by Carl V Phillips
Vaping, smoking, and other tobacco product use are routinely described as “addictive”. As I have pointed out repeatedly, this is a very misleading characterization. (You might recall my major essay – years in the making – on this topic from earlier this year. If you missed it and are reading this, you will definitely want to go read it.) The two sentence summary of the headline point is: All ‘official’ definitions of “addiction” hinge on the behavior being highly disruptive to someone’s functioning – work, social, etc. But tobacco product use has no such effects, at least not for more than a minuscule fraction of consumers.
So the fallback position, in the event that someone recognizes the problem with that word, is to say that tobacco product use produces dependence. But this is barely more accurate and is equally misleading. For those who use these products or advocate for their acceptance, to use of either of these words is to make the rhetorically and psychologically dangerous mistake of adopting the language of one’s oppressors.
“Addiction” evokes the specter of destitute drug users who are entirely focused on getting their next fix, at huge cost to every other aspect of their lives. Use of this word is a calculated choice by Big Tobacco Control, to intentionally invoke that misleading picture. Before the last decade of the 20th century — back when tobacco controllers were honest decent people — they did not say, and often even explicitly denied, that tobacco products should be considered addictive.
But “dependence”, even though it is a technical term with a reasonably established definition, evokes much of the same specter. Technically the word means that someone experiences withdrawal symptoms when and if they cease the consumption. But what kind of symptoms and how strong are they? The dramatic specter is that of a heroin user in severe pain and gastrointestinal meltdown during withdrawal, or the heavy drinker suffering tremors and fever when “DT-ing”. The message contained in the word is that any payback from giving up a drug or other enjoyed consumption choice should be considered as dire as those horrible pains — which is to say, as good a reason for avoiding the choice.
Suffering some cost from ending longstanding pleasant practices is the norm in life, not the exception. We mourn, long, and miss. If we start implicitly conceding that any welfare loss that results from ceasing a rewarding behavior renders that behavior bad, we are implicitly endorsing the worst excesses of the “public health” political movement. There are a million good novels and other stories about love and loss, as well as less cliched combinations of benefiting from something and then suffering the backlash of its absence. This is the price we pay of the best parts of the human condition. Meanwhile, there are a million bad public health papers that implicitly recommend that we should restrict the human condition to shades of dull grey so that we never have to experience the cost of losing something. Before you fall into the trap of whose language you adopt, ask yourself whether you want to live a life as told in stories by Tolstoi, Hugo, George Jones, Gaiman, and various saints, or the stories told by some scolding public health puritans in their ham-handed Discussion sections.
The word “dependence” is not usually intended to include many of those losses I invoked in the previous paragraph, but the point remains that there being some negative effect of quitting is hardly a good reason for never starting. “Dependence” usually implies effects that are dramatic and distinct shortly after we stop “using”, and that are limited to the quasi-physical rather than thought-based (i.e., your body is messed up, or feels like it is, due to what is happening in your brain, rather than merely feeling a longing). Consider the well-known effects of giving up caffeine, or even just skipping it for a day: a splitting headache and difficulty focusing. Being bereft of one’s habitual nicotine also has a collection of these quasi-physical effects.
The two are not all that different from each other, so our lack of social condemnation of coffee must mean that creation of dependence alone — even in this narrow sense of the concept — is not a defining problem in itself. This is to say nothing of our celebration of countless other activities that bring us bouts of unhappiness when they end.
We accept the withdrawal effects for coffee as a simple legitimate cost-benefit calculation. If you choose to become a habitual coffee drinker then you will either need to go through a few days of payback when you quit, or just keep doing it forever. The latter is the preferred choice for most of us. But the dependence has approximately no influence on that choice. Yes, it is true that most of us “coffee addicts” (jk — see above) will occasionally put some effort into getting our “fix” on days when our routine is disrupted, perhaps drinking some swill that is barely palatable merely because of our dependence (i.e., primarily to avoid the withdrawal symptoms). But that is hardly specter-worthy stuff. It is just about a small cost for a day; we plan to resume drinking good coffee according to our routine as soon as it is convenient. The specter that is evoked by “dependence” is that it motivates substantial unfortunate choices.
