by Jim Eldridge
However, Alan wanted to build on the work he had begun, when breaking the Enigma code, in the research he’d seen into the early computers in America, and also with his recent work on Delilah. Alan wanted to build a real and sophisticated Thinking Machine.
In October 1945, Alan joined the staff at the National Physics Laboratory in London, working with the electronic engineers there to create a new breed of computer, using digital technology, and using as a springboard all the mechanical and electronic advances that had happened during the War.
By February 1946, Alan had created a design for an electronic computer – his Thinking Machine – which he called ACE (Automatic Computing Engine). In creating ACE, Alan had realised that the real problem with building an effective Thinking Machine was the speed of operation, and the deciding factor in making a machine that worked fast was in its memory. As he wrote in a talk he gave in 1947:
In my opinion this problem of making a large memory available at reasonably short notice is much more important than that of doing operations such as multiplication at high speed. Speed is necessary if the machine is to work fast enough for the machine to be commercially valuable, but a large storage is necessary if it is to be capable of anything more than trivial operations.
Although Alan’s time at the National Physics Laboratory should have been one of excitement and fulfilment, with the NPL electronics engineers putting his designs and theories into practical form, in reality Alan was frustrated there. For one thing, there was a constant lack of funds, which affected the development of his ideas. Immediately after the war, Britain was suffering economically, and the Government opted to spend what little money it could get on rebuilding the bomb-damaged country.
Alan was also upset at the levels of bureaucracy at NPL, which interfered with his work. He was constantly being forced to attend meetings, and produce organisational paperwork that had little to do with his actual research. Although Alan gained a lot by working in small teams, such as the Hut 8 code-breaking team at Bletchley, and the small three-man team who developed the Delilah project, he was not a ‘company’ person. Bureaucracy and paperwork, especially when he considered it unnecessary and interfering with really important scientific work, annoyed and frustrated him.
So, when, in 1947, he was offered a chance to return to King’s College, Cambridge for pure academic research for a year, he took it.
12
A Computer Called Baby
It was while he was at NPL that Alan started competitive long-distance running again. In 1946 he became a member of the Walton Athletic Club, based in Walton, Surrey. Such was his ability at long-distance running, including marathons, that he was considered as a serious candidate for the British athletics team for the 1948 Olympics. Unfortunately, during this period (1946–8) he suffered an injury which meant he would not be fit for the Olympics, and put an end to his competitive running career.
However, 1948 was to mark a major new chapter in his scientific career.
During his year at Cambridge, Alan was invited to join a brand new project at the University of Manchester, aimed at developing a new kind of computer. The machine was called the Small Scale Experimental Machine or SSEM – though it was called Baby by everyone involved with the project. In May 1948, when his sabbatical year at Cambridge ended, Alan joined the team at Manchester.
At the heart of Baby was the importance of the amount of memory it could store, just as Alan had emphasised in his talk in 1947. The SSEM used a device known as the Williams tube to store computer data. The Williams tube had been invented during the war by Frederick Williams as a means of storing radar images of ships and planes in a cathode ray tube. The images were kept bright and easily visible in the cathode tube by means of electronic pulses passed through the tube. The radar image on an ordinary screen would appear, and then fade. The Williams tube kept those images alive and available in the machine’s memory.
The Williams tube was able to store data at 2Kb (2,000 bits), which at that time was a massive amount of storage. (To put this in context, the computer that powered the Apollo space mission in 1969 had a memory of 64Kb. Computer memory has expanded hugely since then: the memory of a modern smartphone is 256Mb.)
The first program for the SSEM was written by a computer engineer called Tom Kilburn, and was run in June 1948. The program was a mathematical one, aiming to find the highest factor of 218 (262,144), using every integer from 1 to 262,144. To do this, the SSEM had to carry out 3.5 million operations. The operating time for this program was 52 minutes, and at the end the machine had calculated the answer as 131,072 (i.e. half of 262,144).
It was not the answer that created the excitement, but the fact that the machine had calculated it, and had done so by carrying out 3.5 million calculations in so short a time.
Alan arrived at Manchester University towards the end of Tom Kilburn’s successful program using Baby. The question was, what could he contribute to the project? The machine had already been created, with Frederick Williams’ memory device at its heart, and programmers such as Tom Kilburn writing complex maths programs for it. But Alan wanted to see if he could develop Baby to replicate the biological functions of the brain. He wanted to create a machine that wasn’t just a fast-acting supersonic calculating machine, but a device that could reason, and reach a conclusion about a problem – any problem, including artistic or literary issues as well as scientific ones. Alan was aiming to develop something capable of artificial intelligence.
13
AI
In October 1950 Alan published one of his most famous articles, in the magazine Mind. It was called ‘Computing Machinery and Intelligence’ and began: ‘I propose to consider the question: Can machines think?’
Alan went on to describe the problems that he felt a machine, properly programmed, could deal with. To back up his proposition, he put forward what became known as the Turing Test. In this theoretical test, a person (A) was stationed at a computer terminal which was linked to two other terminals, and posed questions by means of sending text. One of the other terminals was controlled by a person, who sat at the keyboard and responded to the questions posed. The other terminal was a computer that answered the questions sent to it with no human interference. Replies from both terminals were also in text.
The Turing Test is whether the person who was asking the questions (A) could decide from the answers which of the two responding terminals was operated by a human being, and which by a computer.
According to Alan: ‘If, during text-based conversations, a machine is indistinguishable from a human, then it could be said to be thinking, and therefore could be attributed with intelligence.’
The sample questions that Alan used as examples for the Turing Test included the following exchanges between A and the two unseen recipients (along with the answers received by A from them):
Q: Please write me a sonnet on the subject of the Forth Bridge.
A: Count me out on this one. I never could write poetry.
Q: Add 34957 to 70764.
A: (After a pause of about 30 seconds) 105,621.
Q: Do you play chess?
A: Yes.
Q: I have K (King) at my K1, and no other pieces. You have only K (King) at K6 and R (Rook) at R1. What do you play?
A: (After a pause of about 15 seconds) R–R8 Mate.
If these questions and the answers are examined and thought about, it can be seen that any of the answers could have come from a human being or a machine, including the one in response to writing a sonnet about the Forth Bridge: ‘Count me out on this one. I never could write poetry.’
In this same article, Alan went on to say: ‘I believe that in about 50 years time it will be possible to program computers so well that an average interrogator will have no more than a 70% chance of making the right identifications after five minutes of questioning.’
It was with this article that Alan started up a whole debate about Artificial Intelligence, not just among the scientific community, but among re
ligious leaders and theologians as well. This was because most organised religions have as their basis the idea that God created human beings, and their capabilities. If a machine could be created that could think and reason in the same way as a human being, then that took away the idea of reasoning beings as only ‘created by a God, a divine being’.
14
Fibonacci Numbers and Order in the Universe
If Alan had upset some of the most religious people with his claim that machine intelligence equal to human intelligence could be created by scientists such as himself, the next phase of his mathematical theories might have confused many. It examined the connection between mathematics and plants, and his research suggested a link between plants and a kind of Mathematical Designer and Creator, although he did not assert this conclusion directly.
His research centred around the patterns on sunflower heads. Alan had noticed that seeds on sunflower heads were arranged in a spiral pattern that followed what was known as the Fibonacci sequence.
In the Fibonacci sequence, each number is the sum of the previous two numbers, starting with 0 and 1. The sequence begins:
0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987 … and so on.
Leonardo Fibonacci was a thirteenth-century Italian mathematician. He was not the first to spot the sequence’s occurrence in nature, but he was the first to bring it to the attention of the Western world. The first known observation of this sequence of numbers occurring regularly in nature was found in in ancient India, in the Sanskrit language. Versions appear in the Hindu-Arabic number systems.
Leonardo Fibonacci’s father was a merchant who travelled around the Arabic world, often taking Leonardo with him. It was on these journeys that Leonardo became aware of the number sequence that occurred so frequently in plants, and in other forms of nature. Leonardo studied mathematics under the leading Arab mathematicians of the time, and in 1202, at the age of 32, he published what he had learnt in a book called Liber Abaci (The Book of Abacus, also known as The Book of Calculations).
The theory that plants, and other aspects of nature, had a mathematical basis to their construction was widely discussed. Leonardo da Vinci studied this phenomenon, and many other mathematicians and scientists from ancient times had made the same observation. Shapes in nature where the spiral form could be defined by the Fibonacci sequence of numbers include the growth of mollusc shells, the shapes of many flowers and fruits (including pineapples and artichokes), the uncurling of fern fronds, the distribution patterns of branches on tree trunks, and the shape of pine cones. In modern times, the Fibonacci sequence has even been detected in the construction of DNA.
Before computers, these discoveries had been the result of physical examination of plants, as scientists counted the facets that made up the spiral shapes. Now Alan was able to apply technology to the problem, using a computer to positively identify the arithmetical Fibonacci sequence in the spiral forms of sunflower heads.
However, when, in 1952, he published his paper linking the spiral forms of the sunflower heads to the Fibonacci sequence, it was still only a theory. It was not until 2012 – a hundred years after Alan’s birth – that his theory was properly investigated and evaluated by the Museum of Science and Industry with the launch of the Turing Sunflowers Project, in which 12,000 people from seven different countries donated sunflower specimens for examination. At the time of writing (March 2013), of the sunflower heads examined and analysed in the Project, 82% had their seeds arranged in a spiral pattern in which the number of rows of seeds followed the Fibonacci sequence.
The implication of Turing’s theory, following those of the Hindu-Arab scholars, Fibonacci, Leonardo da Vinci, and sundry others, is that the elements in nature have not emerged haphazardly, but have a mathematical basis. If that is the case, where did this mathematical basis come from? Did these elements in nature that follow the Fibonacci sequence develop these spiral forms under some pressure of evolution? Or were they designed by some mysterious force?
To some people, the idea that Alan Turing – renowned as a purely analytical scientist, a mathematician – should be involved in something that suggests a ‘creator’ at work appears very inconsistent with his previous research and work. It is often assumed that hard science and spirituality do not have a common ground. Certainly there is no evidence that Alan was interested in any organised or traditional religion. The concept of an all-powerful God does not appear in his work, or his life.
However, there is a lot of evidence of Alan’s interest in what could be called spirituality, or non-physical influences in the world.
The first traces of this can be found in Alan’s writings and letters following the tragic death of his close friend, Christopher Morcom, in 1930. In a letter to his mother shortly after Christopher died, Alan wrote: ‘I feel I shall meet Morcom again somewhere and that there will be work for us to do together.’
In a paper he wrote at this time called ‘Nature of Spirit’, he said: ‘The body, by reason of being a living body, can attract and hold onto a spirit. While the body is alive and awake the two are firmly connected. When the body is asleep I cannot guess what happens, but when the body dies the mechanism of the body holding the spirit is gone and the spirit finds a new body sooner or later, perhaps immediately.’
This may seem like an anti-scientific stance, but in fact it was very much part of the sciences that were emerging at that time, particularly in the areas of quantum physics, where practically anything was considered possible, including the idea of parallel universes existing at the same time, and the concept of a soul or spirit leaving the physical body (called an ‘out-of-body experience’).
Further evidence of the continuation of Alan’s interest in and acceptance of what are sometimes termed ‘alternative sciences’ or ‘the paranormal’ come in his explanatory notes to his Turing Test at Manchester in 1950. He wrote: ‘Unfortunately the statistical evidence, at least for telepathy, is overwhelming. … Let us play the imitation game (aka the Turing Test) using as witnesses a man who is a good telepathic receiver, and a digital computer. The interrogator can ask such questions as ‘What suit does the card in my right hand belong to?’ The man, by telepathy or clairvoyance, gives the right answer 130 times out of 300 cards. The machine can only guess at random, and perhaps gets 104 right.’
From all of this, as well as his interest in the Fibonacci number sequence in nature, it does suggest that Alan believed there was more to life and the universe than what we could physically see or experience. In this he was no different to those earlier scientists who insisted on the existence of electricity in the atmosphere, or magnetism – both scientific facts but which could not be seen by the general observer. In our own time, it is the same with the discovery of black holes, which were considered theoretical at first but subsequently proved to exist. Many now-proven scientific facts were at first dismissed as supernatural.
It is my view that with all of this, Alan was trying to find scientific causes for everything. His research and investigation over many years into the act of reasoning, using machines to explain and recreate the process, were just a part of his trying to explain the bigger question: is there a form of order and structure in the universe?
15
Arrested and Tried
Alan’s 1952 publication showing the connection between Fibonacci numbers and sunflower heads could have led him to even greater discoveries in his overall quest for the bigger questions about the structure of the Universe. But, in 1952, something happened that changed the course of his life dramatically, and, some believe, led to his early death.
As we have seen, from early in his life, since his teens, Alan had realised that he was gay. He never deliberately kept his homosexuality a secret. As we have seen, at Bletchley Park he met and became engaged to Joan Clarke, but he did warn her that he had ‘homosexual tendencies’. Once he had got to know his co-workers, he often revealed the fact that he was gay to them. Many of his friends and cowo
rkers simply accepted it and got on with working with him. Others reacted badly. This was the case when he told Donald Bayley, his teamworker on the Delilah Project, that he was gay. Whether because being gay was illegal or Bayley’s own prejudice, his initial reaction was one of disgust, and for a time it put a strain on their relationship. Eventually Bayley came to accept Alan’s sexuality as just one part of him, and they became good friends.
In December 1951, Alan met a young man called Arnold Murray in Manchester, and they began a relationship. Murray came from a tough working-class background. At the time that Alan met him, Murray was broke, out of work, shabbily dressed, and very thin from a lack of proper food. He was a very different character from men Alan had known previously, most of whom came from his own kind of background: middle class family, public school and university.
Alan and Murray spent a lot of time together at Alan’s small semi-detached house in Manchester, and Alan started giving Murray money to help him out.
In January 1952, Alan’s house was burgled, and he was shocked to discover that an acquaintance of Murray’s, a man called Harry, had carried out the burglary. Alan reported the burglary to the police, and Harry was arrested. He confessed to the burglary, but told the police about Alan’s relationship with Murray, possibly in the hope that giving evidence about their criminal relationship might get him a lighter sentence.
Because male homosexuality was illegal, most men suspected or accused of being gay would deny it. Often this denial was accepted, even if the truth was known, provided it could not be proved. Frequently, the only way a case was proved was when the men involved admitted to taking part in homosexual acts, and this rarely happened, as the men would not want to go to prison. In the 1895 case of Oscar Wilde, for example, most of Society at that time knew he was gay, but they were happy to turn a blind eye to it. He was not arrested until the father of one of his lovers gathered evidence against him and sent it to Scotland Yard.