Addicted to love: What is love addiction and when should it be treated?

Abstract

Recent research suggests that romantic love can be literally addictive. Although the exact nature of the relationship between love and addiction has been described in inconsistent terms throughout the literature, we offer a framework that distinguishes between a narrow view and a broad view of love addiction. The narrow view counts only the most extreme, harmful forms of love or love-related behaviors as being potentially addictive in nature. The broad view, by contrast, counts even basic social attachment as being on a spectrum of addictive motivations, underwritten by similar neurochemical processes as more conventional addictions. We argue that on either understanding of love-as-addiction, treatment decisions should hinge on considerations of harm and well-being rather than on definitions of disease. Implications for the ethical use of anti-love biotechnology are considered.

“By nature we are all addicted to love … meaning we want it, seek it and have a hard time not thinking about it. We need attachment to survive and we instinctively seek connection, especially romantic connection. [But] there is nothing dysfunctional about wanting love.”

– Smith, quoted in Berry (2013)

Introduction

Throughout the ages love has been rendered as an excruciating passion. Ovid was the first to proclaim: “I can’t live with or without you” (Amores III, xi, 39)—a locution made famous to modern ears by the Irish band U2. Contemporary film expresses a similar sentiment: as Jake Gyllenhaal’s character famously says in Brokeback Mountain, “I wish I knew how to quit you.” And everyday speech, too, is rife with such expressions as “I need you” and “I’m addicted to you.” These widely-used phrases capture what many people know first-hand: that when we are in love, we feel an overwhelmingly strong attraction to another person—one that is persistent, urgent, and hard to ignore.

Love can be thrilling, but it can also be perilous. When our feelings are returned, we might feel euphoric. Other times, love’s pull is so strong that we might follow it even to the point of hardship or personal ruin (Earp, Wudarczyk, Sandberg, and Savulescu 2013). Lovers can become distracted, unreliable, unreasonable, or even unfaithful. In the worst case, they can become deadly. In 2011, over 10% of murders in the United States were committed by the victim’s lover (FBI 2011). When relationships come to an unwanted end, we feel pain, grief, and loss. We may even become depressed, or withdrawn from society (Mearns 1991).

These phenomena—including cycles of alternating ecstasy and despair, desperate longing, and the extreme and sometimes damaging thoughts and behaviors that can follow from love’s loss—bear a resemblance to analogous phenomena associated with more “conventional” addictions like those for drugs, alcohol, or gambling. Nevertheless, although we do sometimes use the language of addiction when referring to love, there is at least one major feature that distinguishes love from the kinds of substance-based addictions typically described in the psychological and medical literature: nearly everyone aspires to fall in love at least once in their life. By contrast, nobody yearns to become addicted to heroin (for example), or cigarettes, or slot machines. So it might seem absurd on its face to suggest that there could be a real similarity between lovers and “genuine” addicts. Surely it is all just hyperbole and poetic license?

Addicted to love

Perhaps not. So numerous are the superficial similarities between addictive substance use and love- and sex-based interpersonal attachments, from exhilaration, ecstasy, and craving, to irregular physiological responses and obsessive patterns of thought, that a number of scientific theorists have begun to argue that both sorts of phenomena may rely upon similar or even identical psychological, chemical, and neuroanatomical substrates (e.g., Insel 2003; Fisher, Brown, Aron, Strong, and Mashek 2010; Burkett and Young 2012).¹

The past decade has seen a dramatic increase in published studies on the neurobiology and neurochemistry of romantic love (for overviews, see Savulescu and Sandberg 2008; Earp, Sandberg, and Savulescu 2012). Taken together, these studies suggest that the subjective state (or states) of “being in love” is intimately tied to characteristic biochemical reactions occurring within the brain. These reactions involve such compounds as dopamine, oxytocin, vasopressin, and serotonin and recruit brain regions known to play a role in the development of trust, the creation of feelings of pleasure, and the signalling of reward (Esch and Stefano 2005). The involvement of similar neurochemicals and neural activities in processes associated with addiction has already been well established (Blum, Chen, et al. 2012). Consequently, scientists have begun to draw a number of parallels between the naturally rewarding phenomena associated with human love and the artificial stimulation afforded by the use of addictive substances such as alcohol, heroin, or cocaine (see Frascella et al. 2010).

While the specific nature of these parallels has been described in inconsistent language throughout the literature, two main approaches to conceptualizing the relationship between love and addiction can be usefully teased out. The first approach counts only the most extreme cases (or phases) of love and love-related behaviors as being potential instances of addiction. Research in this vein focuses on sexual compulsions, paedophilia, toxic or abusive relationships, abnormal attachments and unhealthy tolerance of negative life- and relationship outcomes (e.g., Carnes 2005; Reynaud et al. 2010).

The second approach takes a wider view, and counts even “normal” romantic passions as being chemically and behaviorally analogous to addiction (e.g., Fisher et al. 2010; Burkett and Young, 2012). Studies in this vein emphasize the commonality between the experience of someone under the influence of certain drugs and the quite ordinary experience of someone in love—including her “focused attention” on a preferred individual, “mood swings, craving, obsession, compulsion, distortion of reality, emotional dependence, personality changes, risk-taking, and loss of self-control” (Fisher et al. 2010, 51). Burkett and Young (2012, 1) go so far as to defend the hypothesis that basic social attachment – covering the whole course of love-based relationships from initiation to break-up – may be understood as a form of behavioral addiction “whereby the subject becomes addicted to another individual and the cues that predict social reward.”

