Recognition vs understanding

[RobertJasiek]

a general view that proverbs are NOT nonsensical just because you can find another proverb that says the opposite.

Proverbs can be right or wrong. If two proverbs contradict each other, then at least one of them is wrong.

You don't bring out every tool and use them in a hierarchical order

This ideal cannot always be achieved yet because we do not have a complete tool box yet.

difference between just absorbing good shape and this sort of discovery is that the latter is "effortful" and so is more likely to stick in my mind and/or be easier to recall for practical use.

I go a step further: When parsing collections, I am also looking for principles and other general theory to be discovered. (This is not the only way to discover general theory, of course.)

But that involves work.

It does.

At present I think the main problem for most of us is that we don't have the time, capacity or motivation to do the work, and so instead we grasp at the latest proverb or buy the latest book, or hope that genius will be absorbed via the ether when we watch the latest video. I'll be interested to see what Hendriks has to say about that.

Hopefully he says: finding generalising theory is worth the effort.

[jts]

I can't help shake the feeling that the discussion in this thread is glomming about in a swamp of category errors.

For example, "principle" or sometimes "general principles". What are they? I think you could introduce several candidates for that role, and a general principle about general principles can't be true of all of them.

For example: a name or categorization; an instruction given to a beginner; chastisement given to a weak player; a point of interest in a positional judgment; a flag for significance; a narrative description of strong play; a heuristic or algorithm to follow when lost.

So a general principle like "the hane at the head is good shape" or "a rook on an empty file is liquid gold" can function in any or all of those ways. The same can be said for reading/calculation: sometimes this refers to seeing the whole canopy at once, as Bill reminds us, sometimes to clever pruning, and sometimes to the laborious business of peering down every branch. The same goes for intuition... Sometimes this refers to first impulses, sometimes to the way your understanding of a position develops from absorbing the details of it, sometimes to the astuteness and efficiency of one's reading, sometimes to pattern recognition. Even if we limit ourselves to a general statement about the relationship between principles, reading, and intuition in a single board game, we need to start by being specific about what we mean by each term.

[lemmata]

Indeed there seem to be two possible interpretations of "general" being used here.

1) General = "Often true"

2) General = "Always true without exceptions or with an explicit and complete enumeration of the exceptions"

The first is the common usage in everyday speech. The second also appears in everyday speech but appears most frequently in mathematics.

I suggest that the first meaning be adopted in the remainder of this thread (but not for future threads). We can perhaps use the term "universal" when intending the second meaning.

Under such conventions, two general principles may contradict each other but two universal principles may not.

Examples of general principles:

• Hane at the head of two stones
• From a wall of two, extend three
• Play at the point of symmetry

Examples of universal principles:

• A contiguous string of stones surrounding a contiguous string of empty points 4 long and without any bends (i.e., a "straight four") is unconditionally alive.
• A go board is worth 10.32 gobans.

Universal principles are not only fewer in number than general principles, but also more difficult to state properly.

I think that we can all agree that both types are useful as long as we understand them for what they are.

Universal principles can be used as shortcuts to speed up calculation (reading). General principles can be used to find candidates branches to read out first since we cannot read every possible variation. In one popular 4-4 one-space pincer 3-3 invasion joseki, the general principle "hane at the head of two stones" is violated. Someone who does not know this joseki may consider this hane as a candidate move, but will reject it if he has carefully read out the sequences that start with it. Does this mean that the general principle is bad? No. If the hane works with sufficiently high frequency, the player will earn back the time he lost reading out the hane in the long run. Of course, what is a good general principle may change with what kind of moves most people are playing at any given moment in time, but, as amateurs, we do not need to conscious innovators of game play.

If general principles reflect the Bayesian posterior beliefs of collective experience of past go players, then pattern recognition built up by playing games (or replaying pro games) might reflect the Bayesian posterior beliefs of individual experience.

I don't know what further conclusions to draw... It seems that the transmission of go skills is a very difficult task that is still far from being understood. However, this sort of question is not unique to the go world. Effective transfer of cognitive skills is one of the white whales of business education. The case study method, which is the most commonly used, often ends up giving students a bad habit of digging up the solution used in a known case that is closest to the one they are facing. Most MBAs fail to recognize that the case-to-solution correspondence is not a continuous one.

