An engineer's lament

John Billingsley

The intellectual universe is divided into two worlds. One is the world of the arts, the other is known as the world of science - and that is at the heart of my grumble. This world should be shared at least equally between the scientists and the engineers.

When mankind lands on Mars, it will be heralded as a triumph of science. If something goes wrong on the way, it will be blamed on engineering failure.

It has been forecast that the Great Questions of Science will nearly all be answered in the next twenty years. I agree. But I am also convinced that within a century nearly all these answers will have been shown to be WRONG.

The word scientist means ONE WHO KNOWS. A scientist knows, for example, that mixing hydrogen and oxygen and striking a match is likely to prove noisy. But someone who 'knows' that the universe came into being fifteen billion years ago in a big bang is not a scientist, but a credulist - one who BELIEVES. The big bang is a theory, an explanation of observations which holds good until conflicting evidence causes it to be disproved.. It certainly cannot be proved as a fact or a truth- though any amount of evidence might be found to support it. And who knows, it might even be true.

Scientists delight in propounding laws. These laws, however, are just rules of thumb. Voltage is proportional to current - over a limited range. The stretch of a spring is proportional to the tension - but the 'law' runs out of steam long before the spring is pulled into a straight piece of wire. And these laws are very convenient within their limitations, allowing measurements to be predicted and experiments to be second-guessed.

It is when the law takes on the flavour of a universal truth that the scientist is transformed into a credulist. A highly regarded spokesman for the credulists said recently, "Isn't it wonderful that an electron can be both a particle and a wave at the same time?" It's as daft as saying, "Isn't it wonderful that a platypus can at the same time be a beaver and a duck !" The credulists have got it wrong again. The electron is neither particle nor wave. Think again.

Credulists have to have a stream of new theories to believe in. Those same credulists who forecast a coming ice-age during the savage winters of the early sixties are now proclaiming that they have proved that we are going to simmer in a carbon-dioxide soup.

Credulists are hungry for money to 'prove' their theories correct. They are happy to ignore any amount of data which disproves a theory - unless of course the theory has been proposed by someone else.

A favourite credulist's refrain is, "My theory could be proved if only I could find a ..." perhaps a magnetic monopole or a charmed quark. Consider the quest for gravitational waves.

For quarter of a century, the favoured method has been to hang up a large lump of something heavy and wait for it to go 'ping' because of something going on at the other end of the universe - if the universe has an end, that is. As time goes by, the mass is put in increasingly inaccessible and vibration-free places and is made increasingly denser, larger - and of course costlier. The engineers are cajoled to devise instruments to measure more and more delicate vibrations - though they might just as well be asked to measure the beating of an angel's wing.

An engineer will point out that a wave bounces along by transferring energy from one form to another and back again. So light is an electromagnetic wave which flips between electric and magnetic fields many millions of millions of times per second. It carries energy with it and its field drops off only linearly with distance - twice the distance, half the field strength. (It's the power which is inverse-square)

A simple monopole electric field drops off with the square of the distance, a dipole magnetic field drops off as the third power. A gravitational wave, unless it can find a pseudo magnetic partner, will have to drop off with the fourth power of distance when launched by a quadrupole such as a spinning pair of neutron stars. The twirling of a drum majorette's baton in the next state would give an effect orders of magnitude greater.

Of course the object of the experiment might be to prove the existence of just such a partner field which could give a gravitational wave some carrying power. Now consider. Would you hang up a highly charged pith ball and watch it twitch to the electric field of the light from a distant star? Would you even see it twitch to the brilliant light from the sun?! How about the field from the local radio station?

You don't hunt electromagnetic waves with amber and lodestones. So to find gravitational waves you will have to use greater insight than gluing your ear to a dumbbell.

When the experiment fails, the credulist points out that the brilliant new instrumentation devised by the engineer has a practical application, so the money wasn't wasted after all. But the engineer would have invented exactly the same system if commissioned to measure the beat of the angel's wing.

In my impressionable, formative years I fell into the clutches of the mathematicians. They taught me to distrust everything. A proof could stand or fall by whether a line possessed or lacked just one point of an uncountable infinity.

But in mathematics a proof is a proof is a proof. Theorems are tautologies which are ALWAYS true. Theorems, note, not mere theories. They gain their impregnability from the IF which always precedes them. IF all mice have tails AND this is a mouse THEN this has a tail. The theorem behind this line of reasoning is known as modus ponens. An encounter with three blind, tailless mice does not disprove the theorem - only the premise that all mice have tails.

Credulists are not always as logical.

To say, "IF receding stars have red shifts AND this star has a red shift THEN it is receding" is not modus ponens, it is pons asinorum. It is no more logical than to assert that everything with a tail must be a mouse.

A vacation job in engineering taught me that a design could be proved just by showing that it worked.

Engineers are always eager to please. You want to cross the river? They'll build you a bridge. You can't be bothered to walk to the TV to change channels? They'll give you a remote control. They always seem to dance to somebody else's tune.

The only way engineers seem to be able to gain recognition is by failing short of success. We have all heard of Babbage, whose analytical engine was never completed. Who developed the pocket calculator, though? It certainly wasn't the clever marketing man whose name is sometimes associated with it. Somewhere in the TI organisation a team of engineers devised the TMS1000 microcomputer - but who can name them?

Successful inventions vanish into the maw of large corporations and reappear with a name such as Edison, Microsoft, Black and Decker. The real inventors are buried in obscurity. The engineer can only hog the limelight as long as the product or invention is unmarketable.

Problems are cherished much more highly than their solutions. To solve the problem once and for all is as fatal as opening the gift box of chocolates which has been passed round the family for years. The mathematicians are still mourning the loss of the four-colour problem.

So is the argument all one-sided? Can we really do without the credulists?

I suppose not, really. If the engineers insist on dancing to someone else's tune then we have to pay the piper.

And, who knows, one day we may be able to measure the beat of the angel's wing.

This article first appeared in Australasian Science Magazine and was later heard on Radio National's Science Show.