For the preceding segment of this four part geneology, see Part 2
For the preceding segment of this four part geneology, see Part 2
In this third segment of our genealogy we begin to form a connection, and to examine those lesser-known but foundational figures that radiate out from Turing's early work. Perhaps appropriately, given the venue, this second figure leads us to one of the earliest examples of computational art ever produced, though he did not claim the title of artist for himself. This history also moves us forward to those pivotal years surrounding Turing's arrest and death. While Turing underwent a highly visible crisis, Christopher Strachey's work was coming into its own. Once again the connection is tenuous, and little record survives to document more than a passing relationship between these two men, but what remains is a surprisingly poetic attempt to play at the machine.
Christopher Strachey was born in 1916 in Hampstead, England to Oliver Strachey and Rachel (Ray) Costelloe.[i] The Strachey family may be familiar to some, as it has a long and distinguished history in England. Christopher's father Oliver served as an intelligence agent in the First World War and, along with Alan Turing, as a cryptographer at Bletchley Park in the World War II. Christopher's great-grandfather was Sir Henry Strachey, 1st Baronet, and the family has ties back to John Strachey, an associate of the philosopher John Locke.
Perhaps most well known is Christopher's uncle, Lytton Strachey who – along with Virginia Woolf, John Maynard Keynes, and E. M. Forster – was a member of the Bloomsbury Group, a widely influential group of writers and artists living in Bloomsbury, London in the first half of the twentieth century. Lytton is perhaps most famous for his biographical work Eminent Victorians (1918), which defied Victorian bibliographic norms through irreverent, comedic character assassinations of some of the most beloved moral figures of the Victorian era. The Bloomsbury Group is particularly famous for its modern views on feminism, pacifism, and sexuality. Much like Turing, Lytton was open about his homosexuality – at least between friends and other members of the Bloomsbury group – at a time when homosexuality was explicitly illegal. The Strachey family home was located at 51 Gordon Square, and Christopher would have grown up in the middle of the Bloomsbury group's most productive period.
The Strachey home in Bloomsbury.
Appropriately, Christopher Strachey is also best known for a series of literary works. In 1952 Strachey developed a love-letter generator that ran on the Manchester Mark 1 using a random number generating algorithm, predating the ELIZA natural language processing program by twelve years. The project is considered by many to be the first example of algorithmic or computational art, though such claims are always highly contested. As a mathematician and computer scientist, Christopher Strachey was also one of the founders of denotational semantics and a pioneer in programming language design; yet this is not the path Strachey began on as a young man growing up in Bloomsbury among artists and intellectuals.
By most accounts Strachey was an extremely intelligent child but an altogether undistinguished student. Fascinated by puzzles and with a knack for mathematics and logic, he applied these talents only when it suited him, and wound up at King's College, Cambridge for his undergraduate education. While at King's college Strachey would first come in contact with Alan Turing, who was a junior research fellow at the university. According to Strachey's biographer the two met socially and not through what would become a mutual interest in computing, and as such it is unlikely that they discussed Turing's research on computability. As with the infamous Cambridge Apostles, of which Christopher's father and uncle had been members, King's College had a reputation for homosexuality and Marxist politics leading up to World War II. While Christopher was largely uninterested in politics, it was during this time that he seems to have come to grips with his sexuality, leading to a mental breakdown in the last two terms of his third year.
What little information exists on this episode comes from Strachey’s sister. As Martin Campbell-Kelley notes in his brief biography of Strachey, "The reason for his breakdown is obscure, although his sister supposes it may have been a coming to terms with his homosexuality. At all events, he recovered, and the problem did not manifest itself as a breakdown again." The time away from school was spent partly in a residential home for psychotherapy – Christopher's uncle James was a prominent psychoanalyst credited with first translating Freud's works into English and penning his biography – and on vacation in the United States. This is the only explicit mention of Strachey's sexuality, or indeed any personal struggle he may have had with his identity, in any of the historical material I've been able to gather, aside from passing declarative statements that identify him as a homosexual. Again, the extent to which this breakdown functioned as a transformative moment in Strachey's life is unclear, as is the way in which his sexuality evolved and came to affect his life as an adult. Strachey would return to finish his education in the year following the episode, graduating with a disappointing "lower second" that dashed any hopes of a research studentship. Instead he would turn to education, and spent the next thirteen years at various educational institutions performing the role of schoolmaster.
