Would you say you had unprotected sex if…? Sexual health language in emails to a reproductive health website

Abstract

The words and metaphors that people use to describe sexuality and reproductive health reflect experiences with peers, sexual partners, health service providers, and public health campaigns. In this paper, we analyse 1,134 e-mails sent to an emergency contraception website in the USA over the course of a one year period. Through an examination of the terminology used by authors to describe contraceptive methods, sexual intercourse and other sexual acts, we analyse what those terms signify within their textual context. We find that the kinds of risk concerns used in assessing sexual activity – whether evaluating pregnancy risk, disease transmission risk, or moral risk – influence the definitions people give to terms that are multiply defined or whose definitions are culturally contested. This finding emerged clearly in the meanings given to terms for “sex” and “unprotected sex,” which varied widely. We conclude with a discussion of the implications of this finding for research, clinical care, and health education activities.

Introduction

In December 1998, President Bill Clinton’s impeachment by the House of Representatives and subsequent Senate trial stemmed from his deposition statement, “I have never had sexual relations with Monica Lewinsky. I’ve never had an affair with her.” President Clinton famously reiterated the substance of his sworn testimony in a White House news conference in January 1998 when he stated unequivocally, “I did not have sexual relations with that woman, Miss Lewinsky.” When it was later revealed that the President had in fact been the recipient of oral sex, debates about the legal and popular definitions of “sexual relations” ensued.

In 1999, the public health and medical communities became directly implicated in this national debate. In January of that year, the editor of the Journal of the American Medical Association (JAMA) fast-tracked an article focused on the meaning of sexual terms for publication. Entitled, “Would you say you ‘had sex’ if…?” (Sanders and Reinisch 1999), the Kinsey Institute article presented the results of a 1991 survey of college students who were given a list of descriptive sexual acts and asked to choose which of them constituted “having sex.” After the publication of the JAMA article, the executive vice-president of the American Medical Association fired JAMA editor George Lundberg for “inappropriately…interjecting JAMA into a major political debate that has nothing to do with science or medicine” (T.W. Smith 1999, 385). The ensuing outcry from the medical community was coupled with several articles published in prestigious medical journals that contended that the firing of Lundberg was the political act and that Lundberg’s dismissal threatened the credibility and editorial integrity of a major American medical journal (e.g. Horton 1999, R. Smith 1999; see T.W. Smith 1999 for a review). Revealed in this debate was not only that meanings attributed to the language of sexuality and sexual health are variable and contested, but that scientific representations of this variability can be politically charged as well.

Yet the importance of language in sexual and reproductive health cannot be underestimated. Provoking clinicians and health policy-makers to think about the range of what people mean when they talk and think about “sexual relations” has significant implications for providing sexual health counselling, education, and care. Through an examination of over 1,100 emails sent to an emergency contraception (EC) website, the majority of whose users are based in the U.S.A., our study identifies another contested area: what constitutes unprotected sex? We argue that three types of risk concerns -- pregnancy risk, disease transmission risk, and moral risk -- shape not only the language of sex, but the conceptualisation of “protection.” We conclude by reflecting on the implications of our findings for health education and clinical care.

Language, Sexual Health, and Politics

A wide interdisciplinary literature documents the ways that language both reflects and structures thinking about bodily states, health, and sexuality. In their seminal work, Lakoff and Johnson (1980) argued that language uses metaphors of body states to signal mood and affect, spatial organisation, social hierarchies, and political or economic orders. But if language can reflect embodied states, so too does it reflect and constitute particular social orders and ways that cultural systems frame bodies. For example, analysts have examined the linguistic framings of diseases, from cancer and tuberculosis (Sontag 1978) to AIDS (Brandt 1988, Sontag 1989, Treichler 1992, Weiss 1997), Ebola (Joffe and Haarhoff 2002), and SARS (Wallis and Nerlich 2005), and shown how these framings are linked to particular conceptualisations of individual bodies and the body politic. Another way that the language of health is political is in the way that it frames issues and problems and thus shapes the possibilities of health policy outcomes (Annas 1995, Treichler 1992). Language is therefore central ground for both activists trying to influence public health debates and health professionals trying to provide high quality services (Hardon 2006, Hadlow and Pitts 1991, Kleinman 1988, Boyle 1970; Ong, de Haes, Hoos, and Lammes 1995). As a result, it is not always possible to disaggregate the politics from the science or strategies of public health outreach.

