When I open a book written by a former world champion, audaciously titled THE SYSTEM, I expect to encounter some high-class stuff. In this case, however, Dr. Berliner's remedy comes across as more snake oil than chess elixir.

For those unfamiliar with the author, Dr. Berliner is a renowned authority on computers and chess. He finished fifth in the 1957 U.S. championship and won the 5th world correspondence chess championship, which included a famous victory over the Soviet theoretician Estrin. Berliner also designed, built, and programmed the computer chess program Hitech, which was arguably the best computer chess program of the 1980s.

This book seeks to demonstrate a scientific approach to chess and a system for playing the openings that, in the examples cited, lead to a demonstrable advantage for white. The author seeks to explain the concepts behind this system, which stresses board control, with a discussion of each of the pieces and their role in the system.

I found these introductory discussions to be generally simplistic. We learn the same things that most basic texts teach us – knights are best in blocked positions and on advanced squares, bishops are bad when blocked by their own pawns, rooks increase in value as pawns are exchanged. There's little here that most players haven't read before.

Berliner does attempt to embellish his system with some more original material, although even here most players will have been exposed to the concepts before. He explains the importance of keeping options open, not prematurely resolving tension (i.e., the threat is stronger than the execution), not unnecessarily allowing transpositions, etc.

While there are some useful discussions, Berliner often resorts to trying to explain concepts in an overly numerical fashion. Thus, we are given a new evaluation of the pieces that suggest that pawns are worth 1.0, knights 3.2, bishops 3.33, rooks 5.1, and queens 8.8. While this may (or may not) be a more accurate evaluation of the value of the pieces, just what does it mean to the average player that his minor pieces are 0.13 greater in value than the opponent's? Likewise, we are given tables of the values of unpassed pawns in the opening and endings, which range from 0.90 for second rank pawns on the a- and h-files to 0.95 on the b- and g-files, 1.05 for the c- and f-files, and 1.10 for the d- and e-files. There are many others. All I can ask is, “Is it legal for me to bring a calculator to the board for my next game?”

The crux of Berliner's book is his belief that he can, through use of his system principles, prove the best opening move and prove an advantage for white in several mainstream openings. This is quite a claim, and I would expect some pretty deep and probing analysis to prove these facts. Alas, it is not to be, and it leads me to take the bulk of the book with a large grain of NaCl.

Berliner argues that 1.d4 is clearly the best first move. While there is some recent practice to suggest that this may be the case (NEW IN CHESS YEARBOOK suggests that its database of high level games shows that 1.d4 performs better than any other first move), the evidence from practice is far from conclusive. Indeed, some of Berliner's proof seems pretty flimsy. 

For example, Berliner largely dismisses 1.e4 because after 1...e5 2.Nf3 doesn't keep white's options open (it blocks the f-pawn, which is necessary for white to fight for board control). As a result, Berliner concludes that 2.f4 would be the only logical "system" move. He then, without any analysis, concludes that, "Since 2.f4 is not feasible, it is likely that 1.e4 is wrong." First, Berliner provides no analysis to show why 2.f4 is wrong; rather he relies on offhand comments from Fischer (who played 2.f4 from time to time) and the conclusions of Weaver Adams  (I'm not making this up) to back this up.  Where is the rigorous systematic proof for this? Indeed, it's interesting that NEW IN CHESS suggests that the reason that 1.e4 doesn't score as well as 1.d4 is not related to 1...e5, rather it is the move 1...c5. Of course, 2.f4 would be a perfectly logical "system" approach to this defense – except it hasn't performed very well of late after 2...d5 3.exd5 Nf6.

There has been much written of late about adopting a rigorous analytical approach to the openings. John Watson's influential book SECRETS OF MODERN CHESS STRATEGY: ADVANCES SINCE NIMZOWITSCH (click to see SILMAN'S REVIEW) suggests such a perspective. Berliner, on the other hand, argues for adherence to the system even when analysis might suggest otherwise. For example, after 1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 cxd4 Berliner writes that "do not now play 5.Qa4+ and follow with Qxd4. 5.Qxd4 is the move you want to play, and it should be played immediately without giving Black any further developing tempi." That's fine, but if you go beyond the simplistic view that the queen check gives black an extra developing tempo you will find that the tempo is not particularly useful and must be returned to properly develop the interposing piece. Meanwhile, the check allows white to avoid certain black lines. Given that set of circumstances, what is the point of "the system?"

That, however, is mere quibbling compared to my concerns about some of Berliner's analysis of specific lines to back up his belief that he has refuted several popular defenses to 1.d4. For example, regular practitioners of the Grunfeld Defense (1.d4 Nf6 2.c4 g6 3.Nc3 d5), which include world class players like Kasparov, Anand, Shirov and Leko, should be interested in Berliner's view that the defense is busted after 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 c5 7.Bc4 Bg7 8.Ne2 Nc6 9.Be3 0-0 10.Rc1. This was, in fact, played a bit in the 1980s, but it has been determined that after 10... cxd4 11.cxd4 Qa5+ 12.Kf1 Qa3! black has decent prospects. Black's move frees a5 for the knight to move to c4, and it keeps pressure on the Be3, which prevents f3 freeing a square for the king. Its point is shown after white's normal thrust 13.h4 Bg4, and 13.Rc3 Qd6 14.f4 e5! blasts away at the white center. 13.Qb3 allows black to either trade the queens with fair chances or play 13...Qd6 when 14.Bd5 Na5 15.Qb5 (Atalik-Rotsagov, 1997) 15…Qd8! would have given black good chances according to Rowson in another recent Gambit book UNDERSTANDING THE GRUNFELD (click to see SILMAN'S REVIEW).

If this were the only case of quibbling, of course, one could suggest that the theoretical jury is still out here. However, that is generally the case in each of the openings Berliner would cite as proved to be advantageous for white. For example, the author also claims that the King's Indian should be met by the "system" approach with the Samisch Variation (1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.f3). It is true that this is a respected main line, but many white players have been put off it because of problems proving any advantage in the endgame after 5...0-0 6.Be3 c5 7.dxc5 dxc5 8.Qxd8 Rxd8 9.Bxc5 Nc6.  Indeed, Berliner concedes this by suggesting that white should meet 6...c5 with 7.d5.  Unfortunately, this transposes into lines of the Benoni where white's move order with an early f3 makes his chosen development more difficult to achieve. This, of course, is the sort of analytic nuance that doesn't get discussed in this book. Indeed, Berliner's suggestion that the King's Indian player look to the chapter on the Benoni will shed no light on this specific move order. That player will find, for example, no mention of the lines mentioned in NUNN'S CHESS OPENINGS (click to see WATSON'S REVIEW) or Norwood's THE MODERN BENONI, which generally lead to satisfactory positions for black.

Even Berliner suggests that several black defenses to 1.d4 have not yet led to his expected advantage for white. He suggests that the Slav and the Nimzo-Indian Defense are among those defenses. Given the problems noted above (and there are others), I would suggest that there really is very little to the system that can be taken as science.

One should not, of course, entirely dismiss Berliner's findings. He has come up with many important theoretical findings in the past (including the recognition that black had nothing to fear after 1.e4 Nf6 2.e5 Nd5 3.d4 d6 4.Nf3 g6 5.Ng5, which Bobby Fischer most likely took note of before his world championship match with Spassky). Some of his systems, such as favoring f3 against the King's Indian and Benoni complexes, have been adopted with success by world-class players like Seirawan. Still, I doubt that Yasser would claim the theoretical certainty of the advantageous of these systems for white that Berliner claims.

In the end, chess is still an unsolved game. While Berliner is a noted scientist and chess player, I do not believe that he has sustained his theory in this book. I doubt that the average player will get enough out of this book to justify its close study.

If you want another opinion, click to see SILMAN'S REVIEW of THE SYSTEM.