Chess24 Jan Gustafsson on Alpha Zero

John Watson Book Reviews (103)

John Watson Book Review #103 - Challenging Conventional Wisdom

Move First Think Later.

Move First Think Later. |

John Watson reviews and discusses one of the most controversial and provocative instructional books in many years "Move First, Think Later" by Willy Hendriks. I (Mark Crowther) think the book is wonderful and have learned a lot from it. I'm not sure I agree it is for players of all strengths, I think a certain amount of ability is needed to get something out of it. As someone who has struggled with quite a lot of instructional literature over the years this book is a great tonic. I've found strong practical puzzle books as the only thing I've truly got on with. This book has given me many ideas on what to try next. I was pleased to read in the book there is a nice idea on how to use the TWIC games section by choosing random games and positions to produce an almost unlimited amount of exercise material. See John Watson's detailed review below.

John Watson Book Review #103 - Challenging Conventional Wisdom

Move First, Think Later

Move First, Think Later; Willy Hendriks; 256 pages; New in Chess 2012


I've been asked about Willy Hendriks' Move First, Think Later enough times (by people who loved or disliked it) that I thought I'd contribute my own two cents worth. This review turned into something frighteningly long, but that's in large part due to the fact that I have reproduced Hendriks' examples and prose at considerable length to give you a direct taste of the material.

First, let me attempt a general description of the book, which, incidentally, won the English Chess Federation's 2012 Book of the Year award. In Move First, Think Later (subtitled ' Sense and Nonsense in Improving Your Chess'), Hendriks tries to demolish a number of myths (in his view) about how to improve one's chess, as well as to investigate how the mind of a chessplayer works and what insights can be derived therefrom. In doing so, he refers to the cognitive sciences ("cognitive psychology, artificial intelligence, neurology, neurobiology, philosophy and others"), which he says have made "considerable progress." To quote from the Introduction: "From general concepts and theories we are moving towards knowledge on a more empirical and microscopic level, to summarise it briefly and (too) simply. Some of the old questions and new insights of the cognitive sciences form the source of inspiration for this book. Are they of any use for the player trying to improve his chess? Do they shed new light on our different training methods? Or even suggest new and different forms of training?" But what most people will remember from this book, and will likely be either amused or alarmed by, is Hendriks' acidic criticism of famous chess authors and of conventional chess wisdom. Much more on that below.

The book consists of 27 chapters, most of which are essentially essays covering a wide array of subjects. It has quite a few exercises, presented at the beginning of each chapter and then discussed within the chapter, which I think is an attractive and useful method of getting your points across. Hendriks' says, with his usual sense of humour: "If you do the exercises, you will learn the most from this book. Some may say: 'You will learn at least something.' "

The range of topics is impressive - Hendriks likes to jump here and there, touching upon subjects or random thoughts which are only tangentially related to chess. That's fine, even stimulating, and particularly good for browsing. In part, this flitting about may be due to the fact that some of the chapters were originally articles written for a chess column. Nevertheless, a good portion of the book's contents are related to a major unifying idea. That is expressed in various ways in various chapters, but to a large extent comes down to the idea that moves themselves are more important than verbal concepts in chess, and that methods for improvement that go from the general to the particular are flawed. He has several ways of saying this, e.g., "You learn to play good chess by taking in good chess. There is no way to outsmart a diligent student with some clever way of thinking. There is no short-cut route to the best move by some revolutionary way of looking at the position. The strongest players are not following secret protocols."

Hendriks Video Lecture from London Classic

Another version of this can be seen in a lecture by Hendriks at the London Chess Classic, which you can see on YouTube (

Improve without using language?

