critical rationalism blog

Brian Gladish has picked up the importance of Barry Smith’s work on “fallibillistic apriorism” (conjectural knowledge) and the way that this has corrects the unhelpful form of dogmatic apriorism that some of the Austrians picked up from von Mises.

With Smith’s introduction of fallibilism into Mises’s system, some of the distance between it and Karl Popper’s concept of conjectural knowledge was reduced.  This reconciliation has been visible in a number of efforts that attempt to bring Mises’s approach into the methodological housing of Popper and other philosophers of science, notably Imre Lakatos.³  More on that in another post; but, at this moment we have another issue to address — Mises’s claim that economics, and its encompassing science, praxeology, are new sciences unconnected with previous knowledge.  This claim did not sit well with those who believe all knowledge to be connected and do not have an anthropocentric view of the universe.

Taking up Brian’s comment, it took me years to see the significance of Smith’s work, I must have scanned his paper in the volume on Menger that Bruce Caldwell edited because years ago I wrote a summary of Jack Birner’s paper in that volume.

On a tangent to this is the way that Talcott Parsons was practically on the same page as Popper and von Mises when he wrote “The Structure of Social Action” in 1937 but then he lost his way. This paper describes the strange trajectory of a very busy and ambitious scholar who started with Parsons and ended up with “deep culture theory” which incorporates just about every school of social thought wtih the exception of CR and Austrian economics.

The format issue is sorted, don’t ask me how – trial and error:)

http://www.amazon.com/dp/B00BX3ATBS#reader_B00BX3ATBS

From the  “jacket”.

On the topic of jackets, don’t you love the covers which have been produced by Barnes, Catmur and Friends? What a shame that they will never be seen on real books on shelves in the bookshops of the world.

Daniel Barnes, like others including Matt Dioguardi, has been  a long-serving and valued colleague in Popper studies. Thanks!

This guide, the first in the Popular Popper series, provides an introduction to “The Logic of Scientific Discovery”, a book which changed the direction of the philosophy of science in the 20th century. The guide proposes that Popper’s ideas are best understood as a number of “turns” which he introduced. These include the “hermeneutic” or “conjectural” turn, to acknowledge that even our best scientific theories may be false, and the “conventional” or “rules of the game” turn, to account for the social nature of science and allow for the revival of metaphysics within any scientific research program. It also lists the most common misunderstandings of Popper which have confused students of philosophy and diminished his standing in academic circles.

UPDATE: The format problem in LSD is fixed although in both books the links from the table of contents are not working. OK in a preview but not in the live version.

Actually The Logic of Scientific Discovery and The Poverty of Historicism have launched as  Amazon e books but The Logic has a fault in the format and I am hoping the Amazon support people can fix it, or tell me what to do.

But The Poverty is ok in that respect.

The idea is to provide a clear summary of the main points with a little commentary to clarify some of them.

There is a set of Appendices to provide extra information.

I. The Progress of Popper, a summary of life and works.

II. The Popperian “Turns”. The six ways he changed direction, some of which isolated him  from the mainstream of the profession.

III. The reception of Popper’s ideas by the positivists and The Legend, that he was really an eccentric positivist.

IV. The Top Ten standard errors in reading Popper.

V. A case study, the philosophy and methods of economics.

VI. Carl Menger’s problem, how a theory of conjectural knowledge would have helped.

The Viennese Socrates: Karl Popper and the Reconstruction of Progressive Politics.

This book examines Karl Popper’s attempt to develop a political theory that draws upon Socratic fallibilism and commitment to ethical autonomy while preserving progressive sociological insights and commitment to activism. Philip Benesch argues that Popper’s critique of Marxist theory is largely an endeavor to separate its progressive-activist core from its positivist and uncritical-rationalist entanglements. The author defends Popper against the charges of positivism and scientism leveled by the Frankfurt School, among others. Although he is in no sense an apologist for Popper’s commentary on the classical tradition of philosophy, Benesch contends that Popper’s philosophical contribution is of classical breadth and significance and that it continues and advances “the great Conversation” that is the substance of the classical tradition.

Philip Benesch is Associate Professor of Political Science at Lebanon Valley College, Pennsylvania. He has also taught at Bryn Mawr College, West Chester University, and at colleges in London. He received his PhD in political science from the University of Delaware and his MA from the London School of Economics.

In this 166 page volume Dr Carlos E. Garcia (2006) articulates a systematic analysis of Karl Popper’s philosophy of science.

