Interpersonal Communication: a Process of Rule Negotiation Through Metacommunication

Gloria P. Thomas, Baruch College, City University of New York
Gary F. Soldow, Baruch College, City University of New York
ABSTRACT - The purpose of this discussion is to elaborate upon the notion of process in interpersonal communication. An understanding of process requires attention to the underlying rules which link sequential messages. It is argued that these rules are negotiated through a process referred to as metacommunication.
[ to cite ]:
Gloria P. Thomas and Gary F. Soldow (1985) ,"Interpersonal Communication: a Process of Rule Negotiation Through Metacommunication", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 482-486.

Advances in Consumer Research Volume 12, 1985      Pages 482-486


Gloria P. Thomas, Baruch College, City University of New York

Gary F. Soldow, Baruch College, City University of New York


The purpose of this discussion is to elaborate upon the notion of process in interpersonal communication. An understanding of process requires attention to the underlying rules which link sequential messages. It is argued that these rules are negotiated through a process referred to as metacommunication.


One of the most difficult dimensions of interpersonal communication has to do with the issue of process. The notion of process necessarily requires that each person's message convey information relative to the message which preceded it. Consumer researchers have failed to capture the dynamic, sequential nature of process, instead of leaving it to the province of communication researchers. It is important, however, for consumer researchers to pursue this area because we do have a unique subject matter that is unlikely to occupy the attention of communication researchers.

While process is relevant to all interpersonal encounters, the nature of process depends upon certain underlying rules, and these rules are context bound (e.g., Smith 1984). In effect, then, process refers to a linkage between contiguous messages, and this linkage follows from a rule-governed pattern. Understanding such rule-governed patterns has been the pursuit of various fields of inquiry in keeping with their respective boundaries (e.g., organizational communication, therapeutic communication, family communication).

The purpose of the present discussion is to elaborate on notion of process in interpersonal communication in terms of its relationship to rules. These rules comprise an ongoing aspect of the communication process such that interactants do, in fact, negotiate rules through a somewhat abstract level of communication referred to as metacommunication (Watzlawick, Beavin, and Jackson 1967). Metacommunication literally means "communication about communication", and refers to our ability to convey multiple messages simultaneously such that some of these messages refer to the process itself. It is argued here that our ability to metacommunicate allows us to develop the rules underlying our interactions with little explicit reference to them while, at the same time, exchanging information about some topic. Since the rules are likely to be context specific, it is our concern that we begin to understand and develop a typology for them in terms of our own field of inquiry. Thus, because different rules may be operative in various consumer and non-consumer related settings, rules articulated in a general setting would not necessarily be expected to apply to a buyer-salesperson or family buying interaction.

In effect, we are proposing a dual system of communication in which, one, information is exchanged and, two, rules are negotiated. The negotiation of these rules is what allows for the unfolding of the information in an orderly manner. The discussion begins with an illustrative dialogue between two consumers. The paper then proceeds by (a) defining the terms process and rules as they apply to interpersonal communication and tb) discussing the ways in which process and rules are interconnected. This is followed by an examination of the concept of metacommunication and the role it plays relative to communication processes and rules.

An Illustrative Example

Consider the following two dialogues:

Dialogue 1:

Person A: "I want to buy a Mercedes."

Person B: "Have you seen the interest rates?"

Dialogue 2:

Person A: "Have you seen the interest rates?"

Person B: "I want to buy a Mercedes."

To begin, it should be noted that these dialogues could occur between any number of dyads such a two friends, a husband and wife, two strangers, or a buyer and salesperson. The assessment of these dialogues, as suggested, must take into account which context in which they occur. Thus, while we could theoretically consider the relationship between process and rules in terms of any exchange, the above example is intended to highlight an exchange between a salesperson and buyer which, even at an intuitive level, is different from that same exchange between two friends.

Were these two dialogues to be compared in terms of the verbal content, there would be no difference between them. The exchange of information could be measured and quantified, but the measurement would not provide any indicator of the process. The two dialogues, apart from their identical information, are very different. The most significant distinction occurs in the realm of process. The salient issue here is to understand how and why these two dialogues are different.

Intuitively, the first dialogue seems to be reasonable. That is, a statement indicating a desire to own a Mercedes might logically be followed by an inquiry as to awareness of interest rates. The second dialogue, on the other hand, while not nonsensical, does not appear to have the straightforward underlying logic. An inquiry about interest rates does not logically lead into a statement of a desire to own a Mercedes. It can be seen, therefore, that the difference between these dialogues is meaningful in terms of the juxtapositions of their contiguous statements. Necessarily, this implies the two concerns that are related to metacommunications: process and rules. Each will be discussed below.


