This is a wonderful opportunity to indulge in a bit of iconoclasm, to reminisce about times long past and to set the world to rights. I can only hope that you will find it as much fun as I will.
At the end of my schooldays, I was hurled into the bewildering new world of Cambridge University as a mathematics scholar. I skipped the first year and met a philosophy different from anything I had met in school. The most important message seemed to be, "Don't believe in anything without careful testing." Miss Grimshaw left you wondering if any series would ever converge. A line has an uncountable number of points, but whole swathes of pure maths are based on whether a set lacks just one of these points. Another lecturer looked at lines another way. He mapped them onto the surface of a five-dimensional quadric surface.
More anecdote-worthy was the lecturer who taught us the wave equations. This was Fred Hoyle. Even though this was an undergraduate course, it had a sub-plot. By looking at the inductance of all space, he impressed on us that the universe must have been eternal. Then came the second semester - and he impressed on us that there must have been a Big Bang! His theories had flipped completely in the short vacation. That sort of thing shakes your belief that the cosmologists' unanimity actually means that they have got it right - but more about that later.
In the second year I attended lectures preparing us for the wonders of Part III, a postgraduate course. We met Schroedinger's equation, leaving our heads buzzing with a vision of empty space in which swam the solutions of wave equations - ourselves! We met relativity, quantum theory and magnetohydrodynamics.
Other memorable events in those years included a visit to see EDSAC 1. In a small room, researchers sweltered in the heat of thousands of war-surplus EF50 valves. (On the other hand, remembering the rigours of the Cambridge climate they probably thought themselves lucky.) The valve heaters alone took 15 kW.
The working memory of this early computer was built from a tank of mercury - or several tanks, actually. A 'speaker' at one end emitted sound pulses, which travelled as vibrations along the tank, were picked up by a microphone at the other end and fed back to the speaker. So a complicated pattern of data went round and round until some computing result was applied to change it.
To check how the memory was getting on, the pulses were connected to a display. So that a lot of pulses could be seen at once, the display was not on an oscilloscope but on a television screen. With the ability to change patterns on a television screen, the software writers' minds had turned to games. So the occurrence of games on a memory-mapped display screen dates back to the 1950s.
One such game was noughts and crosses, or tic-tac-toe as others like to call it. A 'splodge' moved from square to square, and to write your cross you moved your hand into the beam of the paper-tape reader. Interactive graphics!
To put history into perspective, another Archimedeans' trip was to London to see My Fair Lady.
A leading light in the Archimedeans was John Horton Conway. While we should have been preparing for finals, we spent many evenings together working out the intricacies of tetra-flexagons. Ask me about them some other time.
The long vacation brought the need to scrape some money together. I took a job at Smiths Aviation Division, near Cheltenham. I was put under the wing of Bill Wallace, an engineer in the Differential Analyser group. He had me measuring valve characteristics in no time, and guided me through the design of a simple amplifier.
Then I was given two precious transistors, newfangled devices which were hitting the headlines. I measured their characteristics, built a circuit and it had a magnificent gain. Suddenly one transistor grew hot and became a short circuit. Disaster. Mortified I carried the corpse to confess. Bill's cheery "So you've popped it?" put matters into perspective.
In those days if a transistor became perceptibly warm to the touch, you became worried that it would become victim to "thermal runaway" With modern silicon, a smell of scorching flesh as you touch it still does not mean that anything is amiss - except your decision to touch it!
The second half of this vacation work involved testing and repairing operational amplifiers, hugely expensive precision valve amplifiers which were the essential component of an analogue computer. Nowadays 'op amps' are found, four to a chip, at just a dollar or so.
The following year I went back to work in the same department. A circuit for a general purpose computing cell, one whole transistor plus assorted resistors, had appeared in an article. Sam Marshall wanted the idea tested. I raided all the drawers for sample transistors. By the end of the vacation I had built a four-bit arithmetic unit - it could add, subtract, multiply and divide. It contained three hundred transistors. I was hooked!
This vacation followed my Part II mathematics finals. But now I was far from sure I wanted to devote another year to Part III maths. Instead I transferred to Part II of the Mechanical Sciences Tripos - course D, Electronics. Here a series always converged. If a design looked right and worked, it was right.
At the end of that academic year, in 1960, I helped a friend overcome a problem with his project. He had taken the Control Option of part II and my contribution involved devising a steering system for a small mobile robot.
