Psychology Journal of Language and Social

Journal of Language and Social Psychology
http://jls.sagepub.com Identity Implications of Relationship (Re)Definition Goals: An Analysis of Face Threats and Facework as Young Adults Initiate, Intensify, and Disengage From Romantic Relationships
Steven R. Wilson, Adrianne D. Kunkel, Scott J. Robson, James O. Olufowote and Jordan Soliz Journal of Language and Social Psychology 2009; 28; 32 originally published online Nov 17, 2008; DOI: 10.1177/0261927X08325746 The online version of this article can be found at: http://jls.sagepub.com/cgi/content/abstract/28/1/32
Published by:
http://www.sagepublications.com
Additional services and information for Journal of Language and Social Psychology can be found at: Email Alerts: http://jls.sagepub.com/cgi/alerts Subscriptions: http://jls.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations http://jls.sagepub.com/cgi/content/refs/28/1/32
Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009
Identity Implications of Relationship (Re)Definition Goals
Journal of Language and Social Psychology Volume 28 Number 1 March 2009 32-61 2009 Sage Publications 10.1177/0261927X08325746 http://jls.sagepub.com hosted at http://online.sagepub.com
An Analysis of Face Threats and Facework as Young Adults Initiate, Intensify, and Disengage From Romantic Relationships
Steven R. Wilson
Purdue University
Adrianne D. Kunkel
University of Kansas
Scott J. Robson
Fort Hays State University
James O. Olufowote
Boston College
Jordan Soliz
University of NebraskaLincoln
Identity implications theory (IIT) is applied to analyze how young adults manage identity concerns associated with the goals of initiating, intensifying, and disengaging from romantic relationships. Participants wrote their responses to one of six hypothetical romantic (re)definition scenarios, indicated whether they actually would pursue the relational goal if their scenario were real, and rated degree of threat to both parties face. Responses were coded for positive and negative politeness strategies. Participants in different relational goal conditions perceived different face threats, varied in their likelihood of pursuing the relational goal, and employed different politeness strategies. Relationship (re)definition goal also moderated associations between perceived face threats and goal pursuit as well as politeness strategies. The findings show how multiple goal theories such as IIT can be applied to situations where relational goals are primary as well as how, to varying degrees, identity concerns shape and constrain how young adults pursue relational (re)definition goals. Keywords: romantic relationships; relationship goals; face threats; multiple goals; politeness strategies
Authors Note: We wish to thank Howard Giles and the anonymous reviewers for their insightful feedback on an earlier version of this article. Please address correspondence to Steven R. Wilson, Department of Communication, Purdue University, 100 North University Street, BRNG 2114, West Lafayette, IN 47907-2098; e-mail: [email protected]. 32
Wilson et al. / Relationship (Re)Definition Goals and Face Threats 33
lthough change may be a constant in relationships, some moments in the development or decay of romantic relationships are particularly memorable. Young adults describe events such as going on their first date, meeting their partners parents, or breaking up for a time as turning points that reflected significant changes in their relational commitment (Baxter & Bullis, 1986). When asked to imagine how they would feel when initiating, intensifying, or disengaging from a romantic relationship, young adults report they would feel excited, nervous, fearful, sad, and/or courageous (Kunkel, Wilson, Olufowote, & Robson, 2003). Perhaps, these varied emotions reflect that initiating, intensifying, and ending romantic relationships each are complex situations in which participants risk loosing face (Cupach & Metts, 1994) and must manage multiple, conflicting goals (OKeefe, 1988; Schrader & Dillard, 1998). One framework offering insight into the complexities of relationships (re)definition is identity implications theory (IIT; Wilson, Aleman, & Leatham, 1998; Wilson & Feng, 2007). IIT highlights the unique identity implications associated with seeking specific types of relational change. We compare what threats to face young adults associate with the goals of initiating, intensifying, or disengaging from a romantic relationship; what types of facework (i.e., means of managing both parties face) they employ; and whether associations between face threats and facework vary depending on the particular relational goal. We show how multiple goal theories can be applied in contexts beyond those involving instrumental goals and shed light on some of the microdetails of how relationship (re)definition is accomplished. To set the stage, we review prior research on initiating, intensifying, and disengaging from romantic relationships, describe IIT and explain its relevance to the current project, and forward research hypotheses.
Initiating, Intensifying, and Disengaging From Romantic Relationships

Initiating, intensifying, and disengaging from romantic relationships have been explored as turning points, memory structures, and relational goals. Young adults in Western societies share similar ideas about actions or events that represent meaningful changes in the course of romantic relationships. In their classic study of turning points, Baxter and Bullis (1986) interviewed 80 college students from 40 romantic relationships independently about all of the times that there were changes in the joint commitment level that you can recall (p. 477). Commonly recalled turning points included meeting for the first time, going on their first date, meeting their partners family, deciding to date exclusively, and breaking up for a period of time. Some turning points (e.g., dating exclusively, disengaging) typically prompted explicit talk about the relationship itself whereas others (e.g., going on a first date) did so only rarely.
34 Journal of Language and Social Psychology
Honeycutt, Cantrill, and colleagues (Honeycutt, Cantrill, & Allen, 1992; Honeycutt, Cantrill, & Greene, 1989; Honeycutt, Cantrill, Kelly, & Lambkin, 1998) have studied relational memory structures or sequences of prototypical behaviors that young adults expect to occur as romantic relationships escalate or decay. Relational memory structures allow participants to make predictions about likely future actions and inferences about the meaning of implicit events. Their research shows that meeting for the first time, asking for the others phone number, and going on a first date are actions that college students expect to occur early in romantic relationships, whereas meeting the partners parents and talking about dating exclusively are expected to occur later as relational commitment escalates. Talking about breaking up is expected to occur midway through the process of relational decay, after actions such as arguing about little things and spending less time together. Female and male students largely agree on how typical or necessary these actions are within developing or decaying relationships as well as when they are most likely to occur. Complimenting this work, scholars have also investigated relational (re)definition from the perspective of goals or desired end states that motivate participants actions (Kunkel et al., 2003; Mongeau, Serewicz, & Therrien, 2004; Schrader & Dillard, 1998). Actions such as asking someone on a first date or asking a dating partner to meet ones parents are meaningful, in part, because of what they signal about the current state and possible future trajectory of a romantic relationship. In their investigation of reasons for going on first dates, Mongeau et al. (2004) concluded that reducing uncertainty about the partner, investigating romantic potential, and creating or strengthening a friendship are popular first date goals for both men and women (p. 134). Asking someone to go on a date thus can be seen as an attempt to initiate a romantic relationship, whether the other party is a virtual stranger or a previous platonic friend. Asking a romantic partner to meet ones parents, thus further integrating the partner into ones larger social network, may be seen as a sign of wanting to intensify or escalate levels of relational commitment. Such actions are meaningful because of the goals projected to underlie them. In the goalsplansaction (GPA) model, Dillard (1990, 2004) distinguished primary and secondary goals. Within any interaction, the primary goal is what motivates a person to speak at that point in time and hence explains why the interaction is taking place. The primary goal brackets the situation. It helps segment the flow of behavior into a meaningful unit; it says what the interaction is about (Dillard, Segrin, & Harden, 1989, p. 69). Primary goals are not necessarily instrumental in nature; indeed, when asked to describe situations in which they sought change, respondents include not just instances of trying to change another persons behavior (e.g., task goals such as seeking assistance or enforcing obligations) but also situations where they were trying to change the fundamental nature of their relationship with the other party (Cody, Canary, & Smith, 1994; Rule, Bisanz, & Kohn, 1985). Wanting to initiate, intensify, or disengage from a romantic relationship each is an
example of a primary goal that provides a meaningful frame for participants to understand what is taking place as they interact. Secondary goals are cross-situational concerns that shape/constrain whether and how individuals pursue their primary goal. Dillard et al. (1989) proposed five secondary goal categories: (a) identity, (b) conversation management, (c) personal resource, (d) relational resource, and (e) arousal management goals. Schrader and Dillard (1998) had college students read a scenario illustrating the primary goal of relational initiation, escalation, or de-escalation; recall a similar situation from their own lives; and rate the importance of the primary goal and the five secondary goals in their recalled situation. In each case, mean ratings of goal importance were above the scale midpoint both for the primary goal as well as for 34 secondary goals. For example, conversation management goals, such as wanting to maintain face, were rated as more important than the primary goal by individuals who recalled initiating romantic relationship and just as important as the primary goal by individuals who recalled escalating or de-escalating a romantic relationship. These findings indicate that multiple, potentially conflicting goals typically are present when individuals attempt to redefine a romantic relationship. Although goals research has offered useful insights, important questions remain about the identity implications of pursuing relationship (re)definition goals. Most research has applied the GPA model to instrumental rather than relational goals, and the limited research that has investigated relational goals typically has focused on only a single turning point, such as first dates (Mongeau et al., 2004) or problematic events (Samp & Solomon, 1998), rather than comparing goals underlying distinct turning points. For example, although existing research suggests that concerns about face, or the conception of self that each person displays in particular interactions with others (Cupach & Metts, 1994, p. 3), are salient during attempts to seek relational (re)definition, goals research has not explained why unique threats to both parties face might be associated with the relational goals of initiating, intensifying, and disengaging from a romantic relationship. IIT addresses just this question.
Identity Implications of Relational (Re)definition

