A Rational Solution to the Debate on the Critical Voice in Ethnography of Communication Research

Abstract

This article proposes a way to transcend the debate on the critical voice in Ethnography of Communication (EOC) and qualitative communication research more broadly. First, it demonstrates how EOC’s epistemological paradigm may prevent ethnographers from understanding their subjects fully. Secondly, the article offers a Weberian approach to rational interpretation as a resolution, replacing the concept of “culture” as an a priori explanandum with “practical rationality”. This move demonstrates the feasibility of a unified method in the social sciences capable of dismantling the artificial divide between interpretive and post-positivist philosophies and research designs. Finally, the article provides an illustration of the proposed approach based on some ethnographic data from a volunteer setting of open-source civic software production in Israel.

In a recent publication, Ward et al. (2022) present a critique of Philipsen’s (1975) classic study of masculinity in “Teamsterville” to reinvoke the debate on the critical voice in Ethnography of Communication (EOC). ¹ Specifically, they argue that Philipsen’s approach to EOC known as Speech Codes Theory (SCT) not only prevented him from addressing the fact that the discourse he studied was homophobic, but also concealed from him the role of historical group struggle in the shaping of this and similar discourses. In their words:

While the homophobic speech of Teamsterville men was reported, though unremarked, another essential contextual element was omitted. Philipsen (1975) introduced Teamsterville as “located on the near South Side of Chicago” and “a neighborhood of blue-collar, low-income whites” (p. 13). This terse description slights the intense racial animus that was a prominent feature of Teamsterville life. In the postwar decades, the Great Migration saw an influx of Black Southerners into Chicago’s South Side […] Given what has been learned about whiteness, it seems likely that preserving racial hierarchy was a prominent factor underlying the masculine honor code of Teamsterville.

The present discussion addresses this and similar critiques by explicating the epistemological problem that has triggered their emergence, i.e., the gap between ethnographic description and scientific explanation. For Ward et al. (2022), SCT is incomplete because it prevents one from seeing one’s object of inquiry as an expression of something external to it. In this case, SCT prevented Philipsen from seeing that Teamsterville’s men had a structurally and functionally reasonable motive to develop the honor code he documented – i.e., as a symbolic means by which to negotiate their position of power within the U.S. socio-economic hierarchy, and within Teamsterville itself. To the extent that the limitation of SCT is not a lack of orientation toward social justice but a more fundamental inability to analyze relations of causality, any advancement in the debate will require us to address the following questions: “what is the role of causal explanation in ethnographic interpretation?” And, “how, if at all, are the explanatory factors of critical theory necessary for the accomplishment of this role?” In providing preliminary answers to these questions, the present discussion seeks to intervene into a broader debate – i.e., the question of the possibility for a unified method in the social sciences capable of dismantling the artificial boundary between “interpretive” and “pst-positivistic” ways of thinking and researching. Inspired by the work of Popper (1957), I argue that such a unified method is within hand’s reach, and that its application requires the explanatory factor of collective goals. This factor, in turn, can be used more or less critically in accordance with the questions one asks and the data one has in hand.

The importance of adopting the proposed approach can be seen in research that does not recognize its object as part of a social stratification system which is not cultural or “sui generis” to a locale (Carbaugh et al., 1997). Ward et al. (2022) selected Philipsen’s (1975) study due to the author’s high standing and the work’s relevance. While more recent examples from lesser-known scholars could be provided, this seems unnecessary for the articles’ purpose. Ward et al. (2022) aim was to reexamine the canon in communication pedagogy. The purpose of the present discussion is to provide a new perspective on ethnographic inquiry and ethnographers’ self-positioning within the critical voice debate. As such, it can be considered an example of the writing genre Carbaugh (1989a) proposed as “academic criticism,” whose “objects are communication theories and methods, the locus of criticism being couched within some scholarly community, with the mode of criticism varying from direct to indirect” (p. 270).

To put forth the present criticism in the most direct way possible, the discussion is organized as follows. First, the EOC research program is defined in terms of Carbaugh’s (2007a) paradigm of Cultural Discourse Analysis (CuDA). Second, the epistemological limitation pointed out by Ward et al. (2022) in relation to Philipsen’s study of Teamsterville, and the gap between ethnographic description and scientific explanation at its core, are clarified by Aristotle’s categories of formal and final causation. Third, the required mode of explanation in qualitative inquiry is conceptualized in terms of Popper’s (1957) “zero-method” of rational construction. Fourth, the data by which the article illustrates this method is accounted for. Fifth, the analysis is conducted and its conclusions are drawn.

Rationale

The EOC Paradigm of CuDA

Ethnography of Communication is an approach to language and social interaction that emerged from the linguistic anthropological work of Dell Hymes and his colleagues (e.g., Bauman & Sherzer, 1974; Gumperz & Hymes, 1972; Hymes, 1974). In the early iterations of this program, Hymes sought to construct a cross-cultural taxonomy of social units with the aim of providing a cultural-rhetorical alternative to the then predominant theory of generative grammar (Chomsky, 1965). Within communication, this research program has been further developed in at least three trajectories. First, Philipsen (1987, 1997, 2002) proposed a “speech codes” approach to “cultural communication” that synthesizes between Bernstein’s (1972) notion of language as speech code and Carey’s (1975) notion of communication as ritual. Secondly, Carbaugh (1988, 1996, 2005) followed Gadamer’s (1976) hermeneutic philosophy and Geertz’s (1973) symbolic anthropology in proposing a CuDA approach to the interpretation of symbols for communication practices. Finally, Leeds-Hurwitz and her colleagues (e.g., 1995) have used Scheflen’s (1968) “social communication theory” to study the interplay between competence and performance in communicative occasions. For more recent reviews of EOC, see Carbaugh and Boromisza-Habashi (2015); Boromisza-Habashi et al. (2019); Katriel (2015); Winchatz (2018); Witteborn et al. (2013). A more complete bibliography of EOC is in the making and is partly accessible at: https://nimshav.github.io/EthnoComm-Repository/.

