6th INTERNATIONAL COGNITIVE LINGUISTICS CONFERENCE
STOCKHOLM, July 1999
THEME SESSION: `LANGUAGE AND IDEOLOGY'

Cognitive Linguistics and the Marxist approach to ideology


Peter E Jones, Sheffield Hallam University, Great Britain


1. Preamble: ideology and social theory

All contributors to the theme session on ideology, whatever their theoretical allegiance, would probably accept a view of ideology as having to do with the social function of ideas. That is, ideology has to do with the way in which ideas, expressed in language or some other medium, play a role in justifying and defending or, alternatively, in challenging and opposing existing economic, political or social structures. Assumptions and claims about the role of ideology therefore make sense only within the framework of a broader view of the way society as a whole works. For this we need a social theory within which the complex interconnections within the social whole between ideas (or `discourse', if you will, to use a more fashionable expression) and social activity can be identified and analysed.

Marxism is a social theory and provides one answer to the question of how ideas and other expressions of human imagination relate to action. It also provides a methodology for studying this relation in concrete circumstances. But where do non-Marxist approaches to the study of ideology, for example Cognitive Linguistics, stand on the question of social theory? And if the general premisses of the social theory espoused by CL conflict with those of Marxist theory, does that mean that there is no basis for dialogue between these two theoretical systems? This paper attempts to open a discussion on these issues. It begins with an exposition of the main principles of Marxist social theory and ideology, moves on to examine the social theory implied by CL work on ideology and, finally, attempts to explore the basis for productive engagement between the two approaches by considering to what extent the CL conception of `Idealized Cognitive Model' may allow insights into ideological formations in the Marxist sense.

2. Marxism and Cognitive Linguistics

The question of the relationship of CL to the Marxist tradition is an important one which has not, as yet, been addressed in the relevant literature. In fact, there are no references to Marxism, as philosophy or social theory, in the main works on the philosophy of CL (eg Johnson, 1987; Lakoff, 1987; and now Lakoff & Johnson, 1999). Nor is there any engagement with the long tradition of specifically materialist philosophy which was one of the contributory sources of Marxist materialism. It is rather an extraordinary fact that an approach which proclaims itself as `a challenge to Western thought' (Lakoff & Johnson, 1999) can overlook not only the whole spectrum of Marxist work in philosophy, social theory, history, the natural sciences, economics, and politics but also ignore that immense body of neurolinguistic, psycholinguistic and psychological work (on topics dear to the heart of contemporary `cognitive scientists') from within the Marxist-inspired Vygotskian and Activity Theory traditions. Mainstream western thinking, it appears, can acknowledge and celebrate Darwin's `dangerous idea' (to use Dennett's phrase) while Marx's idea seems to be simply too dangerous to mention.

The situation is all the more ironic in that those issues of central concern to the philosophy of `embodiment' have been in the forefront of debates within Marxism and between Marxism and other scientific currents for more than 150 years. The claim by Lakoff and Johnson (1980: 181) that their account of truth `can be considered an attempt to extend the realist tradition' by dealing with `social and personal reality as well as physical reality' comes nearly 155 years after the founders of Marxism first expounded their `materialistic conception of history', an attempt to extend their materialist philosophical outlook to the study of human social existence and its laws of development. The Marxist and CL solutions to such vital issues are hardly identical, it is true. But I submit that a dialogue between Marxism and CL would, for that very reason, be fruitful on a number of grounds. Firstly, Marxist (`dialectical') materialism does not share many of the features of the `objectivism' or `external realism' which are offensive to CL theorists. Thus, the `real premises' of historical materialism are `men, not in any fantastic isolation and rigidity, but in their actual, empirically perceptible process of development under definite conditions' (Marx and Engels, `The German Ideology' in Selected Works: 25). Secondly, Marxist theory developed through protracted polemical engagement with philosophical trends akin to CL's `experiential realism', whose proponents claimed, as do those of CL, that their position was a `third way' between materialism and idealism, escaping the `defects' of both extreme positions. One such trend, calling itself `empirio-criticism' or `empirio-monism', was taken up by the leading figures of Russian Marxism (including Plekhanov and Lenin) at the beginning of the century. The Marxists argued that philosophical views which take human `experience' as their epistemological foundation inevitably lead to idealist or anti-realist positions in the absence of a consistently materialist (or `objectivist') interpretation of experience. In the course of these arguments, materialist perspectives on such contentious issues as the possibility of objective truth, the relation between relative and absolute truth, and the relationship between sensation and concept, were elaborated which would repay serious study by CL philosophers.

Central to all the philosophical issues at stake between Marxism and CL is the question of objective truth, which is also at the heart of the Marxist treatment of ideology. On this pivotal question, Marxist theory and CL represent different, and to some degree, opposing philosophical orientations. In CL terms, Marxism is unacceptably `objectivist' while, from the Marxist point of view, CL, despite its claims to realism, occupies positions considered to be typical of relativism and idealism. Thus, from a Marxist perspective, the `experiential realism' of CL, like other `experience'-based philosophies, is not a consistent and coherent philosophical system but wavers between materialist and idealist premises, often giving priority to the latter. On the one hand, there is a commitment to `basic realism', ie `to the reality of a world existing independent of human beings' (Lakoff, 1987: 266), a commitment shared with `objectivism' (op.cit: 158-159). On the other hand, there is a denial of the possibility of objective truth or objective knowledge when `the mind reproduces the logical relations that exist objectively among the entities and categories in the world' (Lakoff, op.cit:163), a `God's eye' view' of truth. On the CL view: `Human concepts do not correspond to inherent properties of things but only to interactional properties [ie `experience', PEJ]' (Lakoff & Johnson, 1980: 181). Experience, then, though `constrained at every instant by the real world of which we are an inextricable part' (Lakoff, 1987: 263) does not provide us with concepts and theories which correspond to the properties and interconnections in the external world. These ideas underlie a CL variant of conceptual relativism (not `a total relativism', op.cit: 264) which entertains the `existence of alternative, incompatible conceptual schemes' (ibid) and in which `reality as we understand it is structured by our conceptual schemes' (op.cit: 262). On this view truth is not `absolute, objective truth' but truth `relative to understanding' (op.cit: 294).

