Background and motivation

Action and reflection are two central concepts in information systems. People act until they experience a breakdown, which then leads them to reflect upon their activities (Schön, 1983). To improve the activities, reflections on actions may be anticipatory (before the action), contemporaneous (during the action), or retrospective (after the action). The language-action perspective (LAP) on communication modeling is one of the theoretical orientations that emphasize the importance of action and reflection in designing useable, useful, and legitimate information systems (Goldkuhl & Lyytinen, 1982; Winograd & Flores, 1986). In this perspective, communication is viewed as action. To conceptualize communication actions, most approaches have considered concepts from the speech-act theory (Austin, 1962; Searle, 1969; Habermas, 1984). Speech act refers to an utterance and represents the smallest meaningful unit of communication. Speech-act-based modeling has also been controversially discussed (Suchman, 1994; Winograd 1994; Ljungberg & Holm, 1996).

Another promising concept for conceptualizing communication is the concept of genre, which has been successfully applied in information systems (e.g., Brown & Duguid, 1994; Orlikowski & Yates, 1994, 2002; Clarke, 1996; Ljungberg, 1997; Bergquist & Ljungberg, 1999; Päivärinta, 2001). The concept of genre represents a meaningful pattern of communication, such as a business letter or an application form, which consists of a sequence of speech acts. Genres are socially constructed (Berger & Luckmann, 1966; Luckmann, 2001): they may consequently be socially more or less acceptable or contested. Thus, reflections on genres become more important the more the differences in technological standards, social values, norms, and interests in global contexts interfere within the sphere of genre-based modeling of communication.

In fact, many researchers have articulated the need for systematically reflecting on many issues of information systems (e.g., Hirschheim & Klein, 1994; Mathiassen, 1998; Klein & Hirschheim, 2001; Ulrich, 2001). They argue that any information system entails a multitude of assumptions and claims and serves some interest at the expense of the others. Thus, discursive concepts have been suggested, to promote reflections and discussions among all stakeholders for reaching mutual understanding about the desired features of a system. In extending previous discursive approaches, Yetim (2005) suggests a meta-communication model that provides a wider spectrum of concepts for reflections. However, reflective meta-communication processes have not yet been thoroughly discussed in relation to the concept of genre as a means for structuring (organizational) communication. An exception is the work of Päivärinta (2001), who discusses the potential of genre concept with respect to promoting critical orientation in systems development communication. This work, however, is limited in its spectrum of concepts for reflections. In another empirical work, Yates et al. (1999) describe how genres are explicitly and implicitly structured in practice, although they do not deal with the issue of whether the processes and thus the resultant genre can be regarded as legitimate.

This paper focuses on the concept of communicative genre and addresses the issue of how meta-communication processes guided by discursive-ethical principles can promote a rational and legitimate definition, design, and structuring of genres. It illustrates how a broad concept of rationality that also includes ethical and moral rationality as advocated by Habermas (1984, 1993) can be brought into the process. This paper contends that this fills a gap within the traditional genre research.

Let us first illustrate some global challenges and also justify why we need a discursive-ethical approach to systematically reflecting when structuring genres. Consider as an example the development of a global scholarly communication system with its communicative genres such as 'call for papers' and 'reviews' (Päivärinta, 2001). It is well known that neither establishing some new communication features nor changing existing ones (e.g., changing call for papers or review processes) is alone a task of the system developers, rather it requires the discussion of those involved within corresponding academic community. When developing, for example, a call for papers, the academic community may choose between three potential options: (a) developing a single version of the call for papers by considering only the conventions of one culture (e.g., written in English; a US local time for submission deadline); (b) creating several culture-specific versions (e.g., written in several languages; country specific submission deadlines); or (c) finding 'the best' compromise solution among alternatives. These orientations can be characterized in this sequence as trans-, multi-, and intercultural orientations in design (Yetim, 1998).

The 'free choice' of any of these potential options may be viewed as less problematic in this case but may, however, be legally and/or morally unacceptable in some other cases, for example, in the case of the development of a résumé template for job applications. Whereas 'age', 'place of birth', and 'nationality' are not obligatory elements of a résumé in the U.S.A., they are required for decisions in most other countries. Hence, system developers and other stakeholders may become aware that such situations pose more serious issues than developing a call for papers, including such issues as: What fields should be mandatory within the template? Can its design be multiculturally oriented, that is, with differing mandatory fields for each culture? Who should decide, and in what sort of decision process? And so on.

To deal with these kinds of issues, principles of discourse ethics advanced by Habermas (1990, 1993) can provide an orientation since the main concern of discourse ethics is to legitimize norms. Moreover, discourse ethics claims to be universalistic – applicable to more than one specific culture or epoch. This paper uses the meta-communication concepts – described earlier (in Yetim, 2005) – and the discourse-ethical principles to illustrate how they could add value to previous genre research.

This paper claims to make the following contributions: firstly, it provides a wider spectrum of concepts for critically reflecting on and discursive evaluation of the content and structures of genres and their instantiations. These concepts can also be used to pursue organizational goals in a strategic manner without considering discourse-ethical idealizations. Secondly, the paper contributes to genre research by illustrating how different kinds of meta-communication (ex ante, in-action, and ex post) can be used to legitimate genres in a manner compatible with discourse ethics. It illustrates the discourse-ethical viewpoint concerning the legitimacy of genre structuring processes and thus also the legitimacy of resultant norms and contents of communication, especially in global contexts. Such a meta-communication process has not yet been thoroughly discussed in relation to the concept of genre as a means for structuring (organizational) communication.

This paper is organized as follows: we first present the discourse-based meta-communication model for reflection. Then, we briefly introduce the basic concepts of genre theory and also illustrate usage of the model with an example. In addition, we describe some limitations of genre research and illustrate the discourse-ethical aspects of legitimating norms of genres in global contexts through different types of meta-communication. In the subsequent discussion section, we compare our approach with related approaches and also reflect on its implications for information systems. Finally, some conclusions are offered.

A discourse-based meta-communication model for reflection

Actions can be communicative or non-communicative. Accordingly, reflections can refer to communicative as well as non-communicative actions. Reflections can take place in an individual's mind, but also be dialogical. The term 'meta-communication' used here implies dialogical reflections among actors about communications. Such reflections can take place before, during, and after a communication action. Accordingly, three types of meta-communication are distinguished (Yetim, 2005):

1. ex ante meta-communication, taking place before an action;
2. meta-communication-in-action, taking place during an action; and
3. ex post meta-communication, taking place after an action.

The meta-communication model, which can be used for reflecting in these three situations, is mainly based on Habermas' (1984) discourse theory. Habermas distinguishes between communication action and discourse. He argues that, when communicating, actors raise several validity claims such as comprehensibility, sincerity, etc., and that a communication breaks down when any validity claim becomes problematic. This leads to a reflective mode of communication, that is, to a discourse about the controversial claim. At the reflective discourse level, actors attempt to vindicate or criticize contested validity claims through arguments.

Figure 1 shows how these ideas are reflected in the architecture of a discourse-based meta-communication model (see also Hoppenbrouwers & Weigand, 2000). Whereas in Habermas' discourse theory the discourse level is entered to reflect on the communication action layer, in this architecture the meta-communication layer is entered to communicate about the communication action layer. In order to provide a system for reflections at the meta-communication layer, two levels are further distinguished: the conversation for clarification level and the discourse level. The idea is that the conversation for clarification level provides a structure for systematic conversations about validity claims or potential communication breakdowns, whereas the discourse level provides a structure and orientation for disputing controversial positions related to the validity claims. Distinguishing between these two levels also allows us to separate 'just talking' from argumentative disputes. Further reflections will be provided at the end of this section. At this point, we should note that one might add another layer to the architecture, that is, a meta-meta-communication layer, to allow actors to reflect on the concepts and structures of the meta-communication layer – something that we, however, do not consider here.

Figure 1.

The architecture of meta-communication.

