Not only is it pride month, but tomorrow is the great Alan Turing's 100th birthday, which are good reasons to remember his life and impact. Alan Matheson Turing (1912-1954) was an English mathematician, and a tremendously inventive and original thinker whose work had impact in several disciplines; his interests fell outside conventional categories. He made significant contributions to cryptography, computer science, cognitive science, and other generally geeky subjects. Although the twentieth century is rife with brilliant scientific pioneers, Alan Turing's story also resonates for me because of the courage and integrity he showed living life as a gay man in a society soaked with homophobia. Like Harvey Milk he ended up a kind of martyr. We LGBTI people are richly represented in the lively arts and letters, but it is a challenge to name many LGBTI titans of the nerdy arts. Those of us who share both queer and nerd identities sometimes experience a kind of intersectionality deserving attention, and Turing's life and legacy remind us of just how brightly our tribe can shine.
There are some excellent online biographies and articles about Turing, and I heartily recommend Alan Turing: the Enigma by Andrew Hodges. I want to sketch out his technical achievements but devote a little more attention than typical to his life as a gay man. (Should I even call him gay? If you're more a social-constructionist than essentialist, you might well object; Turing definitely called himself homosexual. But in one short story that's clearly a roman-a-clef autobiography, he calls his counterpart gay. That and brevity are reason enough for me.)
Alan Turing was born with privilege into an upper-middle-class family. He grew up with only occasional contact with his parents, as his father served as an administrator in India. Alan was a disheveled lad, headstrong and fond of solitude. Some have posthumously diagnosed him with Asperger's syndrome. The Turings sent him to English public school at Sherbourne, where he did well at math but poorly at the social arts one needs to win friends and good grades. He indeed made few friends, apart from one, Christopher Morcom, who shared with Alan an enthusiasm for math and science, especially chemistry, astronomy, and physics. Chris became an intense crush for Alan, who "worshipped the ground he trod on"; those unrequited emotions became crushing early sorrow when Chris died of tuberculosis in his teens.
(continued below fold)
Turing won a scholarship to King's College, Cambridge, a relatively gay-friendly environment -- the alma mater of Horace Walpole, J. M. Keynes, and E. M. Forster. Newfound independence suited him: he came out to himself and enjoyed his first sexual friendship, though it wasn't love, and he cut a lonely figure. He studied mathematics and managed to derive an independent proof of the Central Limit Theorem, a result that describes why so many natural random events are distributed according to a bell-shaped curve. Unfortunately the same result had been proved twelve years earlier. Still, it was an amazing feat for an undergraduate. On that basis, after graduation he was elected a Fellow of the college, a sort of researcher-in-residence.
The Turing Machine
As a Fellow he took to mathematical logic, and in 1935 attended a seminar taught by Max Newman, who became a mentor and friend. This seminar inspired him in his first major paper, my favorite, which I wrote about in a separate diary. This paper has a jawbreaker title but is known as Computable Numbers for short, and in it he presented a creative solution to a major open problem of mathematics, proving the nonexistence of an algorithm to generate logical proofs. Nonexistence proofs are often the most demanding kind, requiring the greatest resourcefulness and novelty. My other diary has the details, but I will mention that the methods he devised are still important today, and they are taught in undergraduate computer science. In a sense his methods have become more famous than the conjecture he conquered.
The starring role of Computable Numbers was an idealized model of computation he introduced, now called the Turing machine. Though Turing's description makes it sound like a souped-up stock-ticker, he argues for its universal significance, based on humanity. In other words, his machine models how a human thinks; its work medium is the humble pencil on paper, and it is guided by "states of mind." Turing believed that the human mind is a machine, and works like one; this contentious principle shapes all his published work. In a sense, Computable Numbers holds a kind of abstract sketch of the human mind, a modern Vitruvian Man. Like the Vitruvian Man, his picture encircles us within a boundary line: Computable Numbers shows that some questions are Undecidable, beyond the limit of what can be answered in language. In this way I see his paper as an advance in postmodern thought, like Godel's theorems of a few years earlier.
