Chips with a pinch of salt: Duchamp and Le Lionnais

With typical anagrammatic humour, Marcel Duchamp — in my opinion the most influential artist of the 20th century — described himself as Le Marchand du Sel (“the salt merchant”). Pronounce it out loud and the phonetic joke will become apparent.

Before I started to devote serious attention to François Le Lionnais, the name brought to my  mind only a potato dish. I now realise that Le Lionnais ( to whom I devoted a previous column here ) was a most formidable practitioner of chess, quite the equal of Duchamp, as we shall soon see.

I start with Le Lionnais’ most famous game, a win against Marcel Duchamp in a competitive event. It was interesting enough to attract the attention of the great Ksawery (usually known as Savielly) Tartakower, the Polish-French grandmaster to whom I recently dedicated an  entire column .

It is serendipitously topical that, perhaps in realisation at the major international art gallery travelling retrospective on Duchamp, due for 2026, the British Chess Magazine, in their October issue, have demonstrated a sympathetic harmony for what regular readers will know, has been my long-standing admiration for this French genius. The BCM’s new series on Duchamp will likely be joined by many others, in many varying publications, over the coming year.

Paris 1932, 5th annual tournament, Cercle Caissa chess club

Marcel Duchamp vs. François Le Lionnais

A lucid description by the late Oliver Beck of Seattle, relates the relationship between these two cultural giants of their day:

‘In “Echecs et Maths”, from  Marcel Duchamp: abécédaire, approaches critiques  edited by Jean Clair (Paris, 1977), Francois Le Lionnais claimed that although he and Marcel Duchamp were brought together through Dada and a strong philosophical and artistic affinity, their long friendship ultimately resulted from a shared interest in chess. He then provided an interesting summary of Duchamp’s chess career.

First, a little art history: 

Le Lionnais gave some background information on his win over Duchamp during the Fifth Tournament of the Cercle Caissa in 1932, a game originally published in the January 1933  Wiener Schachzeitung  with Tartakower’s annotations. 

It was this game that prompted Ralph Rumney to interview Le Lionnais about Duchamp’s chess, and that resulted in Rumney’s “Marcel Duchamp as a Chess Player and One or Two Related Matters”, which was originally published in  Studio International  (January-February 1975). 

The interview, which can be found in  The Map is Not the Territory  by Rumney and Alan Woods (Manchester, 2000), is often cited in the many discussions on the correlation between Duchamp’s art and chess, and particularly regarding the common assertion that an avant-garde artist would naturally be a proponent of hypermodern chess. With this in mind, Rumney’s own observations about chess became especially interesting. In his autobiographical work  Le Consul  (Paris, 1999) he described playing chess with Duchamp at a reception following an art event:

He said to me: “Let’s go to the next room.” There was a chessboard created by Max Ernst. I had created a chess set that I preferred to Max Ernst’s. It has now disappeared into nature, but one of my projects before I die is to find it again or recreate it. Duchamp said to me: “Do you know how to play chess? That way we can chat.”

I play chess in a very ordinary way. Chess is not a game of intelligence but a game of memory and calculation. It’s like poker, like almost all the so-called games of chance that humans have invented; it’s as simple as that. If you can remember everything you need, you win. If your memory fails you, you lose.”

After this, to put it mildly, controversial assessment of chess by Rumney, here is that notorious encounter, a game where memory in fact  plays a subordinate role to creative imagination, while accurate calculation is frustrated by the infinite web of variations which chess tends to spin. 

Analysis by Ksawery (Savielly) Tartakower

(from: Eröffnungstheoretische Streifzüge [Opening Theory Forays] #11, in

Wiener S chach- Zeitung , 1933, pps. 24-26, courtesy of Dr Tim Harding)

Additional analysis by  Adam Black  with  Stockfish 17, such interventions being marked (AB).

1. c4 c5 

The symmetrical treatment. Against 1… e5 (which leads to a reversed Sicilian game), Tartakower opines, controversially, that perhaps 2. a3 is the most energetic. Modern theory would prefer the more direct 2. Nc3. 

2. Nc3 Nc6 3. g3 

Instead of this heavyweight ‘Fianchetto System’, the quick opening of the centre with 3. Nf3 and 4. d4 is also commonly used.