The present discussion takes CuDA as a paradigm of all EOC inquiry due to its ability to model phenomena proper to the other two approaches and to provide a methodological procedure for the investigation of any such phenomenon. CuDA’s key term, communication practice, refers to a descriptive category that incorporates three social units. The first unit, communication act, provides ethnographers with a means to describe how people perform individual actions within a given situation (Carbaugh, 1989b, p. 98), and the variable ways in which all others present interpret that performance (Carbaugh, 2007b, pp. 2–4). The second unit of a communication event provides ethnographers with a means to describe culturally recognized act sequences that require two or more individuals for their performance (Carbaugh, 1989b, p. 99), and involve culturally bounded aspects of social life which have a beginning and end (Carbaugh, 2007b, pp. 2–4). Finally, the unit of communication style refers to the systematic ways by which participants organize and select between alternative acts and events (Carbaugh, 1989b). ² Within CuDA, communication acts, events, and styles are treated as referents of symbols for communication practices. Each such symbol is conceptualized in three ways: first, as a denotative symbol that forms a part of an activity system; second, as a connotative symbol that forms a part of a value system; and third, as a regimented symbol that forms a part of a rule system. The activity, value, and rule systems that converge upon a specific cluster of symbols constitute a cultural discourse (Carbaugh, 2007a). In their capacity as elements of an activity system, symbols for communication practices denote the acts, events, and styles by which individuals constitute forms of social organization. In their capacity as elements of a value system, these symbols orient the acts, events, and styles they denote toward shared values and beliefs about personhood, activity, social relations, affect, and dwelling (Carbaugh & Cerulli, 2013). Finally, in their capacity as elements of a rule system, the symbols are templates for the performance and interpretation of social conduct (Carbaugh, 1987). ³ The relationship among these three systems can be summarized as follows: cultural values depend on normative templates for their consistency; normative templates depend on cultural values for their legitimacy; and both cultural values and normative templates depend on the capacity of individuals to coordinate the performance of specific activity patterns in-situ (Leeds-Hurwitz et al., 1995).

The Epistemological Limitations of EOC

With the CuDA paradigm, EOC is especially effective in demonstrating how communication plays a role in shaping all aspects of sociocultural life (Carbaugh, 1995). But what are the epistemological limitations of this approach? And, how could have these limitations impeded Philipsen’s study of Teamsterville? To properly address these questions, let us recall that the structure of human reasoning comprises three basic elements: an inquirer, an object of inquiry, and a frame of reference in which the object of inquiry achieves its minimal level of intelligibility vis-à-vis the inquirer (Chang, 1996, p. x). In order to demarcate something, one needs to posit it against a conceptual background. In quantitative communication research, this background is a theory that specifies relationships among statistical variables. In qualitative research, and especially within EOC, it is a theory of social grammar that provides practitioners with categories and continua of units and components for the classification, modeling and comparison of human (inter)actions and interpretations. From a philosophy of science standpoint, this kind of ethnographic theorization can be classified into the type of causation that Aristotle (1996) called “formal.” As such, EOC assumes that observable phenomena depend for their consistency and legibility as tokens of types on the units and components that pattern them in space and time. Much like a set of geometric forms, these nominals are “abstract entities that exist only in the analyst’s descriptive framework” (Duranti, 1985, p. 201). The object of EOC theorization is therefore a conceptual framework by which to construct a formal-causal explanation of any communication phenomenon. Accordingly, creative development of EOC theory occurs when inquirers discover that their descriptive grammars do not fit with the patterns they attempt to trace. ⁴ The CuDA paradigm as presented above can be taken as the state-of-the-art product of this line of theoretical innovation and development (Scollo, 2011). ⁵

While formal-causal theorization is essential to any ethnographic research, it is limited by its condition of legibility as a distinct type of causation (i.e., one can classify a given explanation as “formal” only in relation to other types of explanation which are “not-formal”). Indeed, Aristotle (1996) constructs the type formal causation only in contradistinction with the types he calls substantial, efficient, and final. In brief, substantial causation is any explanation of an object by that object’s substance; efficient causation is any explanation of an object by that object’s dependency on a prior object; and final causation is any explanation of an object by that object’s teleological end. In conceptualizing one’s object of inquiry in formal-causal terms, one will not be able to see how the three other types of causation might enter its constitution. From its inception, EOC was designed to discover the “Whats” and “Hows” of a communication system – i.e., its composite elements, the patterned use of these elements, and the deep values and beliefs with which such uses are associated. Undoubtedly, the EOC research program has made important contributions to our understanding of human communication. But it is equally clear that scholars working within this approach have been insufficiently equipped to address “Why” questions with regard to their objects of inquiry.

