Daimon. Revista Internacional de Filosofía, nº 84, 2021 pp. 81-95

ISSN: 1130-0507 (papel) y 1989-4651 (electrónico)

http://dx.doi.org/10.6018/daimon.481861

Licencia Creative Commons Reconocimiento-NoComercial-SinObraDerivada 3.0 España (texto legal). Se pueden copiar, usar, difundir, transmitir y exponer públicamente, siempre que: i) se cite la autoría y la fuente original de su publicación (revista, editorial y URL de la obra); ii) no se usen para fines comerciales; iii) se mencione la existencia y especificaciones de esta licencia de uso.

 

Not all speakers are equal:
harm and conversational standing*

 

No todos los hablantes son iguales: daño y posición
conversacional

 

CLAUDIA PICAZO**

 

Abstract: McGowan has provided a linguistic mechanism that explains how speech can constitute harm. Her idea is that utterances routinely enact s-norms about what is permissible in a given context. My aim is to argue that these s-norms are sensitive to the conversational standing of the speaker. In particular, I claim that the strength of the norm enacted depends on the standing of the speaker. In some cases, the speaker might even lack the standing required to enact new s-norms.

Keywords: hate speech, conversational exercitive, conversational standing, silencing, discursive injustice, sexism

 

Resumen: McGowan ha proporcionado un mecanismo lingüístico que explica cómo el lenguaje puede constituir daño. Su idea es que las proferencias de manera rutinaria establecen normas-s sobre qué está permitido en un contexto dado. Mi objetivo es argumentar que estas normas-s son sensibles a la posición conversacional del hablante. En concreto, sostengo que la fuerza de la norma establecida depende de la posición del hablante. En algunos casos, el hablante puede incluso carecer de la posición requerida para establecer nuevas normas-s.

Palabras clave: lenguaje de odio, ejercitativo conversacional, posición conversacional, silenciamento, injusticia discursiva, sexismo

 

---

Recibido: 31/05/2021. Aceptado: 01/07/2021.

* I am very grateful to María José Frápolli, Alba Moreno, Eduardo Pérez Navarro and Andrés Soria for discussions on McGowan’s framework and their feedback on the ideas presented here. I also thank Carlos Jaén for comments on a previous draft.

** Juan de la Cierva postdoctoral fellow, Department of Philosophy I, Universidad de Granada. E-mail: claudia.picazo@gmail.com. Lines of research: communication, context-dependence, silencing. Publications: Picazo Jaque, C. (2019), “Are mental representations underdeterminacy-free?”, Synthese, 196(2), 633–654, and Picazo, C. (2021), “Homophonic reports and gradual communication, Pacific Philosophical Quarterly, DOI: 10.1111/papq.12368.

 

 

1. Introduction

 

During the campaign for the Madrid elections in 2021, a Vox (a current Spanish far-right party) poster was displayed in a metro station. The poster read: “Un MENA 4.700 euros al mes. Tu abuela 426 euros de pensión al mes.”1 Many people, as well as some political parties, took the message to spread racist ideas and to be a piece of hate speech. The poster was thought to be not only disturbing, but also dangerous. McGowan has provided a linguistic mechanism that explains how this kind of speech can constitute harm (McGowan, 2019). According to her, assertions and other speech acts are not only assertions and so on, but also conversational exercitives that routinely enact norms about what is permissible. Vox’s poster suggests that the government is spending too much money on unaccompanied migrant children, thus making it permissible or appropriate to complain about migration laws. Moreover, it might make it permissible to use de-humanizing terms to refer to migrants. It might also make it permissible to treat migrant children as second-class citizens, less worthy of respect than our grandmothers, at least in the conversation to which the poster is a contribution. McGowan’s idea is that these permissibility facts follow from the rules of conversation and the state of the conversational score. If the utterance were not rejected (which it was), it would make certain harmful behaviours permissible (in a given domain).2

My aim is to argue that the norms enacted by a conversational exercitive are sensitive to the conversational standing of the speaker. McGowan holds that conversational exercitives are not authoritative speech acts. Although I agree that they do not require the kind of official authority necessary to perform paradigmatic standard exercitives, I claim that the standing of the speaker can affect the strength of the norm enacted. Some speakers have more power than others over the normative landscape. In some cases, the speaker might even lack the standing required to enact new norms. I argue that this is what happens in some cases of silencing, and perhaps also in other ordinary exchanges. To see that the identity of the speaker, and thus its conversational standing, might affect her ability to enact permissibility facts, let us compare Vox’s action with what would happen if it was not a political party but an anonymous citizen who uttered those words. Imagine that this anonymous citizen reaches an audience as wide as that reached by the poster. The anonymous citizen is interviewed by a television program with the aim of having a sample of the political ideas of random people before the election, and uses the words quoted above. I take it that, even if the reporter did not reject the anonymous citizen’s utterance, this utterance would fall short of enacting at least some of the norms enacted by Vox’s action. Perhaps an anonymous citizen simply lacks the standing to make it permissible to treat migrant children as second-class citizens. Or, as I argue, she is only able to enact a weak norm. Thus, whether a strong or a weak norm is enacted does not only depend on the words uttered and the rules of conversation. It also matters who the speaker is.

