How I quit drinking

Or, On Thought's Determination of Action

DISCLAIMER: Do not make your future contingent on some grad students’ ability to write convincing blog posts. If you might be an alcoholic—specifically, if you drink every day, especially if you know your life would be better if you quit, and especially especially if you feel like you’ve been trying hard and you just can’t do it—talk to your doctor and get support from your loved ones. At worst your doctor will say “You should do XYZ” and then you won’t do it. Maybe you’re afraid of the pain of quitting, but that’s future pain, future pain that will be outweighed by future pleasure. For now, you just have to talk to your doctor. I do not trust in the power of philosophy to cure the average reader’s medical disorder. This post will probably only have a chance to help you if you happen to be weird in the same ways I am.

---

Because every pleasure or pain provides, as it were, another nail to rivet the soul to the body and to weld them together. It makes the soul corporeal, so that it believes that truth is what the body says it is. As it shares the beliefs and delights of the body, I think it inevitably comes to share its ways and manner of life and is unable ever to reach Hades in a pure state; it is always full of body when it departs, so that it soon falls back into another body and grows with it as if it had been sewn into it. Because of this, it can have no part in the company of the divine, the pure and uniform.¹

So, I don’t really believe in weakness of will. Wills can’t be weak—it’s in your power to do everything that is in your power to do, and you can always just do it or not, for any reason you want. Doing something might be painful, and that pain might convince you not to do it, but that’s totally up to you. Making a choice can only be hard in the sense of being painful. Making a choice cannot be hard in the same way your muscles are simply too weak to lift a refrigerator.

I also don’t believe in akrasia, which commonly goes by the name “weakness of will,” though it has the advantage that the nonsense of the label is hidden behind a layer of Greek. Akrasia is defined as not acting the way you want to in respect of your rational volition, or not acting the way you intended to when temptations arise, stuff like that. Most philosophers who believe akrasia isn’t real, like Socrates or Kant,² think akrasia is unreal in the same way square circles are unreal: they’re conceptually impossible. According to Socrates, to recognize something as good just is wanting to do it; people only do bad things because they mistake them for good. According to Kant, it is the nature of the human that his action is determined by thought; nothing else is attributable to his free choice. But I, in contrast, think akrasia is unreal like how unicorns are unreal: they just happen to not exist.

Nevertheless, my own character has been one most philosophers would call “weak.” I’m a will of principle, meaning I don’t do moderation. If there’s reason to do something one day, there’s reason to do it every day, assuming conditions are the same every day. In particular, if a few drinks is a nice way to relax one evening, then obviously it’s a nice way to relax every evening. So, predictably, I became what is apparently considered a “heavy drinker.” I drank literally every day, usually about 4-6 drinks—enough to feel nice, I didn’t want to get wasted or anything. (And apparently, whether you’re a labeled a “heavy drinker” is determined by how many drinks you have per week; if they did that on purpose, it is ingenious).

I don’t think I’m an addict. When I’ve briefly abstained (e.g. during Lent when I was a Christian, and several times for a week after promising God not to drink for a week) the feeling isn’t “I’m sick, I need a drink to feel normal,” it’s “A few drinks would make me feel nice, and I want to feel nice.” So I abstained fine, then went overboard again when I was allowed to.

Nevertheless, while the effects were not catastrophic, they were still bad. My level of drinking cost a lot of money and made health problems more likely down the road, not to mention all the extra calories. Worst yet, I couldn’t really do anything intellectual during the evening, so I was less productive. Time I would have spent reading more philosophy was spent watching Gotham Chess—great content, but I don’t think I’ll be on my deathbed wishing I spent less time reading philosophy and more time watching Gotham Chess. I was aware of all these considerations, but none of those effects hinged on the decision of any single day, so nothing changed. I had “tried” to change, but the temptation to feel nice this evening always won. As significant as all these reasons are, as obvious as it was that my life would be better if I didn’t drink, those mere thoughts never exerted the pull on me that they should have.

Usually, the advice for how to get over problems like mine involves putting in place some external constraints that will make you do what you want. Avoid situations that tempt you, tell a friend to keep you accountable, and so on. But I really don’t want to do anything like that. It’s not the life I want. I want to be able to just do things because I decide they’re the best thing to do. I want to be free, I don’t want to have to coerce myself.

