Myths in Go #1 "Joseki"

[Loons]

Dinerchtein & Younggil's book New Moves gives an interesting glimpse of pro joseki familiarity, analysis and later analysis.

[RobertJasiek]

If you're figuring out the thickness/influence/territory after the joseki is over, that's sort of midgame concepts.

1) Application during creation of a joseki (or a similar corner sequence) is also possible.

2) For determination of the stone difference and the territory, the skills are the same as for the middle game.

3) Influence stone difference is applicable to every joseki but only for some middle game tasks.

4) The relation between stone difference, territory count and influence stone difference is applicable only for josekis (or a similar corner sequences) because a) the (almost-)equality is presumed for the relation, b) the value of an early corner stone is known and applied for the relation, c) valid ratios of territory and influence are known for josekis but not (in general) for other local middle game positions.

So you're still arguing for midgame knowledge.

The method is specific for josekis.

In fact, think about why so many pros seem to say "i don't really know joseki"

1) Understanding of knowledge and positional context embedding can always be deepened further.

2) Huge dictionaries always contain more variations than one already knows.

3) They do not know how to relate territory with influence well because they have not studied my method for this yet.

[Bill Spight]

Studying joseki by memorizing standard sequences is like studying icebergs by photographing what is above the water.

Which is to say; useful! If you're taking into account environmental variables like the densities of icebergs and seawater.

That's what makes this a good analogy.

[Bill Spight]

So you're also agreeing that the middle game is more important than the joseki that got you there. In other words, how you navigate the outcome of your opening will bring you the win more than the opening steps.

Well, the middle game is longer than the opening. But one indicator of the importance a player places on a move is the time he spends on it. By that token, most amateurs do not place much importance on the opening, while most pros do.

Amateur is just that, amateur. These problems apply to even dan players. If you find me just saying that insulting be prepared to be insulted some more, haha.
Dan players make the same fundamental mistakes that kyu players do, just in different ways. This includes joseki.

And that is obvious to you from your vantage point as a professional.

Let's also stress the fact that a "10 point mistake" doesn't exactly always mean 10 tangible points. It could be applied in a way that is realized in thickness or aji. Don't take that number and think that it actually means he's ahead by that much. It's a little bit more vague than that.

Indeed. Endgame gains and losses are the most concrete, while opening gains and losses are the least concrete. That may be why some amateurs discount them.

It is true that large plays emerge in the middle game, but the gains and losses of middle game plays are typically smaller than those of opening plays.

If you don't know what you're getting out of the joseki you're playing, you're doing yourself and your game a great disservice.

Well, as you point out, amateurs don't know much.

If you play within the bounds of what you comfortably can understand, it is better than playing the "correct" move that might lead to a variation you can't handle.

I heartily disagree. I recall seeing a variation of Sakata's that was exactly what I would have played. Both players made good shape, so I figured that it was an even exchange. Sakata thought otherwise. He thought that allowing the opponent to make good shape was not good enough. Sakata chose a different play, even though, as he admitted, he had been unable to read it out.

I realized that in my own games I had chosen plays that, like the variation, led to good shape for both sides. Because I had understood the results, I thought that I had understood the plays. Sakata taught me that I had to look more deeply.

In short, if amateurs stick to plays that they think that they understand, they are making inferior plays. If you look at each of these plays, the loss may not be much, say a point or two per play. But a difference of only one point per play is approximately equivalent to a 9 stone handicap. The problem is not making an occasional slightly inferior play, but adopting an approach of continually and consistently losing points. That is how games are frittered away.

In terms of joseki how often do players say, I didn't make that play because I do not know the joseki! I think that that is a mistake, both in terms of learning the game and in terms of winning games. If you think that a play is right, make it. If you want to play a game you understand, stick to tic-tac-toe.

[Bantari]

In short, if amateurs stick to plays that they think that they understand, they are making inferior plays.

