Schizophrenia: No Smoking Gun

[Note: despite how some people are spinning this, tobacco is still really really bad and you should not smoke it]


Schizophrenics smoke. A lot. Depending on the study, about 60-80% of schizophrenics smoke, compared to only about 20% of the general population. And they spend on average about 27% (!) of their income on cigarettes. Even allowing that schizophrenics don’t make much income, that’s a lot of money. Sure, schizophrenics are often poor and undereducated and have other risk factors for smoking – but even after you control for this, the effect is still pretty strong.

Various people have come up with various explanations. Cognitively-minded people say that schizophrenics smoke as a maladaptive coping strategy for the anxiety caused by their condition. Pharmacologically-minded people say that schizophrenics smoke because smoking accelerates the metabolism of antipsychotic drugs and so makes their side effects go away faster. Pragmatically-minded people say that schizophrenics smoke because they’re stuck in institutions with nothing to do all day. No points for guessing what the Freudians say.

But all these theories have problems. Sure, schizophrenics are often institutionalized, but even the ones at home smoke a lot. Sure, some schizophrenics are often on antipsychotics, but even the ones who aren’t on meds smoke a lot. Sure, schizophrenics are anxious, but we don’t see people with Generalized Anxiety Disorder having 80% smoking rates.

As usual, I’m more biologically-minded, so I find it interesting that some of the genes that most commonly turn up as linked to schizophrenia – especially CHRNA3, CHRNA5, and CHRNA7 – are in nicotine receptors. Indeed, some of them are also the genes identified as risk factors for smoking. Further, there’s a lot of evidence that schizophrenic people actually feel better and have fewer symptoms when they’re smoking. Further, schizophrenics tend to gravitate toward cigarettes with higher nicotine content, and smoke them in ways that maximize nicotine absorption.

It seems like part of the problem with schizophrenia is that the brain’s nicotine system isn’t working well. Smoking supplements nicotine and makes the system run smoother, so schizophrenics feel better when they smoke and continue to do so. This is the widely accepted self-medication hypothesis.

I like this because it’s a really elegant example of…I don’t know what you’d call it…memetic evolution? Nobody knew that nicotine helped schizophrenia, nobody told the schizophrenics that, but they sort of naturally gravitated to an effective treatment for their condition by going in the direction of things that make them feel better, even going so far as to unknowingly gravitate toward cigarette brands with more nicotine. They did all of this before psychiatry had any idea why they were doing it, and in the face of constant protests that it was stupid and useless. This should be a warning to anyone who’s too quick to tell patients that their coping strategies are maladaptive.

But there’s a much more important question here: does smoking cause schizophrenia? How about prevent it?


First, the causation argument. Gurillo et al do a meta-analysis and conclude that “daily tobacco use is associated with increased risk of psychosis and an earlier age of onset of psychotic illness. The possibility of a causal link between tobacco use and psychosis merits further examination”. That is, schizophrenics are already smoking much more at the moment their schizophrenia starts. This suggests that maybe smoking is helping to cause the schizophrenia?

All nice and well, except for a few things. First, this study ignores the possibility that the genes that cause schizophrenia might also cause increased smoking, even though we have some evidence that this is true (actually, it doesn’t ignore this, it mentions it, but uses it as a reason why a schizophrenia-smoking link is more plausible). Second, we know that people who will later develop schizophrenia are seen as kind of odd even before they come down with the disease, and it’s possible that they’re already in some unusual brain state that smoking helps relieve. Third, this study is not controlled – meaning that we’re totally helpless before factors like “people destined to later develop schizophrenia are often poor, and poor people smoke more”.

And fourth, another study shows exactly the opposite.

Zammit et al (thanks to @allfeelsallthetime for the tip) looks at 50,000 teenage Swedish conscripts, then follows them throughout their lives to see which ones do or don’t get schizophrenia. They find that without adjusting for confounders, smokers are more likely to get schizophrenia. But when you do adjust for confounders, smokers are less likely to get schizophrenia, (hazard ratio 0.8, p = 0.003) and heavy smokers are much less likely to get schizophrenia (hazard ratio 0.5)! A dose-dependent relationship was found between smoking and protection from schizophrenia. This is really interesting.

Why do we find such different results from these two studies? The only explanation I can think of is that the second study controls for various factors including cannabis use, personality variables, IQ, past psychiatric diagnoses, and place of upbringing (thanks @su3su2u1 for the tip) and the first study controls for zilch. In fact, we find that the second study’s uncontrolled numbers are not that different from the first study’s uncontrolled numbers, and that the only difference is that the second study then went on to control for confounders and get the opposite result. Controlling for more things is not always better, but controlling for a few things that previous studies and common sense suggest are very relevant is pretty superior to just leaving the data entirely unprocessed. Advantage very much second study.


Unlike certain people on Facebook, I fucking hate science. Let me explain why.

The first study here, Gurillo et al, was published ten years after the second study. Since it is a meta-analysis, it included the second study in it. The authors of the first study definitely read the second study. They just didn’t care.

Nowhere in the first study does it say “By the way, we read this other study that got the opposite results from us, let’s try to figure out why, oh, it was because they controlled for things and we didn’t, maybe that should call our findings into question.” You know what they did do? They listed the second study as finding that smoking increased schizophrenia risk, because the rules of their meta-analysis said they would only take uncontrolled data, and so they did. You can read this entire study, which cites the second study no fewer than six times, without hearing at all about the fact that the second study got the opposite result using likely better methodology.

Then they go on to conclude that:

Cigarette smoking might be a hitherto neglected modifiable risk factor for psychosis, but confounding and reverse causality are possible. Notwithstanding, in view of the clear benefits of smoking cessation programs in this population, every effort should be made to implement change in smoking habits in this group of patients.

Clear benefits! Every effort! Aaaaaaah!

I mean, I know where they (and the Lancet editors, who write a glowing comment backing them up) are coming from. Smoking is bad because lung cancer, COPD, etc. But now we have these things called e-cigarettes! They deliver nicotine without tobacco! As far as anyone knows they carry vastly less risk of cancer, COPD, etc. If nicotine actually prevents schizophrenia rather than causing it, that is the sort of thing we should really want to know. And instead we’re just getting this “We should make schizophrenia patients stop smoking, because smoking is bad”.

Look. I am not going to come out and say that there’s great evidence that nicotine decreases schizophrenia risk. There’s one study, which other studies contradict. I happen to think that the one study looks better than its competitors, but that’s my opinion and I have nowhere near the evidence I would need to feel really strongly about this. But I feel like we are very far from the point where we know enough to be pushing people at risk of schizophrenia away from nicotine, and light-years away from the point where we can use phrases like “clear benefits”.

Possibly I am an idiot and missing something very important. But if this is true, I wish the authors of the new study, and the editors of The Lancet, would have acknowledged the existence of the conflicting study and patiently explained to their readership, many of whom are idiots like myself, “Here’s a study that looks better than ours that seems to contradict our results, but here’s why our study is nevertheless far more believable.” That’s all I ask.

No matter how much of an idiot I am, I can’t possibly imagine how that wouldn’t be a straight-out gain.

PS: Cigarette smoking definitely decreases your risk of Parkinson’s Disease. Parkinson’s is similar to schizophrenia in that both involve dopamine. But schizophrenia involves too much dopamine and Parkinson’s too little, so the analogy could go either direction.

PPS: Tobacco smoking is definitely still bad! Nothing in here at all suggests that tobacco smoking has the slightest chance of not being a terrible decision!

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

264 Responses to Schizophrenia: No Smoking Gun

  1. Typo: “The authors of the first study definitely read the first study.”

    • DanielLC says:

      It’s not wrong.

      • Phil says:

        fwiw, when reading the post originally, I had to re-read that sentence several times to parse it correctly

        My brain at least, wants to associate ‘first’ with ‘written first’ even if it means ‘the study I referred to first’

        if I was an editor, I might suggest rephrasing that sentence, even if its not technically wrong (maybe just refer to the actual study, instead of first and second, just my 2 cents)

      • RCF says:

        By any reasonable standard it is.

  2. Lambert says:

    Can Acadamia make a rule that you are the only person allowed in the peer-review process?

    • paper machine says:

      The peer-review process is too important to have a single point of failure 😉

      • Jack V says:

        But it does seem like the process could be better, it doesn’t do well at weeding out studies like this one which are apparently scientific but may have big underlying problems.

    • TeMPOraL says:

      How about: every peer review has to go through Scott for verification, and if Scott is not available, then it should be posted to Reddit and HackerNews.

      • Phil says:

        It seems like if a study gets taxpayer funding to exist, at the very least, it should be reasonable to demand that it not be published behind a paywall,

        and it might be even better if it had to be published in a public wiki, where the public had access to the data, and anyone who wanted to look into and point out flaws could

        seems like that might be an interesting thing to try to get a congressman or two behind

        • RCF says:

          And also, it has to be registered before any data is collected, with all statistical tests declared ahead of time with the alpha they will use, and with well-defined criteria for when to release the data. Once those criteria are met, the data must be released, regardless of the results, and if the criteria are not met, the data cannot be officially released. No “the results were so promising we terminated the study early and are publishing just the data from that first part”, unless the protocol specifies early termination procedures that were followed.

          • Devilbunny says:

            Mark, my med school’s stats department had a standing offer for any researcher that they would offer pre-, intra-, and post-study counseling in return for a middle-author credit. Few took them up on it.

  3. John Schilling says:

    Take the “memetic evolution” argument back a step. Tobacco would not be as broadly and persistently popular as it has been, if users didn’t consistently find some real benefit to what is otherwise an expensive and inconvenient habit. Yes, it is addictive – but if, say, eating coal turned out to be addictive you still wouldn’t expect it to be massively popular, because how would anyone get hooked in the first place? “Hey, everybody, let’s go around inhaling the pungent smoke from dead plant leaves! Best part is, once you start you can’t stop!”, as a globally winning sales pitch?

    I’m with you on hating science in this particular case. We really need a thorough examination of the costs and benefits of nicotine, with consideration of alternate delivery systems. And we’re not going to get it because of the cognitive bias that comes from Tobacco Is Pure Evil, No Exceptions, itself derived from the good but narrow science of past generations. Zammit et al may not find themselves as unpopular as a climate skeptic at an IPCC meeting, but from your description their work seems halfway down the memory hole already.

    And the practical question: If on further careful examination it turns out that the balance of evidence is on tobacco preventing or alleviating schizophrenia, is it possible to recommend it to your schizophrenic or at-risk patients without losing your job?

    • Scott Alexander says:

      I as a resident can’t sneeze without losing my job if someone wants to take my job away from me. Once I’m a fully qualified doctor, I can if I so choose open my own practice and do whatever I want (within sensible legal limits of eg not overusing controlled drugs or overrecommending disability benefits). That would include recommending nicotine for at-risk schizophrenics – but I’m not sure whether I would do so at this point. “Switches from bad to good after controls in one study” is pretty weak evidence, especially in terms of having to defend myself from lots of people who disagree.

      • Alex says:

        I would think that the better way to look at the issue, at least with the current state of knowledge, is that if you encountered a teen who was smoking you should ask whether the smoking is their primary health problem, or if the smoking is a symptom of schizophrenia.

        Where you go from there I have no idea, but I think it’s less about “Maybe smoking would help?” and more “Hmm, why is this kid smoking? Is it a sign of something else we need to worry about?”

        • Scott Alexander says:

          Just to make it clear: tobacco smoking is still unambigously really bad.

          • John Schilling says:

            As is schizophrenia; fortunately we have alternatives to smoking tobacco as a nicotine delivery system.

            If it turns out that there is a strong positive effect w/re schizophrenia but that it takes tobacco smoking, not just nicotine, well, that would be evidence that God exists and he hates us.

          • anon says:

            If it turns out that there is a strong positive effect w/re schizophrenia but that it takes tobacco smoking, not just nicotine, well, that would be evidence that God exists and he hates us.

            My mother is schizophrenic, and if it turned out that tobacco smoking massively reduced my chance of developing schizophrenia myself I would do it. Schizophrenia is awful. I would definitely prefer dying 15 years early of lung cancer to becoming schizophrenic.

          • Murphy says:


            The only problem is that even if it does work somewhat you could still very well end up both schizophrenic and with lung cancer.

          • anon says:

            I am not a suicidal person, but if I were 100% sure that I’d develop schizophrenia I would seriously consider suicide. It is awful.

            As it is, my risks are quite a lot lower than that I think.

          • HeelBearCub says:

            “Just to make it clear: tobacco smoking is still unambigously really bad.”

            This, combined with the fact that you can also get nicotine gum or other delivery devices over the counter (I’m pretty sure?), seems like the biggest failure point of the post.

            The whole post treats “products produced by RJ Reynolds” as if they are the only way to get nicotine. Why?

          • Marc Whipple says:


            Not only can you get nicotine gum OTC, it is now being marketed as a recreational product. There are signs at my local gas station advertising that they have it inside and that it produces a pleasant feeling. They don’t say anything about smoking or smoking cessation. The product is not labeled as having anything to do with smoking cessation. (I didn’t try any, but I did look at the package.)

            I knew this would happen the first time my MIL tried to quit smoking with nicotine gum, forgot to stop chewing it after the craving passed, and became mildly intoxicated. (She doesn’t use it recreationally, but I knew people would do so and that it would eventually be a marketing point.)

          • Nornagest says:

            This, combined with the fact that you can also get nicotine gum or other delivery devices over the counter (I’m pretty sure?)

            You can, but it’s expensive. It’s marketed as a “smoking cessation tool”, which is to say a quasi-medical device, so the markup is huge even though it’s gotta be dirt cheap to make.

            Dip is cheaply available over-the counter, but while I’m sure it’s far less lethal than cigarettes it carries enormous social stigma in Blue spaces — worse than smoking by a long shot. Probably because of that stigma, there also doesn’t seem to be much widespread appreciation of the gap in lethality.

            I expect e-cigarettes will expand to fill the niche of a cheap, less lethal, non-redneck nicotine delivery system, but I don’t think we can say they’ve gotten there. Manufacturers certainly haven’t settled on a single business model for them yet.

          • Vox Imperatoris says:

            @ Nornagest:

            Dipping tobacco of the kind normally used in the U.S. is pretty dangerous, as far as I’m aware, because it is smoke-cured. It is a major cause of oral cancer.

            Snus, which is popular in Scandinavia, is steamed and therefore (as far as studies can show) pretty much harmless—as a result, there is a wide disparity in Sweden between male and female tobacco-related mortality. This is, of course, because women don’t dip tobacco.

            This finding is fairly well-known, I think, and hasn’t been “suppressed” by the academic establishment.

          • Nornagest says:

            Dipping tobacco of the kind normally used in the U.S. is pretty dangerous, as far as I’m aware, because it is smoke-cured. It is a major cause of oral cancer.

            It’s a major cause of oral cancer, but the rates we’re talking about are a lot lower than the rates of lung cancer you see among smokers: this meta-analysis shows relative risks usually around 2 for American-style dipping tobacco, though with wide error bars. (The relative risk of getting lung cancer among smokers is about 20, though the exact numbers vary with how you pool the data.)

            I don’t think I can conclude from this that Scandinavian-style snuff use is harmless, but it does look less dangerous.

          • Vox Imperatoris says:

            @ Nornagest:

            I don’t think I can conclude from this that Scandinavian-style snuff use is harmless, but it does look less dangerous.

            I am no expert on this (and have never used nicotine of any kind), but that meta-analysis does little to discuss the relative risk of products like snus in particular.

            You’ll note that the relative risk of oral cancer in some of the Scandinavian studies was less than one (i.e. subjects who dipped tobacco were less likely to get oral cancer than those who didn’t!). And the meta-analysis makes no clear distinction between smoke-cured and steam-cured tobacco products.

            But absolutely, no one should take my word for it that it’s harmless.

          • houseboatonstyx says:

            @ Vox Imperatoris
            Snus, which is popular in Scandinavia, is steamed and therefore (as far as studies can show) pretty much harmless—as a result, there is a wide disparity in Sweden between male and female tobacco-related mortality. This is, of course, because women don’t dip tobacco.

