On Media Reports, Politicians, Indirection, and Duplicity

Abstract

We often say one thing and mean another. This kind of indirection (concerning the content conveyed) is both ubiquitous and widely recognized. Other forms of indirection, however, are less common and less discussed. For example, we can sometimes address one person with the primary intention of being overheard by someone else. And, sometimes speakers say something simply in order to make it possible for someone else to say that they said it. Politicians generating sounds bites for the media are an example of this kind of indirection. In this paper, I will explore—via a series of fictionalized examples—these less discussed forms of indirection and consider how such forms of indirection can be duplicitous and misleading.

1 Introduction

We often say one thing and mean another. This kind of indirection (concerning the content conveyed) is both ubiquitous and widely recognized. Other forms of indirection, however, are less common and also less discussed. For example, we can sometimes address one person with the primary intention of being overheard by someone else. And, sometimes speakers say something simply in order to make it possible for someone else to say that they said it. Politicians generating sounds bites for the media are an example of this kind of indirection. In this paper, I will explore—via a series of fictionalized examples—these less discussed forms of indirection and then consider ways that they can be duplicitous and misleading.¹

The paper proceeds as follows. In the next section, we review certain aspects of linguistic communication that will aid our investigation. Then, in Sect. 3, we identify some of the considerations involved when a speaker decides what to actually say in order to communicate what that speaker means to get across. In Sect. 4, four forms of indirection are identified, and then, in Sect. 5, we explore how these forms of indirection can be intentionally misleading. In Sect. 6, a fictionalized example—Ripped from the Headlines—is presented; this example involves duplicitous instances of the two less familiar forms of indirection; it is also inspired by real events and highlights how deceptive the news media can be. Finally, in Sect. 7, we identify ways in which the preceding discussion call into question some widespread but tacit assumptions about communication and conversation.

Our first order of business then is to take a look at the inferential nature of communication. To this task, we now turn.

2 On Linguistic Communication

As is well known to those who study language, linguistic communication is highly inferential. This means that when we decide what to say on some particular occasion and when we are interpreting what someone else means by what they say, we need to do quite a lot of figuring out. It is not a mere matter of simply decoding the meaning of what is actually said. To get at what the speaker means by what they say, we need to perform complex and context-sensitive inferences. Such inferences are so routine, however, that we are typically hardly aware of performing them. Moreover, we are very good at doing it. We routinely perform lots of complex inferences; we do it very quickly, and we are barely aware of doing so.

Let’s focus for a moment on a simple example. Consider the following.

Divorce: Two work colleagues, A and B, are discussing the recent uncharacteristic behavior of their fellow work colleague, Joan. The following exchange takes place:

A: Joan has been jumpy and irritable lately.

B: Divorce is stressful.

In Divorce, B communicates to A that Joan is going through a divorce; Joan is stressed by the divorce process, and this explains the recent change in Joan’s behavior. Notice, however, that B does not actually say any of these things. Instead, B says ‘divorce is stressful’ and A works out that B means all of these things by saying this in this context. H.P. Grice first drew our attention to this phenomenon; it’s called ‘conversational implicature’.² The conventional meaning of the words actually uttered by a speaker (or more technically the fact that the speaker has uttered these particular words with these particular meanings in this particular context) are just one clue among many that a hearer uses in figuring out what a speaker means to get across. In sussing out B’s intended meaning, A would also rely on the conversational context, background information, an ability to make inferences, and the presumption that B is being communicatively cooperative. This is one familiar way that linguistic communication is highly inferential.

There are many further complexities regarding the sorts of inferences involved in linguistic communication but I shall here identify just one that will be relevant to our investigation.³ Levinson (1979) stressed that conversations are embedded within broader social activities and that the norms governing those broader activities inform how we interpret particular conversational contributions. In short, the norms governing the broader activity inform what sorts of contributions are appropriate and this, in turn, guides the interpretation of the utterance in question. This means that the nature of the broader activity is a crucially important part of the context and it guides the identification of the relevant bits of background information. Levinson’s insight enriches Grice’s analysis. Let’s apply this to the above example.

In Divorce, A and B are discussing the behavior of their colleague, Joan. Permissible moves must contribute to an understanding of Joan’s behavior. Such contributions might add evidence that Joan has been behaving out of character, describe how others have reacted to Joan’s behavior, or offer explanations of that behavior, just to name a few. Levinson’s insight is that A interprets B’s utterance in light of these options. So, how we interpret any utterance is guided quite strongly by the norms and expectations of the broader social activity involved; it is not just a matter of conversational norms or of general norms of cooperation; the norms of these broader social activities are crucial as well.

