BY-NC-ND 3.0 license Open Access Published by De Gruyter July 4, 2017

Sexual Orientation and Choice

Saray Ayala


Is there a choice in sexual orientation? [Wilkerson, William S. (2009): “Is It a Choice? Sexual Orientation as Interpretation”. In: Journal of Social Philosophy 40. No. 1, p. 97–116] argues that sexual desires require interpretation in order to be fully constituted, and therefore sexual orientation is at least partially constituted by choice. [Díaz-León, Esa (2017): “Sexual Orientation as Interpretation? Sexual Desires, Concepts, and Choice”; In: Journal of Social Ontology] critically assesses Wilkerson’s argument, concluding that we still lack a good argument for the claim that choice plays a role in sexual orientation. Here I examine Díaz-León’s response to Wilkerson. I introduce what I call the conceptual act theory of sexual orientation, and argue that even if interpretation were not necessary to constitute sexual desires, it is a necessary element to constitute what we call sexual orientation. However, I conclude that even if we agree that interpretation is involved in sexual orientation, it does not follow that there is a choice involved.

1 Introduction

The idea that sexual orientation is not a choice is very common (the “born this way” and “not a choice” slogans so fiercely endorsed in popular culture and in the mainstream LGB political movement), although this is far from unproblematic (Stein 2011; Behrensen 2014). This work discusses William Wilkerson’s recent defense of an alternative, and Esa Díaz-León’s critique of it. Wilkerson (2009, 2013) argues that sexual orientation is at least partially constituted by choice, and Díaz-León (2017) critically assesses his argument, concluding that we still lack a good argument for the claim that choice plays a role in sexual orientation.

My contribution to this debate has two parts. The first part engages Díaz-León’s critical response to Wilkerson. According to her, Wilkerson does not provide a sound argument in favor of the idea that sexual orientation requires interpretation of feelings and desires. In particular, Díaz-León questions the claim that sexual desires require interpretation in order to be fully constituted. I have two comments on this. On the one hand, although I agree with Díaz-León that a metaphysically independent layer of desires might be there prior to any interpretative act, I argue that this layer might not correspond to what we call sexual orientation. In order to build my argument, I draw a parallel with the conceptual act theory of emotions (Barrett 2006; Barrett et al. 2015). I argue that this independently constituted layer of desires could consist of basic affects that need to be conceptualized in a particular way in order to become what we call sexual orientation. Thus I conclude that even if interpretation is not necessary to constitute sexual desires, it is a necessary element to constitute what we call sexual orientation. On the other hand, although I agree with Díaz-León that interpretation does not in general entail constitution, I argue that it could be the case that the specific way in which we conceptualize the affects relevant to sexual orientation could be one that imposes sharp category boundaries and, in such a case, at least some of those feelings would be affected (in a constitutive sense) by how we categorize them.

The second part brings us back to Díaz-León’s conclusion that Wilkerson’s argument does not successfully show that sexual orientation involves choice. This part engages a premise of Wilkerson’s argument that Díaz-León does not address in her critique, which is his claim that interpretation requires choice. I argue that even if we agree that interpretation is involved in sexual orientation, it does not follow that there is choice.

2 Constituting vs. Interpreting

2.1 Wilkerson’s Metaphysical Claim and Díaz-León’s Response

In Wilkerson’s argument for the conclusion that sexual orientation involves choice, the first premise states that sexual desire is partially constituted by interpretation; [1] the second premise states that interpretation requires choice. From these two premises the conclusion follows that sexual orientation involves choice (Wilkerson 2009, p. 100). While my second comment engages the second premise, I will for now focus exclusively on the argument for the first premise and Díaz-León’s response to it. In particular, I will focus on her response to the metaphysical argument that Wilkerson puts forward. [4]

Wilkerson defends the view that our acts of classifying and labeling transform what is classified, in this case, our feelings and sexual desires. And they transform our feelings and sexual desires in a constitutive sense. That is, the way we conceptualize and classify our desires changes their nature. The fact that the same collection of desires or experiences can be classified in different ways seems to work as support for Wilkerson’s claim, the reasoning being that the possibility of different (correct) classifications shows that the desires/experiences are not finished products. The assumption here is that there is only one correct way of classifying something that is already constituted independently of the classification. If different classifications are possible, this means that what is being classified is undetermined.

