Abstract: Somnophilia refers to the interest in having sex with a sleeping person. Using an online sample of 437 participants, the present study provides the first empirical examination of somnophilia, its various forms, and theorized correlates. Participants completed the newly developed Somnophilia Interest and Proclivity Scale, which comprises three subscales (active consensual, passive consensual, and active nonconsensual somnophilia). To test hypotheses about the convergent and divergent validity of different paraphilic interests, participants also completed scales measuring necrophilic, rape-related, and sadistic/masochistic sexual fantasies, rape proclivity, and the need for sexual dominance/submission. Male participants scored higher than females on all scales except the passive subscale. For both males and females, each subscale was associated most strongly with conceptually congruent variables. These results support existing theoretical assumptions about somnophilia, as well as offering newer insights, such as distinguishing between active and passive somnophilia. Limitations and implications for further research are discussed.
Keywords somnophilia, necrophilia, paraphilia, biastophilia, sexual fantasy, dormaphilia
How Common Is Somnophilia?
Somnophilia has been termed a rare paraphilia (Lauerma, 2016). Yet, to the authors’ knowledge, no empirical studies have directly examined the prevalence of somnophilia per se in either community or forensic populations. A study by Joyal et al. (2015) did, however, provide some insight. Using an online sample of 1,516 community adults (799 females; 717 males), they examined the prevalence of 55 different sexual fantasies. Fantasies about “sexually abusing a person who is drunk, asleep, or unconscious” were found to be used more frequently by males than females (22.6% vs. 10.8%, respectively). As Joyal et al. (2015) note, the prevalence of these sexual fantasies in women was statistically unusual (<16%), which was not the case for men. It should be highlighted, however, that the fantasy item was framed in offending terms (i.e., “abusing a person who is drunk, asleep, or unconscious”). Had it been framed in neutral or consensual terms, the results may have been different. Moreover, the fantasy item was not specific to sleep, but instead included two other passive states (i.e., drunk and unconscious). Thus, the specific rate of somnophilic fantasies (involving just a sleeping person) cannot be accurately established, as any participants with sexual fantasies about sex with a drunk person, for example, would have also responded to this item.
Discussion
In this study, we assessed participants’ interest and proclivity to engage in active somnophilia (i.e., having sex with a sleeping person), in both consensual and nonconsensual contexts, as well as consensual passive somnophilia (i.e., being the recipient of sexual activity while asleep). This was achieved using a newly developed measure termed the SIPS, which showed convergence with corresponding sleep-related sexual fantasies. The SIPS data were then used to test the main theorized assumptions about somnophilia present within the literature. These included whether each aspect of somnophilia is associated with (a) necrophilic fantasies, (b) rape-related variables (i.e., biastophilic fantasies and fantasies of being raped, as well as rape proclivity in males), and (c) sexually sadistic/masochistic fantasies and the need for sexual dominance/ submission.
Males were found to have higher scores than females on all SIPS subscales, except for the passive subscale. This extends upon Joyal et al.’s (2015) finding, as the present gender difference was focused more specifically on somnophilia (i.e., being asleep) and included consensual somnophilic acts (not just abusive/nonconsensual behaviors). Given these results, male and female participants were subsequently analyzed separately. For both males and females, correlation and regression analyses showed that active somnophilic fantasies were most strongly associated with both active somnophilic subscales, while passive somnophilic fantasies were most strongly associated with the passive subscale. These results provide convergent and construct validity for the SIPS, along with the factor analysis results.
The findings also provided some support for the link between somnophilia and necrophilia (Calef & Weinshel, 1972; Fedoroff et al., 1997; Peck, 2006; Pettigrew, 2017). That is, necrophilic fantasies were associated with the nonconsensual SIPS subscale in male participants. However, necrophilic fantasies did not remain significant for the active SIPS subscales after the first stage of the hierarchical regression. Future research could look to see whether necrophilic fantasies play a mediating role in the link between somnophilic fantasies and somnophilic proclivity. Also, given that the lack of consent potentially plays a role, future research could examine whether the link between necrophilia and nonconsensual somnophilia is driven by having a sexual interest in passive targets (Pettigrew, 2017, 2019b).
