�Flaming� and Other Problematic Messages: (original) (raw)

An Interactional Reconceptualization of "Flaming"
and Other Problematic Messages

Patrick B. O'Sullivan
Department of Communication 4480
Illinois State University
Normal, IL 61790
(309) 438-2688
posull@ilstu.edu

Andrew J. Flanagin
Department of Communication
University of California, Santa Barbara
Santa Barbara, CA 93106
(805) 893-7892
flanagin@sscf.ucsb.edu

The authors would like to acknowledge the contributions of Marcia Haddigan to this manuscript.

Abstract

Among the many online communication phenomena currently attracting attention, one of particular interest to scholars and lay commentators alike is "flaming." Several researchers have examined instances of "flaming"--hostile and aggressive interactions via text-based computer mediated communication (CMC) channels (e.g., email, discussion groups, and chat rooms)--and have proposed theoretical frameworks to explain possible causes for such behavior. However, a stable and precise definition of "flaming" has yet to be established and there is little consensus on why flaming occurs. We argue that current definitions use imprecise terms and ambiguous examples that fail to recognize multiple possible perspectives in the situated and evolving nature of message interpretations. Consequently, we propose an interactional-normative framework that focuses on interpretations of messages from multiple perspectives in the context of norms of appropriateness. Using the flaming issue as a starting point, this framework is intended as a tool for approaching an array of problematic interactions, online and offline, that include flaming as well as interactions that have been imprecisely labeled "flaming." We discuss tThe implications of this framework for flaming research and the study of other problematic interactions. are.discussed

Introduction

Computer-mediated communication (CMC) is becoming more essential for work and social interaction. Electronic means of communication have been credited with extending the number and variety of people involved in organizational decisions (Huber, 1990; Sproull & Kiesler, 1991), diminishing temporal and physical interaction constraints (Eveland & Bikson, 1988; Kaye & Byrne, 1986), and increasing horizontal and vertical communication in the organization (Hinds & Kiesler, 1995). Although many benefits have been attributed to CMC, there is also a dark side in new mediated communication such as electronic mail and online discussion groups in the form of hostile and aggressive communicative behavior, or what has been dubbed "flaming." This has prompted concerns about whether CMC is beneficial or disruptive in organizations and whether it is constructive or destructive in society.

The importance of CMC and the questions that its use raises have prompted communication scholars to investigate the flaming phenomenon (Dubrovsky, Kiesler, & Sethna, 1991; Kayany, 1998; Lea, O'Shea, Fung, & Spears, 1992; Lerner, 1996; Spears & Lea, 1992; Walther, Anderson, & Park, 1994). A general consensus is that flaming consists of aggressive or hostile communication occurring via computer-mediated channels. However, assessment of flaming research indicates that definitions, when offered, are imprecise within, and inconsistent across, research projects (Lea et al., 1992). Nevertheless, scholars have collectively pushed forward to try to quantify levels of flaming as well as to provide explanations and possible remedies. The lack of a clear and consistent conceptual and operational definition of the concept, however, leaves a notable void in this body of work.

Consequently, there is a pressing need for greater precision in defining flaming and for developing explanations for such behavior based on solid conceptual and operational grounds. Ambiguity surrounding communicative behavior that is perceived as aggressive and hostile can result in a wide array of problems, possibly with substantial negative social and relational consequences. For example, interpersonal conflict may result from discrepant views of what constitutes antisocial behavior. Within organizations the need might be even more acute: Misunderstandings regarding communication behavior can result in personnel discord, splintered team efforts, and legal liability, all of which are clearly detrimental to organizations' ongoing success.

Several scholars have argued that there are specific characteristics of computer-mediated channels that might contribute to the incidence of flaming and other problematic online interactions (Kiesler, Siegel, & McGuire, 1984; Siegel, Dubrovsky, Kiesler, & McGuire, 1986; Sproull & Kiesler, 1991). Although others have questioned both the prevalence of flaming (Lea et al., 1992) and technologically deterministic explanations for it (Walther et al., 1994), the emphasis has remained on flaming behavior as a uniquely computer-mediated phenomenon. However, limiting the notion of flaming behavior to CMC suggests that interactions via computer-based communication systems are so distinctive from the rest of human communication that they have no linkage to other social phenomena. Furthermore, doing so neglects an accumulated body of knowledge on the nature of human interaction beyond mediated channels.

The framework proposed here assumes that what is labeled "flaming" can occur in interactions via other mediated channels (e.g., letters, telephone, fax) or via face-to-face communication. Although conceptualizations of flaming emerged from experience in online interactions, considering it aside from any specific channel (even while acknowledging the role of various channels) allows conceptual linkage to a substantial body of accumulated research on human communication that is useful for forming a precise understanding of the phenomenon.

To develop a framework of problematic interactions, we proceed by first reviewing and critiquing existing conceptualizations of flaming behavior, from both the popular and academic literatures. Based on this critique, we present an interactional-normative framework for problematic interaction. In doing so, we contextualize flaming behavior within communication behavior broadly and situate flaming within the context of problematic interactions online and offline, thereby enriching our understanding of its antecedents and framing its consequences. Finally, we discuss implications of this framework for theoretical and applied issues. Our overarching goal is to provide a framework for more precise conceptualizations of various types of problematic interactions, including flaming, which can be a foundation for continuing research.

Current conceptualizations of "flaming"

Lay definitions. "Flaming" as a concept emerged from discourse surrounding the online community and has been discussed widely in popular and trade journalistic outlets. Common among lay observers is a representation of flaming as a highly negative message that functions something like a metaphorical flamethrower that the sender uses to roast the receiver verbally. Accordingly, flames have been characterized as "incendiary messages" (Tamosaitis, 1991) and "inflammatory remarks" (Bernthal, 1995). Typical descriptions represent flaming as "scathingly critical personal messages" (Cosentio, 1994), "rude or insulting" messages (Schrage, 1997), "vicious attacks" (Dvorak, 1994), "nasty and often profane diatribe" (Chapman, 1995), "derisive commentary" (Tamosaitis, 1991), "vitriolic online exchanges, poison pen letters" (Dery, 1993), or "derogatory, obscene or inappropriate" language (Seabrook, 1994). Stewart (1991) identified "overheated prose" and "brusque putdowns, off color puns, and anonymous gripes" as flames and Tamosaitis (1991) argued that flames are "not constructive criticism nor heated debate."

Whatever the definition, most commentators agree that flaming is an intentional act that occurs via computer-mediated channels. For example, Seabrook (1994) described flaming behavior as "premeditated insults" and Tamosaitis (1991) viewed flaming as being done "purposively" and describes a sender as "someone who delights in inciting trouble." Furthermore, Seabrook (1994) stated that flaming is "a form of speech unique to online communication" (p. 70), although Stewart (1991) was more general, limiting his conception of flaming to "rabid, abusive, or otherwise overexuberant outbursts sent via computer" (p. 26).