As I see it, the difference between authority and standing is a matter of degree. Consequently, it is not clear that standard and conversational exercitives are two separate categories. All exercitives enact norms about what is permissible or appropriate. Depending on the authority or standing of the speaker, these norms might survive the linguistic exchange or be of limited duration, and be stronger or weaker. Once it is admitted that conversational exercitives are sensitive to the standing of the speaker, they can be seen as a weaker variety of exercitives. Exercitives can be ordered along a scale, with standard exercitives at the top and weak conversational exercitives at the bottom.

The plan is the following. In section 2, I present McGowan’s framework with more detail. In section 3, I argue via examples that how powerful norms are depends on who enacts them. Section 4 develops this claim and considers two potential objections.

 

2. McGowan on speech and harm

 

Speech can harm. This much is common ground. Words can be used, among other things, to enact discriminatory laws and to insult. In a series of papers and a book, McGowan has explored a mechanism by which speech can harm (McGowan, 2004, 2009, 2019). 3 According to her, it can do so in two different ways: it can cause harm, and it can constitute harm. Someone’s words cause harm when they change the beliefs of a group and, as a consequence, the group adopts some kind of harmful behaviour—e.g. the group discriminates or marginalizes certain people. Harm constitution is McGowan’s technical term for a specific way of causing harm. Speech constitutes harm when it enacts norms that, when followed, bring about harm (McGowan, 2019, 23–25).

Some speech acts clearly constitute harm in this sense. Imagine the CEO of a company who declares that women are no longer eligible for work. This speech act enacts a norm that makes it impermissible to hire women. If the norm is followed, it will bring about discrimination. Thus, the CEO’s speech act constitutes harm. In Austin’s speech act theory, this kind of speech act is called an exercitive. Austin defines exercitives as a kind of speech act in which speakers exercise powers, rights or influence (Austin, 1962, Lecture XII).4 Exercitives enact facts about what is permissible (in a given domain), and they do so via the speaker’s authority. In the example, the utterance enacts facts about whom it is permissible to hire (in the company). It manages to do so because the CEO has the authority to implement hiring policies for the company. Let us call this type of speech act a standard exercitive.

Together with standard exercitives, one’s words can also constitute harm in more ordinary ways. For instance, sexist and racist remarks made by ordinary speakers under ordinary circumstances may also discriminate and subordinate. McGowan has identified a mechanism by which ordinary speech acts can constitute, and not merely cause, harm: conversational exercitives. Conversational exercitives enact permissibility facts for the conversation (McGowan, 2019, 27–49). The idea is that the rules governing conversations make it the case that speech acts routinely enact facts about what is permissible in the conversation. Most contributions to a conversation modify what is appropriate to do afterwards. It is in this sense that they are exercitives. Some utterances, such as sexist remarks, may make it appropriate to discriminate or subordinate women. Such utterances may constitute harm.

McGowan follows here Lewis’ view on conversational dynamics (Lewis, 1979). Lewis identifies some similarities between conversations and games such as baseball. To begin with, both are activities governed by rules. Moreover, just like baseball games have a score, conversations can be thought of as activities whose development is captured by a conversational score. The conversational score is more inclusive than a game’s score. It doesn’t keep track of who wins, but of everything that is relevant to the development of the conversation. This includes at least what has been said, the topic under discussion, presuppositions, and whatever is needed to resolve context-sensitivity (standards of accuracy, salient objects, etc.). By definition, the score keeps track of everything that is relevant to determine which moves are appropriate, or permissible, at a given point of the conversation.