IIT builds on Dillards (1990, 2004) GPA model as well as Brown and Levinsons (1987) politeness theory. Like the latter theory, IIT assumes that face is composed of two basic wants: the desire to have ones attributes and actions approved of by significant others (positive face) and the desire to maintain autonomy and be free from unnecessary constraints (negative face). People assess their own behavior and the actions of interactional partner(s) in terms of what is implied about both parties, in part, because many actions have the potential to threaten face. According to IIT, people recognize potential face threats that could arise in any situation based on two sources of implicit knowledge. One is the constitutive rules
or defining conditions for speech acts (Labov & Fanshel, 1977; Searle, 1969). When making a request, for example, a speaker implicitly assumes that there is a need for action, a need to request (i.e., the target person was not going to perform the desired action already), the possibility that the target person might be willing or feel obligated to comply (otherwise there is no point in requesting), and so forth. These assumptions have implications for both parties face. By assuming that a target person might be willing to perform an action that the target otherwise would not have performed, for instance, a speaker places some degree of constraint on the targets autonomy (negative face). Because these implicit assumptions are contestable (e.g., the target of a request may not see any need for action or feel willing/obligated to act), performing speech acts also may lead to questions or disagreements that can threaten face (Ifert, 2000; Jacobs & Jackson, 1983). In addition to rules for speech acts, people draw on their understanding of primary goals as a second source of implicit knowledge to identify potential threats to face. IIT assumes that the rules for speech acts have different implications for both parties identities when framed by different primary goals (Wilson et al., 1998). According to the need for action rule, a speaker who makes a request presumes that there is a reason why the requested action needs to be performed (Searle, 1969). A student who says Id like you to meet my parents to a casual dating partner plausibly could be seen as pursuing the goal of relational intensification. If this occurred early in a dating relationship, questions could be raised about whether greater relational commitment actually is needed at this time (e.g., is the speaker trying to intensify the relationship too quickly, perhaps because s/he is insecure or needy). Alternatively, a student who says I think we should take a break from seeing each other for a while to a romantic partner likely would be seen as pursuing the goal of relational disengagement, which almost certainly would lead to questions about why the dating relationship needed to change (e.g., is the partner somehow inadequate). Put simply, the same rule (need for action) has different implications for both parties identities depending on the primary goal defining the interaction. As a second example, a speaker who offers an invitation presumes that the other party plausibly might be willing to accept. If a student asked a classmate would you like to go see a movie sometime? but the classmate appeared reluctant, reasons for the lack of willingness could threaten the speakers face (e.g., is the student not physically attractive). By analyzing what the rules for speech acts imply when framed by the goals of initiating, intensifying, or terminating a romantic relationship, it is possible to predict which face threats might arise in each case (see Kunkel et al, 2003, for a detailed analysis of potential face threats associated with each relationship goal). IIT assumes that speakers often must manage multiple, conflicting goals such as pursuing relational (re)definition while maintaining both parties face. Like politeness theory, IIT assumes that face wants are interdependent and speakers usually have some motivation to mitigate threat to the hearers face. Politeness theory does recognize that face concerns may be set aside when urgency is great; similarly, the
GPA model assumes that when the primary goal (e.g., getting out of an unwanted relationship) is very important, speakers may communicate clearly and directly despite face-related concerns. When the relational partner is perceived to have violated an obligation, speakers may even desire to attack their partners face (Cai & Wilson, 2000). Finally, IIT assumes that a broad range of message features function as facework or actions designed to make what one is doing consistent with face (Goffman, 1967). One element of facework is whether to pursue the relational (re)definition goal. In the language of politeness theory, speakers may choose not to perform the face-threatening act (FTA) when perceived face threat is great; similarly, the GPA model assumes that speakers may decide not to pursue a primary goal when important secondary goals could be jeopardized. When relationship (re)definition goals are pursued, speakers language choices are assumed to reflect concerns about maintaining both parties face. Early research developed typologies of verbal strategies for creating affinity (Bell & Daly, 1984) or disengaging from romantic relationships (Baxter, 1982; Cody, 1982). Although these typologies implicitly recognized the importance of identity concerns (e.g., Codys typology includes both negative identity management and positive tone strategies), they have been criticized for being ad hoc lists without any overarching theoretical framework (Kellermann & Cole, 1994). In this study, we assess the types of politeness strategies that students include in messages designed to initiate, intensify, or disengage from romantic relationships. Politeness is the expression of the speakers intention to mitigate face threats carried by certain face threatening acts toward another (Mills, 2003, p. 6). Positive politeness strategies, such as giving compliments or emphasizing similarities, mitigate against threats to the others positive face wants; negative politeness strategies, such as hedging or apologizing, redress threats to the others negative face wants (Brown & Levinson, 1987). IIT assumes that such verbal strategies may be used to redress threats to ones own as well as to the hearers (relational partners) face. Kunkel et al. (2003) reported the only investigation that has applied IIT to relationship (re)definition goals. In their first study, participants provided open-ended reports of the face concerns and emotions they anticipated after reading a scenario involving relational initiation, intensification, or disengagement. Themes in their responses, along with a theoretical analysis of the rules for speech acts and primary goals, were used to identify eight different potential threats to the participants and their romantic partners positive and negative face. Closed-ended scales measuring these potential face concerns were administered to different participants in a second study. Participants associated unique sets of face threats with each of the three relationship goals and varied how directly they asked for what they wanted in light of these differences. The current study extends Kunkel et al.s (2003) initial investigation in three important respects. First, we use two scenarios per relationship (re)definition goal, whereas Kunkel et al. only used one scenario to provide a stronger test of whether
face threats are associated with a relationship (re)definition goal rather than with just the specific scenario representing that goal.1 Second, we examine facework at a more mircolevel, exploring use of particular linguistic forms of politeness rather than global ratings of message directness. Third, we assess whether associations between perceived face threats and facework strategies vary across relationship (re)definition goals. Given that participants may place more or less emphasis on supporting face depending on the type of relationship (re)definition being sought, it is possible that perceived face threats are a better predictor of politeness strategies for some relational goals than for others. Based on this rationale, we forward four hypotheses and two research questions.
Hypothesis 1: Participants will perceive different potential threats to their own positive and negative face, as well as to their partners positive and negative face, depending on whether they imagine initiating, intensifying, or disengaging from a romantic relationship.
Although this hypothesis is stated nondirectionally, we can predict how perceptions for many of the potential face threats identified by Kunkel et al. (2003) will vary across goals. As suggested by our discussion of speech act rules and primary goals, we expect that (a) participants who imagine initiating a romantic relationship will perceive the greatest threat that they might not appear attractive to the other, (b) those who imagine intensifying a romantic relationship to perceive the greatest threat of appearing overly dependent, and (c) those who imagine disengaging from a romantic relationship to perceive the greatest threat making their partner appear inadequate. Besides perceived face threats, we expect that participants in the three relationship goal conditions also will vary in terms of facework. Because young adults on average place greater importance on face maintenance than on the primary goal when initiating a romantic relationship, whereas they weight both goals about equally when intensifying or disengaging from a romantic relationship (Schrader & Dillard, 1998).
Hypothesis 2: Participants who imagine intensifying or disengaging from a romantic relationship will be more likely than those who imagine initiating a relationship to report that they would actually talk to the other party (i.e., pursue the relational goal) if the situation were real.
Although politeness strategies should be present in most messages seeking relational change given the complex nature of relational (re)definition, the specific strategies used also may vary depending on the relationship goal. For example, an individual might include statements of caring (I care about you very much, but) as a form of positive politeness when seeking to disengage from a romantic relationship, but saying I care about you very much may be too risky for someone who wants
to intensify a romantic relationship yet is uncertain whether his or her casual dating partner feels the same way (Knobloch & Carpenter-Theune, 2004). Given that specific strategies may be more or less suited to different relationship goals, we ask the following:
Research Question 1: Do participants in the initiating, intensifying, and disengaging conditions differ in their use of positive and negative politeness strategies?
Finally, we investigate perceived face threats and facework. Based on politeness theory, we predict the following:
Hypothesis 3: Participants who report that they would not pursue the relational goal if the scenario were real perceive greater overall face threat than those who report that they would pursue the relational goal. Hypothesis 4: As suggested by politeness theory, as the overall level of perceived face threat increases, participants will include positive and negative politeness strategies in their messages more frequently.
Finally, participants may have varying motivation to mitigate face threats depending on the particular relationship (re)definition goal. Hence, we ask the following:
Research Question 2: Does the strength of association between perceived face threats and pursuing the relational goal or politeness strategies vary depending on whether participants imagine initiating, intensifying, or disengaging from a romantic relationship?
Method
Participants
Undergraduate students enrolled in communication courses at two large Midwestern universities (N = 598; 342 women, 255 men, 1 no response) participated in this study. Participants ranged in age from 18 to 53 years with an average age of 21.43 (SD = 3.19). Most were sophomores (n = 116, 19%), juniors (n = 163, 27%), or seniors (n = 287, 48%). In terms of ethnicity, the majority indicated they were European American (n = 443, 74%).
Procedures
To fulfill a research requirement or to receive extra credit, participants completed a Romantic Relationships Goals Questionnaire. Participation lasted between 40 min and 1 hour. After preliminary instructions were provided, each participant received an informed consent form as well as a packet of materials that randomly assigned
him or her to one of six scenarios involving relational (re)definition. Two of the six scenarios instantiated each of the three relationship goals (i.e., initiation, intensification, and disengagement).
Instrumentation
The Romantic Relationship Goals Questionnaire was divided into four separate sections: general background items, relationship goals, message construction, and face threats. Participants also completed an individual-difference measure not relevant to the current report. General background items. The general background items obtained demographic information about the participants, including sex, age, year in school, and ethnicity. Relationship goals. Depending on the version of the questionnaire, participants responded to a hypothetical scenario in which they imagined initiating, intensifying, or disengaging from a romantic relationship. There were two scenarios for each goal (see Table 1; initiation [total n = 202; 92 seen in class scenario, 110 met at friends scenario], intensification [total n = 199; 91 date exclusively scenario, 108 meet parents scenario], or disengagement [total n = 197; 91 boring/avoiding scenario, 106 arguing/alternatives scenario). After participants read one of these six scenarios, they indicated on 7-point semantic differential scales the extent to which they thought the situation was (a) unrealistic to realistic, (b) difficult to imagine themselves in to easy to imagine themselves in, (c) unreasonable to reasonable, (d) something that could never happen to them to something that could easily happen to them, and (e) unbelievable to believable ( = .87). Responses to the five questions were summed and divided by the number of items to retain the original 1 to 7 scale (higher scores = greater perceived realism). Across the six scenarios, participants felt that their situation was realistic (M = 5.68).2 Message construction and pursuing the primary (relational) goal. After responding to the realism items, participants wrote out in detail exactly what they would say to their (potential) partner (Chris) in trying to attain their assigned relational (re)definition goal. After writing their message, participants completed a dichotomous measure of whether they actually would pursue the relationship goal. Specifically, participants were asked to circle no or yes in response to the following question: If this were a real situation, would you actually confront Chris and talk to him/her about the situation? Following this question, participants were given 10 blank lines to record their explanations for why they would (not) pursue the relational goal. Face threats. Participants responded to a total of 40 seven-point Likert scales, with 5 items designed to assess the degree to which seeking relationship change might threaten each of the eight specific face threats identified by Kunkel et al.
Table 1 Romantic Relationship Scenarios: Two Per Goal