3… g6 4. Bg2 Bg7 5. d3 d6 6. e4 

An important decision. The usual (and reliable) moves are 6. Nf3 Nf6 7. O-O O-O, etc., but White would like to steer into the so-called ‘Dresden Formation,’ characterised by the pawn structure c4, d3, e4, and possibly f4. The e4/c4 pattern was introduced by Nimzowitsch in the 1926 Dresden tournament, which he won convincingly ahead of Alekhine, Rubinstein and, indeed, Tartakower himself. Also interesting here is the immediate 6. f4, a concept later taken up by the first Soviet world champion, Mikhail Botvinnik. 

6… f5 

A helpful innovation that aims to provide a boost in counterplay on the f-file, while the usual continuation: 6…Nf6 7. Nge2 O-O 8. O-O Rb8 9. f4 a6 10. h3 cedes White a tangible initiative on the kingside.

7. Nge2 Nf6 8. O-O O-O 9. Rb1 

After completing his piece development, White wants to protect the b2-pawn, sometimes also preparing for the well-known wing-offensive (b2-b4 etc.). 

9… Bd7 

This modest move is unknown to modern theory, but is a natural selection. Alternatively, Black could achieve full equality after: 9… fxe4 10. dxe4 e5 11. Bg5 Be6 12. Nd5 Qd7 (AB)

10. Bg5 Ng4 11. h3 Nge5

In this new position, Black’s knight exerts pressure on the backward d3 pawn.

 12. f4

Expelling the knight from e5 but also blocking in  his bishop on g5. It was first worth considering the following intricate variation: 12. Be3 fxe4 13. Nxe4 Nf3+ 14. Bxf3 Rxf3 15. Nf4 Ne5 16. d4, which promises material gain to White. But does that confer any advantage? 

Analysis position after 16. d4

The  engine assesses the resultant position as level, one can clearly see the number of slips and tactical miscalculations latent in the position. For example:

A) 16… Rxf4 17. gxf4 Nxc4 18. dxc5 Bc6 19. Qg4 Qc8 20. Ng5 h6 21. Ne6 Be4 22. Nxg7 Qxg4+ 23. hxg4 Bxb1 24. Rxb1 Kxg7 (better than 24… Nxe3 25. fxe3 Kxg7 26. cxd6 exd6) 25. cxd5 Nxd6, with an even position; 

B) 16… Bf5 17. Ng5 cxd4 18. Nxf3 Bxb1 19. Nxd4 Be4 20. Re1 b6 21. Nfe6 Qc8 22. Bh6 Bxh6 23. Rxc4 Qxc4 24. b3, and again, the game remains level. (AB) 

12… Nf7 13. exf5 

This exchange (which White has been trying to avoid since move 7, so as not to allow the opponent’s queen’s bishop to attack the backward pawn on d3) has now become necessary in order to escape the threat of 13…h6 14. Bh4 g5, etc., which would lead to material loss for White. Indeed, at this point, the preventive retreat 13. Bh4 would be insufficient due to 13…h6 14. g4 fxg4 15. hxg4 Bxg4, resulting in a comfortable gain of a pawn for Black.

13… Nxg5 

If now: 13… h6, the zwischenzug ( in-between move ) 14. fxg5 follows.

14. fxg5 Bxf5 15. g4 

Once again (as in White’s 12th move), we have a nervous expansion that compromises king safety and which was better avoided. Meanwhile, after the immediate 15. Ne4, we have 15… Qc8 (or …Qd7), then 16. Kh2.

Although 15. g4 is not such a terrible move, the motive attributed by Tartakower provides a precious and telling insight. That the text move compromises king safety by loosening the king’s pawn guardians, is incontrovertible; that it is significantly more nervous than other moves that maintain an equilibrium, like, 15. Qd2 or Nf4, is apparent. The engine also prefers 15. Ne4 over g4, but takes exception with the narrative following. After 16. Kh2: 16… Rf7 17. Qd2 Raf8 18. Rbd1 Bxe4 19. dxe4 Be5 20. b3, neither side can claim an ascendency. (AB) 

15… Bd7 16. Ne4 Rb8 

After intermediate moves (where White unsuccessfully attempted to shift the focus of the battle to the king’s wing), this characteristic rook deployment emerges. It not only removes the queen’s rook from the X-ray attack by the opponent’s fianchettoed bishop but also prepares a wing offensive (b7-b5, etc.)

