Ben Goldacre
Thursday September 9, 2004
The Guardian

· I prostrate myself before you and admit defeat. I’ve been writing this column for nearly two years, and I still haven’t managed to stiff a single multinational cosmetics firm: they’re just too good at constructing legally defensible pseudoscience. I’m picturing huge laboratories and rows of scientists writing incomprehensible but legally sound babble onto their clipboards.

· Cosmetics companies take laboratory data – stuff at a molecular level, the behaviour of cells in a glass dish – and then pretend it’s the same as the ultimate issue of whether something makes you look nice. This amino acid, they say, is crucial for collagen formation. Perfectly true, along with 19 others. No evidence to say that anyone is deficient in it, and, crucially, no explicit claim from the company that rubbing that actual amino acid on your face is what is going to make you look better. That link is made only in the customer’s mind: because the claim that the cream makes you look good is an entirely separate one, made for the cream as a whole, and it’s true, because all creams will hydrate your skin and make you look good. Vaseline, as it happens, also does it rather well, but leaves a greasy sheen. And most cosmetics research, since you ask, comes down to conserving the moisturising properties of Vaseline, but avoiding the greasiness. Diprobase, at less than £10 for a half-litre tub, represents a pretty good stab at solving this problem.

· What about the other magic ingredients? One thing kind of works: cooked and broken-up protein (hydrolysed X-microprotein nutricomplexes, or whatever they’re calling them this month). These are long soggy chains of amino acids, which contract when they dry, and so temporarily contract your wrinkles. That’s temporarily. And all the expensive creams have got that in them anyway. A couple of other things kind of work. Vitamin C, and alpha-hydroxy acids affect skin significantly, although only at such high concentrations that they also cause irritation, stinging, burning and redness: so now they have to be watered down, to pretty useless dilutions. But companies can still name them on the label, and wallow in the glory of their efficacy at higher potencies, because by law you don’t have to give the doses of your ingredients, only their ranked order.

· Now, I’m begging you, find me one that makes a properly fraudulent pseudoscientific claim I can write about without getting sued, and I’ll give you a free tub of Diprobase.