Language, ethnicity, and the nation-state: on Max Weber’s conception of “imagined linguistic community”

Abstract

Methodological nationalism in sociological theory is unfit for the current globalized era, and should be discarded. In light of this contention, the present article discusses Max Weber’s view of language as a way to relativize the frame of the national society. While a “linguistic turn” in sociology since the 1960s has assumed that the sharing of language—linguistic community—stands as an intersubjective foundation for understanding of meaning, Weber saw linguistic community as constructed. From Weber’s rationalist, subjectivist, individualist viewpoint, linguistic community was a result of social actions, not a prior entity as assumed by German metaphysical organicism (and historicist holism). Indeed, Central Europe in Weber’s era was a battlefield of linguistic nationalism(s); in contrast to the national societies of the Cold War period, national borders were unstable and ultimately the multiethnic empires of the region were dismantled after World War I into ethnolinguistic nation-states. Experience of this contemporary reality brought Weber to the core of the relationship between language and politics: A language community is an imaginary one demarcated not by language itself but by conscious opposition against outsiders, with monolingual contexts within borders created artificially by homogenizing policies like linguistic standardization and national education—the first modernity of language. In this way, Weber felt, language can be a means to domination.

The concept of ‘ethnic’ community, which disappears with exact concept-formulation, corresponds in this regard, to a certain extent, to that of ‘nation,’ one of the concepts which vex us with emotional sensations the most....

—Weber [1921a] 1980a, p. 242 = [1968] 1978, p. 395¹

[I]n a nineteenth-century Europe in which Latin had been defeated by vernacular print-capitalism for something like two centuries, these solidarities had an outermost stretch limited by vernacular legibilities. To put it another way, one can sleep with anyone, but one can only read some people’s words.

—Anderson 1991, p. 77

Under ongoing globalization, the field of sociology has been expected to reconsider its methodological nationalism, that is, the identification of national boundaries with societal ones. According to Ulrich Beck, traditional categories of sociology are “zombie-categories” (Beck [2002] 2009, p. 8) with limited capacity to reflect current social realities, since they still rely on the nation-state frame. The practicability of methodological cosmopolitanism in sociological research may be surely questionable. Statistical social data, for example, are often for technical reasons gathered and analyzed in terms of national units. However, the notion that people’s social lives are operatively confined to a national territory, or a national society equated with a homogenous nation-state, is already anachronistic. Since modern life is inseparably woven into a world-wide, diversified system of division of labor and a globally expansive information network, it cannot be maintained under national autarky. In this sense, the actual society inhabited by people today is “the world society” (Luhmann [1971] 1975), extending beyond national boundaries.

Against this background, this article aims to clarify Max Weber’s perspective on language and show how his linguistic views may provide substantial cues for a critical examination of methodological nationalism, particularly in sociological theory. For this purpose, in the following sections, I demonstrate first that Weber viewed the language community as an imaginary construct, a fiction. From a perspective of rationalism, subjectivism, and individualism, he consistently denied the objective reality of such a community as part of a larger rejection of German metaphysical organicism (and historicist holism). Second, I analyze the political character of this linguistic community, with reference to the theoretical implications of Weber’s general concept of community, according to which a language community is demarcated not by language itself, but by conscious opposition against outsiders. Third, to complement these considerations, I discuss the relationship between language and the nation-state in terms of certain social backgrounds. In contrast to the postwar period, which brought the stabilization of national societies and the naturalization of “sharing the same language” inside borders, Weber’s Central Europe was a major battlefield of linguistic nationalism and boundary changes. Finally, I suggest that a gap in linguistic competence, especially in English, can lead to the stratificatory differentiation of the world society beyond national borders.

However, before all of these discussions, in the next section I start with a brief preliminary consideration of a linguistic turn in sociology since the 1960s.²

Preliminary consideration of the linguistic turn in postwar sociology

Sociological theory began to focus attention on language only surprisingly recently, from the 1960s onward. Before that, sociological theory was strongly influenced by Talcott Parsons’s idea that normative values backed by religious (irrational) beliefs orient human actions toward the social order (Parsons 1978, p. 240). However, Parsons’s presupposition of “shared religion” became ever less tenable as religion increasingly was left to individual choice in the postwar era.³ This individualization (or privatization) of religion was a natural development in modern states founded on the separation of church and state, which entails a guarantee of freedom of religious belief. In this context, the secular state required a substitute for religion in order to unite the people of a polity in a way that transcended their pre-modern positions in the feudal hierarchy. Historically, language took on such a role. In a well-known argument, Benedict Anderson describes the contribution language made to the creation of the nation-state, or “imagined political community” (Anderson 1991, p. 6). Vernaculars replaced Latin as administrative languages (“state” language used by and for officialdom) in early modernity, and subsequently became, through the development of capitalism and communication technology, national print-languages that nurtured a sense of unity within their territories (Anderson 1991, chap. 3). “[T]he most important thing about language is its capacity for generating imagined communities, building in effect particular solidarities” (Anderson 1991, p. 133, emphasis original).

It seems no coincidence that Peter Berger and Thomas Luckmann, phenomenological sociologists in the tradition of Weber’s interpretative sociology (verstehende Soziologie), relied on both sociology of religion and sociology of language in their 1966 The Social Construction of Reality. These two research fields served as a pair of wheels for their sociology of knowledge (Berger and Luckmann [1966] 1989, p. 185; see also Tada 2015b). Berger and Luckmann were of course well aware of the significance of religion for human life, but never ignored the empirical fact that religion was fading as people’s common basis for recognizing the world and acting within it. Nevertheless, they did not abandon the idea of a macro foundation common to people in a society and enabling their mutual understanding: They were confident that modern people share natural (or ordinary) language, with its origin in daily life. This idea appears to have been the point at which postwar sociological theory broke away from Parsons and took a linguistic turn, resonating with contemporary philosophical trends (Tada 2015b). Jürgen Habermas’s theory of communicative action would be the zenith of this language-first sociological theory, which postulates language as primal in the public sphere (see Habermas 1981, chap. 3, 6; Habermas 1984; Habermas 1985, pp. 438–445). It is true that Habermas’s academic interests veered in a more political-philosophical direction after the intensive development of the theory in the 1970s and in the 1980s, but the theoretical frame based on language was still maintained. For instance, he makes the following assertion in his hefty book of 1992 Faktizität und Geltung (Factuality and Validity).