Goal Initiation Scenario 1: Seen in Class You and Chris both are students in a class of about 50 students. You have been interested in Chris for a while. In fact, you are sure that you would very much like to ask Chris out on a date. At this point, you are not even sure that Chris knows your name or even who you are. The semester is almost over and you are realizing that if you are going to ask Chris out, you had better do it soon. So, you finally have the courage and you are ready to try to initiate this relationship. You speak to Chris. Scenario 2: Met at Friends You met Chris at a friend's house last week. Even though you only talked to Chris briefly, you are interested in getting to know Chris better. In fact, you are sure that you would very much like to ask Chris out on a date. Earlier this week you asked your friend about Chris, but your friend was not sure whether Chris was seeing anyone else. Tonight you ran into Chris at a party. Chris is about ready to leave, and you are not sure when you will see Chris again. You realize that if you are going to ask Chris out, you had better do it now. So, you work up the courage and are ready to initiate the relationship. You speak to Chris. Scenario 4: Meet Parents You and Chris have been casually dating for about 6 months. You are beginning to realize that you really enjoy being with Chris. In fact, you are starting to think that you may be falling in love. You would really like to try to formalize your commitment and intensify your relationship with Chris. In fact, you have been thinking that the time is right for Chris to meet your parent(s). You realize that everyone involved will likely take this as a sign that your relationship with Chris is becoming serious. You are not sure how Chris will feel about intensifying your relationship, but you have finally gotten the courage to ask Chris to take the step of meeting your parents. You speak to Chris. Scenario 6: Arguing/Alternatives You and Chris have been seriously dating for about 6 months. You are (Continued)
Goal Intensification
Scenario 3: Date Exclusively You have been casually dating Chris for several months. You are beginning to realize that you really like being with Chris. In fact, you are starting to think that you may be falling in love with Chris. You would really like to try to formalize your commitment and intensify your current relationship. In fact, it seems like the time is right for you and Chris to agree to date exclusively (i.e., not date anyone else). At this point, however, you are unsure if Chris feels the same way about you. So, you finally have the courage and you are ready to try to intensify this relationship. You speak to Chris.
Goal Disengagement
Scenario 5: Boring/Avoiding You have been seriously dating Chris for several months. You are starting to
Table 1 (Continued)
Goal Scenario 5: Boring/Avoiding realize that things are not the same as when you started dating. In fact, you are very unhappy with how the relationship has been going. Every time you talk to Chris, you find the conversations uninteresting and boring. Lately, you have been trying to avoid contact with Chris and it's starting to get very awkward. It seems like it might be time to end this relationship. So, you finally have the courage and you are ready to try to get out of this relationship. You speak to Chris. Scenario 6: Arguing/Alternatives starting to realize that your feelings for Chris have changed. It seems like the two of you have been arguing a lot, and you often do not enjoy spending time with Chris. Recently, you have realized that there are other people whom you would be more interested in spending time with. It seems like it might be time to end this relationship. So you get up the courage to try to get out of your relationship with Chris. You speak to Chris.
(2003). Specifically, they indicated the degree to which initiating, intensifying, or disengaging might threaten (a) the partners negative face, whereby the partner might be pressured to comply (e.g., Chris might feel pushed into agreeing with what I want in this situation, = .87), (b) the participants own negative face, whereby the participant might feel that he or she was precluding future relationships with other partners (e.g., By asking this now, I might end up feeling like I was boxed into this relationship, = .86), (c) the participants own negative face, whereby the participant might feel that he or she was losing a desirable current relationship (e.g., By making this request, I could end up feeling I had made the wrong decision, = .83), (d) the partners positive face, whereby the participant might make the partner appear inadequate (e.g., By saying what I did in this situation, I might make it seem like something must be wrong with Chris, = .87), (e) the participants own positive face, whereby he or she might worry about appearing physically attractive to the other party (e.g., I would be very concerned about making myself appear physically attractive to Chris in this situation, = .94), (f) the participants own positive face, whereby he or she might appear to be too forward (e.g., I would be concerned that Chris might think that I was being too forward by talking to him or her in this situation, = .83), (g) the participants own positive face, whereby he or she might appear to be overly dependent (e.g., I could make myself appear very needy by asking what I did in this situation, = .85), and (h) the participants own positive face, whereby he or she might look insensitive (e.g., By saying what I did, I may appear to be insensitive in this situation, = .93). All 7-point Likert scales were bounded by the endpoints strongly disagree (1) to strongly agree (7), with larger scores indicating higher perceived face threat. Items tapping different face threats were intermixed. To ensure that the measure for each face threat was unidimensional, separate principal axis factor analyses were conducted on the 5 items tapping each face threat. In all eight cases, only one factor with an
eigenvalue > 1.00 emerged. Responses to items constituting each scale were summed and averaged by the number of items to retain the 1 to 7 scale.
Message Coding
Written messages were analyzed to identify strategies that mitigated threats to the hypothetical partners positive or negative face. Brown and Levinson (1987) identified 15 positive politeness strategies, 5 of which appeared to occur with some regularity in a preliminary scan of our data and hence were analyzed in the current study. Giving compliments strategy refers to complimenting the partners physical or nonphysical attributes, acknowledging enjoyment of the partners company, or making mere statements of liking. Demonstrating nonsuperficial interest in partners affairs strategy refers to statements that are more substantive than simple how are you and whats going on types of statements because they demonstrate an interest in the partners life and/or affairs. Demonstrating similarity and common ground strategy represents phrases where the participant is stating a similarity between himself or herself and the partner or something they share in common. Making statements of caring or affect strategy involves messages demonstrating that the participant cares deeply about the partner. These messages are more intense than those classified as giving complements because they expressed liking. Avoiding blaming partner strategy refers to messages constructed to avoid blaming the partner for a negative situation or to absolve the partner of responsibility for a negative situation. Examples of these 5 positive politeness strategies appear in Table 2.3 Brown and Levinson (1987) identified 10 negative politeness strategies, 4 of which were utilized in the current study. Hedging strategy refers to words or clauses (e.g., I think, maybe) that make a statement or request more tentative in nature. Managing imposition strategies are those which specifically focus on softening the constraints placed on the partner by explicitly recognizing or minimizing the impact of the imposition, providing different options or a less than definite time frame (e.g., lets go out sometime), or giving the partner the option of not accepting or not having to make an immediate decision. Apologizing for request/imposition are strategies where the participant offers a direct apology to the partner for constraining his or her autonomy. Soliciting partners input in sought directive occurs when the participant explicitly asks the partner for feedback about the request (see Table 2 for examples).4 Two coders (the third and fifth authors) and the second author initially worked together to develop rules for identifying and classifying positive and negative politeness strategies. In total, 60 messages from each relational goal (n = 180) were used during this process. Following this, the two coders independently coded a subset of 20 messages from each relational goal (n = 60) for specific politeness strategies. As a test of reliability, percentage of agreement was computed separately for positive and negative politeness strategies within each relational goal by calculating a ratio of the strategies coded the same by the two coders divided by the total number of strategies coded. For example, if both coders identified two instances of giving
Table 2 Categories and Examples of Positive and Negative Politeness Strategies