---

Now, here’s how I’ve always thought of akrasia. Humans are complicated things, made of multiple particles. High-level descriptions like “Bob believes that p” and “Bob desires that q” will correspond³ to very complicated patterns of what these particles do. “Bob believes that p” will correspond to the pattern in which Bob talks and acts in a way that makes sense conditional on p being true, and “Bob desires that q” will correspond to the pattern where Bob tries to make q obtain and talks like he would try to make it obtain.

But there is no law of physics that says that, just because a human conforms to the “believing p” pattern 90% of the time, they will also conform the other 10% of the time. Obviously. So in a case like that we might say Bob unqualifiedly believes p, because language is imprecise, even though he doesn’t perfectly conform to the concept of believing p. Or maybe Bob conforms to the “believing p” pattern 50% of the time and he conforms to the “believing ~p” pattern the other 50% of the time, in which case we can’t make any unqualified statement about what Bob does or doesn’t believe, and in this case we have someone whose attitudes are inconsistent.⁴ Our beliefs and ultimate desires are up to us, so it’s our job to make ourselves conform to one coherent pattern, but nothing guarantees that we will, in fact, do our job.

So, that’s how akrasia is possible. Bob may truly desire not to have another bag of chips in respect of his explicit thinking (which is to say, his thoughts and words make most sense under that interpretation), but desire to have another bag of chips in the respect that is manifest in his ultimate action. He has a general principle that fails to affect his ultimate action in the right way; that’s akrasia. No law of physics says that just because we truly and sincerely intend to do something then we will execute that intention when the time comes, even if the agent doesn’t retract their intention. It’s our job to make that not happen, but nothing guarantees that we will do our jobs.

I once read a post on LessWrong that I really didn’t like. It argued, essentially, that akrasia doesn’t exist, that what we call such is actually just self-deception about what we really want: “Akrasia, then, is a kind of suffering that arises from identifying with particular desires in spite of having already given them their fair weighting in coming to a choice of action.” I found some parts of the account plausible, but I really disliked the post because it seemed to set up the expectation that it would help with akrasia, but then it turned out the “help” consisted in helping you convince yourself that your akratic actions are the best thing to do all along. The user gives great advice if your goal is to feel better about yourself, but bad advice if your goal is to do the right thing and live a good life. Maybe it’s even good advice if you want to be happy, but it’s bad advice if you want to be worthy of happiness.

However, I’ve actually come to realize the conclusion is correct when it comes to actual cases of supposed akrasia. I think the account of akrasia I gave above describes a possible way agents could be, and maybe a way they actually are to a small extent, but I now doubt it’s what’s going on in ordinary cases where humans succumb to temptation and act against their apparent intentions. Just think of what it’s like to believe both p and ~p: you likely can’t, since you just can’t hold those two thoughts in your head as true at the same time. The moment you imagine believing one, you make it impossible to imagine believing the other. The contradiction goes away the moment you recognize it, even if you don’t know which belief (if either) to stick with. You can imagine conforming to one pattern half the time and the other pattern the other half; maybe when a blonde person asks you whether p you sincerely assert p, but when a brunette asks you you assert ~p. As a more realistic case, maybe you believed ~p, then changed your mind and figured p is true, and then later forgot about your change of mind and then asserted that ~p. Maybe in cases like those, you have contradictory beliefs.

But it seems to me that nothing like that is going on when I succumb to temptation. Believing I should do something—where the “should” here is that of instrumental reason, so it seems fair to say this belief is more or less the same as desiring I do that thing—and then not doing it would seem like a blatant contradiction that vanishes the moment you notice it. Maybe people forget what they sincerely planned to do, but that’s generally not what’s happening when people succumb to temptation. They say “I know I shouldn’t do this, but I really want to.” They say they want to do something, then they don’t do it, all without forgetting; they wish they could just do the thing the rational part of their soul says is right, and then they consistently don’t do it.

