It is currently Wed Jul 17, 2019 7:31 pm

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 66 posts ]  Go to page 1, 2, 3, 4  Next
Author Message
Offline
 Post subject: Recognition vs understanding
Post #1 Posted: Mon Jul 16, 2012 4:18 am 
Oza

Posts: 2334
Liked others: 15
Was liked: 3352
I have a question. Those best placed to answer it will probably be go players who are also chess players. The reason for the question is that a good answer will help authors like me and go teachers.

First the background. Three things have conspired. One is that, as a result of a grandson learning chess, I have started to take some interest again in that game, which I last played over 40 years ago. What 'interest' means here is not playing or studying the game again, but rather reading chess news and interviews. It has been fascinating to chart the parallels with go. Chess seems to have the same debates as go (sponsorship problems, Mickey Mouse time limits, etiquette problems, citizenship debates, etc.), and since they always seem a little ahead of go, what is happening there may happen soon in go. (A couple of possible items: Fischer-type time increments, which seem to be the norm now in chess; a reversion to interest in classical chess, i.e. not blitz, even to the extent of resuming adjournments regardless of the presence of computers). But my reading about chess has also included quite a bit about improving and teaching. There, there seem to be radical developments - more below.

Another factor for me was my interest in the art of commentary, which I have elaborated on in the Life, Games and Commentaries of Honinbo Shuei trilogy. However, as that is a little esoteric I won't say more about it here.

The third (and trigger) factor was that I was reading an oldish book by Go Seigen (How to think about joseki), which seemed to pre-date much of the new thinking about improving in chess. This book is unusual in many ways. One is that there is none of the usual "high level" pro talk about matching the joseki to the fuseki (except in the nugatory case that some josekis involve a ladder). Instead Go talks about the "back alleys" of josekis. As you'd expect, he is scornful of those who just memorise josekis, e.g. as in quoting three senryu (5-7-5 = 17-syllable ancient vox pop poems): (1) For each joseki/Memorised, you will become/Two full stones weaker. (2) The guy who says/'The joseki goes like this...'/He's the one who lost. (3) New joseki learnt,/But agony! His rival/Skips the club that night. (Yes, the first one is the source of the go proverb - so 2 stones is poetic licence, not a pro assessment).

But Go is also scathing of fellow professionals. He doesn't mention them by name, though by tracking his examples with the GoGoD database, or referring to books, you can guess some. In fact there are many. For, in short, he says there are 'many bad josekis'. One is shown below.



This appears in joseki books (e.g. Ishida) as a basic joseki. It also attracts comments to do with a supposed disadvantage for Black in having to deal with the ladder stone. Nonsense, says Go (and in the book he illustrates quite a lot of similar examples for josekis we think of as basic or clever). This is a 'bad joseki'.

Just as a taster, he points out that in tewari terms Black should omit 7 and just play 9. If White then plays 8, Black clasps the ladder stone again, but in that position would he ever then exchange 7 for 10? However, that is by the by. Go's real point is that this is not really how to evaluate a joseki. Nor is understanding fundamental principles, and this is where he foreshadows the new chess thinking. The point about this joseki (and all the others he shows) is that it is possible to play 'in the back streets'. In this case, far from the need to capture the ladder stone being a weakness, it is a strength, because then Black can live in (yes, actually inside) the corner. I leave that as an exercise for the reader, but I say again that Go shows this sort of "back streets = aji" thing repeatedly for josekis that appear in books and which are (wrongly) evaluated in terms of fundamentals such as territory versus thickness. In other words, the primacy goes to tactics - always. This is where, for me, the chess angle comes in.

