When is honesty the best policy?

We already knew that most people have stolen stationery from work or kept quiet when given too much change, while sizeable minorities see nothing wrong in making a car insurance claim for pre-existing damage.

Where the report becomes interesting is that it claims we don’t even agree on what honest behaviour is. In other words, people may be honest when they say that honesty is the best policy — they just don’t agree on the policy. So, for instance, while 92 per cent of women said the fraudulent car insurance claim was dishonest, only 85 per cent of men did.

The reason why this is more than of mere passing interest is that English criminal law rests on the idea that there is a common standard of decent behaviour. The ‘Ghosh’ test — named after the defendant in the 1982 trial that set the precedent — requires that juries consider whether the defendant’s conduct was “dishonest according to the ordinary standards of reasonable and honest people”. If these standards don’t exist, then the whole basis of the law disappears.

There’s a potentially fatal circularity to the Ghosh test: what is honest is defined by what honest people think, but you can only identify who these honest people are if you already know what ‘honest’ means. You can get away with such question-begging if it’s obvious what honesty requires, but it is this assumption that the study challenges.

Actually, I’m not convinced that the research does undermine the sense that there is enough common agreement for us to get by. In the car insurance case, for example, the fact remains that nine out of 10 people agree on what is honest, and 93 per cent of opinion does not vary according to gender. Standards of decency do not need to be universal to be ordinary.

However, it seems to me that the real disagreement is not about what counts as honest, but about what kinds of dishonesty are acceptable. People who fiddle insurance claims, for instance, know that they are lying, it’s just that they think in this instance, it’s OK to do so. And almost everyone does think that dissembling is acceptable sometimes, and not just in the reply to the moral philosopher’s favourite example: “Excuse me, but I’m a psychopathic axe-murderer out to slay your neighbours. Do you know if they’re in?”

Even here, though, there is much more agreement about acceptable standards than disagreement. What’s more, much of the variation seems to be the result of gradual social changes, and so are not just reflections of ethical idiosyncrasies. For example, 93 per cent of over-50s thought it was wrong to buy a dress for a one-off occasion and then return it for a refund, whereas only seven out of 10 under-35s did. This is not surprising. First of all, younger people have grown up in a culture that tells them the customer is king and which has made them much more demanding. Second, shops have increasingly become faceless corporations, whereas for older people, to cheat a shop was to cheat a shopkeeper, who was usually a member of their community.

Another example is how wrong people think it is to copy a DVD or CD for a friend. It is not surprising that people who have got used to digital content being free are less inclined to see this as a form of theft than those for whom films and music have mostly had to be paid for.

Put a more sober gloss on the research and the conclusions therefore seem to be less than earth-shattering. People do disagree about what counts as justifiable dishonesty, but not nearly as much as they agree. And standards do shift over time as society changes, but again, not as much as they remain the same. That means the Ghosh test is not fatally undermined, just shown to be imperfect but serviceable. And there’s not a legal system in the world which can honestly claim to be anything more.