Some reflections on the uses and abuses of theory in linguistic thought

Jon Orman

My purpose in this piece is to offer a few brief thoughts on a series of questions with which I have become increasingly interested in recent months. Linguistics, it seems to me, is awash with theories and theoretical work of various sorts, yet it is not all that often that one comes across any critical scrutiny of the theoretical enterprise itself. The basic question motivating my discussion can be put thus: why theorise language? In other words, what motivates the theoretical enterprise in relation to language and what sustains it? This sets in train a series of further questions. Why is there apparently such faith, amongst language scholars at least, in theory? What does theory contribute to our understanding of a human phenomenon? Is it genuine, previously unavailable knowledge or simply a different and perhaps also professionally advantageous way of talking about it? Are theoretical enterprises generated and propelled by genuine intellectual puzzles or anomalies that challenge our settled conception of the world or more by particular ambitions? Or is it the case that where language is concerned theory tends instead to generate merely its own internal puzzles – academic questions in the most pejorative sense of the word? This raises the issue of how, why and by whom particular theories of language come to be valued, which in turn potentially prompts questions which take on a more sceptical tone such as ‘is it possible to theorise language without somehow giving a distorted view or misrepresenting it?’ or even ‘what do we lose sight of by submitting language to theoretical treatment?’

The belief that any identifiable human or social phenomenon is ripe for some kind of theoretical treatment pervades the humanities and especially the social sciences. Writing in the 1970s, the philosopher A.R. Louch spoke of the ‘lust for theory’ which had taken hold in literary studies. A similarly voracious and in some cases intensifying lust also exists, it seems, in many contemporary areas of scholarly inquiry concerned with language. It is fairly common, for example, to encounter the claim that a particular linguistic phenomenon, especially a recently identified one, remains ‘undertheorised’ – the implication being that unless we have a theoretical account of the phenomenon in question, we do not and cannot properly understand it. By way of example, Pavlenko (2007:35) claims that ‘emotions remain undertheorized’ in Second Language Acquisition research. Elsewhere, Elizabeth Miller (2014:4) claims that ‘agency is often treated simplistically and remains undertheorized in Applied Linguistics research’. Attributions of undertheorisation therefore suggest a lack of scholarly rigour or serious treatment, whereas ‘overtheorisation’ tends more to be a term of lay condemnation of intellectual pretentiousness.

Underlying these considerations is a more basic question, a definitive answer to which is perhaps less forthcoming, namely ‘what constitutes a theory of language?’ Is any kind of reflexive contemplation or connected series of statements about a particular phenomenon automatically to be considered theoretical? How stringent a definition are we to operate with? The philosopher Eric Voegelin (1999:64), for instance, was insistent that ‘[t]heory is not just any opining about human existence in society. It is rather an attempt at formulating the meaning of existence by explaining the content of a definite class of experiences’.

The veneration of the theorist in the Western intellectual tradition goes at least as far back as Aristotle’s Ethics and his discussion of the figure of the spoudaios – often translated as the ‘mature man’ and which in Aristotle’s thought also carries suggestions of seriousness, solemnity and moral virtue. According to Voegelin (1999:139):

The spoudaios is the man whose character has been formed by an aggregate of experiences and who has maximally actualized the potentialities of human nature, who has formed his character into habitual actualization of the dianoetic and ethical virtues, the man who at the fullest of his development is capable of the bios theoretikos.

Quite what the bios theoretikos ought to entail, however, is not always so clear. For example, in his book Theorizing (1973:168-9), Alan Blum provides the following Byzantine rationale for the activity of the theorist:

Theorizing is speaking that shows the Reason of critical speaking to lie in the imperative need to re-assert itself as an instance of faithfulness to Reason itself. The success of theorizing in the deepest sense lies in the fact that it re-directs attention to the grounds of speaking by speaking so as to evoke in its very speech the ideal of addressing grounds as the Good […] From a dialectical perspective, theorizing is the speaking which understands itself as inescapably rhetorical, and which acts upon such an understanding by preparing its very speech as an argument for the Rationality of that commitment. Theorizing then achieves its purest character when it conceives of itself as the speaking that is displaying and arguing rather than the speaking which seeks to create a unity among different arguments and displays. Whereas moderns regarded successful inquiry as that which silenced discussion by producing speech to which one must assent, dialectic treats inquiry as the re-opening of a discussion that routine usage has long covered over. Yet the arguing which theory is is not the giving of reasons, but a showing of the achievement of its authorization as a way of affirming the Rationality of that author.

