From Times Online

Only the bright commit suicide

Does a controversial theory linking intelligence with suicide rates help to explain why so many scientists kill themselves?

THE PAST few months have seen a series of notable suicides by scientists. Yesterday we heard details of the strange pact in which Dr Michael Griffin and his wife Adele took their lives in a Devon hotel. Two days before that, Harold Shipman killed himself. Early this month the body of Richard Stevens, a haematologist, was found in the Lake District; we await the Hutton report into the suicide of Dr David Kelly.

These are disparate cases: no one seeks to associate a mass murderer with a blameless government scientist or a troubled consultant physician. Yet these men had things in common: they were intelligent and analytical, with similar educational backgrounds. Could there be a theme that helps to explain their suicides?

Writers, psychologists and philosophers have long argued over the reasons for suicide. The problem they all face is that statistics on the subject are notoriously slippery. Take seasonal variations. Some people claim that suicide is related to climate and light, and especially to “the winter blues”. It is true that suicidal thoughts reach a peak in January (when the Samaritans get most calls). Yet in the West the peak months for actual suicide attempts are, universally, the spring and summer months of April, May and June (the only geographical exception is the Antipodes, where suicide rates peak in their late spring: November and December).

The deeper you dig, the more confusing it gets. More women claim to have depressed or suicidal feelings, yet more men commit the deed. Could it be, as some scientists argue, that this is because men are less averse to violence, even self-inflicted? There are further complexities: how do we explain the surreal fact that immediately after Marilyn Monroe’s suicide, rates of self-murder in the US rose by 12 per cent? Or that suicide attempts are most common at both extremes of the adult age range, ie, among over-65s and those in their teens and twenties? Or that suicide rates can fall dramatically when the means become less accessible (as happened in Britain when we switched from lethal coke gas to less dangerous natural gas)?

The subject is a swirl of conflicting data into which only the bravest of scientists will venture to step. One who has is Martin Voracek, a researcher at the University of Vienna Medical School. In a paper to be published in the next few months, the psychologist makes a claim that may shed some light on those suicides of scientists. His startling theory is that suicide can be positively correlated with intelligence — in other words, the smarter people are, the more likely they are to kill themselves.

Voracek’s starting point is the fact that suicide is a growing problem in the Western world: someone commits suicide in America every 15 minutes, and the World Health Organisation claims that suicide accounts for at least 2 per cent of Western deaths. The rate of suicide in parts of Sweden has risen by 250 per cent in the past 40 years.

Most sociologists have argued that this is because of the so-called anomie of modern Western life — that something in urbanised, industrialised society alienates us from friends, faith and family, the mainstays of human happiness. Voracek, though, argues that the higher rate of Western suicide could be because people are, on average, more intelligent in the Western world (and are apparently becoming more intelligent still, as worldwide average IQs rise over time — something called the Flynn effect).

To back up his theory, Voracek has taken the controversial tables of national average IQ values published recently by Professor Richard Lynn and his colleague at the University of Ulster, Tatu Vanhanen. But instead of correlating these IQ levels with national GDP per head (as, provocatively, Lynn and Vanhanen have), Voracek has compared the various IQ averages with national suicide rates.

The results are, prima facie, impressive: there is a strong correlation between suicide rates and national average IQ in most of the countries surveyed. For instance, Jamaica, with a low average IQ of 72, has suicide rates of 0.5 for men and 0.2 for women (all suicide figures are per 100,000 person-years). Albania, with an average IQ of 90, has low suicide rates of 2.9 and 1.7. Germany by contrast, with its average IQ of 102, has suicide rates of 21.8 and 8.3; Japan, with an average IQ of 105, has suicide rates of 25 and 12.

Wherever you look, and whatever the culture, the same pattern can be seen: in Azerbaijan, Greece, Kuwait and Chile there are lower average IQ levels and lower suicide rates; in Austria, Korea, Singapore and Norway there are higher average IQ levels and higher suicide rates. One exception is the UK, with a relatively high average IQ (100) and a relatively low (at least for the West) suicide rate — 11 for men and 3.3 for women.

The obvious objection to these statistics is that suicide rates still have nothing to do with brains and everything to do with industrialisation, affluence and modernity — that advanced capitalist societies are less nurturing of the troubled soul. Yet Voracek insists that he has “partialed out” the variables of wealth, rate of divorce, unemployment and average age. He says that even when you make these factors statistically irrelevant, there is still a correlation between national IQ and the rate of self-murder.

Voracek also cites something called the Terman Genetic Study of Genius. This was a study of the entire life cycles of 1,528 gifted Californian children born in 1920-21. One of many fascinating facts revealed by the Terman study was that the suicide rate among these super-bright individuals was 33 per 100,000 person-years — about three times the average rate for the US (which is, anyway, fairly high on a global ranking).

So why should there be such an apparently strong connection between intelligence and suicide? Voracek points to a 1981 study by Denys deCatanzaro, a Canadian evolutionary psychologist. In his research, deCatanzaro posited the idea that for suicide to take place, a certain threshold of self-awareness, of intelligence, must be crossed. Such higher intelligence could only be human, hence the rarity if not impossibility of animal “suicide”.

DeCatanzaro went on to suggest that from an evolutionary perspective — excepting certain religious beliefs and military scenarios — almost the only time it could make Darwinian “sense” for someone to commit suicide was when they became aware, or wrongly convinced, that they were probably not going to have more or any children, and/or that they were becoming a burden to kin who might otherwise go on to have children.

It is a reductionist viewpoint but a powerful one — and using these concepts, Voracek wonders whether anyone who can achieve such a critical and lucid self-analysis is simply more likely to be smart.

But deCatanzaro, though he has yet to read Voracek’s paper, rejects the Voracek theory. For a start, the Canadian thinks that comparative IQ studies such as those of Lynn and Vanhanen are clumsy and “could be viewed as racist”.

DeCatanzaro believes instead that suicide-rate differentials are explainable by — yes — modernity and industrialisation. He adds that there are, anyway, huge problems with national suicide-rate comparisons because the methods and reliability of data collection are so different. Authorities in Roman Catholic countries, for example, are loath to condemn anyone to Hell for the mortal sin of self-murder; hence, perhaps, the low rates of recorded suicide in Latin America.

In this argument deCatanzaro is backed up by Dr Rory O’Connor, senior lecturer in health psychology at the University of Stirling and head of its suicidal behaviour research group.

Dr O’Connor likewise rejects the intelligence/suicide thesis, and also cites the case of Roman Catholicism. “Look at Ireland,” says O’Connor. “Suicide rates have exploded there in the past 40 years. In part this is because suicide has got more common, but even more important is the fact that the manner of recording suicide has changed, has become more assiduous, and this is because the religious culture has changed. Voracek’s thinking is very problematic. Suicide is multi-factorial.”

Yet there is some support for Voracek’s suicide/smartness correlation. Dr Keith Ashcroft, a forensic psychologist based in Manchester, notes that “suicide is indeed more common among the professional classes: scientists like vets and dentists have notoriously high suicide rates”.

In fact, Ashcroft has a telling theory of his own. He thinks that “emotional intelligence” is relevant tothe IQ/suicide question. Ashcroft wonders whether people with very high IQs might tend to have lower emotional intelligence — ie, inferior “people skills”. After all, everyone is aware of the cliché of the lonely genius, unable to fit in with society, who is eventually driven to suicide by his isolation.

It is certainly a plausible image. But given the swirl of conflicting facts, and the sensitivities of the entire subject, perhaps the only thing we should say assuredly is that we need a lot more research.