Social Scripts and Sexual Agency

2.1. Defining Social Scripts

In at least an inchoate way, the term “social script” resonates with how many people think about the social world: Significant parts of our lives seem like performances of roles that are defined by shared cultural resources. Presumably because of this resonance, some philosophers rely on an intuitive grasp of the notion of a social script, without explicitly defining it. But this lack of explicit definition creates a risk that people have in mind different parts of our social world when they use the term. This risk is augmented by the variety of the theoretical roles that social scripts are meant to play: In addition to the templates for short interactions like restaurant orders, the term “social script” has been applied to narratives that people use when shaping their projects and telling their life stories (Appiah 1994; 2005), as well as to stereotypes of Black people as untrustworthy (Stoljar 2015). Once our attention is drawn to these different applications, it is natural to start to worry whether there is a univocal sense to the term across them. Are stereotypes of Black people really the same thing as the narratives behind the arc of a life? Are either of these the same phenomenon as a template for a restaurant interaction? Given that these differences are yet to be tracked in the philosophical literature, some initial groundwork will be helpful and so let us begin by asking: What is a social script? My hope is that addressing this question will have the dual functions of helping regiment the existing philosophical literature and setting up our later analysis of the constraints on sexual agency.

Since “social script” is not a phrase that features in everyday English, we are in the search for a definition of a theoretical term. For such a term, I follow Sally Haslanger (2012: 367-368) in holding that our theoretical goals determine which definitions are the most helpful for us to adopt in a particular project. My goal is to explain how certain social phenomena can constrain people’s agency in sexual interactions with each other. As an example of the interactions that I am interested in, consider the following testimony of a college student from New York, Rachel:¹

I hate admitting how much sex I’ve had because it was “polite” to just let him finish. You read stories of rape and sexual assault but never about your own manners pressuring you into having sex. Sometimes you just don’t want to have sex after all the buildup but there is no way to get out of it without coming off as rude or disappointing your partner, who is probably a good person, not some creepy dude in a club. (Bennett & Jones 2019: 82)

Why did “all the buildup” lead Rachel to think that good manners required her to have unwanted sex? And why did she think declining sex would disappoint her partner? Perhaps, he will feel some disappointment or frustration simply because he was hoping to have sex. After all, it isn’t odd for someone to feel these emotions when social activities conclude earlier than hoped. However, in addition, I suspect that Rachel thought that her partner expected them to have sex, and that she would be breaking off a social activity before it had concluded if she had declined. But if that’s the case, then we should want to know: Why did she attribute these expectations to him and why did she take herself to be midway through a social activity that will later involve sex? To answer these questions and to explain why she saw unwanted sex as polite, my working hypothesis is that Rachel’s attitudes were influenced by social phenomena that stated the “normal” way for casual sexual encounters to go: Specifically, these encounters proceed until the point of male orgasm, and once the encounter has begun, it is abnormal to break off the encounter.

To elaborate the hypothesis that the social world can determine what constitutes a normal sexual encounter, I will build on a mature social scientific research program that explains a wide variety of sexual behavior and misconduct in terms of social scripts.² This program views scripts as the social templates for interactions between different agents (henceforth “interactions” for brevity).³ However, canonical works in the social science literature are not univocal regarding the details of scripts and do not always contain the tight definitions that are best suited for philosophical analysis. And so, it is welcome that within the philosophy literature, we find a thorough account of these templates in the work of Cristina Bicchieri (2006). Conceiving of schemata as “cognitive structures that represent stored knowledge about people, events, and roles,”⁴ Bicchieri (2006: 93; see also 2017: 131-132) states that the

schemata of interest for understanding how norms affect behavior are event schemata that describe appropriate sequences of events in well-known situations. Examples of such schemata are descriptions of what happens at restaurants, soccer games, theaters, and lectures. Consider a ‘lecture schema,’ which contains roles (student, professor) and sequence rules (the teacher enters the classroom, the students seat and prepare to take notes, the lecture starts, the students take notes and ask questions, the lecture ends, all leave the classroom). Schemata for events such as this are called scripts (Schank and Abelson, 1977). A script for a lecture thus describes a stylized, stereotyped sequence of actions that are appropriate in this context, and it defines actors and roles.