Popper’s core catechism is “I may be wrong and you may be right and by an effort we may get closer to the truth”. This is not a slight expression. For one thing by stating firstly “I may be wrong” rather than “I may be right” he assiduously avoids the trap of infinite regress as no reasons are needed to admit error. It is a non-justificationist assertion. Although the assertion about which we may be wrong could be arrived at by any psychological means, Popper avoids psychologism, the effort he is referring to is an effort of logic. Criticism is applied to the conjecture in an attempt to reduce error rather than attempting to confirm. With luck and effort, we may get closer to reality. For Popper there is a reality, non-relative and impersonal, but the inescapable propensity for error prevents us from ever claiming too close a familiarity with same. The notion of truth is timeless, whereas appraisals of corroboration are always indexed to a point in time and a set of accepted test-statements.

Garcia limits his chapter headings to:
1. Introduction
2. Solution to the Problem of Induction
3. Falsifiability
4. Corroboration
5. Verisimilitude, plus a very helpful and concise
Appendix: List of Definitions

As is apparent from the headings, Garcia’s study deals mainly with these cornerstone topics. I recommend the book to both advanced readers and professional philosophy workers. This study is clearly laid out and integrated. Criticisms of Popper over the years have at times been somewhat psychological, criticising him for intemperance towards criticism. He was probably entitled to be testy towards foolish criticism. e.g. Kuhn on one hand absolves him from certain logical sins and then says he may as well have committed them. What defence would work against that? Garcia concludes that Kuhn’s criticism of falsifiability and falsification is inadequate. I recommend that the interested reader carefully test Garcia’s arguments, I think they are strong. Popper, like Schopenhauer, is a lucid writer with an antagonism towards loose argumentation – if one has the patience to follow his trains of thought one might be convinced his was a great mind indeed. Garcia does him justice and I for one have been introduced to fresh perspectives.

Garcia reminds us that Popper considers the distinction between logical probability and corroboration as one of the most interesting findings in the philosophy of knowledge and notes that the logic of probability cannot solve the problem of induction. For Popper, the logical probability of ‘x’ is the probability of ‘x’ relative to some evidence; that is to say, relative to a singular statement or to a finite conjunction of singular statements. Probability gives us information about the chances that an event will occur but it does not inform at all about the severity of the tests that a hypothesis has passed (or failed). Corroboration and degree of corroboration are not equivalent to confirmation and degree of confirmation, or probability, as per Carnap’s logical empiricism. The “probability of a hypothesis”, in the sense of the degree of its corroboration, does not satisfy the laws of the calculus of probability. A highly testable hypothesis is logically improbable.

In Popper’s view science is concerned with intersubjectively testable explanations, a subjective view of probability is problematic. Popper fears that the theory of the probability of hypotheses confuses psychological and logical questions. Is it a probability measure or a plausibility measure? Any unîversal hypothesis goes beyond the empirical evidence. It can be validly tested by seeking counter instances not by collecting supporting examples as this could go on to infinity.

The appendix is instructive for its list of definitions. In this list, Garcia includes the version that he thinks best supports a reading of Popper’s account of science:

Basic statement (also test-statement): a statement that can serve as premise in an empirical falsification.

Corroboration (degree of): the degree to which a hypothesis has stood up to tests.

Event: a set of occurrences of the same kind.

Empirical content (also informative content): The amount of empirical information conveyed by a statement or a theory. Its degree is determined by the ‘size’ of the class of potential falsifiers.

 Falsifiability (or testability): the logical relation between a theory and its class of potential falsifiers. Falsifiability is a criterion of the empirical character of a system of statements.

Falsification: the conclusive demonstration that a theory has clashed with a falsifier.

Falsity content (of x): the subclass of false consequences of x. It is not a Tarskian consequence class.

Occurrence: a fact described by a singular (basic) statement.

Logical content (of x): the class of (non-tautological) statements entailed by x, where x can be a statement or a theory. A Tarskian consequence class.

 Logical probability (of x): the probability of x relative to some evidence; that is to say, relative to a singular statement or to a finite conjunction of singular statements.

Logical strength: increases with content (or with increasing improbability).

Truth content (of x): the class of (non-tautological) true logical consequences of x. It is a subclass of the logical content.

Verisimilitude (of x): the degree of closeness to the truth of x. It can be measured as the difference of truth content minus falsity content.

The Two Fundamental Problems of the Theory of Knowledge (2009) was first published in German in 1979. It is a thick book comprising a collection of drafts and preliminary work from the years 1930 to 1933 for Karl Popper’s first published book Logik der Forschung (1934). The Logik der Forschung was not published in English until 1959 with the badly translated title Logic of Scientific Discovery, rather than Research which would have been less diverting from his message.

Popper seemed to have a habit of writing and then delaying publishing for decades. The delay in publication of the three volumes of the Postscript is the other notable example, written in the fifties and published in the eighties. It is unfortunate for generations of students that Kuhn, Lakatos, Feyerabend and others in the meantime interpreted Popper’s work so separated from his actual writings that at times they may as well have been talking about something else entirely. It is not criticism that is an issue but misattribution of Popper’s views before criticism. It reminds me of the time my Scottish grandfather was at a funeral and could not conjoin the content of the eulogy with the deceased subject, he leaned over and whispered “I think we are at the wrong funeral”.

I recommend the purchase of The Two Fundamental Problems even if one only wants to read the introduction written by Popper in 1978. The introduction is as good a synopsis some of his core themes as I have read, startling in clarity and brilliance.

Some aphoristic points:

1. Plato’s Apology of Socrates is perhaps the most beautiful philosophical work known. Socrates is a little wiser than others perhaps because he knows that he does not know.

2. For Kant, Newton’s theory is not empirically gleaned from phenomena through the senses but is derived from our understanding. Kant believes Newton’s theory is true, Popper adds that it is not necessarily so.

3. Up until the Einsteinian revolution, Newton’s theory was corroborated better than anyone could have dreamed of. For Kant Newton’s theory is justifiable science, and therefore certain knowledge. For Einstein knowledge about reality is uncertain.

4. Fallibilism destroys scientism.

5. Science is the searching for truth: not the possession of truth, but the quest for truth. The idea of truth as manifest is unfortunately widespread.

6. In Popper’s view there is only one theory of truth that is to be seriously entertained: the correspondence theory; namely the thesis that the truth of a statement consists in its correspondence with the facts. The statement, “A cat is sleeping here”, is true if and only if a cat is sleeping here, no matter what language “A cat is sleeping here” (the object language) is spoken in.

7. We must sharply distinguish between the question of whether a statement is decidable (whether we believe we can prove it true or false) and the question of its truth.

8. Universal theories are fundamentally hypothetical or conjectural because they are not decidably true. This does not mean however that they may not be true.

9. The criterion for demarcation (falsifiability) is non-empirical. It was not obtained by observing what scientists do or do not do whether by studying living scientists or by studying the history of science.

10. Theories such as the Einsteinian and Newtonian theories of gravitation have an infinite number of potential falsifiers.

11. I have strongly emphasised in The Logic of Scientific Discovery that there is almost certainly no such thing as indubitable (or final) falsification by observation.

12. The critical attitude is characterised by the fact that we try not to verify our theories but rather to falsify them. Naturally one should not dwell on errors that are easily repaired but, if possible, correct them before embarking on serious criticism.

13. It is not even possible to formulate a principle of induction that is moderately plausible.

14. The fundamental weakness of inductivism lies in its extremely popular but fundamentally erroneous tabula rasa theory of the human mind. We are active, creative, inventive, even if our inventions are controlled by natural selection. The stimulus-response scheme is replaced by the mutation-selection scheme (mutation= new action). Knowledge is not a passive expression of the “data” provided by the senses.

and last but not least

Problem of induction (Hume’s problem) : “Can we know more than we know?”

Problem of demarcation (Kant’s problem) : “When is a science not science?”

I think Popper’s answer to these problems is: we don’t know but we do guess and we can structure our guesses as criticisable and falsifiable.

In 1945 Karl Popper left Christchurch and moved to the London School of Economics where he became the Professor of  Logic and Scientific Method.

His main course was Introduction to Scientific Method and he delivered a series of fifteen lectures on this topic for a decade or so through the 1950s. Mark Notturno became the editor of Popper’s work and one of his tasks was to convert the transcripts of the lectures into a publishable form. The recordings included questions and answers, and the usual false starts and half sentences of the spoken  word, so the idea was to create an “ideal type” of each lecture, drawing on the best parts of the ten copies that were available for each lecture.

As described in the Introduction, the publication stalled but it is now possible to read the first three lectures on line.  Introduction with links to the lectures.

The first lecture on Values began with a welcome from Popper who warned the students.  “I am getting old…my English is deteriorating with age…I am getting more and more inclined to ramble…and I am not a good lecturer either”.

In fact he was a captivating lecturer, speaking without notes, inviting and responding to interjections, inserting asides about his projects and references to significant developments events in science at the time. That was all edited out unless it related specifically to the content of the lecture.