Since the score determines which moves are permissible, any speech act that updates the score enacts new permissibility facts. Following McGowan, it enacts a norm that establishes that certain moves are permissible whereas others are impermissible. Here is an example. Suppose that in a conversation about dogs Mary says “My dog loves chewing on shoes”. This utterance makes Mary’s dog the most salient dog, and this is captured by the score. The utterance establishes new permissibility facts. For example, it is now conversationally permissible to refer to Mary’s dog with the expression “the dog”. Therefore, the utterance is a conversational exercitive.5

There is another way in which an utterance can modify what is permissible at a given point in the conversation. Unlike the rigid rules of games, conversations are governed by rules of accommodation. Rules of accommodation update the conversational score so that the actual moves of the participants count as appropriate. Suppose that Mary meets a stranger who is walking her dog. The stranger’s dog starts chewing on her shoes, and Mary says “My dog also loves chewing on shoes”. This utterance introduces the presupposition that Mary has a dog. This piece of information was not part of the score, but it is necessary for Mary’s utterance to be appropriate. If the utterance is accepted, the score is updated with the presupposition that Mary has a dog (via a rule of accommodation) (McGowan, 2019, 36). Because of that, it enacts facts about what is permissible in the conversation. For example, it is impermissible to question the presupposition afterwards. Again, the utterance is a conversational exercitive.

Both standard and conversational exercitives enact norms establishing what is permissible, but there are important differences between them. Unlike what happens with standard exercitives, speakers’ intentions and hearers’ recognition do not seem to play a relevant role in conversational exercitives. A speaker who successfully performs a standard exercitive (e.g., the CEO that enacts a new hiring policy) has the intention to enact permissibility facts. However, one can perform a conversational exercitive without conscious intention to do so. The intention to contribute to the conversation can be enough (McGowan, 2019, 56). Another important difference concerns authority. According to McGowan, conversational exercitives do not require authority either (McGowan, 2019, 66-67). Although I will go back to this point later, let me note here that the kind of example discussed above suggests that any participant in a conversation is capable of establishing new permissibility facts. By contrast, standard exercitives require authority—perhaps even official authority. Finally, there are differences in scope. Whereas standard exercitives enact norms that can still be in place after the conversation (e.g., hiring policies), the norms enacted by conversational exercitives are restricted to the conversation.

It is worth mentioning that conversational exercitives enact norms of a specific kind. According to McGowan, the norms governing a given activity are of two kinds (McGowan, 2019, 85–86). First, some norms, called g-norms, are global and govern all instances of the activity. For conversations, g-norms include rules of grammar and Grice’s Cooperative Principle and maxims. As an example, think of Grice’s Maxim of Quality. According to this maxim, one ought not to say what one believes to be false or unjustified. This norm governs all conversations (or, at least, all cooperative conversations). Second, some norms, called s-norms, are token-activity-specific. These norms are enacted during the performance of the activity-token and govern only that activity-token. For conversations these include, for instance, norms about the topic under discussion or what is salient. Think again of a conversation about dogs. In it, it is permissible to talk about dogs and impermissible to discuss other topics. Unlike Gricean maxims, this norm (an s-norm) does not apply to conversations in general. Conversational exercitives enact s-norms, that is, norms of limited duration that apply only to the particular conversation.

The notion of a conversational exercitive is a useful tool to analyse norm-governed activities beyond conversations, such as informal social interaction, dancing with a partner or playing improvisational jazz (McGowan, 2019, 83). In these activities, some moves count as appropriate or permissible and others count as inappropriate or impermissible. Here as in games and conversations we can think of a score determining which moves are appropriate. What is interesting about this generalization is that, as McGowan notes, an utterance can be a move both in a conversation and in another non-conversational activity. Moreover, it can, and often does, enact s-norms, or permissibility facts, in both activities.

This framework explains how ordinary utterances like sexist or racist remarks can constitute harm. Despite not being standard exercitives, they can enact s-norms that prescribe harm. Moreover, they can enact harmful norms both in the conversation and in the activity of social interaction (or a more specific activity, as interaction with work colleagues or interaction during a date). Here is one of the examples that McGowan discusses. Imagine the following exchange by two co-workers at the employee lounge (McGowan, 2019, 110):

 

Example 1:

John: So, Steve, how did it go last night?

Steve: I banged the bitch!

 

Steve’s utterance is a contribution to two activities and enacts s-norms in both of them. As a contribution to the conversation, it is a conversational exercitive that makes it permissible to use derogatory terms for women. Besides this, the utterance is a contribution to broader social practices. Here McGowan considers the practice of oppression. As a contribution to the practice of oppression, the utterance also enacts s-norms. The mechanism is the same as in conversations: the g-norms and the state of the score determine which moves are permissible at a given point. Although she acknowledges that we would need to know more about the g-norms of oppression and the score in order to specify which s-norms are enacted, McGowan’s hypothesis is that Steve’s utterance makes it permissible to degrade women (at the environment of the utterance, for a limited time).6 It makes women count as second-class citizens. This makes certain behaviours that would harm women permissible, such as undervaluing women’s work. Thus, Steve’s utterance can constitute harm: it enacts an s-norm such that, if it is followed, women will be harmed. Alternatively, we can see this kind of remark as a contribution to the activity of social interaction at the workplace.