Academic conceptualizations. Definitions of flaming in the academic literature follow popular characterizations closely in that hostile language is an essential element of a flame. For example, flaming is characterized as "direct, sometimes gratuitous, criticism" (Deuel, 1996), as an insult (Herring, 1996), as "hostile verbal behavior" (Thompsen & Foulger, 1996), and as "a hostile, provocative posting" (Kollock & Smith, 1996). Flaming episodes are similarly described as "the hostile expression of strong emotions and feelings" (Lea et al., 1992). However, others characterize flaming in more abstract terms. For example, Parks and Floyd (1996) describe flaming as "verbal aggression," "blunt disclosure," and "nonconforming behavior" and Korenman and Wyatt (1996) characterize flaming as "emotional outbursts." For Thompsen (1996) flaming is an "antisocial interaction" and for Colomb and Simutis (1996) it is "a form of social aggression."

Most academic discussions of flaming implicitly or explicitly link it to CMC and cyberspace (e.g., Kollock & Smith, 1996; Thompsen, 1996; Thompson & Ahn, 1992; Thompsen & Foulger, 1996). A common argument is that CMC, as compared to face-to-face communication, lacks the capacity to incorporate rich contextual and nonverbal cues into communication, filtering out such features as gesticulation, facial expressions, tone of voice, and external environmental signals. This environment of reduced social cues has been posited to lead to correspondingly reduced social constraint and a reduced impact of social norms. The results of this are believed to include greater equality of participation (Siegel et al., 1986), reduction in status differentials (Dubrovsky, et al., 1991), and group decisions that are more polarized and risky than those arrived at via face-to-face communication (Kiesler et al., 1984). In addition, the disinhibition characteristic of such environments is also widely held to increase flaming (Kiesler et al., 1984; Siegel et al., 1986; Sproull & Kiesler, 1991).

Views that claim that CMC exhibits a diminished capacity to convey rich social cues have been dubbed "systems rationalism" outlooks (Lea, 1991), "cues filtered out" approaches (Culnan & Markus, 1987; Walther, 1992), and "social cues perspective[s]" (Spears & Lea, 1992). Spears and Lea (1992) note that the implication is that computer-mediated communication environments lack social cues, foster difficulties in coordination and feedback, cultivate deindividuation, and promote widespread depersonalization. However, the so-called "liberation" claims and flaming behaviors said to occur within these environments have been challenged on several levels (Culnan & Markus, 1987; Lea et al., 1992; Postmes, Spears, & Lea, 1998; Spears & Lea, 1994; Walther, 1992; Walther et al., 1994). In spite of this, these claims continue to enjoy wide support and to reflect a sentiment of the impersonal character of CMC. As noted by Walther et al. (1994), these findings are commonly referenced by others as indicative of differences between CMC and face-to-face communications, in spite of some dubious conclusions and the discovery of important mitigating factors such as longer-term interactions (Walther, 1992) and anticipated future interaction (Walther, 1994).

Critique of Current Perspectives on Flaming

Current conceptualizations of flaming suffer from several shortcomings that inhibit the development of an encompassing view of the perceived hostile communication acts that occur in human interchange. Specifically, perspectives on flaming suffer from (1) ambiguity in conceptual definitions of flaming, (2) an overemphasis on message content versus message context, (3) an assumption that flaming behavior lacks clear functional value, and (4) a narrow view of flaming behavior as a strictly online phenomenon. These shortcomings are obstacles to developing an understanding of flaming that accommodates its richness as a communication phenomenon. Here we address these issues to establish the basis for a reconceptualization of flaming behavior that is more consistent with actual experiences and that takes into account the rich interplay of the sender, receiver, and third-party interpretation of communication acts.

Definitional ambiguity. Despite abundant commentary, analysis, and research, there is little agreement on what flaming is. Among the existing views, flaming behavior is seen as (a) a direct criticism or blunt disclosure, (b) messages containing hostile language or profanity, or (c) messages that are provocative or nonconforming. As a consequence, current characterizations constitute vague, imprecise, and inconsistent criteria by which to reliably identify what is--and what is not--a "flame."

For example, there is a great deal of difference between direct criticism, blunt disclosures, "hostile" language, and provocative or "nonconforming" messages. Blunt disclosures (presumably negative comments about the message recipient) and direct criticisms can be presented in a very caring manner with the intent to help another or to strengthen a relationship, and as such would be inconsistent with current conceptualizations of a flame. One person's "hostile language" is another person's polite reminder, an attempt at humor, or a poorly worded but well-intended message. In short, although there may be many examples of messages that contain one or more of these elements that most would agree constitute a flame, there are also many messages consistent with these characterizations that would not. We argue that these definitions neither individually nor collectively provide a useful standard to reliably identify a flame and we urge greater definitional clarity while accommodating the inherent richness in the concept.

Reliance on message content. Popular and academic characterizations both assume that a flame can be identified based on the content of the message. According to lay commentators, a message that is highly critical, insulting, or rude--or that uses profanity--is a flame. A corollary assumption is that those message characteristics are in fact recognized by the interactants and can be reliably identified by an observer external to the situation. Thus, the assumption is that it would be clear to anyone that a particular message is "inflammatory" (Bernthal, 1995), "scathingly critical" (Cosentio, 1994), "rude or insulting" (Schrage, 1997), a "vicious attack" (Dvorak, 1994), a "nasty . . . diatribe" (Chapman, 1995), "derisive commentary" (Tamosaitis, 1991), or "derogatory, obscene or inappropriate" (Seabrook, 1994).

Similarly, academic conceptualizations of flaming explicitly or implicitly base their definitions on message content. The assumption is that a message that presents direct or gratuitous criticism (Deuel, 1996) or that contains hostile or aggressive verbal behavior (Colomb & Simuits, 1996; Lea, et al., 1992; Parks & Floyd, 1996; Thompsen & Foulger, 1996) can be recognized as a flame not only by the interactants but by observers external to the situation. This assumption is the basis for content analyses used to assess the frequency of flames in online interaction (e.g., Kayany, 1998).

However, the crucial question is: Whose message interpretation determines the message label? In the vast majority of research on flaming, the determination of whether a message is considered a flame is based upon an outside observer's perspective--the commentator, the researcher, or a coder. A third party's interpretation, however, might be very different from that of the interactants (Mortensen, 1997; Ogden & Richards, 1956) due to lack of access to the wide array of contextual factors that are key to the interactants' message interpretation. It is precisely this context that interactants draw upon to achieve some degree of shared understanding through the communication process.1

There are many types of messages that could contain elements that typically distinguish flaming behavior and still not be properly viewed as flames. Consider, for instance, profanity--an easily identifiable language element that has been assumed to be a key indicator of flaming behavior. Clearly, there are many instances of messages that use profanity that could be considered a flame. However, there also may be instances when profanity used in a message would not be considered a flame. For example, the casual use of profane language between close friends can be a marker of relationship closeness. Friends have been known to address each other with hostile or vulgar terms as a form of play or friendly verbal jousting--words that they would never use casually with someone outside their social network. In such instances, the normative language uses within the group would include--and may even require--the use of profanity.

Flaming behavior is thus most often characterized as antisocial because "many researchers have tended to ignore the influence of local group norms, and consequently defined flaming as antinormative" (Postmes et al, 1998, p. 696). However, acts that might be considered to be flaming by a third party may be seen as "good and desirable" within a specific group in which members share norms of interpretation (Postmes et al., 1998). Outside of the social network, where different interactional norms dominate, the exact same language use would likely hold far different, and far more negative, meanings for both sender and receiver.