But as Díaz-León points out, the fact that desires and experiences can be conceptualized and classified in different and even incompatible ways (let us call this synchronic flexibility) “does not entail that those experiences were not fully formed prior to our conceptualizing them in this or that way” (Díaz-León 2017, p. 10). Also, the fact that over time we might change our interpretation of the same sexual desires or experiences (we can call this diachronic flexibility), does not imply that they were not fully formed prior to our interpretative act. Díaz-León illustrates this response by appealing to visual color experiences. The same series of color experiences can be classified in different ways, but this only shows that the same token experience instantiates more than one type of phenomenal state. For instance, the same token of an experience of blue could instantiate the type “blue experience” and “dark blue experience”, and actually many more, like the very type “visual experience”, or more specific types like “experience I have in the evening” (Díaz-León 2017, p. 9). Depending on several factors (e.g. relevant color contrast), we might conceptualize a given experience as “blue experience” and group it together with other blue experiences, or we might conceptualize the very same experience as “dark blue experience” instead, opting for a more fine-grained classification that would distinguish it from experiences conceptualized as “light blue experience”. This is not the same, and it does not entail, that by classifying an experience as “blue experience” we are constituting it as such, and that we would constitute a different experience out of the same stimulus if we instead classify it as “dark blue experience”. That is, variation in how we conceptualize visual experiences does not affect the nature of those experiences. The same applies to our conceptualization of sexual desires and experiences of sexual attraction. We can group our sexual attractions in different ways, depending on factors like the contrast class we find significant (e.g. attraction towards this particular woman rather than that particular man, to women rather than men, to this type of woman rather than some other type), and this does not have to affect the nature of those states.

In Wilkerson’s reasoning we find the assumption that flexibility in the conceptualization of something, diachronic or synchronic, is an indication of its metaphysically weak character. I agree with Díaz-León that flexibility in how we conceptualize something, in any of those senses, does not entail a lack of completion in its constitution. No conclusion about the metaphysical nature of something can be directly drawn from the dynamics of its conceptualization. We can draw a parallel here between the conceptualizing act and the thing conceptualized, on the one hand, and an explanation and the explanandum, on the other. How we explain something (e.g. what type of explanation -causal, teleological, formal, at what level, answering what question) is certainly related to the constitution of the explanandum, but it does not straightforwardly reflect the latter. An adequate explanation in a particular context responds to several factors, e.g. the goals of the inquiry, and these are in principle independent of the constitution of the thing that is being explained. This normative principle is also descriptively correct, e.g. people seem to value explanations that match their goals at the moment (Vasilyeva et al. 2017); the nature of the explanandum does not determine one specific sort of explanation. Similarly, we should not expect that sexual orientation-relevant states determine one single way of categorizing them. Our acts of conceptualization respond to several factors, not only the constitution of that which we conceptualize.

Thus according to Díaz-León, Wilkerson does not provide a conclusive argument that our sexual desires and feelings are partially constituted by our conceptualization and classificatory practices. Moreover, Díaz-León sympathizes with the idea that “in order to make sense of the way we conceptualize and interpret our feelings and desires, we need to postulate a layer of ‘raw’ feelings that are fully determined prior to interpretation” (16). That is, on the one hand Wilkerson’s attempted argument against the existence of a metaphysically independent layer of desires and feelings fails, and on the other, it seems reasonable to assume that such a raw layer is there in order to understand the dynamics of categorization and interpretation of them (e.g. the fact that we sometimes change our interpretation over time, that we sometimes move from an incorrect to a correct interpretation). There are two points I want to make with regards to the question of whether or not sexual desires and feelings relevant for sexual orientation are fully and independently constituted prior to our conceptualizing them. I present each at a time.

2.2 A Raw Layer of What?

First, in relation to the existence of a raw layer, I argue that even if we concede that there must be such a layer of feelings that is fully determined prior to interpretation, this layer might still need interpretation in order to arrive to what we usually understand as sexual orientation. That is, perhaps this raw layer consists of very basic affects that need interpretation in order to become part of someone’s sexual orientation. In order to argue for this possibility, I draw a parallel with the conceptual act theory of emotions proposed by Lisa F. Barrett (Barrett 2006; Barrett et al. 2015). According to this theory, physical sensations become emotional episodes only when they are conceptualized and therefore take on a certain meaning in a particular situation. “[W]e experience an instance of emotion in ourselves, or see it in others, when we conceptualize an ongoing, basic affective state via the process of categorization.” (Barrett 2006, p. 21). That is, emotions are not biological primitives, but are constructed via a categorization process. Barrett (2006) identifies two building blocks of emotions: core affect and the categorization act. Core affect can be defined as “a neurophysiological barometer of the individual’s relation to an environment at a given point in time” (Barrett 2006, p. 31). This state, which responds to very basic clues (e.g. whether a stimulus is harmful or not), can be characterized with two parameters: valence (pleasure or displeasure) and arousal (activation or deactivation). That is, it feels as good or bad, and as more or less activated. By means of conceptualizing the core affect (e.g. as sadness, fear, anger, happiness) the individual gets to be in an emotional state. Importantly, the conceptualization process is tailored to the specific situation that the individual is in, and it is constrained by the conceptual repertoire available to the individual.

