Blackmail: How it Works

Blackmail: How it Works Jake Arends

10004206

Contents

1. Introduction...3

1.1 The Felicity Conditions of Blackmail...4

2. Theoretical Framework...7

2.1 Classical Speech Act Theory...7

2.2 Conversation analysis and identifying speech acts...8

2.3 Types of Speech Acts...12

2.4 Power...17

2.5 Identifying Implicit Speech Acts...20

2.6 Conditional Threats and Promises...22

2.7 Indirect Speech and Politeness Theory...23

2.8 Bluffing...24

2.9 Silence in Dyadic Face-to-Face Interactions...28

3. Data...30

3.1 Storyline...31

3.2 The Scene...31

4. Analysis...35

4.1 A Struggle for Dominance (turn 1 – 5)...36

4.2 Blackmail (turn 6 – 8)...38

4.3 An Embedded Challenge and Surrender (turn 9 – 24)...43

4.4 Back to Blackmail (turn 25 – 28)...47

4.5 More Inquiries and Evaluating the Situation (turn 29 – 40)...49

5. Discussion...52

5.1 The Silences...52

5.2 Identifying blackmail...53

6. Conclusion...56

Blackmail: How it Works 1. Introduction

In my term paper (Arends, 2017) I have conducted a theoretical analysis of blackmail using Speech act theory. Speech act theory recognizes that utterances can do more than present information; they can carry out actions as well. Speech act theory originated in the field of philosophy of language and was first introduced by J. L. Austin (1962). Searle (1969), who was a student of Austin, further developed the theory. Speech act theory views utterances as acts and considers “the production of words and/or sentences as the

performance of speech acts” (Sbisa 2009, p. 231). One of the tasks of speech act theory is “to explain in which senses and under what conditions uttering something can be doing

something” (Sbisa 2009, p. 231). The conditions deemed necessary and sufficient for an utterance to be considered an act are called felicity condition.

The aim of my term paper was to determine what the felicity conditions for blackmail are. During the analysis, I identified three speech acts that are inherent to blackmail:

threatening, promising and demanding. The threat and the promise in blackmail are always of the same kind. Invariably, the blackmailer threatens to share compromising information about the victim, and in one way or another, the blackmailer promises the victim that if certain demands are met, the blackmailer will not share the information. However, the promise does not have to be felicitous because even if the blackmailer does not intend to keep the promise, the actions and/or utterances can still be classified as blackmail. The demands, on the other hand, are not restricted by type or quantity.

While my term paper was entirely focused on the theoretical aspects of blackmail as a speech act, this MA Thesis is aimed at the application of the theoretical insights to data. This paper investigates whether the theoretical model described in my term paper fits the data, or

if it should be revised. In order to do this the paper should examine to what extent and in what way these three speech acts appear in a particular instance of blackmail. The data analyzed is from a blackmail scene in the film Ides of March. I came across the scene during my research on blackmail and found it promising in terms of how much a pragmatic analysis of the interaction might uncover. Further, I found few other blackmail data which were appropriate for the type of analysis I had in mind. Most of the cases of blackmail I

encountered were in written form, and without a reply. As the identification of speech acts is often reliant on context and the utterances surrounding the speech act, these cases were not suitable for the aims of this paper.

1.1 The Felicity Conditions of Blackmail

Before proceeding to the theoretical framework I want to present my previous findings. In my term paper (Arends, 2017) I arrived at the felicity conditions for blackmail by analyzing the felicity conditions for the three speech acts incorporated within it: demanding, promising and threatening. The felicity conditions for these three speech acts and those for blackmail are listed below.

Promise

Searle (1969) provide the following felicity conditions for a promise:

Rule 1. The utterance predicates (or implies) some future act A of the speaker S. (The propositional content rule)

Rule 2. The hearer H would prefer S ’s doing A to his not doing A, and S believes H would prefer S’s doing A to his not doing A.

Rule 3. It is not obvious to both S and H that S will do A in the normal course of events.

(The preparatory rules) Rule 4. S intends to do A.

(the sincerity rule)

Rule 5. The utterance (or implication) counts as the undertaking of an obligation to do A.

(The essential rule) (p. 14-15)

Threat

The felicity conditions for a threat are as follows:

Rule 1. the utterance predicates (or implies) some future act A of the speaker S.

(The propositional content rule)

Rule 2. The hearer H would prefer S ’s not doing A to his doing A, and S believes H would prefer S’s not doing A to his doing A.

Rule 3. It is not obvious to both S and H that S will do A in the normal course of events.

(The preparatory rules)

Rule 4. S wants H to fear a future action of S. (the sincerity rule)

Rule 5. The utterance (or implication) counts as an attempt to instill fear. (The essential rule)

Demanding

Searle (1969) provides a list of the felicity conditions for a request. By adding one

preparatory rule, the request turns into a demand. The preparatory rule that needs to be added concerns the power relation between the speaker and hearer. In the case of a demand the speaker must have power over the hearer. The felicity conditions for a demand are:

act A of the hearer H.

(The propositional content rule)

Rule 2. The speaker S believes H can do A, and it is not obvious to both S and H that H will do A in the normal course of events.

Rule 3. S has power over H (The preparatory rules) Rule 4. S wants H to do A (the sincerity rule)

Rule 5. The utterance counts as an attempt by S to have H do A. (The essential rule)

Blackmail

In my term paper I have shown that the felicity conditions for blackmail are less than the sum total of the felicity conditions for the three speech acts that together make up

blackmail. The total number of felicity conditions for the three separate speech acts (threatening, demanding and promising) amounts to fifteen. However, eight felicity

conditions suffice for blackmail. For example, the promise in blackmail does not need to be felicitous. Thus the sincerity condition for promise can be omitted. Furthermore, in the case of blackmail, there is only one essential rule (rule 7), while the three speech acts each has their own distinct essential rule. The felicity conditions for blackmail are as follows:

1. The utterance/actions predicate that the blackmailer B will share certain information A with a third party C.

Or, the utterance/actions of B create the implication that B will share A with C. 2. The hearer H does not want A to get to C.

3. H believes that B is capable of sharing A with C. 4. B predicates a future action D of H,

Or otherwise implies a future action D of H. 5. B believes that H is capable of doing D. 6. B would prefer H doing D to H not doing D.

7. The action of B counts as an attempt to get H to do D. 8. B implies that if H does D, B will not share A.

(Arends, 2017, p. 18-19)

2. Theoretical Framework 2.1 Classical Speech Act Theory

Both Searle (1969) and Austin (1962), the two founding fathers of speech act theory, have argued that the central function of language is to deliver speech acts. According to Austin (1962) language studies have been focusing on theories of meaning and theories of truth and lost track of what people actually use language for, namely to do things, to accomplish their goals. Instead of treating language as a collection of propositions that are either true or false and study their truth conditions, Austin (1962) suggests we view utterances as actions that can fail or succeed. Austin (1962) distinguishes between three different ways in which an utterance can be said to perform an act. The first he deemed the ‘locutionary act’, which is the act of saying the words with their literal meaning. The second way in which utterances can be considered as acts he called the ‘illocutionary act’. The illocutionary act is the action that the words are intended to perform, such as ask a question or make a request. The illocutionary act is basically the speech act. The third act is called the ‘perlocutionary act’, which is the actual effect the words have on the people who hear it. The perlocutionary act also includes any unintended effects the utterance might have, such as a question causing annoyance because the answer has been previously provided on multiple occasions and the speaker keeps forgetting. In the traditional analysis of speech acts the

perlocutionary act is often ignored since it is beyond the speaker’s control. However, in conversation analysis, the reactions of the hearers are deemed crucial for identifying and interpreting the speech act that solicited the response.

