Quora .

This question originally appeared on Quora. Answer by Gary Taubes.

Surely there is a safe amount of sugar to eat, but we don’t know what it is. And we don’t know whether it’s the same for everyone.

This is one of the reason why I recommend as little sugar as possible. Here’s how I answered that question in the last chapter of The Case Against Sugar to give you a sense of how complicated such a seemingly simple question really is (and I’m using bullet points because it’s the only way I can indent this section):

  • The traditional response to the how-little-is-too-much question is that we should eat sugar in moderation—not eat too much of it. But this is a tautology. We only know we’re consuming too much when we’re getting fatter or manifesting other symptoms of insulin resistance and metabolic syndrome. At that point, the assumption is that we can dial it back a little and be fine—drink one or two sugary beverages a day instead of three, or, if we’re parenting, allow our children ice cream on weekends only, say, rather than as a daily treat. But if it takes years or decades, or even generations, for us to get to the point where we manifest symptoms of metabolic syndrome, it’s quite possible that even these apparently moderate amounts of sugar will turn out to be too much to reverse the situation and return us to health. And if the symptom or complication of metabolic syndrome and insulin resistance that manifests first is something other than getting fatter—cancer, for instance—we’re truly out of luck.
  • The authorities (or self-appointed authorities) who argue for moderation in our eating habits tend to be those who are relatively lean and healthy; they define moderation as what works for them. This assumes that the same approach and amount will have the same beneficial effect on all of us (and that it will continue to work for them as well). If it doesn’t, of course, if we fail to remain lean and healthy or our children fail to do so, the assumption that, naturally again, follows from this perspective is that we’ve failed in our assessment of moderation—we ate too much sugar or our children did.
  • To understand this tautological logic better, imagine a situation in which cigarette smokers who don’t get lung cancer (or heart disease or emphysema) assume de facto that those smokers who do are those who smoke “too much.” They’d certainly be right, but it still wouldn’t tell us what constitutes a healthy level of smoking, or whether such a thing as smoking in moderation even exists. How many cigarettes could be smoked without doing at least some harm to our health, and could thus constitute smoking in moderation? If we say none, we may indeed be right, but now we’ve redefined how we’re willing to work with the concept of moderation. The same logic may also apply to sugar. If it takes twenty years of either smoking cigarettes or consuming sugar for the consequences to appear, how can we know whether we’ve smoked or consumed too much before it’s too late? Isn’t it more reasonable to decide early in life (or early in parenting) that not too much is as little as possible?
  • Recall the thinking of Priscilla White, who went to work in 1924 with Elliot Joslin at his [diabetes] clinic in Boston and oversaw the treatment of the clinic’s pediatric diabetes cases. “No child can grow up without a scoop of ice cream once a week,” White had said, although the translation of this belief into clinical practice would require that the children who got their weekly scoop also had to inject more insulin over the course of their lives than children whose parents and doctors might have taken a stricter approach. Had White known (as she couldn’t at the time) that eating a weekly scoop of ice cream and taking more insulin in response would make children suffer greater complications from their diabetes and die earlier than those who abstained from the ice cream, would that have influenced her thinking? I’d bet that it would have; I’d also bet that she would have wanted to know the increase in disease burden and decrease in longevity per scoop of ice cream consumed, if such a thing were possible—as would the parents—before deciding whether a scoop a week was “too much” for these children. And if these children never ate ice cream, would they miss it any more than would a child who never takes up the habit of smoking miss the opportunity as an adult to indulge occasionally in a cigarette?
  • Any discussion of how little sugar is too much also has to account for the possibility that sugar is a drug and perhaps addictive. Even if “people just act like it is,” as Charles Mann has written, this suggests the possibility that having the opportunity to consume at least some sugar (or ice cream) is only meaningful in a world in which substantial sugar consumption is the norm and virtually unavoidable and everyone does it. Trying to consume sugar in moderation, however it’s defined, in such a world is likely to be no more successful for some of us than trying to smoke cigarettes in moderation—just a few cigarettes a day, rather than a pack. Whether or not we can avoid any meaningful chronic effects by doing so, we may not be capable of managing our habits, or managing our habits might become the dominant theme in our lives (just as rationing sweets for our children can seem to be a dominant theme in parenting). Some of us certainly find it easier to consume no sugar than to consume a little—no dessert at all, rather than a spoonful or two before pushing the plate to the side. If sugar consumption may be a slippery slope, then advocating moderation is not a meaningful concept.
  • We can also try to define “too much” from a population perspective—perhaps too broadly, too myopically. George Campbell’s estimate from the 1960s of seventy pounds of sugar per capita prior to the appearance of a diabetes epidemic may have been reasonable, and the assumption of the 1986 FDA report that forty-two pounds per capita is a safe amount may also have been, but the appearance of a diabetes epidemic and of diabetes itself are two different things. If the fuse of the diabetes epidemic is lit a generation or more before the epidemic explodes, if the predisposition to become insulin-resistant, obese, and diabetic is passed down and amplified from mother to child in the womb, then it becomes far more difficult to establish at what level of sugar consumption a population, let alone an individual, remains healthy, or becomes healthy again if they’re not. What appears to be a population threshold of seventy pounds per capita yearly might actually be a threshold of thirty pounds a generation or two or three earlier. Once we’ve crossed the threshold and are on our way to becoming an obese and diabetic population, it’s likely that we have become different physiologically, that the children in a population that has been consuming a significant amount of sugar for generations have been programmed differently to respond to a sugar-rich environment from those who were born earlier. There may be no going back, or not without drastic changes in our diet. The existing research provides no way to know.

More from Quora: