Chapter    3
Structuralism, semiology and popular culture
■    Structural linguistics and the ideas
of Saussure                                            89
■    Structuralism, culture and myth            94
■    Structuralism and James Bond            102
■    Barthes, semiology and popular culture                                                 108
■    Barthes, structuralism and semiology  109
■     Writing Degree Zero                           110
■    Myths and popular culture                  112
■    Bourgeois men and women novelists   116
■    Structuralism and semiology: some
key problems                                       119
■    Lévi-Strauss's structuralism                 120
■    Roland Barthes's semiology                 123
87
In this chapter i shall discuss structuralism and semiology, both of which, since their rise to prominence from the 1950s, have had an important influence upon the study of popular culture. They have left their mark on other seemingly distinct perspectives like feminism and Marxism, and their concepts and procedures -such as signs, signifiers, signifieds and decoding - have continued to be used in the analysis of popular culture. Unlike mass culture theory or the Frankfurt School, their legacy appears to be more assured and wide-ranging. In view of the way they emerged alongside the increasing concern with theory in social science in the 1960s, and the fact that they express an interest in societies increasingly inundated with popular culture images and representations, it may even prove to be premature to talk about their legacy since their role in the social analysis of culture is still very evident. I shall thus try to outline their basic ideas, and indicate some of their empirical applications and critical drawbacks. I shall also look at the major contentions of structural linguistics in order to show how they represent one of the sources for these approaches. But before doing this, there is one problem which has to be confronted, namely the difference between semiology and structuralism.
If one reads the literature extensively, it becomes apparent that these terms are often used interchangeably. So it might be possible to argue that they mean the same thing, and that there is no problem. However, things are not quite so simple. Structuralism has been defined as a theoretical and philosophical framework which is relevant to the social sciences as a whole. It is said to make general claims about the universal, causal character of structures. Semiology, in turn, has been defined as the scientific study of sign systems, like cultures. For example, the Fontana Dictionary of Modern Thought defines structuralism as 'a movement characterized by a preoccupation not simply with structures but with
88
STRUCTURALISM  AND   SEMIOLOGY
such  structures  as can be  held to underlie and  generate the phenomena that come under observation . . . with deep structures rather than surface structures . . . referable [according to Lévi-Strauss] to basic characteristics of the mind', while semiology is viewed as 'the general (if tentative) science of signs: systems of signification, means by which human beings - individually or in groups - communicate or attempt to communicate by signal: gestures, advertisements, language itself, food, objects, clothes, music, and  the many other things that qualify'  (Bullock and Stallybrass:   1977; pp.  566  and  607).  Structuralism has been identified with claims about the universal character of mental and cultural structures, and their causal effects in giving rise to observable social phenomena. Neither of these claims need be entailed in the adoption of a semiological approach. This is roughiy the usage I will keep to in this chapter. I realise it is not totally satisfactory since it could be argued that, in practice, structuralism and semiology have studied the same kinds of things in the same kinds of ways. Any distinction must therefore remain somewhat arbitrary, but perhaps this is better than a completely relativistic standpoint. It also allows me to suggest that semiology can be used as a method which does not carry the universalising implications of structuralism.
Structural linguistics and the ideas of Saussure
The Swiss linguist Saussure (1857-1913) attempted to establish and develop the discipline of structural linguistics, and on this basis suggested it was possible to found a science of signs.1 In these respects, his ideas played a crucial role in the emergence of structuralism and semiology, and make their intentions and methods a lot clearer, while demonstrating their continuing relevance for the semiological study of contemporary forms of popular culture.
Saussure is concerned with establishing linguistics as a science, but in a particular kind of way. To do this he makes a number of distinctions and definitions which have come to be familiar to anyone acquainted with the academic study of culture.
89
STRUCTURALISM  AND   SEMIOLOGY
Saussure's starting-point is the need to define the object of structural linguistics. For this reason, he draws a distinction between langue and parole, between language as an internally related set of differentiated signs governed by a system of rules (language as a structure) and language as it is made manifest in speech or in writing (language as an accomplished fact of communication between human beings). Langue is, according to Saussure, the object which linguists should study, forming both the focus of their analyses and their principle of relevance. Langue is the overall system or structure of a language (its words, syntax, rules, conventions and meanings). It constitutes all specific languages and it makes it possible for them to be used; it is given and has to be taken for granted by any individual speaker. Langue allows people to produce speech and writing, including words and phrases which may be completely new. It is, in the first instance, this notion of langue which has proved influential, since it is relatively easy to infer from this definition that all cultural systems like myths, national cultures or ideologies, may be described and understood in this way.
Parole, by contrast, is defined and determined by langue. It is the actual manifestation of language made possible by, and deriving from langue. Parole is the sum of the linguistic units involved in speaking and writing. These cannot be studied in and of themselves as discrete, atomistic and historical items. Instead, they provide evidence about the underlying structure of langue. The aim of linguistics, then, is to use speaking and writing as the ways in which the underlying structure of the language, the object of linguistics, can be constructed. This means that its rules and relations can be understood and used to account for the particular uses people make of their language. Linguistics, therefore, involves the study of langue as a system or structure.
This study involves the discovery and scrutiny of the system of grammatical rules governing the construction of meaningful sentences. These rules are not usually apparent to the users of the language who can none the less still utter or write such sentences. As Saussure himself states: 'In separating language from speaking, we are at the same time separating: (1) What is social from what is individual; and (2) what is essential from what is accessory and
90
STRUCTURALISM  AND   SEMIOLOGY
more or less accidental' (1974; p. 14). From a sociological point of view it may seem strange to regard speaking as a non-social act. But from the point of view of structural linguistics and its subsequent influence, Saussure is distinguishing between fundamental and derivative social and cultural structures, between those structures which explain and those which need to be explained.
The second distinction Saussure introduces is that between the signifier and the signified. According to Saussure, any linguistic sign such as a word or phrase can be broken down into these two elements of which it is composed, although this is only possible analytically, not empirically. It is a function of langue rather than parole, and accounts for the capacity of language to confer meaning, a feature which has made it attractive for analysing cultural structures other than language. For Saussure, the meaning of particular linguistic units is not determined by an external material reality which imposes itself upon language. These units do not have a direct referent in the external, material world. This world exists but the meanings which are conferred upon it by language are determined by the meanings inherent in language as an objective structure of definite rules. In this sense, the meanings conferred by language arise from the differences between linguistic units which are determined by the overall system of language.
If language can be understood in this way then so can other cultural systems, and we shall see how this notion has played an important part in the attempt by Barthes to develop semiology as the scientific study of sign systems. It is necesary to stress this in order to appreciate the significance of Saussure's distinction between the signifier and the signified. As I have said, the linguistic sign is made up of both these elements and its meaning is not determined by its external referent. Words like dog or god do not acquire their meaning from their equivalents in the world outside language but from the way language begins to differentiate between them in terms of the ordering of their letters. A letter change gives us an entirely different concept, and probably explains why philosophers of language are so fond of using the example of the word 'dog'. In the linguistic sign, the signifier is the 'sound image', the word as it is actually said or written down,
91
9
B7B
STRUCTURALISM   AND   SEMIOLOGY
and the signified is the concept of the object or idea which is being referred to by the sign. So with our examples of dog and god, the letters you see or the sounds you hear are the signifiers, and the object and idea evoked by these sounds and words arc the signi-fieds. Language confers meaning on both of these words through their linguistic differences and their place in the differentiated categories of animals and supernatural beings.
Since the meanings of particular linguistic signs are not externally determined but derive from their place in the overall relational structure of language, it follows that the relationship between the signifier and signified is a purely arbitrary one. There is no definite reason as to why the notation dog should refer to that specific animal nor god to a supernatural deity. There is no intrinsic, natural or essential reason why a particular concept should be linked with one sound image rather than another. Therefore, it is not possible to understand individual linguistic signs in a piecemeal, ad hoc or empiricist fashion. They have, rather, to be explained by showing how they fit together as arbitrary signs in an internally coherent system or structure of rules and conventions. These signs cease to be arbitrary and become meaningful once they are located within the general structure of the language. Any item can only be understood in terms of this structure. This structure is, in fact, what Saussurc calls langue, and it is not given but has to be reconstructed analytically.
This set of ideas is again higly relevant to the way semiology has been developed as a means of studying popular culture. But it is important to note that the relationship between signifiers and signifieds is not arbitrary in culture as it is in language. This is because, according to semiology and structuralism, there are definite factors associated with conventions, codes and ideologies which determine the association of specific signifiers with specific signifieds.
The arguments put forward by Saussure involve the idea that languages as systems - much like cultures as systems - can only be studied and understood in relational terms. For structural linguistics, and structuralism and semiology more generally, meaning can only be derived from a general objective structure of rules in which
92
STRUCTURALISM   AND   SEMIOLOGY
particular items or units arc differentiated from each other, and derive their meaningful character from their place in this general structure. This structure, be it language or another sign system like culture, is not given empirically but has to be discovered and defined in relational terms.