Positive Politeness Strategies Category Giving compliments Demonstrating nonsuperficial interest in partners affairs Demonstrating similarity and common ground Making statements of caring or affect Avoiding blaming partner Verbatim Examples I think you are a great person. We have had a great time together. How was your week at work? How did you do on the last test? Weve been dating for several months now. My brothers and you would have a lot to talk about. I care about you very much. I love you, and will always love you. Its not you. It is nothing that you did in particular.
Negative Politeness Strategies Category Hedging Verbatim Examples I dont think I have the same feelings for you. We should maybe plan a little get away trip. I was wondering if maybe you would want to go out? I realize this may make you uncomfortable. Maybe lunch or dinner sometime? If you dont want to, you dont have to. We can still be friends. Im sorry but I think its for the best. Im really sorry but things have changed too much. I need to know how you feel about all these things. I was wondering how you feel about that? Would it be cool with you to start dating more often?
Managing imposition
Apologizing for request/imposition
Soliciting partners input in sought directive
compliments and one instance of demonstrating interest in a written message, but only one coder also identified an instance of avoiding blame, then agreement for positive politeness strategies would be 75% (3/4) for that message. To take chance agreement into account, Cohens kappa coefficients were calculated. Fleiss (1981) described kappas over .75 as reflecting excellent levels of agreement. Agreement and kappa coefficients were excellent for positive politeness strategies (initiating: 94%, kappa = .93; intensifying: 87%, kappa = .84; terminating: 95%, kappa = .94) and for negative politeness strategies (initiating: 89%, kappa = .86; intensifying: 86%, kappa = .83; terminating: 86%, kappa = .83).
After achieving reliability, most remaining messages (n = 508), including messages utilized for training, were divided equally between the two coders to independently read and categorize into politeness strategies. As a check for coder drift, the final 30 messages (10 from each relational goal) were read and analyzed by both coders. Once again, agreement and kappa coefficients were excellent for both positive (initiating: 97%, kappa = .96; intensifying: 93%, kappa = .92; terminating: 100%, kappa = 1.0) and negative politeness strategies (initiating: 94%, kappa = .93; intensifying: 86%, kappa = .83; terminating: 93%, kappa = .91).
Results
Descriptive information for perceptions of the eight potential face threats within and across relationship (re)definition goals appears in Table 3. Frequencies and percentages for participants use of positive and negative politeness strategies within and across goal conditions appear in Tables 4 and 5. Coders identified a total of 621 positive politeness strategies in the messages written by the 598 participants. Across goal conditions, 60% of participants included at least one positive politeness strategy in their written message. Coders identified a total of 1350 negative politeness strategies in the 598 written messages. Across goal conditions, 86% of participants included at least one negative politeness strategy.
Relationship (Re)Definition Goals and Face Threats

Hypothesis 1 predicted that participants who imagined initiating, intensifying, and disengaging from romantic relationships would perceive different threats to their own and their partners face. To assess this hypothesis, we conducted a 3 (type of relationship goal) 8 (type of face threat) mixed-model ANOVA with repeated measures on the second factor. Degree of face threat served as the dependent variable. Because Mauchlys test of sphericity was significant, W = .40, 2(27) = 526.99, p < .001, Greenhouse-Geisser corrections to degrees of freedom were used for all within-subjects effects (see Keppel & Wickens, 2004). The ANOVA revealed a small but statistically significant main effect for relationship goal type, F(2, 576) = 17.29, p < .001, 2 = .02, indicating that, across types of face threat, participants perceived greater overall potential face threat when pursuing some goals than others. A Tukey post hoc test revealed that participants who imagined initiating a romantic relationship perceived less overall threat to both parties face than did those who imagined intensifying or disengaging from a relationship, although the latter two groups did not differ significantly from each other (see Table 3). A large main effect for type of face threat also was obtained, F(5.78, 3329.06) = 236.53, p < .001, 2 = .14, indicating that, across goals, some types of face were threatened to a greater extent than other types. Across goal conditions, participants perceived that seeking relationship
46 Intensifying SD 1.34 1.16 1.23 0.80 1.24 1.21 0.86 1.44 0.84 3.75b 3.75b 2.02a 3.61 3.36b 1.70 1.39 0.98 1.30 0.89 2.56c 2.41a 4.50b 3.33 3.53b 1.43 1.07 2.79 1.23 0.81 3.80b 1.98b 1.37 0.87 1.28 1.40 4.60c 3.52c 4.27b 3.90b 1.31 1.43 1.14 1.24 4.84c 2.42a 4.19 2.91 3.88 2.42 3.92 2.90 2.79 3.46 3.30 M SD M SD M Disengaging Across Goals SD 1.38 1.46 1.40 1.31 1.88 1.37 1.64 1.33 0.87 2 for Goal Main Effect .16 .23 .15 .35 .39 .19 .54 .01 .06
Table 3 Descriptive Statistics for Perceived Face Threats Within and Across Relationship (Re)Definition Goals
Initiating
Type of Face Threat
3.48a 2.41a
Impose on partner Preclude future relationships Lose current relationship Make partner appear inadequate Not appear attractive Appear overly dependent Appear insensitive Appear too forward Across face threats
3.26a 1.78a
5.40a 2.55a 1.88a 3.43 3.09a
Note: n = 202 respondents for initiating, 199 for intensifying, 197 for terminating, and N = 598 across goals. Within rows, means for goal conditions with different superscripts are significantly different at p < .05 via the Tukey post hoc test.
Table 4 Frequencies and Percentages of Positive Politeness Strategies Within and Across Relationship (Re)Definition Goals
Initiate Strategy Type Give compliments Demonstrate interest Demonstrate similarity Demonstrate caring Absolve blame Total positive strategies f 71 62 48 47 04 232 % 25 18 23 17 02 73 Intensify f 65 58 38 22 04 187 % 28 18 19 09 02 58 Terminate f 85 15 34 57 11 202 % 33 05 17 34 05 58 Across Goals f 221 135 120 126 19 621 % 29 13 20 16 03 60
Note: f = frequency of occurrence (of strategies of that type in that condition); % = % of messages in that condition containing 1 or more strategies of that type.
Table 5 Frequencies and Percentages of Negative Politeness Strategies Within and Across Relationship (Re)Definition Goals
Initiate Strategy Type Hedge Manage imposition Apologize Solicit input Total negative strategies f 178 115 11 121 425 % 57 41 06 51 83 Intensify f 186 159 18 93 456 % 68 50 08 41 85 Terminate f 236 100 20 113 469 % 69 40 09 46 90 Across Goals f 600 374 49 327 1350 % 61 45 07 46 86
Note: f = frequency of occurrence (of strategies of that type in that condition); % = % of messages in that condition containing 1 or more strategies of that type.
(re)definition created the greatest threat to the hypothetical partners autonomy or negative face (M = 4.19 on the 7-point scale; see Table 3). However, participants also perceived moderate levels of threat to their own negative face (i.e., that they might later regret losing their current relationship) and their own positive face (i.e., that they might appear too forward or not appear attractive to the partner) across goal conditions, indicating that they typically saw some potential for multiple face threats. Consistent with hypothesis 1, the ANOVA also detected a large, statistically significant Type of Goal Type of Face Threat interaction, F(11.56, 3329.06) = 183.91, p < .001, 2 = .22, reflecting that the degree to which specific types of face were threatened varied substantially depending on the type of relationship goal. To interpret
this two-way interaction, separate one-way ANOVAs and Tukey post hoc comparisons were conducted to explore the effects of relationship goals on each of the eight types of face threat (see Table 3). Type of relationship goal exerted a large, statistically significant effect on ratings of 7 of the 8 face threats, explaining 16% to 54% of the variance in participants ratings for these face threats. The direction of differences in mean scores generally is consistent with expectations. For example, participants who imagined initiating a romantic relationship perceived greater threat of not appearing attractive, those who imagined intensifying perceived greater threat of appearing overly dependent, and those who imagined terminating perceived greater threat of making the partner appear inadequate compared to participants in the other two goal conditions. A subsidiary analysis showed that participants sex did not qualify these findings because females and males associated very similar face threats with each of the three relationship goals.5 Two additional follow-up analyses were conducted to clarify the effect of relationship (re)definition goals on perceived face threats. Because perceived scenario realism varied slightly across goals (see note 2), a 3 (goals) 8 (type of face threat) mixed-model ANCOVA was conducted with realism as a covariate. Perceived realism was not significantly associated with degree of perceived face threat, F(1, 573) = 0.94, p = .33, 2 = .002. The main effects for goal type and type of face threat as well as the two-way Goal Face Threat interaction were statistically significant, and virtually identical in size, after controlling for perceived realism. To assess whether perceived face threats were associated with relationship (re)definition goals, as opposed to specific scenarios instantiating those goals, we also conducted separate one way ANOVAs with scenario as the independent variable and each of the 8 face threats as the dependent variable. Tukey post hoc tests were performed across the 6 scenarios to assess whether the 2 scenarios instantiating each goal differed in terms of the degree to which each type of face was perceived to be threatened. For example, we assessed whether the two initiating scenariosseen in class and met at a friendsdiffered in terms of the degree to which participants perceived they were imposing on the other party. We did the same for the two intensifying and the two disengaging scenarios. With 3 goals and 8 face threats, 24 pairwise comparisons between scenarios instantiating the same goal were examined. The two scenarios instantiating each goal condition did not differ significantly in 21 of these 24 cases. Given that we had strong statistical power to detect what Cohen (1988) would label as a medium-size difference (d = .50) in this analysis,6 it appears that perceived face threats are associated with relationship (re)definition goals rather than with the specific scenarios instantiating each goal.
Relationship (Re)Definition Goals and Facework

