Archives of Sexual Behavior

, Volume 41, Issue 2, pp 517–524

Consensual Sex Between Men and Sexual Violence in Australian Prisons


    • School of Public Health and Community MedicineUniversity of New South Wales
  • Tony Butler
    • National Drug Research InstituteCurtin University
    • Centre for Health Research in Criminal JusticeNSW Justice Health
  • Karen Schneider
    • National Centre in HIV Epidemiology and Clinical ResearchUniversity of New South Wales
  • Lorraine Yap
    • School of Public Health and Community MedicineUniversity of New South Wales
  • Kristie Kirkwood
    • Centre for Health Research in Criminal JusticeNSW Justice Health
  • Luke Grant
    • New South Wales Department of Corrective Services
  • Alun Richards
    • Offender Health Services, Queensland Health
  • Anthony M. A. Smith
    • Australian Research Centre in Sex, Health & SocietyLa Trobe University
  • Basil Donovan
    • National Centre in HIV Epidemiology and Clinical ResearchUniversity of New South Wales
    • Sydney Sexual Health Centre, Sydney Hospital

DOI: 10.1007/s10508-010-9667-3

Cite this article as:
Richters, J., Butler, T., Schneider, K. et al. Arch Sex Behav (2012) 41: 517. doi:10.1007/s10508-010-9667-3


Estimates of the incidence of sexual coercion in men’s prisons are notoriously variable and fraught with conceptual and methodological problems. In 2006–2007, we conducted a computer-assisted telephone survey of a random sample of 2,018 male prisoners in New South Wales and Queensland. Of 2,626 eligible and available inmates, 76.8% consented and provided full responses. We asked about time in prison, sexual experience, attraction and (homo/bi/heterosexual) identity, attitudes, sexual contact with other inmates, reasons for having sex and practices engaged in, and about sexual coercion, including location and number of perpetrators. Most men (95.1%) identified as heterosexual. Of the total sample, 13.5% reported sexual contact with males in their lifetime: 7.8% only outside prison, 2.8% both inside and outside, and 2.7% only inside prison. Later in the interview, 144 men (7.1% of total sample) reported sexual contact with inmates in prison; the majority had few partners and no anal intercourse. Most did so for pleasure, but some for protection, i.e., to avoid assault by someone else. Before incarceration, 32.9% feared sexual assault in prison; 6.9% had been sexually threatened in prison and 2.6% had been sexually coerced (“forced or frightened into doing something sexually that [they] did not want”). Some of those coerced reported no same-sex contact. The majority of prisoners were intolerant of male-to-male sexual activity. The study achieved a high response rate and asked detailed questions to elicit reports of coercion and sex separately. Both consensual sex and sexual assault are less common than is generally believed.


Sexual behaviorPrisonsMale homosexualitySexual assaultAustraliaSex survey methodology


Sex in prison is a topic that elicits complex emotional reactions. Definitional issues (What counts as “sex”? When does pressure amount to coercion?) are confounded further by attitudes to homosexuality that make some people see sexual contact between men (especially men who outside prison are heterosexual) as prima facie abusive.

These difficulties are reflected in the research literature on consensual and non-consensual sex in prison. Even outside prison and in representative samples, estimates of the incidence of sexual assault or coercion are notoriously variable, from low rates based on reported cases of rape to high rates that include all reports of unwanted or regretted sex, sexually suggestive touching, or sexualized or homophobic verbal aggression. The methodological and conceptual obstacles inherent in sexual behavior research are only further complicated in a prison context. Such difficulties have led to confusion and debate in estimating the true frequency of sexual coercion. Yet, in order to act to reduce sexual assault or coercion in prisons, and to ascertain whether their actions have succeeded, prison authorities need to be able to define and measure the phenomenon.

