First Links in the Markov Chain
Probability and poetry were unlikely partners in the creation of a computational tool
The Bootless Academician
Markov’s Onegin paper has been widely discussed and cited but not widely read outside of the Russian-speaking world. Morris Halle, a linguist at MIT, made an English translation in 1955 at the request of colleagues who were then interested in statistical approaches to language. But Halle’s translation was never published; it survives only in mimeograph form in a few libraries. The first widely available English translation, created by the German scholar David Link and several colleagues, was published only in 2006.
Link has also written a commentary on Markov’s “mathematization of writing” and an account of how the Onegin paper came to be known outside of Russia. (A crucial figure in the chain of transmission was George Polya, a Hungarian mathematician whose well-known work on random walks is closely related to Markov chains.) The statisticians Oscar Sheynin and Eugene Seneta have also written about Markov and his milieu. Because I read no Russian, I have relied heavily on these sources.
In the accounts of Link, Seneta and Sheynin we find the dénouement of the Markov-Nekrasov conflict. Not surprisingly, the royalist Nekrasov had a hard time hanging onto his position after the 1917 Bolshevik revolution. He died in 1924, and his work fell into obscurity.
Markov, as an anti-tsarist, was looked upon more favorably by the new regime, but an anecdote about his later years suggests he remained a malcontent to the end. In 1921 he complained to the Academy that he could not attend meetings because he lacked suitable footwear. The matter was referred to a committee. In a sign of how thoroughly Russian life had been turned upside down, the chairman was none other than Academician Maxim Gorky. A pair of boots was found for Comrade Markov, but he said they didn’t fit and were “stupidly stitched.” He continued to keep his distance from the Academy and died in 1922.