One of the generally accepted criteria for categorizing a sexual desire as an orientation is its relative stability across time (Moser, 2016; Seto, 2012). However, there is no set quantitative standard for how long a sexual desire must last before it is considered stable or intrinsic enough to be considered an orientation, and it is worth considering what kind of evidence would be required for any candidate orientation to meet that standard and, more to the point of Brotto and Yule’s (2016) article, whether asexuality meets that standard.

More generally, as homosexuality and heterosexuality are the archetypal sexual orientations against which other candidate orientations are compared, it is crucial to carefully consider the analogs between asexuality and homosexuality/heterosexuality. However, there are a number of distinctions between asexuality and homosexuality/heterosexuality that present real complications, both conceptually as well as in terms of operationalization and measurement. I raise these issues not under the presumption of having the answers, but rather in the spirit of outlining the implications of testing and labeling asexuality as an orientation.

First, it is tempting to require some degree of formal, inter-wave longitudinal reliability in self-identification before a candidate orientation is validated. My own work (Cranney, 2016) found relatively low reliability between waves in the Add Health survey. Specifically, of the 14 people who indicated “no sexual attraction” in Wave III, only three went on to do so in Wave IV (Table 2). However, those three constituted 14 % of the 21 in Wave IV who indicated no attraction. In other words, most people who indicate “no sexual attraction” do not do so in the next wave; however, it is clear that the two measures are not completely unrelated.

While it is clear that a strong, inter-wave reliability would demonstrate stability, does a lack of such demonstrated reliability necessarily demonstrate instability? I argue that it does not.

Instrumentally, the measurement of dimensions of asexuality in population-based samples is problematized by a lack of common knowledge about asexuality (Cranney, 2016). To wax anecdotal for a moment, people generally seem to know enough about Greek prefixes to deductively derive the fact that an asexual is a person with little or no sex drive, but, outside of this on-the-spot conclusion, very few people have personal experiences with the asexual community or a self-identified asexual family member, or consciously consider asexuality as an orientational option akin to homosexuality or heterosexuality. This point was brought home to me recently when I was explaining this commentary to a gay relative of mine who mistook asexual to mean something along the lines of pansexual. Furthermore, the romantic/sexual distinction, vitally important in the asexual community, is often conflated among the general population, with people assuming that their sexual orientation matches their romantic orientation. Diamond (2014) raised the possibility that traditional definitions of orientation privilege sexual over relational affection. This may be so. Thankfully, the asexual community and scholarship have already developed frameworks and languages for the romantic/sexual distinction, perhaps because the difference is more salient in the asexual case. While this can complicate measurement, it parsimoniously allows for more precision (a heteroromantic asexual, for example) without privileging one form of desire over the other. These complications do not mean that attempts to measure the stability of asexual desire, behavior, or identity are doomed from the start, but it does mean that there are factors that complicate a direct comparison with figures for homosexual/heterosexual stability.

However, even if these complications were addressed, would it be necessary to require a formal test of inter-wave reliability to justify a belief that asexuality meets the stability criterion? To draw an analog with the case of homosexuality, both the scholarly consensus and the public at large concluded that homosexuality was an intrinsic, long-lasting state long before any formal longitudinal analysis of sexual desire had been conducted. Rather, the consensus largely arose out of people taking at face value the reports of lesbians and gays of the innateness and stability of their sexual desires. Indeed, in the public consciousness, the pendulum may have swung too far the other way, with many people conversely assuming no sexual fluidity across the lifecourse. In the same spirit, there is no reason to hold asexuality to a stricter standard and to not assume good faith on the part of individual asexuals who report a non-problematic, long-running lack of sexual desire. These reports are found both in the qualitative literature and the quantitative literature that uses survey questions in such a way so as to capture only those who reported a lifelong lack of sexual desire (e.g., Bogaert, 2004).

While it is increasingly clear that some people do indeed report a lifelong, dispositional lack of sexual desire that would meet the stability criterion for a sexual orientation, it still leaves unanswered the question of how long somebody needs to report lack of sexual attraction for it to be considered their particular sexual orientation. This is another case where the asexuality/homosexuality analogy can be problematized. To draw on some hypotheticals to elucidate the point, while experimental bisexuality is a reality, I know of no case in the literature where an individual has experienced a very brief change from exclusive homosexuality to exclusive heterosexuality (or vice versa) and back again. However, such events do happen along the asexual/allosexual continuum when, for whatever reason, people do not feel sexual attraction for a time period, running from days to years. For example, if an individual is temporarily on an antidepressant that reduces sex drive to asexual or even gray-asexual levels, is that individual not a heterosexual or homosexual for the week that they are on the drug? Would it count as an orientation change if they were on that drug for months or even years? Most would consider this hypothetical case as being fundamentally different from asexuality as an orientation, but it raises the general question of where exactly along the temporal continuum the boundary line is between a temporary response to external cues and/or medication and a long-running, dispositional state.

The specter of psychotropic manipulation raises another distinction between asexuality and homosexuality/heterosexuality that should be noted. Another generally accepted characteristic of an orientation is that it is innate and stable enough to be impervious to psychotropic or therapeutic manipulation (Moser, 2016). While it is true that attempts to “convert” dispositional asexuals to allosexuality may be as ineffective as attempts to “convert” gays to straight or vice versa, lack of sexual desire can be a side effect of various medications and life experiences. There is no comparable drug that has the side effect of making one homosexual or heterosexual given a baseline level of sexual desire.

One additional brief note along these lines: Brotto and Yule (2016) point out that the distinction between asexuality and disorders of low sexual desire as outlined in the DSM-5 hinges on clinically significant personal distress arising from the lack of sexual desire. However, another category that needs to be considered is the case where lower or nonexistent sexual desire is not considered problematic in itself, but “accompanies” (I am intentionally avoiding the use of the term “comorbid” here) symptoms that are universally considered unhealthy. For example, abnormally low levels of testosterone in men have been associated with low sexual desire, as well as a variety of clearly unhealthy outcomes such as osteoporosis (Shores, Smith, Forsberg, Anawalt, & Matsumoto, 2012). This is another limitation in the comparison between asexuality and homosexuality/heterosexuality: the latter orientations never arise out of unhealthy conditions, whereas lack of sex drive can be a characteristic that accompanies clearly unhealthy conditions, even if no personal distress is caused by the lack of sexual desire in itself. Here it may (arguably) be more accurate to avoid labeling a non-problematic lack of a sex drive as an orientation, even if no personal distress is involved from the lack of sexual desire per se.

It may appear that I have simply offered more complications to the already difficult venture that Brotto and Yule (2016) have undertaken of clearly categorizing asexuality, so here is my concrete position: some variations of lack of sexual desire do indeed meet the criteria for a sexual orientation as discussed by Brotto and Yule. However, there are circumstances where researchers must tread carefully when invoking comparisons to homosexuality/heterosexuality in terms of stability, measurement, and conceptualization, because the parallels with these paradigmatic sexual orientations are not precise in every situation.