Pursuing the relationship goal. Hypothesis 2, which predicted that a larger percentage of participants in the initiating condition would indicate that they would not
pursue the relationship goal if their scenario were real as compared to those in the intensifying and disengaging condition, was assessed via 3 2 2 analysis crossing goal type (initiating, intensifying, or disengaging) with pursuing the relationship goal (no, yes). Although nearly two third of the participants (64%) who imagined initiating a romantic relationship indicated that they would talk to the other person and ask him or her out if their scenario had been a real situation, this rate of goal pursuit was lower than for participants who imagined intensifying (86%) and disengaging from (97%) a romantic relationship, 2(2) = 76.47, p < .001, contingency coefficient = .34. Although men (74%) were more likely than women (59%) to say that they would ask the other person out in the initiating condition, both sexes were more likely to say they would not pursue the relational goal in the initiating as opposed to the intensifying or disengaging conditions.7 When asked to explain why they would not ask Chris out, several participants who said no in the initiating condition expressed concerns about their own positive facefor example, I tend to be a little timid and shy with people that I dont know because I am worried about if they will like me or I am scared of rejection and take it personally. Others implicated negative face, such as one participant who wrote Not wanting to overstep any boundariesespecially since it is unclear whether Chris is dating/seeing anyone else. Politeness strategies. Research Question 1 asked whether participants would vary the specific types of positive and negative politeness strategies they used depending on whether they imagined initiating, intensifying, or terminating a romantic relationship. A 3 5 2 analysis was conducted crossing goals (initiating, intensifying, or disengaging) and positive politeness strategies (giving compliments, demonstrating interest, demonstrating similarity, demonstrating caring, and absolving blame; see Table 4). The analysis was significant, 2(8) = 52.70, p < .01, contingency coefficient = .30, indicating that frequencies for positive politeness strategies varied across goals. Inspection of residuals (expected observed frequencies) revealed that participants in the initiating and intensifying condition demonstrated interest more often than would be expected by chance, whereas those in the disengaging condition did so far less frequently than chance would dictate. In contrast, participants in the disengaging condition gave compliments and demonstrated caring more frequently than would be expected by chance, whereas those in the initiating gave compliments and those in the intensifying condition expressed caring less frequently than chance would dictate. Participants in different conditions varied in terms of the specific forms of positive politeness they tended to use. A 3 4 2 also was conducted crossing goals and negative politeness strategies (hedging, managing imposition, apologizing, and soliciting input; see Table 5). A significant effect for relational goals was obtained, 2(6) = 29.23, p < .01, contingency coefficient = .15, indicating that frequencies for negative politeness strategies also varied across goals. Inspection of residuals revealed that participants in the initiating condition solicited input more often and those in the intensifying condition
did so less often than would be expected due to chance. Participants in the intensifying condition used managing imposition strategies more frequently than would be expected by chance, whereas those in the disengaging condition did so less frequently than chance would dictate. Finally, participants in the terminating condition used hedges more frequently than expected by chance, whereas those in the initiating and intensifying conditions hedged less than would be expected due to chance. Participants in the three goal conditions also differed in their use of specific negative politeness strategies.
Perceived Face Threats and Facework

