Introduction

This paper focuses on reconceptualizing adult romantic attachment theory through the lens of polyamory. Polyamory is defined as “the practice of consensually and with mutual interest negotiating desire for more than one relationship” (Brunning, 2018, p. 514, original italics). Polyamory is one type of consensual non-monogamy (CNM), where partners consent to have additional sexual and/or romantic relationships (Moors et al., 2013). This paper does not discuss CNM more broadly than polyamory as that is a subject outside this paper’s frame of reference. Because polyamory is a type of CNM that emphasizes emotional intimacy with multiple partners (Chapman, 2010), there is greater potential for significant romantic attachments among people who practice it. In contrast, the additional academic literature on attachment and CNM discusses swinging and open relationships, where emotional connection is often limited to the primary couple (Matsick et al., 2014).

While “attachment theory is currently one of the leading theoretical frameworks for understanding interpersonal functioning, relationships, well-being, and personality development” (Fraley, 2019, p. 404), it has been developed through a dyadic framework (Main & Solomon, 1990), excluding individuals who practice polyamory and other types of CNM. This paper examines the assumptions and origins of the following contentions: (i) secure attachments, characterized by high levels of satisfaction and commitment, are only present in the context of monogamy (Moors et al., 2015), and, (ii) stigma surrounding avoidantly attached individuals, seen as desiring non-monogamy and casual sex (Fraley, 2019).

This article provides (i) an overview of polyamory and attachment theory, (ii) discusses attachment theory through the lens of monogamy, (iii) examines dimensions of attachment from a polyamorous perspective, (iv) addresses the stigma toward insecure attachment dimensions and polyamory, and suggests directions for future research. Although it appears that those who exhibit dimensions of high avoidance are attracted to polyamory (Moors et al., 2015), this paper illustrates that the practice of polyamory requires characteristics of secure attachment, while individuals exhibiting dimensions of high anxiety are attracted to it as a means of attaining greater intimacy (Ka et al., 2020).

CNM and Polyamory: Defined and Discussed

It has become evident that polyamory and other types of CNM are being practiced widely. According to Moors et al., (2015), 4–5% of the North American population is engaged in some type of CNM, where all parties consent to multiple romantic and/or sexual relationships. In a sample of 2003 Canadians, Fairbrother, Hart, and Fairbrother (2019) found that 4% were currently in an open relationship while 19.6% had been in an open relationship at some point in their lives. Twelve percent also reported that their ideal relationship type was open.

As mentioned above, the academic literature on CNM focuses on swinging, open relationships, and polyamory. Swinging and open relationships are often characterized by sexual relationships with emotional intimacy reserved for the primary couple (Moors et al., 2017). Swinging often occurs in the context of a social gathering with heterosexual couples, with bisexuality reserved for those identifying as women (Barker, 2011). An open relationship allows a couple to have sex and/or romance outside of their relationship (Barker, 2011). Moors et al. (2015) argue that swinging and open relationships are less likely to include love outside of the primary relationship, meaning that the attachment bonds among people who practice polyamory become more relevant.

The attachment and CNM literature agree that one of the defining features of romantic love is commitment (Klesse, 2006, 2011; Shaver & Hazan, 1988). Commitment is defined as the desire to maintain a relationship (Shaver & Hazan, 1988) and an agreement regarding what each relationship will look like (Klesse, 2006). While polyamorous individuals may appear to be less committed than those in monogamous relationships, crafting relationship agreements can lead to relationship satisfaction not found in monogamy, because each relationship can be tailored to different needs. In monogamy, many assumptions regarding commitment may be taken for granted and never discussed, including marriage and children.

As stated above, polyamory is defined as “the practice of consensually and with mutual interest negotiating desire for more than one relationship” (Brunning, 2018, p. 514, original italics). Brunning (2018) argues that intentionally developing multiple romantic relationships decreases anxiety and insecurity, which could engender secure attachment. In polyamory, partners often discuss difficult emotions such as jealousy and insecurity, as they cannot be ignored. In sharing their experience of these emotions and being heard, an atmosphere of safety and security can be established.

