Background Information: Ingredients of Timeless Decision Theory

Alternate Approaches Include: Self-empathy as a source of “willpower”, Applied Picoeconomics, Akrasia, hyperbolic discounting, and picoeconomics, Akrasia Tactics Review

Standard Disclaimer: Beware of Other-Optimizing

Timeless Decision Theory (or TDT) allowed me to succeed in gaining control over when and how much I ate in a way that previous attempts at precommitment had repeatedly failed to do. I did so well before I was formally exposed to the concept of TDT, but once I clicked on TDT I understood that I had effectively been using it. That click came from reading Eliezer’s shortest summary of TDT, which was:

The one-sentence version is:  Choose as though controlling the logical output of the abstract computation you implement, including the output of all other instantiations and simulations of that computation

You can find more here but my recommendation at least at first is to stick with the one sentence version. It is as simple as it can be, but no simpler. 

Utilizing TDT gave me several key abilities that I previously lacked. The most important was realizing that what I chose now would be the same choice I would make at other times under the same circumstances. This allowed me to compare having the benefits now to paying the costs now, as opposed to paying costs now for future benefits later. This ability allowed me to overcome hyperbolic discounting. The other key ability was that it freed me from the need to explicitly stop in advance to make precommitements each time I wanted to alter my instinctive behavior. Instead, it became automatic to make decisions in terms of which rules would be best to follow.

With that as background, this is how I made it happen:

I was walking home from class along my usual route I had made a habit while doing this of stopping into Famiglia Pizza and ordering garlic knots. I like garlic knots quite a bit, but I also hated being fat and the way being fat made me feel. Things weren’t quite as bad on that front as they’d been a few years before but they were still extraordinarily bad. I thought about my impending solace and thought to myself: You wouldn’t be so fat if you didn’t keep buying these garlic knots every day.

I thought about that for a second, realized it was trivially true and then wondered to myself whether it was worth it. If I never stopped for the knots I would weigh less and feel better, but I wouldn’t have any knots. Even worse, I wouldn’t have any garlic. But would I rather enjoy today the full effect of never having had the knots, in exchange for not having any? Once I asked the question that way the answer came back a resounding yes. I didn’t know how much it would matter, but the calculation wasn’t remotely close. I walked right past the pizza place and never stopped in there for a snack again.

Using this method seemed like the most useful thing I’d come up with in some time, so I quickly extended it to other decisions starting with the rest of my diet. For each meal I would consume, I decided what quantity was worth it and forbade myself from ever consuming more. I motivated myself to stick to that rule in the face of hyperbolic discounting by reminding myself that I would make the same decision next time that I was making now, so I was deciding what action I would always take in this situation. More generally, sticking to the rules I’d decided to follow meant I would stick to rules I’d decided to follow, which was clearly an extremely valuable asset to have on my side.

I used two other major rules in what I like to call the “Don’t Eat So Goddamn Much, Shut Your Pie Hole” diet. The first was to cut down from three meals a day to two and eliminate all snacks except water, cutting my consumption by more than a third. I’d had practice skipping meals in the past and realized that skipping dinner was far less painful than it looked; within a few weeks I stopped getting hungry at night. The other change was to weigh myself daily and alter how draconian the rules were based on my current weight relative to my current baseline. If I was below the baseline, I’d lower the baseline and give myself a chance to cheat a little. If I was above it by too much I would cut out all meal options that weren’t “wins” in the sense that they had more calories than my average.

I tried incorporating exercise into this program but made the discovery many others have made that exercise didn’t correlate with weight loss. Exercise makes you better at doing exercise so long as you keep doing exercise, but it had no measurable effect on my mission so I decided to let that wait until after the mission was complete. Even then I found several exercise programs I tried to be not worth it compared to not having one, or found that they became so over time. Eventually I was able to find a trainer and I remain happy with that aside from the cost. I also considered changing what I ate, but found that beyond cutting out the worst choices that it was neither necessary nor worth the cost.

The last obstacle on the journey was that as I lost more and more I started to feel worse rather than better due to all of the excess skin that doesn’t go away on its own. It was only after I’d lost all the weight and had the resulting skin removal surgery that I suddenly got up and felt genuinely good about how I looked and felt for the first time in my life. I’ve since managed to relax a number of the rules but was never concerned I wouldn’t do what was necessary to keep myself on track.

Since then I’ve used similar techniques and rules in a wide variety of areas of life. It was only years later reading Less Wrong that I realized that I’d effectively been employing inter-temporal Timeless Decision Theory. That realization allowed me to better understand and formalize what I had done, and gave me a better framework for explaining it to others. A common and justified criticism of using TDT in everyday life rather than as a theoretical construct is to ask where one can find another TDT agent, or indeed any agent sufficiently causally linked to you so as to allow you to utilize that link. My answer to that is that whether or not there is someone else you are linked to yourself. You can be that other agent, the recognition of which can allow you to win and win big.

I am fully aware that to a first approximation dieting attempts that follow similar patterns never work. Most people do not have the willpower necessary to sustain them, or otherwise suffer too much to choose to remain on the diet long term. There are powerful forces working against such an attempt. My working hypothesis is that I had five unusual things working in my favor: I have extraordinarily strong willpower in such areas, I already had strong affinity for rule setting and abiding, I fully believed in what I was doing, I had a life situation that allowed me to experience temporary discomfort due to hunger and I thought of all changes from the beginning as permanent. At least some of these advantages are things that can be learned. If anyone is capable of following in my footsteps, it would be Less Wrong readers. In New York’s Less Wrong group especially a lot of us have had success with various different approaches, and I think that developing mental techniques is the best way to enhance your chance of success.


New Comment
242 comments, sorted by Click to highlight new comments since:
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

I'm eagerly waiting the new Cosmopolitan cover with the line "TDT: THE HOTTEST NEW WAY TO WASHBOARD ABS".

I have the opposite history: I was able to stop stressing out about my weight when I realized that, no, I really don't prefer the idea of a thin life devoid of cheesecake and fried things to the life I currently have where I eat whatever I want and am yea big. (This calculation may change if I become more than yea big with age.) Making this tradeoff explicit in my head actually helped me uncover a couple of weird self-denial habits that did not make any sense according to any metric (specifically, I was not having all the legumes and fruit I wanted, out of some confused subconscious notion that they were displacing vegetables... but meanwhile I had already abandoned the difficult project of limiting my chocolate intake.)

I have had significant weight loss without reducing fried things and still having bi-weekly cheesecake. I had MORE weight loss after getting rid of the cheesecake, but I did go from 220 to about 190 with the cheesecake in my diet. (5'10", male)

The traditional American diet is so bad that most people can likely have significant weight loss with trivial loss of pleasure. This is especially true when combined with a human's natural scope insensitivity.

My diet isn't a lot like a traditional American one. I'm a pescetarian, I cook nearly everything I make from scratch or close to it, and while I sometimes eat junky snackfood, I don't do it that often. I also don't consume soda or alcohol. There might be some obvious trivial-loss-of-pleasure alteration to make (and if you think of one, please tell me) but it's not jumping out at me.
Do you drink juice? Maybe try watering it down. Generalising from the one example of me, watered juice tastes just as good as straight juice, as long as you're not drinking them in the same meal. At the level of macronutrients (that is, ignoring vitamins and HFCS), juice really isn't any healthier than soda. (And regarding a later comment, oil isn't unhealthy either; even the USDA admits this now if you look at the fine print instead of the food groups.)
I do drink juice. I add a little water to it, but find that any more than a little (about one part water to five or six parts juice) drops it below a relevant threshold of pleasantness. I realize it is not a health drink, but I really hate water, and I do need some liquid intake. (With meals, it's (skim) milk.)
6Swimmer963 (Miranda Dixon-Luinenburg)
Have you ever tried adding a drop or two of lemon juice to water? Or (and this might sound weird) plain hot water? It depends on how hot it is where you live, of course, but my ex-boyfriend's entire family used to drink plain hot water, and after thinking they were very odd, I tried it and found it quite soothing.
I don't like it when they give me lemon slices in my water at restaurants. Plain hot water sounds fantastically unappealing.
You should try the hot water thing if you haven't actually done so. The cost is pretty much zero. Have you tried filtering your water, or drinking it very cold? Water tastes like what's in it, and the taste varies depending on the source. Most city water tastes like the chemicals they use to clean it, while water from a cheap filter tastes like nothing. Cold water also seems to have less taste, or at least I've found I can't detect the taste of cold water (as in, with ice in it, not just cold from the tap). Seltzer (Or whatever they called carbonated water in your part of the world) is another "almost-water" drink, much like tea. No calories, enough taste to notice, feels like soda.
I like filtered water less than regular tap water from most taps. (I have blind-taste-tested my ability to distinguish them.) Cold water is better. I can't bear to drink anything carbonated; it makes my mouth hurt.
You know what "Low Fat" is? a warning label. Saturated fat is not bad for you. Switch to whole milk. Water is probably best, but if that just doesn't work for you, try seltzer or calorie free flavored waters.
I can't stand whole milk. Even drinking semiskimmed is unpleasant. Seltzer is not an option because I can't drink anything carbonated; it makes my mouth hurt.
Interesting. I started disliking whole milk soon after I switched to soy milk. Some report the same reaction after switching to rice milk. There are "lite" versions which, while not a substitute for water, could be an alternative to juice.
Similar to your comment about diet coke, having drunk reduced fat milk for about fifteen years, whole milk now tastes nasty to me. Why switch to whole milk when it's not inherently preferable?
Taste preferences are malleable, though. I grew up drinking skim milk, and switched over to whole milk when I left for college. I grew a love for the taste of delicious, delicious fats in my drinks. The first step was switching to higher-fat milk in eating bowls of cereal. It's a good way to get used to the taste while tricking your brain into not noticing the 'weird' texture. Switching to a high-fat diet helped too. Whole milk really isn't all that fatty - half and half is better, and heavy cream is a bit much straight (I can drink it straight, but it is way too easy to give my digestive system too much to handle). There's nothing inherent about preferences. If you want to include milk as party of a healthy diet, you may as well prefer healthier milk.
In what way is higher fat milk healthier? Saturated fat may not be more unhealthy than unsaturated fat (although I'm not sure how conclusive the evidence on this is,) but that doesn't mean that adding calories from fat, without subtracting something else, will make your diet healthier.
Sorry, I kind of assumed you were struggling with "I want to improve my diet by replacing carbohydrate calories from skim milk with fat calories from whole milk". The point I was trying to make is that preferences are malleable things, and when you don't care as much about the cost of changing them as compared to the benefits of having 'better' preferences, you can go ahead and change them.
If you're not trying to lose weight, then no reason at all. If you prefer skim milk, then by all means drink it. However, assuming 1. You are lactose tolerant 2. You are trying to lose weight Then you're better off switching to products like whole milk or, better yet, heavy cream, that provide a higher proportion of calories from fat (trans fats excepted, but those aren't typically present in milk) and less to none from carbohydrates and protein. Also, after a little googling, I so see that some people have raised concerns about possible health effects arising from the industrial production of skim milk. It seems farmers don't just skim the cream off like I thought, but add various other things like milk solids and Vitamin A and D back in. At first glance, the evidence of harmful effects from this doesn't seem compelling--mostly extrapolations from animal studies and hypothesized biochemical causal chains that aren't actually tied to overall mortality rates, quality of life, or anything else we care about. However, it is suggestive that this subject might be worthy of further research.
Juice selection may make a difference to the desirable dilution ratio. A little lemon or lime goes a long way!
1Swimmer963 (Miranda Dixon-Luinenburg)
I don't generally drink juice at all. I don't find it all that satisfying, and it often gives me heartburn... I drink multiple pots of tea per day, though.
Heartburn is acid-reflux... there are fruit juices that are less acidic if you want that.
I'm out of ideas then. (^_^)
You mentioned before that you don't drink tea... but fruit teas are actually quite juice-like. Lemon tea or ginger work that way too. Might be worth a shot?
Well, I hate ginger, but I suppose I could afford to look around a little more thoroughly in teaspace to see if there's one I like without massive amounts of sugar dumped in.