One conjecture that feels right: Pattern recognition learning is probably more efficient in games with smaller boards. That is, reviewing a database of 2,000 pro games on a 13x13 board is probably exponentially more helpful for getting better at 13x13 games than reviewing a database of 2,000 pro games on a 19x19 board is for getting better at 19x19 games. I imagine that this has something to do with the attention that pattern recognition has received in the chess world.

[ez4u]

the principle inherent in the position...

What is meant with this?

This is a very interesting question. Of course Reti wrote the original in German "Die Neuen Ideen im Schachspiel" (1922 by the way, rather than 1921 as I wrote above) and the English version is a translation. Clearly in the sentence he is contrasting general rules with 'something' in specific positions. However, it is not clear what!

"Modern Ideas in Chess" also has a kindle edition so you can read the author's preface for free (I was too stingy so far to read the whole thing). To me it has a similar zealot's flavor to what JF has written about Yasunaga's books on the New Fuseki.

[RobertJasiek]

"principle" or sometimes "general principles". What are they?

A general principle in contrast to a principle is one that is (more) generally applicable.

A principle can come in different degrees of certainty:

1) Absolute principle: always true. E.g., the New Semeai Formula yields three principles; one of them is: "In a class 1 semeai, if the difference of 'fighting liberties' is zero, then the semeai is a seki or unsettled."

2) Approximative principle: usually or even almost always true. E.g., "Avoid premature endgame."

3) Guideline principle, which implies a preferred consideration but is by far not always true. E.g., "Play away from thickness."

In contrast to principles, proverbs are wrong too often. E.g., "A player having four corners loses [alternative proverb: wins]."

we need to start by being specific about what we mean by each term.

No. This is already started. We need to CONTINUE the good work of those already being specific about terms presumed in principles. E.g., I am careful about defining (for the intended readership otherwise unclear) terms well before using them in, e.g., principles.

[RobertJasiek]

Examples of general principles:

• Hane at the head of two stones
• From a wall of two, extend three
• Play at the point of symmetry

Argh. These are rather proverbs than principles! "Almost always true" would be an overexaggeration.

Universal principles are not only fewer in number than general principles,

I am working on swapping the balance:)

but also more difficult to state properly.

Yes.

It seems that the transmission of go skills is a very difficult task

Not THAT difficult:) The first 30,000 hours of study were really difficult, but now, for me, it has become mostly a matter of available time. I encourage more people to invest a first 30,000 hours to then contribute much to systematic go theory! Maybe the availability of the research fundamentals described by Conway, Spight, me et al now allows a faster education and 30,000 hours are not needed any longer?

[cata]

I'm not convinced that you're really observing a recent trend in chess books. It's not for nothing that you frequently hear the equivalent of "below 1800|2000|2200, tactics win and lose every game." The fact that simple calculation and pattern recognition are the primary skills is not a secret.

However, it's also always been true that people typically find it more entertaining to read about ideas and planning and stories, so books presenting those are popular, as they are in Go.

[tchan001]

There is a Chinese blog where the author, an amateur high dan, said a professional teacher gave him one simple task to achieve great improvement: Analyze the problems in Xuan Xuan Qi Jing. [Obviously JF will be delighted to hear about this advice since he has just come out with his definitive English version, Gateway to All Marvels.]

The author of the blog piece goes on to decompose a number of problems in XXQJ to try to show the various tesuji and L&D situations hidden within each problem.

The objective of the blog author was not to just publish the answers to the problems, but to try to show the inner workings of the various tesuji in a problem and the process of his journey of discovery.

Through this process, he hoped to

(1) gain a thorough familiarity of fundamental techniques to
(a) build up a sufficient knowledge base

and to
(2) appreciate the various methods and order of sequences to
(a) exercise our own discovery of shapes and defects and to
(b) cultivate feelings and insights for tesuji.

In other words, he hopes to build up a sufficiently large internalized knowledge base from the study of XXQJ and to cultivate the ability to form flexible use of this knowledge base.

For a look at this fascinating series (in Chinese), part 001 starts here: http://wszhdm.blog.sohu.com/43373952.html

The series is from 2007 and lasts about 20 posts long with a few post (012, 013, 016) being no longer available (was once posted as sgf form into the weiqi.cn forum which no longer stores the sgf in the thread he made there)

[edit: guess he's not 7d, so I corrected it as high dan]

[RobertJasiek]

frequently hear the equivalent of "below 1800|2000|2200, tactics win and lose every game."

If this is so in Chess, then it is VERY different in Go.

[RobertJasiek]

the author, an amateur 7d

If somebody does nice work, then his name should not remain hidden. What is his name?

[tchan001]

If you read the blog, it says his name is Isaac Chen

[tapir]

JF mentioned Go Seigen on Joseki. I believe it would be useful to remember that to Go Seigen the category weak players included 5 dan amateurs (and probably some fellow professionals as well). None of the people repeating the anti-joseki mantra in the western amateur scene are as strong as Go Seigen and it achieves a different result among amateur players.

Over-reliance on joseki (the few one learned) is in my experience (among kyu players) often a direct result of the continuous stream of discouragement to study joseki which a new player has to face (e.g. [sl=AZenWayToJoseki]A Zen Way To Joseki[/sl] and countless other SL-pages, a number of strong players who feel above playing proper moves, e.g. those opening on 2-2), so they end up simply relying on the few ones they learned. (People who always play a high approach to the 3-4 point, always play a single line after a certain pincer, always play the shoulder hit to the 3-3 point etc. etc.) It does no good when people with a large opening vocabulary actively discourage weaker players from acquiring a vocabulary, but repeatedly tell them they should just speak freely which is so much better than rote memorization. Confidence to speak freely grows with experience and vocabulary. I find this surprising in a post titled recognition vs. understanding, because the supposed addition to memorizing that turns joseki study into something useful is almost always "understanding". But why not consider joseki study as the exposition to many different patterns that might allow you to recognize them later?

[RobertJasiek]

to Go Seigen the category weak players included 5 dan amateurs

What is the relevance of this? 1) Nowadays, Japanese 5d is ridiculously weak indeed. I do not know about the level of amateur ranks when he made that statement. 2) Playing strength and strength in creating go theoretical understanding are so different that one does not determine the other. 3) If he wanted to imply weak theoretical insight be amateurs, then he should provide his insight so that he might convince somebody that his go theoretical understanding was better than that of those amateurs he characterises as "weak".

IOW, if he referred to PLAYING strength only, then there is no relevance at all for strength in THEORETICAL UNDERSTANDING.

It does no good when people with a large opening vocabulary actively discourage weaker players from acquiring a vocabulary, but repeatedly tell them they should just speak freely which is so much better than rote memorization.

Right, but which people do you mean?

why not consider joseki study as the exposition to many different patterns that might allow you to recognize them later?

Because, e.g., the same pattern can be right or wrong in two different global positional contexts.

[hyperpape]

Over-reliance on joseki (the few one learned) is in my experience (among kyu players) often a direct result of the continuous stream of discouragement to study joseki which a new player has to face (e.g. A Zen Way To Joseki and countless other SL-pages, a number of strong players who feel above playing proper moves, e.g. those opening on 2-2), so they end up simply relying on the few ones they learned.

This is one of those points that makes me just stare and ask "why didn't I think of that?"

[illluck]

Issac Chen (on the left-side panel). I'm not sure if he's 7 dan, though - his info states that his goal is to reach 7 dan and that wish is also found in the part part "...也许能够给其他和我一样有志升到业余7段的同好一些帮助" (... can perhaps help others who also strive to reach amateur 7 dan).


Edit: Opps, didn't see that there was a third page where tchan provided the name already. Also, forgot to thank tchan for the link - it looks very interesting indeed. It's also worth noting that Mr. Chen is likely referring to Chinese 7 dan which is the highest normal amateur rank (generally considered to be professional strength), and from what little I've seen in his blog he'd definitely be a high dan on KGS.