In it he says "In my book I have the suggestion that maybe it's possible to improve your chess without using any language." His point is that first looking at the characteristics of the position (which can be expressed verbally), and then finding moves based upon that general characterization, isn't getting to the essence of what produces quality play. Rather, he says, "moves are not only the outcome of some thinking process, they are very much the input, the starting point. It's not some clever thinking process that can help you find the best move in any position, but it's the enormous amount of knowledge that you bring to the board, good moves, mainly." Arthur van de Oudeweetering in ChessVibes describes Hendriks key idea succinctly when talking about an example: "That's exactly in accordance with his general recommendation: no rule beats a good move." [Actually a quote from the book MC but the review is well worth a look]

I'm admittedly extremely prejudiced in this regard, because it's something I've been saying for years, going back to my very first opening books, in which I put it another way: "The move is the idea". Years later, I attempted to expand upon and illustrate that interpretation in Secrets of Modern Chess Strategy and Chess Strategy in Action. One way of approaching this, as I said there, is to look at the similarity between learning chess and a child learning a language. In language, direct experience, imitation, and accumulation of knowledge is what matters most, not learning formal grammatical rules and then applying them.

But Hendriks' exposition is by no means limited to this concept. He discusses a large number of topics, including trial-and-error, pattern recognition, tactics versus strategy, time-trouble, planning, the illusion of general rules, the role of 'proverbs and maxims', ' free advice', 'critical moments', 'chance in chess' (I'm quoting chapter titles for some of these), blunder-checking, tactics, the opening and strategy, games collections, puzzles, etc. Whether or not the point hits home, he always retains a lively sense of humour and the clear goal of entertaining the reader. However, it's clear that he wants to be taken seriously, so I'll approach the material in that spirit.

Before moving ahead, I should also mention that Hendriks tries to apply modern psychological theories and neurological research to chess. Even though there's some overlap between that and his ideas about chess thought, I found these connections a little tenuous and by no means as compelling as the rest of his exposition. Of course, that may be merely a matter of taste or my lack of sufficient knowledge in the field.

Let me roam through the book and try to present a few of Hendriks' key ideas, criticisms, and thought-provoking claims. Right at the beginning, however, I have a fairly trivial complaint I mention only because the problem is so common in Continental chess books. The very first complete paragraphs in the book provide an example. They read:

"Playing chess can be confronting, and it sure helps if you can look with a smile at your own performances. I have known some players with a longing for perfectionism, who couldn't accept their shortcomings and quit playing. The term 'confrontation' in a sentence like 'playing chess confronts us with the working of our brain' seems a bit strange. But, although it's our own brain, we don't seem to have great access to it. This well-known fact is a major theme (problem) in the whole history of the philosophical and psychological investigations of our cognitive powers."

It's an interesting and thought-provoking introduction, typical of the book's foundational inquiry, but it's also bad English. I've never seen 'confronting' used as a modifier (I seriously doubt that it can be so used), and in any case I'm not quite sure what he means (not 'confrontational', obviously, but 'challenging'? or 'confounding'?). Whatever the case, Hendriks then states that the term 'confrontation' appears in a sentence in which it doesn't appear! Yes, that is indeed 'strange'. To be sure, I greatly admire anyone who can translate a book from his native language to another; it's not something I can do. But why not have a native English speaker (for example, someone from the U.S. or Great Britain), read the book through and point out obvious linguistic errors? Granted, it's a small point, and the prose is in general excellent; but periodically throughout the book, I found that the author's strange word choices got in the way of my reading experience.

Enough grousing. On to the contents. I'll look at the book's first few chapters fairly carefully because they contain the core philosophic issues that I am most interested in. I'll also discuss Hendrik's specific criticisms of chess authors, and point to just a few of his thoughts on other topics.

On conventional chess teaching

In the first chapter, Hendriks sets out his fundamental objection to the great majority of chess teaching by showing a sample of how he thinks a typical training session with a trainer and student proceeds, using this position:

Jaime Sunye Neto


Yuri Balashov

Position after 21...b5

"Trainer: 'You've had the chance to have a look at the position. What's it about, what are the most important characteristics of this position? Paul, do you have an idea?'