Teaspace is huge.

I don't know if this will be relevant for you, but white teas don't get bitter.

We may try that. My wife has the problem that she was given sugared tea when she was a baby and now she simply cannot drink water: soda, syrup, or the best she can manage is making 5 liter tea ($deity bless whichever company still makes huge steel kettles, 4l and 5l ones) with 5 tablespoons of sugar. While 48 kcal per liter is not too bad, about 150-200 kcal drunk a day, I am afraid for her insulin and try to find alternatives. White tea may be one, thanks. Next step would be coffe without sugar. Or finding an artificial sweetener that does not taste as bad as sacharine or aspartame.
3Swimmer963 (Miranda Dixon-Luinenburg)
Personally, I adore fruity teas like cranberry. They manage to taste juice like without actually being sweet. Do you like green tea, chai tea or ginseng? You could also try rooibos; it's an African tea that doesn't have caffeine and tastes a lot milder than black tea without being exactly herbal.
I suspect that I won't find any teas I like that aren't sweet. I prefer my comestibles & potables to be either definitively sweet or definitively not-sweet, and items that have features of one (e.g. a fruit flavor) without being sweet (or while being sweet, in the opposite case) are not pleasant to me.
Maybe you should try some not-supposed-to-be-sweet herbal teas, e.g. rooibos, if you haven't already. Or some real black, white, green, or oolong tea. For real teas, though, you should look up brewing instructions. Most people overcook their tea, oversteeping black tea and using boiling water for green, and it comes out bitter and disgusting. It was a revelation for me when I tried properly prepared tea.
3Swimmer963 (Miranda Dixon-Luinenburg)
Interesting. Both my brother and sister have the same phenomenon: they love candy and desserts, but dinner foods that have any element of sweetness (like beets, sweet potatoes, or even sweet-and-sour sauces) gross them out. You can sweeten most of those teas a little...of course, that means adding calories to something that's essentially calorie-free.

You can sweeten most of those teas a little...of course, that means adding calories to something that's essentially calorie-free.

Even if you put a teaspoon or a pack of sugar in your tea or coffee, it's still 6-7 times less sugar than in a can of soda (and most fruit juices are not much better). The amount of sugar in juices and sodas is insane.

