Roland Barthes, 'Le discours de l'histoire' (1967), in Barthes, Le Bruissement de la langue (Paris, 1984); English translation in Structuralism: A Reader, ed. Michael Lane (London, 1970), and in Comparative Criticism, 3 (1981).
Barthes has less of the system-building ambition of Hayden White; in the end, however, he is the more radical anti-realist. He wants to resist the impression that 'history writes itself' (that 'the facts speak for themselves'), partly by showing the presence of the author and of authorial labour in the text, and partly by unmasking the apparent objectivity of the text as just one more feature of its rhetoric.
His method is essentially to scan the historical text for traces of the writing subject.
These he finds firstly in what he calls 'shifters' (borrowing the term from the linguistician Roman Jakobson): 'testimonial shifters' which mark the presence of the historian as someone who gathers and weighs up evidence from other sources; 'organizing shifters', where the narration diverges from the chronological sequence and reorganizes it according to the needs of the discourse (which thereby tends to establish its own complex temporal structure); 'shifters' marking interventions of the writer or appeals to the listener, or (more commonly) the significant absence of such elements (the avoidance of first-person pronouns, for instance, in order to give an impression of objectivity).
Secondly, looking at the things referred to in the text (persons, events, etc.), Barthes looks for signs of their use by the historians to make meaningful patterns. These can be summarizing labels which give a thematic unity to a complex reality; details which build up certain positive or negative associations around events and figures ('indices'); passages of reasoning ('enthymemes'); or nodal points ('functions') of narrative (such and such an event points forward to the need for one or other resolution). (Note: You will understand this section better if you also read: Barthes, Introduction à l'analyse structurale du récit (1966), Collection Points, 129 (Paris, 1981); English translation in Barthes, Image-Music-Text, tr. Stephen Heath (London, 1977; repr. 1993) and in Barthes, Selected Writings, ed. Susan Sontag (London, 1982), pp. 251-295 (also publ. as A Barthes Reader).)
A final section of his essay discusses ideology in the historical text, which Barthes finds less in the explicit value-judgements by which the text is punctuated and structured than in the more insidious 'effect of the real', which according to him results from a kind of bad faith unique to historical writing: its claim that the objects it describes are simply 'out there' and that it is merely reflecting them, whereas in fact the impression of solid truth history gives is a product of its own rhetorical energy.
k.f. hilliard
c:\kfh\teaching\lectures\histlink\brthesho.doc
29/01/97 17:19