More Ways To Battle The Caro-Kann

Last week, we discussed the first few options to fight against Black's Caro-Kann. 

Now let's take a look at some of the most important ways White can respond to this opening.

The Advance Variation:

The line 1.e4 c6 2.d4 d5 3.e5 has an interesting history. Originally, this was simply regarded as unpromising. After all, before Aron Nimzowitsch popularized the move 3.e5 against the French Defense, that too was not highly regarded; by comparison, the same move against the Caro-Kann leaves the black Bc8 free to develop on f5. Against this Black will lose a tempo if he wishes to play the undermining ..c5, but surely avoiding the bane of the whole French Defense should be more valuable?

Of course some masters used the move 3.e5, but -- particularly after the famous game, Nimzowitsch-Capablanca, New York 1927 (which can be found in the first article of this series) -- it was seen comparatively rarely and without good results for White. In the early days, White tended to play the simplistic 4.Bd3 in response to 3...Bf5, and after the exchange of bishops an "ideal French" from Black's point of view was reached.

The young Paul Keres experimented with the move 4.h4!? in the 1930s, while Boris Kostic of Yugoslavia played some early games with 4.c4 in the 1940s. These two moves sought to avoid the "ideal French" scenario of the earlier games and to take advantage of the tradeoffs from Black's early development of the bishop and relatively useless move ...c6. The first sought to cramp the black bishop, while the second emphasized White's space advantage.

The move 4.h4 featured prominently in Tal's second world championship match with Botvinnik in 1961 (some games from that can be found in last week's article), although it did not become particularly popular -- many players had a problem with such a vague and, at the same time, committal move.

Tal via Wikipedia

The Advance Variation really took off when players of the white pieces began meeting 3...Bf5 with a rather innocent-looking move: 4.Nc3. Then after 4...e6 came the wild 5.g4!? Bg6 6.Nge2:

White's wild play looks too weakening. White throws up the kingside pawns without having any clear spot to put his own king. Nevertheless, by putting Black's light-squared bishop in jeopardy, White is logically taking advantages of the drawbacks of Black's otherwise favorable development of that bishop. White gains space and begins a rapid attack; in the course of dealing with the threats to his bishop, Black often has to delay his own development.

This was first played in 1931 by Benjamin Blumenfeld against Genrikh Kasparian. White lost the game in fairly thematic fashion when his position proved to be too exposed -- as a classicist would expect:

Twenty years later, Viktor Kortchnoi picked up the discarded variation with more success than Blumenfeld had. But the line did not start to be seen frequently in master play until the 1980s, when aggressive players such as Viktor Kupreichik, Gyula Sax, and John Nunn started to use it frequently.

By the late 1990s, this hyper-aggressive variation had become a main line in the Caro-Kann, with frequent battles between top level players. It was used consistently by Alexei Shirov, while Garry Kasparov, Viswanathan Anand, and Alexander Grischuk also used it frequently. It produced a great many fascinating, wild games. In fact, this line was truly a 1990s opening -- Kasparov was the champion, sharp and theoretical lines were deeply investigated, while computers were not yet strong enough to make any kind of convincing verdicts on sharp variations.

Thus we saw some wild games. To counterattack and expose White's weaknesses, Black often had to open the center before his development was complete. Thus you often saw some piece sacrifices to tear open the center.

On the other hand, it was frequently Black who sacrificed a piece, developing a strong attack and gaining several pawns while White hunted down the light-squared bishop:

I distinctly remember the famous Shirov-Nisipeanu game from the Las Vegas 1999 World Cup. Shirov was a favorite player of many in those days (and at the time he was owed a world championship match). The game was a great disappointment, as he had a highly promising position in the opening and the loss knocked him out of the tournament. But in those days, these kinds of games provoked months of fascinating analysis.

In the early 2000s the line reached its height of popularity. But by 2007 or 2008, the line had fallen from favor. Perhaps it was due to the good reputation of the formerly obscure line 6...f6 (featured in the very first game with the line). Or it could be that, in the computer age, players have lost their taste for such concrete, ultra-tactical (and positionally risky) lines.