17. Qd2 Nd4

Stronger is 17… Rxf1+! 18. Rxf1 Qa5 19. Qxa5 (19. N4c3 Rf8 20. Bd5+ e6 21. Rxf8+ Kxf8 22. Be4 Kg8) 19… Nxa5 20. N4c3 Rf8, when Black has a superior pawn structure and the bishop pair in a complex rook and minor piece endgame. (AB) 

 18. Nf4

Presumably intending 19. Nd5 e6 20. Ndf6+ Bxf6 21. Nxf6+ Kh8, with an unclear position; but White could immediately equalise with 18. Rxf8+ Qxf8 19. Rf1 Qe8 20. Kh1 Bc6 21. Nxd4 Bxd4 22. Nc3 Be5 23. Nd5 Bxd5 24. Bxd5+ e6 25. Be4, with chances for both players. (AB) 

  18… Bc6 19. Nd5 ?

Since the player with the white pieces does not possess a far-reaching bishop pair, the line opening: 19. b4 cxb4, etc., would have a very double-edged effect for him. Therefore, he prefers cautious, short-piece manoeuvres instead of a lively pawn offensive.

Nevertheless, this is a premature move. White should first prepare with: 19. Nc3 Qd7, and only now, 20. Nfd5 with 20… Be5 21. b3 Rxf1+ 22. Rxf1 Rf8 23. Rxf8+ Kxf8 24. Ne2 Qe8 25. Qe3 Nxe2+ 26. Qxe2. The drawback to Duchamp’s choice is that Black can now capture on d5, thus creating a dead zone in White’s central installations.

19… a5

The engine considers the text an error; no ifs, ands or buts. After 19… Bxd5 20. cxd5 Be5 21. Kh1 b5 22. Qe3 Qc7 23. Qd2 Qb6 24. h4 Qc7 25. Rbe1 Rxf1+ 26. Rxf1 Qd7, Black threatens devastation on the queen side while White’s kingside activity is moribund and notional. (AB). 

 

20. Rxf8+ 

Not 20. a3 due to 20… Nb3 followed by 21… a4, leading to a blockade.

20… Qxf8 21. Rf1

Accepting the Trojan gift would be ill-advised: 21. Qxa5, specifically due to 21… Bxd5 22. cxd5 Nf3+! 23. Kh1 Qf4 24. Bxf3 Qxf3+. 

21… Qd8 22. Kh1

Missing an opportunity to restore parity. Correct was, 22. Nec3 Qd7 23. Nb6 Qe8 24. Nbd5 Kh8 (24… Be5 25. b3 Qd7 26. Kh1 Kg8 27. Qe1 Re8 28. Ne2 Bxd5 29. Bxd5+ e6) 25. b3 Qd7 26. Qe1 Re8 27. Be4 Be5 28. Kg2 Ne6 29. Qe3 Rf8, when both players will have chances. (AB) 

22… Be5

Both sides continue to underestimate the power of deadening White’s prospects by capturing on d5.  Black should punish White’s last inaccuracy: 22… Bxd5 23. cxd5 a4 24. Qf2 (24. Nc3 Qd7 25. Be4 b5 26. Ne2 b4 27. Nxd4 Bxd4 28. b3 axb3 29. axb3 Ra8) 24… Kh8 25. Ng3 (25. Nc3 Qa5 26. Qd2 Nb5 27. Qe1 Nxc3 28. bxc3 Qxc3 29. Qxe7 Qe5) 25… Qd7 26. Qe3 Be5 27. b3 Ra8 28. h4 Qg8 29. Qf2 axb3 30. Nxb3 31. axb3 Ra3, when White’s queenside is in tatters and any kingside compensation, illusory. Black’s b-pawn will soon be extremely dangerous. (AB). 

23. Qe3 

Pointless would be 23. Qf2 due to 23… Bxd5 24. cxd5 Qe8, etc. The move sets a double trap  for the opponent:

1) 24. Nxc5 dxc5 25. Qxe5, resulting in a pawn gain and exchange of a bishop;

2) 23… Nc2? 24. Qf2, now infiltrating on f7.