So the point you are trying to make is that you think we all should make moves we don't understand instead??
I disagree.

Pros moves are usually better than ama moves. But it is because they follow better ideas and fit these ideas better, not because the stones magically land on better intersections. We should learn from them, but not ape them. And this means - try to grasp the ideas, and then make moves you understand, moves that fit into these ideas. The better we understand, the better we can choose the ideas, and the better we can make moves which fit.

Kageyama in his book said something about pros moves being often the same as beginner moves in certain situations. What differentiates them is that pros understand their moves much better, so even if they make the same plays as beginners, their game is much stronger. Now, if we make the pro moves without understanding them, then we regress to the level of beginners.

Of course, we all should stretch ourselves, always try to find something better than what we know. Like Sakata in the example you gave - he saw the "proper" (or "normal" or whatever) way where both sides made good shape, but this was not enough and he strived to find something better. And I bet that even if he could not read it all out completely, he understood the moves he made. This is what pros do... but I bet often they overstretch instead as well, only the delta is probably much smaller.

So I would say:
1. Try to only make moves you understand, and
2. Constantly strive to improve and refine your understanding.

PS>
This touches a little on the idea I have of western vs. easter teaching. One of the methods would call for making moves you don't understand, and when you make them often enough, understanding will come. Well, its a method... appropriate in some cases, but probably not in the cases of most of us on this forum.

The other methods calls for gaining (at least some) understanding before acting, and this is what I advocate for.

[Bill Spight]

In short, if amateurs stick to plays that they think that they understand, they are making inferior plays.

So the point you are trying to make is that you think we all should make moves we don't understand instead??

No, that is not what I am saying.

First, I am saying to have some humility. Realize how little you understand, and that when you think you understand, you may not.

Second, I am saying that your comfort level is a poor guide.

Third, I am saying back your judgement. Sooner or later you are thrown back on it, anyway. Your judgement can never be sure, even if you are meijin. Accept that fact.

[John Fairbairn]

In short, if amateurs stick to plays that they think that they understand, they are making inferior plays.

So the point you are trying to make is that you think we all should make moves we don't understand instead??
I disagree.

In pre-forum days, when people answered a point by making the point what they wanted it to be instead of what it was, I used to think they were deviously using rhetoric, or playing at being politicians. But in forum days, where people actually quote and highlight the point they intend to (mis-)answer, I am coming round to the belief that there is something deeper going on.

Either way, Bantari's response to Bill is a travesty of what Bill actually said, and in that lies, I think (if you are prepared to look before you leap on me), a clue to the deep theme of this thread, which was famously summed up by Lao Zi well over 2000 years ago: "The Way that can be understood is not the true Way" (a difficult concept exacerbated by exploiting a pun on dao meaning 'way' as well as 'understand/speak of'). He rammed this home by adding that the name that can be named is not the true name.

Or, to put it in terms of Bill's words, if you think you understand you probably don't really understand.

This difficult issue is universal but has been crystallised out mainly ever since western culture met eastern culture in a big way. At first it was described simply as topsey-turveydom, but more rigorous thinkers (and business-school quacks who write books for hoy polloi) have since battled with it by trying to use dichotomies such as analysis/synthesis, eastern/western, strategic/tactical, memorisation/assimilation, Sun Zi/Clausewitz and so on.

In the context of this thread, analysis/synthesis is perhaps the most useful dichotomy, even if a little old fashioned now. People like Robert represent an extreme of the analysis side, but his tendencies are very widely shared - Bantari clearly leans that way, too. But analysis has its (severe) limitations. You can describe as much as you like how to catch a ball: you can measure the height, speed and trajectory of the toss, regulate the speed of the hand, factor in the wind, and so on, and do all this with enormous accuracy. You can even do all this and make a machine that will catch a ball better than a human. But you can't teach a human how to catch a ball that way. Arguably, you can't teach a human how to catch a ball at all - he has to learn it from within. For a human, understanding is not the goal. The goal is just being able to do it. And in the case of catching a ball, trying to understand the process doesn't make acquiring the skill any faster. It can even be harmful as it slows the brain down.

