The blog post today is by Arthur Krieger (Temple University) on his recent paper "Torturous withdrawal: Emotional compulsion in addiction" (European Journal of Philosophy, 2024).
![]() |
Arthur Krieger |
We’ve all heard addiction described as a “compulsion.” People with addictions (here I’ll just say ‘addicts’) often act in ways that are so out of character, so self-destructive, and so contrary to their own stated aims, that we feel the need for a special explanation. Compulsion is the main concept used to explain addictive behavior, particularly in health sciences like medicine and psychology. What exactly is compulsion? That’s a surprisingly difficult question. It is sometimes glossed as continuing to do something despite being aware of its serious negative consequences. But that includes far too much, like using a chair instead of a standing desk at work, or eating an eclair every night despite a recent high cholesterol reading. One traditional view is that compulsion is a consistent inability to “do otherwise,” perhaps due to frequent “irresistible” desires.
The last few decades have seen opposition to the standard health-scientific view that addiction is a compulsion. There are several reasons for this resistance, but first and foremost, addicts often can and do abstain from the kind of behavior that sometimes seems compulsive. Furthermore, while many addicts want not to engage in addictive behavior and intend to abstain from it, it is also true that when they do engage in addictive behavior, it is generally because they want and intend to. Addicts may have conflicting or fluctuating attitudes, but isn’t that often true of non-addicts as well, if to a lesser degree? Framed this way, addictive behavior looks pretty normal after all, and the concept of compulsion can seem unhelpful for understanding it.
Writers who argue against the compulsivity of addiction tend to underrate a major reason for thinking that addiction is compulsive: withdrawal. Opioid withdrawal, for instance, is so miserable that it drives many addicts back to using even when they badly want to abstain. “Anti-compulsion” authors have a couple common responses to this kind of point. First of all, even the worst opioid withdrawals have basically the same symptoms as the common flu—uncomfortable, yes, but hardly so bad as to be compelling. Secondly, many addictions do not have withdrawals with dramatic physiological symptoms. Severe cocaine withdrawals are not physically painful, and you might not think that sex addiction has any withdrawal at all.
One thing to say here is that withdrawals from opioids, xylazine, benzodiazepines, alcohol, and some other drugs, really are excruciating, and completely incomparable to familiar ailments like the flu. William S. Burroughs tells us so much in his first novel, Junky: “You can list the symptoms of junk sickness [i.e. heroin withdrawal], but the feel of it is like no other feeling and you cannot put it into words.” More importantly, in my view, is the fact that the pain of withdrawal is not always somatic or “physical.” All kinds of addiction, whether it be to heroin, cocaine, or sex, involve extremely painful emotional withdrawals: the intense anxiety or agitation of feeling that a fundamental need is going unmet. This affective aspect of withdrawal is increasingly appreciated in addiction neuroscience, and is central to some prominent theories of addiction.
Strikingly, like addiction itself, extremely intense negative emotions are also widely thought to be compelling, including by some influential philosophers. Think of typical “fight, flight or freeze” reactions: under the sway of powerful emotions like anger and fear we can appear to lose control, acting spontaneously, unreflectively, irrationally, and even without being able to recall what we had done (or why). In my paper “Torturous withdrawal: Emotional compulsion in addiction” (European Journal of Philosophy, 2024, doi:10.1111/ejop.12967), I argue that some addictive behavior is emotionally compulsive, in the sense that the intense negative emotion constitutive of withdrawal sometimes makes it psychologically impossible to adhere to the intention to abstain. We already knew from extensive empirical research that withdrawal involves intense negative emotion; that much addictive behavior is motivated by this negative emotion; and that negative emotion makes self-control more difficult. But the possibility of emotional compulsion in addiction has been almost completely ignored in the philosophical literature. After examining philosophical work on emotional compulsion, I found that we lacked the theoretical resources to talk about how intense anxiety might make the selection of an alternative course of action psychologically impossible, in the distinctive way addiction seems to. I therefore developed my own account of what I call “pathodoxastic compulsion.” In pathodoxastic compulsion, intense or consistent emotional pain (“patho-”) can cause us to believe (“-doxastic”) that we are no longer able to try to endure it (i.e. the pain). And since the intention to abstain requires this belief, we become psychologically incapable of maintaining the intention to abstain. This makes us incapable of intentional abstinence. A core feature of this account is that negative emotion can cause the formation of beliefs and intentions in a way that is at least largely non-rational, an important (and mysterious!) fact about human psychology that deserves more attention from theorists and empirical investigators.
“Torturous withdrawal” is not just about pathodoxastic compulsion in addiction, but how we theorize agency in addiction more broadly. When we focus on the familiar idea that addiction is compulsive in large part because of withdrawal, we see that, contrary to what philosophers often assume, addiction doesn’t have to involve consistently out-of-control behavior to be properly considered a compulsion. The fact that a lot of addictive behavior is freely chosen doesn’t mean that addicts can just stop whenever they want. In another paper—“The Epistemic Prerequisites of Reliable Abstinence in Addiction” (Philosophy, Psychiatry, & Psychology, forthcoming)—I argue that addiction is compulsive because it undermines long-term resolution adherence. This is not primarily due to withdrawal, but because many addictive-decision making processes are highly automated. But the fact that addiction makes us prone to individually compulsive behaviors under certain circumstances—like withdrawal and cue-exposure—means that addicts are largely out of control of the larger contours of their lives, until they are quite deep into recovery. While addiction has intrinsic philosophical interest, I hope that pieces like “Torturous withdrawal” encourage theorists in many disciplines, including philosophy, to do more work on what it means to be “in control of one’s life,” to whatever extent that is possible.