Also, you can use Splenda, for no calories at all, and it tastes just fine. I know some people can get downright militant about how awful the stuff is, but they are the same people who buy organic when the term is essentially meaningless, and they seem to hate the thought that you are "cheating" to get deliciousness. I simply say to them "Er, human technology has progressed to the point where I can have, say, a sweet breakfast without consuming any sugar, and I'm going to do so. Cheating has nothing to do with it." I drink tea with it alllll the time, too. :)
3Swimmer963 (Miranda Dixon-Luinenburg)
Does it taste the same as sugar? I've found that diet Coke doesn't taste the same to me as regular Coke, and I would prefer non-sweet tea to sweet but weird-tasting tea. Then again, I like unsweetened tea and coffee. To someone who found them really unpalatable, artificial sweeteners would definitely be worth it.
Diet Coke has a long history of not tasting the same as regular Coke. They even made an ad campaign about it (YouTube) in the late '80s. Only Coke Zero is supposed to taste the same.
And fails, unfortunately.
No, Diet Coke doesn't taste the same as regular coke; but why would you want it to? As best I can tell, a preference for the flavor of sugared Coke over unsugared Coke is simply a learned preference like preferring Catsup over Brown Sauce or vice versa. I switched to Diet Coke many years ago, and these days regular Coke tastes wrong and not as refreshingly delicious to me. Stick with it for a while, and you not only get used to it. You come to prefer it. In less sweet drinks like coffee, I'm not sure I could tell the difference between sugar and other sweeteners. FWIW, I do find that aspartame (Equal) works better in coffee than sucralose (Splenda).
Fair enough - I don't like the syrupyness of regular coke, but I drink diet, although it certainly doesn't taste like real sugar. Although I'd ask if you've used other artificial sweeteners than Splenda, because most taste terrible, but it's an entirely different chemical preparation - sucralose which comes from actual sugar, not dextrose or aspartame which come from tar.
I've always found the "tastes like sugar because it's made from sugar" slogan awfully disingenuous. I mean, yes, it does taste like sugar, and it is made with sugar, but it's a chlorinated sugar compound. The fact that it's safe and tastes like sugar rather than say, rat poison, was hardly a foregone conclusion, and was only discovered in the first place due to a lab mistake that could easily have featured in an obituary. On the other hand, there's no reason a compound made using tar needs to taste bad. In terms of elements, there's nothing in tar that isn't in sugar (at least in significant quantities, provided the tar is clean.) Also, dextrose is a naturally occurring sugar.
There's a study that suggests that Splenda causes a change in blood glucose levels: Not consuming sugar isn't the end goal.
Typical mind fallacy, revved up with a claim that people who say they don't resemble you have something wrong with them-- the latter probably needs its own name, probably something to do with preventing feedback. As it happens, I think Splenda tastes inedibly vile, unlike other artificial sweeteners I've tried, which merely taste somewhat off. I do eat some organic food, in the hopes that it will taste better, but there's also some conventional food (including highly processes stuff) that I like and eat.
Here'a an idea that I've been thinking about for a while, any thoughts? Epistemic status is uncertain: Producers are using the buzz-word "organic" as a form of market segmentation for price discrimination. Since organic food is more expensive and marketed at richer consumers, it is not surprising that producers make an extra effort to improve the quality, even if this quality improvement has nothing to do with the agricultural practices. Consumers are rightly noticing that food marketed as organic tastes better, and are demanding more of it. This leads to a vicious cycle that reduces the efficiency of agriculture, which obviously has implications for global warming, deforestation etc. Everyone are following their incentives correctly, but we end up in an inferior equilibrium because of a self-fulfilling prophecy which forces everyone to use the signal "organic" when they mean "good quality".
Um. Basically, producing organic food forces extra expenses upon you, so the organic food has higher costs. I am not convinced about "higher quality". No, it doesn't. I even ran a blind test on eggs -- bought some supermarket-brand generic eggs, and bought some organic free-range extra-special extra-expensive eggs and did a blind test cooking the eggs a couple of different ways. I couldn't tell the difference. For fruits and veggies, there are a lot of factor which influence their quality and none of them have anything to do with being "organic" or not.
I agree that there are a lot of factors which influence the quality of fruit and veggies, and that they are not causally related to whether the vegetables are organic. However, I am convinced that there is some correlation. For example, I expect that it would be difficult to sell unripe mass produced tomatoes as organic. One objective thing I have noticed is the quality of milk. I have an Aeroccino-machine for frothing milk. When I use milk from Whole Foods (an expensive all-organic food store), it consistently creates a great foam, whereas if I use non-organic milk from a normal supermarket, it usually completely fails. It would be great if someone else who owns an Aeroccino machine could try to replicate this claim at home.. I expect that there is a simple explanation that has nothing to do with pesticides, for example that Whole Foods has a better logistics system that keeps the milk properly refrigerated at all times. However, my point is only that many customers will note that the organic milk from Wholefoods foams, whereas the non-organic doesn't.
I've read about other blind tests which found that people can't tell the difference between fancy eggs and ordinary ones. I have felt a little off after eating very cheap eggs for several days in a row. I've seen consensus that free-range beef tastes better. While I said something nice about the veggies to a farmer at the farmer's market, he said that the big difference was freshness rather than better varieties or growing conditions.
Well, the standard local supermarket beef and beef imported from Australia taste clearly different, though "better" is a matter of preferences. There are probably at least three differences between them: (1) Breed; (2) Feed (mostly or solely grass-fed vs. mostly or solely corn-fed); (3) Physical exercise (real free-range vs. limited free-range vs. factory farming).
It's true that artificial sweeteners mean you can get a sweet taste without consuming calories. Beware the conclusion that they therefore don't cause you to gain weight or have other negative health effects though. There's plenty of evidence to the contrary. I agree that eating healthily doesn't mean having to deprive yourself of all delicious foods. Sadly artificial sweeteners seem to be quite problematic, though some types may be less bad than others.
SIgh......I've certainly seen all the 'evidence to the contrary', or at least a significantly representative amount. This is the long and short of it: artificial sweeteners give taste, not satiety, so you won't be as full as you would if you ate sugar, hence may eat more. Also, if you overestimate the number of calories you're 'saving' using sweeteners, you'll undoubtedly end up eating more, and potentially gaining weight. It's the stereotypical "Ooh, I drank a diet coke instead of a real one, saved 200 calories, so I can have a donut!" Conclusion: pay attention to EVERYTHING you're eating, keeping in mind that you DO have a precondition of 'how much food you need', and do so in a manner that consciously minimizes your biases. It's not that hard, but most people don't take such a holistic approach, and I've not ever seen it specified as a factor in the 'studies' on artificial sweeteners. So, the studies are correct, per se, but you and I can hopefully be a little smarter than's pretty much a problem of overcoming internal bias by acting on as complete info as possible.
In my experience, Splenda (brand name of sucralose) tastes identical to sugar, and every study I've read has failed to find any associated health risks in quantities humans can eat. Some studies suggest that people who drink diet drinks tend not to lose weight due to giving themselves "credit" for drinking them and then letting go on something else for a net increase in calories, but if you commit to treating artificially sweetened drinks as a replacement for normally sweetened ones, I don't think it's likely to do any harm. Personally, I like to use it to sweeten decaff black tea mixed with lots of vanilla.
Don't like beets, don't like sweet-and-sour sauces. Do like sweet potatoes, but only by themselves with butter... if I put them with non-sweets (like other potatoes, or savory spices) then they will be too sweet to go with, and if I put them with sweets (pineapple, marshmallows) then they will be too savory to go with. Contrary things, sweet potatoes. If I'm going to wind up drinking liquid sugar anyway, I may as well go on drinking juice, I think.
Huh, me three with certain types of sweet foods. (For example, I generally really really don't like sweet salads. I prefer non-sweet vinegars, etc... But I'll happily om nom nom on, say, chocolate. (though I do favor dark chocolate))
1Swimmer963 (Miranda Dixon-Luinenburg)
Another healthy eater! I also cook nearly everything I eat from scratch (mainly beans, rice, and lentil stews). My guilty pleasures are a) baking, which I find very therapeutic (although I bring most of what I bake to share at work or choir practice, or else leave it for my roommates) and b) eating whatever junk food I can when it's free.
I don't think I'd call myself a healthy eater. I eat a lot of chocolate, put plenty of sweetness into my desserts, and am not shy about the use of butter and oil. "Scratch" doesn't mean "healthy", although it does tend to minimize certain types of unhealthy.
Why do you think adding butter to your cooking makes it unhealthy?
It's in my mental reference class of unhealthful things. Probably because it is pure saturated fat. (mmmmm.)
2Swimmer963 (Miranda Dixon-Luinenburg)
Okay then... I call myself a healthy eater, but that's mainly because I do cook from scratch, because I happen to like things like lentil stew, and because I adore fruits and raw carrots and that kind of thing. But I probably bake cookies twice a week, and while I don't eat all of them, one of the main reasons I like making them is getting to snack on cookie dough...mmm... Aside from health reasons, cooking your own food saves a lot of money. I could keep my grocery bill to $25/week if I were willing to give up a few luxuries like, well, chocolate.
I like lentils and fruits and (cooked) veggies and whatnot too. I don't do it to save money - I don't even pay for my own groceries now, and haven't for about a year.
0Swimmer963 (Miranda Dixon-Luinenburg)
I'm pretty pleased with myself about it, yeah :)
2Swimmer963 (Miranda Dixon-Luinenburg)
I'm sorry if this is a personal question, but how did you manage that? I'm under the impression that you're fairly close to my age (which is 19).
I'm 22. I live with a friend who is willing to cover our joint grocery bill and not charge me rent in exchange for my charming company and a handful of domestic tasks (I cook, pick up around the place a bit, make some of our grocery trips on my own, do the dishes, stuff like that). Before I lived here I lived with different friends; there was a similar deal in place but there were some unresolved issues about how big a handful of domestic tasks I needed to be doing, so I left. Before I lived with those friends, I lived in Benton House working for SIAI, and got my nibbles out of their budget. Before that I was in grad school, lived off campus with a roommate, and did pay for groceries (splitting the bill for both the apartment and the food with my roommate-at-the-time). I should note that if my current roommate says "begone", I don't have any clever ideas lined up for carrying on this enviable situation - I might be able to move in with my best friend, depending on the timing, but I suspect she would charge me rent and that I'd wind up buying my own food. (I have lived with my best friend before, during two summers before my first and second years of grad school; first time around I had a job and paid rent but not for food, second time around I did not pay her anything; but now her living situation is different and I would be less enthusiastically welcomed and would probably have to make up the difference with money.)
4Swimmer963 (Miranda Dixon-Luinenburg)
You have awesome friends! I live in a shared house with a bunch of girls, but I didn't originally know them (I found the room through an ad on the Internet) and we don't share anything except for toilet paper and dishsoap. I'm not sure if my company is "charming" enough to wangle a deal like yours: I can play extroverted and funny at school or at work, but home is my place and I tend to spend a lot of time in my room with the door wedged shut. Anyway, I'm not sure I would like a situation like that: my instincts for living cheaply are strong, but my instincts for living independently are even stronger, which is why I moved out at 17 even though it wasn't really necessary. I'm glad because it's forced me to mature pretty quickly in a lot of ways, which might not have happened otherwise. If that happens, you'll still have saved a year or more of rent and food, and you'll have whatever extra amount in the bank. Always a huge bonus.
I do have awesome friends :) I consider my life adequately independent in the sense that I do not depend on my family. (I haven't lived with them full-time since I was fifteen, or at all since I was nineteen.) There's an important limit to how independent I can be when I don't drive, though, so I find it valuable to be among people; I may as well enjoy the largesse available under that circumstance. (Current roommate doesn't drive either but has a close friend who does and helps out.) I don't know if I'm actually in a better financial situation than I would have been without this string of fortuitous circumstances. My income is vastly greater than my expenses, but my expenses are almost nil, so said income... is tiny. I'd need to get a job-job and keep it if I were paying for rent and food on my own, and might have more leftover cash that way than I do now. However, I have lots and lots of low-stress free time, which is very valuable to me.
1Swimmer963 (Miranda Dixon-Luinenburg)
If I valued my spare time more, I might have more of it... Instead, my life consists of running from one part time job to the other to school to choir practice, and basically collapsing in bed at the end of the day. Once in a while I wonder if it's possible to permanently damage your creativity with enough sleep deprivation, but then I stay up half the night writing, which answers that question. I would probably enjoy all the things I do more (and do them better) if I did less of them...but choosing school over work isn't an option, and I have a "loyalty problem": once I join something, it becomes really hard for me to leave. (Which is why I'm still in a girls' church choir, 5 years later.)
Ack, that might be a case of the candle burning twice as bright but less than half as long, resulting in a loss in net light emitted. In other words, forcing yourself to do creative work while also sleep deprived might be burning yourself out faster than if you got more sleep.
1Swimmer963 (Miranda Dixon-Luinenburg)
If only it were a matter of forcing myself...then I could decide not to, and get sleep! Usually if I stay up all night, it's because I've been itching to work on a particular story all week and haven't had time and am going insane from pent-up ideas.
Maybe if you just noted down the ideas so you could go back to them later?
2Swimmer963 (Miranda Dixon-Luinenburg)
I do sometimes. But when my brain looks at the week ahead and sees a solidly filled schedule, it oftens decides "screw it, if I don't do this now I'll never have the opportunity." Also, I have creative and non-creative moods, and if I don't write while in a creative mood, I often lose out on the chance.
I tried to do way to much throughout high school; studying cello, working with my dad and at a part time job, and then spending most of the night programming, and once I graduated and finished the projects I was working on I lost the enthusiasm to start anything else for almost a year. Sometimes forcing yourself to just do less things and get at least eight hours a day of sleep is the best decision, even if you hate it and lie awake thinking of everything else you could be doing for the entire time. Its better than turning into a world weary introverted otaku. :p
the glib response for this is, of course "get married" ;)
This is in fact part of my ideal life trajectory, but I'm not presently engaged or even seeing anyone, so there's no timetable in place that I can accelerate on demand.
1Swimmer963 (Miranda Dixon-Luinenburg)
I'm not even sure if it's a good idea to date people with the thought of accelerating your marriage trajectory. It seems to me like a) a lot of pressure, and b) a recipe for ignoring your own uncertainty because you want to get married fast. Of course, I'm biased because my parents waited from 1983 until 1995 to get married, a solid 12 years of courtship, and because my mother's brother, who did marry young and impulsively, had a very negative experience of it.
I have no intention of rushing things, I assure you. (Although if we're going by anecdotal evidence, I have a friend who got married when she was nineteen very shortly after meeting her husband, and they are among the most happily married couples I have met.)
1Swimmer963 (Miranda Dixon-Luinenburg)
Which I know. But emotionally, I think my parents' anecdotes weigh the most heavily on me. (This is something I've noticed: I pick out particular attitudes I have that are a result of my specific upbringing, and then go on holding those same attitudes anyway.)
It's not necessarily a mistake to do this, because attitudes are not truly independent of one another. Given your specific genes/environment/attitudes, one particular attitude may work well for you even if it would work poorly for another person. It's difficult to examine beliefs in any manner other than "one at a time", but a belief which is wrong independently may be useful given your overall set of circumstances/beliefs. example: one may be able to get away with substandard food cleanliness if one is vegetarian.
I'm not really sure what that means. A lot, maybe most, of the restaurant contaminated food outbreaks that I read about involve vegetables. For example these involving tomatoes. Presumably the reason for this is not that meats are inherently cleaner than veggies, but that meats are pretty universally cooked and veggies are often served uncooked.
Vegetables and meats can both be contaminated by pathogens, but differently. Vegetables are grown in dirt and handled by many people. It is therefore common for small amounts of viruses or bacteria to be present on their skin. And a small amount of (for instance) salmonella can cause disease. To avoid this, one should wash vegetables. Neutropenic individuals may need to cook them, but typically they may be eaten raw. Meats (and beans, and cooked rice) are a different story. They can not only harbor bacteria and parasites, but may also provide an excellent growth medium. Therefore, meat doesn't just have to be washed, it also has to be cooked (unless rigorous precautions are adhered to). It also has to be kept cold because airborne bacteria may begin to grow on the meat and reproduce to dangerous levels. Even small amounts of residue may cause this problem. So when do you wash your cutting board? If you are a meat eater, the answer should be "every time you cook." If you are a vegetarian, the answer can be "when it starts to look dirty". A bit of old carrot juice on the board just doesn't constitute the same health hazard that a bit of old steak juice would.
0Swimmer963 (Miranda Dixon-Luinenburg)
Definitely something I do.
Keeping with the theme of "possible trivial-loss-of-pleasure alteration", don't skimp on the butter. Use oil as needed, but substitute olive oil, especially extra virgin, for other vegetable oils, especially corn, canola, and soy.
I deep-fry in canola sometimes, and use it when I need a flavorless oil for cooking, but otherwise I use olive oil, and I use lots of butter when that would be nice.
Given your expressed eating preferences, one possible trivial-loss-of-pleasure alteration to make is to switch to sugar free chocolate such as Chocoperfection or Atkins Hazelnut bars. In general, take any opportunity you have to replace sugar with sugar substitutes such as Splenda or Stevia; e.g. in your morning coffee.
Splenda I've never had any complaints about, but in everything I've tried it in, I've found stevia to taste vile, to the point that I was confused that it saw any use as a sugar replacement at all. My mother had the same reaction to it. Maybe a genetically determined taste sensitivity?
I love that moment when you make the realization that you don't have to suffer and quit. Isn't it wonderful?
Yes, it certainly is.
Throwing in my two cents for the "different things work for different metabolisms" rubric: Dieting reduces my quality of life. On the other hand, I enjoy exercising. So, although I try to limit myself to eating actual junk food on an occasional basis, I basically eat as much as I want whenever I want, and rely on getting extraordinary* amounts of exercise to achieve the body shape I want. Contrary to what everyone else seems to claim every time this comes up on LW -- that exercise is largely irrelevant to weight and diet is everything -- this works just fine for me. At this point I'm probably about 13-14% body fat and 52-54% lean muscle mass (as a male in his mid-30s). *Extraordinary by sedentary US standards, that is... around 7-8 hours of strenuous activity per week, plus ~45 minutes every day walking the dog. Admittedly, I might be able to achieve single digit body fat percentage by really dieting, but it's not at all clear that would actually be healthier, and it would certainly make life less enjoyable.

But would I rather enjoy today the full effect of never having had the knots, in exchange for not having any? Once I asked the question that way the answer came back a resounding yes. I didn’t know how much it would matter, but the calculation wasn’t remotely close. I walked right past the pizza place and never stopped in there for a snack again.

I think the interesting bit here is not the comparison you made (it's described quite explicitly in more than one self-help work I know of), but that you transposed it to the past.

That is, you appear to have effectively said, "If I'd not taken action X all this time, I'd have result Y; would it have been worth it?"

I would expect this to increase the usefulness of the technique for people who view the past in an especially concrete way, and/or have difficulty thinking of the future in concrete terms.

In any case, the improved effectiveness you experienced is almost certainly due to this one bit (improved concrete construal counteracting the discounting effect of an abstract future) rather than to anything to do with TDT.

I think I shall try it myself on some things, and see what happens.

It sounds like you just needed something to convince yourself with. TDT isn't special in this regard. With some inventiveness you could also have used quantum mechanics, evolutionary biology, extrapolated volition, or any number of other LW topics :-)

The advantage of TDT is that it is actually supposed to be a method of choosing how to act. The problem with the metaphor is that CDT and UDT would prescribe the same behaviour in this context.

The metaphor still works, but it's just that it should be about the usefulness of decision theory in general rather than about any particular one.
"I used decision theory to decide." is not the point of the post, and that does not seem like it would make a very good post. What is actually illustrated is a calculation that can be done to help one decide and intuitively grasp the reason for the decision.
7Wei Dai
EV might work. ("If you keep thinking about it until you reach reflective equilibrium, you'll probably realize that you don't really want to eat that garlic bread.") I'm not sure how QM or EB would help. What did you have in mind?