Several researchers have in fact recognized that common social categories are important sources of social influence, irrespective of message content, the richness of media, and the cues that they might convey (Lea et al., 1992; Postmes et al., 1998; Spears & Lea, 1992). Consequently, flaming is rightly viewed as "radically context-dependent" (Lea et al., 1992, p. 108) and recognized as group specific and heavily influenced by normative expectations. From this perspective, researchers wisely advocate an explicit consideration of subcultural norms in order to recognize and contextualize antisocial behavior occurring via computer-mediated channels (Lea et al., 1992).

Value judgments of flaming behavior. Flaming is traditionally conceptualized as negative, anti-social, and undesirable. However, particularly in the absence of a clear definition of flaming, applying a value judgment to an entire category of messages can inject biases into observations and interpretations of those observations. Considering too early the issue of "What ought to happen?" can interfere with a clear assessment of the issue of "What is happening?" and can cause one to lose sight of the functions that such messages might serve. Some scholars studying other types of interactions that are generally branded as socially inappropriate or wrong have had to make much the same argument.

For example, Parks (1982) identified what he called the "ideology of intimacy" in research and theorizing on interpersonal interactions that promoted high levels of disclosure, openness, and "honesty" in relationships. He argued that disclosures could in actuality interfere with intimacy, closeness, and happiness and that some degree of privacy and non-disclosure was functional and even beneficial to successful and rewarding relationships. Similarly, Bavelas, Black, Chovil, and Mullett (1990) argued that equivocal messages were not necessarily to be avoided and could be functional and beneficial depending on the context and relationship. Coupland, Wiemann, and Giles (1991) criticized the "pollyana perspective" of some communication scholars, which views miscommunication and other problematic communication as "bad" and a target for reduction and elimination. Even scholars of deception which, like flaming, is widely considered to be negative, destructive, and anti-social, have argued that deception is not necessarily "bad" and, in some situations for some individuals, can be functional and even socially desirable (e.g., Buller & Burgoon, 1994).

Consistent with these perspectives, we approach the issue of flaming (and related types of interactions) with a focus on how it occurs, why it occurs, and what function it serves, rather than with a preconceived value judgment. Just as there may be anti-social motivations for hostile messages, there may be a number of pro-social motivations and outcomes associated with aggressive or hostile messages. For example, harsh language could be used to provoke a reticent individual into a healthy, constructive conflict. A criticism could be used to establish the sender's credibility by demonstrating a willingness to offer critical comments and not just bland, agreeable feedback.

Channels of flaming. Flaming is generally understood exclusively in the context of online interactions. This makes sense if one assumes that online interaction is somehow so distinctive from offline interactions that certain types of interactions are unique to CMC. However, it is too narrow if one accepts the proposition that online interactions are a subset of the larger category of communication and thus share substantial characteristics. This, of course, suggests that "flaming" can occur offline as well as online. Even given the current conceptualizations of the term, messages that convey hostility, profanity, blunt criticism, etc., are found in face-to-face interactions. When experienced face-to-face, we recognize such messages as displays of hostility, anger, impatience, or candidness instead of labeling them as "flaming." This suggests that theorizing and research about communication online would be useful in understanding communication offline--and that the study of communication offline would be useful in understanding communication online.

Channel can play an important, although not necessarily decisive, role in problematic interactions. Mediated channels can convey fewer social cues and therefore less information to guide interpretation (Daft & Lengel, 1984; Short, Williams, & Christie, 1976), which can contribute to misunderstandings. In addition, using mediated channels that convey fewer social cues might provide a sender with a degree of anonymity. Much of the existing scholarship on flaming adopts the view that the resulting sense of deindividuation contributes to behavior that violates social standards (e.g., Joinson, 1998).

However, the focus of our framework is on individuals' intentions and interpretations based on various levels of norms--interpretations that can be shaped by, but not determined by, channel characteristics. Interactional norms likely exist for the use of specific channels in various types of exchanges, recognizing that individuals evaluate, select, and use various channels for specific goals (O'Sullivan, in press). Individuals can unwittingly or intentionally violate norms regarding message content. Channels that convey fewer cues to guide interpretation may contribute to higher incidences of misunderstandings, or may be chosen specifically by an individual to afford a degree of psychological distance from someone whom they are writing about critically (O'Sullivan, in press). Individuals can also unintentionally or purposefully violate norms regarding channel use by using a channel viewed as symbolically inappropriate for the situation (Sitkin, Sutcliffe, & Barrios-Choplin, 1992).

With so many new channels being utilized for wider types of social interaction there may be more instances of norm violation of channel use now than in the past--or perhaps in the future as norms of channel uses stabilize and spread. However, in all of these instances, the individual is responsible for the choices that they make, not the channel. Thus, we seek to avoid the common tendency in much CMC research to invoke a technologically deterministic approach that focuses on channel characteristics as the explanation for channel effects. Instead, we elevate the importance of individuals' perceptions and choices that they make as they select and use various channels for particular interactional goals (O'Sullivan, in press).

Toward An Interactional-Normative Framework of Problematic Interactions

This section outlines a framework for understanding flaming and other problematic interactions that builds on the critique of existing flaming research. The focus is on interactional aspects of message interpretation from multiple perspectives and the role of various levels of norms that are used to guide those interpretations. This approach moves away from trying to identify flames based on specific message elements (e.g., profanity), assumed emotional states (e.g., hostility), or assumed intent (e.g., aggression). The result is a framework that provides more precision in determining what is flaming behavior, based on recognizing variations in norms and expectations among individuals.

The relational nature of communication. Current conceptualizations of flaming appear to be based on a transmission model of communication, insofar as they assume that communication is the transmission of clear information from one individual to another. This is exemplified by conceptualizations of flaming that assume a third party's interpretation of a message will be the same as that of the interactants. However, as many communication scholars have noted, the communication process is far more complex and nuanced than this perspective implies. As Coupland, Wiemann, and Giles (1991) argued, "language use and communication are in fact pervasively and even intrinsically flawed, partial, and problematic" (p. 3). Similarly, Mortensen (1997) stated that "(T)he subject of human understanding is after all, by definition and of necessity, partial, incomplete, and error-prone" (p. 2). The framework proposed here recognizes that communication is inherently problematic and contextualizes flaming as one among many types of problematic interactions.

Communication ambiguity and equivocality have been conceptualized as residing in the sender's intentions, the receiver's interpretations, the message itself (Eisenberg, 1984) or from an interaction of all three (Bantz, 1983; Putnam, 1983; Putnam & Sorenson, 1982; Weick, 1979). This suggests that message construction should be viewed as a relationalprocess, since communication involves at least message sender(s) and receiver(s). Unfortunately, say Putnam and Pacanowsky (1983), not enough attention has been focused on this relational aspect of meaning construction due to a traditionally "functionalist" view of communication processes.