It would take a great deal of work - as well as a lot of practical experience -- to master the line sufficiently to have good results, and once a record of playing it in tournaments was established in the database, future opponents could easily "take aim" at you with concrete, computer-assisted preparation in one of Black's many options. While this crazy line is certainly not refuted, perhaps you could say that it became too hot to handle. Nevertheless, for players who delight in such positions, the line continues to be used.

At the same time, a new and sounder approach began to take over. In the early 1990s Nigel Short began to specialize in what appears like a completely normal development, but which had hardly been used in master chess up to that time. 1.e4 c6 2.d4 d5 3.e5 Bf5 4.Nf3 e6 5.Be2:

This seemingly innocent way of playing was not appreciated earlier because it appeared that it gave Black exactly what he wanted -- the bishop on f5 was developed outside of the pawn chain and Black otherwise was left in peace. But in recent years, players have begun to realize the strategic problems Black faced.

First of all, the counterattack ...c5 was natural for Black, but it became clear that it was not so easy to safely achieve that. The move ...Nd7 was sometimes needed to support ...c5, but the knight would be on a very poor square in the resulting position. Additionally, with Black behind in development it could be dangerous to play ...c5. White began to meet this thrust in sharp ways, such as with c2-c4, or by capturing on c5 followed by utilizing the d4-square. While in the early days, Short used to meet Black's ...c5 with the quiet c3, the most recent theory has focused on White's sharp methods and Black's attempts to make an early ...c5 work.

Meanwhile, when Black avoids ...c5, White places his hopes on his space advantage. In fact, this means that Black has a complicated jigsaw puzzle to solve on the kingside -- there are more pieces to develop than there are safe squares. For instance, if the black knight goes ...Ne7-f5, then there will be a constant threat of g4; if on the other hand, Black plays ...h6, ...Bh7, and ...Ne7-g6, then the bishop is buried on h7 and the move Bd3 might be dangerous.

A setup with ...h6 and ...g5, enabling ...Bg7 and ...Ne7 weakens the kingside. Playing ...Bg6 and ...Nh6 leaves Black having to worry about Bxh6 at any point. Many players of the black pieces have thus resorted to ...Ne7-c8, allowing ...Be7 and ...0-0; however, with the black pieces cramped White can use a plan with gaining space by a4-a5 and then c2-c4.

As it currently stands, the Short System is seen as one of the main threats to the Caro-Kann's vitality, and a great many top-level games are played in this line. On the other hand, the slow, non-concrete, maneuvering games that result are not to everybody's liking.

The Main Line: 1.e4 c6 2.d4 d5 3.Nd2 (or 3.Nc3) dxe4 4.Nxe4

Throughout the years, the classical development with White's queen knight (and 3.Nd2 has generally been seen as slightly more accurate, since after 3.Nc3 there is the extra option of 3...b5!? -- whether it is good or not) has remained the most popular. As I mentioned in last week's article, Black's new plan of castling kingside in the 4...Bf5 5.Ng3 Bg6 6.h4 line has greatly increased the Caro-Kann's popularity.

In the early and middle 20th century, White often used some particularly aggressive lines against 4...Bf5, such as 5.Ng3 Bg6 6.Bc4, or 6...N1e2. Paul Keres and Mikhail Tal were two top players who used these lines.

Nevertheless, the line 4...Bf5 5.Ng3 Bg6 6.h4 h6 7.Nf3 Nd7 8.h5 Bh7 9.Bd3 Bxd3 10.Qxd3 has long been the main variation of the Caro-Kann. White has continued to look for advantages here in the sharp variations with Black castling kingside. Major developments have been rare, but there have been continuous small improvements in the very rich positions which result.

Black's other possibilities after 4.Nxe4 are 4...Nd7 and 4...Nf6.

In the early days of the 4...Nd7 line, White usually responded with the natural move 5.Bc4 (meeting 5...Ngf6 with 6.Ng5, inducing ...e6) or 5.Nf3. But in the 1980s, the move 5.Ng5 started to be played. The point was to meet ...h6 with Ne6!, due to Black's weakness on the h5-e8 diagonal. There were also some possible piece sacrifices on e6 or f7.

For instance, the following was the decisive game in Kasparov's lost match against the computer Deep Blue. By this point, the line 5.Ng5 line as well-known and explored; Kasparov had used it as White multiple times.