Pursuing the relationship goal. Hypothesis 3 predicted that participants who indicated that they would not pursue the relational goal will perceive greater total perceived face threat compared to those who indicated they would pursue the goal. Research Question 2 asked whether associations between goal pursuit and degree of perceived face threat might vary depending on the particular relationship goal. To assess Hypothesis 3 and Research Question 2, we performed MANOVAs with goal pursuit (no vs. yes) as the independent variable and the eight types of face threat as the dependent variable. MANOVAs were conducted separately within each goal condition. In the initiating condition, the multivariate main effect for goal pursuit was statistically significant, Wilks Lambda = 0.88, F(8, 187) = 3.27, p = .002, 2 = .12, reflecting that participants who indicated that they would not pursue the relationship goal (n = 72) perceived significantly greater overall face threat compared to those who would (n = 127). To clarify this group difference, follow-up independent-groups t tests were conducted comparing participants who would not versus would pursue the goal in terms of how they perceived each of the eight threats to their own and the other partys face (see Table 6). In all eight cases, students who said they would not pursue the relationship goal reported significantly greater perceived face threat than those who would do so. Effect sizes (d) ranged from .31 to .74. In the intensifying condition, a statistically significant multivariate main effect also was detected, Wilks Lambda = .91, F(8, 181) = 2.12, p < .05, 2 = .09, reflecting that participants who would not pursue the relational goal (n = 28) perceived significantly greater overall face threat compared to those who would (n = 162). When t tests were conducted on individual face threats, however, 7 of the 8 comparisons were not statistically significant at p < .05 (Cohens d ranged from .22 to .21). The only significant difference was that participants in the intensifying condition who would not pursue the relational goal (M = 4.23, SD = 1.67) perceived greater risk of appearing overly dependent compared to those who would (M = 3.67, SD = 1.33), t(196) = 2.02, p < .05, d = .42. Students who judged that they would not pursue the primary goal perceived trying to intensify a romantic relationship to be more threatening to one specific aspect of their own positive face compared to those who would do so.
Table 6 Perceived Face Threats by Participants in the Initiating Condition Who Would Not Versus Would Pursue the Relational Goal
Would Not Pursue the Goal Type of Face Threat Impose on partner Preclude future relationships Lose current relationship Make partner appear inadequate Not appear attractive Appear overly dependent Appear insensitive Appear too forward M 3.75 2.61 3.61 1.95 5.66 2.98 2.08 4.04 SD 1.47 1.23 1.22 0.88 1.25 1.45 0.94 1.54 Would Pursue the Goal M 3.29 2.26 3.03 1.67 5.26 2.28 1.77 3.04 SD 1.24 1.07 1.20 0.74 1.23 0.97 0.81 1.23 t 2.36 2.09 3.28 2.36 2.20 3.67 2.47 4.71 Group Comparison p .02 .04 .001 .02 .03 .001 .02 .001 d .47 .31 .48 .37 .32 .61 .36 .74
Note: n = 199 participants (72 who would not do the face-threatening act [FTA]; 127 who would do the FTA).
Nearly all participants in the disengaging condition indicated that they would pursue the relational goal; indeed, only five students indicated that they would not talk with their partner if they were really in the scenario. This occurred despite the fact that seeking to disengage from a romantic relationship was perceived to be the most face threatening of all three relationship (re)definition goals (see Table 3). When asked to explain why they would confront their partner, many participants wrote about the importance of being happy. One participant explained that there is no point in my staying in a dead-end relationship that Im not happy in. Its not healthy and life is too short and a second wrote, Not to sound selfish but if my being in a relationship would not make me happy then why be in one? Several participants stated that being happy was more important than avoiding potential threat to their partners face, in comments such as Its not worth sacrificing my happiness because Im afraid Id be hurting feelings and Someone has to realize that were not right together and you cant waste time to just spare feelings. Several participants talked about the importance of not wasting time (i.e., not imposing unnecessarily on both parties negative face), for example, If it is not working, it is better to know as soon as possible so you arent wasting each others time. As virtually all participants indicated they would pursue the relational goal, we could not compare perceived face threats by those who would not versus would do so in the disengaging condition.
In sum, Hypothesis 3 received strong support in the initiating condition and limited support in the intensifying condition. In response to Research Question 2, type of relationship goal influenced whether students who would not pursue the primary goal perceived greater threat to both parties face (initiating) or only to their own positive face (intensifying), and also whether Hypothesis 3 could be tested at all (i.e., whether a substantial percentage of students believed they would not pursue the primary goal). Politeness strategies. Hypothesis 4 predicted that as overall level of perceived face threat increased, participants more frequently would include positive and negative politeness strategies in their messages seeking relationship (re)definition. Research Question 2 asked whether the strength of association between perceived face threats and politeness strategy would vary depending on the relationship goal. To address Hypothesis 4 and Research Question 2, we analyzed associations between levels of perceived face threat and frequencies for politeness strategies separately within each of the 3 goal conditions. Given that there were 8 types of face threat and 9 politeness strategies, 72 correlations were computed in each case. Because frequency distributions for all 9 politeness strategies were positively skewed, raw frequencies were log transformed before computing correlations (see Keppel & Wickens, 2004). Table 7 displays the correlations that were statistically significant (p < .05, two-tailed) within each of the three goal conditions. Sample sizes ranged from n = 190 to 202 in these analyses. Assuming n = 190 and p < .05 (two-tailed), we had excellent statistical power (.93) to detect associations of the size that Cohen (1988) labeled as a medium effect (r = .24) but limited power (.28) to detect small associations (r = .10). In the initiating condition, 12 of the 72 correlations were statistically significant (see Table 7). These 12 correlations fall between small (r = .10) and medium (r = .24) effects. Nine are positive, indicating that the degree to which participants perceived that they were threatening their own and the other partys face by initiating is directly associated with the frequency with which they used politeness strategies. For example, the more participants perceived that they might be imposing on the other person, might appear too forward, might not appear attractive, and might lose a chance at this relationship, the more frequently they included hedges in their date requests. The only exception to this pattern occurred for the positive politeness strategy showing interest. The more participants perceived that they might appear overly dependent, look insensitive, or lose a chance at this relationship, the less they asked about the other persons day-to-day activities (perhaps because this could have been seen as a ploy leading up to asking the other person out). With this one exception, results from the initiating condition are consistent with the claim that greater perceived face threat is associated with greater use of politeness strategies (Hypothesis 4). In the intensifying condition, only 3 of the 72 correlations were statistically significant (see Table 7). Given the p < .05 significance level, one would expect 3.6
correlations out of 72 tests to be statistically significant simply due to chance, (Type I error) even if perceived face threat and politeness strategies were not related in the larger population. Hence, there is no support for Hypothesis 4 in the intensifying condition. In the disengaging condition, 7 of the 81 correlations were statistically significant (see the correlations in Table 7). Unexpectedly, all 7 correlations are negative, indicating that increased perceived face threat is associated with fewer politeness strategies. Given the correlational nature of these data, it is possible these findings reflect the effect of messages on face threats rather than vice versa. For example, participants who solicited their partners input more frequently when asking to end their romantic relationship may have been less likely to subsequently feel that they had made their partner look inadequate, looked insensitive themselves, or that they might later regret losing the relationship compared to those who did not solicit their partners input precisely because the former group made some attempt to include the partner in the decision. In any case, findings in the disengaging condition run contrary to Hypothesis 4. Regarding Research Question 2, associations between face threats and politeness strategies vary not only in magnitude but also in direction, depending on the particular relationship (re)definition goal.
Discussion
This study applies IIT to analyze identity concerns that shape and constrain how young adults pursue relationship (re)definition goals. Participants read a hypothetical scenario in which they imagined wanting to initiate, intensify, or disengage from a romantic relationship, described what they would say in the situation, and reported perceived threats to their own and their partners face. Findings offer insights about identity concerns that young adults associate with each of these relational goals as well some of the micropractices that are used to manage them. Initially, we summarize the types of face threats and politeness strategies that tend to occur with each relationship goal and then discuss how IIT might account for these findings. Along the way we discuss limitations and directions for future research. Participants in the initiating condition were very concerned about their own positive face, perceiving a potential risk of being rejected by the hypothetical partner who might not find them to be attractive. They also perceived moderate risk of pressuring the other party (see Table 3). About one third of participants in the initiation condition indicated they would not pursue the relational goal (i.e., ask the other person out) if their scenario were a real-life situation. Participants in this condition expressed interest in the other and solicited the others input more frequently than those in the other two goal conditions. Initiation also was the goal condition in which perceived face threats shared the strongest associations with facework. Participants who said they would not pursue the relational goal perceived greater risk of all eight
54 Type of Face Threat Inadequate .16a .17b .14b .16b/.16c .20a .16b .18b .15c .15c .19c .15b .16b .19c .14a .15b .17b Attractive? Dependent Lose Insensitive .14b Forward
Table 7 Correlations Between Perceived Face Threats and Frequencies of Politeness Strategies Within Three Goal Conditions
Politeness Strategy
Impose
Preclude
.16b .16c
.16b
Compliment Show interest Show similarity Show caring Absolve blame Hedge Manage imposition Apologize Solicit input
.16c
Note: n = 190 to 202; impose = impose on other party; preclude = preclude future relationships; inadequate = make other appear inadequate; attractive? = appear attractive to other?; dependent = appear overly dependent; lose = lose current relationship; insensitive = look insensitive; forward = look too forward; total FT = total perceived face threat. Only correlations statistically significant at p < .05 (two-tail) are shown. a. Correlations are from the intensifying condition. b. Correlations are from the initiating condition. c. Correlations are from the terminating condition.
types of face threat compared to those who said they would do so (see Table 6). As degree of perceived threat to their own and the potential partners face increased, participants in the initiating condition included more positive and negative politeness strategies in their written messages. Participants in the intensification condition perceived moderate threat to their hypothetical partners negative face (feeling pressured) as well as to their own positive face (e.g., appearing overly dependent) and negative face (e.g., precluding future relationships, possibly losing the current relationship). Students in this condition were more likely than chance to demonstrate interest in their partners and to manage the extent to which they were imposing on them (e.g., by giving them the option of not making an immediate decision); they were less likely to explicitly ask what their partner thought about their request or to explicitly express caring or affection compared to those in other conditions. Students who said they would not talk with their partner about intensifying commitment if the scenario were real perceived greater threat that they might appear overly dependent, but no greater risk to their partners face, compared to those who said they would talk to their partner. Few associations were detected between perceived face threats and politeness strategies in the intensifying condition. Participants charged with disengaging from their imagined dating relationships perceived the highest overall level of threat to both parties face (see Table 3). These participants perceived moderate risk of constraining their partners autonomy, making their partner look inadequate, losing a relationship that they would later regret, and appearing insensitive. Despite this, virtually all (97%) students in the disengaging condition indicated that they would pursue the relational goal if their scenario were real. Participants in this condition hedged, gave compliments, and expressed caring more frequently than those in the other goal conditions. However, perhaps aware of possible counter-persuasion efforts by their hypothetical partner, they managed impositions and expressed interest in their partners lives less often than would be expected by chance. In the disengaging condition, as perceptions of threat to ones own or the other partys face increased, use of politeness strategies such as apologizing or soliciting input actually decreased (contrary to what would be expected based on politeness theory). It merits note that this occurred even though the disengaging scenarios did not include instances where the partner clearly was responsible for problems with the relationship (e.g., infidelity) where it might have made sense that participants would feel little desire to support their partners face. Because it stresses the importance of analyzing the particular primary goal that defines an interaction, IIT helps make sense of these findings. Phrases such as I wondered if youd want to go to a party with me next Saturday, I dont want to date anyone else but you, and Im sorry, but I dont think we should see each other any more cue up culturally viable explanations for what is going onthat is, relational goals that have different implications regarding the current state and possible future of a romantic relationship. By analyzing what the rules for speech acts imply when
framed by different relational goals, IIT offers testable predictions about which threats to ones own and the partners positive and negative face should be seen as most likely to occur when participants initiate, intensify, or disengage from a romantic relationship. Results from our study, which used multiple scenarios per goal, confirmed many of these predictions, which helps alleviate concern that findings from an earlier study (Kunkel et al., 2003) reflected the particular scenario that was used instantiating each goal rather than differences between goals themselves. Because participants were randomly assigned to goal conditions, we also can be confident that variations in primary goals are causing variations in perceived face threats. By acknowledging that the relative importance put on accomplishing the relational goal versus maintaining face may differ across relational goals (Schrader & Dillard, 1998), IIT also suggests why face threats might lead young adults to avoid pursing relational (re)definition in some cases (e.g., initiating) but not in others (e.g., disengaging). Written rationales from some students in the initiating condition suggest that they would refrain from asking out someone they did not know well because of the possibility of rejection. In contrast, many students in the disengaging condition said they would confront someone they had been dating about their dissatisfaction with the relationship because being happy was more important than the possibility of hurt feelings. Although suggestive, one limitation of the current study is that we did not have participants describe their goals or rate the importance of multiple goals in their scenario. Doing so in future research would provide a more direct test of this assumption from IIT. One surprising finding was that type of relationship goal moderated not only the strength but even the direction of association between perceived face threats and politeness strategies (see Table 7). In the initiation condition, where most significant associations were positive, it seems plausible that participants were (unconsciously) tailoring their messages in light of potential face threats (i.e., face threats verbal strategies). In the disengaging condition, where significant associations were negative, it seems plausible that participants were reporting perceived consequences of direct or blunt messages they had just constructed (i.e., verbal strategies face threats). Because we measured perceived face threats and politeness strategies in a cross-sectional design, we cannot be certain about the direction of causation in either condition. Future research might employ longitudinal designs to investigate possible reciprocal relationships between face threats and facework. Students might keep structured diaries (e.g., Duck, Rutt, Hurst, & Strejc, 1991) about their romantic relationships, describing conversations in which they and/or their partner sought changes in their relationship as well as their goals, concerns, and feelings before and after such conversations. Diary studies also might show the relevance of IIT to relationships with less linear trajectories, such as when an individual asks out an ex romantic partner. Given that prior studies (Wilson et al., 1998, Wilson & Kunkel, 2000) have applied IIT to the goal of giving advice, an alternative approach would be to bring dating couples into the lab, instruct one member of each couple to offer his/her partner advice during a
