Shangri-La Diet: Two Years of Data
Shangri-La diet graph

Graph depicts two years of pre-diet baseline, followed by two years on the Shangri-La diet.

Two-year pre-diet baseline: Sept. 30, 2007 to Oct. 21, 2009
Shangri-La diet: Oct. 22, 2009 to Nov. 14, 2011 (just over two years)
Weight immediately prior to start of diet (ten-day avg.):  222.4 lbs. (100.9 kg.)
Weight as of Nov. 14, 2011 (ten-day avg.):  193.9 lbs. (88.0 kg.)
Total weight lost during this period: Approx. 28.5 lbs. (12.9 kg.)


I lost weight by following an unusual but effective diet that was developed by psychologist Seth Roberts, professor emeritus at the University of California at Berkeley and now based at Tsinghua University in China.  Briefly, the Shangri-La diet involves appetite suppression via the ingestion of flavorless calories in the form of oil and/or sugar water.  The diet is described in a book written by Roberts.  A shorter summary may be found here.  See also here and here.

Here is how Wikipedia describes the putative mechanism behind the diet:

“Roberts’ diet is based on the fundamental principle of a set point—the weight which, according to Roberts, a person’s brain strives to maintain.  When actual weight is below the set point, appetite increases; when actual weight is above the set point, appetite decreases.  Furthermore, eating certain foods can raise or lower the set point.  Foods that have a strong flavor-calorie relationship (such as fast food or donuts) raise the set point, whereas bland foods which are slowly digested (like extra-light olive oil or fructose mixed with water) lower the set point.  Roberts states that the diet is based upon connecting two unconnected fields: weight control and associative learning.  Because of this, the research behind the diet is from multiple fields, ranging from Pavlovian psychology to physiology to rat psychology.”


I first heard of the Shangri-La diet when I read this article in the New York Times back in September, 2005.  The diet seemed so bizarre that the idea of it stuck with me—until, about four years later, I finally decided to try it.

Being 5’ 10” tall and weighing 222 lbs. (1.8 m / 100.7 kg.), I wanted to lose weight, and I was looking for a weight-loss plan that would satisfy the following five criteria:

  1. It wouldn’t require me to change the types of foods that I eat.  I’m a vegan for ethical reasons, and I did not want to stray from that philosophy.  (Note, for example, that it’s very difficult to eat a low-carbohydrate diet without consuming animal products.)
  2. The diet wouldn’t depend on willpower, at least not heavily.  Knowing myself, I would not be able to adhere to a diet that could be sustained only by exerting an iron will (which is, essentially, almost all weight-loss diets).
  3. The diet did not have any obvious characteristics that might make you think that it would be harmful to your health.
  4. The diet wouldn’t require regular exercise.  Unfortunately, I’ve disliked exercise my whole life, and I’ve never been able to maintain a regular exercise regimen for more than a few months.
  5. The diet could be continued indefinitely.

The Shangri-La diet met all five criteria, so I tried it—to some success.

My current routine

I start the day with two large mugs of coffee with stevia (sugar substitute) and some soy creamer.  Around noon, I take a heaping tablespoon of Spectrum virgin coconut oil (it’s a solid at room temperature).  I eat it with my nose pinched shut, so that I don’t taste it.  I follow it immediately with a small glass of water.  About an hour later, I take four tablespoons of flaxseed oil (I use Spectrum Organic, without added lignans), followed by a few gulps of water from a drinking fountain.  I do not pinch my nose shut for the flaxseed oil.  I eat a large dinner around 7:00 PM, and I have a substantial snack (really, it’s more of a second meal) around 10:00 PM.  I don’t eat dessert after dinner, but my snack usually ends with a piece of cake or pie, a chocolate bar, or some cookies.  All the food I eat is always vegan.  I don’t eat between the two meals.  About once per week, I eat a large breakfast (more about this below).

The coconut oil is a fairly recent addition to my diet regimen, as of May 2011.  I added it because I noticed that my weight was slowly creeping upwards.  In email correspondence with Professor Roberts, he suggested adding more flavorless calories to my diet.  I originally took a second heaping tablespoon in the afternoon, but I noticed that doing so reduced my enjoyment of dinner, so I stopped the afternoon dose.