McGowan’s framework explains how a system of oppression can be activated in a local context. The speech act does not create the structural oppressions of our societies. Nonetheless, McGowan’s point is that a more or less ordinary remark uttered by an ordinary speaker might “bring the latent force of that gender system to bear in the local context”, as Simpson summarizes it (Simpson, 2013, 563).

 

3. Conversational exercitives and standing

 

In what follows I argue that the social identity or role of the speaker can affect the strength of the s-norms enacted, even depriving her of the ability to perform conversational exercitives at all. In my view, not all speech acts that can be said to enact harmful norms (including, but not limited to, uses of slurs) are equally harmful. Although McGowan herself does not explore the ways in which the identity or role of the speaker can increase or decrease the normative impact of a conversational exercitive, her framework can be easily modified so as to account for such differences.

Let us start by considering what McGowan says about standing (McGowan, 2019, 66–70). Although conversational exercitives are not authoritative speech acts, McGowan herself notes two ways in which the standing of the speaker is relevant. First, in order to be a participant in a conversation one must have a certain standing. For instance, in order to be recognized as a participant in a conversation about local politics one might need to have some knowledge about the topic. A foreigner might be excluded on the basis that she lacks the required knowledge. Second, expertise, and possibly other things, can play a role in the conversation and contribute to the participant’s standing. If a speaker is recognized as an expert in a certain topic, then her contributions are probably believed and not only accepted for the purposes of the conversation.

I hold that something similar happens with harmful speech. The social identity of the speaker or the role she has in a certain context can contribute to her conversational standing. This, in turn, has consequences for the strength of the s-norms she enacts—how they interact with other norms, whether they also enact norms in the activity of social interaction—as well as on her ability to reject a conversational exercitive or even to enact new s-norms. Moreover, this explains why some speakers’ words have more harmful consequences than others’, and why some speakers seem to be communicatively impaired as a result of other speakers’ behaviour.

 

I will add a second example before analysing how the standing of the speaker affects the strength of the enacted s-norms.

Example 2:

The community of the Capital Slam Poetry contest has always been open, constructive, and free of misogynistic speech. However, one day a poet goes on stage and starts talking about women with venom, saying things like “fucking feminists”. (McGowan, 2019, 117)

 

Classifying the speech acts in Examples 1–2 as conversational exercitives allows us to explain why they are harmful. As I said in the previous section, Steve’s utterance in Example 1 might make it permissible to treat women as inferior at the work place—to slur them in conversation and perhaps also to undervalue their work. Similarly, the poet’s misogynistic speech in Example 2 might make it permissible to rank women as inferior in the contest.

The way in which McGowan introduces conversational exercitives suggests that the enacted norms are equally powerful regardless of who the speaker is. However, I think that it is important to note that the standing of the speaker has consequences at the local contexts in which s-norms are enacted. Although in cases of oppression we can say that the main work is being done by the general rules of oppression (by the activity of oppression to which the harmful utterance is a contribution), the general rule needs nonetheless to be activated via an s-norm. My point is that the normative landscape that emerges at the local context where the s-norm is enacted is sensitive to the speaker’s identity.

To see this, consider two versions of Example 1. Both happen during the lunch break, where workmates and managers are chatting about the weekend. In the first version, Steve and John are young interns and, although their workmates treat them with respect, they do not in general attribute much credence to their contributions or pay special attention to their anecdotes. In the second version, Steve is the manager, and the workers pay special attention to what he says, even when it is unrelated to work. I deem it plausible to think that the same words uttered by the manager would have a bigger impact on the work environment than they would if uttered by the intern. The differences can be appreciated if we focus either on the conversational or non-conversational activity. Regarding conversation, it seems that the manager’s utterance would make it more permissible to use derogatory terms for women. As for the non-conversational activity, the relevant activity could be that of social interaction at the workplace. I take it that, again, the manager’s speech act is more likely to make it permissible to be disrespectful with women.7

Similarly, we can image two versions of Example 2. In the first, the one described above, it is a young, unknown poet who starts using misogynistic speech. In the second, it is the organizer of the contest who goes on stage and reads a misogynistic poem. Again, I think that the illocutionary effect is different whether it is the unknown poet or the organizer who reads the poem. For instance, in the first version the audience could be puzzled by the fact that the unknown poet was allowed to read such a poem, but be hesitant about whether hate speech was permissible. By contrast, in the second version it makes more sense for the audience to assume that hate speech is allowed—after all, it is the organizer himself who is using hate speech. It seems that who the speaker is—her standing—is relevant to how powerful the enacted norm is.