The acute pain of opioid withdrawal is enough, by itself, to keep some people coming back. They may have decided that they are better off no longer getting high in order to recover what they lose as a result of their drug use, and are unwavering in that. But they just cannot get through the intense acute negatives to get there. In the jargon of economics, they are stuck at an inferior equilibrium because of the short-term costs of transitioning to the new equilibrium. This is the specter that the technically-correct but fraught-with-connotations word, dependence, evokes.
We have therapies and reduced-risk maintenance drugs that make a huge difference in getting someone through opioid withdrawal (and also managing some of the less dramatic withdrawal effects that would otherwise continue for months or years). And, oh look, we have those for smoking/nicotine also. The thing is, the former are quite useful while the latter offer approximately no effect on smoking cessation. Did you ever stop to wonder why there is such a huge contrast?
The answer is pretty clear: Withdrawal is a trivial barrier to quitting smoking for most people. Smoking is not about dependence. It was never about dependence. Helping someone “get over the hump” to the new preferred equilibrium is seldom the issue it often is with opioids. The problem is that the new equilibrium is simply not preferred.
(I should pause to make clear that I am not implying that everyone has the same experiences and effects. There are always exceptions to what the majority experience, including some huge outliers. Indeed, the stereotypical scenarios are often far worse than the average experience. Anyone who makes blanket declarations about these behavioral issues (*cough* *tobacco controllers* *cough*) is demonstrating that they are clueless about them.)
When someone decides they want to give up coffee, they whine for a day about the headache and maybe take a Tylenol, and they grouse for a week about being groggy. But they just do it. Having decided they would be better off spending the rest of their life coffee-free, the dependence is not that big a deal.
Why they decided this is a complete mystery to me. I do not wonder how they endure the withdrawal or are willing to go through it — doing so is simply no big deal. No, the mystery to me is why anyone would choose to give up one of life’s great joys. That’s just nuts.
Ok, I will take that back. As noted, I recognize and respect heterogeneities, and realize someone might decide that continued coffee consumption causes them more harm than good. But you see my point: The reason that I and most people do not quit coffee is driven not by the withdrawal symptoms for a week, but being bereft of coffee for the ensuing years. We like it and want to keep doing it. For a large portion of the population, a day that includes consuming coffee is better — on net, after considering all costs — than one bereft of coffee. That is often not true for heroin.
This story about coffee is the same as the story for why most people keep smoking or using other tobacco products. A day with the consumption is better than a day without it. Yes, the first few days without it are worse than either of those. But that temporary extra cost is pretty trivial. For most people, having Tylenol or nicotine gum to “take the edge off” is not going to change the calculation about quitting (whereas methadone might). The real reason for not quitting is not the short-term extra-low welfare, but the continuing loss of welfare for all the time after that. Lots of smoking cessation attempts fail not during the withdrawal period, but months later. It boggles the mind that anyone can be aware of that fact and still think the failure to quit smoking has much to do with dependence.
No doubt you have seen the trope about “nicotine is more addictive than heroin” (by which they mean more dependence-producing). This is absurd at so many levels, not least because the people who say it cannot even offer a definition of addiction (or dependence), let alone a method of quantifying it to make a comparison. It is arguably even more absurd when you realize that this claim is not actually about quantifying addiction or dependence; it is a twisted interpretation of statistics for the success rate for self-declared quit attempts. Some such statistics suggest a larger portion of attempts are successful for heroin than for cigarettes. It is, of course, absurd to conclude this means it is less of a burden to go through heroin withdrawal. How stupid would someone have to be to believe that? What these statistics reflect is: (a) it is much easier to make an unserious quit attempt with smoking so that denominator is inflated and that, more importantly, (b) the motivation to be rid of the day-to-day effects of heroin is much stronger.