Evo bio would say that overeating was more useful in the ancestral environment than it is now, so the brain's signals about desiring food are understandable but mistaken ("retarded" would be an appropriate word). Not sure what QM would say, but I've seen it used to support some weird conclusions.

I suppose you could use MWI as a way of illustrating the decision theory approach: Imagine that there is the you that eats the garlic bread, then the you that doesn't. From there, each will experience many more branches with each passing moment. If you take that forward a while, then you can analyze the amount of utility at each of the two different sets of end leaves, to figure out which branch you want to choose now so as to have the highest chance of ending up in a high-utility sub-branch later.

Yeah, this is the kind of bullshit that I'm talking about :-) A cognitive algorithm cannot "choose" a quantum branch to "continue into", it always continues into both. The perception of choice relies on logical uncertainty about the future output of your deterministic algorithm, not on quantum uncertainty.

Yeah, that's an issue. I suppose I could get around that by emphasizing that it's just a desired branch, and the metaphor would still work. But then again, maybe I should just let it die. :-)
In this case QM helps because it supports normal conclusions. QM tells you eating less deserts makes you lose weight. It just takes longer to do the calculation. Just like TDT typically takes longer to understand that CDT but either work just fine for the purpose of deciding not to eat deserts because you care about future consequences. I suppose the only advantage to either is that thinking in those modes can make folk feel mystic/abstract/deep or otherwise in "Far Mode" so trick themselve into the mode of actually making decisions rather than executing habits.
I will readily agree that it is neither a necessary nor sufficient condition for solving this particular problem.
Assuming that the effects of dieting for a day are very small, it is likely that the utility of not eating knots today is lower than the utility of eating them for every possible future behavior. A CDT agent only decides what it does now, so a CDT agents chooses to eat knots. But an EDT,TDT or UDT agent would choose to diet.

An interesting post. I immediately thought of asking "What habits would I adopt if the long-term effects were in full force immediately?"

I think I have some thinking to do.

Edit: typo.

I like this phrasing.

If I never stopped for the knots I would weigh less and feel better, but I wouldn’t have any knots. Even worse, I wouldn’t have any garlic.

You can buy a bulb of garlic in a grocery store for about 50 cents, and add it to meat and vegetable dishes. It doesn't have to come attached to garlic bread, and it's not the garlic that is loaded with calories.

Garlic powder is even easier, and has the advantage that you can put it on stuff that you didn't make at home if desired without it having that bitey raw-garlic taste.
I'm a big fan of using actual garlic bulbs, but if you get more utility from "dried from a shaker" stuff, try the new Penzey's version. It's far superior to standard garlic powder.
I'm a big Penzey's fan (for one thing, it seems to be the only place I can get ground celery seed), but I cook nearly everything I make myself - I use garlic powder when it's texturally important but otherwise mince my own cloves of the stuff.
0Swimmer963 (Miranda Dixon-Luinenburg)
I don't think Penzey's exists here in Canada, but my favorite store is Bulk Barn. Pretty sure they have dried garlic flakes. They also have about a dozen kinds of chocolate chips, and incredibly cheap beans and lentils. And pure MSG in powder form...I always laugh.
I wasn't cooking at the time, and I certainly wouldn't have any garlic right there and then. However, I will admit I've found other ways to get my garlic needs met including actual garlic bulbs, although you won't find them here that cheap.

Utilizing TDT gave me several key abilities that I previously lacked. The most important was realizing that what I chose now would be the same choice I would make at other times under the same circumstances.

This is similar to the mind hack I am working on to bypass my own hyperbolic discounting.

I assume that I will always make the same choice in similar circumstances. I find that this is a very good approximation of my actual behavior.

I determine the potential consequences of the alternatives in relation to my goals. Sometimes it helps me if I specify the consequences in a way that captures an opportunity cost. For example instead of cost in dollars, I'll consider the cost in terms of new tires for my truck.

I decide what to do -- treating the consequences as though they will occur immediately. In practice I only focus on the top one or two consequences for each alternative -- based on my current value weighting.

For example, every morning at work I am tempted by the pile of donuts in my office's cafeteria.

If I ate a donut every day, in a year I could gain an extra 13 pounds (50 work weeks 5 days per week 180 calories per donut / 3500 calories per pound).

These donuts would cost ... (read more)

It seems to me that one needs to place a large amount of trust in one's future self to implement such a strategy. It also requires that you be able to predict your future self's utility function. If you have a difficult time predicting what you will want and how you will feel, it becomes difficult to calculate the utility of any given precomittment. For example, I would be unconvinced that deciding to eat a donut now means that I will eat a donut every day and that not eating a donut now means I will not eat a donut every day. Knowing that I want a donut now and will be satisfied with that seems like an immediate win, while I do not know that I will be fat later. To me this seems like trading a definite win for a definite loss + potential bigger win. Also, it is not clear that there wouldn't be other effects. Not eating the donut now might make me dissatisfied and want to eat twice as much later in the day to compensate. If I knew exactly what the effects of action EAT DONUT vs NOT EAT DONUT were (including mental duress, alternative pitfalls to avoid, etc), then I would be better able to pick a strategy. The more predictable you are, the more you can plan a strategy that makes sense in the long term. In the absence of this information, most of just 'wing it' and do what seems best at the given moment. It would seem that deciding to be a TDT agent is deciding to always be predictable in certain ways. But that also requires trusting that future you will want to stick to that decision.
This is exactly right, which is why I suggest documenting how you respond to different behaviors. I think that it's only partly deciding to be predictable; it's also noticing in which ways you already ARE predictable. In a lot of aspects of life, there are patterns in your behavior, you just haven't noticed them yet. I pretty much know how little I can eat before it becomes unsustainable/distracting. This is the advantage of actually keeping a record. (I might be able to push below that threshold but at the moment it doesn't seem worthwhile.) I also have noticed that I eat better when constrained by rules than when trying to follow "good judgment." EAT DONUT, in particular, is bad for me. I've also made observations about how I feel on different amounts of sleep, and how many hours of work I can maintain before going crazy. (It's much easier for me to "push" on my work capacity than to "push" on food past a certain point.) In other words: it's worth it to try to know yourself better so that you know what "EAT DONUT" will do to you.
What methods do you usually use to keep track of stuff like this? I can well believe that documenting my calories, sleep, exercise, work, etc. would more than pay for itself in terms of hours and dollars, but I'm a little concerned about the upfront willpower/morale investment...I'd feel a little crazy keeping such rigorous track of what I'm doing, and it would help me overcome that feeling if I knew, in detail, what somebody else had done and why/how it worked for her. I'm especially curious as to whether and how you make any effort to measure the quality of your hours/calories...if you work an hour while distracted, do you count that the same as if you were focused? If you sleep an hour from 10 am to 11 am, do you count that the same as if you sleep an hour from 2 am to 3 am? And so on.
I'm less quantified than I could be, but enough that it makes a big difference over "nothing." I log hours on Joe's Goals; I have to count distracted hours the same as focused ones (what else would I do?) but I make up for it by also counting tasks completed. Sleep I don't log regularly, except that now I notice what time I go to bed (I didn't before!) so that I'm actually aware of how many hours I sleep a night.
At least a partial answer is here.
I worry about making precommitments for many of the reasons you bring up; our natural tendency toward hyperbolic discounting makes sense when we need to reason in the face of uncertain risks. But I have found that when I focus on long term uncertainty I tend to lock myself into my current behavior even if it works against my current goals. To avoid the uncertainty inherent in making a commitment, my approach is about how to make a choice for right now -- based on my current goals. By choosing to not eat a donut right now, I am not deciding anything about my behavior tomorrow. Tomorrow I may have to repeat the same process of reasoning; if my state tomorrow is similar to my state today I will probably make the same choice, but if it isn't similar I may make a different choice. No guilt, no fuss. I am using my assumption -- that I will always make the same choice in similar circumstances -- to help scope and quantify the consequences of my alternatives. In the case of my example it allows me to scale the consequences to a level I can more easily compare to my goals. In a year I want to weigh 10 lbs less than now so eating 13 lbs of calories as donuts appears to work against that goal. This approach allows me to make an immediate decision which supports my long term goals, while only experiencing the actual risk of this specific choice, and not the combined risk of all similar future choices. If I discover unexpected negative consequences from this choice then my state will be changed; which I will take into account the next time I face similar circumstances. For example if I discover that not eating a 180 calorie donut in the morning leads to me eating 300 additional lunch and dinner calories, then clearly I will start choosing to eat donuts in the morning. But in fact I discovered that the opposite was generally true; when I ate a donut in the morning I tended to eat 200-300 more calories during the rest of the day. By lowering my perceived exposure to risk I fel
1Swimmer963 (Miranda Dixon-Luinenburg)
I suspect that some people behave more predictably and/or can predict their own behaviour better than others (I don't think those two things are the same, or necessarily correlated). Which would make it easier to be a TDT agent. Mood stability might be a factor.

I am...very impressed. Currently searching for areas in MY life where I can apply this.

Me, too.

A simpler argument would be noticing that what you're actually doing is not just taking too many calories today, but following a strategy of taking too many calories every day. You don't need TDT to see that, it's a matter of recognizing a precommitment, and making decisions about alternative precommitments (strategies).

Zvi mentioned hyperbolic discounting. What if an agent's preferences are actually described by hyperbolic discounting? Then different versions of the agent in time have different preferences, so they are essentially different agents. Consider just two such agent-moments. Each agent-moment would prefer both not eating garlic bread to both eating, but prefers even more itself eating while the other doesn't eat.

Since they have different preferences and the earlier agent-moment can't physically force the later agent-moment to make a certain choice, the analogy with PD seems pretty good and TDT does seem relevant here.

Indeed, there is nothing irrational (in an epistemic way) about having hyperbolic time preference. However, this means that a classical decision algorithm is not conducive to achieving long term goals. One way around this problem is to use TDT, another way is to modify your preferences to be geometric. A geometric time preference is a bit like a moral preference... it's a para-preference. Not something you want in the first place, but something you benefit from wanting when interacting with other agents (including your future self).
I see.
Indeed, it's an example of the general problem-solving strategy of ascending (or recursing) to the meta level.

The resounding impact of this post was to make me crave garlic knots. Not sure I can get any here though... all the pizza places in Berkeley are either chains or places with limited menus selling nothing but California style pizza (and the best ones are closed on Mondays!).

They can be homemade, you know. Looks like pizza dough plus some pretty basic stuff.
More instant gratification, plz
Pretty sure TJ's sells pizza dough, which is hardly instant, but does remove a few steps.

On the matter of exercise, there are two kinds, represented by walking and weightlifting. I.e., there is the low-intensity long-duration kind (such as walking, bicycling, dancing, aerobics), and there is the high-intensity short-duration kind (such as weightlifting, nautilus, squats and pushups). The book Body by Science describes the latter sort of program, and attempts to find the minimal effective program.