While consumer researchers have demonstrated a continuing interest in the interpersonal communication process through attempts to conceptualize (e.g., Hulbert and Capon 1972), to review (e.g., Capon, Holbrook and Hulbert 1977) and to empirically investigate the area (e.g., Graham 1982), our knowledge remains somewhat fragmented and incomplete. Although studies have addressed everything from verbal content (e.g., Holbrook and O'Shaugnessy 1975) to communicate style (e.g., Sheth 1975), timing (e.g., Chapple and Donald 1947), and nonverbal behavior (e.g., Grikscheit and Crissy 1973), they have failed to capture the dynamic, sequential nature of the process.

In order to study process, one can represent interactions as a series of messages: a1b1a2b2...anbn, where the ai represent the messages of person A and the bi represent the messages of person B. Many static coding schemes attend to both A and B, but the analysis results in portrayals such that A = S ai and B = S bi. While the ai and the bi can represent anything from giving suggestions, asking questions, indicating product attributes , etc., such a portrayal views persons A and B as the sum of their own individual behaviors. As Watzlawick and Beavin (1967) explain, however, "in such cases, any explanation of A's behavior must be intrapersonal" (p. 6), even if the same analysis is followed for person B. Thus if a scheme such as Bales Interactions Process Analysis (1950) is applied .o the above dialogues, the differences between the dialogues is not depicted. Although Bales' scheme is described as a "process" analysis, it has typically been employed to count the frequency of various types of messages to establish summary statistics such as the ratio of instrumental to socioeconomic acts. These indices, however are still rooted in a summary of intra-individual behaviors and thus miss the alternating sequence of the interaction (Rogers and Farace 1975).

A variant of this approach comes closer to capturing the interactive quality of communication. This variant views some criterion behavior of A or B as a function of the other's communication. Thus, in the above notation, B's behavior would be represented as follows: B = f (S ai). Note, however, that, although this variant could be said to be interpersonal, it does not fully capture the sequential nature of the process.

In order to develop a model of the communication process in the true sense of an interpersonal process, A and B have to be seen as interacting entities wherein messages are exchanged and alternated. As such, the analysis must view the interaction as a series of events which progress through time:

a1/b1,b1/a2,a2/b2, (Mark 1970)

Rather than looking at the exchange in two message units, the exchange can also be viewed as a continuous stream as follows:

a1b1a2b2a3b3...anbn (Watzlawick and Beavin 1967, p 6)

Viewing communication in this way is a major step to conceptualizing the process. In other words, an analysis of at relative to bt-1 as well as to bt leads us out of the static mode of simply counting the frequency of behaviors. But, turning again to the above dialogue, it can be seen that, while the sequential nature of the two dialogues is clearly different, it is still not clear as to how or why they are different. What is missing are the rules that relates at to bt-1.


Despite the apparent general agreement that the communication process is rule-governed, the issue of what constitutes a communication rule is rarely addressed explicitly (Hymes 1980). While scholars have defined rules in a variety of different ways, communication scholars tend to view rules as "statements of regularity" which are (1) "followable", (2) prescriptive, and (3) contextual (Toulmin 1974).

The requirement of followability implies that a rule may or may not be followed. As such, followability serves to differentiate the rule from another type of statement of regularity--the scientific law. One cannot fail to follow a scientific law. Water, for example, cannot fail to follow the law which says that it will boil at 212E F. In contrast, a person can fail to follow a communication rule such as "questions call for related responses."

The second requirement, that rules are prescriptive, also serves to differentiate the rule from the scientific law. The requirement that rules are prescriptive implies that behaviors can be evaluated relative to rules. Thus, ignoring a question typically results in some sort of negative evaluation. One cannot, on the other hand, criticize or reward water for boiling at 212E.

The final requirement (that rules are contextual) implies that, while communication rules tend to apply in similar situations, they do not necessarily apply across all contexts (Gottlieb 1968). In interpersonal communication, this situational context may refer to the relationship which exists between the speakers, the settings in which they speak, the topics about which they speak, or even their purpose for speaking (Hymes 1980). Thus, different communication rules may apply between a buyer and a salesperson, between a husband and wife discussing a purchase, and between a husband and wife discussing a non-purchasing topic.