Polyamory is seen as both a practice and an identity. Some individuals only consider themselves to be polyamorous if involved with multiple people (Henrich & Trawinski, 2016). Conversely, some consider polyamory to be a fixed identity, regardless of how many relationships they are in (Robinson, 2013). Non-normative identities are often seen as fixed (Benson, 2017); however, where there is room for multiple relationships, there may be greater fluidity in how relationships appear (Katz & Graham, 2020), including the ways that attachment dimensions manifest. Polyamorous individuals often negotiate desire for multiple relationships within themselves, between partners, and in society. Polyamorous individuals may be going to work, studying, and raising children, leaving limited time for multiple relationships. In society, polyamorous individuals must contend with mononormativity, or the belief that monogamy is natural and right (Kean, 2017).

Definitions of polyamory offered in the academic literature, include “the practice of, belief in, or willingness to engage in consensual non-monogamy, typically in long-term and/or loving relationships” (Rubel & Bogaert, 2015, p. 961). This paper uses the already mentioned definition of polyamory, because it is inclusive of individuals who consider themselves to be single, as well as those in more hierarchical primary and secondary relationships, where the primary couple may share a home, finances, and children, while dating others outside the home (Balzarini et al., 2017). The primary couple may set boundaries around activities such as sleeping over and various sexual acts (Ferrer, 2018; van Tol, 2017). There are also polyfidelitous relationships where partners are closed to other relationships, and relationship anarchy, where no relationship is deemed to be less important than any other, including platonic friendships (Barker, 2011). There are many different forms of polyamory that individuals may practice, however they all share the same basic ideals of commitment, honesty, open communication, and consent by all parties involved. Fraley argues that individuals exhibiting dimensions of secure attachment “communicate effectively in their relationships” (2019, p. 406); given the aforementioned emphasis on communication in polyamory, traits of secure attachment would facilitate the development and stability of multiple relationships.

The perception that polyamory begins with a primary couple who decide to transition from a monogamous to a non-monogamous relationship leads to subsequent relationships being deemed, by mental health professionals and other outsiders, as “something that must be tolerated or used as a means of supporting the dyadic couple” (Bairstow, 2017, p. 346). Often, the primary relationship is viewed as not meeting the needs of either or both people, motivating them to seek out additional relationships (Mitchell et al., 2014). However, Mitchell et al. (2014) found that need fulfillment was high across all relationships in their surveys with 1093 polyamorous individuals, allowing individuals to be committed to, and satisfied by, all their partners. It may be premature for mental health professionals to view the primary relationship as unsatisfactory. In a case study of a polyamorous couple by Kolmes and Witherspoon (2017), both people exhibited dimensions of secure attachment, and demonstrated attributes of a satisfying relationship, including a positive outlook and accepting influence from each other, illustrating that practitioners of polyamory can have secure attachments.

While a psychological perspective is useful in discussing attachment theory and interpersonal relationships, other disciplines more effectively convey the larger, societal picture. Sociological literature elucidates the fact that monogamy and heteronormativity are often tied together (Gusmano, 2019). Roseneil and Budgeon (2004) argue that the norm of the heterosexual couple with children “cannot be taken for granted as the basic unit in society” (p. 140). Bauman (2000) asserts that the postmodern nuclear family loses much of its appeal because individuals have greater autonomy in their relationships; thus, individuals can prioritize many types of relationships, including platonic friendships. This has been the case in the LGBTQ2S + community, where families of choice have become normalized (Blumer & Murphy, 2011). A survey of 1235 individuals with 485 identifying as bisexual, homosexual, asexual, demisexual, pansexual, and queer, found that 33% had maintained or were maintaining a CNM relationship (Cubells-Serra, 2021). Those identifying as bisexual were less likely to buy into myths of romantic love, including beliefs that you can only love one person at a time and that jealousy indicates true love. Klesse (2011) found that there is a large overlap in the UK between the bisexual and polyamorous communities, where love can take many different forms because of the acknowledgement of attraction to multiple genders. In interviewing polyamorous women of all sexual orientations, Sheff (2005) found that polyamory allowed a questioning of heteronormative relationship roles because of the emphasis on equality.

Understanding relationships and attachment through a lens that challenges heteronormativity is instructive. Even Bowlby (1979) acknowledged that adult attachments do not need to include sex and that sex can occur without emotional bonds. While there may be stigma attached to relationships that deviate from the monogamous and heterosexual norm, this does not mean that insecure attachment becomes more salient. Bowlby (1979) recognized that relationships involving a secure base will still include many intense emotions; the ability to handle them requires self-awareness, which is often predicated on traits of secure attachment.