The conceptual act theory of emotion stays in sharp contrast to theories that postulate a certain (small) number of emotions as natural kinds, which everyone will experience independently of the conceptual knowledge they have relative to emotions. According to this natural kinds story, individuals might differ in how accurately they report their emotional states, but they all experience the same emotions. In the conceptual act theory, however, not everyone experiences the same emotions, for this depends on the conceptual knowledge individuals bring to bear when categorizing core affect, which responds to both individual and cultural differences.

If we apply the conceptual act theory of emotion to sexual orientation, we could postulate a core affect module consisting of some sort of neurophysiological state of the individual in relation to sexual-affective affects. This state would also be characterized by valence and arousal, but this time will also have sexual responses as characteristics. Stimuli (e.g. individual persons, body parts, body movements) would cause changes in this state with different degrees of arousal. This conceptual act theory of sexual orientation nicely accounts for variability in granularity and richness of sexual-affective experiences. One possible alternative to this would claim, parallel to the natural kinds theory of emotions Barrett criticizes, that there is a small set of sexual orientations that are given biologically at some primitive level (as dispositions, perhaps?). [5] According to the conceptual act theory, however, the possibility of a rich conceptual repertoire brings with it the possibility of a wide spectrum of sexual orientations. This reading of sexual orientation as a conceptualization act over basic affects that is situated, tailored to the context, also accounts for cultural differences in how sexual orientations are lived and classified. [6]

Three important questions arise here. First, to what extent the conceptualization process that is necessary to arrive to a sexual orientation is or remains a conscious process. It is reasonable to think that as it happens with the conceptualization of affects that become emotions, this process can be automatized and performed unconsciously. [7] Second, what are the relevant kinds of concepts needed? One possibility is to say that you need to have concepts of sexual orientations themselves. A less demanding option is that what you need to have are concepts for sex and/or gender, [8] so you can conceptualize your affects as directed towards particular categories of people (e.g. a sexual desire for people categorized as women). [9]

Finally, there is the question about how desires and concepts are connected. One option here is to claim that having sexual desire and the concepts for sex and/or gender is sufficient, and nothing else is needed, assuming that the question of how the former connects with the latter is uninteresting. Another option is to say that while these might be necessary conditions, they are not sufficient. The concepts of sex and/or gender need to be connected to sexual desires and affects in the appropriate way, otherwise the sexual desire might be addressed to something irrelevant for sexual orientation. In order for sexual desires to be relevant for sexual orientation, the former need to be about the sex and/or gender of someone, at least if we want to keep our analysis of sexual orientation closely related to the way we ordinarily understand it (as something related to the sex and/or gender of people). [10] I hypothesize that the appropriate connection between sexual desire, or what I have been calling affects, and the aforementioned concepts involves a second order process about the desire, which we might well call interpretation. In this process, the desire is represented as desire for a person qua someone gendered in a particular way (e.g. woman, man, genderqueer, agender). According to this reading, requiring the concepts of sex and/or gender entails the presence of interpretation.