In the following sections, we highlight some of the latest thinking on the nature of romantic love considered as an addiction, drawing on behavioral, neurophysiological, and neuroimaging studies of both love and addiction. By doing so, we hope to give a taste of, as well as to clarify, the existing evidence in favor of these differing accounts. Following that, we will attempt to explore some of the moral and practical implications that begin to emerge once we recognize that:

• (a)one can indeed become addicted to love, and
• (b)to be in love is in some sense to be addicted—i.e, to another person

Our main thesis will be that on either understanding of love-as-addiction, there is a reasonable case to be made that, in some instances, “treatment” of love could be justified or even desirable. We will also argue that respecting the lovers’ autonomy should be paramount in any treatment decision. Along the way, we will entertain some possible objections to our views, as well as offer our replies.

The narrow view: addiction as the result of abnormal brain processes

Although scholarly attitudes have been shifting in recent years, the dominant model of addictive drug use—among neuroscientists and psychiatrists, at least—is that drugs are addictive because they gradually elicit abnormal, unnatural patterns of function in the human brain (Foddy and Savulescu 2010). On this ‘narrow’ view of addiction, addictive behaviors are produced by brain processes that simply do not exist in the brains of non-addicted persons.²

One especially popular version of this view holds that drugs ‘co-opt’ neurotransmitters in the brain to create signals of reward that dwarf the strength of ‘natural rewards’ such as food or sex. They thereby produce patterns of learning and cellular adaptation in the brain that could never be produced without drugs (e.g. Volkow et al. 2010). According to this strict account, then, addictive drug-seeking is an aberrant form of behavior that is peculiar to drug addicts, both in form and in underlying function. It follows that natural rewards like food and love can never be truly addictive, and that food-seeking or love-seeking behaviors are not truly the result of addiction, no matter how addiction-like they may outwardly appear.

Other researchers, however, have noted appreciable behavioral similarities between binge-eaters (for example) and drug users, and have flagged a growing body of evidence that is suggestive of neurological similarities as well (Foddy 2011). Sweet food, to take just one example, can elicit a reward signal in the brain as strong as the reward from a typical dose of cocaine (Lenoir et al. 2007). In addition, it can even induce—at least in rats—a withdrawal syndrome as strong as that induced by heroin (Avena et al. 2007). If an illicit drug like cocaine, therefore, can produce ‘abnormal’ brain processes by providing abnormal and chronic reward, then so might an abnormally high natural reward, like the reward one gets from bingeing on food, or from experiencing unusually strong or frequent feelings of love. Given these considerations, a more plausible ‘narrow’ view of love addiction would hold that one can indeed be addicted to love, but only if these abnormal brain processes are present.

Indeed, a number of addiction theorists have argued that otherwise harmless love-related phenomena can qualify as addictive if they take on such an ‘extreme’ or maladaptive form—termed “destructive” love by Fisher (2004), and “unwise” or “desperate” love by Peele and Brodsky (1975). One way to begin to understand love-related behaviors of this “destructive” type is to use the framework of process addiction (Sussman 2010; Timmereck 1990). Process addiction—as opposed to substance addiction—typically refers to an obsession with certain activities such as sex, spending money, eating, or gambling. When a person in love repeatedly seeks contact with another individual—for physical intimacy, attention, or merely to be in the same room—it is often to secure momentary feelings of intense pleasure and to relieve obsessive thought patterns about the object of her passion. If this sort of behavior threatens the individual’s (or another’s) safety, mental or physical health, or incurs serious social or legal costs, it may rise to the level of an addiction (e.g., Sussman 2010).

A further distinction has been drawn by Sussman (2010), following Curtis (1983), between mature love and immature love. Sussman suggests that only the latter may be considered a form of addiction. Rather than permitting mutual growth within the partnership, or contributing to increased self-esteem and well-being in both individuals, immature love is typified by power games, possessive thoughts and behaviors, obsessive concern over the partner’s fidelity, “clinging” tendencies, uncertainty, and anxiety. Love-addicts on this model “feel desperate and alone when not in a relationship,” “continue trying to romance the love object long after the relationship has broken up,” and “replace ended relationships immediately” despite such declarations as “I’ll never love again” (Sussman 2010, 34).

To summarize, a lover might be suffering from a type of addiction (on this narrow view) if she expresses one of a number of abnormal sexual or attachment behaviors—perhaps underwritten by similarly abnormal brain processes—such that her quest for love (1) interferes with her ability to participate in the ordinary functions of everyday life, (2) disables her from experiencing healthy relationships, or (3) carries other clear negative consequences for herself or others. In the case of more ordinary examples of love—i.e., the ones to which most people probably aspire—these feelings, behaviors, and ill consequences are not present, or are present only to a mild or manageable degree.