Attachment Theory: Defined and Discussed

Attachment has been defined as the communication of physical safety through proximity and protection provided by an attachment figure (Blakely & Dziadosz, 2015), which may include biological parents or other caregivers for children, and peers and romantic partners for adolescents and adults. Traditional definitions of attachment are anchored in categories of secure and insecure, the latter including subcategories of anxious, avoidant, fearful, and disorganized. This developed within developmental psychology and originates with John Bowlby. More recent discussions within the discipline of social psychology make use of the concept of dimension, which pertains to the amount and quality of attachment-related anxiety and avoidance within which an individual resides (an in-depth discussion of the conceptualization of the dimensions of attachment is beyond the scope of this article. For that information the reader is referred to the work of Shaver and Mikulincer (2002), as well as Mikulincer and Shaver (2016)).

The distinction between categories and dimensions of attachment is critical, especially for the current discussion. The difference in the two conceptualizations of attachment focuses on the distinction between the existence of (i) static categories and (ii) fluctuating patterns that reflect the quality of attachment (Fraley, 2019; Shaver & Mikulincer, 2002). Work by Shaver and Mikulincer (2002) and Mikulincer and Shaver (2016) integrates both conceptualizations of attachment for a more thorough approach. Their discussion includes the influence of adult, romantic experiences of attachment, which can shift dimensions, and therefore examines attachment categories as more malleable than previously was believed. Hazan and Shaver (1987) also acknowledge that attachment dimensions can change based on the romantic partners one chooses and the environment in which the relationships are anchored. This more fluid conceptualization of attachment is especially important for the current discussion of whether individuals involved in polyamorous relationships are insecurely attached. As Moors et al. (2019) illustrate in their study of individuals with at least two partners, relationships that exhibit dimensions of avoidance or anxiety did not lead to anxiety or avoidance in the others, suggesting that attachments in different relationships are not intertwined.

This paper, as stated, focuses on attachment within adult romantic relationships. Adults come into romantic relationships with an attachment history originating in infancy. The presence, amount, and quality of secure, anxious, and avoidant attachment dimensions is first developed in early relationships with caregivers. Securely attached adults received caregiving that communicated (i) their value as human beings, and (ii) a feeling of being lovable and safe in the world. Secure adults are generally competent in their romantic relationships, which tend to last longer and contain less conflict and more satisfaction than those of insecurely attached adults (Schmitt et al., 2004). Adults who experienced inconsistent and/or abusive caregiving may develop dimensions of avoidant or anxious attachment that they take into romantic relationships. Avoidant adults are often afraid of others depending on them (Fraley, 2019); anxious individuals view themselves negatively and others positively, which may lead to a fear of partners being unresponsive.

While it is evident that secure attachment is very beneficial, theorists now acknowledge that groups with a variety of attachment dimensions experience advantages. Ein-Dor and Hirschberger discuss the fact that individuals will respond to threats in a way that gives clues to others in the environment. Because individuals exhibiting dimensions of high avoidance wish to avoid vulnerability, they will likely be the first to flee (Ein-Dor & Hirschberger, 2016). Individuals exhibiting dimensions of high anxiety are sensitive to environmental cues. If they witness avoidantly attached individuals fleeing, they understand that a threat is present. Securely attached individuals are often good leaders and coordinators; however, studies show that they react more slowly to danger (Ein-Dor, 2014). Thus, they depend on the reactions of insecurely attached individuals.

This discussion is useful when considering polyamory; when individuals are connected with each other because they share partners, there may be different types of threats to each relationship, and individual responses may be anchored in their attachment dimensions. Fern (2020), a psychotherapist supporting the polyamorous community, acknowledges that polyamory is inherently insecure, as multiple breakups may occur in a short amount of time, while strong emotions such as jealousy and envy may be dealt with regularly. If an anxiously attached person is partnered with one or more avoidantly attached people, there may be an imbalance in intimate desire (Hazan & Shaver, 1987). While anxiously attached individuals are sensitive to threats, they may be aware that their partners wish to end the relationships. They may also sense when their partners experience difficulties in other relationships. As mentioned, polyamory emphasizes open communication; while anxiously attached people may personalize problems that their partners are having (Fern, 2020), they are also likely to talk about it (Mikulincer, 1998). If communication amongst multiple partners is required, securely attached individuals would be able to encourage it, in a caring way (Klesse, 2011). Thus, having a variety of attachment dimensions within multiple relationships may be supportive to the group as a whole.