Saussure suggests a conception of langue can be constructed by studying the linguistic signs of parole not as discrete individual items but as pointers to, and signs of, the structure of langue. In this respect, there are two types of relationships which Saussure considers important: syntagmatic, that is the relationships between items in a linguistic sequence, say words in a sentence; and paradigmatic, the relationships between items which might replace each other in a sequence thus altering its meaning, say substituting one word for another in a sentence. To define any item, unit or sign in this manner is to specify its relation to other items, units or signs which can be combined with it to form a sequence, or which contrast with it and can replace it in sequences. In either case, it is the relational character of the structure which enables the item, unit or sign to acquire meaning. This notion helps explain why structural linguistics has influenced the study of culture more generally since it suggests that other cultural systems can be analysed as if they are structured like a language.
The attempt to establish structural linguistics as a science leads Saussure to make a final distinction between synchronic and diachronic analysis. He argues that if the task of linguistics is to reconstruct the langue which makes speech and writing possible at any particular point in time, then synchronic analysis has to be kept separate from diachronic analysis. Synchronic analysis refers to the study of structures or systems at a particular point in time, while diachronic analysis involves the study of structures or systems over time. In Saussure's linguistics, synchronic analysis also entails the reconstruction of the system of language as a relational whole which is distinguished from, but not necessarily subordinated to, the diachronic study of the historical evolution and structural changes of particular linguistic signs and items. To combine these two forms of analysis would undermine the attempt to define the relational structure of a language. Language is viewed
93
STRUCTURALISM  AND   SEMIOLOGY
as a system of interrelated signs which are made meaningful by their place in the system rather than by their place in history. Freezing the system assimilated by speakers and writers at one point in time allows its structural and relational character to be clearly identified without being obscured by extraneous and incidental historical circumstances. Saussure seems to suggest that synchronic and diachronic analyses be kept separate from each other in order that the structure of langue can be determined. But he has been criticised, as have structuralism and semiology, for giving priority to synchronic analysis, and thereby neglecting the importance of historical and social change.
Saussure regards linguistics as a sub-branch of semiology. According to him, semiology is a science which studies the life of signs within society, showing what constitutes signs, and discovering the laws which govern them. Language can be studied as a semiological system of signs which express ideas, and can be understood properly by being compared with other systems of signs. Structural linguistics is one of the first examples of how semiology can be developed. In making this case, he laid the foundations for later attempts to extend the analytic potential of structuralism and semiology to other systems like popular culture.
Structuralism, culture and myth
The form of linguistics developed by Saussure has not, of course, gone unchallenged. Nor has it served as the uncontested paradigm for the study of language ever since. I have outlined some of Saussure's ideas because they have influenced the development of both structuralism and semiology. The importance of Saussure's linguistics will become clearer during this chapter, and is readily apparent in the structuralist concept of structure.
As far as the social study of language is concerned, Saussure's ideas have been heavily criticised.2 His definition of parole, for example, has been rejected for a number of reasons. Speech and writing are in fact social rather than individual phenomena, langue is only ever manifest in parole anyway, and the social nature of
94
STRUCTURALISM  AND   SEMIOLOGY
speech and writing make them open to change and transformation, unlike langue. Saussure only ever sees the latter as being social in character, and his overall theory tends to eschew social change. He also tends to regard human beings as mere mouthpieces for the rules of language which govern their speech and writing. Fairclough, for example, notes that 'language varies according to the social identities of people in interactions, their socially defined purposes, social setting, and so on. So Saussure's individualistic notion of parole is unsatisfactory' (1989; p. 21). Since Fairclough wishes to stress the links between language and power, he continues:
Saussure understood langue as something unitary and homogeneous throughout a society. But is there such a thing as 'a language* in this unitary and homogeneous sense? . .. When people talk about 'English' in Britain, for instance, they generally have in mind British standard English. . . . The spread of this variety into all the important public domains and its high status among most of the population are achievements of standardization... as part of the economic, political, and cultural unification of modern Britain. From this perspective, 'English' and other 'languages' [unlike Saussure's notion of langue] appear to be the products of social conditions specific to a particular historical epoch.
(ibid.)
While langue may be studied for its formal properties, it cannot in fact be understood outside of its particular uses, that is independently of parole. If this is so, it tends to undermine the rationale for the distinction Saussure draws between them.
Despite these and other problems, Saussure's work has formed an important influence upon the emergence of structuralism and semiology. Interestingly enough, Fairclough retains a reformulated sense of Saussure's distinction between langue and parole in that the former refers to 'underlying social conventions' and the latter to 'actual use' (ibid.; p. 22). In order to discuss structuralism, I shall begin with Lévi-Strauss's concept of structure.3
The French social anthropologist Lévi-Strauss (b. 1908) is well known for introducing the concepts and methods of structuralism
95
STRUCTURALISM   AND   SEMIOLOGY
into anthropology, and using them for studying the myths circulating in pre-industrial societies. His version of structuralism is concerned with uncovering the common structural principles underlying all the specific and historically variable manifestations of culture and myth. These structural principles involve the logical and universal characteristics of the human mind whose mental structure lies behind, classifies and generates all the empirical examples of cultural myths which can be discovered. This means that, for Lévi-Strauss, his concept of structure is a theoretical and explanatory one which has, in the first instance, nothing to do with empirical reality. It is not directly available to the senses, and lies behind, while producing, what we can observe. It is thus non-observable but generates or 'causes' that which can be observed. The relationship envisaged here by Lévi-Strauss is similar to that implied by Saussure between langue and parole.
If this structure is unobservable and causal, then it follows that it must be unconscious. Human beings subject to its power are unaware or unconscious of its influences in much the same way that the speakers or writers of a language are unaware or unconscious of its rules. More than this, consciousness often involves the mis-recognition of underlying structural causes and is a poor guide to their defining characteristics. The perceptions of human beings are as likely to mask as reveal these characteristics, and it falls to structuralist analysis to say what they are. Structuralism can do this because it is able to construct a relational model of what this underlying structure is like, even if it cannot be verified by the usual norms of empirical observation. According to the principles of structuralist analysis, it is a model of an underlying reality which has to be constructed. All the parts of this structure are systematically related to each other in the same way that all the units of a language are related to each other, and acquire their distinctive meanings due to their being part of a relational whole. Structuralism refers us to a relational structure like language which is more than the sum of its parts, and it argues that things cannot be studied in their empirical isolation but only in their structural unity.
This model of an underlying, unobservable, generative, relational and unconscious but real universal structure is defined by
96
STRUCTURALISM  AND   SEMIOLOGY
Lévi-Strauss as a logical grid of binary oppositions, combining rational modes of classification. It consists of a determinable number of related elements or oppositions which can be combined or classified as belonging together in a finite number of ways. It thus implies that all cultural forms represent different empirical combinations or symbolic reconciliations of inherent logical oppositions. Empirical examples, in this sense, merely represent secondary expressions or temporary reconciliations of the basic elements to be found in the underlying mental structure of binary oppositions. All cultures represent the logical transformations of the fundamental workings of the structure of oppositions inherent in the human mind. As Lévi-Strauss states in an essay on language and kinship:
If the general characteristics of the kinship systems of given geographical areas, which we have tried to bring into juxtaposition with equally general characteristics of the linguistic structures of those areas, are recognised by linguistics as an approach to equivalences of their own observations, then it will be apparent . . . that we are much closer to understanding the fundamental characteristics of social life than we have been accustomed to think. . . . We shall be in a position to understand basic similarities between forms of social life, such as language, art, law, and religion, that on the surface seem to differ greatly. At the same time, we shall have the hope of overcoming the opposition between the collective nature of culture and its manifestations in the individual, since the so-called 'collective consciousness' would, in the final analysis, be no more than the expression, on the level of individual thought and behaviour, of certain time and space modalities of the universal laws which make up the unconscious activity of the mind.
{1963; p. 65; cf. p. 21)
There is a lot more to Lévi-Strauss's view of structuralism than I can do justice to here. Moreover, what I have said so far may appear to be too abstract, so perhaps a few examples may clarify what he has to say. In his study of totemism, Lévi-Strauss shows
97
STRUCTURALISM   AND   SEMIOLOGY
very clearly how he conceives of his method of working. Totemism refers to the way types of animals or other 'natural' phenomena may be taken to represent a particular social group, say a clan or a tribe, within certain types of societies. Now, according to Levi-Strauss, totemism cannot be explained in terms of any specific example since there is no necessary reason why certain totems should represent certain groups. He rejects utilitarian, functional and integrative explanations and says, in effect, that the relationship between the group, the signified and the totem, the signifier, is, at this level, arbitrary. What he argues instead is that the empirically observable phenomenon is only one possible combination which exists alongside other logical possibilities. These can be discovered if the overall relational structure of possibilities and transformations is constructed. This overall procedure will, in turn, make the phenomenon of totemism intelligible.
