Chess and Logic

[The idea]

Is chess, more than other board games, a game of logic?

It seems to me that chess is simply an extremely complicated and dynamic game based on a very simple set of rules. In other words, it is a game of rules rather than a game of logic.

Are there principles of logic that can be well illustrated by means of chess games?

[The idea]

By "dynamic" I meant to say that sometimes black seems to be winning and then suddenly white seems to be winning and then suddenly black again seems to be winning, and so on.

[T'ai Chi]

Originally posted by The idea
Is chess, more than other board games, a game of logic?


I personally don't think so after reading about and playing trough some of Bronstein's games.

There is a lot of artistry and creativity in it. I think it is a mix of both that and calculation.

[Hand Bent Spoon]

If computers can be made to beat a human opponent 100% of the time (and we are almost there already), then chess must be a game of logic, not one of creativity, or art (beyond the styling of the pieces and board, which I consider to be one of the most beautiful objects created by human beings).

Based on this, I think chess is a game of logic and not one of chance or creativity.

In anoter 10-15 years I would expect they will have developed computers which no human can beat at chess.

[T'ai Chi]

Originally posted by Hand Bent Spoon
If computers can be made to beat a human opponent 100% of the time


Not regularly beating grandmasters. Not in regular timed events. And not without megs of opening books and endgame tables.

Still a far way off, but people have been claiming "eventually" for years (and I don't disagree with that).

[The idea]

Originally posted by Hand Bent Spoon
If computers can be made to beat a human opponent 100% of the time (and we are almost there already), then chess must be a game of logic [...]
Can you provide an argument in favor of the principle that you are relying on?

In other words, can you explain why, if computers can beat humans at a game, then the game should be described as a game of logic?

For example, computers can beat human beings at memory games. Does it follow that there is no distinction between games of memory and games of logic?

Obviously chess is a game of rules. No one can deny that. Maybe I can elaborate on that. There is no need for an umpire. There is no need for any weighing or measuring. A given move is either clearly legal or clearly illegal. It is not difficult for an honest person who has learned the rules to determine whether or not any given allegedly legal move really is a legal move.

There are no "chess judges". There are no "chess lawyers."

[Humphreys]

Most board games are games of rules.

It's surely a game of both rules and logic. You can logically work out what is the best move to make at any given time, whilst sticking to the rules of the game.

Other than that I don't get what you mean.

[Bikewer]

What's our definition of logic in this regard? Chess involves analysing the positions of the pieces constantly, and considering the consequences of each move and also the future moves and responses of your opponents.
It involves planning, strategy, and tactics as well. One has to weigh the merits of losing valuable pieces in order to make even more valuable gains in position.

By all accounts, players at the master or grandmaster level are so involved with the game that they see the board in positional terms almost instantly, thus the "master plays 20 players at once" exhibitions.

Sounds pretty logical to me, as I understand the term.

Computers can do this analysis very rapidly, and compare all possiblilties with the parameters of the chess-playing program that's been written.

I would imagine that a program like Deep Blue could beat 98% of all chess players on a regular basis, and seems to play on a fairly equal basis with the top players in the world.

[chance]

On one chess forum I read, that humans play chess by pattern (position), the computer by brute force (calculating all possible variables, X amount of moves in advance). The logical conclusion, give more and more advances computers is that, a game will proceed thus:

White, (human) Pawn to K4
Black, Pawn to Q3 (“and mate in a maximum of 55 moves”)

And so on, with the moves to mate dropping by one for every ‘correct’ move, and more for every ‘incorrect’ move.

[Cleopatra]

Isn't it wrong to suggest that chess is a game of rules since by definition every game must have rules?

edited to add:

Chance:Exactly, pcs play chess by brute force indeed. I remember some discussions I have read in the FOM list. One of the authors of those messages was Harvey Friedman, if I recall correctly.

[Cleopatra]

I found the relevant discussions in the FOM list.


Chess as Science (http://www.cs.nyu.edu/pipermail/fom/2004-February/007913.html)

Is Chess ripe for foundational exposition/research? (http://www.cs.nyu.edu/pipermail/fom/2004-February/007910.html)

[Cleopatra]

I have been thinking about this thread since last night and browsing my books I attempted to figure out if is it possible to spot to games that could be played only by humans.

Waiting for Lux Ferum to return from Rome ( I know he is intersted in openings) this is what I think.

I believe that we can spot the difference between human and computer chess in certain types of positions.Please note that I have acquired such a program only recently so I am kind of experimenting here.

For example in the QGD Tarrasch-style where a pc plays " Like God" as Kasparov has said after being beaten by such a God and in closed, semiblocked ones like the Big Clamp Sicilian, or Colle-Stonewall, Pc3,d4, e3, f4 where the human simers a slow, long term plan, Ne5, Rf3, Rh3, maybe Bd2-e1-h4, bring over the other Knight... Kh1, Qh5, Rg1 or similar no direct threats, Black lingers a bit on the Queenside... and after 20 such moves White's concentration of forces is such that there is no defense.

I believe that such a broad-lines, long term plan is human chess, and computers can't handle it . I have started to believe that manufacturers skew the openings book to avoid entering into such programs' nightmares.

John van der Wiel won a model game a few years ago .

No tactical skirmishes, Pawn chain, a bit of space gain here, making a Bishop worse there..By move 40 he had a simple breakthrough and a calm win.

[MRC_Hans]

Ultimately, chess is algorithmic, since there is a finite number of possible games. But even ruling out obviously idiotic sequences, the number is astronomical, so in a practical game, intuition plays a geat role. Plus, in human to human games, tactics and bluff.

If/when the ultimate chess computer is built, I expect that it will be possible to find games that will always result in a draw.

Hans

[LuxFerum]

Just because the game can be well described in terms of mathematical equations, that doesn't mean that that is the way we see it.
In fact the way humans see the game is much more efficient than the way that computers do.
Sometimes we really analise the game almost in the same way the computer do, just checking the position in a few moves ahead. But even when we do that, we do it much more objectively than the computer.
We use concepts much more complex than the the computer, something like, pawn structure, domain of the center, pieces coordination, closed position, etc.
Those concepts reduce the amount of computation dramatically. That is why the brain that only compute +-3 moves per second is still able to beat a machine that compute billions of moves per second.

Computers still have trouble with long term strategy. That is why they use a library of openings and end games.
If you play an opening by the book, the computer will follow the exact moves of his library and will not wast any time in the opening.
That is why sometimes people use an uncommon Opening (made up) to play against the computer, that will make the computer lose an extra time in the beginning.
Other tactic is to play extremely closed positions, that way the long term strategy have a bigger advantage.

A position like this one:
http://img2.imgspot.com/u/04/187/14/blackplaywhitewin.gif
Black to play, white wins
Still require more than 1 hour for the computer to find the solution. (in a P3 1giga Hz with Fritz 7)
A lot of humans players that don't play near as good as the computer, are able to find the solution in less time.
The solution is not complex, it is simple, but it is long.

One big advantage that the computer have, is its cold blood. The computer will play very consistently all the game, even if he is under attack, or with one pawn less, or if is the final of a championship.
It won't even feel the difference. Is not uncommon to see players giving up in positions that later is found to be a draw. Even Kasparov gave up in a game with deep blue that was a technical draw. (http://www.research.ibm.com/deepblue/home/may07/news_1.html)

[Jyera]

Originally posted by The idea
Is chess, more than other board games, a game of logic?

It seems to me that chess is simply an extremely complicated and dynamic game based on a very simple set of rules. In other words, it is a game of rules rather than a game of logic.

Are there principles of logic that can be well illustrated by means of chess games?


Does chess, more that other game, improve logical thinking?
What area of logical thinking does playing chess improve?

[Cleopatra]

Originally posted by Jyera
Does chess, more that other game, improve logical thinking?
What area of logical thinking does playing chess improve?

Hello Jyera and welcome to the forum!

Is your question rhetorical? If not could you be more specific on how you use the term " logical thinking"? Do you mean critical thinking? Also, I do not understand what you mean by the " areas of logical thinking".

A year ago I started lurking in discussion lists of mathematical logic and I discovered that most of the top mathematicians are chess players, although one top player and publisher of chess books I know is a taxi driver in NYC ( ex-lawyer)....So, one might see a connection between a mathematically trained brain and chess.