A chess book that caught my eye enough to buy it was "Move First, Think Later" by Willy Hendriks. I haven't yet read all of it, and part of the reason for my question here is to see whether it will be worth making the effort to finish it, but the beginning is certainly though-provoking (and very well written). Essentially, the book seems to have resulted from Hendriks's own experience in which he was a trainer. He would (like every other trainer) present pupils with lessons replete with fundamental principles and he would painstakingly try to instil a thought process in which the pupils examined the position for strengths and weaknesses and then carefully formulated a plan of attack or defence. However, he found he was being disconcerted too often. He would present a lovingly crafted move that (say) involved a retreating manoeuvre ("the sort of moves trainers love" as he put it). Then someone would say, but why not just go for the king at once and play Rxg7? Sure enough that was a killer move. But when Hendriks tried to work out how people came up with these killer moves, he could not elicit any sort of thought process. People just saw the moves. Realising that this was happening even with ordinary good moves and not just sacrifices, he (again as he put it) began to "eat humble pie" and he began to research the topic. What he discovered was that a couple of recent writers had got part of the way there before him. John Watson in Secrets of Modern Chess Strategy was one. The other was Alex Yermolinnsky in The Road to Chess Improvement. Watson apparently stressed the "primacy of the concrete" (i.e. tactics, I gather) and Yermolinsky apparently summed up his advice as "get to work" - the way to reach a higher level was "not to take generalities from books but to struggle with concrete positions". I have not seen either of these books, but from elsewhere I gather that this change in thinking has evolved from chess computers. Many openings evaluated in the past through the standard filter of general principles have been revitalised by tactics found by computers, and top players have (I gather) begun to think in the same, "new" way. I say "new" because this seems to me to be no different from the way go books have been written for donkeys years (i.e. reams of concrete examples or problems rather than pages of general principles).

According to Hendriks, what is required to make good moves at the board is not understanding or planning but simply recognising. The more positions you have seen, and the more times you have seen them before, the better you will be at recognising an opportunity in the game before you. This alters the burden on trainers. It may be that principles still have a role to play, but afterwards. In other words they are no longer fundamental, but are just post-facto tools to help you as you "get to work", but even then the idea seems not really to help you understand per se but to use them to imprint the position in your mind. But again we've heard this before in go: recall the hoary old adage that the way to 1-dan is to play over the moves of 1,000 games.

So now we come to my question. Can anyone offer views beyond my own so far very limited knowledge on how thinking on this topic is developing in the chess world? Are Hendriks, Watson and Yermolinsky going to end up in the "back streets" or are they genuine pioneers? Is Hendriks's book worth finishing? Are the others worth reading?

Forestalling the answers a little, I have to say that my recent fresh look at the chess world (from the standpoint of a writer) is that books on openings appear still to be a fetish for the average player, but there is a strong current (new to me) of game commentaries which focus on explaining the alleged thinking, even to the level of explaining every move. When I did play chess I would have lusted after these. I was strongly put off by commentaries that were full of: 9. Ng3. "In Bad Joseki, 1919, von Spifflewurf played 9.Bb4 followed by [add 500 moves]", and I know I was not alone. But I am now beginning to wonder whether writers who write like that are in some ways closer to the truth: they are "recognising" a position and its close relatives, and not even bothering to try to understand it.

Before fingers rush to the Reply button, may I be allowed to stress that what I would like to see discussed is chiefly LESSONS FOR GO FROM NEW CHESS THINKING ABOUT TRAINING. I am not at all interested in seeing yet another rehash of the topics discussed here before.


This post by John Fairbairn was liked by 6 people: Bill Spight, emeraldemon, gasana, gogameguru, nagano, Phelan
Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #2 Posted: Mon Jul 16, 2012 4:59 am 
Tengen

Posts: 4249
Location: North Carolina
Liked others: 446
Was liked: 696
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 5k
Good question, and I don't know the answer--I suspect I follow the chess world even less than you, but there has been a lot of commentary about how the newest generation of players has been shaped by the presence of computers. In particular, Magnus Carlsen is supposed to be a rather "disorganized" student of the game, and only took on a coach after he had reached an extremely high level. What he did was play constantly against the computer. The idea that computers can speed up the development of young players is pretty ordinary, but the idea that it might change the way they study and emphasize experience/intuition and all of that is a bit more interesting.

Oddly, I've also heard it said that Carlsen has a very positional style, and lacks the really sharp tactical sense of some other top players.

_________________
Occupy Babel!

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #3 Posted: Mon Jul 16, 2012 5:43 am 
Tengen
User avatar

Posts: 4504
Location: Chatteris, UK
Liked others: 1580
Was liked: 651
Rank: Nebulous
GD Posts: 918
KGS: topazg
hyperpape wrote:
Good question, and I don't know the answer--I suspect I follow the chess world even less than you, but there has been a lot of commentary about how the newest generation of players has been shaped by the presence of computers. In particular, Magnus Carlsen is supposed to be a rather "disorganized" student of the game, and only took on a coach after he had reached an extremely high level. What he did was play constantly against the computer. The idea that computers can speed up the development of young players is pretty ordinary, but the idea that it might change the way they study and emphasize experience/intuition and all of that is a bit more interesting.

Oddly, I've also heard it said that Carlsen has a very positional style, and lacks the really sharp tactical sense of some other top players.