That's the central flaw in Robert's arrogance that pros like Maeda need to learn his system to become stronger. It's a good way to become weaker, because the goal of his method is to measure and understand. A pro just wants to be stronger, and as his grade shows, he is significantly further along that path than a 5-dan amateur. And it's nothing to do with having a bigger pool of players in the west. It's mindset. Bill's anecdote about Sakata illustrates that.

One of the famous examples from Lao Zi is the bowl. You can make a bowl that is bigger, prettier, cheaper than anything else, but none of that obviates the fact that the only thing that makes a bowl useful is the empty space you put your cereal in. Sakata eschewed the 'prettiness' and easy result of good shape and looked for the useful empty space instead. He wasn't sure he'd found it, but at least he knew there was something deeper he should be looking for (again putting it in Bill's terms, play the move you feel is right, not the move you think you understand).

There will be those who say, that's all very well but how do I translate that into becoming stronger now. That immediately brings to mind one of T Mark's favourite quotations: the lady (she was always American in his version, but doesn't have to be) who said "God grant me patience - and I want it NOW!!" I came across a more subtle version of that just last week. I got talking to an old guy in a park in Shanghai who was doing taiji, and I remarked that he was doing it much more slowly than anyone, even experts, I've seen doing it in Britain. He smiled and said, "It's not about slow movement; it's about slow mind".

And thinking about that profound remark later, it occurred to me that even those in the west who do genuinely try to understand the synthesis/eastern/assimilation approach of Lao Zi's Dao De Jing have a tendency to focus too much on 'body/doing' over 'mind/not doing' and, more specifically, on Dao (understanding) over De (virtues). Yet there are eastern scholars who argue even that the correct order of the book in ancient times was De Dao Jing, and thinking about the book in those terms can be illuminating. For example, one virtue is "daring not to be first" (how does the sea become the king of all streams - because it lies lower than they do). A good starting point!

I think it is no accident that analysers find chess easier than go, whereas synthesisers are most adept at go. Apart from anything else, the synthesisers (as humans, not machines) have the advantage that they can use the results of analysis, but the analysers find it hard to use the results of synthesis. Presumably this was behind Kasparov's attempt to introduce Advanced Chess where humans were allowed to consult and override computers. Lao Zi himself said this (of the Way, after explaining that we should look at its inner essence, ever hidden but should also look at its outward essence, always manifest): "These both flow from the same source. Though differently named they are both called mysteries. And the mystery of the mysteries is the gateway to all marvels."

[jtman24]

Please go easy on me. I am a beginner and continuing my longtime learning of the game. I want to reply to this from the point of view of a total beginner. When I first picked up the game, I read a lot of concepts of go including Joseki. I learnt the basic moves and most of them are corner Josekis. I then looked at a few pro games and realise that they don't play them except for some occasions with corner moves. Then recently I saw a youtube video of go commentary https://www.youtube.com/watch?v=Ofu1xRwE2_4 and realise that Joseki is not just about corners but moving around the whole board in perfect harmony (which in real play, it is impossible to play perfect harmony because your opponent would not want to be lured into it and I think that's why most games end up with resignations and very few with o.5 points). However, when I study pro games, they don't follow "perfect" Joseki. I think it is because once they start playing certain moves, the opponent breaks it down and he has to find a move to balance it out. To me Joseki is balance of your moves to the opponent's moves. Recently, I also use Smartgo Kifu to study pro games. When I do, I turn on the Fuseki and Joseki options. I find that the Fuseki database shows that it only applies up to moves 8 or 9 and Joseki database analysis only applies up to around moves 28 to 30. Thereafter, it is about middle game and endgame. When I play (note: I haven't played against human opponents yet) against Igowin Pro/MFOG, I mark out generally where I want to move in the first 8 to 9 moves (using basic corner Joseki where necessary) and from 9 to 30, I mark out potential territories (I look for balance and make sure that I don't get boxed in. If I do get boxed, I counter with external moves and making balance shapes there) and I don't use Joseki here. I make sure that I gain as much territories as possible and thereafter attack and defence to gain territories and reduce my opponent's territories and finalise them at the end. I find that it works. Game 7 of Lee Sedol and Gu Li Jubango Match open my eyes to how to play go with balance, whole board connections, good shape, making good shape while running and so forth. Therefore, when I play I use all those concepts. I don't think Joseki. I am constantly discovering many possibilities. I know I have a long way to go.

