Anyone for the Franco-Benoni?

Why play a main line defence that your opponent may know well beyond move 20 when you may be able to surprise him at move two with <1 e4 e6 2 d4 c5!?>

Black's chief expectation is that White will reply 3 d5, leading to Benoni-like central pawn structures, which is why if this defence has any accepted name, it's the Franco-Benoni.

It was played against me in the preliminary round of the 5th European Team Correspondence Championship, which obliged me to research it in depth.

I soon found that authors writing on the French Defence do not consider 2...c5 part of their brief; you won't find it, for example, in "The Complete French" by Lev Psakhis. Benoni books do not bother with it either, for White need not play c2-c4 and if the e-file is opened (which often happens) the central pawn structure is quite different. The line can also arise via 1 e4 c5 2 d4 e6 but again there is little help to be had. In his 1981 book "The Morra (Smith) Gambit", Hungarian IM Janos Flesch did give a brief discussion of the line, pointing out the trap 3 c3 d5 4 exd5? Qxd5! and saying that <3 c3 d5 4 e5> is a French Defence, Advanced Variation: that's true, but Black has alternatives at move three, perhaps <3...cxd4>.

This is uncharted territory, as is the possibility <3 c4 cxd4 4 Qxd4 Nf6 5 Nc3 Nc6 6 Qd1 Bb4> (Dluski-Drazic, Cattolica open 1993). White's main choice is between <3 Nf3>, consenting to an open Sicilian (although <3...a6>, for example, has been seen), or trying to take advantage by <3 d5>. Flesch wrote: "3 d5!.... ensures a healthy space advantage for White. Black cannot hope for a Benoni type counterplay, for White puts a piece and not a pawn on c4." However, he gave no concrete examples to support this argument and 3 d5 gives a kind of middle game that may be unwelcome to many 1 e4 players. In fact it is a form of Old Benoni (also reachable 1 d4 c5 2 d5 e6 3 e4) and is considered briefly under the heading A43 in ECO (the "Encyclopaedia of Chess Openings").

Similar material is to be found in the second edition of Kasparov and Keene's "Batsford Chess Openings" (BCO2). Such sources as do consider the variation generally imply a slight edge to White with 3 d5. I amassed about thirty games, the majority of which went 3...exd5 (3...Nf6, 3...d6 and even 3...e5?! have also been played.) <4 exd5 d6 5 Nc3 Nf6 6 Nf3> or reached the same position by a slightly different route, but in "Informator 61" (published December 1994) there was a game fragment with a new move.

G.Timoshchenko-Zelcic, Passau 1994, had gone <1 d4 c5 2 d5 e6 3 e4 a6 4 Nf3 d6 5 a4 Ne7 6 Be2 exd5 7 exd5 Ng6 8 0-0> (8 h4!? Timoshchenko) <8...Be7 9 Na3> (9 Nc3!? Timoshchenko) <9...Bg4 10 Nd2 Bxe2 11 Qxe2 0-0 12 Nac4 Nd7 13 a5 Re8 14 Ra3 Bg5 15 Qd1 Nde5!> with an unclear position. G Timoscenko vs R Zelcic, 1994

The main discussion that I could find which did not sneer at the Franco Benoni was by American masters Joel Benjamin and Eric Schiller in their "Unorthodox Openings" book. They had found an improvement for Black in one of the principal games cited by ECO. Subsequently I found a 120-page Chess Digest booklet by Andrew Soltis entitled "Franco-Benoni Defense", but it didn't contain any games previously unknown to me in the 1 e4 e6 2 d4 c5 line; at least half the monograph was devoted to lines like 1 d4 e6 2 c4 c5 which normally won't transpose at all to the line that interests us here, that is the 1 e4 variety.

Move order in the Franco Benoni seems flexible, which makes its study even harder, and after <1 e4 e6 2 d4 c5 3 d5 exd5 4 exd5 d6>

click for larger view

Opening Explorer

it was not easy to find grounds to choose between the different knight developments. If White wants to retain the option of c2-c4 then of course he must choose Nf3.

Anyway, my game (against correspondence international master Xaver Steiner) began 1 e4 e6 2 d4 c5 3 d5 exd5 4 exd5 d6 and now I actually chose <5 Nc3>.

However, the second diagram position after <5...Nf6 6 Nf3 Be7 7 Be2 0-0 8 0-0> is just as likely to arise if White develops the king's knight first.

click for larger view

Opening Explorer

The main book line runs <8...Na6>, intending Nc7 before....b6, but <8...Bg4> also comes into consideration. Then 9 Re1 Nc7 10 a4 b6 11 Rb1 (surely not best?) 11...Re8 12 h3 h6 13 Bc4 and now "ECO" gives 13...Bf8 14 Rxe8 Qxe8 15 Be3 which is at least slightly in White's favour (Vaganian-L.Bronstein, Sao Paulo 1977 Vaganian vs L Bronstein, 1977). However, Schiller and Benjamin reveal <13...a6!> which they consider, probably rightly, to be a very significant improvement for Black. The game Bohm-Konikowski, Dortmund II 1981, continued 14 b4 Bd7 15 bxc5 bxc5 16 Qd3 (If 16 a5, with the idea N-a4-b6, then 16...Nb5!) 16...Qc8 17 Bf4 Bf5! 18 Qd2 Bf8 = and Black even won in the end.

Quite a few games featured White playing Bb5+ instead of Be2. This check is sometimes effective in the Modern Benoni but that is usually because it involves variations with e4 e5, not available here. The case for Bb5+ hardly seemed compelling to me; the bishop was liable to be exchanged for a knight or driven back with loss of time.

Danish super-GM Bent Larsen had played this line twice with Black with few problems: 1 d4 e6 2 e4 c5 3 d5 exd5 4 exd5 d6 5 Nf3 Nf6 6 Nc3 Be7 7 Bb5+!? Nbd7 8 a4 0-0 9 0-0 when O'Kelly-Larsen, Palma de Mallorca 1967 went 9...a6 10 Be2 b6 11 Re1 Re8 12 Bf4 Nf8 13 h3 Ng6 = while 9...Re8 10 Re1 a6 11 Bf1 b6 12 b3 Nf8 13 Bb2 Bb7 14 Bc4 Ng6 = was Gligoric-Larsen, Busum 1969.
( O'Kelly vs Larsen, 1967 Gligoric vs Larsen, 1969 )

A more recent grandmaster game Tseitlin-Kuzmin, Cappelle la Grande 1994, had gone similarly until after 9 0-0 a6 10 Be2 Re8 11 Re1 b6 12 h3 Bf8 (instead of Nf8) but this game too was a draw in the end. "BCO2" cites a game going 5 Bb5+ Nd7 6 Nf3 Ngf6 7 0-0 Be7 8 Re1 a6 9 Bf1 0-0 10 a4 b6 11 c4 (Bohm-Johansen, Amsterdam 1983) but that light-squared bishop does not seem to have much of a future here.

The most promising game for White that I could find was Van Wely-Meulders, Brussels zonal 1993: 1 d4 e6 2 e4 c5 3 d5 exd5 4 exd5 d6 5 Nc3 Nf6 6 Nf3 Be7 7 Be2 0-0 8 0-0 Na6 9 h3 Nc7 10 a4 Rb8 (If 10...b6 11 Re1 Bb7 12 Bc4 a6 13 Nh4 Re8 14 Nf5 with advantage in D.Gurevich-Kavalek, USA Champ 1984.) 11 Bf4 b6 12 Bb5! (Only now that c6 is available) 12...a6 (If 12...Nxb5 White could choose between a-file play after 13 axb5 or piece pressure by 13 Nxb5 Qd7 14 Re1.) 13 Bc6! Bd7 14 Re1 Re8 15 Qd2 Bxc6 16 dxc6 (Now we see why White wanted his bishop on f4.) 16...Ne6 17 Rad1 Rc8 18 Bxd6 Rxc6 19 Bg3 with persistent pressure despite the symmetrical pawn structure and piece exchanges. Van Wely vs R Meulders, 1993

If Black prefers 8...b6 from the diagram then I could take as my model the "Informator 48" game Chandler-Mik.Tseitlin, Palma de Mallorca 1989: 9 Re1 Na6 10 Bb5! (threatening Qe2) 10...Nb8 (Not 10...Nc7 11 Bc6 Rb8 12 Qe2+ ) 11 Qe2!? (11 a4 Chandler) 11...a6 12 Qxe7 Qxe7 13 Rxe7 axb5 14 Bg5 Nbd7 15 Nxb5 Nxd5 16 Nxd6! and White won in 45 moves. Chandler vs M Tseitlin, 1989

However, a further surprise was in store for me. My very experienced Swiss opponent met 5 Nc3 by <5...a6!?>, a move I had never seen before. It seemed a good idea to maintain White's space advantage by <6 a4> and postpone major decisions until Black revealed his plans for his king's bishop.

The game then continued <6...Nd7 7 Nf3 g6> at which point I noticed that we had transposed to a 1983 postal game,
Wiedenhofer-Kerinnis, which went 8 Bf4 f6 (ugly but effective; 8...Ndf6!? also comes into consideration.) 9 Bd3 Nh6 10 0-0 (10 h4!?) 10...Nf7 11 Re1+ Nde5 12 Bf1 Bg7 13 a5 0-0 14 Nd2 b5 15 axb6 Qxb6 16 Ra2 Bf5 and Black even went on to win; his firm control of e5 and the lack of an f2-f4 advance negated White's efforts to exploit the weaknesses around e6 and f7. I decided on normal development by <8 Bd3> with a view to early occupation of the e-file by my rook.

I conclude by giving my game in full. Maybe a reader will discover an improvement for White somewhere?

Tim Harding - Xaver Steiner (Xaver Steiner) corr 1995-6

1 e4 e6 2 d4 c5 3 d5 exd5 4 exd5 d6 5 Nc3 a6 6 a4 Nd7 7 Nf3 g6 8 Bd3 Bg7 Black could play 8...Qe7+ forcing 9 Be2 but after 10 0-0 and 11 Re1 it is Black who ends up losing time. <9 0-0 Ne7!> Not 9...Ne5 10 Nxe5 when 10...dxe5? 11 f4 Qb6 12 Kh1 exf4 13 Qe2+ Kf8 14 a5 Qd8 15 Qf2 Bd4 16 Qxf4 gives a good initiative and 10...Bxe5 11 Re1 Ne7 12 Bh6 is also fairly promising. <10 Ne4!?> Note that if 10 Bg5 0-0 11 Qd2 (11 Re1 f6!?) 11...Ne5!? 12 Nxe5 dxe5 White must avoid the trap 13 Bxg6?? f6! <10...Ne5 11 Bg5 0-0!> Black is right to get his king into safety. Obviously 11...Bg4 12 Nf6+ favours White and trying to prevent the knight check by 11...Ng4 12 h3 f5 (12...Bxb2 13 hxg4 Bxa1 14 Qxa1) allows 13 Nxc5! <12 Nxe5> My calculations showed that White is driven back after 12 Nf6+ Kh8 13 Nxe5 dxe5 14 c4 h6! 15 Bh4 Nf5! (but not 15...g5? 16 Bxg5 hxg5 17 Qh5+) while 12 Bf6 Bxf6 13 Nxf6+ Kg7 is also inconclusive. <12...Bxe5 13 Bf6> My idea was to liquidate Black's most dangerous piece and create weaknesses in his King position. <13...Bxf6 14 Nxf6+ Kg7 15 Qd2! > This was the move I was counting on.

click for larger view

15...Ng8!
Black finds the right answer. Unfortunately he is under no compulsion to go in for 15...Kxf6? 16 Qc3+ Kg5 17 Qg7 when White has a very promising king-hunt, e.g. 17...Nxd5 18 h4+! Kg4 (18...Kf4!?; 18...Kxh4 19 Qh6+ Kg4 20 f3+ Kg3 21 Qh2# mate) 19 Rae1! Nf4 20 Re4; I had expected instead 15...Nf5 when I had hopes of some initiative with 16 Ne4 or 16 Ng4.
<16 Nxg8 Rxg8 17 b4> I was trying to keep the initiative while his queenside was undeveloped and maybe force a passed d-pawn or weakness on b-file, since there seemed nothing doing on the kingside. <17...b6 18 Qc3+ f6 19 bxc5 bxc5 20 Rab1 > Maybe 20 Rfe1!? could be preferred but my judgment by now was that Black had conducted a successful defence and that I should be wary of drifting into an inferior position.
<20...Ra7! 21 Rfe1 Kf7 22 Qd2 Re8 > At this point I decided the prudent thing was to haul in the half point while there could still be no doubt about it, so: <23 Qh6 Rxe1+ 24 Rxe1 Kg8 25 Bxg6! hxg6 26 Qxg6+ > and I made my opponent an offer he couldn't refuse (a draw).

click for larger view

Anyway, whatever the merits of 5...a6!?, there would appear no obvious way for White to achieve any substantial advantage by force against 2...c5 and the line clearly deserves more attention from both theoreticians and practical players than it has so far received.

Tim Harding (aka The Kibitzer)
http://www.chesscafe.com/text/kibit...

♙ ♙ ♙ ♙ ♙ ♙
Discussion <Is the Franco-Benoni playable, "busted" or great?<>> http://www.chesspub.com/cgi-bin/yab...

Here's the one game played by Ivanchuk:

Ivanchuk,V (2709) - Kengis,E (2570) A43
Warsaw AIG Life rapid Warsaw (5), 21.12.2002

1.d4 e6 2.e4 c5 3.d5 exd5 4.exd5 d6 5.c4 Nf6 6.Nc3 Be7 7.Bd3 0-0 8.Nge2 Na6 9.0-0 Nc7 10.Ng3 Rb8 11.Qc2 g6 12.f4 b5 13.b3 bxc4 14.bxc4 Nce8 15.h3 Ng7 16.f5 Nfe8 17.Bd2 Bf6 18.Rae1 Bd4+ 19.Kh1 Qh4 20.Rf3 Nf6 21.Qc1 Ngh5 22.Nge2 Ng4 23.Bg5 Nf2+ 24.Rxf2 Qxf2 25.Ne4 Qxf5 26.Nxd4 cxd4 27.Nf6+ Qxf6 28.Bxf6 Nxf6 29.Qf4 Ne8 30.Re7 Ng7 31.g4 a5 32.c5 Rb4 33.c6 h5 34.Kg2 hxg4 35.hxg4 a4 36.Kf3 Rb2 37.Qf6 Rxa2 38.Bxg6 Bxg4+ 39.Kxg4 Rg2+ 40.Kf3 Rxg6 41.Qf4 d3 42.Ra7 Nh5 43.Qd4 Rg3+ 44.Kf2 Rg6 1-0

Oh yeah, it was a rapid game!