Lévi-Strauss therefore constructs a grid of binary oppositions and possible permutations on the assumption that the phenomenon of totemism is an empirical manifestation of the more fundmental and universal tendency of societies to classify socio-cultural things, like groups or tribes, by means of things which are natural, like animals or plants, if totemism is seen as a phenomenon which provides a non-social (natural) representation of the social (cultural), and which can have both an individual and collective existence, and if the natural can thereby consist of categories and particulars, and the cultural of groups and persons, then totemism can be located within these possible combinations of the logically related oppositions between collective and individual existence, and culture and nature. Totemism is made intelligible as a way of transforming the elements contained in the following grid (from Lévi-Strauss: 1969; pp. 84-85):
Nature      Category     Particular Culture     Group          Person
These collective and individual expressions of the binary opposition between culture and nature can be combined and transformed into a number of distinct relational types as follows (from ibid.):
98
STRUCTURALISM  AND   SEMIOLOGY
12                 3                  4
Nature     Category     Category    Particular     Particular Culture    Group         Person        Person          Group
Totemism can thus be understood not as a distinct empirical phenomenon to be found in certain cultures, but as a number of different types which emanate from this classificatory structure of logical oppositions and possible transformations. Lévi-Strauss identifies totemism empirically with types 1 and 2, and says it is only indirectly related to types 3 and 4. It is a phenomenon which consists of certain relations and forms which can be explained only when the complete structure, of which these relations are amongst the other possible combinations, has been reconstructed. For it is this universal and underlying structure, rooted in and articulating the opposition between culture and nature, which causes or gives rise to the phenomenon of totemism, and which allows other possible transformations of the elements involved to occur. Equally, totemism provides a symbolic reconciliation of the opposition between culture and nature in that both are united by being represented through the totem. It is an empirically manifest way in which societies and their cultures mediate the universal relation between culture and nature. Other myths such as those to be found in folklore could in principle be analysed in terms of other universal oppositions, like that between good and evil, or the sacred and the profane.4 Similarly, structuralist analyses have been extended to contemporary forms of popular culture. Before going on to look at this, another illustrative example of the way Lčvi-Strauss uses structuralism to study myths may be useful.
In Structural Anthropology Lévi-Strauss cites a myth to be found amongst the Iroquois and Algonquin indians of North America which he suggests bears a close resemblance to the 'Oedipus legend'. The story concerns incest between brother and sister rather than mother and son, and murder, although not the unwitting slaying by a son of his father; but it does contain the moral that attempts to prevent incest making its occurrence inevitable. In looking at elements of these two myths he asks the questions:  'Is this a simple coincidence - different causes
99
STRUCTURALISM  AND   SEMIOLOGY
explaining that, here and there, the same motifs are arbitrarily found together? Or are there deeper reasons for the analogy? In making the comparison, have we not put our ringer on a fragment of a meaningful whole?' (1977; p. 21). His answer to the last question is yes. However, he does provide a test of his theory since the native north American myth lacks the puzzle or riddle to be found in the Oedipal legend. If these myths are fragments, cultural manifestations, of a meaningful whole, an underlying and causal logical structure, then this element should also be there even if it has been transformed.
This is indeed what Lévi-Strauss discovers. He points out that puzzles and riddles, like that associated with the Sphinx episode in the Oedipus myth, are almost entirely absent among the 'North American Indians'. So if such an element could be found it would demonstrate that he has uncovered 'a fragment of a meaningful whole', and that 'it would not, then, be the effect of chance, but proof of necessity' (ibid,; p. 22). He says that amongst native North American myths only two types of riddles can be found: one where they are told to audiences by clowns whose birth is the result of incest; and one, to be found amongst the Algonquins, where owls ask riddles 'which the hero must answer under pain of death' (ibid. - this is the dilemma Oedipus finds himself in when confronted by the Sphinx). In the story we started with, the incestuous brother, the hero of the myth, murders his double whose mother is a sorceress, a mistress of the owls. This means that not only do we have a transformation of the basic structure of incestuous relations, sister-brother, mother-son, but also of puzzles or riddles which 'present a double Oedipal character, by way of incest on the one hand, and on the other hand, by way of the owl in which we are led to see, in a transposed form, an American Sphinx' (ibid.).
Again Lévi-Strauss has discovered meaningful relations between elements and oppositions - incest and riddles - which are transformed from one myth to another. They are, in turn, suggestive of other possible relations, and arise from an underlying and universal mental structure which 'thinks* these relations and oppositions. Thus he then pursues, for example, the possible permutations of riddles with each other, those questions which
TOO
STRUCTURALISM   AND   SEMIOLOGY
have no answer and those answers which have no question. This takes him on to the death of Buddha and the Holy Grail cycle, where questions which should be asked arc not. He also looks at the relations between sexuality as expressed by incest, and chastity as represented by the actions of the heroes of myths, and tries to locate Oedipal type myths within a wider structure of possibilities. The point of these examples is to unravel the meaningful and logically based mental whole which lies behind them.
This argument is given additional force since it uncovers a structure which manages to prevail irrespective of the influences exerted by specific historical, social or cultural conditions. As Lévi-Strauss concludes: 'it seems that the same correlation between riddles and incest exists among peoples separated by history, geography, language and culture' (ibid.; p. 24). Comparable transformations of myth can be found in societies as far apart from each other, and as structurally distinct, as native north American tribes and the city states of Ancient Greece. If this is so, then the specific features of these societies cannot explain the character of myths. However, the logical structure of the human mind can explain the similarities and transformations detected by structuralism in cultural myths. In a fitting conclusion to his discussion of these myths, Lévi-Strauss writes: 'we have only sketched here the broad outlines of a demonstration ... to illustrate the problem of invariance which, like other sciences, social anthropology attempts to resolve, but which it sees as the modern form of a question with which it has always been concerned - that of the universality of human nature' (ibid.). This is related to his view that:
today, no science can consider the structure with which it has to deal as being no more than a haphazard arrangement of just any parts. An arrangement is structured which meets but two conditions: that it be a system ruled by an internal cohe-siveness and that this cohesiveness, inaccessible to observation in an isolated system, be revealed in the study of transformations through which similar properties are recognized in apparently different systems.
(ibid.; p. 18)
101
STRUCTURALISM   AND   SEMIOLOGY
Structuralism and James Bond
An analysis of contemporary popular culture which shares much in common with the structuralist approach, while not committing itself to all the presuppositions held by a writer like Lévi-Strauss, is Umberto Eco's study of the James Bond novels written by Ian Fleming, A critical consideration of this study can show us how structuralism works as a way of studying contemporary popular culture, as well as the limitations it confronts. A leading contemporary Italian intellectual and semiologist, Eco (b.1932) is well known as a popular novelist, and as one of the first intellectuals to take a serious interest in popular culture at a time when it was not academically fashionable to do so. His study of the Bond novels is perhaps the best known example of his attempts to apply the methods of structuralism to a form of popular culture.5
Eco's concern is to uncover the invariant rules governing the narrative structure of these novels which ensure both their popular success and their appeal to a more restricted and culturally literate audience. The novels are a form of popular culture based upon an underlying structure of rules which ensures their popularity. For Eco these rules are comparable to 'a machine that functions basically on a set of precise units governed by rigorous combinational rules. The presence of those rules explains and determines the success of the "007" saga - a success which, singularly, has been due both to the mass consensus and to the appreciation of more sophisticated readers' (1979; p. 146). This 'narrative machine' presumably connects at some unconscious level with the desires and values of the popular audience, for each cog or 'structural element' of which this machine is composed, is assumed to be related to 'the reader's sensitivity' (ibid.).
First of all, Eco constructs the series of oppositions upon which the novels are based. These oppositions, which are very similar to Lévi-Strauss's binary oppositions, can be combined and recombined with each other, and are 'immediate and universal' (ibid.; p. 147). Their 'permutation and interaction' means that the combination, association and representation of each opposition can be varied, to some extent, from novel to novel. None the less
102
STRUCTURALISM  AND   SEMIOLOGY
they form an invariant structure of oppositions which defines the narratives and ensures the popularity of the novels. These oppositions involve the relations between characters in the novels (for example, between Bond and the villain or the woman), the relations between ideologies (for example, between liberalism and totalitarianism, or the 'free world' and the 'Soviet Union') and a larger number of relations between distinct types of values {for example, 'cupidity-ideals, love-death, chance-planning . .. perversion-innocence, loyalty-disloyalty' (ibid.)). These relationships are worked out by particular characters, the relations between characters and the unravelling of the story as a whole. For example, Bond, in his relations with the villain, represents the ascendancy of the free world over the Soviet Union, and the victory of chance over planning. But whatever the specific transformation of relations between oppositions in particular novels, the underlying structure of oppositions remains the same. Eco traces the nature of this structure and its transformations across the stories to be found in the James Bond novels.