The fact that the rules of communication are contextual has important implications in terms of the eventual articulation of a list of rules. While certain rules may appear to be general (e.g., questions should be followed by related responses), we cannot expect most rules to apply across all contexts. Perhaps as a result of this limitation, a definitive list of communication rules does not exist. The development of such a list is also complicated by the view that communication rules may be implicit or explicit (Shimanoff 1980). In order to capture implicit patterns of regularity, lists of rules would have to be developed empirically and tailored to different contexts.

While their intent was not to develop a list of rules per se, Rogers and Farace (1975) empirically identified patterns of regularity in interactions which were linked to dominance and control. Because these patterns were between contiguous messages, they can be viewed, within this context, as a limited set of rules. According to these "rules", patterns such as abrupt subject changes and interruptions serve to take control of an interaction, while the support of a previous statement, for example, relinquishes control.

A Taxonomy of Rules

Although these exists no definitive list of actual rules, there have been several attempts to taxonomize rule-related behaviors according to various criteria. Typically these criteria are based on the degree to which a behavior consciously follows or does not follow a rule (Shimanoff 1980). One such taxonomy which is relatively inclusive includes the following categories:

1. rule-absent: the behavior is not rule-governed (e.g., blinking an eye).

2. rule-fulfilling (rule-ignorant) behavior: one in which the behavior is rule-governed, but the person in question has no knowledge of the rule.

3. rule-conforming (or rule-error): one in which the person in question has tacit knowledge of the rule but follows (or breaks) it unintentionally.

4. rule-following (rule violation) behavior: a behavior in which the person in question consciously follows (or does not follow) the rule.

5. rule-reflexive behavior: a behavior that consciously follows (or does not follow) a rule as a result of a conscious evaluation of the rule's merit.

This type of taxonomy is quite consistent with the concept of context-bound rules. The negotiation of communication rules within a particular context (i.e., within a particular relationship, for a particular purpose, etc.) involves, among other things, an implicit assessment by both parties about where the behaviors of the other fall within this taxonomy. That is, both parties achieve some sense about the extent to which the rules are followed or not followed intentionally.

In the second illustrative dialogue, for example, there are many different explanations for person B's responding to a statement about-interest rates with a statement about a Mercedes. Depending upon the context, A might view B's comment as "rule-error" behavior (inferring that B unintentionally changed the subject because of some unstated associations made between interest rates and Mercedes in his own thinking. A might also infer that B's response was "rule-violating" (i.e., that B intentionally changed the subject, perhaps as a move to dominate the conversation). The inference made by A would be expected to vary depending upon the type and length of their relationship, whether they were a buyer and salesperson, two friends, etc. In either case, A's response to B will tent to vary depending upon A's interpretation, which in turn will vary as a function of the context.


Shimanoff (1980) argued that communication "requires the use of rules in order to make sense out of otherwise arbitrary representations, rules are central in communicative processes; therefore, above all else, a theory of communication must involve a rule-related explanation...." (p. 230) Our ability to engage in dialogue rests on the existence of mutually shared rules which govern the sequencing of a dialogue. In other words, dialogues must have some system of rules indicating how strings of utterances can be heard as being somewhat connected to one another (e.g., Donohue, Diez. and Hamilton 1984. Ellis, Hamilton, and Aho 1983).

This system of rules is analogous to the system of rules of grammar. Grammatical rules govern how strings of words can be connected to one another to form meaningful units ref erred to as sentences. While many people are not explicitly aware of the rules of grammar, these rules tent to be followed, and violations tent to be noticed. A similar situation exists with respect to the interpersonal communication process where a mutual understanding of the patterns developed in a particular relationship allows messages to be connected to one another such that meaningful dialogue results.

In interpersonal communication, rules of connectedness govern nonverbal as well as verbal sequencing. With respect to regulating the flow of an interaction, paralinguistic cues such as dropping the voice, pausing, or looking to the other serve as a signal that a speaker has finished speaking and is waiting for a response from the other (Exline and Fehr 1979, Harpor, Wiens, and Matarazzo 1978). As Dittman (1977) has argued, "the patterns of conversation which included taking turns and giving feedback to the speaker are superimposed upon language per se" (p. 144). Kent, Davis, and Shapiro (1981) take a similar position when they argue that orderly interaction requires that knowledge of these rules be shared.