Full figure and legend (10K)

Figure 2 presents the concepts of the model. At the conversation for clarification level, we use an extended version of Ulrich's (2001) philosophical staircase for reflective practice. Ulrich has suggested a conceptual framework to support researchers and practitioners in the process of identifying and scrutinizing the diverse issues that they face in any information systems development project, aiming at providing people with information for purposeful action. The concepts – information, knowledge, and rational action – play a central role. The theoretical analysis of these core concepts leads to the staircase, whose steps were originally called semiotic steps (syntactic, semantic, and pragmatic clarity), epistemological steps (expressive, empirical, and normative validity), and practical–philosophical steps (instrumental, strategic, and communicative rationality). Whereas semiotic steps are theoretically grounded in Peirce's (1931–35) work, the epistemological and practical–philosophical steps are grounded in Habermas' (1984) theory of communicative action.

Figure 2.

Concepts of the meta-communication model.

Full figure and legend (62K)

We have extended the staircase by two additional steps (Yetim, 2005): physical clarity, which is related to the media aspects of signs (Stamper, 1996), and esthetic rationality, which deals with esthetic aspects of signs. As diversity in technological standards and esthetic values often causes breakdowns, they are generally regarded as relevant usability issues, especially when designing multimedia information (Smith & Yetim, 2004). In addition, instead of pragmatic clarity, we prefer to use the notion of relevance, since pragmatic aspects, dealing with intentions and the usage of signs, are also addressed by the subsequent steps of the staircase.

Whereas the staircase organizes diverse issues and also provides a thematic structure for conversations on them, the discourse level is entered when controversies arise, which require argumentative examination. At the discourse level, we use different type of discourses and reflective media proposed by Habermas (1984, 1993, 1996). Previous discursive approaches have only considered some of the validity claims and/or discourses. In contrast, this model provides a wider spectrum of concepts by extending the philosophical staircase, which initially involved nine steps and two discourses, to 11 clarification issues and eight discourses.

Let us briefly describe them and their relations. The first three steps of the staircase are interrelated and are concerned with different aspects of the comprehensibility of communication actions. They aim to contribute to the correct understanding of signs. Physical clarity is strongly related to the media (paper, fax, or digital media) through which communicative action is performed. It deals with clarifying the perceptibility or visibility of signs (e.g., 'Can this text be clearly displayed by all browsers?'). Syntactic clarity refers to structures and rules for composing complex signs. Different rules and syntactic conventions may cause breakdowns or misinterpretations that require the clarification of syntactically correct formulations of signs (e.g., 'Should we use the 'day/month/year' or the 'month/day/year' format?'). Semantic clarity deals with clarifying the meanings of signs, that is, of a word or a complex sign (e.g., 'What do you mean by 'meeting time: 8 o'clock'? 8 a.m. or 8 p.m.?'). When proactive or reactive conversations for the clarification of any comprehensibility aspect of signs arise, the explicative discourse serves as a forum to argumentatively examine the controversial positions. A position related to the corresponding validity claim needs to be articulated with reason such as 'This expression is comprehensible/incomprehensible because ...'.

The next step relevance clarifies whether a sign (or the content of a design) is relevant for the purpose of communication (e.g., 'What fields do we need in a résumé template?'). There are many aspects of relevance including thematic, interpretative, and motivational relevance (e.g., Schütz, 1970), and relevance can also be viewed as the 'meaningful relation' of signs or actions for achieving a particular ends (Allwood, 2000). Thus, this step is related to the rationality aspects of communication as well. Yet, the main concern here is the establishment of the claim 'This sign is relevant/irrelevant' by regarding the overall purpose and context of communication. The controversial positions have to be justified in pragmatic discourses.

Clarifying the comprehensibility and the relevance of signs does not guaranty their validity, that is, whether they reflect sincere pragmatic intentions, are free from errors, and in accordance with accepted social norms. Thus, following Habermas (1984), expressive validity refers to the clarification of whether an expression reflects sincere pragmatic intentions (e.g., 'Do we really mean that?'). If the expressive validity of a sign becomes controversial, the reflective medium therapeutic critique can be entered to challenge and argumentatively defend the sincerity of expressions (e.g., by showing the transparency or consistency of self-presentations). Empirical validity deals with clarifying whether it refers to the true (commonly believed) state of affairs (e.g., 'Does this message agree with the facts?'). The theoretical discourse serves as the related forum for disputing and justifying contested truth claims. Normative validity deals with clarifying whether it is communicated in accordance with accepted social norms or laws (e.g., 'Is it legally/morally appropriate to say that?'). In his earlier work, Habermas regarded practical discourses as the place for testing 'both the rightness of a given action in relation to a given norm, and, at the next level, the rightness of such a norm itself. This knowledge is handed down in the form of legal and moral representations' (Habermas, 1984, p 334). Following his later differentiation between pragmatic, ethical, and moral discourses (Habermas, 1993) as well as legal discourse (Habermas, 1996), we regard the legal discourse as the place where the rightness of an action or expression in relation to a given rule or norm is examined, whereas the rightness of a norm itself remains a matter of moral discourses, as described later.

The subsequent higher steps of the staircase are based on the assumption that clarifying comprehensibility, relevance, and validity aspects of signs alone cannot secure rational communication practice. This requires consideration of the successful transformation of signs into effective and efficient communication action and also consideration of the implications for those involved and affected. Following Habermas (1984), instrumental rationality refers to means-end rationality. It deals with the efficient use of means or choosing the most effective means to a given end (e.g., 'Do we need to design so many pages to communicate this content?'). Social aspects are not considered here. Strategic rationality is a purposive, but also a social, concept of rationality. That is, the behavior of other rational actors is taken into account by the choice of the most effective means. It involves egocentric calculation of success, deception, and power in influencing others. A design can be assessed according to its efficacy in influencing social actors (e.g., 'Should we disable the page numbers so as not to put readers off?'). When the empirical basis of instrumental or strategic rationality may become controversial, actors enter theoretical discourse. In this discourse, the claim 'this is efficient/inefficient' for achieving the goal, either in transforming things in objective world (instrumental) or in influencing social actors (strategic), can be argumentatively challenged and justified. In addition, esthetic rationality of a design can be judged according to its beauty, that is, whether a design is in accord with or deviates from culturally established standards of esthetic value (e.g., 'Isn't it too white?'). This type of rationality is added to Habermas' ideal types of rationality. The place for disputing controversial esthetic experiences is esthetic critique. Finally, the concept communicative rationality refers to the communicative achievement of mutual understanding among actors. This highest step of the staircase allows actors to reflect on what they have achieved so far and to seek mutual agreement (e.g., 'Does everyone agree? Any further comments?'). When a communicatively achieved consensus with respect to one or several aspects of a communication action is challenged, actors can enter corresponding discourses.

Before closing this section, some additional reflections and concluding remarks on the meta-communication model should be offered. Firstly, in the model, no single link is established from the staircase to the ethical and moral discourses, since ethical and moral issues can arise in many steps and discourses. According to Habermas (1993), ethical discourse refers to individuals or communities with differing value systems and is concerned with justifying regulations from a cultural perspective, that is, from the perspective of what is 'good' for them. By contrast, duties of justice, that is, whether the corresponding practice is equally good for all, can be rationally justified in moral discourse. In this discourse, the ethnocentric perspective of a particular group expands into the perspective of an 'unlimited communication community'.

Secondly, in Habermas' discourse theory the relationship between validity claims and discourses is intelligible, yet dialogs may become complex, confusing, and possibly inefficient when actors have to move from action level to discourses and also between discourses to check each validity claim. If actors have to enter discourses directly when validity claims are contested, then this raises the issue of where conversations for clarifications (e.g., clarifying questions) can be articulated when actors want to assure themselves of their understanding of the conflicting issues first before entering argumentative disputes. In our model, the separation of conversation and discourse level makes the validity claims explicit and considers the steps of the staircase and the discourses as separate 'discussion spaces'. It allows actors to separate 'just talking' from argumentative disputes, and it can also reduce the information overload for those who are only interested in controversial positions on a specific issue rather than in conversations for clarification, or vice versa. In addition, the clarification level can be used by moderators for facilitating discourses, for example, by clarifying abstract concepts or providing orientations while using the model.