Toward a Procedure of Rational Construction in Communication

What EOC lacks, then, is a scientific method by which to determine why a phenomenon such as Teamsterville’s honor code could have emerged in the first instance. The question is not simply: “why are members of this speech community homophobic?” Rather, the question is: “what motivates these members to use this homophobic discourse in this historical juncture?” To address this question, one must change the epistemological context of reference from CuDA’s formal-causal paradigm to a paradigm premised on principles of efficient or final causation. Within communication, epistemological theories of efficient causation are less likely to fit the bill as they are designed to guide studies that measure effects. The methodological paradigm of these studies is the controlled experiment where temporal relationships between stimuli and responses – or independent and dependent variables – can be tested and replicated. This paradigm cannot be easily integrated into ethnography for at least two reasons. First, it contradicts the traditional commitment of ethnographers to the study of naturally occurring phenomena. Secondly, the empiricist epistemology of efficient causation is incompatible with the idealist epistemology of formal causation as used within qualitative communication research. Whereas the explanans of efficient causation are empirical objects that precede their explananda in space and time, the explanans of formal causation are a priori wholes or ideal types of which the explananda are presumed to be parts or tokens. Ethnography of Communication scholars thus treat observable units of communication as expressions of an a priori system of such symbols and meanings, or a culture (Philipsen, 1997). Culture as a presupposed entity is assumed to be intersubjective (Reijven & Townsend, 2021) – i.e., to exist as shared knowledge within the brains of participants in a given speech community. Anything an actor does or says can thus be interpreted as a display of their general capacity to (mis)use symbols in relation to norms. Cultural interpretation hence forces one to presume a perfectly competent “model speaker” who knows how to perform correctly in any occasion proper to the studied culture (Hart, 2017; Leeds-Hurwitz et al., 1995). ⁶

The Weberian approach to final causation advocated here has a better fit with the interpretive orientation of formal-causal theorization as it substitutes the a priori explanans of culture with that of reason. ⁷ The methodological paradigm of this approach is the microeconomic model. Analytically, it requires one to enter the hermeneutic circle with the presupposition that any cultural form is practically reasonable – i.e., that its usage satisfies the desires of some agent. Human agentive desires are distinguishable into collective and individual goals – i.e., the ends of an individual group and of each individual within it (Hymes, 1972). On general grounds, a satisfactory explanation of collective goals should provide micro-foundations for the object in question (Elster, 1985, pp. 15–16). The most parsimonious final-causal explanation of culture will therefore invoke nothing but rational and selfish motivations among cultural members (Elster, 1985, p. 16). This, however, does not mean that the explanation cannot be critical. For instance, the collective goal of a population within some speech community might reflect a class interest. This interest could be used to explain, for example, the existence of a local discourse that helps members of this population to benefit from the activities of other populations. In order for such an explanation to hold, though, one must show how the presumed beneficiaries manage to overcome the impediment on all social cooperation known as the free rider problem (Elster, 1986, pp. 29, 38; Olson, 1968/2002) – i.e., the default rational choice to sit back and let somebody else perform the work (in this case, the work of symbolic domination). Significantly, this approach would have enabled Philipsen to identify a collective goal that is likely to motivate Teamsterville’s men to display antisocial behaviors, and to explain how these men are likely to coordinate the upholding of such displays across social scenes where these might be costly. More generally, an approach to rational interpretation promises to provide the discipline with a specializing means by which to explain any communication system – including anomic systems such as that of Teamsterville.

Analytically, the proposed approach is a version of Popper's “zero-method” of rational construction. By his account (1957, p. 141):

I refer to the possibility of adopting, in the social sciences, what may be called the method of logical or rational construction, or perhaps the “zero method.” By this I mean the method of constructing a model on the assumption of complete rationality […] on the part of all the individuals concerned, and of estimating the deviation of the actual behavior of people from the model behavior, using the latter as a kind of zero co-ordinate.

Instead of starting from the actual experiences of participants, one takes the imaginary standpoint of a purely practical-rational actor to construct the most efficient cultural discourse by which a multiplicity of such actors will achieve a collective goal. ⁸ The key methodological consideration here has thus to do with the distinction between predictive and diagnostic modes of inquiry. For Popper, scientists should only use predictions due to their greater amenability to falsification. Popperian ethnographers are thus required to construct rational models of the cultural discourses they expect to find, and then test these hypotheses in the field. This requirement is problematic, however, because it ignores the prosaic fact that the need to construct and defend diagnoses is part of the human condition, and is therefore pervasive within the social and natural sciences too. The present discussion thus proposes to think of Popperian rational construction in terms of a continuum between prediction and diagnosis. At best, ethnographic work will fall in the middle as qualitative analysts – like police detectives – can only elucidate the relationships between reasons and actions through an ongoing interplay “between part and whole, between focus and background, between implicit comprehension and explicit interpretation” (Chang, 1996, p. x).

To systematize this interpretive dynamic into a replicable procedure whose results could be tested as hypotheses in experimental settings, the proposed approach further specifies three analytical steps: (a) contextualizing identification of participants’ goals and the conditions that constrain their accomplishment; (b) rational construction of a cultural discourse by which practical-rational actors are likely to accomplish these goals under these conditions; and (c) critical examination of this prediction against ethnographic data. Before turning to illustrate this procedure, let me briefly review the data used in the analysis.