Paul: 'Uh, yes, I would play Rc6 and if he takes it I will have Nd5.'

Trainer: 'Yes, you come up with moves right away. But let's go back to the characteristics of the position, can you say something about them?'

Paul: 'Well, uh, Rc6 threatens to take on d6, I don't see what Black can do about it, if he takes, I take back and Nd5 is coming, what can he do then?' "

This is quite funny, and oh-so-true. Hendriks continues:

"Many chess books are written in the same pedantic tone the trainer is using here. They are based on the idea that you should not try out moves at random, but first take a good look at the characteristics of the position, try to make a more general plan on that basis and only then search for a concrete 'result' at the level of an actual move. This is nonsense. No chess player thinks like this, no one has learned to play chess by thinking like this and even trainers and authors of chess books don't think like this. In many books, however, this prospect is held up to us: if we only take a good look at the characteristics of the position, a good move will come flowing out almost automatically. These authors often forget that in fact they themselves do it the other way round: in the position they have selected to illustrate something, they already know the strongest move. Then they pretend that this move is a logical consequence of their description of the characteristics of the position, whereas they are only adapting those to the move they already know is strong..."

In my opinion the last two sentences are a bit overstated, but in any case, Hendriks has an excellent point which is hard to argue against. Soon thereafter he makes a simple and eloquent summation:

"The thing is: there is no order at all! We don't first judge the position and then look at moves. It all happens at the same time. The explanation for this is the following: you cannot have a meaningful characteristic of a position if it isn't connected with a (more or less) effective move. [emphasis his] We see characteristics of a position and the corresponding moves at the same time, since characteristics that are not connected with an effective move are simply not relevant. We don't see a weakness on f7 if we don't see (at the same time, or earlier) moves like Ng5 or Bxf7. "

Later he says: "Looking carefully at the characteristics, making a general plan, searching for a 'realisation' at move level: this is not the way we play chess, nor the way we learn to play chess. This order can be reversed just as easily: effective moves steer us towards the essence of the position." Hendriks consistently denigrates verbal advice, for example: "Closely related to, and often difficult to distinguish from the dogma of the respectable order, is an even more repellent notion that I would like to describe, somewhat solemnly, as 'the delusion of the lingual protocol'. Some trainers and chess book writers think that they can formulate all kinds of advice in words, and that this advice can then be applied in concrete positions and can help a player find the right move (the right plan). In other words, language in chess can be not only descriptive but also prescriptive. In chess manuals of this type, you often find first a lingual piece of 'advice' of a general character, followed by a (grandmaster) game serving as an 'example'. The suggestion is that what the grandmaster does is not much different than following up on the advice just given." Sorry to go on so long with these selections, but they are wonderfully apt and discerning. Hendriks then launches into one of his unforgiving attacks on chess authors. If you think the following is bad, it turns that even some famous chess authors are treated equally badly:

"An example of this can be found in Carsten Hansen's book Improve your Positional Chess, which is largely based on this delusion. Obviously, a lot of work has been put into this book, but this meaningless advice, delivered in a pedantic tone, makes the work hard to digest. Under the heading 'How to create a weakness' Hansen writes, among other things, the following: [In positions where the opponent has no weaknesses] 'you will have to look at the imbalances that exist on the board and see how you can use them to create a weakness in your opponent's position, either through provocation or through goal-oriented play where you see a way to establish a weakness.' As an 'example' he then gives the following fragment...."

At this point we see Hansen discussing a famous Kasparov game in which Black plays a remarkable and original positional idea based upon a clever tactic. You may recognise this position:

Garry Kasparov


Alexei Shirov

Position after 16.b3

There followed

16...g4! 17 f4 h4! 18 Be3 h3 19 g3 Nc6

with advantage to Black, largely because of the loose light squares in front of White's king.

Hendriks' point is that this singular and exceptional example isn't any help in answering Hansen's question of how to create weaknesses. "It is as if you write a manual on the art of painting, where you claim that with a number of well-aimed but sensitive brush strokes and a good idea of the eventual composition, you can create the finest paintings, and then below this you print a painting by Monet as an example."

Books About Chess Improvement

This rant is typical of other times when he addresses books about chess improvement: rather sarcastic, a bit mean, and nevertheless entertaining (as long as you aren't the target). All in the tradition of great Dutch chess journalism. That doesn't mean that Hendriks' criticism isn't basically correct, both about this and the other works mentioned below. I can hardly disagree, since I've often made the same argument, and Hendriks even credits me for thinking along the same lines. But I also feel that he's missing something, at least insofar as he seems to be condemning these books as a waste of time for the developing player. Hendriks' own investigation (in the chapters relevant to this topic) is aimed at the true nature of chess thinking and chess learning (hardly a small matter, of course, and fascinating). But there seems to be an implication that by getting that nature 'wrong', other books will influence students to think incorrectly and presumably retard their development. I don't believe this. For one thing, I'm not so sure that players are that malleable (after all, it's the trainer in the example above who is convinced by his own pedantry, not the student). More importantly, chess books have value beyond addressing those concerns. Teaching is a broader endeavour than telling the truth. The way in which a book has the most influence on a student's improvement is by keeping them excited about chess and inspiring them to study and learn more. Moreover, I think Hansen's is one of the better chess teaching books in this regard: it has great explanations of positions and themes, a clear style (I guess Hendriks and I will disagree about that one), and most importantly, exciting and/or deep examples for the student. The greater part of the book isn't concerned with telling players how to think (the passage above isn't typical), but with laying out the elements and theory of positional chess in a way that the developing player will understand. In fact, I often recommend this excellent book to students in the 1400-2000 range.

To be fair, Hendriks may even agree with what I've just said, and simply point out that this isn't the subject he is addressing, i.e., that his focus is upon whether or not books which claim to be instructional are telling the truth about the best ways to think about chess, or about how good players actually think, or about how to improve by organizing your thinking in certain ways. That's certainly a valid point, and his is a valuable contribution. I just want to be clear that the same books, even if literally 'wrong', can still be useful, even extremely useful, for a player trying to improve. Hold that thought.

Hendriks moves on to discuss the 'proverb-like pieces of advice' about chess that we grow up, and the problems with general principles. For example, he writes:

"Take the adage 'Meet a pawn push on the wing with action in the centre'. A chess book writer can give plenty of attractive examples to illustrate this. It would be difficult to do any statistical research on this subject matter, but to my mind, an adage like ' After an enemy pawn push on the wing, stay calm and don't do anything crazy' is just as valid. Or, if you like, just as meaningless..."

"Chess is not about the application of general principles that can well be expressed in words, on a slightly more subtle level. It takes place in another domain, where words are hopelessly inadequate...For the trainer, this means that the primacy is with the positions he discusses. Positions are not examples illustrating more general principles- they constitute the actual learning material! A way for a trainer to sin against this is by forcing positions into the straitjacket of a general principle." Again, this is insightful and I believe spot on. Still, my experience with beginners, developing players, and even fairly advanced amateurs, is that 'pretty lies' about general principles are often necessary to orient them in the midst of what seems a chaos of possibilities. That's why I explicitly made the point that my two strategy books were not meant to be teaching tools, but rather descriptive of modern chess. Perhaps a more modest application of Hendriks' insights would consist of advising chess teachers not to wait too long before weaning their students off of generalities and on to specifics.

Chapter 2 begins with an example of how prior knowledge of a manoeuvre is very often the reason we 'know' that it is good, as opposed to any logical train of thought leading us to the move. Hendriks then presents examples of how often the best thinking technique is trial and error, particularly when finding effective tactics in a position. I heartily agree; trial-and-error is often, paradoxically, more efficient than 'understanding' a position. And yet this is a technique which is very often criticized and even ridiculed, e.g., Hendriks disagrees with the author Cor van Wijgerden, who uses a 'step-by-step' method and calls trial-and-error 'Guessing and missing'. He point out that trial-and-error isn't the same as random target practice, because it depends upon the knowledge and experience of the player.

