I will be reviewing one of the most influential books I had early on in my chess development. This is an excellent book by grandmaster John Nunn and Peter Griffiths.
Secrets of Grandmaster Play is at once a best games collection (all of the games are played by GM Nunn) and a treatise on how to analyze. The book contains 24 games, all won by Nunn, with extremely deep analysis. In some cases, the analysis of one move takes up a page or two. However, this is not the kind of long variations that you might be used to. Rather than series of moves with no explanation other than an evaluation at the end, Nunn and Griffiths explain the thinking (either over the board or in later analysis) that went into the creation of the variation. Not only the "best" line - as determined by thorough analysis - is given, but also all the stops on the way to reaching it. Thus you can see the actual thought process, rather than just the results.
In addition to variations, there are a large number of general assessments and explanations of positions with plain language. The student won't be buried in soulless variations. For the first thing, the variations are not soulless when they are illuminated by real words; but also the book is as much about understanding (in particular, the understanding of dynamics) as it is about calculation.
The games cover a wide spectrum. There are a few absurdly wild games (in particular, the Nunn-Anthony game and the first couple of games in the book, as well as Nunn's "immortal game" against Belyavsky). But there are also a fair number of quiet games, such as his games against Christiansen and Andersson.
Where I got it
I don't know, but I do know that I had it for a really long time. I remember analyzing some of my early games and writing them up in the same style as Nunn and Griffiths do. In fact, in the 9th grade (when I was 14) I had an teacher who was at least somewhat interested in chess. We had to write a long paper and I wrote mine on chess, including several games annotated in the same way as Nunn. The games were from my second or third tournament (in Alaska there weren't so many tournaments). No doubt the annotations were completely stupid, but I think I got an "A" anyway. It would be interesting to see that paper now, but I am sure it is long since gone away.
What's good about it
I think I answered this somewhat above. Now let me quote from the introduction:
"In presenting two dozen of John Nunn's best games, and in trying to explain his thought processes as clearly as possible, I should make the point at once that it is not our intention to give the reader an easy time! This book is designed for a wide range of players who wish to improve their chess and are willing to put in some hard work to achieve that end. One of the main aims has been to highlight the differences in approach between a Grandmaster and a weaker player, and to try to narrow the gap. To some extent this comes down to technical matters - more accurate analysis, superior opening knowledge, better endgame technique ad so forth; but in other respects the difference goes deeper and many readers will find that they need to rethink much of their basic attitude to the game. One example of this would be the tremendous emphasis which is placed on the dynamic use of the pieces, if necessary at the expense of the pawn structure, or even of material. This is no mere question of style; it is a characteristic of the games of all the strongest players."
Rather than try to teach the basics of chess in an abstract form, this book is more of glimpse into a top player's mind. In particular, it illustrates how freedom of dogma is necessary to play well. It also shows that chess cannot play superficially. I do think there is a fundamental difference in the way players of different levels think. Such differences in level between a 1400 and a grandmaster are not explained simply by superior knowledge or technique, or even by superior calculation. There should be something fundamentally different in the approach to chess, and this book attempts to bring it to light.
How it impacted me
I think that this book helped me to see that there was much more beyond the most superficial level of chess. In particular, the major emphasis of the book is on dynamic play. Since a lot of basic chess books focus on static features (especially the old books which I was reading) this book would be "part 2" of understanding strategy. Additionally, I surely learned a lot about the concept of compensation from this book. There are a great many positional sacrifices made, and the authors explain the concept in detail. For example, here is an excerpt from page 49, his game against Britton:
"And here is the kind of positional sacrifice which seems so natural and logical when played by a Grandmaster. We can all appreciate that Black is getting plenty of compensation for the exchange, yet we hesitate to take such decisions ourselves. Why? For two reasons, it seems. The first is that although material is only one element in chess (the others being time, space, development, pawn structure etc.) it is the tangible one; and the same player who will throw tempi around with appalling abandon, or neglect his development, will shrink from sacrificing even a pawn unless he can see the consequences with absolute clarity. But Grandmaster have a highly developed sense that material is just one factor in the equation; so in their games it is constantly being interchanged with the other elements. It is not a question of their being more daring than other players, or more prepared to risk losing. To a GM there is no danger whatever in this type of sacrifice; it is simply a matter of technique, a transaction. In this connection I might add that players who are generally regarded as extremely hard to beat tend to specialise in the art of positional sacrifice (Petrosian and Andersson spring to mind).
The second reason is what Dr. Lasker calls 'the certainty of having to apply yourself vigorously' (after sacrificing, that is). In other words: "Its all very well for Nunn to do that, but if I tried it I would soon go wrong and then lose the endgame". Well, it's true that after sacrificing you have to play with a certain amount of vigour. But you have to do that anyway! If you want to stand up to really strong opponents, that is. Look at it this way: in allowing his pawns to be doubled at move 14 Nunn has already made a sacrifice of a kind (his pawn structure), and without vigorous play to follow it up he would have let the white bishop into c4. He also sacrificed at move four (development) and any hesitation after that would have led to trouble. He is always 'sacrificing'; you cannot play this game without doing it."
I have already given two excerpts above, and including a full annotated game would be too long. So I will a short excerpt from one part of his game against Mikhail Tal, on page 104:
"With this move the most instructive positional part of the game begins. Obviously control of the centre is important on all occasions, but here its effect is particularly striking. The advance 12.g4 must be seen as a positional maneuver, not the start of a mating attack. It may eventually result in an attack, but White cannot possible generate any real threats against the king while his knights are far away and while only his queen can participate quickly. For the moment the intention is to gain space and drive the black knight from f6, thus augmenting White's control of the centre.