The narrow view of love addiction is narrow, then, in the sense that it sees only extreme, radical brain processes, attachment behaviors, or manifestations of love as being potentially indicative of addiction—and hence it is thought to be quite rare. For example, Timmereck (1990) has estimated that love addiction of this type may affect between 5-10% of the U.S population. By contrast, “healthy” romantic love, which is assumed to be much more common, is described by scholars such as Sussman (2010) as being benign or even beneficial. Such love is said to have evolved, for example, for adaptive (and still-useful) ends, such as the promotion of procreative behaviors and the facilitation of cognitive and social learning. Reynaud et al. (2010, 262) distinguish between love addiction and mere “love passion” which they describe as “a universal and necessary state for human beings.” And Peele and Brodsky (1975) refer to “genuine” love, which, unlike the self-seeking dependency associated with addictive love, involves a commitment to mutual growth and fulfillment between the partners involved.

As we explore in the following section, however, other researchers, notably Burkett and Young (2012), have begun to highlight the similarities between addiction and even “normal” romantic relationships by emphasizing the common behavioral, neurophysiological, and neurochemical signatures of both.

The broad view: love as addiction

There is a broader understanding of addiction that has been gaining steam in recent years. As two of us have argued, addiction should be considered to be a spectrum of motivation that emerges from the repeated application of any type of reward, including drug rewards, gambling rewards, food rewards, and sexual rewards (Foddy and Savulescu 2006, 2010; Foddy 2011). These appetite-motivations arise in response to reward conditioning, and are, indeed, the evolved mechanism by which we humans and other animals learn to behave in survival- and reproduction-enhancing ways. At the same time, such appetites do not always lead directly to these “evolutionary” outcomes, especially in humans, and even more so in the modern environment we have created for ourselves (see Earp, Sandberg, and Savulescu 2012). Our appetite for food, for example, is not strictly genetically controlled: we are weaned onto it during gestation, and it can wax and wane over the course of our lives, often in ways that run contrary to our real nutritive needs (Foddy 2011). By the same token, we may develop appetites for any rewarding behavior, and these appetites may exceed or fall beneath a level that suits our biological needs, our conscious values, or our personal preferences.

On a broad view, then, addictions are simply appetites: they are felt needs that can be temporarily satisfied, but which become urgent and distracting if one abstains from fulfilling them for too long. Conversely, and perhaps counterintuitively, appetites are simply weak addictions. At least on this account, then, everyone is ‘addicted’ to food, to sex, and to other ordinary substances and behaviors, although most of us are not ‘hooked’ on them to such a degree as to cause us any major harm or distress or to merit the application of a psychiatric diagnosis or treatment (see Foddy and Savulescu 2010, for a sustained defence of this view).

A similar broad approach can be applied to the concept of love addiction. This approach would claim that to love someone is literally to be addicted to them, though perhaps only weakly. In line with this view, James Burkett and Larry Young (2012) have recently argued that romantic relationships experienced universally—from “falling in love” to ultimate separation and subsequent withdrawal—may be considered a form of addiction. To prime the reader for their thesis, they open their seminal paper on this subject with the following vignette:

At first, each encounter was accompanied by a rush of euphoria—new experiences, new pleasures, each more exciting than the last. Every detail became associated with those intense feelings: places, times, objects, faces. Other interests suddenly became less important as more time was spent pursuing the next joyful encounter. Gradually, the euphoria during these encounters waned, replaced imperceptibly by feelings of contentment, calm, and happiness. The moments between encounters seemed to grow longer, even as they stayed the same, and separation came to be filled with painful longing and desire. When everything was brought to an abrupt end, desperation and grief followed, leading slowly into depression. (Burkett and Young 2012, 1)

Does this story describe falling in love or becoming addicted to a drug? Burkett and Young’s point, of course, is that it could equally describe both. Drawing on evidence from animal models along with parallel research in human attachment and the neurobiology of substance abuse, they conclude that there is “a deep and systematic concordance … between the brain regions and neurochemicals involved in both addiction and social attachment” (Burkett and Young 2012, 2).

In other words, substance dependence and everyday romantic bonding have much more in common than their outward psychological profiles. At the level of the brain, the mechanisms underlying pair-bonding in socially monogamous or quasi-monogamous species (such as humans) overlap quite substantially with those involved in reward learning and addiction (see, e.g., Wise 1996). The greatest overlap occurs in neurochemical regions involved in the processing of dopamine (Kelley and Berridge 2002; Burkett and Young 2012) oxytocin (Insel 2003; McGregor, Callagham and Hunt 2008), and other neurotransmitters such as serotonin. As Margolis (2005, x) states: “… through sex [with our partner], orgasm’s serotonin rush and momentary muscular relaxation comprise the most potent and popular drug we have.”

With respect to dopamine, both mating and addictions elicit very similar neurochemical activity, concentrated in the reward circuitry of the brain: sex, orgasm, and all known drugs of abuse stimulate high levels dopamine release in the nucleus accumbens (see Burkett and Young 2012, Kirsch et al. 2006, and Di Chiara et al. 2004 for more information). In fact, the role of dopamine extends far beyond addiction and is linked to a wide range of other processes associated with reward-learning—including eating, drinking, having sexual intercourse, and love (see Burkett and Young 2012, for a review). Some scientists have suggested that this dopaminergic overlap may explain why experiencing love or engaging in sexual activity can feel like a cocaine rush (Blum, Wernel, et al. 2012; Bartels and Zeki 2000).