To see this more clearly, let’s consider a different version of the example, a version involving a different broader social activity in which the very same words are uttered.

Divorce2: C, who is about to go through a divorce, and D are good friends; C has been offered a new job and the two friends are trying to decide whether C should accept that job. The following exchange takes place:

C: It would be exciting but a steep learning curve; the timing isn’t great.

D: Divorce is stressful.

In Divorce2, C and D are discussing whether C ought to accept the new job. Permissible contributions, in this context, then should offer considerations relevant to that decision. Appropriate contributions might add a consideration for or against taking the job, clarify the weight of a reason being discussed, request clarification about one such reason, and so on. In light of these options, D’s utterance ‘Divorce is stressful’ is rightly interpreted as offering a consideration against taking the job. C takes D to mean something like ‘one reason against taking this job is that you are about to get divorced and you won’t be at your best because the divorce process is really stressful, distracting, and exhausting’. So, the uttering of the very same words means very different things in Divorce and Divorce2. And, the norms governing the broader social practice are a very important factor in recognizing the different speaker meanings.⁴

3 Aspects of Speech Production

Analytic philosophers of language have certainly had a lot to say about the inferential nature of linguistic communication but the emphasis tends to be on the interpretation side of things. In other words, the attention has been placed on explaining how we figure out what a speaker means by what a speaker says. Here, I want to focus instead on the production side of linguistic communication.⁵ I’d like to identify some of the considerations affecting how a speaker decides what to actually say in order to get across their intended meaning.⁶

3.1 Multiple Participants

Ordinary utterances are often heard by—or even directed at—more than one person.⁷ And, when this is the case, the speaker crafts what they say in order to affect how each participant interprets what they mean. Consider the following:

Parent1: E and F are parents of spirited young children; the family is about to visit E’s (uptight) mother’s house. E says to F within earshot of their children: We can get frozen dairy delight post-visit so long as behavior pleases.

In Parent1, what E says is engineered to make sense to E’s partner but to go over the heads of their young children. E’s awareness that the children can hear what E says affects how E chooses to communicate with F. E chooses their words so that F can understand the intended message but the children cannot. This sort of thing is very common. Here is another example.

Parent2: J and K are eating dinner with their young children. J announces to the family: Sarah called this afternoon; Aunt Kathy is going on vacation again.

In Parent2, K rightly takes J to mean that Aunt Kathy has fallen off the wagon and is in a rehabilitation program again; the children take her to mean that Aunt Kathy won’t be around for a while because she is going on a fun and relaxing vacation. Here, J crafts what she actually says so that her partner picks up one message and her children another.⁸ Now, it might be tempting to think that K got it right and the kids misunderstood but that’s inaccurate. After all, when J says what she says, she intends for her children to interpret it exactly as they do; the children correctly identify her speaker meaning for them. This is a case where a speaker crafts an utterance in order to communicate different messages to different participants.⁹ Moreover, it is perfectly ordinary. Again, speaker awareness of multiple participants affects which particular words a speaker actually says.

3.2 Multiple Contexts of Interpretation and Multiple Timelines

Speech situations where one’s utterance will be interpreted in different contexts (from that in which it was produced) and at different times in the future are also already familiar and quite commonplace.¹⁰ Here is one such example that will be familiar to many readers.

Many of us are tasked with writing descriptions for our courses and these descriptions are routinely read by our departmental colleagues. Since course descriptions serve multiple purposes, they are read by those same colleagues in different contexts. Sometimes, course descriptions are read in preparation for putting the curriculum together. Other times, they are read in preparation for an evaluation of my job performance. Still other times, they are read in the process of updating the public-facing departmental website. Aware that my colleagues will be reading my course descriptions in these various contexts and capacities, I adjust my descriptions accordingly. I craft those descriptions mindful of the various contexts in which my colleagues will read them. Here again we see Levinson’s insight that the norms and expectations of the broader social activities involved (in this case the course catalogue, the faculty evaluation process, and the public-facing website) guide how we craft our linguistic contributions to these activities.

