Extreme rituals in a BDSM context: the physiological and psychological effects of the ‘Dance of Souls’

Abstract

Participation in extreme rituals (e.g., fire-walking, body-piercing) has been documented throughout history. Motivations for such physically intense activities include religious devotion, sensation-seeking and social bonding. The present study aims to explore an extreme ritual within the context of bondage/discipline, dominance/submission and sadism/masochism (BDSM): the ‘Dance of Souls’, a 160-person ritual involving temporary piercings with weights or hooks attached and dancing to music provided by drummers. Through hormonal assays, behavioural observations and questionnaires administered before, during and after the Dance, we examine the physiological and psychological effects of the Dance, and the themes of spirituality, connectedness, transformation, release and community reported by dancers. From before to during the Dance, participants showed increases in physiological stress (measured by the hormone cortisol), self-reported sexual arousal, self-other overlap and decreases in psychological stress and negative affect. Results suggest that this group of BDSM practitioners engage in the Dance for a variety of reasons, including experiencing spirituality, deepening interpersonal connections, reducing stress and achieving altered states of consciousness.

Full text

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=tchs20
Download by: [67.173.62.19] Date: 14 March 2017, At: 05:55
Culture, Health & Sexuality
An International Journal for Research, Intervention and Care
ISSN: 1369-1058 (Print) 1464-5351 (Online) Journal homepage: http://www.tandfonline.com/loi/tchs20
Extreme rituals in a BDSM context: the
physiological and psychological effects of the
‘Dance of Souls’
Kathryn R. Klement, Ellen M. Lee, James K. Ambler, Sarah A. Hanson, Evelyn
Comber, David Wietting, Michael F. Wagner, Valerie R. Burns, Bert Cutler,
Nadine Cutler, Elwood Reid & Brad J. Sagarin
To cite this article: Kathryn R. Klement, Ellen M. Lee, James K. Ambler, Sarah A. Hanson,
Evelyn Comber, David Wietting, Michael F. Wagner, Valerie R. Burns, Bert Cutler, Nadine Cutler,
Elwood Reid & Brad J. Sagarin (2017) Extreme rituals in a BDSM context: the physiological and
psychological effects of the ‘Dance of Souls’, Culture, Health & Sexuality, 19:4, 453-469, DOI:
10.1080/13691058.2016.1234648
To link to this article: http://dx.doi.org/10.1080/13691058.2016.1234648
View supplementary material Published online: 14 Oct 2016.
Submit your article to this journal Article views: 80
View related articles View Crossmark data

CULTURE, HEALTH & SEXUALITY, 2017
VOL. 19, NO. 4, 453469
http://dx.doi.org/10.1080/13691058.2016.1234648
Extreme rituals in a BDSM context: the physiological and
psychological eects of the ‘Dance of Souls’
Kathryn R. Klementa, Ellen M. Leea, James K. Amblera, Sarah A. Hansonb,
Evelyn Combera, David Wiettinga, Michael F. Wagnera, Valerie R. Burnsc, Bert Cutlerd,
Nadine Cutlerd, Elwood Reide and Brad J. Sagarina
aDepartment of Psychology, Northern Illinois University, DeKalb, USA; bDepartment of Sociology, Georgia
State University, Atlanta, USA; cDepartment of Counseling Psychology, University of Wisconsin-Madison,
Madison, USA; dIndependent Scholar, Tempe, Arizona, USA; ePrimal Edgewalkers Services Inc., Edmonton,
Alberta, Canada
ABSTRACT
Participation in extreme rituals (e.g., re-walking, body-piercing) has
been documented throughout history. Motivations for such physically
intense activities include religious devotion, sensation-seeking and social
bonding. The present study aims to explore an extreme ritual within
the context of bondage/discipline, dominance/submission and sadism/
masochism (BDSM): the ‘Dance of Souls’, a 160-person ritual involving
temporary piercings with weights or hooks attached and dancing to
music provided by drummers. Through hormonal assays, behavioural
observations and questionnaires administered before, during and after
the Dance, we examine the physiological and psychological eects of the
Dance, and the themes of spirituality, connectedness, transformation,
release and community reported by dancers. From before to during
the Dance, participants showed increases in physiological stress
(measured by the hormone cortisol), self-reported sexual arousal,
self-other overlap and decreases in psychological stress and negative
aect. Results suggest that this group of BDSM practitioners engage in
the Dance for a variety of reasons, including experiencing spirituality,
deepening interpersonal connections, reducing stress and achieving
altered states of consciousness.
Early psychiatry viewed bondage/discipline, dominance/submission and sadism/masochism
(BDSM) as deviant sexual practices, rst mentioned by Krat-Ebing (1965) in his Psychopathia
Sexualis, and then interpreted by Freud as a method of expressing repressed desires
(Weinberg 1994). Many modern researchers, in contrast, tend to view BDSM as a fairly com-
mon sexual interest (Faccio, Casini, and Cipolletta 2014; Joyal, Cossette, and Lapierre 2015)
performed by psychologically healthy individuals (Connolly 2006) with favourable psycho-
logical characteristics (Wismeijer and Van Assen 2013). Mainstream depictions of BDSM often
focus on the sexual aspects of its practice, and the majority of research on BDSM has studied
© 2016 Informa UK Limited, trading as Taylor & Francis Group
KEYWORDS
Sadomasochism; extreme
rituals; Dance of Souls;
spirituality; cortisol;
sexuality; USA
ARTICLE HISTORY
Received 24 April 2016
Accepted 6 September 2016
CONTACT Kathryn R. Klement Kathryn.Klement@gmail.com
This article was originally published with errors. This version has been corrected. Please see Corrigendum (http://dx.doi.
org/10.1080/13691058.2017.1256628).
Supplemental data for this article can be accessed here.