23… b5 

A generous response that seemingly ignores the opponent’s threat of pawn gain actually aims for a highly strategic goal:to control the b-file. This creates a tense rivalry between the two open rook files: “b” (for Black) and “f” (for White)!

Yet, once again, slightly better for Black is: 23… Bxd5 24. cxd5 Qc7 25. Qf2 Kh8 26. Nd2 Bg7 27. Qe3 a4 28. h4 Qd7 29. Qh3 Kg8 30. Be4 Rf8 31. Rxf8+ Kxf8, when Black enjoys superior coordination. (AB) 

24. Nxc5 bxc4 25. dxc4 

Stronger is 25. b4 c3 26. Nb3 Nc2 27. Qa7 Kh8 28. bxa5 Bxd5 29. Bxd5 Nb4, with chances for both sides. (AB) 

25… Rxb2 

The infiltration occurs with a simultaneous double threat:

1) The knight gain via 26… Re2 followed by 27… dxc4;

​2) The uncompensated pawn gain via 26… Ra2.

26. Ne4 

An error and a losing move. 26. Nd3 Re2 27. Qg1 Bg7 28. Rf2 Rxf2 29. Qxf2 Bxd5 30. cxd5 Qb6 31. Qf1 Qa6 32. Nf4 Qc8 33. Qf2 Qc5, is still bad, but keeps the game within bounds of salvation. (AB) 

26… Bxd5 27. cxd5 Rxa2 

A small yet crucial gain is present. However, in the subsequent phase of the game, the White player – as expected – attempts various tactical manoeuvres.

28. Nd2 a4 29. Nc4 Re2 ?

Dropping both ball and victory. Black should continue: 29… Bg7! 30. Be4 Re2 31. Qf4 Qf8 32. Qxf8+ Bxf8 33. Bd3 Ra2 34. h4 Bg7 35. h5 Ne2 36. Bxe2 Rxe2 37. hxg6 hxg6; winning. (AB) 

30. Qd3 Bg7 31. h4 Qc8

Black’s initiative and strong passed a-pawn provide excellent winning chances. (AB) 

32. Rf4 Re1+ 33. Kh2 Nb3

Considerably stronger is 33… Rc1! after which, White’s c4-knight is a goner: 34. Nb6 Qb7 35. Nc4 Qb5 36. Re4 (to avoid the deadly pin along the h2-b8 diagonal) Qxc4 37. Qxc4 Rxc4 38. Rxe7 Be5+ 39. Kg1 Rc8, and the presence of an additional knight is sufficient for an easy win. (AB) 

34. h5 Qc5 ??

The additional punctuation may appear harsh, but in a single move, Black has thrown away a completely winning position.

After 34… Nc1 (isn’t this why Black played 33… Nb3?), it is nigh-impossible to find a haven for the White queen: 35. Qa3 (35. Qd2 Re2 36. Qxc1 Be5 37. Kh3 Bxf4 38. Qxf4 a3! 39. Bf1 a2 40. hxg6 hxg6 41. Bxe2 a1=Q and Black is completely winning) 35… Ne2 36. Re4 Bd4 37. Qd3 Rc1 38. Qxe2 (38. Ne3 Qc3 39. Qxe2 Re1 40. Qxe1 Qxe1 41. Rxd4 a3! 42. Re4 a2 43. Nc2 Qh4+ 44. Kg1 Qxg5, and Black can queen his a-pawn at his leisure) 38… Qxc4 39. Qxc4 Rxc4 40. Rxe7 Be5+ 41. Kg1 a3! 42. Re8+ Kf7 43. Ra8 Rc1+ 44. Kf2 Rc2+ 45. Kf1 a2, and wins. (AB) 

35. Bf1

White misses a tactical save. 

