In my life I have regularly been asked, in one way or another, to explain how chess makes you better at life. Many chess related things will try to convince you of this, and I have remained unconvinced. The more compelling explanations, in my opinion, will tell you that it improves your ability to think sequentially or can offer a form of mental discipline, though in my experience the students this works for are people who already have these traits. I’m sure for some these are true. I’d like to talk about something else though. What does chess show us about thinking well?
Chess is an incredibly difficult cognitive activity that, for the most part, does not parallel thinking in other areas all that well. But it is a measurable activity that is very challenging. And while the specific thoughts might not apply elsewhere – an enemy backwards pawn is both a target in itself and often comes with a potential outpost on the square in front of it that can be even more valuable for us to exploit is probably not thinking I can apply to a decision about which car to buy – what better thinking looks like is something we can apply broadly.
In Think Like a Super-GM one thing is made eminently clear. Weaker players spend most of their time thinking about what they can do, while stronger players spend most of their time thinking about what is wrong with their ideas. I believe, broadly, this is emblematic of good thinking. The online grifter will tell you doctors don’t what you to know about this one simple trick, their method is perfect and will solve everything. It may or may not be easy but it has no drawbacks or doubts. Most of us, I hope, can identify this kind of grifting. Compare to a scientific paper that tests a hypothesis and comes back with there is a small but noticeable effect. They had a hypothesis and they checked if real world evidence contradicted it, after having challenged their belief with real world data they did not conclude this one simple trick will change everything, but that what they were looking at has a non-zero effect. Peer review, p-hacking, there are serious issues, but it is an obviously legitimate pursuit in the way that ads I get on youtube are not. One is infused with what if this is wrong, we must test if this is wrong, while the other is borne of asking you to imagine what if it’s right; what if this is a game changing secret.
The weaker chess player spends their time finding ways for their ideas to work, the better player spends their time refuting the ideas they have. The salesmen (in all forms, I mean famous entrepreneurs at least as much as I mean car salesmen) of our world do a remarkable job of convincing people of the former, but it is the latter where the real thinking happens. I’m not saying chess trains us to do this well; I’m saying that we can measure who is better at thinking about chess – an incredibly complex thinking task – and also measure how differently those who are better think, and learn from that about how to think better in other areas. So many of my projects begin with my making a bazillion caveats about how my methodology doesn’t cover x, y or z. And it’d be better if it did! But without certain tools I can’t cover those things, and it’s very important to be aware of all the mental adjustments and flaws in my output that are there. This is unproductive presentation if I want to sell venture capitalists on the Juicero, but it is good thinking if the goal of thinking is to be… good. Chess can show us what good thinking looks like, and our world could use more of it. And more distrust when it isn’t there.
A thing I’ve noticed in chess thinking, made most explicit by Aagaard’s three positional questions, is that better thinking involves questions rather than answers. The beginner is told to develop their pieces, and this is useful advice. They make too many pawn moves, or too many moves with the same piece (often queen or knight) while neglecting the rest of their forces. As a player gets better the phrase may come to feel banal, but we also get better at the relevant skill. We choose more active squares to develop our pieces to, we prioritize the development of some pieces over others (often kingside over queenside so that we can castle, often at least one knight before the first bishop, and often the second knight before the second bishop) and our development fits more harmoniously with the pawn structure, or, even, with augmentations to the pawn structure we intend to make to the future. Carlsen is better at developing his pieces than I am, while I am better at it than a beginner.
What if as a player developed they moved from the declarative develop your pieces to the question what piece development has impact?
When IM John Bartholomew shows the Opera House Game he calls it an efficient masterpiece. Morphy’s games are filled with optimized active development. But if you start scanning his games, for instance this one or that one, you’ll notice he doesn’t develop his queenside bishop immediately, and when he does, it is very rapidly followed by the queenside rook entering the game. Until that rook has a place to go bringing out the bishop is not that impactful. It doesn’t help us get castled and in these e4 e5 structures it doesn’t immediately have a useful square to go to, so, being efficient, he prioritizes other moves over its development. Even in the Opera House game you’ll notice the first time has a free move after playing Bg5 he spends it castling queenside, moving that bishop is about the rook just as much as about the bishop.