In Cambridge the classicists were playing chess fanatically their style though is totally different to the style of the mathematicians.

Although my testimony is anecdological and by definition subjective I have wondered many times if some forms of education make a better chess player and if yes then why jews rule.

Does it have to do with the study of talmud which has a mathematical structure?

[Cleopatra]

I edited my post to add a couple of thoughts, sorry for the inconvience.

[Jyera]

Originally posted by Cleopatra

Is your question rhetorical? If not could you be more specific on how you use the term " logical thinking"? Do you mean critical thinking? Also, I do not understand what you mean by the " areas of logical thinking".


Hi Cleo,

The questions are not rhetorical.

I found it hard to understand the original questions posed by the originator of this thread. So I had attempted to ask another two questions in the hope of opening up the possibility of fruitful discussion. If it was not helpful perhaps you could treat it as rhetorical.

Let me try a second time...

1. Does Playing Chess help to makes a person more logical in the way he or she think?

2. Does "Principle of logic" refered to by the thread starter refers to logical deduction, induction or inference? Is there such things as "principle of logic"?

3. Does it improves the ability of a person to do logical deduction, induction or inference? I thought identifying areas of possible improvement is more meaningful.

4. What other areas of improvement under "logic" are there? So that when people listen to you reason/analyse with logic, they can notice that you have improved and regard you as more logical person than before.

- Jyera.

[The Cats Venm]

Compare chess and Tic Tac Toe. Both are games that involve no luck what-so-ever and have nothing subjective about the way they are played and won.

The only difference is complexity.

Theoretically, a computer playing chess will eventually be able to play a perfect game, to either a win or a draw. Every move could be potentially mapped out, and the best one chosen for any situation. If there is a way to beat the computer, it will be impossible not to given the right sequence of moves.

Chess, played by humans, works because no one can see the whole picture. Tic Tac Toe becomes pointless once the solution in known, only while there is the potential to make a mistake is it fun or challenging.


As for being a game of logic, the most logical way to play would be to find the perfect game and use it every time.

[drkitten]

Originally posted by Jyera



Does chess, more that other game, improve logical thinking?


This question has been studied empirically, and the answer seems to be "no." Teaching people chess doesn't appear to improve their abilities on logical reasoning, nor does logical reasoning appear to improve with practice at chess playing.

I believe there is a well-recognized link between logical abilities and skill at chess playing, but it appears to be interest related -- i.e. people who are good at logic seem to like chess more than average, and thus practice it until they get some degree of skill. But this is like the link between height and basketball playing: playing basketball won't make you tall, but if you're taller than average, you're more likely to be good at basketball, to play it regularly, and to enjoy playing it.

[T'ai Chi]

I find it fascinating that even after chess is conquered by computers, there will be xiangqi, Go, and other games that computers still have a hard time with.

I see these games as ways to of course have fun in human to human interactions, but also in making computers better.

I've enjoyed playing Fisherandom chess a lot.

[Jyera]

(About "Does Chess improves logical thinking?")

Originally posted by drkitten

This question has been studied empirically, and the answer seems to be "no." Teaching people chess doesn't appear to improve their abilities on logical reasoning, nor does logical reasoning appear to improve with practice at chess playing.


I would appreciate if you are able to point me to the empirical study.

I am not be surprise if the empirical studies show that it does not improve logical thinking. But there are certainly many ways to fail to improve logical thinking. Including playing without thought.

Originally posted by drkitten

I believe there is a well-recognized link between logical abilities and skill at chess playing, but it appears to be interest related -- i.e. people who are good at logic seem to like chess more than average, and thus practice it until they get some degree of skill. But this is like the link between height and basketball playing: playing basketball won't make you tall, but if you're taller than average, you're more likely to be good at basketball, to play it regularly, and to enjoy playing it.

I agree.

And to add, it is socially related. If everyone in your community plays chess, you'd pick it up under social influence.

This leads to the next question.
Would you then be more logical than if you were not in a chess playing community?

- Jyera

[Jyera]

After thinking ...
Here's an illustration/explanation.

