I was browsing IM Jovanka Houska's 2007 book Play the Caro-Kann, and while looking through the introductory section on the Panov/Botvinnik Attack I read something incredible. In a subsection called 7th move sidelines, I came across this:
1 e4 c6 2 d4 d5 3 exd5 cxd5 4 c4 Nf6 5 Nc3
5 Nf3 is known as Vallicella's Caro-Kann trap - Black has to watch out for one big trick. Best is simply to play 5...Nc6, transposing to the main line after 6 Nc3, but 5...Bg4? would be a mistake after 6 c5! Nc6 7 Bb5. The point is that Black has big difficulties defending the c6 point; for example, 7...e6 8 Qa4 Qc7 9 Ne5 Rc8 10 Bf4 and White is winning! [p. 76]
There's nothing objectionable about the analysis*; rather, what struck me was the reference to Vallicella's Caro-Kann trap, as if this was standard lore in treatments of the Caro-Kann. How did Bill Vallicella, an outstanding philosophical blogger but a 1500-1700 club player not engaged in publicizing his games, suddenly achieve such fame? I had come across his trap either from an email by him or on a post on his predominantly philosophical blog, but when did a move he may have played but a single time turn into an idea requiring mention in a pretty major new theoretical work?
Houska doesn't cite a source, and I certainly didn't recall seeing it in any published materials, so naturally it was off to Google. Entering "Vallicella Caro-Kann", I discovered the main source, conveniently entitled "Vallicella's Caro-Kann Trap"...and you can, too - just click here. Then laugh.**
* Actually, while I wouldn't disagree with her positive suggestion, I don't believe 5...Bg4 is in fact a mistake; the real error comes later. After, e.g. 7...e5 I don't see a White advantage after 8.dxe5 Ne4 or 8.Qa4 Bxf3 9.Bxc6+ bxc6 10.Qxc6+ Nd7 11.gxf3 exd4, and even the arguably best 8.Nc3 promises little or nothing after 8...Nd7 9.dxe5 Bxf3 (10.Qxf3 d4; 10.gxf3 a6).
** If anyone knows IM Houska personally, please ask her to write me - I'd like to trace the path from Vallicella's idea to her book...............................
(A) 5...e6, when White plays 6.Nf3,
(B) 5...g6, when White plays 6.Qb3, and
(C) 5...Nc6, when White can either accede to the pin after 6.Nf3 Bg4, or else play the sharper but less reliable 6.Bg5.
Instead, the Maverick Philosopher has been utilizing the tricky 5.Nf3. It looks slightly clumsy, welcoming the Black bishop to g4 right away, but his idea is revealed after 5...Bg4 6.c5 Nc6 7.Bb5 e6 8.Qa4 Qc7 9.Ne5 Rc8 10.Bf4, when between the pin on c6, the threat of various discoveries involving the Bf4/Ne5/Qc7, and other, lesser but still significant problems with the Black position, White is winning.
Where did Black go wrong? I've already addressed this to some extent in a post on my previous blog, but as the move order examined there was a bit different than what we find in this game, I'll offer some new comments.
First, on move 5, Black can respond with the three normal anti-5.Nc3 options: 5...e6, 5...Nc6, and 5...g6. Should he do so, I don't see any advantage to be had by 5.Nf3, and there is a possible disadvantage. After 5.Nc3 g6, White's best try for an advantage is 6.cxd5 Bg7 7.Qb3 O-O 8.Be2 Nbd7 9.Bf3 Nb6 10.Nge2, but this variation is obviously impossible once White has placed the knight on f3. After 5.Nf3 g6 6.Nc3 Bg7 7.cxd5 O-O 8.Bc4 Nbd7 9.O-O Nb6 10.Bb3 both 10...Nbxd5 and 10...Nfxd5 have scored very well for Black.
Second, after 5.Nf3 Bg4 6.c5 Nc6 7.Bb5, the confrontational 7...e5 seems to give Black equal chances after 8.dxe5 Ne4 9.b4 Be7 10.O-O O-O 11.Bxc6 bxc6 12.Qd3 a5 13.Nd4 Bd7.
Third, as mentioned in my earlier blog post (linked above), after 7...e6 8.Qa4, the pawn sac 8...Bxf3 9.Bxc6+ bxc6 10.Qxc6+ Nd7 11.gxf3 leaves Black some compensation for the pawn in the form of White's numerous pawn weaknesses and the lack of an obvious refuge for the White king.
In sum, I think 5.Nf3 is objectively inferior to 5.Nc3. However, it doesn't seem that much weaker, and it does come with a nice positional trap, making it a reasonable surprise weapon for the odd game.
The variations above, and a bit more, can be replayed here.