The language of emergency contraception

In the USA, the realms of sexuality, reproductive health, and personhood are particularly contested terrain where both popular and medical terminologies stake out political positions in debates over abortion and contraception (Heriot 1996, Wynn and Trussell 2006a). The intertwining of linguistic strategising in public health education and the politics of reproductive health language is particularly visible in debates over the terminology for “emergency contraception”. In early public health campaigns to educate the public about the contraceptive potential of high doses of hormones taken post-coitally, the term “morning after pill” was used by physician-activists. In 1992, activists made the effort to change the terminology to “emergency contraception” (Trussell et al. 1992, Ellertson 1996, Prescott forthcoming). There were several reasons behind the attempted switch. First, the phrase “morning after pill” implied that the medication could be taken only the day after sex occurred; in fact, studies have shown that emergency contraceptive pills (ECPs) are effective if taken immediately or up to five days after sex (Ellertson et al. 2003, von Hertzen et al. 2002). The phrase “morning after pill” is also imbued with assumptions about when sexual intercourse occurs - in the evening - as well as with broader cultural references to regret and shame. Second, recognising that EC is not as effective as any ongoing method of contraception (Trussell 2007), the term “emergency” was meant to convey the sense that it was a last-resort strategy for preventing pregnancy, and distinguish it from the regular use of postcoital contraception. And third, as medication abortion in the form of the mifepristone (or “RU-486”) and misoprostol regimen became increasingly known to the public, the term “emergency contraception” was meant to distinguish it in the public eye from “the French abortion pill” and place it firmly in the same realm as contraceptives used before or during sex. Yet opponents of hormonal contraception reject the term precisely because of this; groups such as Pharmacists for Life and Concerned Women for America persist in using the term “morning after pill” on the grounds that it is an abortifacient, not a contraceptive, because of the hypothetical possibility that it might interfere with the implantation of a fertilised egg (Wynn and Trussell 2006b, Davidoff and Trussell 2006). This history of the debate over and politicisation of EC terminology is an important backdrop to understanding the language we encounter in e-mails sent to the Emergency Contraception Website.

Data Source and Methods

The Emergency Contraception Website http://ec.princeton.edu is hosted on a Princeton University server. Popularly known as Not-2-late.com the website was launched in October 1994 as one of the first websites to provide medical information on the Internet, long before the Internet had become a widely used resource for information on health and sexuality (Rideout 2001). The peer-reviewed website, jointly operated by Princeton University’s Office of Population Research and the Association of Reproductive Health Professionals provides comprehensive information about EC derived from the scientific literature, and is available in English, Spanish, French and Arabic.¹

EC refers to the use of medications (such as pills containing higher doses of hormones than are found in oral contraceptive pills (OCPs)) or devices (such as the post-coital insertion of a copper intra-uterine device (IUD)) that can be used up to 120 hours after sexual intercourse to reduce, but not eliminate, the likelihood of pregnancy (Trussell, Stewart, and Van Look 2007). Not-2-late.com primarily targets a non-specialist audience of women and men seeking information about, and access to, EC. The website gives users the option of sending questions to James Trussell, the founder of the website. The most commonly asked questions sent to the website about EC are how to use it (22.5%), side effects of taking ECPs (21.2%), pregnancy testing, risk, and symptoms (17.1%), whether EC is needed in a given situation (14.2%), EC access (7.6%), EC effectiveness (4.0%), and how EC works to prevent pregnancy (3.3%) (Wynn and Trussell 2005).

In this paper, we analyse the content of one year’s worth of e-mails sent to the website, focusing specifically on the language used by writers to describe situations and ask questions about sexuality and reproductive health. Between July 1, 2003 and June 30, 2004, 1,134 English-language e-mails were received (excluding spam). We tracked different terms for EC, sex, and various aspects of reproductive health, and examined what those terms mean in context. In contrast with methodologies that use computer programs to do keyword searches on vast quantities of texts (e.g., Seale, Ziebland, and Charteris-Black 2005), our analysis involved reading each e-mail to identify terms, including terms that may not have been anticipated by study authors. Our method also allowed us to identify long phrases and expressions that express a concept, as well as misspelled terms, content that may be missed by computer-assisted searches. This search examined not only the language that was used but also language that was not used (Sontag 1978, Rapp 1988). This approach provides insight into what people feel uncomfortable saying, with important implications for patient-clinician communication and public health education campaigns.