On the function of lectures he said “Lectures are sometimes enjoyable, sometimes boring, but always, in a certain sense, unimportant. The important thing is the work that you are doing yourself.”

On the real aim of a university education.  “I believe that someone is well-educated only if he realizes in great detail how little he knows. And I think that this is really very important. I think that a man who has the feeling that he knows a lot is somehow badly educated. Yes, one can know a lot…but the main point, at least with regard to pure knowledge, is  to recognize the many open problems that lurk in all the knowledge that we have achieved. Without that l would say that you are not really educated…And the more we know and the more our knowledge grows, the more modest we should become about all those things that we don’t know.”

Thanks to Mark Notturno!

Key Issues in The New Knowledge Management (2003), by Joe Firestone and Mark W. McElroy, is for me a welcome library addition. A strong point is the focus on the epistemological issues. The summary of various theoretical underpinnings is clearly laid out.

I have felt uncomfortable with the traditional knowledge pyramid which has a base of raw data, then information, knowledge and a capping of wisdom. The traditional model is superficially seductive. It assumes that pure data are converted into information and then semantically assimilated into a body of knowledge. The question asked by the authors is how can such data be primary, let alone pure? How can perception be primary? Without existing propensities or expectations, agents or their computers cannot perceive anything. An agent’s pre-existing information provides structure to the world of experience. Data are types of information. Without structure experience is not data. What is normally treated as information is in the authors’ view, “just information”, that is to say information with conceptual commitments plus interpretations.

Knowledge is a subset of information (not a superset) that has been evaluated without ever being proven. Knowledge is an outcome of knowledge production and integration processes. It is an object (thing) that is uncertain but testable. Wisdom is knowledge coupled with value judgments and actionable assessments, it has untested metaphysical qualities.

Thus rather than a model based on a pyramid, it seems to be epistemologically more appropriate to picture a Knowledge Life Cycle in which data, “just information” and knowledge are types of information. New data and knowledge are made through this Knowledge Life Cycle from pre-existing information. That is from “just information”, data, knowledge, and problems.

Karl Popper, to whom the authors are indebted, said that all life is problem solving. One might say that all knowledge management is problem solving. Corporations depend on validated information but this is not the same as saying they depend on true or certain information. The critical method is that of making all knowledge claims testable i.e. capable of being falsified by a non-empty universe of test statements. Those claims that have not been falsified are preferred to those that have failed testing, without the luxury of ever being content that the knowledge cycle has found a utopia of certain knowledge.

It is important in knowledge management to reject the notion that the function of knowledge systems is to be a bucket for pure data. The knowledge cycle exists to solve problems and the problems in turn structure the questions to be asked and the information model that is tentatively appropriate.

In the space of this review I have left out a lot of solid content. The book is an extremely valuable resource for its definitions sections alone, for instance it is pointed out that there is no consensus on the nature of knowledge. I agree heartily with their rejection of the venerable but circular “justified true belief” definition so beloved by empiricists who believe knowledge claims can be justified by, rather than tested against, facts.

The New Knowledge Management framework is based on Karl Popper’s worlds of knowledge:
• World 1 knowledge – encoded structures in a physical system e.g. DNA
• World 2 knowledge – tacit, beliefs and belief predispositions in minds about the world, the beautiful and the right that we believe have survived our tests and evaluations
• World 3 knowledge – shareable linguistic formulations, knowledge claims about the world, the beautiful and the right e.g. books, wikis

Popper’s three worlds’ model is particularly useful for conceptualizing information systems. Far too often epistemologists have been blind to objective knowledge and been obsessed with tacit knowledge. It is a blind spot equivalent to humans in a pre-Darwinian age not seeing the evolutionary linkage between naked apes and furry apes.  In this age of understanding of DNA and computer systems the existence of objective knowledge is surely not controversial. Animals and plants and humans know things without knowing that they know them. In fact most of our knowledge is not visible to us in any passage of time. Who would deny that a logarithmic table is knowledge, even though possibly no one memorizes it? Organizational data, information, and knowledge are World 3 objects.

A datum is the value of an observable, measurable, or calculable attribute or experience. Data are more than one attribute value. Information is always provided by a datum or data, because data are always specified in some conceptual context. One should avoid talking about data flowing like water into buckets, rather the agent via its senses acts as a searchlight that codes experience into data.

I recommend this book not only to information technology systems professionals but also to managers in general, psychology and philosophy students. There is much wisdom in it and the meta-context it provides could help prevent the building of systems that meet a dead end due to lack of focus on the cycle of knowledge.