The communicative [reason] is distinguished from the practical reason by the fact that it is neither longer ascribed to an individual actor nor to a macro-subject like state or society. What enables communicative reason is rather the linguistic medium, by which interactions get networked and life-forms structured. This rationality is written into the linguistic telos of communicating (Verständigung) and builds an ensemble of conditions which simultaneously enable and restrict [communicating]. Anyone who uses a natural language for communicating with an addressee about something in the world finds her/himself forced to adopt a performative attitude and to get involved in especial presuppositions. (Habermas [1992] 1998, pp. 18-19, emphasis added)

In this context of discussion, Habermas clearly assumes that “we—who find ourselves in our linguistically structured life-forms” (Habermas [1992] 1998, p. 11) exist previously as a basis for communicating; however, to the present day, such a premise of “us” has apparently not been examined much in sociological theory. In this sense, it seems to have been a theoretical axiom (or an orthodoxy) to think that people shared a language in daily life.

But, what range of people share what language? The important thing is that it is arguable whether people’s sharing of a concrete language as it really exists is truly natural—on the contrary, linguistic homogeneity in modernity turns out to be constructed more or less artificially. As Klaus Schubert says, “In ‘natural’ languages […] there is a good deal of artificiality” (Schubert 1989, p. 9, emphasis in original).⁴ For the establishment of a far-reaching linguistic homogeneity, a particular language must be chosen or selectively cut out from a dialect continuum, standardized by authorities, diffused through the education system and mass media, and finally accepted by the people. In this sense, the construction of linguistic unity has historically been an intrinsic part of the politics of the modern nation-state. The so-called “first modernity” (Beck 1999, p. 2), in which governments powerfully advanced industrialization, demanded linguistic homogeneity, because, as Ernest Gellner (1983, chaps. 3, 6, 10) suggests, a multilingual situation could have hindered efficient cooperation in social (more precisely, intranational) division of labor. At the same time, increasing literacy rates extended and fixed the unitary political space through the vehicle of common written texts, and the rise of communicative labor tied to industrial-structural change also prompted the mono-lingualization of social life.

The assumption of sociological theory that a people (the people of a national polity) share the same language seems to have resulted from these processes of naturalization of a standardized national language.⁵ Particularly after World War II, nations became more homogeneous on multiple axes, owing to broadly improved living and education standards following continued industrialization and economic growth. In addition, the Cold War brought relative stability to national borders. All of these circumstances allowed people to believe that they were living in distinctly bounded, homogeneous societies and to take for granted the territorial congruity of their political and cultural units. It seems that even sociological theorists confused such national societies with life-worlds, thereby perceiving national languages to be virtually a given—second nature.

Besides those historical conditions, there was also a theoretical reason for interpretative or phenomenological sociologists in particular to accept the idea of a shared language: It allowed their original aim of understanding subjective meaning to be replaced by that of understanding linguistic meaning (see Luckmann 1995 = 2002, p. 209; see also Habermas 1981, pp. 377–410; Habermas 1984, pp. 11–13). That is, they believed that intersubjective, public language preceded individual, subjective thought and regulated it; furthermore, this substitution would have been advantageous to their sociological practice, as linguistic meaning was evident to all and much easier to analyze than subjective meaning.

However, Weber, the father of interpretative sociology, himself never assumed language to be the foundation for understanding. This was not because he failed to recognize its intersubjectivity or publicness as a result of the relatively underdeveloped idea of language in his period. Rather, he deliberately kept his distance from such a linguistic view,⁶ as is clear from his concept of language community (Sprachgemeinschaft).⁷ As detailed below, Weber did not admit linguistic community as an entity prior to individuals, declaring instead that it, like ethnic community, was a mere idea that would “disappear with exact concept-formulation.” For Weber, the common basis of understanding was a rationality universal in human existence, which he called “calculability” (Berechenbarkeit).

With these theoretical differences from the sociologists of the linguistic turn in mind, in the following sections I concretely examine Weber’s view of language and its historical background.

The subjective construction of linguistic community

This section clarifies the theoretical position of linguistic community in Weber’s interpretative sociology. Note first that Weber had noticed the essential function of contingency in semantics, as shown in his famous discussion on Beruf (“calling”). According to Weber ([1905] 1963, pp. 63–64, 80–81 = [1930] 2005, pp. 39, 47), an expression of this kind, with a religious connotation of a “task set by God,” occurs neither in languages of predominantly Catholic peoples (Völker) nor in the period of classical antiquity, whereas it does occur in the vernacular languages of predominantly Protestant nations, such as English and German. However, he flatly denied that, for example, the English or German people were by nature endowed with industriousness as a national characteristic (Volkscharakter) reflected in their languages. Against this reified notion, Weber suggested that the “industriousness” should be ascribed instead to contingency in translation:

It is further shown that what is concerned is not some specific, ethnically conditioned property of the languages in question, e.g., the expression of “German national spirit” [germanischer Volksgeist], and that the word [Beruf] in its meaning of today comes from the Bible translations, to be precise, from the translator’s spirit, not from the spirit of the original. (Weber [1905] 1963, pp. 64-65 = [1930] 2005, pp. 39-40, emphasis original)⁸

Translation by nature depends on the translator’s choice. In Weber’s view, Luther’s translation of the Bible, which could have been otherwise than it appeared, happened to bring Protestant vocational ethics affirmed electively with the spirit of capitalism. Industriousness was a quality acquired through the indeterminacy of translation.

This idea of Weber’s has a further implication. As has been known since at least Johann Gottfried Herder and Johann Gottlieb Fichte, the idea had prevailed in Germany that a mother tongue expresses a Volksgeist and therefore that, for example, the German language circumscribes and integrates the German nation.⁹ Albeit euphemistically, Weber defied such an irrational, organicist metaphysics (and historicist holism) by arguing that there is no innate national character manifested in language.

Indeed, Weber clearly professed that the method of his interpretative sociology was rationalistic (Weber [1921b] 1980b, p. 3 = [1968] 1978, p. 6): “[T]he sciences of understanding [die verstehenden Wissenschaften] treat ascertainable regularities of such psychical processes entirely like laws of physical nature” (Weber [1913] 1988, p. 429 = 2012a, p. 274). Weber’s concept of understanding [Verstehen], in contrast to empathy theory [Einfühlungstheorie], aimed at a rational interpretation of subjectivity to explain human action. In this respect, his interpretative sociology was complementary to statistical (quantitative) social research—not only do statistics verify hypotheses about regularities in causal chains; also, subjective meaning and beliefs causally explains statistical regularities in human affairs (see Weber [1913] 1988, p. 437 = 2012a, p. 279). An example of the latter approach is his famous Protestant Ethics and the Spirit of Capitalism, which begins—though its significance might easily be missed—by pointing out the statistical fact of regular differences in vocations, social stratification, and education between Catholics and Protestants (Weber [1905] 1963, pp. 17–30 = [1930] 2005, pp. 3–13). Weber considered this regularity explicable in terms of the “reverse causal connection” (Weber [1920a] 1963a, p. 12): In contrast to materialistic assumptions, such differences are a consequence of subjective, religious beliefs.

