Introduction: Romance Gone Bad

The aim of this article is to shed light on the phenomenon of women being killed in the context of romantic love relationships. Why is a woman assaulted, battered, abused, and finally murdered by a man who claims or has claimed to be in love with her—and most of the time, intensely and passionately in love with her? According to the hypothesis that I would like to explore, this type of murder, which I call “romantic femicide,” can be understood by examining how romantic love, under certain social structural conditions, functions as a relation of domination. I take into consideration, in particular, the structural condition of the gender dichotomy.

“Romantic love” is intended here as a principle, introduced in modernity, describing intimate relationships between free, adult individuals, who are bound by “pure” emotions and affects regardless of their gender, class, race, and so on. Social theorists have argued that, according to this principle, individuals fall in love and engage in intimate relationships for motives that are seen as independent from economic interests, class membership, family and community expectations, and values (for example, Reference GiddensGiddens 1993; Reference HonnethHonneth 2014). This should put lovers in the position of “making their own rules” and of recognizing each other in ways that allow them to be known and appreciated in their irreducible singularity, including being able to enjoy each other's company in ways that enhance and increase their freedom and well‐being.

This idea is not unproblematic. Critics of modernity, capitalism, and patriarchy have convincingly shown how the notion of romantic love is the product of specific social modern institutions and practices (for example, Reference IllouzIllouz 1997; Reference AdornoAdorno 2005); as such, it fulfills the ideological function of justifying and perpetuating a gender‐based division of labor, subordination of women, and discrimination against nontraditional (queer, polyamorous, non‐dyadic, and so on) forms of love (for example, Reference Wilkinson, Hines and TaylorWilkinson 2012; Reference IllouzIllouz 2013; Reference Jónasdóttir and FergusonJónasdóttir and Ferguson 2014). In this article, I will take into consideration one specific, sometimes lethal effect of social structures on relations of love. When confronted with the troublesome question concerning the motives that push men to kill their beloved partners, I argue, a critical understanding of the gender dichotomy and of related forms of domination and oppression might assist in providing us with some answers.

The modern principle of love, which has become part of our everyday language and common understanding, is also reflected in philosophical theories. At first glance, in fact, the entanglement between love and lethal violence might appear to be counterintuitive: love is usually thought of as a bundle of positive feelings and attitudes toward the other. Someone in love is usually depicted as deeply caring for the partner's happiness and well‐being, rather than as doing them harm. In fact, many recent philosophical publications presuppose or explicitly argue that love is a constitutive component of a good, happy, healthy, fulfilling life (see, for example, Reference NozickNozick 1989; Reference HonnethHonneth 1995; Reference FrankfurtFrankfurt 2004; Reference HelmHelm 2009). However, the association between love and hate, aggressiveness, and violence is part of a similarly well‐established narrative about intimate romantic relations. In this story, love is placed within a semantic field pivoting on the idea of irrationality: lovers—of any gender—are characterized as blind, passionate, suffering, unpredictable, mad, irascible, ready for anything in order to seduce or keep the beloved one (Reference Ben‐Ze'ev and GoussinskyBen‐Ze'ev and Goussinsky 2008, 90–91). As a matter of fact, the media representation of romantic femicide often matches this second understanding of love. The murderers of wives, fiancées, girlfriends, and mistresses appear as somehow justified on the basis of the narrative of the “crazy, unhappy lover.” Tragically blinded by the despair of unreciprocated love or by jealousy, a broken male heart can be driven to the most unthinkable atrocities. According to this narrative, the murderer of a woman who is unfaithful or who abandons him is not to be taken as entirely responsible and accountable for his acts (Reference Lipperini and MurgiaLipperini and Murgia 2013, 9–13).Footnote ¹

Theorists and activists fighting against femicide aim precisely at undermining the idea that the killing of the beloved woman might be explained or even justified as an unexpected emotional outburst aroused by exceptional pain and despair.Footnote ² Men killing their beloved ones follow, on the contrary, a structural pattern. Femicide is not (completely) irrational; on the contrary, it responds to an “objective” wrong or distorted logic, which is embedded in the patriarchal system (Reference Caputi, Russell and HarmesCaputi 2001, 203–204). Although individual psychic pathologies cannot be completely ruled out, femicide can be explained primarily by the social, political, and legal context in which it takes place. In this article, I will conceptually reconstruct only a particular portion of such a context, namely the structure of the romantic relation of love. This inquiry is motivated by the lack of attention in current debates to conceptualizations of the specificity of the kind of femicide that is associated with romantic love.

According to its most widely accepted definition, the notion of femicide applies to all forms of killings of women motivated by gender. In the words of Barbara Spinelli, a victim of femicide is murdered “for the sole reason that she is a woman” (Reference Spinelli, Laurent, Platzer and IdomirSpinelli 2012, 42). Several researchers and activists note that femicide is more frequent than homicide: men are murdered more frequently than women, but men are rarely murdered simply because they are men (Reference Radford, Radford and RussellRadford 1992, 10). Its first public appearance can be traced to the US‐based scholar Diana Russell. She first used this word—without defining it—in the context of her testimony about misogynist murder before the International Tribunal on Crimes against Women held in Brussels in 1976. Russell defined femicide for the first time in her book Rape in Marriage as “the killing of women because they are women” (Reference RussellRussell 1990, 286). In 1990, Russell and Jane Caputi adjusted the definition of femicide as “the murder of women by men motivated by hatred, contempt, pleasure, or a sense of ownership of women” (Reference Caputi and RussellCaputi and Russell 1990, 34). One of the most influential moves aimed at institutionalizing the term femicide in the English language is the collection Femicide: The Politics of Women Killing (Reference Radford and RussellRadford and Russell 1992). Its editors believed that the “naming of this extreme form of sexual violence” would have predicated “widespread resistance to it” (Reference Radford and RussellRadford and Russell 1992, xv). Recently however, Russell points out that, regrettably, most (American) feminists continue to use terms—such as domestic homicides—that obscure the misogynist factor in virtually all these crimes (Reference Russell, Laurent, Platzer and IdomirRussell 2012, 19).

Femicide may take different shapes. In its most common form, it corresponds to the endpoint of domestic violence escalation and is perpetrated by male partners or close family members (Reference Chawla, Laurent, Platzer and IdomirChawla 2012, 10). A growing literature has outlined other modalities, such as: racist femicide, homophobic femicide, femicide committed outside the home by a stranger, serial femicide, mass femicide, organized crime related femicide, targeted killings of women at war, female infanticide and gender‐based sex‐selective foeticide, gender mutilation related femicide, and others (see Reference Radford, Radford and RussellRadford 1992, 7; Reference Laurent, Platzer and IdomirLaurent, Platzer, and Idomir 2012, 55–59). According to the aforementioned definition, a victim of femicide is killed “because she is a woman.” The identification of such a lowest common denominator certainly helps to bring into focus the gendered basis of all these murders; as a definition it appears, however, as too vague and abstract. Further investigations have to be undertaken in order to grasp the specific features of each modality of femicide. With reference to romantic femicide, moreover, such a definition is not very helpful. In what follows, I argue that romantic femicide is perpetrated against a woman not exactly “because she is a woman,” but because she refuses to (or cannot) comply with the identity of a traditional woman, and her lover is not able to cope with her challenge to this identity (see Reference SpinelliSpinelli 2008, 21).

As Simone de Beauvoir knew, love may entail a “mortal danger” (Reference de BeauvoirBeauvoir 2011, 725), which does not pertain to the nature of the phenomenon itself, but to those particular social conditions that structure love as a social relation. Following Beauvoir, I will discuss these conditions as rooted mainly in the gender dichotomy. In this context, however, I prefer not to rely on Beauvoir's existentialist vocabulary, which presents the gender dichotomy as a sharp distinction between a “man” elevated to “transcendence” and his “Other,” “woman,” doomed to “immanence.” I would rather reformulate the “man”‐identity and the “woman”‐identity by focusing on the notions of “dependence” and “independence” (see, for example, Reference BenjaminBenjamin 1980; Reference BenjaminBenjamin 1988; Reference FriedmanFriedman 1998, 174, 177; Reference KittayKittay 1999). These two terms are, moreover, crucial, as I will show later, for giving an account of the dynamic of love and its distortion. I define woman (W) as the agent who perceives and understands herself, and is perceived and understood by others, as devoted mainly to others, that is to say, as someone whose main activity and responsibility is that of taking care of the other's needs and desires while putting aside her own needs and desires. According to this interpretation, W is the agent who takes over the management of the other's dependencies, thus attaching herself mainly to a position of dependence on the other.Footnote ³ This dependence makes it more difficult for her to strive for independence, to struggle to be recognized as independent, and thus to exercise power or to gain a power position. On the contrary, man (M) is the agent who perceives and understands himself, and is perceived and understood by others, as focused mainly on achieving and exercising his own independence and being recognized as independent. According to this interpretation, M is the agent who stands in a power position and exercises his power in order to satisfy his needs and desires; his power is thus a way of making his own dependencies less compelling and binding.