Earlier literature reflected an uncertainty about the distinction between consensual homosexual behavior and sexual victimization (Eigenberg, 1992). If one takes an essentialist or “trait” approach to homosexuality, one assumes that same-sex desire is a continuing property of the individual. If one further takes the view that sexual activity only arises through desire for the partner as a sexual object, sexual contact between men who are not usually homosexually inclined can only be explained in terms of victimization. If the assumption about desire is abandoned, the explanatory notion is that of “situational” homosexuality. This idea, often based on what has been called a “hydraulic” conception of male sexuality (or “deprivation theory,” Fleisher & Krienert, 2006), sees male libido as a continuing need that must be satisfied by sex, a tension that must be released through orgasm. In the absence of women, therefore, heterosexual men will seek release through sex with other men. A social constructionist regards sexual identities (i.e., as gay, straight, etc.) as created through social meaning-making, and acknowledges that individuals may change not only their sexual practices but also their sexual identity at different stages of their lives and in different social situations (Eigenberg, 1992; Gagnon & Simon, 1973). Gagnon and Simon argued that homosexual relationships in prison serve as a means to satisfy needs that the prison system does not otherwise meet. In the 1970s, influenced by feminist analyses of the rape of women, rape of men was re-defined as an act of aggression and power rather than primarily one of sexual need (Scacco, 1975, 1982). Research then sought to address where and why this was happening and to document the detrimental effects experienced by victims of sexual coercion in prison.

Reported estimates of the incidence of consensual and non-consensual male-to-male sex in prison and the number of men affected have been highly inconsistent. The reported rates of sexual contact in prison (i.e., the prevalence of reporting having ever experienced it) have ranged between 1 or 2% (Green et al., 2003; Lockwood, 1980) and 65% in one atypical prison (Wooden & Parker, 1982). Similarly, estimates of the proportion of inmates who have been sexually coerced or assaulted range from less than 1% to as high as 41% (Gaes & Goldberg, 2004).

If survey respondents are concerned with being viewed in a socially acceptable and positive manner, they are likely to distort their answers (Tourangeau, Rips, & Rasinski, 2000). Social pressures on men and women differ so that men tend to exaggerate or “round up” and women tend to minimize their sexual activities (Alexander & Fisher, 2003; Catania et al., 1996; Fisher, 2007). Social desirability bias may be an issue particularly for heterosexually identified men when they are asked about sex with men. In addition, sexual victimization may go unreported due to victimhood being highly taboo in hegemonic masculinity. This matter of “spoiled identity,” to use Goffman’s (1968) term, is a pressing practical issue for men who have been sexually assaulted, as the information, if spread through the prison, might not only result in shaming and ostracism by fellow prisoners but also attract further predators (as observed by Banbury, 2004).

Unfortunately, validity checks on self-report data about sex between men or sexual victimization are not possible. If a man fails to report an experience of sexual victimization or what the researcher might view as sexual contact, whether because of embarrassment or social undesirability, or because he did not understand the question, there is no external gold standard with which to compare it for verification. However, different modes of data collection can partially address such biases and methodological difficulties. For example, Johnson, Wadsworth, Wellings, and Field (1994) reported higher rates of disclosure of same-sex activity in self-completion questionnaire booklets than in face-to-face interviews with the same respondents. A self-completion questionnaire removes the need for respondents to report socially sensitive behaviors directly to the interviewer and is likely to result in more valid responses (Catania, Binson, Van der Straten, & Stone, 1995). However, written questionnaires exclude illiterate respondents, and may not be appropriate where the target population has a low education level. Telephone and computer-assisted interviews (e.g., Wolff, Blitz, Shi, Bachman, & Siegel, 2006) may improve the quality of data collected on sensitive subjects such as sexual victimization (Turner, Forsyth, & O’Reilly, 1998; Weeks, 1992). In addition, Gaes and Goldberg (2004) noted that many self-administered surveys on sexual assault in prison have response rates of 50% or even less, putting the generalizability of their results in question.

There are few survey data on sexual assault in Australian prisons (Butler & Milner, 2003). Their culture may be very different from prisons in the United States, where sex is illegal. In 2006–2007, we conducted a computer-assisted telephone interview (CATI) survey of a random sample of male prisoners in New South Wales and Queensland as part of the Sexual Health and Attitudes of Australian Prisoners study. We asked several different questions about sexual contact with other inmates in prison and about sexual coercion. Questions at this level of detail have never been asked before in a representative sample survey of inmates. Taken together, the results yield a rough but robust estimate of the proportion of men in prison who have ever experienced coercion and the incidence of coercion among those men per year of incarceration. The variations between the answers to different questions reveal some of the complexities of the topic.