454 K.R. KLEMENT ET AL.
BDSM in this context. However, despite common depictions in popular culture media, the
practice of BDSM is not always sexual. For example, Wiseman’s (1996) denition of BDSM
includes the purpose of ‘erotic arousal’, but also ‘personal growth’ (10), Baumeister (1997)
proposed a theory of masochism as escape from self and Newmahr (2010) suggested that
sadomasochism is a form of ‘serious leisure’.
The present study sought to explore a setting where BDSM might be practised in a
non-sexual context. To that end, we collected a range of quantitative and qualitative data
at the ‘Dance of Souls,’ a hook-pull/ball-dance held at the annual Southwest Leather
Conference (SWLC) in Phoenix, Arizona. The Dance of Souls is an extreme ritual that involves
temporary piercings where either weights (e.g., bells, fruit) are attached to the piercings or
hooks are placed into the piercings with ropes attached to the hooks to allow pulling against
the piercings. The participants then dance to music provided by drummers.
Historically and presently, many people engage in extreme ritual practices (e.g., re-walking,
body-piercing; Catlin 1867; Fischer et al. 2014). Motivations to engage in these rituals include
religious devotion, sensation-seeking and social bonding (Fischer et al. 2014; Xygalatas, Schjødt
et al. 2013). Extreme rituals are characterised by physically demanding tasks performed within
a social context, usually with fellow practitioners or interested observers. Research has docu-
mented several eects of ritual participation, including physiological synchrony with related
observers (Konvalinka et al. 2011), autobiographical memory decits (Xygalatas, Schjødt
et al. 2013) and increases in prosocial behaviour (Xygalatas, Mitkidis et al. 2013).
Konvalinka et al. (2011) investigated the eects of a re-walking ritual on those engaging
in the ritual and those observing it. Konvalinka et al. measured the heart rates of the observ-
ers and participants, then compared video footage and heart rate readings. They found that
related observers’ heart rates synchronised with the participants’ as they were walking over
the hot coals. This indicates that participation and observation of an extreme ritual may
invoke a type of arousal-related empathy that can strengthen bonds between individuals
who go through and observe the experience.
Xygalatas, Mitkidis et al. (2013) explored how autobiographical memory is altered by engaging
in an extreme ritual. Participants’ physiological arousal was measured via heart rate prior to and
during a re-walking event. After participants completed the ritual, they were asked to recall
their level of experienced arousal. Although the objective heart rate data showed an increase
in arousal from before to during the ritual, participants reported a decrease in experienced
arousal. Xygalatas, Schjødt et al. (2013) suggested that this inconsistency between objective and
subjective data was due to the cultural norm of suppressing emotion during the event.
The Dance of Souls
As discussed above, the Dance of Souls is an annual event at the SWLC. The Dance shares
elements of the Plains Native American Sundance (also known as the O-Kee-Pa ceremony
[Catlin 1867]) and the Hindu Thaipusam festival of the Tamil communities (Musafar 2002;
Pfa and Simons 1973). In these traditions, devotees pierce their esh, using body stress to
induce ecstatic states.
The introduction of hook pulls/ball dances to the sadomasochistic and body modication
communities is often attributed to Fakir Musafar. In 1999, Musafar and his partner, Cleo
Dubois, taught a two-day workshop culminating in a ball dance at the Arizona Power
Exchange. The rst hook pull/ball dance hosted by the SWLC was held in 2004. In 2006,
Elwood Reid, a protégé of Musafar, became the leader of the hook pull/ball dance at SWLC.

CULTURE, HEALTH & SEXUALITY 455
We sought to expand knowledge of extreme rituals through a multi-method investigation
of the Dance of Souls. Consistent with previous research that assessed both objective and
subjective measures of arousal in ritual participants (Konvalinka et al. 2013), we measured
physiological stress through the hormone cortisol (via saliva samples) and psychological stress
through self-report measures of stress, positive aect and negative aect. To investigate eects
of the ritual on social bonding, we measured participants’ reported self-other overlap and also
analysed qualitative data from open-ended questions regarding motivations for and benets
of engaging in the Dance. To examine participants’ conceptualisation of the Dance, we asked
them how sexual, sadomasochistic and spiritual they found the Dance to be. We planned to
examine cognitive changes during the Dance as well, but due to problems with the cognitive
test that we used (the Wisconsin Card Sorting Test showed a large proportion of errors and
discrepant results in the within-subject and between-subjects analyses), the results were unin-
terpretable. Finally, we recorded behavioural observations throughout the event.
We anticipated that the Dance of Souls might oer insights into some of the non-sexual
manifestations of BDSM. Further, we anticipated that prior research on the eects of BDSM
activities might inform predictions regarding the eects of the Dance of Souls. In contrast
to prior studies of explicitly religious rituals (e.g., Xygalatas et al. 2013) or annual community
rituals (e.g., Fischer et al. 2014; Konvalinka et al. 2011), the Dance of Souls takes place as the
capstone of a conference that caters specically to the leather/BDSM community. Within
this community, the aspect of pain may take on a dierent meaning.
Hypotheses
Observed behaviours
Based on our expectation that this event would be considered BDSM activity separate from
sexuality, we predicted that we would observe sadomasochistic activities but not overtly
sexual activities.
The eects of the Dance on participants
Based on Sagarin et al.’s (2009) nding that cortisol increased in bottoms (BDSM participants
who are bound, receiving stimulation, following orders, etc.) during BDSM scenes, and con-
sistent with previous research on extreme rituals (e.g., Konvalinka et al. 2011; Xygalatas,
Mitkidis et al. 2013), we predicted that cortisol levels would increase but that self-reported
psychological stress would decrease from before to during the Dance. Based on Sagarin
et al.’s (2009) nding that self-other overlap increased during BDSM scenes, we predicted
that self-other overlap would increase from before to during the Dance. Based on anecdotal
accounts from past Dance participants that described the Dance as an enjoyable experience,
we also predicted that positive aect would increase and negative aect would decrease
from before to during the Dance.
Participants’ conceptualisation of the Dance
Again, due to our expectation that the Dance was a BDSM activity separate from sexuality,
we predicted that participants would rate the Dance more sadomasochistic than sexual. We
did not have a specic prediction regarding participants’ ratings of the Dance as spiritual.