In the current literature on femicide, at least three types of explanation for the killing of women in the context of romantic relations can be identified: (a) the reification of the woman's mind/body, which becomes the man's property: the cause of femicide is related to the man's belief that it is his right to do with “his” woman's mind/body whatever he wants to (Reference Wilson, Daly, Radford and RussellWilson and Daly 1992; Reference Russell and HarmesRussell and Harmes 2001, 14). (b) Men's power over women: in contexts in which women find remarkable obstacles to attaining power positions and to exercising power within political, economic, and cultural institutions and practices, it is more likely for them to become victims of violence and femicide (Reference YodanisYodanis 2004, 670; Reference Chawla, Laurent, Platzer and IdomirChawla 2012, 11). (c) The denial of a woman's autonomy: the man commits femicide as a result of his incapacity to accept the woman as an independent and autonomous agent, who is capable of and entitled to make decisions on her own and even to affirm her own independence by leaving him and ending the relationship (Reference Lipperini and MurgiaLipperini and Murgia 2013; Reference HoweHowe 2014).

I intend to reformulate and combine these three types of explanation. The argument presented in (b) can be synthetized as follows: the possibility of femicide being perpetrated is rooted in those social structures that put M in economic, political, and social positions that allow them to unilaterally exercise power over W. In this article, I will better specify which conception of power is at stake in gender domination, and how the exercise of power may become violent and come to push violence to its extreme form. According to (c), moreover, femicide is to be understood as the result of the incapacity of some men—namely of those who strongly embody M—to sustain a relation with a woman who exercises her autonomous agency. This hypothesis will be further examined in the following paragraphs, where I suggest an interpretation of individual personal autonomy on the basis of a theory of power. Finally, the central thought presented in (a), men's alleged entitlement to have their partner's mind/body at their disposal, will be incorporated within the framework of the autonomy‐ and power‐based account. Ws leaving or betraying their partners, or, more generally, rebelling against their property status, would in fact display an unbearable degree of independence, autonomy, and powerfulness.

My central hypothesis consists in framing romantic femicide as the extreme form of violence that occurs as a result of an M's incapacity, within a certain gender order, to accept his partner's autonomy and, as a consequence, to deal with a certain dynamic of power inherent in the social relation of love. Such inadequacy is not to be understood simply or merely as an individual flaw or pathology but rather as the result of the structure of the gender dichotomy. I argue for this hypothesis in three steps. First, I try to reconstruct the conceptual structure of a love relationship between autonomous agents—a love, that is, that would not contain “the threat or the possibility of violence” (Reference BenjaminBenjamin 1980, 150). Lovers of this sort would joyfully and confidently depend on their partner and, at the same time, continuously try to affirm their independence of them. Second, I show how such a love relation, or interdependence bond, would entail a power dynamic. Third, I argue that the gender dichotomy splits the interdependence bond so that power is exercised unilaterally, that is to say it becomes the domination of one party over the other—the one representing independence, whereas the other appears doomed to dependence. On the basis of this framework, violence is conceptualized as a means for maintaining and reproducing relations of dominance. Romantic femicide would result, finally, from the unsustainability of such a gender order.

