The Oldest Chess Opening?

Although many chess fans find it boring, in my opinion it's absolutely fascinating that the Giuoco Piano, or Italian Opening, is so popular again at top grandmaster level. Isn't it wonderful that some of the oldest books about modern chess already mention the moves 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5, and that these very same moves are routinely played in top tournaments by the likes of Carlsen, Kramnik and Anand?

But is it the oldest chess opening, as I've heard some commentators say?

Well, it's very old, that's for sure - and it's true that the oldest printed chess book on modern chess, Repeticion de amores e arte de axedres con CL iuegos de partido (Lucena, 1497) already mentions it.

(The front page of Lucena's 1497 book on modern chess)

One of the lines Lucena analyzes goes as follows:

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.d3 Nf6 5.h3 d6 6.Bb5 ("pinning the knight") 6...a6 7.Ba4 and in this position Lucena mentions not only 7...Bd7 and 7...b5 but also 7...Rf8 followed by 8.Nc3 Kg8 - an ancient 'king's leap' which was typical for the early days of modern chess. (It's interesting that super-GM Wei Yi has recently played this line with White a few times, although he preferred 6.c3.)

You can't really go further back than Lucena when it comes to modern chess opening theory, because Lucena's is the earliest known book devoted to modern chess, defined by the increased strength of the queen and bishop. (An earlier book on modern chess, was unfortunately lost 200 years ago.) However, there is one older game featuring modern chess rules in which a different opening was played: the Scandinavian Defense.

This 'game' (I've put 'game' between quotes because it's an invented game) is from the Scachs d'Amor manuscript, about which I've written before, and which is dated between 1475 and 1480. This manuscript, its poem and the associated game are regarded as the 'birth' of modern chess - arguably the most important milestone in the entire history of modern chess. 

So, the Scandinavian (although it wasn't called like that, of course) is definitely older than the Italian Game if you go by the evidence. 

Can we go even further back in time? 

Well, what about medieval chess opening theory? There's surprisingly little known to us about this subject. One of the main reasons is that medieval chess was a rather slow game (the queen and bishop were both very weak pieces) and it didn't really make sense to analyse positions in which play developed very slowly. There were simply too many ways to develop one's pieces without interfering with the opponent until a later stage in the game.

Instead, pre-modern treatises about chess tended to focus on chess studies and problems, many of which were rather intricate and difficult for us to understand because of their peciliar stipulations. For instance, problems often stated that certain pieces were not allowed to move, or that mate needed to be given with a specific piece or in a specific order.

Some things can, however, be said about pre-modern chess opening theory. The great chess historian Murray (A History of Chess, 1913) writes: 

[T]he mediaeval player possessed very little knowledge of the relative values of the pieces, or of the underlying principles of play. Here he remained far behind the Muslim players of the 10th century. He, of course, recognized that the Rook and Knight were the strongest of the chess forces, and he relied almost entirely upon them when he had brought them into play. But he knew very little of the value of the other pieces. The main use that he made of his Queen was to keep her in close attendance on the King to interpose her when the opponent's Rook checked from the other side of the board. The Muslim masters manoeuvred from the first to secure a road for the Queen into the heart of the enemy's position, the European kept her near home. (...) 

Murray then mentions the so-called Crakow Poem or De Ludus Scaccorum (probably late 13th or early 14th century, though the manuscript dates from a late time, around 1422), in which a number of chess 'openings' (or, rather, 'set-ups') are described. Many involve early play of the a- and h-pawns to make way for a precursor of the 'king's leap', whereby the king would be 'protected' on the queen- or kingside in a 'hut'. For instance, one such opening scheme goes as follows (note that bishops, while limited in range, could jump over pieces under the old rules):

1.e4 2.d4 3.Ne2 4.h4 5.a4 6. Bh3 7.Ba3 8.g3 

(Miniature from the 'Carmina Burana' codex, around 1230, which may show the formation of a 'hut' on the king's side.) 

Murray also mentions several German sources (from the early 16th century) describing pre-modern chess rules in which 'huts' are made, for instance the so-called 'Iron Ward' (eisern Hutt), which features a king's leap and a queen's leap as well: 1.h4 2.e3 3.Bh3 4.Nf3 5.Kg1 6.Qf1

Murray mentions that in the Schachzabel Spiel (ca. 1520), the beginner is advised "always to begin by moving the Queen's Pawn" and that it is "usual to move it to d4." He continues with the following important conclusion: 

We may accordingly regard the Queen's Pawn Opening as the regular commencement in the older European chess - a conclusion which is supported by the description of games which are to be found in someof the mediaeval romances. (...) Two miniatures of early positions of games in progress also support the popularity of the Queen's Pawn Opening in the Middle Ages.

And that's about it, really. Apart from some obscure pawn sequences, initiated by the play of the d-pawn, no concrete opening 'theory' appears to have made it down to our day and age.

Or has there? A rather casual remark from Murray, a bit further in the chapter, got me thinking. What if some of the modern chess opening theory is really a remainder of older, medieval theory? Here's what Murray writes: 

It has sometimes been supposed that the so-called Damiano Gambit is a survival of old chess. It may be so as far as 1.e4 e5 2.Nf3 f6 - quite a good defence in the older game, since 3.Nxe5 was utterly pointless. 

This strikes me as a very sensible remark. (Actually, it seems to be 'common opinion' among chess historians.) The Damiano Gambit, or Damiano Defense, has always been the joke of chess opening theory. 2...f6 - seriously? And what's even more curious, is that there is a lot of theory about this line. If you look in the old books on chess, you will find pages of analysis of the consequences of 3.Nxe5 fxe5 4.Qh5+ which, as everybody knows, simply wins for White. 

But in the old game, where queen and bishop were almost powerless pieces, the weakening of f7 and the a2-g8 diagonal wasn't a big deal. And I suspect that the weakness of the f7 point, now something that every schoolkid learns on Day 1, wasn't a concept that was well understood anyway in those early days. Take the following analysis from the second regla of Lucena's book:

The line 1.e4 e5 2.Nf3 f6, then, was probably included in so many early books on modern chess because it was opening theory under the old rules: a common opening in the medieval game. In those pristine days of modern chess, people still played it, thinking it was a decent way to defend the e-pawn. Which it wasn't anymore - not under the new rules.

Damiano himself, whose book Questo libro e da imparare giocare a scachi et de le partite was published in 1512 (i.e. 15 years after Lucena), wrote of 2...f6 that it was "the worst" way to defend the e-pawn. (It was only decades later that his name got to be associated with this inferior line, by Ruy Lopez.)

(The first page of the 4th edition of Damiano's book, ca. 1530)

Damiano went on to analyze 3.Nxe5 fxe5 4.Qh5+ (although he gave 4.Qh6 in earlier editions of his book, much to the amusement of some later authors) 4...Ke7 5.Qxe5+ Kf7 6.Bc4+ d5 7.Bxd5+ Kg6 8.h4! after which there is no defense for Black. This line still stands today as the variation's refutation.

Of course, everything after Black's second move is new opening theory. Under the old rules, as Murray noted, 3.Nxe5 didn't make sense. Nothing is known about how play continued under the old rules - not even in the Damiano Defense.

So, that's all we have: two moves of documented 'old' opening theory - we don't know how play continued after that. But it does explain why there's such a thing as the 'Damiano Defense' at all. And why it was analyzed, even more than 500 years ago. It also emphasizes that the chess masters of the past weren't 'crazy' at all to analyze such obviously bad openings.

They analyzed exactly the right opening moves, both under the old rules and the new. Any modern chess tournament still proves it.