Diversity, linguistics and domination: how linguistic theory can feed a kind of politics most linguists would oppose

Nick Riemer
University of Sydney & HTL, Université Paris-Diderot

Riemer illustration 1

Antonio Gramsci, a co-founder of the Italian Communist Party and one of the twentieth century’s most prominent intellectuals. Gramsci studied linguistics and wrote about linguistic topics throughout his life. ‘Study, because we’ll need all your intelligence’.

What connections might linguists’ professional activities have to politics? Most recently, the question has been posed by the collective self-dismissal of the Lingua board and the journal’s metamorphosis into the open-access Glossa – a welcome attempt to break the monopoly of profiteering multinationals over the dissemination of research. Initiatives like Glossa or Language Science Press are much-needed, and all too rare, instances of scholarly activism against the widespread ‘enclosure’ of knowledge characteristic of our age (Riemer forthcoming). As such, they are compatible with the ‘vague form of liberal progressiveness’ that Hutton (2001: 295) has identified as the ethos of contemporary linguistics. But how might other aspects of linguistics as an institution fit, or not, into this frame? What can we say about how linguistics might relate to characteristic progressive priorities like support for diversity, opposition to discrimination and domination, commitment to democracy, and to the overall political contexts in which efforts to advance those priorities are situated?

There’s been little shortage of critical discussion of linguistics’ ideological and political valencies, though it has often come from sources other than linguists themselves.[1] Linguists have, in fact, on the whole been strikingly reluctant to direct against their own discipline the kinds of critique that swept over the rest of the humanities in the final third of the last century. Linguistics’ scientistic pretensions act as a strong brake on any attempt even to think in critical terms about the epistemic status of the discipline’s results, let alone to explore the field’s wider political effects or determinants.[2]

Reflection on both, however, is important, in the interests of disciplinary self-awareness at least. Not just that, though: linguists who identify with the ‘vague form of liberal progressiveness’ mentioned by Hutton, or whose political sympathies lie further to the left, have an interest in thinking not just about how social and political factors influence linguistics, but also about how what they do as linguists might feed back into the societies to which they belong. Like other academic corporations, linguists probably mostly have a strong sense of their own distinctness. But we are nevertheless a part of the body politic, and our professional activities influence it in various ways.

Formatting human nature for the market

Ideological critique of linguistics, especially of ‘core’ domains like grammar, phonology, semantics and pragmatics, has often focussed on what we might call the discipline’s overwhelming individualistic rational universalism. This emerges clearly in the assumptions that students are encouraged to assimilate when they are instructed in the discipline’s basic concepts and procedures. If we had to detail these assumptions, we might come up with a list like this:

Assumptions about people

  • individualism: as a cognitive or psychological faculty, language is understood to be, at base, an individual phenomenon;
  • rationalism: speakers and hearers are to be understood as essentially rational agents (the emotional dimension of personhood, by contrast, unambiguously playing second fiddle);
  • uniformity: the biological identity of the human species is reflected in the fundamental identity (commensurability) of human languages.

Assumptions about linguistics’ relation to language

  • objectivity: there is a fact of the matter about the structure of language: a unique and unambiguous level of semantic content; a unique representation of syntactic and phonological structure; a unique information structure, and so on;
  • reducibility: the diversity of observed utterances in any domain of linguistic phenomena are realizations of a much more restricted template (grammar, underlying forms, phonological structure, etc.), capturable in a unique analytical metalanguage, where cultural and cognitive diversity bottom out;
  • formalizability: language can be described through formal (or quasi-formal) rule systems;
  • transparency: this formalization is intuitive and shallow, since the theorist’s L1 can be used to express the rules assumed to underlie language without being enriched with an extensive apparatus of technical concepts. For example, the definitions of thematic roles and the protocols for subject-assignment that they participate in make reference to ordinary language terms like ‘move’, ‘action’, ‘place’, or ‘possession’; the definition of lexical aspectual classes are about commonsensical notions like ‘bounded’, ‘instantaneous’ and so on. Wierzbicka and Goddard’s Natural Semantic Metalanguage approach is a particularly obvious instance of this phenomenon. Not all theories are as averse as NSM to technicality – think Minimalism, or phonological theory. But the generalization holds for those domains of linguistics most prominent in standard undergraduate syllabuses – which, as I’ll explain in a moment, is the part that counts. As a wise linguist once commented to me, ‘Linguistics ain’t rocket science’.

Assumptions about the epistemic status of linguistics

  • scientific authority of the discipline: as a result, linguistics is ‘scientific’ and linguists detain an intellectual authority that allows us to say what people are like insofar as their linguistic practices are concerned, without mastering technical competencies of anything like the degree of complexity necessary to the ‘hard sciences’.

Probably none of these principles would be accepted without qualification by all linguists – in particular, any linguists for whom language is an abstract object or system strictly independent of its psychological manifestations will reject the whole first category. Nevertheless, it seems to me the list as a whole fairly captures the essential mindset that the vast majority of students are encouraged to adopt in their early encounter with the discipline. (It’s this early encounter which is most relevant, since most students don’t hang around long enough to be exposed to the inevitable nuancing the ideas undergo later: if we want to explore the ideological effects of linguistics, we need to look at undergraduates, not PhD students.)