Love as Interdependence Bond

What would a love relationship between autonomous agents look like? The feminist paradigm of mutual recognition (Anerkennung), I argue, can provide us with some precious insights for addressing this question. Feminist theorists such as Simone de Beauvoir (Reference de Beauvoir2011) or Reference BenjaminJessica Benjamin (1988; Reference Benjamin1995) have been able to give an account of the ambivalent nature of love from a recognition‐theoretical account: such ambivalence indicates that modern love could be a milieu of structural oppression, exploitation, subjection, and violence, but also a flourishing, fulfilling, empowering experience. In Beauvoir's words, when “it will be possible for the woman to love in her strength and not in her weakness, not to escape from herself but to find herself, not out of resignation but to affirm herself,” love will be for women (as for men) a “source of life and not a mortal danger” (Reference de BeauvoirBeauvoir 2011, 724–25). According to Beauvoir, this form of love, “founded on reciprocal recognition of two freedoms,” would allow lovers—regardless of and beyond their gender—to experience themselves as themselves and, at the same time, as their other, without abdicating their “transcendence,” without “mutilating themselves”: “for each of them, love would be the revelation of self through the gift of self and the enrichment of the universe” (723).

Like Beauvoir, Axel Honneth has also thought of the love relationship in terms of mutual recognition. His contribution is important since it highlights how dependence, neediness, and vulnerability do not necessarily represent means of subjection and oppression but also possibilities of freedom, flourishing, and, as I will later show, empowerment. In Honneth's view, in order to affirm, discover, express, and transform themselves, lovers should reciprocally recognize each other as needy and vulnerable beings (Reference HonnethHonneth 1995, 95; Reference HonnethHonneth 2014, 141–42, 151). More precisely, by mutually recognizing the partner's needs and desires and aiding in (partially) satisfying them, each partner places the other in the condition of becoming independent (transcendent, in Beauvoir's vocabulary) and autonomous. Roughly stated, autonomous and independent agents are able to differentiate themselves from the other(s) and thus to develop a certain awareness of their desires and needs, to elaborate and pursue goals and projects, thus realizing and expressing themselves.

According to this picture, the condition for self‐realization lies in the relation with the other. Dependence on the other is the condition for independence and autonomy. Each agent is deeply dependent on the confirmations, care, physical appreciation, and satisfaction provided by the partner, and it is precisely such mutual reliance that makes separation and therefore self‐realization possible.Footnote ⁴ The entanglement of the moment of independence with the moment of dependence makes the social relation of love an interdependence bond. As such, love could be conceived of as a social relation that enhances individuals’ possibilities of self‐realization, flourishing, and transformation.

The conception of love as interdependence cannot but remain an illusory ideal, attainable only by a few, until the traditional dichotomy between M and W is overcome.Footnote ⁵ In the last section I will try to show how the interdependence bond is systematically disrupted and split by gender dichotomy, and how this may have lethal consequences for W. Before coming to articulate the explanatory framework for romantic femicide, a further step is needed. In the next section I will argue that love as interdependence is not to be considered to be as peaceful and harmonious as it may often appear.Footnote ⁶ In order to better grasp how dependence and independence are entangled, love has to be reconstructed as inhabited by a power dynamic.

Power(s)

Different modalities of power contribute to set such a dynamic in motion. To begin with, I propose to define power as an individualizing exercise. Exercising power, in this sense, is what allows individuals to differentiate themselves from the other, to affirm and express themselves as independent and autonomous agents and to be recognized as such. In the social relation of love, the exercise of this kind of power coincides with the lovers’ ability to figure out and to affirm their needs, desires, goals, and projects independently of the partner's.

As previously argued, however, individualization presupposes the other's positive response. Only if the other recognizes my needs, desires, goals, and projects, and tries (at least to a certain extent) to aid me in satisfying them, can I fully realize and express myself as a strong, autonomous, and independent being. Paradoxically, power as an individualizing (self‐affirming, self‐expressing) exercise cannot be conceived of as an individualistic exercise in the sense that it is performed by one individual; it depends rather on the interaction with the other and their responses (recognition, care, willingness to satisfy one's needs, and so on). On this account, then, my need of the other's recognition and care bestows power on the other. My dependence on the other's positive response gives the other power over me. It is the other's positive response to my exercises of power (to my attempts to become and to be independent and autonomous and to express myself as such) that, while empowering me, empowers the other as well. As a result, the dynamic of power between lovers should correspond to a mutual empowerment process.Footnote ⁷ I use the modal verb “should” since, as I noted before, this account of love fails to describe what a love relation between W and M looks like. In this section, I will refer to A and B, namely to lovers who do not identify themselves according to the gender dichotomy. Subsequently, I will turn to W and M.