There is some empirical evidence for these assertions. Clarke (1975) had subjects reconstruct dialogues in which the order of speaking turns was randomized. Subjects were able to perform this task, and Clarke argued that this lent support to the notion that dialogue is governed by a structure of shared rules. At the same time, Kent, Davis and Shapiro (1981) found that, when the structure of a dialogue was interrupted by banning questions, a third party was not able to reconstruct the dialogue. With respect to the two dialogues posited above, we would hypothesize that, if subjects were told the two utterances were random, upon reconstructing the dialogues, the subjects would show a preference for the first dialogue.

There is also empirical evidence indicating that rules are context-bound (as defined above) in that they tend to be negotiated as a relationship develops. Kent, Davis, and Shapiro (1981), for example, found that banning questions was more disruptive on the dialogue of strangers than on the dialogue of mutual acquaintances. This is suggestive that strangers have to spend more time developing the boundaries and functions of rules than do people with interpersonal experience with one another. This is corroborated by a large body of literature that has found that behaviors directed explicitly at the negotiation process become more simplified and efficient throughout interaction (Ruesch 1973, Scheflen 1963, Vine 1970). It appears that, although there is a need for continual negotiation between interactants regarding the process of their interaction (Duncan and Fiske 1979), the effort devoted to this negotiation diminishes over time (Derlega, Wilson, and Chaiken 1976). (It can be noted that Watzlawick, Beavin, and Jackson 1967, among others, have argued that the explicit negotiation of rules will increase again in distressed relationships.)


What has been posited thus far is that interpersonal communication, as a process, is sequential, and that the elements of the sequential development depend upon one another because they are connected through rules. The interaction process appears to involve a continuous negotiation regarding how the rules will be enacted. Our ability to conduct this negotiation process while simultaneously exchanging information about some topic is argued to be a function of our ability to metacommunicate.

The definition of metacommunication as "communication about communication" can be deceptively simple. Some scholars have argued that metacommunication is crucial to the control of interactions (Watzlawick, Beavin, and Jackson 1967). The essence of the concept is that every instance of communication contains multiple messages. Some of these are explicitly spoken and, as such,become part of the actual information exchanged. Others are implicit in the grammatical form of verbal messages, and still others are nonverbal. The levels of communication which communicate about the negotiation of the interaction process are metacommunicative.

It is probably more typical that metacommunicated messages are communicated through some means other than the explicit verbal content of the dialogue. In most cases, interactants do not need to verbalize about how they are negotiating the process. As a result, it is our ability to communicate simultaneous messages that allows us to negotiate procedural rules without constantly verbalizing about what we are doing.

Consider, for example, the contrast between the two dialogues which referred to interest rates. The dialogues contain identical words, yet the negotiation of procedural rules is quite different. The difference can be seen in the metacommunicated messages. In the first dialogue, the metacommunicative negotiation becomes almost trivial. Person A initiates a conversation by making a statement to which Person B responds with a related question. The second dialogue, however, contains an apparently abrupt change of subject. A change of subject is metacommunicative because it refers to a rule which dictates that questions should be followed by related responses. In this case, the metamessage is that Person B has failed to follow the basic rule. In order to continue the interaction, Person A needs to assess whether this change of subject was rule-ignorant, rule-error, or rule violating (Shimanoff 1980). One way in which Person A might handle this assessment is by a direct statement (e.g., "Why did you change the subject?") . Note that, in this case, A must interrupt the flow of information. If B responds in kind with another direct metacommunicative message (e.g., "I didn't mean to change the subject."), the conversation would no longer be about interest rates and buying intentions, but would evolve into a metacommunicative discussion about the rules of the interaction.

Alternatively, A may negotiate indirectly by changing the subject back to interest rates in order to test B's response. This negotiation also represents metacommunication by conveying the implicit message that Person A will not accept B's change of the subject (i. e., B's failure to abide by an accepted rule). Given this type of message, B must respond by breaking another rule (e.g., failing to answer, changing the subject) or by submitting to A's challenge (e.g., Rogers and Farace 1975). In either caseS B's response will help A to assess B's original intent.

An additional alternative for A would have been to accept B's metacommunicated message without question and to continue to talk about a Mercedes. In this case, the response to B's statement metacommunicates that A is willing to proceed according to B's definition of the rules without questioning B's intent.