People may have various motivations for following scripts, which can either explicitly feature in conscious reasoning or operate subconsciously. Over time, these motivations can lead to certain behaviors becoming engrained at the level of habit so that they are performed automatically. For example, lecturers and students typically internalize a social script for a lecture and hence follow it on autopilot (Bicchieri 2006: 97-98). As well as guiding people’s behavior, scripts have implications for people’s beliefs: If people view each other as following the script, then they will use the script as the basis of their expectations of each other’s future behavior (96).

To illustrate, consider the restaurant script. The social script involves a representation of a certain situation, namely the table at the restaurant. The script also involves representations of the roles of the server and the diner. Finally, it contains prescriptive norms of requirement, permission, and prohibition, which apply to people as occupants of these roles. These norms designate some ways of carrying out the interaction as “normal” and other ways as “deviant” by specifying specify behavior that must, may, or must not be performed at various stages of the interaction. For example, in the restaurant script the server must ask what the diners want; they may ask for anything off the menu; they must not ask for a completely different meal that doesn’t feature on the menu. I understand the normativity in this script—and indeed of all social scripts—to be conventional. Likewise, I understand “deviant” in a non-pejorative manner, such that behaving deviantly is simply behaving in a way that is conventionally abnormal. But as we will discuss in detail in §3, when certain contingent conditions are met, the conventional normality of an action can take on a particular significance for an agent. For example, if a diner would feel embarrassed or ashamed by acting in a way that is deviant according to the culture that they are in, then they will be motivated to conform with the social scripts that govern their restaurant interaction.

Generalizing from this example, we can follow Bicchieri’s (2006) account to construct a definition that recognizes situations, roles, and prescriptive norms. To capture these, I propose the following definition of a social script as a blueprint for a specific interaction:⁵

Social Script Definition: A social script for an interaction between different agents is a cluster of a representation⁶ of a situation, representations of roles, and prescriptive norms. The norms determine what people in the roles may, must, and must not do in the situation in order to participate in the interaction in the normal way.

When I say that the norms govern what certain people “do,” I aim to include both speech-acts and non-verbal behavior. According to this definition, certain roles are governed by a social script.⁷ For some interactions, such as forming a queue in a minimally egalitarian society, the same script applies to members of all social groups. So the “person joining a queue” role is one that anyone can fill. Meanwhile, other scripts designate roles partly by group membership. For example, in patriarchal societies, heteronormative sexual scripts can be heavily gendered, with clear “man” and “woman” roles.⁸

2.2. Distinguishing Interaction Blueprints from Global Norms and Other Conceptions of Social Scripts

By governing interactions between different agents, social scripts’ norms differ from other social norms, such as a global norm for men to be stoical. That global norm applies to more than just interactions. It also governs solo activities, like pushing through the pain barrier in private exercise, and patterns of thought and emotion, like suppressing thoughts or feelings that make one vulnerable.⁹ That said, social scripts can specify determinate ways of satisfying these global norms that apply to social roles. Hypothetically, the global male stoicism norm could stand alone, with individuals using situational judgment as to which behaviors satisfy the norm. In actuality, this local norm is supplemented by scripts for particular interactions. An example would be a conversational script that specifies that general questions about a man’s life are met with sanguine responses that focus on the less intimate aspects of his life rather than personal issues, problems, or feelings.

These interaction blueprints are analytically distinct from the global norm insofar as the latter would also govern solo activities (e.g. pushing through the pain barrier in private exercise) and patterns of thought and emotion (e.g. suppressing painful memories). Still, we need not presume that this analytic distinction tracks a deep ontological difference as the interaction norms and the global norms can be thought of as species of a broader genus of social norm. As such, what I am proposing is primarily a terminological choice: I propose that we restrict our use of the term “social scripts” to the interaction blueprints. In making this proposal, I allow that nothing of substance turns on which terminology we use to refer to various norms. Still I note two disadvantages of broadening “social script” to refer to any social norm. First, we already have the term “social norm” to refer to the more general category. Second, I have learned in conversation that some people see the lack of specificity of global norms as contrary to the role of scripts in specifying standardized forms of behavior. To illustrate this point, let us consider an analogy between theatrical scripts and social scripts. This analogy is loose because there are differences between paradigmatic theatrical scripts and social scripts. For example, paradigmatic theatrical scripts have a linear structure and specify verbatim sentences for actors to utter while leaving non-verbal behavior up to them and the director. Meanwhile, paradigmatic social scripts have a branching structure, range over non-verbal behavior, and specify the rough contents of communication while leaving agents with discretion as to which natural language sentences express these contents. Now, consider an improvisational theatrical performance in which actors retain artistic license about how to interact on stage so long as they follow general directions about their assigned characters (e.g. “your character is a stoical person”). Since some people find it natural to describe this improvised performance as “unscripted,” they correspondingly find it unnatural to use “social script” to refer to a global social norm for male stoicism.