As I have previously argued, dependence and independence mutually presuppose each other: the exercise of power of agent A, which is required for the affirmation of A's independence, both brings forth and presupposes the exercise of power of agent B, on whom A is dependent—and vice versa. Following the logic of the interdependence bond, A and B are both trying to become and to be independent of the other; at the same time, they need each other for their full self‐realization. Hence the exercise of power of agent B, which is required for the affirmation and expression of B's independence, both brings forth and presupposes the exercise of power of agent A, on whom B is dependent. A and B become (and continue to be) independent and autonomous by exercising power; power is thus to be first of all conceived as an individualizing exercise. A's individualization, namely her affirmation of independence and autonomy, makes A dependent on B; such a dependence empowers B—and vice versa.

By drawing on the current literature on power, two definitions of power may be introduced here to better explain such a dynamic, namely “power to,” PT, and “power over,” PO (see, for example, Reference WartenbergWartenberg 1990; Reference AllenAllen 1999; Reference LukesLukes 2005; Reference Allen, Forst and HaugaardAllen, Forst, and Haugaard 2014). In this context, PT can be defined as the capacity to be an agent, namely to figure out and satisfy one's own needs and desires, to set one's own goals, to elaborate and to pursue one's own projects. PO is framed rather in terms of a social relation, in which one agent is said to have (or to be in a position of) power over the other.

There is a further sense in which power is constitutive of the social relation of interdependent love. As we have seen, A is in a position of power over B insofar as A is able to affect the ways in which B becomes and is independent and autonomous (namely B's PT). Such an effect works through recognition. Recognition here does not mean a mere acknowledgment of the partner's needs, desires, goals, and projects, nor does it mean a mere appreciation or positive evaluation of those. In the sphere of love, recognition means something more material and concrete. B needs A in order to be recognized as an independent and autonomous agent in the sense that B needs A to be a positive factor in B's self‐realization: A is the one who satisfies, at least partially, (some of) B's needs and desires; moreover, A is the one who aids B, at least to a certain extent, to set B's goals, to elaborate B's projects, and to pursue them.

This more concrete and material idea of recognition that underlies the lovers’ involvement in each other's life provides a sense of the power that can be exercised over the other. Such PO may assume a considerable weight. A may affect B insofar as A acknowledges or attests to B's needs, desires, goals, and projects, or insofar as she approves of them. Furthermore, A may also affect B insofar as she is concretely involved in the satisfaction, pursuit, and realization of B's needs, desires, goals, and projects. Such an involvement could mean that A takes an active role not only in the interpretation but also in the shaping and constitution of B. A's power over B can be then framed as shaping, or constituting power.

Note that, although A is placed in such a “strong” power position over B, this does not necessarily undermine B's independence and autonomy; on the contrary, A's PO is empowering insofar as it provides the “material,” so to speak, by relying on which B can exercise independence and autonomy (and thus PT). Independence and autonomy consist, broadly speaking, in self‐realization and self‐flourishing, namely in the satisfaction of one's own needs and desires, and in establishing and pursuing one's own goals and projects. Needs, desires, goals, and projects are not, however, the product of agents’ individual, independent minds and bodies, but rather emerge within the (multiple, interconnected) relations with others, including love partners. Relatives, friends, lovers, colleagues, acquaintances, or even strangers are a constant, limitless source of inspiration, learning, and influence: it is thanks to the interactions and encounters with these others that we become able to understand our needs and desires, to opt for some goals and to elaborate projects, thus realizing and expressing ourselves. Social relations of this type are empowering, however, provided that PO does not suppress or block the partner's PT. A's shaping and/or constituting power over B is empowering as it allows, or at least as it does not prevent, B's resistance against A's power. B still has to be in the condition to affirm needs, desires, goals, and projects independently of A.