After 35. hxg6! the position is miraculously equal: 35… Qg1+ (no better is 35… Be5 36. gxh7+ Kg7 37. Nxe5 Qg1+ 38. Kh3 Re3+ 39. Bf3 Qh1+ 40. Kg3 Qg1+ etc.) 36. Kh3 Nd4 37. gxh7+ Kh8 38. Qg6 Re3+ 39. Nxe3 Qxe3+ 40. Rf3 Nxf3 41. Qe8+ Kxh7 42. Qh5+ Kg8 43. Qe8+, and again, there is nothing more than a perpetual check. (AB) 

35… gxh5

Another blunder which this time, potentially gifts White, not equality, but a significant advantage. After 35… Be5!! 36. Nxe5 dxe5 37. Rf3 (37. Qg3 Re3 38. Rc4 Rxg3 39.  Rxc5 Nxc5 40. Kxg3 Ne4+ 41. Kg2 Nd2 42. Ba6 a3) 37… e4 38. Qc3 Qxc3 39. Rxc3 Rxf1 40. Re3 Rf2+ 41. Kg1 Rf4 42. Kg2 Rxg4+ 43. Kf2 Rxg5 44. hxg6 hxg6, Black has a totally won endgame. (AB) 

36. gxh5

The last tactical surprise opportunity presented itself with 36. Qf5 (threatening 37. Qf7+ Kh8 38. Qe8+) 36… Kh8

37. gxh5 Nd2 38. Nxd2 Re5 39. Qg4 h6 40. Rxa4 Qf2+ 41. Bg2 Rxg5 42. Ra8+ Bf8 43. Ne4 Rxg4 44. Nxf2, when White would have been a knight for a pawn ahead in a winning endgame. (AB) 

36… Re5

Correct was 36… Be5 when Black was once again winning: 37. Nxe5 dxe5 38. Rf5 Nd4 and Black is on the verge of victory. After the text, Black’s advantage is tangible, but not conclusive.( AB) 

37. Rf5

The comedy of errors is on a long run. Had White deigned to venture, 37. Bh3, he would have equalised the position: 37… Rxg5 (37… Re1 38. Rf1 Rxf1 39. Be6+ Kh8 40. Qxf1 Qd4 41. Qf7 Be5+ 42. Nxe5 Qxe5+ 43. Kh3 Qc3+ 44. Kh2 etc.) 38. Be6+ Kh8 39. Qf1 Rxh5+ 40. Kg2 Rg5+ 41. Kh3 h5 42. Rf5 Qg1 43. Qxg1 Rxg1 44. Rxh5+ Bh6 45. Rxh6+, and White will likely need to sacrifice the knight for one or both of Black’s remaining pawns, to draw. (AB) 

37… Rxf5 

And after six consecutive horror-show shuffles by both contestants, Black suddenly sees the light, and a way to proceed without further let or hindrance. (AB) 

38. Qxf5 

The current endgame phase remains tactically intriguing. Considering the previous course of the battle, especially from a strategic standpoint, it is now time to apply various manoeuvres.

38… Nd4 39. Qe4 a3 40. Bd3

Threatening 41. Qxh7+ and (in response to 40… a2 because of 41. Qxe7!

  40… Nf3+ 

A beautiful decoy sacrifice that aims to gain an important tempo. Quite different would be 40… a2 due to 41. Qxe7!

41. Qxf3 

Forced.

41… a2 42. g6 

Or 42. Qe4 Qf2+ 43. Kh3 a1=Q 44. Qxh7+ Kf8, and Black wins.

42… hxg6 43. Bxg6 Bf6 44. Qg2

 The last try.

44… a1=Q 45. Bh7+ Kxh7 46. Qg6+ Kh8 47. Qe8+ Kg7 48. Qg6+ Kf8  White resigns 0-1

There is no defence: 49. Qh6+ (49. Qg2 Qxc4 50. Qf2 Qg4 51. h6 Qc1 52. Qg3 Be5 53. Kg2 Qxg3 checkmate) 49… Bg7 50. Qf4+ Kg8 51. Ne3 Be5 52. Kg2 Bxf4 53. h6 Qb2+ 54. Kf3 Qxe3+ 55. Kg4 Qbe2+ 56. Kh4 Qg3 checkmate.

What sort of player was Marcel Duchamp? The  Bulletin of the French Chess Federation  of 1924 described him thus: 

“  … étant donné son jeu profond et solide… sa froideur imperturbable, son style ingénieux… font de lui un adversaire redoutable ”. (“Given his profound and deep play… his imperturbable coolness and his ingenious style, together these qualities make him a redoubtable adversary.”)

Note those opening words:  étant donné , they were to  recur in Duchamp’s artistic career. Many people have been puzzled by the similar title of Duchamp’s final installation from 1946. It is my belief that they hark back to this first official recognition of and encomium to Duchamp’s chess skills. 