One of the aims of speech act analysis is to determine under what conditions speech acts succeed. Austin (1962) named these conditions ‘felicity conditions’. These conditions include appropriate subjective states and appropriate circumstances. Searle (1969)

systematized Austin’s concept of the felicity condition and identified four main types: the propositional content rule, the preparatory rule, the sincerity rule and the essential rule. The propositional content rule addresses the locutionary act of the utterance. It delineates what type of content the utterance must have for the speech act to be successful. However, this rule has proven to be problematic since it is possible for speech acts to be carried out through indirect speech. This means that speech acts can be performed with a wide variety of

utterances that have different kinds of propositional content. Furthermore, there are idiomatic means by which to express speech acts which bypass the propositional content rule. Gordon and Lakoff (1971) have noted that it is possible to communicate a speech act indirectly by including one of the felicity conditions for the speech act in the propositional content of the utterance. For example, when a person tells a waiter that they would like a cup of coffee, they are expressing one of the preparatory conditions for the speech act of requesting, namely the speaker's preference towards having coffee. The preparatory rule states what kind of

circumstances would be deemed appropriate for the speech act, which also can include certain subjective states of the addressee. In the case of a promise, it is important that the recipient desires what the speaker promises. The sincerity rule and the essential rule describe what kind of subjective state the speaker must be in for the speech act to be felicitous.

2.2 Conversation analysis and identifying speech acts

While speech act theory has its roots in the philosophy of language, conversation analysis (CA) emerged in the sixties as a subfield of sociology. Conversation analysis is closely related to ethnomethodology, which is “an effort to study the methods in and through which members concertedly produce and assemble the features of everyday life in any actual, concrete…setting” (Maynard & Kardash, 2007, p. 1483). The main idea behind

ethnomethodology is that experience is not chaotic, but rather has an intelligible structure which is provided by the actors, and there is no need for prior description or generalizations to discern the structure (Maynard & Kardash, 2007, p. 1483). Conversation analysis applies this idea to language in interaction. The participants in a conversation construct the activity they are participating in together by “understanding and making sense of the actions initiated by the other participants as well as his/her own” (Gonzalez-Lloret, 2010, p. 59). This is in contrast to traditional speech act analysis where the speaker performs the speech act on his own, and the hearer is nothing more than a passive recipient. The difference arises because traditional speech act theories are based on a mentalist view of knowledge, and thus interprets conversations as encoded messages that are "transmitted from speaker to hearer through language" (Gonzalez-Lloret, 2010, p. 61). CA, on the other hand, maintains that meaning in encounters is “socially and sequentially constructed by the participants of the interaction” (p. 61). One of the strengths of CA is that it can identify speech acts without having to refer to speakers intention, which is often not present in the data, and as a result involves mostly guess work.

Another concern of speech act analysis is the identification of speech acts. The issue with the identification of speech acts is twofold. Firstly, as already discussed, speech acts are often in the form of implications. Secondly, there is often no one-to-one mapping between an utterance and a speech act. It might be possible that one speech act is performed across

multiple utterances and oftentimes the reverse is the case, one utterance performs multiple speech acts. Consider the following example presented by Levinson (2016) in which a young girl negotiates her allowance with her mother:

Virginia: But- you know, you have to have enough mo:ney¿ I think ten dollar’sud be good. ←Proposal (0.4)

Mom: .hhh Ten dollahs a week? ←Repair-I, Q, pre-challenge Virginia: Mm hm. ←Repair, A, Go-ahead

Mom: Just to throw away? ←Repair-I, Q, Challenge, Pre-rejection (0.5)

Virginia: Not to throw away, to spe:nd. ←Repair, A, Defense

Most of the utterances in this exchange serve multiple functions. The mother’s answer to her daughter’s proposal signals that she believes her daughter is asking for too much, since the prosody expresses incredulity, while also requesting clarification regarding the frequency of the allowance (p. 207). Her daughter responds by answering her mother’s question in the affirmative and simultaneously giving her mother the go-ahead. The mother, in turn,

continues by challenging her daughter’s proposal, signaling that she will reject it, and asking for further clarifications regarding what her daughter would do with the allowance.

Levinson (2016) discusses four different approaches that are used to identify speech acts. One could identify speech acts by relying on natural language. In natural language, there are many words for speech acts such as, promise, declaration, greeting, invitation, rejection, etc. However, the list is not exhaustive; speech acts have been identified that do not have a vernacular name (p. 203). Furthermore, there might be languages with unique speech acts, but no name for these speech acts. Thus, "natural language terms are a poor guide" (p. 203). A second approach is to use the felicity conditions of speech act to define identify them. This

method also has its issues since it is difficult to come up with the right set of felicity conditions for each speech act. Furthermore, when felicity conditions are used to identify speech act subtle differences between speech acts that fall into the same family of speech acts will be overlooked (p. 203).

Another way of characterizing speech acts is by using the responses elicited. This is the preferred approach in CA (Levinson, 2016, p.204). If different utterances elicit similar responses, then all those utterances would be categorized as the same speech act. The idea supporting this approach is that speech acts often come in ‘adjacency pairs'. Adjacency pairs are a pair of utterances, one for each speaker, that are related to each other in such a way that the initial utterance ‘basically determines what the next person can say' (Jones, 2012, p. 57). This type of relationship is one of ‘conditional relevance,’ which means that the meaning of each of the utterances is dependent on the other.

Schegloff and Sacks (1973) first introduced the term ‘adjacency pair’ in their analysis of the structure of closings. Each adjacency pair has a first pair part (FPP) that initiates an exchange and a second pair part (SPP) that is responsive (Schegloff, 2007, p. 13). However, an adjacency pair cannot consist of a combination of any FPP and any SPP. The FPP “creates a ‘slot’ for the next person to ‘fill in’ in a particular way” (Jones, 2012, p. 57). The second speaker has a limited amount of choices available that corresponds to the FPP, and out of these options, some SPPs are expected responses to the FPP. These SPPs are deemed preferred responses. Examples of speech acts that come in adjacency pairs are, ‘question/answer,' ‘greeting/greeting' and ‘offer/acceptance or rejection'.

Levinson’s fourth approach is very closely related to the third and therefore also reliant on the underlying assumption of conversation analysis that conversations are orderly (Jones, 2012, p. 57). This approach relies on the sequential position the utterance occupies in an interaction to identify what kind of speech act the utterance performs.

When attempting to determine the speech acts, it is important to use a combination of these four approaches. If there are more arguments to support the interpretation of an

utterance as a particular speech act, then there is more chance that the interpretation is correct.