456 K.R. KLEMENT ET AL.
Methods
Participants
Participants were attendees at the 2012 SWLC who had signed up for the Dance of Souls
(not all conference attendees participated in the Dance of Souls). Out of the approximately
160 participants in the Dance of Souls, 67 individuals contributed at least one measure to
the study.
All demographic questions were open-ended. Responses were independently coded
by two researchers (discrepancies resolved through discussion). The sample contained
33 women, 26 men and 3 transgendered individuals (5 participants did not report gender
identity). The mean age was 49.16 (SD = 10.34). Most participants identified as Caucasian
(n = 57); 4 reported being bi-racial and 5 indicated other (e.g., ‘Spanish gypsy’). Sexual
identities included heterosexual (n = 19), homosexual (n = 11), heteroflexible (n = 10)
and bisexual (n = 12). Participants’ BDSM roles were relatively evenly split among tops
(n = 25), bottoms (n = 21) and switches (n = 16). Most participants reported their reli-
gious affiliation as being outside mainstream monotheistic religion; the modal response
was ‘spiritual’ (n = 12).
Procedure
During the conference, the researchers set up a table in a well-trafficked area to invite
individuals to enroll in the study. Interested individuals were given an informed consent
form to read, and if they agreed to participate, they chose a unique ID to track their
participation throughout the study. It was emphasised that participants were under no
obligation to complete all of the measures. Participants were then given the pre-Dance
survey.
At the beginning of the Dance, a table was set up in the room where the Dance was being
held. Participants standing in line to be pierced were asked to contribute saliva samples for
cortisol testing. A researcher would give a vial to the participant and instruct the participant
to remove the swab, hold it under their tongue for one minute, deposit it back into the vial,
and write their ID on the vial. Vials were then stored on ice in a cooler.
After the piercing was concluded, the data collection table was moved to the side of the
Dance area. Throughout the Dance, researchers collected further saliva samples and admin-
istered during-Dance surveys.
The post-Dance survey was available on paper immediately following the Dance and
online three months later when the conference organisers sent an email to all attendees
with a link to the survey.
Materials
Pre-Dance survey
The pre-Dance survey began with demographic questions, including gender, sexual orien-
tation, age, race/ethnicity, religion/spirituality and BDSM/leather role(s). Participants were
asked to identify the role(s) they were planning on fullling during the Dance (Dancer with
hooks, Dancer with weights, Unpierced dancer, Observer, Drummer, Piercer or Dance leader).

CULTURE, HEALTH & SEXUALITY 457
They were also asked if they had participated before in a similar event and which role(s) they
played. Two open-ended questions asked: ‘Why are you participating in the Dance of Souls?’
‘What do you hope to gain from the experience?’
Participants were then asked to rate the extent to which they would use the word ‘we’ to
describe their relationship with other people in the Dance (1 = I denitely would not use the
word ‘we’ to describe my relationship and 7 = I denitely would use the word ‘we’ to describe my
relationship). Following this was Aron, Aron and Smollan’s (1992) Inclusion of Other in Self
(IOS) scale. Participants were shown seven sets of two circles in various degrees of overlap;
one circle represents the self and the second circle represents other(s). Participants were
asked to choose which set of overlapping circles best describes their relationship at the
present moment with other people in the Dance. The pre-Dance survey concluded with the
Positive and Negative Aect Scale (PANAS; Watson, Clark, and Tellegen 1988). The PANAS
asked participants to rate the extent to which they felt 20 emotions at the present moment
on a scale of 1 (very slightly or not at all) to 5 (extremely). Two additional feelings were included:
‘sexually aroused’ and ‘stressed’.
During-Dance survey
The during-Dance survey began with the PANAS, ‘sexually aroused’ and ‘stressed’. Participants
were then asked in an open-ended format what they were doing just before starting the
survey and their perception of the challenge of that activity and their skill level in that activity
on a scale of 1 (none) to 5 (very high). These questions were designed to assess ‘ow’
(Csikszentimihályi 1990). The last question asked participants to share any current thoughts
or feelings in an open-ended format.
Post-Dance survey
The post-Dance survey rst asked participants to identify which role(s) they took on during
the Dance. If they were a dancer with hooks or weights, they were asked how many weights
or hooks they used. Participants were then asked to describe their activities during the Dance.
Next, participants were asked in an open-ended format, ‘How do you feel about the Dance?’
‘Did it go well?’ ‘Did it go poorly?’
Participants were then asked ‘How much did you lose yourself during the Dance?’ (1 = not
at all and 5 = very much), ‘How spiritual was the Dance for you?’, ‘How sexual was the Dance
for you?’ and, nally, ‘How sadomasochistic was the Dance for you?’, all on a scale of 1 (not
at all) to 5 (extremely). Following this was the scale assessing participants’ use of the word
‘we’ to describe their relationship with others at the Dance, the IOS and retrospective versions
of the PANAS and the ‘sexually aroused’ and ‘stressed’ questions. Finally, demographic ques-
tions were included for participants who had not completed a pre-Dance survey.
Saliva samples
Saliva sample assays were performed using immunoassay kits purchased from Salimetrics,
LLC.
All saliva samples were stored on wet ice during the collection procedures and then
at -65 °C for later analysis. Immediately prior to assaying, samples were thawed and

458 K.R. KLEMENT ET AL.
centrifuged (1500 x g at 3000 rpm for 15 min) to separate saliva from any other matter.
Each sample was analysed for cortisol immediately, without freezing the sample again.
Each sample, standard or assay diluent, was pipetted (25 μl) into plates pre-coated with
antibodies for cortisol. The standard for cortisol ranged from 0.012 μg/dl (0.33 nmol/l)
to 3.0 μg/dl (82.77 nmol/l), with an average 2.62% inter-assay coefficient of variation for
the cortisol controls. All samples were assayed in duplicate: r(49) = 0.95, p < .001. The
immunoassays were conducted in accordance with the directions from the Salimetrics,
LLC hormone kits. Briefly, the conjugate was added to each well with the sample, stand-
ard or assay diluent. After 60 minutes, the plate was rinsed and 200 lμl of tetramethylb-
enzidine solution was added. The reaction was stopped after 30 minutes with sulfuric
acid and the plate read with a BioRad E1A Reader Model 2550 within 10 minutes at
wavelength 450 nm.
Results
Preparation of data
There were four administrations of survey materials. Any time prior to the Dance, participants
could complete the pre-Dance survey. The during-Dance survey was available only during
the Dance. After the Dance, paper post-Dance surveys were available to participants. Three
months later, all Dance participants were sent an email with a link to an online version of
the post-Dance survey. Some participants completed the post-Dance survey for the rst
time online while others completed the post-Dance survey for a second time. We attempted
to measure levels of positive aect, negative aect, psychological stress and sexual arousal
that occurred during the Dance in two ways. Items measuring these constructs appeared
on the during-Dance survey and retrospectively on the Post-Dance survey. No signicant
dierences were found in the scores for participants who reported positive aect, negative
aect, psychological stress and sexual arousal in real time and retrospectively (all ps > .09).
Thus, the during-Dance and post-Dance items for aect, psychological stress and sexual
arousal were averaged into one during-Dance variable set.
Because many participants completed only a subset of the measures, analyses were per-
formed both within-subject and between-subjects. In most cases, the within-subject and
between-subjects analyses led to similar results. To avoid redundancy, the results of the
within-subject analyses will be presented and any substantive dierences between the
results of the within-subject and between-subjects analyses will be noted.
Observed behaviours
Dance participants received temporary piercings from experienced volunteers in sanitary
conditions, and had the option to receive hooked piercings in their chest and/or back, or to
receive lament piercings on which weighted objects (e.g., bells, gemstones) were hung.
Music was provided by a drumming group, and pierced and non-pierced dancers were
invited to move to the music. Participants were in various states of undress; the majority of
dancers who had received hooks were topless. Some hooked dancers connected their ropes
in pairs or in groups as large as eight to pull against each other. Participants expressed a
variety of emotions, yelling, grunting and crying at dierent intervals. No overt signs of

