Reflections on Paul Feyerabend

The best lecture I’ve ever attended was given by Paul Feyerabend, in the Old Theatre of the LSE, in 1967. Two hours before he started, every seat was taken, at least a hundred more were sitting in the aisles, and closed-circuit tv had been set up in separate rooms. People had travelled from abroad, down from Inverness and up from Plymouth to be there, tipped off by a network relying only on letters and phone calls. It was a rowdy affair – cheers, whistles, clapping, feet-stamping from the audience; outraged, hardly-contained protests from members of the LSE philosophy department who wriggled uncomfortably in their seats in the front row, waving their arms, mumbling insults. This was the apotheosis of the old-style university celebrity gig. There were so many such gigs in those days – visiting stars who kept their audience enthralled but never silent – and I wonder if they don’t belong to a long-gone era of university life, snuffed out by modern concerns for political correctness. The only event I can remember that came close to rivalling Feyerabend’s performance was when Joan Robinson, a Marxist economist at Cambridge, came to the LSE a year before to give a brilliant critique of Keynesian economics. She brought the house down, finishing with a wonderfully-cutting, satirical riff aimed at the economics department of LSE. “What makes you laugh”, she concluded, “is the sheer effort they exert to make the whole silly story sound plausible”. The audience rose to give her a standing ovation, while the department members stormed out.

Feyerabend started his academic life in Vienna in the late 1940s. He changed from physics to philosophy, and made plans to study with Wittgenstein in Cambridge, but Wittgenstein died before Feyerabend arrived in England, so Feyerabend ended up, rather begrudgingly, with Karl Popper as his PhD supervisor. He became a big fan of Popper’s falsificationist view, though ideas of the “incommensurate” nature of rival theories of science (a term made famous a decade later by Thomas Khun and which we will hear a lot more about) already concerned him. In a nutshell (a very cracked one): Feyerabend thought Popper was right epistemologically to stress the asymmetry between truth and falsehood (we’ll never know if a theory is true, but we can know if it’s false), but he wasn’t sure that rival theories were comparable. He went back to Vienna, torn between academia and his desire to be an opera singer, and then went to the University of California, Berkeley, where his first really influential paper was published: “An Attempt at a Realistic Interpretation of Experience” (1958). In the paper, he lambasted positivism and supported Popper’s falsificationist view.

It’s important to point out here that Feyerabend’s criticism of positivism referred to a school of philosophy started by Comte (1848) and brought to a disastrous end by the Vienna School in the 1920s. No scientist these days calls themselves a positivist, despite which, the label “positivist” is now commonly used by badly informed so-called scholars in SLA to refer, disparagingly, to those in their field who base their studies on empirical evidence and rational argument. When you read current stuff written by postmodern, social constructivists like Larsen-Freeman, Lantolf, Thorne, Block and so many other intellectual imposters,  “positivists” are the bad guys, the ones suffering from “science-envy”, the ones choking real progress in our understanding of second language learning by adopting an outmoded “scientific paradigm”. Whatever merits their arguments might have, and I see none, let it be perfectly clear that using the term “positivism” to refer to scientific method displays an ignorance of both philosophy in general and the philosophy of science in particular.

Where were we? Ah yes, Feyerabend. He ended up back in London in the early 1960s, where he became good friends with Imre Lakatos, second in command of the Philosophy department of LSE, nominally headed by Karl Popper, but in fact run by Prof. John Watkins. Popper by this time had become world famous, thanks to the publication of Conjectures and Refutations (Popper, 1963), a much more accessible version of The Logic of Scientific Discovery (Popper, 1959), popularly referred to by everybody I knew as “LSD”, a drug we all took far too much of, but which never dented our  allegiance to English beer. By the time Feyerabend gave his lecture, he was already disenchanted with the philosophy department of LSE, which he considered a suffocating environment, something close to a cult, a cabal with increasingly closed ranks devoted to defending the Popperian canon against all critics. I had some personal reasons to agree with him.