Even though it is hard to identify speech acts in academic analysis, people in

everyday conversation do not seem to have any problems with interpreting the speech acts of the people they converse with. Of course there are instances of miscommunication, and people do not always understand each other correctly, but given the amount of speech acts people have to process every day (approximately 5000; Levinson, 2016), and the small amount of time they have before they need to give an adequate response (about 200 ms; Stivers et al. 2009), they do an excellent job. While there is variation across languages and cultures regarding the average gap between turns of verbal interactions, the gap does not last for more than half a second in general. The range of variation is 250 ms from the mean (Stivers et al. 2009), which seems like hardly enough time to consciously think about what the other person said, logically arrive at a conclusion about what the intended meaning of the utterance is, which is often not directly reflected in the propositional content of the utterance, and prepare a response. Thus universally people can subconsciously interpret speech acts rapidly and accurately, and one of the tasks of speech act theory is to spell out this process.

2.3 Types of Speech Acts

Another puzzle in the field of speech act theory regards the types of speech acts there are. Is there a limited set of speech acts or can a new speech act be performed on the spot? According to Austin (1962), speech acts are based on social conventions and these very across time and geographical regions. If Austin (1962) is correct, there could be an infinite amount of different types of speech acts. However, since speech acts are based on cultural conventions, a single speaker would not be able to perform a new speech act on his own.

Novel speech acts would have to emerge organically from conventions of social encounters. Searle (1962), on the other hand, differentiates between five general categories of speech acts: Assertives, Directives, Commisives, Expressives and Declarations. All speech acts that involve telling someone about a given state of affairs (e.g. The Red Sox won their game) would be categorized as Assertives. Directives are speech acts that try to get people to do things (e.g. Open the window, please). A promise would fall under Commisives, since the speaker commits themselves to doing something. Expressives are speech acts where the speaker expresses how they feel. When parents’ names their newborn child, they change something in the world. The child was nameless before her utterance, but their utterance gives the child a name. Such a speech act would be considered a Declaration.

There are several issues with Searle’s (1969) method of categorizing speech acts. Firstly, conversation analysts have identified many speech acts (Pre-closings and repair initiators) by looking at data of naturally occurring language that do not fit into any of Searle's five categories (Levinson, 2016). Secondly, Searle's categorization does not take cultural variation into account (Levinson, 2016). Different cultures have speech acts that defy Searle's (1969) categorization. The approach in conversation analysis is not to presuppose what types of speech acts there are, but rather let the data show what the function of the utterances is. Be that as it may, it might be the case that the data will not solve the issue of categorizing speech acts. The theoretical framework with which speech acts are analyzed to a large extent determines the conclusion that will be drawn from the data, and there are critical differences between classical speech act theory and speech acts in conversational analysis. For instance, speaker’s intent, which is crucial in classical speech act theory, is mostly not touched upon by conversational analysts.

A different way to categorize speech acts would be to distinguish between lower-order and higher-lower-order speech acts. Grice (1989) made this distinction. He writes, “speakers

may be at one and the same time engaged in performing speech-acts at a different but related levels” (Grice, 1989, p. 362). According to Grice, lower-order speech acts would include “saying, telling, asking” and higher order speech acts involve “adding, contrasting, concluding, which comment on the lower-order ones” (Ifantidou, 2001, p. 51). However, Ifantidou (2001) disagrees because to her “it seems pretty clear” that contrasting and

concluding are not examples of higher-order speech acts since they are analyzed by Griceans as basic speech acts (p. 52). I wish to propose another way to distinguish between higher and lower-order speech acts. All speech acts that can be performed in a single turn-at-talk would be categorized as lower-order. These would include most of the speech acts identified by CA, such as repair-initiators, pre-closings, etc. Speech acts spanning over multiple turns at talk would be deemed higher-order speech acts.

For example, the speech act of saying goodbye can consist of multiple utterances, each performing ‘smaller’ speech acts, as in the following example:

Johnson:…and uh, uh we’re gonna see if we can’t uh tie in our plans a little better. Baldwin: Okay//fine.

Johnson: ALRIGHT? Baldwin: RIGHT. Johnson: Okay boy. Baldwin: Okay. Johnson: Bye//bye. Baldwin: G’night.

(Schegloff & Sacks, 1973)

The whole sequence from Johnson’s “ALRIGHT?” to Baldwin’s “G’night” could be described as the act of saying goodbye. Johnson’s “ALRIGHT” functions as an initiation of the closing section, since it is an invitation to close the conversation, and Baldwin accepts the

invitation in the subsequent utterance. The ensuing ‘Okay boy/okay’ is a pre-closing

adjacency pair, which in turn is succeeded by proper closings. Given these points, the act of saying goodbye can be seen as a higher order speech act, consisting of multiple lower order speech acts. Similarly, blackmail may be a higher order speech act.

Yet, it could be argued that only the closing adjacency pair in the example involves saying goodbye, and the utterances leading up to saying goodbye are to be treated as separate speech acts since they serve a different purpose. Therefore, it might be the case that the example of saying goodbye was not convincing enough. In that case, consider the ‘knock knock’ joke. It is widely accepted that joking is a speech act (Jones, 2012, p. 64). In order to identify a joke in a conversation, CA would rely on the response the joke elicits to categorize it as such. Responses that would be solicited from a joke are laughter if the audience found the joke enjoyable, or a remark such as, “that is not funny,” when they do not find it

particularly amusing, whereas, a ‘knock knock’ joke solicits a different type of response from the other participant. In the ‘knock knock’ joke tradition one person (the person with the intention to tell the joke) initiates the joke by uttering, “knock knock,” and the other

participant responds by asking, “who’s there?”. The person telling the joke would answer the question with an utterance that is supposed to be interpreted as a first name, and the other participant would inquire into the person’s last name by repeating the utterance and adding ‘who?’ Only then would the ‘joke teller’ deliver the punch-line.

The reaction solicited by the punch-line would reveal that the preceding utterance is a joke. However, that single utterance does not constitute the complete joke. Even though a single speaker initiates the telling of the joke and delivers the punch-line, the joke actually emerges from the interaction between the two participants. A ‘knock knock’ joke cannot be told if the other participant does not cooperate. The speech act of joke telling in the case of a ‘knock knock’ joke would include four utterances, spanning over four different turns-at-talk,

by two different participants. Kasper (2009) notes that, meaning in an interaction results "from participants' ongoing, contingent interpretive work during jointly pursued practical activities" (p.36). Something similar might be in effect in the case of blackmailing. Unlike ‘knock knock’ jokes, blackmail does not have a set structure of paired responses. It would be possible for someone to blackmail someone else in a single turn-at-talk, without any

interaction from the victim. For instance, a single e-mail or text message can constitute the whole act of blackmailing. Be that as it may, even in cases where one text message is