That's interesting, because I heard one of the most intriguing aspects of his development was his _lack_ of computer usage (http://www.chessbase.com/newsdetail.asp?newsid=7778 for example).

My understanding was he felt that they could be as much of a handicap as a help due to over-reliance (amongst other things). However, he seems to strongly advocate the "recognition" school of thought, and seems to consider working in detail on concrete "real" positions as the absolute must for improving.

One of the chess authors I've always liked a lot was Jeremy Silman, who aside from transcribing huge numbers of player comments from his students and responses and thoughts, does most of his teaching by presenting game positions, asking the reader to distil it into viable general game plans that fit the pieces on the board and their position, then seeing how to make those plans come to fruition or whether they would be forestalled by an opponent's equally viable general plan coming to fruition first (in which case, back to the drawing board etc).

All my limited experience with Go professionals on certain positions appear to also be similar - that the recognition aspect of the game trumps completely when pitted against amateurs (even strong ones). The instancy with which they'll have picked what appears to be the right move seems to be what makes them stand out, and reading is used as a confirmation tool rather than a decision making tool.

I'm unsure whether there's a big Chess groundswell towards this sort of thinking in teaching, but chessbase is probably the best place to look. Many of Andrew Martin's videos and the tutorial DvDs by other top GMs may offer some insights as to whether or not it seems to have become common.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #4 Posted: Mon Jul 16, 2012 6:01 am 
Tengen

Posts: 4249
Location: North Carolina
Liked others: 446
Was liked: 696
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 5k
Hmm, perhaps I misremembered. But on the other hand, there's a difference between using computers for analysis, database searches and all that, vs. having the internet or computer opponents constantly available, which is what I was thinking of. So a kid like Carlsen can sit at home and just play (or replay games) without going out and searching for a teacher or anything like that.

_________________
Occupy Babel!

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #5 Posted: Mon Jul 16, 2012 6:55 am 
Gosei

Posts: 1366
Liked others: 348
Was liked: 386
Rank: 5d
GD Posts: 1000
I'm not sure I understand your question, John. Hypothetically imagine a player who has played through thousands of pro game records. When that player actually plays a game are you saying that recognition of positions, rather than invoking general principles, should guide move choice? We routinely condemn players who, by rote, play "joseki" moves because they don't understand the reasons for their moves. Usually this means they would follow the joseki moves in positions where that joseki is a poor choice. Presumably they wouldn't see such moves in the pro games they have played through. But what happens when their opponent plays a move they've never seen before (shin te)? If they don't have some sort of general principle to guide them how do they decide their move? By "back street" tactics? Maybe what you mean is that the player, by examining many many pro games, would unconsciously have internalized the "principles". This might be what go pros refer to as "feeling".

Concerning training, it would seem that the trainer would just supervise the student's playing through games, perhaps answering questions. And game reviews would consist of correcting mistakes but not expounding general principles. It reminds me of what Kato Masao said about being a disciple in the Kitani dojo, namely that Kitani didn't actually teach, he just watched the disciples play each other, and somehow they became stronger.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #6 Posted: Mon Jul 16, 2012 7:46 am 
Oza

Posts: 2334
Liked others: 15
Was liked: 3352
Quote:
I'm not sure I understand your question, John. Hypothetically imagine a player who has played through thousands of pro game records. When that player actually plays a game are you saying that recognition of positions, rather than invoking general principles, should guide move choice?


I'm saying that this appears to be the new thinking. Recognition may come from mere repetition but that's bad recognition. Good recognition comes from "effortful" struggling with real positions. Presumably some sort of principles are inculcated in that process, but the point seems to be that they are not fundamental in the sense of coming first (i.e. Move First, Think Later). Again presumably, the brain does all of this subconsciously (the Carlsen interview, amongst other articles, seems to suggest that). Since, however, the end result depends on the input (garbage in, garbage out), a trainer, or a Kitani, can be valuable in ensuring the input is good. As far as I can see (and this applies to the Kitani example you gave), the best guidance is not "here is a list of essential principles you must learn first" but a very vague instruction such as "it's now time for you to study the games of Shuei". Obviously many chess and go trainers could see that as detrimental to their livelihoods, so may speak against it, but it does seem to fit all the cases of pros reaching the top in chess and go.

Trainers needn't worry as I would say there is a strong market among players who never will and don't want to reach the top, and just want to learn to enjoy the games more. But even there, the chess examples seem to suggest that trainers of these less serious players are (or should be) shifting towards "recognition" teaching, though perhaps in a less pure form than might be used by inseis.