Thirdly, although the staircase metaphor suggests examining validity claims step by step, we consider this one of many usage options. In many situations, actors may need to articulate only some of the breakdowns. Moreover, while clarifying issues step by step, using discourses along the staircase may not be efficient. For example, a longer controversial dispute on the comprehensibility of a sign in explicative discourses may be waste of time before examining its relevance in pragmatic discourses. Therefore, actors can communicate along the steps of the staircase, and at the same time put controversial issues in related discourses, but leave them undecided until they have risen to the highest step, 'communicative rationality'. Open issues can then be retrospectively clarified in related discourses. This can also be done in different rounds as in Delphi techniques (for some ideas, see Yetim & Turoff, 2004).

Finally, as mentioned before, the concepts of the model are mainly theoretically grounded. In addition, the steps of the staircase are used to categorize a set of research-based usability guidelines (Yetim, 2006). This experiment shows that the concepts of the staircase provide useful distinct purpose categories, while further research on empirical grounding of concepts would be of value. Nevertheless, the assignment of utterances to the most appropriate category will be an issue of interpretation since they are intentionally and contextually determined and can be multifunctional (Allwood, 2000).

Meta-communication on genres

Fundamental concepts of genre theory

The concept of genre has a long tradition in rhetorical and literary analysis (Bakthin, 1979/86). A number of researchers use the notion of genre as typified social action in several disciplines including cultural, rhetorical, communication, and design studies (Miller, 1984; Bazerman, 1994; Brown & Duguid, 1994; Orlikowski & Yates, 1994; Berkenkotter & Huckin, 1995; Günthner & Knoblauch, 1995; Luckmann, 2001). Although there is no universally accepted definition of genre, most studies regard a genre as a historically and culturally specific pre-patterned solution to recurrent communicative problems (Orlikowski & Yates, 1994; Luckmann, 2001). Genres – such as letters, résumés, announcements, etc. – are fixed patterns of communicative action, and they structure communication by creating shared expectations of the purpose, form, and content of the interaction, thus easing the burden of production and interpretation. Another important aspect of genres is that they transmit knowledge of conventional 'solutions' for a variety of communicative problems that appear in social interaction in human societies (Luckmann, 2001).

Interrelated genres form a genre system (Bazerman, 1994), which constitutes a more coordinated communicative process. For example, journal articles are often realized through an interlinked sequence of genres including a submitted manuscript, peer reviews, and an editorial decision letter, enacted in a typical sequence. Like individual genres, genre systems also create expectations about the purpose, content, form, participants, time, and place of communicative interaction (Orlikowski & Yates, 2002). The genre repertoire (Orlikowski & Yates, 1994) or communicative budget (Luckmann, 2001) of a community indicates its established communicative practices and thus reflects the common knowledge, expectations, and norms that members of the specific community share about communication.

An illustrative example of meta-communication on a genre instance

Let us first illustrate by means of a simple example how the model can be used for conversations on genres. Table 1 shows a conversation of a team of instructors about the syllabus of a course that has to be offered in different countries. Note that this example is partly constructed for illustration purposes and does not serve as an empirical validation of the model. Since the purpose of a syllabus as a genre is to communicate information about the course, the communication by instructors about the 'syllabus' communication' can be viewed as a kind of meta-communication.

Table 1 - An example conversation on a genre instance.

Full table

The example illustrates the thematic diversity and complexity of meta-level conversations. After entering the meta-communication, actors question the perceptibility of the signs (physical clarity), express the need to clarify a syntactically common convention for date formats (syntactic clarity), challenge the understandability of the meaning of '9/11' in global contexts (semantic clarity), and the relevance of the suggested topic for the course (relevance). The conversation continues by expressing a doubt about the sincerity of the expression of feelings for the whole team (expressive validity), questioning the correctness of the description of assignments (empirical validity), and disputing the appropriateness of the submission of assignments with respect to institutional rules (normative validity). In the subsequent parts of the dialog, actors challenge the efficiency and effectiveness of the planning of course activities (instrumental and strategic rationality) as well as the esthetic appearance of the syllabus (esthetic rationality). Finally, actors reflect on what they have achieved so far and assure themselves of their mutual understanding with respect to the changes needed (communicative rationality).

The example also shows that the conversation may consist of diverse types of contribution such as clarifications, questions, position statements, etc., and that not all of the contributions necessarily lead to controversial debates. As mentioned before, the places for disputing controversial issues with arguments are the corresponding discourses, which we have neglected in the illustration for reasons of simplification. Discourses will come up when we discuss more controversial issues in relation to the use of meta-communication for the legitimacy of genre norms.

To draw an interim conclusion, this simple example is chosen to illustrate how the concepts of the 'conversation for clarification' level can be used to check the instantiation of a genre. The meta-communication model deploys a range of validity claims and related discourses and thus facilitates the evaluation of genres with respect to whether they provide comprehensible, relevant, and valid information (or knowledge), and allow rational communication and action. At this point, we can claim that the model, as illustrated so far, makes a contribution to current research on genres by extending the concepts for reflecting on genres. The model also allows systematic and meaningful structuring and organizing of meta-level conversations on genres and genre instances. In practice, these concepts can also be used to further organizational goals in a strategic manner.

In the following, we will focus on using genres in a manner compatible with the discourse ethics. We claim that addressing the legitimacy of genres fills a gap in the genre research. To justify our second contribution, we will first discuss global challenges and the gap in genre research, and then illustrate how different kinds of meta-communication (ex ante, in-action, and ex post) can be used to legitimate genres from a discourse-ethical point of view.

Global challenges and the gap in genre research

If we take communicative genres as socially constructed solutions (Berger & Luckmann, 1966) that organize and routinize the ways of dealing with particular communicative problems, it seems quite obvious that different cultures may construct different solutions for specific communicative problems. Thus, the repertoires of communicative genres vary from culture to culture as well as from one epoch to another. Several empirical researchers found cultural differences in the function and structures of genres (Günthner & Knoblauch, 1995; Günthner & Luckmann, 2001).

To mention some structural differences and challenges, in Chinese argumentation, for example, proverbs and idioms fulfill an important function in supporting arguments (also in academic texts). They allow speakers to demonstrate their classical knowledge and present their own assertions as part of traditional and still valid collective wisdom. By contrast, in the Western context books on style may advise against using proverbs. Concerning the inner structure of genres, the discursive organization of the Chinese genre of request letters, for example, generally reveals a preference for providing reasons first, before the main point (the request) is stated, and conforms to the following schema: salutation, preamble (face work), reasons, and then the request itself (Kirkpatrick, 1991, p 198). Thus, in contrast to English request letters, Chinese not only tend to place the reason before the request itself, they also engage in extended face work, which forms an integral part of their requests. Changing the order, by moving the request to the beginning, results in a letter or request marked as direct and possibly impolite. Concerning the outer (social) structure of a genre, cultural differences may appear, for example, in respect to who has access to, is competent in, or is allowed to perform particular genres (including gender-related differences). Finally, the situative aspects of genre performance (e.g., technical and/or social elements) may influence genre's use.

Since users habitually apply their genre conventions to the new media (Yates et al., 1999), these differences are also valid in computer-mediated communication contexts (Yoshioka et al., 2000). Moreover, these 'structural differences' and asymmetries in genre-related knowledge may not only lead to misunderstandings in intercultural contacts but also have negative implications for actors, that is, be taken to reflect individual incompetence or malice (Günthner & Luckmann, 2001). In addition, if we take into account that communicative genres transmit knowledge of conventional 'solutions' for a variety of communicative problems, then this demands the treatment of their entire spectrum of contents (information, knowledge), structural and normative aspects in a reflective way. Given the global diversity, it is a particular challenge to determine what knowledge is relevant and should be transmitted by genres (see, e.g., Nonaka & Takeuchi (1995) as well as De Long & Fahey (2000) for the role of culture in shaping assumptions about what and which knowledge is worth managing).