I would be reluctant to take too confident a stab at trying to say what this is all supposed to mean – is Blum saying that the ‘moderns’ (Weber? Marx?) erred by trying to find solutions to puzzles whereas the proper role of theory is to proliferate discourse? In any case, the reader is obviously expected to agree that theorizing is a very important activity indeed and presumably one reserved only for select members of the intellectual clerisy. Further support for the theorist can be found in Edward Said (2000:210), who is emphatic in claiming that ‘of course it is ridiculously foolish to argue that “the facts” or “the great texts” do not require any theoretical framework or methodology to be appreciated or read properly.’ It is notable in passing that Said feels no need to place embarrassed quotation marks around the word ‘properly’ as he does around ‘the facts’. Said would obviously feel little sympathy with Stanley Fish’s claim, no doubt to be taken with a fair pinch of salt, that theory is ‘an inconsequential activity’.

Peter Sloterdijk (2012:37), however, raises the prospect of a far less reverent view of the theorist when he asks:

What if the much lauded theoretical virtues really derive from secret weaknesses? What if they’re based on a questionable compensation for stubborn defects, or even the morbid inability to face the facts of life without embellishment and evasion? Does homo theoreticus really come from such a good background as he has assured us from his earliest days? Or is he actually a bastard trying to impress us with fake titles?

What sort of person, then, might homo theoreticus linguae be?

Wittgenstein and the ‘anti-theory’ position

The most renowned, if not entirely consistent, expression of an anti-theory position in respect of human behaviour and hence language is to be found in Wittgenstein’s later work, especially in the Philosophical Investigations and On Certainty. Wittgenstein saw the urge to theorise as arising from what he called the ‘craving for generality’ (das Streben nach Allgemeinheit) and the desire to imitate the methods and secure the prestige of the natural sciences. For Wittgenstein, there was a fundamental difference between two forms of understanding, which he called ‘scientific’ and ‘philosophical’. Scientific understanding is of the theoretical or nomothetic sort – concerned with the discovery of general principles and operating with a mode of explanation based on causality and prediction whereby explanation consists either in an appeal to regularities or in bringing individual cases under some law. Philosophical understanding, however, is for Wittgenstein avowedly non-theoretical and ad hoc. He famously says that philosophical understanding ‘consists in seeing connections’ (PI, § 122). A favourite example to illustrate this point is that of what it is to understand a piece of music. Wittgenstein makes an explicit connection with language, saying that ‘understanding a sentence is more akin to understanding a piece of music than one may think’. However, he goes on to make the point that the evidence for understanding in either case is ‘imponderable’ and therefore resistant to the generalising requirements of theory.

One slightly odd side-effect of Wittgenstein’s view of the divide between science and philosophy was his insistence that by placing a prohibition on philosophers generalising from specific cases to establish general laws or principles, one therefore had to renounce the possibility of explanation and make do with pure description. Indeed, some of Wittgenstein’s most resonant metaphilosophical remarks are on the issue of explanation. In the Philosophical Investigations, for instance, he says that ‘[t]here must not be anything hypothetical in our considerations. We must do away with all explanation, and description alone must take its place’ (PI, § 109), and ‘[p]hilosophy simply puts everything before us, and neither explains nor deduces anything’ (PI, § 126). Perhaps the most striking of all is to be found in the terse opening remark of the PI, namely ‘[e]xplanations come to an end somewhere’ (PI, § 1).