CULTURE, HEALTH & SEXUALITY 459
sexual arousal were observed. The dancing lasted approximately three hours, after which
participants began to have their piercings removed.1
The eects of the Dance on participants
Table 1 displays the mean levels of sexual arousal, positive aect, negative aect, self-
reported stress and cortisol before and during the Dance as well as the mean levels of ‘we’ness
and self-other overlap before and after the Dance. Table 2 displays the correlations between
these variables as well as participants’ conceptualisation of the Dance as spiritual, sexual
and sadomasochistic. Pre/during and pre/post correlations were signicant for sexual arousal,
‘we’ness, self-other overlap and positive aect. Thus, for example, participants who were
higher in sexual arousal before the Dance tended to remain higher in sexual arousal during
the Dance.
Sexual arousal
Participants reported a signicant increase in sexual arousal from before the Dance (M = 2.24,
SD = 0.94) to during the Dance (M = 2.78, SD = 1.11), t(33) = -3.75, p = .001. Additionally, there
was a signicant positive correlation between post-Dance ratings of how sexual and sado-
masochistic the Dance was and sexual arousal both before and during the Dance (all ps < .05).
‘We’ relationship, self-other overlap
There was not a signicant dierence between participants’ use of the word ‘we’ to describe
their relationship to others pre-Dance (M = 5.54, SD = 1.50) to post-Dance (M = 5.56,
SD = 1.39), t(23) = -0.09, p = .93. Participants did, however, report a signicant increase in
self-other overlap pre-Dance (M = 3.7, SD = 1.49) to post-Dance (M = 4.87, SD = 1.55),
t(22) = -3.57, p = .002.
Positive and negative aect
Contrary to our hypothesis, participants did not report a signicant increase in positive aect
from pre-Dance (M = 39.06, SD = 6.72) to during-Dance (M = 40.44, SD = 6.34), t(33) = -1.39,
p = .175.
Negative aect, however, decreased signicantly from pre-Dance (M = 13.45, SD = 5.28)
to during-Dance (M = 11.17, SD = 1.63), t(32) = 2.59, p = .014. Thus, even though levels of
Table 1.Within-subjects means for pre-Dance, during-Dance and post-Dance measures.
Note: The variation in sample size is due to the fact that not all participants completed all of the measures.
Variable
Pre-Dance
Mean (SD)
During-Dance
Mean (SD)
Post-Dance
Mean (SD) n p
Sexual arousal 2.24 (0.94) 2.78 (1.11) - 33 .001
‘We’ 5.54 (1.5) - 5.56 (1.39) 23 .929
Self/other overlap 3.7 (1.49) - 4.87 (1.55) 22 .002
Positive affect 39.06 (6.72) 40.44 (6.34) - 33 .175
Negative affect 13.45 (5.28) 11.17 (1.63) - 32 .014
Stress 1.6 (1.11) 1.15 (0.42) - 32 .022
Cortisol .18 (0.15) .46 (0.55) - 12 .064

460 K.R. KLEMENT ET AL.
Table 2.Correlations for self-report measures.
*Indicates significance at the .05 level.
**Indicates significance at the .001 level.
Measure 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. How spiritual –
2. How sexual −.26 –
3. How sadomasochistic (SM) −.20 .28 –
4. Sexual arousal pre-Dance −.09 .49* .65** –
5. Sexual arousal during-Dance −.20 .76** .35* .69** –
6. ‘We’ pre-Dance .32 −.14 .10 .01 .02 –
7. ‘We’ post-Dance .48** .01 −.28 −.32 .01 .69** –
8. Self/other overlap pre-Dance .53** .17 .15 .16 .16 .62** .30 –
9. Self/other overlap post-Dance .42** .11 .03 −.10 .09 .70** .59** .46** –
10. Positive affect pre-Dance .29 −.50* −.35 .21 .04 .15 .04 .33* −.30 –
11. Positive affect during-Dance .23 .08 −.07 .26 .25 .11 .26 .24 .13 .61** –
12. Negative affect pre-Dance .03 .20 −.06 −.18 −.10 −.24 −.29 −.13 .09 −.41** .08 –
13. Negative affect during-Dance .01 −.13 .06 −.08 −.29* −.26 −.19 −.21 −.12 −.29 −.06 .28 –
14. Stressed pre-Dance .08 .26 −.18 −.19 −.06 −.32* −.17 −.18 .14 −.37** .15 .74** .50** –
15. Stressed during-Dance .12 −.26 −.15 −.17 −.20 .01 −.05 .02 .07 −.02 −.01 .05 .51** .26 –