The empirical-oriented research facilitates designers' understanding of how communicative genres generally function, shape, enable, or constrain action within the communicative practice of communities. However, they do not provide orientations on how to design legitimate genres or structures for the routinization of action in (intercultural) contexts where norms and conventions compete. For example, in the information systems field, Yates et al. (1999) describe how genres are explicitly and implicitly structured. However, they do not deal with the issue of whether the processes and thus the resultant genre can be regarded as legitimate. Päivärinta (2001) discusses the potential of genre concept with respect to promoting critical orientation in systems development communication among stakeholders. Yet, this work remains limited in its spectrum of validity claims and does not fully consider discourse-ethical viewpoint. In the next section we are explicitly concerned with the legitimate structuring of communication concepts, and describe meta-communication processes according to discourse ethical principles.

Legitimating genres through meta-communication: a discourse-ethical view

Discourse ethics

The theoretical approach to discourse ethics advanced by Habermas in several writings (1990, 1993, and 1996) enjoys the widest critical attention. Habermas's (1984) previous analysis of communicative action (i.e., an action oriented to reaching mutual understanding) provides the background of his discourse-ethical deliberations. The goal of discourse ethics is to clarify the normative foundations of human action, communication, and interaction. Discourse ethics claim to be universalistic, applicable to more than one specific culture or epoch. Habermas's position is that under conditions of irreducible pluralism, there seems to be no alternative to locating the normative basis for social interaction in the rational structure of communication itself. Discourse ethics is supposed to overcome the problems of Kantian ethics, namely the idiosyncratic nature of the process of checking the validity of norms. Habermas assumes that the rightness of norms can be rationally grounded and regards the legitimating of norms as a matter of a public discourse where shared understanding about the generalizability of interests can be tested. He replaces Kant's categorical imperative with a procedure of moral argumentation.

According to discourse ethics, 'just those action norms are valid to which all possibly affected persons could agree as participants in rational discourses' (Habermas, 1996, p 107). The practical discourses are central to discourse ethics. Habermas notes that these discourses can be used (a) for the justification of norms as well as (b) for the application of norms. In the case of justification, the principle of universalization plays a central role. It 'compels the participants in discourse to examine contested norms in view of foreseeably typical cases, in order to determine whether the norms could meet with the considered agreement of all those affected' (Habermas, 1996, p 162). However, such discourses of justification cannot take into account ex ante all the possible constellations of future situations, and thus the application of norms calls for an argumentative clarification in its own right, that is, it requires discourses of application that address the question of context-sensitivity by considering all the relevant features of the current situation (Habermas, 1993). In other words, several norms can be justified from the perspective of universalizability and can claim prima facie validity. However, only in their application to particular concrete cases will it transpire which of the competing norms is the most appropriate in the given context. Consult also Günther (1993) for the sense of application discourses.

Instantiation of meta-communication types within the scope of genre theory

The characterization of the three different types of meta-communication was based on whether the meta-communication is triggered before, during, or after an action. An action can be, for example, the uttering of a sentence, and thus reflections can take place before uttering (e.g., 'I will tell him that ...'), during the production of an utterance (e.g., 'His face is becoming red, should I continue telling?'), and after having expressed it (e.g., 'Was it really appropriate?'). In practice, facilitating meta-communication for each single action may not be desirable, not to mention effective, and a sequence of actions or activities will most likely be taken as a basis for drawing the lines between meta-communications. For example, communications on this paper before its submission to the review can be viewed as an ex ante meta-communication and communications during the review process as meta-communication-in-action, whereas critiques and comments related to it after its publication as an ex post meta-communication. So, there are many ways to implement the three kinds of meta-communication in practice, also in the context of genre theory.

Yet, the way they are implemented is significant for the rationalization and legitimating of genres. If we drew the line between meta-communications taking place when (1) defining of or reflecting on prototypes before action, (2) contextualizing them, that is, creating instances and acting 'with' them, and finally (3) having acted 'with' them, it would make the significance of the discourse-ethical differentiation between justification and application discourses more visible.

When implemented in this way, each of the three types of meta-communication can be initiated in many sub-situations: ex ante meta-communication can be triggered in situations when actor(s) plan to introduce a new recurrent communicative situation or changes to an existing one (e.g., suggesting a new online 'security dialog' for online flight reservations requesting personal data). In such cases, actors anticipate possible breakdowns regarding the functional and structural differences of genres at the internal and social level and suggest solutions in the form of prototypical specifications. This includes addressing issues such as: What is the purpose of the genre? What constitutes its internal structure? Who are legitimated to use it? and so on.

Meta-communication-in-action characterizes reflections in application contexts, for example, when actors instantiate genres for acting with them (e.g., 'What should be written in the field "Date of Habilitation" in an application form if applicant's country does not have such a post-doctoral academic degree?'). Meta-communication-in-action can be initiated by many types of breakdowns that may occur due to deviation between prototypes and instances with respect to circumstances such as local established rules, changes in local conditions such as plan and desires of actors, new established interpersonal relationships, etc. Consequently, actors deal with issues concerning the appropriateness of genres in action contexts (see also, Orlikowski et al., 1995; Bergquist & Ljungberg, 1999; Päivärinta, 2001; Yetim, 2002).

Finally, ex post meta-communication refers to reactions or reflections on experiences made after the execution of communicative actions 'with' genres. It should be noted that the line between meta-communication-in-action and ex post meta-communication is not always easy to draw since action is a continuous process. In the case of a paper submission, one might consider the communications during the review process until the decision of the paper's acceptance or rejection as meta-communication-in-action, the reactions to the published paper as ex post meta-communication. Similarly, in the case of the syllabus example reflections during the creation of the instance as well as students' reactions during the semester can be regarded as meta-communication-in-action. The reflections on the syllabus at the end of the semester can be characterized as ex post meta-communication. In the case of a job application, all the reflections until the official confirmation of receiving the application can be viewed as meta-communication-in-action. In other words, ex post meta-communication starts when the activity ends to which the genre is related, and it usually has no direct influence on the current activity and can just help to improve genres or instances for future actions.

These brief reflections on how the three types of meta-communications can be meaningfully implemented within the genre theory should suffice to illustrate the use of the meta-communication model within the scope of these characterizations. However, since we have already illustrated the usage of the staircase by means of the syllabus example, the following illustrations will primarily concern the use of practical discourses and make use of the discourse-ethical differentiation between discourses of justification (as procedures of ex ante meta-communication) and discourses of application (as procedures of meta-communication-in-action).

Illustration of meta-communications: justification and application discourses

Consider the Chinese and English request letters (mentioned earlier), which differ regarding their internal elements as well as their sequential organization: <Salutation, Preamble, Reason, Request> vs <Salutation, Request, Reason>. If the enactment of any of them leads to breakdowns in intercultural contexts and thus initiates ex post meta-communication, actors can use the model to express their comments and critiques of its comprehensibility, its empirical and normative validity (e.g., the form of address), its instrumental rationality (e.g., the efficiency of the organization of the content), or its strategic rationality (i.e., in terms of its effectiveness for getting what is requested). Actors can enter related discourses to debate controversial positions.

Becoming aware of the diversity of conventions in letter writing and of its possible negative implications for their relationship, actors may regard it as 'good for all' to establish a common convention for intercultural correspondence, that is, to specify the intercultural genre of a request letter. This is the situation where a 'common interest' for the regulation of their interaction exists. At the meta-communication level, one can regard this shift from critical reactions to regulating actions as a transition from ex post meta-communication to ex ante meta-communication:

(a) Ex ante meta-communication: justifying genres – The task of actors in ex ante meta-communication is to suggest, discuss, and justify prototypical elements of the corresponding intercultural genre. Since actors are not concerned with the instantiation of a genre with concrete data, they do not need to consider the first three steps of the staircase dealing with comprehensibility of the content at the physical, syntactic, and semantic level. Actors can start on the 'relevance' step of the staircase and first clarify what is needed for the purpose of a request letter. Argumentative justification of controversial positions on the relevance of suggested elements is to take place in pragmatic discourses. While justifying the relevance of an element (e.g., the relevance of a 'preamble' in a request letter), actors may refer to values (e.g., to politeness or face-saving). Others may question its relevance or purposefulness by referring to a different value such as sincerity, and they may, for example, argue that the more information provided before the request, the more it risks being interpreted as a sign of insincerity within their cultural contexts (Clyne, 1994).

When such value-related conflicts occur in pragmatic discourses and need to be resolved, actors have, in principle, the following options to continue the debate (Figure 3):

1. reflecting on which of the competing values is good for them and should be taken as a basis for orientation when they aim to construct a single genre prototype for all; or
2. clarifying whether constructing some variations of genres informed by the existing diverse values are acceptable to all before conducting a value-discourse; or
3. discussing how a single genre prototype should ideally be under the unchangeable condition of the existence of value pluralism.

Figure 3.

Options for dealing with value conflicts in discourses.

Full figure and legend (15K)

Thus, actors need to make a procedural decision about which of the three options should first be exploited when value-related conflicts occur in pragmatic discourses. The first option suggests a kind of rational value choice as proposed by Klein & Hirschheim (2001). Conducting rational discourses on competing values in ethical discourses can help to clarify values. However, as the authors also remark, it does not guarantee a resolution of value conflicts at organizational level, not to mention at global level. After a value-related discourse actors may still have to deal with remaining values, and thus may pursue the other options.

The second option suggests a moral discourse on the acceptability of diverse genres. Whether diverse orientation is obligatory, permissible, or forbidden is a basic moral issue. Actors cannot know the acceptability of such a regulation before examining its consequences and side effects for those affected in moral discourses. Classifying actions in moral discourses as permissible means that members of different communities can weigh the merits of that action with respect to their values and preferences in ethical discourses, for example, members of a community sharing a value can separately reflect on how they can realize a shared value in their culture-specific prototypes (e.g., how politeness or face-saving can be realized). If diverse orientation is not acceptable, actors can exploit a solution in the sense of option 3.

The third option suggests determining the right solution under conditions of value pluralism. While cooperatively searching for a regulation of the intercultural genre, actors suggest solutions (e.g., concerning the required elements of a letter and/or their sequencing), which they regard as good for all and thus just. Actors also express their positions on a suggested regulation by considering its 'consequences and side effects' for their interests and value orientations (e.g., 'Salutation should not be used since existing diverse expectations have created many miscommunications'). From a moral point of view, actors have to adopt the perspective of others (in the sense of ideal-role taking) and transcendence of merely particular interests in an argumentative search for a solution rationally acceptable to all.

At this point, the issue arises of how can actors conclude from many diverse interests to a single right norm that shall regulate their 'common interest'. When searching for the right norm in moral discourses, the universalization principle serves as a kind of 'bridging principle' or pragmatic inference rule. According to Habermas, this principle is intended to function like the principle of induction in empirical inquiry (Heath, 2001). For example, in an empirical argumentation actors conclude that a suggested theory t1 is true by appealing to a certain number of cases and showing that, given the relevant facts (f1, f2, and f3), theory t1 explains all these facts better than any other (i.e., Facts Induction Theory). In moral argumentation, actors conclude that a suggested norm n1 is right, given the interests of all those affected (I1, I2, and I3) and given that the norm n1 satisfies all these interests better than any other (i.e., Interests Universalization Norm). As one accepts a theory when it accounts for the facts, one accepts the norm when one sees that it adequately balances the interests of all those affected. The task of actors is searching for such a rationally acceptable norm.

The example of a request letter illustrates the conflicts that can arise due to socio-cultural value differences. In other cases, for example, in the case of a résumé, conflicts can occur at the legal level. Actors may first become aware of the differences in national laws after the normative validity of a résumé has been challenged due to its obligatory elements (age or nationality), and subsequent controversial debates in legal discourses have been conducted. Consequently, the rightness of the corresponding laws themselves may also be challenged and thus subjected to moral discourses. Discourses should not be conceived as day-to-day activities that can lead to immediate changes. In practice, actors may find themselves faced with the problem of establishing short-term pragmatic solutions, if necessary through negotiations.

Finally, justifications in ex ante meta-communication concerns not only the specification of the typical elements and structures of genres but also the rationality and legitimacy of the sequencing of several genres for organizing a larger communicative activity, that is, for building systems of genres. In practice, the shaping of genres and genre systems will be heavily affected by the involved actors (e.g., representatives), that is, by their expertise and knowledge of cultural differences as well as by their interests and openness to change. Since actors cannot foresee all future situations when justifying the norms of genres in ex ante meta-communication, the application of genres calls for an argumentative clarification in its own right to close the gap that normally had to remain open in the justification.

(b) Meta-communication-in-action: applying genres – Meta-communication-in-action refers to reflections taking place in application contexts and concerns the appropriateness of enactment or instantiation of genres. The following issues may pose themselves in action contexts:

1. What genre is appropriate and thus should be enacted (e.g., requesting first the self presentation of other or presenting oneself, that is, writing a request letter or submitting a résumé)?
2. What variant of a genre is appropriate (e.g., using an intercultural request letter or a culture-specific one)?
3. What instantiation of a genre is appropriate (e.g., how should it be verbalized)?

Addressing issues (1) and (2) requires the consideration and interpretation of all the relevant features of the action situation in order to determine what genre or variant is the most appropriate one, whereas (3) presupposes a consensus concerning the appropriate genre, and only requires clarification of its content. It is a matter of context whether actors first discuss their interpretations of the situation in order to decide what genre is appropriate, or whether such discussions emerge after an inappropriate proposal of genre application.

The syllabus example in Table 1 has already illustrated how the staircase of the model can be used to clarify the appropriateness of the instantiation of a genre. Therefore, the following illustration is restricted to some relevant aspects of application discourses and their differences to justification discourses, guided by the ideas of Günther (1993). Consider the following proposal, for which a proponent can also put forward the warrant if required:

• (CLAIM) We should use the English request letter.
• (DATA) We will send it to a team of medical scientists living in the U.S.A.
• (WARRANT) One should respect the culture of others.

An opponent can dispute the appropriateness of the use of the request letter in different ways, for example:

1. By contesting the semantic clarity of the data expressed as a reason, for example, expressing doubts that it belongs to the semantic extension of the warrant in order to be put forward as a reason for the claim. It deals with the identity of meaning between those expressions included in the WARRANT and those included in the DATA. For example, the fact that the team's address is in the U.S.A. does not agree with the meaning of respecting the culture of others. In this case, actors can enter the explicative discourse to explicate the word usage.
2. By contesting the empirical validity of the DATA claimed to be relevant in the description of the situation, for example, whether the team is in the U.S.A. If necessary, the truth of the statement must be checked in a theoretical discourse, to assure whether the condition asserted by the proponent is present.
3. By contesting the normative validity of the description of the situation, for example, by showing that considering other relevant data would legitimate the enactment of other genres. For example, the opponent could object: 'Why do you rely only on the fact that the receivers live in the U.S.A., and not on the fact that the medical team consists solely of Indian and Chinese researchers?' The legitimacy or appropriateness of the selection needs to be disputed in legal discourses, which can also include appropriateness argumentations in moral discourses.

Application discourses must determine which descriptions of the facts are significant and exhaustive for interpreting the situation in a disputed case and thus which genre is most appropriate. In application contexts, several norms and semantic variations of norms may compete. They are interpreted and thematized in moral discourses with respect to their appropriateness in a given situation. Even though accepted norms and prototypical specifications of genres may be challenged, their validity is not justified in application discourses. For example, the norm 'You ought not to lie', which may be accepted by all as valid after a universalization test in justification discourses, can be thematized for its appropriateness when the medical team reports the requested results of a medical test to a terminally ill patient. In this knowledge transfer from medical team to the patient, not telling the truth about the patient's condition may be viewed as considerate treatment rather than as a lie (Günther, 1993). Yet, this does not challenge the validity of the norm itself, rather its appropriateness in the action situation.