Finally, neuroimaging support for an overlap between love-addiction and drug-addiction comes from studies in which participants have been exposed to images of their romantic partners during scanning. These images evoke not only self-reported feelings of love and positive affect but also show heavy activation in brain’s reward regions (Aron et al. 2005; Bartels and Zeki 2000; Young 2009; Fisher et al. 2006).

These are just a few of the neurochemical and functional similarities between “normal” love attachment and drug addiction that have been noted by Burkett and Young as well as by other researchers who argue for the “broader” view of love addiction. There are many others as well. But what about differences between love and addiction? Surely the numerous “parallels” between these phenomena—at both the behavioral and neural levels—should not be taken to mean that they are strictly equivalent. In the following section, we examine some of these differences, and assess their implications for the “broad” vs. “narrow” debate.

What about differences?

Although there is a significant amount of “concordance” between love and addiction on several levels of analysis, there are some important differences between naturally rewarding activities (such as those associated with being in love) and the artificial stimulation of the reward system that occurs during drug use. As we will see, however, these differences are less significant—in terms of establishing a distinction in kind between love-based and drug-based addiction—than they may seem at first glance.

One such difference, at the level of the brain, concerns the duration of the effects of the stimulus. The release of signalling molecules in case of love-related experiences—such as sexual intercourse—may not be as long lasting as the analogous release stimulated by the use of addictive drugs (Esch and Stefano 2005). Natural rewards are highly controlled by feedback mechanisms, which may lead to a quicker return to “baseline” by promoting (for example) an aversion to engaging in the same rewarding activity immediately after the pleasant stimulation (Small, Zatorre, Dagher, Evans, and Jones-Gotman 2001). For example, the building-up of sexual desire often precedes a sexual act (e.g., Heaton and Adams 2003), after which it quickly decreases and then takes time to recover its initial level of intensity. Addictive stimulants, on the other hand, often rebuild high levels of desire immediately after drug consumption (e.g., Nestler 2005). From this it is often argued that drug-seeking activities override other motivations more easily than activities associated with the pursuit of “natural” rewards (Esch and Stefano 2004).

Even granting this view, however, it must be remembered that the neurodynamics of even “conventional” addictive drugs are not all the same, since their mechanisms of action vary so widely (Nestler 2005). Non-stimulant drugs such as alcohol, opiates, and benzodiazepines, for example, are not as quick to rekindle cravings as stimulants are, and they can even induce an appetitive regret that is similar to the regret lovers experience as “post-coital malaise” (cf. Le Moal and Koob 2007; Norville, Sweeney, and Elliott 2010). More importantly, the feedback mechanisms that control the processing of “natural” rewards are not always reliable, nor are the neurological signals from natural rewards always weak. Some 3% of American adults suffer from binge-eating disorder, in which their complex satiety system, which should alert them to the state of being full, is so dysregulated that these individuals can regularly eat enormous quantities of food in a fevered “binge’ (Foddy 2011). When these same individuals binge on sugary food, they can experience a level of neurological reward that surpasses a dose of a drug such as cocaine (Lenoir et al. 2007). The evidence, therefore, suggesting that drugs of abuse are inherently better-suited to causing addiction than are other types of reward is mixed at best.

How else might “being in love” be taken to differ from being addicted to certain drugs? For one thing, while drug addiction is a circumscribed problem, affecting a mere fraction of the global population, romantic love is a universal phenomenon, emerging from basic, evolved sub-systems that helped our ancestors pursue mating opportunities with preferred partners (see Savulescu and Sandberg 2008). In other words, love is deeply bound up with reproduction, which is the engine of natural selection (Blum, Chen, et al. 2012). Drug use, by contrast, serves no clear function with respect to survival or reproduction, and it has often been described as ‘hijacking’ mechanisms underpinning the natural reward functions of the brain, supplanting more ‘adaptive’ behaviors with entirely artificial ones that are irrelevant to genetic fitness (Hyman 2007). Taking a complementary view, Fisher (2004) has argued that romantic love might be considered a ‘constructive’ form of addiction when one’s love is reciprocated; whereas it is presumed that drug addiction would never be constructive. And while we can flourish without ever taking drugs, we cannot do so if deprived of food, or, indeed, some measure of love and human affection.

However, these considerations do not entail that love addiction, food addiction, and drug addiction are different in kind. Binge-eating is in fact extremely detrimental to one’s chances for long-term survival, just as extreme forms of love-related phenomena—such as an insatiable appetite for sex—can lead to such negative outcomes as interpersonal problems, sexually transmitted diseases, loss of employment, or even imprisonment (Carnes 2005). No extreme pattern of reward-seeking, whether it is for food, sex, romantic love, or drugs, is likely to promote a person’s chances of flourishing in the modern world—and any normal capacity can be taken to extremes (see Earp, Sandberg, Kahane, and Savulescu 2014).

Moreover, although a moderate appetite for so-called ‘natural’ rewards is quite plainly beneficial, it would not be unreasonable to make a similar claim regarding certain drugs to which people can sometimes become addicted—at least if we view pleasure as being one important source of the good. Nobody strictly needs drugs to flourish, but in some circumstances, and for some people, some non-therapeutic drugs could certainly be considered compatible with human flourishing if taken within reason, such as the moderate consumption of alcohol.