Therefore, in order for a subject to have sexual desire for someone that counts as relevant for that subject’s sexual orientation (that is, sexual desire for people in virtue of being gendered in a particular way), the subject needs to, on the one hand, possess the concepts of sex and/or gender, and on the other, interpret a particular collection of affects as related to the sex and/or gender of the person(s). [11] This is importantly different from Wilkerson’s idea that desires themselves are constituted by interpretation. What I am defending here is that while desires/affects are themselves constituted independently of any interpretation, the selection of some of those desires/affects and their conceptualization as related to the sex and/or gender of someone is an interpretative process necessary for sexual orientation. Like the properties of a country when we make a map of it, affects are there prior to our interpretation. But only when we select some of the properties of a country (e.g. its province boundaries and city limits, its roads, its geological properties, or its bike paths) and make them salient for our mapping-purposes, do those properties become part of a political, geological, road, or bike map. Similarly, it is only when we make a selection and interpret some of the affects as directed towards the sex and/or gender of someone, that those affects become relevant for sexual orientation. Interpretation does not have (neither a causal nor) a constitutive effect on the affects, but rather on what we call sexual orientation. The analogy with mapmaking reveals that selection is a critical part in the interpretation process. As it happens in Jorge Luis Borges’ tale “Del rigor en la ciencia” (Borges 1960), a 1:1 scale map that contained a representation of every single thing in reality and coincided in every detail with it would be of no use, it would not even count as a map. Similarly, what counts as sexual orientation according to our ordinary use of the term is neither any random selection of sexual affects, nor all the sexual affects we feel, but a selection of those that are related to the sex/gender of others. [12]

The view that I am proposing here is also different from Díaz-León’s. In her view, desires and concepts are connected in a different way. She says:

once someone possesses these concepts of sex/gender, and has sexual desires that are appropriately connected to those concepts, this is sufficient for the content of those desires to amount to sexual attraction for men and/or women, qua men and/or women. It is true that one can have these desires only when one possesses those concepts, but this does not entail that one can have those desires only when one interprets those very desires as being a desire-for-men or desire-for-women. That is to say, according to the account of sexual orientation that we are assuming here, what determines someone’s sexual orientation is a matter of her sexual desires for men and/or women, where the subject needs to conceptualize humans in terms of their sex and/or gender. But this does not require the subject to interpret her sexual desires themselves as sexual desire for men and/or women. (Díaz-León 2017, p. 18–19)

In her view, what I need in order to have a sexual orientation is a desire for S, and independently, conceptualizing S as a woman/man/genderqueer. Therefore, I happen to be attracted to S qua a woman/man/genderqueer as a consequence of these two separate mental moves, so to speak. What I propose is that in order for desires to be part of (our ordinary notion of) sexual orientation, the desires themselves need to be conceptualized in certain ways, having to do with the sex and/or gender of people.

To sum up, even if we agree that sexual orientation involves a raw layer of desires that is metaphysically autonomous and entirely determined prior to any conceptualization exercise, this layer might not coincide with what we identify as sexual orientation, but rather with one of its composing elements (i.e. affects). Therefore, even if Díaz-León is right in her vindication of this metaphysically independent layer, interpretation has not been ruled out yet.

2.3 Imposing Categorical Boundaries

My second point engages the choice of visual experiences as an illustration for Díaz-León’s defense of the metaphysical independence of experiences (relative to our acts of classifying them). With the appeal to visual experiences, Díaz-León makes the general point that the possibility of different ways of categorizing an experience does not necessarily entail that the experience is not fully formed prior to the categorization act. But she also makes a more specific point about one particular dimension of classification, i.e. granularity. According to this more specific idea, the possibility of having different options along the granularity dimension in our categorization (i.e. fine-grained “dark blue experience” vs. coarse-grained “visual experience”), does not entail that the experience itself is affected in its constitution (that is, labeling it as “dark blue experience” does not make it a more fine-grained experience than labeling it “visual experience”). And this is because granularity in the conceptualization of experiences is independent of the granularity of the experiences themselves. This specific point about conceptualizing experiences in more or less fine-grained ways seems uncontroversial. It is the general point about categorization more generally that is our main interest here and the one I find controversial in Díaz-León’s reasoning.