Method

Field-Sites and Data

The group whose cultural discourse is analyzed here is The Public Knowledge Workshop (PKW), a voluntary association of software developers who undertake to build internet websites through which detailed information pertaining to government activities and officials may be made more transparent and accessible to the public. As apparent from the group’s website, its mission is to provide civic actors such as investigative journalists with specialized software tools for the enforcement of public accountability. ⁹ Participants develop these software through small project teams known by the local term “electroknights,” which is sometimes abbreviated to the English word “eKnights.” ¹⁰ Within PKW, there may be multiple projects operating simultaneously on completely separate civic tools, each of which is referred to as its own eKnight. The volunteers in the different eKnights are associated administratively to PKW, but function mostly within the confines of their individual teams. As independent units of sociality, these teams organize around the developers who started the projects. For the most part, these project founders have working relations with at least one highly committed co-developer. A second and wider circle of participation consists of slightly less committed and productive volunteers who regularly attend the development meetings or otherwise participate in the projects online. The number of these individuals is relatively small (one to five persons). At the time of my fieldwork, participation in PKW was anchored in six such initiatives. Five teams regularly attended the development meetings, and one team worked primarily online. The activities of these six teams are documented in the online GitHub platform of open-source software production, and therefore are easily accessible.

Data Collection Procedures

The data used in the analysis derive from an ethnographic study conducted between 2013 and 2017. In 2015, PKW would hold two weekly development meetings in a high-tech tower rented by the Google corporation in Tel Aviv, and in a high-tech “hub” at the National Library of Israel in Jerusalem. Additionally, PKW administrators would organize occasional “hackathons” – i.e., intense development meetings that encompass an entire weekend. ¹¹ Finally, one of PKW participants – a professional programming teacher – started in 2014 a personal initiative he called “hackita,” an amalgamation of the English word “hack” and the Hebrew word “kita” or classroom. Hackita is an 8-week programming course that focuses on the technological means by which most eKnights are produced. ¹² In total, I attended development meetings in Tel Aviv 10 times for 3–4 hours per visit, and the hackita meetings in Jerusalem six times for 6–8 hours each. This attendance provided me with opportunities to participate in a small open government data conference, a variety of public lectures, and other meetings between the eKnight founders and hackita students. My naturalization into this volunteer setting also provided me with crucial access to participants with whom I could conduct more in-depth interviews. Altogether, I managed to talk with 10 individuals, eight of whom were active or veteran participants in one of the eKnights. The other two were members of PKW's administrative body. ¹³

Results

The concrete product of this article is a three-step procedure that will allow ethnographers and other qualitative inquireres to address “Why” questions and thus to gain final-causal knowledge about their objects of inquiry. In illustrating this procedure, the following analysis shows that PKW’s cultural discourse of work can be best explained by the collective goal of project founders to maximize utility under conditions of time scarcity and command-inability. The analysis further indicates how the steps under consideration could resolve the epistemological issue raised by Ward et al. (2022), and to provide the discipline with a potential means by which to gain insight into the raison d'être of any cultural discourse.

Step 1: Contextual Identification of Participants’ Goals and Limiting Conditions

An implementation of the proposed procedure can be best accomplished through the construction of a model that explains how cultural discourses operate as means to practical ends. Hence, in the first analytical step, the ethnographer seeks to gain knowledge about participants’ collective goals and the limiting conditions that constrain their accomplishment. An identification of participants’ goals requires an examination of the history, politics and sociology of the speech community under study, including through the conduct of pilot interviews with key informants. The aim of these moves is to develop and test hypotheses about the differing goals and interests of specific populations within the speech community studied. For example, an examination of Teamsterville’s history and social character could have provided one with the notion that the identity of being “white,” “blue-collared” and “man” is under threat, and that this could motivate participants to engage in some form of collective action. To further specify this form, one could have looked into the limiting conditions that constrain participants’ behaviors and choices – i.e., the physical, psychological, and social boundaries of the speech community under study. If Ward et al. (2022) are correct, then Philipsen could have guessed that Teamsterville’s men are constrained in such a way that they will be inclined to preserve the status of their shared identity through an intensification of this identity’s expressive order (rather than, for example, through a geographic relocation).

In the case of PKW, it is easy to recognize the centrality of eKnight founders in the respective projects. These individuals, however, do not share a single collective goal, and, from the standpoint of PKW’s administration, there is no reason they should as their individual goals sufficiently motivate them to lead the production processes into completion. Indeed, I found that project founders volunteer in PKW for a variety of reasons that include the solution of problems that bother them personally, the satisfaction of “itching” curiosities about various technical and social issues, the reclamation of civic agency in what they sometimes perceive as an apolitical social environment, the acquisition of new technical skills, and the creation of professional alliances that could help them promote their careers in the high-tech industry. That which unites the eKnight founders is therefore not a common utility, but rather the common interest of practical-rational actors to gain their different utilities at minimal production costs.

