ForumHelpSearchMy ProfileSite InfoGuests InfoRepertoireLinks
The last couple of months have seen the peak of the chess year in regards to number of games played each week and which appears to have also created an influx of new chess books that are being published in droves at the moment. All of this makes it a little difficult to keep up with the latest trends and recommendations. Therefore in this month's update I will present a feature on a particular line in the King's English Four Knights. Next month's column will conclude our investigation into this particular line. The line has been recommended for Black as part of a repertoire in the new "Play 1...Nc6!" by the German International Master Christoph Wisnewski. If you don't mind playing offbeat lines and the Queen's Gambit Chigorin looks sufficiently interesting for you to wheel it out as Black, then the book could well be worth another look.
In the future, whenever I see a recommendation against openings that fall under our subject of Flank Openings, I will examine the recommendations closer to see if they hold water. In upcoming features, I will discuss some of Richard Palliser's recommendations from his recent book "Beating Unusual Chess Openings", where he takes aim at the English, Reti, King's Indian Attack, The Bird, The Nimzo-Larsen Attack and just about any other opening that we flank opening enthusiasts enjoy.
On a final note before I move over to this month's games, I can inform you that I have finally started receiving the page proofs for John Donaldson's and my "Strategic Opening Repertoire for White - 2nd Edition"; from here onwards it should be a relatively speedy process before it becomes available in stores.

Download PGN of August '07 Flank Openings games

King's English Four Knights: 4...Nd4!? - Part I

In this theoretical feature I will discuss the following line: 1 c4 e5 2 Nc3 Nc6 3 Nf3 Nf6 4 g3 Nd4!?:

As mentioned in the introduction to this month's update, this variation is recommended in Wisnewski's new repertoire book "Play 1...Nc6!", where he boldly claims the variation allows Black to equalize comfortably! While not agreeing entirely and finding some flaws in his coverage, I do agree with him to the extent that it is one of Black's better options on move four. In my "Guide to the English Opening: 1...e5" (Gambit 1999), I wrote the following: "This idea of Korchnoi's, dating from 1972, constitutes one of Black's best defences against 4 g3. Black wants to exchange a set of knights, believing that by doing so the black defence will be easier. On the other hand, Black invests two tempi in the project, but the nature of the position makes it difficult to take advantage of this."

The first try for White is 5.Nxe5, which is almost universally condemned as bad move on account of 5...Qe7 6 f4 (6.Nd3?? Nf3 mate is of course embarrassing) 6...d6 7 Nd3 Bf5:

but as John Watson pointed out back in 1979, White can try 8 Kf2!, and things are not as bad as they seem. This has been tested in Bauer - Dorfman, where I have added some supporting analysis, including some answers to the recommendations by Wisnewski as well as some improvements for Black.

The second option for White is 5.Nh4!?, which can be met with 5...c6 - see Gulko-Hansen in the PGN archives and 5...d5!? - see Korchnoi-Svidler, also in the PGN archives.

The main line is 5.Bg2 Nxf3+ (also 5...d6 has been played a couple of times, including in an old game, via transposition, Petrosian-Smyslov, Budapest 1952; Black has yet to face any problems in this line) 6.Bxf3 Bb4:

The alternatives: 6...Be7, 6...c6, 6...d6 and 6...Bc5, will be discussed in part II of this feature.

Now White has a number of moves available:

In the main game Jobava - Aronian we will discuss some of the lesser alternatives, including 7.Na4, 7.Nd5, and 7.d4, while 7.d3 will normally transpose to lines after 7.0-0 0-0 8 d3. Of these moves, 7.d4 is clearly the most interesting, although it appears Black is okay.

After 6...Bb4, White has three main line moves:

  1. 7.Qc2
  2. 7.0-0
  3. 7.Qb3

Of these moves, the latter is considered the slightly more accurate, but the other moves also deserve their due. Please note that there is an element of line A and C transposing into line B, as well as the other way around.

A) 7.Qc2

This is a solid move which does little to challenge Black right away:

In Karpov - Vyzhmanavin we will examine this move in detail.