If all our categorization practices were along the granularity dimension, then Díaz-León’s general conclusion that experiences are not affected by how we categorize them seems to follow. There are, however, other dimensions of classification that do not illustrate this general point in the same clear-cut way, and that might jeopardize the conclusion Díaz-León wants to arrive to. In particular, there is a dimension of classification that, in my opinion, undermines the conclusion that the granularity criterion so neatly conveys. I am referring to how sharp we take the boundaries of categorical types to be. Let us call it the category boundaries dimension. Now let us take an example and see how conceptualization might work along this other dimension. Let us consider a stimulus consisting of a series of consonant-vowel syllables, artificially generated, that gradually moves along a continuum with 14 different values. The two endpoints (i.e. values 1 and 14) are /be/ and /ge/. The interesting question is how we would categorize the in-between sounds (i.e. values 2 to 13) as the series moves from /be/ to /ge/. On the category boundaries dimension, there are two main ways we could categorize that set of stimuli. First, we could categorize it as a continuum; this means that the stimuli values in between the two endpoints (i.e. values 2 to 13) will be categorized as slightly different from each other, drawing a categorization slope; second, we could instead group stimuli into well-defined category groups drawing a very sharp boundary between them (e.g. we could group values 2 to 4 together into one category, and 4–9 into a different category, and so on). Empirical research suggests that people opt for the second categorization style when presented with the continuum syllable series described above (Liberman et al. 1957). Instead of perceiving subtle differences as the stimulus values gradually change from one end point to the other (from /be/ to /ge/), people tend to perceive instances of different discrete categories. That is, perception changes dramatically from hearing the sound as /be/ to hearing it as /de/ (around stimulus value 4), and then again around values 9 and 10 people rapidly go from hearing /de/ to hearing /ge/. This case is an example of what in the psychological literature is called categorical perception.

One way to read this result is that our categorization act (i.e. the fact that we draw sharp categorical boundaries) does affect the stimuli in a profound way. We impose category boundaries on the continuous series, and that makes us hear a paradigmatic /be/ up to a certain point, right until the moment the sound dramatically shifts to become a paradigmatic instance of a different category, and so on, instead of hearing in-between, ambiguous syllables for the in-between stimulus values. If we opted for the first, continuum way of categorizing the syllable series, we could confidently say that our auditory experience is entirely determined by the sound, for ambiguous stimulus values would be categorized as ambiguous. But imposing sharp category boundaries makes our categorization act part of what defines and constitutes the experience. While paradigmatic /be/ and /ge/ are not transformed by our categorization act (i.e. the two end points), in-between cases are clearly affected. Now, what if we categorize sexual desires and feelings similarly to how we categorize speech stimuli? The category boundaries dimension of classification does not leave things as clear-cut as the granularity dimension Díaz-León chooses to illustrate her point, or so I argue. It seems possible now that how we group stimuli matters a lot for how our experiences of them are constituted.

It is an open question which of the above cases is a better illustration of how we categorize our sexual desires, either the visual experience example or the /be/ - /ge/ case. If the latter, then it is not clear we can conclude that our categorization practices do not affect the constitution of our sexual desires. If we were to categorize sexual orientation-relevant affects in a categorical way, similar to how we categorize speech stimuli, we would group slightly different affects into a few discrete categories, and would feel only those paradigmatic categorical affects. If so, the ambiguous in-between affects for which we perhaps do not have concepts for, would be assimilated into one of the categories we already have created.

Independently of this, the main point Díaz-León defends still holds, and that is that the possibility of different ways of conceptualizing an experience does not by itself entail that the experience is undetermined prior to categorizing it, and therefore that a conceptualization act is necessary to finish up its constitution. Interpretation does not entail constitution. That being said, I emphasize that it could be the case that the specific way in which we conceptualize the feelings and affects relevant to sexual orientation could be one that imposes sharp category boundaries and, in such a case, ambiguous feelings (although perhaps not unambiguous ones) would be transformed by how we categorize them, turning them into one or another type of experience depending on where our categorization act draws the boundary.

3 Interpretation vs. Choice

My second comment engages Wilkerson’s second premise, which says that interpretation requires choice. The main point I make is that the existence of interpretation does not imply choice; only ideal interpretation requires choice. According to Wilkerson, “if interpretation is forced, then it loses its meaning as interpretation” (Wilkerson 2009, p. 109). That is, in a process of interpretation there must be alternatives in order to be a proper interpretation, even if the number of them is small. Wilkerson illustrates this idea with the example of a literature scholar who interprets the work of some author by choosing amongst possible alternative interpretations. There is, however, another possible reading of what interpretation consists of that sees interpretation as figuring out the meaning or dynamics of something, as making sense of it. This other reading might or might not involve choosing amongst different alternatives. Let me illustrate this with an example.

Imagine you need to activate a machine you have never encountered before. In order to do that, you need to make sense of it, understand what at least some of its parts do, figuring out how it works. In this process, you might or might not have access to different alternative ways of how the machine works. If you are not given an instructions manual, you would probably consider different alternative hypothesis of how the machine works, and test them, until you arrive at some successful interpretation. If you are given the instructions manual, however, you will probably follow it without considering different alternatives. In both cases you are interpreting the machine, in the sense of making sense of it (I assume that in following the instructions manual you understand something about how the machine works, that is, you do not just merely do what the manual says without understanding anything). This reading of interpretation as figuring out or making sense does not assume, in contrast to Wilkerson’s reading, that choice is necessary in interpretation. And this reading fits nicely with the above considerations about a conceptual act theory of sexual orientation: in applying available conceptual resources, we figure out what our sexual-affective affects are, we make sense of them, without necessarily choosing amongst several alternative interpretations.

Wilkerson worries that if choice is not part of the interpretation process, then we are risking loosing freedom. I think, however, that we do not need to assume choice in interpretation in order to guarantee freedom of some sort. First, we do not have to eliminate choice altogether from the process of interpreting sexual orientation-relevant affects. Rather, we can limit choice to ideal cases, in which the interpretation process includes consideration of different alternatives. Still, there will be cases in which interpretation happens without choice. Second, even if we say that interpretation involved in sexual orientation often lacks choice, this does not mean we are not free in some important sense. The main point here is to avoid assuming that freedom requires the existence of alternate possibilities. Harry Frankfurt’s famous examples were designed to show this (Frankfurt 1969), and more specifically, that the freedom-relevant condition to hold an agent morally responsible does not require that they could have done otherwise. In line with this, we could appeal to John M. Fisher’s distinction between regulative and guidance control over one’s actions (Fisher 1994) and apply it to interpretations of sexual orientation-relevant affects. While regulative control does require control over which possibility out of a number of alternatives becomes actual, guidance control does not; and the latter is sufficient for freedom. Guidance control requires something different; on the one hand, that the agent’s actions issue from their own mechanism, which needs to be responsive to reasons, and on the other, that the agent takes responsibility for that mechanism, that it is their own. Fisher’s distinction, like Frankfurt’s examples, concern freedom as long as it is a requirement for moral responsibility. Even though we are not interested in moral responsibility here, we can still benefit from their work to illustrate a possible avenue to keep freedom in the picture of sexual orientation even if we eliminate the requirement of having access to alternative possibilities. Thus we can apply the distinction between regulative control and guidance control to the interpretation of sexual affects, instead of actions more generally. If we agree with the idea that only guidance control is required for freedom, we can say that when interpreting sexual orientation-relevant affects, we can have freedom even if we do not have access and control over alternative interpretations. We can have guidance control over our interpretation instead. This picture allows us to distinguish between cases of interpretation of sexual affects that reflect genuine freedom, and those that do not. If in the interpretation of their affects, someone does not own the process of interpretation, but is instead clearly forced (by internal and/or external factors) to one specific interpretation, then we can say that this person has no freedom in the process of creating her sexual orientation. But if the person interprets their affects in a way that meets the stipulated definition of guidance control, then we can say that freedom was part of the picture, even if the person did not have access to or actually considered alternative interpretations. A detailed discussion of the conditions for owning a process of interpretation lies beyond the scope of this paper. As a first step, we can sketch two ideas. First, in a free (although choiceless) conceptualization we are in principle capable of changing our conceptualization in case of new affects that are incongruent with the current one, as opposed to stretching the current conceptualization to fit whatever new affects we might have. This condition corresponds to Fisher’s idea that the free agent’s mechanism issuing their actions needs to be responsive to reasons. Second, and corresponding to Fisher’s second condition, we are capable of reflecting on the matter, and take responsibility for our interpretation. [13]

One indirect source of support for the idea that we have something like Fisher’s guidance control over the interpretation of our sexual orientation-relevant affects is the fact that if we were to require instead regulative control in order to attribute freedom and choice, that is, if in order to have choice and freedom in the picture we were required to have access and control over alternative interpretations, then we have to say that most heterosexual people are not free and do not choose. This would make Wilkerson’s proposal only valid for people with a non-heterosexual orientation. I speculate that most people do not consider alternative interpretations of their affects before establishing their sexual orientation as heterosexual. This happens because heterosexuality is the default interpretation. Most, if not all of the schemas, norms and conceptual resources available that can be used in interpreting affects facilitate a heterosexual interpretation. Heterosexual interpretation is the default, and default options have a lot of grip in human decision-making. [14] If I am right in my speculation and the explanation I just gave for it, requiring access to, and/or actual consideration of, alternatives would dramatically lower the number of people who could be said to be free in the process of negotiating their sexual orientation. Also, it would mean that only those who conclude a non-heterosexual orientation were free and had a choice in Wilkerson’s sense. A proposal about whether or not choice is involved in sexual orientation should be, however, applicable to any sexual orientation, unless we make the additional claim that there are qualitative differences among, e.g. heterosexual and non-heterosexual orientations, so that the latter but not the former involves choice. Thus, Wilkerson’s insistence on interpretation requiring choice could jeopardize his own conclusion that there is an element of choice (and with it, a hint of freedom) in sexual orientation. If most people for most of the time go with the default interpretation of affects and this is read as a process lacking in freedom and choice, then Wilkerson would have to choose between abandoning his requirement to keep choice and freedom in the picture, or accepting that his conclusion that sexual orientation involves choice (and freedom) does not correctly describe how things actually work.

