In recent years a considerable degree of interest has developed in the semiology of the cinema, in the question whether it is possible to dissolve cinema criticism and cinema aesthetics into a special province of the general science of signs. It has become increasingly clear that traditional theories of film language and film grammar, which grew up spontaneously over the years, need to be reexamined and related to the established discipline of linguistics. If the concept of ‘language’ is to be used it must be used scientifically and not simply as a loose, though suggestive, metaphor. The debate which has arisen in France and Italy, around the work of Roland Barthes, Christian Metz, Pier Paolo Pasolini and Umberto Eco, points in this direction.

The main impulse behind the work of these critics and semiologists springs from Ferdinand de Saussure’s Course in General Linguistics. After Saussure’s death in 1913 his former pupils at the University of Geneva collected and collated his lecture outlines and their own notes and synthesized these into a systematic presentation, which was published in Geneva in 1915. In the Course Saussure predicted a new science, the science of semiology. ‘A science that studies the life of signs within society is conceivable; it would be part of social psychology and consequently of general psychology; I shall call it semiology (from Greek semeion “sign”). Semiology would show what constitutes signs, what laws govern them. Since the science does not yet exist, no one can say what it would be; but it has a right to existence, a place staked out in advance. Linguistics is only a part of the general science of semiology; the laws discovered by semiology will be applicable to linguistics, and the latter will circumscribe a well-defined area within the mass of anthropological facts.’ Saussure, who was impressed by the work of Durkheim in sociology, emphasized that signs must be studied from a social view-point, that language was a social institution which eluded the individual will. The linguistic system—what might nowadays be called the ‘code’—pre-existed the individual act of speech, the ‘message’. Study of the system therefore had logical priority.

Saussure stressed, as his first principle, the arbitrary nature of the sign. The signifier (the sound-image o-k-s or b-ö-f, for example) has no natural connection with the signified (the concept ‘ox’). To use Saussure’s term, the sign is ‘unmotivated’. Saussure was not certain what the full implications of the arbitrary nature of the linguistic sign were for semiology. ‘When semiology becomes organized as a science, the question will arise whether or not it properly includes modes of expression based on completely natural signs, such as pantomime. Supposing the new science welcomes them, its main concern will still be the whole group of systems grounded on the arbitrariness of the sign. In fact, every means of expression used in society is based, in principle, on collective behaviour or—what amounts to the same thing—on convention. Polite formulas, for instance, though often imbued with a certain natural expressiveness (as in the case of a Chinese who greets his emperor by bowing down to the ground nine time), are nonetheless fixed by rule; it is this rule and not the intrinsic value of the gestures that obliges one to use them. Signs that are wholly arbitrary realize better than the others the ideal of the semiological process; that is why language, the most complex and universal of all systems of expression, is also the most characteristic; in this sense linguistics can become the master-pattern for all branches of semiology although language is only one particular semiological system.’

Linguistics was to be both a special province of semiology and, at the same time, the master-pattern (‘le patron général’) for the various other provinces. All the provinces, however—or, at least, the central ones— were to have as their object systems ‘grounded on the arbitrariness of the sign’. These systems, in the event, proved hard to find. Would be semiologists found themselves limited to such micro-languages as the language of traffic-signs, the language of fans, ships’ signalling systems, the language of gesture among Trappist monks, various kinds of semaphore and so on. These micro-languages proved extremely restricted cases, capable of articulating a very sparse semantic range.

Many of them were parastitic on verbal language proper. Roland Barthes, as a result of his researches into the language of costume and fashion concluded that it was impossible to escape the pervasive presence of verbal language. Words enter into discourse of another order either to fix an ambiguous meaning, like a label or a title, or to contribute to the meaning what cannot otherwise be communicated, like the words in the bubbles in a strip-cartoon. Words either anchor meaning or convey it.

It is only in very rare cases that non-verbal systems can exist without auxiliary support from the verbal code. Even highly developed and intellectualized systems like painting and music constantly have recourse to words, particularly at a popular level: songs, cartoons, posters. Indeed, it would be possible to write the history of painting as a function of the shifting relation between words and images. One of the main achievements of the Renaissance was to banish words from the picture-space. Yet words repeatedly forced themselves back; they re-appear in the paintings of El Greco, for instance, in Dürer, in Hogarth: one could give countless examples. In the twentieth century words have returned with a vengeance. In music, words were not banished until the beginning of the seventeenth century; they have asserted themselves in opera, in oratorio, in lieder. The cinema is another obvious case in point. Few silent films were made without inter-titles. Erwin Panovsky has recollected his cinema-going days in Berlin around 1910: ‘The producers employed means of clarification similar to those we find in medieval art. One of these were printed titles or letters, striking equivalents of the medieval tituli and scrolls (at a still earlier date there even used to be explainers who would say, viva voce, “Now he thinks his wife is dead but she isn’t” or “I don’t wish to offend the ladies in the audience but I doubt that any of them would have done that much for her child”).’ In Japan, ‘explainers’ of this kind formed themselves into a guild, which proved strong enough to delay the advent of the talkie.

In the end Barthes reached the conclusion that semiology might be better seen as a branch of linguistics, rather than the other way round. This seems a desperate conclusion. The province turns out to be so much ‘the most complex and universal’ that it engulfs the whole. Yet our experience of cinema suggests that great complexity of meaning can be expressed through the images. Thus, to take an obvious example, the most trivial and banal book can be made into an extremely interesting and, to all appearances, significant film; reading a screenplay is usually a barren and arid experience, intellectually as well as emotionally. The implication of this is that it is not only systems exclusively ‘grounded on the arbitrariness of the sign’ which are expressive and meaningful. ‘Natural signs’ cannot be so readily dismissed as Saussure imagined. It is this demand for the re-integration of the natural sign into semiology which led Christian Metz, a disciple of Barthes, to declare that cinema is indeed a language, but a language without a code (without a ‘langue’, to use Saussure’s term). It is a language because it has texts; there is a meaningful discourse. But, unlike verbal language, it cannot be referred back to a pre-existent code. Metz’s position involves him in a considerable number of problems which he never satisfactorily surmounts; he is forced back to the concept of ‘a “logic of implication” by which the image becomes language’; he quotes with approval Bela Balazs’s contention that it is through a ‘current of induction’ that we make sense of a film. It is not made clear whether we have to learn this logic or whether it is natural. And it is difficult to see how concepts like ‘logic of implication’ and ‘current of induction’ can be integrated into the theory of semiology.