B) 7.0-0 0-0

With two options for White:

B1) 8 d3 (here 8 Qb3 transposes to the lines C21-C23 below, while 8 Qc2 transposes to line A above), and here Black has a variety of moves available, these are covered in our main game Huzman - Svidler.

B2) 8 Bg2 isn't White's most dangerous option, yet in Jobava - Mchedlishvili White won in a mere 25 moves. We will examine all of Black's 8th move alternatives.

C) 7.Qb3

Now Black has tried several things:

C1) 7...a5; C2) 7...Bc5 and C3) Lesser alternatives: 7...Be7, 7...Ba5, and 7...Qe7.

C1) 7...a5 is considered to be better for White, and there haven't really been any developments in this line for years. The current theoretical situation is covered in Sunye Neto-Morovic Fernandez.

C2) 7...Bc5

Now we reach a major fork in the road, because the play can develop in completely different directions within a few moves, yet still allow for a number of transpositions:

C21) 8 0-0 0-0 (and 8...c6) 9 Na4!?
C22) 8 0-0 0-0 9 d3 c6 (and minor alternatives) 10 Bg2
C23) 8 0-0 0-0 9 d3 h6 10 Bd2
C24) 8 d3 c6 (and 8...0-0 without 9 0-0) 9 g4 or 9 Bg5
C25) 8 d3 h6 (without 9 0-0 0-0 - C23)

C21) 8 0-0 (8 Na4 will normally just transpose our main line after 8...Be7 9 0-0 0-0) 8...0-0 (8...c6 has been played a couple of times - see the notes to the main game) 9 Na4!?

is an interesting alternative to the main lines (and one that is ignored by Wisnewski) below - see Kosten - Pert. In the notes to this game, you will also find 9.Rd1!? which has only been used a couple of times, but White has obtained the better chances each time.

C22) 8 0-0 0-0 9 d3 c6 (and minor alternatives) 10 Bg2

This is covered in Lautier - Sebag. In my opinion, White doesn't achieve any real chances of obtaining an edge out of the opening with this line.

C23) 8 0-0 0-0 9 d3 h6 10 Bd2

(and some other 10th move alternatives for White) is in my opinion quite harmless for Black - see Ljubojevic - van Wely.

C24) In the line 8 d3 c6 (8...0-0 9 0-0 transposes to the lines above, but 9 Bg5 and in particular Watson's 9 g4!? are worth a try) 9 Bg5 (9 g4 can be found covered in the games Vallejo Pons-Gelfand, Pamplona 1999 and Aronian-Gelfand, Saint Vincent 2005 in the PGN archives) 9...Be7 10 Rd1 ended up offering White an edge in our main game Kovacevic - Potkin, but Black should be able to improve. However, as one of the lines that Wisnewski has failed to mention in his book, White should definitely give it some consideration.

C25) 8 d3 h6 (but only the lines without 9 0-0 0-0, which is covered under C23), and now 9 g4 c6 10 h4 and 9 h4 c6 10 g4 leads to the same interesting position, which can be found discussed in the main game Mikhailuk - Sagalchik:

C3) Lesser alternatives: 7...Be7, 7...Ba5, and 7...Qe7. The first move is hardly ever seen, the second has only been used by Garcia Palermo and with him off the chess scene, the move departed as well. Both these moves offer White the better chances; that leaves 7...Qe7 which has done reasonably well and deserves more attention:

We will look at all of these lines in our main game Topalov - Karjakin.


In the main lines, Black appears more or less okay, and as such this represents a good alternative for Black to the more popular 4...d5 and 4...Bb4, both of which are heavily loaded with theory. For those who don't mind sharp positions, White should opt for 7 Qb3 Bc5 8 d3 and 9 g4, whereas the more positionally inclined players should probably go for 8 0-0 0-0 9 Rd1. Finally, White can also try 5.Nxe5, which forces Black to play sharp, complicated lines, very much unlike the rest of the lines in this variation.

See you next month, Carsten


Please remember to point out and send your games to me. Drop me a line at the Flank Openings Forum, or subscribers can write directly to