Of course, in this case, serving the different purposes at hand requires communicating effectively with different audiences. The course catalogue is addressed to potential students; the faculty evaluation process is directed at the reviewers involved, and the website aims to communicate with the general public. And, when my colleagues interpret my course descriptions, their judgments are guided by how they think those audiences might interpret what I have written. Here, we see one way that these axes interconnect; we see how awareness of multiple participants and multiple contexts can interact.

In short and unsurprisingly, when a speaker is aware that her utterance will be interpreted in different contexts and at different future times, this affects how that speaker says what she says.

3.3 Other-Agent Reuse

Another feature of linguistic communication, the awareness of which affects production, is the possibility that our utterance will be re-used somehow by other agents.¹¹ Gossip can be repeated; tweets can be retweeted; Facebook posts can be commented on and shared; speech acts that are recorded can be reported on and replayed for new audiences.¹²

Awareness of other-agent re-use is already a familiar phenomenon and it is part of perfectly ordinary language use. To see this, consider the following:

Malevolent Gossiper: L and M are neighbors. M is well known to enjoy causing friction between people. Whenever L speaks to M, L is aware of this and is therefore very careful about what she says and how she says it. L does not want any part of M’s trouble-making.

In Malevolent Gossiper, L is aware that what she says will be repeated; L is also aware that M’s agenda is different from her own. In light of these considerations, L adjusts what she says when speaking to M. Awareness of the possibility that M will re-use her words (in order to stir up conflict), L is careful to avoid saying anything that M could re-use to cause trouble. Again, awareness of this feature affects production. In Malevolent Gossiper, the speaker aims to avoid producing strings of words that are well-suited to M’s trouble-generating re-use.

Sometimes, however, speakers aim to produce such strings that will be re-used. Here is an example that will be familiar to many faculty members.

Medical School Letter: Professor N has been asked to write a letter of recommendation for student O to go to medical school. At their home institution, a committee writes a composite letter stating the case for each medical school candidate and, in that letter, the committee quotes from the individual faculty letters. (The individual letters are also a part of the student’s application file.) Aware of this practice, Professor N strives to write a letter with good quotable bits.

In Medical School Letter, Professor N is aware of how the committee works and, wanting to produce a letter that is useful to that committee and thus to the candidate, Professor N crafts the letter so that it contains short quotable bits of relevant positive information. Again, we see that a speaker’s awareness of the possibility of other agent re-use affects the production side of things. And this is as it should be.

4 On Various Axes of Indirection

In this section, I shall briefly revisit some already familiar forms of indirection before drawing attention to two less familiar forms.

4.1 Already Widely Recognized Axes of Indirection

When discussing the inferential nature of linguistic communication in Sect. 2, we saw one very familiar kind of indirection. In Divorce, B says one thing (namely ‘divorce is stressful’) in order to communicate something else (namely that Joan is going through a divorce; that her divorce is stressing her out, and it is causing her to be irritable). Notice also that, in Divorce, B means what she literally says (namely that divorce is stressful) but B’s primary reason in saying anything at all is to get across what is implicated and not what is actually said. The meaning of the words actually uttered function primarily as a clue to enable A to figure out B’s primary meaning. Conversational implicature then is a form of indirection. What is literally explicitly said is not the primary communicative point. What is inferred from what is literally explicitly said is. This is indirection at the level of content. It concerns the message that the speaker primarily intends to convey.

Indirection at the illocutionary level is also quite ordinary and highly theorized.¹³ To see this, consider the following.

French Fries: P and Q are dining together and sharing a plate of French fries. At a certain point, P says to Q: Can you pass the salt?

Here, P poses what is literally a question about Q’s physical ability (to pass the salt). But, we all know that P’s primary aim is to request that Q pass the salt. In French Fries, the direct speech act (the question) is merely a tool for enabling the hearer to figure out the primary point (of requesting that the salt be passed). Such indirect speech acts are quite common.¹⁴

4.2 Two Additional Axes of Indirection

Implicature and indirect speech acts are both familiar and widely discussed. Other forms of indirection, however, are significantly less so. Consider first indirection with respect to other agent re-use.

Sometimes what others will do with one’s speech action is the primary point of performing that speech action in the first place. Here is an example.

Medical School Letter2: Professor G has been asked to write a letter of recommendation for student H to go to medical school. At their home institution, a committee writes a composite letter stating the case for each medical school candidate and, in that letter, the committee quotes from the individual faculty letters but those faculty letters are not a part of the student’s submitted application. Only the internal committee actually sees the letters from individual faculty members. Aware of this practice, Professor G strives to write a letter with good quotable bits but takes little care with the rest of the letter.