Wittgenstein’s repudiation of the possibility of explanation stems from his faith in what may be seen as a traditional scientific model of explanation, which involves his upholding of the fact-value distinction and the empiricist doctrine that proper explanation is a matter of exhibiting causality – causality being fundamentally a property of the movement and collision of physical matter – and hence amenable to scientific explanation through the invocation of general principles and laws. Additionally, a key feature of scientific theories is that they can be evaluated as true or false – in other words, they make realist claims about the world. It is this chain of thought which leads Wittgenstein to reject the possibility of ad hoc or non-theoretical explanation.

Although there are strong grounds for taking issue with Wittgenstein’s upholding of the fact-value dichotomy in respect of human action (see Orman, 2017), he clearly identifies the two central notions when it comes to theory, namely generalisation and explanation, or rather explanation through generalisation. It is important therefore to distinguish between two forms of generalisation: on the one hand, generalisation simply as a factual summary of individual cases (non-theoretical) and generalisation as a strategy of explanation (theoretical).

Elements of Wittgenstein’s work are, however, not entirely faithful to its overall spiritual tenor. Indeed, it is significant that what are arguably the most problematic and unattractive elements of his later thought are precisely those which have received the most enthusiastic uptake within linguistics (especially pragmatics in the form of Speech Act Theory), the reason being that they provide the potential ingredients for generalising, theoretically driven methodological enterprises. Despite his aversion to science-aping theory, Wittgenstein – at least on some highly authoritative interpretations – actually succumbs to an explanatory generalisation of his own in the form of his assertion that language is a rule-governed form of action. In other words, the concept of rule-following is generalised to all instances of language, a move which, if accepted, opens the way for the kind of methodological investigation which has fortified linguistics self-image as a bona fide science of language.

Theory and linguistic science

This scientific conception of theory is evident in many of the key models of modern linguistics. For example, Saussure’s theory of language is based upon the reductive, general, exceptionless postulate that langue – the determinate language system – constitutes the sole object of the linguist’s investigation. If it is not langue, then it falls outside the bounds and concerns of the theory and the science. So, in Saussure’s reductionism, we already glimpse some recognition that key elements of language – he called them the subject matter of an external linguistics – are beyond the scope of any theoretical enterprise. It is perhaps unfortunate that Saussure did not live to develop these insights further in his promised but unrealised work on a linguistique de la parole.

Saussure’s model of linguistic communication works on the basis of a simplistic general model operating on causal principles. In brief, A has a particular thought which triggers a corresponding sound pattern in his/her brain, the brain then transmits an impulse to the organs of phonation, sound waves travel from A’s mouth to B’s ear, which then triggers the corresponding psychological association of the sound pattern and concept in B’s brain. This constitutes Saussure’s answer to the general ontological and hence theoretical question of what communication is. It also provides a precise general, i.e. universal, answer to the question of what constitutes understanding – understanding is simply a case of two or more individuals matching the same mental content to the same sound patterns or signs. Under such a model, there are no grey areas or indeterminacy of comprehension – understanding either does or does not occur. Saussure therefore provides generalisable answers to the questions ‘What is a language?’, ‘What does linguistic communication consist in?’ and ‘What is understanding?’ – in fact these last two questions turn out to be synonymous on Saussure’s account.

Now, the obvious question when confronted with a theory claiming to be the basis of science is simply “Is it true?” or perhaps more generously “How true is it?” Clearly to pass muster as science, Saussure had to maintain that his theory and its constructs constituted realist claims about entities in the world. Saussure makes the point in the Cours – and here he is in agreement with Wittgenstein – that a science must advance empirical propositions, yet he famously never identified a single real-life example of a langue as stipulated by his theory. He also offers no criteria for what would constitute empirical verification of his theory.

It is therefore interesting to reflect at this stage on the rationale for Saussure’s theorising. What motivates it? It has been suggested by critics – most notably Roy Harris – that Saussure’s theoretical enterprise is motivated less by a genuine intellectual puzzle or a void in knowledge but by professional ambition and the perceived necessity to set up linguistics as an autonomous science distinct from those other areas of inquiry which also deal with questions of language. It is of course true that the question of how to go about doing this generates theory-internal puzzles of its own and Saussure showed enormous ingenuity in addressing these, but in taking the course he did he arguably reduced the activity of linguistic theorising to an arid, scholastic exercise.

One theoretical approach whose accompanying rhetoric is very much focused on its alleged scientific solving of an intellectual puzzle or set of puzzles is/was Chomsky’s theory of Universal Grammar. The supposed puzzle in question is that of how children are able to acquire mastery of their native language in spite of the limited data available to them – the so-called ‘Poverty of the Stimulus’ argument. What is of interest here is the basic form of Chomsky’s explanation, which allows it to claim nomothetic theoretical status. In order to explain away the puzzle, Chomsky appealed to certain postulated innate, biologically endowed, universal features of language. For example, the claim that all languages are organised in terms of phrases which have ‘heads’, ‘arguments’ and ‘adjuncts’ and that no language that was not organised thus could possibly be a natural human language.

Sociolinguistics and theory

The essential link between theory and generalisation or abstraction is also evident in sociolinguistics, a field which appears to be developing an increasingly insatiable appetite for theoretical reflection. For example, in the recent book Sociolinguistics: Theoretical Debates Nikolas Coupland (2016:6) writes as follows:

Like all theory, theory in the social sciences and sociolinguistic theory should still involve abstracting away from particular data contexts and instances. But distinctively from classical scientific theory, it might also be conceived as providing a guide to social action (the moral agenda of ‘theory’ returning to prominence), so that explanation may be insufficient in itself as a priority.

There are several things to say in this connection. Firstly, it points to an underlying metatheoretical assumption in which abstraction – which amounts to generalisation – is deemed to be a legitimate explanatory move in relation to language. Compare this with Deleuze’s claim that ‘abstractions explain nothing, they themselves have to be explained’ (1995: 146). However, it seems a quite natural assumption to make for a field of inquiry such as sociolinguistics concerned as it is with the distribution and function of repeatable linguistic and other semiotic forms, a concern which finds theoretical expression in the materialist notion – consistent with the neo-Marxist ethic which pervades the discipline – of linguistic resources. The essential point here is that as soon as one has something which can exist in two places at once – the same form here, the same form there – one can begin generalising beyond the individual case and embark upon theory construction.

However, it also points to a tension at the heart of sociolinguistic theoretical endeavours, namely the twin desires to on the one hand make scientifically respectable and explanatory valid generalisations and, on the other, to pursue a particularistic socio-political, moral or, dare I say it, ideological agenda. Problems potentially arise when the truth claims of the theory are seen to contradict or fail to provide sufficient support for the moral-ideological and – just as crucially – methodological agenda. The question, then, is which aspect takes theoretical priority? It is significant, I think, that despite its theoretical promiscuity, sociolinguistics won’t just hop into bed with any old theorist or philosopher. This is not so much a question of high standards, however, – why, even Derrida still gets a look-in – as a reflection of a rather limited, ideologically determined taste. The preferred theoretical bedfellows in the case of sociolinguistics tend increasingly to be those neo-Marxist, poststructuralist thinkers who nourish its basic ideological and methodological conceit, which in turn fortifies its practitioners’ claims to expertise in matters linguistic.

The foregoing considerations point to two potential dangers for theory construction in relation to language and communication, namely those of overgeneralisation and imprecision. The issue of generalisations, or statements cast in general form, which turn out to admit exceptions – in other words they turn out not to be general – is a fundamentally problematic one, albeit one that is rarely acknowledged as such, for the credibility of any theoretical enterprise. These are what are sometimes called – in the sciences at least – ceteris paribus generalisations. Essentially they are overgeneralisations. A less lofty way of describing them would be statements which are true except when they’re not. A simple example would be something like ‘People eat when they are hungry.’