To clarify the implications of this definition, and to ward off potential misunderstanding with anyone who uses the term differently, let me emphasize some points of contrast with other conceptions of social scripts in the philosophical literature. As I mentioned at the outset of this section, the term “social script” has also been applied to life-narratives and identity-based stereotypes. Let us consider in turn the differences between these conceptions of social scripts and my conception of templates for interactions.

First, in developing an account of the role that shared social resources play in the construction of identity, Kwame Anthony Appiah (2005: 22; see also 1994: 159-160) defines “social scripts” as narratives that people use “in shaping their projects and in telling their life stories.” Specifically, these are the narratives that are provided by collective identities—the identities of “kinds of person” including gay people, Black people, and Americans, as well as butlers, hairdressers, and professors (2005: 65).¹⁰ While Appiah’s paradigmatic narratives concern major life events—and indeed, Appiah often uses “scripts” and “life-scripts” interchangeably—my blueprints concern specific interactions that frequently, though not exclusively, are limited to a short timeframe of a matter of minutes or hours.¹¹ Moreover, there is a respect in which Appiah’s conception is broader than mine insofar as narratives include solo actions and projects, as well as experiences, emotions, and events that are distinct from actions.¹² Finally, there is also a respect in which Appiah’s (2005: 23) definition is narrower than mine insofar as “narrative form entails seeing one’s life as having a certain arc, as making sense through a life story that expresses who one is through one’s own project of self-making.” While these narratives focus on meaningful events and share formal features, characteristic of “novels and movies, short stories and folktales” (22–23), these restrictions do not apply to interaction blueprints, which can also apply to humdrum encounters. For example, when queues of cars merge on a highway, this interaction is governed by a blueprint whose function is to facilitate efficient coordination rather than to enable people to tell the stories of their lives by drawing on their collective identities.

Second, in developing an account of how people’s autonomy can be limited by internalizing negative ideologies that they do not endorse, Natalie Stoljar (2015: 106–107, 117–119) uses the term “social script” to refer to a variety of phenomena, including a negative stereotype of Black people as untrustworthy and a culture’s evaluative ranking of Black people as having inferior social status.¹³ By contrast, these phenomena are not covered by my definition of “social scripts.” On the one hand, prescriptive norms of requirement and permission are distinct from evaluations of social groups as superior and inferior. On the other hand, while blueprints’ norms specify interactions that can be stereotyped, these norms are distinct from other stereotypes, including stereotypes of kinds of people e.g. a pejorative stereotype of Black people as untrustworthy.¹⁴

But while I distinguish interaction blueprints from these phenomena to clarify my terminology, I also want to flag that these phenomena are connected in important ways. For example, if a social group is evaluated as inferior, then this can explain why there are prescriptive norms for interactions that disadvantage or disrespect members of that group. Moreover, if there exists a stereotype for Black people as untrustworthy, then this can explain why there are scripts for members of other groups to act in ways that express a lack of trust. Indeed, because social roles are multi-faceted, I do not mean to suggest that my terminology uniquely carves the social world at its metaphysical joints. Rather, I think that different terminological choices may be more or less helpful for projects with different theoretical goals. For the purposes of this article, my choice is made primarily in order to pick out a suitable object for our inquiry into how the social world structures sexual agency.