From the point of view of the person considering whether to add exercise to his program, one of the major advantages of the Body by Science program is... (read more)

That assumes a very straightforward calculation of being willing to 'pay' time to fufil various outcomes. Personally, I'd find walking a lot more enjoyable and it would have clearer other benefits: it certainly makes you good at walking without getting tired/sweaty, just like weightlifting makes you good at lifting weights. I'd rather be good at walking. Different people also find different routines easier. Some might find 12 minutes easy to find and fit in, but others would get into the swing of walking to work, whereas a 12 minute slot of intensive weights would not be enough to click as part of their routine.
I am making an educated guess about a key factor of people's costs, and I am addressing that factor. Time is, in fact, very limited. After you subtract work and sleep and other necessities (bathroom activities, kitchen activities, meals) there is very little free time left in the day. I am also writing from the point of view of someone who is not yet sure what the outcomes are. I touched on outcomes and I will touch on your claims about outcomes, but I deliberately am trying to write from the point of view of someone who does not know, because I don't want to set myself up as an authority, I merely want to reason with the reader on the basis of what the reader himself already knows or can work out for himself. That's you. I am addressing myself to the writer of the essay who abandoned exercise (for the duration of their weight loss) and to anyone like them. Since they abandoned the exercise program, evidently they did not find it sufficiently enjoyable to do it. The exercise program was in all likelihood walking. Or, if weights were included, it was in all likelihood one of the more common, more time consuming resistance training programs. The reality is that walking is not enjoyable to many people. Intense resistance training is, admittedly, even less enjoyable, but it can be fantastically brief. That's biased. Under what circumstances are the benefits clearer? If you want to be fair, you either take a position of ignorance with regard to both, or you take a position of knowledge with regard to both, in which case either their benefits are both clear, or they are both unclear. To treat the benefits of walking as known and the benefits of resistance training as unknown is slanted. This implies that resistance training only makes you good at resistance training, which is wildly false. After starting squats, I found my ability to walk was transformed. I really do not want to get into this, in part because it's TMI about myself. So I'll leave it at that. I would
Fair enough: I was taking your post as a general 'this is the best way to approach exercise', but as advice in the particular context it makes a lot of sense.
0Swimmer963 (Miranda Dixon-Luinenburg)
Does intense-but-brief resistance training have the same positive effects on the cardiovascular system as aerobic exercise? This is an important reason why I exercise; because it's generally established that regular exercise has benefits for the heart independent of weight.* Does resistance training have the same effect? I guess probably if you do it intensely enough to raise your heart rate, which I don't; I find my "aerobic" workouts (usually hour-long swims) much more intense than the weight-lifting portion of my workout, which I don't consider as important. *Wise, F. (2010). Coronary heart disease--the benefits of exercise. Australian Family Physician, 39(3), 129-133. Retrieved from EBSCOhost.
The answer from Body by Science is yes, very much so. I can drown you with quotes from the book if you like. Sadly, Amazon seems not to have fully implemented page numbers into the Kindle highlighting feature. The overall argument as I understand it (my understanding may be faulty) is this: 1) Low intensity exercise does not do any part of you including your heart much good because it fails to significantly stimulate the kind of adaptation you want. You need to get your heart pumping hard if you want to trigger an adaptation, and low-intensity aerobics does not do that. High intensity aerobics does, but that is (3). 2) Long-duration furthermore potentially does your body much bad through wear and tear and accidents (don't forget probability of accident per second is multiplied by time to get total probability). As you'll see in the quotes, they claim (and have evidence for the claim, but you need to consult the actual book for full details) that exercising for long periods does not give you additional body-adapting benefits over and above exercising for short periods. So the long duration is (a) mostly wasted (except I suppose for direct calorie burning, but they emphasize that exercise is overrated as a calorie burner) and (b) potentially harmful. 3) High intensity (this can be high intensity aerobics such as the stationary cycle and it can also be high intensity resistance) does your body good (including your cardiovascular system, because your heart is working hard) by stimulating adaptation (for example, muscle building, but not only that). You can get much more information on the benefits of high intensity exercise (for all parts of you including your heart) if you google "high intensity interval training". As a consequence of (1), (2), and (3), they recommend high intensity short duration exercise with a long rest period in between (for your body to recover and build). They make the interesting additional point that steady-state activity (which is necess
This claim seems almost absurd to me. What evidence is used in support of this? Are any studies cited?
No, that would give you the expected number of occurences (assuming independence, which is likely a bad assumption in this case). If the probability of injury in a second is P (given no previous accidents) and you exercise for T seconds, or until an accident occurs, then the probability of an accident occuring during the session is (1 - (1-P)^T), which is the probability of not having the conjunction of T accident free seconds.
I accept the technical correction. But for P close enough to zero, the probability remains nearly linear up to large T. So while you are technically correct I do not think you have refuted my statement as a very close approximation for a typical session lasting an hour or several. Given that, and given the relative difficulty of thinking about an exponential function, I prefer my formula for conveying the essential point in a comprehensible way.
It would be better to make the use of approximation explicit, and to specify the domain in which the approximation is a good one, like "don't forget probability of accident per second is multiplied by time to get approximate total probability, given small total probability".
No no, I am saying I erred outright. I simply used a heuristic and forgot it was a heuristic. You corrected me. That's why I had not identified it as an approximation initially. But that said, I find the correction to make the reader's task difficult, and my heuristic retains approximate validity, so I pointed that out in my reply.
1Swimmer963 (Miranda Dixon-Luinenburg)
Maybe that's why I gasp and feel like I'm dying every time I try to run, despite being in supposedly good shape. On the other hand, conditioning my body to run efficiently would be very useful in terms of transportation (using your own body to walk or run places is free!) Can your muscles tell? I'm often sore the day after doing weights in the gym, but also the day after biking, or running, or playing tag with children (if I haven't done any of these things in a while.) I swam with one of my friends who doesn't swim regularly, and pushed her hard, and she was pretty sore the next day.
I presume they would say that your muscles can tell whether the exercise is low-intensity or high-intensity. However, the quote that you are replying to is specifically comparing high-intensity leg press to high-intensity stationary bike, i.e., two forms of high-intensity exercise which involve comparable amounts of work. I don't recall that they specifically addressed that comparison. My overall sense of it, based on all I've read (including online discussions), is that high-intensity is good either way - either stationary bike or resistance.

Wow. Super relevant to my situation, at a BMI which is close to 40, and has climbed from 25 over the years. I make changes that feel sustainable and then they disappear.

I have questioned whether I would be happier as a skinny person not eating frozen pizzas for dinner than as a fat person eating frozen pizzas and ice cream. I really like eating frozen pizzas and ice cream. I have at best a theoretical liking of being more attractive and having more of an opportunity to have sex with non-fat girls, and to ski a whole day, and to not face the embarrass... (read more)

After listening to Gary Taubes' Why We Get Fat, I'm curious to know three things about your weight-loss diet:

  1. Did you eat more or fewer grams of carbohydrates per day?

  2. Did you eat more or fewer grams of fat per day?

  3. Did you eat more or fewer grams of protein per day?

I cut certain deserts out entirely, which effectively I am sure means I reduced protein intake less than the other two, but I reduced all three dramatically.
Gary Taubes basically says that one generally loses weight if and only if one eats fewer carbs, so this is some evidence for his claim (not strong evidence, since it's consistent with most other models of weight loss too).
I would consider my experience zero evidence for or against Gary Taubes' claim, since every model of weight loss predicts that anyone successfully doing what I did would lose weight. Sounds like a very strong claim, but the word "generally" is murky.
1Swimmer963 (Miranda Dixon-Luinenburg)
Um... Is that logical? If you eat 3000 calories a day of fat and protein, wouldn't you still gain weight?

If you eat 3000 calories a day of fat and protein, wouldn't you still gain weight?

Why would I? The only thing thermodynamics tells us is that calories place an upper bound on how much weight a person can maintain/gain. The actual amount of weight gained/lost depends on the operation of regulatory mechanisms.