Survey participants were randomly selected from lists of inmates generated shortly before the recruiter team visited each prison. The sample size represented approximately 14% of the male inmate population in 2007 (Australian Bureau of Statistics, 2007).

A number of inmates in isolated and remote settings, such as work camps, were not covered by the sampling technique, due to logistical difficulties in providing telephone access and post-survey support to participants in these locations. Prisoners in transitional centers, police court cell complexes, and periodic detention centers, were not part of the survey. Prisoners were ineligible if they had insufficient English, were acutely ill or in crisis, could not provide consent, or had previously participated in the survey at another prison. They were recorded as unavailable if they were being transferred between prisons, were in court or hospital, could not be released from their work duties, or were deemed too dangerous by prison authorities. Further details are given in Richters et al. (2008).

Of 3,915 prisoner numbers selected, 288 inmates were ineligible and 1,001 unavailable; the remaining 2,626 were invited to take part, of whom 537 refused and 2,089 were interviewed (participation rate, 79.6%). Seventy-one interviews were incomplete or could not be matched with Corrective Services data, leaving a sample of 1,118 men in New South Wales (NSW) and 900 men in Queensland (76.8% of those invited to take part).

Response rates were over 99% for questions asked of all participants. A few questions addressed to subsets of respondents had higher refusal rates of up to 5%.

Participants were between 17 and 78 years (median, 32 years). This is similar to the general prisoner population in NSW and Queensland (32.9 years), but skewed towards younger ages compared with the general adult population. In total, 21.6% of participants identified as Aboriginal and/or Torres Strait Islander (NSW 18.3%, Queensland 25.6%). This is similar to the indigenous proportion of the total male prisoner population in those states (NSW 20.0%, Queensland 26.1%), allowing for the somewhat wider definition used by NSW Department of Corrective Services (DCS), but much higher than the general adult population (NSW 2.2%, Queensland 3.4%) (Australian Bureau of Statistics, 2007).

For 39% of inmates, this was their first time in an adult prison. The total length of time that inmates had spent in prison (including previous sentences) ranged from less than a month to more than 10 years (median, 3.5 years).


The selected inmates were given a full verbal explanation of the study by the recruiter and a printed information sheet and consent form, which informed them that certain demographic and offence data would be obtained from the Corrective Services. Participation was voluntary and written consent was obtained before the interview. After their interview, participants were seen by the recruiter, who referred them for health care or counseling if required. Each participant received A$10 paid into his prison account.

Study codes, including an inmate identifier, were used to link survey responses with the data for each individual provided by Corrective Services. Once all surveys were completed and data linked, the data records were de-identified and automatically numbered.

Interviews were conducted by female telephone interviewers at Taverner Research in Sydney. No participants requested a male interviewer. The inmate was in a private place (generally a room with a window) so officers could see him but not hear what he said, and the phone call was not monitored by Corrective Services. Interviews averaged about 30 min in length (range, 19–60 min).

Ethics approval was provided by the NSW Justice Health Human Research Ethics Committee (GEN5/05) and ratified by the University of New South Wales Human Research Ethics Committee (HREC 05045). The NSW Department of Corrective Services Ethics Committee (Ref. 05/0882) recommended approval of the study, which was approved by the Commissioner of Corrective Services as required by the Crimes (Administration of Sentences) Act for all research conducted with inmates. The Queensland Corrective Services Research Committee also approved the study.


The questionnaire was based on that used for the Australian Study of Health and Relationships (Smith, Rissel, Richters, Grulich, & de Visser, 2003a), with minor adaptations of question and response wording to allow for the lower literacy of this sample and the addition of further sections on experiences in prison. Questions used for this analysis are available from the corresponding author on request.


Data were analyzed using SPSS Statistics 17.0. Chi-square analyses were used to generate p values where necessary to identify associations between variables. For estimation of sexual coercion incidence rate, lengths of time in prison were imputed in the middle of the response category (e.g., 1–6 months was coded as 3.5 months) or as indicated by the distribution of inmates across a category reflected in the inmate census (Corben, 2008).