In Medical School Letter2, Professor G knows that the letter they write will not be read by the admission committee at the medical school. Professor G also knows that the internal committee is only interested in lifting quotable bits from their letter. In light of this, Professor G seeks to produce a letter with good liftable bits but Professor G does not waste energy polishing or crafting the rest of the letter. In Medical School Letter2, Professor G’s primary point in writing the letter at all is to generate liftable bits for the committee to quote. And, given how this system operates, this is as it should be.

Another (related) potential axis of indirection concerns the intended audience. A speaker might say something to one person with the primary aim of being overheard by someone else. Such cases are familiar enough. Here are two:

Courtroom: A lawyer questions a witness on the stand. The lawyer already knows what the witness will say in response to each of the questions posed. The primary purpose of the lawyer’s questions is to be overheard by the jury.¹⁵

Smoker in Line: R and S are in line to get into a night club. The person behind them in line is smoking. Knowing that S does not have any cigarettes, R says to S: Oh, how I wish I had a smoke. You got any?

In Smoker in Line, R speaks to S but R’s primary aim is to be overheard by the smoker behind him in line.¹⁶ R says what he says hoping that the smoker will overhear and give him a cigarette. R’s primary audience is the smoker and not S the addressee. Here, R might not care one way or another whether the smoker behind him recognizes his intention to be overheard by that smoker. In other cases, however, a speaker does not want this intention recognized. In the next section, we explore that sort of case.

5 Duplicity with Respect to Indirection

Let’s face it. Language is sometimes used to manipulate. Speakers can outright lie.¹⁷ More often, however, we mislead by indirection. That is, we say something literally true but misleading. We aim to verbally mislead by relying on implicature to communicate (without actually saying) something that we do not believe to be true.¹⁸ In what follows, I would like to explore how indirection with respect to the two less-theorized forms of indirection can be used to mislead. We shall consider how duplicity with respect to one’s intended overhearer and with respect to a speaker’s intentions regarding other-agent reuse can each be exploited in order to deceive.

5.1 Covertly Intended Overhearer

Sometimes a speaker does not want an intended overhearer to know that an utterance is intended for them. Consider the following.

Santa Claus: Within earshot of their daughter Nora, T says to his partner U: I sure hope the kids settle down tonight; Santa only comes if the kids are asleep.

In Santa Claus, T addresses his husband U but T’s primary aim is to get a message to Nora, the intended overhearer.¹⁹ Unlike Smoker in Line, where the speaker is indifferent as to whether the intended overhearer (the smoker behind her) realizes that the message is intended for them, T definitely does not want Nora to realize that she is the intended overhearer. Following Bach and Harnish, T’s intention to get that message to Nora is covert.²⁰ With covert intentions, the speaker’s aim depends on that intention not being recognized.

To further illustrate this notion of covert intentions, consider an act of lying.

The Lie: V and W are in an exclusive romantic relationship; V correctly suspects that W has been cheating. During a discussion about the suspected infidelity, W says: I have never cheated on you.

In The Lie, W asserts that W has never cheated on V. W’s intention to assert this is recognized by V and W wants that intention to be recognized by V. This is how communication works. Since W is lying, however, W also knows that what he is asserting is false and W intends to deceive V about W’s infidelity.²¹ Although W intends to deceive V, W definitely does not want V to recognize that intention (to deceive). Intentions to deceive are covert intentions since the success of the speaker’s aim to deceive depends on that intention not being recognized.

Let’s now bring this back to Santa Claus. Here, T addresses his partner U but T’s primary aim in speaking at all is to get a message to Nora, the overhearer. Although Nora is the primarily intended audience for T, T does not want Nora to recognize this. For this reason, Nora is (what I am calling) a covertly intended overhearer.

Before we explore how Nora interprets T’s utterance in Santa Claus, it makes sense to first consider a different case. Suppose instead that T had addressed Nora directly and said “I sure hope you settle down tonight; Santa only comes if the kids are asleep”. In this case, Nora would know that the message is intended for her; she would also know that her father says what he says primarily in order to get her to go to sleep. T’s aim to affect Nora and to get her to go to sleep are foremost in Nora’s mind as she interprets what her father says to her.

