Chess for Zebras – a Starbucks-Gambit publication

6/30/2007 – What is the best environment for a chess grandmaster to write an instructive book? The cafes in South-West London, and Starbucks in Barnes. "I find I work best when surrounded by benign-looking people who don’t want to talk to me," says GM Jonathan Rowson, who produced the very successful Gambit book "Chess for Zebras". Jonathan gives us a frank insight into the creative writing process.

ChessBase 14 Download ChessBase 14 Download

Everyone uses ChessBase, from the World Champion to the amateur next door. Start your personal success story with ChessBase 14 and enjoy your chess even more!

Along with the ChessBase 14 program you can access the Live Database of 8 million games, and receive three months of free ChesssBase Account Premium membership and all of our online apps! Have a look today!


Chess for Zebras

By GM Jonathan Rowson

Why Zebras? “If you are reading this book, you are very unlikely to be a Zebra, if only because Zebras can’t read. Nor can they play chess. In the absence of literate chess-playing zebras, you may therefore be wondering whether my publishers misjudged their market when they accepted my proposal to write this book.”

So goes the first paragraph of Chess for Zebras, which was released towards the end of 2005. Thankfully the English edition has done very well over the last 18 months, both critically and commercially, so Gambit’s gamble paid off. Nonetheless, back cover blurbs never give the whole story of what goes in to writing a book, so I am grateful to have the chance to give some background for those who haven’t yet overcome their fear of zebras, or who have just noticed the new German edition, Schach fűr Zebras, and are wondering whether to buy it for their friends in Sallmannshausen.

The title may remind you of the late Simon Webb’s Chess for Tigers, and many have indeed seen my title as a kind of ironic twist on that popular classic. They are not far wrong, for while Chess for Tigers is a ‘feel-good’ book in title and content, Chess for Zebras is more like a ‘feel-different’ book that is supposed to knock you off balance slightly, and the content, though hopefully reader-friendly, should make you think more deeply about the game, and the complexity of chess improvement. Indeed the book’s title subtly exemplifies the main message, which can be summed up in one line: “Improvement begins at the edge of your comfort zone”. (Note that it is the edge for a reason. I don’t want to wrench you away from all that is fine and familiar, but rather to keep you on that edge, enjoying the sensation, neither bored nor bewildered, but supported and challenged.)

The author of Chess for Zebras: GM Jonathan Rowson

My original motivation was actually to write a book about White’s alleged first move advantage. I agreed with Adorjan’s basic ‘Black is OK!’ ideas (though, I hasten to add, without the messianic zeal) and felt I could make a more sophisticated case for the first move advantage being some sort of illusion, perpetuated by linguistic conventions and a misreading of statistics. This was early in 2002, it was cold outside, and there was no prospect of significant income in the foreseeable future, so I just had to think of a suitable title for the endeavour and I wouldn’t have to worry about how to pay the rent for a while.

A good friend suggested a cover with Arnold Schwarzenegger in Terminator mode saying: “I’ll be Black”, a slight modification of his immortal line, but I feared this may increase racial tensions in California, so I needed something else. It occurred to me that you couldn’t write a book about Black without making extensive reference to White, so I felt something quintessentially Black and White would be ideal. Chess for Badgers felt too much like Wind in the Willows, but Chess for Zebras seemed to have the necessary allure, and Gambit agreed, on the condition that it had a sensible subtitle.

What’s it about?

A few months later life took a new direction, and I went to study at Harvard for a one year masters degree, on their ‘Mind, Brain and Education’ programme. I learned a great deal about the neuroscience and psychology of learning, and much of it resonated with my own understanding of the challenges of improving at chess. I became particularly interested in the issue of players becoming stuck at a certain age and level, and unable to budge despite their best efforts. This is a curious phenomenon, because in most domains you become progressively better with experience, but with chess we typically see rapid growth in youth, followed by what might be called ‘cerebro-sclerosis’- a hardening of the categories. We become stuck by our own ideas and habits, and all the new information we take in is filtered through a limiting set of assumptions.

