

Well done, Carnap! Well done.

---------------------------

General Research Notes:

(1.) Traditionally academic science has been divided into two broad categories, "experimental / empirical science" and "theoretical science."' Perhaps a third category is also relevant, "data analysis." It would seem that many of the poor results from otherwise good experimental / empirical research falls into this later category. For example, experimental science might generate valid data regards the question "Is a particular vitamin or mineral supplement good, bad, or indifferent prescription for the general population?" Or, "Is drinking coffee good, bad, or indifferent to overall health?" The credibility of science becomes strained in answering such questions much less more complicated ones. The public often might come to the conclusion that "If you don't like the current answer, just wait a few months and research will conclude the opposite." Unfortunately, enormous research biases seem to come into play concerning issues of analysis, more so than in the primary narrowly focused research. Such questions, due to the quantity of variables involved, might not be answerable even with ideal research. Research biases only add to the difficulties. While various neutral organizations do as well as they can, often via newsletters, in presenting the relevant output from analysis by those doing the primary research, much more might be done in terms of eliminating any biases. It makes no sense to depend on research from those who have a vested interest in the end result. Even without biases involved, whether or not even AI computers can choose and answer the relevant questions is difficult to know.

(2.) Who is to judge whether or not any epistemology --- scientific, religious, (or even business management, political, etc.) --- is right or wrong.? William of Ockham has long been noted by scientists as having inspired what has become known as Ockham's / Occam's razor, alternatively known as the rule of parsimony, q.v. A question on a recent Jeopardy tv show with Alex Trebek noted that William of Ockham was also a significant religious philosopher. The ideas that Occam arrived at seem to be very much along the lines of the mathematical proof of Godel, above. Occam's more religiously oriented rule, however, is more general and relates, if I have interpreted him correctly, that religious belief also cannot be proved but rather ultimately is dependent upon some improvable assumption. I.e., belief is belief.

(3.) The Reluctant Messenger website, q.v., provides an interesting definition based on a synthesis from several major religions. Def.: "God is the indescribable, uncreated, self existent, eternal all knowing source of all reality and being." While there would not seem to be much that natural science can do in terms of exploring the concept, such a concept would appear to at the very least be compatible with the naturalistic thinking expressed by Godel, Occam, and Carnap above.

(4.) P.A. Cox in his THE ELEMENTS: THEIR ORIGIN, ABUNDANCE, AND DISTRIBUTION ([1989], 1994) reports that positivist philosopher Comte, above, believed it would never be possible to ever know the composition of the Sun or other celestial bodies. In 1835, Comte wrote:

"We understand the possibility of determining their shapes, their distance, their sizes and motions, whereas never by any means will we be able to study their chemical composition, mineralogic structure, and not at all the nature of organic beings living on their surface."

Well, apparently "never" is not such a very long time as researchers have indeed quite satisfactorily been able to discover --- thanks especially to the discovery of mass spectrometry by Aston in 1921 --- the elemental composition of even bodies far removed from our own solar system. Now, if we can just spot those little green (or red) folks on Mars who are endowed with enough green (or red) foliage that eating for the purposes of vitamin C ingestion and all other nutrients is entirely unnecessary. Obviously these Martians are making use of nanotechnology to slightly strengthen the strong nuclear force. This allows the Martians to manufacture helium from readily available hydrogen in sufficient quantities to cause themselves to be light enough (without being explosive) to move around. Even with the reduced Martian gravity --- about 1/3 that of Earth --- the increase in buoyancy is quite useful.

Carnapian Survivals and Computerized Successes

Positivism is dead. Most of the philosophy in this book is now obsolete, but it has relevance today following the passing of the brief but popular anti-analytical Kuhnian fad.

Carnap rejected the idea of a machine for creating theories, which reference unobservables. And in his Logical foundations of probability (1950) he concluded that there cannot be an "inductive machine", i.e. a computer system, into which the scientist can input all the relevant observation sentences, and then get an output consisting of a system of empirical laws that explain the observed phenomena. He only believed that given observation e and hypothesis h, there can be an inductive machine which will mechanically determine the logical probability or degree of confirmation of h on the basis of e.

It is regrettable that the computer age had not begun thirty years earlier, because Carnap's linguistic-analysis constructionalist approach and his idea of semantical systems could have found evident application in contemporary computational philosophy of science - of course with large and important modifications to accommodate both contemporary pragmatism and modern systems design.

There have been many computerized discovery systems, sometimes called "artificial intelligence systems", as found in Herbert Simon's book Scientific Discovery: Computational Explorations of the Creative Processes. I created a discovery system for social science, and found similarities between Carnap's ideas and the pragmatic concepts in my system design.

Perhaps the reader of this review interested in computational philosophy of science will permit me to share some personal experience in the linguistic-analysis approach. A summary of my modifications and similarities to Carnap's approach, which are discussed in my books titled Introduction to Metascience and History of Twentieth-Century Philosophy of Science, is as follows:

1. Unlike Carnap scientists never use the Russellian symbolic logic for the expression of their theories. The object-language theories constructed by my discovery system are expressed as mathematical equations of the type actually used in the relevant science.

2. The computer language constituting the discovery system is the metalanguage expressing a mechanized generative grammar in the program.

3. The semantical rules that describe the semantical interpretation of the object-language statements are sentences that are both analytic and synthetic like Quine's analytical hypotheses or discursive postulates. They might also be viewed as similar to Carnap's reduction sentences, which he says determine only "part" of the meaning of theoretical terms.

4. The state descriptions are the computer system input and output expressed in the object language, and they reveal the semantical changes produced by the discovery system.

5. The theory of information is similar to Yu Shreider's semantical metatheory, and the state descriptions are identified with Shreider's concept of a thesaurus. Thus the amount of information communicated depends on the degree of transformation between his initial thesaurus and the outputted theory that must transform his thesaurus for the user to understand the new theory, and the psychological resistance to a new theory is large if the amount of information communicated is large.

Had history been kinder to Carnap, I believe that notwithstanding his positivist pessimism about a theory-making "inductive machine", he would have contributed to computational philosophy of science, perhaps anticipating Herbert Simon's Stahl and Dalton systems.

Even more significantly computational philosophy of science might have taken the linguistic-analysis turn, which I prefer, instead of its now popular psychologistic turn, which I think is misconceived. I believe that today's computational philosophy of science would be better served were discovery systems construed as language-processing constructionalist systems producing new semantical state descriptions, and that the legacy of Carnap can contribute some needed perspective for twenty-first century philosophy of science.

Readers are invited to Google my book at my com web site philsci, which offers free downloads. Also see my ebook Philosophy of Science: An Introduction.

Thomas J. Hickey

The value of this book is that it contains a summary of the views of Carnap in his last years, but it is readable not only for specialists. The "introduction" really menas that it is not so technical. You may read this book if you are interested in special relativity and some philosophy, or if you are curious about the scientific method and like to think abot it.

A cultural cornerstone. Must be read by those who are interested in science and its history.