There are varying estimates of the percentage of people who fall into the different attachment dimensions; according to Allen and Baucom (2004, p. 476), who sampled both university students and the wider community in a large southeastern city in the United States, 26% of the students and 37% of the community members were securely attached, 36% and 19% respectively were anxiously attached, and 17% and 26% respectively were avoidantly attached. Romantic relationships exert powerful influences and have the ability to shift attachment dimensions, either increasing or decreasing anxiety and avoidance (Shaver & Mikulincer, 2002). This is important for the current discussion, as multiple romantic relationships may shift attachment dimensions to a large extent; thus, blanket statements regarding the attachment categories of polyamorous individuals may be unfounded.

A Monogamous View of Attachment Theory

Attachment theory was crafted using a dyadic framework (Main & Solomon, 1990). The Strange Situation experiment that led to the creation of infant attachment styles included pairs of middle class children and, typically, their mothers (Ainsworth et al., 1978). Thus, relationships where the dyad is challenged become ignored at best and stigmatized at worst within this framework. More recently, there is an acknowledgment that children can become attached to multiple caregivers, including fathers, grandparents, and childcare providers (Howes & Spieker, 2008). As there has been a steady increase over time in the number of working mothers, many children are placed in childcare early. In their study of Israeli children in childcare during their first year of life, Sagi et al. (2002) found that “those experiencing infant-caregiver ratios of 3:1 or less had a security rate of 72%, compared to 57% for infants experiencing higher infant-caregiver ratios” (Howes & Spieker, 2008, p. 323). Children become attached in different ways to different caregivers; however, they may also construct internal working models based on which caregiver is most present in their lives (Bretherton, 1985). This may lead to both secure and insecure attachment dimensions in different relationships. Other studies have found that when children have multiple caregivers, they are more likely to form positive peer relationships (Van IJzendoorn et al., 1992). Being able to connect and attach with caregivers, no matter who they may be, is important for a child’s development. Given that children attach with multiple people, this suggests that adults may enter into multiple relationships where attachment becomes salient.

While children develop a script based on their attachment history (Waters & Waters, 2006), this may require revision if they embark upon polyamory in adulthood, if they were raised in a monogamous context. There is evidence that attachment dimensions remain stable (Waters et al., 2000); however, individuals may vary in different dimensions depending on the relationship. For instance, studies have found that individuals may be securely attached to their caregivers while being insecure in their romantic relationships (Fraley et al., 2011a, 2011b).

Hazan and Shaver (1987) argue that attachment within romantic love is influenced by earlier attachment patterns between children and their caregivers. Just as Hazan and Shaver assume the primary relationship in childhood is limited to two people, mother and child, they also make an assumption that important romantic attachments consist of a bond limited to two people. In this way, one of the formative articles in the field conceptualizes adult romantic love as pair bonding, leaving no room for multiple romantic relationships. When someone in a monogamous partnership develops an emotional relationship with someone who is not their partner, this is often seen as infidelity (Zola, 2007). The discussion of attachment as pair bonding may frame polyamory as infidelity, despite the fact that all parties consented. Thus, there is a need to reconceptualize attachment theory in a way that allows for multiple relationships.

Couples with at least one partner living within insecure attachment dimensions often experience negative communication patterns, such as “demand-withdraw and mutual avoidance” (Domingue & Mollen, 2009, p. 678). Demand-withdraw communication is defined as a state where one partner seeks change in the relationship and the other responds by walking away or changing the subject (Domingue & Mollen, 2009). This often occurs in a relationship where the partners exhibit insecure attachment dimensions. Allen and Baucom (2004) argue that individuals with attachment dimensions that include high anxiety are likely to engage in infidelity because they often desire more intimacy than their partner can provide. Conversely, individuals with dimensions that include high avoidance are not concerned with the consequences of their own infidelity because they do not fear becoming attached to a non-primary partner. This argument has been supported by other authors (e.g. Fish et al., 2012). It illustrates that there is a discrepancy between the needs of those in different attachment dimensions, which may not be communicated.

Monogamy as Seen Within Attachment Theory: Empirical Research

The above discussion alludes to the fact that individuals living within dimensions that include both high anxiety and high avoidance may be unable to articulate what they need in a relationship, which could result in lower relationship satisfaction. Tucker and Anders (1999) reported that women exhibiting dimensions of anxious attachment described lower relationship satisfaction, while men living within dimensions of anxious attachment were often inaccurate at perceiving their partner’s feelings. Others have illustrated that individuals living within dimensions of insecure attachment struggle to trust their partners (Brennan & Shaver, 1995; Hazan & Shaver, 1987; Simpson, 1990). Where trust is not present, sustaining any type of relationship becomes difficult. In contrast, Simon and Baxter (1993) found that securely attached individuals were more likely than insecurely attached individuals to use reassuring communication strategies with their partners; this is critical in an environment where emotions such as jealousy are communicated on a regular basis. Being able to reassure one’s partners that their wellbeing is important may be vital in polyamorous relationships.