It is important to note, however, that this piece sacrifice is far from clear. In fact, the Chilean grandmaster Julio Granda Zuniga had defended the position after the better 8...fxe6 both in one of the earliest games with the line  in 1992, as well as less than a month ago, in the Reykjavik Open tournament.

Nevertheless, most players have been scared off by this sacrifice, and thus 7...Bd6 is usually preferred. White has then sought an advantage in the line 8.Qe2 h6 9.Ne4 Nxe4 10.Qxe4. At some point the line 10...Qc7 (keeping the knight on d7 where it supports ...c6-c5) 11.Qg4 Kf8 was "hot theory."

Moving on, White's approaches to 4...Nf6 5.Nxf6+ gxf6 or 5...exf6 have always varied over the years. Against the former, White has often played systems with g2-g3. However, one of the main things to point out has been the development of a strategically clear plan where White plays 6.Nf3 Bg4 7.Be2 followed by 0-0 and at some point Nh4, forcing the trade of bishops, deadening the g-file with g2-g3, and then breaking through with c2-c4 and d4-d5. In the event of the black e-pawn capturing on d5, White can occupy f5 with the knight.

Against 5...exf6, White's main system has been 6.c3, followed by Bd3, Qc2, Ne2, sometimes with 0-0-0.

Odds and Ends:

Besides these main variations, there are a number of sidelines which White can play against the Caro-Kann. First of all, there is the King's Indian Attack, with 2.d3. Compared to King's Indian Attacks against the French or the Sicilian, here the move ...c6 means that Black is more likely to play for ...e5 rather than ...c5. Tal won a famous game against Smyslov with this system:

And Leonid Stein was famous for using 2.d3 -- especially with his trademark b2-b4 move.

Nevertheless, White's slightly passive setup with a lack of direct pressure on Black makes the KIA against the Caro-Kann regarded as little more than a practical weapon -- a way to reach a full-play position and avoid forcing variations.

Next, there is 2.c4. This challenges the very notion of the Caro-Kann. However, after 2...d5 3.exd5 cxd5 4.cxd5 Nf6 the game frequently transposes to the Panov-Botvinnik attack. White can try moves such as 5.Qa4+!?, attempting to interfere with Black's development and hold on to the d5-pawn, but evidence shows that Black is okay. In addition to 2...d5, there is 2...e5 (usually transposing to the Old Indian) or 2...e6!?, as we was in a game by Nimzowitsch from the first article of this series.

Finally, two recent -- and quite eccentric -- lines have been played. One is 2.Nc3 d5 3.Qe2!?

This was first played by a few players in the 90s- - particularly Semyon Dvoirys. Later, Vadim Zvjaginsev, a specialist in unusual opening lines (for instance, it was he who pioneered 2.Na3!? against the Sicilian), began to use it.

The general idea is to induce Black, by the threat of 4.exd5 cxd5 5.Qb5+, to break the tension in the center. After 3...dxe4 4.Nxe4, White can hold back the d-pawn and prepare to quickly castle queenside, as in the above game. In addition, 4...Nd7 is strongly prevented!

Black's main alternative is 3...d4, after which the knight can now retreat to d1. White hopes to reach a good kind of reversed King's Indian. This line has been popular among attacking players who like to avoid the main lines.

A final unusual and recently rather popular line is 2.Ne2!?

This very strange-looking move was used early on by some of history's most creative players: Savielly Tartakower, David Bronstein, and Bent Larsen. The basic idea is to meet 2...d5 with 3.e5, when White has a kind of Advance Variation where he is ready to immediately meet developments of Black's light-squared bishop by some knight maneuvers.

For instance, after the typical (but probably inaccurate) move 3...Bf5, White has 4.Ng3 Bg6 5.h4 h6 6.h5 Bh7 7.e6!, a frequently seen kind of sacrifice. This 2.Ne2 variation has been championed by Nigel Short in recent years.

As you can see, there is something for everybody in combatting the Caro-Kann -- but in some ways White's huge choice is a problem. Personally, I have played practically every line against the Caro-Kann at some point.

In a way, Black just creates his pawn chain and then puts the ball in White's court. Both attacking players and positional players have their options, but in all cases Black remains solid and flexible.