Probably the most well-known example from the history of linguistics of the identification of a generalisation which turned out not to be exceptionless was Grimm’s Law concerning the first Germanic Sound Shift – the exception was famously noted by Karl Verner, who realised that Grimm’s Law was only valid when the stress fell on the root syllable of the Sanskrit cognate word. As a result, so-called Verner’s Law was formulated to account for this exception in accordance with the Neogrammarian creed that ‘sound change laws admit no exceptions’.

So what other kind of ceteris paribus generalisations can one identify in the work of linguistic theorists? Some clear examples can be found in the work of Mikhail Bakhtin, a theorist in whom there has been a surge of interest in recent years within sociolinguistics, a development which elicits sideways glances from literature colleagues for whom Bakhtin fell off the fashionable theorist radar some moons ago. However, when read as theoretical statements, some of Bakhtin’s pronouncements must be classified simply as false or incoherent. The following examples are taken from his famous article The Problem of Speech Genres (1996).

Neutral dictionary meanings of the words of a language ensure their common features and guarantee that all speakers of a given language will understand one another.

One might think it remarkable that someone who can make such a statement can possibly be taken seriously as a linguistic theorist at all. A more complex problem is presented by the following example:

When selecting words, we proceed from the planned whole of our utterance, and this whole that we have planned and created is always expressive.

Bakhtin presents this piece of psychological speculation as a general statement about language, although he does not say how he arrived at the insight. If we leave aside the by no means theoretically unproblematic assumption that linguistic communication is simply a matter of selecting words, it strains credulity to suppose that we always plan our utterances in advance, as if everything we say follows on from some inner rehearsal process. What about spontaneous, unplanned utterances, where we literally don’t know what we are going to say until we’ve said it? Now if we drain Bakhtin’s claim of its generality, it reduces to little more than the commonplace observation that sometimes we plan what we say. How often we do so it would appear very difficult to say.

In addition to this kind of overgeneralisation, Bakhtin’s work also contains a series of what one might call open-ended generalisations. Take the following, for instance:

A speech genre is […] a typical form of utterance; as such the genre also includes a certain typical expression that inheres in it. In the genre, the word acquires a typical expression. Genres correspond to typical situations of speech communities, typical themes, and, consequently, also to particular contacts between the meanings of words and actual concrete reality under certain typical circumstances. Hence also the possibility of typical expressions that seem to adhere to words. The typical expression (and the typical intonation that corresponds to it) does not have that force of compulsoriness that language forms have…

This whole passage hangs on the undefined term ‘typical’. Bakhtin does not supply any general criteria for determining what constitutes a ‘typical’ utterance, expression, situation, theme, intonation etc. Nor does he provide any examples. Furthermore, the notion of typicality can be said to be theoretically enigmatic insofar as something may be typical or atypical in an indeterminate number of respects – an inherently open-ended concept. It is also a question about which people can be reasonably expected to disagree. Consensus cannot be assumed. The problem, then, is that if we are unable to definitively identify a typical utterance, word, expression etc., other than stipulatively, it then becomes impossible to identify the theoretical object, i.e. the genre, which is supposedly elucidated by appealing to it in the first place. The only way to then identify a genre is, similarly, by means of stipulation.

One further consequence of the vagueness and stipulative nature of Bakhtin’s formulation is that it provides it with insulation against refutation. We cannot make the general claim that a speech genre is NOT a typical form utterance because we have no general criteria for saying what a typical form of utterance is. However, such insulation is bought at the price of explanatory significance. So this leads to the question: If the theory doesn’t explain, what does it do?