Results

Frequency of terminology

In Table 1, we provide quantitative results as well as information about what terms appeared on the website itself. We have grouped these terms into ten categories and provide the total number of emails containing any specific term related to the general category: EC (710), other contraceptive methods (446), sexual acts (422), bleeding (321), ejaculation (117), sexual organs (92), sexual fluids (94), reproductive processes (58), foetus/embryo (24), and sexually transmitted infections (STIs) (5). As a single email may include multiple categories, the number of emails in each category does not sum to the total number of emails received. Table 1 lists the terms used in the emails and provides the number of emails containing each specific term.

Table 1.

A list of the most frequently used terms in e-mails sent to the website

Category	Emails (proportion of emails received)	Terms used on website	Most common terms used by authors of emails	Emails where term is used
Emergency contraception	710 (62.2%)	emergency contraception, EC, emergency contraceptive pills, ECPs, brand names	ECP brand names	163
			emergency contraception, emergency contraceptive	145
			morning after pill	128
			the pill, this pill	125
			ECP	119
			EC	96
			emergency contraception pill	56
			EC pill	30
			the emergency pill	16
			other terms, including: 72-hour pill, 24-hour contraceptive, postcoital pill, pills of the day after, the night after kind of pills, Plan B form of the pill, morning pill, after pills, emergency anti-pregnancy drug, and high dosage pill	90
Other contraceptive methods	446 (39.3%)	birth control pills, progestin-only pills, combined pills, oral contraceptives, regular birth control pills, contraception, other forms of contraception, pill and pills, condom, IUD, copper-T, brand names	oral contraceptive brand name	152
			condom	144
			birth control	76
			the pill, the Pill	75
			birth control pill	67
			contraception, contraceptive	31
			protection	19
			the shot	10
			injection, injectable	10
			the patch, birth control patch	10
			other terms, including: withdrawal, Depo, IUD, diaphragm, OCs, OCPs, BCPs, female condoms, the coil, Copper IUD, spermicide, and “that stuff that kills the sperm”	80
Sexual acts	422 (37.2%)	sex, intercourse, sexual intercourse, sexual activity Non-consensual sex: sexual assault, rape, forced to have sex	sex	312
			intercourse	85
			sexual intercourse	14
			doing it, did it	6
			vaginal intercourse	3
			the sexual act	2
			other terms including: love-making, that act, hooking up, relations, slept together, infidelity, oral sex, anal sex, hand job, fingered, doing foreplay, outercourse, the whole sexual thing, enter me, “kinda had sex,” “I went in like twice but I didn’t ejaculate,” incident, anything, and session.	47
Bleeding	321 (28.3%)	bleeding, period, menstrual period	period	256
			bleed, bleeding or blood	78
			menstruation	24
			cycle	20
			spotting	16
			menstrual period	3
			menses	2
			other terms, including rag, flow, my courses, a little stain, and shedding lining.	35
Ejaculation	117 (10.3%)	ejaculation	ejaculate, ejaculation	81
			come, came	15
			cum (verb)	11
			orgasm	2
			other terms, including climax, shoot the sperm, busted (“busted in me” or “busted a nut”), discharged, and injectculation.	14
Sexual organs	92 (8.1%)	(none)	male organs:
			penis	12
			it	5
			other terms: dick, mine, private, him (as in “when I felt him he was dry”)	4
			female organs:
			in, inside	66
			vagina	20
			vulva	3
			other terms, including private, kooch, buttox area, and “down there.”	7
Sexual fluids/ pre-ejaculation	63 (5.5%) 31 (2.7%)	preejaculatory fluid, sperm	sperm	38
			precum	15
			preejaculation	8
			semen	8
			preejaculatory fluid	4
			cum (noun)	3
			cervical mucous	3
			other terms, including preejaculate (noun), bodily fluids, and “seamen of man.”	23
Reproductive processes	58 (5.1%)	ovulation, implantation, fertilization, tubal transport	ovulation	35
			fertilization	16
			conception	9
			implantation	6
			other terms, including impregnation, enpregnanted egg, and embedding.	5
Fetus or embryo	24 (2.1%)	fetus	baby	11
			fetus	7
			other terms: the unborn baby, unborn child, it, incoming baby, the pregnancy, and human embryo.	6
STI	5 (0.4%)	sexually transmitted infections	STD	5

