Metamagical Themas is an eclectic collection of articles written for Scientific American during the early 1980s by Douglas Hofstadter, and published together as a book in 1985 by Basic Books (ISBN 0465045669) .

The subject matter of the articles is loosely woven about themes in philosophy, creativity, artificial intelligence and important social issues. The volume is substantial in size and contains extensive notes concerning responses to the articles and other information relevant to their content.

Major themes include: self-referentialism in language; art and logic; discussions of philosophical issues important in cognitive science/AI; and lengthy discussions of the work of Robert Axelrod on the prisoner's dilemma and the idea of superrationality.

The concept of superrationality and its relevance to the Cold War, environmental issues and such is accompanied by some amusing and rather stimulating notes on experiments conducted by the author at the time.

Many other topics are also mentioned, all in Hofstadter's usual easy, approachable style. Another feature is the inclusion of two dialogues in the style of those appearing in Gödel, Escher, Bach. Ambigrams are mentioned.

There are three articles based around the Lisp programming language, where Hofstadter first details the language itself, and then shows how it relates to Gödel's Incompleteness Theorem.

The title is an example of wordplay: it is an anagram of Mathematical Games, the title of Martin Gardner's column that Hofstadter's column succeeded in Scientific American.

French Edition

Metamagical Themas was also published in French, under the title Ma Themagie (InterEditions, 1988), the translators being Jean-Baptiste Berthelin, Jean-Luc Bonnetain, and Lise Rosenbaum.

Unfortunately, the wordplay was lost in the French title, and replaced with a much weaker one, about Math and Magic. The translators had contemplated Le matin des metamagiciens, which would have been a play on Jeux malins des mathematiciens (respectively, The Dawn of Metamagicians, and Clever Tricks of Mathematicians). However, the publisher found it too elaborate, so we are left with the simpler one.