In 1965, the philosophy of science was offered as a new option for undergraduates doing the B.Sc. (Econ) degree (before then, it had only been a postgraduate degree). I was among the first cohort to take up the option. Once in, I was delighted by all the clever people I mixed with, delighted by my increasing grasp of Popper (particularly his brilliant reply to my hero Hume), and delighted to meet Paul Feyerabend in the Three Tuns bar, where he regularly held court. I got drunk more than once with Paul and Imre, where all the rows going on within the department were discussed with outlandish candor. Inspired by their rants against W.W. Bartley, whose book The Retreat to Commitment attempted to rescue Popper’s critical rationalist view from its critics, I, still an undergraduate, presented a paper to the department suggesting that Bartley had done nothing useful to address the problem of infinite regress. “It’s still turtles all the way down” I concluded, much to the horror of Watkins, who couldn’t quite believe his ears, but couldn’t muster a good off-the-cuff reply either. He asked me to present a second paper, this time with “the seriousness and rigor one might expect of such a …. bla bla bla”. I did so, and a couple of years later, out came Watkins’ paper “Comprehensibly Critical Rationalism” in the Philosophy journal (Watkins, 1969), where my argument was dressed up with the necessary obscurantist finery, with no acknowledgement of my contribution.

Such was – and still is – the world of academia, but anyway, Feyerabend (not present at either of my presentations) was particularly scathing of the way Watkins and others in the department behaved. At one of our boozy sessions in the Three Tuns bar, Paul suggested that a very bright post grad. student –  Jane, we’ll call her – should take my final exams for me, while he, Imre and I went off on a fishing trip.  Jane was up for it, Imre too, but I bottled out, missed the fishing trip, did the exams myself, and ended up with a mediocre 2.2 honors degree, tho my “essay on a philosophical subject” – on Bartley – got a first. “Never mind”, Lakatos told me “you’ll do a doctorate with us”.  Later that year, 1968, I was involved in “disturbances” at the LSE which resulted in my being named as one of six “ringleaders”, served with an injunction forbidding me access to the LSE, and charged with criminal damage to LSE property.

Where were we? Ah yes, Feyerabend. The lecture. The Old Theatre of LSE must have been a theatre at some point, because it had a stage. At least 20 metres long, I’d say, with a huge blackboard almost as long as its only prop. So on comes Paul Feyerabend, the archetypal German army officer, blonde, blue eyes, and all that and all that, with a very noticeable limp (a war wound, of course). He takes a bit of chalk and draws a line along the blackboard. He stops the line in the middle, makes a tiny space, and continues the line to the end. Then he strides back to the middle, bangs the tiny gap he’s made in the line, and says, in a very loud voice: “Zat, my friends, is Popper’s contribution to the philosophy of science!”.  Instant uproar! Shouts of “Bravo!” and “Nail him!” mingle with “Bollocks!”, “Rubbish!” and “Out!”.

What followed was a tour through the history of science, mostly physics, which aimed to question the status of ‘observation statements’, the basic pillar of science. Observation statements are supposed to be ‘empirical’ – objective statements about the things in the world that we observe through our senses and measure with reliable instruments. But they all turn out to be ‘theory laden” – shot through with unestablished assumptions about how the bits fit together. Feyerabend argued that a theory could easily be constructed using the conceptual apparatus of classical physics that would be just as comprehensive and useful as the classical apparatus, without coinciding with it – after all, the concepts of relativity include all the facts in Newtonian physics, yet the two sets of concepts are completely different and bear no logical relation to each other. So the suggestion that the meaning of observation language is determined by pure observation is wrong. In a body of knowledge, no part can be appraised individually, since each one is connected to others; therefore, there is no theory without observation, nor observation without theory. The meaning of an observation statement is determined neither by the pragmatic conditions in which a language is used, nor by the phenomenon that makes us assert it is true. To quote the 1958 paper, “the interpretation of an observation language is determined by the theories which we use to explain what we observe, and it changes as soon as those theories change”.

And thus, Feyerabend argued, we meet the problem unresolved by Popper, pace the famous Khun versus Popper confrontation at Bedford College, London, in 1965, of incommensurability: “I interpreted observation languages by the theories that explain what we observe. Such interpretations change as soon as theories change. I realized that interpretations of this kind might make it possible to establish deductive relations between rival theories and I tried to find means of comparison that were independent of such relations”.

Now, I have to pause again in my no doubt frustrating account of the lecture to tackle the term ‘incommensurability’ alluded to above. The term was coined by Feyerabend and Kuhn, and it is a term that modern constructivist, postmodernist, relativist intellectual imposters took to their feeble souls for succor, and used to create the worst body of academic work ever seen since the invention of the printing press.