In terms of extra-discursive constraints, participation in PKW is limited by an economic condition of time scarcity and a political condition of command-inability. The condition of time scarcity results from the structural opposition between work and leisure at the foundation of modern industrial society (Turner, 1982). To volunteer in PKW is to give up recreational activities that serve as mental respite from the type of work modern actors are required to perform in their professional, domestic, and civic capacities. This not only explains how voluntary production time in PKW is scarce, but also why participants are constrained by the political condition of command-inability. PKW's volunteers are free from hierarchical structures of command and control, and specifically from those found in their workplaces, because their practices of software production take place in a short time-span where each of them is required to make decisions freely and perform work of their own choosing. Insofar as participation in any voluntary project is mostly a leisure time activity, it requires participants to “do what they want” and thereby constitute a libertarian relationship where the obligation to respect one’s teammates’ freedom of choice principally overrides any consideration of solidarity. One can therefore speak of a distinct social order that arises within the modern realm of leisure to suspend institutional statuses, positions, and roles so as to constitute a libertarian heterotopia where each participant allows the others to exist as autonomous individuals par excellence. This alternative social order that precedes and constrains the choices of individuals can be defined in terms of two reciprocal norms: voluntary participation and voluntary selection of tasks (cf., Weber, 2004, p. 62).

A categorical imperative to abide by these norms calls attention to the free rider problem. Insofar as any rational actor wants to enjoy the fruits of other people’s labors, all participants in any given eKnight are expected to sit back and wait for somebody else to work in their place. In this case, the initiative breaks down as no one writes the initial lines of code that constitute an eKnight. While this problem is valid, it does not threaten the existence of most such initiatives for two reasons. First, voluntary software projects rarely take shape within a social setting where the ratified members of a project team are already gathered and ready to begin the production process. To the contrary, the typical project emerges as a personal task of one developer who seeks to gain an immediate benefit by pursuing it. If this benefit is gained, no transition to a software initiative is needed and so is unlikely to occur. If, however, such individuals realize that they cannot complete the project on their own, or that the costs of such endeavors are too high, they will be motivated to publish their source codes so as to attract volunteers who can help them. The second reason is that the cost of sitting and waiting for somebody else to perform a desired task is often greater than the cost of giving up the minimal amount of leisure time necessary for the completion of this one task. These two reasons highlight the distinction between the substance and force of individual goals. While two equally competent developers H and S may desire the same benefit, H’s desire could be stronger than S’s or vice versa. One could therefore expect to find a positive correlation between the degree to which developers desire specific benefits and the amount of leisure time they are willing to give up in order to gain them. In other words, programming tasks will not be distributed equally, and this unequal distribution could be partly explained by the relative force of participants’ motivations to carry out the projects into completion. A quantitative examination of the division of labor in the project that brought about the formation of PKW in 2011 confirms this prediction (Appendix 1).

To the extent that developers do not expect somebody else to offer them something in return for a set amount of work, one can speak of the social arrangement of a “hive.” In this constellation, self-interested work at a distance maximizes each participant’s benefit from the collective effort of the group regardless of how much code each programmer contributes. The free rider problem is partly resolved by a mode of uncooperative-cooperation whereby participants give up the pleasures of comradeship in order to accomplish their individual goals. This social arrangement, however, is particularly vulnerable to a risk of distraction whose features are articulated by Elad, an eKnight founder, as follows:

(1)	Interview with an eKnight founder (1/7/2016)
1.	Ela:	What I’m finding most disrespectful […] is people who waste my time.
2.	XXX:	What does it mean to waste your time?
3.	Ela:	Now this is an important point because it is very painful on a daily basis. […] The actual situation today is that you constantly have to make presentations about the project to people who think it’s cool, while knowing that two hours later you’ll not hear from eighty percent of them. […] And since I’m both a project founder and the main developer, so all this time I’m in a meeting- and there are periods when the development meeting is the only time I’m available to work on [the project], so like there’s no progress and it’s very very frustrating.
4.		[…] When you meet a new volunteer he feels like okay I came to contribute and change the world and I’m like one of one hundred. [And I think,] as if you would stay long enough to try writing a useful code. And that’s before I even know if he is able to write good code.
5.		[…] The experience of volunteering in the workshop became very frustrating to me once I was dragged into this preoccupation. Once I saw that I invest time in it and it doesn’t bring any impact. And then I found myself saying like fuck it I don’t accept volunteers. […] And even when I do, I am very unwelcoming. I can say that I started with the approach that one should be welcoming, and I’m still trying to be more communicative. And let’s say that my pride is that I do manage to operate volunteers more than other leaders. But it’s just wearing out […] and frustrating, and doesn’t return itself. […] In this sense, when a new volunteer arrives, it doesn’t pay to […] invest time in him.

In the desirable scenario, an eKnight founder learns about the existence of a contributor only within the framework of code contribution. This practical condition functions to ensure that founders will only enter into social interactions with developers who have already demonstrated their value to the project. However, this condition cannot be easily reproduced offline when curious programmers arrive at the meetings just in order to see what their peers are doing (L:3–4). It is precisely here where one finds the problem of social cooperation proper to PKW's volunteer setting. On the one hand, project founders have an interest in increasing the number of contributions to their eKnights. On the other, these individuals are aware that their chance of encountering newcomers at a development meeting who are not only willing and able to volunteer, but are also ready to work as much and as hard as they do, is approximately 4%. The practical question for PKW's project founders then becomes: “how can one maximize the chances of singling out the “ideal volunteer” among newcomers to the development meetings?” Importantly, this question, and the need to resolve it by discursive means, is relevant to a specific population within PKW’s speech community – i.e., the in-group class of project founders. To the extent that these individuals share a basic desire to enjoy the fruits of their labors, they must also share an interest in reducing the time wasted on volunteer recruitment. Knowing this in advance allows the ethnographer to develop a predictive model of the cultural discourse by which these individuals qua practical-rational actors are likely to resolve the problem.