conversation about what was going on in their lives, and assess how the advice was given as well as goals and perceived face threats before, during, and after the conversation (for similar designs, see Jones & Wirtz, 2006; Sillars, Roberts, Leonard, & Dun, 2000). Such a design would shed light on how participants adapt (perhaps without awareness) message features in light of perceived face threats while also constantly updating perceptions of face threats in light of anticipated and perceived responses from their partner to what they already have said and done. Caution should be exercised in generalizing findings from our college student sample to the dating relationships of single adults 30 years and older (Mongeau, Jacobsen, & Donerstein, 2007) or to other societies where relational turning points and hence the goals inferred from behaviors might differ. We also used hypothetical scenarios and gathered written messages stripped of paraverbal features that influence judgments of politeness (Dillard, Wilson, Tusing, & Kinney, 1997; Laplante & Ambady, 2003). Actual episodes of initiation, intensification, and disengagement would feature dialogues rather than monologues and might be comprised of multiple conversations. Despite this, we are confident that our detailed coding of politeness strategies in written messages offers insights into what dialogues about relationship (re)definition would be like; indeed, prior research demonstrates moderate correspondence between qualities of the messages people write in response to hypothetical scenarios and what they say during role plays or naturalistic interactions (e.g., Applegate, 1980, 1982). Our findings thus clarify the identity concerns that young adults associate with three relationship (re)definition goals and show how to varying degrees these concerns shape and constrain what verbal strategies are used to pursue each goal.
Notes
1. Half of the data analyzed in this study (i.e., responses to three of the six scenarios) were reported in Kunkel et al. (2003, Study 2), whereas the other half was gathered for this study. Aside from gathering new data, we report several findings that were not presented in Kunkel et al.s earlier report, including analyses of politeness strategies and whether associations between perceived face threats and facework vary depending on the relationship (re)definition goal. 2. To assess whether perceptions of realism varied across scenarios instantiating the three goals, we conducted a one-way ANOVA with realism scores as the dependent variable and goal type (initiating, intensifying, disengaging) as the independent variable. Data from the two scenarios instantiating each goal were collapsed. Type of goal exerted a small but statistically significant effect on realism ratings, F(2, 593) = 13.25, p < .01, 2 = 04. A Tukey post hoc test revealed that participants who imagined disengaging (M = 5.93) perceived their scenario as more realistic than those who imagined intensifying (M = 5.69), who in turn perceived their scenario as more realistic than those who imagined initiating (M = 5.40). Despite this main effect, participants in all three conditions perceived their scenario as realistic. For example, 89% (178 of 201) of participants in the initiating condition rated their scenario at the scale midpoint (4.00) or higher on the 7-point realism scale. 3. In terms of correspondence between Brown and Levinsons (1987) positive politeness strategies and the five strategies that we coded, our demonstrating nonsuperficial interest strategy is derived from their exaggerate interest and intensify interest strategies, and our demonstrate similarity strategy is
derived from their presuppose/assert common ground strategy. Our give compliments and make statements of caring strategies are derived from their express approval as well as give gifts strategies. Brown and Levinson noted that the latter includes not only tangible gifts but statements that address the others desire to be cared about. Finally, our avoid blaming partner strategy corresponds loosely to their avoid disagreement strategy in that blaming the other is likely to lead to disagreement. 4. In terms of correspondence between Brown and Levinsons (1987) negative politeness strategies and the four strategies that we coded, our hedge strategy is derived from their extensive discussion of hedging, our manage the imposition strategy is derived from their minimize imposition, our apologize strategy is derived from their apologize strategies, and our solicit the partners input strategy is derived from their dont assume the others willingnessquestion. 5. To assess possible sex differences, we conducted a 3 (goal type) 2 (sex) 8 (type of face threat) mixed-model ANOVA with repeated measures on the last factor. A small main effect for sex, F(1, 572) = 12.98, p < .001, 2 = .01, reflected that across goals and types of face threats, male students (M = 3.46) perceived seeking any type of relationship (re)definition to be potentially more face threatening than did females (M = 3.19). The three-way Goal Sex Type of Face Threat interaction also was statistically significant, F(11.56, 3305.06) = 2.98, p > .001, 2 = .01, though the effect also was small. Follow-up 3 (goal) 2 (sex) factorial ANOVAs were run separately for each of the 8 face threats. Small, but statistically significant, two-way interactions were obtained for 3 of the 8 face threats: making their partner appear inadequate, F(2, 579) = 3.10, p < .05, 2 = .01; not appearing attractive, F(2, 591) = 4.29, p < .02, 2 = .01; and appearing too forward, F(2, 590) = 6.17, p < .01, 2 = .02. Post hoc analyses revealed that the effect of goals on perceived face threat in two of these three cases was virtually identical for females and males. In the third case, females perceived greater risk of appearing too forward in the initiation (M = 3.48) and intensification (M = 3.65) than in the termination (M = 3.04) condition, F(2, 338) = 6.09, p < .01, 2 = .04, whereas ratings by males did not differ significantly across goals (M = 3.36, 3.56, and 3.79 in the initiating, intensifying, and terminating condition), F(2, 252) = 2.02. p = .14, 2 = .02. Because this was the only case where sex qualified which goal conditions differed significantly, it appears that female and male students associated largely similar sets of potential face threats with each relationship (re)definition goal. 6. The Tukey post hoc test adjusts the critical value for pair-wise comparisons to reduce inflation of the family-wise Type I error rate when conducting all possible pair-wise comparisons (Keppel & Wickens, 2004). Given this adjusted critical value, statistical power is lower than it would be for an independentgroups t test. However, even if we assume that the critical value for each pair-wise comparison corresponded with an alpha level of p < .01 rather than .05, and also n1 = 90 for the first scenario instantiating each goal and n2 = 100 for the second scenario instantiating the same goal (the approximate sample sizes in each goal condition), statistical power to detect a medium-size difference (d = .50) between pairs of scenarios still was .80. 7. To assess possible sex differences, we conducted 3 (goal) x 2 (dont/do pursue the relational goal) 2 analyses separately for female and male students. For females, the effect of goal type was significant, 2(2) = 62.79, p < .001, contingency coefficient = .40, reflecting that only 59% of females in the initiating condition indicated they would pursue the relational goal as compared to 87% and 97% of females in the intensifying and terminating condition. Although less pronounced, the effect of goal type also was significant for males, 2(2) = 18.24, p < .001, contingency coefficient = .26, reflecting that 74% of males in the initiating condition indicated they would pursue the relational goal as opposed to 85% and 98% of males in the intensifying and terminating conditions.
References
Applegate, J. L. (1980). Person- and position-centered communication in the day-care center. In N. K. Denzin (Ed.), Studies in symbolic interaction (Vol. 3, pp. 59-65). Greenwich, CT: JAI.
Applegate, J. L. (1982). The impact of construct system development on communication and impression formation in persuasive contexts. Communication Monographs, 49, 277-289. Baxter, L. A. (1982). Strategies for ending relationships: Two studies. Western Journal of Speech Communication, 46, 233-242. Baxter, L. A., & Bullis, C. (1986). Turning points in developing romantic relationships. Human Communication Research, 12, 469-493. Bell, R. A., & Daly, J. A. (1984). The affinity-seeking function of communication. Communication Monographs, 51, 91-115. Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage. Cambridge, UK: Cambridge University Press. Cai, D. A., & Wilson, S. R. (2000). Identity implications of influence goals: A cross-cultural comparison of interaction goals and facework. Communication Studies, 51, 307-328. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. Cody, M. (1982). A typology of disengagement strategies and an examination of the role intimacy, reactions to inequity and relational problems play in strategy selection. Communication Monographs, 49, 148-170. Cody, M. J., Canary, D. J., & Smith, S. W. (1994). Compliance-gaining goals: An inductive analysis of actors goal types, strategies, and successes. In J. A. Daly & J. M. Wiemann (Eds.), Strategic interpersonal communication (pp. 33-90). Hillsdale, NJ: Lawrence Erlbaum. Cupach, W. R., & Metts, S. (1994). Facework. Thousand Oaks, CA: Sage. Dillard, J. P. (1990). A goal-driven model of interpersonal influence. In J. P. Dillard (Ed.), Seeking compliance: The production of interpersonal influence messages (pp. 41-56). Scottsdale, AZ: Gorsuch Scarisbrick. Dillard, J. P. (2004). The goals-plans-action model of interpersonal influence. In J. S. Seiter & R. H. Gass (Eds.), Perspectives on persuasion, social influence, and compliance gaining (pp 185-206). Boston: Allyn & Bacon. Dillard, J. P., Segrin, C., & Harden, J. M. (1989). Primary and secondary goals in the production of interpersonal influence messages. Communication Monographs, 56, 19-38. Dillard, J. P., Wilson, S. R., Tusing, K. J., & Kinney, T. A. (1997). Politeness judgments in personal relationships. Journal of Language and Social Psychology, 16, 297-325. Duck, S., Rutt, D. J., Hurst, M. H., & Strejc, H. (1991). Some evident truths about conversations in everyday relationships: All communications are not created equal. Human Communication Research, 18, 228-267. Fleiss, J. L. (1981). Statistical methods for rates and proportions. New York: John Wiley. Goffman, E. (1967). Interaction ritual: Essays on face-to-face behavior. Chicago: Aldine. Honeycutt, J. M., Cantrill, J. G., & Allen, T. (1992). Memory structures for relational decay: A cognitive test of sequencing of de-escalating actions and stages. Human Communication Research, 18, 528-562. Honeycutt, J. M., Cantrill, J. G., & Greene, R. W. (1989). Memory structures for relational escalation: A cognitive test of the sequencing of relational actions and stages. Human Communication Research, 16, 62-90. Honeycutt, J. M., Cantrill, J. G., Kelly, P., & Lambkin, D. (1998). How do I love thee? Let me consider my options: Cognition, verbal strategies, and the escalation of intimacy. Human Communication Research, 25, 39-63. Ifert, D. E. (2000). Resistance to interpersonal requests: A summary and critique of recent research. In M. E. Roloff (Ed.), Communication yearbook 23 (pp. 125-161). Thousand Oaks, CA: Sage. Jacobs, S., & Jackson, S. (1983). Strategy and structure in conversational influence attempts. Communication Monographs, 50, 285-304. Jones, S., & Wirtz, J. G. (2006). How does the comforting process work? An empirical test of an appraisal-based model of comforting. Human Communication Research, 32, 217-243. Kellermann, K., & Cole, T. (1994). Classifying compliance-gaining messages: Taxonomic disorder and strategic confusion. Communication Theory, 4, 3-60.
Keppel, G., & Wickens, T. D. (2004). Design and analysis: A researchers handbook (4th ed.). Upper Saddle River, NJ: Pearson/Prentice Hall. Knobloch, L. E., & Carpenter-Theune, K. E. (2004). Topic avoidance in developing romantic relationships: Associations with intimacy and relational uncertainty. Communication Research, 31, 173-205. Kunkel, A. D., Wilson, S. R., Olufowote, J., & Robson, S. (2003). Identity implications of influence goals: Initiating, intensifying, and ending romantic relationships. Western Journal of Communication, 67, 382-412. Labov, W., & Fanshel, D. (1977). Therapeutic discourse: Psychotherapy as conversation. New York: Academic Press. Laplante, D., & Ambady, N. (2003). On how things are said: Voice tone, voice intensity, verbal content, and perceptions of politeness. Journal of Language and Social Psychology, 22, 434-441. Mills, S. (2003). Gender and politeness. Cambridge, UK: Cambridge University Press. Mongeau, P. A., Jacobsen, J., & Donerstein, C. (2007). Defining dates and first date goals: Generalizing from undergraduates to single adults. Communication Research, 34, 526-547. Mongeau, P. A., Serewicz, M. C. M., & Therrien, L. F. (2004). Goals for cross-sex first dates: Identification, measurement, and influence of contextual factors. Communication Monographs, 71, 121-147. OKeefe, B. J. (1988). The logic of message design: Individual differences in reasoning about communication. Communication Monographs, 55, 80-104. Rule, B. G., Bisanz, G. L., & Kohn, M. (1985). Anatomy of a persuasion schema: Targets, goals, and strategies. Journal of Personality and Social Psychology, 48, 1127-1140. Samp, J. A., & Solomon, D. H. (1998). Communicative responses to problematic events in close relationships I: The variety and characteristics of goals. Communication Research, 25, 66-95. Schrader, D. C., & Dillard, J. P. (1998). Goals structure and interpersonal influence. Communication Studies, 49, 276-293. Searle, J. R. (1969). Speech acts: An essay in the philosophy of language. Cambridge, UK: Cambridge University Press. Sillars, A., Roberts, L. J., Leonard, K. E., & Dun, T. (2000). Cognition during marital conflict: The relationship of thought and talk. Journal of Social and Personal Relationships, 17, 479-502. Wilson, S. R., Aleman, C. G., & Leatham, G. B. (1998). Identity implications of influence goals: A revised analysis of face-threatening acts and application to seeking compliance with same-sex friends. Human Communication Research, 25, 64-96. Wilson, S. R., & Feng, H. (2007). Interaction goals and message production: Conceptual and methodological developments. In D. Roskos-Ewoldsen & J. Monahan (Eds.), Communication and social cognition (pp. 71-95). Mahwah, NJ: Lawrence Erlbaum. Wilson, S. R., & Kunkel, A. W. (2000). Identity implications of influence goals: Similarities in perceived face threats and facework across sex and close relationships. Journal of Language and Social Psychology, 19, 195-221.
Steven R. Wilson (PhD, Purdue University) is a professor in the Department of Communication at Purdue University. His research interests focus on interpersonal and family communication, persuasion and social influence, conflict and aggression, and politeness and identity management. He is the author of Seeking and Resisting Compliance: Why People Say What They Do When Trying to Influence Others (Sage, 2002) and approximately 50 articles and book chapters on these topics. Adrianne D. Kunkel (PhD, Purdue University) is an associate professor in the Department of Communication Studies at the University of Kansas. Her research interests include emotional support/coping processes in personal relationships and support group settings, romantic (re)definition processes, and sex/gender similarities and differences.
Scott J. Robson (PhD, University of Kansas) is an assistant professor in the Department of Communication Studies at Fort Hays State University. His major research interest is in relationship transformation and maintenance strategies. James O. Olufowote (PhD Purdue University) is an assistant professor in the Department of Communication at Boston College. His research focuses on influence processes, the roles and uses of power, and communication and participation in decision making. His previous works can be found in journals such as Management Communication Quarterly, Western Journal of Communication, and Health Communication. Jordan Soliz (PhD, University of Kansas) is an assistant professor in the Department of Communication Studies at the University of Nebraska-Lincoln. His research interests are in the area of interpersonal, family, and intergroup communication.