As to how recognisers cope with positions they have never seen before, in a glib way we can say they don't. That's why they lose and go away and learn to recognise even more positions. Once a player is truly on his own, maybe general intelligence becomes a major factor?

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #7 Posted: Mon Jul 16, 2012 9:28 am 
Tengen

Posts: 4784
Liked others: 0
Was liked: 648
1) Of course, the actual position has to be taken into account. Principles presuming a simplified position need to be modified. Principles working with a dynamic position can still be applied.

2) Go Seigen makes a premature conclusion "bad joseki" because he studies only partial aspects but misses other evaluation methods. According to chapter 3 of Joseki 3 Dictionary, the ratio is 3.3 and the joseki type 'ordinary sharing of territory and influence'. This qualifies it as a "good joseki", and, if anything, this ratio has a tendency to favour the territory player, i.e. White. While there can, of course, be positional contexts in which Go's tewari or not removing the cutting stone aspects become relevant, this cannot be said for all positional contexts. Otherwise, my evaluation theory is a strong indication for the "good joseki" side of the argument.

3) Of course, studying by examples only is bad, and since you start this discussion again, you get it again, regardless of your preference not to get it again. If you don't like reopened discussion, then do not reopen it! Studying by examples only is bad because there are simply too many shapes and reading sequences. Principles and other generalising knowledge prune efficiently. It does not surprise me that Go Seigen or others do not trust that much in principles: they do not study generalising theory enough. Without proper study, they will, of course, come to the wrong conclusion that it would be an inefficient approach. If they cannot develop mighty general theory on their own, then they must study the general theories invented by others, e.g. me. Go was good at co-inventing New Fuseki, but afterwards his generalising inventions have not been that great.

4) Of course, one can learn from computers.

5) "working in detail on concrete 'real' positions as the absolute must for improving" says little as long as one does not mention the study tools. Recognition only? Generalising theory only? Or both?

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #8 Posted: Mon Jul 16, 2012 9:48 am 
Honinbo

Posts: 8591
Liked others: 2511
Was liked: 2962
Two comments:

First, the Fred Seki was something that I was taught was not joseki, and that was long before the Ishida came out. It may well have been joseki in the 19th century, or something. Science marches on.

Second, the importance of seeing has been around for a long time, too. IIRC, Krogius mentions it in Psychology in Chess, and maybe it is also in Chess: The Mechanics of the Mind. The emphasis may be new. :)

I first learned about the importance of seeing in bridge, from Victor Mollo. In go, I have always said that seeing is better than reading. That may be a first in go pedagogy. :)

Edit: I would not call seeing recognition. :)

Edit 2: I got the term, seeing, from Mollo. I don't think that the chess books use that exact term. :)

_________________
There is one human race.
----------------------------------------------------

The Adkins Principle:

At some point, doesn't thinking have to go on?

— Winona Adkins

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #9 Posted: Mon Jul 16, 2012 10:45 am 
Gosei

Posts: 1366
Liked others: 348
Was liked: 386
Rank: 5d
GD Posts: 1000
John Fairbairn wrote:

Recognition may come from mere repetition but that's bad recognition. Good recognition comes from "effortful" struggling with real positions. Presumably some sort of principles are inculcated in that process, but the point seems to be that they are not fundamental in the sense of coming first (i.e. Move First, Think Later). Again presumably, the brain does all of this subconsciously (the Carlsen interview, amongst other articles, seems to suggest that).


I like this distinction between effortful practicing and mere repetition. Children learn to talk through just doing it. They receive a lot of correction in the process but they aren't given general principles (like put your lower lip touching your front upper teeth when you want to say an "f" sound). I believe learning to talk is an effortful process. A goal is constantly visible (audible?) so the learner has something to aim for.

I think that "effortful" must include some sort of attention to and awareness of the goal. In the Kitani dojo Kitani's presence in the practice room must have made the disciples focus their attention.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #10 Posted: Mon Jul 16, 2012 11:05 am 
Lives in sente

Posts: 1228
Liked others: 98
Was liked: 262
I thought that Botvinnik covered this sort of thing.
Anyway, can't beat Kotov :) I had forgotten that he has this "Kotov Syndrome" we see so much in amateur fuseki. 40 minutes on move, and it's still pants :)

_________________
North Lecale

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #11 Posted: Mon Jul 16, 2012 11:20 am 
Tengen

Posts: 4784
Liked others: 0
Was liked: 648
Bill Spight wrote:
the Fred Seki


What is this, please?