CULTURE, HEALTH & SEXUALITY 461
negative aect were fairly low going into the Dance, participants still reported feeling less
negative aect during the Dance. Positive aect pre-Dance was signicantly negatively cor-
related to pre-Dance negative aect, r(50) = -.41, p = .003. In addition, during-Dance negative
aect was signicantly negatively correlated with sexual arousal during-Dance, r(53) = -.29,
p = .03.
Psychological stress
Participants reported a reduction in self-reported stress pre-Dance (M = 1.60, SD = 1.11)
to during-Dance (M = 1.15, SD = 0.42), t(32) = 2.41, p = .022. There was also a significant
negative correlation between participants’ ratings of feeling stressed and the use of the
word ‘we’ to describe their relationship to others pre-Dance, r = -.32, p = .021. Additionally,
a feeling of pre-Dance stress was significantly related to feelings of positive and negative
affect pre-Dance, r = -.37, p = .007, and r = .74, p < .001, respectively. Taken together
these results indicate that those who felt more stressed going into the Dance also felt
less positive affect, more negative affect and less connected to others before the Dance.
This trend also continued into the Dance with significant correlations between dur-
ing-Dance negative affect and pre-Dance stress, r = .50, p = .003 and during-Dance stress,
r = .51, p < .001.
Physiological stress
Although self-report measures of feeling stressed decreased pre-Dance to during-Dance,
there was a marginally signicant increase in cortisol pre-Dance (M = 0.18, SD = 0.15) to
during-Dance (M = 0.46, SD = 0.55), t(12) = -2.04, p = .064. A between-subjects analysis of
cortisol, which enabled us to include participants who provided only one saliva sample,
revealed a signicant increase in cortisol pre-Dance (M = 0.20, SD = 0.14) to during-Dance
(M = 0.46, SD = 0.52), F(1, 45) = 6.96, p = .011. During-Dance cortisol correlated negatively
with during-Dance positive aect, r(12) = -.70, p = .011, and positively with during-Dance
negative aect, r(12) = .61, p = .037.
Flow
Jackson, Eklund and Martin (2010) identied nine facets of ow including challenge-skill
balance. On the during-Dance survey, participants indicated the level of challenge of their
most recent activity and their skill level on that activity. High endorsement (4s and 5s) on
both items was interpreted as ow. According to these criteria, relatively few participants
(16%) were in a state of ow. However, anecdotal responses from participants suggested
that challenge-skill balance was less central to their experience of the Dance than other
(unmeasured) facets of ow (e.g., loss of self-consciousness, transformation of time, autotelic
experience). Thus, although we observed little evidence of challenge-skill balance, more
general conclusions regarding ow would benet from assessing multiple facets of ow.
Furthermore, because participants completed the during-Dance survey when they were
taking a break from the Dance, it is likely that they were not in a state of ow while they were
lling out the survey.

462 K.R. KLEMENT ET AL.
Participants’ conceptualisations of the Dance
Sexual, sadomasochistic, spiritual
After the Dance was over, participants were asked how sexual, sadomasochistic and spiritual
they found the Dance to be (see Figure 1). Participants reported that they found the Dance
signicantly more spiritual (M = 3.85, SD = 1.17) than sexual (M = 2.30, SD = 1.14) or sado-
masochistic (M = 2.59, SD = 1.22), t(39) = 4.32, p = .001, and t(39) = 5.37, p < .001, respectively.
There was no signicant dierence between ratings of the Dance as sexual versus sadomas-
ochistic, t(39) = -1.28, p = .21.
Ratings of ‘we’ post-Dance and self-other overlap pre- and post-Dance were signicantly
positively correlated with ratings of how spiritual participants found the Dance (all ps < .001).
Open-ended responses
Coding
Two questions from the pre-Dance survey (‘Why are you participating in the Dance of Souls?’
and ‘What do you hope to gain from the experience?’) and comments from the during-Dance
and post-Dance surveys (‘Please write down any current thoughts or feelings you would like
to share’) were coded.
The open-ended data were analysed using inductive analysis (Braun and Clark 2006).
Inductive analysis is a ‘process of coding the data without trying to t [them] into a pre-
existing coding frame’ (83, emphasis original). Thus, the data were analysed for any common
themes that emerged during reading. First, seven graduate students, led by one trained in
qualitative coding, read through all of the responses. After a group discussion, a coding
scheme was created and agreed upon by all seven coders. The coding scheme consisted of
17 codes that the members of the group felt encapsulated the themes mentioned within
the responses. These codes were: altered consciousness; celebration; connectedness to
others; energy; future expectations; generalised aect; growth; knowledge; met expecta-
tions; negative aect; positive aect; release; service; sexuality; spirituality; transformation;
1
1.5
2
2.5
3
3.5
4
4.5
5
How spiritual was the
Dance?
How sexual was the
Dance?
How sadomasochistic
was the Dance?
Post-Dance Ratings
Figure 1.Spiritual, sexual and sadomasochistic ratings of the Dance.

CULTURE, HEALTH & SEXUALITY 463
and unmet expectations. Next, each coder read all the responses and noted whether a
response mentioned each of the codes (yes/no). Another group discussion followed and
any coding disagreements were resolved; thus, inter-rater reliability was not calculated due
to nal convergence on the categorisations.
Content themes
Figure 2 shows the number of times each theme was mentioned across all three open-ended
questions. The top three mentioned themes were: spirituality, connectedness to others and
altered consciousness. Regarding spirituality, this seems to align with the results from the
quantitative analysis where participants rated the Dance as more spiritual than sexual or
sadomasochistic. The qualitative coding also demonstrated that people saw the Dance as
an opportunity to bond with others; this is consistent with the increase in self-other overlap
pre- to post-Dance ratings from the surveys. Importantly, one of the least-mentioned themes
was sexuality and, again, this is consistent with the quantitative ratings.
Pre-Dance responses
On the pre-Dance survey, the most frequently mentioned theme was spirituality; many par-
ticipants described how the Dance was a way to commune with higher powers and experience
a spiritual journey. The other most commonly mentioned themes were connectedness to
others and release. One participant mentioned each of these themes: ‘For the ecstatic expe-
rience, which is always spiritually cleansing and renewing for me, and the chance to share
this with my chosen family/tribe’
2
. Another participant further described how the Dance could
be a bonding experience as well as cathartic: ‘It is a wonderful way to exchange energy not
just with my owner but with my tribe. It is transcendental and amazing. It is also a great way
to release the negative.’ Even though many participants desired, and anticipated, their Dance
experience to be spiritual, one participant humorously described how this might not always
happen: ‘I have had visions and deep spiritual experiences. Or not. But it’s fun anyway.’
0
5
10
15
20
25
30
35
Figure 2.Frequency of coding themes for open-ended responses.