However, the conjunction of basic realism with a denial of objective truth (in the above sense) involves a logical contradiction typical of contemporary neo-Kantianism. Basic realism in fact already presupposes a `God's eye view', since it asserts that the real world exists independently of human beings and therefore independently of all human experience. Who is there on the CL view - other than God - to speak of the existence (and the properties) of a world beyond all possible and actual human experience? The materialist perspective, on the other hand, proceeds consistently from the premisses of basic realism. Human experience proves to us that material reality exists, demonstrating that the `inherent properties' of material reality, including its existence outside of, prior to and independently of us, are in principle knowable and can be discovered through experience. The concept of `experience' itself needs critical re-evaulation in this connection. By `human experience' is meant first and foremost social practice, the practical transformation of the external word by the organized social collective. Experience is therefore itself an objective, material process subordinate to the laws of material reality existing independently of human experience.1 Such a view does not necessarily imply that there is `one true or correct description of a system of phenomena' as no attempt is made to prescribe a priori the form that objective knowledge must take. But objective knowledge is possible, since knowledge is the conscious form in which humans, as objective beings amongst others, relate to reality. Human beings, of course, have their own purposes and needs distinct from the natural world but these purposes are realized in that world through activity. The matter of nature, whose intrinsic properties are independent of and essentially indifferent to our purposes, is the source and material of human life-activity; in bringing this material into the service of our own needs we begin to discover its intrinsic, objective properties, thereby developing images, ideas, and ultimately concepts and systems of theoretical thinking in which the objective phenomena of material reality and their law-governed interconnections are represented. Practice itself is the test of correspondence between idea and reality.

Truth, however, is not a finished once-and-for-all-time state, but a process. A particular natural scientific theory, for example the theory of evolution, does not constitute the final and absolute truth of the matter. The correspondence between a scientific theory and the reality it depicts is always conditional, approximate and relative to the system of objective interactions revealed by historical practice. But if the theory is not absolutely true, neither is it absolutely false; if the cupboard is not full, neither is it bare. There is a growing kernel of truth within the theory of evolution to do with the facts, the processes and the mechanisms of development and differentiation of organic life - a kernel which will never be refuted. Truth and falsehood are dialectical `opposites' and must not be counterposed in a formal and mechanical way. A theory may therefore be true only within certain limits, but within those limits absolutely and objectively true.2 From such a standpoint, the CL perspective on relativity of knowledge and its arguments from the existence of `alternative conceptualizations' of objects cannot be considered convincing. Such phenomena merely provide evidence of the multiplicity and density of interactions and interconnections between all phenomena of nature. They demonstrate, in the terminology of `critical realism' (Bhaskar, 1979), `ontological depth'3 as well as the relative autonomy, the relatively independent scope for action, of different dimensions and aspects of the material whole.4

3. Ideology in Marxist social theory

Marx and Engels referred to their approach to the understanding of society as `the materialistic conception of history', thereby claiming that there was a way of approaching social structure and social change in a way consistent with materialist assumptions. I should stress that this approach is hardly a good example of what Lakoff and Johnson (1999) call `a priori philosophising', since it emerged through a critical rethinking of the history of philosophy informed by a close study of the natural sciences and from an enormous volume of empirical work in history, in economics and politics and, not least, from the experiences - successes and failures - of political struggle, guided by these ideas, on an international scale. Marx himself speaks of `the general result [my emphasis] at which I arrived and which, once won, served as a guiding thread for my studies' which he summarises briefly as follows:

`In the social production of their life, men enter into definite relations that are indispensable and independent of their will, relations of production which correspond to a definite stage of development of their material productive forces. The sum total of these relations of production constitutes the economic structure of society, the real foundation, on which rises a legal and political superstructure and to which correspond definite forms of social consciousness. The mode of production of material life conditions the social, political and intellectual life process in general. It is not the consciousness of men that determines their being, but, on the contrary, their social being that determines their consciousness' (Selected Works One: 503).

As Engels commented, `this apparently simple proposition, that the consciousness of men depends on their being and not vice versa, at once, and in its first consequences, runs directly counter to all idealism, even the most concealed. All traditional and customary outlooks on everything historical are negated by it. The whole traditional mode of political reasoning falls to the ground' (op.cit: 509).

The radical consequences of such a view for thinking in general flow from the fact that `men, who produce their social relations in accordance with their material productivity, also produce ideas, categories, that is to say, the abstract, ideal expressions of these same social relations' (op.cit: 524). And therefore:

`Morality, religion, metaphysics, all the rest of ideology and their corresponding forms of consciousness, thus no longer retain the semblance of independence. They have no history, no development; but men, developing their material production and their material intercourse, alter, along with this their real existence, their thinking and the products of their thinking' (op.cit:: 25).

For Marx and Engels the principle of the primacy of the economic within the social whole remained a fundamental tenet of their conception, although they ridiculed `the fatuous notion of the ideologists that because we deny an independent development to the various ideological spheres which play a part in history we also deny them any effect upon history' (Selected Works: 701). Thus:

`It is not the case that the economic basis is cause, is solely active and everything else is only a passive effect. Rather, there is an interaction which takes place upon the basis of the economic necessity which ultimately asserts itself' (Letters: 282).

`Political, legal, philosophical, religious, literary, artistic etc development is based upon economic development', Engels noted, but nevertheless these `all react upon each other and also upon the economic base' (ibid).

Essential to the Marxist treatment of ideology is the notion of `inversion' and the associated idea of `false consciousness'. While Marx, in line with his general theory, emphasised that the `abstraction, or idea...is nothing more than the theoretical expression of those material relations which are their lord and master' (Grundrisse: 164), it is not the case that this relation between ideas and material relations is transparent to the social actors whose life process is at issue. Rather, things appear to be quite the reverse: social relations appear to be the consequence of ideas with ideas as the motive force behind social development:, eg:
`the struggle beween the classes already existing and fighting with one another is reflected in the struggle between government and opposition, but likewise in inverted form, no longer directly but indirectly, not as a class struggle but as a fight for political principles, and so distorted that it has taken us thousands of years to get behind it' (Engels, Selected Works: 696).

In this way, the real movement of the social whole is `inverted' in ideology. Ideology `is a process accomplished by the so-called thinker consciously, it is true, but with a false consciousness. The real motive forces impelling him remain unknown to him; otherwise it simply would not be an ideological process. Hence he imagines false or seeming motive forces' (op.cit: 700).

Ideology, as a form of social consciousness, is not merely dismissed as deliberately concocted falsehood, although there is also plenty of such in circulation. Instead, the source of ideological notions is viewed as inextricably tied up with the narrow and historically limited scope of social practice itself. An ideological view, such as bourgeios ideology, is a view of society from the standpoint of a particular social class acting in accordance with its own interests. From this viewpoint, a series of categories of phenomena are observed and abstracted, eg `wages', `profit', `capital', `rent', etc in the region of political economy.These abstractions are not subjective illusions or fantasy. They `are formed not only in the consciousness of an individual of bourgeios society but in the reality itself of the economic social relations which he contemplates' (Ilyenkov, 1982: 127). In other words, these phenomena are real - they are `there' on the surface of the actual social process itself and observable to all: `The things given in contemplation to an individual of bourgeios ("civicä) society are superficially exactly the way they seem to him' (ibid)5. The ideological work is to be found in the account and interpretation of these phenomena and their relationships, specifically in their mystification where the `mechanism of mystification consists in the collapsing of social facts into natural ones' (Geras, 1990: 216) thereby giving them `an idealistic explanation' (Geras, op.cit: 218). For Marx, such categories constitute merely the `outward appearance' of the workings of the economic system. They cannot be accepted uncritically as the categories out of which a science of political economy should be built, even though they may offer the starting point for analysis: `That in their appearance things often represent themselves in inverted form is pretty well known in every science except Political Economy' (Marx, in Geras, op.cit: 208). Thus, `vulgar economy', as Marx puts it, `everywhere sticks to appearances in opposition to the law which regulates and explains them' (in Geras, op.cit: 207) and consequently merely consitutes a theoretical apologia or rationalization for the existing economic forms.

In the case of political economy, then, ideology is the reflection in ideas of the material interests of a ruling class, a reflection in which the outward appearances of the economic forms in which its own interests are expressed are seen and presented in mystified fashion as naturalized, as the product of `human nature' (in our genes, perhaps), as eternally valid, universal `civilized values'. This viewpoint `is of course consolidated, nourished and inculcated by the ruling classes by all means available' (Grundrisse: 165), by the vulgar fetishisation of the immediate forms of appearance of economic processes, but also by more sophisticated `philosopohical' attacks on the very possibility of scientific knowledge capable of penetrating appearances to get at the inner interconnections within the system.

From the Marxian standpoint, a materialist science of society is premissed on `the necessity of constructing reality against appearances' (Geras, op.cit: 209), a theoretical viewpoint which coincides with the practice of that social class exploited by capital and therefore struggling against it. The result of Marx's empirical economic analysis was a system of theoretical categories expressing the `inner nature of capital' (Marx in Geras, ibid) - use-value, exchange-value, labour power, labour, surplus value, capital etc - through which the law-governed movement of the economic system as a whole, including those forms in which the workings of the system manifest themselves to immediate experience, could be theoretically reproduced.

4. Cognitive Linguistics and social theory

What social theory informs Cognitive Linguistic treatments of ideology? To my knowledge, no detailed exposition of a social theory exists, but it is possible to infer one from claims made in the CL literature. Lakoff, for example, argues:

`Governments are real. They exist. But they exist only because human beings conceived of them and have acted according to that conceptualization. In short, the imaginative products of the human mind play an enormous role in the creation of reality...In the case of social and cultural reality, epistemology precedes metaphysics, since human beings have the power to create social institutions and make them real by virtue of their actions' (1987: 208).

This view, in the absence of further explanation or contextualization, implies that in the overall dynamic of the social process it is human conceptualization which is primary: ideas (`epistemology') guide actions which produce social structures (`metaphysics'): social consciousness determines social being, the inverse of the Marxist proposition. This implicit social theory appears to be common to many CL discussions of ideology. Bruce Hawkins (`Ideology, metaphor and iconographic reference'), arguing that `making sense of the sociopolitical phenomenon of oppression is a task for the historian and the political scientist' sees the role of `cognitive scientist' in the analysis of ideology as involving `see[ing oppression as a conceptualized social order imposed upon a particular sociopolitical formation. That social order emerges from a belief system in the mind of the oppressor'. The `sociopolitical phenomenon' of oppression, he claims, `has its roots in the mind of the oppressor'. Similarly, Willem Botha (`The deictic foundation of ideology with reference to African Renaissance') argues that ideology `emanates from a person's (group of persons) cognitive system', and Harry Howard (`Does the central nervous system impose biased representations on cognition?') sees ideological biases in the expression of gender relations as `the result of a vector-space representation that the central nervous system imposes on cognition'. The social theory implicit in such claims is itself ideological, in Marxist terms, precisely because it turns upside down the relationship between ideas and social reality, seeing in the former the cause or source of the latter and, furthermore, in some cases leading to an ahistorical and naturalistic view of ideas as the product of the body or brain independently of social circumstances. This view is in fact characteristic of many schools of thought whose primary focus is the study of ideology by linguistic means (eg Teun van Dijk's `Critical Discourse Analysis').

Nevertheless, despite these clear differences, there is, arguably, sufficient congruence and complementarity between the two traditions to allow a productive engagement. For CL, human experience in the world is the source and motivation for `imaging' and thinking processes, rather than innate biological mechanisms. CL makes the claim that cognition is not based on language, but rather, `language is...based on cognition' and `depends upon the nature of thought' (Lakoff, 1987: 291). It follows that language cannot constitute a barrier, boundary or limit to human thinking (as is the case with some idealist philosophical trends), although the distinctive properties and processes of human language are themselves a constituent of human cognitive activity and influence and shape cognitive processes in many ways. These propositions are, within limits, commensurable with a Marxist perspective in which human cognitive activity, whether embodied in language, artistic images or some other mode, is viewed as a form of `ideal' (Ilyenkov, 1977) or conceptual `modelling' of the natural and social worlds which arises from and is inseparably connected with human activity in the world.
Similarly, there may well be common ground in the understanding and treatment of metaphor, although there are clearly different perspectives on the significance of metaphor for an understanding of cognition. This is not the place for a detailed critical engagement with CL views on this. The Marxist tradition, unlike forms of `objectivism' attacked in Lakoff (1987) and Lakoff and Johnson (1999), has always recognised, and emphasised, the cognitive function of imagination and, therefore, of metaphor (eg Farman, 1994). To acknowledge the role of imagination in the cognitive process is not to imply that what is thought, if metaphorical or metaphorically expressed, is ipso facto imaginary in the subjectivist sense, or that such forms of thinking `do not mirror nature' (Lakoff, 1987: 371).

However, even within the `philosophy of embodiment' there appears to have been a subtle shift in emphasis on the role of metaphor as a cognitive instrument. Thus, Lakoff and Johnson (1999) repeatedly stress that philosophy and science cannot do without metaphor and, consequently, to `set out the defining metaphors of a philosophy is not necessarily to critique it....Identifying philosophers' metaphors does not belittle them. (1999: 542-543). Justification for this attitude is based on a notion of `aptness' for metaphors in the sense that `metaphoric theories can have literal, basic-level entailments' (op.cit::91), `they can entail non-metaphorical predictions that can be verified or falsified' (ibid, my emphasis). In other words, metaphor is not simply a relation between cognitive elements (construed in the broadest sense) but a more complex phenomenon to do with the relationship between theory (the cognitive) and the reality studied and transformed by scientific practice on the other. Here the emphasis shifts from the link between metaphorically expressed ideas and `bodily experience' to the theoretical soundness of such ideas in relation to `converging evidence' from as many independent domains of scientific investigation as possible.6 This emphasis now brings us close to the view of metaphor embraced by more conventional forms of scientific realism as for instance, Harr_:

`The process by which originally metaphorical descriptions are subsequently shown, by "ontological experimentä, to constitute accurate factual accounts of how nature works is characteristic of progress in the natural sciences' (1961, quoted in MÿhlhSusler, 1995: 281).

Furthermore, perhaps the most significant point, for our purposes, is made in connection with such formulations as `a gene for aggression' in science writing. In their discussion of such cases, Lakoff and Johnson argue: (1999: 217):

`There is, of course, a difference between conceptualizing and reasoning about the world according to this metaphor and actually believing that the metaphor is a truth'.

If one knows that the metaphor which one is using for `conceptualizing and reasoning' is untrue then this implies a critical distance between the thinking process in general and the linguistic (semantic-metaphorical) structure serving as the vehicle of thinking. From a materialist point of view this means that one must not equate or directly identify the semantic network, including the metaphorical structure of concepts, with cognition, ie the reasoning process itself through which new knowledge develops.7

It may well be the case, then, that there is more common ground between the Marxist and CL views of cognition than is apparent at first sight. There is some reason to believe that a collaboration between the two might be possible in the area of analysis of ideology. Indeed I will suggest that there is a space for productive collaboration in investigating what Engels referred to as the `concept-material', or in other words, the semantic structures and mechanisms, through which ideologies are expressed and developed. Engels argued that:

`Every ideology, however, once it has arisen, develops in connection with the given concept-material, and develops this material further; otherwise it would not be an ideology, that is, occupation with thoughts as with independent entities, developing independently and subject only to their own laws' (Selected Works: 628).

And, further:

`The historical ideologist (historical is here simply meant to comprise the political, juridical, philosophical, theological - in short, all the spheres belonging to society and not only to nature) thus possesses in every sphere of science material which has formed itself independently out of the thought of previous generations and has gone through its own independent course of development in the brains of those successive generations' (op.cit: 700).

Consequently, while Marxism and CL may well disagree on how to explain the social function of ideas - ie the role of ideology, properly speaking - there may be scope for dialogue about the internal, semantic and conceptual side of this `concept-material'. The distinction made Purvis and Hunt (1993: 476) between `discourse' and `ideology' may also be useful here:

`If "discourseä and "ideologyä both figure in accounts of the general field of social action mediated by communicative practices, then "discourseä focuses upon the internal features of those practices, in particular their linguistic and semiotic dimensions. On the other hand, "ideology directs attention towards the external aspects of focusing on the way in which lived experience is connected to notions of interest and position that are in principle distinguishable from lived experience'.

Thus, the concepts and methodology of CL may be usefully applied to the internal semantic systems of ideological discourse, supplementing and concretising the Marxist analysis of ideologies in terms of historically specific relations between social being and social consciousness.


5. Theory and ideology in the `concept-material' of economic discourse

From a discussion of general philosophical and theoretical positions we move to the examination of a particular conceptual system, specifically a theoretical model of economic processes developed in post-modern cultural theory in Baudrillard (1981) and taken up by proponents of `Critical Discourse Analysis' (eg Fairclough, 1997). This model, ostensibly grounded in an attempt to extend or `correct' basic concepts and categories of Marxist economic theory based on claims about changes in the nature of commodities, in effect amounts to a repudiation of that theory. These alleged changes, it is argued, involve a transformation of the relations between economic and ideological phenomena resulting in a dynamic of social processes incompatible with the classical Marxist insistence on the primacy of the economic within social evolution. Within this revised conception, the ideological realm, with discourse as its most important vehicle, assumes a much more important, and possibly preeminent, role as agency of social control and social change.

Baudrillard explicity develops a critique of historical materialism, rejecting the distinction between `base' and `superstructure' and, with it, the primacy of material production with respect to ideology. Criticising `the artificial distinction between the economic and the ideological' (op.cit:143), he argues that ideology is not something which expresses a primary, independent realm of material economic processes but, instead, `ideology is actually that very form that traverses both the production of signs and material production' (ibid). Ideology, Baudrillard claims, lies `in the logic of the commodity' and consequently:

`Ideology can no longer be understood as an infra-superstructural relation between a material production (system and relations of production) and a production of signs (culture, etc), which expresses and masks the contradictions at the "baseä. Henceforth, all of this comprises, with the same degree of objectivity, a general political economy (its critique), which is traversed throughout by the same form and administered by the same logic' (ibid).

The basis for this view, as Baudrillard makes explicit, is an identification of `the logic of the commodity' with `the internal logic of the sign', ie with `the relation of [signifier] to [signified]' (ibid). Baudrillard's arguments are based on equating the commodity as a unity of use-value and exchange value (in Marxian terms), with the sign as a unity of signifier and signified. Like the commodity, he claims, the sign `can function as exchange value (the discourse of communication)' and `as use value (rational decoding and distinctive social use' (op.cit:146). But it is not just that the sign is like a commodity; the commodity works like a sign: `Like the sign form, the commodity is a code managing the exchange of value. It makes little difference whether the contents of material production or the immaterial contents of signification are involved, it is this code that is determinant: the rules of the interplay of signifier and exchange value' (ibid).

Let us take a step back and look more closely at Marx's concept of the commodity. This concept forms part of an entire theoretical system or model of the inner dynamic of capitalist production and it is, ultimately, with respect to this system that Marx's concept of the commodity must be understood.8 Suppose we take this concept as an `Idealized Cognitive Model' (following Lakoff, 1987). This ICM is a `propositional model' (op.cit:: 113). Commodities are defined as things which `have a dual nature, because they are at the same time objects of utility and bearers of value' (Capital: 138). The commodity, then, is a unity of use-value (UV) and exchange-value (EV). Roughly, UV is whatever the commodity is used for, EV how much it is worth. EV (or `value' for short) is so called because, in its simplest form, it is realised through the actual act of exchanging one commodity for another: person X with commodity A exchanges A for commodity B belonging to person Y (and, of course, vice versa). While UV is based on the actual physical properties of the commodity, EV is determined by the socially necessary labour time involved in its production. Exchange value is a historically specific, concrete form of appearance of economic relations (of relations between people). It is this form, expanded and generalized to all the constituent elements of productive activity, which constitutes the capitalist mode of production.

Is the relation between UV and EV a semiotic relationship as claimed by Baudrillard? There are, certainly, implications in Marx's analysis, and in his discussion of `the language of commodities' (Capital:143), for a `Marxist semiotics' (cf Jones, 1991). Ilyenkov, for example, argues that, in Capital the "dialectic of the transformation of a thing into a symbol, and of a symbol into a token, is...traced...on the example of the origin and evolution of the money form of valueä (1977: 273). However, it is quite wrong, as I shall attempt to demonstrate, to identify the relation between UV and EV with that of signifier to signified.

When commodities are exchanged, it is the (`socially necessary') time it took to make them which determines the proportions in which they are exchanged. In the act of exchange, when A is exchanged for B, one could say with justification that B represents the value of A, and vice versa. Each commodity, at the same time, acts as representative of the value of the other. As Marx puts it:

`By means of the value-relation, therefore, the natural form of commodity B becomes the value-form of commodity A, in other words the physical body of commodity B becomes a mirror for the value of commodity A' (Capital:144).

But this `representation' is complex. When A and B are exchanged, then each is actually the measure of value of the other. If, for example, one pair of boots exchanges for (`is worth') one sack of corn (and vice versa), then the sack of corn supplies the measure of value of the pair of boots (and vice versa). B represents the value of A, then, only in so far as B itself embodies an amount of socially necessary labour time which is the equivalent of the amount contained in A. It is their value equivalence that both makes possible and is expressed or realised in the exchange itself. The dual nature of the commodity, as a use-value and a value, gives it a mysterious and contradictory character which arises not from the workings of human imagination as such but from the very form that the social production of wealth assumes under capitalism:

`Not an atom of matter enters into the objectivity of commodities as values; in this it is the direct opposite of the coarsely sensuous objectivity of commodities as physical objects. We may twist and turn a single commodity as we wish; it remains impossible to grasp it as a thing possessing value. However, let us remember that commodities possess an objective character as values only in so far as they are all expressions of an identical social substance, human labour, that their objective character as values is therefore purely social. From this it follows self-evidently that it can only appear in the social relation between commodity and commodity' (Capital: 138-139).

Marx himself likens this `social relation between commodity and commodity' in which one appears as the measure of value of the other not to the semiotic bond within the sign9 but to the way in which some objects are used to measure the physical aspects, eg the weight, of others:

`A sugar-loaf, because it is a body, is heavy and therefore possesses weight; but we can neither take a look at this weight nor touch it. We then take various pieces of iron, whose weight has been determined beforehand. The bodily form of the iron, considered for itself, is no more the form of appearance of weight than is the sugar-loaf. Nevertheless, in order to express the sugar-loaf as a weight, we put it into a relation of weight with the iron. In this relation, the iron counts as a body representing nothing but weight. Quantities of iron therefore serve to measure the weight of the sugar, and represent, in relation to the sugar-loaf, weight in its pure form, the form of manifestation of weight. This part is played by the iron only within this relation, i.e. within the relation into which the sugar, or any other body whose weight is to be found, enters with the iron. If both objects lacked weight, they could not enter into this relation, hence the one could not serve to express the weight of the other.' (op.cit:148-9)

At this point, however, `the analogy ceases' since in `the expression of the weight of the sugar-loaf, the iron represents a natural property common to both bodies, their weight' whereas in the expression of value commodity B `represents a supra-natural property: their value, which is something purely social' (ibid).

Marx goes on to show how the `value form' of commodities undergoes a series of metamorphoses with the expansion and intensification of commodity production, resulting in the money form of value wherein one commodity, namely gold, takes on the function of universal equivalent and acts simultaneously both as measure and as general representative or symbol of the value of all other commodities. Significantly, however, Marx himself warns against spurious analogies between money and other phenomena, including language:

`(To compare money with blood - the term circulation gave occasion for this - is about as correct as Menenius Agrippa's comparison between the patricians and the stomach.) (To compare money with language is not less erroneous. Language does not transfrom ideas, so that the peculiarity of ideas is dissolved and their social character runs alongside them as a separate entity, like prices alongside commodities. Ideas do not exist separately from language. Ideas which have first to be translated out of their mother tongue into a foreign language in order to circulate, in order to become exchangeable, offer a somewhat better analogy; but the analogy then lies not in language, but in the foreignness of language.)' (Grundrisse: 162-163).

Let us now return to Baudrillard's conceptual reworking of the commodity as a sign (and vice versa). The sign, he argues `can function as exchange value (the discourse of communication)' and as use value (rational decoding and distinctive social use)'. Here the term `exchange value' expresses a metaphorical mapping of the meaning of `exchange' within the propositional model of the source domain to the target domain of communicative `exchange' using signs. It is this first mapping which allows Baudrillard to apply terms proper to economic phenomena to communication, eg in the very title of Baudrillard (1981) - `the political economy of the sign'. This also licences the application of the term `use value' to the processes of understanding and interpreting signs. Furthermore, once the metaphorical sense of `exchange' has been applied to semiotic processes, it can then be turned back onto the commodity itself, which is seen `like the sign form' as `a code managing the exchange of values' (op.cit:146). In fact, the commodity can now be presented as a particular instance of the general field of signs with the linguistic sign, perhaps, as the `prototypical case'. This gives us a `metonymic model' of the category of semiotic phenomena (Lakoff, 1987: 84) where the `target' concept (the commodity) is presented in terms of another concept (the sign) which comes to stand for the whole category. In this way a complex conceptual inversion has taken place: commodity and sign have been equated via a metaphorical mapping from economics to communication and on that basis the commodity has been re-conceptualized as a sign. Thus, the concreteness of both phenomena (commodity and sign) has been dissolved in a particular type of abstraction represented by the new meaning given to the word `exchange', thus: `A critique of general political economy (or a critical theory of value) and a theory of symbolic exchange are one and the same thing' (op.cit: 128, my emphasis). This semantic shift in effect succeeds in destroying the concept of commodity. The crucial theoretical (and real world) distinction between economic and communicative phenomena, between material productive and semiotic processes, is erased. Let us explore this distinction in more detail.

Commodities are literally exchanged; they change hands (to use a more basic, bodily metaphor). Signs, eg words, do not. When person X exchanges a pair of boots for a sack of corn, X is no longer in possession of the boots which now belong to person Y. If person X `exchanges' greetings with person Y, these persons have not given up possession or use of these forms of greeting. In this metaphorical use of `exchange', an aspect of the form of the concrete phenomenon - the way in which commodities mutually `represent' to one another their common social substance in the patterned reciprocity of the act of exchange - has been abstracted while its specific, genuine economic content (labour-time) has been lost. A commodity is a use-value and an exchange-value; but its exchange-value does not stand for or represent its use-value, nor does its use-value stand for or represent its value. Its value, rather, represents the (abstract) socially necessary labour time the commodity embodies. By contrast, signs in general have no exchange value; they are not produced as commodities (but see below). Indeed, they are not `produced' at all in the economic sense. They `exchange' as a function of their meaning, not as a function of the time (or physical or mental effort) involved in their `production'. For the same reasons, signs (or their signifieds) are not use-values, since they are not `consumed' in the economic sense (although see below); they do not get worn out, or used up etc. The equation of the commodity with the sign is, therefore, entirely spurious since the metaphorical mapping of EV and UV carries the `concept-material' expressed in those terms beyond the legitimate bounds of the theoretical system (the source ICM) in which they have their precise and strict application. The corollary is that the concepts of EV and UV, by being applied to signifying activity in general, assume a supra-historical dimension, rather than being seen as valid only for a historically specific mode of production.

Baudrillard's arguments also figure in the work of such Critical Discourse Analysts as Norman Fairclough, who makes the following claims (cf Jones,1998 for detailed discussion):


The specific `concept-material' (ICM) of Marxian economic theory, and its materialist foundations, is again challenged by this claim about `a shift in the nature of commodities' (my emphasis), about their becoming increasingly `linguistic in nature' (my emphasis), resulting in the `linguistic penetration' of the economy. These arguments, stemming essentially from Baudrillard (1981, cited by Fairclough, op.cit) mistakenly imply that a shift in the type of commodities being produced can have an effect on the nature of commodities in general, culminating in the breakdown of the distinction between economic and semiotic processes. Fairclough implies that the proliferation of commodities which are `linguistic in nature', such as a textbook (op.cit: 8), somehow subverts the intrinsic logic of economic production and knocks it onto a new, `semiotic' track . But this is in fact a misconception. Whether a commodity is a `semiotic' object like a textbook, or a non-semiotic object like a carrot is irrelevant to its nature as a commodity. As commodities, the textbook and the carrot are identical despite their differences as use-values. As Marx emphasises in relation to use-value:

`its particular content...was completely irrelevant to the definition of the commodity. The article destined to be a commodity, and hence the incarnation of exchange-value, had to gratify some social want or other, and had therefore to possess some useful qualities. Voil tout' (Capital: 979).

The textbook, like any other commodity, is a unity of exchange-value (represented in its price) and use value (it is written in order to be read). Its value is determined by the quantity of socially necessary labour time involved in its production and its use-value, though `linguistic' or `semiotic', is `conditioned by the physical properties of the commodity, and has no existence apart from the latter' (Capital:126). Therefore, a sign or system of signs (eg a textbook) may indeed be a commodity, but its nature as a sign (ie the entire sum of signifier/signified relations making up its sign nature) has to do with its distinctive use-value as a commodity and not with its nature as a commodity. Consequently, however much the production of textbooks, films, TV programmes etc expands, in all this there is absolutely no shift in the nature of commodities or in the economic logic according to which the production of the means of life under capitalism moves.

6. Conclusion and prospects

This paper has attempted to explore the relationship between a Marxist and a CL approach to the study of ideology. While the two approaches differ, arguably irreconcilably, on key philosophical and theoretical issues having to do with the source and role of ideas within the social process, there is, possibly, scope for meaningful dialogue on the analysis of the `concept-material' from which theories and conceptual models (ICMs) are constructed. As illustration, the conceptual framework of Baudrillard's `political economy of the sign' has been analysed in relation to that of Marx's economic theory. The Baudrillard model relates to the Marxist ICM through the appropriation of key elements of the terminology of the latter but the `concept-material' of the Marxian theoretical system is altered by the application of metaphorical and metonymic processes to key concepts. This results in a model of a general domain of semiosis or ideology in which the distinction between economic and ideological production is erased and which is, therefore, conceptually incompatible with the source ICM and with the philosophically materialist world view within which the latter is developed. Thus, within the limits discussed, the concepts and methods of CL, despite differences in philosophical and theoretical orientation, can provide useful, auxiliary conceptual tools for ideological analysis and critique.

According to the argument presented here, the semantic `stretching' of the Marxist concepts takes them beyond their legitimate scientific limits. This raises important issues to do with the relationship between the conceptual and the semantic, between cognition (in the materialist philosophical sense) and such processes as metaphor, metonymy etc. Clearly, in some cases metaphor may help us to achieve crucial new insights into a phenomenon and in others it may undermine the theoretical and scientific integrity of a conceptual system. We have already noted the CL view that the identification of metaphors within conceptual systems does not by itself constitute a critique of those systems. Indeed, it would appear valid to emphasise a more general principle: the existence of `metaphorical' concepts within a theory tells us nothing about the adequacy of that theory. The role of metaphor in theory construction or in cognition more generally depends on its `aptness', which in turn depends on the explanatory power (the truth) of the entire conceptual system within which it functions. Where Marxism and CL obviously differ in principle is in their understanding of the source and process of development of such systems. On the Marxist view, these systems develop not from bodily experience understood naturalistically, but from social practice - the specifically human mode of life activity based on social production.

References

Abbreviations

Selected Works: K Marx & F Engels, Selected Works, Vol 1, Moscow: Progress, 1969
Selected Works One: K Marx & F Engels, Selected Works in three volumes, Volume One, Moscow: Progress, 1969
Capital: K Marx, Capital Volume 1, Harmondsworth: Penguin Books, 1976
Grundrisse: K Marx, Grundrisse, Harmondsworth: Penguin Books, 1973
Letters: K Marx & F Engels, Letters on `Capital', London: New Park, 1983