[EdLee]

( Ninja'd by Bill, John, and jtman. )

For a human, understanding is not the goal. The goal is just being able to do it.

The ball example is great. My understanding, or at least the wording of it, is slightly different --

The physics and math of catching a ball represent one level of understanding, (A). (Perhaps, we can call this an intellectual understanding. )
Being able to catch a ball is another level (B), a mind-body understanding.

It's not whether (A) is a higher level than (B), or vice versa.
The point is that (A) and (B) are two different levels, or spheres, of understanding.

More than once, people have brought up a similar analogy of the "theory fighter with zero actual combat experience"
versus the street fighter with years of actual combat experience -- like the ball example, they too represent two different understandings.

Bantari, the river-monk analogy has a similar meaning to the ball analogy --
it's also about the intellectual understanding, and the mind-body understanding.

[joellercoaster]

Or, to put it in terms of Bill's words, if you think you understand you probably don't really understand.

"If you think you understand quantum mechanics, you don't understand quantum mechanics"?

[shapenaji]

Or, to put it in terms of Bill's words, if you think you understand you probably don't really understand.

"If you think you understand quantum mechanics, you don't understand quantum mechanics"?

I don't know if this analogy is applicable. My understanding from my time with the subject was that, in learning quantum, there's an incentive to run too far with some of its startling revelations, and turn it into a Heisenbergian black box that hides things people don't understand.

Quantum Mechanics is still, at its core, a system which can be applied. I'm not sure that learning go falls in the same category.

[Aidoneus]

For a human, understanding is not the goal. The goal is just being able to do it.

The ball example is great. My understanding, or at least the wording of it, is slightly different --

The physics and math of catching a ball represent one level of understanding, (A). (Perhaps, we can call this an intellectual understanding. )
Being able to catch a ball is another level (B), a mind-body understanding.

It's not whether (A) is a higher level than (B), or vice versa.
The point is that (A) and (B) are two different levels, or spheres, of understanding.

I have been following this and your understanding thread with considerable interest. At the risk of sounding like a broken record, many of the ideas you have started to articulate were addressed by Robert Pirsig in Zen and the Art of Motorcycle Maintenance. In particular, he had much to say about Classical versus Romantic (or, as John Fairbairn says, analytic versus synthetic) understanding or aspects of Quality. And he speculated that his pre-electroshock self's undivided, undifferentiated concept of Quality might be the same as the Dao, though his post-electroshock self has reservations. Another, nearly forgotten, contributor to this subject was Henri Poincaré, who wrote quite a bit about "pre-intellectual" awareness; e.g., from The Monist, vol. 20 (1910) https://archive.org/details/jstor-27900262

Thank you for your stimulating discussions, I'll now go back to my observatory.

[Bill Spight]

Here is an example (one that I have used before; my apologies to those who are seeing it again) that can be taken to support often’s idea that players should stick to what they understand instead of making correct plays that they do not understand.