I capture this through the idea that norms can be stronger or weaker, that is, that they can have different degrees of strength. We can find examples of norms with different degrees of strength in the activity of social interaction. For instance, there is a norm that makes it impermissible to make bad comments about the physical aspect of our interlocutors, such as making a remark that they have a blemish on their nose or that they have gained weight. This norm is very strong when it comes to strangers and probably weak when it comes to close friends or family. This might happen also with orders or standard exercitives. When a mother orders her child to do his homework, she thereby enacts a strong norm. If it is an elder brother who utters the same words, we can say that he also enacts a norm, but that this norm is somewhat weaker. My view is that something similar can be said about conversational exercitives and s-norms. The s-norm that it is permissible to use derogatory terms for women is strong when it is enacted by the manager but weaker when enacted by the intern. This does not mean that only managers, or those with official authority, can perform conversational exercitives, but only that not all conversational exercitives are alike. In a conversation among a group of friends who see each other as conversational peers, perhaps all utterances are conversational exercitives. However, this might not always be the case.

Examples 1–2 show that s-norms can be more or less powerful depending on who enacts them. They also suggest that some speakers might even be unable to enact certain s-norms, despite of being recognized as participants in the conversation. A plausible explanation why the unknown poet’s move fails to clearly have the same consequences as the organizers is that the first, but not the second, falls short of enacting a sexist s-norm. Although this conclusion can be disputed for Examples 1–2, there are more radical cases. In what follows I propose to interpret the phenomenon of silencing8 as showing that some speakers (such as women) are sometimes unable to enact s-norms.

McGowan argues that her framework provides a mechanism by which pornography (or, better, uses of it) can silence the speech of women. Consider a third example:

 

Example 3:

A man and a woman are on a first date and he starts a pornographic film in which the actress says “No” as a way to sexually excite the actor. Afterwards, the woman says “No” with the intention to refuse sex, but the man doesn’t take her to seriously intend to refuse. Even though she says things like “No! I mean it!” he thinks she is doing that as a way to sexually excite him. (This example is based on examples by McGowan (2019) and Langton (1993)).

 

The woman in the example is communicatively disabled (Maitra, 2009). Although she intends to refuse, and utters the words that are typically used to refuse, she is not recognized as being serious in her refusal, that is, as really intending to refuse. She is silenced. Silencing is a term of art. A silenced speaker is not unable to utter words, but she is nonetheless unable to be recognized as seriously performing the speech act that she intends to perform.9

In the example, we can think of the pornographic film as enacting an s-norm that prescribes silencing. This norm makes it permissible to interpret a woman saying “No” as not being serious, and it makes it impermissible to interpret her words as a refusal. Or, to put it otherwise, the norm establishes that “No” doesn’t count as a refusal.10 This is McGowan’s explanation of the case. However, this cannot be the whole story. What remains to be explained is why the woman in unable to enact a new s-norm that overrides the silencing norm, that is, why is it that (the use of) the pornographic film is able to enact s-norms but the woman who says things like “I mean it” is unable to reverse them. My hypothesis is that, because of her social identity, the woman’s conversational standing is insufficient to perform certain conversational exercitives. In the situation described, the pornographic film is more authoritative than the woman. Whereas the film is recognized by the man as enacting an s-norm about how to interpret “No”, the woman is not equally recognized as rejecting that norm or as enacting a new s-norm to the effect that her use of “No” is a refusal.

In their analysis of the mechanisms by which pornography can silence women, Langton and West draw a useful parallelism with Lewis’ example of a conversation between a master and a slave (Langton & West, 1999). In a language game between a master and a slave the difference in power results in a difference in the speech acts they can correctly perform. When the master says things like “It’s now permissible for you to do x” he thereby enacts a new permissibility fact. This move is correct. But the slave cannot do the same. He lacks the authority to modify what is permissible. According to Langton and West, the speech of women might in some contexts resemble that of the slave—although they are able to utter words, they lack the power to perform certain speech acts they ought to be able to perform. For instance, they might be unable to refuse.