464 K.R. KLEMENT ET AL.
During-Dance responses
The most frequently mentioned themes from the during-Dance survey were altered con-
sciousness and positive aect. Two participants reported ‘Connection through the drums
and the tribe and the shared energy that overrides any pain and opens my heart and mind
to a focus beyond the physical’ and ‘I am in an amazing happy place. Serene yet joyous’.
Post-Dance responses
After the Dance was over, the most frequently mentioned themes were connectedness to
others and altered consciousness. One participant eloquently described how the experiences
of other aected them: ‘Observed energetic releases of others, saw emotions stir in many;
tears, anger, screaming, laughter, pure joy. Many had variety of dierent experiences. I got
to share in all of them, and feel their release.’ Another participant talked about how the Dance
created feelings of connectedness and an altered state of mind:
All throughout the dance, I felt very present and connected with everyone else there, but also
recognised the work I was doing was my own work and needed my attention. I also felt closer
to the spirits of those who have walked this earth before us and the Universe’s energy abuzz
inside me. It was beautiful and powerful.
Discussion
The goal of this study was to examine an extreme ritual within a BDSM context. To this end,
we examined the activities, eects and conceptualisations of the Dance of Souls, a ritual in
which BDSM practitioners received temporary piercings with weights or hooks attached
and danced to music provided by drummers. We used mixed methods to determine whether
the observed activities and their eects aligned with previous research on extreme rituals
in religious and community contexts. First, we observed the behaviours and compared them
to prototypical sexual and BDSM behaviours (please see supplemental materials for this
discussion). Second, we tested whether the Dance of Souls had similar eects on participants
as observed in past research on extreme rituals and BDSM scenes. Third, we asked partici-
pants how they conceptualised their behaviours.
Observed behaviours
We posited that participants in the Dance would demonstrate more sadomasochistic behav-
iour than sexual behaviour, and the behavioural observations supported this hypothesis.
Community denitions of BDSM maintain that sex and sexuality are not necessary elements
of BDSM behaviour. Our behavioural observations did not provide evidence for physical
sexual arousal (i.e., penile erections) and no overt sexual acts were observed. Thus, we inter-
preted the behaviours of participants to indicate that people participated in the Dance
primarily for non-sexual purposes. This might be due to the original intent of the ritual and
the norms and expectations that Musafar imparted when he and those inspired by him
brought the Dance to the SWLC. Some other common behaviour throughout the Dance
included frequent touching, rubbing and talking between Dancers; participants were clearly
providing physical and emotional support to one another as they went through the expe-
rience. Some BDSM community members purport that the desire to feel close to others is a

CULTURE, HEALTH & SEXUALITY 465
central purpose behind BDSM behaviour. It appeared to us that the Dance brought people
together, both physically and psychologically.
Eects of the Dance
We predicted that participants’ cortisol levels would increase from before to during the
Dance, and this was supported, consistent with the results from Sagarin et al. (2009) and
with Konvalinka et al.’s (2011) ndings that extreme rituals can increase physiological arousal.
However, although physiological stress (as measured by changes in cortisol before to during
the Dance) increased, psychological stress decreased. We suggest two explanations for this
seeming paradox.
First, the change in cortisol may stem from the physical pain that participants were under
-
going. Thus, although participants, particularly those who have a high pain tolerance, may
not have experienced the temporary piercings as particularly painful, their bodies registered
the sensations as pain. Second, the experience of the Dance itself could have moderated the
link between physiological and psychological stress by reducing the negative emotions that
might have otherwise accompanied the physical pain. Consistent with this explanation, par-
ticipants reported a reduction in negative aect from before to during the Dance. We suggest
further that this dissociation between a physiological marker of stress (i.e., cortisol) and the
psychological experience of stress may be evidence of an altered state of consciousness.
The fact that participants perceived the Dance as lowering their psychological stress
suggests another potential motivation for participation in these types of activities: stress
reduction. In his Theory of Masochism as Escape from Self, Baumeister (1997) identied stress
reduction as a primary motivation for masochism. And although the majority of Dancers did
not self-identify as masochists, the painful activities of the Dance were accompanied by
reductions in self-reported stress and negative aect – results that align well with Baumeister’s
(1997) theory. This is particularly interesting while also considering that positive aect did
not increase from before to during the Dance. Taken together, one interpretation of these
results is that the Dance did not make people feel better, but rather, it helped them feel less
bad. Consistent with this is research examining the relationship between anxiety and ritual
behaviour (Liénard and Boyer 2006) including studies that nd increased ritual behaviour
in reaction to anxiety-provoking situations (Lang et al. 2015) and reduced anxiety among
ritual performers in high stress situations (Sosis and Handwerker 2011). It is also possible
that some participants entered dissociative psychological states brought on by the pain
they experienced from the piercings.
However, we also must be cautious in interpreting the cortisol results, due to the small
within-subject sample size and due to the nature of cortisol. Because cortisol levels uctuate
throughout the day, it is possible that the time of day is responsible for the changes prior
to and during the Dance. Also, the changes could be the result of movement. While we have
seen similar eects in other examinations of extreme rituals and BDSM scenes (e.g., Sagarin,
Lee, and Klement 2015), there are many potential confounds in the present study.
We also predicted that self-other overlap would increase from before to during the Dance,
and we found support for this hypothesis, again consistent with Sagarin et al. (2009) and
Xygalatas, Mitkidis et al.’s (2013) ndings that extreme rituals can increase social bonding and
prosocial behaviour. This result paralleled our behavioural observations in that there was a
dierence between participant interactions pre- to post-Dance. Prior to the Dance, many