This argument is linked to the idea that there is an invariant sequential structure underlying the novels. Eco compares this to 'play situations' or 'games' in which each initial 'move' gives rise to a countermove and so on, pushing the story forward. The prevalence of games of chance in the novels occurs 'because they form a reduced and formalized model of the more general play situation that is the novel. The novel, given the rules of combination of oppositional couples, is fixed as a sequence of "moves" inspired by the code and constituted according to a perfectly prearranged scheme' (ibid.; p. 156). Abbreviating slightly, this 'invariable scheme' can be detailed as follows:
A M moves and gives a task to Bond;
B Villain moves and appears to Bond . . .;
C Bond moves and gives a first check to Villain or Villain
gives first check to Bond; D Woman moves and shows herself to Bond; E Bond takes Woman . ..; F Villain captures Bond . . .;
103
STRUCTURALISM  AND   SEMIOLOGY
G Villain tortures Bond . ..; H Bond beats Villain . . .;
I   Bond, convalescing, enjoys Woman, whom he then loses.
(ibid.)
This scheme is invariant in that each novel must contain all these elements or 'moves'. It is demanded by the narrative structure of the novels and explains their popular success. However, these basic elements need not appear in this sequence. In fact, Eco goes to great lengths to show the range of variations possible. In this sense, paradigmatic relations are more fundamental to the structure which articulates the novels than syntagmatic relations. The sequence may change but the structure remains the same. And it does so, as Eco tries to demonstrate, irrespective of the many 'side issues' or incidental features which may be introduced to add colour and variety to any particular novel.
It is the coming together of these two structures of binary oppositions and premeditated moves which, for Eco, accounts for the popular attractions of the novels. The incidental features or 'collateral inventions' play their part in this success, especially amongst more 'sophisticated' readers. According to Eco:
The true and original story remains immutable, and suspense is stabilized curiously on the basis of a sequence of events that are entirely pre-determined .. . there is no basic variation, but rather the repetition of a habitual scheme in which the reader can recognize something he has already seen and of which he has grown fond ... it, in fact reconfirms .. . [the reader] in a sort of imaginative laziness ... by narrating . . . the Already Known. . . . The reader finds himself immersed in a game of which he knows the pieces and the rules — and perhaps the outcome - and draws pleasure simply from following the minimal variations by which the victor realizes his objective.
(ibid.; p. 160)
This passage is interesting in that it combines the concepts of structuralism with a picture of the audience consistent with that
1 04
STRUCTURALISM  AND   SEMIOLOGY
presented by the mass culture critics and the Frankfurt School. As Eco continues: 'the novels of Fleming exploit in exemplary manner that element of foregone play which is typical of the escape mechanism geared for the entertainment of the masses' (ibid.; p. 161). A theory which relies for its explanatory power upon the concept of an underlying and unconscious structure is clearly liable to underestimate the significance of the role of the audience in understanding popular culture.6 This view is evident in Eco's account of Fleming's use of ideology. He argues that the ideologies to be found in the novels are determined by the demands of mass culture. Fleming's reliance on cold war ideology, for example, derives simply from his endorsement of 'the common opinions shared by the majority of his readers* (ibid.). Eco suggests that 'Fleming seeks elementary oppositions; to personify primitive and universal forces, he has recourse to popular standards' (ibid.;
p. 162).
Another aspect of Eco's structuralism, one consistent with that of Lévi-Strauss, concerns the universal character of the structure which lies behind and explains the popularity of the Bond novels. Eco sees the narrative structure of these novels as representing a modern variation on the universal theme of the struggle between good and evil. This struggle, which for Eco defines Fleming's Manichaean ideology even if it is the result of opportunism, forms a fundamental binary opposition. The Bond novels are comparable to fairy tales in which a knight (Bond), under the orders of a King (M), goes on a mission to destroy the monster, such as a dragon (the Villain), and rescue the Lady (the Woman). Both types of story involve transformations of the basic elements embodied in the binary opposition between good and evil. They express a universal structure of basic oppositions which, because it is universal, will ensure popular success. Both the Bond novels and fairy tales are successful because they are universal in their underlying connection with the eternal conflict between good and evil.
The popular success of the Bond novels is accounted for by the idea that the mass audience is unknowingly in tune with the universal themes which arc evoked. However, there are more discerning readers who are very aware of why the novels appeal to
105
STRUCTURALISM  AND   SEMIOLOGY
them. These readers are conscious of the mechanics of the novels, and capable of grasping the more subtle and esoteric allusions in Fleming's writing (ibid.; p. 163). Eco is in fact identifying a culturally stratified audience for the Bond novels, one divided between the mass popular readership on the one hand and a highly sophisticated cultural elite on the other. He thus elaborates upon the kinds of references to be found in the novels which appeal to the tastes and temperament of the culturally literate reader. He notes, for example, the resemblance between the physical description of James Bond and that of a typical Byronic hero (ibid.; pp. 171-172; see also pp. 169-170). 'By now', he writes,
it is clear how the novels of Fleming have attained such a wide success: they build up a network of elementary associations to achieve something original and profound. Fleming also pleases the sophisticated readers who here distinguish, with a feeling of aesthetic pleasure, the purity of the primitive epic impudently and maliciously translated into current terms and who applaud in Fleming the cultured man, whom they recognize as one of themselves, naturally the most clever and broadminded.
(ibid.; p. 163)
Structuralism therefore leads Eco to argue that the structure of the novels positions particular kinds of readers, masses and elites in terms of particular kinds of attraction, elemental primitivism and cultural sophistication. In view of this, it is curious that Eco finally comes to recognise the importance of readers who are not determined in their reading by the structure of the text. He argues that 'since the decoding of a messsage cannot be established by its author, but depends on the concrete circumstances of reception, it is difficult to guess what Fleming is or will be for his readers' {ibid.; p. 172). But without this 'definitive verification' what is the point of the analysis Eco has carried out? If it is the reception of the novels by their readers which determines their meaning, then what is the point of uncovering their invariant structure of binary oppositions? What value do structuralist analyses have if cultural meanings are derived from the 'society that reads', from 'the concrete
1 06
STRUCTURALISM  AND   SEMIOLOGY
circumstances of reception'? These circumstances tend to involve socially and historically specific patterns of cultural production and consumption, and are certainly not defined by the invariant structure of narrative or the logic of binary oppositions. To suggest that the influence exerted by the universal structure does not determine how and why people read the texts which it generates, is to call into question the value of structuralist analyses. This is all the more surprising since Eco starts from the assumption that by identifying this structure it is possible to account for the popularity of the texts being studied.
There is an obvious confusion here over the role of readers or audiences. Are audiences determined in their 'readings' by a universal structural principle, or arc these 'readings' determined by the social, cultural and historical circumstances of their audiences? This problem is one which, I think, can be found throughout the work of structuralists and semiologists. Another way of thinking about it is to ask the questions: do audiences themselves decide upon their understandings of cultural forms?; or does the analyst or theorist decide for them?; and do the latter take account of the former in arriving at their interpretations of popular culture?
This problem is compounded by the ahistorical nature of Eco's structuralist analysis. His explanatory principle is, after all, an invariant, static and eternal structure. As Bennett and Woollacott argue, there are in fact no fixed, universal and ahistorical codes; readings of popular culture are always organised in historically specific contexts. They point out how difficult it is to make sense of the James Bond novels without taking into consideration their 'inter-textuality' (Bennett and Woollacott: 1987; chapter 3). By this they mean that the popular cultural phenomenon of James Bond has to be assessed in the context of a range of 'texts', or cultural forms and media, including, most significantly, the series of James Bond films as well as the novels. They also indicate how readers come to novels with some prior cultural knowledge, and suggest that the codes developed in the context of reading the British imperialist spy thriller formed an important aspect of the cultural knowledge readers would have brought to their interpretations of the Bond novels.
1 07
STRUCTURALISM   AND   SEMIOLOGY
They even speculate that some working-class readers would have read them in terms of the codes associated with detective fiction. Similarly, Denning argues that the emergence of codes associated with tourism and pornography in the 1960s was a crucial reference-point for audiences of both the novels and the films (Denning: 1987; chapter 4).
