As it happened, returning to London last winter from the Hastings Tournament, I found myself on the same train as that long time friend and well-known chess writer Leonard Barden, who was traveling with his friend, Raymond Keene. With the possible exception of Jonathan Penrose, Keen is the strongest English player and a good grandmaster possibility.
As you have probably guessed, in no time at all we found ourselves engaged in a lively conversation about any chess subject which came to mind. Rather unexpectedly, at least for me, just when we were debating the constituents of natural talent in chess, Barden produced the following position on his pocket set:
"Try this one, Miro," he said with a mischievous smile. "Try what?" I asked, whereupon he proceeded to explain my task - and yours if you want to try it. As he did, I will first tell you what the puzzle is and only then divulge its purpose.
First you have to transfer the Knight, making only legal Knight moves, from a1 to a8, stopping on the way at every square which is not occupied or controlled by a Black pawn. (All will soon become obvious, it is not required that any square be visited only once.) The Black pawns remain in their present positions throughout. The Knight must first aim for b1, then c1, and so on along the first rank until it reaches h1. For instance, a simple method of making the first objective is to play the Knight from a1 to c2, a3 and b1.
The second step, from b1 to c1, is more difficult because b2, c3 and d2 are occupied or controlled by a pawn. After reaching h1, the Knight must continue by going to h2, then to f2 (but not g2, of course), then to other available squares on the second rank as far as a2. The next phase is from a3 to h3, and then to zig-zag, one rank at a time, until the Knight reaches a8.
The idea is to complete the task as quickly as possible, so it will be helpful to have a friend time you and count your errors (each time you land on a square occupied or controlled by a pawn, you must add ten seconds to your score). Of course, you may time yourself, but keep in mind that the value of the test is completely negated if you cheat!
When you have finished the test, wait a few minutes and do it again.
And now: Why should you go to all this trouble just to jump all over the board with that damn Knight?
As Barden told me later, this was an experiment which originated in Czechoslovakia and which could be defined perhaps, as the nearest approach to a chess equivalent of an intelligence test for anyone, grandmaster or novice, who knows the moves of a Knight and a pawn. Revised in Czechoslovakia, it was tried there several years ago on a large national sample of school children who played chess. The test was timed and showed that four boys did far better than all the others in their age group, with one of the four significantly faster than all the others. The boy who performed best is now a world title contender: Vlastimil Hort. The other three, Kavalek, Jansa and Smejkal, are currently grandmasters or international masters.
Why, you may ask, should the time you spend on this test measure your chess IQ? The test measures rapid sight of the board plus some elements of flexible thought and pattern recognition, I was told. These are, as shown by other controlled experiments, important aspects of chess skill. Adriaan de Groot, a psychologist and former member of the Dutch chess team, showed players of varying strength an unfamiliar position from an actual game for five seconds and then asked them to reconstruct it from memory. Differences in chess ranking showed dramatically, with former world champion Euwe recalling the position perfectly, while average players made a dozen or more errors.
A player's speed at the test the first time measures his natural skill, while the second performance of the task measures his learning ability. Some people improve only slightly on the second attempt while others are much quicker the second time.
Now, of course, we come to a fundamental question: How long should a talented player take on this test? To this one could reply that there is no "solution" as there are to normal puzzles. But there are some clues. According to believers in this exercise in chess intelligence, if you did this test in under five minutes you are probably a potential international player (watch out for that word "potential" - there's a lot more to chess than mere talent!); less than four minutes may make you a budding grandmaster. That would seem to be the conclusion drawn from the results of several leading British and Czech masters, whose scores show a high correlation to their known playing strengths. Penrose and Hort took two minutes, Harston three, Smejkal, Jansa and Kavalek three and a half, Keene four, Clarke four and a half, Barden, Kottnauer and Webb five.
Strong club players should take around seven minutes. But to console readers who were much slower than this. Barden mentioned that one well-known chess personality (whose reputation shall not be sullied in these pages!) got the Knight from a1 to b1 and c1 with much difficulty, then abandoned the task in disgust! The test may also reveal a player's pattern of thinking. Smejkal, a brilliant but occasionally erratic player, hesitated for the best part of a minute in deciding on a route from f1 to g1, then rattled off an 11-move Knight tour between these squares. On the fourth and fifth ranks, there are often two equally obvious ways of getting from one square to its neighbor; you can reach e4 from f4 via e6 and c5 or via h3 and f2. In these situations, according to Barden's study, Kottnauer, an energetic and positive player, always chose the forward route, while Clark, a solid defender, selected the backward route. The best "learning ability" result, shown by a better score on the second run, was achieved by Keene, who improved from four minutes on the first attempt to two and a half on the second.