Poe’s Secret

Writer Edgar Allan Poe (1809–1849) is famous for his short stories of the mysterious and the macabre. His popular tale “The Gold-Bug,” published in 1843, is often cited as one of the best works of fiction that turn upon a secret message.

Poe had a longstanding interest in cryptology. When he became editor of Graham’s Lady’s and Gentleman’s Magazine in Philadelphia, he offered to solve cryptograms sent to him by readers and, while waiting for responses, wrote an essay about secret writing for the July 1841 issue of the magazine (see http://www.eapoe.org/works/essays/fwsw0741.htm).

When Poe ended the contest 6 months later, he claimed to have solved the 100 or so legitimate ciphers that readers had submitted. In his final “addendum” on the topic, published in the December issue, Poe included two ciphers from “a gentleman whose abilities we very highly respect” (see http://www.eapoe.org/works/essays/gm41sw03.htm). He attributed the ciphers and accompanying notes to W.B. Tyler and challenged readers to solve the puzzles. Poe never published the solutions.

No one seems to have paid much attention to the ciphers until 1985. That’s when Poe scholar Louis A. Renza of Dartmouth College suggested in an essay that Tyler and Poe were the same person. English professor Shawn Rosenheim of Williams College took up Renza’s theory and, in his 1997 book The Cryptographic Imagination, furnished additional circumstantial evidence pointing to Poe as the author of the two ciphers.

Meanwhile, in 1992, intrigued by Rosenheim’s ongoing research on the topic, Terence Whalen took a break from working on his dissertation to tackle the puzzle. Now at the University of Illinois at Chicago, he solved the first of Tyler’s cryptograms, which consisted of a long string of various typographic symbols. The key to the solution was Whalen’s recognition that the three-symbol pattern “, † §” —repeated seven times in eight lines—represented the word “the.”

The deciphered passage proved to be lines from the 1713 play Cato by English essayist Joseph Addison (1672–1719). Whalen’s solution, however, did not settle the question of who created the cipher.

The second cipher, which used several different symbols for each English letter in the text, was much more difficult.

Determined to obtain a solution, Rosenheim offered a prize of $2,500 for deciphering the cryptogram. The contest received a substantial boost in 1998 when Jim Moore of Bokler Software in Huntsville, Ala., which builds cryptographic components for software developers, created a Web site devoted to the puzzle (http://www.bokler.com/eapoe.html).

Rosenheim and Moore fielded hundreds of inquiries from around the world. The puzzle was finally solved in July by Gil Broza, a software engineer in Toronto. It turned out that the number of different symbols for a given letter depended on the relative frequency with which that letter appears in English text. So, there were two symbols standing for “z” and 14 symbols for “e.” Given the brevity of the cipher, would-be sleuths couldn’t count on applying information about letter frequencies to crack the code. Broza assumed that the symbols in the cryptogram were properly broken up into English words and proceeded from there, eventually using a computer to check for patterns and sort through various possibilities.

Broza’s solution revealed that the original cipher contained more than two dozen mistakes, which had been introduced by the typesetter or the puzzle’s originator. The deciphered text begins, “It was early spring, warm and sultry glowed the afternoon. The very breezes seemed to share the delicious languour of universal nature. . . .”

That certainly doesn’t sound like Poe, but the text echoes themes that he favored. The passage is probably taken from some unknown novel or story of the period, Rosenheim says. It’s still possible, though not certain, that Poe himself composed both ciphers.

There is something mysterious even in the decrypted cipher, Rosenheim adds, not only because we do not know who enciphered it but also because it reminds us of the uncanny and limited immortality writing sometimes affords.