Thus, following foundational concepts grounded in the work of such scholars as Blumer (1969), Mead (1934), and Ogden and Richards (1956), we argue that an outside observer cannot reliably know what the essence of a message means to the sender or the receiver (Bradac, Hopper, & Wiemann, 1989). Furthermore, just as an outside observer might misinterpret the meaning of a particular interaction, the interactants themselves cannot necessarily have perfect knowledge or understanding of one other through communication. Pearce and Cronen (1980; Cronen, Chen, & Pearce, 1988), for example, note that the communication process required a complex coordination of efforts among interactants to determine message meanings. Drawing on prior understandings of constitutive rules and the contexts of the interaction, individuals come to an assessment of the most likely interpretation and determine the appropriate response. Incorporating this assumption helps to develop a more precise conceptualization of flaming while situating it within a set of other types of problematic interactions. Ultimately, we propose to ground flaming behavior within an interactional perspective, with an explicit understanding of normative communication behavior, to better understand flaming and other problematic interactions in online--and face-to-face--communication environments.

Interactional norms. Individuals are guided by communicative norms that help them to convey and interpret messages. Norms can be identified at a cultural, local or group, and relational levels. Grice's (1975) maxims, for example, identify a set of interactional norms applicable to North American culture and Brown and Levinson's (1987) classic analysis of politeness examines in depth "universal" norms of language use to manage face in social discourse. Social etiquette rules (e.g., Miss Manners, see Martin, 1997) also describe behaviors that are appropriate (and by implication those that are inappropriate) in various social situations.

The emergence of computer-based interactions has also given rise to relatively codified rules of online conduct. For example, "netiquette" guides provide standards for acceptable interactional practices when conversing on the web (Crump, 1998; Glassman, 1998; Shea, 1994) and are intended to establish the bounds of what is appropriate and what is inappropriate in a variety of online interactions. In this regard, experience is crucial to recognize online norms and to act appropriately (McLaughlin, Osborne, & Smith, 1995).

Concurrently, and heavily interwoven with cultural norms, norms grounded in local social structures such as a specific organization or other social network can influence individuals' message construction. For example, it might be normative in an organization for managers to use highly formal language when interacting with each other and especially with employees. For another organization it might be normative for managers to use informal language and even engage in teasing and joking with each other and with employees. Of course, formal and informal socialization processes are keys by which an organization conveys to new members the various norms that they are expected to follow (Jablin, 1987). Local interactional norms may be codified into employee manuals or manifested in the unwritten rules that a newcomer learns by trial and error or with the assistance of a veteran organization member. Furthermore, technologies are often both the conduit and the source of socialization processes in organizations (Waldeck & Flanagin, 1999).

Within broader cultural and local norm systems, individuals can also develop norms distinctive to specific relationships. These norms might be consistent or inconsistent with local or cultural norms, and they might not be apparent or decipherable to anyone outside of the relationship. Relational partners bring a set of expectations based on previous relationships and on other influences such as parents, peers, and other sources (e.g., mass media) (Planalp, 1985). One of the central processes of early relationship development is learning and negotiating expected behaviors to increase one's ability to predict the other's behavior (Berger & Bradac, 1982; Berger & Calabrese, 1975). Most likely, there is mutual--though not necessarily equal--influence of each partner on which norms apply within the relationship (Duck & Pittman, 1994). These norms provide a base for individuals to use in interpreting the meaning of each other's messages within the relationship.

Interactional norms, then, serve as a guide to the formation of messages and their interpretation. Messages that are perceived as consistent with norms are deemed appropriate. Of course, a message can be appropriate at one level and inappropriate at another level (such as relational norms that differ from cultural norms). Both the message sender and message recipient presumably evaluate a message's appropriateness or inappropriateness: The sender in the process of creation, and the recipient as he or she strives to draw meaning and understanding from message content.

Norm violations. Although individuals typically avoid norm violations, rules can be broken. There is a range of goals that could motivate someone to intentionally violate interactional norms. For example, one may violate norms to attract attention, to display opposition, or to demonstrate independence. Norms, and their potential violation, can thus be seen as a resource that individuals can use in pursuit of their interactional and relational goals (Burgoon & Hale, 1988).

Alternatively, of course, norm violations can be entirely unintentional. Unintentional violations can be understood as a misalignment of norm sets: Individuals (say, newcomers to a social network) hold their own set of understandings about appropriate and effective communication that may not overlap substantially with those of the social network. Socialization (or trial and error) is the means by which norm sets can become more accurately aligned.

In addition, individuals can vary in their communication competency (Spitzberg & Cupach, 1984, 1989; Wiemann, 1977). For instance, some might be quite able to construct clear messages that are interpreted fairly closely to the sender's intent while others may have a tendency to compose messages that are obscure or that are interpreted differently than the sender intended. Of course, there is a wide range of competencies possible for different types of interactions, in different relationships, using different channels--even within a particular individual as well as across great numbers of people.

The role of normative expectations in "flaming." As already discussed, messages are formed and understood within a framework of interactional norms that evolve over time (Brown & Levinson, 1987; Grice, 1975; Planalp, 1985). Flaming behavior is extremely complex, however, because cultural, local, and relational norms coexist, may conflict, may differ from one person to another, change over time, and differ from one medium to another. For example, the seemingly straightforward statement "That dress looks good on you" could mean that the speaker believes that the woman wearing the dress made a wise fashion choice (advice among friends), that she looks bad (when the wearer know the dress is, in fact, stained or badly wrinkled), or that that she is being sexually harassed (when said by an individual with a history of propositioning female employees). Interpretation would be guided by each person's understanding of the norms he or she perceives as salient in the situation.

The "Interactional Norm Cube"

Based on a consideration of the relational nature of communication, multi-level norms acting to guide interactants' message formation and interpretation, and the consequences of norm violations and expectations, we identify here a framework (the "interactional norm cube") to contextualize communication as it implicates flaming behaviors. Table 1 identifies the possible combinations of message interpretations, as a function of various message sender/receiver and third party viewpoints, and as informed by normative cues. Working from message sender, message receiver, and "objective" third party perspectives, Figure 1 considers messages in terms of their perceived appropriateness (on a continuum from highly appropriate to highly inappropriate) and graphically represents the relationships of three perspectives on a specific message's appropriateness or inappropriateness. The highly inappropriate end of each one of the continua is where flaming behavior has traditionally been located, although this distinction has typically been made from a view of only one perspective at a time (i.e., sender, receiver, or third party).

The consideration of these three perspectives, however, coupled with an understanding of the appropriateness of messages based on their normative context, yields the eight possible conditions represented by the cube in Figure 1 (represented also in Table 1).**2 **This approach departs from past conceptualizations of flaming behavior that are often based on a single perspective without consideration of interactional norms. Thus, this framework is intended to expand our understanding of flaming by an explicit consideration of the interactional richness and normative-contextual cues that guide message formation and delivery as well as reception and interpretation.

The rows in Table 1 correspond to the octants in Figure 1. Table 1 identifies possible combinations of message interpretations and explains the effects of norms on message interpretation (via the comments and examples provided in the table). Each of the eight rows in the table represents a specific communication circumstance in which the interpretation of a message is a function of the sender's, receiver's, and a third party's perception, as influenced by interactional norms. Thus, the rows of Table 1 provide a view of various types of interactions, including flaming behavior, understood in this context. The first three cells of each row illustrate whether each person interprets the message as socially and normativelyappropriate or, alternatively, as a transgression (i.e., inappropriate). Such interpretations have implications both for shared understanding of flaming and other problematic interactions as well as for the outcomes of that behavior (for example, through possible personal, social, or legal repercussions).