Finally, there is the fact that love must be requited for it to deliver its full benefit. Lovers rarely regret being in love if the love is returned, and indeed a significant part of the suffering associated with falling in love stems from romantic rejection, or withdrawal of love—rather than from love itself. Substance addicts, by contrast, are never rejected by their drug in the same way, so perhaps there is a difference in this regard as well. Even so, there are some potential parallels on this dimension that complicate the notion that love addiction and drug addiction are totally incommensurable phenomena.

For example, consider the fact that the difficulty of obtaining drugs can be the source of many (if not most) of the problems that drug addicts experience. Access to drugs is constrained by scarcity, cost, illegality, and social stigmatization in addition to any biological (side) effects. Some of the destructiveness of drug addiction occurs because an addict is unable to afford her drug and turns to crime; because her friends and family reject her for taking a drug; or because she cannot obtain her drug and goes into withdrawal. The addiction itself is not clearly the cause of most of the associated harms. Therefore, we can draw at least a tenuous analogy here, too, between a lover who is rejected and a drug user who cannot access her drug of addiction.

Taken together, these considerations show that while there are indeed some differences between love-based and drug-based addictions, these differences may have more to do with the frequency of problems at the population level, or with the typical degree of reward-stimulation involved in particular cases, than with any clear incommensurability in kind.³

Broad or Narrow?

The behavioral, psychological, and neurophysiological evidence concerning love, love-related phenomena, drugs of addiction, and the parallels between them, paint a very complicated and hotly-debated picture. Just within the drug addiction literature—holding “love addiction” to the side—there is little agreement about whether the “narrow view” or the “broad view” of addiction is to be preferred. In other words, there is (as yet) no meaningful consensus about whether it is the case that addictive drug-related desires are (a) of an unnatural form that is not present in the non-addict (the narrow view); or (b) whether they are on a continuum with “regular” appetitive desires, and therefore simply a strong form of the same underlying phenomenon (the broad view). The available neuroscientific and behavioral evidence simply cannot settle the question firmly one way or the other.

The same is true regarding “love addiction.” Should only the most extreme, compulsive, or harmful love-related phenomena be considered potential forms of addiction? Or, as Burkett and Young (2012) have recently suggested, might it be the case that there is a neurological equivalence between quite ordinary experiences associated with falling in love—and even basic social attachment itself—and addictions of a more conventional stripe? Much more work is needed, we suggest, both empirical and conceptual, to arbitrate between these differing views.

At a minimum, however, the evidence we have discussed in this article suggests that drug addiction, on the one hand, and at least certain love-related experiences or behaviors, on the other, can reasonably be understood to be equivalent phenomena at the level of the brain, underwritten by the same neurophysiological processes. If this is correct, it cannot be the case that (narrow) addiction is a phenomenon confined to addictive drugs. Holders of the narrow view must instead claim that addiction is a term we can apply to any person who has undergone chronic and unusually strong (pleasurable) experiences—intense drug-reward in the case of drug addicts, food binges in the case of binge-eaters, or powerful personal attachments in the case of love addicts.

Based on this premise, in the next section, we will explore some of the practical and ethical implications that arise when we acknowledge that love has (or can have) addictive characteristics, at least on the narrow view, but also possibly on the broad view as well.

Ethical implications: two mysteries

The science and philosophy of addiction—generally speaking—have sought to solve two basic mysteries. The solution to these mysteries, in the case of love-addiction as much as for any other purported form of addiction, will have important practical and ethical implications.

First, we have sought to learn whether, or to what degree, those who suffer from addictions are capable of abstaining from or moderating their reward-seeking behavior; and second, we have been very interested to know whether—and how—we can help people to recover from addiction using various treatment modalities. We will consider these “mysteries” in turn as they relate to the notion of love addiction in the following sections.

Autonomy and responsibility

The first longstanding mystery concerning addiction in general is whether addicts are capable of abstaining from or moderating their drug use or other problematic behaviors. The solution to this mystery would have some indirect implications for the medical treatment of addicts, but it has much stronger philosophical and ethical implications: for example, is it reasonable to force addicts to abstain? It is reasonable to hold them morally or criminally responsible for their drug-taking? What about for illegal actions they committed in pursuit of their drug? These deep empirical and conceptual problems date back at least as far as the time of Plato, who wondered how a responsible person could continue to choose courses of action that she would predictably come to regret.

Naturally, this issue does not apply to love in the same way as it does to substance addiction or to other problematic behaviors. We do not ordinarily choose to love someone (at least not consciously) and it would be a hard thesis to defend that we should be held responsible for falling in love—even though such an occurrence can have very far-reaching, and sometimes destructive, consequences for those involved. We cannot criminalize falling in love, and though history is replete with stories of people having been punished for falling in love with the “wrong” person, such punishment would seem cruel or even absurd by today’s standards.⁴

However, while the formation of love seems to be at least largely involuntary, there is a real question regarding how autonomous one’s behavior is once one is, in fact, in love. People who are in love make choices every day about how to express their feelings: whether to seek proximity with their loved one, or physical affection; whether to make expressions of their attachment a matter of public notice, and so on.

If addictive love is nothing more than a symptom of abnormal brain processes (i.e., the narrow view), then the choices and behaviors it elicits might be considered to be inauthentic or in some sense non-autonomous reflexes of those deviant processes. Hence, proponents of the narrow view of substance addiction have frequently argued that addicts lack control over their actions and are not fully autonomous (Hyman 2007). This idea is also reflected in the common concept of “crimes of passion”— and indeed the law has traditionally taken such passion into account in sentencing decisions.