(;FF[4]SZ[19]GM[1]AP[GOWrite:2.3.46]ST[2]CA[ISO8859-1]PM[2]GN[Do you know Joe]HA[7]C[*** This was played one afternoon at the Nihon Kiin Yaesu branch at Tokyo Station. I asked for this handicap placement because I did not know how to play in an open corner.]BR[4 kyu]PB[Bill Spight]RE[B+R ]FG[259:]AB[pd][pp][dp][ji][kj][jk][ij]PW[Unknown ]WR[2 dan]
;W[cd]
;B[ec]
;W[hd]
;C[*** I had picked up this much joseki. Not that I understood it, OC.]B[ef]
;W[ee]C[*** A correct play that White did not understand. Should he not have played it?]
;C[*** My bad.]B[de]
;W[dd]
;B[ed]
;W[fe]
;C[*** So far, so good. At this point I am not entirely sure of the play, but I think that these are the next few moves.]B[ce]
;W[be]
;B[bd]
;W[bc]
;C[*** A tesuji I had learned the previous day.]B[ad]
;W[ac]
;B[cc]
;W[ae]
;C[*** I think he was surprised by the fact that I knew this play.]B[bd]
;W[ad]
;B[dc]
;W[bd]C[*** The rest is a haze. But I did capture the 8 stone dumpling and he resigned.]
)

But I take it to support my claim that we do not really understand plays that we think we understand. So to only make plays that we understand is too restrictive a rule. Do we really want to say that a 2 dan should not have played ? Leave that move to the pros?

[Bill Spight]

Another example. I know I'm tooting my own horn. But hey! why not?

As a 2 kyu I had learned some joseki. Here I chose to back my judgement and deviate, with disastrous results.

(;FF[4]SZ[19]GM[1]AP[GOWrite:2.3.46]ST[2]CA[ISO8859-1]PM[2]GN[Back your judgement 1]HA[4]BR[2 kyu]PB[Bill Spight]RE[W+ ]FG[259:]AB[dd][pd][pp][dp]PW[Unknown]WR[4 dan]
;W[fc]
;B[jc]
;W[df]
;B[ef]
;W[eg]
;B[ee]
;W[cc]
;B[dc]
;W[cd]
(
;C[*** I knew that this was not joseki, but I backed my judgement. See variations.]B[ce]
;W[de]C[*** The rest I do not recall. It was a disaster for me. Would I make the play again? Well, maybe.]
)
(
;C[*** Joseki, but I did not know that.]B[dg]
)

(
;C[*** The book play. Based, I suppose, on making plays that you understand. But I did not like the empty triangle.]B[de]
)

)

Seven years later I faced the same situation, this time against a pro. The pro is unknown because it was at a go school where a number of pros were each playing against two or three students at once. You just sat down where directed.

(;FF[4]SZ[19]GM[1]AP[GOWrite:2.3.46]ST[2]CA[ISO8859-1]PM[2]GN[Back your judgement 2]HA[4]BR[3 dan]PB[Bill Spight]RE[? ]FG[259:]AB[dd][pd][pp][dp]PW[Unknown pro]WR[7 dan]
;W[fc]
;B[jc]
;W[df]
;B[ef]
;W[eg]
;B[ee]
;W[cc]
;B[dc]
;W[cd]
;C[*** Seven years later I still liked this move. I backed my judgement and played it again.]B[ce]
;W[de]
;C[*** This time I saw this play. Something, BTW, that the 4 dan had not seen during the post mortem of the previous game.]B[be]
;W[cf]
;B[cb]
;W[ed]C[*** Something I had not seen.]
;B[ec]
;W[fd]
(
;C[*** Not sure where to play. See variation.]B[ff]
;W[db]C[*** Let's make a deal.]
;B[eb]
;W[bb]
;B[da]
;W[bd]
;C[*** Looks OK to me.]B[fb]
)
(
;C[*** Maybe here?]B[ge]
)

)

Well, I backed my judgement again, and this time the result looked OK to me. I anticipated that the pro would comment on my discovery and maybe pat me on the back. But not a word. I did ask him about , and he said that either my play or the variation was fine. A while later I saw the very sequence that we had played in a go magazine. It turned out that in those seven years it had become joseki.