What further sense can be made of this passage? Given its failure to specify any determinate theoretical entity, it could be read as expressing the commonplace and, again, rather unexciting point that particular uses and forms of language are associated with certain types of situation. The question then becomes ‘why do we need a theorist of language to tell what we surely – in some fundamental sense and however explicitly articulated – already know?’ If we didn’t somehow know it, we would very much be the communicatively incompetent ‘social monsters’ which Dell Hymes famously imagined. One answer which suggests itself – and this might be considered part of a wider postmodern rationale for theory – is that the constructs or terminology of a theory provide the basis for talking about phenomena in new, different ways. Theory then becomes, in effect, a way of redescribing in sometimes exuberant vocabulary what we either already must know or could at least conceivably come to know pre-theoretically. Theory, it seems, may be a disguise for platitude.

There is an important methodological issue here too. If a theoretical claim has some high degree of generality, it becomes almost trivially easy to find instances of data which conform to it. Cue the spectre of such intellectual curiosities as Bakhtinian analyses of teenagers’ text messaging and Foucauldian Facebook studies. On the other hand, if a claim is generally valid, i.e. universal, then all instances conform to it and no confirmation via individual cases is needed because the generalisation already serves to identify the phenomenon in the first place – as is clearly the case with Chomsky’s universal grammar theory, for example. In this sense, then, it becomes a formal or logical claim.

The consequences of this were noted by the Norwegian social psychologist, Jan Smedslund (1979:9), who noted that:

Theories which aim at being testable and empirically valid must fit the local conditions and, hence, cannot be general. Theories which aim at being general cannot fit particular local conditions and hence cannot be testable and empirically valid. Their validity has to be purely formal. […] This means that we must stop believing that our data are relevant for and support or refute general theories.

The only way to get around this difficulty is to treat as constants those aspects of communicational situations which, in reality, are inherently variables, in particular questions of value and meaning.

The activity of theorising may, then, become a pretext for interpretive enterprises which nevertheless often parade as empirical. A clear expression of this tendency is to be found in the work of anthropologist Clifford Geertz associated with the so-called ‘hermeneutic turn’ in anthropology and noted for what he called an ‘interpretive theory of culture’ based around the notion of Thick Description (TD) (Geertz, 1973a, 1973b) and a ‘semiotic concept of culture’. Geertz takes up some curious positions. On the one hand, he is quite explicit in claiming to do science yet he asserts that for him the aim of anthropology is not to solve puzzles as such but rather to achieve ‘the enlargement of the universe of human discourse’ (1973a:14) – he also says elsewhere that anthropological writings are ‘interpretations’ and ‘fictions’.

The role of theory, then, for Geertz is to make possible what he calls Cultural Interpretation. Culture with a capital ‘C’. For example, he makes explicit mention of seeking to understand Moroccan culture as a ‘theoretical entity’ (1973a:37). He says elsewhere that the ‘essential task of theory […] is to make thick description possible’ (1973a:25), TD being ‘the close-reading of a culture’ with the suggestion being that ‘proper’ TD is not possible without theory. It is also apparent that Geertz is not especially bothered by what one might call the truth value of a theory. Instead, theories merely have to be useful, which Geertz glosses as ‘throwing up new understandings’ (1973a:27). Once they stop being useful, they are presumably abandoned and one moves on to the next theory.

It is telling, then, that one of Geertz’s most renowned theoretical contributions or devices comprises a metaphor, and a rather ethnocentric one at that, namely that of ‘CULTURE AS TEXT’, which itself springs from the idea that the key to understanding culture is to apprehend the ‘symbolic meaning’ of social actions – a notion which Ernest Gellner (1992:30) rather nicely labelled a ‘conceptual intoxicant and instrument of self-titillation’.

Now, one key characteristic of metaphors is surely that they are not straightforwardly evaluable in terms of being true or false. Rather, metaphors are more naturally assessed in terms of how illuminating, fruitful or productive they are – and this of course can only be done in relation to specific purposes. As he makes clear, Geertz’s purposes are fundamentally methodological – in his case a kind of unending textualism. Texts beget texts in order to serve the stipulative goal of expanding the universe of discourse and it is in this way that his theoretical entities take on a self-justifying aspect. One significant effect of the ‘culture as text’ metaphor is, therefore, to reify and render culture static, as a product, a text, rather than a process, a move which obviously makes the ethnographer’s task of symbolic interpretation rather easier – the methodological benefit. One of Geertz’s critics, Heather Love (2013), puts it well when she talks of Geertz’s ‘freezing of behaviour into culture’ in order to ‘clear the way for interpretation and bring together data gathering and textual analysis.’

