[Dialogue Macrogame Theory Home Page]

The Status of Analyses in Dialogue Macrogame Theory

The purpose of this section is to identify in part the significance of successful DMT analysis of a particular dialogue.

Why analyze dialogues? There are many research purposes for seeking to understand dialogues better. Linguistic, psychological, sociological, literary, historical and other sorts of research regularly seek to gain the fullest possible knowledge of what happens in dialogue. Each of these fields has multiple perspectives on dialogue, and no single method of analysis is in any sense complete, not even complete relative to a single field. Dialogue game analysis provides a supplement to other methods.

Dialogue and Analysis as Human Activities

Participating in dialogue is a human activity that involves cultural and linguistic conventions, the personalities and knowledge of the participants, and many facts about the social histories and lives of the participants.

Similarly, DMT analysis is a human activity. Most commonly, dialogues are analyzed based on video or audio recordings along with transcripts, and most commonly the participants knew that recordings were being made. Often the analyst thus has access to the interaction that is comparable to that of the original participants. If the participants are strangers to each other or only casual acquaintances, then the analyst is like the silent observer.

Analysis involves subjective human judgment. It is continually answering the question: "What do you think they are doing now?", answering it only relative to the narrow framework of DMT. So the most that a written analysis can achieve is an account of what the analyst thought. For the participants there was a tacit comparable question: "What do I think my counterpart is doing now?" which had to be considered in a much broader way.

The Nature of Completed Analyses

One way to represent the subjectivity of analysis results is to see them all as representations of what was plausible to the analyst, i.e. the analyst saying "Within the DMT framework that was defined before analysis began, I see this as the most plausible account." (If several accounts are equally most plausible, then that set of accounts is the analysts' result.)

The DMT framework is open for revision and extension, and so each analysis must be identified with the particular definitions of games and other constructs that were used.

A DMT analysis of a dialogue is in effect an assertion by the analyst that particular definitions apply to the dialogue in particular ways. The framework is intended to make it unnecessary to consult the analyst in order to determine what assertion an analysis makes. In other words, DMT analysis is designed to produce definite results.

An analysis of a dialogue is called successful if it gives an account for every element of the dialogue. It is partially successful if there are some portions accounted for and some not accounted for.

Experience with applying DMT in this way makes it clear that some dialogues can be fully analyzed, others partially analyzed and some cannot be analyzed even in part. These distinctions, and for partially analyzable dialogues with the identification of the parts, provide a crude differentiation of dialogues into classes. Where analysis succeeds, of course, it provides much more information.

The Linguistic Status of an Analysis

A loose analogy can be made between DMT analyses and judgments of grammaticality. Like grammaticality judgments, DMT analyses are artifacts of technical inquiry. They are thus more like data to be accounted for than like an explanatory account. They say nothing about cause -- why an analysis turned out as it did. The term "Theory" in "Dialogue Macrogame Theory" is used, as in many other frameworks, to designate a descriptive approach that is methodical, systematic and examinable.

The analogy can be extended to the problems encountered in the two activities. Analysts disagree. Culture divides. Context (or the attempt to imagine some null context) influences the activity far beyond examinability. Also, both involve categorization of phenomena that may not bow down to set theory. DMT has all of the problems that are raised with grammaticality judgments, and surely some of its own as well.

Interpersonal Coordination and Macrogames:

Macrogames are essentially collections of goals that are activated and dismissed together rather than individually.

We expect that for some kinds of interaction, especially when social dominance is absent, both parties are effectively in control of the course of the interaction. That means that both parties have the right to propose changes, and both parties have the right to prevent particular changes. For changes in what goals will be pursued, each change must be proposed and ratified. One can imagine that it could be very awkward to propose and ratify each change of goal -- personal and joint -- individually. Grouping goals into sets that are activated and dismissed together facilitates smooth, nearly invisible coordination.

Because the games are conventions of interaction, the particular combinations do not need to be spelled out before they are used.

Because the conventions arise out of repeated use, the goals of the individual parties converge to being complementary.

Because the games account only for collaborative interaction, they include joint goals that are achieved by the pursuit of the complementary goals of the two parties.

The result is potentially smooth coordination and cooperation, in which the management of the course of the interaction is generally outside of the focus of attention. Thus we see that this account of collaborative activity in dialogue depends of conventionality, goals and the grouping of goals.


Given a natural dialogue which has produced a "successful" DMT analysis, what can be said about the significance of that analysis? Several things:

For collections of analyses, including the collection on this website, more can be said:

Back to the Definitions of Particular Games page: [ Definitions of Particular Games: ]

Back to the Abstract Definition of Games page: [ What's a Game? ]

Back to the Entities and Terminology page: [ words and things ]

Send email to Bill Mann: bill_mann@sil.org