On rare occasion, I've said confusing things about sexual orientation, nearly implying that it is a choice. And of course it isn't a choice. But there is a complication that confuses many people: Sexual orientation has multiple definitions.
Take, for instance the word "gay". "Gay" can have any number of definitions, but I will simplify it to these three:
- Desire definition: A gay person is someone who is attracted to members of the same sex but not the opposite sex.
- Behavior definition: A gay person is someone who forms romantic and/or sexual relationships with members of the same sex but not the opposite sex.
- Identity definition: A gay person is someone who identifies as gay.
The Real, the Useful, and the Really Useful
In some sense, the desire definition is the "real deal", and the other two definitions are just proxies. Behavior and identity are far too contingent, and can be changed by mere circumstance. The desire definition, on the other hand, encapsulates something essential about the group, and gets to the heart of what's really going on.
However, I hesitate to go too far in assuming that the desire definition is the "real deal" in all its aspects. For instance, I believe that "attraction" is actually a whole set of distinct phenomena, and we simply categorize them together because it is useful to do so. As an analogy, take the word "weather". Weather describes a whole set of distinct phenomena including wind, rain, and temperature, but we find it useful to put them all in the same section of the newspaper. But perhaps it could have been different; we could have put temperature in one section, with another section for cloud cover and stargazing. Similarly, "attraction" could be divided up in any number of ways, or even joined with other essential qualities. But we divide it up the way we do because that is the most useful to most people.
In conclusion, the desire definition is not entirely about what is "real" but also about what is useful. And once we admit that usefulness is important, we must take the other definitions seriously, because they are also very useful.
The behavior definition is most obviously useful for some medical issues. For example, if we're studying the role of gay men in the spread of HIV, the only important issue is who they have sex with. It clearly doesn't matter if they identify as hetero, or if they have no attraction to men. A related issue is the ban on gay blood donors. The ban actually applies to "men who have sex with men", not to gay people, because they want to make it explicit that only behavior matters.
I think the behavior definition is also useful on a personal level. By the desire definition, I'm in a gray area, so I could spend quite some time describing the details of orientation. But at some level, many of the details are unimportant, and the only important question is, am I "available" for a same-sex relationship? Interested guys need to know. (The answer is yes.) This is not a question about desire, but about my future behavior.
The identity definition is perhaps the most useful in everyday situations. If someone says that they're gay, that's what you go by. For one thing, you want to be respectful of people's wishes. A person's identity is determined by a particular combination of behavior and desire; you do not get to decide what that combination should be, they do. Furthermore, it is very presumptuous to think that you know a person's orientation better than they know themselves. When it comes to individuals, self-identity is one of the most reliable indicators of sexuality. (In large groups, it is less reliable because there are systematic biases affecting self-identity.)
But the identity definition confuses people because it is rarely explicitly stated as a definition. This is probably because when you state it as a definition, it sounds really silly. A gay person is someone who says they're gay. But why do they say they're gay? The identity definition is not meant to be a stand-alone definition, but it is still very prevalent and useful.
A Matter of Choice
The question of whether sexual orientation is a choice is a highly politicized one. By the desire definition, sexual orientation is not a choice, but by the other two definitions, it is a choice. So it is not surprising that different political groups choose to emphasize different definitions.
I think the desire definition is the most relevant one in determining whether sexual orientation is a choice. The desire is the primary cause of both the behavior and identity. We cannot go any further down the causal chain, because the cause of the desire is too obscure.
At the same time, I also assert the right for people to choose their behavior and identity. I think in most situations, it's advisable to choose behavior and identity that align with desire. But it's not an absolute moral obligation. Also, it's not always possible, since the underlying desire tends to be more nuanced than our range of choices for identity labels and behavior.
Some groups, such as the Catholics, are okay with gay people, but not okay with gay behavior. The Catholic Church correctly understands the distinction between what is chosen and what is not chosen, and applies their understanding to their rules. And yet, I still consider their rules morally backwards. It just goes to show that conceptual understanding is not enough.
Definitions of Asexuality
Since my expertise is in asexuality, I should comment on how the different definitions apply to asexuality.
When the asexual community was younger, certain leaders decided to emphasize the difference between asexuality and celibacy. If you rarely or never experience sexual attraction to others, then you are asexual. If you don't have sex with anyone, then you are celibate. I believe this was a very shrewd political move, because it clearly distinguished between the desire and behavior definitions by giving them different words entirely.
There is a great danger that the behavior definition could erase asexuality. After all, lots of people are perpetually single. Lots of asexuals are in relationships, both sexual and nonsexual. There's not any particular behavior which clearly corresponds to asexuality. So if we used the behavior definition of sexual orientation, it's not clear where asexuality would fit in, if anywhere. (Similar concerns affect bisexuality, causing bisexual erasure.)
Therefore, the behavior definition is strongly discouraged. But it's still occasionally used sparingly in combination with the desire definition.
The identity definition, of course, is used all the time.
It's much less clear which definition is used for romantic orientation among asexuals. There's a lot of disagreement about whether an "aromantic" is someone who never experiences romantic attraction to people, or someone who just doesn't ever pursue relationships. It's essentially a dispute between the desire and behavior definitions of romantic orientation.
Many asexuals intuitively understand that the desire definition of romantic orientation is the best. But at the same time, the behavior definition is obviously very useful. A lot of asexuals say that they have romantic attraction, but avoid relationships altogether. Clearly, the fact that they avoid relationships is just as important, if not more important than the presence of romantic attraction.
To make things more complicated, the desire definition of romantic orientation is not clearly "real". There are all sorts of different things that people classify as romantic attraction, and it's not clear that they belong in the same category. There is a significant minority in the community that doesn't identify by any romantic orientation.
Given the lack of consensus on definitions, the identity definition tends to prevail.
(Some ideas in this post were inspired by Kenji Yoshino on bisexual erasure, via Asexual Explorations)