The first four rows of Table 1 are all instances when the originator/sender intends the message to be consistent with relevant norms; hence, the message is interpreted as appropriate by the originator/sender in each case. In addition, each of these rows is distinguished by specific combinations of message interpretations (appropriate or transgression) by the recipient and third party members. The last four rows are all instances when the sender intends the message to transgress relevant norms. Again, each of the last four rows is further distinguished by specific combinations of message interpretations (appropriate or transgression) by the recipient and third party members. Each row and corresponding cube quadrant will be described in turn.

The first row of Table 1 represents the situation where the sender, receiver, and a third party all interpret the communicative act as an appropriate messages consistent with each individual's understanding of interactional norms. This represents non-problematic interactions, and corresponds to messages located in Octant A in Figure 1.

Row 2 of Table 1 (corresponding to Octant B in Figure 1) describes the occasion where the sender and receiver believe the message is appropriate but a third party views the message as a transgression. In such instances, sender and receiver share the view that the message is consistent with specific local or relational norms that are not shared by the third party. In practice, this can be illustrated by sarcastic or joking messages that are understood as such by the interactants. In organizations, such messages may involve language that violates policy or legal standards on their face even through they are perceived as appropriate to interactants. Consequently, although a very low likelihood for complaint exists from the interactants themselves, there may be consequences for such violations of policy or legal standards.

Row 3 in Table 1 (corresponding to Octant C in Figure 1) describes the situation where the sender's message is intended to be consistent with appropriate behavioral norms, and is perceived by third parties as consistent with such norms, but the recipient perceives the message as a transgression. This indicates a misalignment of norm sets of the recipient, which could be a result of a misinterpretation due to lack of shared cultural, local, or relational norms. This may represent instances when an individual is oversensitive to evaluations or other comments that are generally viewed as valid and appropriate criticisms. Alternatively, this could result from candid evaluations that are intended as constructive or legitimate, based on the sender's relationship to the receiver. In practice, this could represent such examples as when a manager's accurate and appropriate evaluation of an employee's inferior work product is viewed as a personal attack.

Row 4 (corresponding with Octant D in Figure 1) represents the situation in which the sender intends his or her message to be appropriate but the message is perceived to be inappropriate by both the recipient and a third party. In general, this represents a misalignment of the sender's norms with widely-accepted norms. In practice, this could be illustrated by incidents when a sender is insensitive to how others would interpret a specific message, with possible consequences for creating conflict, particularly because legal or policy norms and expectations have been violated.

Rows five through eight in Table 1 all represent situations where the sender intends that his or her message violates interactional norms. Row 5 (corresponding to Octant E in Figure 1) describes the situation where the sender's intention to violate communication norms is not viewed as a transgression by the recipient nor third parties. In practice, this may be because the sender lacks an understanding of the norms he or she is attempting to violate, uses too high a degree of subtlety, or exhibits communication incompetency. In cases such as this "failed flame," although the sender intended to violate norms it is unlikely that any negative result will follow from the message since it was not perceived as intended. One has not offended someone who takes no offense.

Row 6 (corresponding with Octant F in Figure 1) identifies a situation when a sender intends to violate communication norms but the receiver does not view the message as a violation. This might be due to a lack of sensitivity on the part of the recipient (someone with "thick skin") or to the full meaning of the message escaping the recipient's perception. However, if the message were shared, delivered in a public arena, or detected through monitoring of communications, a third party would view the message as a transgression. In practice, this might represent a "missed flame" that has possible social or legal consequences within some organizational environments. This might describe situations when the sender constructs a message knowing the receiver would not perceive it as a flame but knowing also that others would "get it"). Consequences could be that the sender would face criticism by others in his or her social network or possibly disciplinary action in formal organizations that monitor interactions between employees for harassment or conflict. However, the likelihood of penalties arising from this intended transgression could be minimized because of the recipient's failure to appreciate the violation and likely failure to complain.

Row 7 (corresponding to Octant G in Figure 1) identifies instances when sender and receiver share an understanding that communication norms were violated but the transgression is not apparent to a third party. This illustrates what might be called an "inside flame" shared by sender and receiver but not understood by an outsider who lacks an understanding of the shared local or relational norms salient to the interactants. A comment that, on its face, appears to be innocuous to someone unfamiliar with the specific relational history and norms operating between the sender and receiver could be understood as pointed criticisms or hurtful comments to the interactants themselves. Consequences would likely be limited to the relationship since the transgression is not apparent to those outside the relationship.

Finally, Row 8 (corresponding to Octant H in Figure 1) represents the quintessential "flame." In this case, the sender intends to violate norms, the receiver perceives the message as a transgression, and a third party also recognizes that the message is a violation. Personal, organizational, legal and other consequences may result. In many instances, the consequences would likely be negative for the sender. However, repercussions might not be universally negative, such as when a sender's goal is to get under someone's skin or to demonstrate an unwillingness to follow established norms--or even to hurt another or to end a relationship. Violations can be strategic and functional, even though they are generally viewed as antisocial behavior.

Discussion

The framework proposed here provides a means to more precisely define a true "flame" as well as a means to distinguish flames from a number of other possible types of interaction. Based on the present perspective, then, a "flame" is a message in which the creator/sender intentionally violates (negotiated, evolving, and situated) interactional norms and is perceived as violating those norms by the receiver as well as by third-party observers. This definition recognizes that interpretations emerge from the various interactional norm sets used to evaluate messages. It distinguishes true "flames" (Octant H) from unintentional violations of interactional norms (Octants A-D) as well as from "missed flames," "failed flames," and "inside flames" (Octants E-G). Each of these eight types of problematic interactions has distinctive consequences for personal, relational, and organizational outcomes of the interactions. Clarifying these distinctions contributes to the study of flaming by providing a means to sort out true flames from other problematic interactions and provides an improved opportunity to assess more accurately the incidence, causes, consequences of, and solutions to, a range of problematic interactions--including flaming.

This approach incorporates the sender's intent in problematic interactions. A focus on the sender's intent to flame helps to distinguish attempted flames from other types of problematic interactions in which the sender's intent is not to violate interactional norms. Unintentional violations of norms are not considered flames but instead are viewed as misalignments of norm sets at various levels. These types of episodes are more precisely understood as miscommunication (Coupland, Giles, & Wiemann, 1991; Mortensen, 1994, 1997) rather than flaming.