According to Bennett and Woollacott, Eco is talking about genre. They argue, however, that he tends to conceive of it as a structure of relatively fixed, textual conventions, whereas they would rather see it as a social and textual set of culturally variable expectations, orientations and values which circulate between producers and consumers, and which change over time (1987; pp. 76-90). Eco's study contains no sense of history and very little sense of the society in which the James Bond novels have been produced and read, and then made into films. Yet texts never exist in splendid isolation. They become significant when they arc located within the social relationships which produce and consume them. It could equally be concluded that the vocabulary of readers, texts and readings is misplaced as a way of analysing popular culture.7
Barthes, semiology and popular culture
The semiological study of popular culture probably owes much of its reputation and importance to the writings of the French literary critic and semiologist Barthes (1915-1980), and in particular to his book Mythologies (originally published in 1957). In the studies and theoretical arguments which make up this book, Barthes sets out a way of interpreting popular culture which has, with some notable revisions, been much discussed and highly influential ever since.8 Although it will not be the only work of Barthes's that I shall refer to here, it is arguably the most significant. Before I do consider it, I need to make some general points about Barthes's semiology, and outline some of the points made in Writing Degree Zero, the book he wrote before Mythologies, since it clarifies a great deal of his subsequent work.
1 08
STRUCTURALISM  AND   SEMIOLOGY
Barthes, structuralism, and semiology
The general points which need to be made here tend to echo those made about structuralism, except that semiology does not rely upon the idea that there is a universal structure underlying sign systems. In view of this, it is more capable of analysing social change. The signs and codes it refers to can be regarded as historically and culturally specific. It does, however, insist that it is these codes and signs which make meaning possible and thus allow human beings to interpret and make intelligible the world around them.
The wider significance of semiology can perhaps be gauged by the way Barthes later clarified his aims in writing the pieces which make up his book Mythologies: 'I was dazzled by this hope: to give my denunciation of the self-proclaimed petit-bourgeois myths the means of developing scientifically; this means was semiology or the close analysis of the processes of meaning by which the bourgeoisie converts its historical class-culture into universal nature; semiology appeared to me, then, in its program and tasks, as the fundamental method of an ideological critique' (1988; p. 5). This serves as a useful indication of the intentions of his book, and the general premises upon which it was based, even if it also hints at Barthes's subsequent reluctance to conceive of semiology as a systematic science.
As with structuralism, the first point which needs to be noted is that semiology is defined as a science of signs, in keeping with Saussure's original suggestion. It not only possesses a notion of ideology against which the truth of science can be measured, but it promises a scientific way of understanding popular culture. This allows it to be distinguished from the arbitrary and individualistic impressionism of liberal humanist studies of culture, as well as from those approaches which rely upon aesthetic discrimination and 'good taste'.
Semiology argues that material reality can never be taken for granted, imposing its meanings upon human beings. Reality is always constructed, and made intelligible to human understanding by culturally specific systems of meaning. This meaning is never 'innocent', but has some particular purpose or interest lying behind
1 09
STRUCTURALISM  AND   SEMIOLOGY
it, which semiology can uncover. Our experience of the world is never pure or 'innocent1 because systems of meaning make sure that it is intelligible. There is no such thing as a pure, uncoded, objective experience of a real and objective world. The latter exists but its intelligibility depends upon codes of meaning or systems of signs, like language.
These codes and signs are not universally given, but are historically and socially specific to the particular interests and purposes which lie behind them. It is in this sense that they are never innocent. Meaning is not something which is given or which can be taken for granted. It is manufactured out of historically shifting systems of codes, conventions and signs. Semiology is concerned with this production of meaning, with what Barthes calls 'the process of signification*. Just as culture cannot be seen as being universal, neither can it be seen as being divorced from the social conditions in which it is to be found. Rather, it tries to present itself in this way when it is really historically and socially fixed. As Barthes writes in Mythologies, the function of myth is to 'transform history into nature' {1973; p. 140). This will become clearer when we look at his analyses of myth, as will his general assumptions if we look briefly at his book Writing Degree Zero (1967; originally published in 1953).
Writing Degree Zero
In this, his first book, originally published in 1953, Barthes was concerned with the French classical style of writing. This style, which emerged in court society in the seventeenth century, prided itself upon clarity and preciseness of expression, and set itself up as a universal model or standard for all writing. This French classical style of writing came to be considered, by the nineteenth century, as the only correct and rational way to write, an inevitable and 'natural' style which simply and unambiguously served to reflect reality. During this period, this model of lucidity came to be legitimised as a universal paradigm of human communication.
This is not, however, how Barthes wishes to view it. For a start, despite its claims to universal validity as a model of clarity
1 10
STRUCTURALISM   AND   SEMIOLOGY
and as a natural way of representing reality, it begins to disintegrate from the mid-ninteenth century onwards. In this process, it is challenged by a multiplication of styles, for example writing as a craft or job, self-conscious literariness, and 'writing degree zero', to which we shall return. The reasons for this disintegration underpin Barthes's critique of the French classical style. He argues that it is wider social forces and class interests which govern the formation and transformation of writing styles. The emergence of new class interests and conflicts result in the breakdown of the classical style. Barthes interprets this style, despite its pretensions, as an aspect of the rise of bourgeois hegemony, and thus as a 'class idiom'.
For Barthes, French classicism, irrespective of its pretensions, is neither neutral and universal nor natural and inevitable. Instead, it has to be located in its historical and social contexts. As such, it is central to the rise of bourgeois hegemony between the seventeenth and nineteenth centuries, and to the emergence of challenges to this hegemony from the 1850s onwards. According to Barthes, the classical style is rhetorical in character, motivated by the 'permanent intention to persuade'. It is the style of the law courts and the political campaign, aimed at changing opinions and ensuring the acceptance of the bourgeois view of the world. It has thus not simply been a reflection of reality but an attempt to shape conceptions of reality. It has not been neutral and universal, nor natural and inevitable, but historically specific and socially constructed, rooted in a particular set of class interests. Its meaning has not been given but produced, not 'innocent' but 'guilty*. French classicism is another 'myth' which tries to transform the historical into the natural in the interests of the bourgeois class.
This is, for Barthes, a feature of all writing. 'Writing degree zero' is a style developed in order to reject the idea of politically committed writing. It values writing which is colourless, transparent and neutral, blank and impersonal. It pretends to be as asocial and ahistorical as possible. In a way, it is not a style at all. But this is not in fact possible according to Barthes. For him, all writing is a form of fabrication, a way of making things up, which therefore cannot avoid these signs of fabrication or style.
1 1 1
STRUCTURALISM  AND   SEMIOLOGY
Furthermore, all writing is ideological and cannot avoid being so. Writing is never just an instrument of communication, an open way of addressing people. It is rather a product of certain social and historical circumstances and certain power relations, and cannot escape their influence. Non-ideological writing, writing which presents itself as being beyond ideology, is for Barthes shown to be an illusion by his investigation of French classicism and 'writing degree zero'.
Myths and popular culture
Barthes carried these ideas further in his book Mythologies, which contains both a series of short essays on various examples of popular culture, originally published in magazines, and an outline of the concepts and methods of semiology which he uses to analyse the examples. It is the latter we shall consider first. Myths are forms of popular culture, but they are also more than this according to Barthes. We have to find out what is really going on, and to do this we have to turn to semiology.
'Myth is a system of communication, that is a message', Barthes writes, 'a mode of signification, a form', 'a type of speech .. . conveyed by a discourse. Myth is not defined by the object of its message, but by the way in which it utters this message' (1973; p. 117). This means that the concepts and procedures of semiology can be applied to the study of myths. To understand this we need to remind ourselves of the claims semiology makes. Barthes notes that 'any semiology postulates a relation between two terms, a sig-nifier and a signified' (ibid.; p. 121), a distinction elaborated by Saussurc as we have seen. There is also a third term in this, the sign itself (be it linguistic or mythological), which contains the signifier and the signified. Barthes wishes to use this argument to study myth, and he gives an initial and preliminary example of how this might be done.
The case he has in mind is a bunch of roses which can be used to signify passion. 'Do we have here, then', Barthes asks, 'only a signifier and a signified, the roses and my passion? Not even that: to put it accurately, there are here only "passionified" roses. But
1 12
STRUCTURALISM   AND   SEMIOLOGY
on the plane of analysis, we do have three terms [even if empirically there is only one thing, the roses]; for these roses weighted with passion perfectly and correctly allow themselves to be decomposed into roses and passion: the former and the latter existed before uniting and forming this third object, which is the sign' (ibid.; pp. 121-122). In other words, the roses are a signifier of a signified which is passion, something signified by the roses sent to a loved one. The bunch of roses can thus be analytically if not empirically broken down into a signifier, the roses, a signified, passion, and a sign which combines and is not separate from these two components, the roses as a sign of passion. Here, passion is the process of signification. The fact that this attribution of meaning - the roses signify passion and not, say, a joke or a farewell -cannot be understood simply in terms of the system of signs, but has to be located in the context of the social relationships in which the attribution of meaning occurs is a problem which semiology finds it difficult to deal with. It is similar to the problem Sassurian linguistics has in dealing with language independently of the contexts in which people actually use language.