It is important to recall in this regard that Weber supposed that human action should be called social when meaningfully related to others’ behaviors,¹⁰ but also that social action is nevertheless generally oriented toward the subjectively estimated chances of given outcomes for the actor’s own action, as well as expectations of and for others’ particular behaviors (see Weber [1913] 1988, p. 441 = 2012a, pp. 281–282; Weber [1921b] 1980b, p. 1 = [1968] 1978, p. 4). Hence, the prototype of social action was purposively-rational action (zweckrationales Handeln), in which individual actors are convinced that, as far as standing on the usual relevance (context) of meaning (Sinnzusammenhang), they can rationally pre-calculate others’ behaviors with a certain probability (see Weber [1913] 1988, p. 441 = 2012a, pp. 281–282).¹¹ In other words, the more purposively-rational an action becomes, the more regular it becomes and, in a sense, moves toward determinism. For instance, one has no other choice but to learn seriously in order to pass an exam, under purposive-rationally thinking. In this frame, the modern autonomous individual who rationally sets her/his own end and means is easy to understand.

Thus, Weber formed an action theory predicated on rationality, subjectivity, and individuality (as distinct from irrationality, intersubjectivity, and collectivity); he maintained this stance even when discussing language. This becomes clear if we contrast Weber’s linguistic views with, for instance, those of Ferdinand Tönnies, who assumed the superiority of community (Gemeinschaft) to society (Gesellschaft) in people’s mutual understanding. According to this German sociologist, nearly a decade senior to Weber, community was a natural unity based on “essential will” (Wesenwille), and society only a transitory existence produced by “arbitrary will” (Kürwille) through agreement and contract: “Community is the continual and genuine living-together [Zusammenleben], society just a temporary and apparent one. Accordingly, community itself should be understood as a living organism, society as a mechanical aggregate and artifact” (Tönnies [1887] 1922, p. 5). Tönnies defended this notion in terms of language:

The true organ of understanding (Verständnis)¹² […] is language itself […]. Language has—as all know—neither been invented nor, as it were, arranged as a means or tool by which one makes oneself understood. It is itself a living understanding, and simultaneously [that understanding’s] content and form. As is the case with all other conscious expression activities, linguistic expression is the spontaneous outcome of deep feelings and prevailing thoughts, and does not serve as an artificial means—which a natural lack of understanding would underlie—for the purpose of enabling one to make oneself understood. (Tönnies [1887] 1922, p. 20, emphasis original)

Thus, Tönnies virtually identified community as such with linguistic community, believing that a tacitly pre-shared language (a mother tongue) creates an affective bond of community and that a linguistic commonality enables people to make agreements or contracts in an explicit manner (see Tönnies [1887] 1922, pp. 19–23).

In contrast to this organicist idea, Weber firmly held to his rationalist, subjectivist, and individualist standpoint.¹³ In his view, language community is not a pre-existing foundation for something else, but a product of individuals’ speech acts; and therefore pre-sharing of a national trait is not required for the establishment of a language community. Even objective commonalities, such as the same pronunciation, are unnecessary; rather, the important condition for the formation of a linguistic community is that individual actors orient their speech behaviors to the chance of being understood by a certain range of others: This range, in which actors can on average (therefore rationally) expect their speech behaviors to have meaningful, to some degree predictable, effects on others, is consequently circumscribed as a “language community” (see Weber [1913] 1988, pp. 455–456 = 2012a, p. 290). Simply put, a language community is founded on actors’ subjective expectations of and for communicating (Verständigung). It exists only if one can rationally expect others’ comprehension when talking to them.

In this sense, language community is not a presupposition but a result of social actions: After speaking to others and being apparently almost totally unable to make oneself understood, one may conclude that their language is different from one’s own and that there is accordingly no linguistic community between oneself and them (see Tada 2010, p. 188). George Yule (2014, p. 256) concisely defined a speech community as a “group of people who share a set of norms and expectations regarding the use of language.” However, the sharing of verbal norms and expectations can be confirmed only ex-post, in actual communication. As Nobuharu Tanji (1996, chap. 5, sec.) suggests, it is not because speakers belong to the same language community that they can communicate smoothly, but because they (appear to) communicate smoothly that they can be said to belong to the same language community. Hence, shared linguistic norms prior to speech acts are merely a kind of semblance taken for granted under certain conditions.

Weber ([1913] 1988, p. 453 = 2012a, p. 289) gives as another example the use of money, by which users orient their actions toward the expectation that others will gain benefit from receipt of money, an expectation grounded in the other’s interests as well as one’s own. Note that money-using actions are not oriented to pre-enacted or explicit agreement (see Weber [1913] 1988, p. 453 = 2012a, p. 289); the use of money is instead required despite, or rather because of, the lack of a common order such as an agreement or plan regarding how to satisfy demands with goods. (Under the genuine communist regime, money would become unnecessary in theory because of the perfectly planned production and full rationing system).

The market is the ideal-typical complex of such money-using actions (Weber [1913] 1988, p. 453 = 2012a, p. 289). Due to others’ “blackboxness” (intransparency), purposively-rational actors must take rational account of others’ interests to maximize their own self-interest. The market emerges from these kinds of individual actions. Weber clearly states that the market is “the case as a result of meaningful relatedness of the action of the money user” (Weber [1913] 1988, p. 453 = 2012a, p. 289, emphasis original). Thus, it is not a pre-existing entity, simply appearing “‘as if’ it had been achieved through orienting to an order that satisfies all participants’ demands” (Weber [1913] 1988, p. 453 = 2012a, p. 289, emphasis added).

Weber extended the fictiveness of this “as if” to language community. This was a logical consequence of his interpretative sociology, as his starting point remained individuals’ rational actions. By implication expressing a refusal to reify the linguistic community, he states:

A language community is, in the ideal-typical “purposively-rational” limiting case, presented through social action’s numerous particular acts oriented to the expectation that the other person achieves an “understanding” of one’s subjective meaning. This occurs massively [massenhaft] among a multitude of people through a semantically similar use of certain externally similar symbols, and somehow approximately “as if” the speakers oriented their behaviors to purposively arranged grammar rules. (Weber [1913] 1988, p. 45 = 2012a, p. 289, emphasis added)¹⁴

As Stuart Hughes (1958, p. 310) notes, “In the social and cultural world, he [Weber] had found, a fixed reality was undiscoverable.” Thus, Weber regarded even the language community as a semblance emerging from a certain number of speech acts by individuals, rather than an invariable, a priori reality underlying people’s linguistic activities.