            Any dire warnings against making one’s own ‘nicotine water’ starting with soaking a pinch of snus/chewing tobacco/whatever in, say, half a gallon of warm water?

            I looked into vaping but getting away from the flavorings etc seems to involve handling dangerous concentrations of nicotine (as well as expense).

          • Vox Imperatoris says:

            @ houseboatsonstyx:

            No clue how dangerous it is to make “nicotine water” like that.

            If what you want is flavorless nicotine vapor, are you sure you can’t buy it in normal concentrations? I kind of doubt it.

            Diluting your own from industrial concentrations is sort of dangerous, I agree. You could screw up and kill yourself.

          • Nornagest says:

            I’ve heard of nicotine tea being used as a suicide method in prison, but for all I know this could be an urban legend.

          • houseboatonstyx says:

            @ Vox Imperatoris

            I’d rather avoid vapor at any stage, actually, and just make something approaching ‘nicotine water’ (without the expense, legal/import issues, and whatever else the manufacturer might put in it).

            So, anyone — ever tried soaking a pinch of snus/chewing tobacco/whatever in room temperature or cold water?

          • Richard says:


            The quickest way to get some nicotine water is to buy organic pesticide and dilute it. It is made by brewing “tea” on the nerves in the tobacco leaf which contain too much nicotine to be useful for smoking tobacco. This was a DIY process before organic farming became mainstream.

            (Of course, understanding exactly how boiling a plant to get hold of a complex chemical makes it in any way less a chemical is beyond my mental capacity, but so is a lot that goes on in the organic farming movement.)

            If you plan to drink it or similar, and have reservations with respects to drinking pesticide, you can go with brewing tea on a cigar. Good cigars have the advantage over cigarettes of containing only the tobacco leaf with no added anything.
            I sacrificed a cuban to try this out, and found that drinking is a much more efficient nicotine delivery system than smoking, so you should get several doses from a single cigar.

          • nydwracu says:

            IME, tobacco products and e-cigs have very different effects — e-cigs drop my perceived mental ability by about as much as a beer, without any of the benefits of a beer, whereas actual tobacco products have the opposite effect. (When I was in college, I was in the habit of buying cigars when finals season came around, and dipping snus before lectures that I thought would be interesting.)

            I’ll sacrifice a snus (a snu?) to a mug of hot water sometime and report back.

          • houseboatonstyx says:

            @ Richard

            Thanks. I googled, put ORGANIC CIGAR on my shopping list. (Am now earwormed with “We Three Kings of Orient Are”.)

            @ nydwracu

            Will be interested in your results. A few scraps of pipe tobacco in 1/4 cup of hot water was pleasant to smell, possibly nostalgia.

    • Deiseach says:

      It’s a hell of a mess, isn’t it? About the only thing I gathered from those competing studies is that nicotine and schizophrenia seem to be linked because the genes that make you susceptible to needing/wanting/being deficient in nicotine or whatever nicotine does for you (?) seem to be connected with being at risk of developing schizophrenia.

      I think there’s genuinely well-founded doubt about recommending smoking to schizophrenics (“Hey, you should smoke and not only should you smoke but you should smoke a lot of the highest nicotine content you can get!”) because exactly lung cancer etc. (Lung cancer is a fucking miserable way to die, I can tell you that from watching a family member die from it – and yes, due to smoking).

      It’s also a bit dodgy for medical professionals to recommend a poison (which is what nicotine is; Golden Age detective stories were very fond of having amateur gardeners make concentrated doses of nicotine to bump off people they wished to bump off) to mentally vulnerable people. I imagine surviving family members would sue the arses off any doctor or hospital that encouraged or even made known to a schizophrenic that smoking would help them, in the event of that person then developing a smoking-related disease.

      It’s like sickle cell anaemia and malaria, isn’t it? Sickle-cell is bloody awful, but is it better than malaria? Possibly.

      • nope says:

        Here is a short list of poisons, as per the definition: “substances that cause disturbances in organisms, usually by chemical reaction or other activity on the molecular scale, when an organism absorbs a sufficient quantity”, and just to be extra nice we’ll limit our criteria to only those which can actually kill you:
        – caffeine
        – aspirin
        – bananas
        – water

        Let’s use words more judiciously from now on, shall we?

        • Kaminiwa says:

          Is it the slightest bit plausible to *sneak* a fatal dose of any of those in to someone’s diet? I’m pretty sure bananas are safe unless you’ve tied them down and are force feeding them, and I know that’s the case with water. These are not assassination tools like Deisach was talking about. Yes, the word “poison” was used, but context matters…

          Now, like, tylenol, that seems like a much better metaphor here.

          • nope says:

            My point is that while Deiseach is technically correct in his/her use of the word “poison”, he/she would be correct almost no matter which subject he/she was talking about, because nearly everything that does anything can be a poison. Using “it’s a poison!” as a point against it is a bad idea, because it’s an emotional word people have Feelings about, and worse it conflates things that I’ve literally never heard of anyone dying from an overdose of, like nicotine or bananas, with things like strychnine or arsenic which people have actually succeeded in killing other people with. Golden Age detective stories do not actually count as evidence of anything in the real world. The examples were not intended as direct analogies, they were intended to represent the broadness of the category.

          • nope says:

            Also relevant,

            “Pure powdered caffeine, which is available as a dietary supplement, can be lethal in tablespoon-sized amounts.”


          • Anony says:

            >Is it the slightest bit plausible to *sneak* a fatal dose of any of those in to someone’s diet?

            Someone with a heart condition and a penchant for dessert beverages might not notice a large amount of caffeine in a Starbucks drink.

          • Vox Imperatoris says:

            Nicotine has (rarely) been used in real life to kill people.

            Mainly because, for a brief time in the 19th century, it was an undetectable poison. (They found a way to detect inorganic poisons like arsenic before they found a way to detect organic poisons.)

            Also, the fact that there is no bright line between a poison and a non-poison does not mean that the word “poison” has no useful meaning, or that you can’t tell that it would not include water.

          • Vaniver says:

            The story of how nicotine poisoning became detectable is fun, but I can’t seem to find a link. As I recall, there was a prominent case in which it was pretty obvious that someone had murdered their family with concentrated nicotine for the money, but the government couldn’t prove it, and so an organic chemist decided that This Would Not Stand and spent a month or so of frenzied effort figuring out how to detect it (and yep, the suspected perp had definitely murdered them with nicotine).

          • Vox Imperatoris says:

            @ Vaniver:

            I read the same story. Didn’t Scott link it or something?

          • Deiseach says:

            nope, you are very nearly issuing me a dare 🙂

            Tobacco is related to the nightshade family of plants, which contains very toxic plants and also useful food crops like potatoes.

            Potatoes can contain solanine – you are warned not to eat green potatoes and to cut out the green portions or discard any such potatoes (I admit, I’ve ignored this advice and suffered no ill effects) because it can result in mild poisoning:

            Solanine poisoning is primarily displayed by gastrointestinal and neurological disorders. Symptoms include nausea, diarrhea, vomiting, stomach cramps, burning of the throat, cardiac dysrhythmia, nightmare, headache and dizziness. In more severe cases, hallucinations, loss of sensation, paralysis, fever, jaundice, dilated pupils, hypothermia and death have been reported.

            Again, I admit I’ve never heard of anyone dying from eating green potatoes, but the point remains that even common, in everyday use, plants of that family can do you harm if misused. Nicotine was extracted and used as a pesticide in gardening, from whence the murder mystery plots originated. Indeed, it was used in at least one successful murder – successful, that is, in that it did cause death but unsuccessful in that the murder was eventually proven.

            So while “poison” may be emotive, I maintain that it is still factually correct. Indeed, to cite Professor Bostrum, it is an empirical fact! 🙂

            Toxicologists estimate that a fully smoked cigarette delivers about 1 mg of nicotine to the lungs; this compares to a lethal dose estimate of 30-60 mg.

            There’s a reason people referred to heavy smokers as “killing themselves by nicotine poisoning” even before the link with lung cancer and other diseases was known!

          • nope says:

            @Deiseach, of course you were factually correct. I wasn’t disputing that. The substances I listed *can* all actually kill you, given the right doses, and three on that list are actually somewhat easy to kill yourself with if you want to. And practically, it’s easier to kill yourself with aspirin or caffeine than it is to smoke 30-60 cigarettes in a short enough period of time to actually kill you. However, I doubt you would be so up in arms about recommending aspirin or coffee to schizophrenics because of how “poisonous” they are. Nicotine has gotten a bum rep because Smoking is Evil, when the most Evil parts of smoking actually don’t have anything to do with nicotine itself, because nicotine is more or less equivalent to caffeine in most real senses.

          • nydwracu says:

            Anglos didn’t eat tomatoes for a while because they thought they were poisonous, for the fairly sensible reason that all the parts we don’t eat are.

            There’s a similar story with yaupon, a plant related to guayusa and yerba mate that grows in the South — the natives used it as part of a purification ritual (i.e. an ingredient in an emetic), so until recently everyone thought it caused vomiting. It doesn’t help that its scientific name is Ilex vomitoria.

            (I haven’t managed to get my hands on any yet, but I’m hoping it becomes popular someday, for purely nationalistic reasons. I hope it’s as worthwhile as guayusa — which has the two advantages of not containing any tannins (the chemicals that make black tea in restaurants universally awful, because they always brew it too long) and being almost completely tasteless, meaning you can flavor it however you want.)

          • “Anglos didn’t eat tomatoes for a while because they thought they were poisonous”

            Do you have a source, preferably a primary source, for that? I have seen claims of that sort made about various foods, and they sometimes turn out to be false.

            Tomatoes are described in 1544 as being eaten in Italy. The earliest English reference I know of is in 1604 referring to them in the West Indies. In the mid-18th century there are references to tomatoes being eaten in England.

          • Vox Imperatoris says:

            @ David Friedman:

            I don’t know if this counts exactly, but I just happened to run across it today. The widely-read Victorian cookbook, Mrs. Beeton’s Book of Household Management contains a passage stating: “The whole [tomato] plant has a disagreeable odour, and its juice, subjected to the action of the fire, emits a vapour so powerful as to cause vertigo and vomiting.”

            The passage has been noted to run contrary to the statement on the same page: “The Tomato is a wholesome fruit, and digests easily”. Maybe she is talking about the juice of the leaves in the first part, or something? But that seems implausible, and the Wikipedia article mentions that this was criticized in contemporary times as being an example of careless editing and/or inattentive plagiarism.

            She also expresses a negative opinion on mangoes (taste like “turpentine”), lobster (“indigestible”), garlic (“offensive”), and potatoes (“suspicious; a great many are narcotic, and many are deleterious”). And she recommends boiling pasta for an hour and forty-five minutes (!), which some people considered evidence that she had never actually cooked many of the dishes in the book.

        • Mary says:

          Short version:

          It is the dose that makes the poison.

          Oldest principle of toxicology.

      • Nicotine doesn’t have to come from burning tobacco. The serious medical negatives seem to all be associated with burning tobacco, not with nicotine. So your lung cancer point is irrelevant to someone who recommends vaping, or nicotine gum, or nicotine patches.

      • Ninmesara says:

        Sickle-cell anemia is NOT better than malaria!
        Sickle cell doesn’t even protect from malaria. In fact, it increases mortality in malaria infections.

        Sickle cell trait, on the other hand (the condition where a patient has the heterozygous form of the disease, with only a single mutated gene) does in fact protect against malaria.

        As with other recessive diseases, sickle cell trait (SCT) is much more common than sickle cell anemia (SCA). This means that areas where malaria is prevalent, there is a large number of SCT patients that are protected while the fewer patients with SCA are doubly screwed: they have both the inherited hemolytic anemia of SCA and the acquired hemolytic anemia of malaria. The morbid beauty of the situation is that the prevalence of the mutated gene will increase to the level where the combined reproductive fitness of those suffering from SCT will be balanced out by the loss of reproductive fitness from those with SCA (both from the disease itself and from the increased mortality of malaria).

        This paper is a great review on the subject. The writing is as simple as the subject allows, and some parts are probably informative even to a lay audience with no knowledge of medicine but with an interest in population genetics.

        For those who don’t want to read the full paper, some highlights:
        > On one hand, heterozygotes for the sickle gene (AS) are relatively protected against the danger of dying of malaria, as now firmly established through a number of clinical field studies from different parts of Africa. […] On the other hand, patients who are homozygous for the sickle gene and therefore suffer from sickle cell anaemia (SCA) are highly susceptible to the lethal effects of malaria

        > A recent population study carried out in Kenya has shown that malaria is no more common in SCA children than in controls: however, the mortality of SCA children who had malaria was about 10 times higher than in controls.

        And last but not least, the last sentence from the paper is possibly the most poignant sentence I’ve ever read from a scientific article:
        > If, as doctors, we have a professional obligation towards all of our patients, for those with SCA we have an added human obligation, if we consider that they carry the genetic burden that has helped human populations to survive in malaria-endemic regions of the world.

        Beside the epidemiological data, the paper explains the mechanisms for malaria protection in SCT patients, which are really interesting but are not readily accessible to a lay audience.
        If anyone is interested, I might try to “translate” the content of this paper to a lay audience on my blog. The theme is interesting but I imagine some people might be put off by the medical jargon.

    • dndnrsn says:

      I’m a non-smoker, but every now and then I will have a cigarette. It often coincides with drinking. With no nicotine tolerance, a single cigarette gives a pleasant headrush, and honestly a very mild high.

      Couldn’t it just be that, as with a lot of other addictive substances, it’s pleasant until you’ve built up a tolerance, and then you can’t stop? People do a lot of disadvantageous stuff because it feels good, at least at the beginning.

    • vV_Vv says:

      Yes, it is addictive – but if, say, eating coal turned out to be addictive you still wouldn’t expect it to be massively popular, because how would anyone get hooked in the first place? “Hey, everybody, let’s go around inhaling the pungent smoke from dead plant leaves! Best part is, once you start you can’t stop!”, as a globally winning sales pitch?

      Because it was high status. These specific dead plant leaves weren’t easy to obtain and process, making smoking a luxury. Consider the ornate ceremonial pipes of the Native Americans, or the elaborate waterpipes created in India and Persia.

      Even in recent times, in the Western culture, smoking cigars or cigarettes was considered high status until perhaps 30-40 years ago.

      Now cigarettes and even most types of cigars are cheap mass-produced commodities, smoking pipes are old fashioned objects and smoking is generally seen as low status.

      • I remember back to the sixties. Cigarette smoking was status-neutral. Cigar smoking *might* have been high status if you made a big deal of it with good cigars, but I don’t think it defaulted to high status.

        • LHN says:

          I just miss the sixties, but my impression (+memory of the 70s) is that cigars were decidedly unfashionable. The prototypical cigar-smoker was middle aged or above, middle class, and definitely not “with it”. They might still have status, but the cigar didn’t really add to it, and if anything suggested that they were dinosaurs. (Pop culture example: J. Jonah Jameson in Spider-Man.)

          It wasn’t till maybe the late 80s that I remember cigars starting to become aspirational again. About the same time that the martini did, or maybe a little before.

      • RCF says:

        Is snorting cocaine high status? Something being expensive is clearly not sufficient to explain it becoming popular. I’m not convinced that it’s even a significant contributory factor.

        • Vox Imperatoris says:

          It definitely was (and to a good extent, still is) in the right circles.

          And I don’t just mean the Sherlock Holmes days. Even in the 70s and 80s. And even today, it’s the drug of movie stars—it may be a bit seedy, but it’s not like meth, which is the drug of rural poor people.

      • Vox Imperatoris says:

        Because it was high status. These specific dead plant leaves weren’t easy to obtain and process, making smoking a luxury. Consider the ornate ceremonial pipes of the Native Americans, or the elaborate waterpipes created in India and Persia.

        But that just puts the question back one step to: why did it become high-status?

        Sure, people do weird things to mark themselves as high-status, but those things typically have a basis in their use independently of status effects. For example, gold is naturally thought beautiful and is very easy to work into elaborate shapes. Diamonds are beautiful and extremely durable, making them more useful than, say, quartz—which is just as beautiful but very easy to scratch.