If this is the correct view, then we might think that it would be reasonable to try to eliminate the problematic feelings and behaviors associated with addictive love, since they are merely the symptoms of disease. And we might even believe that we could be justified in using coercion or force to prevent a love addict from being near the object of her desire. An analogous view is held by scholars such as Caplan, who have argued that we are justified in overriding a drug addict’s choices in the name of restoring her autonomy (Caplan 2008). Indeed, we, too, have previously argued that in some domestic abuse situations, including Stockholm Syndrome-like cases of attachment between the victim and her abuser, coercive intervention may possibly be justifiable (Earp, Wudarczyk, Sandberg, and Savulescu 2013).

The broad view, by contrast, argues that even the strongest, most negative forms of love are merely extremes of an authentic emotion. Hence it is possible to argue, from these grounds, that even those people who suffer from harmful extremes of love may be fully autonomous in their behavior. On this kind of view, any possible treatment modality would then differ along certain dimensions. The goal should not be to eliminate the feelings of love entirely, since those feelings are authentic aspects of the person’s mind and personality; but rather it should be to moderate them instead. Likewise, treatments should never violate the autonomy of the person-in-love, nor should they involve coercion or force of any kind. Two of us have advanced a similar point of view regarding treatment for substance addiction (Foddy and Savulescu 2010).

Treatment and recovery

These considerations bring us to the second “mystery” concerning addiction, namely whether treatment is appropriate or advisable and whether recovery is possible. Although one would not normally think of offering “treatment” to individuals who are in love, once we begin to realize that at least some cases of love and love-related phenomena are similar to behavioral or substance addictions—in form, function, as well as in effect—then the possibility becomes worth taking seriously. Indeed, the idea of an anti-love remedy or a “cure” for love has deep historical roots. As we noted in a recent paper (Earp, Wudarczyk, Sandberg, and Savulescu 2013, 3): “references [to such ‘cures’] may be found in the writings of Lucretius, Ovid, Shakespeare, and many others, and are tightly linked to the notion that love or infatuation—under certain conditions—can be just like a serious illness: bad for one’s physical and mental health and, in some cases, profoundly damaging to one’s overall well-being.”

In other words, the possibility of treating love—given the right sort of circumstances—may not be so far-fetched after all. Importantly, the way in which we approach this possibility will depend upon whether we subscribe to the broad view or the narrow view of “love addiction” as outlined above. Yet as we show in what follows, the general conclusion of such an analysis may not differ very much in the end, regardless of the view one takes. That is, love addiction—however it is conceived—would seem to be an appropriate candidate for treatment in at least some circumstances. Or so we argue in this section.

Recall: the narrow view considers that love is only an addiction (or at least addiction-like) in rare or abnormal cases, generated either by pre-existing pathophysiology or by chronic ultra-strong “doses” of love-based reward. On this kind of view, love addiction is a neurobiological disorder that has no place in a healthy or flourishing life, and it follows that we ought to offer love addicts some measure of treatment or support. Furthermore, on the narrow view, we might expect that the best course of action is a standard psychiatric treatment modality, in which we try to restore normal neurophysiology using psychiatric drugs and/or cognitive therapy (see Earp, Sandberg and Savulescu 2015). For example, just as heroin addicts are sometimes given oral naltrexone to block the pharmacological effects of their drug, we could use oxytocin antagonists to reduce the reward an individual receives from being close to another person (see Earp, Wudarczyk, Sandberg, and Savulescu 2013).

According to the broad view, by contrast, everyone who loves is on a spectrum of addictive conditions: being addicted to another person is not an illness but simply the result of a fundamental human capacity that can sometimes be exercised to excess. On this kind of view, it can be objected that love—no matter how destructive—is never an appropriate object of psychiatric treatment. A similar notion is that we now know that homosexuality is within the range of normal human variation and is therefore not appropriately considered a “disease.”

Yet this is where the distinction between broad and narrow begins to break down. On both views, that is, the primary difficulty from an ethical point of view is to determine how we ought to distinguish the good kinds of love from the bad: the innocuous or even the beneficial from the dangerous and the harmful. As one of us has argued, in cases of mental illness it can be very difficult to sort pathology from normal function, especially in the grey areas between extremes (Savulescu 2009). Current theories of brain function are nowhere near advanced enough to tell us whether a particular individual’s emotions fall within statistical norms; and even if science could tell us such a thing, we would still need to make value judgments in order to define which part of the statistical curve is desirable. For example, we arbitrarily define anyone with an IQ two standard deviations below the mean as being intellectually disabled—and hence deserving of special treatment—but we could just as easily have put the bar at one standard deviation, or three.

Nearly every psychiatric disability or disease represents an extreme of a smooth spectrum of behavior, function, or capacity. Not all sad people are depressed, and not all energetic people are manic. Our definitions of psychiatric illness, therefore, are essentially holistic rather than purely naturalistic; they are based inevitably on judgments of value—or in other words, ethical judgments. At base, these judgements relate certain states of biology or psychology to well-being: when a state either constitutes or creates harm, it may come to be defined as a disease and thereby mark itself out as a candidate for treatment.⁵

Given this inherent value-ladenness, we suggest, in defining some condition as a disease or a disorder, we should consider a range of different outcomes that would result from the application of such a label. We must consider who we think should have excuses for their behavior; who should receive support from friends or from the state; who should be an object of medical research; and (above all) who should receive treatment. These outcomes must be assessed on a range of moral and practical grounds, such as the suffering experienced by the afflicted person; the degree to which her ‘addiction’ undermines her ability to act on her preferences or execute her plans; and the extent to which it compromises important social relationships, impedes the development of her talents, impoverishes her interaction with the world, and so on.

In line with this view, consider the definition of substance dependence from the latest available (at the time of writing) edition of the Diagnostic and Statistical Manual of Mental Disorders from the American Psychological Association, the DSM-IV: “When an individual persists in use of alcohol or other drugs despite problems related to use of the substance, substance dependence may be diagnosed” (American Psychiatric Association 2000, emphasis added). As this quotation makes plain, the existence of “problems”—a normative concept—is inherent in the definition of the disorder. In other words, it matters fundamentally whether harm, difficulty, or ill-consequences are associated with the reward-seeking behavior: the reward itself is not the problem. Furthermore, and consistent with our arguments in this paper, addiction theorist Stanton Peele (quoted in Curley 2010) has suggested that next edition of the DSM—volume 5—should include “life-harming, compulsive” involvement with sources of reward such as sex and food in the category of addiction. Here, too, the notion of “life-harm” is built into the very concept of the disorder in question.

What insight do these considerations bring to our understanding of love addiction and its potential for being an appropriate subject of treatment? The message by now should be clear: regardless of whether we understand the love-related phenomenon to be the result of abnormal brain processes, or simply the manifestation of a strong appetitive desire, the key determinant for labelling and treatment should be the degree to which the individual is harmed by the love through its deleterious impact on her well-being. In the next section, we examine this perspective in greater detail.

Love addiction and well-being

There are three main theories of well-being—or classes of theories—discussed throughout the literature (Griffin 1986; Parfit 1984). How one relates love addiction to well-being, and therefore to treatment, will depend upon the theory (or theories) of well-being one finds most convincing.

The first class of theories concerning well-being are hedonistic theories, which are defined in terms of mental states. The simplest account of this type of view is that happiness, or pleasure (understood broadly as a mental state) is the only intrinsic good, whereas unhappiness or pain is the only intrinsic bad. More complex hedonistic views include a greater plurality of states of mind as possibly contributing to (or constituting) well being: for example, Freud is reputed to have refused analgesia when dying of cancer—although he was in great physical pain—on the grounds that he preferred to be able to think clearly in a state of torment than foggedly in a state of drug-induced comfort (see Griffin 1986).

Yet however one construes⁶ this hedonistic mental-state view, it is clearly possible that a person could prefer to exist in a rapturous state of love, even though it might yield a number of adverse consequences in other areas of her life, due to its irrefutably high, intrinsic hedonic value. Indeed, in Western societies, being in love is widely considered to be an extremely valuable state, and possibly constitutive of a good life all on its own. This notion is captured in the ideal of “dying for love” with the implication that such a love might even be the very meaning of life.

The second class of theories concerning well-being are desire-fulfilment theories. According to these theories, well being consists in having one’s desires fulfilled. They give weight to individual values and yet they account for the plurality of values that might differ across individuals: economic theories commonly employ a related notion of value, and such accounts are widespread in philosophy and in the social sciences in general. On the most plausible desire-fulfilment theories, desires should be informed (of the relevant facts) as well as freely devised—that is, not forced upon one in any way—in order to count toward one’s well-being.

Could love addiction be consistent with well-being on this desire-fulfilment account? It does seem possible. Specifically, if a person desired, freely, to exist in a state of extreme passion—even granting certain negative outcomes—while nevertheless being fully informed about the nature of addictive love, how it might affect her brain (and so on), then love addiction and well-being might indeed coincide in such a person. On this view, treatment might not be appropriate, even though negative consequences were present.

Finally, there are objective list theories of well-being. According to these theories (sometimes called substantive good or perfectionistic theories), certain things can be good or bad for a person—and thus contribute to her well being—whether or not they are desired and whether or not they lead to a “pleasurable” mental state. Examples of the sorts of things that have been proposed as being intrinsically good in this way are: gaining knowledge, having deep personal relationships (including being in love), engaging in rational activity, and developing one’s abilities. Examples of “objectively” bad things include: being betrayed, being deceived, and gaining pleasure from cruelty.

On this type of theory, it might be harder to square love addiction with any genuine sort of well-being—especially on the “narrow” view. That is, one could plausibly argue that only “normal” or “healthy” or “constructive” kinds of love are objectively constitutive of well-being, whereas love that is extreme, compulsive, resultant from abnormal brain processes, or that carries negative consequences for one’s health or for other social relationships is objectively bad for one. If this were the case, then such a love could potentially be deserving of some form of treatment under the right conditions.

Yet what theory of well-being should one accept? Unsurprisingly, philosophers have long noted that each of the theories of well being outlined above capture something important and intuitive about what is needed to live a good life, but all have problems as well. Accordingly, many philosophers opt for a composite theory in which well-being is seen as requiring at least certain aspects of all of the theories. On this composite view, then, well-being is constituted by engaging in objectively worthwhile activities, which we desire, and which provide us with pleasure or other valuable mental states (Savulescu 2007).

To summarize, love will clearly be an important component of any plausible theory of well-being. However, on a composite view, especially one which incorporates aspects of the objective-list account, love that entails insufferable pain, that frustrates other important desires, or that stops one from engaging in objectively worthwhile activities, might reasonably be taken to compromise well-being. The best life is not one that consumed by destructive or mal-adaptive forms of love, but is rather one in which love finds a robust harmony with other sources of the good.

Implications for treatment

What does all of this mean for treatment? With respect to the narrow view of love addiction, it means that we will need to make an ethical judgment about how narrow the diagnostic category should be. How strong or destructive does love have to be to qualify as addictive love? On the broad view, we face a similar challenge. Although it is that case that, on this view, everyone who loves is technically addicted, only some subset of cases should be judged to be appropriate candidates for treatment because of their effect(s) on other aspects of our well-being. Though the two views take radically different stances on the neurobiological underpinnings of problematic love, it is still the case that for both views it is the negative consequences or harm of the ‘bad love’ that determine whether the person or people involved should be considered appropriate subjects for the application of some remedy.

To reiterate: the primary diagnostic criterion for love addiction, in our view, should not depend very much on whether we agree with the narrow or the broad interpretations of the available neuropsychological evidence. Instead our allocations of medical and social care should be informed by an appeal to how much a person is made to suffer (or to experience other threats to well-being) through his or her experience of love (Kahane and Savulescu 2009). If we determine that medical or social intervention could be of benefit to a person, then it will be at least potentially appropriate to offer some degree of treatment or support.

How would it work? The prospect of anti-love biotechnology

Treatment of love addiction, like any other form of addiction, could take many forms. The most plausible starting place would be “traditional” therapies such as professional counselling, cognitive-behavioral techniques, psychoanalysis, or some combination of these and other widely-used treatment modalities that work primarily on the psycho-behavioral level (see Earp, Sandberg, and Savulescu 2015). At the same time, considering the recent surge of research focusing on possible neurobiological sources of love addiction, it may soon be possible to devise adjunctive drug-based therapies that could facilitate treatment of problematic forms of love by working on underlying neurochemical substrates. In a recent paper (Earp, Wudarczyk, Sandberg, and Savulescu 2013), we identified four conditions for the ethical use of such anti-love biotechnology:

• (1)the love in question is clearly harmful and needs to dissolve one way or another;
• (2)the person would want to use the technology, so there would be no problematic violations of consent;
• (3)the technology would help the person follow her higher-level goals and commitments instead of her lower-level feelings;
• (4)it might not be psychologically possible to overcome the perilous feelings without the help of anti-love biotechnology—or at least non-biotechnological methods had been already tried or thoroughly considered

In addition, we have suggested that any drug-based treatment of love or love-related phenomena should be undertaken only under the guidance of a trained professional, and only once the safety and efficacy of such treatment had been established via rigorous clinical testing (Wudarczyk, Earp, Guastella, and Savulescu 2013). Moreover, we have argued that such technologies should not be used on minors, i.e., before an age of consent (see Earp, Sandberg, and Savulescu 2014).

Concluding remarks

In this article, we have argued that there is now abundant behavioral, neurochemical and neuroimaging evidence to support the claim that love is (or at least that it can be) an addiction, in much the same way that chronic drug-seeking behavior can be termed an addiction. And we have argued that no matter how we interpret this evidence, we should conclude that people whose lives are negatively impacted by love ought to be offered support and treatment opportunities analogous to those that we extend to substance abusers.

It must be acknowledged that by suggesting any sort of equivalence between love, which is culturally admired above all other emotions (or relational states), and addictive drugs, which are deeply and almost universally demonized, we are raising a comparison that many will find off-putting or even offensive. In Western society, we hold love—or even just falling in love—in very high esteem. To speak of romantic passion and destructive drug use in the same breath might only serve to conjure images of punishing people for falling in love, stigmatizing them, or forcing them out of their lover’s arms and into a treatment program. We do agree that these would be inappropriate, and even dangerous ways to treat people who are suffering from a harmful sort of love or love gone bad.

But if falling in love is not so different, behaviorally or biologically, from developing a drug habit or a binge-eating problem, then perhaps this recognition could afford us an opportunity to deeply interrogate our views about the inherent harms of addiction. Perhaps, instead of taking on the view that love must be harmful if it is addictive, we could take the reverse view, and wonder if even substance addictions might be not always be strictly harmful under the right sorts of conditions. At least in principle, it would seem that drug addiction could even be good for us if the drug were plentiful, safe, and legal, in much the same way that, on the converse, romantic love may be bad for us if the object of our affection is cruel, or unfaithful, or uninterested. If the evidence shows that the two phenomena are identical or even just substantially overlapping in nature, then perhaps we should revisit our attitudes to both. The ultimate goal, whether the subject is drug-use or romantic passion, should be to identify those cases in which the behavior and its related phenomena cause harm and suffering to those involved. And any treatment that is pursued—on either the narrow view or the broad view of addiction—should be undertaken in such a way that the decision-making autonomy of the lovers is given maximal consideration.