Weber in this sense clearly grasped the role of subjectivity in language community construction. Even though speakers may externally appear to use the same language in the same way, they actually, as suggested in the above quotation, use similar language in similar ways, each one with their own idiolect (individual dialect),¹⁵ and thorough observation should always show speakers’ personal differences in vocabulary, pronunciation, and even grammar. People simply, mutually, ignore differences in their language use, since they believe in their membership in the same language community. That is, the range of “same” depends on one’s perspective. In this perspectivist sense, the language community is founded on subjective belief,¹⁶ according to which its boundaries alter. As Pit S. Corder says:

The concept of ‘a language’ is a matter of social psychology. A speech community is made up of people who regard themselves as speaking the same language. […] In other words, a speech community is defined in terms of its beliefs, not its language. (Corder [1973] 1993, p. 53, emphasis original)

With this insight in mind, the language community may be characterized as a type of imagined community: imagined linguistic community.¹⁷ As a reality constructed through subjectivity, the language community both expands and contracts in the imagination and can correspondingly be either inclusive or exclusive in different contexts. That is, it is an ideal type existing in the actual world, changing over time.

Language community as a political product

As stated above, Weber’s interpretative sociology set social action as its immediate object of observation. His radical rationalism prohibited him from assuming any non-empirical collective entity, as reflected by the concept of consensual action (Einverständnishandeln), which is discussed below.

Weber regarded social actions as consensual when the objective probability exists that an actor’s expectations regarding others’ behavior will be treated as valid (appropriate) by these others despite no agreement (see Weber [1913] 1988, p. 456 = 2012a, p. 291). In this respect, his idea of consensus (Einverständnis) means virtually the same as “common sense” in everyday life: Consensus exists if a certain chance (probability) of an action’s having a particular consequence is taken for granted. Notable is that this consensus concept does not imply approval, solidarity, or understanding; it means “simply to ‘submit to’ [Fügung] the accustomed because it is accustomed” (Weber [1913] 1988, p. 471 = 2012a, p. 299, emphasis original). As typical instances of consensus in this sense, Weber offers multiplication tables drilled into a child’s head and a tyrant’s orders imposed on subjects. These are bindingly valid even though their basis or purpose is never understood: “Not by rational contemplating but by practiced (imposed) empirical cross-checking, one confirms whether one has calculated ‘correctly’ in the consented manner” (Weber [1913] 1988, p. 471 = 2012a, p. 299).

This Weberian concept of consensus may equate, in the terms of phenomenological sociology, to what is called the “epoché of the natural attitude.” Weber himself remarks that people are in the main interested only in the expectations relevant to their practices (see Weber [1913] 1988, pp. 471–472 = 2012a, pp. 299–300; see also Weber [1921b] 1980b, pp. 10–11 = [1968] 1978, pp. 21–22). For instance, we use an electric tram or hydraulic lift without knowing the natural-scientific details of their construction (Weber [1913] 1988, p. 471 = 2012a, pp. 299–300), and conventionally obey laws without knowing their precise wording or necessarily their spirit: Once a new legal order acquires consensus and becomes entrenched, the original meaning that the lawgivers had purposively-rationally thought is forgotten or altered (see Weber [1913] 1988, pp. 471–472 = 2012a, pp. 299–300). As the complexity and differentiation of social life increases, it becomes more universal to bracket things outside of everyday practices (see Weber [1913] 1988, p. 472 = 2012a, p. 300). Hence, as phenomenological sociologists presume intersubjectivity to be a life-worldly (cultural) given, everyday consensus may be seen as the basis for understanding other people.

However, Weber himself never accepted the idea of something collectively pre-shared for understanding. He warned not to identify concepts like social action or consensus with the image of “living with and for one another” (Mit- und Füreinander) and insisted that consensus does not immediately mean “exclusiveness” against foreign others (Weber [1913] 1988, pp. 463–464 = 2012a, pp. 294–295). In his opinion, even a genuinely intercultural battle like the Mongol invasion of Europe contained fragments of a consensual communitization (Einverständnis-Vergemeinschaftung): Battle is a social order that foes, respectively thinking of some meaning of their own actions, communally form together (see Weber [1913] 1988, pp. 463–464 = 2012a, p. 295). As a corollary of rationalism, subjectivism, and individualism, Weber thought that every community is what individuals achieve through communitizing social actions.

Even if it is a matter of so-called “social objects” [soziale Gebilde] like “state,” “church,” “cooperative,” “conjugality,” and so on, social relation exists exclusively and merely in the chance that an action which, according to the content of meaning, connects with others in some definite way took place, takes place, or will take place. This is always to hold in order to avoid a “substantial” interpretation of those concepts. Sociologically, a “state,” for instance, cease to exist as soon as the chance that certain kinds of meaningfully oriented social action occurs has disappeared. (Weber [1921b] 1980b, p. 13 = [1968] 1978, p. 27, emphasis original)

Weber applied this view to language community as well.

Nonetheless, the linguistic community seems to have no correspondent in Weber’s typology of communities: It is normally neither a rational purposive association (Zweckverein); nor is it an official institution (Anstalt), such as the state (a kind of political community) or church (a kind of religious community), predicated on a binding power structure or rational statutes; nor is it in itself an associative grouping (Verband) with a consensual power order, such as a household community (Hausgemeinschaft) under a master of the house (see Weber [1913] 1988, p. 466–467 = 2012a, pp. 296–297). For these reasons, language community might appear to have grown naturally and existed as a transsubjective, fundamental entity. However, as with money use, language must be grasped in terms of individuals’ purposively-rational, materialistic, and strategic actions, as primarily an instrument for satisfying personal interests.¹⁸

The commonality of language, which is created through the same kind of traditions from the family and neighboring environment, makes easier in the highest degree mutual behavior, that is, the foundation of all social relations. However, it itself does not yet mean the communitization [Vergemeinschaftung, “community-building”], but only makes easier the intercourse [Verkehr] inside the group in question, namely, to make easier the genesis of societizations [Vergesellschaftungen, “society-building”], first among individuals, not in their property as linguistic comrades [Sprachgenossen], but as other kinds of interested people. Orientation toward the rules of common language is primarily, therefore, only a means of communicating [Verständigung], not the semantic content of social relations (Weber [1921b] 1980b, pp. 22-23 = [1968] 1978, pp. 42–43, emphasis original).

The above idea can be illustrated by the status of English as the lingua franca in the present era of globalization. English certainly helps speakers of diverse mother tongues to communicate, but in this regard serves mostly (e.g., in international relations) as a relatively neutral tool to achieve speakers’ own (generally political or economic) aims. Such purposively-rational use of English hardly amounts to an English-language community of comrades, and non-native English-speakers participating in transitory intercourse in business English retain a sense of belonging to their own language communities.