Perhaps action and desire are different from factual belief in a way that makes blatant contradictions easier to maintain. That’s what I believed; we evolved from animals that just automatically do whatever they’re most tempted to at the moment, and it stands to reason that adding thought on top of that won’t automatically endow us with motivations that will cause thought to win. But I don’t think this is what’s happening. If I’m super excited to be in possession of my $300 wad of cash, but then a guy says he’ll stab me if I don’t give it to him, then I’m not going to be remotely tempted to keep the money; all good for the evolutionary explanation. But if the guy says he’ll stab me in ten years if I don’t give him the cash now, and I really am certain he’ll do so, I still won’t feel remotely tempted to keep the money, even though it’s only my reasoning about my long-term interests that speaks in favor of giving the money and my short-term animal interests that speak in favor of keeping it. But these are the same kinds of desires at play when it comes to deciding whether or not to drink; I think if I don’t drink I truly will get more pleasure from that than I’ll miss out on, and yet I drink. One desire just happens to be for a long-term good.⁵

So, even though I think it’s a conceptual possibility that agents don’t act as they sincerely intend to, it’s doubtful that that’s happening when people succumb to temptation. I believe, rather, that when we do as we are tempted, we are doing what we really, truly, consistently, and unchangingly doing what we want to do. We then also lie to ourselves about what we really want; when it comes to the level of explicit thoughts, we say to ourselves that we really wanted to do the other thing, but ended up succumbing to some “weakness” that is neither explicable in terms of our reason nor in terms of external compulsion. Such was true of me—I really, truly believed that having more pleasant evenings and social gatherings was worth the money, health risks, and less time spent doing things worth doing. I told myself that I wanted to stop because “I should to stop” seemed like words that form a true sentence, but not because I looked at the relevant considerations and concluded that I should do what follows from them. But thinking words in a certain order will not affect your actions unless those words actually constitute thought about how to act. This is a simple and natural view of things, the sort of thing you’d antecedently expect to happen. I believed that the negatives were worth the positives, while also wanting other words to be the ones that appear in my head, and while having associated feelings that made the words seem like something I believe.⁶

Where the post referenced above goes wrong is by supposing that, just because people are doing what they most want to do all along (which I now agree with), that means all your desires got a “fair weighing” in your deciding how to act. But weighing is hard! You need thought to do that. If you superficially think “I shouldn’t do this” and then you don’t do it, your thought is not determining your action, and you are not using Reason to determine what the best thing to do is. Rather, the thing you conclude is the best thing to do is just what you’re automatically drawn towards independent of Reason, and there’s no basis for thinking that the thing you’d do automatically is what a “fair weighing” of your desires would have you do. What motivates you, rather, is an unfair weighing of your desires, one that gives an unfair advantage to the desires that can motivate you without you having to think.

---

But, from the title, you know the story has a happy ending. So how should one get out a conundrum like mine? Well, the usual way you come to get true beliefs. You look at the relevant considerations and see that they imply something, and then congrats, you have a new belief. Introspectively, it is obvious that the negatives of drinking as I did do not outweigh the positives. Therefore, I change my mind about what is desirable. As of now, drinking is not something that tempts me for more than a moment. My mind habitually suggests it to me from time to time, but drinking is an obviously incorrect action, it won’t get me what I want, so I don’t do it.

Imagine if people thought akrasia was a thing when it comes to belief about matters of fact. (Yes, people do in a sense believe things against their better judgment, but they can generally only do that by also deceiving themselves about whether that’s what they’re doing, which is not like akrasia). “Yeah, I grant that p and that q follows from p, so I really *want* to conclude that q, but I just don’t end up doing that!” We may say: “Concluding that q shouldn’t be an additional thing you do. Just see that p, and that q follows from p, all of which you grant, and then you’ll just see that q is true.” They reply: “I find no fault in your reasoning, but my mind is just weak. I always plan to apply modus ponens, but when the time comes, I just don’t do it.” It would seem that the only reason this person doesn’t conclude that q is because of a mental block of their own invention: their prior judgments about how to infer don’t give the desired result because of their strange state of mind, in which they think under the assumption that something else has to cooperate with their intellect before they can conclude q given that they accept that p and that q follows from p. Once they operate under this conception of themselves, it will be a self-fulfilling prophecy that they do not assent to conclusions whose sufficient conditions they claim to grant.

The character’s only chance for improvement is that they accept that there’s no way they truly understand what’s meant by p and what’s meant by q following from p if they do not, upon consideration, end up believing q. They have, rather, ended up deceived in thinking they really do grant the premises and that the conclusion follows from them. Once they get rid of this mental block, they can say “Okay, now I see that I was hesitant to accept q because I wasn’t really convinced that it follows from p. Now I can think through the reasons why it does; q does seem to follow, so I guess q is true.” But so long as they conceive of themselves as being capable of self-consciously failing to perform an inference, they will never be able to re-open the question of whether q really follows from p (since they think they already believe that!), the result of which would be inferring that q.