considered as blackmail, it may rely on previous exchanges between the two parties involved. Consider the following message that was sent to a middle school teacher in Dallas by a student that she had had sexual intercourse with: "Anit [expletive] watch ima start getting the pics and everything ready ... to show the cops right I anit playing" (Rajwani, 2017). While the contents of the message makes it so that it would be unreasonable to interpret the message as anything but blackmail, the blackmail is incomplete as the demand is missing. The demand was probably made in an earlier message, or in another instance of communication. In any case, context is needed to interpret the message as blackmail. As readers of the message, we immediately imagine a possible context in which such a message may fit, and we base our interpretation on that. Even so, CA maintains that actions emerge in a “precise moment” and a “precise context,” which our imagination would only approximate (Gonzalez-Lloret, 2010, p. 60). Conversation analysts stress that it is important to look “at speech acts not as isolated utterances but rather as sequential evolving actions” (Gonzalez-Lloret, 2010, p. 62). It may be the case that certain speech acts are made up of multiple single utterance units. Thus,

blackmail, and the ‘knock knock’ jokes, should be treated as sequentially evolving actions. What makes blackmail more difficult to analyze, is that, unlike ‘knock knock’ jokes, blackmail is not scripted and does not have a definite beginning and end. In other words, the contextualization cue (Erickson & Shultz, 1981) in the case of ‘knock knock’ jokes is

obvious, since one person always initiates the joke with the exact same utterance, whereas in blackmail this is not the case. Contextualization cues are signals that give the other

participants in the interaction information about how the subsequent utterances should be understood (Erickson & Shultz, 1981).. Contextualization cues signal what type of

interpretive frame - a person’s understanding of how words and actions are to be interpreted in a given situation - a person should be using in the encounter. Goffman (1974) distinguishes between two kinds of frames. The primary framework is the set of expectations people have about “the overall activity” in which they will be engaged (Jones, 2012, p. 21). For example, the set of expectations people bring to a coffee date is different than those they would bring to an interview, which affects how they interpret the actions of others for the duration of the event. Even during each interaction, there is variation in regards to how utterances are supposed to be construed. Sometimes someone tells a joke or makes a sarcastic remark, and these have to be interpreted in another way than the other utterances in the interaction. These smaller changes occurring within an interaction are called interactive frames. Due to the dynamic nature of interactive frames, contextualization cues are “particularly important in the role they play in helping us manage and negotiate” them (Jones, 2012, p. 21).

2.4 Power

In my term paper (Arends, 2017) I briefly explained the difference between a request and a demand. The difference lies in the third felicity condition for demanding (S has power over H). However, I did not elaborate on the type of power in question, nor did I give a precise definition of power, or any way to measure power, other than stating that the speaker in the case of a demand must be able to penalize the hearer if the action is not carried out (Arends 2017, p. 14). It appears that “few linguists explicitly discuss the conceptual nature” of power (Spencer-Oatey, 1996, p.22). Furthermore, Kucherenko (2016) has shown that it is

common in communication studies to use "power as a self-explanatory concept," so I am not alone in making this oversight (p. 97).

Nonetheless, in order to conduct a proper analysis, we must be able to identify a demand and to do that we need to define power and find out how it is exercised or discernible in interactions. It appears that in the social sciences the debate on power is still ongoing, and there are as yet no definite answers to questions such as, “what it is, where it is located and how it can be analysed” (Davis et al., 1991, p. 7 & Thornborrow, 2002, p.5). According to Locher (2004), “power can be exercised in any interaction involving two or more

interactants” (p. 9). The Foucauldian concept of power, likewise, proposes that power can be constructed within social interactions (Thornborrow 2002, p. 15). Power in this sense is relational, contestable, dynamic and contextual, since power structures can change over time and the participants in an encounter can attempt to seize additional power, and some succeed (Locher 2004, p. 39). The question that arises is: how does power operate in social

interactions?

Fairclough (1989) presents different devices that can be used by the powerful participant in an interaction to deprive the subordinate participant of the chance to freely contribute to the conversation (p.135). The four tools Fairclough (1989) introduces are: Interruption, Topic control, Enforcing explicitness and Formulations. Even though Fairclough (1989) composed this list while analyzing spoken discourse in institutional settings, I would argue that these devices can also be used outside of institutional discourse. One of the main features of institutional discourse is that it involves asymmetrical participant roles in forms of speaker/hearer rights and obligations. Yet, this feature is not unique to institutional discourse. Even outside of institutional settings it is often the case that power is not equally distributed amongst the interactants. Another way in which power relationship might be noticeable in interactions is by what kind of ‘face’ strategies each participant uses.

According to politeness theory, in hierarchal encounters the subordinate participant would likely use independence strategies, while the dominant participant would more likely use independence strategies (Brown & Levinson, 1987). Involvement strategies involve using informal language, first names, being direct and talking about ‘us,’ whereas independent strategies are characterized by hedging, formal language, using titles and apologizing (Jones, 2012, p. 21). Identifying which participant’s speech includes more elements of involvement strategies and which participant makes the most use of Fairclough’s devices helps in

establishing what type of power structure is at play in an interaction.

Dahl (1957) provides the following definition of power: A has power over B to the extent that A can get B to the something that B would not otherwise do (p. 202-203). Bachrach and Baratz (1970) propose three conditions that are necessary for there to be a power relation: “a) there is a conflict over values or course of action between A and B; b) B complies to A’s wishes; and c) B does so because he is fearful that A will deprive him of a value or values which he regards more highly than those which would have been achieved by noncompliance” (p.24). These three conditions expand on Dahl’s (1957) definition of power, by providing context and explaining why B would submit to A.

The issue with Bachrach and Baratz’s (1970) and Dahl’s (1957) definition of power is that they maintain that A’s power is only recognizable through B’s behavior. Locher (2004) notes that various scholars in the field of power have deemed such definitions too narrow. One of the reasons why these type of descriptions are not sufficient is that they fail to account for situations in which B resists A’s influence, while A has power over B. Furthermore, these notions of power do not take into account the pressure that is put on B by A before B decides to act. Bachrach and Baratz’s (1970), however, describe two types of power that do not rely on which course of action B takes, namely, latent power and potential power. While latent power hinges on wealth, status, and access to resources, potential power is identifiablewhen

“the recognition of the possibility of future sanctions results in ‘exercise’ of power in the present” (Bachrach and Baratz’s, 1970, p. 26). In the case of blackmail, the information that the blackmailer has about the victim gives them potential power. Alldridge (1993) notes that “the power of a secret rests with its potential revelation" (p. 368).

2.5 Identifying Implicit Speech Acts

Goffman (1959) observes, “many crucial facts lie beyond the time and place of interaction or lie concealed within it” (p. 1). Whereas the facts that lie beyond the time and place of the encounter are mostly found in the knowledge of the world and assumptions the participant share with each other, those that lie concealed within it can be located in implicit speech acts. To identify an implicit speech act in a conversation, we would need to

incorporate Grice's (1975) cooperative principle (CP) and the four conversational maxims that follow: the maxim of quality, the maxim of quantity, the maxim of relevance, and the maxim of manner (p. 49). Grice (1975) maintains that conversations are at least to some extent "cooperative efforts," and that certain conventions are guiding what is deemed an appropriate contribution to a conversation. These conventions are expressed in the four above mentioned maxims.

