The search for intelligent life

The IQ test is 100 years old - but do its multiple-choice sums and sequences really measure anything, or is it just a way to make money? Matthew Sweet examines the bizarre world of cleverness, and attempts to join Mensa, the club for smart alecs

There were 24 of us in Room 509 of Birkbeck College in London - and nobody needed to be told to take the lift to the fifth floor. A teenage girl with "luscious" in sequins across the seat of her jeans; a middle-aged lady with an impressive array of pencils; a boy who might have been captain of the rugby team; a father and son who wouldn't have been out of place in that documentary about the American National Spelling Bee competition. Not at all the gaggle of elbow-patched former maths teachers you might expect to encounter at an IQ test. Twenty-four of us, all with something to prove - and given the opportunity to do so only because the French government, exactly one century ago, decided that they needed a way of weeding "borderline subnormals" out of the mainstream education system.

There were 24 of us in Room 509 of Birkbeck College in London - and nobody needed to be told to take the lift to the fifth floor. A teenage girl with "luscious" in sequins across the seat of her jeans; a middle-aged lady with an impressive array of pencils; a boy who might have been captain of the rugby team; a father and son who wouldn't have been out of place in that documentary about the American National Spelling Bee competition. Not at all the gaggle of elbow-patched former maths teachers you might expect to encounter at an IQ test. Twenty-four of us, all with something to prove - and given the opportunity to do so only because the French government, exactly one century ago, decided that they needed a way of weeding "borderline subnormals" out of the mainstream education system.

"If you get in the top 2 per cent you can join Mensa," muttered the invigilator, an agreeably downbeat man sporting a tarbrush-beard-and-polo-neck combo. "But it's a funny bloody group. I'm supposed to sell it to you, but let's just say that if you join you have to really want it..."

Most of us did, it seems. Otherwise, we'd have stayed in to watch Test the Nation and not paid £9 to sit in a classroom for an hour sweating over logic puzzles involving fruit and cartoon faces and geometric shapes. It didn't necessarily follow that we were mad keen to attend their annual bash in Manchester - a lecture on police forensics, a disco and a Mongolian-style barbecue, entirely in the company of other people who are terribly good at multiple-choice questions. But we wanted, at least, to have our cleverness confirmed by a letter from the Mensa Testing and Admissions Co-ordinator, informing us that we were in possession of an IQ of 148 or more - putting us in the same genius category as Albert Einstein, Carol Vorderman and Sir Clive Sinclair.

This is the supreme irony of the IQ test. It was designed by its makers to distinguish between differing degrees of "idiocy, imbecility and moronity" in children, and was later championed as a way of selecting "feeble-minded" candidates for incarceration or sterilisation. But, by 1946, when Mensa was founded by an eccentric Australian named Roland Berrill, it had become a tool to detect adult geniuses and the organising principle of a bizarre social club.

You need to be pretty smart to comprehend the history of the intelligence test. It was conceived by Sir Francis Galton, a fabulously eccentric 19th-century polymath who explored Ovamboland in southern Africa, pioneered the criminological use of fingerprints, was the first to identify the anticyclone, and asserted that the female intellect was inferior to the male on the grounds that no woman he knew could make a decent cup of tea or coffee.

Galton was convinced that sensory acuity was the key indicator of human intelligence. He set up an Anthropomorphic Laboratory in South Kensington, crammed with machines to test reaction times. (Cannily, he charged his 9,000 subjects threepence each to supply him with free data.)

But it was his protégé James McKeen Cattell who undertook the most energetic practical work. Cattell spent many happy hours pressing rubber-tipped compass points into the foreheads of his subjects and noting how long it took them to complain. The fun ended in 1901 when Clark Wissler, one of Cattell's graduate students, squished Galton's theory by showing that there was no correlation whatsoever between high scores in Cattell's tests and high academic achievement; a hypersensitive forehead was no guarantor of straight As.

The term "intelligence quotient" doesn't emerge until 1912, when the German academic William Stern named the figure produced by dividing a subject's so-called "mental age" by chronological age. (Four years later, an American counterpart, Lewis Terman, contracted the phrase to one of the catchiest acronyms in science.)

But the tests used to truffle out the numbers used in these equations were formulated exactly 100 years ago by Alfred Binet, a self-taught French psychologist with a passionate interest in hypnosis and a moustache primped into a glossy propeller of hair. Binet's tests are the grandparents of the modern IQ test, giving rise to job-interview psychometrics, sterilisation programmes and the Countdown conundrum.

"They were the first tests that really worked to any substantial degree, because they did make some effective diagnoses," notes Raymond E Fancher, the author of The Intelligence Men: Makers of the IQ Controversy (1985) and a professor of psychology at York University, Ontario. "But Binet was quite sceptical about the quantification of the results. So it's ironic that the IQ, the unitary number of intelligence, became so strongly identified with his name."

Like many in the history of psychology, the inside of Alfred Binet's head was a fairly peculiar place. His principal childhood memory involved his father's attempts to toughen him up by making him lay hands on a dead body. His squeamishness forced him to abandon his medical training, yet he spent much of his time collaborating with the playwright André de Lorde on bloody melodramas for the Grand Guignol theatre in Paris. (His most celebrated work, A Crime in a Madhouse - evil nuns, killer hunchbacks and eye-gouging - dragged its chains across the boards for four decades.)

And the logic problems Binet set for his 10-year-old subjects suggest that his morbid tastes influenced his psychometric practice. One test paper asked candidates to explain the inconsistencies in a series of sentences, such as: "Someone said that if I should ever go desperate and kill myself, I will not choose Friday, because Friday is an unlucky day and it will bring unhappiness," and: "The body of an unfortunate girl was found, cut into 18 pieces. It is thought that she killed herself." (When questioned about this, Binet insisted that French children found such things amusing.)

Binet asked his group of schoolchildren to perform 30 tasks of increasing complexity. He passed a lighted match to and fro in front of a subject's face, and recorded whether the child's eyes followed the flame. He asked them to indicate various body parts, repeat a series of three digits, memorise simple sentences and give functional definitions of words such as "horse", "fork" and "table". Trickier items required children to reproduce drawings from memory or construct sentences including three given words, such as "Paris," "river" and "fortune". The toughest elements demanded that subjects repeat a sequence of seven digits, name three rhymes for the French word obéissance and explain what might be happening in a neighbour's house if they saw a doctor, a lawyer and a priest visit in turn.

Binet saw intelligence testing as a form of medical exploration "comparable to a person's making a colonising expedition into Algeria, advancing always only upon the map, without taking off his dressing gown". He died in 1911 at the age of 54, with the journey half-completed. But his death prevented him from seeing some of the uses to which his ideas would be put: turning public vanity into cash, and advancing the cause of racist eugenics. The two most prominent names in the next generation of intelligence testers, Lewis Terman and David Wechsler, both got rich by selling IQ tests through the post.

Bigotry, though, has been as strong a force as greed. At one of the first meetings of Mensa, a motion was proposed to make black candidates ineligible for membership. (Roland Merrill, the founding chairman, employed a successful wrecking amendment that switched the prohibition to cover "green men with yellow stripes".) In the late 1950s, William Bradford Shockley, who bagged the Nobel prize for physics for inventing the transistor, contended that the lower grades customarily scored in IQ tests by African-American students suggested that the problem might be rectified by a mass sterilisation programme. He was burned in effigy on campus for his trouble.

Opponents of Shockley's views have argued persuasively that the discrepancies in the results were attributable to social and economic factors. Stephen Jay Gould's study, The Mismeasure of Man (1981) exposed the ways in which bigotry has shaped attempts to assess the intelligence of different cultural and racial groups.

But IQ tests remain a useful tool for racist ideologues. In their notorious book, The Bell Curve (1994), Charles Murray and Richard Herrnstein used dodgy IQ data - largely taken from studies funded by the neo-Nazi Pioneer Fund - to "prove" the intellectual superiority of white people. Go to the website of Richard Lynn, Professor Emeritus at the University of Ulster - a man who believes, he has said, not in "genocide" but in the "phasing out" of "incompetent races" - and you'll find reams of comparable racist pseudoscience. His latest article, he asserts, "presents new evidence showing conclusively for the first time that lighter-skinned blacks have higher IQs than darker-skinned blacks [which] supports the theory that the proportion of white ancestry is a determinant of the intelligence of African-Americans". Binet, you suspect, must be positively gyroscopic in his grave.

"Binet's tests," Raymond Fancher observes, "can be very useful in a clinical situation, because they give a much more differentiated view of an individual's abilities. You get much more than just a raw IQ score. And if you're trying to predict the success or failure of an individual, you're probably better off having that information than by simply going on random guessing. But they're far from perfect. They certainly don't measure creativity. The real spark of originality is not easily measured."

On the way out, I collar another candidates: Mary Rayner, the lady with the pencils. She solved the puzzles in a Mensa newspaper advertisement, and was pleasantly surprised at how well she did. "My friends and my GP laughed at me when I told them I was coming to do the test," she says, cheerfully. "It's a British thing, isn't it? It's somehow not decent to succeed. But I did it to prove that my brain still works."

Her analysis of Mensa's system of IQ tests shows a cool logic that doubtless upped her score. "It's all about money, isn't it? They want your nine pounds. But if I pass I'll join and go to a meeting to see what it's like. If they turn out to be a bunch of weirdos, I'll be naughty and cause a lot of consternation."

When my letter turns up 10 days later, it proves not to be an invitation to chew over logic problems by the Mongolian barbecue. "We thank you sincerely for your interest and co-operation, and regret that we cannot invite you to join on this occasion." But I've only missed the pass mark by one percentage point - which means that, at most, my IQ is only two units beneath those of Mensa members such as Garry Bushell, Sir Jimmy Savile and Jamie Theakston - the creamy top of our intellectual élite.

"Under the rules of membership," coos the letter, "you are not allowed to retake this test for 12 months." But it seems that if I'd like to try again, and pay another fee, and return to Room 509 for more fun with fruit and little faces, I'm welcometo have a bash. What's 24 x 12 x £9? You don't need to be a genius to work that out.