Psychology Journal of Language and Social

Uploaded by

Copyright:

Available Formats

Psychology Journal of Language and Social

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Psychology Journal of Language and Social

Uploaded by

Copyright:

Available Formats

Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Identity Implications of Relationship (Re)Definition Goals

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 33

Initiating, Intensifying, and Disengaging From Romantic Relationships

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

34 Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 35

Identity Implications of Relational (Re)definition

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

36 Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 37

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

38 Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 39

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

40 Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 41

Table 1 Romantic Relationship Scenarios: Two Per Goal

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

42 Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 43

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

44 Journal of Language and Social Psychology

Table 2 Categories and Examples of Positive and Negative Politeness Strategies

Apologizing for request/imposition

Soliciting partners input in sought directive

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 45

Relationship (Re)Definition Goals and Face Threats

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Type of Face Threat

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

5.40a 2.55a 1.88a 3.43 3.09a

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 47

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

48 Journal of Language and Social Psychology

Relationship (Re)Definition Goals and Facework

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 49

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

50 Journal of Language and Social Psychology

Perceived Face Threats and Facework

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 51

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

52 Journal of Language and Social Psychology

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 53

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

Wilson et al. / Relationship (Re)Definition Goals and Face Threats 55

Downloaded from http://jls.sagepub.com by Vida Shaghaghi on April 19, 2009

56 Journal of Language and Social Psychology