After Computable Numbers, he visited Princeton but soon returned to England before the outbreak of war, and in 1938 joined the Government Code and Cipher School. During the war the codebreaking effort turned into an operation of huge strategic importance and the darkest secrecy. The center of operations was at Bletchley Park, Buckinghamshire, and there Turing and other theorists worked on cracking the codes of the Nazi military communication. It's an exciting, complicated story told by Hodges and, if you want even more details, see David Kahn's Seizing the Enigma. Cryptanalysis and cryptography strike me as a sort of modern alchemy: one turns lead into gold (that is, gibberish ciphertext into informative cleartext) or vice versa. Like alchemy, it's not an intuitive system, and the serious players want practical results.
The atmosphere was intense and collegial. Turing's friends nicknamed him "the Prof" for his brilliant but absentminded otherworldliness. He had a grating high-pitched laugh and used a piece of rope in place of a belt. But his quirkiness was tolerated because he was so good at his job. He loved theory but he relished the engineering aspect of the task too, that of taking a theoretical solution and watching it become real. In these roles he was the undisputed alpha-nerd of Bletchley.
Turing and others exploited vulnerabilities intrinsic to the Axis cipher system (called Enigma) and blunders made by enemy personnel. The team devised electromechanical systems -- slow, clacking computers called Bombes -- to search for message keys. They effected an industrial revolution in cryptanalysis, which formerly had relied on manual methods using index cards and specially perforated sheets. Practical results varied: new analysts, new breakthroughs and new bombes helped the cause, but Nazi countermeasures and mere bad luck sometimes caused months of doldrums. Towards the end of the war the GC&CS employed thousands of staff, "but if anyone was indispensable to [cryptanalysis of Naval communiques] it was Turing," wrote Hugh Alexander.
After the war Turing was in a difficult spot. In addition to his theoretical contributions to cryptanalysis, he was a part of the engineering of cutting-edge computing machines. To a certain extent he helped implement them, with a working knowledge of electronics. Yet he had to keep strictly silent about nearly all those achievements, since they were official secrets. This technical "closeting" must have been extremely frustrating. He went to work for the National Physical Laboratory and developed detailed hardware and software plans for one of the first general-purpose computers, but his superiors there found ways to ignore him. Admittedly, Turing articulated his vision as a "thinking machine," not just a calculating engine; it would need not only programming, but also "experience." (Maybe to NPL administrators he sounded like a hopeless crank, but I think he was just a decade or more too farsighted. In 1960, Frank Rosenblatt received a much warmer reception when he enthused about comparable ideas.) Fortunately Max Newman obtained him a faculty position at Manchester University, which was a more supporting environment, and he partially realized his quest for a universal computing machine there.
Intelligence: the Turing Test
Turing's next major accomplishment was a paper on the theme of thinking machines, "Computing Machines and Intelligence," published in 1950 in the journal Mind. You might know of this one: it was the source of the Turing test. The Turing test is an empirical, operational definition of thinking, a looks-like-a-duck test for intelligence. In other words, how do you know if someone or something is intelligent? You know because its output shows intelligence.
Turing's idea to establish that was a remote interview where a judge instant-messages with two interviewees, a human and a computer pretending to be human. The judge types in questions to both. By reading their responses she tries to determine which one is the machine. Turing argues that all three are, in fact, machines, even if two are flesh and bone -- which is probably the whole point of the paper, though the Turing test is now the more famous idea. There is plenty to criticize in this paper and in its interpretation, but it certainly has inspired serious and not-so-serious thinking about the philosophy of intelligence and minds.
Although the test is defined as a dual-subject interview, there are lots of similar variations possible, any of which is likely to be called a Turing test nowadays. I don't imagine that Turing foresaw that the judge might be a computer, but that's what has happened in CAPTCHAs, the commercially most important application of the Turing test. Plausible chatterbots like ELIZA have been (temporarily) duping us primates since 1966. If we redefine the test so that the judge does not interview, but rather evaluates a static work -- a painting or a text -- then it becomes a very difficult challenge indeed. Another partway-there step was the Jeopardy-playing Watson. The computer agent was not concealed, but I was still pretty impressed with how well it understood questions and produced answers. Was it thinking? What say you?