Usually, you don’t have to think about your beliefs in order to correct them. If I think it’s raining, and then I see it’s dry outside, I don’t necessarily think, “Oh, I have this false belief that it’s raining, I better correct it.” I just see the dry ground and figure it’s not raining; my thoughts are focused on the world, not on my mental states. But if, as in the case of apparent akrasia, people have a narrative about themselves that causes their practical thought and their actions to not be corrected in the right way—“I truly believe it would be better for me if I chose A, but I chose B instead out of weakness”—then they are no longer able to say “A is better for me than B, therefore I hereby choose A.” They can’t say that until they realize they liked B better all along, realize they thought B was better for bad reasons, and then change their mind and choose A.

So, assuming you are like me and you frequently succumb to so-called “weakness of will,” and assuming you are the kind of person who thinks they can improve their character by thinking about the nature of the will, what should you do? I’ll lay out again the steps that actually happened to me, and hope that they resonate with some people. So, I would advise:

(i) Remember that instrumental rationality is about getting what you already want. Being moral is hard; you have to sacrifice things you want for something else just because you ought to. Being instrumentally rational, at least to the extent that you do not succumb to temptation, should be easy. Over-drinking, not exercising, not going to the dentist, etc. when doing otherwise would make your life better should seem—and in fact are—as insane as refusing to light a $100 bill on fire when you know someone will pay you $1000 to do so. No matter how much you like having $100, that will not be a difficult thing to do. Doing something that makes your life worse is an incorrect action, and you shouldn’t entertain the possibility of doing it any more than you would entertain believing something you know to be incorrect.

(ii) Become convinced of the main theses of this post—conditional, on course, on them being true (do not believe false things because they will make your life better; none of this will work if you’re not really convinced anyway). Specifically, come to the realization that the reason you’re lying to yourself about what you think, that you’re doing what your doing because your actual evaluation was that the course of action you chose is best, and that the thoughts of yours that concluded in words like “I shouldn’t do this” were not really practical reasoning. They did not constitute thought about how to act. They were words you had go through your head because they are formally implied by other words that seem reasonable.

(iii) Now that you’re honest about what you’ve taken to be best, you can re-open the question of what to do, genuinely recognizing that you are reevaluating a suspect belief of yours, and it could go either way. In my case, this step was instantaneous, since the reasons were pretty decisive. If you’re already viewing your action as akrasia, then things are likely similar for you. Now that you are truly thinking about what to do, weigh the reasons and conclude on acting the way that will get you most of what you want.⁷ Do not just go through the usual words that end in “I should really stop drinking/etc.” Actually reason through the issue, knowing this is practical reasoning, which starts with facts and goals and ends with action.

(iv) Assuming yours is not a particularly unclear case regarding how the reasons bear on the issue, you have now come to a very certain conclusion about What To Do. Acting in another way would be as insane as refusing to be paid $1000 to light $100 on fire just because $100 is a lot of money.

---

Again, as of now, drinking is just an incorrect action for me. Now that I’ve weighed the considerations with the sincere intent to figure out how to act, there’s no reason why I would drink, any more than I would get stabbed because I’d really like to keep my $300. And in this case, I could not have corrected my false belief that drinking is desirable if I did not admit I believed it in the first place.

I know I can be dramatic, and sometimes I mistakenly overstate my views. But the facts are what they are: I drank 4-6 drinks a day, I knew it made my life worse, I felt I couldn’t stop, then I thought the above thoughts, and now I don’t drink, and it feels easy and free.⁸ Now, these words will likely fail to have the same effect for you unless you happen to be inclined to think about akrasia similarly to myself; even then, they might just not hit you the same way they hit me. But, really, nothing seems to have happened except my thinking the above. I really feel like a new person, not from not drinking, but from the control I have over myself. There have only been a handful of times in my life where I felt this good, free, and powerful.

