I have argued that “sexual orientation” has a number of related by slightly different meanings: sexual orientation as a legal category, as a social category, and as a scientific category. Depending on which meaning of “sexual orientation” we’re dealing with, the issues involved in determining/deciding if asexuality is a sexual orientation or not are somewhat different. In this post and the next, I intend to address it as a scientific category.
Given the current state of research on asexuality--and the research on sexual orientation more generally--I do not think it is remotely possible to answer the question of whether asexuality is a sexual orientation in anything but the most tentative way. Asexuality has received very little research, so there is very little data to work with. Sexual orientation more generally has received quite a bit more research, but things are very uncertain there as well.
Regarding “sexual orientation” more broadly, there are a number of open questions. To restate a few from my last post: What is “sexual orientation” is measuring? How should it be operationally defined? Is sexual orientation in men the same thing as sexual orientation in women? Should sexual orientation include all of the factors involved in a person’s pattern of sexual attraction (e.g. personality features, physical features, etc.) or be limited to the part gender plays in their sexual attraction? Is "sexual attraction" more about a person's patterns of sexual attraction or about patterns of sexual arousal? Is all sexual attraction part of sexual orientation, or does some sexual attraction stem from some other cause? What is "sexual attraction" anyway? Are heterosexuality, homosexuality, and bisexuality each somehow unified groups in some scientifically important way or are they just conglomerations of merely superficially similar features lumped together? How useful a concept is "sexual orientation"? There are opinions on these questions but not much consensus.
And if it's not clear what "sexual orientation" is, then that makes answer the question of whether asexuality is one or not even more difficult.
One question of fundamental importance is "Scientifcially, how useful of a category is asexuality?" Thinking about this problem, along with think about Nelson Goodman’s New Riddle of Induction, which I learned about at around the same time, has gotten me thinking a lot about categories. Ultmiately, what seems to make a category useful is that identification of someone or something as a member of some category liscenses (probabilistic) inference about them/it with respect to properties other than that which defines them as a member of that category. Thus, if I know someone is from Korea, they probably speak Korean, they probably have a certain ammount of knowledge about Korean, etc. In meeting people for the first time, identifing such categories can be very important for ascertaining what knowledge/beliefs are liekly to be mutual and waht aren't. (If I'm talking to a linguist, I can assume certain knowledge about linguistics that I cannot assume for others.) Likewise, if I see an apple, I can infer it's probably good to eat (unless it looks like it rotting or something) but if I see a table, it's probably not good to eat.
With something like asexuality, there seem to be three possible types of correlates of interest: causes, effects, and other. To illustrate these, let’s use biological sex. There is a strong correlation between having two X chromosones and being anatomically female. Now, it’s certainly not 100% (there are a number of intersex conditions that make exceptions; also there is SRS), but it’s a pretty strong correlation. This enables people to understand and research causal mechanisms involved. Another way of thinking about the category deals with effects. If we take the (extensionally somewhat different category though largely overlapping) category of women, we can think about effects: what are the effects of being a woman in some particular cultural context? Here, it’s usually not going to be effects that happen to every single woman that are of concern: rather, it’s going to be tendencies and likelihoods people are interested in, or issues that are more often of concern for women than non-women, etc. The third type are going to be things that happen to be correlated but no one really knows why. Typically, there are some of the most common found in research, and people are interested in trying to figure out what causal mechanisms might be invovlved in creating some difference.
With respect to asexuality, I think we have these three things involved. How scientifically useful the category is, I think, depends on how strong the causal and/or effectual correlations turn out to be. This raises an interesting point: causally, what we call asexuality may well be (in my view, almost certainly is) a whole bunch of things just sort of lumped together into one category. However, the social consequences (effects) of lack of sexual attraction (whatever “causes” might be involved) seem to be similar despite this. To a large extent, it is the similarities, overlaps, and criss-crossing of shared experiences that has led to the creation of this identity category in the first place. And once the category is created, a community of discourse and an identity is created. This results in common vocabulary and concepts that people use to understand themselves and to communicate. (If I'm talking to other asexuals, I can generally assume knowledge of certain words that most people probably won't know. I can make jokes based on shared sorts of experiences, etc.)
What this means for scientific theories of asexuality is simply that the usefulness of the category “asexual” is dependent on what we’re trying to understand (which is true of pretty much all of our other categories anyway.) Ultimately, it’s the cause-based-categories and the effect-based-categories that are the most important and the most interesting—if the goal is to understand how stuff works. But, where humans are concerned, establishing cause and effect are extraordinarily difficult.
And here’s another thing: suppose that we were able to figure out what causes asexuality/asexualities. Would this create the danger that those caused-based-categories are seen as somehow “real” asexuality, and anyone who didn’t fit those, their asexuality is somehow less real? Even if, at the level of experience, what they’re feeling is very similar to those proclaimed to be “real asexuals,” even if the social effects are very much the same? There is no a priori reason to expect cause-based-categories to exactly coincide with effect-based-categories. I don't know the answer to this, but it is a matter worth considering.