Feyerabend met Thomas Kuhn in Berkley, California, in 1960, and it was Feyerabend who first suggested to Kuhn the idea of incommensurability. Feyerabend was already a star, and Kuhn was relatively unknown. Feyerabend read Khun’s work in 1959, and the two got together the following year. They got on well, but there was always, right from the start, tension between Khun’s sociological view of the history of science and Feyerabend’s more philosophical interest in, well, epistemology – he was, after all, a frustrated Wittgenstein scholar. In fact, Kuhn had better academic credentials – a Ph.D. in theoretical physics versus Feyerabend’s Masters in astronomy – but Feyerabend was surely the better thinker, the better philosophical mind, let’s say. Feyerabend’s initial interest in Kuhn’s work was inspired by his interest in quantum physics, and he wanted to pick Kuhn’s brain about its observational base.

Out of time. I have medical issues to attend to. Join me soon for the concluding Part Two, which will include references.

What Any Fule Kno

Here’s a recent Twitter exchange:

Me: (edited version) Why not use appeals to reliable evidence, and rational argument to discuss the design of a TBLT syllabus, rather than all this hand-waving rhetoric.

Scott Thornbury: Rational argument suggests there is no One Way to teach SLs, because learners, teachers, contexts, institutions, cultures & languages themselves are all different. Nailing a method (however evidence-based) to the cathedral door ignores this infinite variety. Ask me. I tried! 🙂

If you read the whole thread you’ll see that Scott stayed calm and I didn’t. No surprises there, then. I wrote saying sorry to Scott, who was as gracious as ever in accepting my apology.

To the issue, then. I’ve be irked (sic) that Scott’s reply has been “liked” and re-tweeted so many times. So irked, in fact, that I use my blog to repeat, everso politely, that it’s more hand-waving rhetoric. Scott implied that I was trying to persuade people that TBLT is the “One Way” to teach SLs, whereas, in fact, I was doing no such thing. The tweet ends with a particularly adroit rhetorical flourish: Scott knows just how wrong I am to do what he wrongly accuses me of, because he’s done it himself.

The analogy with Luther’s famous nailing of his 95 theses to the church door in Wittenberg is an apt one. Luther was challenging the abusive practices of the Catholic Church, a powerful body which in those days had almost as much reach as the ELT establishment (IELTS, British Council, Cambridge Assessment, Pearson, etc.) enjoys today. He was complaining about the selling of plenary indulgences, certificates signed by the Pope (or one of his approved representatives), which gave those languishing in purgatory a reduced sentence, or even, if the sum were big enough, got them out of there and sent them straight to heaven. Today, many of us (including Scott, if somewhat erratically) challenge the huge power of the ELT establishment, and their right to sell certificates which can have similarly dramatic results – albeit here on earth. And like Luther, we, the ELT rebels of today, demand the end to so much wrong-headed, self-serving interference with, and regulation of, our primary goal. Luther sought salvation through more direct access to God; we seek the rather more mundane goal of a decent level of proficiency through more direct access to holistic, unimpoverished use of the L2.

But, as Luther might have said, and as any fule kno, it’s complicated: there is no One Way. And not just, in the case of instructed second language learning, because “learners, teachers, contexts, institutions, cultures & languages themselves are all different”, but because we don’t know enough about the psychological process of learning an L2. Scott likes to stress the social factors; I think that although there’s still a lot we don’t know, the reliable results of psycholinguistic research should inform syllabus design. As I’ve often said, folllowing Popper, it’s much easier to be sure about what’s wrong than about what’s right, and we know that concentrating on the explicit teaching of items of grammar and lexis is inefficacious. In any case, it’s no more than a platitude to say there’s no One Way, and it’s no more than rhetoric to accuse proponents of TBLT of suggesting that they’ve found it. The argument for Long’s TBLT syllabus design is that it’s more efficacious than the syllabus implemented by coursebooks. Research in SLA supports this view. The fact that the research findings are largely ignored is explained, I suggest, by the influence of commercial interests that treat ELT as a commodity.

So Long’s TBLT syllabus makes no ridiculuous claim to be the One Way. It offers an alternative. It’s open to endless modification according to local conditions, and it’s the subject of on-going research among scholars and increasing discussion among teachers. I repeat: it should be discussed on the basis of the evidence and rational argument.

When Luther marched up to the church door, hammer, nail and inky parchment in hand, he didn’t have One Way in mind, he wanted reform. There’s a decent argument to be had that it made things worse. I hope we get the chance to see how TBLT fares, once it sweeps away coursebook-driven ELT.

SLB: Task-Based Language Teaching Course No. 3

I’m delighted to say that we already have enough participants to ensure that the third edition of our course will run. This is a shameless attempt to persuade you, dear reader, to join us.