Because the mythic process of signification is not totally comparable with that associated with language, Barthes uses other concepts to analyse myths. According to Barthes, myth 'is a second-order semiological system' (ibid.; p. 123). This means it relies upon signs in other first-order systems like language (and horticulture, as with roses?) in order to engage in the process of signification. A sign in a first-order system, a word, a flower or a photograph, becomes a signifier in the second-order system of myth. Myth uses the language of other systems, be they written or pictorial, to construct meanings. Myth thus becomes a metalanguage because it can refer to other languages, and so necessitates the use of new if comparable concepts.
These concepts are established by Barthes through his most famous example, and it is worth quoting him at length:
I am at the barber's, and a copy of Paris-Match is offered to me. On the cover, a young Negro in a French uniform is saluting, with his eyes uplifted, probably fixed on a fold of
1 1 3
STRUCTURALISM   AND   SEMIOLOGY
the tricolour. All this is the meaning of the picture. But . . . I see very well what it signifies to me: that France is a great Empire, that all her sons, without any colour discrimination, faithfully serve under her flag, and that there is no better answer to the detractors of an alleged colonialism than the zeal shown by this Negro in serving his so-called oppressors. I am therefore . . . faced with a greater semiological system: there is a signifier, itself already formed within a previous system (a black soldier is giving the French salute); there is a signified (it is here a purposeful mixture of Frenchness and militariness); finally, there is a presence of the signified through the signifier. . . . French imperiality.
(ibid.; pp. 125-126 and 128)
While retaining the analytical value of the distinctions made by structural linguistics, Barthes suggests that for studying myths it is more appropriate to avoid confusion. Therefore, the signifier becomes form> the signified concept and the sign signification. In the example cited, we have the form of the black soldier saluting the French flag, the concept of French military strength and the signification of the grandure and impartiality of French imperialism all embodied in the photograph, and revealed by semiological analysis.
Using these concepts and this example, Barthes argues that myth works through the particular relationships between form, concept and signification. The form of this specific myth of French imperiality, the black soldier, is taken from one system, his real history, which gave him his meaning, and placed in another system, that of the myth, which is precisely designed to deny his history and culture, and thus the real history of French colonial exploitation. What motivates this 'impoverishment of meaning' is the concept of French imperiality which gives another history to the soldier, that of the grandure and impartiality of French colonialism. The soldier is now made to function as a sign of French imperiality. As Barthes puts it, emphasising the process of signification: 'The French Empire? It's fust a fact: look at this good Negro who salutes just like one of our boys' (ibid.; p. 134).
1 14
STRUCTURALISM  AND   SEMIOLOGY
For Barthes, 'signification is the myth itself (ibid.; p. 131), the coming together of form and concept in the cultural sign. But the form does not hide the concept, or make it disappear as some theories of ideology tend to insist. Barthes writes: 'myth hides nothing: its function is to distort, not to make disappear . .. there is no need of an unconscious in order to explain myth ... the relation which unites the concept of the myth to its meaning is essentially a relation of deformation ... in myth the meaning is distorted by the concept' (ibid.; pp. 131-132). Unlike the linguistic sign, the 'mythical signification ... is never arbitrary; it is always in part motivated' (ibid.; p. 136), and this motivation of form by concept brings us to the social and historical characteristics of myth.
Barthes notes that 'if one wishes to connect a mythical schema to a general history, to explain how it corresponds to the interests of a definite society - in short, to pass from semiology to ideology' (ibid.; p. 138), it is necessary to become a semiologist and to understand 'the very principle of myth: it transforms history into nature' (ibid.; p. 140). As with his analysis of French classical writing, Barthes argues that myth has to be understood in terms of the way it functions to transform that which is socially particular (the interests of the bourgeois class) and historically specific (the structure of capitalist societies) into something which is natural and inevitable, about which nothing can be done because it has always been the case (for example, 'The French Empire? It's just a fact'), when it is in fact an historically specific structure of imperial power.
I shall try to illustrate this point about the function of myth in a moment. What has to be noted here is that one consequence of Barthes's idea that myth serves to naturalise history is how this process is seen to influence consumers by naturalising their reactions to the myth:
what allows the reader to consume myth innocently is that he does not see it as a semiological system but as an inductive one ... the signifier and the signified have, in his eyes, a natural relationship . . . any semiological system is a system
1 15
STRUCTURALISM  AND   SEMIOLOGY
of values; now the myth-consumer takes the signification for a system of facts: myth is read as a factual system, whereas it is but a semiological system.
(ibid.; p. 142)
Although myth is not an unconscious process, its consumers, according to Barthes, take it at face value, and accept it as natural and inevitable. They need semiology to tell them that myth is a system of meaning which cannot be taken for granted. The semiological interpretation of myth proceeds on the assumption that readers will understand myth in the way the theory presumes they will. It does not therefore take account of how people actually interpret myth, for if myths are so effectively mystifying, how can they be so easily demystified?
Bourgeois men and women novelists
In a later work, Elements of Semiology,9 Barthes refined his understanding of the relationship between the signifier, the signified and myth by drawing a distinction between denotation and connotation. On one level, the meaning of popular cultural signs is self-evident. They are what they are or what they appear to be, an advert, a photo of a black soldier, a bunch of roses and so on. In other words, they denote something to us, they present it to us as a matter of fact: this is a photo of a soldier, an advert, a bunch of roses. Denotation refers to those things which appear to us as natural and which we can take for granted.
But the task of semiology is to go beyond these denotations to get to the connotations of the sign. Doing this reveals how myth works through particular signs. In this way, the constructed, manufactured and historical location of the myth can be discovered. The connotations of myths can thus be identified: this may appear to be a bunch of roses but it connotes passion; or this may appear to be a photo of a black soldier saluting the French flag, but it really connotes the grandure and impartiality of French imperialism. The methods of semiology reveal the ideologies contained in cultural myths.
1 16
STRUCTURALISM  AND   SEMIOLOGY
Barthes is concerned with the role of myth in modern society, how it is constructed and sustains meaning as a systematic force. His intention is to get behind the process of mythical construction in order to reveal the real meanings which are distorted by myth. This involves moving from meanings that are taken for granted, which make things appear natural and inevitable, to meanings that are rooted in historical circumstances and class interests, moving, as he puts it, 'from semiology to ideology' (1968; p. 139). Although there is a great deal of novelty and interest in Barthes's notion of semiology, his conception of ideology seems to be more consistent with cruder Marxist versions of the concept in that the myths of popular culture are viewed as serving the interests of a bourgeois class. I shall look at this theory in the next chapter, but it is useful to see what Barthes has to say about the issue.
It is characteristic of bourgeois ideology, according to Barthes, to deny the existence of a bourgeois class. He writes: 'as an ideological fact, it completely disappears: the bourgeoisie has obliterated its name in passing from reality to representation, from economic man to mental man' (1973; p. 150). This is a class with no name because myth functions as ideology to ensure that it is not named. For example, the myth of the nation guarantees the anonymity of the bourgeoisie by representing everyone as citizens. This is part of the more general tendency for bourgeois ideology to focus upon the figure of universal 'ratf«', thereby dissolving the reality of social classes: 'the fact of the bourgeoisie becomes absorbed into an amorphous universe, whose sole inhabitant is Eternal Man, who is neither proletarian nor bourgeois .. . the whole of France is steeped in this anonymous ideology ... is dependent on the representations which the bourgeoisie has and makes us have of the relations between man and the world' (ibid.; pp. 153 and 152).
Barthes is therefore led to the conclusion that the function of myth as ideology is defined by the fact that bourgeois ideology lies at the very heart of myth in modern society. 'The flight from the name "bourgeois"', Barthes writes:
is not therefore an illusory, accidental, secondary, natural or
1 17
STRUCTURALISM  AND   SEMIOLOGY
insignificant phenomenon: it is bourgeois ideology itself, the process through which the bourgeoisie transforms the reality of the world into an image of the world, History into Nature. And this image has a remarkable feature: it is upside down. The status of the bourgeoisie is particular, historical: man as represented by it is universal, eternal.
(ibid.; p. 154)
This allows Barthes to fill out his understanding of myth. Myth transforms history into nature, which is exactly the function of bourgeois ideology. Myth thus facilitates the tasks of bourgeois ideology. It becomes 'possible to complete the semiological definition of myth in a bourgeois society: myth is depoliticized speech' (ibid.; p. 155). Politics is here understood in its wider sense of the totality of power relations, and myths are said to vary in the extent to which they are political. None the less, Barthes sees politics as being central to the understanding of myth because it accounts for the transformation of the historical into the natural. Since this is precisely what bourgeois ideology does, myth has to be conceived of as the outcome of the interests of the bourgeois class.
Barthes conducts a similar analysis of gender, and we can use this example to conclude our exposition of semiology. Yet again he takes as an example a photo in a magazine. Since the signs of popular culture are, at first sight, self-evident and all around us, we don't have to look very far for examples of how myths work. For Barthes, it is partly because modern bourgeois society is flooded with cultural signs that semiology is so important. This time his example is of a photo of seventy women novelists. What is interesting from Barthes's point of view is the fact that these women are also identified by the number of children they have. The photograph and its caption denote a group of women writers who are also mothers. The connotation is, however, what interests Barthes and he identifies this as the attempt, by the conjunction of women novelists and mothers, to make the role of women as mothers appear to be primary, natural and inevitable, whereas it is in fact historically and culturally specific. Women may succeed in being novelists, but the connotations of the photo and caption distort
1 1 8
STRUCTURALISM  AND   SEMIOLOGY
this to imply that women are much more fundamentally and naturally concerned with motherhood. Equally the photo and caption can be seen as the signifier, the signified of which is the natural role of women to be mothers, irrespective of whatever else they do or aspire to, such as being novelists.
Barthes interprets this myth in terms of
the eternal statute of womanhood. Women are on the earth to give children to men; let them write as much as they like; let them decorate their condition, but above all, let them not depart from it. . .. Let women acquire self-confidence: they can very well have access, like men, to the superior status of creation. But let men be quickly reassured: women will not be taken from them for ail that, they will remain no less available for motherhood by nature.
(ibid.; pp. 56-57)
Myth is again seen by Barthes to function by transforming history into nature. This time the role of women as mothers is made to appear natural and inevitable, the related connotation being that the power and dominance of men is equally natural and inevitable. The myth exhorts women as follows: 'Love, work, write, be businesswomen or women of letters, but always remember that man exists, and that you are not made like him; your order is free on condition that it depends on his; your freedom is a luxury, it is possible only if you first acknowledge the obligations of your nature* (ibid.; p. 58).
Structuralism and semiology: some key problems
I have already hinted at some of the problems these perspectives face, particularly in my outlines of the arguments of Saussure and Eco. I now want to clarify these problems by looking at some of the criticisms that can be made of Lévi-Strauss's structuralism and Barthes's semiology.
1 19
STRUCTURALISM  AND   SEMIOLOGY
Lévi-Strauss's structuralism
A familiar complaint about Lévi-Strauss's ideas concerns their lack of empirical validity.10 A number of related criticisms can be made here. It can be claimed that Lévi-Strauss's theories are supported by a highly selective and very partial use of examples, that they are simply not based upon sufficient evidence or that they are so constructed as to be impervious to any kind of empirical refutation. These claims may appear strange in that Lévi-Strauss's work is full of examples, but critics insist that these are only admitted if they are favourable to his case, and if they divert attention away from cases which might refute his theories. For example, his analysis of totemism is only possible because he confines this phenomenon to the study of myth, and does not consider how it works in relation to kinship systems. Also, his analysis of Oedipal myths is only successful because he selects those features of the stories which suit his case, and ignores others which contradict the notion that they are expressions of a universal mental structure. Related to this is the criticism that the myths he refers to cannot be interpreted in the logical manner he suggests, but must be understood in terms of the way they function in specific historical societies.
The claim that Lévi-Strauss ensures his theories are closed to empirical refutation is closely linked to the criticism that his ideas are overly abstract and theoretical. His concern with the mental structure which lies behind the myths he studies leads him to engage in cerebral exercises rather than empirical research. More than this, his notion of structure can be regarded as too abstract, its very abstractness allowing him to reach the conclusions he does. The more abstract an idea is, the more vague it is, and thus the more closed it is to empirical refutation. The abstractness of his notion of structure is equally related to its definition as a mental or psychic phenomenon. Lévi-Strauss's structuralism is marked by idealism and reductionism in that the variety and complexity of myths are reduced to the mental structure of the human mind. There are said to be two problems with his argument. First, it neglects the material processes of production whereby societies reproduce themselves, and thereby reproduce their cultures. Second, it neglects the complexity and historical and social sped-
120
STRUCTURALISM  AND   SEMIOLOGY
ficity of cultural phenomena because it reduces it all to a simple mental structure. It thus fails to provide an adequate explanation of this, complexity and specificity, and cannot account for the things which it is trying to explain except by ignoring their specific character.
Another way of appreciating this problem is to look at the claim that structuralism presents an ahistorical approach to the study of culture. We have already seen how Saussure distinguishes between synchronic and diachronic analysis. We have also seen how difficult it is to maintain this distinction in practice: it is difficult both to disentangle the uses of language over time from the formal rules which are used by speakers at any particular point in time, and to treat such rules as simply static and fixed norms. With Lévi-Strauss's work we confront this problem more directly in that he does not appear to recognise this distinction. His almost exclusive concern appears to be with synchronic analysis, uncovering the hidden and unconscious mental structure which gives rise to the myths we can observe. Insofar as his work dispenses with history, it confronts the same kind of difficulties experienced by Saussurian linguistics. Downplaying the importance of history means that the problems posed for any analysis of popular culture by historical variations in societies and cultures are simply not addressed. Indeed, it could be argued that it is impossible to understand the formal structures of language or myth outside of their social and historical contexts.
These problems are linked to the deterministic view structuralism has of the role of the subject or human agency. The major determinant of cultural myths is the logical structure of the human mind, and this exerts its power irrespective of any particular social or historical context. This means it must also exert its power irrespective of the efforts of human subjects to impose their meanings on their social world, and to attempt to alter them in different ways. Just as it is impossible to understand culture without taking its history into account, so it is equally impossible to understand it without taking account of human agency. Culture is produced, consumed and interpreted in definite social and historical circumstances, and these processes involve human agency and human
121
STRUCTURALISM   AND   SEMIOLOGY
intentions, the attribution and transformation of meaning, which cannot be spirited away at the mere mention of a universal structure. The variations in meaning entailed in the production and consumption of culture tend not to support the contention that fixed and immutable universal oppositions make cultural myths possible. The very fact that meanings can be contested suggests that the importance of the human subject cannot be so easily dismissed as structuralism argues. Rather, human agency has a leading part to play in any attempt to develop the study of popular culture. This may not aspire to the systematic qualities demanded by structuralism, but it is more in keeping with the empirical character of popular culture.
The problems structuralism has in dealing with the role of human agency can equally be detected in the importance Lévi-Strauss ascribes to the unconscious in his explanations. As we have seen, the mental structure exerts its power irrespective of the role of subjects who are unaware of what is happening. This raises questions about the empirical validity of this argument. How is it possible to validate the causal influence of something which is unconscious? If the mental structure remains unconscious we must presumably remain unaware of it, and cannot therefore talk about it in any meaningful empirical sense? Alternatively, if we can claim to demonstrate its existence how can it be unconscious?
Lastly, we can indicate the analytical difficulties associated with Lévi-Strauss's understanding of the binary opposition between nature and culture. He sees this as a basic logical opposition lying behind, and causing, the temporary reconciliations between nature and culture to be found in myths like totemism. Yet how clear and basic is this opposition? How can it be conceived of as a component of a universal mental structure lying outside specific societies and cultures, when it can only be defined in cultural terms? The concept of nature within particular societies is not 'natural' but something which is culturally defined. Lévi-Strauss does refer to the ways in which the distinction between nature and culture varies between societies, for example with respect to definitions of edible and inedible food (1970). However, rather than trying to account for this in comparative historical and sociological terms,
122
STRUCTURALISM  AND   SEMIOLOGY
he reduces it to an invariant mental structure. Clearly all societies are confronted by a nature which they have to deal with. Therefore their cultural definitions of nature can be seen as the ways they understand nature and make it meaningful. Nature can never therefore be 'innocent'; it exists as a reality which is interpreted by a society's culture. This proposition is in keeping with the arguments of semiology, which does not appear to deny the importance of culturally specific definitions of categories like nature and culture.
Roland Barthes's semiology
It can be suggested that the version of semiological analysis developed by Barthes is preferable to structuralism since it is historical in scope, and tries to relate the signs of popular culture to social forces and class interests. Barthes's approach has had a crucial influence upon popular culture studies, and yet it faces certain problems which I wish to consider in closing this chapter.11
For a start, it is hard to say whether Barthes's analyses of myth fare any better than those of Lévi-Strauss when it comes to the problem of empirical validation. While semiology, like structuralism, is presented in principle (at least in Barthes's earlier work) as a rigorous scientific method, this is not carried over into its practice. What validity does Barthes's interpretation of a particular cultural item possess? He does not attempt to indicate why his interpretation is to be preferred to others. For example, he suggests that roses signify passion. But how can he validate this conclusion, and say they should be understood in this way, and not as a way of signifying a joke, a farewell, or a platonic thank you? How do we discriminate between these interpretations? What evidence could a semiologist call upon to back up Barthes's interpretation? Similarly, semiologists are fond of referring to the codes which lie behind, or are embodied in, a particular sign or myth, but rarely if ever produce evidence of this code independently of the sign or myth under consideration. The fact that, later on, Barthes argues that texts are polysemic in being open to different interpretations hardly gets us very far.12 He is not presumably
123
STRUCTURALISM   AND   SEMIOLOGY
arguing that texts are open to an infinite number of interpretations, nor that all interpretations are equally accceptable. So why should one interpretation be preferred to another? And why should some interpretations of signs be rejected?
This lack of attention to empirical validation is evident in another problem with semiology, which concerns the attribution of meaning to myths. Reading Barthes, it is difficult not to be persuaded by the style and polish of his arguments. Yet there is little effort on his part to be empirically persuasive, so it is necessary to be sceptical about the claims he makes. One of the aims of semiology is to show how the meaning attributed to a particular myth is systematic and not arbitrary. But it can be argued that the opposite is the case. Semiology wants to demonstrate that the meanings uncovered by its approach are systematic in that they possess a comprehensive structural form and are prevalent within the society in which the myth is found. However, if the analysis is confined to the sign itself and the problem of empirical validation is ignored, it is difficult to see how this claim can be substantiated. How do we know, for example, that the conclusions offered by semiology are not the result of the subjective impressions of the analyst but an objective uncovering of the systematic structure of meaning? Indeed, is semiology better viewed as a form of textual appreciation or literary criticism than as an objective social science?
Perhaps a brief example will clarify this point. Williamson has tried to apply semiology to the analysis of magazine advertisements, and in her first analysis of an advert in her book, one for car tyres showing a car on a jetty, she writes:
the jetty is supposedly here as a test of braking power; it provides an element of risk. . . . However, the significance of the jetty is actually the opposite of risk and danger .. . the outside of the jetty resembles the outside of a tyre and the curve is suggestive of its shape . . . the jetty is tough and strong . .. because of the visual resemblance, we assume that this is true of the tyre as well. In the picture, the jetty actually encloses the car, protectively surrounding it with solidity in the middle of dangerous water; similarly, the whole safety
124
STRUCTURALISM  AND   SEMIOLOGY
of the car and driver is wrapped up in the tyre, which stands up to the elements and supports the car.
{Williamson: 1978; p. 18}
This analysis is dependent upon the idea that the jetty represents a place which is strong and safe, and that this is an expression of a wider cultural code. How else could the signification of the jetty work? But why should we assume that people will regard a jetty as a place of safety no matter how strong and secure it may appear? In fact, Williamson's attribution of meaning, which equates the jetty with safety, is totally arbitrary. Accordingly, the implication that it is indicative of a cultural code is unfounded.
There are some related problems associated with the semi-ological analysis of popular culture. Much is made of Barthes's distinction between denotation and connotation. It is argued that myth works because we see the denotations of a particular sign or myth but its connotations remain hidden until they are revealed to us by a semiologist. Yet is there such a thing as pure denotation? Are not the connotations of a sign as clear, if not sometimes more clear, than its denotations? It can be argued that significations like adverts, photos or roses never exist as simple denotations but are always overlain by some kind of cultural interpretation. In this sense, they never lack meaning. Formally it is possible to define signs in a technical or denotative manner; but signs always exist in some kind of social context and are always available to some kind of interpretation. Moreover, in so far as interpretations of signs are offered in terms of their connotations, without being backed up by independent evidence, there is no reason why the connotations of a particular sign should not be readily apparent. After all, Barthes sits down in the hairdressers, sees the photo of the soldier on the cover of a magazine and quickly works out its connotations. The connotations of myths are more contestable than Barthes realises, but what we are concerned with here is the fact that the connotations of cultural signs and myths may not be so hidden or as difficult to discern as semiologists argue.
As I indicated above, a major problem with the semiological study of signs is the way it neglects the contexts in which signs are
125
�226�����81367033073164837458925
STRUCTURALISM   AND   SEMIOLOGY
used as forms of communication. The question raised here is whether signs can be adequately understood if they are divorced from the contexts in which they are used and interpreted. For example, how can we know that a bunch of roses signifies passion unless we also know the intention of the sender and the reaction of the receiver, and the kind of relationship they are involved in. If they are lovers and accept the conventions of giving and receiving flowers as an aspect of romantic, sexual love, then we might accept Barthes's interpretation. But if we do this, we do so on the basis not of the sign but of the social relationships in which we can locate the sign. Moreover, if we accept the interpretation of the sign Barthes proposes, and he makes no attempt to indicate the social relationships in which it is to be found, how do we know that intentions and relationships are involved which do not concern passion? The roses may also be sent as a joke, an insult, a sign of gratitude, and so on. They may indicate passion on the part of the sender but repulsion on the part of the receiver; they may signify family relations between grandparents and grandchildren rather than relations between lovers, and so on. They might even connote sexual harrassment. The point here is that it is impossible to interpret signs adequately unless their contexts of use, and the social relationships which confer meaning upon them, are taken into consideration. Semiology does not recognise that meaning is not a quality of the sign itself but of the social relationships in which it can be located.
This point about the social context of signs can be taken further if it is defined a bit more precisely. For a start, signs are implicated in social relationships in that they have to be produced in order to be culturally available as signs. It is a familiar complaint about semiology that it ignores the context of production. Cultural signs, like magazines, are produced by industries on the basis of their marketability and profitability. They are among the commodities which are produced, circulated and consumed in a capitalist society. In view of Barthes's understanding of the bourgeois interests served by the ideology to be discovered in cultural signs, it might appear to be the case that his semiology is not incompatible with this Marxist perspective. However, for the latter
126
STRUCTURALISM   AND   SEMIOLOGY
it is the way production generates meaning which is crucial rather than the decoding of signs, which ignores the context of industrial production. This problem is, in part, dealt with by introducing the concept of encoding. The idea here is that cultural producers consciously or unconsciously (usually the latter) instill meanings into cultural products which are then decoded or interpreted by audiences in relatively diverse and independent ways which are none the less, in the final analysis, in keeping with some general dominant ideology. The concepts of semiology are used to render the Marxist theory of ideology less deterministic and instrumental. However, this still tends to underestimate the ways in which what is produced is itself subject to conflicts and negotiations, and how the meanings produced may not be uniform, consistent, unambiguous or reducible to a coherent dominant ideology.13
This, in turn, raises the issue of the decodings made of signs by the people at which they are directed. The basic question here is why should the interpretation of signs and myths offered by semiology be accepted if they take no account of the interpretations placed upon them by their audiences? On what grounds can semiologists argue that their understanding of popular cultural signs is adequate if it neglects those groups who consume these signs. In part, this relates back to the fact that semiology fails to tackle the problem of justifying empirically its interpretations. In part, it also relates to the fact that semiology neglects the social relationships in which signs are produced and consumed. But it equally concerns how the meaningful character of popular culture can be determined. It would seem to me to be the case that this cannot be done without researching the part that audiences play in arriving at interpretations of forms of popular culture. These forms cannot possess any level of social meaning unless the meanings generated by audiences are taken into account. If a text has no audience, how can it have any meaning? It is only the fact that popular culture attracts audiences which gives it sociological importance.
I have indicated how, in my view, Barthes has a fairly crude view of ideology. He sees it as functioning in the interests of the bourgeoisie. It is this theory which introduces the concept
127
STRUCTURALISM  AND   SEMIOLOGY
of ideology into semiological analysis since the connotations and signifieds of signs are, in the end, reduced to bourgeois ideology.14 To appreciate the problematic character of this concept of ideology it is necessary to take into consideration the arguments of the next chapter.
Further reading
Barker, M. (1989) Comics: Ideology, Power and the Critics, Manchester,
Manchester University Press (chapters 6 and 7). Barthes, R. (1968) Elements of Semiology, New York, Hill and Wang.
------(1973) Mythologies, London, Paladin Books.
Craib, I. (1984) Modern Social Theory, London and New York, Harvester
Wheatsheaf (chapter 7). Culler, J. (1983) Barthes, London, Fontana. Dyer, G. (1982) Advertising as Communication, London and New York,
Methuen (chapter 6). Fiske, J. and Hartley, J. (1978) Reading Television, London and New
York, Methuen. Leach, E. (1970) Lévi-Strauss, London, Fontana. Lévi-Strauss, C. (1969) Totemism, llarmondsworth, Penguin. Sturrock, J.  (ed.)  (1979)  Structuralism  and Since,  Oxford,  Oxford
University Press. Woollacott, J. (1982) 'Messages and meanings', in M. Gurevitch et al.
(cds), Culture, Society and the Media, London, Methuen.
128
Chapter   4
Marxism, political economy and ideology
■     Marx and ideology                                 130
■     Marxism and political economy           136
■     The limits of political economy            142
■    Althusscr's theory of ideology arid structuralist Marxism                             146
■     Althusser's Marxism: economic determinism and ideology                      155
■     Gramsci, Marxism and popular
culture                                                       160
■     Gramsci's concept of hegemony           165
■    Conclusions: Marxism, Gramscian Marxism and popular culture               171
1 29