A contextualizing identification of participants’ goals and the limiting conditions that constrain their accomplishment is thus shown useful in identifying a focal collective goal that provides the raison d’être of a cultural discourse to be found as prominent in the community under study. In the case of Philipsen (1975), this analytical step could have helped one identify the collective goal of Teamsterville’s men to preserve the status of their shared identity within their intra-communal gender relations and the U.S. socioeconomic hierarchy more generally. Similarly, in the case at hand, this analytical step has been shown useful in identifying the collective goal of a specific in-group class – i.e., PKW’s project founders – to improve their process of volunteer recruitment. In both cases, the factor of collective goals provides a means for constructing scientific predictions and explanations.

Step 2: Rational Construction of the Requested Cultural Discourse

The second analytical step develops a predictive model of the cultural discourse by which members of the focal population are likely to accomplish their collective goal. In some cases, such as that of PKW's in-group class of “eKnight founders,” the discourse in question emerges to solve an instrumental problem. In other cases, such as that of Teamsterville's in-group class of “men,” the discourse arises to provide for more symbolic or political ends. Assuming that such differences in participants’ collective goals matter less than the means by which the goals are accomplished, and building on the results of the previous step, the question becomes: “if members of the focal population are rational, informed and self-interested actors, and if these actors were to achieve the stated collective goal under the limiting conditions of this particular social setting by means of constructing a cultural discourse, what will this discourse be?” While the logical theorems that analysts might construct in addressing this question cannot be too specific, they can provide falsifiable models by which to guide the inquiry.

In order to construct such a model in the case of PKW, one must first recognize that the new volunteers who constitute the minority of viable participants in the offline development meetings (i.e., the elusive 4%) and the new participants who enter the eKnights online through acts of code contribution belong to the same category of persons. Secondly, one will have to identify the traits that make members of this category desirable. Finally, one will have to imagine some effective means by which to examine the qualities of these traits among newcomers to the development meetings. Under the limiting conditions of time scarcity and command-inability, the required traits can be deduced as follows. First, any practical-rational eKnight founder will want to work alongside competent programmers who will not require continual guidance throughout the development process. Hence the trait of competency. A second, related trait is proactivity qua the ability to observe the norms of voluntary participation and voluntary selection of tasks. In selecting independent and self-reliant developers for their teams, eKnight founders will increase their available time to focus on essential tasks. Last is the trait of assiduousness or the capacity to take on tedious programming tasks and carry them out to completion.

This model of the ideal volunteer as a competent, proactive and sedolous programmer would surely provide project foundrs with a means by which to describe the characteristic person they like to work with. However, in order to single out volunteers who approximate the ideal type in-situ, founders will also have to create means for the expression and evaluation of the desired traits of competency, proactivity and assiduousness. In CuDA terms, these traits belong to the discursive hub of personhood and should thus be conceptualized as elements of a local value system. To the extent that values do not exist on their own but are ingrained in practice, one must also presume an activity system by which participants operate, display and sanctify the values. And to the extent that elements of this activity system are reiterative, one must also postulate a rule system that helps participants regiment practices as normative templates for social conduct. In approaching the ethnographic data, then, one can expect to find at least one symbol that denotes a specific practice by which participants bring the virtues of competency, proactivity and assiduousness into a sharp relief.

Step 3: Model Testing

To properly examine this predictive model is to conduct a blindfolded analysis of PKW’s entire discourse of work from the bottom-up. While this analysis was conducted prior to the present discussion, it goes much beyond the scope of a journal article. Moreover, to present the analysis in full is to sidestep the aim of this article. What is at stake here is not the particular features of PKW’s discourse of work, but the epistemological implications of our gained ability to engage with a predictive mode of inquiry. Hence, for example, using this mode of inquiry in Philipsen’s (1975) case would have allowed one to arrive at a scientific theory that explains why Teamsterville’s men display homophobic attitudes. This theory, in turn, would have allowed one to contribute to the social sciences more generally – e.g., by specifying how certain people are likely to behave under conditions of identity threat.

In accordance with these considerations, the analysis conducted below focuses on a single symbol for communication practice that makes the virtues of competency, proactivity and assiduousness salient. Rather than presenting an exhaustive interpretation, the analytical aim is to show that ethnographers can anticipate their objects of inquiry. To anticipate the object of a cultural discourse is to make a falsifiable claim about its teleological end on the assumption of complete rationality on the part of all the individuals concerned. If the claim is confirmed (as is the case here), then one has a scientific theory of one’s object of inquiry. If the claim is falsified, one will have to explain the deviation of the actual discourse of participants from the model discourse (e.g., by psychological factors specified in the literature). Such explanations, in turn, can be best formulated as hypotheses in subsequent research.

The particular symbol for communication practice examined here is “la’asot cod” (to do/make code). As an element of an activity system, this symbol denotes a preferred communication style which is often contrasted with the pejorative style denoted by the term “la’asot ra’ash” (to do/make noise). This opposition is readily available from the following commentary by Liat who served as a community coordinator for PKW.

(2)	Interview with a Community Coordinator (12/24/2015)
1.	XXX:	Because maybe there’s no continuity and there’s no internal dynamics that generate an agenda?
2.	Lia:	I think it depends. Did you happen to talk to yoav?
3.	XXX:	[…] he is on my interviewee list, I just haven’t got to him yet.
4.	Lia:	He is a really really good guy. I learned a lot from Yoav. […] Lots and lots. Because he told me one of the most interesting things.
5.		[…] He told me, I know that most of the code will fall on some three people. There will be people who come and go, but like […] twenty percent will do and eighty percent will make noise.
6.		Like ((chuckling)) there’s a huge difference between la’asot cod and la’asot ra’ash.
7.		[…] And he was right.

In this excerpt, Liat, who did not have prior experience participating in voluntary software production, indicates that Yoav played a significant role in introducing her into this scene (L:4). The most significant lesson she learned was that the division of labor in the typical eKnight is extremely imbalanced (L:5). To explain this imbalance, Liat follows Yoav in referring to the Pareto principle, according to which one should expect to find that 20% of the participants in an eKnight will make 80% of that project’s source code while making 20% of the “noise” (or verbal interaction) surrounding the production process. The other 80% are expected to write only 20% of the source code while being responsible for 80% of the “noise” made by the team. From here we learn, first, that PKW participants are aware of the limiting conditions under which they operate. Secondly, one hears the communication practices that mediate this awareness. The most productive and dedicated volunteers refer to themselves as persons who “do/make code” in contradistinction to their less productive peers who mostly “do/make noise.” ¹⁴

Such hearable oppositions between terms for communication styles are often used as means for the organization and evaluation of other terms that denote locally recognizable acts and events. This is clearly the case in PKW as apparent from the following excerpt where Yoav contrasts between the styles of “la’asot cod” and “la’asot ra’ash” to differentiate between proper and improper usages of other acts and events within the development meetings.

(3)	Interview with an eKnight Founder (1/10/2016)
1.	XXX:	And if you can go in, you know more specifically, what the structure of the development team is, what roles there are, how it organizes in terms of work itself, how many people are there at any given time?
2.	Yoa:	[…] There are many people who ask you lots of questions ((chuckles)), they make lots of discussions and you don’t see code.
3.		I’m very much in the business of show me a code.
4.		[…] I think that one of the problems of people who come to open source is that they ask for permission […] and feel that they are expected to come up and suggest something, get an approval, go on and do it, and then it will be merged into the project.
5.		It’s not like that! You are supposed to come and see if there is something here that bothers you, that you think can be bettered. Do it. If you want to propose beforehand it’s okay. But […] don’t expect to have a prolonged discussion. Do it and […] if it is good it will get in.
6.		By the simple fact that you did something good, you contributed to the project in seconds.
7.		[…] In open source, do not ask for permission and do not expect somebody else to do something. Just step forward and hand it in.

As apparent from this excerpt, Yoav defines the symbol of “showing code” (L:3) as a proper alternative to pejorative labels of face-to-face acts and events such as “asking questions” (L:2), “making discussions” (L:2), and “asking for permission” (L:4). The valued symbol of “showing code” refers specifically to the first move in the act sequence by which a code contributor requests a project founder to accept their contribution via a technical act called “pull request” (L:4). In this sense, a volunteer who only “shows code” to a project founder within a development meeting is no different from any participant in a voluntary software project that runs online. As apparent from L:2, asking a founder “lots of questions” about their project, or leading a project team into “prolonged discussions” are undesirable because they come at the expense of scarce programming time.

What, then, is the character of the virtous volunteer who only “does code” in PKW development meetings? The typical answer to this question can be heard in the next four lines of excerpt 5 where Yoav compares between different modes of volunteering in the eKnights:

8.	[…The best case] is that a person comes and really gets into the code […] And this is the rarest case. And there are people like that. […] The guys I worked with on [a specific eKnight] whom I didn’t know beforehand- so let’s say that one of them is a superb programmer who is very strong in the field of software testing. And tests are something that programmers don’t like to write, and it is hard– there were no tests there. And he came and built for it a proper testing framework at a really serious level. So here is a guy who came in and made a very significant contribution.
9.	[…] In the best case you have someone who also has the initiative, and he also works and makes the features, and also makes them properly. But this is what everyone is looking for. It’s very rare.
10.	The worst case is that you come up with ideas. Like you’re not doing them. I don’t want to say that this is the worst, because many times people came up with ideas, and the ideas were good, and I did them. But […] if you become someone who always comes up with ideas […] and gives suggestions for improvement but do not do anything, so your situation is pretty bad.

On this account, the most desired volunteers are “superb programmers” who take the “initiative” to “really get into the codes” of PKW’s eKnight founders with the aim of locating and resolving product issues that “bother them” personally, and of “showing” these solutions to the founders through a practice locally known as “code donation.” Accordingly, the least desired volunteers are persons who “come up with ideas and suggestions for improvement” with the expectation that somebody else will work in their place.