466 K.R. KLEMENT ET AL.
participants engaged in before care in dyads or small groups. After the Dance, participants
laughed and talked within a larger group, particularly at the scheduled dinner following the
Dance. It seems likely that the physical and psychological support we observed participants
providing for each other throughout the Dance helped people feel closer to one another.
Finally, we also investigated how sexual arousal changed from pre-Dance to during-Dance.
As our goal was to examine an extreme ritual within a BDSM context we did not initially
predict changes in sexual arousal, but participants did report a signicant increase in sexual
arousal from before to during the Dance. It is important to point out, however, that the
during-Dance level of sexual arousal was still below the scale midpoint. Furthermore, as
noted earlier, we did not observe any signs of overt physical sexual arousal.
Conceptualisations of the Dance
Our third hypothesis predicted that participants would conceptualise the Dance as more
sadomasochistic than sexual. Although this was somewhat supported (in that participants
rated the Dance as non-signicantly less sexual than sadomasochistic), the primary concep-
tualisation was neither sexual nor sadomasochistic; it was spiritual. The primacy of this con-
ceptualisation appeared in both the closed-ended questions (in which participants rated
the Dance as signicantly more spiritual than sadomasochistic or sexual) and the open-ended
questions (in which spirituality emerged as the most frequently mentioned theme). This
suggests that many participants interpreted the Dance as primarily a spiritual experience.
One reason for this might have to do with the inuence of Musafar on the Dance of Souls.
The expectations of participating in such a ritual might be also impacted by the promotional
conference materials. A follow-up study at another regional BDSM conference might be
helpful to investigate whether the spiritual component is universal or region-specic.
The Dance as a spiritual experience
Although the scientic literature has not focused on sadomasochism as a spiritual experience,
the BDSM community literature has long recognised the spiritual aspects of BDSM activities.
In The New Bottoming Book, for example, Easton and Hardy (2001) observe:
Today we are also seeing the emergence of S/M [Sadomasochism] often referred to in this
context as ‘Sex Magic’ – as a spiritual practice. The combination of ritual with S/M, and the use
of strong sensation and sometimes opening the skin to achieve transcendent states, have led
to a potent combination of S/M practice with spiritual seeking. (8)
Likewise, in Ties that Bind, Baldwin (1993) also discusses SM as a spiritual experience:
It is not outrageous to suggest the possibility that those of us who pursue ecstatic spiritual or
mystical experiences through SM/leather/fetish actions may be the early forerunners of a new
spiritual tradition. (239)
It seems likely that contextual factors play an important role in shaping how participants
experience an event such as the Dance of Souls and, in this case, the extent to which the
event is seen as spiritual. The Dance of Souls takes place at a BDSM/Leather conference
known for its strong spiritual focus. It seems plausible that this spiritual focus infused par-
ticipants’ interpretation of the Dance. Alternatively, given the historical roots of hook-pulls
in Native American and Hindu spiritual traditions, it is also possible that such rituals are more
likely than other activities to be interpreted within a spiritual framework.

CULTURE, HEALTH & SEXUALITY 467
The Dance as a means of achieving altered states
The qualitative reports from participants also indicated that seeking and achieving an altered
state of consciousness was another motivation for participating in the Dance. In fact, an
altered state of consciousness was the third most frequently mentioned theme from the
qualitative data; participants reported feeling high, seeing visions and feeling connected to
a higher power. Reports of altered states also coincided with reports of spirituality, suggest-
ing connections between extreme meditations or ecstatic visions that have been recorded
in religious texts, Native American sun-dance rituals, Hindu Thaipusam festivals and rituals
such as the Dance of Souls (Jilek 1982; Xygalatas 2013). In Urban Aboriginals: A Celebration
of Leathersexuality, Mains (1984) noted that body piercing can be ‘a journey marked by fetish
and mana, shaman, ritual and trance’ (9).
The Dance as a way to connect with others
The second most frequently mentioned theme within the qualitative data was connected-
ness to others. This result nicely paralleled our behavioural observations as well as partici-
pants’ ratings of signicant increases in self-other overlap from before to during the Dance.
It is worth noting that in Sagarin et al. (2009), participants rated self-other overlap with their
scene partner, whereas in this study participants rated the degree of overlap with other
people in the Dance. This suggests that self-other overlap can be expanded to a community
and that people can feel increased psychological closeness with a group of individuals. In
fact, many people used the word ‘tribe’ in their open-ended responses. One participant
reported that the Dance was ‘now more about others, not myself’, while another said they
‘connected deeply during dancing with several people’. In a community that is often stig-
matised by the mainstream, connection to others could be a motivation for continued mem-
bership. Being a part of the BDSM community in a larger context, and participating the Dance
of Souls, could be a way to express that membership and dedication to the community, as
well as a chance to bond with fellow members.
The Dance as a method of reducing stress
Finally, the reductions in negative aect and self-reported stress suggest that the Dance also
functioned as a method of reducing stress and, potentially, as a means of escaping the self
(Baumeister 1997). These eects are perhaps best illustrated by the open-ended report of
one dancer:
The experience was quite calming. The degree of pain I allowed myself to have physically allowed
my hyper-analytic brain to focus/center on the present. With the exception of knitting and cro-
chet, pain this intense is the only way I can truly fall into this state of peace/have a quiet mind.
Conclusion
Through observational, self-report and physiological data, we documented the ‘Dance of
Souls’, an extreme ritual conducted within a BDSM context. To the extent that the Dance of
Souls represents an instance of sadomasochism outside a sexual context, our data challenge
the prevailing academic perspective that links sadomasochism inextricably with sexuality,