As argued in the previous section, language community is subjectively demarcated. Let us examine further the question of the conditions required for such demarcation. The answer must lie elsewhere than in language itself, because using the objectively same(−seeming) language does not always build language community. On this point, Weber has made a suggestive remark: The subjective construction of community originally has a political nature. He states, “Only the genesis of conscious oppositions against the third parties can create the same kind of situation for the participants in the linguistic commonality, community feeling, and [enduring type of] societizations whose conscious ground of existence is the common language” (Weber [1921b] 1980b, p. 23 = [1968] 1978, p. 43, emphasis added). Simply put, the most crucial thing for community formation is awareness of the “us” (inside) versus. “them” (outside) distinction. Any kind of community comes from such a (in Rogers Brubaker’s terms) “politics of belonging” (Brubaker 2015, pp. 132–135).

Language, in this conception, is merely chosen as an objective(−appearing) basis to justify the distinction between “us” and “them.” From Weber’s viewpoint (see Weber [1912a] 1988a, p. 484; Weber [1921/22] 1980, p. 515), the “we-consciousness” of a language community is based not on the sharing of language itself, but on political destiny in a broad sense, examples being the way the German-Alsatians had a sense of community not in general with the Germans, speakers of the same language, but with the French, with whom they shared political memories of the French Revolution; Polish people in Upper Silesia (Oberschlesien) having formerly had no Polish national feeling, rather regarding themselves as Prussian; and Baltic Germans eschewing unity with the German Empire (Weber [1921a] 1980a, pp. 242–244 = [1968] 1978, pp. 395–398; Weber [1921/22] 1980, pp. 528–529). Weber explains these instances in terms of national feeling (Nationalgefühl), insisting, “[O]ne must not grasp ‘nation’ as ‘cultural community’” (Weber [1912a] 1988a, p. 485; see also Konno 2007, pp. 334–335). However, these cases also make it clear that an objective element like a common mother tongue is not essential for a sense of belonging together.¹⁹

Insofar as language serves to define “us” versus “them” subjectively, it becomes politically significant. To demarcate insiders, linguistic differences between them and outsiders are often arbitrarily exaggerated or even newly constructed. Since language boundaries themselves, as mentioned above, are not always distinctive, whether two linguistic variants are regarded as the same language’s different dialects or themselves different languages depends on speakers’ own sense of identification (see Coulmas 1985, p. 32). If a political community declares an independent state, its linguistic variant could also declare an independent national language. As linguistic boundaries can vary in such a way, the discussion in the previous section might have to be slightly amended: The range of intelligible communication does not directly reflect the range of language community. For instance, the two major Albanian dialects, Gheg and Tosk, are not perfectly mutually intelligible, but speakers perceive themselves as speaking a common language; in contrast, although Danish, Norwegian, and Swedish are mostly mutually comprehensible, they are perceived as different languages (Billig [1995] 2014, p. 32; Ruhlen [1987] 1991, p. 277). In this regard, Michael Billig states:

[T]he boundaries between languages, and the classification of dialects, have followed the politics of state-making. Where national boundaries are established, then, the differences in speech patterns [on] either side of the boundary are more likely to be seen as belonging to distinctly different languages by the speakers themselves, their national centres and the world in general (Billig [1995] 2014, p. 33).

Thus, people may belong to the same nation, but not because they speak the same language. Rather, after political (mostly national) borders are demarcated, “languages themselves have to be ‘imagined’ as distinct entities” (Billig 1995, p. 36)²⁰—people must speak the same language because they belong to the same nation.²¹ Such nationalist ideology seems to be a major reason why many nation-states feature linguistic unity within their borders: Coterminousness of national territory and linguistic community is artificially produced by political opposition to the outside. And this “artificially” has two meanings: First, linguistic boundaries are subjectively established against the outside in the form of the belief in a shared language; second, national territories are actually homogenized by language planning or even by linguistic, if not ethnic, cleansing. Since language community is imagined and created in this double artificiality, it must be regarded as a selective achievement of human action—at least partly ideological, or value-rational action, which appears, according to Weber ([1921b] 1980b, p. 14 = [1968] 1978, p. 26), to be irrational from the purposively-rational standpoint. In nationalist ideology, linguistic homogeneity is itself an eigenvalue to be fulfilled, and in this regard, Alastair Pennycook’s following statement seems to get at the heart of the matter: “Language is located in social action and anything we might want to call a language is not a pre-given system but a will to community” (Pennycook 1994, p. 29, emphasis original).

The theoretical orthodoxy that language and its community form a life-worldly given structure for social action thus becomes too naïve to maintain. It is true that Weber also conceived of language community as a kind of structure into which “one is normally born and educated” (Weber [1913] 1988, p. 466 = 2012a, p. 296). However, if linguistic community is assumed to be a natural product, it fails to reflect reality. The language communities into which today’s people are born and educated are/were often designed by nation-states.

The trend by which sociological theorists take for granted the sharing of language seems itself to derive from the postwar epoché. As stated earlier, the nation-state gained relative stability under economic and educational development and additionally the Cold War. Therefore, people, including sociologists, ceased to think back on their national identity and instead continued to believe that they lived in a national society. National languages also became a given to the majority, as the “mother tongue” within each nation. As Michael Billig ([1995] 2014) indicates, nationalism came to be supported by banal practices, not by conscious choices. The absence of the nation-state is far more unimaginable today (Billig [1995] 2014, p. 77, 95)²²; it has been naturalized, its artificiality forgotten.²³

In contrast, late nineteenth- and early twentieth-century Central Europe, where Weber lived, remained in a turbulent state of nationalism. Under such circumstances, this rational sociologist believed that even the boundaries of ethnic communities were not natural—that contrary to the common conception that ethnic communities were objectively defined in terms of race, common ancestor worship, religion, and language, they were actually rooted in people’s collective imagination (see Yamazaki 1999, pp. 11–13, 18).²⁴ He argued explicitly that ethnicity is artificially (künstlich) classified on the basis of subjective belief, and that the political community, above all, awakens the belief in ethnic commonality (see Weber [1921a] 1980a, p. 237 = [1968] 1978, p. 389; see also Isajiw 1974, p. 116). In summary, ethnic groups are formed not through features objectively recognizable as common to the people making them up, but through people’s self-recognition of themselves as a “political remembrance-community” (politische Erinnerungsgemeinschaft) (Weber [1921a] 1980a, p. 238 = [1968] 1978, p. 390; Weber [1921/22] 1980, p. 515). War is a typical case: A sense of “us” as a common unitary ethnicity can originate even when people are caught up, as a third party, in warfare between foreign powers (see Smith 1991, p. 27). As shown above, “common political destinies” (Weber [1921/22] 1980, p. 515; see also Weber [1905] 1963, p. 81 = [1930] 2005, p. 47; Weber [1912a] 1988a, p. 484) accentuate ethnic differences to others—and some of these differences become the core of modern national consciousness.²⁵