Expressed in terms of Toulmin's (1958) scheme of argumentation, appropriateness argumentation refers only to the relation between CLAIM–DATA–WARRANT; reasons which affect the BACKING of the validity of the norm (WARRANT) do not play a role. Appropriateness is guided by whether the norm, in all its semantic variants and in relation to all other applicable norms, is applicable to a situation described completely. The validity depends on the fact that the norm can be universalized in view of the evidence (BACKING) put forward for its general observance.

Separation of the justification discourses from application discourses relieves actors of having to justify genres and principles in each situation by considering the interests of all those affected as required by the universalization principle. In order to resolve norm conflicts in application contexts, the one appropriate criterion for justification of the singular judgement under restrictive conditions of scarce time and incomplete knowledge is the criterion of coherence (Günther, 1993), which requires a coherent interpretation of all the applicable valid norms in a situation.

To summarize, the application of appropriate genres as well as appropriate instantiation of genres will not only be shaped by technical, economic and social factors, they are also highly dependent on the interpretation and definition of situations by actors in action contexts. Conflicts between concurring interpretations, which are not soluble, may also lead to doubt of the universalizability of a norm and consequently to a discourse on the justification of its validity.

Discussion

The approach presented so far raises further issues that go far beyond what can be achieved in a single paper. To provide additional discussions and reflections, this section compares the discursive structuring of genres with non-discursive approaches, then reflects on the link of the genre-based approach to the LAP, and finally addresses further implications for information systems.

Discursive-ethical vs non-discursive structuring of genres

The structuring of genres through meta-communication is related to the genre structuring processes investigated by empirically oriented genre research. Yates et al. (1999) identified two general processes of genre structuring in an electronic medium, which they labeled explicit and implicit structuring. The process of explicit structuring includes planned replication (i.e., predetermined action intended to reproduce prior forms of social interaction within a new medium), planned modification (i.e., predetermined action intended to create some changes in the status quo), and opportunistic modification of existing genres (i.e., purposeful changes introduced in response to an unexpected occurrence, condition, or request). The process of implicit structuring includes migration (i.e., unreflective action that reproduces genres established in one medium or community within another medium or community) and variation of existing genres (i.e., results in a departure from established forms of social interaction by introducing some changes in the genres enacted within the new context).

These processes are related to the three types of meta-communication. Explicit structuring as deliberate action corresponds to structuring through meta-communication. In particular, 'planned replication' and 'planned modification' as predetermined actions are closely related to ex ante meta-communication, whereas 'opportunistic modification' of existing genres as a response to an unexpected occurrence or conditions in use contexts is related to meta-communication-in-action. The process of implicit structuring, which includes 'migration' and 'variation' of existing genres, may cause breakdowns that thus – if articulated – may lead to ex post meta-communications.

The meta-communication as a deliberative action to explicitly structure genres does not eliminate implicit structuring. The suggested model provides the means to deal with resultant structures deliberately if they are experienced as breakdowns and thus not accepted. It advocates a deliberative practice among those concerned, that is, incorporating their knowledge and cognitive inputs. Moreover, this approach, grounded in discourse-ethical principles, values legitimate processes for creating and shaping genres. It explains the legitimacy of resultant genres by means of procedures and communicative presuppositions that, once they are legally institutionalized, justifies the supposition that the processes of making and applying genres lead to rational outcome. In this respect, the explicit structuring through meta-communication model distinguishes itself from those that are describing status quo processes. From an ethical–moral point of view, it makes a difference whether the genres that may require actors to accept disadvantages are the outcome of a legitimating, deliberative process or whether, on the contrary, they merely emerge as side effects of programs and processes motivated by other, private, interests. This applies to culture-specific genres as well as to intercultural ones.

The notion of intercultural genre as used here requires further clarification. It does not include those patterns that may (unconsciously) be enacted by a bicultural actor and that may consist of a mixture of contents (e.g., languages) or structural elements from two cultures (e.g., Rehbein, 1994). One might regard them as cultural patterns of those having similar bicultural socialization. From the perspective of discourse ethics, we have to distinguish genres whose structuring is based on consensus or mutual compromise as a result of intercultural discourse among those affected from those genres that are not. Implicit and explicit structuring processes described by Yates et al. (1999) may also take place under conditions of asymmetrical relations between cultural groups, rather than representing the ultimate form of intercultural discourse under conditions of 'free speech situations'. We might call such genres for intercultural contacts 'transcultural genres' to emphasize the non-discursive character of their emergence (Yetim, 1998). Thus, intercultural genres, as they are defined here, incorporate another dimension of validity and legitimacy to regulate interaction 'between' cultures in contrast to transcultural ones.

Beyond the legitimating aspect, meta-communication processes can also lead to new patterns of intercultural communications that may not be identified by the current genre research. Although there are a number of competing definitions of genre, genres typically emerge from use contexts when communicative activities are repeated over time. Thus, only those communicative activities can be identified as genres that have taken place several times. However, communication patterns for intercultural contexts can also be shaped and legitimated ex ante for anticipated action contexts in order to provide solutions to cope with differences and possible contradictions (or incommensurability) between cultural conventions and expectations. In this sense, the notion of intercultural genre resulting from discourses does not only characterize a validity dimension but also refers to such patterns that are constructed for anticipated future use situations. Accordingly, an intercultural genre repertoire consists of genres or new patterns that are legitimated in intercultural discourses for use in intercultural contacts.

As the composition of intercultural genres is dependent on the outcome of discourses, we cannot a priori specify the characteristics of such genres. Potential outcomes of intercultural discourses could be: (1) no result achieved (e.g., due to miscommunication); (2) dominance of one culture in the resultant genre; (3) a mixture of elements from two or more cultures; or (4) new elements/structures that emerged from reflections, which may or may not be traced back to the cultures involved. Yet, intercultural genres do not necessarily remain 'intercultural', they may end up being incorporated into the dominant culture.

The link to the LAP

As briefly mentioned at the beginning of this paper, within the LAP different communicative actions are categorized mainly by using speech-act classification schema (Searle, 1969). Organizations are regarded as networks of conversations, and conversations consist of networks of actions (Winograd & Flores, 1986). Thus, within the LAP there is an interest in recurrent speech-act patterns. Conversation for action as the principle model for interaction specifies the network of moves involved in the interplay of speech acts (e.g., requests and commissives) directed toward cooperative action. In a genre-based approach, a genre is viewed as a purposeful organization of a sequence of speech acts, whereas a genre system refers to a purposeful organization of a sequence of genres. They play a coordination role among people and between people and their tasks (Orlikowski & Yates, 2002).

Speech-act and genre approaches are interrelated and can also enrich each other (e.g., Bazerman, 1994; Berkenkotter & Huckin, 1995; see also Ljungberg, 1997). As a text document contains many acts, speech acts could be used to articulate the web of intertextually linked documents as genres. Depending on the recognizable overriding illocutionary force, actions with a document as genre can correspond to Searle's five categories of speech acts (1969), that is, when the text effects a law (declaration), or makes an application (directive), or contractually binds you (a commissive), or presents a scientific claim (assertive), or conveys outrage in response to governmental action (expressive). As Yoshioka et al. (2001) illustrate, speech acts can be used as purpose categories such as inform, request, propose, decide, respond, record, etc. to organize a system of genres in a sequence that coordinate collaborative activities. A genre may have multiple purposes. For example, the memo genre can be used to inform its readers and record information and it may also be used for directing an order. Like a genre, a genre system also provides expectations about its socially recognized purposes. Yet, a genre system as a whole usually has a different purpose than its constituent genres. For example, a ballot genre system, which is used to poll opinions and test consensus among the participants, might belong to the 'decide' purpose. In the ballot genre system, the ballot questionnaire genre might belong to the 'inform' and 'request' purpose categories because it is used to inform group members about issues and to request their replies. The ballot response genre might belong to the 'response' category, whereas the ballot result genre might belong to the 'inform' and 'record' purpose categories (Yoshioka et al., 2001).