Few longitudinal studies examine attachment from infancy to adulthood (Cassidy, 2000). However, Waters et al. (2000) studied 60 participants, measuring attachment dimensions of infants at 12 and 18 months old and again after they had entered university. At 12 months, 58% of the participants were securely attached, 24% were avoidant, and 18% were anxious; as adults, 50% were secure, 32% were avoidant, and 18% were anxious (Waters et al., 2000). This study has several flaws: there is no discussion regarding why the percentage of anxiously attached individuals did not change. The discussion around why attachment dimensions change revolves around stressful events in the parents of the participants, including their divorce or death, parent(s) having a life threatening illness or psychological disorder, and physical or sexual abuse by a family member. The authors mention that if a participant’s mother had no stressful events, attachment dimensions remained the same 72% of the time; however, when mothers did have stressful life events, both securely and insecurely attached individuals changed classifications. Finally, the sample was composed of lower to upper middle class individuals. Thus, there is a need for diversity, while recognizing that events in the lives of the participants, including the ending of significant relationships, can shift attachment dimensions.

Kirkpatrick and Hazan (1994) measured changes in attachment dimensions and relationship status over a four year period. The sample consisted of 146 women and 31 men, with a mean age of 41. At the beginning of the study (T1), 88 participants reported themselves as secure; 73 of those participants remained secure four years later, or T2 (Kirkpatrick & Hazan, 1994). For those who reported insecure attachment dimensions, they were more likely to shift. Initially, 46 reported that they were avoidant and 38 reported that they were anxiously attached. 13 of the avoidant participants became secure and five became anxious, while 13 of the anxious participants became avoidant and six became secure (Kirkpatrick & Hazan, 1994).

Participants were given a variety of options with regard to relationship status, including “(a) seeing someone; (b) seeing more than one person… (e) living with a relationship partner; (f) engaged to be married; (g) married” (Kirkpatrick & Hazan, 1994, p. 127). The results show that avoidant individuals were more likely to report that they were seeing ‘more than one person’ (undefined), which indicated a preference for relationships with less intimacy, stigmatizing individuals who are purposefully seeking out multiple relationships. Individuals exhibiting dimensions of high anxiety were more likely to report at T1 that they were looking for a partner. At T2, they were usually living with someone. They were more likely than securely attached individuals to be with the same partner at T2 (Kirkpatrick & Hazan, 1994), indicating that anxiously attached individuals can have stable, intimate relationships.

Simpson et al. (2011) studied attachment and empathic accuracy in a sample of 95 married couples, who were brought into a laboratory to have a discussion about jealousy, and minor and severe relationship problems. Anxiously attached individuals “displayed higher empathic accuracy when the problem was relatively more severe compared to less anxious individuals, who showed the opposite pattern… Among couples who discussed jealousy, no interaction between anxiety and problem severity was found” (Simpson et al., 2011, p. 9). This may explain how anxiously attached individuals can be polyamorous, which often requires discussion of difficult emotions such as jealousy. While being empathically accurate with multiple partners can be exhausting, it provides an opportunity to have a level of intimacy that anxiously attached individuals crave.

While much of the attachment literature and empirical research discussed above has only been applied to heterosexual individuals, the practice of polyamory often involves questioning heteronormativity; a discussion of how attachment theory applies to the LGBTQ2S + community is merited. Studies have shown that minority stress, which can lead to physical and mental health issues (Meyer & Frost, 2013), leads to insecure attachment dimensions in LGBTQ2S + individuals. Mohr et al. (2013) assert that gay men are more likely to exhibit dimensions of high anxiety, compared to lesbians, because their violation of gender norms draws more resistance. Landolt et al. (2004) showed that gay men who did not conform to gender norms during childhood experienced negative judgment from their fathers and peers, which led to anxious attachment. However, when gay men are accepted by their peers, the link between parental rejection and insecure attachment is weakened.