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #12 Posted: Mon Jul 16, 2012 11:22 am 
Tengen

Posts: 4784
Liked others: 0
Was liked: 648
gowan wrote:
Children learn to talk through just doing it. They receive a lot of correction in the process but they aren't given general principles


They are given, learn and apply general principles (e.g., grammar rules) implicitly, later at school partly explicitly.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #13 Posted: Mon Jul 16, 2012 11:24 am 
Tengen

Posts: 4784
Liked others: 0
Was liked: 648
Javaness2 wrote:
we see so much in amateur fuseki. 40 minutes on move, and it's still pants :)


We also see 40 minutes thinking by amateurs with good results.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #14 Posted: Mon Jul 16, 2012 11:29 am 
Lives in sente

Posts: 1228
Liked others: 98
Was liked: 262
RobertJasiek wrote:
We also see 40 minutes thinking by amateurs with good results.


Is that relevant? Kotov said that if you think too long about a move, then you forget your initial analysis, allowing the potential of a colossal blunder. Wise advice. Chess training books are so much better at giving tips like this, they simply do not exist in (western) books.

Anyway, to hop back on topic, I don't believe the ideas John quotes are new.

_________________
North Lecale

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #15 Posted: Mon Jul 16, 2012 11:53 am 
Tengen

Posts: 4249
Location: North Carolina
Liked others: 446
Was liked: 696
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 5k
Fred Seki: http://senseis.xmp.net/?UserCreatedGoSlang

_________________
Occupy Babel!

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #16 Posted: Mon Jul 16, 2012 1:44 pm 
Tengen

Posts: 4784
Liked others: 0
Was liked: 648
Javaness2 wrote:
Kotov said that if you think too long about a move, then you forget your initial analysis, allowing the potential of a colossal blunder.


I cannot know if that always applies for Chess players. It does not always apply for go amateurs.

Quote:
Chess training books are so much better at giving tips like this, they simply do not exist in (western) books.


There are indeed no relevant (for playing strength) Western language go psychology books yet.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #17 Posted: Mon Jul 16, 2012 5:58 pm 
Lives in sente

Posts: 1210
Liked others: 733
Was liked: 231
Rank: OGS 1d
KGS: illluck
Tygem: Trickprey
OGS: illluck
RobertJasiek wrote:
Javaness2 wrote:
Kotov said that if you think too long about a move, then you forget your initial analysis, allowing the potential of a colossal blunder.


I cannot know if that always applies for Chess players. It does not always apply for go amateurs.



"Does not always" is not very useful wording :p

There's a well-known Chinese saying that "Lengthy deliberations result in terrible moves" (长考出恶手) and it's often seen in game reports. I haven't really noticed it myself because I usually play pretty quickly and my play is full of terrible moves regardless. I do think that the idea of forgetting the initial analysis to be a good explanation, though.

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #18 Posted: Mon Jul 16, 2012 7:41 pm 
Tengen
User avatar

Posts: 5245
Location: Banbeck Vale
Liked others: 928
Was liked: 1322
Rank: 1D AGA
GD Posts: 1512
Kaya handle: Test
RobertJasiek wrote:
Bill Spight wrote:
the Fred Seki


What is this, please?


Fred <> Joe

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #19 Posted: Mon Jul 16, 2012 10:51 pm 
Lives in gote

Posts: 370
Liked others: 91
Was liked: 252
Rank: Weak
Hi John. It seems that what you are describing as "effortful struggles" that give rise to pattern recognition is essentially how we study life-and-death and tesuji. We do these problems repeatedly and gain pattern recognition skills that allow us to come up with candidate moves fairly quickly. The rules of L&D permit a few widely applicable generalizations, but, for the most part, it is difficult to come up with a general algorithm to determine life-and-death in most situations. Thus, repeated problem solving and resulting pattern recognition skills form important bases for in-game L&D determination. So I would say that the pattern-based teaching is already ocurring in the go world. Wouldn't you agree?

Top
 Profile  
 
Offline
 Post subject: Re: Recognition vs understanding
Post #20 Posted: Mon Jul 16, 2012 10:52 pm 
Tengen

Posts: 4784
Liked others: 0
Was liked: 648
illluck wrote:
"Does not always" is not very useful wording


As long as neither side of the argument has evidence for relative frequency, there is no better wording:)

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 66 posts ]  Go to page 1, 2, 3, 4  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group