In the account of love I have outlined so far (love as interdependence), both partners depend on the recognition, affirmation, and aid provided by the other; both try to be independent and autonomous and to be recognized and confirmed as such. If exercises of power are reciprocal, the relation between A and B can unfold as a mutual empowerment process. This is not, however, how love usually manifests itself in our world. In many cases, love does take place in relations of domination and violence.

Domination, Violence, and Romantic Femicide

As many theorists of power have pointed out, the power dynamic that involves two (or more) agents is embedded in objective, super‐individual orders or structures. One can think of these as networks of material resources, interrelated norms, (more or less legally enforced) institutions, codified practices, shared beliefs, values, ideologies, and narratives that constitute, organize, and fix agents’ characteristics and positions. One particular structure can be identified as the gender dichotomy. What are its effects on love—and power—relationships?

As Jessica Benjamin suggests, the sharp distinction between the two gender identities splits the intertwinement between dependence and independence that constitutes what I have called the interdependence bond, thus polarizing the side of dependence onto one party (on the W) and the side of independence onto the other (on the M). The power dynamic between A and B is blocked, and cannot unfold as a reciprocal empowerment process. As a result, the romantic relation becomes domination of M in relation to W (Reference BenjaminBenjamin 1995, 185; see also Reference BenjaminBenjamin 1980, 153–54).

In this type of relation, both M and W can in principle exercise individualizing power. However, M's individualizing capacity is stronger and more effective than W's; moreover, M has more and better chances to exercise it. Furthermore, both M and W have, in principle, power over the partner, since they both need the other in order to be confirmed and recognized as autonomous and independent and to realize and express themselves. In the dichotomous and polarized organization of gender, however, M and W achieve self‐realization in different ways. On the one hand, M's needs, desires, goals, and projects are recognized and approved not merely by W but by a number of other agents outside of the love relationship as well. Not only W but also a number of other agents aid M in satisfying his needs and desires and in his pursuit of his goals and projects. Such a “multi‐lateral” and “multi‐level” recognition enjoyed by M curbs W's power over him. On the other hand, W's needs, desires, goals, and projects are structured in such a way that they mostly need M's confirmation and recognition in order to be satisfied and pursued. Moreover, W cannot rely on many other agents who would support her goals and projects. This asymmetry empowers M and disempowers W.Footnote ⁸ For the same reason, M can affect the partner's autonomy and independence more extensively than W can. M's acknowledgment and approval count and weigh more than W's do; moreover, M is in the position of exercising a more permeating shaping and/or constituting power than is W.Footnote ⁹ Thus, the dynamic of power between M and W is not reciprocal but results in the domination of M over W. This implies that W is not in the position to exercise her counter‐power against M.

In a relation of domination, W's capacities and chances to affirm themselves, to autonomously and independently figure out and pursue their needs, desires, goals, and projects, and to influence and affect their partners tend to shrink, to be obstructed and blocked. In a traditional romantic relationship, both M and W do recognize each other, but the kind of recognition at play here works asymmetrically. While W recognizes M essentially as an autonomous and independent agent, M recognizes W in ways that tend to suppress or to obstruct W's autonomy and independence. Certainly, M does appreciate, like, support, and adore W; his modes of recognition do not, however, lead to W's empowerment. On the contrary, W's different recognition activities (including her care [and] work for him) allow him to boost his autonomy and independence.

The asymmetry inherent in traditional love, however, never makes W entirely heteronomous, subjugated, and dependent.Footnote ¹⁰ W's margins for power are constituted rather as the object of ongoing negotiations and more or less open struggle. Ws involved in love relationships, although subjected to domination, are still agents. Hence, their powers of individualization and self‐affirmation cannot be thus completely drained away. This allows them, at least in principle, to exercise their counter‐power and to find ways to satisfy their needs and desires and to pursue their goals and projects outside love relationships. In the modern age, the (partial) institutionalization of the norms of human equality and the valuing of individual autonomy have increased the possibility for W to challenge the relations of domination she finds herself in. Although complying with a traditional gender identity, many Ws do aspire toward a love to be experienced as a mutual empowerment process.