Meta-communication processes can be useful in the design of new and redesign of existing communication processes. The discursive concepts can not only be used for reflecting on and legitimating genres but also for speech acts and speech-act patterns. As for genres, there are also cultural differences in the use of single speech acts (e.g., Clyne, 1994). Thus, we encourage further investigations on designing and understanding of structures that help to act as well as to reflect appropriately, especially in interorganizational contexts.

Further implications for information systems

The approach has also implications for other research areas in the information system field. To mention a few: Firstly, genres are useful concepts for capturing and communicating knowledge (Orlikowski & Yates, 1994). Meta-communications on genres can help to legitimize what knowledge should be captured and communicated, whereas reflections on the genre instances can ensure that validated knowledge enter in the organizational knowledge base. As knowledge management practice often includes diverse interests and power relations (Marshall & Brady, 2001), meta-communication processes can not merely facilitate that diverse voices be heard but may also reduce different types of uncertainty in the mind of individuals, for example, uncertainty due to incomplete communication or domain knowledge. Moreover, building an intercultural genre repertoire, in addition to a culture-specific one, can help managers and designers to be aware of and learn about cultural differences, which may avert some miscommunications. Agreement on some intercultural genre norms may relieve actors from 'spontaneous' choices as to how to communicate what, when, and to whom.

Secondly, the meta-communication concepts can also be used for critical evaluation of information systems as they can enrich current critical approaches to evaluation (e.g., Klecun & Cornford, 2005) by a set of usability issues. Practitioners and users can systematically reflect on them during the planning (ex ante), development, and implementation (in-action) as well as using of information systems (ex post). In addition, the model provides purpose categories, which can be used to organize usability guidelines for interaction design (e.g., Yetim, 2006). As the approach generally allows iterative and reflexive interpretation of the data, it would be of value to investigate further how well it fits within the stream of different action research in information systems (e.g., Baskerville & Wood-Harper, 1998).

Finally, this approach also supports the recommendations of research on computer-mediated cooperation, which emphasizes the importance of building conventions for successful cooperation. Without explicit structuring, one would expect implicit structuring, which might leave team members from different cultures or functional and institutional background with misunderstandings and a lack of a common framework for communicating (Yates et al., 1999). On the other hand, as Mark's (2002) empirical study shows, it is a paradox that a group forms agreements for conventions believing that it will improve their cooperation but does not strictly adhere to these agreements in the actual working practice, leading to overhead and process loss. Thus, the study argues in favor of a specific set of awareness information requirements to promote active learning about the group activity in order to support the articulation of conventions. Feedback is regarded as a powerful mechanism for shaping and learning about group behavior. In the same line, our approach values ex ante meta-communication for evaluation and justification of the norms and conventions, but also assumes the difficulty of theorizing about or anticipating the appropriate conventions. It thus pleads for the support of meta-communication-in-action as well as ex post meta-communication for articulation of breakdowns.

Conclusions

This paper has focused on the concept of communicative genre and addressed the issue of how meta-communication processes guided by discursive-ethical principles can promote a rational and legitimate definition, design, and structuring of genres. Such a meta-communication process has not yet been thoroughly discussed in relation to the concept of genre. Thus, this paper has made the following contributions to the genre research: firstly, it has provided a wider spectrum of discursive concepts for critically reflections on and discursive evaluation of the content and structures of genres. The meta-communication model arranges a range of validity claims and related discourses and thus facilitates the evaluation of genres as to whether they provide comprehensible, relevant, and valid information (or knowledge), and allow rational communication and action. In addition, the model allows systematic and meaningful structuring and organizing of meta-level conversations on genres and genre instances. It can also be used in a strategic manner, without accepting the discourse ethical idealizations. Secondly, it demonstrates how different kinds of meta-communications (ex ante, in-action, and ex post) can be used to legitimate genres in a manner compatible with discourse ethics. It illustrates the discourse-ethical viewpoint concerning the legitimacy of genre structuring processes and thus also the legitimacy of resultant norms and contents of communication, especially in global contexts.

As the approach also shows how communication practices need to be modified for the special requirements of discourse ethics, some barriers and limits of realizing discursive approaches in practice should be finally reflected on. First of all, the concepts of discourse theory themselves leave room for a number of philosophical challenges and further issues that could not be addressed in this work in their entirety. Readers may consult Honneth & Joas (1991) and White (1995). Discourse as an essential component of democratic legitimacy involves a trade-off between the efficiency of continued conversations and the goal of mutual understanding to produce shared contents, norms, and conventions. Achieving complete rationality in discourses about validity claims depends on the extent to which the ideal speech situation has been approximated at the organizational, national, and international level. Discourse on validity claims cannot always secure consensus and eliminate all disagreements by 'the force of better arguments'. In these contexts, one might question the universal usability of a framework that is mainly based on Habermas' discourse theory. In many cultures, people may use practices other than discourse to resolve disputes. However, any procedure will need to make significant use of language, since it has the expressive resources that people need to deliberate about their practice (Habermas, 1992; Heath, 2001). As long as language itself does not break down, the underlying practice of argumentation will always be available as a mechanism for resolving disputes. In other words, discourse is always present. Thus, against a background of increasing complexity and cultural pluralism, the paper argues in favor of discursive modes of conflict resolution and offers some basic issues that need clarification. Their negotiation may allow actors to find commonalities and create common orientations or a compromise, which integrates different views and values, in the sense of interculturality, but also leaves room for diversity (at least partly) in the sense of multiculturality, where it is morally justifiable.