I am not sure I am convinced by this argument, for the following reasons: If you think of calorie content / thermodynamics as an upper bound on how much energy can be extracted from the food, you have to make an argument for what happens to the unused energy. Even if you are in a biochemical state where not all the energy is used, there is still energy floating around in your body in the form of carbohydrates, fat and protein. I can think of three possible mechanisms for what happens to this extra energy, and I am not convinced by any of them: (1) Calories are excreted unused in their original form. However, I don't think this happens to a meaningful extent (2) If there is excess fat, nutrients are broken down to molecular constituents in a less efficient mechanism of cellular metabolism, ie, producing less ATP. This is a little more plausible than 1, but I think it would be evolutionary maladaptive to reduce the fuel efficiency of your engine unless it was absolutely necessary. Note that there are cases when the body does reduce the fuel efficiency (such as anaerobic metabolism), but I can't see how this applies here (3) (Added): If there is excess fat, the body begins to run processes that are not strictly necessary, thus using more fuel. However, I am not sure what these processes would be, or why they would be triggered by fat and not carbohydrates. I find it plausible that increasing fat intake will help you lose weight due to regulatory pathways such as insulin, but I think this pathway operates almost exclusively through changes in appetite. I fail to see any arguments why we cannot use thermodynamics (calorie input/output) as a very good approximation of predicted weight change. EDIT: This comment is being downvoted. I am happy to delete it if it doesn't add to the discussion, but it would help me immensely if someone could explain why my reasoning is wrong... EDIT2: I am not sure if I misunderstand the karma system, but I don't think you are supposed
(4) Calories are excreted unused not in their original form. What do you think shit is made of? When dried out, it will burn -- that's calorific value right there. Everyone takes in more calories than they turn into heat and motion. (5) Thermogenesis. If that's impaired, you won't burn as much fuel as it takes to maintain body temperature, but you may not even notice, because you'll do other things to keep warm instead. Come to think of it, accumulating an insulating layer of fat will also dampen the effect. CICO is about as helpful as MIMO -- matter in, matter out. In is easy to measure, out not at all easy.
There is obviously thermodynamic energy in food which is not absorbed in the gastrointestinal tract. Fiber is an example of this. Energy which is not absorbed is not listed on the nutrition label of food. When I say 'calories', I mean the biochemically available energy in absorbed macronutrients such as fat, carbohydrates, protein and alcohol. Nobody doubts that thermogenesis uses energy. This is a special case of my mechanism 3. It is part of the 'energy used'. Again, if you want to convince me that you can eat 3000 calories of fat without gaining weight, you would have to make an argument that the proportion of fat in my diet has a causal effect on thermogenesis, ie, that my body will start running additional thermogenesis because I ate fat instead of carbohydrates. Your claim that CICO is as helpful as MIMO is clearly ridiculous, and if you truly believe this, then supermodels eating tissue paper are more rational than you, as their beliefs will lead to more accurate predictions. The point I am trying to make is that our body is an efficient engine due to evolutionary pressure, that energy doesn't just disappear (if it did, we would observe large amounts of unmetabolized macronutrients in urine), and that even if CICO is not the whole picture, it explains a very large part of the variation in body weight observed in human populations
On a previous occasion when this topic came up, I posted this anecdote. Now, as you might imagine, there were medical reasons for that episode. Or rather, there were concurrent medical events with no obvious connection: acute ulcerative colitis, a disease of the large intestine only. Most nutrition is extracted by the stomach and small intestine, which were unaffected. So what was going on there? What made my digestive system so inefficient for several years following the initial attack? So there's a lot of room for variation in digestive efficiency. (For those who know the last-resort treatment for ulcerative colitis, I'll just add that I recovered without surgery.) What is more ridiculous about MIMO than CICO? Conservation of matter, can't argue with that. You don't get to be a supermodel unless you can stay thin. Some people can, no-one doubts that, and some people just are, without taking any effort. And I'll take Eliezer's word that nothing has worked for him over anyone's assertion that because they can't see how something could happen, it doesn't happen. CICO is only one part of the picture, and its abundantly clear from experiences of dieting that it's of little explanatory value on its own, and of practical value to only a subset of people.
OK, I see the point. But multicellular life evolved as thermodynamic engines, not as fusion plants. Over billions of years, cells were surviving based on how efficiently they could extract thermodynamic energy from macronutrients, to power intracellular processes. This is what we are optimized for. If we had been able to use fusion power in our evolutionary past, MIMO would be a more appropriate level at which to draw your map.
It all depends on what you mean by "very good approximation." There's an entire cottage industry in medicine that revolves around developing weight prediction models; none of them get good results even assuming one knows much more data than simply calorie intake. I suspect this is possibly the source of a few downvotes. Since this is superficially a site on rationality and science, every once in a while the doctrine of Calories In, Calories Out (CICO) rears it's ugly head. People who have actually looked into the situation know that it's a drastic oversimplification, but experience has shown it's usually not worthwhile to convince adherents of CICO of the complexity of the problem. Here is a list of some violations of CICO.
Thank you, that was helpful. Note that I don't disagree with anything in that Mayo Clinic article. The point about "pounds of fat, muscle and water" is obviously true and does not contradict anything I said. The points about "metabolic rate" and "response to reduced calories" just seem to say that sometimes it is difficult to estimate the "calories out" part of the equation, and that it is endogenous to the system. This is also obviously true. I still find it difficult to believe that we can affect the metabolic rate to an extent that matters in the final analysis, based on the fat/protein/carbohydrate content of our diet..
Did I misunderstand your grandparent post? It sounded like you were looking for an explanation as to why CO is hard to quantify. I disagree that this is a fair rephrasing of the article. A correct rephrasing would be "It is always difficult to estimate the CO part of the equation." What would convince you otherwise? When I posted my grandparent response, I wasn't in a position to link to the various body weight modelling studies that have been done, but I could do so if you'd think it might convince you.
OK. I'll accept your rephrasing. Let us assume that "calories out" is always difficult to estimate and depends on a lot of factors such as muscle mass and total calorie intake. I took the original comment to mean that we can eat very large amounts of fat and protein, because our bodies would somehow react, in response to the proportion of different nutrients in our diet, and change how efficiently we use energy. I find it difficult to believe that this would explain much of change in body weight. I find it much easier to believe that it would change our appetites and thus reduce calorie intake. I am certainly willing to update my priors if someone convinces me of a plausible mechanism by which proportion of each nutrient alters efficiency of energy use..
I think this can actually happen to a very great extent depending on how much the person normally eats and burns, and how quickly they consume it, and is the main mechanism by which e.g. competitive eaters generally avoid becoming obese.
By which mechanism do these nutrients get excreted? Urine? Bile? Non-absorption? My impression is that carbohydrates in urine is something that we only see to a significant extent when blood glucose concentration is at diabetic levels. Protein and fat in urine occurs, but it doesn't seem to me that this happens to an extent where it can make a difference to the total energy picture I don't think excreting them through bile would work, the nutrients would just be reabsorbed further down the gastrointestinal tract. It is possible that at very high intake levels, there is significant non-absorption of fat. Maybe this happens in competitive eaters, but I am not convinced it can explain much at fat intake levels seen in ordinary people..
2Manfred (Bam.) So, under normal conditions you're pretty efficient (~4% of your calories just get pooped back out), meaning that something like metabolic rate just swamps poop-energy-content as an interpersonal variable.
This article says that there is some non-absorption of fat in healthy people, and much greater non-absorption in people with cystic fibrosis. If you want to convince me that I should consider this when I choose the fat/carbohydrate/protein content of my diet, you would have to make an argument that the percentage of fat that is not absorbed is a function of my diet, ie, causally related to what I choose to eat. I'm not saying this is not theoretically possible, but my intuition tells me that the variation in absorbtion that is caused by diet, is unlikely to have a major impact in the final analysis
So, Tim Ferris has done a couple of demonstrations where he ate about 20,000 calories in the course of 24 hours. The vast majority of that is not absorbed. You may have had in mind the limited claim that macronutrient ratio has a small effect on the percentage of calories absorbed, which seems reasonable for normal macronutrient ratios, but quantity seems important, as well as more detailed chemical composition. For example, I don't produce enough lactase to digest normal American quantities of milk consumption without chemical assistance, and so if I continued to drink a glass of milk each day, the amount of calories that made it into my bloodstream would be predictably lower than the amount of calories put into my mouth. So while the CI calculation can be complex, it seems obvious to me that the amount of calories you put in your mouth is a good upper bound. (I don't think this is seriously contested by anyone, but it's worthwhile to establish that it's not seriously contested.) The system dynamics may mean that in some cases a higher total number of calories in leads to a lower maintenance weight, and so just lowering intake is not always the right solution.
Comments which mention the importance of calories are reflexively downvoted around here. I think many people are confused between what CICO actually says (your energy balance determines your weight loss or gain) and what their image of CICO -- conveniently made out of straw -- says (there is a magic fixed number of calories, if you eat less than that magic number you'll lose weight).
Typically the conversations are downvoted based on the actual expressed claims in those comments. Most people who make thermodynamics references do in fact say stupid things out of ignorance.
My limited experience -- that is, actual empirical data available to me -- suggests this is not the case when the topic is CICO. Most people who mention dieting or human metabolism do in fact say stupid things out of ignorance.
5Swimmer963 (Miranda Dixon-Luinenburg)
I had noticed this. Personally, I'm very confused about the causes-of-obesity issue. To me it's obvious that if you eat less or burn more, you will lose weight. It's complicated by regulatory mechanisms; eating less causes your body to conserve energy by slowing the metabolism, and physical exercise increases appetite. And I think it's likely there are genetic set points that affect both body type (weight) and appetite. Then there's the fidgeting thing. Then there are the low/high carb theories and studies where weight is modulated by changes in regulatory pathways, and the "fructose poisoning causes fatty liver causes metabolic dysfunction" theory. And stuff like "metabolic syndrome" and Type 2 diabetes. Then there are people who are fat and eat half what I do. In the end, I have no idea how the human body regulates weight and calorie intake, but my body seems to do it fine.
Well, it is complicated. Think of weight regulation as a three-layered cake :-) The bottom layer is physics and CICO holds. The only way to lose weight is to spend more energy than you consume. The middle layer is biochemistry. CICO still holds, but the energy output is a function of a large number of inputs (from genetic makeup to what kind of food do you eat). All the metabolic issues, insulin, leptin, ketosis, etc. live here. The top layer is the mind. CICO still holds and all the biochemical mechanisms from the middle layer still hold, but now we add all the mental stuff -- preferences, compulsions, eating-for-comfort, anorexia, eating as a displacement mechanism, etc. And the cherry on top is that people are different. They have different metabolisms which work in different ways, they react differently to the same stimuli. No solution works for everyone and it looks likely that no solution even works for most. The only way out is to personally experiment and find out what works for you, for your personal, unique, and strange body and mind. And we haven't even touched the question of whether weight is the right metric to use (consider the alternatives, e.g. body fat % or general health even understood in a limited sense as absence of disease and normal metabolic markers). So yeah, complicated it is.

The only way to lose weight is to spend more energy than you consume.


The laws of thermodynamics don't require a fat cell to release lipids because you're hungry or exercising; the fat cells can just physically not react until your muscles run out of glucose or your brain overrules your attempt to starve yourself to death. Similarly, there's no rule that fat cells can't die or shrink and the waste be dumped out through urine.

Thermodynamics is not any more useful than quantum mechanics in understanding obesity. It is moralizing disguised as an invocation of natural law.

