Why I love exams

This week, thousands of students will turn over the paper, pick up a pen - and try to recall what they've learnt. Why do so many dread this moment? Timed tests are bliss, insists Michael Bywater

This time of year, hell descends on teachers and students alike. The blood drains from the face. There are tears and antihistamines, queues outside the library, stomach-aches, headaches, eye-burn and brain-knacker. Pens are chewed, elaborate colour-coding schemes devised, and midnight oil burned like Eyjafjallajökull. The advent of summer brings no joy. Things are frantically revised that were all-too-often never vised in the first place. Parents who have not written an essay since their own exams suddenly start dispensing mad advice. Students comfort themselves, saying "It's only a multiple choice" or "Einstein got a Third"; those with academic intentions tell themselves that A E Housman failed his finals, yet after more than 10 years as a mere patent clerk was elected Professor of Latin at UCL. Fifteen-year-olds face a comprehensive horror of assessment which, if carried out by the Intelligence Services, would involve accusations of extraordinary rendition, if only to the local school hall. Eighteen-year-olds are winnowed (so they, wrongly, believe) for life. Thousands of teenagers will never become doctors, lawyers, city traders – hurrah – or teachers, because everything now requires a 2:1 degree or better. It must, for them, feel like a simple choice: fail now, or fail later.

It is, in short, that annual festival of the cruel and the unusual: the exam season.

Was ever anything so important so apparently unfair? Is anything so critical to progress in life also so destitute of wit and of joy? Why does it have to be so damned unpleasant, and why so little fun (summed up by All Souls' dropping of the famous Essay, three hours on one word; "Miracles" it might be, or "Bias")?

Perhaps it was always so; but the most joyful three hours of my entire education was the question in the Cambridge entrance exams (do they still exist?): "Can an omnipotent being create an object too heavy for him to lift?" Bliss. An actual invitation to obliquity, showing-off, rhetorical posturing and just plain general buggardliness.*

Contrarily, the most unpleasant three hours were the Part I Anatomy paper. Which is why I am here now, instead of in a hospital somewhere, making things better.

Exams.

Now the well-meaning spectres of fairness and transparency (poor preparation for an unfair and often opaque life, perhaps) have triumphed, and the world is grown grey with their breath.

Neither the omnipotent being nor the course and relations of the saphenous vein were fair. The first indulged my glibness, the second exposed my ignorance. They were unfair just as Sats are not fair now (being dumped everywhere) and the 11+ wasn't fair (it really wasn't fair); just as O-levels and GCSEs and AS and A-levels, as university exams and multiple choice and practicals and the 16th century defence-from-the-pulpit of German doctoral candidates weren't fair, and the viva voce wasn't fair (even though it saw Oscar Wilde refuse to stop translating the Gospel, extempore, into classical Greek on the grounds that he wanted to know how it ended) and the Baccalaureate, which many schools are now introducing instead of A-levels, will soon be declared unfair.

The easiest charge to level at the whole notion of exams is that they are, by intention, by design and by outcome, elitist. But of course they are; yet it's curious, and interesting, to note that etymologically "examine" and "educate" share a semantic ancestor in two Latin words – agere and ducare – both of which in this context have the implication of a "leading-out". You might argue that they are in profound conflict here. If education is the drawing-out of the individual person and the intellect, then standardised examination is a nonsense. You might equally argue that, if standardised examination is not a nonsense, then "education" is playing false to its linguistic ancestry, and that the only thing a standardised exam can test is how well a student has been taught, and learned what she has been taught.

But standardised exams are what we have, in the name of fairness. Huge efforts are put into making sure marking schemes and pass grades are equitable – or, at least, legally defensible, where the outcome is significant. Sometimes, experts in the subject are polled as to what percentage of students should be able to answer a question correctly; sometimes, the approach is reversed: students are given test questions and the percentage who get them right is calculated. Both have the objective of fairness; but both ignore a fundamental point made by the director of, I think, Shrek, who was asked on radio whether he wasn't worried by the other big CGI-animation movie that was coming out that summer. "You don't understand," he said; "this is Hollywood. For my movie to succeed, it is not necessary that the other guy's movie fail."

Alas, exams aren't Hollywood; for you to succeed, it is necessary that I fail. Work around it how you will, that is a given of the exam system, and it has been since the Chinese Imperial examinations were established in the early seventh century AD. I fail; you are District Preferred Man, the Shengyuan; and he – lucky fellow – is Zhuangyuan, first among that year's Imperial elite. All started with the same kit: a cubicle and two boards (serving, together, as a bed and, apart, a writing-desk and bench). Ink and paper were provided. And, of course, the questions. Arithmetic, literature, ritual, music, archery and equestrianism. The Six Arts. And then the divisiveness could begin.

The result – intentionally – was a conformist elite culture throughout the entire vast nation, and perhaps things are not that different here and now. We are still trying to identify the best in whatever definition of "best" the culture happens to favour at the time. How can such a procedure not be elitist?

It seems to be inherent. Try as we might, we seem unable to create the unarguably fair exam. Examiners over the centuries have tried every imaginable strategy. But the only truly accurate indicator of ability and success in life remains hindsight, the privilege of the vicar and the obituarists (and even they disagree).

Nor does it seem to help much how we examine. Multiple choice questions – leadenly abbreviated to "MCQ" – set the teeth on edge in those people – I am one of them – who writes in order to find out how much (if anything) they know and what they think. The MCQ may be choice but it ain't that multiple. It's like a bullying barrister: "Just answer yes or no", when there are damn few questions outside formal logic which can be answered with an ugly binary chop. There's no nuance, no shading, yet neither is it foolproof. When I took the written exams for my pilot's licence, my then instructor told me that, if in doubt, tick the longest answer. Try it for yourself, you non-flying non-meteorologists. What is the Adiabatic Lapse Rate? (a) Pilot losing concentration; (b) The rate of decrease in atmospheric temperature with increasing altitude, in conditions of thermal equilibrium; (c) The average accident rate, excluding acts of God?

Quite right: (b) it is. Congratulations. And it's worth noting that the pass mark – the "cutline", as the professionals call it – in pilots' written papers is 70 per cent. That means that the chaps at the pointed end quite possibly don't know 30 per cent of the theoretical stuff. Now sit back, relax, and enjoy the flight. Conversely, the timed essay petrifies the less fluent. They may know more; they may understand more. But if they can't get it out, they've had it. Sunk. Marking schemes are carefully designed to roughen the smooth edges of the glossier candidates' prose, to see through the rhetoric to the knowledge and understanding beneath, but it's hit or miss. Markers, though it may appear otherwise to candidates, are only human. Keith Hopkins, the distinguished ancient historian, used to tell his undergraduates that anyone who could make him laugh in their Finals paper would automatically go up a class. He was merely being explicit about the normally-unspoken.

Then, too, there are the practicalities, not least the extraordinary difficulty of just simply writing for three hours at a stretch. Most people now type. For them, the physical labour of moving the nib over the page for three hours, perhaps twice a day in the exam season, is literally crippling. One of the best pieces of advice you can give a student is to write everything by hand for at least the term preceding the exams. It's hard to think in agony.

It's also hard to think, period. The other major practicality of exams – particularly in the humanities, though narrative writing still has a vital place in the sciences and science-based crafts like engineering and medicine – is working out (a) what you think or know and (b) how to say it. Here the tyranny of the Outline still rules and students are often advised to write out a hierarchical structure before they even begin to answer the question.

But there's a problem there, too, which is that writing is very often a way of finding out what you know, or think, or think you know. The outline or the spidergram is not so much a way of structuring an essay as a method of clearing the mind before writing. As one who writes to put bread on the table (which may simply mean I'm useless at anything else, so my experience is of limited value) I go in for spidergrams like nobody's business. Often – I have a wonderful program called Tinderbox which indulges me in all this – I then turn the spidergram into an outline, or possibly a tree-chart, or a three-dimensional map of links and notes-within-notes. Yet once the first sentence is written, the spidermindmapstormchartogram finds itself, poor thing, pushed to one side and seldom consulted thereafter. One sentence sets up the next; a paragraph sparks another idea; the train of thought takes a branch line. Curiously, afterwards, if I check the outline, I'll find I have covered most if not all of the ground, but the topography is very different. Most writers I know report similar experiences.

When I teach (because engaging with clever young minds is probably the most rewarding thing the world has to offer) that's one of the three pieces of advice I offer: do your outline first, then put it to one side and don't consult it until you've finished. (The other two are: whenever you reach the bottom of the page, check the question before you go on writing, to make sure you are actually answering it; and – this was told to me by the author Caron Freeborn, with whom I have often taught undergraduates – before you go in to the exam, work out the five points you are burning to tell the examiners.)

And the relatively new kid on the block, continuous assessment, seems on the surface fairer than concentrated examination; yet it places at a disadvantage those whose intellectual makeup is that of the performer rather than the craftsman. You wouldn't want to hear Stephen Hough practising Rachmaninov – the endless repetition of tiny details until he not only gets them right but (the mark of the professional) is unable to get them wrong; nor are you invited to. But continuous assessment is not dissimilar. Some need to have the big picture before they can isolate the crucial details; continuous assessment, with its shifting emphasis on the buzzwords and box-tickings du jour, can trip them up badly. I know. I'm one of them.

Questions like my omnipotent being or the All Souls' Miracle are probably the most unfair of them all. They rely on a timed, one-shot-only display of virtuosity and scholarship. They are, most unfashionably in our times, essentially ludic. They demand playfulness, mental agility, controlled digression, the slantwise view, the unexpected shift in focus. Do any exams now invite the display of such qualities? Or do we no longer value them? Write your answers on one side of the paper only. You may begin ...

... now.

* My conclusion: no, but he could create an even more omnipotent being who could.