The maxim of quality holds that the contributions should be genuine, meaning that speakers should say what they believe to be true. According to the maxim of quantity, the contributions should be neither more nor less than is required. The maxim of relation

suggests that speakers make contributions that are appropriate to the situation. The maxim of manner states that the contributions should be clear and unambiguous. A speaker can flout one of the maxims in order conversationally implicate a meaning without having stated it explicitly. The hearer would notice that one of the maxims are being flouted, and would ask himself the following question: “how can his saying what he did say be reconciled with the supposition that he is observing the overall CP?” (Grice, 2007, p. 49). With the information

that the hearer has about the situation together with his general reasoning ability, the hearer should be able to work out the speaker’s intended meaning.

Be that as it may, Grice’s (1975) cooperative principle has been criticized, since it fails to account for the fact that social interactions involve both cooperation and conflict (Pinker, 2007, p. 833). This is certainly the case in the act of blackmailing. If a conflict is involved then how much influence do the maxims have on the contributions of the

participants? For instance, according to the maxim of manner clarity should be aimed for in communication, however, “clarity is only a measure of communicative competence, if the individual has as his or her goal to be clear” (Eisenberg, 1984, p. 230). In the framework of strategic communication, the strategic intent of the participants is what guides their

contributions, rather than the CP (Dulek & Campbell, 2012). The focus is “on achieving the sender’s predetermined aim” (Dulek & Campbell, 2012, p. 123). In this, it is similar to speech act theory, where speaker’s intent takes center stage. Nonetheless, the strategic

communication framework does not give us the necessary tools for identifying implicit speech acts. For that, we still need Grice (1975).

Nevertheless, how can we use a tool that relies on purposely flouting the maxims of the CP at opportune moments to convey hidden messages in an interaction where some of the maxims, such as the maxim of quality (e.g. the speaker lies about finding a note in a

blackmail attempt), is not adhered to by one of the participants, and rather than cooperating one of the participants is trying to deceive and coerce the other? Firstly, Grice (1975) notes that the participant do not have to share the same ultimate aim in a conversation to adhere to the CP, even a common immediate goal in an interaction such as identifying themselves "with the transitory conversational interest of the other," would suffice (p. 48). Secondly, even though the speaker misleads the hearer by lying about the note, for the most part, they

adhere to the CP since the speaker needs to get his message across to achieve his goal in the encounter, and the CP is essential to any form of communication.

2.6 Conditional Threats and Promises

The threat and promise in blackmail are conditional, meaning that they have an ‘If-then’ structure. According to Beller (2007) “conditional promises and threats are speech acts that a person uses to manipulate another person’s behavior" (p. 209), and this is precisely how they are put to use in the act of blackmailing. Beller (2007) explains how conditional promises and threats are related:

The speaker always announces (explicitly or implicitly) that he or she will react positively after the addressee has shown the desired behaviour, and negatively after an undesired behaviour. The promise focuses on the cooperative sequence, whereas the threat focuses on the non-cooperative one. In this sense, both schemas are

complementary, that is, the threat schema may also be read as a promise not to punish the addressee if this person cooperates (Beller, 2007, p. 213)

If a conditional threat can also be interpreted as a promise, does it make sense to separate these two speech acts? Four classical inferences are used to analyze people's understanding of conditionals:

Modus Ponens MP: ‘‘If P, then Q’’ and ‘‘P’’ – infer ‘‘Q’’

Denial of the Antecedent DA: ‘‘If P, then Q’’ and ‘‘not P’’ – infer ‘‘not Q’’ Affirmation of the Consequent AC: ‘‘If P, then Q’’ and ‘‘Q’’ – infer ‘‘P’’ Modus Tollens MT: ‘‘If P, then Q’’ and ‘‘not Q’’ – infer ‘‘not P’’

(Beller, 2005, p. 210)

Only the Modus Ponens and the Modus Tollens are logically valid inferences. Be that as it may, research has found that people draw not only the MP and MT inferences but also the DA and the AC inferences. What is more, these inference studies have found that people

make the DA and AC inferences more when presented with a conditional threat or promise then they would if they were presented with a universal conditional or a conditional tip or warning (Beller, 2005, p.210). One way to make sense of these findings is to assume that people are interpreting the conditional threat and the conditional promise as biconditionals. In biconditionals, all four of the classical inferences are logically valid. What further supports the idea that people are interpreting conditional promises and threats as biconditionals is that “most people accept the complementary conditional ‘‘If not-P, then not-Q’’ as following from inducements of the form ‘‘If P, then Q’’” (Beller, 2005, p.210). In other words, the promise is implicit in the threat and vice versa.

2.7 Indirect Speech and Politeness Theory

People rarely just “blurt out what they mean in so many words,” instead they use indirect speech (Pinker, 2007, p. 833). Indirect speech is a common feature of language; it might even be a language universal (Brown & Levinson, 1987). Indirect speech is used in a variety of situations. According to Jones (2012) “people don’t always say what they mean, and people don’t always mean what they say because language is by its very nature, ambiguous” (p. 3). Be that as it may, even when there is the option in language to be direct people do not make use of it and choose a less direct form of communicating instead. Why would people make this choice? One of the most widely accepted answers to this question in the field of pragmatics is provided by politeness theory (Terkourafi 2011). According to politeness theory, indirect speech offers the speakers the ability to get their message across without “constantly putting their relationships on the line” (Terkourafi, 2011, p. 2861).

Another reason why people might choose to be indirect and use implication instead of a more direct way of conveying their message, is to evade responsibility for what they say (Simpson & Mayr, 2010, p. 43). This can be quite useful for a blackmailer since the act of

blackmailing is punishable by law in most countries. If blackmailers are not explicit, it becomes more difficult for others to charge them with blackmailing, since indirect speech is often open to multiple interpretations. Moreover, implications are cancelable, which means that the utterance of a speaker can imply something, but the speaker can cancel the

implication by adding, “but I do not mean it like that” (Jaszczolt, 2009, p. 260).

Pinker et al. (2007) use game theoretical models to show how plausible deniability could be a reason why rational actors would opt for indirect speech in certain situations. The situation that is presented in Pinker et al. (2007) involves a driver who is pulled over for a ticket by a police officer and wants to bribe the officer into not giving him the ticket. If the driver is too direct with his offer he might get himself arrested by an honest officer. To avoid this outcome, the driver would use indirect language so that the officer cannot charge him with bribery because it would be cancelable.

2.8 Bluffing

When people enter a social encounter, they may have different intentions regarding how they want the other people present to perceive them. A person may want others to think well of them, or to think that they think well of them, or they may wish to expose their honest emotions, yet, it is also possible that they want others “to obtain no clear-cut impression” (Goffman, 1959, p. 3) regarding how they feel, and they might even want to mislead them, as is the case in bluffing. Goffman states that “regardless of the particular objective which the individual has in mind and of his motive for having the objective, it will be in his interest to control the conduct of the others, especially their responsive treatment of him” (p. 3). Someone can try to accomplish this by acting “in a thoroughly calculating manner,

expressing himself in a given way solely in order to give the kind of impression to others that is likely to evoke from them a specific response he is concerned to obtain” (p. 6). In the case of bluffing the aim is for the other person to believe the bluff and back down, instead of

calling the bluff. A bluff is a type of lie, one that is used strategically to make one appear stronger than one actually is. Thus, it involves deception by a false show of confidence.