Over the course of time, I also found that I was playing better, which I attributed to breaking down a lot of the preconceptions that held me back when I was younger. Indeed by the time I started writing the book properly I was already close to 2600 (ridiculously close in fact) and British champion, so I felt I was on to something worth sharing. The difficulty was that I now had about three books inside me, but one outstanding contract, with a peculiar title, had to be upheld, so I started looking for more information about zebras to see if I could find a unifying theme.

In early 2005 I was lucky to find a book in a second-hand shop with the title: “When you hear hoof beats, think of a zebra”. This is a Sufi saying to encourage mindfulness of the dangers of natural assumptions (hoof beats = horses). The relevance of this saying is captured by the sub-title: ‘Thinking Differently about Black and White’. ‘Black and White’ refers not only to the first move issue (part three of the book), but also conventional wisdom about chess improvement (part one) and chess in general (part two).

How and Why was it written?

Most of the writing of this book happened in cafes in South-West London, Café Nero and Starbucks in Barnes in particular, because I find I work best when surrounded by benign-looking people who don’t want to talk to me. The photo below gives some idea of my usual writing habitat, but some of the more serious chess analysis took place at home.

Vaio and Starbucks – the working environment of a chess author

The problem with most chess books, as I see it, is that they could have been written by anybody. The author frequently disappears in analysis and positional platitudes, and when he or she does appear it is with a hackneyed joke, usually followed by an exclamation mark, in case you missed it. Since the lifeblood of chess is moves and variations, and most readers care primarily about improving their results, the brute facts of the game inevitably clip the wings of even the most adventurous author. Nonetheless, I have always felt that if you are going to allow trees to be cut down in your name, you should make a contribution that is as distinctive and personal as possible.

Central ideas in the book include: the difference between knowledge and skill and why we should place more emphasis on improving our skill, the important relationship between chess understanding and storytelling, how to cultivate concentration, why chess seems so difficult, why we struggle to understand non-material factors when evaluating positions, what stops us from defending better, and whether there is any basis to the claim that White starts the game with some advantage. These themes, and more, are illustrated with games from players of a variety of strengths, and include lessons I have learned from my own games with players like Aronian, Adams, Morozevich, Short and Miles, and conversations with luminaries like Dr. Hübner, Dvoretsky and Yusupov.

 The following excerpt comes from a chapter designed to show why defending can be fun:

Gdanski – Arkell
European Clubs Cup, Neum 2000

The situation looks pretty grim for Black. He is a pawn down and White seems to have good control of the position. Given two moves – say Kg2 and Nd4 – Black’s position will become hopeless. However, at this very moment Black has a surprising resource and it begins with one of the most extraordinary moves I have ever seen.


I feel I should give the reader a few quiet moments to make sense of this move before adding any thoughts of my own. 33...g5!! illustrates, I think, the core reason why Kotov’s canonical method of calculation, in which we start our thinking by looking for candidate moves, is flawed. 33...g5!! could never be an original candidate move because it only begins to make sense after looking at the position and finding that the more conventional approaches are not working, and then gaining some insight into why they are not working. Indeed, although ingenious and highly impressive, the move is actually quite logical once you start trying to solve Black’s problems. You only have to examine the more obvious (though still quite creative) attempt to free Black’s position 33...c5!? 34 Nxc5 Bxc5 35 Qxc5 Qd1+. Now you find that there is a check and you might also see the idea of meeting 36 Kg2 with 36...Nf4+ followed by ...Qg4+ if White takes. Then your heart sinks when you see that there is no perpetual due to 36 Kh2. But then, while some would give up on the whole idea of ...c5, Keith thought to himself: if only there were a way of checking the king on h2, and he managed to reject the move 33...c5 but hold on to the idea. Black needs a check on the h-file for the check on d1 to become a perpetual, and therefore has to get rid of the pawn on h4. Hence 33...g5!!.