This difference in power is what is lacking in McGowan’s analysis. In the silencing cases the silenced speaker is not on a par with the other participants. She is not recognized as a conversational peer. We can interpret Example 3 in two different ways. First, the power imbalance can result in the woman being unable to reject the speech act that enacts the silencing norm. This is what happens if, right after watching the pornographic film, the woman says something like “Women never say ‘No’ meaning ‘Yes’” and her utterance is ignored. Second, we can consider that the woman is unable to enact new norms that allow her to make her communicative intentions manifest. Because of the time lapse between the enactment of the silencing norm by the film and the woman’s attempt to reverse the situation, I think that Example 3 fits better the second interpretation. When she says “I mean it”, the woman tries to enact a new positive s-norm to the effect that her saying “No” counts, in that context, as a serious and sincere refusal. This norm is in conflict with the one enacted by the film and therefore, if the woman’s norm is enacted, the other one will be overridden. Using McGowan’s framework plus conversational standings, we can see the woman’s move as an attempt to perform a conversational exercitive that is unsuccessful due to lack of conversational standing.11

It is important to note that this inability to reverse the situation is a central feature of silencing. The problematic instances of silencing do not involve misinterpretations that are quickly corrected. The phenomenon is typically persistent. In this sense, imagine a version of Example 3 in which after saying “No” the woman adds “I mean it! Stop!”, and the man stops and apologizes. Although we can imagine that the man took “No” not to be a refusal (even if only for a few seconds), he soon realizes that the s-norm enacted by the pornographic film is no longer valid (if it ever was). In this case one could perhaps say that the woman is silenced for a short moment, but since the communicative disablement is not persistent I take it to be a borderline case. The more problematic instances of silencing seem to involve some kind of persistence.

The conclusion is that the identity of the speaker matters both to the strength of the norm enacted and to the ability to perform conversational exercitives. We can capture this through the notion of conversational standing. The conversational standings of the participants reflect the ranking among them for the purposes of the conversation.12 It reflects the fact that some (probably many) conversations involve a hierarchy. This hierarchy can be inherited from social hierarchies, and certainly social identities and roles play an important part, but I leave it open how the conversational standing of a participant is determined and on what exactly it depends. For instance, the conversational standing could be affected by the participant’s performance in past conversations. If she has proved to be trustworthy, that might grant the speaker a high standing regardless of her social identity. Moreover, it is interesting to note that one’s conversational standing might vary across conversations. For instance, being a woman might be relevant in some conversations where the participants have sexist attitudes, but not in others where sexist beliefs and stereotypes play no significant role. Also, one can be an expert in one topic but not in others and thus have a higher standing in conversations about the first topic and a lower standing in conversations about the second. Anyway, we can assume that who a speaker is (a man or a woman, the manager or the intern, an unknown poet or the organizer of a contest, someone who has proved to be trustworthy or a liar) affects the power she has over the conversational score and, consequently, over what is permissible or appropriate in the conversation and other activities the utterance might be a contribution to. In this sense, we can say that conversational exercitives are sensitive to the conversational standing of the performer. The next section explores how conversational standing has an impact on the s-norms enacted.

 

4. Conversational standing and s-norms

 

In order to make sense of the examples in the previous section, we can think of the enacting of s-norms as dependent on conversational standing. My proposal is that we think of conversational exercitives as involving a threshold for standing and then allow for s-norms to have different degrees of strength, where the degree of strength of a particular s-norm depends again on the standing of the speaker who enacts it.13

Concerning the threshold, we have already seen that it takes a certain standing to be able to enact s-norms, i.e., to perform conversational exercitives (or some specific kind of conversational exercitive). The participant’s conversational standing must meet a given threshold. The standing of the silenced speaker in Example 3 falls below the threshold. Silenced speakers are not recognized as conversational peers, and are unable to enact the s-norms that would allow them to make their communicative intentions available. Minimally, as McGowan notes, one has to be recognized as a participant in the conversation. But often something else is needed. One’s utterances must also be recognized as contributions to the conversation, and not ignored, considered irrelevant or pointless. In order to perform conversational exercitives we need that the participants in the conversation are “willing and capable of hearing us” (Dotson, 2011, 238).

As far as degrees of strength are concerned, we have already seen that the strength of an enacted s-norm depends on the standing of the speaker. Some norms are stronger than others. To put it otherwise, some utterances make some actions more permissible than others. For instance, the organizer of the Slam Poetry contest’s use of misogynistic speech makes it more permissible to use derogatory terms for women than the use of the same words by the unknown poet. In general, the higher the conversational standing of a participant, the stronger the s-norms she enacts will be.

