DDG 4.Nc3 and Various

Outside the main lines, Black has a couple of interesting options after 1.e4 e6 2.d4 d5 3.c4 dxe4 4.Nc3. Especially 4...e5 is a fascinating move.
  • 4...Nc6
  • 4...e5 (Pietilš Variation)
  • 4...c5 (Tarrasch Defence)
  • 4...Be7 (Charousek Variation)
  • 4...c6 transposes
  • 4...Various


I have met this strong-looking move only a few times. It often transposes to more common variations, but the following moves may also lead to unique lines.

4...e5 (Pietilš Variation)

4...e5!? looks paradoxical in that Black moves the e-pawn for the second time. The move was invented by Jari Pietilš, a colleague of mine.

The position resembles the BDG, Lemberg Defence, also known as the Lemberg(er) Counter-Gambit, except that White has played c4 for free. On the other hand, the c4 move has drawbacks: it takes two potential squares from the king's bishop, and it would weaken White's king after queenside castling.

The best way to utilize the c4-pawn is 5.d5.


Other moves that have been tried:

A) 5...Bf5

B) 5...f5 6.f3 Bc5 7.fxe4 Bxg1 8.Rxg1

C) 5...Nf6 6.Bg5?! (6.Nge2) h6 7.Bxf6 Qxf6 8.Nxe4 Bb4+ 9.Nd2 O-O 10.a3 Bxd2+ 11.Qxd2 Rd8, 0-1 in 26, Heikkinen - Pietilš, corr. 2003-2004. The lesson learned is that White should retain the Queen's Bishop, instead of trying to regain the pawn.

D) 5...Bb4?? 6.Qa4+ wins a piece.

The Ellis Gambit 1.e4 e6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 e5 might also transpose to this line.

The following Lemberg Defence lines have ideas similar to this DDG variation:

1.d4 d5 2.e4 dxe4 3.Nc3 e5

4...c5 (Tarrasch Defence)

4...c5 5.d5 exd5

5...Nf6 6.Bg5 Be7 7.Bf4 Qa5 8.f3 O-O 9.Qd2 exd5 10.cxd5 Bf5 11.fxe4 Bg6 12.Nf3 Qb4?! (12...Re8) 13.a3 Qb3 14.Bd3 c4 15.Bc2 Qb6 16.e5 Qxb2? 17.Ra2 Qb6 18.Be3, 1-0 in 53, Heikkinen - Tuomala, 1999.

6.cxd5 transposes surprisingly to the QGD Tarrasch Defence, Marshall Gambit: 1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 exd5 5.e4 dxe4 6.d5. Janowski, Grob and Bronstein played it as White; Tarrasch and Euwe met it as Black. See the index of transposed games.

The reversed position can be reached in the Albin Counter-Gambit: 1.d4 d5 2.c4 e5 3.dxe5 d4 4.Nf3 c5 5.e3 Nc6 6.exd4 cxd4. Tartakower was the most famous player of the 4...c5 variation. "Many masters were addicted to gambling, notably Tartakower, Janowski and Marshall", wrote The Even More Complete Chess Addict (1993). The DDG is gambling.

In The Complete Book of Gambits, Keene gave 2 stars out of 5 to this gambit, meaning "doubtful". 1 star would have been "unsound". For example, the BDG got 3 stars: "playable to both sides".


The main lines are 6...Nf6 7.Bg5 (7.f3 transposes to the DDG Keres Variation) and 6...f5 7.Bf4, which is covered in Junejew - Shimarjow, 1978. Marshall gambiteers do not seem to continue the gambit with f3, however.

4...Be7 (Charousek Variation)

4...Be7 5.Nxe4 Nf6 6.Nxf6+ Bxf6 7.Nf3 c5 8.Be3 cxd4 9.Bxd4 Qa5+ 10.Qd2 Qxd2+ 11.Kxd2 Bxd4 12.Nxd4 O-O 13.Rd1 Rd8 14.Kc1 e5 15.Nb5 Rxd1+, 0-1 in 59, Afshar - GNU Chess, 1996.

The game actually started as 1.d4 d5 2.c4 e6 3.Nc3 Be7 (Charousek Variation, QGD) 4.e4. Afshar's surprising comment in post-mortem was: "I believe I lost this game here?" First I thought he must be joking, but then I tried to improve White's play: 6.Nc3 looks solid but stupid; 6.Qd3 takes the bishop's square; 6.f3? Nxe4 7.fxe4 Bb4+ and 8...Qh4+; and finally 6.Bd3!? looks desperate gambit with no real compensation. Something must have gone wrong before that!

In the original QGD game, theory gives 4.Nf3 Nf6 5.Bg5 O-O 6.e3 - no problem for White. White just does not play e4 because it would leave d4 defenceless. Suddenly I realized that by playing 4.e4, White is actually offering a pawn sacrifice! After 4...dxe4, White can choose between 5.f3, 5.Be3, 5.Nge2 or 5.g3. But the point is that 5.Nxe4 is simply a bad move.