In both of these indirect metacommunicative cases, A can communicate about the process without interrupting the flow of information (i.e., verbal content) with explicit reference to the negotiation. This is because in both cases, the metamessage is handled through the form of message relative to the preceding message (e.g., a statement that is nonresponsive to a question). This form dimension of verbal communication is commonly referred to as relational communication (Soldow and Thomas 1984). Rogers and Farace (1975) and Donohue (1981) have shown that the relational dimension of communication plays an important role in establishing control in interactions. Perhaps because of the efficiency of the dual (form and content) message system, interactants rely heavily on the relational dimension for reference to and negotiation about rules. In fact, existing empirically based relational communication coding schemes (e.g., Rogers and Farace 1975) can be viewed as preliminary attempts to establish lists of rules. To the extent that an empirically identified pattern (e.g., change of subject, question non-answer) is associated with a move toward dominance, for example, we can infer an underlying rule that these responses are appropriate when dominance is called for and inaPpropriate when it is not.

An additional channel for handling the metacommunica-messages in the above dialogue would have been nonverbal communication. In fact, were this an actual interaction, nonverbal messages would have contributed to the inference/negotiation process. The nonverbal cues accompanying B's original subject change, for example, would have helped A to make the original determination regarding whether B had ignored or intentionally violated a rule. Person A might have, in turn, relied heavily on nonverbal cues to return a metacommunicative message. By failing to look at B, A might have metacommunicated a lack of acceptance without engaging in an obvious change of subject (e.g., Harper, Wiens, and Matarazzo 1978).

Regardless of the specific options chosen, one would expect the inferences made by A and B at each stage of the negotiation to vary according to the context in which the dialogue takes place. If, for example, A and B are a salesperson and a consumer with no ongoing relationship, the occurrence of an explicit verbal metamessage would probably be less likely. If, on the other hand, A and B represent a distressed couple, we would e. ct more direct metamessages following a rule violation (Watzlawick, Beavin, and Jackson 1967). Another typical effect of A and B's relationship would be on A's inference regarding B's change of subject. If the relationship is an ongoing one in which B has typically dominated, the topic change may represent an established pattern of behavior (an idiosyncratic rule) which A need not further investigate.

It should be clear from this discussion that the processes involved in interpersonal communication are extremely complex. The purpose here was to illuminate this complexity with the eventual hope of providing a framework for understanding different types of consumer-related interactions.


An understanding of interpersonal communication as a sequential process requires an examination of how each message is linked to contiguous messages. It has been argued that rules underlie these linkages. While communication rules represent regularities, they also vary somewhat according to the type and length of the interactants' relationship, the purpose for the interaction, etc. As a result, each interaction involves a negotiation of its own underlying rule structure. What allows interactants to engage in this negotiation is the ability to communicate multiple simultaneous messages to metacommunicate about the process without interrupting the flow of information.

The above conceptualization of interpersonal interaction has several important implications for consumer researchers. First, we cannot rely completely on communication theorists for our knowledge of the interpersonal interaction process. To the extent that rules are context-bound, consumer researchers need to investigate specific communication rules which may exist in consumer-relevant situations. Given preliminary findings that the rules which are negotiated during interactions pertain to dominance and control, this may be particularly important for the study of consumer-salesperson, husband-wife buying, and opinion leader-opinion seeker dyads.

Because the identification of rules is an empirical problem, there is a clear need for future research which attempts to identify these patterns of regularity in communication sequences. The identification process involves initial research in which dialogues are analyzed in order to identify rules. Additional research will then be required to examine the extent to which these rules vary according to context. Eventually, we may be able to identify the dominant member of various consumer dyads simply by examining underlying patterns of interaction. In addition, we might expect findings that certain patterns of negotiation tend to be used by successful salespeople and opinion leaders.

A crucial requirement of empirical research which attempts to examine communication rules is that it approach interaction as a sequential process. Specifically, this approach entails the analysis of each message (at) relative to its contiguous messages (bt-1) and (bt) in order to identify implicit patterns.

Methods which are limited to counting explicitly stated messages cannot reveal implicit metacommunicative messages, and thus miss the implicit negotiation patterns which underlie the process. As a result, research which relies on static coding schemes can never capture the dynamic process through which messages are linked.


Bales, Robert F. (1950), Interaction Process Analysis, Cambridge, Sass.: Addison Wesley.

Burgoon, Judee K., David B. Buller, Jerold L. Hale, and Mark A. deTurck (1984), "Relational Messages Associated with Nonverbal Behaviors," Human Communication Research, 10 (Spring), 351-378.

Capon, Noel, Morris Holbrook, and James Hulbert (1977), "Selling Processes and Buyer Behavior," in Arch G. Woodside, Peter D. Bennett, and Jagdish N. Sheth (Eds.), Consumer and Industrial Buying Behavior, New York: North-Holland.