A possible way to see how the strength of an s-norm matters is to explore how the norm interacts with other norms that are in place during the conversation or in the non-conversational activity. We can imagine a workplace in which usual courtesy norms apply in everyday conversations. According to these norms, it is impermissible to disrespect others, e.g., to insult them or use derogatory terms. Take now the first version of Example 1, in which Steve is an intern. Steve’s subordinating s-norm is in conflict with courtesy norms. I think that it might be the case that the new s-norm is not strong enough to override those courtesy norms. All things considered, it is impermissible to use derogatory terms for women, even though there is a sense in which Steve’s utterance might have made it permissible. However, things might be more complicated when it is the manager who utters the sexist remark. In that case, the s-norm might be strong enough to override the courtesy norm at stake or to take priority over norms of courtesy. All things considered, it might be permissible, after the manager’s utterance, to use derogatory terms for women.

Something similar can be said about the non-conversational activity. Imagine now a workplace in which social interaction is governed by an egalitarian norm to the effect that all workers are to be treated in equal terms, all equally respected. In this workplace, the s-norm enacted by the intern might be insufficient to cancel this norm. But again, the manager’s s-norm, being stronger, could result in non-egalitarian social interaction, with women counting as second-class workers.14 However, it will not be strong enough to override anti-discriminatory laws. Laws, that is, legal norms, are stronger than the kind of norms that are enacted in conversation. Not only is their duration longer and their scope wider, but they also take priority in case of conflict. A manager cannot enact an s-norm that overrides a law. What the manager can do is make certain actions socially acceptable.

Finally, once the standing is in the picture and it is admitted that conversational exercitives can enact stronger and weaker norms, one can see the relation between conversational and standard exercitives as a matter of degree. The more authority the speaker has in a given domain the stronger will be the norms she is able to enact. In the paradigmatic examples of standard exercitives, a special kind of authority will be required.

In what follows I discuss two potential objections to the view presented here. The first objection concerns the threshold. I have claimed that some speakers lack enough standing to perform conversational exercitives, that is, to enact s-norms. The example I used in support of this claim is one in which the speaker is silenced, using Langton’s terminology. It could be objected that whether or not her interlocutors are aware of it, and regardless of their subsequent behaviour, the woman has enacted an s-norm. She has performed a conversational exercitive despite the fact that her interlocutor fails to recognize the new permissibility facts. This is because the enactment of norms depends only on the rules of conversation and not on the attitudes of the participants in the conversation. Given the rules of conversation, she has enacted an s-norm. This objection touches upon difficult questions regarding the notion of conversational score (and of common ground) and the determination of its state at a given point of the conversation. Two broad strategies can be followed here: one can, in the Gricean spirit, explain the score through the propositional attitudes of the participants, or one can, in a more Austinian tradition, think of communication as a conventional practice governed by rules.15 The objection assumes that the second strategy is preferable here, or at least that it should not be discarded. However, my point can be seen as concerning the way in which the theoretician can identify the rules of conversation. A reasonable way to find out whether all speech acts enact s-norms is to consider whether all speech acts are recognized as modifying what is permissible. If it turns out that some speech acts are not treated as changing the permissibility facts, then one reasonable explanation is that the rules of conversation are such that those speech acts do not enact s-norms. In the silencing cases, and in particular in Example 3, the woman’s words are intended to make it impermissible to interpret an utterance of “No” as a non-serious refusal. However, they are treated as not enacting this s-norm but, rather, as re-enacting a previous s-norm to the effect that it is permissible to interpret utterances of “No” as not being genuine refusals. Thus, using this example as a way to discover the rules of conversation, there are reasons to think that the rules of conversation are such that some speakers cannot perform certain conversational exercitives (in certain contexts).

Still, it could be objected that it is not the case that the speaker lacks the standing to enact an s-norm but, rather, that the norm is enacted and then rejected. Generalizing this idea, one could hold that all speakers perform conversational exercitives but that some of them are automatically rejected. I think, however, that this would be an odd description of Example 3. In that example, the woman’s words (“No! I mean it!”) are not rejected, but rather reinterpreted as not being serious. It is not the same to misinterpret the force or the seriousness of a speech act as to reject it.

A more problematic case would be one in which the speaker is not silenced but her words nonetheless fail to enact new permissibility facts. I suggested that this could happen with the intern’s utterance in Example 1 or with the unknown poet in Example 2.16 I think it would not be implausible to hold that some sexist remarks are not powerful enough to enact subordinating s-norms, and, in this sense, that some speakers lack the standing to perform conversational exercitives (in some contexts). The objection would apply to these cases more straightforwardly than in the cases of silencing. Here, it can make sense to interpret the example as involving an s-norm that is enacted and then rejected. However, it is not completely clear what the dynamics would be, given that, as Examples 1–2 are described, we can imagine that there is no explicit rejection and that what simply happens is that the other interlocutors do not see themselves as being allowed to use derogatory terms for women.

