A Game of Letters

Born on April 13, 1899: Alfred Mosher Butts, inventor of Scrabble®.

If you play Scrabble®, there are probably a few letters you groan to find in your tile rack. Maybe you find the "V" hard to play, or the "C." But when you look at point values, the "C" is worth only 3, while the "H" (arguably much easier to play) is worth 4. Of course, the three big-ticket letters, X, Q, and Z, are prized possessions in a game and, used wisely, can be combined with spaces that multiply the points of a tile or a word to generate high word scores. Beyond the point values, there is the issue of how many of each letter appears in the tile set. There are, for example, 12 E's in the game, 9 A's, only 8 O's, 2 of popular consonants like M and H, and only 4 S's.

Have you ever wondered how the values and letter quantities were determined? Alfred Mosher Butts, who invented the game in 1938 as Criss-Cross Words, spent a great deal of time analyzing text samples during the development of the tile set. Attempting to pinpoint how common each of the 26 letters in the alphabet is in the English language, Butts manually tracked through the distribution of individual letters in text from sources like the New York Times. Based on his letter-frequency assessment, Butts established the breakdown of letters included in the set of 100 tiles and developed the corresponding point value distribution used in both Scrabble® and an earlier version of the game, Lexiko, both of which used the same point-based tile sets.

Making Connections

Butts' assessment of the English language was driven by his desire to create a challenging word-based game. But similar studies are carried out by scientists and scholars to analyze samples of writing, determine authorship and historical accuracy, and pinpoint other linguistic, lexicographic, and etymological trends. Students curious about lexicography, or even the newer discipline of stylometry, defined as "the science of measuring literary style," can conduct their own analyses of literary samples—either by hand, as Butts did, or using a variety of computer-based tools, programs, and algorithms. The Computer Sleuth: Identification by Text Analysis project from the Computer Science area lets students dive into this word-worthy area of research.