There are different versions of TBLT, including  “task-supported” and “hybrid” versions. They all emphasise the importance of students working through communicative activities rather than the pages of a coursebook, but we think the best is Mike Long’s version, which identifies ‘target tasks’ – tasks which the students will actually have to carry out in their professional or private lives – and breaks them down into a series of ‘pedagogic tasks’which form the syllabus. In the course, we consider

  • how to identify target tasks,
  • how to break these down into pedagogic tasks,
  • how to find suitable materials, and
  • how to bring all this together using the most appropriate pedagogic procedures.

What does the course offer? 

It’s an on-line course about Mike Long’s version of TBLT, consisting of twelve, two week sessions.

In the course, we

  • explain the theory behind it;
  • describe and critique Long’s TBLT;
  • develop lighter versions for adoption in more restricted circumstances;
  • trace the steps of designing a TBLT syllabus;
  • show you how to implement and evaluate TBLT in the classroom.

We value interaction and debate. In this edition we have made changes to try to ensure that we ‘walk the talk’, by basing the course on tasks, and by making sure that the participants’ needs drive the course.

Why Do it?

Because it offers an alternative way of doing ELT whose time, we think, has come. Current economic conditions, affected by Covid 19 and its likely successors, push the move towards more local trade and commerce. This move is a powerful force which can help to dislodge the one-size-fits-all global approach to ELT. Now is the time for a new more local, tailor-made approach to ELT, where global General English courses are replaced by TBLT courses. But not just any TBLT courses. Rather, courses which start by finding out what a certain group of learners need to do in English, and then help those learners to reach their objectives by leading them through a series of sequenced pedagogic tasks designed to give them the input and practice they need to perform target tasks in the real world.

From the classroom, to the small groups of students in in-company training courses, to the single private student, TBLT offers a better alternative to the coursebook, and it offers teachers the basis for developing a portfolio. Whether you’re a teacher with a few years of experience, or a DoS, or a syllabus / materials designer, this course will help you re-tool for what lies ahead.  

The version of TBLT we explore offers a viable, not uptopian, option to coursebook-driven ELT. It respects robust findings in SLA research about how people learn an L2. It is guided by a philosophical tradition of libertarian education. It is more rewarding for all those directly concerned. Efficaciousness is our criterion: we seek a more efficacious approach to teaching, an approach which overcomes the weaknesses of coursebook-driven ELT.  Coursebooks implement a syllabus which falsely assumes that teaching about the language should take prime place. This leads to too few opportunities for students to learn for themselves by engaging in relevant communicative tasks where they learn by doing. The result is that learners don’t get the procedual knowledge they need for fluent, spontaneous use of the language. Hence, these courses are ineffacious – they don’t deliver the goods. Does Long’s TBLT result in more efficacious teaching? We believe so. We’ll argue our case, but we expect challenges from participants.

And that’s another reason to do this course. One of the best things about the two previous editions of the course was the quality of discussion among participants. We’ve had experts in special fields (e.g. air-traffic control), directors of studies, and teachers with less than 3 years experience all shooting the breeze in the discussion forums, interrogating the experts, and submitting end of module written work worthy of publication in academic journals. Neil and I defend our corner, but we’ve been persuaded to move our postion considerably as the result of persuasive arguments by participants.  Our course is not a sales pitch for Long’s TBLT – he wouldn’t want it be – it’s a platform for lively discussion about it.

When is it? 

It starts on October 16th 2020 and ends on April 15th 2021.

What are the components of the Sessions?

  • Carefully selected background reading.
  • A video presentation from the session tutor.
  • Interactive exercises to explore key concepts. 
  • An on-going forum discussion with the tutors and fellow course participants.
  • An extensive group videoconference session with the tutors. 
  • An assessed task (e.g. short essay, presentation, task analysis etc.). 

Who are the tutors?

Neil McMillan and I (both experienced teachers with Ph.Ds) do most of the tutoring, but there will also be tutorials by Roger Gilabert, Mike Long, Cathy Doughty and Glenn Fulcher, all widely-recognised experts in their fields. 

How much work is involved?

About 5 hours a week.

 The course sounds very demanding.

Well, it is quite demanding – it’s an in-depth look at a thoroughly-described syllabus. However, we’ve extended the length of the course, and we now offer different options – see “More Info.” below. Reading is non technical, the video presentations are clear, participation in on-line discussions is very relaxed, and the written component is practical and short.

Free Preview

Click here to see Why TBLT?    Use the Menu bar on the left to open “Why TBLT?”. Work your way through the unit, or just watch the presentation.   

For More Information About The Course Click Here