It’s hard to say what it takes to come up with a move like this. Even after this explanation, you might still be left with the feeling that you could never find the hidden idea. In this sense 33...g5!! is a good example of the relevance of the hindsight/foresight distinction introduced in Chapter 1. Keith’s idea of playing for perpetual check directly from the original position is a remarkable feat of conjuring, and something that would not occur to the vast majority of players during practical play. Yet such moves often spring from situations such as these where your determination to defend as tenaciously as possible gives you the glimmer of an idea, and then you just need to tweak it slightly, and get a little cooperation from the opponent to make it work.

34 hxg5

Apparently just after Keith played 33...g5, Gdanski, who was rated 2557 when the game was played, looked bemused and just whipped the pawn off. Taking is the best move, but he should have treated 33...g5 with a little more respect.



Now liquidation is a serious threat because most of the 4 vs 3 positions will give Black excellent drawing chances. Even if they don’t look too pleasant for Black, they are much more pleasant than the position before 33...g5, so White has more reason than Black to be upset.

35 Nxc5?

This was not the game continuation but I wanted to make the main idea prominent. White didn’t fall into the trap quite so directly. He now saw Black’s devious idea and deviated from this main line, but Black drew without problems in any case: 35 Kg2 cxb4 36 axb4 Qa2! 37 Bc1 Bxb4 38 Qc8+ Kg7 39 Nd4 Bc3 40 Nxe6+ (after 40 Nf3, 40...Qa1! is the only move to keep the balance; 40...Qe2? 41 g6! and 40...Bb4? 41 g6! are both favourable for White) 40...fxe6 41 Qd7+ Kh8 42 Qe8+ Ó-Ó.

35 Bc3!? is probably best, exchanging the relatively ineffectual bishop. Then White’s winning chances and Black’s drawing chances seem about equal.

35...Bxc5 36 Qxc5 Qd1+ 37 Kg2 Nf4+!

The hidden detail: Black forces a draw. Curiously, when I showed the original position to a student, Gordon Rattray, he saw 33...g5 after a little prompting but rejected it, because in the process of visualizing the resulting variations he made a persistent miscalculation. He saw this far, but couldn’t see how to deal with the fact that after ...Qg4-h5+, the white king will have the g3-square. In fact, White will not have the g3-square, because Black’s queen is not limited to light squares and can also give check on h4.

This is another good example of what happens when we confuse ideas with variations. If you frame the idea of perpetual check in terms of ...Qd1-g4-h5+ you might run into problems because your perceptual system might filter out moves like ...Qg4-h4+ as a ‘separate idea’, not associated with the perpetual. This is partly a problem with the linear thinking. If we are not careful, one idea follows another and our thoughts will be determined in advance, locked in motion by the way we began our thinking process.

38 Kh2 Qh5+ 39 Kg1 Qd1+ 40 Kh2 Qh5+. Black draws by perpetual check. A truly brilliant defensive conception from Arkell.

Who am I?

I am a Grandmaster from Scotland, living in London and married to Siva from South India. I am at ease with being thirty, but still haven’t figured out what to do with my life and increasingly wonder if I ever will. I studied at Oxford and Harvard and am currently a PhD student at Bristol University with an interest in the concept of wisdom from the perspective of enactive cognition, but I suspect I am too intellectually flirtatious to ever be a serious academic. I write a weekly chess column for The Herald in Scotland, a regular review column for New in Chess magazine, and occasionally try to write about something other than chess, with moderate success. I do a fair amount of private tuition in London, mainly with young players, but it was my experience of teaching adult players that inspired the first section of Chess for Zebras.

Chess-wise, I consider myself a semi-professional, which is a way of encouraging people to pay me for playing while excusing myself from the theoretical grind that typifies more dedicated Grandmasters. I consistently tease myself with the prospect of being 2600, but tend to drop a few points just in time for the next rating list to keep the creative frustration alive.

Discussion and Feedback Join the public discussion or submit your feedback to the editors


Rules for reader comments


Not registered yet? Register