This approach also recognizes that interactional norms are not imposed intact from the outside but rather emerge through negotiation over time within cultures, social networks, and relationships. This negotiation process can be viewed as a process of realigning norms as perceptions of appropriate and inappropriate behaviors evolve (or are reinforced) as a result of communicative exchanges. The process of "flaming" thus includes the creation, transmission, and interpretation of a message that is perceived as violating norms from multiple perspectives, which can be the start of norm negotiation and evolution. From a sender's viewpoint, the recipient's or third party's reaction to an intended "flame" (or lack of reaction) provides information about the sender's alignment of his or her norm set vis-à-vis others' norm sets. Reaction (or lack of reaction) to a message not intended as a violation also provides information that can guide a sender's assessment of appropriateness. A similar process can occur for the recipient and a third party, resulting in adjustments to norm sets for them as well. Over time, various parties might learn to be more effective in their message construction and interpretation which may result in increased communicative competence about how and when to be consistent with, or to violate, norms.

A contribution of this framework is that it identifies the origins of some problematic interactions and suggests steps to address issues arising from such instances. Viewing unintentional transgressions as norm misalignments suggests the need to reassess and renegotiate understandings about appropriate interactional behavior. It does not suggest, contrary to strategies often employed, that communication should be censored or that individuals should be censured for initiating such messages. Rather, it suggests the need for adjustments in socialization processes. In relationships, adjustments might include explicit discussions about what each partner views as proper and acceptable language, interactional styles, conversation topics, and interactional settings for certain types of interactions. In organizations, adjustments might include individualized coaching for improved communication practices, instituting training programs, and improving orientation processes for new employees such that acceptable, normative communication behavior is made explicit.

Remedies might also include mutual influence on the specific aspects of the guiding norm set. In relationships, both partners might adjust their expectations to accommodate the other's preferences. In organizations, individuals might help shape a particular organization's norms of appropriateness rather than an organization imposing its norm set on individuals. In each case, the goal would be to better align one or more sets of expectations about appropriate interactional behavior toward a common standard. These steps would likely not eliminate "flaming" activities, since some may continue to intentionally violate norms for various goals, but it could reduce the incidence of related, unintentional problematic interactions and avoid confusing non-flames with true flames.

As noted earlier, this framework is intended to apply to non-mediated as well as to mediated interactions. Because CMC and other forms of mediated communication (telephone, letters, answering machines, etc.) are best understood as a subset of human communication, this analysis takes advantage of what we know about human communication and applies it to mediated communication. In addition, what we learn about mediated communication can illuminate antecedents, processes, and outcomes of non-mediated communication. Moreover, the framework also highlights the important role of channel inall interactions (mediated and face-to-face) and further emphasizes the importance of considering interactional channel when studying communication.

What requires further examination is how norms regarding the interactional channels that we use can also contribute to problematic interactions. Although this analysis focused on message content, norms also exist for interactional channels (O'Sullivan, 1997), which likely interact with message content. For example, some may feel that electronic mail is an inappropriate channel for critical messages and that face-to-face is the appropriate way to present criticisms (see, for example, Daft & Lengel, 1984). Consequently, norm violations could occur based on the message and/or the channel, while the channel concurrently shapes the interaction in ways that can contribute to problematic interactions.

There are a number of potential consequences of adopting this framework for the study of flaming. For example, studies examining the frequency of flames in various interactional settings might need to be reassessed in light of the possibility that what have been counted as flames might actually be other types of problematic interactions (e.g., Kayany, 1998). Assessing the frequencies of true flames, as well as other quasi-flames and various types of miscommunication, would be useful in helping to understand more precisely the prevalence of these types of problematic interactions.

This framework could also stimulate rethinking models that seek to explain causes of flaming. For example, the SIDE model (Lea et al., 1992; Postmes, et al., 1998; Spears & Lea, 1992) is one of the more dominant models developed to explain the social implications of CMC. The model emphasizes the interaction between social factors (e.g., social identity) and channel factors (e.g., degree of anonymity of various forms of CMC). The interactional-normative framework proposed here could help ensure that those explanations are focusing on true flames (or at least instances when there is an intent to flame) rather than messages that might have been labeled a flame but are more accurately characterized as a type of miscommunication. In addition, the framework and the assumptions that provide its foundation can serve to re-emphasize the strategic choices than individuals make regarding both message content and channel use (O'Sullivan, in press). As Postmes et al. (1998) noted, although the cognitive aspects of the SIDE model have been examined, the strategic elements have not been fully explored. This approach can help to balance the existing emphasis on channel effects apparent in the SIDE model and provide a direction for elaboration of the strategic elements.

For example, individuals can assess and assert some control over the degree of anonymity or social distance that will exist in the impending interaction through their selection of the channel of interaction as well as what they reveal or disclose. Therefore, the degree of anonymity that might be experienced could be actively shaped by the interaction initiator, which could have consequences for the degree to which that individual feels deindividuated. Similarly, individuals have the ability to shape others' perceptions of them by using a particular channel and revealing, obfuscating, or concealing information about themselves salient to their own and other's perceptions of group status (O'Sullivan, in press). These choices might be crucial for understanding perceptions of interactants' social identify and thus attitudes and behaviors that emerge from those perceptions.

This framework might have utility in other, related, areas as well. By taking this broad interactional and relational perspective, the framework can inform analyses of other types of problematic interactions, such as harassment in its various forms. Harassment has been a highly visible topic in recent years, particularly in the context of workplace relationships between males and females (Bowes-Sperry & Tata, 1999; O'Donohue, Downs, & Yeater, 1998). Social and legal discussions have often focused on the challenges of defining "harassment." Viewing harassment as a communicative issue suggests the need to draw on communication-based analyses.

Applying the present framework, examples of harassment (sexual or otherwise) parallel examples of flaming. As in true flames, true harassment would require the intent to harass, the perception of harassment, and third party perception of the action as harassment. Just as there can be "missed flames," "failed flames," and "inside flames," there can be "missed harassment," "failed harassment," and "inside harassment." The lack of intent to harass directs attention toward misalignment of norm sets, which represents various sources of miscommunication, rather than harassment. This framework thus provides a diagnostic tool for understanding various problematic interactions and a direction for remedies. In addition, it highlights the role that the channel might play in harassment-related interactions in shaping message interpretation and in the concurrent assessment of channel use norms.

Similarly, this framework might be of use in examining hate speech and other types of harmful messages (Leets & Giles, 1997). As with flaming, assuming that a third party's interpretation of message content as harmful is the proper interpretation can result in mischaracterizing such language as "harmful" when it could be something else. For example, although there is language that many would say is unquestionably racist, such language might be used freely within some groups and interpreted benignly as appropriate language play. Applying the interactional-normative framework to such situations could help distinguish more precisely what is actual hate speech from "failed hate speech," "missed hate speech," "inside hate speech" and various types of unintentional miscommunication. Potential consequences or remedies could then be pursued appropriate to the sender's intent and others' interpretations.

Overall, locating communication behavior in an interactional-normative context expands our understanding of flaming behavior to accommodate communication acts as normative and situated. In view of the potential negative outcomes of flaming, the misinterpretations that may arise from communication that is incorrectly perceived as flaming, and the increasing use and importance of mediated channels for contemporary personal and organizational communication, this understanding is both timely and useful.