References

1. Allwood J (2000) An activity-based approach to pragmatism. In Abduction, Belief and Context in Dialog (BUNT H and BLACK W, Eds), pp 47–79, John Benjamin Publishers, Amsterdam.
2. Austin JL (1962) How to do Things with Words. Clarendon Press, London.
3. Bakthin M (1979/86) The problem of speech genres. In Speech Genres and Other Essays (EMERSON C and HOLQUIST M, Eds), pp 60–102, University of Texas Press, Austin.
4. Baskerville R and Wood-Harper T (1998) Diversity in information systems action research methods. European Journal of Information Systems 7(2), 90–107. | Article |
5. Bazerman C (1994) Systems of genre and the enactment of social intentions. In Genre and the New Rhetoric (FREEDMAN A and MEDWAY P, Eds), pp 79–101, Taylor & Francis, London.
6. Berger P and Luckmann T (1966) The Social Construction of Reality. Penguin Books, Harmondsworth.
7. Bergquist M and Ljungberg J (1999) Genres in action: negotiating genres in practice. In Proceedings of the 32nd Annual Hawaii International Conference on Systems Sciences (SPRAGUE RH, Ed), IEEE Computer Society, Los Alamitos, CA.
8. Berkenkotter C and Huckin TN (1995) Genre Knowledge in Disciplinary Communication: Cognition/Culture/Power. Lawrence Erlbaum, Hillsdale.
9. Brown JS and Duguid P (1994) Borderline issues: social and material aspects of design. Human–Computer Interaction 1, 3–36.
10. Clarke RJ (1996) The persistence of systems in organizations: genre analysis of systems decommissioning. In Signs of Work (HOLMQVIST B, ANDERSEN PB, KLEIN H and POSNER R, Eds), pp 59–106, Walter de Gruyter, Berlin.
11. Clyne M (1994) Inter-Cultural Communication at Work: Cultural Values in Discourse. Cambridge University press, Cambridge.
12. De Long DW and Fahey L (2000) Diagnosing cultural barriers to knowledge management. The Academy of Management Executive 14(4), 113–128.
13. Goldkuhl G and Lyytinen K (1982) A language action view of information systems. In Proceedings of the Third International Conference on Information Systems (GINZBERG M and ROSS C, Eds), pp 13–30.
14. Günther K (1993) The Sense of Appropriateness: Application Discourses in Morality and Law. State University of New York Press, Albany.
15. Günthner S and Knoblauch H (1995) Culturally patterned speaking practices – the analysis of communicative genres. Pragmatics 5, 1–32.
16. Günthner S and Luckmann T (2001) Asymmetries of knowledge in intercultural communication: the relevance of cultural repertoire of communicative genres. In Culture in Communication (DI LUZIO A, GÜNTHNER S and ORELETTI F, Eds), pp 55–85, John Benjamin Publishers, Amsterdam.
17. Habermas J (1984) The Theory of Communicative Action: Reason and the Rationalization of Society, Vol. I, Beacon Press, Boston, MA.
18. Habermas J (1990) Moral Consciousness and Communicative Action. MIT Press, Cambridge, MA.
19. Habermas J (1992) The unity of reason in the diversity of its voices. In Postmetaphysical Thinking: Philosophical Essays (HABERMAS J, Ed) The MIT press, Cambridge.
20. Habermas J (1993) Justification and Application. MIT Press, Cambridge, MA.
21. Habermas J (1996) Between Facts and Norms. MIT Press, Cambridge, MA.
22. Heath J (2001) Communicative Action and Rational Choice. MIT Press, Cambridge, MA.
23. Hirschheim R and Klein K (1994) Realizing emancipatory principles in information systems development: the case for ETHICS. MIS Quarterly 18(1), 83–109.
24. Honneth A and Joas H (Eds) (1991) Communicative Action. Essays on Jurgen Habermas's The Theory of Communicative Action. The MIT Press, Cambridge, MA.
25. Hoppenbrouwers S and Weigand H (2000) Meta-communication in the language action perspective. Proceedings of the Language-Action Perspective on Communication Modeling, September 14–16, 2000, Aachen, Germany.
26. Kirkpatrick A (1991) Information sequencing in mandarin letters of request. Anthropological Linguistics 33(2), 183–203.
27. Klecun E and Cornford T (2005) A critical approach to evaluation. European Journal of Information Systems 14(3), 229–243. | Article |
28. Klein H and Hirschheim R (2001) Choosing between competing design ideals in information systems development. Information Systems Frontiers 2(2001), 75–90. | Article |
29. Ljungberg J and Holm P (1996) Speech acts on trial. Scandinavian Journal of Information Systems 8(1), 29–52.
30. Ljungberg RJ (1997) Organizations and conversation. Proceedings of the 20th IRIS Conference, Hank, Norway.
31. Luckmann T (2001) On the methodology of (oral) genres. Keynote Lecture, Symposium on Genres, University College, Oslo, May 13–16, 2001.
32. Mark G (2002) Conventions and commitments in distributed CSCW groups. Computer Supported Cooperative Work 11(3–4), 349–387. | Article |
33. Marshall N and Brady T (2001) Knowledge management and the politics of knowledge: illustrations from complex products ad systems. European Journal of Information Systems 10, 99–112. | Article |
34. Mathiassen L (1998) Reflective systems development. Scandinavian Journal of Information Systems 10, 67–118.
35. Miller CR (1984) Genre as social action. Quarterly Journal of Speech 70, 151–167. (revised in Genre and the New Rhetoric (FREEDMAN A and MEDWAY P, Eds), pp 23–42, Taylor & Francis, London.
36. Nonaka I and Takeuchi H (1995) The Knowledge-Creating Company. How Japanese Companies Create the Dynamics of Innovation. Oxford University Press, Oxford.
37. Orlikowski W and Yates J (2002) Genre systems: structuring interaction through communicative norms. Journal of Business Communication 39(1), 13–35.
38. Orlikowski W, Yates J, Okamura K and Fujimoto M (1995) Shaping electronic communication: the metastructuring of technology in the context of use. Organization Science 6, 423–444.
39. Orlikowski W and Yates J (1994) Genre repertoire: examining the structuring of communicative practices in organizations. Administrative Science Quarterly 39, 541–574.
40. Päivärinta T (2001) The concept of genre within the critical approach to information systems development. Information and Organization 3, 207–234. | Article |
41. Peirce CS (1931–35) In Collected Papers, Vol. I–V (HARTSHORNE C and WEISS P, Eds), Harvard University Press, Cambridge.
42. Rehbein J (1994) Rejective proposals: semi-professional speech and clients' varieties in intercultural doctor-patient communication. Multilingua: Journal of Cross-Cultural and Interlinguage Communication 13(1–2), 83–130.
43. Schütz A (1970) In Reflections on the Problem of Relevance (ZANER R, Ed), Yale University Press, New Haven, CT.
44. Schön D (1983) The Reflective Practitioner: How Professionals Think in Action. Basic Books, New York.
45. Searle JR (1969) Speech Acts. An Essay in the Philosophy of Language. Cambridge University press, London.
46. Smith A and Yetim F (Eds) (2004) Special issue on 'global human computer systems: cultural determinants of usability'. Interacting with Computers 16(1), 1–91. | Article |
47. Stamper R (1996) Signs, information, norms and systems. In Signs of Work (HOLMQVIST B, ANDERSEN PB, KLEIN H and POSNER R, Eds), pp 349–397, Walter de Gruyter, Berlin.
48. Suchman L (1994) Do categories have politics? The language/action perspective reconsidered. Computer Supported Cooperative Work 2, 177–190. | Article |
49. Toulmin S (1958) The Uses of Argument. Cambridge University Press, Cambridge.
50. Ulrich W (2001) A philosophical staircase for information systems definition, design and development. Journal of Information Technology Theory and Application 3, 55–84.
51. White S (Ed) (1995) The Cambridge Companion to Habermas. Cambridge University Press, Cambridge.
52. Winograd T (1994) Categories, disciplines, and social coordination. Computer Supported Cooperative Work 2, 191–197. | Article |
53. Winograd T and Flores F (1986) Understanding Computers and Cognition; a New Foundation for Design. Ablex Publishing, Norwood, NJ.
54. Yates J, Orlikowski W and Okamura K (1999) Explicit and implicit structuring of genres in electronic communication: reinforcement and change of social interaction. Organization Science 10, 83–103.
55. Yetim F (1998) Interkulturalität und informatische gestaltung – eine interdisziplinäre annäherung. Informatik-Spektrum 21, 203–212. | Article |
56. Yetim F (2002) Designing 'communication action patterns' for global communication and cooperation: a discourse ethical approach. Proceedings of Xth European Conference on Information Systems, June 6–8, 2002, Poland.
57. Yetim F (2005) A discourse-based meta-communication model for collective sense making. Proceedings of the Language-Action Perspective on Communication Modeling, September 19–20, 2005, Kiruna, Sweden.
58. Yetim F (2006) Bridging diversity: a deliberative approach to organizing and application of usability guidelines. 14th European Conference on Information Systems, June 12–14, 2006, Göteborg, Sweden.
59. Yetim F and Turoff M (2004) Structuring communication processes and enhancing public discourse: The Delphi method revisited. Proceedings of Language-Action Perspective on Communication Modeling. Available at: www.scils.rutgers.edu/~aakhus/lap/Proceedings.htm.
60. Yoshioka T, Herman G, Yates Y and Orlikowski W (2001) Genre taxonomy: a knowledge repository of communicative actions. ACM Transactions on Information Systems 19(4), 431–456. | Article |
61. Yoshioka T, Yates Y and Orlikowski W (2000) Community-Based Interpretive Schemes: Exploring the Use of Cyber Meetings within a Global Organization. Center for Coordination Science, MIT, Boston, MA.

Acknowledgements

I would like to thank anonymous reviewers, the special issue editors, and Nancy Pouloudi for their helpful comments on an earlier version of this paper.

About the author

Dr. Fahri Yetim received his B.S. in Computer Science (1986), M.S. in Information Science (1989), and Ph.D. in Information Science (1994) in Konstanz, Germany. He held a DAAD professorship at the Marmara University Istanbul, German Department of Information Systems (1997–2000), and was a visiting professor at the New Jersey Institute of Technology, U.S.A. (2001–2004). He has published several papers on global aspects of information systems, organized workshops, and co-edited a special issue of the journal Interacting with Computers.