Strachey's draughts program.
Things began to change in January of 1951 when, through a mutual friend, Strachey received an introduction to Mike Woodger of the National Physical Laboratory (NPL). At that time the NPL was one of three institutions in the UK constructing computers – in this case the Pilot ACE, a preliminary version of the full Automatic Computing Engine or ACE, which had been designed by Alan Turing. Inspired by his visit, Strachey immediately began work on a program to make the Pilot ACE play draughts (checkers). He also worked on a program that would allow the machine to do its own coding, a self-reflexive gesture that reflected Strachey's interest in logical puzzles. The following spring he learned of the Ferranti Mark I computer at the University of Manchester, for which Alan Turing had written the manual. Through his earlier connections with Turing, Strachey managed to acquire a copy of the manual and began reprogramming his draughts program for the new machine.
The Manchester Mark I computer.
Strachey would visit Turing in Manchester twice in the second half of 1951, and on his second visit he was given access to the Mark I to try out his program. Over the course of an intensive session that began in the early evening and lasted through the night, he was able to get the program mostly working, and on running to completion "it finished with a characteristic flourish by playing the national anthem on the 'hooter.'"[ii] In fact during his visit Strachey programmed the Mark I to play a number of songs, including "Baa Baa Black Sheep" and "In The Mood" - which were captured for BBC radio in the autumn of 1951. While his love letter generator would come the following year, and is perhaps more strictly a computational artwork, these tunes are considered to be one of the earliest examples of computer generated music, produced by a total novice and programmed in the course of one evening.[iii] The speed and ease with which Strachey appeared to work the Mark I cemented his reputation overnight, and he would soon become known as the man who wrote "perfect programs," which would lead to a job offer at the National Research and Development Corporation (NRDC) the following year.
In June of 1952, Strachey began his position at the NRDS. With a lack of projects to occupy him at the start of his employment, he kept himself busy by building his own programs to entertain himself. Then, beginning in August of 1953, short notes began appearing on the notice board of the Manchester University Computer Department. They appeared to be letters of love and adoration addressed to an unnamed, genderless other, signed only with the initials M.U.C.
M.U.C., it turns out, stood for Manchester University Computer, and the letters were the product of an algorithmic generator that Strachey had written in his spare time. Each letter follows a similar structure, and is full of melodramatic Victorian overtones, with pet names like "honey," "jewel," and "moppet" along with other saccharine and yearnful descriptives. The letters were constructed via a generative algorithm that produced a variety of orders and combinations. In "There Must Be an Angel: On the Beginnings of the Arithmetics of Rays", David Link describes its execution in detail:
Apart from position commands like carriage return ("CR"), line forward ("LF"), and spaces ("spaces" or "sp"), the algorithm prints two salutations ("Add." = address). Then it enters a loop, which is carried out "5 times" and, depending on a random variable ("Rand"), follows one of two alternative paths. One generates a sentence following the syntactic skeleton "You are my—Adjective (adj)—Substantive (noun)"; the other path gives "My—[Adjective]—Sub- stantive—[Adverb (adv)]—Verb (verb)—Your—[Adjective]—Substantive" (the static words are underlined, the optional words are in square brackets). [...] Each phrase ends with a "Full stop". After the programme leaves the loop, it closes with the ending "Yours—Adverb (in the schematic this is given erroneously as 'Adj')—MUC."[v]
Previous scholarship by Andrew Hodges and others has suggested that the letters – surviving examples of which conspicuously lacked any variation of the word "love" – might have indicated a negotiation with the terms and legitimacy of desire, and a fascination with or alienation from love. More recent work done by David Link[vi] and Noah Wardrip-Fruin[vii] in the Strachey archives – in which the love letter generator is well documented – shows that in fact the original list of words that the computer could pull from via random number generation did include several variations on the word love, there simply were no examples of such letters in wide circulation.