In many cases, Ms suffer because of their partner's attempts to abandon them, or just because Ws fail to support or encourage them, or give attention and care to something or someone else. How can the traditional dichotomous structure “defend” itself against W's striving for power? It is at this point that violence enters the picture. Violence perpetrated by Ms against their partners may thus be seen as fulfilling a threefold function: first, to contain, repress, and stop W's PT; second, to prevent W from resisting and rebelling against M's PO; third, to make M's dependence on W less compelling, thus curbing and neutralizing W's PO.

Gender violence in romantic relations can be of different types: physical, psychological, sexual, or economic (see, for example, Agnello Reference Agnello Hornby and CalloniHornby and Calloni 2013). As Kimberlé W. Crenshaw has extensively shown (for example, Reference CrenshawCrenshaw 1991), in order to understand the overlapping and multi‐layered ways in which violence works, not only gender but other social categories as well—race, class, sexuality, and so on—have to be taken into account. Individuals’ identities have in fact “multiple grounds” and cannot be reduced solely to gender factors. Underemployed women of color, or immigrant women who do not speak the institution's language, for instance, are likely to be more dependent on their partners and to have fewer possibilities of rebellion and separation than would a white, middle‐class professional. Theoretical analysis, empirical research, and activism cannot conflate or ignore qualitatively and quantitatively relevant “intragroup differences” (Reference CrenshawCrenshaw 1991, 1242). In the context of romantic relations, further investigations should be aimed at showing how racism, class, or age differentials, social norms of sexual desire, health issues, and so on contribute to restrict and block W's margins of autonomy and power, thus fixing and reproducing the dichotomy between M and W. In other words, the split between dependence and independence is conditioned, reinforced, and perpetuated by factors that are not only identified with gender dichotomy but intersect with it.Footnote ¹¹ Taking into consideration categories such as race, class, and age would help to specify different degrees and modes of agency and power, which lovers could rely upon in order to cope with their dependence and gain (or lose) their independence. In what follows, however, I will continue to concentrate on agents characterized solely by gender identities (M and W).

In the last part of the article, I argue that the maintenance, consolidation, and reproduction of relations of domination through violent means does not always correspond to the reinforcement or enhancement of M's power. The extreme escalation of violence seems, in fact, to give rise to a paradox: it is aimed at restricting and blocking W's autonomy and independence and therefore at increasing M's unilateral power (domination) over her, yet, it leads to the decrease and emptying of M's power (see also Reference BenjaminBenjamin 1980, 150–51).

The paradox can be explained on the basis of the logic of power that I have previously outlined. In the previous section, mutual empowerment was presented as a central dynamic of free love, namely of love understood as interdependence, which in most cases is not realized as such. In most cases, a romantic relation is a relation in which M dominates W; in some cases, M has to rely on violent means in order to maintain domination (in some other cases, moreover, interdependence works only for a relatively short period of time). In order to explain romantic femicide, however, the idea of interdependence as mutual empowerment again has to be taken into account. I propose, in fact, to see this kind of murder as resulting from the unsustainability of the distortion of the mutual empowerment logic. Although such distortion constitutes the most common form of romantic relation, it sometimes becomes unbearable, and disruptive in the most tragic sense. The more W's margins for independence, autonomy, and power are restricted, the less power she can exercise in order to acknowledge, approve, and actively contribute to satisfying and pursuing M's needs, desires, goals, and projects. The disempowerment and loss of independence and autonomy that affect one partner, W, could then affect M negatively as well, thus giving rise to the aforementioned paradox. To put it bluntly: when mutual empowerment does not take place, the consequences may be fatal. The disempowerment of W could lead finally to the disempowerment of M.Footnote ¹² In order to combat this, and to maintain his (unilateral) power over W, M becomes more and more violent. Violence increases insofar as it seems not to fulfill its purpose, and M continues to experience his powerlessness. The mortification, humiliation, sexual harassment, beating, and torture of the female partner is aimed at ultimately annihilating her individuality, integrity, independence, autonomy, and power (both PT and PO). The intensification of violence, however, cannot provide M with the power he was seeking in the first place; on the contrary, by exerting violence he becomes more helpless, dreading his dependence on his partner, incapable both of leaving W alone and of recognizing her power, independence, and autonomy.

Murder is the final, desperate attempt to regain mastery—which represents the peak of the paradox, since the master needs the other, a living other, in order to perform as master. This argument does not claim that every distortion of mutual empowerment leads to femicide. Asymmetrical dependence and unequal chances of independence represent, as I have said, the normal state of affairs for romantic relations under conditions of gender dichotomy. My claim is a more modest one: romantic femicide can be explained by pointing to a distortion of the logic of interdependence that cannot be sustained.

One might object at this point that in traditional romantic relations M does not rely very much on W and on her contributions to his happiness, well‐being, and self‐realization, since M has other arenas (work, politics, friends, other Ws) where he can satisfy his needs and desires, set and pursue his goals and projects, realize and express himself. In other words, in relations of domination, W's autonomy would not be relevant for M. According to this objection, love relations should not matter very much for M. This assumption, however, contradicts empirical evidence showing how violent partners, who then become perpetrators of femicide, are deeply emotionally attached to their victims. As Aaron Ben‐Ze'ev and Ruhama Goussinsky show, men who kill their female partners are very often deeply, desperately, obsessively dependent on their victims (Reference Ben‐Ze'ev and GoussinskyBen‐Ze'ev and Goussinsky 2008, 43). This can be explained, on my account, by the fact that, by refusing to accept their own dependence, and at the same time to recognize and to enhance women's autonomy and independence, they deny themselves the possibility of empowerment. By destroying W's PT and PO, they destroy the possibility of gaining power for themselves.

Another objection that can be raised consists in pointing out that gender dichotomy has arranged (love) relationships between M and W in ways that are de facto favorable for M: W's disempowerment can namely result, and does in fact very often result, in M's empowerment, rather than in the deprivation of M's PT and PO. Most revealing, Jónasdóttir has expounded on the mechanisms that allow men “to exploit women's capacities for love and transform these into individual and collective modes of power over which women lose control” Reference Jónasdóttir, Jónasdóttir, Bryson and Jones(Jónasdóttir 2011, 49). According to Jónasdóttir, love corresponds both to care and eroticism (or “erotic ecstasy”), and is viewed as a creative/re‐creative, transformative power, essential for those practices that make life possible. The exploitation of women's love power puts men in a position of domination over women (50). Exploitation consists in the fact that although the man, thanks to the woman's loving care, is put in the condition of experiencing erotic ecstasy, of therefore enhancing his self‐confidence (his PT, I would say), the woman is not taken care of by the man and thus is denied one vital component of socio‐sexual relations, namely ecstasy:

On this account, exploitation does not necessarily involve “coercion or abuse” (52); it therefore does not have to rely on violence.

This explanation clarifies one mechanism that allows M to take advantage of his dependence on W, that is, how he manages to increase his PT (individualizing, self‐affirming, self‐expressing power) in virtue of (and not by breaking with) his dependencies ties. Jónasdóttir's account, moreover, shows how love power represents a vital basic force, essential for human life's natural and social reproduction (Beauvoir's “source of life”). Such force cannot be and has never been completely and permanently taken away from women. What happens then if, and when, Ws lay claim to fully enjoy eroticism and to become recipients of care? If and when they want their sexual and emotional concrete needs to be recognized? Such demands push Ws to lower the caring activity for their less supportive and empowering male partners, to loosen their dependent ties, even to betray or leave them. Confronted with his own dependence, in order to recover his control and mastery and to contain W's “rebellion,” M may resort to violence. Violence, in some cases, may escalate—pointlessly. Although violence can manage to destroy W's PT and, as a consequence, W's PO, this does not enable M to regain PT. M's power remains in fact dependent, at least partially, on his partner's power.

As cases of romantic femicide show,Footnote ¹³ a way out of such a deadlock—which would imply acknowledgment of M's dependence on W, may be fatal. The deadlock consists in the fact that violence reveals itself not as an expression of power but of weakness, powerlessness. Ms who commit femicide are people who depend and have depended very much on their victims, and who were not capable of accepting their dependence and making it into a source of power. Violent attempts to maintain and restore unilateral mastery by disempowering Ws may prove to be not only ineffective but also tragic. Romantic femicide is the sign that the “normality” of gender domination, which is the result of the structure of the gender dichotomy, sits on a disruptive, lethal potential.