Let’s now go back to the original version of Santa Claus in which Nora is merely the intended overhearer. Suppose further that Nora does not realize that she is the intended overhearer. In such a case, Nora believes that she is merely overhearing her two fathers speaking to one another. In this case, her father’s intention to get a message to her and her father’s intention to get her to go to sleep are no part of Nora’s interpretation. Rather, Nora operates on the assumption that T’s primary aim is to get a message to U. Since T’s utterance takes for granted that Santa Claus exists, Nora takes this utterance as further proof that each of her dads sincerely believe in Santa Claus.²² Moreover, since her father asserts that Santa only comes when kids are asleep and since having an effect on her is no part of her father’s aim in speaking (at least as far as Nora is concerned), she takes this as very good evidence that Santa only comes when kids are asleep. In light of all this, Nora decides to go to bed early.

Granted that the precise inferences that a child will make in a situation like this can vary and be contested, this much is clear: whether Nora realizes that she is an intended overhearer surely makes a difference as to how she interprets what her father says. In short, T’s aim (of getting Nora to go to sleep early) is much more likely to be successful if Nora is an overhearer, rather than the addressee, and if Nora does not recognize that she is the intended overhearer.

Covert intended overhearers are not uncommon. Here is another case.

Saving Face: X, Y, and Z are co-workers. Z is thrifty but criticized by some co-workers as cheap. There had recently been a big promotion; the office sent flowers to the promoted person; X and Y contributed; Z did not. X worries that Z mistakenly believed that the company paid for the flowers. Knowing that Y already knows this and intending for Z to overhear, X says to Y: You contributed for the flowers right? You might not realize this but the company did not pay for them.

In Saving Face, X wants to get a message to Z without Z realizing it. Although X could speak directly to Z, it might be embarrassing for Z and it might be awkward for X. So, in order to avoid a potentially socially disruptive experience, X speaks to Y with Z as the covertly intended overhearer.

As these examples demonstrate, covertly intended overhearers are common enough. When speakers have covertly intended overhearers, speakers are being sneaky, deceptive, and manipulative. In some cases (e.g. Santa Claus and Saving Face), such manipulation seems permissible or at least not obviously wrong. As we shall soon see, however, others cases are more fraught.

5.2 Covertly Intended Re-use

Sometimes a speaker does not want others (either the addressee or those interpreting their utterance) to know that their primary aim in producing the utterance is for other agents to reuse it. To see this, consider the following example:

Malevolent Gossiper2: L and M are neighbors. M is well known to be a malevolent gossiper and L is well aware of this. Although L is also a trouble maker, L has managed to keep this under wraps; that is, L has managed to manipulate people without those people realizing it. Aiming to stir up trouble once again, L says things to M with the primary aim of having M repeat them. L is careful to word their utterance in a way that will enable L to deflect responsibility down the road when things get ugly as planned. In order to wreak the intended havoc, it is important to L that neither M nor those who will hear M’s gossip realize that L aims to have their words repeated.

In Malevolent Gosspier2, L relies on the fact that M will repeat what they say and, in this case, L’s primary aim in saying anything at all to M, is to produce words that M can repeat. L does not want those who hear the gossip to realize that L’s primary aim was to have their words repeated in this way. Malevolent Gossiper2 is a case of other agent re-use indirection. Moreover, L’s intention that M re-use their words is covert. L does not want M or those who hear M repeat what L has said to realize that L intends for their words to be repeated. If that were to happen, L’s cruel intentions would be less likely to be realized.

Many political dogwhistles appear to work this way. The now infamous Billie Horton ad, for example, was originally a small-scale local advertisement but it was generated with the intention that it be picked up and spread by the media. Other-agent reuse is how that dog whistle was designed to work.²³

As these above examples demonstrate, these two forms of indirection can be covert and their being covert can facilitate a non-admirable form of duplicity. In the following section, we will consider a fictional case that involves these two forms of covert indirection.

6 The Fictionalized Media Case

Media reports can involve both of these types of indirection. Moreover, such cases appear to be increasingly in frequency. And, in light of how misleading such indirection can be—especially when it is covert—we need to be on our guard. Let’s get to the example.