A stereotypical failed attempt to quit heroin involves the overwhelming discomfort of withdrawal dragging someone back. A typical failed attempt to quit smoking is saying (perhaps on the first day or perhaps a few months later), “this turned out to be a really bad time to quit, so I will just do it later — not a big deal.” This results in quips along the lines of “quitting smoking is easy — I’ve done it dozens of times.” No one says this about heroin.
(Aside: it actually is a bit of a bigger deal to delay quitting smoking than most people think. Just a few months delay poses as much risk as a lifetime of using a low-risk alternative.)
Perhaps you recall
Leslie Nielsen [correction: Lloyd Bridges] in Airplane, who comments that the dire stress of the day made it “the wrong week to stop smoking” (as he lights up). The joke comes when he reprises “the wrong week to stop” about sniffing glue and taking speed. It is funny because the first iteration, about smoking, is reasonable. If the cost of functioning better during an acute crisis is to delay smoking cessation for another week, then yeah, do that. But the idea of backsliding about stopping a devastating and (genuinely) addictive “hard drug” — one which is probably not exactly going to improve one’s crisis management like smoking would — is not so reasonable.
Of course, it is not just a matter of day-to-day welfare. Someone might like (on net) the daily effects of smoking while not liking the net lifetime effects (the latter, of course, include end-of-life diseases). This creates a challenging tradeoff, one that is absent for coffee: trading off short-term net benefits versus long-term costs. It is important to keep in mind that there is nothing at all unusual about this challenge. Many or most efforts to get people to change their behavior “for their own good” consists of trying to get them to do something that will make them better off over their lifecycle, even though it makes them worse off today. It is the whole “invest rather than consuming everything now” thing. (Indeed, there is a good case to be made that the only legitimate “for their own good” efforts are described by this, and the rest are just someone trying to impose their own preferences on others. But that is a topic for another day.) This is a challenge, but once again it has nothing to do with dependence.
No doubt some of you are now thinking, “but most smokers want to quit.” Really? Do they? That popular trope contains its own disproof: they smoke. All they have to do to quit is to not smoke; therefore, they pretty clearly do not want to quit. You might want to run a four-hour marathon, but are physically incapable no matter how hard you try. Avoiding smoking is not like that. It involves only the physical act of not lighting up, which is possible for anyone who chooses it.
That common utterance is not without meaning, however. (Well, sometimes it is without meaning. It is based on survey responses, many of which are genuinely meaningless –they are just the respondent dutifully reciting the only socially-acceptable answer to the question.) The trope means something, just not what it says. I believe that most often it means that someone wants to want to quit, even though they do not want to quit. (A “second-order preference” — for more, see this paper.) That story is clarified with the more intuitive observation, “they want to get the net benefits from smoking this year, but do not want the net costs from a lifetime of smoking; they thus would prefer to not have such a short-term preference, but they cannot override it despite recognizing it is contrary to their long-term interests.” Once again, however this has nothing to do with dependence.
I have observed at length (including in the previous link) that substitution of low-risk products is the perfect solution to exactly these challenges. If someone wants to want to quit, we should help them out by offering a substitute which keeps enough of the benefits that they really do want to quit. Their preferences then align with their second-order preferences, and also align with what they are “supposed” to do. Win-win-win. Happiness and health.
Pharmaceutical nicotine (NRT) works best for smoking cessation when someone discovers that they are satisfied using it every day as a substitute for smoking. It is approximately useless when used “on label” to merely overcome dependence, because dependence was never the real challenge. Long-term use can align someone’s preferences. Of course, using better products — vaping or snusing — works far better still.
Ah, but that is not “real” quitting because someone is still “addicted” or dependent. And herein we are back to the self-defeating danger of adopting the language of one’s oppressors. When you adopt their characterizations of the situation, even if for purposes of arguing about it, you have already lost. Falling into this trap is easy to do, and having done it it is difficult to avoid internalizing that you accept key premises of your own oppression. It happens to slaves, prisoners, hostages, the impoverished, and the otherwise discriminated against. It has happened to smokers over decades. And I have watched it happen to many vaping advocates quite rapidly. The concession of the absurd claim that smoking is somehow bad and/or happens merely because of dependence, and thus it is not mostly about normal preferences, concedes the legitimacy of a potent but absurd line of attack on vaping.