Hendriks then roughs up legendary chess author and teacher Jeremy Silman. It's amusing that these two come down so far apart, although by using Silman's 1999 The Amateur's Mind as his source, Hendriks misses out on the more complete and nuanced How to Reassess Your Chess, Fourth edition. The latter is Silman's masterpiece of chess instruction, aimed primarily at the 1400-2100 crowd, and deserves a place in every student's library. Here's Hendriks' characterization 'For Silman, the essence of looking at a position comes down to noting the main 'imbalances', which should bring the player onto the right track.' This is fair enough; compare Silman's own statement in HRC, 4th edition: 'I tossed out anything and everything that I felt distracted from the book's real purpose: mastering the imbalances and allowing them to guide you to the correct plans and moves in most positions. On top of that, I also integrated quite a bit of chess psychology into the lessons..." With regard to this last, it's interesting that Hendriks delves into the psychology of chess decision-making as deeply as anyone has, but not much into the psychology of the sporting element in chess, for example, he doesn't talk about how to become a tougher competitor. These things arguably have to do with chess improvement as well. That's not a criticism, just an observation.

Hendriks' objection to Silman's work has a familiar ring. In The Amateur's Mind, Silman gives a position in which he says the key imbalance is bishop-versus-knight.

Ron Gross


Jeremy Silman

White to play

It turns out that two of his students play poorly because, he says, they fail to recognize the imbalance; and other student, rated 1000, recognizes that the bishop should be retained (although playing poorly thereafter). Hendriks argues that for the improving player, "saddling him up with a 'bishop-knight imbalance' seems to be a bit over his head. If there is a constant danger of dropping both bishops and knights, then appreciating the difference between them seems to be a minor issue. But my main point here is that you need to have some idea of possible continuations to be able to estimate the value of the knight on b6. You need moves like a2-a4(-a5). Look and you will see versus trial-and-error and b2-b3 in your repertoire to make use of this bad knight or even to see that it's bad. Important characteristics combine with effective moves. If you have no clue about plans like a2-a4-a5 (maybe in combination with Bg5, to prepare Nd5) or of Black's possibilities of playing ...Nc4 or not, then the notion of the bad knight on b6 is meaningless. To Silman it will be appalling, but I think it works in the other direction: you see effective moves (or, as in the case of the knight on b6, the absence of promising moves) and then conclude what the most important 'imbalance' is. Quite on the contrary, Silman formulates his main motto like this: "Before you get carried away, let me remind you: DON'T look at individual moves! In fact, never calculate until you understand the basic components (imbalances) of the position.'' Here we have in a nutshell the misconception of the 'look and you will see' doctrine. I think for a chess player it is almost impossible to look at a position without looking at individual moves. In any case I would not advise you to do so, because you would be depriving yourself of a very effective way to get at the essence of the position."

Once again, I have to agree with Hendrik's specific objection. In many positions, for example, there are several major imbalances and it's beyond the ability of the top player, much less the amateur, to understand the proper weighting of each imbalance. But Silman's books are deservedly among the most popular and admired in the world, and very many players have stated that his works were directly responsible for their improvement. Why is that, if imbalances aren't at the heart of chess understanding? For one thing, I don't think you can underestimate how engaging Silman's writing is; while writing in a lively style, he finds typical and therefore instructive examples. A reader who is given something to focus on in a position, and is directed to its interesting features and challenges, will inevitably absorb a great deal of the kind of knowledge Hendriks himself values, namely pattern recognition and familiarity with the moves that 'go' with a position. No one will improve as quickly as an eager learner, and Silman's prose motivates and entertains like no other author's. For most people, learning without joy or inspiration isn't very productive.

In Chapter 3, Hendriks shows a variety of tactical motifs. This is in a sense one step removed from trial-and-error, because you are given a specific thing to look for in a position, for example, line-opening moves, or what he calls pieces 'looking through obstacles', before resorting to trial-and-error to find the way to exploit the motif. He gives an example of the latter from one of his pupil's games:


White to play

Here White played 25 Qxg4, gaining a pawn, instead of 25 Be4! dxe4 26 d5. Henriks explains that "My pupil did not really estimate the value of the white bishop on c3 working towards the king on h8. 'Seeing' this indirect attack is essential for starting to look for ways to get the pawn on d4 out of the way and finally finding the move Be4 (maybe after looking at moves like e2-e4 or c4-c5 with the same line-opening ideas) ." In a later game the same student utilizes a variant of this tactic.

One way to look at this is that it's a form of pattern recognition, and thus consistent with the most important asset a chessplayer has at his disposal. Hendriks might like that interpretation. But the other and more compelling one is that Hendriks would have his student do a 'general' look for the abstract condition of the bishop vis-à-vis the king on h8 before finding the specific move Be4, and that looks suspiciously like 'Think first, move later' to me! I personally think that pure tactics are not the best example of 'move first, think later'. In a majority of cases, pattern recognition and trial-and-error suffice; but in others a conscious verbally-motivated attempt to, say, look for the opponent's loose pieces (this is pretty handy!), or to look for pieces 'looking through obstacles', might well help to find the correct move.

Looking at direct attacking moves and doing blunderchecks are good examples. Hendriks isn't happy with systematizing these procedures, and in fact devotes some space to questioning their efficacy; but in fact they are in a different category of 'think first' than some of the too vague techniques he discusses. When you are told to look for imbalances, for example, Hendriks might say that there are too many imbalances and their importance is impossible to weigh accurately, so the technique is unworkable. I think that at least he can make that case (see below). But you can't say the same about, say, looking at every check, capture, and move that creates a threat to mate or win a piece. Doing so isn't an ambiguous task, and it would help to solve a very large number of the positions Henriks discusses. Look at this one, for example:

Leonid Starozhilov


Stanislav Bogdanovich

White to play

Hendriks says: " Essentially this is a very simple puzzle, but the key move 33 Qh1!, moving backwards with the queen, is a rare move. A few days ago I used this puzzle as a warming-up exercise in an internet session, but to my surprise my student needed a lot of time and hints to finally find the move. By the way, giving a hint here is difficult without letting on the solution straight away. Essentially, no useful remarks about this position can be made, except 33.Qh1!."

Well, I approve of the general point ('the move is the idea'), as well as the notion that you can't depend upon rules like 'always look at backward queen moves' without needing hundreds of such rules. But doing a quick check by mentally moving your pieces around to available squares and taking a quick look at any checks, captures, or forcing moves which arise takes very little time and significantly increases the chance of finding 33 Qh1, which after all threatens checkmate. Since there are only three non-trivial moves of this sort (dxc5, Bxd5 and Qh1) such a scanning of moves is realistic, even if it naturally isn't any guarantee that you'll recognize the right idea. In other words, one 'useful remark' would be that White should briefly look at all his moves to see if any of them threaten give check, capture something, or threaten mate. A very few of such modest and unsubtle thinking techniques can be useful, in my opinion, especially if there is no need for assessment attached. So I think that Hendriks' is correctly prejudiced against looking for 'forcing moves' in the broader sense Charles Hertan uses the term (see below), but that looking for directly forcing moves in the very narrow sense (as well as a simple blundercheck against moves by your opponent, before you pick up your piece to move) can be very useful and improve your results. But there are very few such techniques, and some best apply when it's time to sit back and clear your head anyway ('Are there any loose pieces?' and 'Can I improve the position of my worst piece?' are the only two I would recommend for such relaxed moments). By contrast, a 'checklist', even of 8 items (for example), probably clutters the mind too much and doesn't help in most cases anyway. Yet much longer checklists are given or implied (when you've finishe reading all their 'tips') by many chess authors.

On the other hand, I was shaking my head in agreement with Hendrik's assessment of Charles Hertan's 'forcing moves' technique, which I feel is relatively useless, especially the silly idea trying to locate the 'quiet forcing move' (not involving a check, capture, or direct attack, for example). After demonstrating a contradiction in one of Hertan's examples, he continues: "I'm not trying to find fault with everything, but this minor inaccuracy hints at what seems to be the central problem of Hertan's main concept: to determine what is the most forcing move (actually, what is the best forcing move, that is what the above example seems to point at) , you already have to take a good look at it. So we get caught in a circular mechanism: we are advised to analyse the most forcing moves first, but to determine which ones they are, we have to analyse them first! This is a real problem, since in most (tactical) positions, there are a great many moves that are more or less forcing. You might expect that Hertan would have some problems with tactical positions, in which the best move is not a violent, but a so-called quiet move. However, he overcomes this problem with the clever catch of the 'quiet forcing move' ! One of his examples is the following fine shot....

Teimour Radjabov


Ruslan Ponomariov

Position after 37.h4. Black to play.

In the manoeuvre ...Qd2-e2-f3-g2 mate, Radjabov found a very convincing way to end the game.

37 ... Qe2! 38.Qh5 g4 0-1

If you look at the position from the perspective of forcing moves, you'll get quite a list and may not end up with ....Qe2 at all. To name a few: . .Rf7, . . .Ng4, . . .g xh4, ...Nd1, ...Qxf2+, ...Qc2, ...Qxa2, ...Qd7, and, less likely but also very forcing: ...Qb2, ...Qxb4, ...Qc l +, ...Qdl+, ...Qe1+, ...Nf5, ...Nf1, ...Qd5 (yes, that one too) . It's funny that the computer's second choice, 37...g4 (with the same mating idea but preventing in advance White's queen coming to help via h5 or e6), will also not be on a list of forcing moves. If we allow quiet moves to be called forcing, then our list will grow even further. The set of forcing moves will become almost as large as the set of legal moves. So the whole concept seems too crude to guide you to the best moves: almost all moves that 'do something' can be called more or less forcing - and hell, there are a lot of them.

But we have to admire the clever marketeer: developing a new and flashy concept, coining [sic] all exclamation mark moves with it, adding a lot of enthusiasm, and the final result looks very convincing."


Hendriks criticizes the works of a variety of other authors, perhaps overzealously. But Andrew Soltis' The Wisest Things Ever Said About Chess is obviously not meant to be a serious guide to improvement, and I think Hendriks knows this; he uses the book not to criticize, but to reinforce his point about the inutility of words of wisdom as opposed to concrete moves. For example,

"Proverb 1 2 0 states: 'Don't worry about finding the best move. Seek always to find a good move.' A few pages later we read (proverb 1 2 4) 'If you see a good move, look for a better one.' Since these are as good as contradictory, they have no use as prescriptive advice. But with hindsight, they may be great commentary on a situation. Soltis' book is a very good read and at the same time proof that these 'wisest things' have very little practical value. The main reason for this, and also the main theme of this book, is the fact that in chess, as in real life I guess, you need to understand the gist of the situation. And the content in chess is so rich, so complicated and multi-dimensional, that all our written rules, maxims and protocols fall completely short in handling it. This is sad news for those who open a chess book hoping to have the 'rules', 'secrets' or 'fundamentals' explained to them and be supplied with a clear protocol to handle every position. But essentially it is good news, since if this were possible, our game would be rather poor and dull."

Pattern Recognition

Chapter 4, entitled 'Recognizing the Similar', has to do with pattern recognition (as do Chapters 5 and 6),. This is universally conceded to be an essential part (I think the most essential part) of mastering chess. But he takes it further: What are these patterns? Complete positions, parts of positions, big parts, small parts, moves, a series of moves taken together, parts plus moves? Concrete moves, abstract ideas, abstract parts, concrete parts plus abstract ideas? Or something else? In any case, to my mind it is misguided to identify these patterns as the characteristics of the position, on the basis of which some intellectual activity (searching) produces moves. Moves are as much 'pattern-like knowledge' as are the statics of a position. I think that throwing out 'successful' moves is the main activity of the chess playing department in our brain. Regardless of the position, randomly? Not completely, I guess, although I think we will all consider Bxf7+, at least briefly, even when that bishop is our only piece left and no Ng5 or any other follow-up is available."

On Knights and Bishops

Finally, I should comment upon Hendrik's discussion of bishop pair-versus-knight pair, since he disagrees with my own remarks in Secrets of Modern Chess Strategy, as follows: "But on the subject of the supposed supremacy of the pair of bishops over the pair of knights, I think he missed an important point.... Watson gives database statistics on 2 bishops versus 2 knights. I repeated his research (with rated players only) and got an overall figure of 58 % for the bishops (versus 42% for the knights) . This was on a total of more than 61,000 games, so that should give a reliable outcome. These numbers are roughly the same as Watson's findings. What Watson doesn't mention, or didn't take into account, is the rating difference: the players with the bishops were on average 36 points stronger! This is in itself remarkable, but it also has consequences. Do stronger players like the bishops better, or are they better at getting them? Food for thought, though more important is the following consequence: with this rating difference, the players with the bishop pair are supposed to score 55% ! If I can translate this directly to the overall result (objections can be made against this) , only a score of 5 3 % versus 4 7 % in favour of the bishops remains. Not without significance, but not shocking either."

As a fan of the knights, Hendriks goes on to try to explain away the 3% difference as well! But I'll stop there to make two points. First, he's right that my analysis was not particularly strict or well thought out (I'd completely redo it today), and that I certainly should have used performance rating rather than raw percentages. Guilty as charged. But in reality, I think the two bishops' inherent advantage is much greater than his or even my figures. That's because experienced players, and especially strong ones, tend to allow their opponents the bishop pair versus their knight pair mainly when they have a serious compensating advantage, e.g., outposts, a big lead in development, a clearly superior pawn structure, and/or more space. I can't prove that statement without further research, but if you think about the various openings in which one side voluntarily takes on the knights, you'll see what I mean. So on a neutral playing field (other positional and tactical factors being equal), I would suggest that the bishop pair has a large advantage indeed. Of course, that's still only an opinion until someone does the grunt work of examining large numbers of positions with this material imbalance.


I've finally come to a stopping point. There are another 21 chapters, full of discussions both serious and light-hearted, and covering a wide variety of subjects including those I mentioned at the beginning of this review. As always with New in Chess, the production quality is excellent. Regardless of what you think of Hendriks' style and philosophy, or of his opinions of other authors' works, this is a fun and absorbing read. It won't surprise you that I recommend it to players of all strengths.

PDF sample from the book

There is a PDF sample from the book available

New in Chess Magazine 2018 issue 1

Chess and Bridge Shop Titled Tuesday

Modern Chess Learn Chess without a coach

ChessBase Ad 5 MyGames

American Chess Magazine 5

Ginger GM - Chess Grandmaster Simon Williams

Contact Mark Crowther (TWIC) if you wish to advertise here.

The Week in Chess Magazine

Send a £30 donation via Paypal and contact me via email (Email Mark Crowther - I'll send you an address for a cbv file of my personal copy of every issue of the games in one database. Over 2 million games.

Read about 20 years of TWIC.

Read about issue 1200.

TWIC 1215 19th February 2018 - 3658 games

Read TWIC 1215

Download TWIC 1215 PGN

Download TWIC 1215 ChessBase