Yet, a person is not in total control of all the information others receive from them. The false show of confidence could be uncovered. The participants in an interaction receive impressions from each other via expressions (Goffman, 1959). Goffman distinguishes between two types of expressions people produce in social situations. There are expressions people give, which include all the information that the person intends to convey to the other participant(s), and there are expressions that are unintentionally given off, which are

“accidental forms of information transmission” (Wilson & Wharton, 2006, p. 1560), and a person giving this information off might not be aware that they are doing so.

Wilson & Wharton (2006) make a similar distinction between signs and signals, terms they borrow from Hauser’s (1996) study on animal communication: “Natural signals, like linguistic signals, are genuinely coded and inherently communicative; natural signs, by contrast, are interpreted by inference rather than decoding, and are not inherently

communicative at all” (Wilson & Wharton, 2006, p. 1561). To illustrate the distinct role these two concepts play in human communication Wilson & Wharton (2006) asks their readers to consider the difference between shivering and smiling. Both shivering and smiling convey information about a person’s state of being. Shivering is a physical reaction to low

temperatures, and thus a person could infer that someone who shivers feels cold, yet, this is not the purpose of shivering. Thus shivering is an example of a sign. In contrast, smiling evolved specifically for its communicative function (Wilson & Wharton, 2006). Smiling is a signal. Whereas one can intentionally smile, one could not intentionally shiver. Wilson and Wharton (2006) more or less equate signs and signals to showing and meaning in language and argue that no aspect of communication falls under either one or the other, but are rather located on a showing-meaning continuum. The primary difference between these pairs of

contrasting terms is that one (give, signal and meaning) is intentionally communicative while the other (give off, sign and showing) is coincidentally so.

Goffman (1959) observes, “it is plain that many performers have ample capacity and motive to misrepresent the facts” (p. 58). Someone who bluffs makes use of this capacity to construe a situation in which the other(s) would think that he is in a position of advantage, while this is not the case. The bluffer presents a false front. Goffman (1959) notes that

“When we think of those who present a false front or 'only 'a front, of those who dissemble, deceive, and defraud, we think of a discrepancy between fostered appearances and reality. We also think of the precarious position in which these performers place themselves” (p. 59).

A bluffer is in a precarious position since others could catch him in the act. In any encounter people seek to acquire information about each other (Goffman, 1959). They are not motivated only by pure curiosity, since the information can have practical value and this would be of interest to them. It helps them to make sense of the situation, and enables them to know what to expect of them, and what they might expect in return. It is in the interest of someone who is being bluffed to find out the truth about the situation. People do not limit themselves to the information others voluntarily choose to give, but rather they make use of their inference skills to interpret what people give off as well, the signs.

The term bluffing has its origins in 19th _{century American poker (Online etymology}

dictionary). It might be derived from the Dutch word bluffen, which means to brag, but it entered the English language through poker. In poker a player bluffs by betting while holding a weak hand. The player signals with his bet that his position is strong. Be that as it may, the bet alone does not suffice for a believable portrayal of strength, the person’s demeanor must show it as well; bluffing includes a performance. One of the ways in which poker players catch bluffs are by identifying ‘tells’. A ‘tell’ in the poker jargon refers to certain aspects of a

players behavior that provides information about the strength of the hand they are holding. These behaviors would be subconscious; hence players are not aware of what aspects of their behavior is telling. Thus tells are a kind of information the players unknowingly give off; they are signs.

To avoid giving off ‘tells’ a bluffer tries to make sure “that as many as possible of the minor events in the performance, however instrumentally inconsequential these may be, will occur in such a way as to convey either no impression or an impression that is compatible and consistent with the overall definition of the situation that is being fostered” (Goffman, 1959, p. 51). In poker, this means that the players try toning down all aspects of their behavior and try be as consistent as possible with their actions. If there is no variation, it becomes more difficult for their opponent to read them. Facials features are particularly salient at a poker table; the players are seated close to one another and are able to analyze each other’s face in great detail. Therefore, the ability to keep a straight face at a poker table is considered essential if one would want to be successful at the game, which is why a face that does not show any emotion has come to be referred to as a ‘poker face’.

It is not only in the poker world that people believe it is possible to recognize a lie based on subtle signals the liar subconsciously displays. Holm (2010) observes that this belief is widespread in fiction as well, and provides two examples to demonstrate his point. One is from Angels and Demons by Dan Brown, where Sir Leigh Teabing believes he has

recognized a tell. Teabing says, “Your poker face needs work my friend. It took me several seconds, but I can see now that you are lying.” (Brown, 2004, p. 553). As a second example, he mentions Raskolnikov, the protagonist in Dostoevsky’s (1866) “Crime and Punishment”. Raskolnikov is paranoid since he believes that a police officer can tell that he is lying.

Despite how widespread the belief in people’s ability to recognize truth and lies are, research has shown that people are poor lie detectors. A meta-analysis of 40 different studies has

shown that the average accuracy rate for lie detection was only 44 percent (Vrij, 2000). Nonetheless, this result does not change the fact that people still believe that others might be able to tell if they are lying, and adjust their behavior accordingly. Furthermore, when a person is faced with a bluff they are forced to make a decision, and this they make based on both the information that has been intentionally communicated to them and the information they acquire by interpreting the signs.

2.9 Silence in Dyadic Face-to-Face Interactions

All interactions are organized by turn-taking (Sacks, Schegloff & Jefferson, 1974). When one person is speaking the other will generally remain silent until it is their turn to speak. Yet, there are instances where neither person speaks; instances of silence. Not all silences are the same. Jensen (1973) observes that while we are “aware of the gift of

communicating via sound and visual symbols, few of us are fully sensitive to the reality that absence of sound – that is, silence – can also perform a number of highly significant

communicative functions” (p. 249). The major distinction to be made when it comes to silences in interactions is between communicative silences and non-communicative silences.

Saville-Troike (1982), for example, declares that “Silent communicative acts are to be distinguished from the pauses which may or may not occur in conversational turn-taking” (p. 144). She also states that “when silence has an intentional communicative function it may be analyzed as one of the forms a ‘speech’ act may take, and should be considered along with the production of sentence tokens as a basic unit of linguistic communication” (p.144). How is a person to know whether a silence that occurs in an encounter is communicative or not? According to Kurzon (1998) speaker’s intent is key. He asserts that “[f]or silence to have meaning in the linguistic sense, the speaker must have an intention, […] a meaning that may be expressed in words. But when the speaker has no intention behind his behavior, we may

refer to this silence as unintentional and therefore meaningless” (p. 8). However, as discussed earlier in section 2.2, speaker’s intent is not readily identifiable in the data, and as a

consequence, it is problematic to analyze. On the other hand, Bilmes (1994) suggests that social conventions regarding how interactions should proceed help determine which silences are communicative, when he writes, “where the rule is ‘speak’, not speaking is

communicative” (p. 74).