The second objection is that the thesis that s-norms can be more or less powerful, and that some speakers might even be unable to enact s-norms, rests on an illocutionary/perlocutionary conflation. Illocutionary acts are acts like asserting, making a request or warning. They concern the force of the utterance. Following Austin (1962), the performance of an illocutionary act is the performance of an act in saying something. Perlocutionary acts concern the effects of the illocutionary act, what we bring about by saying something. For instance, an illocutionary act can convince or frighten. One could object to the view that I have presented that what I identify as differences in degrees of permissibility or strength of norms is something that belongs to the perlocutionary act. The idea would be that there is a difference between the utterances of the intern and the manager, of the unknown poet and the organizer of the contest. Perhaps the manager and the organizer not only enact a discriminatory or subordinatory norm, but also persuade their interlocutors to discriminate or subordinate women. This might very well be the case. Certain speakers, because of their position, are more persuasive than others. However, I think that McGowan’s claims, and the intuitions her examples elicit, concern the permissibility facts enacted, not whether the audience is persuaded. I see no reason for treating the intuitions elicited by the two versions of Examples 1–2 sketched in section 3 differently. Thus I take it that, when the intuitions concern whether two utterances make it equally permissible to discriminate or subordinate, or whether one is more powerful than the other, then those intuitions concern the permissibility facts enacted, and not how persuasive the utterance is or other similar effects.

 

5. Conclusions

 

I have argued that not all speakers are able to enact the same permissibility facts. Whether some words enact a strong or a weak s-norm, or no s-norm at all, does not only depend on the words uttered and the rules of conversation. In general, a speaker’s ability to enact s-norms, as well as the strength of the s-norms enacted, depends on her conversational standing. Differences in conversational standing can explain differences in discursive power. A high conversational standing allows the speaker to enact strong s-norm, whereas a low standing might only allow her to enact weak s-norms or even deprive her of the ability to perform the intended conversational exercitive. When it comes to hate speech, who the speaker is usually matters to how harmful her words are. Thus, we can see the difference between standard and conversational exercitives as a matter of degree.

 

References

 

Austin, J. L. (1962), How to Do Things with Words, Oxford: Clarendon Press.

Ayala, S. (2016), “Speech affordances: A structural take on how much we can do with our words”, European Journal of Philosophy, 24(4), 879–891.

Ayala, S. (2018), “A structural explanation of injustice in conversations: It’s about norms”, Pacific Philosophical Quarterly, 99(4), 726–748.

Caponetto, L. (2021), “A comprehensive definition of illocutionary silencing”, Topoi, 4(1), 191–202.

Dotson, K. (2011), “Tracking epistemic violence, tracking practices of silencing”, Hypathia, 26(2), 236–257.

Kukla, R. (2014), “Performative force, convention, and discursive injustice”, Hypatia, 29(2), 440–457.

Langton, R. (1993), “Speech acts and unspeakable acts”, Philosophy and Public Affairs, 22(4), 293–330.

Langton, R. & West, C. (1999), “Scorekeeping in a pornographic language game”, Australasian Journal of Philosophy, 77(3), 303–319.

Lewis, D. (1979), “Scorekeeping in a language game”, Journal of Philosophical Logic, 8(1), 339–359.

Maitra, I. (2009), “Silencing speech”, Canadian Journal of Philosophy, 39(2), 309–338.

McGowan, M. K. (2004), “Conversational exercitives: Something else we do with our words”, Linguistics and Philosophy, 27(1), 93–111.

McGowan, M. K. (2009), “Oppressive speech”, Australasian Journal of Philosophy, 87(3), 389–407.

McGowan, M. K. (2019), Just Words: On Speech and Hidden Harm, Oxford: Oxford University Press.

Mikkola, M. (2011), “Illocution, silencing and the act of refusal”, Pacific Philosophical Quarterly, 92(3), 415–437.

Picazo, C. (2021), “Homophonic reports and gradual communication”, Pacific Philosophical Quarterly, DOI: 10.1111/papq.12368.

Picazo, C. (forthcoming), “Are utterance truth-conditions systematically determined?”, Inquiry.

Popa-Wyatt, M. & Wyatt, J. (2018). “Slurs, roles and power”, Philosophical Studies, 175(11), 2879–2906.

Simpson, R. M. (2013), “Un-ringing the bell: McGowan on oppressive speech and the asymmetric pliability of conversations”, Australasian Journal of Philosophy, 91(3), 555–575.

Stalnaker, R. (1978/1999), “Assertion”, in Context and Content, Oxford: Oxford University Press.

Stalnaker, R. (2014), Context, Oxford: Oxford University Press.

 

Notes