It is often said that the national consciousness of Germany, Weber’s homeland, was formed by appealing to language: Those who share the German language are the German nation. Herder, Fichte, and later Karl Wilhelm von Humboldt were major representatives of this idea. However, as mentioned above, many native German-speakers did not identify with the German nation. As James Crawford also points out, a notion that a language determines a natural political unit was a “historical phantasy” (Crawford 1992, p. 239). Hence, Weber kept a distance from the Germanic definition of the nation, and at least in later years, “refused to define nation, with Herder, as a language community” (Mommsen 1974, p. 54), contrary to his view in early life.²⁶

Accordingly, the Weberian point of view also refuses to define a language community by a language. Using the objectively same(−seeming) language does not necessarily lead to linguistic communitization. Despite admitting common language to be the most direct means of generating affinity, a sense of community, and ultimately national feeling, readily facilitating the sharing of cultural heritage and meaningful intelligibility among people (see Weber [1921a] 1980a, p. 238 = [1968] 1978, p. 390; Weber [1921/22] 1980, p. 528), Weber never regarded language community as an ontological entity with defined, invariable shape:

A concrete language community […] has boundaries somewhere (mostly fluid) at each time. That is to say: Normally, not all human beings who exist in general, but only a certain number of human beings, whose boundary is often quite unclearly determinable, can be taken into consideration in the “expectations” as an—actual and potential—participant in consensus at each time. (Weber [1913] 1988, p. 463 = 2012a, pp. 294-295, emphasis added)

Thus, a language or its community is demarcated depending on people’s subjective expectations. As Danny Law notes, “Linguistic difference matters, but that difference is not just an independent objective feature of the world” (Law 2014, p. 168). A sense of speaking the same language or of belonging to the same linguistic community is none other than the “ideological construction of sameness” (Law 2014, p. 164), which can be enforced afterward by (or in spite of) linguistic facts.

Toward the re-turn from lingualism

As discussed above, a common language does not directly imply the creation of linguistic community, but simply facilitates communication, which in this context is to say societization. Even if people can communicate smoothly with each other, they do not always feel themselves to belong to the same linguistic community.²⁷ Furthermore, sharing the same language is not a necessary condition for communication, because unsuccessful communication due to linguistic difference is also communication. In this sense, the speaker can know only after communication what language the other uses.

Here again, Weber avoided reifying language community by virtue of his confidence in rationalism, subjectivism, and individualism. Indeed, assuming that language community pre-exists as an invariant entity is metaphysical and scientifically insupportable; instead, it is the dynamics of linguistic communitization that should become the object of sociological investigation. For example, as Weber ([1913] 1988, pp. 462–463 = 2012a, pp. 294–295) suggests, the extent to which social actions using a particular language are open or closed to other people can vary from situation to situation:

[T]he participants in a language community have normally, for instance, no interest in excluding third persons from the consensus […]; and market-interested persons are also often interested exactly in an “expansion” of the market. Nonetheless, both a language (as sacred, hierarchical, or a secret one) and a market can be monopolistically “closed” through consensus and societization. And on the other hand, although the participation in the specific social action of concrete political power organizations is normally closed through societization, it also can be kept widely open (for “immigrants”) precisely owing to the interest of power (Weber [1913] 1988, p. 463 = 2012a, p. 295; see also Weber [1921b] 1980b, p.24 = [1968] 1978, pp. 44-45).

Indeed, there are a large number of inclusion and exclusion issues that concern language, such as linguistic assimilation, linguistic imperialism, multilingualism, and “language wars” (battles or competitions for linguistic hegemony).²⁸ Sociology, as the study of society, must be capable of dealing with these matters, and therefore cannot naively postulate that language and its community are a natural, stable, and perhaps peaceful foundation for social actions. Language and its community are rather an object for sociology to research, and, for this purpose, the nation-state seems to be one of the most useful analytical lenses. That is, as stated above, the language community does not necessarily coincide with the state, an officially institutionalized political community; nevertheless, the modern nation-state typically tries to match itself to the language community. This “nationalist equation of one language = one nation = one state” (Auer 2005b, p. 8) is at the root of many sociolinguistic issues, and its obstinacy has been illustrated relatively recently in certain ex-Communist countries in Europe, sometimes peacefully, sometimes not.

According to Anthony Smith (1979, p. 58), from the eighteenth century to the present day language has remained one of the prime objects of efforts at purification in the service of nationalism. The phenomenon of linguistic affiliation’s becoming central to one’s social identity is very often a result of nationalist ideology, and is not universal in human history (see Law 2014, pp. 171–173). With modernization, the basic principle of state construction shifted from religion to language (see Tada 2015b). This change had not only an ideological but also a practical reason: For the sake of industrialization and militarization, modern states needed to make it easy for people to communicate with each other, and homogenized their working language for this reason. Even an ideal “public sphere” (Öffentlichkeit) would not have been immune to such linguistic homogenization. Usually, the term “rationalization” would remind us foremost of the disenchantment process from religion to science. However, parallel to that kind of rationalization, which secularized people’s conception of the world, linguistic rationalization also occurred through the standardization of language and the spread of schooling, as the modern state needed a “disciplined” common language to achieve “national mobilization” efficiently. Urbanization, too, encouraged people of different origins to use a common language as they mixed.

With linguistic diversity thus reduced, daily life became monolingualized, and the first modernity of language was established. This then became the background against which sociological theorists shifted their axis, beginning in the 1960s, from the Parsonian to the phenomenological school, that is, from the theory that looked for the root of social integration in the religious idea to one that emphasized the vital role of public language in understanding others’ meaning and in the constitution of the social order.

Unfortunately, this “linguistic turn” in sociological theory appears to have been out of step with actual changes in the world. In those days, the second modernity was already dawning, whereby the uniformity-orientation of the first modernity was reflexively revised. Such a theoretical shift in sociology would have been ideologically sympathetic to the second-modern movements, for example, ethnic, racial, and anticolonial movements,²⁹ but it relied practically on methodological nationalism or the image of the national society. The linguistic turn in sociological theory, while rejecting religion as a remnant of pre-modern state identity, still presumed a shared language, one of the core elements of the modern nation-state identity, to be the intersubjective foundation of society.

Needless to say, the relationship of language to the modern nation-state as discussed thus far in this article is but an ideal type, with many variants and exceptions. Furthermore, this article lacks the scope to go into further detail on the history of sociology since World War II and the place of language therein. However, it is already evident that lingualism—a linguistic reductionism that optimistically conceives of language, in its sharedness, as the solution to problems of meaningful understanding and the social order—is insupportable.³⁰ In fact, language itself causes a range of social problems, particularly coupled with the nation-state frame. For instance, minority language groups may be easily relegated to a lower social position if they have less than full command of a common national language. Such structural discrimination occasionally spurs oppressed people into mini-nationalisms, leading them to seek self-government as a unique linguistic community or even independence as a country with its own national language.