Bruneau (1973) distinguished between three types of silences: Psycholinguistic, interactive and socio-cultural. For this study only the first two types are relevant. Psycholinguistic silences are non-communicative. The instances of silence that are considered to be psycholinguistic are: Hesitations, stutters, sentence incompletions, etc. (Ephrat, 2014). Usually, there is no communicative intent behind these types of silence, but they might still give off information. According to Ephrat, psycholinguistic silences are instances of communication breakdown. In Lyon’s (1972) analysis of silence, a similar distinction is made of which he states that such incidences belong to performance, and have no communicative function. Interactive silences, on the other hand, “are highly bound to the nature of the message sharing process” (Bruneau, 1973, p.28). Bruneau notes that interactive silences are usually longer than psycholinguistic silences.

Bruneau (1973) further subdivides interactive silence into six different kinds, of which five are relevant for the current analysis: Decision-making, drawing inferences, exerting control, intense emotions, and maintaining or altering personal distance. Decision-making silences include the decisions regarding who talks, “questioning of the meaning of the previous speech” and “decisions about clarification or qualification of recent or past messages” (p. 28). Silences resulting from drawing inferences involve the time participants take to form “judgements about the many possible meanings of a message” and “about the character, motive, (and) personality” of the other participant (p.29). These judgments are

made based on inferences drawn. Silence can be used to exert control in various ways. Silence can not only be used to command attention, but it can also “create an ambiguous void in an interpersonal relationship – allowing others to either project inference or question their previous judgements about the relationship and thereby establish a new relationship” (p. 30). Bruneau (1973) explains that the structure of interactive silences in a conversation can bring the sort of “authority-subordinate” relationship to light (p. 31). According to him, “the initial burden of speech is often the burden of the subordinate” (p.31). Some intense emotions, such as “grief, sorrow, and great disappointments” (p. 34) can be shown with silence. Further, silence is a common reaction to outbursts of violent emotions, such as in an insult. The reason Bruneau (1973) gives for this is surprise. Finally, silence can also function as a means to maintain or alter interpersonal distance. Bruneau (1973) suggests that long silences can make words more meaningful and that by talking less a person could make their words so intense that they seem almost tangible.

3. Data

There are different reasons why gathering data regarding blackmail is troublesome. Firstly, blackmail is often conducted in private and contains sensitive information. People go to great lengths to cover up their interactions so that the contents do not spread beyond the concerned parties. Furthermore, blackmail is a criminal offense. For these reasons, the people involved, be it the victim or the blackmailer, will not gladly volunteer to record the

conversation on tape and make it available for research purposes. Therefore, I have chosen to conduct an analysis of a blackmail scene in the film Ides of March. However, since the data comes from fiction, I must rationalize my choice. According to Fairclough (1995) and Kosetzi (2008) media not only shapes our society but is also shaped by our society, since media aims to provide representations of the world. Media intends on showing

Unless a character has human qualities, it is not easy for us to relate to them. Both Stephens (1992), who works in the field of children’s fiction, and Intintoli (1984), a soap opera analyst, have stated “that for fiction to make sense to its audience, it must draw on some version of reality” (Kosetzi 2008, p. 229). Even though the encounter analyzed in this paper is scripted, the communicative strategies that are used by the actors to pursue their character’s goals are derived from strategies that are employed by real people in real situations.

3.1 Storyline

The protagonist of the film is Stephen Meyers, who is played by Ryan Gosling. He is employed by Governor Morris, played by George Clooney, as a junior campaign manager. Morris is the Governor of Pennsylvania, and he is running in the Democratic primary elections. Morris is portrayed as a smooth-talking, pro-abortion, anti-death sentence, ideal Democratic presidential candidate. He values integrity and wants a clean campaign. Morris’ senior campaign manager, Paul Zara, convinces Morris to fire Meyers for having met the campaign manager employed by Morris’ preeminent rival, Senator Ted Pullman, without first having informed his superiors. During the campaign, Meyers became sexually involved with one of the interns, Molly Stearns, who also happens to be the daughter of the chairman of the Democratic National Committee. One night, when Molly was asleep in bed beside him, Meyers picks up her phone by mistake and recognizes Governor Morris’ voice on the other end. Molly later tells him that she had a brief sexual affair with him and was now pregnant and that she needs money for an abortion. Meyers gives her the money and drives her to the abortion clinic, however, he does not return to pick her up. That same night she overdoses on painkillers. Meyers manages to sneak her phone into his pocket when he returns to her room.

3.2 The Scene

Stephen Meyers arranges a clandestine meeting with Governor Morris in the kitchen of a closed restaurant where he plans to use the information he has acquired during his

relationship with Molly to coerce Governor Morris into firing Paul Zara and giving him Zara’s position as senior campaign manager.

In the transcription, each turn-at-talk is designated with a number for ease of reference during analysis. Turn-at-talks are get a number instead of each line because the turn-at-talk is the unit of analysis in this paper. Turn-taking is crucial for the organizational structure of interaction and takes a central role in the study of ‘adjacency pairs,’ which in turn is

indispensable for the identification of speech acts. Furthermore, the analysis makes use of the reactions solicited by utterances for assessing what type of acts are being performed, and reactions happen over turns-at-talk rather than over lines of transcription.

Transcription

( ) : contextual information, nonverbal/paralinguistic behaviors

= : immediately adjacent utterances/ actions

[ : overlapping utterances/actions

 : rising intonation, often indicating question

BROTHER : emphasis through pitch, tone or volume

- : unmarked pause

(2 sec) : exact timing pause

M: Governor Morris (George Clooney)

S: Stephen Meyers (Ryan Gosling)

(1:18:45 – 1:23:35) (Beginning of scene)

 Morris and Stephen maintain eye contact almost throughout the entire conversation.  Both Morris and Stephen keep a straight face.

 Stephen never raises his voice, and all his utterances are almost devoid of tonal variation.

(Stephen walks into a barely lit restaurant kitchen after hours, in a slow pace, and tries to locate Governor Morris)

1.M: Seems like you have something you want to talk to me about=

S: (=turns his face and locates

Morris)

2.S:(4 sec) Does it seem that way (walks toward Morris, who is behind a stainless steel kitchen shelve full of cooking material, so not visible to Stephen)

3.M:(4 sec) You got something-or you think you have something you want to get off your chest, then why don’t you

4.S: (2 sec)(Still walking at a slow pace towards where Morris is standing)How about I do the talking and you do the listening (Stops walking as soon as he can see Morris, stands c. four steps away from Morris)

5.M: (3 sec)Okay

6.S: (1 sec) (walks towards Morris as he talks) As of tomorrow there's going to be a few changes in your campaign-Paul is out(Stops walking)-I am your senior campaign manager-I'll draft a statement-Campaign got to a point where we need to make some changes-you can put your own words in there=

7.M: =Why would I do that

8.S: -Because you want to win- Because you broke the only rule in politics-You want to be President, you can start a war, you can lie, you can cheat, you can bankrupt the country, but you CAN’T FUCK the interns-they'll get you for that

9.M: (3 sec, smiles)What do you think you have Stephen A troubled young girl tells you a story=

10.S: =A troubled young PREGNANT girl=

11.M: =Is that what she told you

12.S: Who needed cash for an abortion=

13.M: =What did you do, give her money Maybe she just

needed cash and you were the perfect guy to hit up 14.S: -(slightly tilts his head down and looks Morris in the eye)

S: Is that your best play=

15.M: =You need a job that bad Stephen

M: You come in here with your DICK in your hand, you've got NOTHING.