To sum up, it is possible to hold that interpretation is an important part of sexual orientation without necessarily holding that choosing among alternative interpretations is required. If, moreover, we were to concede that interpretation is always involved in sexual orientation (e.g. because we endorse something like the conceptual act theory of sexual orientation), we can still say that interpretation does not always imply choice (in the sense of having access to alternatives). Only interpretation in ideal cases requires so. Now, we can ask what a case of ideal interpretation looks like. As mentioned above, in such a case the agent has access to alternatives. One way of spelling out what this access involves is to say that in ideal cases, the interpretation of sexual affects is connected to other possible conceptualizations. This is similar to how Robert Nozick (1981) spells out what understanding is, as opposed to explanation. While explanation “locates something in actuality, showing its actual connections with other things”, understanding locates it in a network of possibility, showing the connections it would have to other nonactual things or processes.” (Nozick 1981, p. 12). If we translate this insight into our discussion, what we have is that an ideal interpretation connects the actual conceptualization of sexual affects to possible alternative ways we could have conceptualized them.

One extreme way of reading ideal interpretation is to say that this access to alternatives necessitates the absence of a default option. That is to say that the presence of a default option trumps access to this network of possibility. A less extreme reading sees the existence of a default option as a significant moderator of what alternatives we consider. This second reading aligns with recent research in cognitive psychology about what alternatives we consider when confronted with different situations (see e.g. Hitchcock and Knobe 2009; Philips et al. 2015). Not only statistical but also moral and social norms affect which alternatives we consider to be relevant in particular contexts. Applied to the conceptualization of sexual affects, this would mean that the existence of a norm (in this case, heterosexual interpretation) affects which alternatives we will consider.

4 Conclusion

Wilkerson’s argument for the claim that sexual orientation requires interpretation and therefore choice, takes issue with how sexual desires themselves are constituted. He claims that sexual desires require interpretation to be fully constituted, and for his defense he relies on the idea that sexual desires (and experiences) can be conceptualized and classified in different ways. But as Díaz-León points out, (synchronic and diachronic) flexibility in the conceptualization and classification of desires and experiences does not entail that those desires and experiences are not fully constituted prior to their classification. Here I have explored a different argument for the presence of interpretation, one that does not conclude that there is choice in Wilkerson’s sense. First, something like the conceptual act theory of emotions could be true of sexual orientation. According to it, there are core affects that are metaphysically independent of our conceptualization practices, but there is still a step in between them and sexual orientation. This step involves a second-order process about those affects that most likely entails concepts for sex and/or gender, so affects are about persons in virtue of being sexed/gendered in a particular way. This second order process looks like interpretation. Second, I pointed out the possibility that in the interpretation of affects relevant for sexual orientation, we impose category boundaries in a way that not only impact those affects causally, but also constitutively. The presence of interpretation, however, does not necessarily take us to the conclusion that choice is involved in sexual orientation. I argued that interpretation does not require considering or even having access to alternatives. If so, we need to explore a different avenue to say that choice, as commonly understood in the relevant debates, is involved in sexual orientation.


Previous versions of this paper were presented at two “Minorities in Philosophy” workshops (University of California, Los Angeles and Stanford University chapters), and at a Philosophy Club meeting at California State University, Sacramento. I wish to thank the organizers of those events (Olufemi Taiwo, Adam Simon, and Selena Trinies, respectively), and audiences for their questions. I would also like to thank two anonymous reviewers of this journal. Finally, special thanks go to Esa Díaz-León, for insightful and fun conversations on sexual orientation and many other topics during the last 21 years.


