Towards skillful construction

Teaching Doesn't Exist, But Learning Does

A Sketch

There is a certain view, which I have found useful, that postulates “teaching doesn’t exist, but learning does”.

A teacher might think that they are teaching you to speak better when they correct your pronunciation of a word, but you might instead be learning that “speaking up leads to embarrassment and should be avoided at all costs”. Such a teacher has provided data for the student to learn from that includes their words, intonation, expression, demeanor, and more. Such a student has done the best that they can to learn from this data.

The most common model for teaching is “shoving knowledge into students heads”. The problem is that the knowledge that is shoved is often not what is learned. This becomes especially clear when teachers, or those who would teach, become upset at apparently intransigent students, and try even harder to force the knowledge into their heads without changing their way of teaching. This seldom works, and from the view of “no teaching, only learning”, this makes sense. The teacher has just modified that data that they are providing to students by adding something like “and teacher is angry that you don’t understand this” to it. It is of little surprise that such a modification doesn’t make learning easier.

Given the common model of teaching, discussions on who has the right to teach or what teachers ought to teach are often actually about who ought have the power to discipline students and control their bodies (e.g. when they can use their mouths, or stand and walk). People seem to seldom realize that, within such a system, it is not usually the putative content which is being learned, but the structure, discipline, and worldview of the system.

Different people learn different things from the same data. This is almost tautological – one of the fundamental things that it means to be a different person is to respond differently to the same situation, and to view, understand, or regard the same thing differently from others.

There is an extreme version of this view, which says stuff like:

While there is truth to this view, it simplifies away many important details and dynamics.

Some data can be bad and destabilizing. For an entertaining sketch of this, imagine if this started describing all of the details of you, the reader’s, life, culminating in this present moment, and then said that the reason that this sketch was able to do this is that life is a simulation and nothing is real. Obviously, this could be extremely destabilizing, even though it is just data.

Some data is easier to learn from than other data. Some data results in more profound shifts than other data. But it is all data, and you are learning from all of it.

Some of the most generative books that I have read have been books that I vehemently disagree with. These books were so generative because the process of engaging with them greatly clarified what I thought and why. Though I think that they were largely wrong, I think that they were very useful data. This points to the fact that for data, correctness is distinct from usefulness.

To deeply assimilate data, one usually needs to substantively engage with it. Engaging with data might involve turning it around, playing with it, deconstructing it, reconstructing it, acting it out, or any number of other things.

If you can’t engage with the data, it is very hard to actually assimilate it. Exposure to the data is not sufficient for assimilation.

The complexity of a learner’s own understanding usually far exceeds the complexity of the examples that they are learning from. Thus, anyone who presents examples usually has relatively little control authority for determining how someone learns from their examples.

On the flip side, sometimes others can be dangerously good and changing our information environment, and are able to “take us apart and rewire us” through their control of our information environment.

Controlling an information environment usually involves not just presenting an individual with new data, but preventing them from being presented with certain data. Hence the importance of social isolation to manipulation.

While effective manipulators actively prevent you from getting data that they don’t want you to have, the most effective manipulators make it so that you actively avoid engaging with data that they don’t want you to engage with.

Collin,