Arrest and trial
"He who has committed iniquity, shall not have equity." (Legal maxim)
AGENT SMITH: What good is a phone call if you are unable to speak?It seems like a truism to say that marginalized people are disproportionately the victims of crime. I don't mean hate crimes specifically; just everyday criminal predation. Pierre Seel (1923-2005) recounts that, as a youth in Alsace, his watch was stolen in the midst of a brief romantic encounter with another man. The watch had been a gift from his godmother, so Seel reported the theft to the police. Instead of getting his watch back, his name was recorded on a police registry of known homosexuals; that in turn led to his arrest, torture, and internment in a Nazi concentration camp. What good is police protection when you can't use it?
In January 1952, Alan Turing's home was burgled, and like Seel's watch, the stolen oddments were taken as a consequence of a liaison. Turing had connected with a young working man named Arnold Murray, and they had enjoyed a rather tentative affair. Murray unfortunately boasted of this to a lightfingered acquaintance. Turing reported the burglary to the police, who quickly detected that, in addition the larcency, something queer had been going on. On questioning, both men admitted they had slept together. They had no good option but to plead guilty to Gross Indecency, and to hope for a lenient sentence.
Turing was not very secretive about being gay: Hodges says he had a habit of outing himself to his colleagues, "because he did not wish to be accepted or respected as the person he was not." But the trial implied a level of public exposure and obloquy far outside his comfort zone. Turing's family had been unaware that he was gay, and he had to tell them himself. They chose to stand by him, but not without disgust and anger.
Turing had a strong ally in his mentor Max Newman. After the indictment, Turing bluntly told Newman all about it over lunch one day in the cafeteria -- in a particularly loud voice so as to be overheard. Newman was surprised but supportive, and agreed to testify as a character witness. The Manchester administration fussed and considered firing Turing, but Newman as head of the math department said he wanted him to remain, and that was that. When Newman did testify, he was asked if he would deign to receive the defendant in his home. He answered that indeed he already had; Alan was a good friend of the family.
A half-century earlier, Oscar Wilde had been sentenced to prison at hard labor on the same charge of Gross Indecency. But progressive jurisprudence in the mid-1950's began to view homosexuals as diseased, rather than criminal. This (ahem) enlightened view was not yet universal, and Turing was presented with a choice between prison or one year of "organo-therapy" (estrogen injections) and parole.
[DR BENWAY:] "For example . . . for example . . . take the matter of uh sexual deviation." The doctor rocked back and forth in his chair. His glasses slid down his nose. Carl felt suddenly uncomfortable.
"We regard it as a misfortune . . . a sickness . . . certainly nothing to be censored or uh sanctioned any more than say . . . tuberculosis."William S. Burroughs, Naked Lunch, 1959
Organo-therapy was intended to eliminate libido temporarily. Today we would call it chemical castration, but at the time it was seen as therapeutic, not punitive. Its rationale is infuriating to read because the reasoning is so lacking. Some endocrinologists in the 1940s had been curious what would happen when they injected sex hormones into people. So they tried it -- androgens for some, estrogen for others. Their subjects did not always consent, but who cares? Turns out, estrogen squelches male libido, and makes men grow breasts. Perfect for the grossly indecent.
It was certainly unfair; yet superficially "scientific" intrusion was common then. Lots of terrible things occurred in mid-century (and later) under the banner of improving society and individuals with science. (Fortunately, medical research now includes better mechanisms to prevent ethical atrocities, though IRBs are as subject to corruption as any other institution.)
During the Lavender Scare, one facepalm-inducing justification for the systematic firing of LGBT government employees was that they would be disproportionately susceptible to blackmail. (Susceptible to blackmail, of course, because of the repercussions of being exposed. The solution is, obviously, more repercussions.) Turing thus lost his security clearance, which he was using as a part-time cryptology consultant with the British Government Communication Headquarters. He had enjoyed his Bletchley years, and I think it likely he worked on similarly rewarding problems at GCHQ, until they slammed the door.
