Oh dear! -- I really had better go back to sleep! Many thanks for your replies, Ben and Alumbrado, and please accept my apologies for what was a truly weird case of 'autopilot' thinking and daft confusion!

In the game I cited, White of course didn't play Nc3 and Be3 at all, but Be2 and 0-0!! -- he met Black's system by developing his kingside, as Davies in fact recommends (p. 128)! Davies says that, after 5 ...g6 6 Nf3 Bg7 7 Be2 0-0 8 0-0, 8 ...Be6 can now be met by 9 Nbd2. Lavrov played instead 9 b3, and the game appears to show that here the ...c5/ ...Nxe5 tactic does not work. My blunder arose from my confusing in my mind THIS position with that occurring after the moves I cited in my post, in which White's QB is of course on e3!! Here, according to Davies (p. 129), the 9 ...c5/ ...Nxe5 tactic is on.

This just shows the dangers of dabbling in chess -- plus not even bothering to get a board out -- when tired! I'm duly chastened, but maybe good will come of my laziness if I've opened up a discussion on an interesting line. The first question here might be, how should Black respond to Davies's suggestion? Since after 8 ...Be6 he has to contend with both 9 Nbd2 AND 9 b3, how about playing instead 8 ...de 9 fe c5, intending 10 ...Bg4? This, I suppose, transposes to 5 ...de/6 ...g6 lines, which have received some coverage on ChessPub. Anyway I'd love to know what you think and what, if anything, you reckon might call into question either of these fascinating lines.

I can't see anything much for Black. As far as I can see best play is something like 13.Ne5 Ne5 14.Qd8 Rd8 15.Be5 Be5 16.fe Nd7 17.Bf3 Rab8 18.Re1 b5 19.Na3 when I don't think Black has anything really for the pawn. It's difficult to see how he can avoid that line without just being worse.

Personally I have played 5...g6 but I don't really trust it.

Ben Hague