I wanted to move my White Pawn to the square e4.
I checked that there is a Black Knight on square f6.
According to my knowledge of how a Knight could move, I DEDUCE that the Black Knight can capture my Pawn on square.

I could also DEDUCE that the opponent will likely capture my Pawn if I place it on e4. This is because I do not have any other piece that is able to provide a threat to capture on square e4.

I therefore decide not to move my Pawn to e4.

I think this shows that chess involve exercising deduction.
So ...

Chess improve skill of deduction.
Deduction is one of the logical reasoning process (deduction, inference, induction).
Therefore chess could improve logical reasoning (thinking).


- Jyera.

[Cleopatra]

Jyera, what you describe above is not necessarily the outcome of improved logical thinking but it can just demonstrate a better familiarity with the rules of the game along with the improved skills in playing the game that practice brings.

It's practice and studying of openings and previous games that make you a better player

[drkitten]

Originally posted by Jyera
After thinking ...
Here's an illustration/explanation.

I wanted to move my White Pawn to the square e4.
I checked that there is a Black Knight on square f6.
According to my knowledge of how a Knight could move, I DEDUCE that the Black Knight can capture my Pawn on square.

[

... [snip] ...

Except that this kind of deductive reasoning is not how people, expecially chess experts, appear to play chess. Instead, it's played as a pattern recognition game, where the expert looks at the board, "remembers" similar situations from previous games, and selects moves that are similar to the moves that worked in those previous situations. I'd describe it more as "analogical" reasoning than as "logical" reasoning.

What makes a chess expert an expert is precisely this ability, and the fact that s/he doesn't have to "deduce" anything about the board. This applies in much broader domains than just chess -- if I'm driving and a light turns red in front of me, I don't "deduce" that the red light means stop, "deduce" that, since the car I'm driving right now is a standard-shift, the brake is in the center, and then press on that pedal. No, I simply hit the brake, which at this point I can do WITHOUT need of deduction, since I more or less just "know" (read, "remember") how to react to a red light. I touch-type the same way; I can't tell you which letters are adjacent to the 'h' key, but I can type at 120wpm. My "knowledge" of the typewriter keyboard has passed beyond the stage of deductive logic and into a level of expertise where I can simply do the task at hand, without having to reason about it.

[Jyera]

Originally posted by Cleopatra
Jyera, what you describe above is not necessarily the outcome of improved logical thinking but it can just demonstrate a better familiarity with the rules of the game along with the improved skills in playing the game that practice brings.

It's practice and studying of openings and previous games that make you a better player

I agree that to remember the rules are a matter of memory.
But if it is the first time I place a Pawn on e4 which is threaten by a Knight on f6, I do think that I am forced to exercise my deductive skill in order to avoid capture. Even if it does exercise deductive skill, is it significant?

I can agree that practice and studying of openings and previous games make you a better player.

But I'm keen to explore how does it make me better?

Does it make me more logical and therefore I improve in the next game. Or does it make me more intuitive so that I win in the next game?

My objective is not to win games, but to find out how playing chess can benefit logical thinking, so that improvement in logical thinking the the chess player can be useful in other areas. Eg. doing well in logic in school.

- Jyera.

[Jyera]

Originally posted by drkitten
Except that this kind of deductive reasoning is not how people, expecially chess experts, appear to play chess. Instead, it's played as a pattern recognition game, where the expert looks at the board, "remembers" similar situations from previous games, and selects moves that are similar to the moves that worked in those previous situations. I'd describe it more as "analogical" reasoning than as "logical" reasoning. ...snipe...

drkitten, I agree with you.
But consider this.

Some people who teaches chess to young children touted that it improves logical thinking which helps with doing maths in school.

And it seemed to be a universally accepted concept that
"Playing chess help to make you more logical in your thinking."
(Sorry I don't have any stats)

Is that wrong?
What's wrong?

- Jyera

[drkitten]

Originally posted by Jyera


drkitten, I agree with you.
But consider this.

Some people who teaches chess to young children touted that it improves logical thinking which helps with doing maths in school.

And it seemed to be a universally accepted concept that
"Playing chess help to make you more logical in your thinking."


