OK. The precautionary principle isn't a good way to make policy, and has almost certainly been used to make some bad policy in the past. However, how do we apply the insurance method when we really don't know what the risk is? In the case of, say, measures to restrict paedophile activity, we can probably quantify them and decide whether they are worth taking on a cost/benefit basis.

Any way of quantifying the risk posed by global warming? Stern attempted to quantify the cost, which again ran into trillions. You can quibble over his methodology, but in the absence of a better one, we should give it consideration. It's possible that reducing our energy demand, our use of fossil fuels and our carbon emissions is unnecessary and will have an unintended downside. However, the weight of scientific opinion is rather against it.

In this situation, "we don't really know, we should not use the precautionary principle" sounds like an excuse for ignoring unambiguous warnings on this issue.

Pav, London,

An excellent piece Jamie! These days society seems obsessed with 'eliminating' (or at very least minimising) risk - not only is that impossible, but it's not even desirable. Instead we should be 'managing' risk, taking informed risks where the potential reward outweighs the probability and scope of a negative outcome.
The trouble is that this requires us to take individual responsibility, something which is unfortunately going increasingly out of fashion, especially with our government!

Dominic Graham de Montrose, London,

Jamie, your insurance principle is wrong. When we take out insurance, we knowingly make a losing bet with the insurance company. If we pay £100 to insure a £200,000 house against fire, we can be confident that the risk of the house burning down is less than 1 in 2,000. However, the loss of the house would be such a disaster, compared with the loss of £100, that taking out the insurance is still a good idea. This is the precautionary principle. However, one needs some knowledge of whether the disaster to be insured is realistically possible (the house burning down) or impossible (being abducted by aliens) and one needs to know the cost of the insurance.

Frank Upton, Solihull,

You know, I've been puzzling for years about Jamie Whyte and about how anyone so obviously intellectually ill-equipped should have thought himself to be a 'philosopher'. All is revealed. In New Zealand the term is used for anyone whose lips don't move when they're reading.

eric campbell, harrogate, uk

DDT was used to combat malaria, but resistant mosquito strains developed, so DDT use was discontinued.

Ian Rawlings , Dorset , Dorset

It should be published while some copyright control exists, but ONLY WITH AN ANNOTADED TEXT, so that it can be viewed as a disgusting historical document.

Unfortunately, it is available in Britain, as exposed by the TV programme "The War on Britain's Jews", despite copyright ownership. Extreme British muslims regard it as a near sacred text, and we have ignored this barely subterranean Jew hating for far too long.

Incidentally, as an adolescent I learned that de Sade's 120 Days of Sodom would turn one into an insatiable sex maniac by page 10, and the Archbishop of Canterbury's permission was required to access the one copy, kept in the British museum.
Now freely available, only a scholar of anti - Enlightenment literary curiosities, can get much beyond page 10.

martin bright, Bracknell,

And its even worse than that.... In many cases the "precautionary principle" has a bias twoards the current state of affairs, so that it is applied to doing something, but not "business as usual". So even if there is uncertainty on both the current activity AND the proposed one, the current one is always presumed to be better, because any potential new risks are imagined to be worse than current ones. In fact in many cases the actual risks of current activities are ignored or assumed to continue. Some of the arguments about seatbelts fell into this category - the percieved risk of being trapped in a burning car, overcoming the actual risk of going through the front window, the same goes for smoke hoods on aircraft, and the use of clot-busting drugs after heart attacks. The uncertainty of the new system - which can often be estimated - is allowed to overwhelm the current risk.
We still do sometimes say "she'll be right" unironically, but only about sport.

Dave Parry, Auckland, New Zealand

Ian Kemmish: "In the example of GM crops, it is shown to favour the status quo (that is, inaction) over action."
But that is also a bad policy, which proves the PP is generally a bad thing.
Why is it bad in favouring the status quo over GM crops? Because the status quo is not cost free. Decisions are not made in a vacuum, but opposed to alternative decisions. The GM status quo condemns millions to future starvation; it is not a case of "we do nothing and everything is fine" against "we plant GM and the ecosystem may be damaged." It is a case of "we do nothing and millions will die" against "we plant GM and the ecosystem may be damaged."

Frederick Davies, Oxford, UK

Chris from Edinburgh: reading your comment carefully, "What criteria for robust evaluation of risk would the author seek to apply?" seems to translate as "But climate change is too scary!"