6Swimmer963 (Miranda Dixon-Luinenburg)
Obesity rates used to be low. They're now higher. The most obvious changes are higher food availability (and different food availability, read refined sugar/high fructose corn syrup/superstimuli fast food deliciousness), and more sedentary lifestyles. There may be other subtler changes, like the permanent psychological effects of being exposed to food advertising from a young age, and a million things that we don't know yet, but there's something out there, in the physical world, that has changed. And it's not liposuction–that, and gastic bypass surgery, etc, didn't exist a hundred years ago–their invention apparently hasn't reduced obesity rates. In summary, people moralizing about how obesity is just "calories in calories out" aren't doing anything to solve the problem. But saying that thermodynamics is "moralizing disguised as an invocation of natural law" is just pointing out how not to solve the problem–it's not helpful either unless you suggest an alternate solution. Or a list of 20 different things to try, at least 1 of which should work for >99.9% of the population. Or a drug that can target some mysterious fat cell receptor to make them cooperatively release energy during exercise/dieting. Or a special diet that empirically does the same thing, even if no one knows how or why. Or a way to raise children so that they have the same obesity risk as children 200 years ago. Or a way to at least treat the negative cardiovascular health benefits of obesity and make it harmless. Or liposuction. Etc etc etc. I don't think people will get so mad about "moralizing" when this problem has a better solution. ...Oh, and a society that doesn't massively penalize people for being chubby as long as their cardiovascular health is good would be a massive step in the right direction. As a normal weight girl who used to train in the pool every day, but still spent most of high school thinking I was fat and unattractive because of media images of models, I have a particular pet
7Eliezer Yudkowsky
In previous centuries, people rich enough not to have to worry about calories were rarely fat, certainly not at anything like modern rates. Food types have changed. Calorie supply seems like as much a red herring as the number of pirates or global warming. Here's someone fat enough to be a circus freak one century earlier: "Thermodynamics" doesn't explain that change. Some significant number of people a century ago could afford to eat as many calories as they wanted. Also, are we supposing that the circus freak was exceptionally rich?
This statement surprises me. I had always heard that obesity was a sign of social status, but I never thought to wonder about the actual statistics for obesity among the rich. I'm finding statements like "common among the rich," which suggests to me that rates are probably comparable to the modern American rate, but I'm not finding numbers. It appears that medieval monks were obese at three times the rate of the general population [src], but I'm having trouble finding the actual paper or the actual rates. And here's a famously fat nobleman. I read recently about Dionysius of Heraclea, who grew so fat that he could not eat, and eventually so fat that he could not breathe. (There are several other examples there of people who grew too fat to move across the centuries, including a Roman senator who was only able to walk when two slaves carried his belly for him.)
A snippet of the linked article stands out. One Michael Edelman, weighing 1200 lb at his heaviest, in the end made serious attempts to lose weight. My emphasis: Now, it's clear from that article that all of these people ate massive amounts, but then, one has to ask why they did that. Just what are the causal connections here? What are the causal arrows going into the "overeating" node? To say "gluttony" is just giving a name to one's ignorance and mistaking it for knowledge. What actually distinguishes someone to whom a triple chocolate muffin with chocolate sauce and a chocolate-coated chocolate flake on top with extra chocolate is a temptation to be resisted, from someone to whom it is not a temptation? What other arrows are going into the "obese" node? What arrows are coming out of it and where do they point? I don't think anyone knows the answers to these questions.
The first place I would look is hunger and satiety hormones. It wouldn't surprise me if their ability to tell themselves they're full is broken, and so they're hungry all the time, so they eat all the time.
That's also what The Hacker's Diet says.
I've heard this of Africa, but not of Europe. The reason that in preindustrial times millers were stereotypically fat is because it was assumed they pilfered a portion of the corn that farmers brought to them for milling. Friars (think of "Friar Tuck") were stereotypically fat because they lived well (or were thought to) by visibly freeloading on the community. Neither class was well thought of for this. Being fat was taken as a sign not of status, but of idleness, sloth, greed, and gluttony. As indeed it continues to be taken to this day.
So, nobles, merchants, monks, and millers were all more likely to be fat than farmers, and all were seen by farmers as idle, greedy, gluttonous sloths. It's not clear to me why you think that means farmers saw them as having low social status, rather than resenting their high social status.
I think that by “status” he meant what Yvain calls “social power”, whereas you mean what Yvain calls “structural power”.
5Eliezer Yudkowsky
A famously fat nobleman? LIke, that was the fat guy from the seventeenth century?
I'm not sure. My point was more than Galen knew about people so fat Jerry Springer would want to put them on his show today, and the heaviest guy you see at a sci-fi convention is comparable to the heaviest guy you would see in the Roman Senate. I'd be way happier with "this is the percentage of monks that were obese in 1400s Britain" to compare with "this is the percentage of Americans that were obese in 2000s America." From the qualitative descriptions I'm seeing, the rich were obese at broadly similar levels to Americans today, and I'm having trouble finding quantitative descriptions. Are you aware of data I'm not aware of? The trouble is that routine weighing of individuals wasn't common until industrial times, records are spotty anyway, and so we're forced to look at individual accounts in most cases. For example, Rubens painted lots of overweight women, so we know they were around, but statistics of an artist's models says more about the artist than about the general population. Similarly, a 19 year old that weighs 500 pounds is a very rare event, even today. The primary reason we don't call them fat freaks and put them in circuses is because making fun of abnormal people in person has become less acceptable, and if you do it on television you can find people that weigh more like a thousand pounds.
I am not sure the outliers (or the tails of the distribution in general) are relevant here. We know that there are metabolic disorders leading to obesity. It's a pretty good bet that the 500-600 lbs people are metabolically different from the rest of the population and that was as true in the Roman times as it is now. The real question is not whether the 500 lbs people existed in the olden times, sure they did. The real question is why does it seem that 250-300 lbs people were rare in pre-industrial ages and are rather common now (yes, I don't know of good data on the prevalence of obesity before XIX century either...). I don't think it's mostly a calorie availability issue. I don't have a strong opinion on the cause, but if pressed I'd probably point to a confluence of factors including sedentary lifestyles, taste superstimulation and calorically dense foods (mostly refined carbs), stress, etc.
I agree that this is the real question; what surprised me was Eliezer's confident empirical statement on a subject where all the weak data I have points in the opposite direction. It looks to me like the historical data suggests that calorie availability and sedentary lifestyles might be the primary explanations (and of the two, I would expect calorie availability to have a larger impact).
For comparison, this is what it takes to be the fat girl from the first decade of this century. She made the national news when she weighed 63 stone and had to be taken out through the wall of her house to go to hospital. 1 stone = 14 pounds.
And are the kinds of food our ancestors ate no longer available today?
Plenty of foods available today not available to our ancestors, such as semi-dwarf wheat.
Or Coke, for that matter. But if the reason why we are fat and our ancestors were thin is that there are foods we have and they didn't, and we don't want to be fat, we can just not eat those foods. “[We are fatter than our ancestors because] food types have changed” only entails that you can't affect your weight through your diet if you cannot choose to eat what your ancestors did.
Everything we eat has been bred for thousands of years. Does any of it have enough in common with our ancestors' diets that "eat only that" can work? I suppose one might look at what wild primates eat in the present day to answer that. Part of that is "smaller primates", so that still might not be the way to go.
I meant “ancestors” on the timescale of one or two centuries (the time it took for the prevalence of obesity to rise from negligible to sizeable), not megayears. By “food types have changed” EY was referring to (I assume) availability of industrial superstimulus foodstuffs full of high-fructose corn syrup and whatnot.
Ok, I had thought this was going in the direction of the whole paleo thing. Eating as we ate a couple of centuries ago looks much more doable, at the individual level. (Changing the whole society would be a whole different thing.) But perhaps "eat food, mostly plants, not too much" is already one of the things EY has tried?
3Eliezer Yudkowsky
If you can eat "not too much" without your fat cells starving you to death, you're probably already thin. I haven't tried "mostly plants" because it's vastly underspecified and I'm not particularly interested in being told afterward that I ate the wrong plants.
Probably he has; but, unless the fraction of “metabolically disprivileged” people like him has been rising a lot in the past couple centuries, I guess that the rising prevalence of obesity means there are a sizeable number of people who haven't tried that (seriously enough).
0Eliezer Yudkowsky
English can be surprisingly ambiguous at times.
A simpler solution is just taking a piss. But do you really want to go into highly specific and precise definitions which take care of all technicalities? I don't think you have anything to win there. No they don't. But they require you to lose mass. Notably, CICO does not claim you'll lose fat -- it claims you'll lose weight and in most cases some of that weight will come from fat and some from muscle. I disagree. I think it was Taubes who compared CICO to telling an alcoholic that it's drinking alcohol that makes him an alcoholic. He used it as a put-down, but I'm totally fine with the metaphor. Understanding that is only the first step on a long and twisty road, but you have to make it. Otherwise people tend to believe that alcoholism can be fixed by, say, switching to drinking port from drinking whisky, or that as long as they take supplement X they can eat all they want. I think for you (and many other people) CICO became associated with a moralizing stance of "you just need to exercise self-control to stop shoving things into your mouth and then your weight will be fine" -- but it says no such thing (and the stance is stupid, of course).
Thermodynamics does rule out the case where people claim to stay fat and metabolically active enough to be alive despite eating very few calories, which apparently has been a thing.
Mm... I guess what this would be a case of I agree with the connotations of what you're saying, but not with the explicitly stated form, which I'd say goes a bit too far. It's probably more fair to say "energy-in - energy-spent - energy-out-without-being-spent = net delta energy" is part of the story, simply not the whole story. It doesn't illustrate the ways in which, say, one might become unwell/faint without sufficient energy-in of certain forms, even if one already has a reserve of energy that is theoretically available to their metabolism, for example. It's probably a useful thing to keep in mind when trying to diet, for those that can usefully diet that way, but it's not the whole story, and other info (much of it perhaps not yet discovered) is also needed. (And certainly using it as an excuse to moralize/shame is completely invalid.) But I wouldn't call it useless, merely insufficient. What is useless is to pretend that there aren't really important variables that can influence the extent to which one can usefully directly apply the thermodynamics. (People who ignore the ways that other variables can influence the ability to usefully apply the thermodynamic facts and thus condescendingly say "feh, just eat less and exercise more, this is sufficient advice for all people in all circumstances" are, of course, being poopyheads.)
Depending on what you mean by "consume", that statement is false. What about a converse statement? I'm confident this statement would be untrue in practice. Sure, the person could undergo reverse-liposuction, or retain more and more and more water, or have a fusion reactor in her spleen, but the laws of thermodynamics restrict possible scenarios to implausible ones.
They don't, but the reason we evolved fat cells in the first place was that they released lipids allowing our ancestors to survive during periods of scarcity who otherwise wouldn't have. Of course there may be people for whom this mechanism is broken, but I doubt that John Walker, who claims to have lost a sizeable fraction of his body weight without surgery and not gained it back for decades, is lying.
I can't think of any way to answer you correctly and yet also briefly, because Taubes's ideas are not easy, at least not easy for me, to put into a nutshell. Therefore what I am about to say should be taken as no more than a very crude, and greatly exaggerated, approximation of Taubes's theory. Here it is: if you eat more carbs, you turn yourself into a Zucker rat. If you eat fewer carbs, you stop being a Zucker rat. What is a Zucker rat? I'll let Taubes describe the Zucker rat:
1Swimmer963 (Miranda Dixon-Luinenburg)
I suppose the point is less that you can lose weight on 3000 carb-free calories a day, and more that you can't lose weight if you're eating carbs. Just out of interest, what's the proposed mechanism by which carbs turn us into Zucker rats?
Briefly and crudely, carbs affect insulin, insulin affects how greedy and stingy the fat cells are. Greedy fat cells grab energy-carrying molecules, effectively starving the rest of the body. Stingy fat cells are reluctant to let go of the energy they've stored, keeping the rest of the body starved. A starved body is simultaneously hungry and lethargic, for obvious reasons. This in effect reverses the usual causal picture. Greedy stingy fat cells cause a person to feel hungry and lethargic, which causes a person to be inactive and eat a lot. The picture that most people have in their minds is the reverse: a person who eats a lot and who exercises little will, as a consequence of these two vices, get fat. Taubes argues that these so-called vices are a symptom of starvation, which is caused by fat cells hoarding energy, which in turn is caused primarily by high insulin. To break the vicious cycle, cut out the part of the food which spikes insulin, and that is primarily the carbs, and specifically certain kinds of carbs which are rapidly digested. To repeat, while I'm trying to give the best answer I can, it's only an approximation of his argument.
0Swimmer963 (Miranda Dixon-Luinenburg)
Based on what I know about biochemistry and metabolism, that sounds reasonable. Have they done any studies on humans (not mice)?
Taubes wrote a 600-page book on the science, most of it involving humans. I'm out of my depth at this point - you would need to consult the book, either Good Calories Bad Calories, or one he wrote more recently for a wider audience that seriously trims back on the science. But if you want the science you want the earlier, bigger book.
The amazing thing (to me) is that Taubes' two books are not original or personal studies, nor does he claim otherwise. Instead, they are exhaustive reviews of the published dietary research, in which he looks at what was found, and the conclusions that should be drawn. It is (in my opinion) one of the most egregious examples of confirmation bias that the establishment researchers and the government (USA) chose to conclude from these same studies those things that supported their established views, in spite of their own evidence to the contrary. I conclude that adoption of Taubes' findings into our lifestyles would have more positive impact on healthcare (at least, in the USA) than anything else I know. I acknowledge that SIAI President Michael Vassar has said to me: "I met Taubes and he seemed (almost certainly) sincere but not all that bright. Definitely not very erudite and not all that good at philosophy of science." I suggest that readers review Taubes' credientials and determine, as I have, whether he is likely qualified both to understand and to write about the topic. I have been an Atkins devotee for years, and my chemistries reflect Taubes' conclusions (anecdotal). I acknowledge that some very intelligent people (e.g. Yudkowsky) believe that some particular individuals are mysteriously "metabolically challenged", and may respond differently, although I am not aware of the studies to confirm this.
I'm not a nutritionist, but the theory as I understand it is that shifting the balance of calories away from carbohydrates primes your metabolism by changing the pattern of insulin secretion, making your body more likely to break down its stored fats in order to keep blood glucose levels up. High-fat, high-protein foods also tend to feel more filling for a given number of calories, and leafy vegetables are physically bulkier, which might also contribute to their perception as less fattening.

This is really impressive, and a great example of how rationalists really do will. From what I gather, even the great Eliezer himself have not been able to take this idea so seriously! Truly an example for us all.

9Eliezer Yudkowsky
rolls eyes My dear fellow, I assure you that I take TDT very seriously. Some of us cannot lose weight that way. See "Beware of Other-Optimizing", linked at the top of the post. I am beginning to think that metabolisms vary about as much as minds.

Meeeeep D: covers from the higher status individual and tries to look harmless

Then remembers Eliezer is probably on crockers rules

... Sorry, I just had a really low prior for how someone not eating and using lost of energy could not lose weight given thermodynamics.

The fascinating thing about this situation is that Eliezer is about as high status here as it's possible for a human being to be in a non-religious group, and it's still extremely difficult for him to get people to take what he says about his experiences with food and exercise seriously.


It's VERY hard to accept that your own anecdotal experience doesn't apply to everyone. Especially in nutrition/diet/exercise, where basically the only information we have available is anecdotal. (Well, there are medical studies, but they tend to test only not-very-strenuous diets and exercise routines.) It takes a while before you notice that there are real physiological variations. There are people who can't run without joint pain; there are people who can't go three hours without eating or they'll faint; etc. There really are constraints that we don't set ourselves. It's not always an easy thing to accept.

