On Teaching: Polanyi’s Paradox

I was recently reading an article about ALPHAGO, the computer that’s been programmed to be the best Go player ever. In the article, the author described Polanyi’s Paradox, which can briefly be described as “know[ing] more than [you] can tell.” One example the article cited was face recognition– most humans are incredibly good at this, but we cannot describe how to do this (and it is difficult to program a computer to do so, as seen in the infamous “gorilla” tagging incident last year). In the case of professional GO players, it struck me as somewhat funny and somewhat interesting that the players simply cannot explain how they are so effective (and consistently so) at playing the game well. I imagine some musicians (particularly improvisational musicians) may be in the same boat.

On Teaching: Polanyi’s Paradox?

So what does this all have to do with teaching? I worry that a lot of effective teachers experience Polanyi’s Paradox, even though effective teaching requires multiple, diverse skills. I have to admit that I, myself, am a better researcher and writer than teacher, so I do very consciously what I think many teachers have automated over the years. For example, if a student is interrupting, I still go through an internal monologue where I coach myself to ignore the first student, then praise another student for raising a quiet hand while calling on them, then going back to the first student directly after if they have now raised their hand. Teachers who have been teaching for several years likely don’t go through such a language-heavy internal monologue, thus over time, they may not be able to explain what they do, or (more importantly) how and why they do it.


Familiarity with research on effective classroom environments is important

Even if a teacher is self-aware and reflective enough to be able to explain their internal monologue, many teachers may not be able to provide a consistent framework or research evidence about their teaching.

Aside from being a stickler for evidence-based instruction, an inability to explain why something a teacher is doing is effective makes generalization difficult, and there is absolutely no way that any professional development or teacher training program could address every single scenario that is likely to arise during teaching.

To be clear, PD and teacher training programs shouldn’t address “every single scenario” since teachers shouldn’t teach all scenarios as being different; there are a handful of methods that are effective for vocabulary development, for behavior management, for reading instruction, etc. thus those are all that need to be taught explicitly. Think about the analogy of vocabulary learning: it is complex, but not every word is explicitly taught, so students need to be explicitly taught about prefixes/roots/suffixes, explicitly taught essential content words, and explicitly taught how to apply context clues, but then students consciously or “automatically” (aka subconsciously) will learn thousands of words independently).


More research is needed & that research needs to more accessible to ALL teachers

Why do I say I “worry” about Polanyi’s paradox? For one, it perpetuates the myth of great teachers being born not made. Several research studies have countered this assumption (which is a relief since otherwise, any teacher training or effectiveness research would be irrelevant). In addition, the idea of internal professional development is appealing for certain purposes– it is often more relevant (since everyone knows the students, it makes linking theory to practice easier) and it celebrates the talents of people at your institution (which is valuable for morale and for teacher retention). Do note, that strictly local PD is also ineffective (see Ilana Horn’s excellent post about the need for diverse types of professional development, both internal and external).

If you are able to describe how and why you teach as effectively as you do, not only can you prove to be an effective mentor or instructor, but you are more likely to be consistent in applying those strategies or frameworks. Think of pedagogical frameworks as being like the law– there are certainly some grey areas and times when multiple laws need to be considered simultaneously, but it is still a set of rules that determine what is “right” and what is “wrong” (or in the case of teaching, what is “effective” and what is not). I am not against experimenting with new methods of teaching, but I am against not having pedagogical knowledge as this harms students (i.e. teachers don’t know what they’re doing) and teachers (i.e. teachers feel that they do not know what they’re doing).

All of this to say that I value the presence of “Pedagogical Content Knowledge” in TPACK as

it is not enough to know your content and to know about technology; if you are unable to teach your content and apply technology effectively, all of that is irrelevant!

Stay tuned for my reflections on teaching and Higher Education, which is an area often criticized for lacking Pedagogical Content Knowledge.


What are your thoughts, world? Is it too much to ask that effective teachers be able to explain why their teaching is effective?

Leave a Reply

Your email address will not be published. Required fields are marked *