The range of terms used varied widely by category: email authors used more than 30 terms or phrases to describe sexual acts but used only the term STD (sexually transmitted disease) when referring to STIs. Formal or medical terminology predominates in most categories. For example, while we found a creative variety of terms describing sexual acts, including doing it, love-making, making love, hooking up, slept together, infidelity, oral sex, fingered, doing foreplay, anal sex, and outercourse, the list is clearly dominated by the terms sex, intercourse, and sexual intercourse.

We also found that people were most likely to use avoidance strategies in writing about sex and genitals. Indeed, writers were almost as likely to refer to male genitals as “it” or “him” as they were to write “penis.” And writers showed considerable preference for vague references to “inside,” “down there,” or “private” in describing female genitals over terms like “vagina” (Table 1).

However, there are limits to a quantitative approach to describing language used in these e-mails. First, writers may mimic what they read on the website. The fact that more people use the term “emergency contraception” than the “morning after pill” may not reflect their everyday terminology so much as the fact that they recently saw that phrase on the website. Thus Table 1 lists not only the terms used in the e-mails but also the terms used on the website itself. The predominance of the term emergency contraception in these e-mails when other research suggests that morning after pill is a much more widely used popular term reinforces other research that has demonstrated how the language of health is formed in contexts of mutual mimesis (Pinto 2004).²

Second, this research reflects only the e-mails written by people who arrived at the EC website. It says nothing about the language of people who never reached the website. This distinction is important because, at the time these e-mails were sent, the website itself was biased towards using medical language, and was highly ranked by most search engines for searches using the term emergency contraception, but fared considerably less well in searches for popular terms such as morning after pill. In short, mere use of this website is a language filter. Thus a more robust way of understanding the language of sexual health in these e-mails is not just to enumerate the terms authors used but rather to examine the range of this terminology in the broader context of their concerns about sexuality and reproductive health.

Contextualising the terminology and interpreting meaning

Writers of the emails in our study used a variety of modifiers and qualifiers when asking questions about sex. For example, one e-mail writer wrote, “We didn’t have full out sex, he just put it in and out.” Another said, “My girlfriend and I kind a had sex, I went in like twice.” And in the following e-mail, the writer gives different meaning to “sex” as opposed to “sexual intercourse”:

HI I needed some advice I had unprotected sex this past Sunday and Monday…. and kind of fooled around but there wasnt any bodily fluids past out of the body last night… I was wondering if I was protected for Monday and Sun after taking the pill last night the pill be Plan B … I dont know if I should try and get some more Plan for what happened last night even tho I know me and my boyfriend didnt have sexual intercourse nor pass any fluid… please get back to me I am very worried and dont know what to do… ³

Such language and the use of modifiers that can be variably interpreted suggest a modern lack of consensus over what constitutes sex and sexual intercourse. Just as former President Clinton’s statements about sexual relations and the JAMA controversy sparked debate about the definition and meaning of sex, unpacking the language of sex in these e-mails reveals how language is an (imperfect) indicator of how notions of sex vary and are informed by moral, medical, legal, and political interpretation (Binson and Catania 1998, Mann 1999, Quirk, Rhodes, and Stimson 1998, Sanders and Reinisch 1999, Seal 1997).

Interpreting the meaning of terms for sex in these e-mails calls for a close examination of textual context. The first two e-mail examples are from writers who are not certain whether to classify as sex actions which involved vaginal penetration but did not involve ejaculation, in the context of articulating questions about whether EC was needed. In the context of assessing risk of pregnancy entailed in a specific act, confusion over what to call a sexual act that briefly involved penetration and did not lead to ejaculation has its own logic. Qualifiers such as “full out” or “kinda” reflect writers’ uncertainty over not only whether EC is needed but also what occurred during sex: not only were many unsure whether a specific act could lead to pregnancy, some writers were not even sure whether or not ejaculation had occurred. In contrast, the third e-mail, which describes past actions as constituting unprotected sex but not sexual intercourse, reflects the writer’s anxiety about possible pregnancy risk even in the absence of penetration, and it is this fact that may lead to her ambivalence over what label to give to the sexual activity that occurred.