Geertz’s insistence on the ‘culture as text’ metaphor has been described as the call sign of a ‘positivist in despair’ (Yoshida 2014:57) – the idea being that if anthropology cannot be an empirical science, then it must be literature, pure interpretation, therefore anything goes and issues of accountability to the embarrassing notion of truth fall away. However, Geertz’s hermeneutic enterprise is undermined by his linguistic metatheory which is, though, based on realist foundational claims. In perhaps the most explicit and concise statement of his philosophy of language, Geertz states that ‘culture is public because meaning is’ (1973a:12). In other words, Geertz’s subscribes to a traditional Western metaphysics of communication which regards as a fundamental requirement for communication the availability of a communal, shared linguistic code. However, this requirement cannot be coherently reconciled with a desire for endless, free interpretation because the postulation of a code implies the existence of rules, systems, determinate forms and meanings etc. which, in theory, actually do the interpretive work for us. The idea of endless interpretation due to textual indeterminacy of meaning cannot coherently be reconciled with a code-based view of language because a code is a concept which implies inherent determinacy. A code is a construct which contains its own interpretation. The code just has to be known. Or rather, to know the code is already to know its interpretation, irrespective of context.

Similar considerations can also be applied to Derrida’s famous theoretical slogan ‘there is nothing outside the text’ – again a theoretical statement which is methodologically motivated insofar as it is seen to license the endless proliferation of interpretative enterprises. However, if one generalises the implication of Derrida’s remark to all of language or even just to all written language, namely that interpretation of the intentions of the speaker or author are irrelevant to the determination of meaning, the conclusion is patently absurd. It is noticeable too that Derrida failed to live up to his own dictum and was notably upset, for instance, by John Searle’s supposed misunderstanding of what he meant in their renowned disagreement. Once again, it is important to point out that Derrida’s hermeneutic theory is based on a particular realist foundational claim, namely the iterability and citationality of linguistic signs, which supposedly lend them their intention-flouting character. Consequently, as with all theories, any refutation of Derrida’s theoretical enterprise should begin by addressing these metatheoretical issues.


After these mostly critical comments, I want to end by briefly outlining what might be considered a positive case against the attempted explanation of specific communicative acts by means of theory and this is tied up with a view of language as moral action and as individuals as moral agents.

Now to offer such account on principled grounds inevitably requires a particular metatheory. One of the very few theoretical generalisations that Integrational Linguistics makes about language and communication is that it is radically indeterminate. The paradoxical nature of this generalisation is that it is one which appears to preclude the possibility of any further ontological generalisations. As Pablé & Hutton (2015:18) note, the integrational position is that ‘there is nothing in human society, psychology or biology that determines language in general.’ They go on to say that ‘to argue that language is indeterminate is […] to make a general assertion about the absence of decontextual authority over what words mean’ (p.19). To this one might also add the absence of any decontextual authority over what actions mean. In other words, the only determinacy which can be achieved is the unique determinacy of a particular individual’s interpretation or evaluation in context. Furthermore, contexts do not submit to theoretical generalisation because:

to generalize is to assert that some fact holds regardless of context. But our grounds for making […] judgments as to the meaning, the significance, the purpose of an action are restricted to individual cases. For that is where meaning, significance and purpose reside.
(Louch 1966:215)

