Prostitution and the Argument from Sexual Autonomy

(Previous Series of Posts)

I have recently been looking into the ethics of prostitution. This started last week with my analysis and exposition of Ole Martin Moen’s article “Is Prostitution Harmful?” which appeared in the Journal of Medical Ethics. Those of you who read the resultant series of posts will know that I had promised to look at some criticisms of Moen’s piece which appeared in the same journal. Having read those criticisms, and having reflected on their contents, I have now decided against this course of action. I don’t think the criticisms are sufficiently substantive to merit any extended discussion.

So, instead, I’m going to look at a more substantive argument in favour of the prohibition of prostitution. This one coming from Scott Anderson’s article “Prostitution and Sexual Autonomy” which appeared in the journal Ethics back in 2002. (For what it’s worth: Anderson was one of the people who responded to Moen’s article and in his response he points to the argument made in this earlier piece).

The main focus of Anderson’s article is what might called “classical heterosexual prostitution”, i.e. the case in which a man pays a woman for sex. Although prostitution can take other forms, I don’t see anything too problematic about focusing on this case since it tends to be the norm. Anyway, Anderson’s article starts by contrasting the “liberal” and “radical” views on the regulation of this type of prostitution. Roughly speaking, these views can be characterised in the following manner:

Radicalist View: Prostitution is an inherently degrading and dehumanising institution, one which perpetuates social inequality between men and women. As such, it should be prohibited and restricted. This view is associated with certain members of the radical feminist movement, e.g. Andrea Dworkin.

Liberal View: There is nothing inherently degrading and dehumanising about the commodification of sex. What’s more, many of the harms that are currently associated with prostitution could be eliminated if it were decriminalised/normalised (i.e. treated the same as any other form of human labour). This view is associated with the likes of Martha Nussbaum.

Anderson’s stated goal is to provide an argument in favour of the prohibition of prostitution that addresses the concerns put forward by the radicals, while also taking seriously the suggestions of the liberals. The result is the argument from sexual autonomy, which holds that prostitution should be prohibited because its normalisation would undermine sexual autonomy.

In the remainder of this post, we will look at the supposed good of sexual autonomy; reconstruct the argument from sexual autonomy; and finally consider some problems with that argument.


1. Sexual Autonomy and Sexual Intimacy
I want to start with an interpretive point, one which may influence the overall strength of Anderson’s argument. Ostensibly, that argument is spelled out in terms of the desirability of preserving sexual autonomy. As a result, one might think it is solely about the good of sexual autonomy. There is, however, some room for confusion about the ultimate value judgments that motivate the argument. At least, I find myself somewhat confused about it even if others aren’t.

It seems to me that Anderson’s argument appeals, in the first instance, to the inherent value of sexual intimacy, and only secondarily to the value of sexual autonomy (see comments on pg. 759 for example). Since these two concepts are central to what I am about to say it will be worth defining them:

Sexual Intimacy: A type of human good achieved through intimate (maybe affectionate) sexual relations.

Sexual Autonomy: The right/capacity of each individual to decide the who, when, where and how of their sex lives.

The definition of sexual intimacy does not presuppose what we previously referred to as the “significance view” of sex. In other words, it does not assume that sex must be romantically significant in order to be valuable. That’s why the “maybe affectionate” bit is in brackets. The definition of sexual autonomy is pretty straightforward, and maps onto definitions of autonomy in other domains.

Reading between the lines somewhat, I think Anderson believes that the goods of sexual intimacy and sexual autonomy are linked in an important way. Roughly, he believes that one can only really gain access to the good of sexual intimacy if one has a robust right to sexual autonomy. Thus, protecting the latter serves as a pathway to the former. This is pretty standard liberal thinking: give people the freedom and capacity to choose and they will coordinate on the types of sexual interactions that are most valuable to themselves.

The problem, of course, is that this understanding of the connection between intimacy and autonomy might be wrong. It might be that a more paternalistic approach to sexual choice ensures greater access to the good of sexual intimacy. (I’m not suggesting that this is likely, merely that it is possible). Thus, if one interprets Anderson’s central argument as being about the link between autonomy and intimacy, it may seem weaker than it needs to be.

The solution may then be to interpret the argument as being about the intrinsic good of sexual autonomy, i.e. about the notion that sexual autonomy is good, in and of itself, irrespective of whether it leads to greater sexual intimacy. That is how I am going to interpret the argument. So in what follows I will not focus on the consequences of achieving sexual autonomy.


2. The Argument from Sexual Autonomy
To illustrate the intrinsic value of sexual autonomy, and the problems that might be associated with its erosion, it is worth considering how differently we approach sex as compared with other social gifts or exchanges. Anderson uses the example of a man and woman going on an expensive dinner date. Within some communities and cultures (increasingly historical, I suspect), this process follows certain social norms and expectations. Typically, the man pays for the dinner. In return for this, he may expect the woman to accede to (at least some of) his sexual overtures later in the evening…

…however — and this is crucial — the cultural norm/expectation does not extend so far as to allow the man to demand that the woman accede to his overtures, or to justify him in feeling aggrieved if she does not, or to allow others to look down upon her if she does not. This is rather different from other informal and formal social exchanges. For example, if I took a friend out for dinner, paid for it, and later asked him to return the favour by helping me with some aspect of my work, I might be more justified in feeling aggrieved if he refuses to accede to my request. Other informal exchanges incur similar obligations of reciprocity (e.g. buying rounds of drinks among friends). And, of course, formal exchanges incur even greater obligations. For example, the contractual exchange of goods and services creates legally enforceable rights and obligations.

The question, then, is why do we treat sex differently? Why are we so unwilling to impose obligations and demands in relation to sexual performance? The answer, Anderson suggests, is that we appreciate the good of sexual autonomy. We don’t impose obligations because we don’t want to interfere with or coerce people in their sexual choices. We recognise that this is something that should be entirely up to them.

The danger, according to Anderson, is that if we start to treat sex like other forms of commodified human labour, this right to sexual autonomy will start to dissipate. After all, other forms of labour reduce autonomy in other domains. If I choose to become a highly-paid corporate lawyer, I know that I will be subject to many demands on my time that I’m not entitled to refuse. Likewise, if I choose to become a prostitute, my sexual autonomy will be compromised. Furthermore, given the economic necessity of paid employment, this could have a broader impact on sexual autonomy, outside the class of “willing” prostitutes.

We can set out the argument in the following manner:

• (1) Sexual autonomy is an important human good.

• (2) The good of sexual autonomy is only available to us if there are certain restrictions placed on acceptable sexual demands and obligations (support: dinner date example)

• (3) If prostitution is legalised/normalised, those restrictions would/could be eroded.

• (4) Therefore, if prostitution is legalised/normalised, we will lose an important human good (viz. sexual autonomy).

The key to this argument is, of course, premise (3). In its current form it faces two problems. First, we need some more detailed reason for thinking that legalisation/normalisation undermines the restrictions needed for sexual autonomy. Second, we need to address the “would/could” equivocation. Using the term “would” suggests that this is something that is likely to occur; using the term “could” suggests that it is merely possible. Which is it and does this make a difference?


3. Defending the Argument from Sexual Autonomy
To defend premise (3), Anderson asks that we take seriously the liberal claim that sex is unexceptionable; that it should be treated like any other type of commodified human labour. Anderson thinks that three general infringements on sexual autonomy emerge if we take this claim seriously, each of which gives rise to a set of specific infringements. They are:

1. We will be expected to engage in sexual activity when certain economic incentives are present: Anderson gives three examples of this: (a) employees may find that the performance of sexual services is part of their contract or employment or that it is expected in order to achieve promotion; (b) those on welfare may be expected to engage in sex work if it is available; and (c) contracts for sexual labour may become enforceable in the same way as other contracts.

2. We will have less control over the who, what, where, when and how of our sexual lives: Again, three examples are given here: (a) large and aggressive corporations will start marketing and selling sexual services, and in the process will closely monitor (standardise) the sexual performance of their workers; (b) workers with sexual duties may be required to comply with non-discrimination statutes with respect to their clients and co-workers (i.e. they won’t be able to pick and choose who they have sex with); and (c) sex workers will become subject to government health and safety standards, and risky sexual practices (both on and off the job) may be prohibited.

3. Our sexual attitudes and values will be changed: Two examples are given here: (a) aggressive marketing and advertising of sexual services will alter the marketplace of ideas in relation to permissible sex; and (b) special vocational training and careers advice may target those who seem differentially suited to sex work.

The last category of infringements is a little odd to me, but later in the article Anderson says that the effect on sexual autonomy in these cases is more subtle and indirect. The point is that our sexual values and attitudes will become distorted by the changed legal landscape. In particular, we may find that the marketplace of ideas is saturated by sexual imagery and sex-related advertising. This may mean that we are constantly forced to confront the sexual aspects of our identities. And this would be bad because there are at least some occasions on which we would prefer not to attend to those aspects of our identities. Of course, one could argue that this is already true (to some extent) — that the marketplace of ideas is saturated by sexual imagery — but Anderson distinguishes the current marketplace from the hypothetical one on the grounds that at the moment sex is used as an instrument to sell other things, not as the thing that is itself being sold.

Now, I have some problems with Anderson’s examples. It seems to me that some of the things he points to might actually be advantages of legalisation; and, of course, the examples encourage us to attend to the negative more than the positive, which is always a problem with thought experiments like these. Nevertheless, I can agree that some of the examples would involve an unwelcome erosion of sexual autonomy. What’s really at issue for me is whether any of things he points to are likely, or necessary consequences of legalisation. He explicitly denies that they are:

My point in offering these scenarios is not to suggest that any of them is likely to come to pass in the wake of moves to normalize prostitution. Rather, these scenarios are simply exhibits of what it would mean to take the bold claim seriously—to treat sex as just another way to use the body to make a living… 

(Anderson, “Prostitution and Sexual Autonomy” p. 763)

I find this odd. As I understand it, the argument from sexual autonomy functions as a type of slippery slope argument: if we do one thing, unwelcome consequences follow. But if those consequences are not particularly likely, or, more importantly, if there are obvious ways in which to avoid them, then the argument is significantly weakened. Could we not have special exemptions from sex work (akin to, say, the conscientious objections available to other workers — in fairness, Anderson mentions this possibility)? Could the market for sexual services not be regulated like the market alcohol or cigarettes? In many countries, there are severe and increasing restrictions on the advertisement of these products, so they no longer control the marketplace of ideas (if they ever did). Why couldn’t something similar be done with prostitution.

In summary, I think Anderson’s argument is an interesting one, and I think the good of sexual autonomy is one that we should cherish and preserve. Nevertheless, I’m not convinced that the argument in its current form provides any strong reason for favouring the legal prohibition of sex work. We would need a lot more evidence for thinking that legalisation leads to the unwelcome compromises that Anderson identifies before we could become convinced of that. And surely some such evidence is available? Prostitution is legal in several European countries and I’m not aware of any significant impact on sexual autonomy that has resulted from this, (despite some false reports to the contrary). If anyone knows of more detailed research, please share in the comments section.