Playing a blitz match against a stronger player six or so years back I noticed I was often concluding development, then looking for a plan. He, on the other hand, paused earlier rather than automatically bringing that queenside bishop to a plausible square, and often achieved large advantages in the late opening/early middlegame against me. I was developing my pieces, while he was figuring out which moves were more and less useful to make. He was asking does this move help? Do I need to commit this piece to that square now or is the pawn structure going to change and a better square will become preferable? He was asking questions while I was following rules. Sometimes his questions would conclude that yes, he should just finish development, sometimes they wouldn’t. The question had the chance to support what blindly following the rule would say, but it might also direct you another path.
When Aagaard asks you to think what is the worst placed piece this doesn’t give you the answer to the position. But it might help you notice that you’ve left your queenside bishop alone, and that including it in operations might make all the difference. Or maybe you might conclude that said bishop is just a move away from relevance on any number of squares and not the problem, rather you might note that your knight on f3, which looks very respectable, isn’t doing anything and needs a new home. Or maybe you look sadly upon your a1 rook and conclude yes, my bishop does need to come out, to free the rook as Morphy ordains. It does not give an answer, it gives direction, that might or might not lead us to developing our pieces. It is what thinking looks like.
I think, perhaps, we might do better in chess, and in life, if we transformed many of our guidelines into questions. This would not work at all for people who suffer from decision paralysis as it makes everything into an obligation to think rather than just to do, but if something is complex than asking questions directs us better than following broad heuristics, and every heuristic has questions nestled in it to be explored. Knights before bishops is a useful guideline. Knights are slower than bishops and more close ranged, they need help getting close to the action. It is also, often, more clear where they belong. Knights sit best at the edge of our pawn structure on squares like f3 and c3, then they usually wait to see how the action plays out, but a nearby perch is best to adapt to whatever happens. Bishops are fast and if a square is clearly best they can get there quickly. Committing them very early is often unnecessary. There are lots more little understandings here, but the details aren’t the point. The point is we might ask ourselves where is my piece best placed? and notice that the knight has an obvious square while the bishop it’s not so clear, as Lisitsin, I think, said back in the 50s, if you have two moves you know you want to make and one is committal and one is not, make the un-committal move first. Though of course you and I might prefer to ask ourselves which of these moves I desire is less committal and also are either of these moves urgent such that the opportunity to play them will disappear?
If we want to improve at something difficult then our understandings need to grow. The broader space of questions, that have the ability to both convince us of something or refute it can be far more useful than declarative rules, even if we understand the nuances of those rules. Follow the money is not a question, but it acts similarly to one. It invites us to investigate a direction easily forgotten that can help us contextualize what’s going on. However it has no ability to refute the idea that fiscal interests are not the determinant factor in a situation. That mostly does okay, fiscal interests are often determinative. But even with such a useful bit of advice, surely it’d be more useful as a question…
Thinking to refute ourselves and thinking in questions does not mean we cannot conclude declaratives. It describes the process of getting there. In the end, Kasparov makes a move. He has considered the strengths and flaws of the possibilities, but something is concluded. He has asked, perhaps not explicitly, the position at hand the questions he needs to ask it to find the best answer, and answer it he does. I do not mean to suggest at any point here that good thinking should lead to ambiguous conclusions clouded with doubts. It means we’ve thought about the doubts, that we’ve taken the doubts seriously and are careful about what our conclusion can and can’t address. Chess is an extremely nuanced, complex and difficult game, that, nonetheless, has better and worse choices, has clearly correct answers. To better get to those answers more often, we should think better. To train ourselves to do so more broadly we should spend more time in the shower not arguing with others, but arguing with ourselves, trying to refute ourselves. And when we see an interesting direction on an issue we hadn’t considered we should figure out what question we might try asking ourselves regularly that’d lead us to see find that angle without aid. We can practice asking that question, seeing when it applies and when it doesn’t, and over time we’ll find that applying it has become reflexive and no longer needs conscious thought.
raskerino chess 