But we need a little more justification than that for 12.g4. Suppose Black tries to break open the centre - the standard response to the threat of a flank attack? In fact a closer look reveals that White is effectively restraining both ...e5 and ...d5 here (and even ...b5, which might affect the centre by driving the knight from c3). We shall have more to say about this matter in a moment; but there is one final point: if the loss of tempo in playing 6.g3 and then 12 g4 seems hard to accept, remember that the fianchetto development (compared to the commoner Be2) has already gained time for White in the sense that ...d5 has always been prevented. Therefore White does not now have to waste another move to achieve that aim.
So all these things justify 12.g4; but no more than that can be claimed. Black's game is still quite sound and it is only his ensuing mistake which leads him into such trouble.
Tal observes that 13.g5 would be a first-rate positional blunder, ruining everything for White, because 13...hxg5 14.fxg5 Nh7, followed by ...Ne5, would give Black a rock-solid position on the K-side (control of the centre square e5!); but he doesn't notice that g5 can be prepared by h4. He should at all cost have prevented any further expansion by playing 12...Nd7!, after which White would have nothing better than to continue with straightforward development: 13.Be3 b6 14.Qe2 Bb7, for example, with a double-edged middle-game and approximately equal chances.
It seems that at least one of the watching Grandmasters was inclined to laugh at this! But Nunn knows exactly what he is doing: the further pawn advance is not only logical, but even strategically decisive, because from now on Black will be unable to obtain any decent play for his pieces. After g5, which cannot be averted, ...hxg5 can now be met by hxg5, without yielding the point e5. In fact it would be most unwise for Black to exchange pawns: with the opening of the h-file his king really would be in trouble (Qh5, then Rf3-h3, or Kf2 and Rh1). In a moment, then, Black will have a choice of knight retreats: to e8 where it will interfere with his rooks and leave his K-side exposed (13...Rfd8 14.g5 Ne8 would expose it even more), or to h7, defending the K-side, but leaving the knight permanently marooned. Tal decides that ...Nh7 is the lesser evil.
Perhaps you are still worried because White's row of pawns looks unwieldy? How is he going to control events when the real fighting begins? What if Black sacrifices a pawn to break up the centre? That has happened frequently enough in such positions; in the resulting open game the white king could be seriously harassed down the diagonals.
Well, Nunn makes the interesting comment that he hardly calculated a single variation more than a couple of moves deep during the entire game. And here surely lies the secret. He does not need to analyse in great detail. In a moment Black will indeed make the pawn sacrifice ...d5; and White, with his pieces established in the centre, will then maintain tactical control by working out a series of short variations. If he starts out with a firm grip on the centre, he never need permit the black pieces any dangerous activity. At move 13 he will have concentrated on Black's pawn breaks to make sure no mischief is possible, namely (a) 13...e5? 14.f5 (or even g5 at once) followed by g5 and Nd5, and Black is hopelessly weak and cramped; (b) 13...d5 14.exd5 Nxd5 15.Nxd5 exd5 16.g5, winning a pawn safely. The actual choice, 13...b5, is not a sacrifice (14.axb5 axb5 15.Rxa8 Rxa6 16.Nxb5 Qb6+ and 17...e5 would be disastrous); so White just pushes the knight back and develops."
One thing I could say is that the book is definitely verbose. That may be seen as a downside or an advantage, depending on your point of view. The authors definitely do a lot of telling, rather than showing. By contrast, a book I reviewed earlier, Endgame Strategy by Mikhail Shereshevsky has very terse explanations.
Considering that this is a glimpse into the mind of a grandmaster, there is always the question of what he saw at the board versus later analysis. The book does not contain only Nunn's thoughts at the board, but also many hours of later analysis. On the other hand, Nunn does not try - as many players do - to exaggerate what he saw at the board. Frequently the authors point out what was seen at the board and what was later analysis. Also the introduction deals with this, warning the reader that they should not imagine that a Grandmaster could analyze to the depths that are in the book. He follows this by saying "I shudder to think what would happen if readers laboured under that illusion and tried to play accordingly!" Nevertheless, there is still a lot of unclarity about this.
What level of player should read this book?
At first glance the book is very intimidating. Whole pages devoted to one move, long complicated variations, extremely crazy games...you would think it were only for very advanced players. But looking at it more closely, you see that - while the games are very complex - the points being emphasized and illustrated are very simple in most cases. I think this is the result of how the book was written - Peter Griffiths is a vastly weaker player than Nunn and is a professional chess teacher. Nunn analyzed his games deeply and the explained them to Griffiths, who put the explanations in his own words. Thus we have the grandmaster's thoughts and the 2200-rated teacher's explanations. Somehow it works.
If you look at the above excerpt, you can see this. The authors emphasize fairly basic concepts - that 12.g4 is not, contrary to appearances, a direct attack on the king, but rather an maneuver to fight for the center by dislodging the knight from f6; that White must not play 13.g5 and yield the e5 square (something any average tournament player would know); that control of the center is crucial in allowing White to absorb Black's pawn sacrifice.
All in all, I think the book would be best for players from 1200-2200. However, that player must be dedicated and want to improve in chess. Because the book isn't really for casual reading.
What you should eat/drink while reading this book
English tea and scones. In order that the tea doesn't leave a ring on your table, and also because John Nunn - since stopping competitive play - is a three-time world champion of problem solving, you should probably use a table coaster with chess problems on it, made by Caissa Commerce (picture below).