A common retort to the “not real quitting…you are still addicted/dependent” attack is that it is better to remain dependent than suffer the health effects of smoking. There is also the ongoing debate about the relative dependence-producing properties of different products. Such rhetoric effectively concedes that (a) dependence on nicotine or tobacco products is a problem in itself and (b) dependence is genuinely a substantial motivator in this context.
Point (b) is clearly false, as I have noted. I hope I have also already made the point, regarding (a), that facing some cost of withdrawal from a behavior is not necessarily a big deal, and certainly not an evil. It describes caffeine and countless other socially-accepted choices. If we extend beyond the quasi-physical aspects of dependence, it describes most everything that we like to do. If suffering a bit for giving something up is a problem that warrants condemning the behavior, then all pleasure is bad. You miss your kids when they are away. You are distracted by daydreaming about last week’s vacation. Unless we are talking about truly dire symptoms, “dependence” is just name-calling.
Moreover, the concession further concedes that the greater the benefit a behavioral choice (and thus the greater loss from giving it up), the worse the behavior must be.
I was inspired to write this by a tweet in which someone asserted that vapers are less dependent than smokers because e-cigarettes do not deliver the MAOIs — basically antidepressants — that are delivered by smoking (or snusing). This was presented as if it were an advantage of vaping. But if one of the benefits of smoking for someone is the MAOIs (or any of the other psychoactive chemicals that are absent from vaping, which are usually ignored because only nicotine and the MAOIs are well-understood), then losing them by switching to vaping is a cost, not a benefit. That does not mean that switching is a bad idea on net, of course. But losing this benefit is a downside, not an advantage. If someone is much more willing to give up vaping than they were to give up smoking, that could mean that there was something uniquely evil about smoking. But the more likely story (given that the supposed diabolical force remains unidentified) is that there was something better about smoking.
Having one or more advantages over vaping does not mean that smoking is a better choice on net, of course. But we should not allow what is good to be spun as a cost. Recall the Tolstoi-versus-puritans contrast. People smoke because they like to smoke. People vape because they like to vape. People are faux-“addicted” because they prefer to consume rather than not consume. When “dependence” is used to invoke a specter, but is really just used to mean “someone likes this aspect of the consumption choice and so is reluctant to give it up”, it is a backdoor way of denying that preferences and welfare matter. It implies that enjoying a behavior is bad, at least if some puritanical minority objects to the behavior based on their personal pique. Conforming to their language usage helps them cement that implication.
[Endnote: I would be remiss if I did not acknowledge that a very small number of tobacco controllers, those who consider themselves more sophisticated, argue that the dependence in question runs deeper. Note that this has nothing to do with the dominant rhetoric, and so does not change any of the above. But they argue that past nicotine use creates a deficit in terms the very benefits that it delivers, and withdrawal thus includes not only acute symptoms but being stuck with that deficit. That is, the claim is that nicotine product users experience improved focus, cognitive performance, calmness, etc. from consuming, but most or all of that is just getting them back to the level they would have been at had they never smoked or otherwise used nicotine products. They are running fast just to stand still. The claim here is that nicotine dependence is like someone’s dependence on a prosthetic foot: not only does withdrawal include immediate problems, but it also includes permanent disability because the product was filling in for an acquired loss of normal functioning.
I will assert, without going into detail here, that I have looked at what they cite as evidence for this “it is all just running to stand still” claim, and find it completely unconvincing. (You might not be surprised to learn that a lot of the problem is tobacco controllers failing to consider confounding.) Still, at least this assertion makes a defensible case that the “dependence” in question is genuinely something to really worry about, unlike the trivial effects the word usually refers to in this context. However, even if this were true, it could argue in favor only of not starting nicotine product use (or rather, it might argue in favor — there are tradeoffs). Once someone has already built up this supposed deficit, it would be a reason to keep smoking or using a substitute product, and to never become abstinent.]