Sexual Identity, Attraction, and Experience

Asked “Do you think of yourself as: 1 Heterosexual or straight; 2 Homosexual (gay); 3 Bisexual…?”, interviewees responded as shown under “Identity” in Table 1.
Table 1

Reported sexual identity, attraction, and experience





 Heterosexual or straighta









 Not sure, undecided or other




 Only to females, never to males



 More often to females, and at least once to a male



 About equally often to females and to males



 More often to males, and at least once to a female



 Only to males, never to females



 Never felt sexually attracted to anyone at all




 Only with females, never with males



 More often with females, and at least once with a male



 About equally often with females and males



 More often with males, and at least once with a female



 Only with males, never with females



 No one



aOne man in this category also identified himself as “queer”

Responses in Table 1 were roughly similar to the general community (e.g., Smith, Rissel, Richters, Grulich, & de Visser, 2003b), though prisoners were somewhat more likely than men surveyed in households to report bisexual identity (3.1% vs. 0.9%), sexual attraction to both males and females (7.8% vs. 6.2%) or experience with both males and females (12.9% vs. 6.0%). All three transgender participants identified as bisexual, though one of them said she was only attracted to males and had only had sexual experience with males; the other two were attracted to and had experience mostly with males (i.e., at least once with a female).

Experience of Sexual Coercion Outside Prison

Asked about times outside prison, “Have you ever been forced or frightened by a male or a female into doing something sexually that you did not want to do?”, 270 men (13.4%, 95% confidence interval (CI) 11.9–14.9%) said this had happened to them; this is referred to below as “sexual coercion.” This rate was much higher than in the general community (4.8%) (de Visser, Smith, Rissel, Richters, & Grulich, 2003).

Same-Sex Contact in and out of Prison

Table 1 showed that 263 men reported ever having had some form of sexual experience with another male (“any kind of contact … that you felt was sexual”). Later in the interview, after the questions about sex with women and regular relationships, a check question elicited nine further men who reported some sexual contact with males (total 272, 13.5% of sample). Most of these 272 men had had same-sex contact outside prison, and 112 (5.6% of the total sample, CI 4.6–6.6%) had had some sexual contact with a male in prison (Table 2).
Table 2

Setting where male participants reporting any same-sex experience had had sexual contact with males



% of men with same-sex experience

% of total sample

No sexual experience with males



Outside prison




Both inside and outside prison




In prison








Sexual Contact with an Inmate

Later in the interview, as part of the section about life in prison, participants were asked about sexual contact with another inmate, and 144 men (7.14%, CI 6.02–8.26%) reported this. The largest source of the discrepancy between this figure and the 112 men who earlier admitted same-sex contact in prison was that 24 of the 144 men had not disclosed any same-sex experience early in the interview. Twenty-two of the 24 identified as heterosexual and two as bisexual, and almost all (23) said they had been sexually attracted only to females. (The other one said he had felt sexually attracted mostly to females.)

The other main source of the discrepancy was 16 men who initially said they had had sexual contact with males only outside prison, but later reported contact with another inmate. Two men who had had sexual contact with males both inside and outside prison said no to the question about contact with an inmate; perhaps they had had contact with men other than inmates. Interestingly, 10 of the 26 men who identified as gay did not have any sexual contact with another inmate.

The length of time the 144 men had been in prison before their (first) sexual contact ranged from 2 days to 18 years (median, 2 years). For 29 men, it occurred during their first time in an adult prison. Most men (124) said it was consensual, 17 said it was not consensual, and three said they “sort of” agreed to it. For 70 men, it was their first sexual experience with a male. The question was not asked of 48 men who had already indicated that their first same-sex sexual experience was before the age of 18.

These 144 men had had sexual contact with 1–50 other inmates (Table 3), apart from one man, a long-term prisoner, who reported 2,500 sexual partners, including 450 for anal intercourse. Most men had some form of sex with one to five partners, many of them with only one man. Four participants had oral sex with 41–50 partners, and one man had anal sex with 21–30 partners—this was one of the transgenders, who also reported 50 oral sex partners.
Table 3

Number of inmates with whom participants had had sexual contact (for 144 men who had had any sexual contact with another inmate)

 Number of partners


Any contact


Anal intercourse






















































Don’t know/Can’t count/Refused







Table 3 also illustrates that for many sexual encounters presumably only manual sex (stimulating, touching or rubbing the penis with the hand) was involved; 75 of the 144 men never had anal intercourse and 30 did not have fellatio. Men who identified as heterosexual were less than half of those who reported anal intercourse, but 41 of the 54 who reported “any contact” with only one partner. Of the 69 men who reported anal intercourse, 22 had used a condom at least once for sex with another prisoner (21 of whom were in a NSW prison, where condoms are provided from free dispenser machines).