This commentary sufficiently indicates the salience of competency (being a “superb programmer”), proactivity (being a person who “takes initiative”) and assiduousness (being a person who “really gets into the code”) as moral criteria of character assessment in this speech community. At minimum, we find that PKW’s project founders are aware of the traits that make volunteers in the group's semi-industrial setting desirable, and that to assign a higher economic value to these traits is to sanctify them as cultural values. It could therefore be the case that members of this in-group class have contributed to the facilitation of “false-consciousness” whereby their collective goal translates into a normative system of justifications, or an ideology. As Ward et al. (2020) indicate, a similar “gender-ideology” was most likely hearable in Teamsterville's communal function at the the time of Philipsen's (1975) investigation.

Discussion

The overall aim of this article has been threefold: to propose a way by which to resolve the debate on the critical voice in EOC; to provide qualitative inquirers with a scientific method of final-causation by which to explain their objects of inquiry; and to provide evidence for the possibility of a unified method capable of dismantling the opposition between “interpretive” and “post-positivistic” types of research design in the social sciences. The following remarks briefly attend to each goal.

Contribution to the Debate on the Critical Voice in EOC

So far, the debate on the critical voice has been a struggle between two conflicting perspectives: one that is phenomenological and bases all interpretation on the viewpoints of participants, and the other that approaches data through a critical lens focused on social justice. Proposed solutions include focusing on the local critiques of participants as “organic intellectuals” (Carbaugh, 1989a; Gramsci, 1971; Katriel, 2015; Katriel & Shavit, 2013; Lie, 2020; McLaughlin, 1996). However, these solutions are only partial as the native discourses of participants may not reflect their integration into larger social structures as seen in the case of Teamsterville’s men (Ward et al., 2022). Incorporating a critical mode of inquiry as an optional phase of CuDA (Carbaugh, 2007a) is also limited because it does not tell one how to select among the available critical frameworks, leading back to the original dilemma. The present discussion has attempted to transcend this deadlock by showing that the epistemological horizon of EOC might not allow one to see critical aspects of one’s object of inquiry. If, in fact, one’s object functions strategically, for example, as a mode of subjugation, one would like to know this not only for moral and political reasons, but also for epistemological ones. To use an analogy, while ethnographers are heavily equipped to provide a precise description of a person “crying” along with a manual for how to “cry” properly in this culture, they have little scientific means by which to determine why this person is crying or why people in this culture are expected to cry in such a manner. It will be therefore productive to assume that Philipsen did not identify the rational goal underlying Teamsterville’s honor code because such rational goals, which can reflect structural-functional class interests and other kinds of group struggle in society, fall outside the hermenutic formal-causal horizon of EOC theorization.

Contribution to Qualitative Communication Research

As part of this effort, I was tasked with developing a final-causal framework and a methodological procedure that can be applied in EOC and qualitative communication research to explore the factors of individual and collective goals from a hermeneutic perspective of practical rationality. The analysis presented above demonstrates that interpreting discursive forms as symptoms of practical reason can lead to the creation of falsifiable models that can predict aspects of cultural discourse. These models are not limited to the study of anomic phenomena that involve group conflict, and are therefore not necessarily critical. As demonstrated, a rational interpretation can provide insights into morally neutral phenomena such as the cultural discourse used by PKW project founders to improve volunteer recruitment.

In developing these claims, the present discussion aligns with the efforts of qualitative communication scholars to explain observable forms of interaction by associating them to macro-social structures. For example, within interactional sociolinguistics, Shrikant (2022) and Shrikant and Musselwhite (2019) have recently linked forms of talk to categories of race/ethnicity and economic ideologies respectively. Within EOC proper, Katriel (2021) introduced a framework for theorizing forms of defiant speech as part of a language ideology specific to the Israeli political-activist subfield. Similarly, Boromisza-Habashi and Fang (2020, 2022) have theorized the market value of communication forms within the economic field of contemporary global capitalism. The approach to rational interpretation proposed here advances this trend by providing scholars with a way to verify their micro-macro associations. Given that any act of compliance or resistance to a dominant discourse or an institution carries costs, a full explanation of either kind of practice will require a micro-foundational examination. The question of how this could best be done will be explored in future publications.

Contribution to the Unification of the Social Sciences

Does this mean that rational interpretation could work in any situation or speech community? At this preliminary stage, it is challenging to provide a definitive answer as further empirical research is required. However, what can be concluded is that the capacity of the proposed procedure for falsification and replication helps bridge the gap between qualitative and quantitative research methods and their associated “interpretive” and “post-positivist” philosophies. To date, pedagogy in the social sciences accepts this artificial and largely unwarranted opposition and the division between descriptive and explanatory research it creates (e.g., King et al., 2002). Returning to Popper’s “zero method” of rational construction, I have attempted to challenge the prevailing consensus by highlighting the importance of adopting a unified standard for knowledge production in the social sciences. Unless others join this effort, the current discussion will lack a contemporary comparison. And while the idealist approach to rational interpretation advocated here may seem to conflict with the empiricist approach to efficient causation that dominates our discipline, the two approaches are actually complementary. For example, in this article, I have argued that a specific cultural discourse is designed to optimize the accomplishment of a specific collective goal by logical necessity. If I am right, any class of participants who encounter the problem discussed here will develop a cultural discourse that resembles the one analyzed above. This hypothesis can be further tested in experimental settings. Critical insights from such quantitative research should be, in turn, integrated into the assumptive bases of rational interpretation. This proposal, too, should best be developed in and qualified by future research.