If only that were the situation. There's the generalizing from the one example which is your own experience, and then there's the generalizing from the one example that everyone is telling you is the real truth. There's a complex, highly socially supported mind-blocking ideology which goes way beyond generalizing from one example. One is the common belief that "it's just a matter of "calories in, calories out", which eliminates a huge amount of biological detail. Another is "I didn't say it was easy", which is a way of blurring out the huge range of the amount of difficulty involved. There seems to be an underlying belief that everyone is in some sense really fairly lean, so that any apparent health problems caused by (or causing!) weight loss can be ignored in favor of the Platonic truth of the ideal body. It's not just politics is the mind-killer-- so are status markers.
1Luke Stebbing
For how many people was it extremely easy? I maintain a healthy weight with zero effort, and I have a friend for whom The Hacker's Diet worked perfectly. I thought losing weight was a matter of eating less than you burn. Then I read Eliezer's two posts. Oops, I thought. There's no reason intake reduction has to work without severe and continuing side-effects.

Bleah. Still not used to this high status thing.

But seriously, if you are not metabolically privileged, what happens if you try this is that your body shuts down and goes into starvation mode instead of losing weight. Your fat cells do not release fat under any circumstances, though they're happy to hoover up blood sugar so you always feel tired. We're not talking "feeling hungry", we're talking that you stop feeling hungry and lie down, feeling very very cold and having a hard time moving. Literal starvation, instead of your fat cells releasing fat. I've never tried starving myself that much (I worry that it will cause my brain to cannibalize irreplaceable neurons or something, the way the rest of the body cannibalizes muscle) but I've just recently watched that happen to someone else who tried to lose weight by not eating and wasn't metabolically privileged enough to get away with it.

A calorie is not a calorie. The thermodynamic theory of metabolism is a fucking lie. And it seriously does wear away on your nerves like sandpaper, after a while, to be blamed for it, when the exact same diet can make one person thin and cause the other to blow up like a balloon...

Eh, just read "Beware of Other-Optimizing."

We're not talking "feeling hungry", we're talking that you stop feeling hungry and lie down, feeling very very cold and having a hard time moving.

Have you had your thyroid hormone levels checked? Lethargy, feeling cold, and weight gain/inability to lose weight are ALL symptoms of hypothyroidism. Basically, without enough thyroid hormone telling your cells to be active, your metabolism shuts down. Just a thought.


The thermodynamic theory of metabolism is a fucking lie.

Not so much a lie as 'inapplicable'. Energy balance and mass balance are still true. What happens so that that balance is maintained is highly variable.

I'm convinced that this is a science that doesn't exist yet. (Or, at least, isn't established yet.) Sharing information anecdotally is better than not sharing it, but we haven't worked out the principles yet. I've experienced "starvation mode" and I'm sure as hell not interested in eating that little for any longer than a day or two. I think (with low confidence, though I've seen some physiological just-so stories) that things like exercise and macronutrient breakdown affect whether or not you go into "starvation mode" on a given number of calories. But, regardless, it sucks and people are quite sensible to avoid doing things to their bodies that make them feel like dead dogs. Edit: I am not a doctor, but the other standard thing to get checked is testosterone levels.
To generalize a bit: I believe that people's bodies make choices about what to do when handed some calories-- the calories are allocated to heat, movement, immune system, fat, thinking, etc., and very little is known about how to affect how calories are allocated. Treatment for thyroid problems can help, I think-- if that happens to be your problem and your doctor is more astute than most.
One thing that presumably massively influences calory-allocation may be the the prevalence of brown fat in some adult humans. I found a lot of unusual observations about my (privileged) metabolism suddenly explained by the presumed presence of it. See There are some biases that make me favouring this explanation, and there is contradicting evidence, so I am not so sure anymore.
Agree. Several times I've attempted to study how the body metabolizes energy, how to lose weight, how to gain muscle, and so on. There does seem to be a huge amount of variability between persons, and I personally found it harder to get decently confident answers about such things than about other complex phenomena like procrastination and happiness. I will not be writing about metabolism or weight loss or muscle gain anytime soon. Too complicated and not a field of my expertise.
Just what kind of a calorie deficit were you running when you experienced this?
1Eliezer Yudkowsky
I think she was probably on 1200cal/day or something like that? Maybe less? Naturally, eating more hadn't produced weight loss, so she went lower, which naturally also failed to produce weight loss.
Now, you say "she", and that's important. For women, their weight fluctuates a lot more throughout the day simply due to water intake and excretion. I think it's possible that she was losing weight in the form of body fat but it failed to show up on her scale. What is recommended is that one measures their weight as a weighted moving average. There's an app on the hacker's diet site that does just that. 1200 cal/day sounds extremely low unless this person is very small. However, that doesn't really tell me anything unless you can also give me her height and weight at the time. The calorie deficit is what's important. Also, was she exercising on top of this 1200 calorie diet?
There are a lot of people who assert that they cannot lose weight on 1200 calories per day. Normally such people assert that it is their metabolism. When such people are metabolically tested, it is invariably discovered that their metabolisms are perfectly normal and they eat far more calories than they realize. Of course the claim being made by Eliezer's friend is a bit different. It's that if she has enough of a calorie deficit to lose weight, her fat cells will not give up enough energy to make up the deficit. So that she will feel terrible but not lose any weight. While I concede that there may be people out there like that, it's a pretty extraordinary claim for any individual to make. For one thing, even if your fat storage system is working perfectly normally, it will be uncomfortable to run a caloric deficit especially in the early days of a diet as your body adjusts. Such discomfort is widely reported among all dieter, successful or not. So how can the person possibly know that she isn't experiencing the normal discomfort experienced by all dieters? I would want a medical diagnosis before concluding that something was so seriously wrong with a person's fat storage system. Here's a question: Did the individual successfully lose weight in the past (even if they later regained)? If so, that's a good indication that their fat storage system is working properly.
My suspicion is that she neither experienced ordinary discomfort nor does she have a faulty metabolism. Rather, it's possible that her weight loss strategy was far too extreme. A caloric deficit of more than 25% is considered very dangerous. If she did cut her calories that far, then it's little wonder why she went through hell. Add that to the random variation in her weight caused by water and then it's obvious why she'd given up on trying to lose weight.
Is the caloric deficit inherently dangerous or is it that people usually cut the wrong things from their diet? Do you think there are significant dangers to an otherwise healthy person who gets all the micronutrients they need during the deficit and does it only for a month or two?
Yes, an extreme caloric deficit would be dangerous to anybody. If the body can't make up the difference between the energy expended and energy eaten by burning fat, it will go into starvation mode, slow down, start eating muscle mass and eventually the internal organs.
I'm not sure I understand why the body would eat internal organs on a two month diet when there's plenty of fat and muscle to burn, or why losing some muscle mass would be dangerous.
The heart is made of muscle tissue, and the digestive system is lined with it.
Yeah, smooth muscle and heart muscle, different kinds of tissues from skeletal muscle. I doubt the body has trouble differentiating them.
Look, dude. I'm not a doctor, and I can't really tell you what exactly happens to your body if you have an extreme calorie deficit. Nonetheless, every medical professional will tell you that you shouldn't do it.
Yes now that I think about it that's the most likely explanation. I've been informally researching diet and weight loss for nearly two years now. One thing I've informally observed is that self-deception is a big problem in dieting. Thus when failed dieters report on their failure, they have a tendency to underestimate their caloric intake; they also have a tendency to assert that there is something wrong with their metabolism.
How long did you endure that? I kind of get the same when I start a weight-loss diet, but it usually only lasts about three days for me (provided I don't do anything drastic such as reducing my calorie intake by more than 25%). I don't think that's likely to happen. Neuroplasticity has been widely demonstrated even in adults, hasn't it?

In Breakdown of Will George Ainslie describes a mechanism through which this sort of behavior can be obtained within the framework of hyperbolic discounting. He says

The will is a recursive process that bets the expected value of your future self-control against each of your successive temptations.

In this view, thinking now in terms of TDT (or whatever) gives you enough evidence for future self-control such that abstaining from garlic knots is the correct decision even for a hyperbolic discounter.

"The most important was realizing that what I chose now would be the same choice I would make at other times under the same circumstances."


Wow, this is a really powerful statement that has immediately begun helping me with my procrastination issues. 

This reminds me of the character based self improvement programs of guys like Benjamin Franklin. The first step is deciding who you want to be. Then take them one at a time, and practice them. You become who you want to be by acting like it until it becomes a habit.

Covey is similar, though he stresses being clear on the choice of who you want to be more than developing into who you want to be through habit. Without that clear vision of who you want to be readily available, it's hard to overcome the easy feel goods of the moment.

My bias is toward Franklin'... (read more)


The rule of thumb that helped me lose 120 pounds is this: long term (due to being not fat) happiness is greater than than short term, ephemeral (due to eating lots of pizza) happiness.

To put this in mathematical terms, I consciously try not to use hyperbolic discounting when thinking about the effects of my actions.

Great post. You can think of lots of on-going choices like this; whether to be nice to colleagues, for example, or making an effort to drive efficiently in a similar way. Accruals based accounting also gives you similar results, if you can manage the trick of actually assigning some amount of extra weight, unhappiness etc. to each unnecessary food item consumed.

FWIW I used this method successfully to get in the habit of flossing my teeth every night. It seems like every night after dinner I am tired and I have the same urge to skip flossing and just read a book, watch tv, hang out with my family, etc.

I started telling myself that whatever I decided to do was probably what I would end up doing every night. To my surprise, this worked like a charm.

I think flossing is especially amenable to TDT because you are in basically the exact same scenario every day, day after day after day.

With dieting it's probably tr... (read more)

Can imagine losing weight only with tdt

Wow, fancy, for a bot.

The title sure reminds one of Egon Spectowsky's all-time self-help classic, How I Rose From the Dead in My Spare Time (and So Can You!)

I'd love to hear more success stories like this from you and other people but only if you/they also talk about the background reasons that made the final successful step possible. You did this in your last paragraph, which is extremely helpful!

But sometimes you don't eat the junk food because you've been eating junk food for the past week. Sometimes you stop procrastinating for a while because you've been procrastinating for the past month.

Thus your decision is affected by recent history, which is different every day. Still, if TDT considerations help one take up the strategy of not committing some vice if one hasn't committed it in in one's recent history, that's very useful.

(I've thought of these things before, and TDT hasn't cured me of akrasia)


To me, keeping the weight off after reaching your "maintenance weight" is the real challenge. To keep it off for 5 years or more would be truly impressive. How long have you kept it off?


In your article , you say that

"The first was to cut down from three meals a day to two and eliminate all snacks except water, cutting my consumption by more than a third."

It may have worked, but it was probably not the most healthy way. Losing weight isn't all about how much you eat, but also what you eat. I'd suggest adding a couple of "healthy snacks" every day. These might be a piece of fruit or a nice glass of milk. That's healthy, tastes fine and may further motivate you to remain abstinent of other, less healthy snacks.

After all, ... (read more)

Yes, it has been pointed out by many people. I'm most definitely alive and well, and I agree that vitamin pills seem like a good idea for someone doing something of this type. However, I have heard many arguments both for and against the concept of snacks, both in terms of motivation and also in terms of physical effects. I don't have any idea what the answers are on that for most people, but I know that this was the right answer for me on a motivational level. My best guess is that snacks are bad for carbohydrate addicts such as my past self, and are clearly unnecessary for those who are on very low-carb diets based on my personal observations. And as far as the "healthy snacks" issue goes... I will simply say that I still have plenty of problems to solve and that this involves many of them, but that the circumstances there aren't likely to help anyone else so I didn't involve them. I've discussed them with the NYC group and am working on them.