3.1. Normality and Deviance

Social scripts can scaffold our agency by creating new options that facilitate coordination. For example, a restaurant script creates a conventional way for a diner to communicate their order to a server. In the absence of a script, the diner and server may still find a non-conventional way to communicate, but the conventional option has distinctive benefits when it comes to efficiently facilitating coordination: The diner and server do not need to spend time and effort thinking either about how they will manage their side of the interaction or about how the other person is likely to behave. Nor do they need to engage in a meta-discussion of how the conversation will take place. Similar benefits are on offer when it comes to sexual coordination. For example, if two people want to flirt, kiss, and have sex with each other, then they can achieve these goals more efficiently if there is a social script that creates conventional options for these behaviors. By contrast, it is likely to be more difficult and awkward for people to approach these encounters in cultures that lack hook-up scripts—in much the same way that it can be more difficult and awkward for people to discuss how they would like to have sex in the absence of scripts for how these conversations are to go.

But the very creation of conventional options can affect other options that already exist. For once a script marks out some options as normal, it also thereby marks out other options as deviant. As I mentioned when elaborating my definition of social scripts in §2, this normality and deviance is merely conventional. As such, it need have no special significance for any agent. For example, if someone is entirely indifferent as to whether they behave in deviant ways, and deviant behavior has no social consequences for them, then the deviance of an option will have no significance for how they deliberate. However, things are different for many agents precisely because they care about avoiding deviance in itself or because they are in circumstances in which there will be social consequences for behaving deviantly. For example, Kate Manne (2018: 169) notes that in certain contexts, deviant behavior can be “perceived as ‘off,’ off-putting, peculiar, and creepy.” In other contexts, deviant behavior can indicate a lack of competence or experience. For example, Emma Atherton (2021) discusses why people may follow scripts that escalate sexual activity:

In my work in sex education, one of the most common things I encounter is fear about not being normal with respect to sex or sexuality. Sex scripts give a blueprint for “normal” sexual interactions and so give a blueprint for what is to behave like and be a “normal” sexual agent or a “good” sexual partner, what “normal” desires are, what it is to be properly “functional” as a sexual agent, and so on. Escalation scripts tell us that “normal” sexual agents “follow through”, “good” partners don’t start what they can’t finish, there’s something wrong with you if you don’t want to follow through, and that sexually sophisticated people aren’t so prudish as to stop at “only” making out. Many people want to be sexually competent and normal.¹⁶

Moreover, deviance can take on a particular significance when it connects with specific aspects of people’s identities. For example, when heteronormative scripts specify gendered roles, Atherton (2021) notes that people

can be invested in performing to the script, in part, because they are invested in doing their gender properly. Failing to act in accord with the script can threaten someone’s sense of being a masculine, sexually normal man, or a feminine, sexually normal woman.

Since some people are averse to behaving deviantly, particularly for their gender, the existence of a social script imposes costs on options that the script marks as deviant: In the absence of the script, these options would simply be neutral with respect to what is designated as normal. These costs may enter practical reasoning that has a cost-benefit analysis structure, as cons for these options to be weighed against any pros. But I suspect that more commonly the pressure to behave normally affects someone’s practical reasoning in a different way: When someone considers deviant options in conscious reasoning, they may have attendant feelings of shame or inhibition. This emotional discomfort creates a psychological pressure to eliminate the options from their list of alternatives that they are deliberating between. At the limit, this will be manifest in these options not entering deliberative consideration at all. Additionally, the deviance of these options can affect someone’s agency in a further way. Even if someone terminates their conscious reasoning in an advance intention to perform the deviant option, then they may find that when the time comes to perform the action, they experience a form of weakness of will that manifests in a failure to execute their prior intention (Holton 1999). This is characteristic of inhibitions that people feel with respect to other social norms governing public interactions. Compare, for example, the experiences of researchers conducting a norm-breaching experiment, designed by Stanley Milgram. Following the experimental design, these researchers formed prior intentions to ask strangers for their seats on the subway, without giving any reason, but found themselves unable to follow through on their plans when the time came. One researcher reported, “I was afraid that I was going to throw up … I really did feel sick to my stomach,” another reported, “I start to ask for the man’s seat. Unfortunately, I turned so white and so faint, he jumps up and puts me in the seat,” and Milgram himself related that “the words seemed lodged in my trachea and would simply not emerge” (Luo 2004). Similarly, I suggest that when people are particularly invested in behaving in normal ways, they often find that they face visceral inhibitions that prevent them from following through on intentions to perform deviant actions.

