So for my IB course I’m currently in the process of writing an essay for my Theory of Knowledge component of my diploma. If you have the good fortune of being a stranger to the International Baccalaureate (oh how I envy you), ToK is essentially the IB’s attempt to introduce philosophy into the general course, except with a hyperfixation on epistemology. In other words “how (and why) do we know what we know”. Part of my essay is about analyzing the difference between knowing something and understanding it, and this post is essentially my method of writing down some of my thoughts so they may come across more clearly when I actually write them in my project. So it’ll be probably more raggedy and incoherent than usual (in case you thought I couldn’t devolve my coherence further) but here we go.

Our old philosophy pal Plato saw knowledge as Justified True Belief. That is, in order to know something, you must believe it to be true, it must be actually true, and your belief must be founded on some valid reasoning. This is the baby-intro definition of knowledge and hence what I’m using for my IB submission. Gettier showed that this definition had a few flaws but do we really need that level of critical thinking? For IB? Really?

Now understanding is a much more interesting concept to…well…understand. There has been significantly less philosophical literature on the nature of understanding than on the nature of knowledge, and I have no clue why this is the case. Regardless, my favorite interpretation is Wilkenfeld’s idea of representational manipulability. Essentially, the idea is that in order to say you understand an object, you must have a mental representation of it in your head somewhere, and you must have the ability to manipulate that representation to a certain extent as well..

Take the image above, for example. In order to know the equations for tension, mass and force, you simply have to believe that they are true and have a good reason for doing so (e.g confirmed by experiment). But to understand them, you must be able to relate those equations to a mental representation of a pulley system, and furthermore you must be able to manipulate that mental image so as to realise multiple different examples of the pulley system.

I think that’s a neat way of defining understanding for a couple of reasons. 1), it gives a good reason for why understanding should be emphasized so much more than simple knowledge. Knowing an equation is useful only in the context of that equation, but when you understand it, you can make use of that knowledge in other aspects of life as well. 2) It interestingly shows us that, although computers may know more than most humans ever will, none of them (at least as of today) have the understanding capability of a human. This is why computers find it tough to tackle new problems they have not been programmed for. Even when making use of Machine Learning in AI, those programs don’t work as well once you feed them an input slightly different than what they have been “trained” for. This is, of course, because they don’t have an abstract mental representation that they can arbitrarily manipulate of the thing they are analyzing.

I think this distinction between the two is also important for educators to make. Reflecting back, I realise a lot of the exams I’ve had throughout school have tested knowledge far more than understanding. To the extent that, if you just fed a bunch of definitions into some elementary program, it could probably score 60-70% on those exams. Is it really all that imperative to test knowledge even nearly as much as understanding, let alone more than? In my view, exam and syllabus creators need to start catering their classes, their courses, their tests to emphasis understanding things more than just knowing them.

Anyways that’s my little ramble on the topic, but unfortunately my procastinatory detour from essay writing must come to an end, and I have to get back to finishing this up. You know how it is, you understand, right?