The results also revealed that nonconsensual active somnophilia (and consensual active somnophilia to a lesser degree) is associated with biastophilic sexual fantasies in both males and females (as well as rape proclivity in males). These results provide support for the view that somnophilia is linked to an interest in nonconsensual sex (Lauerma, 2016; Pettigrew, 2019a). However, the results of the hierarchical regression revealed that rape proclivity in males and biastophilic fantasies in females remained significant independent variables for nonconsensual active somnophilia only. Thus, while rape-related variables are correlated with each form of somnophilia, they may play a more central role in nonconsensual somnophilia. These findings suggest that it may be beneficial to distinguish between an interest in consensual and nonconsensual somnophilic behavior. Although both are primarily driven by somnophilic fantasies, having an interest and proclivity to engage in biastophilic behaviors may shape someone’s somnophilic interest so that it includes a lack of consent.
Sadistic fantasies and the need for sexual dominance were correlated positively with both forms of active somnophilia. Interestingly, however, the hierarchical regression showed that using sadistic fantasies less frequently was a significant independent variable of nonconsensual somnophilia in both males and females. This perhaps suggests that an interest in more aggressive sexual acts is not linked to nonconsensual somnophilia. In light of these results, it could be argued that that nonconsensual somnophilia is driven by an underlying interest in “passivity, power, and the elimination, in degrees, of the possibility of rejection” (Pettigrew, 2017, p. 353), rather than an overt interest in sexual aggression. Further research is needed to understand the role of sexual sadism in relation to somnophilia. Perhaps it functions as a mediator between somnophilic fantasies and consensual somnophilia proclivity while playing less of a role in the proclivity to engage in nonconsensual somnophilia.
As expected, passive somnophilia was associated most strongly with passive-oriented variables. For example, sexual fantasies about being the recipient of sex while asleep was the strongest correlate in both males and females. In addition, fantasies of being raped were associated with passive somnophilia, particularly in females. It is possible that those who are aroused by being the passive recipient are striving to be totally submissive to their partner (Knafo, 2015). Indeed, the need for sexual submission and the use of masochistic fantasies were both strongly correlated with the passive subscale. Also, masochistic fantasies emerged as a significant independent variable in the hierarchal regression for females. However, passive somnophilic fantasies remained the strongest independent variable in the regression analysis for both males and females. Of course, all these data are correlational and so do not imply causal relationships between the variables.
Limitations
Although this study provides some important and useful insights into somnophilia, a number of limitations should be noted. First, we did not ask participants whether they actually engage in consensual somnophilic behavior. Doing so would have offered the opportunity to compare those who act upon their interest against those who do not. Further to this, it would have been useful to have asked participants where they had seen the study posted. From our data, we were unable to isolate and quantify how many participants were collected from fetish sites and forums related to somnophilia. This information would have helped to establish an approximate prevalence rate of somnophilia within the general population versus those recruited from somnophiliarelated sites. Moreover, the study was advertised as being about somnophilia and so the biases that come with recruiting a self-selected sample are likely present in the results. Thus, some caution is warranted when drawing conclusions from the findings. Related to this, many of the participants in this self-selected sample are unlikely to have a sexual offending background as such individuals commonly have internet use prohibitions. Therefore, the results cannot be generalized to those who have actually engaged in paraphilic offending of this nature. However, it should be noted that such prohibitions are not universal and may not always be followed by the person with a history of sexual offending. Thus, future research is needed to gain a wider picture of somnophilic-driven offending behavior.
In addition, the consensual active and passive somnophilia subscales of the SIPS were each comprised of two scenarios, both of which involved a romantic partner. The nonconsensual subscale comprised only one scenario, which involved a nonpartner (i.e., housemate). Thus, the SIPS could be amended to include a nonconsensual scenario involving one’s romantic partner for consistency. Similarly, future researchers may want to consider investigating nonconsensual passive somnophilia by developing and using a relevant scenario. We chose not include this in the present study as it was deemed conceptually problematic (i.e., having a proclivity for being a recipient of nonconsensual somnophilic behavior). However, as some participants reported fantasies of being raped, it is possible that some individuals will also harbor an interest in (rather than a proclivity for) nonconsensual passive somnophilia—especially within a self-selected online sample.
As the SIPS is a self-report measure, it is susceptible to socially desirable responding. Future research should look to examine the SIPS’s relationship with an impression management measure, as well as its convergence with an indirect (i.e., response-latency) measure of somnophilic interest. Doing so, along with a knowngroups comparison, will aid toward validating the SIPS further. A final potential issue, as with many online studies, is that some participants may not respond plausibly, may not pay attention, and/or may respond spuriously in order to get to the end of the study. Although the Mahalanobis Distance analysis helped identify participants who were outliers across multiple measures, this issue was not explicitly accounted for within the design of current study (e.g., adding questions that can highlight implausible responding, asking whether the participant answered truthfully, using attention checks). This is recommended for future research on this topic.
Posts