3.2. Social Obligations

Social normalcy does not only matter to people in its own right. In addition, it can influence the social obligations that people attribute to themselves and others. A significant part of this phenomenon concerns the way that social scripts shape people’s expectations of each other’s behavior and hence mesh with social obligations not to disappoint others’ expectations. By “social obligations,” I mean the obligations that people have according to a particular culture or set of shared beliefs. The precise nature of social obligations strikes me as culturally specific, but I assume that social obligations connect with people’s motivations in two characteristic ways.¹⁷

First, an individual is typically accountable to others for breaching a social obligation, and the prospect of being held accountable is a further source of motivation for them. (We will explore some aspects of this accountability in §3.3 on sanctions.)

Second, social obligations can be internalized with the consequence that people take these obligations to provide them with weighty—and sometimes decisive—non-instrumental reasons for action. This internalization can also have the consequence that they often feel negative emotions, such as guilt or shame, when breaching a social obligation. However, in saying this, I do not mean to imply that people are necessarily correct in taking social obligations to provide non-instrumental reasons for action. Similarly, I do not mean to imply that social obligations are moral obligations. For example, in a patriarchal culture, women may be socially obligated to defer to men, even though it is false that women are morally obligated to defer, and it is false that they have non-instrumental practical reasons to do so.¹⁸

As a way into this topic, let us consider how people’s sexual-decision-making is influenced by the social obligation to be polite. Earlier, we saw that Rachel had sex “because it was ‘polite’ to just let him finish,” since “there is no way to get out of it without coming off as rude or disappointing your partner” (Bennett & Jones 2019: 82). Other students had similar experiences. For example, a Massachusetts student, Courtney, recounted having sex “not because [she] had to under some form of coercion, but simply because it was the polite, lady-like thing to do,” while a New York student, Meaghan, recalled thinking that having “let it go too far now…it would be rude to stop him” (68, 71). Since facts about politeness are social facts, these experiences have to be understood in terms of their social contexts. This should prompt us to ask why these students saw declining sex as rude.

I propose that the explanation concerns social scripts. To make this precise, let us distinguish two ways in which social scripts intersect with politeness norms. One way is that a script can specify a conventional way to express respect.¹⁹ For example, a social script can designate handshaking as a way to express respect upon meeting someone. With respect to sexual decision-making, that possibility strikes me as far less common than another way that scripts structure politeness: Behavior can be seen as rude because it violates an independent social obligation, and a script partly explains why the behavior violates this obligation. In the context of casual sexual encounters, people are often sensitive to a social obligation concerning the expectations that they give each other: they are concerned not to have “led on” their partner. In general, one of the ways that social scripts shape people’s interactions is by guiding people’s expectations of each other’s behavior. Applying this to the sexual realm, when one person follows the earlier parts of a script, such as flirting or kissing, their partner can come to view themselves as having been given a legitimate expectation that they will also follow the later parts of the script—i.e. intimate sexual activity resulting in male orgasm.²⁰ Moreover, once it is common knowledge that people form these expectations, people are mutually perceived as giving each other these expectations voluntarily. This can explain why if someone ultimately declines to have sex, then they may be perceived as having led the other person on and indeed may view their own behavior in this way. It can also explain why people feel that they have entered into a sexual “contract” that they are no longer unilaterally free to exit. As a New York student, Livia, related, “Entering the dating and hookup scene with low self-esteem and little knowledge led to many encounters of an ‘icky’ nature: I didn’t know how to stop them, once started, and often felt as if I was contractually obligated to take the guy to the end and expect nothing in return.”²¹

The social obligations that are structured by social scripts can be deliberatively significant in their own right. Someone can feel intrinsically pressured to meet what they take to be their social obligations. And they can feel pressured by others’ beliefs that they have these obligations.²² Indeed, people can also be influenced by others’ expectations, even when they neither think that these expectations are appropriate nor view themselves as responsible for creating these expectations. For example, someone may recollect the various ways in which they have indicated that they do not want an encounter to escalate from kissing to sex, even though their partner has been inattentive to these signals. Accordingly, this person may judge that they have not given their partner a legitimate expectation of sex, while realizing that nonetheless their partner has this expectation. Even though this person sees their partner’s expectation as baseless, they can still feel pressured to meet the expectation.