Why highlight these assumptions? Because many of them fulfil rather an obvious ideological function: they reinforce a model of personhood – a model of what people are like – particularly compatible with the requirements of contemporary ‘globalized’, capitalist economies. Just like the other ‘human sciences’ (see Riemer (2015) for more), linguistics contributes to one of universities’ most essential roles: ideologically ‘formatting’ students into the atomized, normalized, and rationalistic subjects that best match market norms.

Human nature as idealized by linguistic theory – individualized, intellectualist, rule-following and uniform – embodies the perfect participant in technocratic capitalist economies. If the predictable (i.e. rule-following), rational, conformist individuals presupposed in linguistic models of speaker-hearers really existed, they would be model consumers and employees:

Students who major in linguistics acquire valuable intellectual skills, such as analytical reasoning, critical thinking, argumentation, and clarity of expression. This means making insightful observations, formulating clear, testable hypotheses, generating predictions, making arguments and drawing conclusions, and communicating findings to a wider community. Linguistics majors are therefore well equipped for a variety of graduate-level and professional programs and careers.
(Linguistic Society of America, ‘Why major in Linguistics?’)

Given the attractiveness to graduates, if only for reasons of job-security, of careers in complex organizations (multinational and other businesses, government departments, media organizations, etc.), the similarity between the constrained, rule-based reasoning linguistics students are trained in and Weber’s principles of bureaucracy starts to look like not such an accident:

(1) All official actions are bound by rules with the official subject to strict and systematic control from above.
(2) Each functionary has a limited and defined sphere of competence.
(3) The organization of offices follows a principle of hierarchy with each lower one subordinate to each higher one.
(4) Candidates are selected only from the basis of technical qualification: ‘They are appointed, not elected’.
(5) Officials are salaried and have no right of ownership over their job: ‘The salary scale is graded according to rank in the hierarchy: but in addition to this criterion…the requirements’ of incumbents’ social status may be taken into account.’
(6) The office is the sole, or at least primary, occupation of the incumbent and it constitutes a career: ‘Promotion is dependent on the judgement of superiors’
(summary of Max Weber, Economic and Social Organization, quoted by Blackburn 1967: 177-8)

In advancing a procedural, rule-based approach to the complexity  of language use, Linguistics education normalizes a Taylorist conception of work. Through its psychologism (cognitivism), it suggests that language, a quintessentially social phenomenon, can be best understood as an individual one. This has a clear ideological utility, as Margaret Thatcher (‘there is no such thing as society’) would have understood. As Alex Callinicos observes, ‘it is at least arguable that social stability depends not on the subordinate classes’ belief in the legitimacy of the status quo but on a fragmentation of social consciousness which prevents them from developing a comprehensive perspective on society as a whole’ (1990: 116). By focussing attention inwards onto the linguistic ‘soul’, linguistics does just that. In impressing its ‘assumptions about people’ on students, it encourages them to internalize a model of personhood that fits the market economy like a glove.

Our language, our meanings: the only ones that exist

Title page of Porte Royale

Title page to the 1660 Port Royal Grammar: General and reasoned grammar, containing the foundations of the art of speech, explained in a natural and clear manner; the reasons for what is common to all languages and the principal differences that are encountered; and several new remarks on French.

Linguistics’ assumptions about its own relation to language (the second category in the list) play a different role. Ideas about ‘reducibility’, ‘objectivity’ and ‘transparency’ confine languages within a centralizing and universalizing dynamic, the aim of which is to boil the diversity of a speech community’s linguistic practices down to a single template called a ‘grammar’ – and, frequently, then to claim that all languages can be understood with reference to a unique master-set of categories (Universal Grammar, ‘the basic blueprint that all languages follow’, to adopt the terms of Fromkin et al. 2010: 18). Language structure, including meaning, is presented as a unique, determinate object open to empirical methods aspiring to the (imagined) epistemology of the natural sciences.

It is in my own field, semantics – crucial to understanding other domains of structure – that the effect of these assumptions are clearest. Semantics is predicated on the belief that the linguist’s own L1 – English, French, Chinese – can serve as an adequate representational medium for the meanings of all languages.[3] In other words, if my semantic theory is mentalist, as most are, I can use an only minimally enriched version of my own native tongue to show what you, regardless of what language you speak, have in your head – both what you mean, and the conceptual operations on which this lies. And I can do so ‘scientifically’ – in such a way, that is, that people have to believe me. The linguistic imperialism of a world in which English is everywhere is replicated in the domain of linguistic theory. I am the authority over what is happening in your head.

This brings some uncomfortable consequences: even though ‘exotic’ languages may be configured differently from English (or French, or Chinese), nevertheless, at base, they can all be enclosed within English (or French, or Chinese, or whatever the ‘substratum’ is for the linguist’s theoretical metalanguage). Linguists’ mainly first-world L1s are not, then, languages just like any other, into which other speakers’ L1s can be translated in rough and ready, contextually variable ways for a variety of purposes. They are, instead, hegemonic master-codes in which fixed, context-independent, explanatory representations of exotic meaning can be given once and for all.

It is one thing to affirm, undeniably, the translatability and intercomprehensibility of different languages for a wide range of purposes and interactions. But it is quite another to assume that this reflects an underlying psychological (as opposed to neurological) identity and that a single language – most often English – can provide the explanatory metalanguage in which anyone’s meaning can be theoretically represented. From ‘explain’ and ‘represent’, it is only a small symbolic step to ‘possess’, ‘control’ or ‘dominate’:

Enlightenment stands in the same relationship to things as the dictator to human beings. He knows them to the extent that he can manipulate them. The man of science knows things to the extent that he can make them. Their “in-itself” becomes “for him.” In their transformation the essence of things is revealed as always the same, a substrate of domination. This identity constitutes the unity of nature.
(Horkheimer and Adorno 2002: 6)

It is, for the worldview encouraged by linguistic assumptions, more or less impossible that other people’s concepts might escape the representational capacities of our own linguistic categories. We can theoretically explain everything. Difference is abolished: our intellectual sovereignty knows no frontier. Our language, our meanings, the student learns, are, in a sense, the only ones that exist. The formulae of linguistic theory conjure into being a homogeneous and uniform world. Is it a coincidence that this is the world that offers the ideal market for mass-produced consumer goods and the perfect terra nullius for the West’s economic and cultural expansion? As Augustine (City of God, 19.7) put it: ‘the imperial city has endeavoured to impose on subject nations not only her yoke, but her language’. If linguistic results are representative, one could be excused for thinking that ‘globalization’ – the relentless spread of multinational capitalist ascendancy – is no more than manifest destiny.

Linguists don’t make good racists

No sooner is this kind of suggestion raised than it risks being indignantly denounced for its ‘absurdity’, ‘lack of balance’, ‘irresponsibility’ or similar heinous crimes. And it’s certainly true that many linguists, as Hutton has noted, are explicitly opposed to the kind of politics I’m claiming our discipline can tacitly support. But this in no way disqualifies my analysis: there would be no such thing as ideology if political intentions and beliefs corresponded perfectly to actions.

Another consideration that goes against the analysis I’ve been offering comes from the fact that the totalizing, imperialist subtext of linguist theory that I’ve claimed to detect is certainly not the only one that students will retain from their linguistic studies. The intellectual climate associated with linguistics is, for the most part, broadly progressive and, in particular, anti-chauvinist: ‘Looking more closely at languages, and in particular at languages that might seem exotic to us, can make us more tolerant’, one introduction (Gasser 2012) claims. The opposition to prescriptivism drummed into students from an early stage is the most concrete manifestation of this attitude. The discipline also encourages such values as curiosity, logical rigour, and problem-solving. While all of these are clearly prized by the reigning economic order, they may also be enlisted for counter-cultural, critical purposes. It’s certainly not to linguistics graduates that one looks, for instance, for the most likely candidates for blinkered racists. Surely, then, it’s arbitrary to single out the discipline’s putative negative effects, as I’ve been doing?

Objections like these are reasonable – in part. They don’t show that the ideological effects I’ve identified don’t exist, but they do suggest that they’re not the only ones at work. To answer them, and to properly appreciate linguistics’ role in producing the kinds of graduates modern states need, we have to concentrate on the form in which the list’s assumptions are presented to students. Here, a less ambiguous picture emerges.

From domination in theory to domination in practice?

Syntax tree

A tree for the sentence ‘The beautiful girl from Brazil found syntax books in the library’. Source.

From the outset of their studies, students learn that language is to be approached in a highly systematizing and totalizing way. They are instructed in an essentially reductive and classificatory approach to human diversity that defines a single normative model of language (the ‘language faculty’, ‘linguistic universals’, ‘grammatical structure’ etc.) to which linguistic diversity is referred. The rules, generalizations, and categorizations that students learn to make throughout the discipline (and not just in its core subfields) all tend in a single direction: almost exclusively, intellectual effort is devoted to bringing complex facts under the scope of general rules, and to reducing the kaleidoscopic manifestations of language to the operations of a unique and singular structure.

Students in different parts of the discipline learn, for instance, that event predicates should universally be sorted into a handful of types (Vendler’s); that the basic unit of language is the sentence (utterance, turn, proposition – it doesn’t matter: the point is that there has to be a basic unit); that some sentences should be considered grammatical, others ungrammatical; that propositions are the base of meaning; that concise ‘conversational maxims’ or principles like them govern our social use of language; that ‘information’ observes a ‘topic-comment’ structure; that speech acts are constative or performative; that a limited palette of classifications can describe the interpersonal meanings conveyed by texts; that the multiplicity of words’ uses can be reduced to their unique ‘meaning’ or ‘meanings’ and that meaning itself is reducible to some set of ‘conceptual primitives’ – and so on.

Behind the diversity and complexity of human linguistic behaviour, there lies, students are encouraged to believe, a single underlying power – abstract linguistic ‘reason’, deriving from the constants of psychology, human nature, or the essential properties of the linguistic ‘system’ itself. Concepts like lexical aspectual categories and the others mentioned are not typically presented as partial, interpretative perspectives on linguistic facts, useful for some purposes and not for others. Instead, they are reified into the permanent essence of linguistic structure. Students are trained, and tested, in formal operations of reduction and analysis far more than in hermeneutic ones of interpretation or complexification. Linguistic diversity is presented as what is left over once winner-takes-all generalizations have swept up as many particulars as possible.[4] The diverse linguistic aspects of human life are presented as the rational products of underlying rule systems. To accomplish this, language is heavily idealized: except in a handful of domains, what is studied are ‘grammars’, ‘vocabularies’, ‘language families’ – imaginary, idealized constructs remote from the realities of situated linguistic ‘performance’.