1. Plato, Phaedo, 83d-e.

2. The keen among you will reply that Kant thinks akrasia is real, and in his work it goes by the name of frailty (see the Religion, 6:29). But “Kant believed p, therefore Kant did not believe ~p” is not an infallible inference. I feel forced to conclude that the discussion of frailty is one of those cases where he just didn’t put two and two together. He defines “frailty” as “the general weakness of the human heart in complying with the adopted maxims.” His so-called Incorporation Thesis states that “[the power of choice] cannot be determined to action through any incentive except so far as the human being has incorporated it into his maxim (has made it into a universal rule for himself, according to which he wills to conduct himself)” (6:23-24). It follows from the latter that if an agent voluntarily succumbs to temptation, they ipso facto have adopted a maxim permitting such, rather than such being a case of failing to “[comply] with the adopted maxims.” A remaining possibility is that the agent has two contradictory maxims; without going too into detail, I doubt Kant himself would be happy with that claim (his arguments for rigorism would not go through if we allow that possibility). If anyone knows a way to reconcile these statements of Kant’s, I really want to know.

3. I said correspond.

4. Just in case you’re unaware, the ~ stands for “not.” So p might stand for “it is raining,” and ~p might stand for “it is not raining.”

5. Donald Davidson has a view of akrasia on which the agent is not guilty of a logical contradiction (1969, “How is Weakness of the Will Possible?”). He thinks the aktratic agent makes a judgment that, in light of all the considerations (or perhaps in light of some particular ones), one option is “prima facie” better than another, all while failing to conclude that option is unconditionally better than the other. I think, as is common in Davidson’s papers, his case rests on formally clever stipulations that only give the appearance of solving the deeper issue. Specifically, here, the problem is that if it is evident that akrasia exists at all, it’s comparably evident that akrasia exists in the sense of agents really, unconditionally judging one option is better than another any not doing it (going by SEP, unsurprisingly, it looks like I’m not the first person to whom this problem occurred). More generally, insofar as I am ever akratic, there are cases where I really come to a practical conclusion and then don’t act on it, all without changing my mind. That is, there would seem just as well to be cases where the same thing I conclude in abstract deliberation is the same thing I reject with my action. Therefore, I do not think our intuitions about akrasia can be faithfully secured by any view on which aktratic action does not involve making a logical error. (I am, however, going to argue that our intuitions are not to be vindicated.)

6. Of course, this is not how things actually felt. They felt the way you’d expect them to; the thoughts came along with the feeling that I really do desire to stop drinking, along with what appeared as an intention to stop. But thoughts plus feelings do not constitute desires, if “desires” are understood as the attitude that is manifest in what one wills.

7. I would be remiss to not add, “unless you want things forbidden by the Moral Law.” This post is for decisions that primarily just affect your life.

8. I might go back to social drinking in the future, since that seems like a natural and psychologically realistic limit to set. But only after I’m used to not drinking and I can be sure I’m only doing it because I think it’s best.

Also, in case you think my success is really is evidence that I’m right about the will, I don’t want to be giving a biased perspective. Things seemed to happen to me as I just described them; I’ve “tried” to quit drinking and failed, all my circumstances the same, the only apparent difference being my realizations above. But this is still an interpretation of my experience, which is not infallible. Moreover, I intended to keep my choice to myself for about a week (and this will have been posted only after a week), to make sure the change was affected by nothing other than my own will. But I was so excited about my realizations that I gestured towards the above with some friends, then I felt obnoxious keeping it a secret, so I told them I quit and how. But that does mean I have additional incentive to not drink—the usual pressure you get from telling others about a commitment you made, and the desire that I won’t prove my philosophizing to them to be fake. It doesn’t feel to me that the latter are really motivating me—and if they were, then similar things would have motivated me sooner—but you are not me, and so take things with a grain of salt. (I’ve been tempted a few times in the past week—nothing felt like it was motivating me except that drinking is an incorrect action, and these external considerations didn’t even occur to me).

Two other causal factors contributing to this change include (i) my writing the post against causal decision theory which concerned the unity of the will, and (ii) reading Andy Masley’s post about drinking. On the latter, the article is very good, but I don’t think the considerations in it (being so indirect and uncertain) had much of a motivational pull on me. On the former, it merely expresses views on the will I’ve had for a long time. But I think these two things happening together is what caused me to have the above realizations, which in turn caused me to stop drinking. Truly, I was putting my trash bins away, and it all hit me in an instant, and now I am free.