As Stephen Barbour ([2000] 2007b, p. 17) argues, a national language is one of the major components of nationalism. He says:

While the linguistically homogeneous state is extremely rare, and while a high proportion of languages are actually not sharply distinct from others, the demand for the linguistically homogeneous nation and the clearly distinct national language has become a standard part of nationalist ideology (it goes without saying that such ideology demands that nation and state be coterminous; in fact it regularly merges these two concepts) (Barbour [2000] 2007b, p. 14).

As suggested in this cited passage, it is incorrect to assume that a homogeneous linguistic community underlies the nation-state. On this issue, Karl Deutsch made a suggestive point as early as the 1940s, referring to a series of The Statesman’s Year-Book: “The national languages of today appear not only as a cause, but as a result, of national differentiation” (Deutsch [1942] 1968, p. 605). According to him, the number of “languages” in Europe increased proportionally from the nineteenth century to the beginning of the twentieth century with the growth in the number of states, as nationalist ideology imagined languages one after another during that period. Concretely speaking, the number of languages in Europe rapidly grew from 16 in 1800 to 30 in 1900, and to 53 in 1937, while the number of sovereign states in Europe increased from 15 in 1871, to 21 in 1914, and to 29 in 1937 (Deutsch [1942] 1968, pp. 599–600, 606). “So far as the language factor is concerned, the bulk of the evidence shows for the years from 1800 to 1941 a steady increase in the diversity and strength of nationalistic feeling” (Deutsch [1942] 1968, p. 600).

Central Europe was the major notable battleground for such linguistic nationalisms because in this land of “belated nations,” nationalism took the form of ethnolinguistic nationalism, and the notion of the national language was already well rooted during the second half of the nineteenth century (see Kamusella 2011; 2001, p. 239; and also Anderson 1991, chaps. 5–6).³¹ In this context, it is also no coincidence that Esperanto was created in the second half of the nineteenth century in Central Europe. Ludoviko L. Zamenhof, the father of this “international” neutral language, intended to reconcile opposing ethnolinguistic groups with its widespread use, and says that his devising of Esperanto—this name means “one who hopes”—was motivated by his childhood experience of ethnolinguistic hostilities among Russians, Poles, Germans, and Jews in his hometown, Białystok, in Poland, under the Russian Empire (Zamenhof 1929, pp. 417–418).³² However, the vast multiethnic empires of Central Europe (the German, Austrian-Hungarian, and Russian Empires) were eventually dismantled due to World War I into smaller nation-states based on ethnolinguistic identity. The expansion of linguistic nationalism in Central Europe was thus, for Weber, a decidedly contemporary phenomenon, the experience of which led him to perceive language as an ideological apparatus for nation- and nation-state-building. He writes as follows:

Today, in the age of language wars [Sprachenkämpfe], “language community” is particularly valid as [national commonality’s] normal basis. […] In fact, “nation-state” became nowadays conceptually identical with “state” based on language uniformity. Great numbers of political associative groupings actually have such modern character on a “national” basis in this linguistic meaning. Aside from them, there are also political associative groupings that contain several language communities. […But they] usually, not always, give preference to one language for the political intercourse (Weber [1921a] 1980a, p. 242 = [1968] 1978, p. 395).³³

Weber was clearly aware that the linkage between the name of nation (natio in Latin) and a national language emerged in modernity (Weber [1912a] 1988a, p. 486), and, as the above excerpt suggests, he did not overlook the relationship of politics to language. This seems to be a corollary of Weber’s consistent viewing of the nation-state not as something metaphysical but as the “worldly power-organization of the nation” (Weber [1895] 1988, p. 14 = 1994, p. 17). Symbolically, he took language as the first instance or opportunity to clarify the phenomenon of domination (Herrschaft) in human affairs (Weber [1922] 1980, p. 541 = [1968] 1978, p. 941). According to him, the elevation of a given dialect as the language of officialdom (Kanzleisprache) had frequently, as in Germany, had a decisive impact on the development of extensively integrated literary-language-communities; conversely, as in the case of the Netherlands and Dutch vis-à-vis German(y), political separation often establishes the final differentiation of languages; furthermore, controlled dominance over the schools fixes the kind and superiority of the official school language.

These examples show that language is closely but unnoticeably associated with a state’s governing systems, like administration and education. This further implies that language is constantly exposed to artificial selection by these systems of power.³⁴ Deutsch states that “[t]he acceptance of a common national language contains an element of choice” (Deutsch [1942] 1968, p. 603). If this were the case, it would be less and less sensible to presuppose language as a natural, certain foundation for understanding meaning. Linguistic homogeneity, at least in modernity, is rather a social construct that must itself be investigated by sociology.

Perspectives: sociology of language in the age of the world society

As is well known, Martin Luther rejected the Latin linguistic dominance of the church in the sixteenth century, translating the Bible into German.³⁵ Historically, this incident represented a turn from sacred language to the secular vernacular, out of which emerged national language.³⁶ Today’s linguistic nationalism may, however, be rooted more immediately in revolutionary France. French nationalism was an amalgam of civic and ethnic components (Smith 1991, p. 13), of the universal and the particular (see Billig [1995] 2014, pp. 24–25)—the revolutionists insisted on standard French throughout the land, at least partly in pursuit of a mission of civilization backed by self-perceived French cultural supremacy (see Calvet [1974] 2002, chap. 7). In this ethnocentric sense, even French nationalism seems to have included ethnolinguistic elements, although it is usually characterized as a purely civic nationalism.

In any case, language was territorialized by the nation-state (Auer 2005a, p. 11). Monolingual environments within borders are more or less a political product. On the other hand, even dialectal communities in provincial areas might remain, in a broad sense, a political construct based on inhabitants’ counter-identity in relation to a capital city or central government using the standard language. But as long as individuals always speak their own idiolects, there is in theory no community featuring a pure linguistic homogeneity. Owing to individual variation in vocabulary, pronunciation, and grammar, the idea of a group of people sharing exactly the same language through socialization is unsustainable; but in the real world, a sense of belonging to a particular linguistic community hides such idiolectal differences.

An individual’s linguistic identity is formed on the basis of political grounds, not linguistic ones, because, as Law indicates, “[O]ur linguistic choices say something to others about us” (Law 2014, p. 169). On this point, the rise of linguistic nationalism in modernity seems to represent a shift “from Gesellschaft to Gemeinschaft” (Wallerstein [1986] 1991, p. 75).³⁷ A small-scale community does not grow to a large-scale society; nor is a community ontologically real while society is unreal. Rather, communities like that of the nation-state were sought after the emergence of the world society, because people needed novel anchorages for their survival and psychological security (see also Bolz 2001, p. 39).