Gelli’s Dialogue on the difficulty of ordering the [Florentine] language, published in 1551 as an appendage to Giambullari’s Language spoken and written in Florence. Source.

It is only because they have been idealized that languages admit the generalizations about them that students are taught to draw. Generalization and idealization are, of course, central to intellectual activity and could not somehow be disappeared from linguistics (or, for that matter, from any other kind of rational enquiry). But they can be conducted and presented to students in a number of different forms, of which the totalizing, universalizing form present in linguistics is just one. This particular mode of idealization would be perfectly fine – if it actually worked. Really worked. But that’s exactly the problem: despite the high inherent interest and intellectual richness of the analytical posits and theoretical categories I’ve mentioned, the discipline hasn’t been able to agree that any of them does actually do what it’s meant to. They are all, precisely, preliminary hypotheses about underlying structure, yet to be accepted by the entire community of linguists, and the objects of sometimes furious disagreement among them. They are, in addition, heavily dependent on the way in which performance data is idealized in the first place. Are the Berlin and Kay colour generalizations really true? What about the thematic hierarchy? Are simple noun phrases really always monotonic? Can interpretation really be reduced to a Q-principle and an R-principle? Is information always to be conceived of as either old (topic) or new (focus)? The answers depend on a multitude of little decisions about how you idealize and normalize a chaos of variable performance data into the imaginary constructs of a ‘language’ or a ‘grammar’. These are creative decisions informed by a host of considerations over which opinions can legitimately differ.[5] Elevating any one of them into the fact of the matter about language is unwarranted.[6]

The theoretical ‘results’ taught to linguistics students in their opening years of study do not command anywhere near the level of disciplinary consensus as the results taught to undergraduate science students. Yet, more often than not, linguistic theories are presented to beginning students as ‘scientific’ and hence as enjoying an authority that’s similar in kind to those of the natural sciences – not as great, certainly, but in the same league. Some linguists may hesitate to make that claim directly, but it is nonetheless always there in the background (just look at a random sample of departments’ ‘why study linguistics’ pages if you want to confirm this).

Students quickly learn, then, that linguistic experts can claim a ‘scientific’ authority for their own favourite candidate theory even in the absence of any disciplinary consensus. A student taught by Chomskyans will be taught the correctness of generativism and the folly of cognitive grammar. Cognitive linguists trade on the authority of science for their own entirely subjective analyses. Systemic-functional linguists likewise claim a uniqueness, necessity and objectivity for their own proposals. For all the live-and-let-live civility of most linguistics departments, the discipline often resembles a slowly churning snake-pit of competing academic ‘lobbies’ (I owe the expression to Rastier 1993: 155), each presenting its own way of understanding language as correct and, more often than not, uniquely so.

This competition is not just over immaterial, intellectual stakes. Since theories are the instruments of careers, they crucially concern the acquisition and exercise of institutional power, embodied in good marks for students and successful careers for academics. Education in linguistics comes to prefigure the clash of interests in society: students learn that they have to fight for their corner, and that they can deploy the full ideological power of claims of scientificity, reason, empirical responsibility, etc., to do so.

Politics, linguistics, or both?

Politics, linguistics, or both?

Neither oversensitivity nor contrarianism

I’ve suggested that students are encouraged to generalize and theorize about the human world in a highly idealized way subject to only loose empirical controls, and strongly open to discretionary, preferential choices, while at the same time arrogating ‘scientific’ authority to this activity. The effect of this, it seems to me, is to accustom them, in the symbolic order, to the kinds of arbitrary domination that they will both perpetuate and endure through their position in society after graduation. Once ejected from university, students must be able to rationalize for themselves the varied forms of arbitrary domination and exploitation, along class, race, and ethnic lines, among others, that the economy entails. They will both enact and undergo the exploitation of the neoliberal economy, justified, in the face of all evidence, as the only possible horizon for the organization of human societies. Can we be sure that the ‘training’ they receive in linguistics, with all the properties I’ve described, dispensed at the very moment they are preparing to enter the labour market, does not play a role in normalizing this kind of unjustifiable domination? I think there’s a serious risk that the kinds of justification for the theorizing undertaken – or, rather, the licence to claim the authority of science for what are essentially discretionary and unoperationalized interpretations – prepare students for the arbitrary and unjustifiable experience of hegemony in society. I develop these ideas, with a focus on the humanities in general, in Riemer (forthcoming), cited earlier.

But, in the end, why worry about linguistics’ ideological feedback on society? Wouldn’t it be better for linguists who are concerned about politics to spend time doing direct political work (involvement with social movements, parties, campaign groups, etc.) rather than wasting it denouncing their discipline for effects like the ones discussed here – many of which, on my own admission, are highly contradictory, refracted and attenuated?