Shorey (2016) illustrates that trans individuals are more likely to come out if they are securely attached to their caregivers. If they are accepted, they are more likely to have secure adult relationships. In their study of 70 trans patients, Colizzi et al. (2013) found that 46% reported living within dimensions of high avoidance, 22% were anxious, and 30% were secure. After undergoing hormone replacement therapy, individuals exhibiting dimensions of high anxiety experienced the greatest drop in cortisol levels, meaning that they were less susceptible to stress.

While minority stress can challenge attachment security in LGBTQ2S + individuals and relationships, there is also contrary evidence. Questioning heteronormativity allows for greater equality and freedom within queer relationships (Hammack et al., 2018). With this freedom comes the ability to create and enter into diverse relationship forms, including polyamory. For individuals who undergo gender transition, this also challenges traditional notions of fixed gender identities (Hammack et al., 2018). Increasing recognition exists of sexual fluidity in all genders: “Diamond’s (2008) longitudinal study of sexual fluidity in a cohort of sexual minority women found that 67% experienced changes in the target of their sexual attraction over a 10-year period” (Hammack et al., 2018, p. 14). In studying men and women, Katz-Wise (2015) found that 62% of their female sample and 52% of their male sample experienced sexual fluidity. Regardless of how individuals identify, “we need a flexible paradigm that recognizes the diverse ways in which individuals engage with normative discourses of sex, romance, and relationships” (Hammack et al., 2018, p. 28). This paradigm includes recognition of polyamory, for people of all genders and sexual orientations.

Attachment Dimensions as Related to Polyamory and CNM: Empirical Research

As discussed in the introduction, there is a belief that those exhibiting dimensions of insecure attachment, especially avoidant individuals, are attracted to polyamory and CNM. However, few empirical studies have correlated attachment dimensions, interest in, and practice of polyamory and CNM (Fern, 2020). It was not possible to find literature regarding swinging, open relationships, and the attachment dimensions of those who practice it. The research regarding infidelity and attachment is mixed; some studies illustrate that avoidantly attached individuals are more likely to commit infidelity (DeWall et al., 2011), while others have found that a partner’s anxiety will lead to infidelity (Russell et al., 2013).

Moors et al. (2015) conducted two studies: one which sampled 1281 heterosexual monogamous individuals and one which sampled 1308 people, 15% of whom were either swingers or in polyamorous relationships. The first study found that “avoidance correlated positively with attitudes toward CNM and willingness to engage in CNM” (Moors et al., 2015, p. 228). In the study that included those in CNM relationships, those individuals were less likely to express dimensions of high avoidance than those in monogamous relationships, and anxiety was not a significant factor for either type of relationship (Moors et al., 2015). In addition, in a study of 108 individuals who identified as lesbian, gay, bisexual, or pansexual, results found that those expressing dimensions of anxious attachment held negative attitudes about CNM; however, they still engaged in CNM relationships (Moors et al., 2017).

Ka, Bottcher, and Walker mention that “anxious individuals may be likely to agree to engagement in CNM, in order to seek additional affection and attention from multiple partners. Also, they may be willing to engage in CNM, at their partner’s request” (2020, para. 30). Ka et al. (2020) had similar findings to the above study; the majority of their 140 participants lived within dimensions of secure attachment and those exhibiting dimensions of anxious attachment still practiced polyamory and other types of CNM. Although written from a monogamous standpoint, Hazan and Shaver (1987) acknowledged that individuals who exhibit both secure and anxious attachment dimensions agreed with the idea that feelings can change throughout a relationship. Thus, anxiously attached individuals may have a degree of fluidity within their relationships that allows for multiple relationships that shift over time.

Morrison et al. (2013) examined the attachment dimensions of 390 people, 47.9% of whom identified as polyamorous and 29% of whom identified as monogamous. Fifty-two percent of the polyamorous individuals and 44% of the monogamous individuals were securely attached (Morrison et al., 2013). Ten percent of the polyamorous individuals and 19% of the monogamous individuals exhibited dimensions of avoidant attachment, while 20% of the polyamorous individuals and 13% of the monogamous individuals exhibited dimensions of anxious attachment. As the polyamorous sample exhibited secure attachment at a higher rate and avoidant attachment at a lower rate, the authors concluded that their polyamorous sample was more comfortable with intimacy than the monogamous sample (Morrison et al., 2013). However, there is no further discussion as to why the polyamorous sample exhibited dimensions of anxious attachment at a higher rate.