Endnotes

1 An exception to the focus on independent interpretation of message content is Thompsen�s (1996; Thompsen & Foulger, 1996) approach. Thompsen and Foulger (1996) noted that a message "is not a flame until someone considers it a flame" (p. 227). They explained that what distinguishes a message as a flame is based less on the characteristics of the message than on the meaning attributed to it. However, subsequent discussion of flaming did not explicitly develop the idea that qualities of a message are determined by the interactants themselves, and cannot be identified reliably by third parties.

back to text

2 However, it should be noted that identifying octants is a simplification of real life since the axes are actually continuous measures of appropriateness. The lines indicating the boundary between appropriate and inappropriate messages is dotted to indicate that the distinction is not always clear. Messages can range from highly inappropriate to mildly inappropriate to questionable yet appropriate, to highly appropriate. Messages that would be located close to the dotted lines would be especially interesting to study and likely prompt confusion and perhaps great effort on the part of the recipient to clarify the meaning. These messages may be best understood as possible examples of strategic ambiguity (Eisenberg, 1984) or intentional equivocation (Bavelas et al., 1990). The octants are thus presented as a graphic representation to help to categorize the various message interpretations and to clarify the links between Figure 1 and Table 1.

back to text

References

Bantz, C. R. (1983). Naturalistic research traditions. In L. L. Putnam & M. E. Pacanowsky (Eds.), Communication and organizations: An interpretive approach (pp. 55-72). Beverly Hills: Sage.

Bavelas, J. B., Black, A., Chovil, N., & Mullett, J. (1990). Equivocal communication. Newbury Park, CA: Sage.

Berger, C. R., & Bradac, J. J. (1982). Language and social knowledge: Uncertainty in interpersonal relations. London: Arnold.

Berger, C. R., & Calabrese, R. J. (1975). Some explorations in initial interactions and beyond: Toward a developmental theory of interpersonal communication. Human Communication Research, 1, 99-112.

Bernthal, K. (1995). Online transmission of inflammatory remarks. PC Novice, 6, 39-40.

Blumer, H. (1969). Symbolic interactionism: Perspective and method. Englewood Cliffs, NJ: Prentice-Hall.

Bowes-Sperry, L., & Tata, J. (1999). A multiperspective framework of sexual harassment: Reviewing two decades of research. In G. N. Powell (Ed.), Handbook of gender and work (pp. 263-280). Thousand Oaks, CA: Sage.

Bradac, J. J., Hopper, R., & Wiemann, J. M. (1989). Message effects: Retrospect and prospect. In J. J. Bradac (Ed.), Message effects in communication science (pp. 294-317). Newbury Park: Sage.

Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language use. New York: Cambridge University Press.

Buller, D. B., & Burgoon, J. K. (1994). Deception: Strategic and nonstrategic communication. In J. A. Daly & J. M. Wiemann (Eds.), Strategic interpersonal communication (pp. 191-223). Hillsdale, NJ: Lawrence Erlbaum Associates.

Burgoon, J. K., & Hale, J. L. (1988). Nonverbal expectancy violations: Model elaboration and application to immediacy behaviors. Communication Monographs, 55, 58-79.

Chapman, G. (1995, April 10). Flamers. The New Republic, 13.

Colomb, G. C., & Simutis, J. A. (1996). Visible conversation and academic inquiry: CMC in a cultural diverse classroom. In S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social and cross-cultural perspectives (pp. 203-222). Amsterdam: J. Benjamins.

Cosentino, V. J. (1994). Virtual legality. Byte, 19, 278.

Coupland, N., Giles, H., & Wiemann J. M. (Eds.). (1991). "Miscommunication" and problematic talk. Newbury Park, CA: Sage.

Coupland, N., Wiemann, J. M., & Giles, H. (1991). Talk as "problem" and communication as "miscommunication:" An integrative analysis. In N. Coupland, H. Giles, & J. M. Wiemann (Eds.), "Miscommunication" and problematic talk (pp. 1-17). Newbury Park, CA: Sage.

Cronen, V., Chen, V., & Pearce, W. B. (1988). Coordinated management of meaning: A critical theory. In Y. Y. Kim and W. B. Gudykunst (Eds.), Theories of intercultural communication (pp. 66-98). Newbury Park: Sage.

Crump, E. (1998). Writing online: A student's guide to the Internet and World Wide Web (2nd Ed). Boston: Houghton Mifflin.

Culnan, M. J., & Markus, M. L. (1987). Information technologies. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational communication: An interdisciplinary approach (pp. 420-443). Newbury Park, CA: Sage.

Daft, R. L., Lengel, R. H. (1984). Information richness: A new approach to managerial behavior and organization design. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior (Vol. 6, pp. 191-233). Greenwich, CT: JAI Press.

Dery, M. (1993). Flame wars. South Atlantic Quarterly, 92, 559-568.

Deuel, N. R. (1996). Our passionate response to virtual reality. In S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social and cross-cultural perspectives (pp. 129-146). Amsterdam: J. Benjamins.

Dubrovsky, V. J., Kiesler, S., & Sethna, B. N. (1991). The equalization phenomenon: Status effects in computer-mediated and face-to-face decision-making groups. Human Computer Interaction, 6, 119-146.

Duck, S., & Pittman, G. (1994). Social and personal relationships. In M. L. Knapp & G. R. Miller (Eds.), Handbook of interpersonal communication (pp. 676-695). Thousand Oaks, CA: Sage.

Dvorak, J. C. (1994). The flaming of Madison Ave. Marketing Computers, 14, 22.

Eisenberg, E. M. (1984). Ambiguity as strategy in organizational communication. Communication Monographs, 51, 227-242.

Eveland, J. D., & Bikson, T. K. (1988). Work group structures and computer support: A field experiment. Transactions on Office Information Systems, 6, 354-379.

Glassman, A. (1998). Can I fax you a thank-you note?Berkeley: Berkeley Publishing Group.

Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (eds.), Syntax and semantics, Vol. 3 (pp. 41-58). New York: Academic Press.

Herring, S. C. (1996). Two variants of an electronic message schema. In S. C. Herring, (Ed.), Computer-mediated communication: Linguistic, social and cross-cultural perspectives (pp. 81-106). Amsterdam: J. Benjamins.

Hinds, P., & Kiesler, S. (1995). Communication across boundaries: Work, structure, and use of communication technologies in a large organization. Organization Science, 6, 373-393.

Huber, G. P. (1990). A theory of the effects of advanced information technologies on organizational design, intelligence, and decision making. In J. Fulk & C. W. Steinfield (Eds.), Organizations and communication technology (pp. 237-274). Newbury Park, CA: Sage.

Jablin, F. M. (1987). Organizational entry, assimilation, and exit. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organizational communication (pp. 679-740). Newbury Park, CA: Sage.

Joinson, A. (1998). Causes and implications of disinhibited behavior on the internet. In J. Gackenbach (Ed.), Psychology and the internet: Intrapersonal, interpersonal and transpersonal implications,(pp. 43-60). San Diego, CA: Academic Press.

Kaye, A. R., & Byrne, K. E. (1986). Insights on the implementation of a computer-based message system. Information and Management, 10, 277-284.

Kayany, J. M. (1998). Contexts of uninhibited online behavior: Flaming in social newsgroups on Usenet. Journal of the American Society for Information Science, 49, 1135-1141.

Kiesler, S., Siegel, J., McGuire, T. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123-1134.