Quite aside from the fact that most linguists don’t have time to properly devote themselves to politics,[7] the answer, it seems to me, is no. If linguistics was a hard science – if, in other words, theorizing was governed by discipline-wide protocols that produced objectivity and consensus – then there would certainly be grounds for simply accepting whatever procedures enquiry demanded, regardless of their apparent ideological import. But that is not the case. For no linguistic subfield do we have a single best theory accepted across the discipline: we do not even have any agreement on how to define linguistics’ object of study. If I can be allowed some self-quotation, the conclusion I’ve reached elsewhere about semantics applies more generally:

As a human “science”, semantics concerns a sphere which is intrinsically bound up with the behaviour of autonomous creatures with their own pluralistic ways of being and understanding. In such a domain, it is not immediately clear that theoretical insight is best obtained by objectifying reduction, assimilating meaning to a unique object open to empirical methods deriving from the study of the objective world, instead of by pluralistic interpretation, assimilating the study of meaning to that of higher-level socio-cultural manifestations. Cultural anthropology, literary history and sociology – all three empirical disciplines which offer explanations, and not just descriptions, of their objects of study – do not aim to produce unique and reductive analyses of their explananda; it is no more obvious that semanticists should try to uniquely characterize the literal meaning of an expression, than it is that literary historians should try to uniquely pin down the single correct interpretation of a canonical text.
(Riemer 2016: 4)

We should resist the attempt to impose a single vision of language, meaning or human nature, when this vision is empirically contested, and, crucially, unoperationalized. We should also hesitate to accustom students to the instrumentalization of truth claims in the service of institutional power struggles. I’m not sure whether we should say that science is, all other things being equal, progressive – but it seems reasonably clear that scientism isn’t. Assuming, on very narrow grounds, that the diversity of human ‘languaging’ can be enclosed in the totalizing schemes of our models, and claiming the warrant of science for these schemes despite the absence of disciplinary consensus over them, don’t look to me like an intellectual trajectory likely to foster the ‘bond of peace’ that Augustine, immediately following the quotation given earlier, saw in linguistic uniformity.

These concerns have been growing on me over some years, as a result of the undergraduate teaching in semantics and pragmatics for which I am responsible. These are fields of enormous intellectual richness and interest – but I worry about the world-view that, in their traditional form, they can reinforce. Oversensitivity? Contrarianism? Inadequate trust in students’ discernment? I don’t think so. The world is not in a good way, either socially or environmentally. As people responsible for educationally preparing the next generation, we cannot think too deeply about what kind of societies we are helping, in our small way, to form.


I’m grateful to James McElvenny and Jacqueline Léon for having discussed these ideas with me, and to James McElvenny for his careful editorial suggestions. Neither should be assumed to agree with anything argued here.


[1] This is a blog-post, so I’m not going to document or reference any of these claims extensively or, often, at all. A more scholarly presentation of some of these ideas is currently in the works.

[2] Given this, it’s no surprise that many practising linguists really don’t like it when you try to call their models’ objectivity or scientificity into question. As attested by the scarcity of serious metatheoretical, foundational theorizing in linguistics, the discipline’s culture is strongly positivistic.

[3] It would take far more space than I have here to develop and justify this claim properly. Anyone interested should consult the first and last chapters of my (2005) for a general defence of the interpretative and therefore hermeneutic (non-objective) nature of semantics.

[4] It’s interesting to note from this angle how in typology ‘universals’ don’t have to be universal at all – they can be statistical. But linguistic diversity is still approached as a hunt for what is universal, and particulars are of interest only to the extent that they enrich more general schemes.

[5] For example, decisions about what counts as the ‘simplest’ theoretical model cannot be settled objectively – see Ludlow (1999).

[6] Compare the situation in the hard sciences. Here, the theoretical models taught to undergraduates can be deidealized in order to accomplish two key empirical goals: making accurate predictions about actual events, and producing machines that can effectively ‘work’ in the real world. The traditional theories of core linguistics allow them to do neither: since the utterances we produce bear only a tenuous resemblance to the normative structures which serve as the basis of linguistic theory, no one can yet deidealize linguistic models to show how they actually relate to observed linguistic behaviour. See Riemer (2009) for discussion relevant to syntax; in semantics, consider the simple fact that theories of meaning always depend at some point or another on a distinction between the literal and the metaphorical, but we have no idea of how such a distinction might properly be drawn. It’s true that, as James McElvenny comments, fields like speech recognition and synthesis, machine translation, and data mining represent applications of theoretical approaches to language. But these typically use empirical and often statistical models remote from the centre of either descriptive or theoretical work in the discipline. See for instance Jurafsky & Martin 2007 Speech and Language Processing: An introduction to natural language processing, computational linguistics, and speech recognition.

[7] I’d be keen to hear of exceptions beyond the obvious example of Chomsky. Geoffrey Sampson, for instance, has been a Conservative UK council member. And according to Ben Braithwaite, ‘Former Prime Minister of Trinidad and Tobago, Kamla Persad-Bissessar once worked as a lecturer in the Department of Linguistics and English at the UWI’s Mona campus in Jamaica, and President, Anthony Carmona was lecturer in the Department of Language and Linguistics in Trinidad and Tobago. Sir Colville Young, Governor-General of Belize, and Dame Pearlette Louisy, Governor-General of St Lucia, are trained linguists. Basically, if you want to be a world leader, do linguistics’ (https://languageblag.com/2016/03/06/why-study-linguistics/).


Blackburn, Robin. 1967. A brief guide to bourgeois ideology. In A. Cockburn and R. Blackburn (eds) Student Power. Harmondsworth: Penguin.

Callinicos, Alex. 1990. Against Postmodernism. New York: St Martin’s Press.

Fromkin, Victoria, R. Rodman, N. Hyams. 2010. An Introduction to Language. 9ed. Wadsworth: Cengage Learning.