Here is the background. Judge Nelson has been nominated to the Supreme Court of the United States and is being questioned by members of the Senate Judiciary Committee as part of her confirmation. One line of questioning concerns the sentences Judge Nelson handed down in some of her child pornography cases. In one particular case, a certain convict received a relatively short sentence due to requirements of the relevant law, the age of the defendant, the defendant’s lack of any prior offenses, and the comparatively low number of images that were part of the case. During Senator Ivy’s questioning of her, Judge Nelson patiently explains why the sentence was what it was. After doing so repeatedly, the following exchange takes place:

Senator Ivy: And, do you regret that sentence?

Judge Nelson: No. As I explained, I followed the law; that’s my job as a judge.

Senator Ivy: So, you don’t regret allowing that monster right back into society in order to re-offend and scar our children.

When a speech action is recorded, as these hearings were, others can report on these speech actions and can even re-use them in a variety of ways. Moreover, the media routinely does so. In light of this, consider the following example.

Ripped from the headlines: Ritenews personality states: Senator Ivy pressed Judge Nelson about Nelson’s apparent leniency with respect to her sentencing of convicted child pornographers. Ritenews then cuts straight to a video of Senator Ivy saying to Judge Nelson “So you don’t regret allowing that monster right back into society in order to re-offend and scar our children.” Ritenews then turns straight to an opinion segment about how liberal judges undermine American values.

6.1 Some Complexities and Further Detail

Ripped from the Headlines involves fairly complex communicative acts: a Ritenews newscaster speaks and then re-uses the speech action of Senator Ivy (during the confirmation hearings). Here, one might well wonder who the speaker is.²⁴ Various agents at Ritenews (e.g., the newscaster, the writer, the editor, the media conglomerate) are decent candidates for being the speaker of the re-use of Senator’s Ivy’s utterance. There are also really interesting philosophical questions regarding what sort of speech act is being performed with such re-uses.²⁵ Such a re-use might be an act of direction quotation, an endorsement of the re-used speech act, a drawing attention to that speech act, or something else entirely. Additionally, there are fascinating legal issues concerning the broadcaster’s liability for any false or misleading information.²⁶ Although the Ritenews reuse is ripe with such philosophical complexity, in what follows, I want to focus on Senator’s Ivy’s utterance instead.

Before proceeding, it is prudent to specify further details of the case. First, when Senator Ivy said what he said during the confirmation hearing, we shall suppose that his primary aim was to produce a sound bite for the media. This assumption is plausible in light of the uncooperative nature of what Senator Ivy says. After all, what he says ignores what Judge Nelson repeatedly says about the reasons for that particular sentence. Understood this way, Senator Ivy’s utterance is an instance of other-agent re-use indirection. His primary reason for saying what he said was to produce something that the media could re-use.²⁷ Second, this suggests that—and I further stipulate that,—Senator Ivy’s main aim was not to communicate with his actual addressee, Judge Nelson, but to reach the viewership of Ritenews. Understood this way, Senator Ivy’s utterance also involves audience indirection. Third and finally, Senator Ivy does not want Ritenews viewers to know that they are the intended overhearers and he does not want them to know that his primary aim in speaking to Judge Nelson was to produce words that would be re-used by the media. Senator Ivy’s intentions with respect to both intended overhearer and other agent-reuse are thus covert.

6.2 Potential Duplicity

Let’s now consider how Senator’s Ivy’s utterance might mislead. Given the many aspects of this speech action, the nature of the inferences involved in interpreting it, and the complexity of its epistemic assessment, there are a wide variety of ways that a viewer could be misled. In what follows, I shall focus on just some of them. In particular, I shall explore how this utterance might be interpreted by viewers who do not recognize Senator Ivy’s covert intentions. In other words, we shall focus on the interpretation of viewers who do not recognize either that they are the intended audience or that Senator Ivy’s primary aim is to produce words to be re-used by the news media.

In thinking about how Senator’s Ivy’s utterance would be interpreted, it is worth highlighting at the outset that Senator Ivy’s utterance (“So you don’t regret allowing that monster right back into society in order to re-offend and scar our children”) involves a presupposition trigger (i.e. the word ‘regret’).²⁸ As such, it presupposes the truth of what follows. In particular, it presupposes that Judge Nelson allowed that monster right back into society in order to re-offend and scar our children. As is well known, presuppositions can be a sly way to introduce controversial content and to do so as if it is not controversial.²⁹ By presupposing something, one treats it as if it is already known (or would be readily accepted by all participants); in short, one treats the presupposed content as (what is called) not-at-issue. Moreover, it is more socially disruptive and thus more difficult to question not-at-issue content.³⁰ These are just some of the reasons that make utterances that introduce such presuppositions potentially sneaky.