In short, the contexts where assessments of sexual activity take place - whether pregnancy risk, disease transmission risk, or moral risk - influence the definitions people give to terms that are multiply defined or whose definitions are culturally contested. By “moral risk” we mean a concern that people have about sexuality that is grounded in cultural or religious ideals about who in a given society should be having sex, what kind of sex they should be having, and with whom. Thus surveys that reveal that some abstinence-pledging teens do not count oral or anal sex in their definitions of “sex” and virginity (Brückner and Bearman 2005) suggest that they give a different moral weight to penile-vaginal sex over anal and oral sex. When the definitions given to sex by teenagers attempting to mitigate the moral risk of sexuality are later evaluated by analysts estimating risks of exposure to STIs, it is unsurprising that there is a gap between the way each group defines sex. In contrast, consider the following e-mail:

I don’t really know how to explain myself because this is the first time something like this has happened. On Sat. my boyfriend and I slept together, but I since we had no protection, I asked him to keep his boxers on. And ejaculation did not occur. I’m currently on my menstral cycle, is there a chance that I might be pregnant? Should I take emergency contraceptives?

Here we find the opposite situation: instead of declining to label an act of brief vaginal penetration sex, this writer uses the phrase “slept together,” which is often used as a euphemism for penetrative sexual intercourse, to describe an act that did not entail ejaculation and apparently—since the man was wearing cloth underwear—did not involve penetration, either. In trying to explain how the writer then could worry about pregnancy risk, we hypothesise that her moral definition for the sexual activity that has taken place has influenced her perception of the act sufficiently that she extrapolates from moral risk to suspect pregnancy risk. However, we cannot rule out the possibility that this may simply represent an abysmal lack of knowledge about how pregnancy occurs.

Fully 4% (47) of all the e-mails came from women who reported that they were regularly and consistently taking some sort of hormonal contraceptive, yet wanted to know if they should take EC. Many of these questions reflected dual method users who experienced a condom break/slip, as the following example illustrated:

Could I get pregnant if I’m on the three month shot ? because I had sex but the condom broke. Do the shot keep you from getting pregnant ?

Nearly half of these (22) described having engaged in “unprotected” sex. In other words, these writers were concerned about pregnancy risk after engaging in sex while ostensibly using hormonal contraceptives correctly (we did not include in this number writers who were concerned about pregnancy risk after missing oral contraceptive pills).

Just as the blurring of lines between moral and pregnancy risk may lead to different definitions of sex, this finding seems to suggest that evaluation of risk in one area - disease transmission - has spilled over into concern about risk in another area - pregnancy. Cognisant of the risks of disease transmission entailed in sex without a condom, the writers worry that they are unprotected in another realm as well, despite the fact that they are taking hormonal contraceptives, which, when correctly taken, are considerably more effective in preventing pregnancy than condoms.

This confusion over what constitutes protection from unintended pregnancy is undoubtedly the unintended effect of public health campaigns designed to educate the public about protecting themselves against STI transmission. The barrage of admonitions to engage in “protected sex” by using a condom apparently leaves some people with the impression that sex without a condom is therefore unprotected sex, regardless of what they are trying to protect themselves from, be it disease transmission or pregnancy. This perception was perhaps best articulated in the following question:

I have a question about how unprotected sex is defined as when pertaining to pregnancy and not STD’s. For example, if my partner didn’t use a condom but I’m on birth control is that unprotected sex when it comes to pregnancy and should one still take EC?

This is a poignant reminder that the language chosen to convey public health messages structures thought in subtle and even unconscious ways—ways that may be unanticipated by the designers of public health information campaigns. There are important implications for clinical care: when a patient tells a provider that she has had unprotected intercourse or even “sex,” it is critical for the provider to solicit more information about what those phrases mean to the patient. There are also important implications for research on sexual risk behaviour (Quirk, Rhodes, and Stimson 1998, Groes-Green 2009).

Discussion

The definitions given to unprotected sex in these data are particularly interesting because the terminology of “protection” is relatively less discussed than the definition(s) of sex, even though the language of protection has become increasingly politicised in the USA. In 1999 (the same year as the Clinton impeachment trial), then-Representative Tom Coburn, a Republican from Oklahoma and a physician, lobbied Congress to legislate that condom packaging be required to carry a warning that condoms provided “little or no protection” against human papillomavirus (HPV or “genital warts”), a common STI (Boonstra 2003, 2005). In response, in 2000 the National Institutes of Health (NIH) convened a panel to review the evidence that condoms protect against eight STIs, including HPV, and reported that there was an absence of evidence that condoms protect against HPV. Yet Coburn and his supporters concluded that an absence of evidence equated with an evidence of absence, with organizations like Concerned Women for America proclaiming that “condoms do not protect against human papillomavirus” (CWA 2004), and such interpretations appear to have informed the application of federally funded abstinence-only sex education curricula under the George W. Bush Administration, in which educators are not permitted to discuss contraceptive methods except in terms of their rates of failure in protecting against both STIs and pregnancy (Sonfield and Gold 2001).⁴

Although the language included in many of the emails received suggest that the writers are adolescents, we have no systematic demographic information on writers. While other websites have attempted to capture such information by requiring writers to provide certain information before submitting a question (see Gainer et al. 2003), this requirement may act as a barrier to those who seek confidential information about sensitive matters and want to take refuge in the relative anonymity of e-mail. Not requiring writers to provide personal information may thus increase confidence and willingness to write, but it also means that we cannot correlate language use with age, education level, native language, cultural background, income, or place of residence. All of these may be important determinants of both knowledge and misinformation, language use, and sexual and contraceptive practices, as well as the ways people conceptualise their own bodies and interact with perceived experts or medical professionals.⁵ Lacking such data, we cannot systematically analyse the important relationship between such social indicators and language in these e-mails. However, we do know that 70% of all users of the website are based in the USA.⁶

But there are also limitations on efforts to correlate language used for talking about the body, sex, and medicine with demographic indicators. Seeking fixed categories and correlations between cultural background and language use fails to capture the way that many people are in the process of learning about sexual activities and reproductive processes and the ways that their language reflects this. This problem is especially true for young people, and for cultures and societies that attempt to deny knowledge of sexuality and contraceptive processes to the sexually uninitiated, such as the contemporary USA, where there has been a recent trend towards replacing previously candid and accurate sexual education in public schools with curricula intended to inculcate an ethic of abstinence from sexual activity.7 The issue that health care providers must grapple with when serving patients of varying ages is not only that young peoples may have their own sub-cultural jargon for speaking about sexuality and reproductive health, but also that they represent a group of individuals who are at the point of initiation into new sexual practices, and therefore may lack the social and cultural knowledge and language for communicating about sexuality and reproductive health that others have already acquired. Thus correlations between language use and age, culture and ethnicity, or class background may fail to capture the processual nature of learning about sexuality and reproduction as well as the processes by which individuals learn how to describe such topics with language, whether to themselves, to friends, or to ‘experts’ or medical professionals (Anderson, Santelli, and Morrow 2006, Groes-Green 2009).

Interacting with the Emergency Contraception Website is a part of this process. Some writers may decide to send an email after minimal interaction with the website, in which case they may draw on language that they use in discussing matters of sexuality and health with peers and sexual partners, or they may use idiosyncratic terms, or they may draw on past experience in dealing with medical providers. On the other hand, many writers have decided to write an email after reading about EC on the website, in which case they may adapt their linguistic practice to mimic the terms and language used on the website. The persistence of writers in using terms such as “the morning after pill,” which was not found on the website itself at the time these e-mails were received, is an indicator of the persistence of popular linguistic practices in the face of attempts by public health education campaigns to literally “spread the word” using other terminologies.