Gasser, Michael. 2012. How Language Works. The Cognitive Science of Linguistics. 3ed. http://www.indiana.edu/~hlw/index.html

Horkheimer, Max and Adorno, Theodor W. 2002. Dialectic of Enlightenment. Philosophical fragments. (Edmund Jephcott, tr.). Stanford: Stanford University Press.

Hutton, Christopher. 2001. Cultural and conceptual relativism, universalism and the politics of linguistics: dilemmas of a would-be progressive linguistics. In René Dirven, Bruce Hawkins, and Esra Sandikcioglu (eds.) Language and Ideology: Cognitive Theoretical Approaches. Amsterdam: John Benjamins, 277-296.

Ludlow, Peter. 1998. Simplicity and Generative Grammar, in Stainton, R. and Murasugi, K. (eds.) Philosophy and Linguistics. Boulder: Westview Press.

Rastier, François. 1993. La sémantique cognitive: éléments d’histoire et d’épistémologie. Histoire Épistémologie Langage 15: 153–187.

Riemer, Nick. 2005. The Semantics of Polysemy. Reading Meaning in English and Warlpiri. Berlin: Mouton.

Riemer, Nick. 2009. Grammaticality as evidence and prediction in a Galilean linguistics. Language Sciences 31: 612–633. https://www.academia.edu/15577226/Grammaticality_as_evidence_and_as_prediction_in_a_Galilean_linguistics

Riemer, Nick. 2015. How to justify a crisis. Jacobin Magazine, October 5. https://www.jacobinmag.com/2015/10/refugee-crisis-europe-zizek-habermas-singer-greece-syria-academia/

Riemer, Nick. 2016. Semantics – a theory in search of an object. In N. Riemer (ed.) The Routledge Handbook of Semantics. Abingdon: Routledge, 1–10. https://www.academia.edu/15576800/Introduction_Semantics_a_theory_in_search_of_an_object._In_Nick_Riemer_ed._The_Routledge_handbook_of_Semantics_Abingdon_Routledge_2015_

Riemer, Nick. Forthcoming. Academics, the humanities and the enclosure of knowledge: the worm in the fruit. Australian Universities’ Review 2016. https://www.academia.edu/17869935/Academics_the_humanities_and_the_enclosure_of_knowledge_the_worm_in_the_fruit

How to cite this post

Riemer, Nick. 2016. Diversity, linguistics and domination: how linguistic theory can feed a kind of politics most linguists would oppose. History and Philosophy of the Language Sciences. https://hiphilangsci.net/2016/05/11/diversity-linguistics-and-domination-how-linguistic-theory-can-feed-a-kind-of-politics-most-linguists-would-oppose

Tagged with: , ,
Posted in Article, Linguistics
18 comments on “Diversity, linguistics and domination: how linguistic theory can feed a kind of politics most linguists would oppose
  1. This is horrid, scientistic, self aggrandizing, intellectual elitist language! It reminds me of Noam Chomsky’s non-political writing! It is like Chomsky writing in his extremely specialized, technical lingo! But, if Noam Chomsky is trying to make political points or actually writing political ideas, he does, actually speak fairly down to earth language, about his politics! But, I did get a few ideas, to grab onto, for my comments, were burning to get out! The paper talks all that jazz, about scientism & critiques of something! But the ideas that just killed me, were some nonsense, at the very opening passages! Something about language being individual centric, as if the “socialist” needs to get the social properties of language, back into the linguistics practice! The individual centric bias was dominating the very art of language itself! It struck me as extremely odd, that anybody would assume language is purely or primarily an individual centric art! Even the most conservative libertarian, individualist social science teachers from the Vienna Law school’s laissez faire, classical liberal economists, akin to Ludwig Von Mises, would not say language is primarily individual centric! The whole point of language is communicating with social neighbors! But this whole attacking of capitalist social progress, as some static obstacle to real clear social progress, is never set in plain language! It is clearly not front & central to the scholar, that he has any great social concern, as for letting the individual social producers, be free to trade & accumulate the fruits of their labours! But, all the scientistic smarter than thou, quality of communism, is just plain assumed! The Socialist smart set, are so much more intelligent, than the social mass! The Socialist intellectual elites’ ability to do radical surgery on the social order, is just plain assumed! That fits well, in this article! It is planning or writing up a preparatory plan for doing radical surgery, on the social culture of middle class human capital owners! That is ,the bourgeois owners of their own minds, inspirations, social fellow feeling animating of their psyches with the golden rule sense of looking in the mirror! That is, after all, the best description of that profit motive, so demonized by capital “S” Socialism! After all, if I were to ask god for an extra hand or robot, that can read my mind & not break down or trip up, at the time & place, I need that helping hand, I would be describing the nature of human capital, animated by the golden rule! So, that golden rule, pays off in golden returns, on that investing in social helping hands of mine, for my neighbors! But, not a single word of my simple social concepts, is on the agenda of this openly, happy to be proudly asserting himself, as a communist expert! The writer makes use of the term scientistic or scientism! I have been away from communist writers for so long! I thought only bourgeois, laissez faire individualist social praxeology used that term! The conservators of individual social- economic liberty, use scientism to describe Socialist state-ism’s elite expert planners or social engineers!