This might seem to suggest, as Wittgenstein held, that there is therefore no basis for explaining human action. However, this is to adhere to an inappropriate paradigm of explanation. Furthermore, it is belied by the simple observation that we usually do manage to explain people’s actions to our satisfaction on an ad hoc basis without recourse to generalities. Indeed, following Louch, I would claim that to identify and describe an action, linguistic or otherwise, is already to offer an explanation of it, albeit not a theoretical one. This is on account of the fact that determining what it is people are doing – the most basic question of social inquiry – is tied up with the recognition and attribution of their motives, purposes, intentions and desires etc. and assessing their performances in the light of these. It is, for example, to distinguish between deliberate, intentional acts and involuntary or unintentional movements of the human body, a distinction which is often obscured by the catch-all term ‘behaviour’. To identify an action as an action, i.e. as purposive, and then further as an action of a particular sort, is to recognise it via an act of individual, context-bound interpretation as a moral act and the individual performing it as a moral agent. When it comes to human actions, the fact/value distinction collapses and it becomes impossible even to talk about actions without recourse to evaluation and appraisal.

In his brilliant yet largely unheralded book Explanation & Human Action (1966), A.R. Louch makes the dramatic claim that ‘psychology and social science are moral science. Ethics and the study of human action are one.’ By extension and granted a view of language as action, one inference of this remark is that linguistic inquiry too might more meaningfully be conceived as a branch of ethics which places the unique, theory-defying individual at the centre of its concerns.


Bakhtin, M.M. 1996. The Problem of Speech Genres. Speech Genres & Other Late Essays. Eds. Caryl Emerson and Michael Holquist, 60-102. University of Texas Press. Austin.

Blum, Alan. 1973. Theorizing. Heinemann. London.

Coupland, Nikolas. 2016. Introduction. In N. Coupland (ed.) Sociolinguistics: Theoretical Debates, 1-34. Cambridge University Press. Cambridge.

Geertz, C. 1973a. Thick Description: Toward an Interpretive Theory of Culture. In C. Geertz, The Interpretation of Cultures, 3-30. New York. Perseus.

Geertz, C. 1973b. Deep Play: Notes on the Balinese Cockfight. In C. Geertz The Interpretation of Cultures, 412-453. New York. Perseus.

Gellner, Ernest. 1992. Postmodernism, Reason and Religion. Routledge. London & New York.

Louch, A.R. 1966. Explanation and Human Action. Blackwell. Oxford.

Love, Heather. 2013. Close Reading and Thin Description. Public Culture, 25 (3, 71): 401-434.

Miller, Elizabeth. 2014. The Language of Adult Immigrants: Agency in the Making. Multilingual Matters. Bristol.

Orman, Jon. 2017. Explanation and theory in linguistic inquiry. Empedocles: European Journal for the Philosophy of Communication, 8 (2): 167-188.

Pablé, Adrian & Chris Hutton. 2015. Signs, Meaning and Experience. Integrational Approaches to Linguistics and Semiotics. Mouton De Gruyter. Berlin.

Pavlenko, Aneta. 2007. Emotions and Multilingualism. Cambridge University Press. Cambridge.

Said, Edward. 2000. The Edward Said Reader. Eds. Moustafa Bayoumi and Andrew Rubin. Granta Books. London.

Sloterdijk, Peter. 2012. The Art of Philosophy. Wisdom as a Practice. Columbia University Press. New York.

Smedslund, Jan. 1979. Between the analytic and the arbitrary: a case study of psychological research. Scandinavian Journal of Psychology, 20: 1-12.

Voegelin, Eric. 1999. Modernity Without Restraint: The Political Religions, The New Science of Politics, and Science, Politics, and Gnosticism. University of Missouri Press. Columbia, MO.

Wittgenstein, Ludwig. 2009 (1953). Philosophical Investigations, eds. P.M.S. Hacker and J. Schulte. Wiley-Blackwell. Malden, MA.

Yoshida, Kei. 2014. Rationality and Cultural Interpretivism. Lexington Books. New York.

How to cite this post

Orman, Jon. 2018. Some reflections on the uses and abuses of theory in linguistic thought. History and Philosophy of the Language Sciences.

Tagged with: ,
Posted in Article, Linguistics, Philosophy

Leave a Reply

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

You are commenting using your 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: