THE MESSENGER - the Saavedra myth exposed
PS 1 December 2001: See item 151 in my Open Chess Diary

One of the greatest stories in chess literature is the Saavedra myth. A sensational discovery, made in August 2001 by the Irish chess historian David McAlister, gives me reason to summarize it again.

White to play and win
G. Barbier and F. Saavedra
Glasgow Weekly Citizen, 1895

1.c7 Rd6+ 2.Kb5 Rd5+ 3.Kb4 Rd4+ 4.Kb3 Rd3+ 5.Kc2 Rd4(!) Now 6.c8Q Rc4+ 7.Qxc4 is stalemate, and 6.Kc3 Rd1 7.Kc2 Rd4 is a repetition of moves. 6.c8R! Ra4 7.Kb3 and White wins.

The background of this celebrated position had long been unclear (it was attributed to Lasker; a game position; a composition by Saavedra), until the Dutch endgame composer and writer John Selman reconstructed it in 1940, in his article 'Who was Saavedra?' The 'Saavedra position' turned out to be a chance composition by the chess columnist of the Glasgow Weekly Citizen, Barbier, inspired by a game Fenton - Potter, London 1875. However, Barbier thought it was a draw after 5...Rd4, and published it as such on 4 May 1895. In his column of 11 May, he gave the solution with the stalemate. A few days later, a reader, the Spanish priest Fernando Saavedra, came to the Glasgow Chess Club and showed the winning Rook promotion to the three chess players present, Barbier among them. Selman: 'The surprise was great when Saavedra told Barbier that the so-called draw was a win for White, and the surprise of those present was transformed into enthusiastic bewilderment when Saavedra demonstrated his find for the first time!'
    Such underpromotions, Selman adds, were not commonplace at the time.
    In his next column, of 18 May, Barbier wrote: 'A member of the Glasgow Club, the Reverend Saavedra, has pointed out a win for White. The position is a very remarkable one.' Finally, on 25 May, (see The Discovery of the Saavedra, elsewhere on this site), Barbier published the winning Rook promotion, adding: 'This position is one of the most remarkable end games we have seen for years.'
    Thus Saavedra, a mediocre player, became a chess myth: he achieved immortality with his only surviving move.

A few days ago, I received an email from David McAlister whose discovery changes this story completely. In researching the career of James Alexander Porterfield Rynd (1847-1917), the first Irish chess champion, he had found a chess column that Porterfield Rynd had published in the Dublin Saturday Herald on that same 25 May 1895.
    Under the heading Clontarf to the Fore (Clontarf being a Dublin suburb and the name of a prominent chess club at the time), he commented on Barbier's "excellent contribution to the Weekly Citizen" of the week before. "It is right to say," Porterfield Rynd wrote,

"that (...) Barbier's latest version acknowledges the finishing touch to have come from the Rev. F. Saavedra, formerly a member of the Clontarf Club, (...) but now a valued member of the Glasgow Club. Now as all the points of this instructive ending practically occurred three or four years ago at the Clontarf Club in one of a number of simultaneous games played there by your contributor, it can hardly fail to be of interest to note the Clontarf position and its concluding moves, as hereinunder given.

Black, Lieutenant-Colonel W. Lynam

White, P.R. to move

1.P-B7! RxPch 2.K-Kt4 R-K5ch 3.K-B3 R-K8 4.K-B2 R-K5! One of the Colonel's ingenuities - to procure stalemate by a check at KB5 if P Queens. 5.P-B8 (becoming a R) R-KR5 6.K-Kt3 Resigns"

So on 25 May 1895, two premieres of the Rook promotion coincided - in a Glasgow paper, and in a Dublin paper. But the Dublin game had preceded the Glasgow composition by a few years.
    Inevitably, the thought of a mystification arises. Porterfield Rynd could have found the solution to Barbier's 18 May position by himself, and have liked it enough to publish the Rook promotion as his own idea.
    Several arguments plead against that.
- Porterfield Rynd was a respected and prominent member of the Dublin establishment. He was a barrister who, for that reason alone, could not afford to publish fabrications, especially not with himself as the hero.
- And certainly not at the expense of a colonel. (He did not invent the colonel. In a 1892 Chess Player's Annual and Club Directory, "Lieutenant-Colonel Wm. Lynam" was mentioned as the President of the Clontarf Chess Club.)
- He would have falsely accused a priest of trying to steal his idea.
- He would have been exposed immediately.
- There wasn't any special reason to steal the move - there was no famous Saavedra position yet.
- He wouldn't have invented a simul game. Fact is stranger than fiction; how were people going to believe he had seen something that deep at such a frivolous occasion?
- He wouldn't have given 1.f7 an exclamation mark. That exclamation mark adds a touch of naiveté and sincerity. If Rynd had invented the position, he would probably have chosen 1.Kg4 (Kg6 wins, too) as the first move, because it is part of the central idea. 1.f7 is just a practical move that you can play without thinking - ideas come later.

I realize the last two arguments are double-edged, but the others are not.

But suppose Porterfield Rynd did invent the game. How would that have happened?
    He sees the 18 May Glasgow Weekly Citizen chess column with a position in which his old club-mate Saavedra is said to have found a 'remarkable win'. He says to himself: "If that patzer can do it, I can do it, too," - and his breath is taken when he finds the Rook promotion. How beautiful. Why didn't he think of something like that himself? But - he just did think of it himself, it is his idea. But if he publishes it as a solution to Barbier's position, he will just be one among many solvers. To really make it his own, he must invent a game in which he played it. That is not so difficult, and as an opponent he chooses the old Colonel - he won't mind or even notice; he forgets his games as soon as he's played them. And if he makes it a simul game played at the Clontarf Club, a few years back, when Saavedra still came there, readers will think the priest got the move from him, and is now showing it around in Glasgow as his move.
    No... it will never work, he will be found out before the ink of his lie is dry, he will probably be disbarred, but who cares - to have been seen as the creator of such an endgame, if only for a few hours...

It is a possible scenario - but not a very likely one. The assumption that the Rynd vs. Lynam game is genuine allows a simpler story.
    That Saavedra, before coming to Glasgow, had lived in Dublin (he went back there, and died there in 1922) was already known to Selman. As McAlister discovered, he had been a member of the Clontarf Chess Club at the time Porterfield Rynd played the miracle move there. Saavedra not having known about this move, and later discovering it by himself, defies probability.
    It must have happened like this. When Saavedra, in Glasgow, saw the (mirrored) Clontarf-position in the paper, wrongfully presented as a draw, he rushed to the club to tell about the beautiful winning move that he knew. That he hadn't found that move himself was a detail that was lost in the ensuing bewilderment and enthusiasm.
    Or maybe Saavedra didn't mention that - but there really is no reason to accuse him of dishonesty. The point is that when Barbier finally published the Rook promotion on May 25th, it was not Saavedra's move, but Rynd's move that Saavedra knew about.

One can imagine how Porterfield Rynd, on the other side of the Irish Sea, felt when he saw Barbier's 18 May column about a remarkable move that Saavedra was supposed to have found in a mirrored version of his game: "Why, that scoundrel is stealing my move!" And he rushed to the Herald's offices to set the matter straight.
    What remains is the remarkable coincidence of Saavedra having been close to two births of the 'Saavedra position'. Amazingly, that perfectly natural position had never occurred in a known game until McAlister found Rynd vs. Lynam - the game in which the whole study was invented. The 'Saavedra' can be stricken from all endgame study anthologies - it was a game.

This is a fresh discovery - many questions remain, most of which will never be answered. How did Porterfield Rynd react when after 1902 his beautiful win vs. Lynam became widely known, but was ascribed to Saavedra or Lasker? Did he, or anybody else, ever publish that game again? When Saavedra returned to Dublin, did he and Rynd meet again and discuss the case? Did Saavedra try to pass as the discoverer of 6.c8R, or was he just an enthusiastic messenger? Was Saavedra present when that simul was played - could he even have been one of the other participants? Did Barbier see Rynd's 25 May column, and did Saavedra and he discuss it? Did Rynd and he discuss it? Why didn't Porterfield Rynd publish his Lynam game as soon as he had played it? - but McAlister already answered that one: Rynd's chess column started in March 1892; he must have played that game before then.

Porterfield Rynd first became chess champion of Ireland at the age of 18, in 1865, and except for the years 1886-1892, he remained champion until 1913. In 1885, his initiative led to the creation of a first Irish Chess Association. He won a match against Burn, and drew one against Mason, but he was always an amateur. He was a prominent barrister, a good piano player, he composed melodies, played tennis, rowed, swam and played billiards, he invented Diamond Chess, a variant that is still sometimes played - and he discovered one of the most famous and beautiful moves in chess.
    As to Saavedra - his myth is unsurpassable now. He became immortal for a move he had just heard about.

© Tim Krabbé, 2001

Top of the page | Main chess page | Main page