"If I don't diet I'll just keep gaining"

This is the fear that keeps the cycle going, and it deserves a straight answer. If the diets go, what's left? Nothing between you and the biscuits. The weight goes up and never comes down, and at least the diets slow it.

The evidence points the other way, and I want to be careful with it, because it's observational and it can be read more than one way.

Kirsi Pietiläinen and colleagues studied over four thousand Finnish twins, following their weight from sixteen to twenty-five and asking how many times they'd deliberately lost more than five kilograms. People who'd dieted once were more likely to be overweight at twenty-five than people who hadn't, and people who'd dieted twice or more, more likely still. The obvious objection is that people who diet are people who were going to gain anyway, and dieting is a marker rather than a cause. So the researchers looked at identical twins where one had dieted and one hadn't. The dieting twin was heavier at twenty-five, by a small amount, with no difference between them at the start. Same genes, same upbringing, one difference, and the one who dieted gained more. The authors' reading was that dieting mostly reflects a tendency to gain weight, and that on top of that, dieting itself may add a little.

Dianne Neumark-Sztainer and colleagues followed two and a half thousand American teenagers for five years. Those who'd been using unhealthy ways of controlling their weight at the start had gained about one unit of body mass index more than those who hadn't, were around three times as likely to be overweight, and were about six times as likely to be binge eating with a sense of losing control. Those were teenagers and the weight-control methods included some severe ones, so I wouldn't carry the figures over to an adult's low-carb fortnight. But the direction is the same as the twins, and the same as Mann's review. Dieting, followed forward in time, goes with more weight, not less.

So the fear has it backwards. The diets aren't the brake. On the evidence, they're part of the acceleration, through the mechanism in chapters 3 to 9. Restriction, hunger, break, write-off, regain, and a fresh start that begins with a last supper.

What's left when they go isn't nothing. It's the ordinary week, and I'll describe it properly in chapter 19. For now I'd only say that "if I don't diet I'll keep gaining" has been tested on you a dozen times, in the sense that you've dieted a dozen times, and you can look at the envelope and ask whether the dieting years were years of losing.