Baudrillard, J (1981) For a Critique of the Political Economy of the Sign, Telos Press
Bhaskar, R (1979) The Possibility of Naturalism: A Philosophical Critique of the Contemporary Human Sciences, Brighton: Harvester
Bohm, D (1957) Causality and Chance in Modern Physics, London: Routledge
Farman, I P (1994) Imagination in the structure of cognition [in Russian], Moscow: IFRAN
Fairclough, N (1992) Discourse and Social Change, Polity Press
Fairclough, N (1997) `Discourse across disciplines: discourse analysis in researching social change', AILA Review, 12: `Applied Linguistics Across Disciplines', 3-17
Geras, N (1990) `Essence and Appearance: Aspects of Fetishism in Marx's Capital', in B Jessop & C Malcolm-Brown (eds) Karl Marx's Social and Political Thought: Critical Assessments, London: Routledge
Harr_, R (1986) Varieties of Realism: A Rationale for the Natural Sciences, Oxford: Blackwell
Harr_, R (1961) Theories and Things, London: Sheed & Ward
Ilyenkov, E V (1977) `On the concept of the "idealä', in Problems of Dialectical Materialism, Progress
Ilyenkov, E V (1982) The Dialectics of the Abstract and the Concrete in Marx's `Capital', Moscow: Progress
Ilyenkov, E V (1997) The Dialectics of the Abstract and the Concrete in Scientific-theoretical Thinking [in Russian], Moscow: Rosspen
Jones, P E (1991) Marxism, Materialism and Language Structure: Basic Principles, Sheffield: Pavic
Jones, P E (1998) `Critical Discourse Analysis as Social Theory', in Future Perfect?, Proceedings of the Association of Media, Communications, and Cultural Studies Conference, Sheffield 1997, Sheffield Hallam University
Lektorsky, V A (1984) Subject, Object, Cognition, Moscow: Progress
Lakoff, G (1987) Women, Fire, and Dangerous Things, University of Chicago Press
Lakoff, G & Johnson, M (1980) Metaphors We Live By, University of Chicago Press
Lakoff, G & Johnson, M (1999) Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought, New York: Basic Books
Johnson, M (1987) The Body in the Mind, University of Chicago Press
MÿhlhSusler, P (1995) `Metaphors others live by', Language and Communication, 15 (3): 281-288

Endnotes

1 Cf Lektorsky's (1984: 94) critique of Husserlian philosophy: `The attempt to place the subject in the "centreä of the cosmos and to deduce the objectiveness of the world from the characteristics of the individual subject was not a success, for the subject proves to be included in a certain system of objective dependencies from the very outset'.
2 Cf Bohm (1957: 31-32): `To pursue our analogy further, we may say that with regard to the totality of natural laws we never have enough views and cross-sections to give us a complete understanding of this totality. But as science progresses, and new theories are developed, we obtain more and more views from different sides, views that are more comprehensive, views that are more detailed, etc. Each particular theory or explanation of a given set of phenomena will then have a limited domain of validity and will be adequate only in a limited context and under limited conditions. This means that any theory extrapolated to an arbitrary context and to arbitrary conditions will (like the partial views of our object) lead to erroneous predictions. The finding of such errors is one of the most important means of making progress in science. A new theory, to which the discovery of such errors will eventually give rise, does not, however, invalidate the older theories. Rather, by permitting the treatment of a broader domain of phenomena, it corrects the older theories in the domain in which they are inadequate and, in so doing, it helps define the conditions under which they are valid (e.g. as the theory of relativity corrected Newton's laws of motion, and thus helped to define the conditions of validity of Newton's laws as those in which the velocity is small compared with that of light). Thus, we do not expect that any causal relationships will represent absolute truths; for to do this, they would have to apply without approximation and unconditionally.'
3 `In fact one finds in science a characteristic pattern of description, explanation and redescription of the phenomena identified at any one level of reality. But only a concept of ontological depth (depending upon the concept of real strata apart from our knowledge of strata) enables us to reconcile the twin aspects of scientific development, viz. growth and change....Moreover, only the concept of ontological depth can reveal the actual historical stratification of the sciences as anything other than an accident. For this can now be seen as grounded in the multi-tiered stratification of reality, and the consequent logic - of discovery - that stratification imposes on science' (Bhaskar, op.cit: 16).
4 `First of all, our basic starting-point in studying the laws of nature was to consider the processes by which any one thing comes from other things in the past and helps to give rise to still other things in the future. Now this process cannot be studies in its totality which is inexhaustible, both in its quantitative aspects and in the complexity of its details. However, it is a fact, verified by human experience transmitted through our general culture since even before the beginnings of civilization, as well as by the experience of many generations of scientists, that parts of the processes described above can be studied approximately, under specified conditions, and in limited contexts. This is possible because there is an objective but approximate autonomy in the behaviour of these various parts of the process relative to any particular context (Bohm, op.cit: 29).
5 Compare Geras (op.cit: 217) `If then the social agents experience capitalist society as something other than it really is, this is fundamentally because capitalist society presents itself as something other than it really is'.
6 The following comment by Lakoff and Johnson (1999: 89) appears unexceptional from a materialist standpoint: `Many scientific results are stable...This is also true of the science of the mind. We are not likely to discover that there are no neurons or neurotransmitters...We maintain that they deserve to be called "resultsä because of all the converging evidence supporting them. The existence of so many forms of convergent evidence demonstrates that what we take as specific results are not merely the consequences of assumptions underlying a particular method of enquiry.'
7 Thus the following comment by Lakoff and Johnson (op.cit: 104)on the use of metaphor within `cognitive science' appears quite unproblematic from a materialist point of view: `When we speak of "neural circuitryä, we are, of course, using an important metaphor to conceptualize neural structure in electronic terms. The circuitry metaphor is used by the neuroscience community at large and is taken as providing crucially important insights into the behaviour of the brain. "Truthsä about the neural level are commonly stated in terms of this metaphor. We mention this because the neural level is seen quite properly as a "physicalä level, and yet much of what we take as true about it is stated in terms of the metaphor of neural circuitry, which abstracts away from ion channels and glial cells'.
8 `One can always come to an arrangement or agreement over the sense or meaning of a term; but things stand quite differently in relation to the content of a concept. Although the content of a concept is directly disclosed as the `meaning of a term', these are definitely not one and the same thing' (Ilyenkov, 1997: 404).
9 Marx specifically comments on the relation between commodities and symbols in the following passage (amongst others): `In this sense every commodity is a symbol, since, as value, it is only the material shell of the human labour expended on it. But if it is declared that the social characteristics assumed by material objects, or the material characteristics assumed by the social determinations of labour on the basis of a definite mode of production, are mere symbols, then it is also declared, at the same time, that these characteristics are the arbitrary product of human reflection' (Capital: 186-187).