Chapple, Eliot D. and Gordon Donald (1947), "An Evaluation of Department Store Sales People by the Interactive Chronograph," Journal of Marketing, 12 (October), 173-185.

Clarke, David (1975), "The Use and Recognition of Sequential Structure in Dialogue," British Journal of Social and Clinical Psychology, 14 (April), 333-339.

Derlega, Valerian J., Midge Wilson, and Alan L. Chaiken (1976), "Friendship and Disclosure Reciprocity," Journal of Personality and Social Psychology, 34, 572-578.

Dittman, Allen T. (1972), "Developmental Factors in Conversational Behavior," Journal of Communication, 22 (4), 404-423.

Donohue, William A. (1981), "Development of a Model of Rule Use in Negotiation Interaction," Communication Monographs, 48 (June), 106-120.

Donohue, William A., Mary E. Diez, and Mark Hamilton (1984), "Coding Naturalistic Negotiation Interaction, Human Communication Research, 10 (Spring), 403-426.

Duncan, Starkey and Donald W. Fiske (1979), "Dynamic Patterning in Conversation," American Scientist, 67 (1), 90-98.

Ellis, Donald G., Mark Hamilton, and Lynn Aho (1983), "Some Issues in Conversation Coherence," Human Communi- cation Research, 9 (Spring), 267-282.

Exline, Ralph V. and 8. J. Fehr (1978), "Applications of Semiosis to the Study of Visual Interaction," in Aron U. Siegman and Stanley Feldstein (Eds.), Nonverbal Behavior and Communication, Hillsdale, N. J., Lawrence Erlbaum.

Gottlieb, Gidon (1968) Logic of Choice: An Investigation of the Concepts of Rule and Rationality, New York: MacMillan

Graham, John L. (1982), "Facial Expression and Marketing Negotiations," Proceedings, American Marketing Association Educator's Conference.

Grikscheit, Gary M. and William J. E. Crissy (1973), "Improving Interpersonal Communication Skill," MSU Business Topics, (Autumn), 63-68.

Harper, Robert G., Arthur N. Wiens, and Joseph D. Matarazzo (1978), Nonverbal Communication: The State of the Art, New York: John Wiley.

Holbrook, Morris B. and John O'Shaughnessy (1975), "Influence Processes in Interpersonal Persuasion," Proceedings, Association for Consumer Research.

Hulbert, James and Noel Capon (1972), "Interpersonal Communication in Marketing," Journal of Marketing Re- search, 9 (February), 27-34.

Hymes, Dell (1980), "Forward," in Susan B. Shimanoff, Communication Rules: Theory and Research, Beverly Hills: Sage.

Mark, R. A. (1971), "Coding Communication at the Relationship Level," Journal of Communication, 21, 221-232.

Rogers, L. Edna and Richard V. Farace (1975), "Analysis of Relational Communication in Dyads," Human Communication Research, 1 (June), 222-239.

Ruesch, Jurgen (1973), Therapeutic Communication, New York, Norton.

Scheflen, Albert E. (1963), "Communication and Regulation in Psychotherapy," Psychiatry, 26, 126-136.

Sheth, Jagdish (1975), "Buyer-Seller Interaction: A Conceptual Framework," Proceedings, Association for Consumer Research, 382-386.

Shimanoff, Susan B. (1980), Communication Rules: Theory and Research, Beverly Hills: Sage.

Smith, Mary J. (1984), "Contingency Rules Theory, Context, and Compliance Behaviors," Human Communication Research, 10 (Summer), 489-512.

Soldow, Gary F. and Gloria P. Thomas (1984), "Relational Communication: Form Versus Content in the Sales Interaction," Journal of Marketing, 48 (Winter), 84-93.

Toulmin, Stephen E. (1974), "Rules and Their Relevance to Human Behavior," in Theodore Mischel (Ed.), Understanding Human Behavior, Oxford: Blackwell.

Vine, Ian (1970), "Communication by Facial Visual Signals," in J. H. Crook (Ed.), Social Behavior in Birds and Mammals: Essays on the Social Ethology of Animals and Man, London: Academic Press.

Watzlawick, Paul and Janel Beavin (1967), "Some Formal Aspects of Communication," The American Behavioral Scientist, 10 (8), 4-8.

Watzlawick, Paul, Janel Beavin, and Don Jackson (1967), Pragmatics of Human Communication, New York: Norton.