Some people also believe that sleeping with your bed aligned in a north-south direction will improve your health. All three beliefs appear to be incorrect when studied carefully.

There was also a common belief in the 19th and early 20th centuries that teaching schoolchildren classical languages (Latin and Greek) would improve their logical thinking. This also appears to be false.

The world is full of common but incorrect beliefs. Part of the task of scientists (in the case of much of the chess studies, the psychology department at Carnegie-Mellon University, I believe) is to sort through common beliefs and determine which ones are actually true.

[drkitten]

Originally posted by Jyera


I agree that to remember the rules are a matter of memory.
But if it is the first time I place a Pawn on e4 which is threaten by a Knight on f6, I do think that I am forced to exercise my deductive skill in order to avoid capture. Even if it does exercise deductive skill, is it significant?



Not really, no.



I can agree that practice and studying of openings and previous games make you a better player.

But I'm keen to explore how does it make me better?

Does it make me more logical and therefore I improve in the next game. Or does it make me more intuitive so that I win in the next game?



It appears to give you a better perceptive ability to see what is important and unimportant about a given position, and a better memory for similar positions from which you can select the best move.

[Jyera]

Originally posted by drkitten

quote:
--------------------------------------------------------------------------------
Originally posted by Jyera
I agree that to remember the rules are a matter of memory.
But if it is the first time I place a Pawn on e4 which is threaten by a Knight on f6, I do think that I am forced to exercise my deductive skill in order to avoid capture. Even if it does exercise deductive skill, is it significant?
--------------------------------------------------------------------------------
Not really, no.

It appears to give you a better perceptive ability to see what is important and unimportant about a given position, and a better memory for similar positions from which you can select the best move.

Perhaps it is not significant to adult or experienced players.

In the case of a young child, just developing his thinking skill, I think that (the deductive exercise) is significant?
It might be like a baby learning to take the first step in walking.
The first steps are often the most agonising but most important.

After some thought, I realized that even for adults can benefit. Provided they attempt to improve their chess skill. They have to attempt to improve by anaysing chess position explicitly and logically, instead of relying on intuitive "auto-pilot".

There are plenty of chess concepts beyond basic movement of pieces according to the rules.
Some as suggested by LuxFerum,
"pawn structure, domain of the center, pieces coordination, closed position, etc."
Others like Pin, Skewer, Fork, DiscoveryCheck, DoubleCheck.

Each of these concepts is like a new weapon in the hand of the player. Attempting to explore and wield these "weapon" ANALYTICALLY, challenges the player and develop new skill while exercising memory retention and deduction.

[Jyera]

Originally posted by drkitten

Some people also believe that sleeping with your bed aligned in a north-south direction will improve your health. All three beliefs appear to be incorrect when studied carefully.

There was also a common belief in the 19th and early 20th centuries that teaching schoolchildren classical languages (Latin and Greek) would improve their logical thinking. This also appears to be false.

The world is full of common but incorrect beliefs. Part of the task of scientists (in the case of much of the chess studies, the psychology department at Carnegie-Mellon University, I believe) is to sort through common beliefs and determine which ones are actually true.

In the face of incorrect belief ("chess improves logical thinking", what would be our advise and action?

Should we adopt an attitude to ban people from touting that chess is a good game that improves analytical skills and logical thinking?

[drkitten]

Originally posted by Jyera


Perhaps [...]

It might be like [....]

After some thought, I realized that [...]



There are many beliefs you can imagine ('realize') in the privacy of your own armchair. Do you have any evidence to support any of these beliefs?

I was (re)reading The Mouse on the Moon last night, in which the author comes up with a clever plot idea based on a previously unknown atomic particle with variable charge --- it can switch from positive to negative and back, thus allowing huge amounts of energy to be released from ordinary materials such as iron filings. This new particle, found only in expensive wine, can not only solve the world's energy problems, but provide a rocket fuel that would work for commuter flights to the moon.

There's only one problem. This book is fiction. All the actual evidence we have says that this particle doesn't exist.

Similarly, all the evidence we have is that playing chess does not improve logical thinking. Rather than trying to figure out how it might (but doesn't), you might try to figure out what chess playing does do.