Moreover, as discussed in Sect. 2, when interpreting a conversational contribution, we are guided by the norms and expectations of the broader social practice to which that contribution adds. And, as presented, Senator Ivy’s utterance purports to be primarily a contribution to the confirmation hearings. Viewers who do not recognize that Senator’s Ivy’s primary aim is to produce sound bites for the media will interpret his utterance as primarily aiming to contribute to the questioning of Judge Nelson as part of the confirmation hearings. And, appropriate contributions to that activity would need to be relevant to the identification of Judge Nelson’s professional credentials, professional performance, and character. In light of this, such a viewer would interpret Senator Ivy’s utterance as an apt contribution to that activity and would likely infer all sorts of (false and unwarranted) things, including—but not limited to—the following: Judge Nelson was in fact lenient in this case; Judge Nelson does not regret that leniency; and Judge Nelson has a callous disregard for the children harmed by child predators.

Ritenews’ framing of Senator Ivy’s utterance seems to be journalistically irresponsible and intentionally misleading. Although the newscaster’s introduction of the clip is true and warranted (Senator Ivy did question Judge Nelson about the apparent leniency of her sentences in some of her child pornography cases), that framing is nevertheless misleading. This is because relevant information about the conversational context is left out. There is no mention of Judge Nelson’s repeated explanations (arguably even justifications) for the apparently lenient sentence in question. As a result, the uncooperative nature of Senator Ivy’s contribution (which is after all good evidence of his insincerity) is thus masked by the misleading framing by Ritenews.

Furthermore, it is no accident that what is inferred here fits with and thus reinforces, a certain right wing narrative. Some right-leaning media outlets, for example, portray left-leaning judges as disregarding the law and deciding cases based on their own personal beliefs; in short, the right sometimes portrays liberal judges as legislating from the bench. For Ritenews viewers who believe this to be true, Judge Nelson’s apparently lenient sentence in this case will not be regarded as a consequences of law or other appropriate considerations and it will also not be taken to be a mistake or a one-off instance. Rather, for those viewers, the apparently lenient sentence in question will be interpreted as the direct result of Judge Nelson’s personal believe that that was the correct sentence. Additionally, many Ritenews viewers will be aware of the Pizzagate conspiracy theory, according to which high-ranking Democrats are deeply involved in the sex trafficking of children. Since a liberal Supreme Court nominee is here portrayed as intentionally and callously lenient on child predation, some viewers could take such inferences as evidence in support of that theory. In fact, Senator Ivy’s utterance appears to be crafted in order to tap into these two negative characterizations of liberals.³¹ Moreover, it even seems as if it is designed to agitate and not to inform.

Notice additionally that neither the Ritenews newscaster nor Senator Ivy say anything that is actually false. But, by exploiting the pragmatics of linguistic communication and by violating principles of relevance and cooperation, both Ritenews and Senator Ivy are able to convey misleading messages without actually lying; they rely on the Ritenews viewership to trust them and to make the intended inferences all on their own.

In order to avoid being misled in these ways, a viewer would need to be on guard. Such a viewer would not assume that Ritenews was providing all the relevant information; and such a viewer would not assume that Senator Ivy’s utterance was a sincere cooperative communication that was aimed primarily at Judge Nelson in an earnest attempt to evaluate her credentials. In short, such a viewer would operate on the assumption that what is actually said is true but refrain from inferring much of anything else from what is said. And, given the inferential nature of communication, such a stance is not sustainable. As we have seen, most of what is communicated is inferred rather than explicitly stated.

Since there is ample reason to believe that misleading reporting and the disingenuous production of media sound bites by politicians occur on all sides of the political aisle, we all need to be vigilant consumers of “news” media. And, such vigilant requires considerable time, energy, and cognitive work.

7 Some Pressure on the Orthodoxy

Before concluding, I shall offer a very brief exploration of how some of the considerations offered here call into question certain fairly widespread but tacit assumptions about how conversations work.

First, in some circles anyway, we have a tendency to think of conversations as discrete temporally-extended events with a definite beginning and a definite end. When I see my neighbor Sally in the supermarket, for example, our conversation begins and then ends right there in the produce section. Of course, things are not so simple. That conversation with Sally probably relied on shared background information and thus picked up on previous conversations in which we have participated. And, when we think about Senator Ivy’s utterance in Ripped from the Headlines, things are even more complex. Given that Senator’s Ivy’s primary aim is to generate sound bites to be consumed by the viewers of media outlets, it seems wrong to treat his utterance as merely a contribution to his conversation with Judge Nelson.