After a haze of May Balls and bumping races I realised that I was a graduate. I needed a job.
I became a Graduate Apprentice at Smiths. I learned to drive a lathe, to conduct time study assessments and look into all the workings of a factory. At the time it seemed a terrible waste of time, but I have exploited the experience ever since.
After the 'purgatory' of the factory, I moved to the research labs. I worked in a humble way on the first autolanding autopilot. It was based on magnetic amplifiers. I moved around each three months, in another department investigating energy losses of watch balance wheels and in Theoretical Studies finding the motivation for the analogue computers of the Differential Analyser group.
After completing the two-year programme I weighed up a choice of jobs - to stay at Smiths or work at GCHQ, the secret communication headquarters. Having been 'positively vetted' I was shown round - and met my old Cambridge maths supervisor hidden deep in the cryptography department. I was shown the computer floor, cut deep in the hillside. The manager (I can only remember him as W60) bemoaned the fact that he had the world's very first computer, Colossus, and because of its top-secret codebreaking purpose, nobody would let him put history straight. (Now of course that story has been told.)
Now I had time for the important things in life. I met Ros. She worked in GCHQ in the top secret inner sanctum, sending mysterious packages to spies around the world. Despite this conflicting attraction, I had chosen to stick with Smiths.
I joined the Simplex autopilot group, to work on the design of a lightweight, low-cost autopilot, SEP6. My work started with a powerful servo system, including the design of a thyristor amplifier, for driving the aircraft control surfaces. It went on to the electronics to implement the control equations, then by exploiting my familiarity with Theoretical Studies I managed to get the equations changed to my own novel non-linear system. (This was afterwards patented.)
I married Ros in May of 1964. The project was coming to an end, indeed I had flown in the 748 aeroplane with my control equations in command. What next? The choice seemed to be USA or Cambridge. Cambridge won.
John Coales had recently 'retired' as research director of Elliots, just a few years younger than I am now. He was starting a new life, building up a thriving research group on control theory. I was told to buy a copy of Zadeh and Desoer, "Linear Systems Theory, the State Space approach." I studied it in my last month or so in Smiths.
A lot of things started coming together. Instead of representing a control system with a whiz-bang set of transfer functions, a set of state variables can be found, each ruled by a simple 'rate of chance' equation. This should not have come as too much as a surprise, because these state variables were exactly represented by the outputs of the amplifiers of an analogue computer.
On the other hand, the book also contained all the mathematical elaboration you could wish, discussing whether two systems were really the same and how you determine that their states correspond.
In July, Ros and I started a delayed honeymoon. After six weeks of roaming round Europe on the cheap - and a week competing in the finals of a talent contest - we settled into a flat in Bateman Street and I started research. The flat had been repossessed by Trinity Hall at the end of a hundred year lease - needless to say, it wasn't in the best order!
I was a research scholar of Trinity Hall and every Wednesday night all the attached researchers and their partners met for dinner in the JCR. There was an Australian couple, Brian and Anne Young and there was also Stephen Hawking and his fiancee, Jane. They had become engaged in Oxford, before Stephen had contracted MS. (The diagnosis was much later revised to motor neurone disease.) At this time Stephen had advanced symptoms, but was still walking and talking reasonably normally. Jane felt that she really must go ahead with the marriage and look after Stephen for the three years or so everyone thought he had left.
That spring, there seemed to be an epidemic. Young wives confided secretly to each other that they were keen to 'start a family' and a complete set of joyous announcements were made almost simultaneously.
In October, Berry Anne was born and at a christening in the College Chapel, Jane became one of her godmothers. It is a pity in a way that Bob Runcie was no longer the college chaplain - he went on to become the Archbishop of Canterbury. Well if you've names to drop, it's a shame not to drop them.
Research went well. The topic was Predictive Control. If you can identify the state equations of the system you want to steer, you make a simulation which will run a thousand or so times as fast. You set the state of the model to correspond to the state of the system, and then you let fly. You can predict the future behaviour of the system for any pattern of inputs you care to apply.
Now 'all' you have to do is to automate the choice of the inputs you wish to try and automate the selection of the 'best ones'. To show it in action, I developed a simple 'ball and beam' experiment which was too hard to control by hand but which rolled the ball obediently straight to the target under automatic control.
In 1966, John Coales was President of IFAC, so the machine, 'Cupid' (Cambridge University Predictive Iterative Device) was shown off in the Festival Hall, London. Ros designed some beautiful posters to explain it, which were reproduced in my thesis.