Alternative explanations are, of course, also possible. For example, the “given” references, may lean in a completely different direction. Pointing the viewer back to specific geographies of Paris, and Duchamp’s time there before his emigration.

It was bad luck for Duchamp that one of his best known games is that loss to Le Lionnais from Paris, 1932. This game happened to be annotated (see above) by the celebrated Franco-Polish grandmaster, Ksawery Tartakower, in the authoritative Austrian magazine,  Wiener Schachzeitung . This brought the game a certain prominence, but Le Lionnais himself was also not slow to trot it out (e.g. in his interview for  Studio International , 1975) whenever he was asked about Duchamp and chess. 

It should be added that Duchamp had drawn a game with grandmaster Tartakower himself in 1928, and later in the Olympiad at Hamburg 1930, he held the American grandmaster and World Championship contender Frank Marshall to a draw. Having already beaten him in a simul in 1923, in 1929, he again beat Koltanowski, the four-time Belgian Champion and one-time World Record holder for the greatest number of opponents faced in a blindfold simultaneous display. The tactical tricks Duchamp produced in that game, his evident love of paradoxical solutions, bowled over his distinguished opponent with extreme speed.

The opening moves were as follows:

White: George Koltanowski
Black: Marcel Duchamp
Paris Tournament, Round 8, 1929
Indian Defence

1. d4 Nf6
2. c4 e6
3. Nc3 d6
4. e4 b6

 

This type of development of the Queen‘s Bishop was closely associated with Aron Nimzowitsch, the celebrated iconoclast, chess teacher and grandmaster of the 1920s. Indeed, the whole strategy of holding back the Black centre pawns was enthusiastically advocated by Nimzowitsch in his books and articles.

Much of the accepted wisdom about Duchamp as a chess player stems from the 1975 interview with Le Lionnais in Studio International. There Le Lionnais stated:

“ In Duchamp’s style of play I saw no trace of a Dada or Anarchist style, though this is perfectly possible. To bring Dada ideas to chess one would have to be a chess genius rather than a Dada genius. In my opinion Nimzowitsch, a great chess player, was a dadaist before Dada. But he knew nothing of Dada. He introduced an anti-conformism of apparently stupid ideas which won. For me that‘s real Dada. I don’t see this Dada aspect in Duchamp’s style… Duchamp applied absolutely classic principles, he was strong on theory – he‘d studied chess theory in books. He was very conformist which is an excellent way of playing. In chess, conformism is much better than anarchy unless you are a Nimzowitsch, a genius. If one is Einstein, one says the opposite to Newton, of course; if one is Galileo, one contradicts everyone, but otherwise it’s safer to be a conformist .”  

I suggest, however, that Le Lionnais  was simply unaware of the degree to which Nimzowitsch, author of the incredibly influential book,  My System , and a potential Candidate for the World Championship by 1927, had shaped Duchamp’s style. Even the opening of the initial game that Duchamp lost with the white pieces to Le Lionnais (1. c4 c5 2. Nc3 Nc6 3. g3 g6 4. Bg2 Bg7 5. d3 d6 6. e4) had been introduced by Nimzowitsch at Dresden in 1926. 

Similar instances of Duchamp borrowing Nimzowitsch’s opening ideas proliferate throughout his games. To conventional players of his day, Nimzowitsch appeared iconoclastic, anything but a conformist. From their chess styles, Le Lionnais comes across as the solid conformist, while Duchamp clearly admired Nimzo’s iconoclasm, a quality he most definitely promulgated in his art and sought to emulate in his chess. 

 

Ray’s 206th book, “  Chess in the Year of the King  ”, written in collaboration with Adam Black, and his 207th, “  Napoleon and Goethe: The Touchstone of Genius  ” (which discusses their relationship with chess) can be ordered from both Amazon and Blackwells. His 208th, the world record for chess books, written jointly with chess playing artist Barry Martin,  Chess through the Looking Glass ,  will be available from Amazon before Christmas. 

A Message from TheArticle

We are the only publication that’s committed to covering every angle. We have an important contribution to make, one that’s needed now more than ever, and we need your help to continue publishing throughout these hard economic times. So please, make a donation.