Ian Kemmish: this is not about the choice between action and inaction. If inaction carries a high cost, it might as well be called action, it makes little difference.

What we are talking about is reacting to unknowns - the effect of new technology, or alleged impending dooms that need reacting to - as if we knew all about them, particularly in cases where where some expenditure or loss is supposed to bring overall benefit ("safety").

The opposite irrationality would be one where something of which we should not be certain is supposed to protect us, removing the need to take normal (reasoned) safety measures - superstitious reliance on a talisman.

Felix, Nottingham,

The article presents a false dicotomy - that the probilities for an action are fixed and known, or are unknown and unknowable. If I were to drive home drunk, I don't know EXACTLY how much more likely I would be to crash, and nor does anyone else as it depends on too many factors. I can make sensible estimates based on (other peoples') prior experience however and decide to get a cab. Similarly with climate change - the probabilities aren't known exactly, but there is a reasonable, tested expectation that the release of CO2 beyond a certain point has an adverse impact on the climate, and to ignore this expectation becuase there is still uncertainty in the figures might be just as reckless as drink driving.

Matt, Coventry, UK

I think i am safe,at lest in my country .

Ou Weiping, Xi'an, CHINA

An interesting piece that nonetheless utterly fails to convince. The author's construct, that the PP is only evoked under his definition of uncertainty, is simply untrue. The PP is most often quoted where risk is perceived using the best available information to be very significant, even if not fully quantified, and action is therefore justified. What criteria for robust evaluation of risk would the author seek to apply? Is he seriously suggesting that application of the PP is not appropriate in the case of the hazard posed by runaway climate change, induced by anthropogenic sources?

Chris, Edinburgh,

There seems to be a fallacy here.

In the example of alien abduction, the precautionary principle is shown to favour action over inaction, and it's claimed that this is why it's a bad principle.

In the example of GM crops, it is shown to favour the status quo (that is, inaction) over action.

You can't claim that the PP is unqualifiedly bad whenever it recommends inaction just because, when it recommends action, the action it recommends sometimes appears to be bad to most rational people.

That may be a Bad Thought, but it's certainly not Clear Thinking.

Ian Kemmish, Biggleswade, UK

Quite right. Only a jury of morons would convict or acquit having considered only the case for the prosecution, or only the case for the defence. Yet most of us happily do that on a whole raft of issues (not just GM crops and Global Warming) without being in the least bit clinically stupid. My guess is it's "group-think" - as the Chinese proverb has it, "one dog barks at something, and a hundred dogs bark at the sound".

Ken Leyland, Liverpool, U.K.

At least since the fall of communism we could have been educating ourselves in clear thinking & economics based on a common understanding of human nature. It is a tregedy that there is wide-spread denial of this opporuntiy & need.

Ali Murray, Solihull,

Good article but your example of alien abduction assurance is a little confused. What matters is whether your chances of being abducted in any given year are more than 1 in a million. This determines whether one year's worth of cover (not a policy lasting several years) is worth more or less than £1 (not $1!)

Ed W, London,

You are the author of "Bad Thoughts: A Guide to Clear Thinking" and yet your thinking is outrageously muddled. I find that quite funny.

Andrew, London,

If one were to apply the precautionary principle to war on terror, well, the result would look line the Japanese internment of WW II. Not a very good idea for a whole host of reasons.

Mike, New York, NY

Risk, probability, hazard, chance - all word thrown about with little regard to definition. In my industry (Oil exploration and production) we base many decisions on "risk" - both whether to drill, and whether a particular course of action should be followed, or not. Risk is a defined term - it is the product of both Probability (how likely something is to happen) and Outcome (how bad will this outcome be). These two items are plotted on a matrix (generally 5x5) - so crossing a quiet country lane on foot has low risk (Outcome of being hit by car - death, probability of being hit by car - low) - Crossing a motorway on foor is high risk - Outcome is the same (death) - probability is very likely.

mark, Houston,

Incisive, well argued and controversial as ever. The paragraph on the difference between risk and uncertainty should be taught in all schools and businesses and nailed to every politican's hand. We live in a society dominated by people whose only response to any kind of uncertainty or risk is to ban or restrict. It can take years (as with DDT) to see the detrimental effects.

Tim, London,

Bjorn Lomborg, if I remember rightly, has used the apt term 'statistical murder' to describe the crime of taking decisions intended to reduce one risk while ignoring other, much greater risks. The scandalous ban on DDT, already cited by Nick from Rotherham, is a good example of that. Lomborg showed convincingly that banning pesticide use in the US, as advocated by some because of its alleged health risks, would in fact be likely to cause 20-30,000 additional deaths every year because of the increase in price of fresh produce and resultant drop in consumption.

Stephen Morris, Shrewsbury,

So well put!!

Ever realise our society is becoming ever more obsessed with RISK AVOIDANCE? The way politicians and mass media talk, soon we'll be led to think developing agoraphobic tendencies is the only sensible, prudent and cautionary way to conduct our lives.

Any new disease is blown up into a potential epidemic of bubonic plague dimensions, anything revealed as potentially harmful to the environment must be stopped, eradicated, destroyed, NOW! You should get insured against this, that or the other, ASAP.

Paranoia over risk avoidance is so beautifully complemented by the "growing threat" of terrorism that Western governments are having a field day.

Maybe all this is a way of deflecting attention from the real facts, from realities that are NOT uncertain, and for which our governments are accountable? Maybe this is a way to exert control over the populace - under the guise of "protecting" us by "putting safety first"?

Let's wake up and not be bought in.

N.H. , Southampton , UK

Well explained and argued. Does anybody have Al Gore's e-mail address so I can send him the link to this article?

Kevin Browne, Reading, Berkshire, England

It all comes down to what's measurable and what's not measurable. If it's measurable it matters because the policy maker can be held responsible if it goes wrong. If the consquences are unmeasurable nobody can be blamed directly. The fear of blame or litigation is what drives policy making and "the precautionary principle" is just impressive - sounding language used to cover up this sorry fact.

Baxter Lindsay, Preverenges, Switzerland

So now we know one way *not* to make decisions in the face of uncertainty (i.e. blindly invoking the precautionary principle).

I look forward to the following articles, where Jamie provides a similar analysis of the "she'll be right" principal, and goes on to tell us how we *should* make decisions in the face of uncertainty.

Munin, Edinburgh,

The usual application of the principle is to make sure that the course chosen is riskier.

This happens because the risks of the alternative are never properly assessed in the same way as the risks of the proposal.

My own parents, for instance, would not allow us to be taught to swim, because they were not willing to incur the slightest risk that we might catch polio. then thought by some to be transmitted in some cases in swimming pools, lakes and rivers.

Some years later, before I had taught myself, I found myself drowning in a local river which I had been pushed into. A passer-by miraculously pulled me out.

The risk of being a non-swimmer in a countryside filled with lakes and rivers, was something they had never thought about.

Usually when people invoke the precautionary principle to justify some course of action which avoids a risk, they assume that there is a risk free alternative course. There never is.

Frederick DesLauriers, London, England

"Banning GM crops, for example, will increase starvation in the third world." Do you have any hard evidence of this? Is there any scientific basis for this wild statement, or is it just a case of wooly thinking?

Obviously GM multi-nationals would like people to belive this, but there is no eveidence whatsoever that GM crops prevent starvation anywhere. In practice they have often made the situation worse, as large GM companies try to control, for their own financial gain, what crops and what seeds are used by small farmers.

Mark Snegg, Johannesburg, South Africa

The precautionary principle has probably cost the lives of over 30 million Africans, sacrificed to ridiculous Western posturing over DDT. Bad science and bad thinking created bad policy, which condemned all those people to death from malaria - when a use of DDT would have helped control mosquitos. Good article, Jamie.

Nick, Rotherham, UK

You say:

"The big lie behind the precautionary principle is the idea that we can identify safe options even when we are profoundly ignorant of the probable outcomes."

But we are not ignorant of the outcome when that outcome is keeping things exactly as they are, or changing them to the way they once were. The precautionary principle is then saying that we should favour a certain outcome over an uncertain one. So for GM crops, the precautionary principle would argue against banning them, since we know that at the current level they are beneficial, but would also argue against expanding their use, since we don't know what negative outcomes may arise.

The real problem with the precautionary principle is that it is profoundly conservative and pessimistic. It denys the possibility that change can be improvement.

william, Ilkley,