The first time I backed my judgement and made a play that I did not understand, and got creamed. The second time I backed my judgement and made a play that I did understand, with success. Right? Well, not exactly. I understood the play better the second time around, but I had not read out the full joseki sequence. If I had stuck to plays I understood, I would not have made the play either time. If I had stuck to published joseki, I might have made the play the second time, if I had kept up sufficiently. But I still think that it was better to play joseki without playing joseki. And in fact, if I had not been backing my judgement all along, I doubt whether I would have made 3 dan by then.

[Bantari]

In short, if amateurs stick to plays that they think that they understand, they are making inferior plays.

So the point you are trying to make is that you think we all should make moves we don't understand instead??

No, that is not what I am saying.

First, I am saying to have some humility. Realize how little you understand, and that when you think you understand, you may not.

Second, I am saying that your comfort level is a poor guide.

Third, I am saying back your judgement. Sooner or later you are thrown back on it, anyway. Your judgement can never be sure, even if you are meijin. Accept that fact.

I agree with the above, pretty much. And none of it runs against what I have said.

But in my defense, going back to what you said previously, it still seems to me that if you advocate against making moves we (think we) understand, you are basically saying that you advocate for making moves we don't (think we) understand, or at least for moves where our perceived understanding is not really a factor. This is what I disagreed with. If this is not what you said, as it seems from what you say now, then there must have been a misunderstanding on my part.

Of course, understanding is a spectrum not a binary, and I don't think anybody ever can say they "understand" something in its entirety. Not amateurs, and not the pros. And I don't think anybody here meant full understanding in this context, I certainly did not. There is always only understanding of a certain, higher or lower, level. Its more of a philosophical than practical question - can we really say we fully *understand* anything? Can anybody?

And yet we still have to use whatever limited understanding we have to make the decisions we make. This has nothing to do with humility, except maybe that humility should prompt you to realize that you need to continuously work on reaching deeper understanding.

Next you speak about "backing" your judgement - but this is exactly my point! When you make moves you (think you) understand, you can put forth arguments to back up your decisions - even if the arguments are weak or faulty. If you simply make "correct" moves you don't understand, you can put no arguments forth, all you can say is "I saw a pro play this move, no clue why" as your justification. And I think this is a very poor justification. Even if you *know* a move is more correct. What's more, this justification will not help you make any follow-up moves, correct or not.

But then - can you ever *know* a move is "correct"? Its same difference as with understanding... The humility you speak of usually prompts me to admit that I have no clue, and realistically, the pro who played it probably also was not quite sure, truth be told. So by making such moves, all you are really doing is not playing more "correctly" (although, strictly speaking, maybe you are), but implementing ideas of somebody else, ideas you don't know, don't understand, and cannot follow up logically. How is that good, I don't know.

Either way, Bantari's response to Bill is a travesty of what Bill actually said, and in that lies, I think

I don't see it like that. The post Bill was responding to made a statement that, I paraphrase, it is better to play a move you (think you) understand than to play a move you think is correct but which you do not understand (and so cannot logically follow up with appropriate moves.) At least - this is how I understood PP's words, excluding the little word "comfortable", as I said in my previous post.

Well, I happen to agree with often and disagree with Bill, provided my interpretation of what was said and meant was correct. If my interpretation was not correct, then my apologies, although I do not really think apologies are necessary, did not try to offend nobody, just voicing an opinion which happens to disagree with what I read Bill's opinion was. How is this a "travesty"?

To reiterate - my opinion is that it is better to make moves which you (think you) understand - i.e. ones with some kind of ideas behind them - then moves which you just seen pros play and so suspect might be "correct" (or more correct than your moves, which might or might not be true), but which you have no understanding of and so cannot follow them up in a coherent way. Provided you combine this with the "humility" Bill mentioned and the attitude that you need to constantly work on improving your understanding.

If we all agree on that, sweet. If not, that's ok too.