16.S: Then how did I get in here(2 sec) You're right Mike, there's nothing. There is no voice messages, no text, no e-mails, no pictures, no tape, and YET-I'm standing right here

17.M: Well go home (Morris turns away and prepares to leave)

18.S: (1 sec) She left a note

19.M: (Suddenly lifts his face and looks Stephen back in the eye)(9sec) How would you know that

21.M: (1 sec)I thought it was an accident

22.S: I don’t know Mike, You are going to have to tell me, you’re the last person she S: Note says she’s pregnant, got an abortion-

S: and doesn’t want to hurt you=

23.M: =Then why would she leave a note that could ONLY do

that

24.S: (Shrugs) She's twenty

25.M: (3 sec) What do you want

26.S: Paul's gone-today-I set a meeting with you and Thompson and you promise him the ticket-You get three hundred and fifty-six pledged delegates-you get North Carolina-you take office-and you make right what so many people have made wrong-All the things we both believe in

27.M: Well I don't believe in extortion, I don't believe in tying myself to YOU for the next EIGHT years=

28.S: =Four years-let's not get ahead of ourselves

29.M: (3sec)huh(5 sec) What were you doing in her room

30.S: Housekeeping, before you made page one=

31.M: =(shakes his head) no, you were fired, you

were out, what were you doing in her room

32.S: I was fixing a situation-I got a call, someone from the hotel thought I could help and they called me

33.M: (Raises his voice) So you were off the campaign, but you thought it was important to FIX THINGS (5 sec, half smiles)It was your voice the other night on the phone-you were fucking her-what did you do Stephen

34.S: (4 sec)I loaned her some money and I gave her a ride

35.M: I BET you did(2 sec)(swallows)There's no note

36.S: Okay, Mike, there's no note=

37.M: =Show it to me

38.S: You think I brought it here(1 sec) That's how people get hurt

39.M:(4sec)(swallows, exhales through his nose)(2sec)Okay-let's play this out(1 sec)You're not wearing a wire because everything you've suggested is illegal-and you wouldn't fare too well in federal prison-you were fucking her and she told you about her situation-there are no

records or conversations between the two of us-and since she's had an abortion, no DNA evidence-So-if there's no note, then it's your word against mine-YOUR word, A FIRED, DISSGRUNTLED, EMPLOYEE, or a sitting governor (4 sec) There's no note

40.S: (6 sec)It's your call Governor. (End of scene)

4. Analysis

I want to begin this section with some general remarks regarding the participants’ overall demeanor throughout the interaction and suggesting what role this might play in helping them achieve their interactive goals. As noted before introducing the transcription, the participants maintain eye-contact for almost the entire duration of the interaction.

Maintaining eye-contact can be a sign of strength and confidence, while looking away can be seen as a sign of weakness. Since gaze is not one of the central concerns of this paper, we will leave it at that, and interpret sustaining eye contact in this scenario as a form of holding one’s ground. It was also noted that both participants keep a straight face during the

conversation. In section 2.8 it was discussed that poker players do this to minimize the amount of information they unknowingly and unwillingly give off to their opponents. The participants in this interaction employ a ‘poker face’ for similar reasons. Meyers does not want to be caught bluffing, and Morris does not want to show weakness since it would be a liability. In other words, both Meyers and Morris put up a front to hide any insecurity they might feel that could betray the strength of their position.

Furthermore, while Meyers goes to the meeting with a plan and knowing what to expect, Morris does not. Nonetheless, Morris brings certain expectations to the meeting; he has a specific interpretative framework which he would use to make sense of Meyers’ actions. Various features of their encounter suggest that Meyers has wrongful intentions: they meet alone and at night in an odd place. With this in mind, Morris can already begin to suspect blackmail even before his encounter with Meyers.

4.1 A Struggle for Dominance (turn 1 – 5)

In the first few exchanges, both participants attempt to take control of the

conversation. It is a contest for power, which is eventually won by Meyers, when he says, "How about I do the talking and you do the listening," (turn 4) and Governor Morris

concedes with his reply, "Okay" (turn 5). It could be said that Meyers has won the contest for power in the initial stages of the conversation because the first speech act that solicits a ‘preferred response' is his. What the two interactants appear to be doing in the initial stages of their encounter is renegotiate their relationship. Meyers used to be employed by Governor Morris, which means that they were in a hierarchal relationship, with Meyers being subordinate to Morris. However, Meyers wants to be the dominant participant in this particular exchange and attempts to assert his dominion, while Governor Morris does not want to relinquish his power.

The preferred response to a request is compliance, yet when Morris starts the conversation by asking for clarification regarding the nature of their meeting, Meyers' responds with the question, "Does it seem that way?" (turn 2). The question challenges the observation made by Morris and engages with the literal meaning of Morris’ utterance, which is the locutionary force of the utterance, instead of with Morris' intended purpose, which is the illocutionary force. In conversations, it is usually the illocutionary force that people respond to. Morris returns the favor by giving an inadequate response to Meyer's question. The preferred response to a question is the expected answer. Instead, Morris responds by issuing a command. However, Meyers does not back down, and instead of complying with the demand, which would be the preferred response, Meyer's ignores the demand and offers a suggestion of his own. Morris finally acquiesces, and the string of adjacency pairs without preferred responses comes to an end when Morris accepts Meyer's proposal by saying, "Okay." However, before giving in and agreeing to Meyers’ suggestion Morris takes a

moment of silence. One of the five functions of silence in interactive communication which was discussed by Bruneau (1973) was to draw inferences regarding the meaning of previous utterances and the motives of other participants. It seems that Morris takes the opportunity to assess the situation before giving his reply.

Governor Morris is the first to talk, and thereby he tries to assert that he is in control of the conversation. Talking first in an encounter gives the speaker different types of control. The first is that the first speaker decides the direction of the exchange, which is topic control. Secondly, the initial utterance restricts the type of responses the other participant can make. Furthermore, by speaking first, it may seem as if the first speaker has power over the second speaker since the utterance solicits a response from the other participant. On the contrary, Bruneau (1973) maintains that the subordinate participant usually speaks first. Governor Morris initiates the conversation by saying, "Seems like you have something you want to talk to me about" (turn 1) which is not only a formulation but also an instance of enforcing explicitness since with this utterance Morris is trying to get Meyers to tell him what their meeting is going to be about.

Meyers resists Governor Morris’ attempt to take charge of the situation by taking a long pause. Meyers’ silence is communicative since he breaks the social conventions by not giving an immediate response. Bilmes argues: “where the rule is ‘speak’, not speaking is communicative” (1994, p. 74), and the rule here is definitely to speak. By not responding immediately, Meyers signals that he himself decides when he speaks, and not Morris. Another interpretation of his silence is that he is using it to add weight to his words and utterances. Bruneau (1973) also postulates that silence can be utilized as a means to exert control, especially when it comes to renegotiating relationships. Furthermore, instead of giving the response Morris was trying to elicit, which would be to cut to the chase and tell Morris what he came to talk about, he responds with an empty question, "Does it seem that