As Weber asserted, the state is an officially institutionalized community.³⁸ Since it can never be a society, a spontaneously emerging order, methodological-nationalistic identification of state with society is clearly a conceptual confusion. However, this confusion arose for a reason: Nationalism itself became a secular religion that gave social cohesion and identity to the growing “middle-class masses” relieved from traditional bonds in the postwar era (see Hobsbawm [1983] 2012, pp. 263–268, 303).³⁹ Although Weber mentioned the church as another example of an officially institutionalized community, religious community was destined to decline with modernization. Indeed, Weber himself said that the Puritan spirit had already in his time escaped from the “iron cage” [stahlhartes Gehäuse; steel-hard shell] (Weber [1905] 1963, pp. 203–204 = [1930] 2005, pp. 123–124). Contrary to Parsons’s supposition, religion was ceasing to be the shared norm upon which action was based, and was becoming instead a consumer good left to individual choice. Under these circumstances, combined with the relative stability of national borders under the Cold War, the nation-state offered itself as a major resource to supply social cohesion and identity. We may perhaps dare to summarize this development in Parsonian terms, but with a key substitution: Nationalism replaces religion as the irrational element to foster social integration within borders; it itself becomes the civil religion of the nation.

In many developed countries, the frame of the nation-state is so self-evident that people tend to conceive of their living in a definitely bounded national society; and territorialized language appears to be the biggest contributor to this naturalization of the nation-state, as it usually feels like an inherent, shared aspect of the nature of the national society. This is why sociological theorists came to view language, not religion, as the common determinant of people’s world-view, taking national language for natural language. They failed to question the idea of the pre-existence of the language community.

Weber’s individualist view of language may fit today’s reality far better. As noted above, he did not reify the linguistic community at least partly because of his contemporary context. Likewise, in our present days of globalization, as more and more people move across borders, nation-states are becoming less and less monolingual. Furthermore, the sub-politics of linguistic minorities seeking official recognition of their language rights can no longer be ignored. If the government compels the whole population to learn a particular language, it can lead to language wars. We are living in the second modernity of language.

On the other hand, the linguistic unification of the world society is also progressing. English is rapidly gaining influence in non-English-speaking countries. To borrow Anthony Giddens’s (1991, p. 18) terms, English has become a kind of “symbolic token” that enables “disembedding” beyond the local frame of a national society. In fact, like Latin or French in earlier times, the world-wide expansion of English may even be giving rise to a so-called New Medievalism: a stratified society crossing national borders. Those in this upper(−middle) layer can construct borderless human networks through their use of the world language, perhaps finding greater familiarity and solidarity with foreigners equal in intellectual and cultural cultivation and similar in taste than they find with others of their own nationality (see also Gellner 1983, p. 113)—English-based cosmopolitanism. Yet, those in the lower layer might also cross borders with relative ease by taking jobs abroad that do not require high levels of English proficiency. Consequently, (lower-)middle-class people seem most likely to be left behind in their home countries, specialized for the domestic labor market through national education and language, which previously functioned as a barrier to immigrants or guest workers entering the white-collar labor market of a host country, but nowadays hinder middle-class members from entering the global labor market. The middle tier in English-speaking countries may also feel themselves facing stiff challenges from overseas students and cheap foreign workers with adequate English proficiency. The anxiety of this declining class regarding their everyday life and livelihood has been widely observed to be a cause of the recent spread of nationalism and “populism” in developed countries, which may seem much like the dying scream of a disintegrating national society: In the era of globalization, the (lower-)middle-class could fall into the lowest class because of their low mobility.

Class issues derived from language are, of course, not the sole factor in the current unsettled global situation. However, some people undoubtedly feel pressured by English linguistic imperialism,⁴⁰ concerned that their mother tongues might be devalued to mere minority or local languages in the global context, losing their prestige as national languages even inside national borders, which would mean a loss of their social status as well. Norbert Bolz says, “In the Global Village of the internet, mother tongues [languages other than English] appear as a blockade on world communication” (Bolz 2001, p. 55). It goes without saying that this is not limited to cyberspace.

Thus, language, like religion, has increasingly become an individual choice—a kind of individualization (or privatization) of language. For the sake of social and economic interests, an enormous number of people personally choose English foreign language education for themselves or their children. Indeed, the enthusiasm for learning English is a societal movement that has evolved from the bottom up (Wright 2000, p. 250), because, as Stephen Ryan states, “Language learning is primarily concerned with how people regulate their future or future oriented actions” (Ryan 2006, p. 38). On the other hand, as Ryan argues in detail, learning English can be also correlated with a new self-identity. According to him, young people who eagerly learn English today are apparently motivated at least partly by the desire to obtain recognition of and legitimacy for their membership in an imagined global (language) community, rather than purposively-rationally by some direct socioeconomic benefit. Aside from an individual’s local identity, an ideal, transnational, global self-identity can be constructed in part by learning English and using it as an objective means to present such a self in certain domains, decoupling English from particular Anglo-American countries or contexts, whereas eager learners of English in days past might have held, for example, a cultural admiration for the United States, and value-rationally targeted the country as a community to access through learning English. In this global era, the change from such a nation-state-based world picture also occurs in terms of English and English learning itself. The target community with which the current younger generation of English learners identifies in the imagination can be de-nationalized, although this may in turn itself be part of a “new medievalization” process.

In parallel to such a de-nationalizing tendency among individual learners, a paradoxical phenomenon has also been emerging from the frame of the nation-state: Some governments prefer to instill English into their own people for the sake of their survival and prosperity in global competition. This nationalistic Anglicization from above may bring leftists or liberals around to ethnolinguistically advocate for their national languages. As Anderson (1991, p. 134, 148) suggests, anyone can in principle learn any language, although temporal limitations can prevent its full acquisition. However, mandatory English education at the cost of one’s mother tongue would inevitably entail emotional conflict, just as the ethnic community or nation is “one of the concepts which vex us with emotional sensations the most” (Weber [1921a] 1980a, p. 242 = [1968] 1978, p. 395). To begin with, if such an Anglicization from above succeeded well, it would be highly likely that the language policy would produce the contrary effect—a brain drain from the country.

In either case, the current globalized era no longer allows sociological theories to assume naively the collective sharing of the same language on the presupposition of a homogeneous “national society.” Since we live in a heterogeneous global society, sociology cannot securely postulate linguistic intersubjectivity for mutual understanding. Such a methodological-nationalistic axiom is itself a social product of the first modernity. Today, language must be dealt with as a sociological research object; and to study various social issues and phenomena regarding language, Weber’s rationalist, subjectivist, and individualist standpoint should be revaluated, as it enables us to look at languages and their communities in terms of their dynamics and to recognize them as social constructs.