Around this time we had a distinguished visitor from Moscow - a strange, secretive place at that time. He described his mathematical tour-de-force in designing a fuel-optimal controller for a soft, unmanned lunar lander. This lost some of its gloss in that two or three probes were already embedded deep in the lunar surface.
I had a bee in my bonnet at that time about the virtues of sub-optimal control rather than optimal. The fuel optimal strategy involves turning on the motor at the last possible moment, then burning at the maximum possible thrust until the vehicle lights gently on the surface. If the motor is a second or two late starting up when the vehicle is falling at several kilometres per second, 'lights gently' is not really the correct expression!
I calculated the extra fuel required if you switch on the motor several seconds early, then taper off as you near the surface. It was negligible. Moreover any fuel carried all the way to the surface is of absolutely no further use - it is the end of a one-way trip. This was suggested to the visitor with great diplomacy, he returned to Moscow - and the next probe landed safely! Now I am not making any direct claims to have helped the Russian space-race...
All good things come to an end, including the pampered existence of a research student. I started to look round for an avenue back into industry, but was tempted instead with the choice of a University Demonstratorship or a Research Fellowship.
I took the demonstratorship, a sort of assistant lectureship. One of my earlier teaching duties was to create a control course for the final year, electrical. I decided to emphasise the state-space approach - still fairly novel. To give it added 'zing' I called the course 'Chronosophy' - 'thinking about time'. I am sure I coined the word before Ursula LeGuin used it in a novel - the way to check this is to look in the Cambridge Reporter of the time.
The state-space idea is on the edge of a lot of ancient metaphysical and philosophical discussions. If you have the complete set of differential equations for the Earth, or even for your immediate environment, plus a complete set of initial conditions, then in principle you could say that the future could be completely foretold. Of course it is out of the question except for the few simple systems the engineers like to play with - and not all those behave exactly as you might imagine.
A later teaching duty was to set up a control option in 'Part II (general)'. The syllabus had already been approved. It included everything that was in the one-year post-graduate control course. It proved a tough challenge for one-quarter of an undergraduate final year!
The Control Group moved almost at once to the disused University Press building in Mill Lane, with an air conditioned computer room in the basement. A PACE 231R (Mark 5) analogue computer was stitched onto an Elliot 4130 computer to make a 'powerful' hybrid computer. Later on, there was a great event when the memory of the 4130 was increased to 64K!
To make the system useful for industrial analysis, there was in the budget a 'data logger', for which an analogue to digital converter and a hugely noisy tape deck had already been ordered. I enthused about the possibility of including a minicomputer in the system, and so inherited a host of problems. The ruggedised tape deck needed a 400 cycle three-phase supply, not a common thing to find around. I tracked one to earth in a London surplus store for fifteen pounds! We selected a computer, a Raytheon 703, with all of 4K 16-bit words of memory. An interface was designed by ERA, a consultancy group in Aachen, Germany. We waited eagerly for the system to re-enter the country.
It became stuck in Customs.
When we eventually argued it free, we found that someone had stood on the teletype and broken the paper tape reader - the only automatic way to load a program into the computer. I had to 'punch' the first test program in by hand, on a row of sixteen buttons.
We had a tape deck, so it seemed sensible to use it and I wrote an operating system to hold programs and data on magnetic tape. As you can imagine, having to squeeze into 4K, alongside a Basic programming system and run-time overlay, it was not huge!
Joined by researchers David Hedgeland and John Moughton, we achieved great things on that 703, writing in assembler language. We proved the feasibility of a typesetting system driven by a scanned laser beam for the Monotype Corporation, and they financed a three-year research programme.
An early decision was to try another computer system. Computer Automation had just launched the LSI2, and were selling it as the 'Naked Mini' at a price of only one thousand pounds for a 4K memory version. It actually had 8K, half being disabled by a link on the board. That link was the first thing to go. We developed displays based both on storage screens and on direct driving of a colour television monitor. (Clive Sinclair 'borrowed' our technique for the ZX80, after one of his technicians visited the group.)
We also bought one of the very first floppy disk mechanisms, built an interface and designed that into the operating system too. With software character generation for the screen driver, disk filing system, keyboard interface, low-level interface to a daisy-wheel printer and a heap of experimental input-output, the whole system still fitted inside eight kilobytes.