Kollock, P., & Smith, M. (1996). Cooperation and conflict in computer communities. In S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social and cross-cultural perspectives (pp. 109-128). Amsterdam: J. Benjamins.

Korenman, J., & Wyatt, N. (1996). Group dynamics in an e-mail forum. In S. C. Herring (Ed.), Computer-mediated communication: Linguistic, social and cross-cultural perspectives (pp. 225-242). Amsterdam: J. Benjamins.

Lea, M. (1991). Rationalist assumptions in cross-media comparisons of computer-mediated communication. Behavior & Information Technology, 10, 153-172.

Lea, M., O'Shea T., Fung, P., & Spears, R. (1992). 'Flaming' in computer- mediated communication: Observations, explanations, implications. In M. Lea (Ed.), Contexts of computer mediated communication(pp. 89-112). New York: Harvester Wheatsheaf.

Lerner, M. H. (1996, November). Flaming to framing: A reconsideration of the relational aspects of negotiation in terms of the demands of asynchronous text-based media. Presented at the annual meeting of the National Communication Association, Chicago, IL.

Leets, L. L., & Giles, H. (1997). Words as weapons: When do they wound? Investigations of harmful speech. Human Communication Research, 24, 260-310

Martin, J. (1997). Miss Manners' basic training: Communication.New York: Crown.

Mead, G. H. (1934). Mind, self, & society. Chicago: University of Chicago Press.

McLaughlin, M. L., Osborne, K. K., & Ellison, N. B. (1997). Virtual community in a telepresence environment. In S. G. Jones (Ed.), Virtual culture: Identity and communication in cybersociety(pp. 146-168). Thousand Oaks, CA: Sage.

Mortensen, C. D. (1994). Problematic communication: The construction of invisible walls. Westport, CT: Praeger.

Mortensen, C. D. (1997). Miscommunication. Thousand Oaks, CA: Sage.

O'Donohue, W., Downs, K., & Yeater, E. A. (1998). Sexual harassment: A review of the literature. Aggression & Violent Behavior, 3, 111-128.

Ogden, C. K., & Richards, I. A. (1956). The meaning of meaning (8th ed). New York: Harcourt, Brace, & World.

O'Sullivan, P. B. (1997, May). When "rich" channels are a poor choice: Information management and channel preference factors.Presented to the International Communication Association's annual conference, Montreal.

O'Sullivan, P. B. (In press). What you don't know won't hurt ME: Impression management functions of communication channels in relationships.Human Communication Research, 26_, xxx-xxx._

Parks, M. (1982). Ideology in interpersonal communication: Off the couch and into the world. In M. Burgoon (Ed.), Communication Yearbook, vol. 5 (pp. 79-107). New Brunswick, NJ: Transaction Books.

Parks, M., & Floyd, K. (1996). Making friends in cyberspace.Journal of Communication, 46(1), 80-97.

Pearce, W. B., & Cronen, V. (1980). Communication, action, and meaning. New York: Praeger.

Planalp, S. (1985). Relational schemata: A test of alternative forms of relational knowledge as guides to communication. Human Communication Research, 12, 3-29.

Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social boundaries: SIDE-effects of computer-mediated communication.Communication Research, 25(6), 689-715.

Putnam, L. L. (1983). The interpretive perspective: An alternative to functionalism. In L. L. Putnam & M. E. Pacanowsky (Eds.),Communication and organizations: An interpretive approach (pp. 31-54). Beverly Hills. CA: Sage.

Putnam, L. L. & Pacanowsky, M. E. (1983). Communication and organizations: An interpretive approach. Beverly Hills, CA: Sage.

Putnam, L. L. & Sorenson, R. L. (1982). Equivocal messages in organizations. Human Communication Research, 8, 114-132.

Schrage, M. (1997). Mr. Bozo, meet Miss Courtesy Worm.Computerworld, 31, 37.

Seabrook, J. (1994). My first flame. The New Yorker, 70, 70-99.

Shea, V. (1994). Netiquette. San Francisco, CA: Albion Books.

Short, J., Williams, E., & Christie, B. (1976). The social psychology of telecommunications. London: John Wiley.

Siegel, J., Dubrovsky, V., Kiesler, S., & McGuire, T. W. (1986). Group processes in computer-mediated communication. Organizational behavior and human decision processes, 37, 157-187.

Sitkin, S. B., Sutcliffe, K. M., & Barrios-Choplin, J. R. (1992). A dual-capacity model of communication media choice in organizations.Human Communication Research, 18_,_ 563-598.

Spears, R., & Lea, M. (1992). Social influence and the influence of the "social" in computer-mediated communication. In M. Lea (Ed.), Contexts of computer-mediated communication(pp. 30-65). London: Harvester-Wheatsheaf.

Spears, R., & Lea, M. (1994). Panacea or panopticon? The hidden power in computer-mediated communication. Communication Research, 21, 427-459.

Sproull, L., & Kiesler, S. (1991). Connections: New ways of working in the networked organization. Cambridge: MIT Press.

Spitzberg, B. H., & Cupach, W. R. (1984). Interpersonal communication competence . Beverly Hills, CA: Sage Publications.

Spitzberg, B. H., & Cupach, W. R. (1989). Handbook of interpersonal competence research. New York: Springer-Verlag.

Stewart, D. (1991, September). Flame throwers: Why the heated bursts on your computer network? Omni, 13, 26.

Tamosaitis, N. (1991). Getting flamed isn't funny. Computer Life, 1, 207-208.

Thompsen, P. A. (1996). What's fueling the flames in cyberspace: A social influence model. In L. Strate, R. Jacobson, & S. B. Gibson (Eds.), Communication and cyberspace: Social interaction in an electronic environment (pp. 297-315). Cresskill, NJ: Hampton Press.

Thompson, P. A., & Ahn, D. K. (1992). To be or not to be: An exploration of E-Prime, copula deletion and flaming in electronic mail. ETC: A Review of General Semantics, 49, 146-164.

Thompsen, P. A., & Fougler, D. A. (1996). Effects of pictographs and quoting on flaming in electronic mail. Computers in Human Behavior, 12, 225-243.

Waldeck, J., & Flanagin, A. J. (1999, November). Technology use and organizational socialization. Presented at the annual meeting of the National Communication Association, Chicago, IL.

Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19, 52-90.

Walther, J. B. (1994). Anticipated ongoing interaction versus channel effects on relational communication in computer-mediated interaction. Human Communication Research, 20, 473-501.

Walther, J. B., Anderson, J. F., & Park, D. W. (1994). Interpersonal effects in computer-mediated communication: A meta-analysis of social and antisocial communication. Communication Research, 21, 460-487.

Weick, K. (1979). The social psychology of organizing(2nd ed.). Reading, MA: Addison-Wesley.

Wiemann, J. M. (1977). Explication and test of a model of communicative competence. Human Communication Research, 3, 195-213.

Figure 1. The "Interactional Norm Cube":

Note: The numerical labels of each octant correspond with the "Figure 1 Octant" column in Table 1, which provides details on each octant. Table 1. Message interpretation as a function of sender/receiver/third party perspective and normative context.

back to text

Table 1: Message Interpretation from Multiple Perspectives