    • nickriemer says:

      Thanks for both your comments. Sorry you found the language elitist. I think it’s fair to say that linguistics is fundamentally individualistic since mental representation is at the base of practically everything. Any social effects that exist get routed through the psychology. Meanings, for instance. It’s not essential for my case that every single subfield in the discipline studies those mental aspects – the point is that, especially as it’s presented to undergrads, it’s all understood as fitting together in a way that presupposes the individual at the core of things.

  2. I was impressed with the writer’s point ,that linguists tend to be more social people friendly, than not! His point makes sense to me! I am an incomplete undergrad! I loved historical comparative linguistics! I could stare at maps or charts of language families, along with geological & ethno-linguistic, religious,, genographic & anthropological family trees, forever! Sorry, your lingo, like Noam Chomsky’s technical lingo, is way over my head! I hope I do not write, like your paper here, when I am trying to write for normal folks! I did not mean to insult you, though! I wish I was able to have a dictionary ready for your article! My passion is dictionary roots of Indo-European & all else I need to look up! But thanks for trying!

  3. Lameen Souag says:

    Interesting post, but I’d take issue with its framing in terms of progressive politics. To the extent that your critique is accurate (and there’s certainly some truth in it), linguistics students are being taught how to delude themselves into thinking that they understand more than they actually do, and how to confuse postulates with facts. I would suggest that that would be a huge problem even if there were no “experience of hegemony in society”, and even if life outside the university gates corresponded perfectly to your preferred vision of a political utopia. Such practices contravene much more basic and widely accepted principles than any of the “characteristic progressive priorities” you list: in particular, the principles that truth and understanding are valuable, and that universities are supposed to promote them. Why not criticise them as such, then, rather than identifying the problem in terms of potential political effects that you yourself describe as “highly contradictory, refracted and attenuated” ?

    A more specific point: Most of your list of assumptions in the first part seem more or less plausible, but your description of “transparency” strikes me as very strange indeed. The fact that technical concepts are ultimately defined in terms of commonsensical ones is just as true of physics or mathematics as it is of linguistics – indeed, if they weren’t, it’s hard to imagine how they could be meaningfully defined at all. And if there’s one thing linguistics is not short of, it’s an extensive apparatus of technical concepts – “noun”, “verb”, “preposition”, “phrase”, “phoneme”, “bilabial”, “wh-movement”, “cognate”… Moreover, perhaps my experience is atypical, but I would have thought that the “domains of linguistics most prominent in standard undergraduate syllabuses” are, precisely, syntax and phonology.

    • nickriemer says:

      Thanks very much for your comment. I agree that it’s not a good thing to present postulates as facts. The reason for analyzing this in political terms as I’ve done is that it offers an explanation for why this is such a tendency in the discipline. Linguists often instrumentalize theoretical ideas for the purposes of survival in the field, and presenting them as facts is part of this. You might argue that other explanations are available that avoid the political ones. I’m not sure I’d agree. Linguistics is something that people do, and the way they do it is governed by considerations that are highly political (or, at least, social, which I take to be synonymous).

      On transparency, my claim isn’t that linguistics has no technical terms ; just that the degree of technicality is lower than it is in the hard sciences. I think you can see that simply from the greater comprehensibility of linguistic theory to novices. It’s often not too much effort to paraphrase a ‘technical’ linguistic proposal in ordinary language, and to do so entirely accurately. This isn’t the case in the hard sciences, I don’t think.

  4. Sebastian Nordhoff says:

    I remember Johanna Nichols saying (quote from memory): “Why is it easier for me to understand a random paper in, say, Earth Sciences, than a theoretical linguistics paper written on the a language of a family I have been working on for the last 20 years?”. This as a comment to your assertion: “It’s often not too much effort to paraphrase a ‘technical’ linguistic proposal in ordinary language, and to do so entirely accurately. This isn’t the case in the hard sciences, I don’t think.”

    Especially in generative syntax (which is what is most often taught), explaining what, say, a Specificier is in “ordinary language” is more difficult than explaining a black hole. YMMV.

    The generative enterprise as whole has indeed a very individualistic nature and does not like language variation, language change, and any other kind of imperfections that arise from actual people communicating with each other. However, the fields of language variation and change, language contact, and basically everything under a broad umbrella of “functionalism” would not seem to fall under the claim of linguistics being individual-centered. The same is true of “Evolutionary Linguistics”.

    I would appreciate a differentiation between generative approaches and linguistics. It appears to me that many of your interpretations are true for what I call “truth-conditional linguistics”.

    • nickriemer says:

      Thanks for commenting. It’s a fair point: generative linguistics is often defiantly unparaphrasable. But it’s something of an outlier in the discipline: there are plenty of ways to do syntax differently, and they conform much better to my generalization. And the bits that undergraduates encounter are less technical than cutting edge papers.

      I’m not sure it’s right to say that everything that’s broadly functionalist is less individualist. Take Hopper and Thompson’s classic approach to parts of speech, according to which nouns function to introduce ‘participants’ and ‘props’ into the discourse. That function is one that’s seen as serving the speaker’s communicative intentions – which takes us right back into classic, subject-oriented individualism. It’s very hard to get away from this, and it it’s hard to see how it could be bypassed entirely. But there’s a hyper-concentration on individualism in linguistics which seems unhelpful to me (semantics is the best example — meanings are things in your head).