Men who had had sexual contact with another inmate were asked about reasons why they might have done this; the majority reported that it was for pleasure, but protection was also a motivation for some (Table 4).
Table 4

Reasons agreed to by 144 participants who had had sexual contact with another inmate




For pleasure



To avoid being physically or sexually assaulted by someone else, i.e. for protection



To pay for drugs



To pay for other goods (like food etc.)



To repay any other debt



None of the above



aAdds up to more than 100% because participants could answer yes to more than one reason

The last time these 144 men had had sexual contact with another inmate ranged from 1 day to 35 years ago (i.e., including the period when condoms were not yet available in NSW). A total of 126 men said they had consented to it on this occasion and 15 had not. The ages of the men this happened with ranged from 18 to 61 years (median, 28 years). Practices engaged in are shown in Table 5.
Table 5

Sexual practices engaged in by 144 male participants at last sexual contact with another inmate




Manual sex (respondent’s penis)



Manual sex (respondent’s hand)



Insertive fellatio (respondent’s penis)



Receptive fellatio (respondent’s mouth)



Insertive anal intercourse (respondent’s penis)



Receptive anal intercourse (respondent’s anus)



Orgasm (respondent)



aTen of the 15 NSW participants used a condom

bSeven of the 16 NSW participants used a condom

Sexual Assault in Prison

A third (32.9%) of prisoners feared before they came to prison that they would be sexually assaulted there (Table 6). Far fewer (7.1%) were currently frightened of being sexually assaulted in prison. A similar proportion (139, 6.9%) had actually been threatened with sexual assault in prison, some of them on repeated occasions: nearly half (57 men) had been threatened once, 59 men 2–8 times, and 18 men more frequently (10–300 times); five men could not give a number.
Table 6

Sexual coercion in prison: perception before prison and experience in prison (n = 2018)




Before coming to prison, worried about being sexually assaulted in prison



Currently scared of being sexually assaulted in prison



Threatened with sexual assault



Forced or frightened into unwanted sexual activity



Fifty-three (2.6%) prisoners said they had been forced or frightened into unwanted sexual activity in prison (Table 6), of whom nearly half (25 men) reported that this had happened once, 15 men said it happened two or three times, nine men up to six times, and three men said it happened more frequently (10–30 times); one could not give a number.

Cross-tabulating the 144 men who reported any sexual contact with an inmate with the 53 who had been sexually coerced in prison, we found that 26 men who had been coerced said no to the sexual contact question. Conversely, 117 of the 144 men who had had sexual contact with another inmate in prison had not been coerced.

Features of First Experience of Sexual Coercion

The 25 men who had been coerced once were aged 18–36 when this happened; 11 of them were under 20 at the time. For those who had been coerced more than once, the first event occurred when they were between 18 and 62; 11 of them were likewise under 20.

Victims had generally been in prison a few months when this happened: for four men, 2–3 days; seven men, 1–3 weeks; 27 men, 1–9 months; 14 men, 1–6 years.

All but one respondent said where this had happened: for 31 participants, it happened in a cell (also toilet block for one of them), for 12 in the showers, for two in the common area, and for six in the yard. One respondent said “Wherever they think they can put their hands on you, in my cell, in other cell[s]. It does happen in the secure units where I am.”

For 35 men, there was one assailant, but 18 reported more than one, ranging from two to 30 (median three). For 14 of the 53 men, the assailant (or one of the assailants) was a cell mate.

Only 16 victims reported this event to a member of staff, and 23 (including 11 of those who reported it) later sought help from or talked to someone else about it; three men spoke to more than one person. Ten men spoke to a counselor or psychologist, and three spoke to a correctional facility officer. Other people spoken to included family and friends, the police, and health and welfare services.

Incidence of Sexual Coercion

The total number of incidents of sexual coercion reported in the survey was 170 over approximately 10,316 man-years of incarceration, giving an incidence rate of sexual coercion in prison of approximately one sexual assault per 61 prison-years. For those who had been in prison for less than a year, there were 19 incidents over approximately 313 prison years, i.e., one sexual assault per 16.5 prison-years.