Moors, Ryan and Chopik (2019, p. 108) studied 357 people, all of whom had at least 2 partners and found that the “mean levels scores of avoidance and anxiety for both partners were lower than established norms for avoidance and anxiety.” The authors mentioned that the first relationship, generally longer than any subsequent relationships, was more secure, casting doubt on the idea that those who enter into polyamory are unsatisfied by their primary relationship. When participants lived within dimensions of high avoidance or high anxiety, their relationship satisfaction was lower. The authors supported the notion that attachment is not static; if one relationship was higher in avoidance or anxiety, the other was often secure.

In addition, there are other studies which measure satisfaction within polyamorous and other CNM relationships. Garner et al. (2019) studied 189 individuals, where 57.7% identified as polyamorous, 13.8% in an open relationship, and 10.1% identified as swingers. On the five-point Relationship Assessment Scale (RAS), those in CNM relationships reported a mean score of 4.12, demonstrating high relationship satisfaction for those individuals (Garner et al., 2019). Studies of gay men in open relationships have similar findings; according to Rubel and Bogaert (2015, p. 977), gay male couples in open relationships “had higher dyadic adjustment, as measured by the DAS, as compared to gay male couples in which at least one partner had broken the agreement to be monogamous”. Where discussion about an open relationship had taken place, couples reported greater relationship satisfaction. Mitchell (2014, p. 1458) sampled 361 gay male couples, 84% of whom had sexual agreements that they adhered to, and 31% of which had one or both members engaged in casual relationships with other men. Mitchell (2014, p.1462) concludes that “having a sexual agreement may help both partners of the couple improve their communication and related relationship skills over time, which in turn, may help them have more satisfying and fulfilling relationships”. When a couple decides to become non-monogamous, they are required to communicate about outside relationships, which strengthens the primary relationship.

The empirical research suggests that individuals exhibiting dimensions of avoidant attachment are not likely to engage in polyamory and other forms of CNM. In fact, polyamorous and other CNM relationships have some of the characteristics associated with secure attachment, including constructive communication and high levels of satisfaction. Lessin et al. (2005) argue that effective communication leads to a greater sense of safety within relationships. When individuals exhibiting dimensions of insecure attachment become mired in demand-withdraw communication, for example, they may experience difficulties in articulating their needs in multiple relationships. Finally, individuals exhibiting dimensions of high avoidance often find it difficult for others to depend on them, which is not conducive to multiple satisfying relationships.

Unpacking the Stigma Toward Insecure Attachment and Non-Monogamy

Discussions related to dimensions of avoidant and anxious attachment require more nuance. In their study, Fraley et al., (2011a, p. 619) found that “people who tend to be more avoidant in their relationships with their mothers, for example, also report being more worried about their mother’s availability and responsiveness”. This suggests overlap between avoidance and anxiety, as anxiously attached individuals are often concerned about others’ availability. As well, anxiously attached individuals may become attracted to avoidantly attached individuals (Levine & Heller, 2010), despite the fact that neither of these individuals can meet each other’s needs. Regardless of whether an adult is monogamous or polyamorous, they will likely have a number of relationships in their lifetime (Schacht & Kramer, 2019). While a person may have grown up to be insecurely attached, romantic partners have the power to change that. In this sense, partners who complement each other’s attachment dimensions are more likely to have satisfying relationships: for example, a securely attached person partnered with an avoidantly attached person may be comfortable with limits on intimacy.

Fern (2020) found that many behaviours attributed to insecure attachment dimensions are present within polyamory and other types of CNM, including “casual sex, one-night stands, sex outside of marriage, multiple sexual partners, partaking in bondage, voyeurism, exhibitionism and even sexting” (p. 118). Attachment researchers use these behaviours to stigmatize CNM, assuming that intimacy is not present if a person is away from their primary partner. However, individuals engaging in sex outside of their primary relationship and BDSM may still “play in intentional, highly attuned, connected and meaningful ways” (Fern, 2020, p. 119). Much like polyamory, there is evidence that BDSM practitioners are predominantly securely and anxiously attached (Ten Brink et al., 2021).

As stated, there is little discussion regarding why individuals exhibiting dimensions of anxious attachment practice polyamory and other types of CNM. Fern (2020) believes that anxiously attached individuals are more aware of their feelings and those of their partners, and that they value “connection and togetherness” (p. 62). Ein-Dor and Hirschberger (2016) assert that anxious attachment leads to sensitivity to environmental cues, which may be useful in situations with a complex web of relationships. According to Ein-Dor, “when insecure team members are in a reassuring environment that accepts them and let them feel safe and trusted, their challenging relationship-related perceptions and behaviors might be turned into advantages” (2014, p. 5). In situations where anxiously attached individuals experience safety and can turn to multiple people, their needs are more likely to be met and they can provide care for others.

