THE SYSTEM
Author: Hans Berliner
Gambit Publications Ltd (1999)
176 pages
$24.95

Reviewed by John Watson

I can’t blame Gambit for signing up Hans Berliner to write THE SYSTEM, a book which purports to advance the state of chess theory, and “to go a long ways towards solving” the game itself. Berliner, after all, is a former world correspondence champion and one of the most successful of the early computer chess programmers. One might safely assume that he has something to contribute. Nevertheless, THE SYSTEM turns out to be a lazily written collection of dogmatic claims, banalities, and awful analysis, colored throughout by an arrogance that fails to disguise the book’s ultimate shallowness. In fact, I have seldom seen a book with fewer redeeming qualities, and wouldn’t bother to review it at all if it weren’t for a few issues. First, any book that makes such sweeping claims should have those claims addressed. Second, many people have sent me email requesting me to assess the book; it apparently has a kind of seductive charm. Lastly, most of the reviews to date have been far too generous to Berliner, in my opinion, crediting him with originality he doesn’t possess and a scientific approach he doesn’t follow. More importantly, they seem to overlook the book’s fundamental failure to prove its thesis: that after 1.d4, White can achieve a winning advantage; in fact, as I will indicate, Berliner doesn’t succeed in making progress against a single Black defense!

For reasons of space, I’m going to concentrate on the only testable part of Berliner’s work: the analysis. I won’t bore you with his assertions of superior insight, e.g., his condescending treatment of Bronstein, Lasker, and Fischer, or his statement that his theory “is as real as the theories of Gravitation and Evolution.” Nor will I critique the few attempts he makes to describe what “The System” actually is; it doesn’t take very long to realize that there is no “system”, but rather, a few stereotyped and timeworn rules with no clue as to how to apply them or assess their relative weight, and so many gaps that one feels that only Berliner himself could identify a “System move”, and then merely by fiat! Even in the rare case in which he tries to make a concrete theoretical point, e.g., when he assigns the pieces his own set of values, the attempt is marred by omissions. In this case, Larry Kaufman’s recent article in Chess Life did a much better job of assigning such values, since Kaufman takes into account the fact the pieces values vary depending upon what other material remains on the board.

The most important claim of THE SYSTEM is that, with correct play, White can win by force after his first move. That move, according to Berliner, “must be” 1.d4. Because he continually asserts that his is a “scientific” system (although something less scientific can hardly be imagined), I want to subject his theory to a scientific criterion, i.e., is there any verifiable evidence supporting its validity? Since the only such evidence Berliner submits is opening analysis, let’s look at that as deeply as a book review allows. I should say in advance that I have looked at every variation which Berliner advocates, and while I can’t fit in all my analysis here, my conclusion is that not a single system he suggests wins for White, and most allow easy equality or better chances for Black.

Berliner starts with the Queen’s Gambit Declined, beginning 1.d4 d5 2.c4. Here, by the way, 2.Nf3 is “wrong” (as are all other second moves), a fact that Bobby Fischer “learned” from Berliner but, mysteriously, “Bobby didn’t do anything with this knowledge.” Too bad, he might have turned into a strong player. Anyway, after 2.c4 e6 3.Nc3, Berliner doesn’t even discuss 3...Be7, when his favorite plan of playing for e3 and f3 is no longer appropriate; certainly he needs to show some plan against this preferred move-order of several world champions. Moving on, after 3...Nf6 4.cxd5 exd5, Berliner says that, “Black has a very bad position!” He continues 5.Bg5 c6 (Berliner has Black playing this now or on the next move; thus, we never know how he “wins” against other Black setups, e.g., lines with ...Nbd7, ...Bb4 and ...c5) 6.e3 Be7. At this point, Berliner doesn’t mention 6...Bf5, but he does give 6...h6 7.Bh4 Bf5 8.Qf3 Bg6 9.Bxf6 Qxf6 10.Qxf6 gxf6 11.Nge2 “with a wonderful endgame for White.” Well, the equivalent ending after 6...Bf5 7.Qf3 Bg6 8.Bxf6 etc. is not the sort of thing Black rushes into (he has few winning chances). But Berliner is claiming a forced win for White, and having discussed this position over the years with GMs and IMs, I can tell you that they are not at all sure of its assessment (at least one GM and one IM think that it’s a draw with best play). If White doesn’t win here, “The System” would be invalidated right off! Needless to say, Berliner should show a way for White to play this position and obtain a clear advantage, but he doesn’t begin to do so.

Okay, after 6.e3 Be7, Berliner gives the standard main line 7.Qc2 0-0 8.Bd3 Nbd7 9.Nge2 Re8 10.0-0 Nf8 11.f3, saying that “the best reply” is 11...Ng6 “!” (he doesn’t look at 11...Nh5, and I think 11...g6!? is underrated, having played against it) 12.e4 dxe4 13.fxe4 Be6! (no analysis of 13...Ng4) 14.Bxf6! Bxf6 15.e5 Be7 16.Rad1 Bf8, when White stands “considerably better”, although he fails to suggest a move here. In this line, 15...Bg5 is certainly of interest, but the main point is that Berliner is essentially just presenting established theory in a line many world-class players are willing to enter as Black. Are these positions really winning for White? I play them myself for White, and I don’t think so. This is a typical Berliner ploy: he follows established variations at less depth than conventional theory goes, skipping various important defenses. Then he takes positions which theory calls equal or slightly better for White and assesses them as clearly better for White, adducing no evidence! This is neither “scientific” nor particularly honest.

Amazingly, that turns out to the Berliner’s best chapter. From here on, it’s all downhill. In the next chapter, Berliner’s anti-Grunfeld suggestion is taken from Polugayevsky (later refined by Shirov): 1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 c5 7.Bc4 (Berliner dismisses 7.Nf3 outright due to 7...Bg7 8.Be3 Nc6 9.Bc4 Qa5, not even bothering with the main line 8.Rb1 or, for that matter, with 7.Be3 and 8.Qd2) 7...Bg7 8.Ne2 Nc6 9.Be3 0-0 10.Rc1. Berliner says that after 10.0-0, “Black can just about force a draw” by 10...cxd4 11.cxd4 Na5 12.Bd3 Nc6, not even looking at 13.Bb5, for example. Instead, he calls 10.Rc1 “a winning continuation”. Nearly every other reviewer of this book has already pointed out that 10...cxd4 11.cxd4 Qa5+ 12.Kf1 Qa3! (discovered by Ilya Gurevich) put 10.Rc1 out of commission some years back. Berliner fails to mention 12...Qa3 (or other good ideas for Black), and I would only add that when I analyzed these lines in depth with IM Larry Remlinger (who played 10.Rc1 and, like every other strong player, gave it up), we found it hard for White to even equalize, much less try for an advantage. Even if you could “patch up” White’s position after 12...Qa3, the whole variation simply fails to give serious chances for an advantage.

Next, Berliner tackles the King’s Indian Defense. After 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6, he explains that 5.Nf3 and 5.Be2 “can’t be right”, 5.Bg5, 5.Bd3 and 5.Nge2 are mistaken, and 5.g3 and 5.h3 are both “ridiculous”. Wow, if only the world’s top players knew about this! Then every KID would begin with Berliner’s “refutation”, 5.f3 0-0 6.Be3. Okay, let’s play along. 5.f3 and 6.Be3 went completely out of fashion at the highest levels when Black (led by Kasparov) discovered the gambit line 6...c5. Mountains of theory on this move still indicates that Black is doing fine here. What does Berliner say about this, the current main line? Absolutely nothing! Hmm. Just out of curiosity, I went to check 6...Nbd7, which I’ve been studying recently and which I think ultimately equalizes. Again, nothing! Now there’s a “System”! You refute a system by ignoring all the standard solutions. What does Berliner consider? He barely mentions 6...e5 7.Nge2 c6 8.Qd2, when the Kasparov favorite 8…Nbd7 9.0-0-0 a6 10.Kb1 b5 is dismissed in one line: “11.Nc1 exd4 12.Bxd4 leading to an excellent position for White!” Amazing. Again, theory barely starts at this point, and its further course indicates equality for Black. Given Kasparov’s many wins for Black here (Kramnik did manage one draw against him), one wonders if the System is an attempt to equalize as White?! And how about 6...Nc6? Now there’s a line White can make a good stab at refuting (I know; I wrote a book on it); but Berliner can only come up with the feeble 7.Rb1 (to which he appends “!!”). Well, for once it’s an original idea, but he doesn’t bother to analyze any obvious sequences at all, e.g., he gives 7...Bd7 8.b4 and says “Now, in order to avoid being swept away, Black is practically forced to play 8...e5 9.d5 Nd4 10.Nge2...” But Black doesn’t play ...e5 and ...Nd4 in this line before White comments to Bd3 or until White has moved his king’s knight a few times, e.g., from e2 to c1. A 6...Nc6 player would avoid 8...e5 and play something like 8...a6, meeting 9.Bd3 with 9...e5 10.d5 Nd4, 9.b5 with 9...axb5 10.cxb5 Na5, and 9.Nge2, perhaps, by 9...b5 (9...e6 is also interesting) 10.cxb5 axb5 11.d5 Ne5 12.Nd4 e6! with excellent compensation.

Is there ANY Black defense to 1.d4 that The System can win against? Berliner next tackles the Semi-Tarrasch: 1.d4 d5 2.c4 e6 3.Nc3 Nf6 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 c5 7.Nf3. After four pages of analysis almost completely irrelevant to the main lines, Berliner admits that his approach fails for White, and says “it may well be that the correct System move is not 7.Nf3, but instead 7.Rb1,” and stops there! Four pages of irrelevant analysis, and nothing on 7.Rb1! Plunging on, the Tarrasch Defense is next in line, with 1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 exd5 5.Nf3 Nc6 6.g3 Nf6 7.Bg2 Be7 8.0-0 0-0 9.Bg5 cxd4 10.Nxd4 h6, when Berliner calls 11.Be3 “illogical”, and explains that after 11.Bf4, “White can continue with Qa4 followed by Rfd1 obtaining an overwhelming position. I can find no real defense against this procedure.” Gee, how come Petrosian (versus Spassky) and Belyavsky, Smyslov, and Karpov (versus Kasparov) missed this in their matches, opting for that illogical move instead? World Championship history might have been altered forever! Of course, Berliner has again simply picked a well-known move and announced an assessment; it’s amazing that he didn’t just look up the published analysis on 11.Bf4 (including Kasparov’s own), nor look at the hundreds of master games. To call this “overwhelming” is just silly; in fact, I agree with theory that Black can equalize.

The Queen’s Gambit Accepted is next. Berliner recommends 1.d4 d5 2.c4 dxc4 3.e4. As usual, he refers only superficially to the rich praxis of the past tens years concerning that move. For example, in the topical line 3.e4 e5 4.Nf3 exd4 5.Bxc4 Nc6 6.0-0 Be6 7.Bxe6 fxe6 8.Qb3 Qd7 9.Qxb7 Rb8 10.Qa6 Nf6 11.Nbd2 Bb4 12.Nc4 0-0 13.a3, Berliner gives only 13...Be7 for Black, when the main areas of contention has been Anand’s 13...Bc5, and, for example, 14.b4 Nxe4! of van Wely-Sutovsky, Wijk aan Zee 1997, which favored Black. This is typical Berliner: suggest a well-known variation, cut the variation off way before theory does, and simply declare a clear White advantage where no one else does. The rest of the 3.e4 QGA is worse still: he doesn’t even mention the critical lines 3...Nf6 or 3...c5, and after 3...Nc6 4.Nf3 Bg4 5.Bxc4 e6 6.d5 exd5 7.Bxd5, he omits the main move 7...Qf6.

But surely he can find something against the much-maligned Benoni? How about the Taimanov System with 7.f4 and 8.Bb5+? Now that might actually gain an advantage. Instead, Berliner sticks with a well-established line, considered harmless by theory, involving e4/Bd3/Nge2/0-0 and (usually) f3. Of course, it’s hard to verify that this “refutes” the Benoni, since Berliner only cites one game with a defense even he admits is bad. Once again, he skips most of Black’s normal defenses (five of which end in “=” or “unclear” in ECO), and offers no analysis to counter Black’s most popular response.

Berliner extends his streak of failures with his Benko Gambit solution. He gives 3...b5 a “?!” and sweepingly rejects most of White’s fourth moves, settling upon 4.cxb5 a6 5.f3 “!” as a refutation. I’ve played 5.f3 for years (in fact, I’ve played several of Berliner’s recommended lines, which is one reason I know that they don’t lead to the clear advantages he claims), and I’m convinced that White can maintain only a small edge in several variations, whereas Black may well equalize in others. What happens when we consult Berliner about those lines? As usual, half of Black’s normal moves are missing entirely. Worse, in the critical 5...e6 variation, Berliner skips the important 6.e4 c4!? entirely, and after 6.e4 exd5, he opts for the punchless 7.exd5, well-known to give White nothing. All of us who play 5.f3 have looked hopefully into 7.exd5, hoping for a simple way to avoid the mega-mess following 7.e5, but any serious analyst can see the shortcomings of White’s pawn structure. Typically, after 6.e4 exd5 7.exd5, Berliner doesn’t consider Black’s main move, 7...Bd6! (he only analyses the well-known error 7...Qe7+?; also missing is 7...d6, which I believe equalizes), after which theory finds Black doing quite well, e.g., 8.Qe2+ Kf8 9.Nc3 cxb5 and 10.Nxb5 Ba6 or 10.Be3 c4!. And yet Berliner blithely concludes: “If the aggressive 5...e6 fails, as it apparently does, then Black has noting better than 5...axb5 6.e4 Qa5+....”, going on to do a butcher job on that familiar line as well.

On and on it goes. Berliner admits that, having “solved” the above Black defenses, he hasn’t quite “found any refutation of the Nimzo-Indian Defense, but [he is] close.” How thrilling. Similarly, he’s not sure if he’s finished off the Slav Defense yet (in Berliner’s world, these admissions qualify as modesty), but he’s leaning towards 1.d4 d5 2.c4 c6 3.Nc3 Nf6 (no need to address 3...e6, by the way) 4.cxd5 cxd5 5.f3 “!”. At least this is Berliner’s own idea, but the move has serious positional drawbacks, and there is more than one Slav line in which Black intentionally provokes f3 by ...Bg4 and then retreats with ...Bc8 or ...Bd7. Anyway, after 5.f3, Black has several good moves, but a very simple one is 5...e6, when White has a poor variation of the Exchange Slav (due to the f3 move) unless he tries 6.e4 dxe4 7.fxe4, when 7...Bb4 puts a lot of pressure on White’s center; I suspect that White has nothing or less than nothing in that case. Alternatively, Berliner himself says that 5.f3 Nc6 6.e4 dxe4 7.d5 Ne5 8.fxe4 e6 “yields no concrete advantage for White”.

Okay, you get the idea. So why would anyone write such a book? One might suppose that Berliner just enjoys making grandiose claims for moves he’s barely looked at. But no: again and again, he claims to have put “hundreds of hours” into these positions, supplemented by computer analysis. And his exposition of the System seems to date back to at least some newspaper columns written in the late 1960s or early 70s, so there is no reason to doubt his assertion that this book represents a lifetime of effort. No, I’m afraid that Berliner is simply a lazy analyst with huge gaps in his chess understanding, and all his bragging about his advanced ideas and “solving chess” is simply laughable in the face of the actual work he presents us. If it weren’t for his overbearing style, deceptive presentation, and obvious disrespect for the great chessplayers past and present, I might just write off Berliner’s book as nonsense. As it is, I think that it is a careless and dishonest effort that represents the very worst in chess writing.

For collectors of chess kitsch, or for those postal players that will blindly stand by Berliner come hell or high water, click to buy (or get more information about) THE SYSTEM