      As to language variation and change, language contact and so on, agreed. But my generalizations concern the core architecture of linguistics, i.e. the conception of the nature of linguistic structures (grammar, semantics etc.) that come into contact, change, vary, etc.

      • qirmesboxer says:

        I find it hard to conceive of a “communicative intention” as being individualist, given that the intention necessarily involves an addressee, i.e. a person other than ego. In order to realise your communicative intentions, you will have to make do with the other folks around you.

        The definition of semantics as “things in your head” sounds very pre-Wittgensteinian to me.

        In the last paragraph of your reply, you fall prey to the very current you criticized before: the core architecture of linguistics is the conception of structures. What if the core architecture of linguistics were not structures but communicative goals/intentions (aka “functions”)? This would make the whole field less solipsist.

  5. Tulishen says:

    This is a great post, and I’m glad to see finally how Mr. Riemer’s approach to semantics finally matches his progressive politics. I had been curious about that.

    What I find amusing is that Chomsky’s politics are hardly non-progressive! He has been the most prominent leftist academic for decades.

    Chomsky’s rationalist linguistics have been adopted by people both in the left and the right (e.g. Pinker). I see no reason why the sort of Wittgensteinian approach that Mr. Riemer take cannot be both used by people in the left and the right too.

    Rationalism, especially when it blossomed in the Enlightenment, was a fundamentally progressive movement, and replacing it by a more social understanding of cognition, which I agree is well overdue, should by necessity change the approach to progressive politics as well.

    • nickriemer says:

      Thanks very much. The salient thing about Chomsky, of course, is that his politics and his linguistics are only tenuously related, if at all. And I agree: ideas about language are open to different ideological uses at different times. The ideological effect of a theory isn’t intrinsic to it, but is a function of how it is or could be put to work. I don’t think that what I’ve called the totalizing impetus behind a lot of linguistic theory is intrinsically regressive. I just think that in the present social circumstances it can contribute to the wrong kind of politics, and so should be presented to beginning students cautiously and with the appropriate qualifications.

      • Lameen Souag says:

        Those beginning students may well have their own ideas about what constitutes the “wrong kind of politics”, and even if they don’t, they certainly didn’t come to the class to imbibe ours. If students come to believe that we select what we present to them based on its potential political impact rather than its usefulness in understanding language, they will have every reason to distrust us. And what you’re saying is, at the least, very susceptible to such a reading.

  6. nickriemer says:

    @Lameen Souag: Such a reading wouldn’t have any basis in the text: I don’t claim anywhere that the theoretical content of linguistics courses should be governed by political considerations. I just say that that content has possible political corrolaries, and that, as a result, linguists should think about the way they present it — in particular, they should consider whether they’re dressing conjectures up as facts. It’s not possible, in my opinion, to think that undergraduate instruction, of all things, can be conducted in a neutral way that doesn’t engage students’ political beliefs at some level. The fantasy of a neutral, purely theoretical environment is just that, a fantasy.

  7. nickriemer says:

    @qirmesboxer: The existence of a hearer doesn’t make things less individualistic: H is just the mirror image of S. Speaker’s meaning and hearer’s interpretation are two separate, isolated things. Information is conceived of as something which passes from one individual to another.

    V relevant of you to raise Wittgenstein. It’s always struck me that semantics is thoroughly anti-Wittgenteinian as a field. I discuss (the later) W. and consider the implications for semantics in the 2005 book cited in the post.

  8. qirmesboxer says:

    Well, S and H have to agree on a meaning, and the meaning is negotiated during communication . If S intends M and H interprets M’, there is surely a relation between M and M’, albeit not the relation of identity. If M and M’ diverge to much, there are misunderstandings, and repair is initiated. The whole process of communication is cooperative, not isolated. Furthermore, for communication to succeed, both S and H have to have some Theory of Mind. The idea that S packages a messages, sends it off, and hopes for the best is misconceived (although probably widespread in truth-conditional semantics). I agree that much of semantics is not particularly concerned about humans and their agency in the world.

    Not sure where that definition of information comes from, and in how far the concept of “information” is relevant here. “Information” in the sense of transmitting facts about the world to other people is one use of language, but assertions are not the only speech acts out there.

    • nickriemer says:

      Communication is cooperative, but at every point meaning is theoretically conceived of as an individual mental representation: that’s the nub of the question.

  9. […] Nick Riemer writes about learning the English language for both native speakers and English Second Language Learners, and he suggests the written word’s grammar confines. Riemer writes that writing is taught “in an essentially reductive and classificatory approach to human diversity that defines a single normative model of language to which linguistic diversity is referred.” Riemer further argues the exclusion of other Englishes diminishes the language in general and reduces “the kaleidoscopic manifestations of language to the operation of a unique and singular structure.”   […]

Leave a Reply to socialdynamos Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Upcoming events

28-30 June 2023
Faro, Portugal
International Inter-association (History of Language Teaching) Conference
Language teachers, methodologies and teacher training in historical perspective

4-6 September 2023
Universidade de Trás-os-Montes e Alto Douro, Portugal
2023 Annual Colloquium of the Henry Sweet Society
What counts as scientific in the History of Linguistics?

6-9 September 2023
Universidade de Trás-os-Montes e Alto Douro, Portugal
XXXII. International Colloquium of the “Studienkreis ‘Geschichte der Sprachwissenschaft'” (SGdS)
Controversies in the history of linguistics

%d bloggers like this: