Concept Analysis in Programming Language Research: Done Well It Is All Right (To appear in Onward’17)

I just submitted my camera ready version of Concept Analysis in
Programming Language Research: Done Well It Is All Right
, a methodology essay which has been accepted at Onward’2017. I will probably write about it more extensively later.

Here is the accepted version, for personal use (not for redistribution): PDF (copyright 2017 Antti-Juhani Kaijanaho, exclusively licensed to ACM).

Antti-Juhani Kaijanaho. 2017. Concept Analysis in Programming Language Research. In Proceedings of 2017 ACM SIGPLAN International Symposium on New Ideas, New Paradigms, and Reflections on Programming and Software (Onward!’17). ACM, New York, NY, USA, 14 pages. (the DOI link will not work until October).

Ramblings inspired by Feyerabend’s Against Method, Part I: My road to Feyerabend

About 15 years ago, I studied for an exam on the philosophy of science, a required (and very much anticipated) part of my minor in philosophy. I must have learned about Karl Popper and his falsificationism, which did not really appeal to me. There was one thing that hit me hard, one that remained stuck in my mind for more than a decade: the idea of paradigms, which consisted of a hard core (usually immutable) and a protective belt (easily changed to account for discovered anomalies), and of scientific revolutions which occasionally happen (usually by generational shift), replacing one hard core with another.

About a decade later, I started debating the methodology and foundations of science with my colleague Ville Isomöttönen. At some point I suggested we both read an introductory textbook on the topic to inform our discussions. I believe we read Peter Godfrey-Smith’s excellent Theory and Reality.

In this book, learned again about paradigms, and noticed that there were several philosophers I had conflated together. Thomas Kuhn talked about paradigms, but the idea of hard cores and protective belts comes from Imre Lakatos, who did not talk about paradigms but used his own term, that of a research programme. Then there was Paul Feyerabend, who was basically crazy. Or that’s how I remember my reaction of reading about him in Godfrey-Smith’s textbook.

This was around the time I started working on the research that became my licentiate thesis. Very early on, one of my advisors, Dr. Vesa Lappalainen, asked me to explain what evidence is. That turned out to be a very difficult question to answer; I continued reading and pondering about it until I submitted the thesis, and even beyond that point. I figured that the philosophy of science probably has an answer, but I cannot really base my discussion of it in a thesis solely on introductory courses and textbooks. I needed to go read the originals.

The first original book on the philosophy of science I read during this period was Thomas Kuhn’s The Structure of Scientific Revolutions. I also borrowed from the library a copy of Karl Popper’s The Logic of Scientific Discovery, of which I was able to finish only the first chapter at this time. Kuhn was very interesting, and I finally realized how thoroughly I had misunderstood him from the secondary sources; his arguments made quite a bit of sense, but his insistence of at most one paradigm in each discipline was obviously false. Popper’s falsificationism is obviously true, but also severely inadequate.

Very early on during the licentiate thesis study, as I was doing preliminary literature research on evidence-based medicine (EBM), I came across the blog Science-Based Medicine, and particularly their post series critiquing EBM (start from Homeopathy and Evidence-Based Medicine: Back to the Future Part V). From this and other sources, I learned of Bayesian epistemology, which I started reading about over the next couple of years. As I have written previously on this blog, it is my current preferred theory of epistemology.

This Spring, some months after the licentiate thesis was approved, I traveled to Essen, Germany, for a three-month research visit at the University of Duisburg-Essen. Two very significant things happened there: I wrote a substantial part of my doctoral dissertation (currently pending public defense) and I spent quite a bit of time discussing the philosophy and methodology of science with Dr. Stefan Hanenberg, who had been one of the examiners of the licentiate thesis. The topics of those discussions probably had something to do with that the chapters I was writing there dealt with philosophy and epistemology.

During that time, I finally read Imre Lakatos’s work on the philosophy of science (The Methodology of Scientific Research Programmes) and on the philosophy of mathematics (Proofs and Refutations), both of which were eye-opening. Lakatos spends a lot of time in the former construing and critiquing Popper, and that discussion allowed me to understand Popper the first time ever (though I recognize it’s from Lakatos’s version of Popper); I finally read Popper’s Logic of Scientific Discovery properly also at that point.

The discussions with Dr. Hanenberg frequently came back to Paul Feyerabend and his Against Method. I knew it well enough from secondary sources to know that I was not going to cite it in the dissertation, and so I did not read it at that point. The time to do that was once the dissertation was submitted.

My next post will discuss my actual reactions to the book, as I just finished it yesterday.