In sum, for all of these reasons, we caution that, although the quantitative enumeration of terms used may be suggestive of certain linguistic patterns in describing sexual practices and reproductive health as pertains to contraceptive use, they cannot be read as representative of any specific population (other than the users of this particular website). However, the linguistic practices that we observe suggest more generally a range of possible thinking and writing about sexuality and reproductive health that medical professionals, especially those who serve diverse communities and adolescents, may encounter. And it is important to examine outliers, including individuals’ unique language for describing sexuality and reproductive health. Just as Malterud and Bærheim’s (1999) study of the language used to describe urinary tract infections (including the phrase “peeing barbed wire”) revealed symptoms that were not mentioned in medical textbooks and literature reviews, so can individual descriptions of sexual health reveal experiences and interpretations that may be unanticipated by health professionals (see also Gilliam, Gay, and Hernandez 2006).

Conclusions

Our examination of language in e-mails sent to this reproductive health website revealed a wide and creative variety of terms used to describe sexuality and reproductive health, while at the same time showed that certain terms, such as sex and intercourse, are hegemonic. Studying terminology in context reveals that a single given term for sexual intercourse can vary widely in meaning between writers, ranging from vaginal-penile sex in which ejaculation occurs to proximate contact of genitals through layers of clothing and without ejaculation.

An increasing body of literature has examined the differences between everyday language and medical language and the extent to which medical professionals and non-professionals differently imbue terminology with their own disciplinary and idiosyncratic content (Bourhis, Roth and MacQueen 1989; Boyle 1970; Hadlow and Pitts 1991; Ong et al. 1995, Tsang and Ho 2007). Avoidance of terms may also be a display of public modesty for writers who feel that certain terms cannot be decently used in public discourse or out of the context of sexual activity. In mimicking the language of the website or otherwise showing a preference for formal and medical terminology over everyday language for sexual and reproductive processes, writers may be putting linguistic distance between passionate sexual activity and requests for dispassionate analysis of the consequences of that activity, or they may seek to avoid the judgment of the faceless person they hope will read and answer their questions, or they may be trying to appear knowledgeable, yet the meaning they attribute to such terms is often idiosyncratic and at odds with medical definitions of these terms.

Finally, one of the most striking findings of this study was the relatively large number of writers who used the term unprotected sex to refer to sex without a condom but with correct use of hormonal contraceptives. Here in these e-mails we see a fascinating slippage between people’s sense of pregnancy risk and disease risk, and more broadly between ideas of physical and moral risk associated with sex, highlighting the extent to which the discourse on sexuality operating at the intersection of family planning, medical science, and political debate is dominated by an aura of danger and the paradigm of risk control. Indeed, this discursive orientation towards the risk paradigm of sex is reflected in the language of the EC website, where sex is framed in terms of pregnancy prevention, safe and unsafe practices, and protected vs. unprotected sex.

These e-mails challenge traditional definitions of culture as something that is shared across a population – a definition that is implicit in many applications of the cultural competence frame (Lee and Farrell 2006). Health professionals are becoming increasingly attentive to language and culture in an attempt to understand and provide better care to their patients (McQueen and Henwood 2002), and calls for medical professionals to attain cultural competence in order to better care for minority and underserved patient populations have become so widespread in medicine that they are codified by the American Medical Association (AMA 2004, Betancourt 2004). The evidence of our study, however, points to the extent to which language surrounding controversial topics (such as sex) varies and is contested, even within cultures. One clear intra-cultural divide in the language of sexuality and reproductive health is that between health professionals and lay people. Yet it is apparent that a range of other flows of linguistic practice surrounding sexuality, from political debates over sex education and condom labelling to international health education campaigns surrounding sexually transmitted diseases, influence how people think and talk about and sexual practices (Pigg 2001).

There are clear implications for public health research, health education, and clinical practice. Research that asks questions about sexual practices can be reliable only if there is clear consensus between the researchers and the research subjects about how those sexual activities are defined. Sexual health education campaigns must take into account the range of popular expressions that can be used to express a single concept, while simultaneously keeping in mind the variability in meaning that may be attached to a single term. Clinicians need to similarly be attuned to the range of expressions that patients may use to describe their sexual lives and reproductive health. They also need to listen closely to their patients, and solicit narratives of events, symptoms, and patient interpretations of their health. Further, they need to pay attention not only to what patients are saying but how they are saying it and how patients present themselves in their statements. Only in dialogue will the meaning that patients attach to words and indeed to their health become clear.