It's not clear what precipitated the end of the story. In school he and Chris had been fond of chemistry, especially a certain experiment that suddenly turns dark blue after a puzzling delay. Maybe the estrogen therapy started a hidden, inner countdown. Honestly I think that explanation is too simple, and since we don't do organo-therapy anymore, it's very exculpatory. Biographer David Leavitt describes the post-parole period as a "slow, sad descent into grief and madness," but he certainly did not go mad in the sense of becoming catatonic, incoherent, or even noticeably depressed. He seemed to be the same queer duck he always was. Having survived the quack treatment, he had moved on with his life. Turing was working productively at Manchester, pioneering a sort of computational biology with affinities to chaos theory. In his personal life he maintained his friendships, saw a psychotherapist, dabbled with writing semi-autobiographical short stories, and occasionally traveled to the continent on holiday to enjoy short, intimate relationships. So to his friends, his suicide was out of the blue. He left no note, unless you count a half-eaten apple: a callback to Snow White or the garden of Eden.
Alan Turing did great things in his bold, odd, too-short career, exploring the intersection of logic, intelligence, and the natural world, merging theory and practice. Also he maybe saved England from the Nazis. It's taken awhile for Turing's story to get the attention it deserves, but Hodges's insightful 1983 biography was a breakthrough. Today the highest honor in the field of computer science is called the Turing Award. There are many plaques and streets and statues honoring him. In 2009 Turing was selected as a Gay Icon by the (English) National Portrait Gallery. In that same year he received an official apology from prime minister Gordon Brown. (He certainly deserved an apology -- better than an apology -- but I struggle to understand the point. What about an apology to the hundreds of other people tried for Gross Indecency in 1951? What about India, Kenya, Egypt? But I digress.)
The intersection of queer and nerd identities is still difficult. Sociologists Erin Cech and David Waidzunas published a study in 2010 about the challenges that LGB engineering students face, by employing "coping strategies which can require immense amounts of additional emotional and academic effort." I want to end with a few quotes from their interviewees (quoting from Cech and Waidzunas, "Navigating the Heteronormativity of Engineering," Engineering Studies, 2010).
'It's like the military, like I have to give 110%. Last quarter, I had this study group, and I think two of the guys feel very threatened that I'm gay . . . . Every time we go into a study room or something like that, they'll try and find a seat the farthest away from where I'm at. And, they'll talk to me about nothing but structural engineering. They'll still meet with me, but they wouldn't want anything to do with me outside of work . . . . And, even though I have tough questions, I feel like I can't ask my questions because of those two students . . . I really don't want them to see me as gay and stupid.' [Juan]
'When you got into a professional engineering environment, you leave your private self at home, and you keep it all professional, and when you leave that environment, you can go back to your private self at home.' (Lisa)
'Before I was out . . . I happened to laugh in a very gay way and [another student] mocked my laugh in the same kind of gay way that I laughed and then asked me, with obvious hostility, "are you gay?" Other people were around him at the same time. And I said "no," I was not gay.' (Respondent looks visibly uncomfortable at this point; he grimaced as he said this line, tugged at his shirt, shifted in his seat between "no" and "I was not gay.") 'Fortunately the conversation ended there, but he really hurt me by that. He made me feel unsafe, and denigrated based on my sexuality. I never forgave him for that.' (Eric)On Saturday, I invite you to remember Turing's hundredth birthday: have a glass of beer, maybe play a game of chess, or just try to discern which of your Facebook friends is actually a machine. If Alan were here he would give a long, loud alto laugh and say that they all are.
Eric confided that he was just coming out to himself at this stage in his life, and fear of this type of interaction pressured him into passing for another year and a half.
Jin Wicked's drawing, "The Universal Turing Machine," is copyright 2002 by Jin Wicked and used with permission. XKCD comic is by Randall Munroe and used under the CC BY-NC-2.5 license. Photo of Bletchley Park is by Draco2008 and used under the CC BY-2.0 license. Photo of Turing statue is by Sjoerd Ferwerda and used under the CC BY-SA-3.0 license.