        Fine wine really does have complexities of flavor which can be admired by connoisseurs (though I personally can’t stand dry wines and don’t think much of beer, either). Golf is an entertaining pastime which does require a good deal of skill and combines athleticism with the aesthetic pleasures of walking through what is essentially a garden. Beluga caviar has a unique flavor which some people actually do admire.

        Now, once they do get popular, you get people whose natural tastes don’t incline that way but feel obligated to join in in order to fit in. For instance, Emperor Nicholas II of Russia didn’t like brandy very much, so he invented the still-common “Nikolaschka” cocktail to cover up the taste with a lot of lemon and sugar.

        Of course, there are some outliers like “fine” water (which doesn’t taste any different from regular water, unless we’re talking about sparking mineral water, which certainly does) and “fine” vodka (which is a tasteless liquid that is indistinguishable from well-filtered industrial ethanol—and even there, there are allegedly certain distinctions in the “mouthfeel” of vodkas made from certain plants vs. others). But those fashions generally don’t last as long.

        The point is, tobacco smoking probably wouldn’t have become high status if it had no intrinsic value or pleasure-producing capacity. And it clearly does: it both stimulates and relaxes, people who are accustomed to the taste find it pleasurable and distinguish between different varieties, and the rituals associated with it (such as the gentlemen donning smoking jackets and retiring to the smoking room, where they compare their choices of pipes and sources of tobacco) promote social bonding. It also has health costs that outweigh these things, but that doesn’t mean it has no benefits.

        • Deiseach says:

          Tobacco smoking seems to have been touted (when it was introduced into Europe) as being medicinal in nature; tobacco is a herb and herbs are drugs and medicines. People getting addicted (if we can use so strong a term) to the nicotine content then continued smoking, and would probably have defended their use against criticism (and there was a lot of criticism of smoking as a disgusting, barbaric habit) on the grounds that “it’s good for your health”. After all, it took a long time and a lot of controversy before the link between lung cancer and smoking was accepted as proven.

          It became fashionable and think of what people do in the name of fashion. Combine smoking with the coffee-houses, and you have high-status behaviour which is on the cutting-edge of being cool, hip, connected to the cultural and political elite – of course people copied it.

          I’ve never in my life trying smoking, I’ve always found it too disgusting to even contemplate, but as far as I can tell nobody who has ever first tried a cigarette liked it and they had to persevere to acquire the habit. Why do people want to acquire the habit? Why do teenagers want to be thought old enough to do things like drink, smoke and have sex? Why do people think smoking is glamorous and cool (the amount of admiring reblogging of black and white photos of movie stars from the 40s and 50s smoking, along with modern actors smoking, I’ve seen on Tumblr genuinely surprises me: most of these are younger generations who would have been raised on the message that “smoking is bad”)?

          • LHN says:

            The transgressiveness is probably a factor. Something worthy of Orwellian erasure from historical fiction and pop culture aimed at kids must be cool.

            And most kids haven’t run into someone with lung cancer or emphysema. Nor can you smell a smoke-filled room in a photo.

            I hate cigarette smoke, and have smoked a number of cigars countable on one hand. (Which I enjoyed enough that I’ve avoided doing it on occasions less momentous and sporadic than, e.g., the birth of a niece/nephew.) But subtracting the hanging miasma and the health effects, I see the attraction of the look, ritual, and accessories.

            Smoking also seems to have been something of a social boon for introverts at work and parties (based on anecdotal discussions with smoker friends, at least): it gives you something to do other than stare at the floor, and bumming a cigarette or asking for a light is an easy conversation opener.

            And once they were driven outside, winding up on the sidewalk for a smoke was a natural chance for people in different departments and places in the office hierarchy to interact on comparatively neutral terms. If it weren’t for all the smoke and the cancer (and these days, the expense), I’d have been tempted to take it up.

            (Though perhaps less so when it’s 5F/-15C out, as it is here now.)

          • nydwracu says:

            as far as I can tell nobody who has ever first tried a cigarette liked it

            I did, but my dad smoked awful-smelling cigarettes, so I was expecting worse than I got.

            (Then I picked up the habit, then I quit for a year to make sure I could, then when the year was up I picked it back up, then one day I forgot to buy a pack and realized I didn’t care, then my roommate got a coupon for some snus with one of his packs and I picked that up instead. Then he switched to dip, which was odd since he was a Jew from New York. Carried around a Sobe bottle, spat into it, forgot a few times and tried to drink from it.)

            Also, there’s a ritual element to it that all smokers understand.

          • Marc Whipple says:

            I have smoked exactly half of one cigarette in my entire life and have no desire to increase my count, but I have smoked cigars and pipes, off* and on, for more than twenty years. The smoking ritual’s importance is often underestimated, as are the tactile and visual elements. Research has indicated that people do not enjoy smoking as much if they cannot see the smoke.

            I had a good friend who was rapid-cycling bipolar, and who took a daily prescription regimen that would literally have killed me. The pills helped, but frankly I sometimes suspected that taking her cigarettes away would have caused her to totally lose her cheese much faster than going off the drugs. You could see her stabilize when she got them out, tapped the pack, lit up, and took the first few drags.

            While vaping seems likely to be much safer, in days gone by when I had a friend who needed to quit smoking cigarettes I would urge them to try a pipe. It was often helpful, as the pipe-smoking ritual is even more involved and calming than the cigarette-smoking ritual, and smoking a pipe is a very enjoyable experience for many. People who only smoke pipes experience far fewer health problems. (People who smoke cigars are somewhere in between, although closer to pipes.) IMO this is because they are much less likely to take the smoke into the lungs, as well as the fact that pipe tobacco is much less adulterated than cigarette tobacco (and there’s no paper.) IME people who switch from cigarettes to pipes (or cigars) often still inhale, but not nearly so much most of the time.

            *Mostly off. I have never experienced nicotine addiction.

        • Echo says:

          If the King of England is calling smoking a filthy, disgusting habit for the urban poor, does it count as high status?
          Because that was the actual situation.

          • Vox Imperatoris says:

            He was calling it a filthy, disgusting luxury of the rich.

            Tobacco wasn’t extremely cheap.

      • Mary says:

        Lots of fads have been high status — temporarily.

    • There’s real benefits and real benefits. Humans have a huge variety of customs, norms, clothing conventions, food laws, metaphysical beliefs, and so on. Conforming to them brings real benefits, for some values of real benefits. But it is not the kind of rock hard mind-independent reality that the hard sciences deal with, tje sense of real benefit in which it is reall ybeenficial to eat nutritious food and breathe an oxygen rich atmosphere. It is more the middle ground kind of thing the social sciences deal with.

  4. James Babcock says:

    This is exactly how I feel when reading the scientific literature on sodium, and I think most of the same forces are in play.

    • yarbel says:

      Care to elaborate?

      • 1angelette says:

        I’d suspect this is probably about the use of alkali metal salts to treat mood disorders. Lithium is the element most classically associated with treating bipolar disorder in particular, as it dampens mood swings of both types, though mania much more severely, leading to complaints of listlessness, cognitive impairment, and generally a cure worse than the disease.

        While much heavier alkali metal elements are pretty toxic to ingest, like Cesium being radioactive, you are probably aware that potassium and sodium are possible to ingest without immediately dying. Potassium is a fairly uncontroversial mineral, but sodium is where things get complicated. Plenty of bipolar people swear by the benefits of sodium chloride or table salt in totally preventing depressive episodes and stabilizing manic episodes. This just doesn’t get much airtime because “sodium causes high blood pressure”, blah blah blah, who cares. Most people I know personally who tried this tended to have a preexisting predilection towards salt, similar to this post. Now this is probably a lot safer an avenue to pursue if you aren’t already struggling with weight/hypertension or taking intensive medication. If you’re just a twentysomething in unidentified malaise who has been too scrupulous to indulge in salt, knock yourself out.

        • Anonymous says:

          Side point : caesium as found in nature is non-radioactive, and only mildly toxic.

        • Scott Alexander says:

          Any link for the sodium/bipolar thing? That sounds absolutely fascinating, and pretty believable since sodium is so close to lithium, but I’m having trouble finding any record of anyone making that claim.

        • onyomi says:

          This is my first time hearing about the sodium thing, but I have wondered sometimes if my constant desire to eat a gallon of hummus has anything to do with the tryptophan found in tahini. Now I wonder if it isn’t maybe the tryptophan and sodium…

        • Orphan Wilde says:

          Anyone who finds themselves craving salt should seriously consider trying iodine supplements.

          I had an insane predilection towards salt. Then one day I figured out my body was actually trying to get as much iodine as possible. When I’m not on absurd levels of iodine supplements, white rice can cause my blood sugar to drop dangerously low (hypothyroidism, probably?). (And iodine touched to my skin vanishes nearly immediately. Whatever my thyroid is doing with iodine, it demands a lot of it.)

  5. Pku says:

    Can you get the same nicotine dose through nicotine patches or something instead of e-cigarettes?

    • Tyler Hansen says:

      Sure, but why would you want to? E-cigarettes are just as safe and more fun. If you’re already smoking, it’s the best choice for substituting your cigarette smoking habit.

      (If you’re trying to imply that e-cigarettes aren’t safe, then please stop. How effective of a quitting mechanism is much more important than the error bars on e-cig safety)

      • Vox Imperatoris says:

        (If you’re trying to imply that e-cigarettes aren’t safe, then please stop. How effective of a quitting mechanism is much more important than the error bars on e-cig safety)

        What exactly do you mean by this?

        If you mean that e-cigarettes are much safer than regular cigarettes and that it’s almost certainly better to switch to them than keep smoking regular cigarettes, I agree. And if you mean that there should be no restrictions at all on e-cigarettes, I also agree.

        But if you mean that we shouldn’t investigate the question of whether e-cigarettes might be slightly more unhealthy than patches because the media might exaggerate the danger in order to scare people, I definitely disagree. That reflects a sort of principled dishonesty and obscurantism which I think is not in accord with the spirit of this blog.

        There are certain theoretical mechanisms by which I could imagine that e-cigarettes might pose a minor danger, mechanisms which are not present in patches. So whether any of these mechanisms actually do pose a real danger seems to me a question worthy of investigation.

        • Vaniver says:

          The claim is that e-cigs are somewhere between 95% and 100% safer than cigarettes, and if we suppose that nicotine patches are 100% safer than cigarettes, then it is a net win to use e-cigs over nicotine patches if the success rate of e-cigs as a smoking cessation mechanism is sufficiently higher than patches (since someone who goes on patches and continues smoking but would have stopped on e-cigs loses way more than the person who could have quit on both but quit using e-cigs).

          • Vox Imperatoris says:

            In that case, you can just tell people: if you want to be totally safe and maximize your expected health, go with cold turkey, if you can’t handle that, then patches, then e-cigs, then dipping tobacco, and only if all else fails, cigarettes. (If you’re taking fun into account, the situation may indeed be different.)

            Nicotine patches, as far as I’m aware, don’t make you start smoking cigarettes as opposed to e-cigarettes. It is indeed irresponsible to leave e-cigarettes out as an option.

            But all of this is scarcely related to the question of whether, if used, nicotine patches might indeed be slightly safer than cigarettes.

          • Tyler Hansen says:

            >In that case, you can just tell people: if you want to be totally safe and maximize your expected health, go with cold turkey, if you can’t handle that, then patches, then e-cigs, then dipping tobacco, and only if all else fails, cigarettes.

            Except that what actually maximizes your health is e-cigs, if you take into account P(smoking a year from now | try quitting with e-cigs) compared to P(smoking a year from now | try elaborate quitting decision tree)

          • Vox Imperatoris says:

            @ Tyler Hansen:

            This is an abuse of probability theory.

            You can’t just blindly apply generalized probabilities to your own case like this. There is no reason why going cold turkey or using patches acts cross-purpose to using e-cigarettes as a backup or in any way makes you more likely to smoke cigarettes. Especially if you buy an e-cigarette and keep it around the house for “emergencies”.

            The way they formulate the statistics is: probability that the subject successfully quit using patches vs. continued smoking. But if you are using e-cigarettes as a backup, the relevant statistic is: probability that you successfully quit using patches vs. you have to try e-cigarettes vs. that doesn’t work either and you smoke. And they haven’t done studies on that one.

            I’m not even arguing that people shouldn’t use e-cigarettes. I think it’s entirely reasonable. But you can’t use probabilities like that.

        • Tyler Hansen says:

          >But if you mean that we shouldn’t investigate the question of whether e-cigarettes might be slightly more unhealthy than patches because the media might exaggerate the danger in order to scare people, I definitely disagree. That reflects a sort of principled dishonesty and obscurantism which I think is not in accord with the spirit of this blog.

          We should investigate, sure, but we have enough information to justify pro e-cigarettes as the best course of action in terms of public health policy. E-cigs are safe, or at least close enough that having one fewer smoker is certainly worth whatever health costs e-cigs have on twenty users. We also know that if we replace people who quit using a patch with people who quit using an e-cig, we’ll go from something like 15 non-smokers to 20 exclusive e-cig users (again, huge error bars with numbers here – as a random first readable result from google,

          In terms of behavior, what I want out of us is to recommend e-cigarettes as being the best choice for smokers, given what we know about smoking cessation effectiveness and the relative health risks.

  6. xtmar says:

    There’s one study, which other studies contradict.

    I think the biggest takeaway from reading you and a few other places is that unless you have physics level certainty in something, the general answer to most things is “it depends on what you control for, and even then we think it’s A, but it might be ~A”. While I don’t mean to be anti-science or anti-knowledge, it seems like we should be a lot more circumspect in how we apply our knowledge, especially when we try to create new edicts that are somewhat hard to reverse.

    • Ilya Shpitser says:

      Yup. And you aren’t being anti-science or anti-knowledge. As John Tukey said:

      “The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.”

  7. Apropos of this … . I have never smoked. Now that e-cigarettes provide nicotine without most of the problems of ordinary cigarettes, should I seriously consider starting? Preventing Parkinson’s would be a big gain, although how large the effect would be this late I have no idea. And I gather there are other benefits.

    You might want to think about doing a post on the general subject of net costs and benefits to e-cigarettes for people not currently addicted to smoking.

    • Scott Alexander says:

      I think the answer is “there’s some benefit in that it seems to be a pleasant stimulant, it’s too early in the day e-cigarette-wise to know what the risks are, but usually there are some, and the graveyards are littered with people who underestimated the risks of a seemingly okay drug, not to mention the people who took their advice”

      But as always, Gwern’s thoughts are fairest and most rigorous:

      • Gwern seems on the whole positive. I had thought that nicotine itself was responsible for the heart problems, but apparently not. It can raise blood pressure, which one would want to check.

        I was thinking of vaping, but gum or patches seem like a simpler approach for the non-smoker.

        • Inifnite Light says:

          I take Good Sense Nicotine Lozenges. I find them very enjoyable and simple to use. I do not smoke.

        • zensunni couch-potato says:

          If David Friedman starts vaping I’ll be less concerned about the nicotine than about whether the world can handle such a concentrated dose of libertarianism.

          • Vox Imperatoris says:

            We very possibly might not have Atlas Shrugged today if it hadn’t been for nicotine and amphetamines.

          • John Schilling says:

            Might we have had something rather like “Atlas Shrugged”, but with maybe one-third the energy and stamina put into writing John Galt’s monologue?

        • Deiseach says:

          Or people could go back to taking snuff which apparently (if Wikipedia is to be believed) does not have a greater risk than ordinary smoking of developing cancer, and even avoids lung cancer.

          • onyomi says:

            But using snuff looks ridiculous, whereas smoking looks cool. Vaping looks less cool than smoking, but more cool than snuff or chewing tobacco. Chewing tobacco is cooler than snuff, but only cooler than smoking if you are a professional baseball player.

        • onyomi says:

          There was a study recently which associated vaping with the deadly “popcorn lung” (so-called because someone got it by deeply inhaling the fumes from microwaved popcorn every day), though I think one can find vaping formulations without the diacetyl chemical which apparently causes that problem.

          • eqdw says:

            Diacetyl is used to make buttery flavours. Stick with fruit and mint and you’re fine. There’s even some sites (I don’t have the link handy) that will list flavours by safety, for exactly this reason.

          • HeelBearCub says:

            Popcorn lung was, IIRC, most associated with the workers in the plants that made microwave popcorn, not people who eat lots of microwave popcorn…

          • HeelBearCub says:


            Allow me to quote from the article you linked:
            “”Popcorn lung” refers to a potentially fatal respiratory disease that mostly develops in workers at flavoring plants who inhale diacetyl, the chemical that provides the butter flavoring and smell.”

          • onyomi says:

            Yeah, but it came to my attention because of the unusual story about a guy who got it just from eating microwave popcorn every day and inhaling deeply.

            Not that it’s okay for factory workers to develop health problems, but it is also well-known that people working with chemicals in factories are at greater risk and so has not, so far as I know, been much of a story. It was, however, a news story that someone got this problem just by inhaling fumes from popcorn microwaved at home.

            My point is, if daily inhalation of microwave popcorn fumes can cause lung disease and some vaping flavors include the chemical which causes these problems, then there’s reason to believe that heavy use of those vape flavors might similarly cause a problem, even in people not working in factories.

          • HeelBearCub says:

            On the point of whether it is a good idea to include diacetyl in eCigs, I completely agree that it is not. As an aside, the fact that manufacturers seem to see fit to do so would appear to be a strike against trusting any health claims made by the manufacturers.

            I was mostly just picking a minor nit, as your post contained something which I knew to be (essentially) not true. Perhaps that was pedantic, or perhaps I needed to make clear that I wasn’t attacking the main point.

          • Orphan Wilde says:

            Many (most?) brands of cigarettes contain diacetyl, in concentrations much higher than e-cigs. It may be that some of it is destroyed by the higher temperatures (not sure?), but cigarettes still deliver higher concentrations than e-cigs.

            So as far as safety goes, e-cigs still beat the pants off of analog cigarettes.

          • Vox Imperatoris says:

            @ Orphan Wilde:

            Interesting! That’s certainly relevant. Here’s an article with links to some research on the subject. Incidentally, it says smoking has not been linked to “popcorn lung”.

            My main concern is: when the government (and private organizations misguidedly relying on government advice) bans e-cigarettes indoors, or when the government tries to ban e-cigarette flavors that “appeal to children”, or prevents advertising of them as being a safer alternative to smoking—how many people does that discourage from switching from cigarettes to e-cigarettes, and how many deaths result from that?

          • Douglas Knight says:

            Orphan, wikipedia claims that cigarette smoke has higher concentration of diacetyl than e-cigs, ie, after the heat has done its destruction.

      • onyomi says:

        It does seem to have that rare quality of being both stimulating and yet somehow relaxing. I have also heard friends report that combining caffeine with smoking produces a very productive mindset–hence, perhaps, the image of the author drinking coffee while chain smoking.

        I would try it but for concerns about addiction, etc. mentioned below.

    • Vox Imperatoris says:

      I would definitely be interested in such a post.

  8. On the general problem of people slanting their work to be on the side of the angels, I had an old post in a different context:

    • HeelBearCub says:

      @David Friedman:
      You have inserted global warming into a great deal of unrelated posts and threads lately.

      I really have to wonder what you are doing. It pattern matches to a few possibilities.

      Broadly, I would say this is against the spirit that motivated the beginning of Scott’s “Reign of Terror”. When you insert your favorite pet subject into as many places as possible, it grates.

      • Vox Imperatoris says:

        You said this more diplomatically than I was tempted to.

      • onyomi says:

        Not to speak for David, but the post isn’t really about global warming per se; it’s about the pattern he’s noticed with respect to certain issues like alcohol, smoking, and global warming–the pattern of wanting to reassure the reader that one is “still on the side of the angels” even when the data in a particular study seems to show, for example, that nicotine may be good for you, alcohol may be good for you, global warming may have some positive effects, etc. (the one about moderate drinking being good for you seems to have become fairly well accepted, however, and fits a different pattern I’ve noticed: the “people love good news about things they were inclined to do already” bias which I think is also a big force behind the recent popularity of high-fat diets).

        If David used global warming as an example it’s probably because that’s an issue he’s written a lot about on his own blog. If he wants to point to some examples of a phenomenon he’s already written about without having to rewrite then he’s necessarily going to end up linking posts tangentially related to his “pet” issues (moreover, he’s going to have more examples of any given intellectual phenomenon within the realm of his “pet” issues because those are the issues he likes to research and debate).

        • Nornagest says:

          I sometimes find myself wishing for a Journal of Irresponsible Science. Occasionally you end up wanting to research something that seems to tack against environmental or public health concerns, or falls outside the bounds of middle-class decorum, or requires feeding hallucinogenic cacti to undergrads without their knowledge or consent.

        • HeelBearCub says:

          If I hadn’t noticed this four times in about the past month, I wouldn’t have commented on it.

          I’m not sure whether Vox was agreeing with the observation, but if so, that would seem to be a fairly strong point in favor of the observation being true, and my mild consternation fair.

          • onyomi says:

            I also notice that when discussion of literature comes up, David is highly apt to bring up Rudyard Kipling. Presumably because he really likes Rudyard Kipling. Does this also grate? And if not, is the only difference that global warming is contentious and Rudyard Kipling is not? To avoid grating, must one limit mention of contentious subjects even when not contending?

            I’m not just asking these questions rhetorically: I tend to think one should be able to have meta-discussions which use contentious things for examples without actually delving into or even having an opinion on the contentious thing. But I can also understand how that may be difficult in practice.

          • Vox Imperatoris says:

            @ HeelBearCub:

            Yes, I was agreeing with you.

            @ onyomi:

            Oh, come on.

            You make a fair point that this particular comment isn’t about global warming per se.

            But still, I am not alone in observing that almost every thread here ends up having a little argument about global warming, usually involving David Friedman, usually with James Picone responding to him. It gets tiring.

            Okay, I don’t have to read them. Fair enough. But it still doesn’t need to be in every freaking thread.

            Whether Friedman meant that way or not, his comment is bait for someone in favor of the left-wing establishment position on global warming to come in and say, “No, the analogy does not apply to global warming.” And then Friedman or somebody else argues, “Ah, but it does,” and we have another little debate on global warming.

            Rudyard Kipling is neither here nor there. I’m not saying Friedman or anyone else should walk on eggshells. None of this would apply if this didn’t predictably keep happening.

      • In this particular case, the “insertion” was done almost four years ago, when the blog post I linked to was written.

        • HeelBearCub says:

          @David Friedman:
          Holy cow that strikes me as obtuse.

          You chose to link to a post on global warming in this thread. That didn’t happen four years ago.

          Now it grates even more.

          • “You chose to link to a post on global warming in this thread.”

            The first four paragraphs of the post I linked to are about alcohol as an example of the problem of people wanting to appear to be on the side of the angels—in that case against alcohol—which is one of the things being discussed here in the context of nicotine. I then gave a second example of the same pattern in the context of global warming.

            You have an odd definition of “a post on global warming.”

          • Nornagest says:

            You have an odd definition of “a post on global warming.”

            Some topics attract more attention than others. If the restaurant down the street comes out with a sandwich made from turkey, Swiss cheese, and fifty grams of pure crystalline capsaicin, I doubt many people are going to end up focusing on the turkey and swiss.

          • HeelBearCub says:

            @David Friedman:
            I apologize for the sentence fragment.

            It’s a post on global warming in amongst other things.

            Is that better? More accurate?

            And if so, does it make my statement weaker?

            Edit: And I will also note that the word “warming” is in the link and the title of the article. So it is fairly clear it was a central topic in the post.

          • “And if so, does it make my statement weaker?”

            It’s not a post on global warming. It’s not even a post on alcohol. It’s a post on the problem of people being unwilling to say positive things about something that is widely viewed as bad, even if those things are true. I said so when I gave the link and it should be obvious to anyone who reads the post without an axe to grind.

            [For any curious, here is the link to the post:


            The problem is illustrated with two examples. I start with the example that caught my attention–doctors who are unwilling to tell their patients that a low level of consumption of alcohol is good for them even though they believe it is true. I go on to a parallel involving global warming.

            The issue of people being reluctant to say good things about something thought of as bad was directly relevant to the discussion, since it concerned the fact that smoking, better perhaps the consumption of nicotine in other forms, might be good for schizophrenics, but the possibility had not been much explored—because everyone knew that smoking was bad.

            I only have one post on that subject on my blog, so I linked to it. Your position seems to be that I shouldn’t link to a relevant post because one of the examples in the post happens to involve global warming and you think I am too interested in global warming.

            That makes no sense to me, and suggests that you are the one who is dragging in global warming where it is irrelevant—as a reason to object to my linking to a relevant post. Your description of that post as “a post on global warming,” which it isn’t, similarly.

            [Psychiatrist shows patient a picture of a triangle, asks him what it reminds him of.

            Patient: “Sex.”

            Picture of a square. “Sex.”

            Picture of an X. “Sex.”

            Psychiatrist: “I think you have a problem.”

            Patient. “I have a problem? You’re the one showing me the dirty pictures.”]

            If the post had been written as a comment on this blog you would have a point, because I could be seen as using that particular example to bring up global warming in a discussion where it was irrelevant. But, as I pointed out, it was a post on my blog from nearly four years ago, and one relevant to the discussion thread.

            Unless you want to argue that I shouldn’t have taken opportunities to mention global warming on my own blog, your objection makes no sense.

          • HeelBearCub says:

            I really would like other people to go read what Friedman wrote and what he is now claiming about it.

            @David Friedman:
            What you are describing is not how that post reads, at all.

            There are two topics: alcohol, which is the preamble, and global warming, which is the conclusion.

            You establish a pattern with alcohol: “Good” people won’t admit positive things about something that is “evil”.

            This observation, by the way, is not a particular novel observation. But if you had wanted to write a post on the pattern, that certainly could be elucidating. You could mention other things that match these kinds of moralistic white lies. Smoking, with nicotine as a known anti-depressant, would have been another easy example. There are surely others. Perhaps defensive gun use. People in your outgroup. Global warming could even be used as an example of the general pattern.

            But rather than do this, go through all sorts of examples of the general pattern, you go immediately to the conclusion. Scientists are ignoring positive effects of global warming because it is just like “demon rum”.

            It’s not a post about a well known pattern of behavior. It’s a post about how climate scientists are displaying the pattern of behavior.

          • @HeelBearCub:

            I not only want people to look at the original post, I provided a link to it so as to encourage them to do so. They can then decide for themselves which of us is obsessed with global warming.

          • houseboatonstyx says:

            Before all this gets buried in the sands of later threads, I’d just like to thank Friedman and Picone for some of the most civil and informative sub-threads in the last few weeks. Imo, AGW (one way or another) is a more important topic than guns, abortion, size of government, etc etc, none of which latter are going to change any time soon, so it is good to have such sensible comments on AGW from opposiing parties. As for ‘gets tiring’, those F/P threads are quite easy to scroll past when they get excessively informative (if there can be such an excess on a such a geeky forum).

    • isionous says:

      Just so the other side is presented here: David Friedman’s links to old blog posts that mention global warming and David Friedman’s new comments that mention global warming do not grate me; I also do not suspect any reason for the direct and oblique mentions other than the fact that David Friedman finds the topic interesting.

  9. onyomi says:

    For a short while, I was very interested in the nutrition ideas of Linus Pauling and his disciples, though I now consider it to be mostly quackery. These are the people who advocate taking massive doses of vitamin C all the time, and who also, as it happens, view schizophrenia as some kind of niacin deficiency. I even read a few books by therapists of this ilk who claimed to be able to treat all sorts of psychiatric problems through addressing of supposed nutrient imbalances. In one such book, a therapist who worked with violent criminals noted that they all seemed to drink huge amounts of coffee and to add huge amounts of sugar to their coffee besides. This, he implied was some sort of coping strategy for whatever imbalance was also, at least in part, causing the undesirable behaviors.

    On the one hand, I am now pretty anti-vitamin. I think most people today are, if anything over-nourished, and that attempting to fine-tune our nutrition by taking nutrients in an unnatural form and quantity is far more likely to cause an imbalance than anything else.

    That said, I also think there is probably much to be learned in these sorts of subtle or not-so-subtle preferences like, it seems, of schizophrenics for smoking, and, supposedly, violent criminals for caffeine and sugar. It seems like it could be very fruitful to continue to explore this sort of thing with respect to many other problems and behaviors, including, perhaps, overeating.

    • nope says:

      Criminals are disproportionately ADHDers. ADHDers are known to self-medicate with caffeine before official diagnosis and treatment. This presumably explains the link.

      • onyomi says:

        Oh, interesting. That does make sense. It’s also always been interesting to me that ADHD gets treated with stimulants. I wonder if this is because stimulants have a paradoxical effect on them, or if it’s more like what seems like “hyperactivity” is actually a lack of the energy necessary to concentrate.

        • nope says:

          Well, it’s important to remember that the name of a class of drugs is often somewhat misleading in terms of actual function. Amphetamine and methylphenidate (ritalin) are stimulating for normal folks because for non-pathological dopaminergic systems, additional dopaminergic activity (and a bit of noradrenergic activity) cause a subjective energy boost. However, dopamine’s role in the brain is mostly *not* about energy. Specifically, in the prefrontal cortex, where these drugs are primarily acting, dopamine seems to be very important for executive functioning, especially affect regulation and reward sensitivity. A major problem in ADHD is inhibition of non-task-relevant responses, which in the hyperactive subtype takes the form of getting overly excited about all sorts of random things. The medications more or less help to channel that into concentration on a single thing at a time.

  10. Is it plausible that someone might actually do a randomized controlled trial on cigarettes? My guess is that that would be considered unethical?

    • gwern says:

      Probably, yes. But if you are creative and you spend more than 5 seconds seriously thinking about the problem, you can probably think of decent ways (aside from the obvious one already mentioned by Scott of using nicotine itself, which is by all sane accounts many times safer than cigarettes and hence much less unethical).

      Here’s one for free: there are trials for getting schizophrenics to quit smoking (presumably run by people who think that nicbacco is the devil and the correlations are nonsense); if schizophrenia symptoms are indeed relieved by nicotine, then the quit rate should both be lower than in anti-smoking trials for regular people and symptoms worsened. This will be a trifle difficult because the researchers will try to downplay any negative results. Here’s a second: anti-smoking trials in general are also ‘smoking for preventing schizophrenia’ trials if you look at them from the opposite perspective; follow the successful quitter groups and look for increased schizophrenia rates. (This is somewhat like the debates over lead and violence: ‘how can we possibly study the effects ethically with a randomized trial? We can’t poison people with lead deliberately!’ Geez guys, just do random lead remediation efforts. You don’t need to increase lead to do causal inference, you can also decrease it.) This is even harder because there will not be many such followups nor will they break out later schizophrenia diangoses, but it is ethically fine. Personally, I would say that if you care about this and you don’t want to mess with nicotine or reanalyses, the best option may be a third one: Mendelian randomization. We have dozens of GWAS hits for tobacco or nicotine, and even more for schizophrenia, AFAIK; it would not surprise me if a few of them formed useful instruments for a MR analysis, and since you are not administering any intervention but standing back and letting Nature take its course, there are no ethical concerns.

      • Bram Cohen says:

        One could also do a study on the efficacy of switching from smoking to vaping for schizophrenics vs. non-schizophrenics. If the schizophrenics really are self-medicating they should have a much easier time switching and staying switched.

      • Murphy says:

        I can think of a problem re: following up on anti-smoking trials.

        I’m not saying this is reality, just a hypothetical.

        Lets imagine a population with 2 kinds of people.

        Type A, unlikely to develop schizophrenia, no special relationship with nicotine.
        Type B, likely to develop schizophrenia, special relationship with nicotine that makes them far less likely to quit.
        Where A is a large group and B is a small group.

        If both groups were equally likely to quit smoking then just following up on anti-smoking trials would work.

        On the other hand, if they’re not then your group of successful quitters will secretly contain far more group A people as a proportion while your unsuccessful quitters will secretly contain far more of group B people.

        Roll on a few years and it will look like quitting smoking prevents schizophrenia even if the relationship is the other way round and schizophrenia makes it harder to quit smoking.

      • thisspaceavailable says:

        I see a problem with that, but given your history of outrageous incivility, I don’t see any point in trying to have a discussion with you.

        • Doctor Mist says:

          I’m trying to figure out why adding this comment was better than not adding this comment…

          • Anonymous says:

            I don’t understand what he means, but he seems to think it’s necessary and true.

          • thisspaceavailable says:

            What is your recommendation when one has comments on a post, but the writer of the post is so lacking in civility, intelligence, and/or sanity to be capable and/or willing of having a coherent conversation, and instead goes around making bizarre, contradictory, and incoherent assertions, then denying that they said them and responding with verbal abuse to requests for clarification?

  11. nope says:

    Don’t schizophrenic patients usually develop negative symptoms before positive symptoms? And aren’t these more often overlooked? Nicotine (and stimulants generally) have a rather obvious implication for the function-impairing symptoms like brain fog and low energy that are often more troublesome than the actual hallucinations and delusions, and fail to be treated by antipsychotics.

  12. Corey says:

    The nicotine-psychosis link seems testable in other ways. You’d expect to see measurable increases in schizophrenia diagnoses, mental hospital admissions, Risperdal prescriptions, or certain kinds of crime after cigarette taxes are raised, for instance.

    • Deiseach says:

      After reading this post, I read an account of a death by suicide where the person had “suddenly” become “psychotic and paranoid” a few weeks before the event, had made attempts at self-harm/threats of suicide, was admitted to hospital and in the cock-up of being transferred from the ER to the mental heath ward was allowed to go outside “for a cigarette” where they threw themselves over a disabled access ramp and eventually died of the resulting head injuries.

      The mention of smoking plus description of what sounded like schizophrenia (they had cut car tyres to pieces looking for listening devices) reminded me of this post, and maybe that sudden onset wasn’t so sudden after all, if they’d been self-medicating with smoking to control the initial symptoms for a while and were gradually getting worse.

      • houseboatonstyx says:

        @ Deiseach
        “maybe that sudden onset wasn’t so sudden after all, if they’d been self-medicating with smoking to control the initial symptoms for a while and were gradually getting worse”

        The self-medication may have been keeping down the symptoms, and a sudden attempt to stop smoking may have brought on the sudden worsening.

    • Scott Alexander says:

      I worry it’s too small an effect to find. Suppose a cigarette tax decreases smoking by 10%, that your study is for 10 years during which 20% of the population is in the risk window for schizophrenia, that 1% of people get schizophrenia, and that smoking increases the chance of schizophrenia by 30%. I think then you’re down to something like 0.01% more of the population getting schizophrenia, sounds really hard to notice in inherently noisy data.

      Also, I predict (without great evidence) that the will-become-schizophrenics are the least likely to be turned off smoking by a cigarette tax, since they might be getting a lot more benefit from smoking than other people.

  13. H.E. Pennypacker says:

    Grammar Nazi:

    “But if this is true, I wish the authors of the new study, and the editors of The Lancet, would have acknowledged…”

    “would have” should be “had”.

    • Vox Imperatoris says:

      Yes, “would have” is the conditional present perfect and refers to a present state. “Had” is the past perfect and (naturally) refers to a past state.

      For example, “If I had put all my money into Bitcoin in 2009, I would have made a million dollars.” If I had (in the past) put all my money into Bitcoin, I would now (in the present) be in the state of having made a million dollars.

      The main purpose of using the past perfect in the sentence Scott gives is to make the temporal relationship clearer. “But if this [were: subjunctive] (now) true, [optional “then”] I [optional “would”: awkward, but this construction carries the sense of “would” in any case] (now) wish the authors of the new study and the editors of The Lancet [had] acknowledged (in the past) the existence […]”

      On the other hand, there’s nothing strictly wrong about “would have acknowledged”. But besides not being the traditional construction, it’s not the best choice with regard to tense. You don’t care about the present state of the scientists; you care about their past state.


      I’m really not this much of a Grammar Nazi in real life. I just wanted to explain the thought process and logic behind this rule.

      • Creutzer says:

        I just wanted to explain the thought process and logic behind this rule.

        You didn’t, really. Explain it, I mean. There is simply a rule of English grammar that says you use the subjunctive mood or past tense (which are identical except for “were”/”was”) in the antecedent clause of a conditional sentence and the conditional mood (formed with the auxiliary “would”) in the consequent clause. “would have been” and “had been” are exactly as past as each other – in fact, both are perfects, contributed by the “have”. There is no place for rationalisations based on temporal relations or stativeness. In fact, if English used the conditional mood in the antecedent clause, “would have acknowledged” would be exactly what you’d see in the sentence in question. The tense is entirely fine.

        • Vox Imperatoris says:

          There is simply a rule of English grammar that says you use the subjunctive mood or past tense (which are identical except for “were”/”was”) in the antecedent clause of a conditional sentence

          This is a case of the past subjunctive, which is not the same as the past tense in general. It uses a past-tense form, but “If I were you” can carry a present meaning.

          The meaning being conveyed is distinct from the superficial form of the words used to convey it.

          In the same way, there is also the present subjunctive, which does not necessarily carry a present meaning. For example, “I asked that this project be completed yesterday.”

          English doesn’t have special subjunctive endings, let alone mandative vs. counterfactual endings, but it uses the forms it does have to express a real distinction between the two.

          “would have been” and “had been” are exactly as past as each other – in fact, both are perfects, contributed by the “have”.

          No, they are not. “Would have” is present perfect. “Had” is past perfect, i.e. pluperfect.

          The present perfect refers to a present state arising out of a past action. Past perfect refers to a past state arising out of an action completed before that time.

          This is independent of whether the clause is conditional.

          In fact, if English used the conditional mood in the antecedent clause, “would have acknowledged” would be exactly what you’d see in the sentence in question. The tense is entirely fine.

          I’m not exactly sure what you meant to say here. As I read it, this is a non sequitur. “Would have acknowledged” is the consequent clause, not the antecedent. I don’t see what the form of the antecedent clause has to do with it.

          It is true that English has no such form as “would had acknowledged”. But English nevertheless expresses conditionality in the consequent clause by means of the “had acknowledged” construction.

          The “simple (and arbitrary) rule of English grammar” you point out is not relevant, since “would have acknowledged” is not wrong by that rule, and the rule does nothing to explain why it is not the best choice.

          English expresses the subjunctive counterfactual by means of “If I were you”. In some dialects, it does this simply with the form “If I was you”. And, like other languages, it could do it with “If I would be you”. But any way it chose to do it, that would still be a subjunctive counterfactual.

          I’m not trying to advocate extreme prescriptivism here. There’s nothing objectively wrong with, for instance, African-American dialects of English. For instance, the sentence, “They don’t think it be like it is, but it do,” is very complex grammatically and uses the habitual be. (The meaning is: “They are under the mistaken impression that this state of affairs is unusual. In fact, it is quite usual.”)

          However, one generally assumes in a piece like this that the author meant to use standard English.

          • Creutzer says:

            “Had” is past perfect, i.e. pluperfect.

            “had been” is ambiguous between pluperfect (i.e. past perfect) and perfect subjunctive. When embedded under “wish”, it’s perfect subjunctive. Look at the semantics: “I wish you had done this” has no past perfect semantics – there is just one step into the past, which is caused by the perfect auxiliary “have”. Turning “have” into “had” does not, in this case, yield a second step into the past. “would have” involves exactly the same number of steps into the past – one, through “have”. The “would” just makes it conditional mood (or maybe it’s a modal quantifier, but that’s a whole other story).

            I can’t give you a Wikipedia citation for this because Wikipedia does traditional grammar and traditional grammarians don’t think in those terms. Actual linguists do, so all I could cite is research papers.

            The “simple (and arbitrary) rule of English grammar” you point out is not relevant, since “would have acknowledged” is not wrong by that rule, and the rule does nothing to explain why it is not the best choice.

            Sorry, my mistake for talking about the antecedents of conditionals when the issue is actually the complement of counterfactual “wish” (which is also a context that requires the subjunctive).

            However, one generally assumes in a piece like this that the author meant to use standard English.

            Oh, absolutely! I’m not disputing that your correction was appropriate, I’m merely saying you didn’t give the right explanation, because the issue is not about tense in any way, it’s just about whether the embedded context in question happens to require the subjunctive or the conditional.

            In the same way, there is also the present subjunctive, which does not necessarily carry a present meaning.

            Yeah, it’s an unfortunate thing about the word “subjunctive” that it’s used for very different forms. This present subjunctive, which is actually formed with a silent modal verb in American English (think of it as “I asked that this project [should] be completed”) and doesn’t exist in British English is very different from the subjunctive that’s used in conditional antecedents. They’re sometimes called “present subjunctive” and “past subjunctive” for historical reasons, relating mostly to the morphology, but this makes little sense in relation to their synchronic semantics.

          • Vox Imperatoris says:

            @ Creutzer:

            “had been” is ambiguous between pluperfect (i.e. past perfect) and perfect subjunctive. When embedded under “wish”, it’s perfect subjunctive. Look at the semantics: “I wish you had done this” has no past perfect semantics – there is just one step into the past, which is caused by the perfect auxiliary “have”. Turning “have” into “had” does not, in this case, yield a second step into the past. “would have” involves exactly the same number of steps into the past – one, through “have”. The “would” just makes it conditional mood (or maybe it’s a modal quantifier, but that’s a whole other story).

            I see what you’re saying here, and I partially agree, but I don’t think I fully agree.

            It’s true that you can’t say “I wish you would have done this,” while you can say “I wish you had done this.” And it is somewhat arbitrary. But they are both perfect. You can’t say “I wish you have done this,” either. The difference is that “had done” is a past perfect subjunctive, while “(would) have done” is a present perfect subjunctive.

            The tense, in my opinion, has something to do with it. You wish about a past state (prior to your wish), not a present state. And present perfect does refer to the present state of a person in a roundabout way: the present state of having done something in the past.

            If you were to express a conditional sentence about a present state, you would use “would” or “would have”. Like the previous sentence itself, or my earlier example: “If I had (in the past) put all my money into Bitcoin in 2009, I would have (now) made a million dollars.”

            Sorry, my mistake for talking about the antecedents of conditionals when the issue is actually the complement of counterfactual “wish” (which is also a context that requires the subjunctive).

            You are right, insofar as “wish” can only be used with a past perfect subjunctive like “had” and, arbitrarily, not with a present perfect subjunctive.

          • H.E. Pennypacker says:

            You guys are good at making grammar extremely complicated but then I guess that’s what most linguistics does.

            Why is it correct to say ‘I’ll buy the cheese if you’ll buy the wine’?

            You shouldn’t put a future form in the ‘if’ clause but in this example it’s OK. We make an exception for offers but there’s no real reason behind that as far as I can see.

          • Vox Imperatoris says:

            @ H.E. Pennypacker:

            Why is it correct to say ‘I’ll buy the cheese if you’ll buy the wine’?

            You shouldn’t put a future form in the ‘if’ clause but in this example it’s OK. We make an exception for offers but there’s no real reason behind that as far as I can see.

            In this case, “if” doesn’t take the subjunctive because it’s not counterfactual.

            For example, “If the Russians can put a man in space, then so can we.” It is known that the Russians can, in fact, put a man in space, so this construction takes the indicative. If this were dubious, we would say, “If the Russians were to be able to put a man in space, then so would we.”

            In your case, I’m not doubting that you will buy the wine. We could rephrase it as “If you will buy the wine, then I will buy the cheese.” (And your original example, the inflection, in speech, or the question mark, in writing, indicates that it is a suggestion.)

            (And you can use this in other examples that take the future indicative. For example, “Since he will die next month, I will inherit his estate.” “If”, despite often meaning the same thing as “since”, just normally can’t be used in this context, i.e. to introduce a conditional based on a future fact that is certain.)

            On the other hand, there is no future subjunctive form. “Should” is generally used to express this meaning. “If you should buy the wine, then I should/would buy the cheese.” (And this is a completely distinct meaning from “should” as in “ought to”.) This is despite the fact that “should” is actually a past-tense form.

          • Creutzer says:

            Why is it correct to say ‘I’ll buy the cheese if you’ll buy the wine’?

            This is actually a really hard question, even more so as other languages like German don’t allow this sort of thing. I don’t know of any analysis of this. It feels somehow subtly semantically different from “I’ll buy the cheese if you buy the wine”, but I can’t quite put my finger on it. Something about your buying the wine causing me to buy cheese vs. the fact that you will buy the wine causing me to buy the cheese, but then what’s even the difference between the two? “I’ll buy the wine if you’ll buy the cheese” allows for a situation where I buy the cheese before you buy the wine, knowing that you’ll buy wine – but so does, I think, the simple “I’ll buy the cheese if you buy the wine”.

            The counterfactual version of it, “I’d buy the cheese if you’d buy the wine (but I know you’re not going to buy wine” also seems kind of okay.

          • H.E. Pennypacker says:

            @ Vox Imperatoris

            I’m not asking why the sentence doesn’t take a subjunctive but why it’s OK in that example to use “will” in the “if” clause of a conditional*.

            For example you couldn’t say ‘If it will rain, I will take an umbrella’. Both clauses are about the future but only the main clause should take a future form.

            But ‘I’ll do it if you’ll do it’ or ‘If you’ll buy the cheese, I’ll buy the wine’ is fine.

            @ Creutzer

            I know what you mean about the slight semantic difference. I’m leaning towards that fact that the extra ‘will’ emphasises the fact that we are making this decision now, it wasn’t planned in the past. We often use ‘will’ for decisions or suggestions made at the time of speaking rather than things that are decided in advance. ‘I’m going to cook dinner’ suggests I’ve already decided to cook it and the matter is settled. ‘I’ll cook dinner’ is a suggestion I’m making now.

            Bonus question:

            Describe a situation where it would be correct to say ‘if it’s going to rain, I’ll take an umbrella’.

            *I don’t know proper linguistics terminology because all my understanding of grammar comes from teaching English as a foreign language.

          • Vox Imperatoris says:

            @ H.E. Pennypacker:

            For example you couldn’t say ‘If it will rain, I will take an umbrella’. Both clauses are about the future but only the main clause should take a future form.

            It doesn’t really violate any particular rule; it’s just unidiomatic.

            Present-tense forms can often carry a future meaning, as they do in this case. Or in, “I am going to France next week.” It’s somewhat stilted to say, “I will go to France next week.”

            “If” in this kind of case implies that you’re talking about the future, since they can’t very well do something different this very moment. So the “will” is superfluous and therefore left out.

          • Anonymous says:

            “What do you take with when you go to work?”
            “If it is going to rain, I take an umbrella.”

            My favorite English oddity along these lines is the baseball counterfactual tense “Posada catches that ball, we win.” (In a post game news conference or similar.)

          • H.E. Pennypacker says:

            @Vox Imperatoris

            I would say it certainly does violate a rule. The rule is that you don’t put future time clauses in the ‘if’ clause of a conditional. There are exceptions to this rule which is what we’re discussing.

            Of course what we mean by ‘rule’ is open to interpretation. When I was about 10 I had a teacher who would always correct us when said something like ‘me and my friends…’, making us say ‘my friends and I…’. We were just confused by this because it’s perfectly normal to say ‘me and my friends went to the shop’ despite being against some version of the rules. It’s the same with ‘whom’ – I used to sometimes tell students who used ‘whom’ that although they were technically correct it sounded weirdly formal to use it. ‘Less’ vs. ‘fewer’ is another classic example

            But to get back to the example at hand I can’t remember hearing native speakers consistently use a future form after ‘will’. I have heard dialects where people consistently use ‘would’ for unreal conditionals: ‘if he would stop being such an arsehole, I would give him back his money’. I think normally they also make the “mistake” that Scott makes above eg. ‘I wish you would have told me’.

            I also don’t think that ‘if’ necessarily implies the future. ‘If it’s raining, I’ll take an umbrella’ could mean ‘if it’s raining now’ (I don’t know if it’s raining I have to check first) or ‘if it’s raining later’ (obviously I’m not sure whether or not it will be raining later).


            Nice example, I was thinking of a situation where someone had just told it’s going to rain but yours works too.

            I didn’t know they used conditionals like that in baseball, I’d only seen it here on the other side of the Atlantic in football (soccer to you) punditry: ‘if he gets his head on that, it’s going in the back of the net’ also in post-game analysis.

          • Winter Shaker says:

            ‘Less’ vs. ‘fewer’ is another classic example

            Does it bother anyone that we only draw that distinction between countable and uncountable nouns when talking about a reduction or a smaller quantity/amount? We don’t have the same arguments about ‘more’ vs. ‘manyer’ 🙂

          • Marc Whipple says:

            @Winter Shaker:

            It didn’t until JUST NOW.

            Thanks heaps.

  14. Joe says:

    Apparently smoking cures everything!
    I’m happy I was able to quit but maybe I should start again in my 60s?

    • I wouldn’t trust any study that gets reported in the daily mail.

      • Anonymous says:

        This is (vulgar signalling dressed up as) a “Hitler was a vegetarian”-tier argument, and is undeserving of this forum.

        • Scott Alexander says:

          Maybe, but I generally find the Alzheimers-nicotine link to be much sketchier than either Parkinsons-nicotine or schizophrenia-nicotine.

        • Whatever Happened to Anonymous says:

          On the one hand, you’re absolutely right.

          On the other… Fuck the Daily Mail.

          • Vox Imperatoris says:

            I wouldn’t trust anything I read in the Daily Mail. On the other hand, it’s not the study authors’ fault if they get reported there.

          • I don’t trust the Daily Mail, but that just means that if I see something that might be interesting there, I see whether it’s reported in better sources.

          • Chevalier Mal Fet says:

            Perhaps we should rephrase the heuristic to “I wouldn’t trust anything I read *only* in the Daily Mail”?

          • John Schilling says:

            Wait, this Daily Mail? Because I have it on good authority that at least some of the work that shows up there is legit…

            But yes, worth checking to see where else it was published.

      • Joe says:

        My apologies I didn’t realize The Daily Mail was an untrustworthy site.

        • Nornagest says:

          It’s probably about one part actually untrustworthy and one part tribal. The Daily Mail is kind of like the British equivalent of Fox News.

          That said, popular science reporting always sucks. Take a grain of salt no matter where you see it, especially if it’s heartwarming or dramatic or touches remotely on politics.

          • Vox Imperatoris says:

            It’s probably most similar to the New York Post in terms of newspapers.

          • JBeshir says:

            Daily Mail is kind of its own political thing; it has formally endorsed politicians from both major parties at different times, although it’s generally felt to lean right. It’s probably well described as “authoritarian”. Very big on law and order in general.

            It’s best understood as playing to its audience- the middle aged, middle class people who want those hoodlums off their lawn and think everything is going downhill, roughly. It provides them with a mixture of ‘clickbait’-style scary/dramatic/surprising stuff, stories about government gone wrong in outrage inducing ways (being soft on bad people is a favourite), and its own brand of outrage porn built around finding good stories about bad people and writing them up, names and enough detail to find their address included. Didn’t pick up the nickname “The Daily Hate” for no reason.

            It’s specifically known for being bad at science, because of its fondness for shock and attention-grabbing, so best to look for other sources on it. I think of it as kind of Gawker for a different era and different demographic.

  15. Justin says:

    I tried to find data on other dopamine-associated disorders and smoking prevalence. Only found this so far: (OCDers have *lower* smoker rates than general population)

    Specifically I am interested in Tourette’s. Apparently there were some studies that suggest nicotine can improve the effect of neuroleptics. I have a mild form of Tourette’s and prefer to pass on the neuroleptics, but do feel some benefit from nicotine gums. But I haven’t done proper experiments. Ironically I have a trouble forming a nicotine habit.

  16. jy3 says:

    “No points for guessing what the Freudians say.”

    A cigar is just a cigar?

    It’s because cocaine is unfashionable?

  17. hanson's ghost says:

    Since no one else wants to come right out and say it – the Blue Tribe likes signalling against smoking tobacco for whatever reasons (they’re immaterial here), and e-cigarettes are especially objectionable to them because they invalidate the standard anti-tobacco dogma while still clearly being “smoking”.

    • onyomi says:

      I maybe nicotine suffers from being the key ingredient of a thing which everyone knows is very bad for you, but which seems not, itself, to be all that bad for you. If I understand, it’s mostly the tar and other smokey stuff which causes the lung cancer and emphysema, though I’m not sure if that’s also true of the heart disease and other problems.

      As to the cost-benefits of nicotine by itself, we are probably all biased against it. I will say that what worries me about it is just how intensely addictive it seems to be for some people. I have never tried it because I’ve seen my mother try and fail off and on to quit smoking all her life. I have no problem drinking responsibly, and have even successfully tapered off supposedly more addictive things like daily clonazepam, but I still am reluctant. Then again, the addictive power of nicotine may (or may not?) be overblown, again because of the huge campaign to fight smoking. I know a number of people, including my sister, who seem to be able to smoke very occasionally but without transitioning to daily, addictive use.

      To those who have smoked and/or drunk heavily and/or used heavier drugs, would you say nicotine is more addictive than say, caffeine or alcohol? Or even more addictive than say, benzodiazepines? Than… cocaine (I’ve even heard it said, though I doubt it)?

      I do also think that part of the addictive quality is just the having something in your mouth with something to do. Something to do with nervous energy eating, sex… I think people gain weight after quitting smoking primarily for this reason.

      Of course, if something isn’t actually bad for you or is even good for you, then the fact that it’s addictive may not be so much of a problem, but the general idea of being dependent on a substance is unappealing.

      • eqdw says:

        Last year I saw the Gwern link that was posted above, and it convinced me to start vaping. This was around the time I realized I may be undiagnosed ADHD, and had not yet seen a psychiatrist about this.

        My one major concern was addiction. Both my parents quit smoking when I was 14, and it was _hard_ for them. However, the Gwern post mentioned that, essentially, there are chemicals in tobacco (I want to say MAOIs, or whatever is in grapefruit that fucks with MAOIs) that make nicotine more addictive. ~1000x more addictive.

        I’ve been vaping for a year now. At a relatively low dosage. I am definitely addicted; I accidentally bought nicotine-free juice the other day, and definitely noticed when I loaded it. But it’s not a big deal. I can go a week without vaping, no significant negative effects. I’ll notice a lag in my energy, and my appetite will spike, but otherwise I’m fine.

        Compare with caffeine. If I don’t watch it, I’ll drink more and more coffee until I’m up to 8 cups a day. I try to keep a cadence where, at that point, I taper back down to zero for a week, to reset tolerance and the like. One time, I tapered too fast. The worst migraine in my life and I went home sick. This lasted three days, though by day 2 I realized what happened and drank some coffee. Do not cut from 8 cups to 0, cold turkey. You’re gonna have a bad time.

        So, in a nutshell: nicotine without tobacco is like a less addictive coffee. P.S. coffee is way, way more addictive than people think it is

        • HeelBearCub says:

          Anecdotally, this was the case for me with coffee…

          until I started taking extended release Adderal for adult ADHD.

          Now, I like coffee and I still drink it every morning. But I don’t continually up my caffeine intake. I rarely have more coffee in the afternoon. If I go a day without coffee I don’t get withdrawal symptoms.

          • eqdw says:

            I left this detail out, since I didn’t want to get on a tangent, but in November of last year i got a prescription for 18mg Concerta, and I drastically reduced my coffee consumption without even thinking about it. It just…. happened.

          • Anonymous says:

            Counter anecdote:

            Amphetamine initially reduced my caffeine consumption, but now I need MORE caffeine when I have taken amphetamine rather than not taken amphetamine, otherwise I feel cloudy-headed.

            This might be more a result of amphetamine tolerance rather than interaction with caffeine. After long-term use, amphetamine feels counterproductive to me.

        • I suspect that responses to caffeine vary a lot.

          I don’t normally drink coffee, but I typically consume something close to a gallon a day of diet coke or coke zero, which I believe contains as much caffeine as a lot of cups of coffee. It doesn’t keep me awake and I don’t observe any withdrawal symptoms when I stop consuming caffeine for several days.

          • Vox Imperatoris says:

            Same here.

            I try to keep it under a gallon, 🙂 but I drink a lot of soda. Maybe one to two quarts per day. I can drink it just before bed, and it doesn’t mess with my sleep. Only with my teeth.

            Also, caffeine doesn’t really stimulate me much, and I never get caffeine withdrawal headaches. I have to take around 400 mg of caffeine pills (which is a lot of caffeine) before I notice it.

            I don’t completely exclude the possibility that I am I just bad at introspection in this matter, however.

          • Nornagest says:

            Coke Zero has 34 mg of caffeine per twelve-ounce can, so a gallon of it has 360 mg; coffee’s caffeine content varies widely, but as a ballpark figure that’s about what you’d get from two mugs of strong joe or one large gas-station travel cup full of oversteeped gritty sludge. That’s a fairly moderate level of consumption, as coffee goes; I drink more, and I don’t think of myself as a heavy coffee drinker.

          • onyomi says:

            I believe the phenylalanine produced by artificial sweeteners is a mildly addictive dopamine stimulator independent of the caffeine content. Perhaps for this and other reasons (not likeing overly sweet beverages), I actually prefer the experience of drinking diet soft drinks to that of drinking the original, caloric flavors (that is, even if I had no desire to limit caloric intake I’d still drink the diet).

            I get most of my caffeine from tea, but don’t notice any particular withdrawal symptoms other than mild fatigue and sleepiness if I quit. I do, however, notice that when I go for a long time without caffeine I can wake up in the morning without feeling groggy more easily. When I’m accustomed to having caffeine in the morning I feel a strong urge to drink some soon after getting up in order to help me feel alert.

          • Deiseach says:

            I seem to be less sensitive to caffeine, though because I’m not a coffee drinker it’s hard to tell 🙂

            I get my caffeine from Coke and occasionally those energy drinks, which never give me that quick jolt of energy they promise (I presume mainly from the sugar plus caffeine): even when I tried a high caffeine one it did nothing for me.

            On the other hand, I get chocolate cravings shocking fierce (to the point of “If I don’t get something chocolatey right now, murder will happen”). So people definitely have idiosyncratic reactions, and one person’s “can give it up any time I like with no ill-effects” is another’s “only way to quit is cold turkey and that’s a nightmare”.

          • eqdw says:

            12 oz coke: 35mg

            8 oz home brewed coffee: 160mg

            Starbucks Grande coffee: 330mg

            If you’re drinking a gallon a day of coke, you’re getting serious caffeine, but the per-unit caffeine content is much much lower

          • houseboatonstyx says:

            @ Deiseach
            On the other hand, I get chocolate cravings shocking fierce (to the point of “If I don’t get something chocolatey right now, murder will happen”).

            What is your preferred chocolate delivery system, and what else may be in it? In the US we conveniently have chocolate bars labeled from 85% to 62% chocolate, and of course there’s Hershey’s cocoa powder without any sweetener.

          • Orphan Wilde says:

            Caffeine didn’t affect me until I was ~24.

            Not “Didn’t affect me much” – didn’t affect me. At all. It didn’t mess with my sleep, it didn’t make me energetic, it didn’t do anything.

            Around when I turned 24, it very abruptly started working, and working very well.

          • Deiseach says:

            “Preferred chocolate delivery system” is a lovely way to phrase it 🙂

            I should wax poetic about boutique chocolatiers here, but the plain truth is I’m not fussy, I’ll take anything from chocolate-coated digestives to bars of whatever is at the sweet counter.

            Though I have to say, since Kraft (Mondelez) took over Cadburys, I notice a definite downgrade in the quality/flavour of the product as well as them introducing all kinds of new formulations which are not as good as the original lines: no, I don’t like Oreos particularly and smooshing them up into a chocolate bar doesn’t make them any more appealing. They even introduced Philadelphia brand soft cheese with Cadburys’ chocolate which was disgusting and which, thankfully, has been discontinued.

            I used to like Green and Black’s range but again, since they were taken over by a larger company, they seem to have diminished that bit in quality.

            What is a complete burst of gorgeousness and something I very, very rarely indulge in (because if I didn’t try and maintain some control, I’d spend every penny on their goods) is Lorge Chocolatiers range.

        • Deiseach says:

          Re: addictiveness of tobacco/nicotine – it’s also that we’ve been growing and developing varieties of tobacco for a good while now, so we’ve artificially selected for more addictive crops than ‘natural’ tobacco usage would have provided. It’s the usual human thing of “If a little is good, then even more is going to be great”.

          Same way cannabis varieties are being bred to give ever more potent highs, which makes me think older studies about “cannabis use is relatively harmless” are outdated because people are not using the same varieties anymore.

        • Nornagest says:

          It might be worthwhile to distinguish between physically addictive (will give you withdrawal symptoms) and habit-forming (will give you cravings) here. Heavy coffee use gives awful withdrawal symptoms but it’s not particularly habit-forming; nicotine is strongly habit-forming but doesn’t seem to provoke as much in the way of withdrawal. At least not on its own; I’ve heard claims that non-nicotine compounds are responsible for a lot of cigarettes’ addictiveness, and I can’t speak for them.

          Which one’s harder to quit probably depends on how much willpower you have and how good you are at tapering off.

          • Saal says:

            Permit me to strongly disagree here with ‘doesn’t seem to provoke as much in the way of withdrawal’. I come from a rural southern area where all the boys started dipping or smoking at the 12-14 age range, and I’ve smoked a pack a day since shortly after I began at about that age (I’m in my mid twenties now). I’ve quit cold turkey several times, and every time I’ve experienced severe headaches, sweats, anxiety, mild jitters, occasionally nausea, etc. That’s on top of the gripiness, which is perhaps more accurately termed viciousness. I don’t know if the body of data on smoking cessation shows what you claim above, but that has most certainly not been my experience.

      • wysinwyg says:

        To those who have smoked and/or drunk heavily and/or used heavier drugs, would you say nicotine is more addictive than say, caffeine or alcohol?

        There are “social smokers” and others who smoke tobacco without becoming addicted. An uncle told me about a friend of his who smoked 2 packs a day until he decided cigarettes had become too expensive and then just quit cold turkey without any problems.

        Similarly for alcohol, but probably in different proportions — some people can have an occasional cocktail or glass of wine without any problems, and some people have serious problems with alcohol addiction.

        eqdw mentioned how addictive they find caffeine to be. Personally, I’m extremely sensitive to caffeine. I like the taste and the smell of coffee, and I enjoy the euphoria from caffeine, but a typical cup of coffee makes me so anxious that I inevitably regret having imbibed. As a result, I usually don’t have enough caffeine to have any kind of withdrawal symptoms afterwards.

        OTOH, I smoked tobacco for 12 years and just recently quit. (I had ended up smoking loose leaf so it was probably not quite as addictive as smoking cigarettes, per Gwern below.) So I found nicotine to be more addictive than any other drug I’ve tried, including cocaine.

        Except maybe marijuana, which isn’t supposed to be addictive at all but which I suspect I’ll find even harder to quit than nicotine. So I guess my point is that everybody’s different.

        • eqdw says:

          eqdw mentioned how addictive they find caffeine to be. Personally, I’m extremely sensitive to caffeine. I like the taste and the smell of coffee, and I enjoy the euphoria from caffeine, but a typical cup of coffee makes me so anxious that I inevitably regret having imbibed. As a result, I usually don’t have enough caffeine to have any kind of withdrawal symptoms afterwards.

          It’s a fine line between enough coffee to get my work done effectively, and too much so that I start to panic. I play this game of roulette entirely too much

          Except maybe marijuana, which isn’t supposed to be addictive at all but which I suspect I’ll find even harder to quit than nicotine. So I guess my point is that everybody’s different.

          I don’t find coffee psychologically addictive, at all. When I quit it too quickly (and i routinely quit it), I will get the worst withdrawal symptoms, but even at that point I don’t seen to associate it with lack of coffee, or feel an urge to drink coffee.

          Marijunana on the other hand. Not chemically addictive in the slightest. But very behaviourally addictive to me. If I have any, I’m going to smoke it every night until it’s gone. No moderation strategy works. None. So now my strategy is “my card expired so I guess I can’t buy any”. This works.

          Addiction’s a funny thing

      • Not My Ususal Handle says:

        To those who have smoked and/or drunk heavily and/or used heavier drugs, would you say nicotine is more addictive than say, caffeine or alcohol? Or even more addictive than say, benzodiazepines? Than… cocaine (I’ve even heard it said, though I doubt it)?

        For me at least cocaine was a strange one. Ten minutes after I did a line I would pretty strongly want another line. An hour after finishing a bag I’d still be thinking about maybe getting more. But the next day, I’d be able to think about it pretty rationally and it was weeks or months between the times I’d use. I ended up stopping altogether when I lost touch with that crowd and never really had any pangs.

      • houseboatonstyx says:

        @ onyomi
        To those who have smoked and/or drunk heavily and/or used heavier drugs, would you say nicotine is more addictive than say, caffeine or alcohol?

        When I was smoking three packs a day, I’d walk a block for a pack (not a mile), but then did some breath meditation and they quit me with no cravings etc. I take very little caffeine; dropping it makes me sleepy, but no discomfort.

      • zensunni couch-potato says:

        I smoked about a pack and a half a day until I took up vaping 6 months ago. Since then, I’ve had maybe 10 cigarettes total.

        For me, smoking is far more addictive than anything else, including cocaine. I’ve done cocaine maybe two dozen times over the past three years. Cocaine is very addictive while I’m doing it, and the crash is really bad. But the next day, and for several weeks after, I usually have little desire to do it, precisely because the crash is so unpleasant.

    • Scott Alexander says:

      Maybe we can cut down on both the unnecessary politicization and the overuse of “blue tribe”?

    • Faradn says:

      It’s probably more of a class thing.

    • vV_Vv says:

      I don’t think it’s a “blue tribe” thing. Nicotine is a psychoactive chemical which is higly addictive, and there is a large group of people, not clearly divided on the political spectrum, which considers the consumption of such substances intrinsically immoral.

      Think of alcohol prohibition, or the prohibition of most kinds of other drugs. There is a group that supported or supports these bans, probably based on some concept “suppression of vice”. You can call it “anti-wireheading”, if you want.

      The fact that the main method of assumption of nicotine, namely tobacco smoking, has very severe health risks used to be a boon for these people: instead of saying “you should not smoke because it is an immoral corruption of Virtue”, they can say “you should not smoke because it is a dangerous poison that is bad for your health”. E-cigs, which deliver nicotine in an apparently safe way, ruined this wonderful talking point.

      • Vamair says:

        I don’t really understand why, but seeing this point being ruined feels so heartwarming. Even though I don’t smoke or vape and never had.

    • nydwracu says:

      It’s generational. A lot of Blue Tribe people I know under 30 smoke.

    • Nornagest says:

      Insofar as this is a tribal thing, I think we’re looking at a tribal dislike of the people most likely to vape rather than a taboo against smoking as such.

      But tribalism isn’t the first place I’d look for an explanation.

  18. Neanderthal From Mordor says:

    One statistics that always baffled me is that smokers have far more car accidents than non-smokers.
    The study above also finds higher suicide among smokers.
    Of course, the most probable correlation and the one given in the study above is that more accidents are made by smokers being distracted from driving by their cigarettes and unfortunately the study doesn’t ask if the drivers were actually smoking when the accident happened.
    But another explanation would be that these 2 thing are not causal to each other, but related. Maybe smokers tend to have a psychological profile that makes them more prone to both smoking and driving poorly.

    • Mary says:

      Smoking is an anti-depressant, and so probably is attractive to people who are likely to commit suicide because of depression.

    • Deiseach says:

      I think the distraction thing is very likely; my mother was a smoker and had to smoke while driving, and with the distraction of getting a cigarette, lighting it, etc. she nearly swerved off the lane a couple of times.

      Doing that at speed on a busy motorway seems like a good way of getting into a car accident.

      • baconbacon says:

        Even more patient smokers who try not to smoke while driving will probably be in a rush to get where they are going.

    • onyomi says:

      Well nowadays almost everyone knows smoking is really bad for you (in fact, they may even have an exaggerated impression of how bad it is for you). Therefore, if you start smoking today you may be more likely to have a high time preference, or otherwise be prone to rash decision making.

  19. Maybe it’s time to take a detailed look (brain scans, blood tests, maybe even heaven forbid, interviews) at the short term effects of nicotine on schizophrenics vs. typical non-schizophrenics vs. people who would be expected to have schizophrenia but don’t.

  20. divalent says:

    Just a WAG/random thought: the 2nd study was in a population of swedish “conscripts”, which I assume means military, and so presumably we are talking very late teenagers, not younger ones. And I assume that if you actually have schizophrenia, you won’t be a conscript.

    So maybe smoking causes schizophrenia (in a “schizophrenia-prone” subset of the population), and heavy smoking causes it faster. So the subset of “schizophrenia-prone” individuals becomes relatively more depleted in the heavy smoker conscript population relative those that smoke less (or not at all), as many of them got schizophrenia already (and so can’t become a conscripts). So fewer of them get into the study to get schizophrenia later.

    • Cliff says:

      I don’t think people get schizophrenia as young teenagers?

      • Whatever Happened to Anonymous says:

        Early onset Schizophrenia is rare, but it is a thing.

        • divalent says:

          I’m no expert, but the Wikipedia article on it says that in 40% of males that get it “the condition manifested itself before the age of 19”. (But only 23% for women. But if this was military conscripts, likely it studied mostly males.)

  21. hnau says:

    I’m having trouble wrapping my head around the implications of the Zammit study. If smoking alleviates schizophrenia, I would have expected more smokers to be schizophrenics *after* adjusting for confounders– since the idea seems to be that schizophrenics (and pre-schizophrenics) learn that consuming nicotine helps. Is the theory that schizophrenia depends on a lot of risk factors, and after compensating for those risk factors, smoking turns out to help? But apparently those some risk factors also make people more likely to smoke– thus the correlation before removing confounders. And in that case it seems like we should be seeing the benefits of smoking even before analyzing the confounders.

    I guess the conclusion to draw is that the confounders create a lot more risk for schizophrenia relative to the reduction in risk that smoking provides, but I’d appreciate it if someone who knows statistics better could help disentangle this.

  22. TheNybbler says:

    Two things make me reach for my (metaphorical, sorry, been on the earlier thread) gun when reading studies. One is case-control studies and the other is meta-analysis. Both are just minefields of statistical gotchas even before you get into the other problems.

    Meta-analysis has gotten extremely popular, but I think this is to the detriment of science. The whole point of a controlled study is it’s controlled. If you took two studies which were the only ones in their study area performed, identical other than using disjoint sets of randomly-chosen subjects, then you could put the data together and have a valid meta-analysis. That’s never the case.

    • baconbacon says:

      Volume is generically better in science. A study of 10 people is fraught with danger no matter how rigourous you are in design. Larger numbers cut out a ton of problems which are impossible to deal with on a small scale.

  23. Jacobian says:

    I remember reading something that people who are genetically less likely to suffer bad long-term side effects of nicotine and smoking actually smoke more. Basically, if your body “knows” that smoking isn’t as bad for you then you’re likelier to smoke, which would reinforce the idea that schizophrenics smoke because it’s good for them. Has anyone heard of this research ? I can’t figure out a combination of Google search terms that wouldn’t spam me with “smoking is bad, nicotine is addictive” pages.

  24. Steve Sailer says:

    The Unitarian church near my house hosts a lot of different 12 step programs and there are always people out front on smoke breaks. They finally had to install a full size trash can for all the cigarette butts.

  25. jjbees says:

    I once volunteered with the mentally ill for some months, including many people with schizophrenia.

    At one mental hospital, smoking was banned, and sometimes all the patients would talk to me about is how badly they wanted to go home and smoke.
    Perhaps I am naive, but I always thought it was inhumane to take away the one pleasure of a person’s life. If you have schizophrenia it’s very likely that you are alienated from your friends and family, can’t get a good job and make money, and your body and mind are unreliable, so that even inside your own head, you can’t be secure. Smoking seems to provide a momentary respite and relief from the constant torment, and with the advent of e-cigarettes it just seems nasty and spiteful to withhold it. I feel pretty strongly about this.

    • God Damn John Jay says:

      “with the advent of e-cigarettes ”

      Nicotine gum is and has been a thing for years. You’re not the first to say this, and I am nitpicking, but if there is a place nitpicking is allowed its here.

      • CoCo Khalizar says:

        Yeah. Why is that more in the news? Is it the spit that one does? It’s probably that. E-smoking *looks* cool. So that’s something.

    • Cadie says:

      Also, when people are having an acute crisis or otherwise are in need of intensive care in a psychiatric unit, it’s a safe bet that they’re in a bad state, and their problem is urgent and serious. Cigarette smoking isn’t good either, but it’s not going to do a lot of harm in the short-term. Wouldn’t allowing occasional cigarette breaks or better yet e-cigs help the patient feel better and possibly even improve treatment compliance?

      A friend of mine who was a heavy smoker went into rehab for a different addiction and was forced to stop smoking for the duration of treatment. She hated this, and I was mad about it too, because it seemed like a terrible idea. When you’re addicted to something illegal that can kill you fairly quickly and isn’t hard to overdose on, recovering from that addiction is way more urgent and important than quitting smoking. My intuition is that waiting until recovery is fairly stable before attempting smoking cessation would usually be better, because then you’re not trying to solve two big problems at once and making it harder to stick with the first recovery. For other mental illnesses, like schizophrenia, I’d guess the smart thing to do is get the patient stable on a treatment regimen, and THEN try to reduce or eliminate the cigs, when it will be easier for them to stick with and not threaten their treatment plan.

    • Anthony says:

      When California banned smoking in all hospitals, the *staff* at the state mental hospitals protested loudly.

      • onyomi says:

        Because the staff wanted to smoke, or because their patients became harder to handle when they weren’t smoking?

        • Marc Whipple says:

          Not sure that’s an “or” question. 😉

          But based on my extensive reading about such places*, I’d bet a moderate sum on the latter, with some “the patients will be happier if they can smoke even if it’s not good for them.”

          *While I have actually read a bit, this is a joke meant to make clear I’m not qualified to have an actual opinion. 😉

        • Anthony says:

          The second, mostly. Most medical people know that they shouldn’t smoke, even if they can’t make themselves stop. So they won’t generally lobby for their right to smoke (especially indoors).

          But aside from the general calming effect smoking has on mental patients, access to cigarettes can be very useful for social control.

  26. Acedia says:

    I read a few studies last year purporting to show that nicotine (not smoking, nicotine specifically) lowers testosterone levels in men by a non-trivial amount, which was enough to ward me off getting into vaping.

  27. Rith says:

    Back in my day, we did not have ADHD. We had coffee-drinking smokers.

    Medications are only good in the right context. Is insulin a good thing? For a kid to play with recreationally? Tobacco is no different. Nothing to play with recreationally, but for the small number of people who are not properly medicated for certain things that smoking alleviates, it is a godsend.

    The gut has its own brain and that brain is very rich in acetylcholine receptors that are very sensitive to nicotine. IBS may for some people be something like schizophrenia of the gut.

    And it really needs to be remembered that nicotine is probably more of a limiter on smoking than the point. The point is the MAOI content. It’s not just the MAOI’s in the tobacco, it is the instantaneous exquisite control of smoking. So chewing tobacco is less precise, less satisfying, and less desireable. Nicotine patches / vaping just help keep the cognitive withdrawal effects at bay–they do nothing to support the MAOI levels that are the real reason the remaining smokers (at this point, because smokers can hardly get a job other than self-employment anymore) still smoke.

    I have had multiple doctors including both of the two psychiatrists I have seen in my life, and come to think of it, both of the psychologists I have seen in my life, recommend that I continue smoking.

    They said, you have ADHD, a handful of anxiety disorders, IBS, and arguably mild bipolar. To date in life you have managed all these and been a very high functioning individual with just smoking, coffee and sugar. And a lot of life-structure gimmicks.

    To treat you with prescriptions instead of these, you’d need for starters, a stimulant for adhd, which would make your IBS worse because you’d eat even more poorly, and that would obviously make your anxiety worse. And the worse anxiety would make your IBS worse, which would make your anxiety worse. Etc. Then you’d need a lot of benzodiazepines to take the edge off all that anxiety. Which might well mean you need more stimulants. Your brain will fight these also, of course, so it is not going to be easy to regulate dose. And there is the marginal manic depression you suffer from, so pretty sure we’re going to need to mix SSRI’s or something similar in there. And or lithium. By the time we get the cocktail of drugs just right, if we ever do, you’ll be taking a mixture that has probably never really been imagined let alone tested on a rat. But it is a safe bet it is more hazardous to your overall health than smoking and drinking coffee with sugar as you do now. If you can get by with smoking, and drinking coffee with a lot of sugar, by all means, keep doing that with my blessing.

    Another overlooked thing with smoking is that it releases glycogen from the liver, so it helps to manage blood glucose levels. This is where the sugar comes in. Smokers need sugar periodically to recharge the glycogen reserve. In the course of this, smoking leads to more inadvertent intemittent fasting, another health benefit. Intermittent fasting is the best way I have found to manage IBS–presumably allowing time to heal or calm down or something between meals.

    Sugar also has the wonderful property that it triggers insulin and that cleanses the smaller amino acids from the bloostream so big ole tryptophan can saunter across the blood brain barrier. Especially if you take vitamins in the morning, with coffee and milk and lots of sugar, you are dosing yourself with everything your brain needs to make serotonin, a poor man’s (gentle) SSRI.

    Thanks for bringing up this topic, Scott. There are numerous poorly understood ways that tobacco can become part of a self-medication regimen and just yanking that out from under people is not always a good thing, however well-intentioned.

    • Scott Alexander says:

      Although I know that tobacco contains MAOIs and MAOIs treat depression, I’ve never heard the MAOIs in tobacco considered in a pharmacological context before. It seems that they can’t be “real” MAOIs since they don’t kill you in various unpleasant ways for consuming perfectly ordinary substances. Are they MAOBIs? Reversible MAOIs? I’m not sure exactly what they are, exactly what’s going on, and what psych effects we should expect them to have, and I should look that up. I bet it’s impossible to find good data on whether they actually do anything considering the confounding effect of nicotine.

      • Rith says:

        “Recently, positron emission tomography imaging has shown that smokers have a much lower activity of peripheral and brain MAO-A (30%) and -B (40%) isozymes compared to non-smokers.”

        The presumptive causes include:

        Harmaline is a central nervous system stimulant and a “reversible inhibitor of MAO-A (RIMA)”.[2] This means that the risk of a hypertensive crisis, a dangerous high blood pressure crisis from eating tyramine-rich foods such as cheese, is likely lower with harmaline than with irreversible MAOIs such as phenelzine.

        The harmala alkaloids are psychoactive in humans.[1] Harmaline is shown to act as an acetylcholinesterase inhibitor.[3] Harmaline also stimulates striatal dopamine release in rats at very high dose levels.[4] Since harmaline is a reversible inhibitor of monoamine oxidase A, it could, in theory, induce both serotonin syndrome and hypertensive crises in combination with tyramine, serotonergics, catecholaminergics drugs or prodrugs. Harmaline containing plants and tryptamine containing plants are used in ayahausca brews. The inhibitory effects on monoamine oxidase allows dimethyltryptamine (DMT), the psychologically prominent chemical in the mixture, to bypass the extensive first-pass metabolism it undergoes upon ingestion; allowing a psychologically active quantity of the chemical to exist in the brain for a perceivable period of time.[5] Harmaline forces the anabolic metabolism of serotonin into normelatonin or n-acetylserotonin, and then to melatonin, the body’s principle sleep-regulating hormone and a powerful antioxidant.

        Several alkaloids that function as monoamine oxidase inhibitors (MAOIs) are found in the seeds of Peganum harmala (also known as Harmal or Syrian Rue), as well as tobacco leaves including harmine, harmaline, and harmalol, which are members of a group of substances with a similar chemical structure collectively known as harmala alkaloids. These alkaloids are of interest for their use in Amazonian shamanism, where they are derived from other plants. The harmala alkaloid harmine, once known as telepathine and banisterine, is a naturally occurring beta-carboline alkaloid that is structurally related to harmaline, and also found in the vine Banisteriopsis caapi. Tetrahydroharmine is also found in B. caapi and P. harmala. Dr. Alexander Shulgin has suggested that harmine may be a breakdown product of harmaline.[1] Harmine and harmaline are reversible MAOIs of the MAO-A isoform of the enzyme, and can stimulate the central nervous system by inhibiting the metabolism of monoamine compounds such as serotonin and norepinephrine.

        • Scott Alexander says:

          Thanks! That pretty much resolves all my questions. I think there are people who take harmaline as a supplement; I wonder if that is a good idea. And I wonder if people should start including liquid harmaline in e-cigarettes somehow, assuming that’s possible.

          • Von Sylvester says:

            I’d certainly be interested. I miss the hypomanic intellectual drive, stamina, and more frequent moments of elation that heavier smoking seemed to promote and which I later rightly or wrongly attributed to the MAOIs. (On the other hand, I like the greater sense of self-possession that’s emerged post-cessation, even if I’m bit jaded.)

          • Leif K-Brooks says:

            (The blogging system seems to be eating my comments, so I’m trying for a third time. Sorry if I end up with duplicates…)

            There is “whole tobacco alkaloid” e-cigarette liquid available that should contain it. However, there is a synergistic effect between nicotine and MAOIs that greatly increases addiction (gwern has citations on his nicotine page), so that’s a downside to consider.

  28. keranih says:

    Very interesting.

    A point of interest – Smoking is bad because lung cancer, COPD, etc. Yes, we know that now – and we also know that a larger effect of smoking is early death due to heart attacks (more so than lung cancer, which smokers get only if they don’t die earlier from heart attacks.

    But the process of convincing people – including the medical hive mind – that smoking was really not good for you at all in the face of the pleasure people got from smoking and sharing tobacco socially was very long and difficult. (The process of legally normalizing SSM by contrast was at light speed.) I don’t have it to hand in this moment, but there was an article in one of the 1950’s SF magazines (Amazing?) – an opinion column on the science of smoking hazards that just ripped apart the whole notion that smoking was that horrific.

    With that in mind, it really should not surprise me as much as it does to find that the public health establishment is emotionally dead-set against ecigarettes and is quite willing to push them to be legally classified as the same hazard categories for tobacco cigarettes. But it does surprise me, and makes me mad, because I have family members struggling with smoking addiction, and ecigs works where nothing else has.

    (E cigs open up whole new social patterns for smokers – they can spend entire hours with people who are not smokers! Comfortably and happily! I am not sure why people are finding this so hard to ‘get’ – I think there is something to the social group signaling thing to it.)

    Tenth and last – tobacco is a nasty poison used to kill insects before we invented safer things like DDT and organophasphates. Don’t use it.

    • HeelBearCub says:

      The fact that eCigs were developed and are marketed and sold by the same companies that develop, market and sell cigarettes substantially reduces the prior probability estimate of the safety of the product.

      The medical establishment doesn’t reject various other forms of nicotine replacement. Their skepticism of eCigs then seems justifiable.

      • Vox Imperatoris says:

        The fact that eCigs were developed and are marketed and sold by the same companies that develop, market and sell cigarettes substantially reduces the prior probability estimate of the safety of the product.

        Were they? That’s not my understanding of the situation. I thought they were invented by small-time companies and entrepreneurs. Wikipedia confirms this, saying they were invented by a Chinese guy named Hon Lik:

        Hon Lik, a Chinese pharmacist and inventor, who worked as a research pharmacist for a company producing ginseng products, is credited with the invention of the modern e-cigarette.[32] Lik quit smoking after his father, also a heavy smoker, died of lung cancer.[32] In 2003,[133] he thought of using a high frequency, piezoelectric ultrasound-emitting element to vaporize a pressurized jet of liquid containing nicotine.[138] This design creates a smoke-like vapor.[32] Lik said that using resistance heating obtained better results and he said the difficulty was to scale down the device to a small enough size.[139] Lik’s invention was intended to be an alternative to smoking.[140]

        Hon Lik patented the modern e-cigarette design in 2003.[1] Lik is credited with developing the first commercially successful electronic cigarette.[141] The e-cigarette was first introduced to the Chinese domestic market in 2004.[32] Many versions made their way to the U.S., sold mostly over the Internet by small marketing firms.[32] The company that Lik worked for, Golden Dragon Holdings, changed its name to Ruyan (如烟, literally “Resembling smoking”), and started exporting its products in 2005–2006[32] before receiving its first international patent in 2007.[138] Ruyan changed its company name to Dragonite International Limited.[142] Lik said in 2013 that “I really hope that the large international pharmaceutical groups get into manufacturing electronic cigarettes and that authorities like the FDA in the United States will continue to impose stricter and stricter standards so that the product will be as safe as possible.”[139] Most e-cigarettes today use a battery-powered heating element rather than the ultrasonic technology patented design from 2003.[6]

        Maybe “Big Tobacco” has jumped in now to get in on the action. But that’s just what you would expect them to do when a new product threatens to undermine their market share. Or, to be fair, maybe you’d expect a large company to keep its head in the sand until the last possible moment—but it’s not inconceivable that tobacco companies have learned from the past.

        • HeelBearCub says:

          So, I think we may be both correct.

          Here is a WSJ article that indicates that the market split is between “cig-alikes” and vaping units.

          It looks like Big Tobacco is pushing for regulatory capture of the broad vaping market. The vape-store/vape-unit market is more DIY, with product development lifecycle as low as 10 weeks.

          Neither of those make me feel particular sanguine that actually consumer health is much of a goal of those who are in it for profit.

          • Vox Imperatoris says:

            Do you think consumer health is ever “the goal” of people who are in an enterprise to make a profit?

            That’s such a weird criticism. People who want to make a profit want to provide what the consumers are willing to pay for. The extent to which they are concerned about health and willing to pay to be informed of and mitigate the risks is the extent to which consumers are concerned about health and willing to pay to be informed of and mitigate the risks.

            Do you think consumer health is “the goal” of the people who run the FDA? No, their goal is to get promoted / not get fired. And that has more to do with “don’t cause any scandals” than “maximize consumer health”. Though, like profit and health, the two are not entirely unrelated.

            If you run the FDA and tell people to smoke as many e-cigarettes as they can, well, that could cause a scandal—if they do, contrary to current evidence, turn out to be bad; or if they’re not bad but get attacked anyway. If you just give the general line about “Well, they’re unproven and might be unsafe, so we prohibit their being marketed as providing health benefits,” that’s not going to cause any scandals.

          • How feasible is it to have consumer health as a primary goal? It’s possible to not be grossly negligent, but (as shown at this blog) it’s hard to be sure of what’s actually good for people.

          • HeelBearCub says:

            You quoted the inventor of the originally vaping unit, who clearly did have better consumer health as a goal. I took that as an implicit criticism of my statement that we should reduce our priors on the likelihood of eCigs being safe.

            This all in a thread where someone is complaining that the medical establishment isn’t immediately jumping on the eCig bandwagon, so to speak.

            My understanding of nicotine replacement solutions (gum, patches, etc.) is that they were developed by the pharmaceutical industry. I don’t think it is a stretch to say that improved consumer health is a goal of the pharmaceutical industry. And they also do studies, a lot of them, which help increase confidence in the effects of the product.

            Take all that together, and I think it is perfectly reasonable for the medical establishment to take a “wait and see” attitude about how safe eCigs and vaporizers actually are.

          • Vox Imperatoris says:

            @ HeelBearCub:

            If there is a demand for a “safer cigarette” (and there surely is), then improved consumer health is also a goal of “Big Tobacco”. In the exact same sense that improved health is a goal of “Big Pharma”—who I do not think are any less driven by “greed”.

            The issue with electronic cigarettes is that it is against the law to advertise them as a safer cigarette.

            I don’t think “Big Tobacco” just wants to kill its customers. I’m sure, all else equal, it would prefer them to be healthy and live longer to buy more cigarettes. If there is a product which is just as desirable to customers as a cigarette but has no negative health effects, they can expand their user base.

            They can shout it from the rooftops: “Everything you wanted in a cigarette, with none of the drawbacks! Why quit, when you can buy them now, for only $9.99?!”

          • keranih says:

            This all in a thread where someone is complaining that the medical establishment isn’t immediately jumping on the eCig bandwagon, so to speak.

            Not what I meant/mean. My apologies for not being clear.

            I mean to complain that the medical establishment is immediately jumping on the OMG IT IS A CIGARETTE AND EMITS A GAS BAN IT BAN IT BAN IT bandwagon. Which is exactly what the conversation looks like, in person.

    • Cliff says:

      Tobacco is a plant bro

  29. Having talked to some homeless schizophrenics, the chances that the cigarette connection is just correlation seems high–homeless people smoke a lot, homeless people are schizophrenic a lot. (That could also explain the high % of income spent on cigarettes, if we’re talking about a guy who gets breakfast at the soup kitchen and occasionally spends a few begged dollars on a pack of cigarettes.)

    But that doesn’t really tell us why homeless people smoke in the first place. I suspect it helps control food cravings, and homeless folk probably don’t make the best life decisions. But maybe it does make their brains feel better. Certainly alcohol can reduce the symptoms of schizophrenia, (actually, I think it might be severe hangovers caused by alcohol; my memory here is vague,) so why not cigarettes?

  30. “It seems like part of the problem with schizophrenia is that the brain’s nicotine system isn’t working well. Smoking supplements nicotine and makes the system run smoother, so schizophrenics feel better when they smoke and continue to do so. This is the widely accepted self-medication hypothesis.”

    By nicotine system do you mean acetylcholine system?

    • Scott Alexander says:

      Nicotinic receptors are one kind of acetylcholine receptor, but they’re very different from the other type and I kind of think of them as a separate system. I don’t know if that’s pharmacologically correct or if there’s an official definition of “system”.

  31. Guy says:

    “No points for guessing what the Freudians say.”

    “It’s just a cigarette, lay off”?

  32. Michael K says:

    Don’t forget:
    Cigarette smoking: an underused tool in high-performance endurance training

  33. SUT says:

    Based on anecdotal evidence, there are significant differences in people:
    {Weed=Yes, Tobacco=No} vs. {Weed=No, Tobacco=Yes}

    The difference is greater* than even between TotalAbstainers(No,No) vs. TotalIndulgers(Yes,Yes)
    *greater: more relevant to forecasting mental health

    Based on the above propositions, I wonder if it’s even possible to control for Canabis use with ordinary statistical tools.

  34. onyomi says:

    If one is looking for an alternative chewable stimulant that may cause throat cancer but at least doesn’t cause lung cancer, I can recommend:

    I am, of course, sort of joking, but I did actually enjoy using these when I lived in Taiwan. They give you a different kind of energy and mental clarity as compared to caffeine and nothing is better if you have a cold. People always thought it was highly amusing/weird when I bought them as they are mostly associated with lower class native Taiwanese as opposed to post-49 immigrants and certainly not foreigners. I did not find it addictive in the sense that, when I came back to the US and had no access to them, it didn’t seem like any sort of hardship.

    • Stan le Knave says:

      Bloody hell.

      Theres a childrens book series i read when I was younger, Mortal Engines by Phillip Pullman (which I anticipate a lot of the audience here would enjoy, even as adults)

      One of the main characters is an aviatrix who is described as having red-stained teeth from chewing “betel nuts” – I’d always assumed that was some science fictional thing. Nope, turns out they’re real.

      You learn something every day it seems.

      • Marc Whipple says:

        Yep. Same thing happened to me when I read Ayo Gurkha, in which one of the antagonists is described as constantly chewing betel nuts and spitting the juice crudely. Ten-year-olds from farm towns in Iowa are often unfamiliar with such things. I didn’t realize the nuts were a pharmacological thing, I thought he was just a glutton.

  35. mobile says:

    PPS. Smoking is good for ulcerative colitis.

  36. Anonymous says:

    This is a very interesting hypothesis, thanks for writing about it. I don’t think you’re being very fair to Gurillo et al. though. They do not ignore the self medication hypothesis, they discuss it below. And they don’t conclude causality, they simply state what the results are and what their hypothesis is. As for the “clear benefits” line. Perhaps that was overbold, but it seemed to be referring to the effectiveness of smoking cessation programs themselves and the already known benefits (that’s what was cited in that sentence), not with respect to schizophrenia or symptoms.

    In view of all our findings, we think that the earlier onset of psychosis and higher risk in smokers of developing psychosis (albeit based on few studies) calls into question the self-medication hypothesis. The excess of smoking nicotine by people with psychosis has been assumed to be secondary to several factors associated with psychosis, subsumed under the umbrella term self- medication. Nicotine use has been suggested to reduce
    cognitive deficits and symptoms (such as visuospatial working memory and attention deficits,81 working memory, and selective attention),82 P50 inhibition,83 sensory gating, smooth pursuit, and antisaccadic eye movements.84 Nicotine has also been postulated to reduce sedating and other effects of antipsychotic drugs28 and to diminish negative symptoms of psychosis. Use of nicotine at a young age could be attributed to self- medication for anxiety in individuals at the prodromal stage of illness. However, a 2008 review stated that the tobacco industry monitored or directly funded research promoting the self-medication hypothesis, in particular, biological research.85 Furthermore, direct behavioural evidence has failed to show attentional benefits with nicotine by comparison with placebo, in smokers with schizophrenia versus controls.86

    And I don’t blame them at all for being suspicious of the self medication hypothesis – if I’m not mistaken, decades ago people seriously proposed that smoking was healthy because it was a form of stress relief. Should they have tried to address the other study in detail? It would have been nice, but its not necessarily within the scope of the paper – which is to establish a correlation that merits further investigation.

    • Scott Alexander says:

      I don’t think I said they “ignore” the self-medication hypothesis (I did say they ignore the ability of genes to confound their results). But I don’t think they do it justice either.

      Their spin on the self-medication hypothesis is “The fact that schizophrenics smoke early on means we don’t need the self-medication hypothesis”.

      I would like them to address the possibility that the reason schizophrenics smoke early on is the self-medication hypothesis, eg what they say with “Use of nicotine at a young age could be attributed to self-medication for anxiety in the prodromal stage of illness”, although if I understand this correctly we’re not talking about “prodromal” per se but rather a more lifelong tendency. They just dismiss this with “But some studies were funded by the tobacco industry”, even though as far as I can tell the vast majority aren’t.

      The current leading paradigm in this field explains away their results and contradicts what they say – they should have more than a sentence about this!

  37. technicalities says:

    they spend on average about 27% (!) of their income on cigarettes.

    To get the “27%” figure the source study for this seems to count only the public money the patients received. Since they were all outpatients this is maybe an overstatement.

    The sample was also only n=78 and seems to be drawn from a single hospital population, which means social confounders (if they knew each other and smoke together, etc).

  38. Jason says:

    Writes lengthy impassioned blog post that includes the phrase Aaaaaaah! Shamelessly types the following words:

    >I have nowhere near the evidence I would need to feel really strongly about this.

    • jimmy says:

      Are you implying that you cannot tell the difference between feeling strongly about nicotine being a good/bad thing for schizophrenia and feeling strongly about the appropriate confidence levels given the evidence?

  39. Kevin B says:

    I have a friend who says she did research on a team that found that asthma medication is a confounder in the cigarette/parkinson’s relationship. Anyone know about this?

  40. ArgleBargleZarg says:

    Doesn’t smoking cigarettes ease symptoms of ulcerative colitis? Is there any link between UC and schizophrenia?