Barrett, Lisa F. (2006): “Solving the Emotion Paradox: Categorization and the Experience of Emotion”. In: Personality and Social Psychology Review 10. No. 1, p. 20–46. Search in Google Scholar

Barrett, Lisa F., Christine D. Wilson-Mendenhall and Lawrence W. Barsalou (2015): The Conceptual Act Theory: A Road Map. In: Barrett L. F., and J. A. Russell (Eds.): The psychological construction of emotion. New York: Guilford, p. 83–110. Search in Google Scholar

Bettcher, Talia Mae (2014): “When Selves Have Sex: What the Phenomenology of Trans Sexuality Can Teach Us About Sexual Orientation”, In: Journal of Homosexuality 61. No. 5, p. 605–620. Search in Google Scholar

Borges, Jorge Luis (1960): El Hacedor. Madrid: Alianza. Search in Google Scholar

Dembroff, Robin (2016): “What is Sexual Orientation?”. In: Philosophers’ Imprint 16. No. 3, p. 1–27. Search in Google Scholar

Díaz-León, Esa (2017): “Sexual Orientation as Interpretation? Sexual Desires, Concepts, and Choice”; In: Journal of Social Ontology. 3. No. 2, p. 231–248. Search in Google Scholar

Díaz-León, Esa (forthcoming): “Sexual Orientations: The Desire View”. In: Maitra, Keya, and McWeeny, Jennifer (Eds.): Feminist Philosophy of Mind. Oxford: Oxford University Press. Search in Google Scholar

Fischer, John Martin (1994): The Metaphysics of Free Will. Oxford: Blackwell. Search in Google Scholar

Frankfurt, Harry (1969): “Alternate Possibilities and Moral Responsibility”. In:Journal of Philosophy 66, p. 829–39. Search in Google Scholar

Gigerenzer, Gerd, Peter Todd, and The ABC Research Group (1999): Simple heuristics that make us smart. New York: Oxford University Press. Search in Google Scholar

Gigerenzer, Gerd (2008): “Why Heuristics Work”. In: Perspectives in Psychological Science 3. No. 1, p. 20–29. Search in Google Scholar

Hitchcock, Christopher and Joshua Knobe (2009): “Cause and Norm”. In: The Journal of Philosophy 106. No.11, p. 587–612. Search in Google Scholar

Liberman, A. M., K. S. Harris, H. S. Hoffman and B. C. Griffith (1957): “The discrimination of speech sounds within and across phoneme boundaries”, In: Journal of Experimental Psychology 54, p. 358–368. Search in Google Scholar

Nozick, Robert (1981): Philosophical Explanations. Cambridge, MA: Harvard University Press. Search in Google Scholar

Philips, Jonathan, Jamie Luguri and Joshua Knobe, (2015): “Unifying Morality’s Influence on Non-Moral Judgements: The Relevance of Alternative Possibilities”. In: Cognition 145, p. 30–42. Search in Google Scholar

Stein, Edward (1999): The Mismeasure of Desire: The Science, Theory, and Ethics of Sexual Orientation. Oxford: Oxford University Press. Search in Google Scholar

Stein, Edward (2011): “Sexual Orientations, Rights, and the Body: Immutability, Essentialism, and Nativism”. In: Social Research 78. No. 2, p. 633–658. Search in Google Scholar

Vasilyeva, Nadya Y., Daniel Wilkenfeld and Tania Lombrozo (2017): “Contextual Utility Affects the Perceived Quality of Explanations”. In: Psychonomic Bulletin and Review, p. 1–15. Doi: 10.3758/s13423-017-1275-y. Search in Google Scholar

Vernallis, Kayley (2013): “Bisexual Marriage”. In: Power, N. Halwani R., and Soble, A. (Eds.): The Philosophy of Sex: Contemporary Readings, 6th edition. Lanham, MD: Rowman and Littlefield, p. 215–32. Search in Google Scholar

Wilkerson, William S. (2009): “Is It a Choice? Sexual Orientation as Interpretation”. In: Journal of Social Philosophy 40. No. 1, p. 97–116. Search in Google Scholar

Wilkerson, William S. (2013): “What is ‘Sexual Orientation’?” In: Power, N. Halwani R., and Soble, A. (Eds.): The Philosophy of Sex: Contemporary Readings, 6th edition. Lanham, MD: Rowman and Littlefield, pp. 195–213. Search in Google Scholar

Published Online: 2017-7-4

©2017, Saray Ayala, published by De Gruyter.

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.