The Importance Of Unlearning - Part 2

Introduction

In Part 1, we discussed what 'unlearning' is, and my first attempts to apply unlearning to my own chess game (after hitting my first real plateau, at the 1800 rating).
⁣
While any experienced player can benefit from the process of 'unlearning', I find it to be an especially important tool for adult improvers.⁣
⁣
One reason for this is that while a 'growth mindset' comes naturally to young players (unless continually suppressed by their environment), adult improvers are more likely to have a 'fixed mindset', and see chess improvement as the acquisition of knowledge, or the application of formulas, rather than as a skill for continual deliberate practice (the way we see driving, swimming or martial arts, for instance). ⁣

How We Learn Chess Ideas

⁣
There are various ways this can manifest, but I will use a relatively straightforward example.⁣
⁣
Early in our chess development, we appreciate that the queen is the strongest piece on the board. We draw the natural conclusion that the queen is the best attacker and try to attack using the queen at every opportunity. ⁣
⁣
However, we discover that 'Scholar's Mate' is not so effective after all, and develop more mature ways of building up an attack. We understand that we will just lose time and lose the initiative if our queen is constantly getting chased by the opponent's pieces with tempo. ⁣

Memorization (Exact Moves) vs. Generalization (Principles)

⁣
If we focus only on remembering that exact move sequence, we don't generalize the principle that the queen is a relatively ineffective attacker in many instances. ⁣
⁣
Concretely, because the queen's power means that we won't, in a vacuum, give up the queen for a less valuable defender, but also because, if our queen is attacked by a rook, defending our queen with another piece won't stop the threat. ⁣
⁣
A simpler way to express this principle is that pawns are the most effective attackers, because as the least valuable pieces on the board, they are the most dispensable, and can be pushed, exchanged, or sacrificed to remove key defenders and clear the way for our more valuable pieces to enter the attack.⁣

The Costs Of Overgeneralization

⁣
But if we are so fixated on simplifying positions down to a principle, and then trying to apply this principle everywhere (rather than seeking to master the skill to the point where we correctly apply it unconsciously), we will miss many great opportunities to bring our queen into the attack:⁣
⁣
- To fork two undefended pieces at once (especially useful in queen vs. multiple piece positions);⁣
- To set up a queen + bishop or queen + rook battery to force concessions (like a weakening of the enemy king's pawn shelter);⁣
- As a 'supporting actress' to mop up the attack after the less valuable attackers have cleared the key defenders out of the way. ⁣
⁣
By not getting attached to one end of the spectrum, and realizing that the truth usually lies somewhere in the middle, we broaden our horizons and keep our chess brain open to absorb new ideas (that may ostensibly contradict our existing knowledge). ⁣

Summary/Conclusion

⁣
You now know the 'what' and the 'why' of unlearning, but how exactly do we implement it?⁣
⁣
We already covered the first two steps:⁣
⁣
1. Appreciating that some parts of our current knowledge may not be supporting us;⁣
⁣
2. Identifying what needs to be unlearned, and what should replace it; ⁣
⁣
Frankly, the second one is not easy to do by yourself. As the late Richard Feynman put it, 'The first principle is that you must not fool yourself, and you are the easiest person to fool'. ⁣
⁣
I will share the third step with you in Part 3, see you there!