Like most of the early developers of the Norwegian variation, IM Arne Zwaig is now a member of my chess-club, Oslo Schakselskap. One of his main contributions was his investigation of the line 7...f6 (after 1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 b5 5.Bb3 Na5 6.0–0 d6 7.d4) which borders between being an independent line and a pure move-order finesse. Black offers White to exchange his light-squared bishop for Black’s knight on g8 rather than the a5-knight. This may at first glance seem advantageous for White who prevents short castling and avoids doubled b-pawns, but on closer inspection it’s not at all so clear, as the open a-file after ...Nxb3 is one of White’s major assets in the mainlines.
If White accepts the offer, the game will take an independent course. However, if he declines with 8.Nc3 (as Greet recommends in his Play the Ruy Lopez), play will almost always transpose to the mainline after 8...Nxb3 9.axb3. So in reality the effect of Zwaig’s move-order is to avoid the 7...Nxb3 8.axb3 f6 9.c4 line and maybe to tempt White into inferior lines.
Let’s have a look at 8.Bxg8!? Rxg8 (Dia):
9.Nh4!?
This opens a path to h5 for White's queen and is a natural attempt to demonstrate that Black's kingside has been weakened. The main alternative is 9.b4 (to which I may return in a later entry) but White has also tried 9.a4, 9.Nc3, 9.Qe2, 9.Ne1, 9.c3 and 9.dxe5.
9...Qd7
This indirectly protects the h-pawn so that Qh5 now can be safely met by ...g6.
10.a4
White must try to stir up things early as Black's bishop pair tends to become a significant factor if the game develops quietly.
a) 10.dxe5 dxe5 11.Qh5+ g6 12.Qf3 Be7 13.h3 Bb7 14.Rd1 Qc6 15.Nc3 f5 16.g3 fxe4 –+ R.Rodriguez-Alvarez Ibarra, San Sebastian 1996
b) Fritz' suggestion 10.Nf5 g6 11.Ne3 must be taken seriously as 11...c6 12.d5 c5 13.Bd2 Nb7 14.a4 b4 15.Nc4 is very good for White.
c) 10.Nc3 and now:
c1) 10...Qg4 11.Nd5 (11.Nf3 Bb7 12.a4 b4 13.Nd5 0–0–0 14.Re1 f5 was unclear in Ostojic-Johannessen, Amsterdam 1969) 11...Ra7 12.Nf3 +=
c2) 10...c6 11.b4 Nc4 12.a4 Bb7?! (12...Qg4 looks like an improvement, e.g.: 13.Qxg4 Bxg4 14.dxe5 dxe5 15.axb5 Bxb4 and Black is clearly better) 13.Qd3 g6 14.Nf3 Rg7 15.d5 c5 16.axb5 axb5 17.Rxa8+ Bxa8 18.Bh6 +/- Y.Gruenfeld-Moen, La Valetta 1980
10...b4 11.c3 Rb8 12.Nd2 g6 13.Nhf3 c5 14.dxc5 dxc5 15.Qe2 Rg7 16.Ne1 b3 (Dia)
17.g3?!
17.f4 seems to be a better attempt and may be sufficient for a small advantage. Black's king will be exposed but may escape danger via f7 and g8.
17...Qc6 18.Ng2 c4 19.Ne3 Be6 20.Nd5?!
Tactically this is a correct pawn sacrifice; if Black accepts it immediately his king will be dangerously exposed. However, strategically White is not able to support this knight so in a relatively short perspective this move will either be an incorrect pawn sacrifice or a double tempo loss.
20...Rf7 21.f4 Bc5+ (Dia)
It's already obvious that Black is better - mainly because his king now is ready to run to f8.
22.Kg2
After 22.Ne3 Qd6 Black is more active and has the bishop-pair as a bonus.
22...Rd8
Also the direct 22...Bxd5 23.exd5 Qxd5+ seems to be sufficient for a clear advantage.
23.Nf3 Bxd5 24.exd5 Rxd5 25.Be3 Bxe3 26.Qxe3 Rd3 27.Qe2 Re7 28.Rae1 e4 29.Nd4 Qd5
Now Black's advantage is beyond doubt, and it's hard to see how White can keep his position together.
30.Qg4 f5 31.Qg5 Nc6 32.Qf6
32.Nxc6 Qxc6 33.Kh3 Rd2 is no better.
32...Nxd4 33.cxd4 Rd2+ 0–1 Hecht-Zwaig, Raach 1969 (34.Kh3 e3 35.Rg1 Qf3 decides immediately)