468 K.R. KLEMENT ET AL.
and they provide insight into a number of non-sexual motivations for extreme rituals and
sadomasochistic activity including experiencing spiritual journeys, achieving altered states
of consciousness, reducing stress and feeling closer and more connected to others.
Future directions
The current study was designed to examine the experiences of individuals participating in
one occurrence of the Dance of Souls. Thus, our conclusions are necessarily restricted to this
group of participants during this particular Dance. Future research might investigate other
communities and their experiences with extreme rituals, such as individuals in the body
modication community or other BDSM communities that are not inuenced by Fakir
Musafar’s philosophy. In addition, although we assumed that the rules and norms of the
SWLC (which prohibit illegal or recreational drug use and which discourage inebriation)
ensured that very few, if any, Dance participants were under the inuence of mind-altering
substances, future studies should assess this.
Notes
1. A fuller description of the behaviour observations can be found in the supplemental data for
this paper.
2. In this context, ‘tribe’ refers to a cohesive group of BDSM practitioners (e.g., a ‘leather family’).
Acknowledgements
We thank Leslie Matuszewich for her help with cortisol assays, the Dance of Souls leaders and partic-
ipants and the Southwest Leather Conference for their openness and generosity, and the Northern
Illinois University Presidential Commission on Sexual Orientation and Gender Identity for funding.
Disclosure statement
No potential conict of interest was reported by the authors.
Funding
This work was supported by Northern Illinois University.
References
Aron, A., Elaine N. Aron, and D. Smollan. 1992. “Inclusion of Other in the Self Scale and the Structure of
Interpersonal Closeness.” Journal of Personality and Social Psychology 63 (4): 569–612.
Baldwin, G. 1993. Ties That Bind: The SM/Leather/Fetish Erotic Style. San Francisco: Daedalus.
Baumeister, Roy F. 1997. “The Enigmatic Appeal of Sexual Masochism: Why People Desire Pain, Bondage,
and Humiliation in Sex.” Journal of Social and Clinical Psychology 16: 133–150.
Braun, V., and V. Clark. 2006. “Using Thematic Analysis in Psychology.” Qualitative Research in Psychology 3:
77–101.
Catlin, G. 1867. O-Kee-Pa: A Religious Ceremony; and Other Customs of the Mandans. Philadelphia, PA:
J. B. Lippincott.
Connolly, Pamela H. 2006. “Psychological Functioning of Bondage/Domination/Sado-Masochism
(BDSM) Practitioners.” Journal of Psychology & Human Sexuality 18: 79–120.

CULTURE, HEALTH & SEXUALITY 469
Csikszentimihályi, M. 1990. Flow: The Psychology of Optimal Experience. New York: Harper Row.
Easton, D., and J. Hardy. 2001. The New Bottoming Book. Emeryville, CA: Greenery Press.
Faccio, E., C. Casini, and S. Cipolletta. 2014. “Forbidden Games: The Construction of Sexuality and Sexual
Pleasure by BDSM ‘Players’.” Culture, Health, & Sexuality 16: 752–764.
Fischer, R., D. Xygalatas, P. Mitkidis, P. Reddish, P. Tok, I. Konvalinka, and J. Bulbulia. 2014. “The Fire-
Walker’s High: Aect and Physiological Arousal Responses in an Extreme Collective Ritual.” PLoS
ONE 9: 1–6.
Jackson, Susan A., Robert C. Eklund, and Andrew J. Martin. 2010. The Manual for the New Flow Scales.
Menlo Park, CA: Mind Garden.
Jilek, Wolfgang G. 1982. “Altered States of Consciousness in North American Indian Ceremonials.” Ethos
10: 326–343.
Joyal, Christian C., A. Cossette, and V. Lapierre. 2015. “What Exactly is an Unusual Sexual Fantasy?” The
Journal of Sexual Medicine 12: 328–340.
Konvalinka, I., D. Xygalatas, J. Bulbulia, U. Schjodt, E.-M. Jegindo, S. Wallot, G. Van Orden, and A. Roepstor.
2011. “Synchronized Arousal between Performers and Related Spectators in a Fire-Walking Ritual.”
Proceedings of the National Academy of Sciences 108: 8514–8519.
Krat-Ebing, Richard von. 1965. Psychopathia Sexualis. Translated by F. S. Kla. New York: Stein & Day.
(Original work published 1886).
Lang, M., J. Krátký, J. H. Shaver, D. Jerotijević, and D. Xygalatas. 2015. “Eects of Anxiety on Spontaneous
Ritualized Behavior.” Current Biology 25 (14): 1892–1897.
Liénard, P., and P. Boyer. 2006. “Whence Collective Rituals? A Cultural Selection Model of Ritualized
Behavior.” American Anthropologist 108: 814–827.
Mains, G. 1984. Urban Aboriginals: A Celebration of Leathersexuality. San Francisco, CA: Gay Sunshine
Press.
Musafar, F. 2002. Spirit + Flesh. Sante Fe, NM: Arena Editions.
Newmahr, S. 2010. “Rethinking Kink: Sadomasochism as Serious Leisure.” Qualitative Sociology 33:
313–331.
Pfa, G., and Ronald A. Simons. 1973. Floating in the Air, Followed by the Wind. Documentary. USA:
Michigan State University and University of Malaya.
Sagarin, Brad J., B. Cutler, N. Cutler, Kimberly.A. Lawler-Sagarin, and Leslie.L. Matuszewich. 2009.
“Hormonal Changes and Couple Bonding in Consensual Sadomasochistic Activity.” Archives of Sexual
Behavior 38: 186–200.
Sosis, R., and W. P. Handwerker. 2011. “Psalms and Coping with Uncer- Tainty: Religious Israeli Women’s
Responses to the 2006 Lebanon War.” American Anthropologist 113 (1): 40–55.
Watson, D., L. A. Clark, and A. Tellegen. 1988. “Development and Validation of Brief Measures of Positive
and Negative Aect: The PANAS Scales.” Journal of Personality and Social Psychology 54: 1063–1070.
Weinberg, Thomas.S. 1994. “Research in Sadomasochism: A Review of Sociological and Social
Psychological Literature.” Annual Review of Sex Research 5: 257–279.
Wiseman, J. 1996. SM 101: A Realistic Introduction. San Francisco, CA: Greenery Press.
Wismeijer, Andreas A. J., and Marcel A. L. M. Van Assen. 2013. “Psychological Characteristics of BDSM
Practitioners.” The Journal of Sexual Medicine 10: 1943–1952.
Xygalatas, D. 2013. “Eects of Religious Setting on Cooperative Behavior: A Case Study from Mauritius.”
Religion, Brain & Behavior 3: 91–102.
Xygalatas, D., P. Mitkidis, R. Fischer, P. Reddish, J. Skewes, Armin W. Geertz, A. Roepstor, and J. Bulbulia.
2013. “Extreme Rituals Promote Prosociality.” Psychological Science 24: 1602–1605.
Xygalatas, D., U. Schjødt, J. Bulbulia, I. Konvalinka, E.-M. Jegindø, P. Reddish, Armin W. Geertz, and
A. Roepstor. 2013. “Autobiographical Memory in a Fire-Walking Ritual.” Journal of Cognition and
Culture 13: 1–16.doi:10.1163/15685373-12342081.