Second, there is also a tendency to think of conversations as involving just two participants, who each take turns speaking. We all know that talk turns often overlap and many conversations involve more than two participants.³²

Third, yet another tacit assumption problematized here is sometimes called illocutionary monism; it is the idea any utterance in context constitutes at most one primary speech act.³³ That a single utterance can be more than one illocutionary act is familiar enough from the standard account of indirect speech acts discussed in Sect. 3.1. Asking ‘Can you pass the salt?” while dining, for instance, is both a direct question and an indirect request. But, according to this account, the request is the primary speech act; the question is performed in order to perform the request. And, according to illocutionary monism, there cannot be more than one primary illocution, an illocution that is primarily intended by the speaker and conventionally realized.

Following Lewiński (2021), there is good reason to suppose that a single utterance can constitute multiple primary illocutionary acts.³⁴ One way to see this, involves multiple hearers.³⁵ Let’s revisit one of our examples.

Parent2: J and K are eating dinner with their young children. J announces to the family: Sarah called this afternoon; Aunt Kathy is going on vacation again.

As argued in Sect. 3.1, one and the same utterance conveys different messages to different participants. When J says this, J conveys to K that Aunt Kathy has fallen off the wagon and will be going back into rehab but J communicates something different to their children (namely that Aunt Kathy will be away for a while, off on a relaxing fun holiday). Such multiplicity of content is uncontested.

Adding to the case, however, demonstrates the plausibility of a plurality of primary illocutionary acts. Suppose, for example, that J and K had been discussing the potential need to split a family trust that was set up to support both K and his sister Kathy, but whose funds had mostly been spent on Kathy’s substance abuse treatments. During the course of those discussions, K insisted that Kathy was better and would not need any more treatment but K also acknowledged that should Kathy need more treatment, changes should be made to the structure of the trust. In light of this background, when J communicates this latest episode with Aunt Kathy, J is instructing K (perhaps even ordering K) to change the family trust. So, one and the same utterance constitutes two different primary illocutions; it is both telling the children that Aunt Kathy is going on vacation and instructing (or ordering) K to change the trust.

The considerations offered here also place some pressure to expand the scorekeeping framework for conversational kinematics.³⁶ The scorekeeping framework is a way to track the context of a conversation.³⁷ Inspired by David Lewis, the score tracks everything that is relevant to a conversation, both in terms of evaluating that conversation and in terms of its proper development. This Lewisian conception of score is a highly inclusive conception and it will track many things.³⁸ In fact, by definition, it tracks everything relevant. This will include—among other things—the topics of conversation, the scope of quantifiers, whose turn it is to speak, what is taken for granted by participants, and plenty more besides. This kind of scorekeeping framework is highly influential in analytic philosophy of language in general; it has also proven useful as theorists have sought to account for more and more social phenomena in language use.³⁹

One way that the considerations offered here place pressure on the scorekeeping framework is temporal. Ripped from the Headlines shows that the primary point of an utterance can be fairly far in the future from the time of its utterance. In order to capture the full complexity of utterances like Senator Ivy’s, the score must then extend (more) through time. Ripped from the Headlines also involves several participants who are placed quite far apart from one another both spatially and temporally. A truly comprehensive score then would need to track all of this. Finally, we have also seen how broader social practices (to which conversational contributions add) are crucially important in interpreting them. Consequently, a genuinely comprehensive notion of score would also need to track any and all such social practices. This suggests that what is really needed is one big score; a score that tracks all (potentially interacting) conversations and all (potentially influencing) social practices across all languages and cultures. Fully motivating—not to mention developing—this all-encompassing notion of score is certainly beyond the scope of this paper but the considerations here are at least suggestive.

8 Conclusions

Indirection with respect to both content and illocution are familiar and highly theorized. Other forms of indirection are less so. We have here identified two less discussed forms of indirection and considered how these forms of indirection can be exploited in order to deceive. We have also explored a fictionalized case involving the media’s reuse of a politician’s utterance; this case—Ripped from the Headlines—involves duplicitous use of these two forms of indirection. Our discussion demonstrates how misleading such indirection can be and how very laborious it would be to always guard against its deception.