3.3. Sanctions

Because social scripts provide determinate content to certain social obligations, these scripts in turn can also explain patterns of sanctioning: Sometimes, when people break from social scripts, they get punished by others for doing so.

The connection between scripts and social obligations is crucial when it comes to sanctioning. We might think that we can get by without the connection if we are tempted by the simple thought that script-breaking is sanctionable as a form of social deviance. But this thought is too simple in light of the fact that scripts merely state the normal ways for encounters to happen and not all unconventional behavior is sanctioned. Instead, some unconventional behavior may simply be seen as oddball rather than as meriting punishment. And in other contexts, unconventional behavior can be welcome. For example, Nick Riggle (2017) argues that it is “awesome” to engage in unconventional behavior that creates social openings for people to express their individuality.

Nor does the simple thought account for the fact that some transgressions are sanctioned more punitively than others. To explain this variation, it is necessary to attend to the social context of script breaking. An important part of this context is constituted by people’s beliefs about whether script-breaking is also the breach of an independent social obligation and, if so, how grave that breach is. These beliefs matter because obligation breaches may be seen as independently sanctionable, and the appropriate sanctions are typically seen as proportionate to the gravity of the breaches. As we saw in the last sub-section, social scripts can shape what someone is socially obligated to do: By shaping people’s expectations of each other’s behavior, these scripts provide specific content to the general obligation not to disappoint expectations. Since a breach of this obligation makes some vulnerable to sanctioning, social scripts influence the sanctions that people face.²³

These sanctions include not only physical harms but also social retaliations.²⁴ These costs are bad for people when avoiding an unwanted sexual encounter means breaking from a social script. And these burdens are politically significant when people face these sanctions in virtue of belonging to a social group since “[c]ultural norms and informal practices that impose unfair burdens on or create disproportionate opportunities for members of one group as opposed to another are oppressive” (Haslanger 2012: 315).²⁵ This scenario is realized when heteronormative sexual scripts intersect with gender norms and stereotypes in ways that constrain women’s ability to avoid unwanted sex and thereby constitute a patriarchal subordination of women’s interests for the sake of men’s sexual interests and pleasures.

Empirical investigation would be required to determine at which times and places, social scripts and norms disproportionately constrain women’s sexual agency.²⁶ Since a general investigation of this matter exceeds what is feasible in this article, I will focus on one phenomenon as an illustrative example: social stigmatization based on gendered stereotypes. This phenomenon is exemplified by a college student, Leanne’s, recollection that “by the time clothes were shed, it would’ve been awkward to stop,” explaining that she “didn’t want to be labeled a prude, a tease” (Bennett & Jones 2019: 81). Here, the gendered stereotype of a “tease” is a woman who does not intend to have sex with a man and yet intentionally causes him both to desire having sex with her and to think that sex is probable. Thus, to be labelled a “tease” is to be stigmatized as untrustworthy and inconsiderate.²⁷ When a woman breaks from a background sexual script according to which her prior flirting and kissing is followed by more intimate sexual activity, this gendered stereotype becomes salient as a possible interpretation of her behavior. Being labelled as a “tease” may lead to additional social sanctions such as angry behavior from a partner and even violence. But even without these additional consequences, the stigmatization in itself functions as a sanction insofar as people care about how others think about them.

This stereotyping can be partly explained by how, as we have seen, social scripts can shape people’s expectations of sex. In addition, the application of this stereotype can partly be explained by a background social script determining which types of behavior call for explanation. To see how this can go, compare a sexually conservative university where the operative heteronormative script is for people to kiss and flirt but not have sex. There, a woman’s declining sex after flirting would not be interpreted as behavior characteristic of a “tease.” Instead, in virtue of conforming to a local script, this behavior would be accepted as normal and as requiring no special explanation. By contrast, when there is local social script for an encounter that escalates in intimacy, and someone breaks from the script, this usually raises the interpretive question of why they did so. In the context of a hookup, relative strangers typically have little more than generic, culturally received expectations about one another, and so their interpretations are particularly likely to be influenced by stereotypes that are locally available.²⁸ Thus, the fact that ending a sexual encounter constitutes breaking from a social script can explain why interpretive questions are raised about women’s declining sex and hence how they can be stereotyped as “teases” for doing so.²⁹