Why does the psychological literature hypothesize that individuals exhibiting dimensions of avoidant attachment are attracted to polyamory? Although such individuals may be attracted to polyamory because they perceive that they can dilute closeness through the ability to have multiple relationships (Moors et al., 2017), the practice of polyamory involves emotional intimacy and the ability to be vulnerable, which may be difficult for those who exhibit dimensions of high avoidance (Brunning, 2018). Shaver and Hazan argue that individuals exhibiting dimensions of high avoidance are less willing to provide care and they are not comfortable with “intimate interaction” (1988, p. 487), due to a history of often neglectful and/or abusive caregiving. Thus, approaching polyamory while living with dimensions of high avoidance may prove challenging.

Individuals who enter into polyamory practice it in different ways, as mentioned above. However, due to the emphasis on romantic and sexual monogamy within psychology and allied mental health fields, polyamory is viewed as insecure and unfulfilling, regardless of the attachment dimensions of the individuals involved (Fraley, 2019; Zola, 2007). Even when a couple enjoys a secure relationship and they openly decide to become polyamorous (Kolmes & Witherspoon, 2017), subsequent relationships are viewed with suspicion. This ignores the fact that the additional relationships bring joy to the couple, which will often enhance the attachment. Where there is understanding that individuals engaging in polyamory can exhibit dimensions of secure attachment, we can begin to take apart the assumption that non-monogamy is an inherently insecure path.

Directions for Future Research

While attachment theory is important for understanding relationships, it is time to revise it to include relationships outside of the dyadic norm. The Adult Attachment Interview (AAI) is considered to be the gold standard in measuring adult attachment (Krahé et al., 2018). Developed by George, Kaplan, and Main (1996) in order to understand the continuity of infant attachment dimensions into adulthood, it does not account for the possibility that adults had more than two parents or that they enter into relationships that differ from the monogamous norm. As discussed above, there are few longitudinal studies that examine the continuity of infant attachment dimensions into adulthood and that account for stressful events in adults’ lives that do not involve their parents and caregivers (Cassidy, 2000). Any new interviews or studies that are developed should account for the fact that most adults have multiple relationships that end, whether or not they are polyamorous. It should also be recognized that some children are raised by individuals who are not their parents, and that some children have more than two parents.

While it seems clear that individuals of varying attachment dimensions are attracted to polyamory, further research is needed. There is little analysis and empirical research concerning why individuals exhibiting dimensions of high anxiety enter into polyamory. Understanding their motivations and how they fare within polyamorous relationships would address this question. For individuals who are currently in polyamorous relationships, large scale research on attachment dimensions is needed. For securely attached individuals, research can explore how their attachment dimension impacts their partners who are insecurely attached, and vice-versa. Studying how individuals of varying attachment dimensions communicate within polyamorous relationships, and which strategies are most effective would deepen our understanding of attachment dimensions. Fraley et al., (2011a, 2011b) argue that when individuals experience a large shift in their attachment style, it is temporary; understanding how that impacts relationships would provide a deeper understanding of polyamory.

Conclusion

This article has begun to unpack the assumption that secure attachment only exists within monogamous relationships. As illustrated, practicing polyamory often requires characteristics of secure attachment, as such individuals are more likely to communicate productively and be self-aware. Having the ability to constructively articulate complicated emotions such as jealousy may be the key to successfully navigating multiple relationships. However, understanding why a person enters into polyamory is also critical; if a person wishes to have multiple casual connections, they should not be negatively judged for doing so if they are honest and open. In an age where infidelity still carries a stigma, having the bravery to enter into polyamorous and other CNM relationships should be acknowledged.

While the psychological literature is steeped in mononormativity and judgment regarding insecure attachment, unpacking these biases will take time. In general, caregivers make a great effort to ensure that they develop healthy relationships with their children, whether biological or not. Although children may exhibit dimensions of secure attachment as a result of consistent and responsive caregiving, they may have adult relationships that erode it. According to Stuchell (2013), monogamy is “found in only 15–18% of societies worldwide” (p. 839). Given the prevalence of non-monogamy, judgment should not be levied upon those who decide to openly engage in multiple sexual relationships. Understanding and unpacking the origins of this stigma will allow this field of research to move forward.