I don’t keep track of calorie counts, but my dinner is larger now than it used to be before I started the diet.  To give some sense of the size, at a Thai restaurant I typically eat a bowl of soup, two fresh vegetable spring rolls, and an entrée with vegetables and tofu.  My 10 o’clock snack is about half as large, or maybe a bit more than half.  I feel quite full at the end of the dinner and following the snack.

Routine at the start of the diet

For the first two months or so, I used extra-light olive oil instead of flaxseed oil, and I would follow the olive oil with a 16 oz. glass of sugar water (9 teaspoons of sugar dissolved in lukewarm water).  I would sip the water slowly, over the course of about an hour.  One day I didn’t have any sugar, and I noticed that my appetite was sufficiently suppressed by the olive oil alone.  Soon after, I tried flaxseed oil and found it comparable to the olive oil.  I stayed with the flaxseed oil because of its possible health benefits.

Other thoughts

  • The first few days I tried the diet, I really didn’t experience much of a reduction in my appetite.  I did manage to avoid eating until dinner-time, but I had a hard time of it.  I was discouraged and started to think that the diet wouldn’t work for me.  However, I did continue it, and I started drinking the water more slowly (I had initially gulped it down quickly).  Taking many sips over an hour seemed to do the trick.

  • On Seth Roberts’s home page, a dieter named Kathy Sierra wrote, “Within three days, I was actually forgetting to eat”. I never experienced such a profound effect.  The diet certainly reduced my hunger to the point where my appetite was manageable, but I would still experience some “break-through” hunger, especially if I felt bored or stressed.  Also, seeing (or, particularly, smelling) appetizing food would bring a temporary wave of hunger.  I find it very difficult to be around other people when they are eating lunch.

  • When I first started the diet, I noticed that I enjoyed my dinner quite a bit more than usual.  This effect continues to this day.  I also eat guilt-free, in the sense that I don’t need to worry about controlling the portion size or stopping before I’m fully satiated.  I eat as much as I want.  On days that I eat breakfast or lunch, my dinner is much less enjoyable than usual.

  • Other Shangri-La dieters have reported a reduction in their cravings for junk food.  I have not found that to be the case.  I’m quite fond of foods with strong, spicy flavors, and I continue to eat things like spiced nacho chips, jalapeño-flavored almonds, and chocolate with chilis.  I don’t snack between my two meals.

  • In certain circumstances, I’m forced to eat lunch, as when there is a family function that I need to attend, or when a meeting at work takes place over a meal.  Luckily, though, such events typically happen less than once per month.  When I know in advance that I will be eating lunch, I don’t take the flaxseed oil or coconut fat.

  • I’ve experimented with other oils.  Udo’s DHA oil blend doesn’t work very well, possibly because it has a fairly strong nutty flavor.  Total EFA (vegan) from Health from the Sun worked as well as the regular flaxseed oil from Spectrum.

  • My advice to those who would try this diet is to stick with it for at least a few weeks, even if you don’t experience dramatic appetite suppression from the start.  Try varying some of the parameters (amount or type of oil, amount of sugar water, speed with which you consume it, etc.) to see if you can hit upon the right combination.

Remaining concerns

  • My initial goal weight was 175 lbs. (79 kg.), but I was not able to achieve that goal.  Nevertheless, I’m reasonably happy with the outcome, especially given the relative ease with which I was able to lose weight.  If I were younger and single (I’m 46 years old and married), I would probably try harder to lose the remaining twenty pounds.

  • The variability in my weight has increased over time, possibly because I did not eat breakfast at all for the first eight or nine months that I was on the diet.  I now eat a large breakfast once (or, sometimes, twice) per week, usually on my days off from work.  When I eat breakfast, I usually do not take the coconut fat or flax oil on that day.

  • Perhaps the biggest drawback to the diet is that I don’t eat with my colleagues at work anymore.  Lunchtime was an opportunity to relax, socialize, and catch-up on events.  I can’t sit with people while they eat—it’s too difficult for me—so I typically work right through lunch.  (On the plus side, though, I get more work done and save money that would have been spent on food.)

Overall impressions

I am certainly glad that I encountered the Shangri-La diet, and I’m grateful to Seth Roberts for having invented it.  Losing 28 lbs. / 13 kg.—and keeping it off for two years—is no small accomplishment (for me, anyway).  I plan on continuing with this diet indefinitely.

I’m hardly an authority on the subject, but if people have questions, they can contact me at

Note:  Most of content in the rest of this site has not been updated since 2004.