Responses to seven attitude statements are shown in Table 7. Most prisoners were intolerant of male-to-male sexual activity. And despite the difficulty with the double negative, most prisoners disagreed with the statement “Choosing to have sex in prison doesn’t make you gay.”
Table 7

Agreement (“agree” or “strongly agree”) with sexual attitude statements (n = 2018)




If two people had oral sex, but not intercourse, you would still consider that they had had sex together



Having an affair when in a committed relationship is always wrong



It’s OK for inmates to have sex in prison then go back to their partners



Sex between two adult men is always wrong



Sex between two inmates in prison is always wrong



Choosing to have sex in prison doesn’t make you gay




Most men (95%) identified as heterosexual and few (7%) had had any form of sexual contact with another inmate. Those who had done so mostly had few partners and the sex often consisted only of manual stimulation. The last time they had sexual contact with an inmate, most did so for pleasure, but some did it for protection and a few to pay for goods or a debt.

In their time in prison, 2.6% of men had been coerced (“forced or frightened”) into an unwanted sexual act, some of them repeatedly. About half of these men did not report sexual contact with a male, suggesting that sexual assault is not the same thing as non-consensual sex. Gaes and Goldberg’s (2004) critical review and meta-analysis of reported sexual victimization prevalence and incidence estimates from 14 studies, most of which were based in American prisons, resulted in a conservatively estimated prison lifetime prevalence estimate of 1.9% of prisoners. Estimates were predominately based on reports of assault and non-consensual sex (although some of the studies asked about attempted assault and sexual pressure) that were bound by varying times (between one year and total length of time spent in prison). This low percentage estimate has been supported in more recent studies of rape in prison (Beck & Harrison, 2007; Jenness, Maxon, Matsuda, & Sumner, 2007; Wolff et al., 2006). However, studies that include sexual pressure and sexually suggestive touching in the definition of sexual coercion still produce higher estimates (Gaes & Goldberg, 2004).

This survey achieved a high response rate of 77% of available and eligible inmates. As inmates are frequently moved between work camps and larger prisons, and between courts, hospital, and prison, it is unlikely that the omission of unavailable prisoners biased the findings. However, it is possible that dangerous prisoners, who could not be moved to take part in the interviews for safety reasons, might be less likely to experience sexual coercion because other prisoners were afraid of them. The inclusion of their responses would have had the effect of slightly reducing rather than inflating our estimates of the prevalence of experience of sexual coercion in prison.

Given the random sampling, large sample, and high response rate, the estimates of proportions of the whole group (e.g., the percentage who had experienced coercion) are likely to be robust. However, we should not overgeneralize from the more detailed results. When we concentrate on detailed answers from small groups, such as subsets of the 53 men who reported coercion in prison, or the few who had multiple consensual sexual partners, we are exploring individual narratives, not a generalizable result. We cannot be sure that they are typical of all NSW and Queensland prisoners. Also, some were talking about events that happened a long time ago.

NSW and Queensland prisons hold 58% of Australia’s prisoner population (Australian Bureau of Statistics, 2007). Although there is considerable demographic variability between states and territories, NSW and Queensland overall are roughly representative of Australia as a whole, with both highly urbanized and remote populations contributing to the prisoner population.

The privacy of speaking on the telephone is likely to have assisted disclosure of stigmatized behaviors. Indeed, in relation to sexual coercion in prison, if we take literally the 53 participants’ statements about whether they told anyone or sought help, then 25 men told us about an incident of sexual coercion that they had not previously revealed. In the 2001 NSW Inmate Health Survey (Butler & Milner, 2003), in face-to-face interviews, 17/747 men (2.5%) reported consensual sex (compared with 126/2018 (6.2%) in our survey, talking about the first time they had sexual contact with an inmate) and only three (0.4%) reported non-consensual sex (15/2018 (0.7%) in our survey). The lower rates compared with our findings may also have been due to the Inmate Health Survey’s more limited definition of “sex” as “vaginal, anal or oral.” Our findings confirm the often expressed conviction that reported sexual assaults are only a minority of actual events: only 30% (16/53) of prisoners who told us they had been sexually coerced in prison had reported it to a member of staff.

Mindful of the need to obtain permission from correctional authorities and cooperation of officers to implement the survey, we did not specifically ask in the interview whether participants had been sexually coerced by, or had consensual contact with, an officer or other member of staff.

No other surveys have asked both about sexual contact (whether consensual or non-consensual) and separately about sexual coercion. The discrepancy in the results is illuminating. Even with a broad definition of sex, carefully worded so as to include non-penetrative practices, men often do not report sexual coercion when asked about their sexual experiences. Some commentators interpret this as underreporting due to social desirability pressure or stigma. Our belief is that from the victim’s point of view, a sexualized threat or assault is not “sex” or even “sexual contact.” Although it is possible that participants may conceal the truth or be unwilling to tell us things, it is certain that we get the wrong answer if we ask the wrong question.

Men who identify as gay or bisexual and who are happy to play the receptive role in sexual encounters with straight men may occupy a “non-masculine” role analogous to that of transgenders. An example is the discrepancy in Table 5 between the proportion of encounters involving sex given or received for manual and oral sex. This could be explained by underreporting of receptive sex because of stigma, but is also consistent with a sexual economy in which a few inmates “service” a larger number of other men.

Although condoms have been available from free dispenser machines in NSW men’s prisons since 1996 (Yap et al., 2007), condom use for anal intercourse was low. It is particularly important that the highly sexually active inmates have access to condoms and are aware of the need to use them.

Less-educated, lower-income people in blue-collar jobs tend to be less liberal in their sexual attitudes (Rissel, Richters, Grulich, de Visser, & Smith, 2003). Our survey showed that male inmates had a strong commitment to sexual exclusivity in committed relationships, disapproved of sex in prison, and were disinclined to excuse non-coital practice as “not real sex” or to regard sex in prison as something that does not count. Interestingly, despite the low average age of the sample, 76% counted oral sex as “sex.” This is slightly but significantly (p < .01) higher than in the general population, among whom 72% of men aged 16–59 agreed with this statement (Rissel et al., 2003). Young people are far less likely to regard oral sex as “sex” than the over-40s (Richters & Song, 1999; Rissel et al., 2003).

Most strikingly, the prisoners were much more disapproving of homosexual contact than the general population: 62% of prisoners regarded sex between two adult men as “always wrong,” but only 37% of men aged 16–59 years outside prison took this view (Rissel et al., 2003). Prisoners who find abstinence unworkable and who seek “situational” sex (even non-coital sex) with other inmates are operating in a very intolerant environment.

This survey showed that both consensual sexual contact between inmates and sexual coercion (“forced or frightened”) were much less common than the popular view of ubiquitous “situational homosexuality” and rape in male prisons would suggest. Inmates were much more disapproving of male-to-male sexual contact than other Australians, and most disagreed with the proposition that “Having sex with another inmate in jail doesn’t make you gay.” A small minority of inmates, mostly gay or transgendered, adopted—or had thrust upon them—a sexually available, non-masculine role. Understanding the dynamics of this sexual pattern requires in-depth qualitative research.

Our use of computer-assisted telephone interviewing was successful and liked by the participants. Results confirmed that surveys need to distinguish clearly between sexual practices (e.g., manual sex, oral sex and anal intercourse) so that non-coital contact can be easily disclosed. Researchers should not assume that prisoners will report sexual coercion when asked about their sexual experiences or contacts.


We are grateful to Tony Falconer, Josephine Belcher, Azar Kariminia, Max Saxby, Fred Ropp, Jenny Douglas, Joanne Holden, Ariane Minc, Debbie Pittam, and the members of the advisory committee for help in the early stages. Our thanks to all the DCS staff and correctional officers who assisted, recruiters Iain Perkes, Jessica Pratley, John Samaha, Kathy Prime, and Patricia MacAlpine and other Justice Health nurses, and staff and interviewers at Taverner Research. The study was funded by the National Health and Medical Research Council (Project Grant 350860) with additional funding from NSW Health, NSW Justice Health, the New South Wales and Queensland Departments of Corrective Services and the UNSW Faculty of Medicine.

Copyright information

© Springer Science+Business Media, LLC 2010