We had a true personal computer system, long before the expression was coined. We wrote text editors, allowing dissertations to be produced, we played music, later linking both computers in a sort of 'midi' fashion, we encoded acoustic signals and analysed them. We also started to commercialise, in a very amateur way. We would buy a complete set of peripherals for a 'client', such as the British Ship Research Association, and while we 'integrated' their system we had an extra research machine to play on at zero budget.
Soon we were properly funded, however, and bought ten Naked Minis to build into a network. This was in 1971, when networks were also a thing of the future.
In 1971, our music-playing computer was on show at the Festival hall at the IEE centenary - John Coales was now president of that. In fact exhibitions had become a common pastime, including a robot 'head' which was shown at the Institute of Contemporary Arts and later went on to a US tour, plus other gadgets shown at the Institute of Physics and a host of inventors' exhibitions.
We founded Digital Design of Cambridge, made some money and then dissolved it. David Hedgeland completed his dissertation, typesetting it on the laser prototype and then took a senior position in the Monotype Corporation. The computer system had become the heart of the Acoustic Telescope, sold to research groups in France and Germany. John Moughton helped found Toltec Ltd to carry on digital consultancy, including acoustics work. The Professor of Acoustics was invited to join, then was not given the power he expected, so he founded Top Express in direct competition. Things were getting complicated.
Since 1968 I had been a Fellow of Sidney Sussex College, with a free lunch every day and an additional stipend for supervising undergraduates. There were five feasts a year and Fellows had the onerous task of attending a dinner every Friday night where we drank up the bin-ends of the Latour '45, the Yquem '45 and other bottles thoughtfully laid down in their youth by our elders.
It turned out that John Conway had also become a fellow of the college. He regaled us with the wonders of his new game of 'Life'. A Research Fellow was John Stuart. His topic was gravitational waves. He pointed out what should have been obvious, that waves need two 'state variables' to get along. He put it that there had to be a companion field which was to gravity what magnetism is to electricity.
When Stephen asked me over to DAMPT to discuss a Cambridge version of the Berkeley experiment, I had some misgivings. The idea was to hang up a huge mass of metal and wait for it to go "ping" when the gravitational wave from a collapsing star hit it. Stephen wanted an ultra-sensitive microphone to mount in the middle. This seemed to me like hanging up a huge electrically charged ball and waiting for it to tremble when an electromagnetic wave - such as the light from a star - hit it.
I didn't have much to contribute by way of miraculous microphones and none of these experiments has really paid off yet. I still think that we should be looking for the companion field instead.
Then came the chance of a Readership at Portsmouth Polytechnic for an amazing increase in salary.
With two young children - and a newborn baby - there was a lot of appeal in living in a seaside resort. Even so, I am often asked, "Why would you ever want to leave Cambridge?"
I suppose that the real answer was the pressure of history, the factor which makes Cambridge so appealing to people who don't live there. Year after year at the Foundation Feast we would pass around the loving cup and intone, "In piam memoriam Fundatricis nostrae," under the watchful gaze of Oliver Cromwell's portrait. The college had been founded in 1595 to refute slanderous allegations made against Lady Frances Sidney, Countess of Sussex. (The college motto is "Dieu me garde de calomnie.")
Oliver was born some three years after the college was founded and became a fellow commoner at the age of seventeen. During Britain's later brief experiment with becoming a republic, this saved the college from a lot of the ravages which befell many others. Quite recently his head was buried in a secret location near the college chapel, three hundred years after he lost it.
In this sort of perspective, the nineteen years which had passed since I had first come to Cambridge seemed a mere pinprick in time. It seemed as though the ultimate success would be to become preserved in the archaeology of Cambridge like a fly in amber.
Instead I had a chance to be part of something new, as Portsmouth Polytechnic struggled towards becoming a university.
I had to start afresh to build up a research group. An early member was Harjit Singh, and we persuaded the director of the Dutch Singer factory to fund us to develop a new mechatronic printer. Courses for industry, introducing managers to the wonders of the microprocessor, led to a host of new contacts and a lot of consultancy. I fondly remember giving such a lecture in about 1977 in which I foretold that one day IBM would announce, "The mainframe is dead. Long live the microcomputer." In the audience there was a strong IBM turnout, from their European centre in Havant.
A neighbour, Eric Woolley, was a member of the Management faculty. He had a request from the catering department of the local hospital. They wanted a machine to read and analyse the menu choices of hospital patients. It was not really too hard to design a mechatronic card reader and the software to go with it. So was born B&W Electronic Systems Ltd (I was the B), which is still trading, for many years both the largest British supplier of computers to hospital caterers and the only computer company in the Portsmouth Yellow Pages.
With the SERC Robotics initiative, research money came flowing in and the group grew rapidly. An early project was "The Craftsman Robot", using sensory-based control. Later on a system for computer assisted quality control won multi-million support form the European ESPRIT initiative.
Arthur Collie took leave from Turnright Controls, then part of the Tube Investments group, to join me to build up research on walking robots. A management buy-out of Turnright Controls from the Tube Group turned out to be less than successful. We joined forces in setting up Portech Ltd from the ashes, selling walking robots to the nuclear industry, test equipment to manufacturers and nowadays turning over several million dollars per year.
Berry Anne went to St Hugh's College, Oxford, then started to build a career with BBC Television. In 1991 she married Tim Dadd. Jane Hawking came to the wedding on her own. It was at that time that we learned that following Stephen's success with his book, there had been a marriage break-up.
Richard went to Peterhouse, Cambridge, gained honours in maths, then vanished into the attic to write software for some Government clients. William was all set to take his O-levels, living in Cheltenham with his grandmother and we suddenly realised that we had been in Portsmouth for sixteen years.
Winters seemed colder, the windsurfing season seemed ever shorter. The spread of the Micromouse Contest around the world had taken us to places like Singapore, Hong Kong and Japan. A 'side-trip' to Australia taught us that there were warmer places to be in January! Australia beckoned!
We ran away from all the children - but there was no escape. William followed us within months, covering himself with glory at TGS and getting a first at UQ. At first we enjoyed hearing Berry Anne's BBC World Service science series here in Toowoomba - even if a bit faint and scratchy. However it was not long before she landed a job in Melbourne as a producer with the ABC. Richard set up a business in Malaysia and another base in Sydney. Now the whole family is Australian, though we are all 'global' in many ways.
The vision-guidance research centred in the NCEA is being exploited through Case IH, a major US tractor manufacturer.
From the experience of running an international conference in Portsmouth and a workshop in Athens, I dared to institute an annual conference on Mechatronics and Machine Vision in Practice. It was first held in Toowoomba in 1994, in Hong Kong in 1995, in Portugal in 1996 and back in Toowoomba in 1997. This year it will be in Nanjing, China.
I am on the Control Board of the Institution of Electrical Engineers in London. They fly me over there for meetings twice a year. My son-in-law Tim is on the board of an American company and my son Richard shook Dr Mahatir's hand when receiving an award for his Kuala Lumpur company.
The world has become very compact indeed.
The only 'down' side to the move is that I now have to take my begging bowl to the Australian Research Council, which seems to move in a mysterious way. One of these days I might tap some of their special academically-scented money, but in the meantime the earthy industrial sort seems to spend just as well.
One of my hobbies has been setting the world to rights with articles in popular science journals and on Ockham's Razor and the Science Show. Berry Anne has joined in the competition and is two Ockhams ahead!
One of my perennial targets is the fear industry. There seems to be more money in scaring people about cellular phones, about microwave ovens, about the 'millennium bug' and especially about the 'greenhouse effect' than is put into the whole of research that is actually useful. Afterwards I will give anyone prepared to listen a half-hour sermon on the dozen or so fallacies in the greenhouse dogma.
Another favourite theme is to taunt the scientists with their lack of logic - caused by their earnest desire to prop up some outworn theory as though it is proven fact. Here is an example.
Stephen's popular book on cosmology (the early parts of which seem to lean heavily on James Jeans' "Riddles of the Universe") has started to make an appearance in a television version. The first program showed the obligatory police car with siren which changes pitch as it rushed past, introducing the Doppler Shift. Then, as ever, the following argument was trotted out: If a star is receding, its light will be shifted in frequency towards the red. Distant stars are observed to have a red shift which increases with distance. Therefore they are receding and the universe is expanding.
Look at that logic closely. It is the old 'pons asinorum'. Cats have tails. This animal has a tail, therefore it must be a cat.
It might well be a cat, but it could equally be an elephant.
The expanding universe is one theory of many, and its ramifications have a lot of flaws, especially the 'Big Bang', which assumes that everything started from a single point. Now if you lump the mass of the universe into a small space it will form a 'Black Hole', something Stephen has thought about very deeply. It will have a gravitational field so strong that nothing can get out, not even at the speed of light.
Personally I like the idea of curved time. Light from a star a billion light-years away has been travelling for a billion years. Does the one-second of time then, which determined the frequency with which the light started, really have to map into one second 'now'?
Put on the spot, the scientists wriggle with a get-out, "There were different laws during the first few microseconds". (In fact it would take a good few minutes for anything to get beyond the event horizon, even if it could).
I must say that it shows extreme cowardice to run away from the possibilities of infinity and eternity.
We have to decide what game we are playing.
Descartes tried to be the ultimate sceptic and got stuck at "Cogito ergo sum." From pure logic he could only assert his own existence, because he was doing the asserting. Everything and everybody else might well be a figment of his imagination. We have to commit to some measure of belief if we are to get any further.
If you take the ultimate creationist approach, everything could have been created just five minutes ago. Maybe your initial conditions were set up with memories of all your personal life histories and of having heard me start talking, but in reality nothing of this life actually existed any earlier. That approach is a bit of a dead-end, too.
A 'sensible path' can be selected, but it is a matter of choice. The sensible path if you are a scientist is to consider all the observations and devise the simplest set of rules which can explain them. If you have to say that the rules changed at half-time, that is unacceptable cheating. By all means devise experiments which will test your rules, but remember that unless those experiments give results you do not expect you will learn nothing. You will just pour another layer of concrete onto your prejudices.
A successful experiment is a failed experiment. Of course this does not apply to long-shot 'quests' like SETI, the search for extraterrestrial life. If that one succeeds the fear industry will certainly rake in the cash - though it might be some other sort of creatures which make a killing.
As an engineer, I enjoy taking things apart to find out how they work, how they were designed. It is called 'reverse engineering'. You find that little features you might not have noticed, such as one of the ridges in fly-net framing, has a huge part to play in making everything fit neatly together.
It is fun to try reverse engineering the universe.
It seems to me there are lots of things which push coincidence a bit too far. You can ask why the oxide of hydrogen, mass eighteen , is liquid water while the oxide of carbon is a gas - and is nearly three times as heavy. Yes, of course there is a nice theory to wrap around it, there has to be because it is so. Why is water densest at four degrees, so that lakes freeze not from the bottom, but from the top and so give the fish a chance?
A lot of these points can be answered by serendipity. Were they not so, we could not exist in our present form and would not be asking the question. Our counterparts might instead perhaps be asking some question about molten silicon!
But take another question which is not life-threatening, the apparent sizes of the sun and the moon. The orbits are tilted, so that eclipses are scarce. Often the sun appears larger than the moon, so that when an eclipse happens it is annular. But just in a while, in a very infrequent while, the moon will appear larger than the sun and the eclipse will be total.
Imagine a grant application to the ABRC - the Ancient British Research Council. "I need ten thousand man-years of effort and a large building site. What will it do? Why, it will predict eclipses. Thank you very much, we'll have Stonehenge finished in a few decades or so." The magic of predicting eclipses was powerful indeed - even with the huge probability that the British weather would be cloudy.
If total eclipses happened every fourth Thursday, would they hold the same mystique?
The creationist approach, a universe built planet by planet, is just too simplistic. I like to imagine God saying, not "Let there be light" but "c-squared del-squared phi equals d-squared phi by d-t-squared". Let there be that very special kind of light which can propagate over billions of light-years without dispersing. I like to imagine God setting up all the laws very carefully and then waiting, perhaps for eternity, to see things eventuate in the way they had been planned.
Mind you, I like to imagine God in other ways too. In asserting that God created Man in his own image, it follows that God must have many human qualities. The quality which is all too often ignored is the sense of humour. Maybe the Earth really was once at the centre of the universe. Then a few smart alecs thought they had everything taped, so a few 'adjustments' were made. Maybe there was a time when earth, air, fire and water were elements enough and a later time when atoms lived up to their name and were unbreakable.
Michaelson, Morley and others performed experiments to measure the speed of the Earth relative to the aether, by measuring the speed of light in various directions. Perhaps God worked another quick fix so that the speeds turned out to be the same in all directions and said, "Let them try to sort that one out!"
Later on, perhaps God said, "Nice one, Albert Einstein."
It's an amusing thought, but it's no foundation for a serious scientific investigation!
A more sobering thought is of the host of clues which
are right under our noses and which we are too blinkered to see.