Schematic of Strachey's love letter program.
Rather than examine the love letter generator in terms of identity, Wardrip-Fruin chooses to view it as a literary project despite the mechanical, even comical tone of these letters. In other words, he attempts to analyze the process of the generator rather than the content of the letters, to understand the materiality of the technical object rather than the meaning of its output. This is a particularly interesting method, one that is especially valuable for the study of computational systems, which function through mechanical processes in which authorship is neither a privileged site to be investigated nor – as Roland Barthes so famously suggested – evacuated. Ultimately this turn suggests that, as Jeremy Douglass puts it in "Machine Writing and the Turing Test," "the true message of this love letter is 'this is a love letter'"[viii] - in other words, that the process by which this message is constructed and conveyed is of greater interest than the content of the message itself.
Ultimately Wardrip-Fruin concludes that the generator is "a process designed to fail that employs a thesaurus-based set of word data and that can result in particularly inhuman surface texts." Thus, "we can see the generator as a parody, though its operations, of one of the activities seen as most sincere by the mainstream culture: the declaration of love through words. That is, [Wardrip-Fruin sees] the love generator, not as a process for producing parodies, but as itself a parody of process."[ix] The letters lack the subtlety and complexity of, for example, the parody of Victorian morality played out by members of the Bloomsbury Group thirty-five years earlier, but this is not where the parody lies. Instead it is a parody of the process of producing love letters, of producing love through this highly formal yet deeply affective medium. It is in this sense a queer critique of normative expressions of love, enacted through a kind of generative, computational performance, through a purposefully deficient simulation.
The interface for artist-researcher David Link's recreation of the love letter generator.
In his biography of Alan Turing, Andrew Hodges writes of the love letter generator, that "[t]hose doing real men’s jobs on the computer, concerned with optics or aerodynamics, thought [it] silly, but [...] it greatly amused Alan and Christopher."[x] It is interesting and perhaps appropriate that what might be considered the first work of computational art was a kind of joke, a critique of "real" epistolary writing and "real" love by means of automation through digitization. It is even more fascinating that it seems to have come from a queer history - not of "passing" as has been suggested with regards to Alan Turing's work on gender and artificial intelligence in the Turing Test, but of camp and the ostentatious performance of "authentic" affect.
[i] The majority of biographical information on Strachey has been taken from Martin Campbell-Kelly's "Christopher Strachey, 1916-l975: A Biographical Note", published in the IEEE Annals of the History of Computing, Volume 7, Number 1, January 1985.
[ii] Campbell-Kelly, Martin. Ibid. p. 25.
[iii] According to the BBC, That honour goes to a third machine called CSIRAC, Australia's first digital computer, which "stunned" audiences with a rendition of Colonel Bogey. That said, no recordings of the CSIRAC music have thus far been found.
[iv] This image and the one that follows are taken from the Strachey archives and reproduced in David Link's essay, cited below.
[v] Link, David. "There Must Be an Angel: On The Beginnings of the Arithmetics of Rays" p. 20. <http://www.alpha60.de/research/muc/DavidLink_RadarAngels_EN.htm>
[vi] Link, David. Ibid.
[vii] Wardrip-Fruin, Noah. "Digital Media Archaeology: Interpreting Computational Processes" in Media Archaeology: Approaches, Applications, and Implications. Erkki Huhtamo and Jussi Parikka, eds. Berkeley: University of California Press, 2011.
[viii] Douglass, Jeremy. "Machine Writing and the Turing Test: From writing to writing system, in accordance with a queer theory of identity and a reception theory of art" <http://www.english.ucsb.edu/grad/student-pages/jdouglass/coursework/hyperliterature/turing/>.
[ix] Wardrip-Fruin, Noah. Ibid. p. 316
[x] Hodges, Andrew. Alan Turing: The Enigma. London: Vintage Books, 1992. p. 478.