CHAPTER TWO
creator and interpreters
93
Creator and Interpreters
Don't take it as a matter of course, but as a remarkable fact, that pictures and fictitious narratives give us pleasure, occupy our minds.
Wittgenstein1
The negative view of language that emerged in French thought during the 1960s and 70s was the culmination of one strand in the general upheaval of attitudes which characterised those years — in some areas, for ,some thinkers. Certainly that upheaval left its mark on the human sciences (theories of language, literature, philosophy, politics, anthropology), and some of its effects on Shakespeare studies can now be seen. But to anyone who uses libraries and bookshops, keeps an eye on publisher's lists, reads newspapers and journals {other than those of a specific party or group), talks to fellow-teachers and students — to such a person observing the whole spectrum of current intellectual activity it will be dear that the so-called 'critical revolution' was the work of a small group of writers, and has been followed up by a relatively small group of readers and publicists. The great majority of scholars, critics, book-reviewers, journalists, general readers, teachers, and students have remained unaffected by the forward march of critical theory, as it has advanced from Paris to Baltimore, and from New Haven to Cardiff.
The proportionally small number of people involved is not an argument . for ignoring the various challenges to accepted attitudes, or for dismissing them with the response 'empty vessels make the most noise'. Ignoring the things you don't like won't make them go away, and refusal to take part in debate with theories that challenge your own merely exposes you to the charge that you cannot defend your theories, either because you're unaware that you have any, or because they are so feeble or contradictory as to be indefensible. (This charge now applies, in fact, to the exponents of Lacan, Derrida, Althusser et Cie., who regularly ignore or dismiss any criticism as 'ignorant' or 'reactionary'.) In this chapter I want to give a reasoned, defence of a theory of literature which has survived the 1960s revolt, and will continue to do so, since it actually functions as, a theory, being derived from the object studied and having a genuine explanatory role. It has also produced some powerful analytical work from a number of different fields in recent years (philosophy of art, literary history, speech-act theory, discourse analysis), with an impressive convergence of thinking on the major; issues of meaning, intention, interpretation, representation. The gap;
between this tradition, developing fruitfully in several directions, and the anti-svstems, locked in their 60s configuration, grows wider every month.
On the one hand are those who see' a work of literature as an organisation of words set down by an author at a specific point in time according to the expectations of a particular literary convention. (He may, of course, subvert these conventions, as Sterne does in Tristram Shandy, but to appreciate that novelist's wit we still need to know what he was subverting.) In fictional works, which give a representation of human beings interacting in situations and in ways not unlike life as we know it, our involvement is with the characters conceived of within their 'possible worlds'2 as free agents, responsible for their own actions, towards whom we express reactions which are both ethical and emotional (love, hate, fear, disgust). To those who share this: conception of a writer creating-fictional worlds which can touch our emotions and moral feelings, reading works of literature is both an enjoyable and a serious activity to which they are prepared to devote a great deal of time, energy, substance. They believe that the experience will enrich but also disturb their lives, provoke them to reconsider human situations and reactions — including their own — and not necessarily leave them in a state of reassurance.3 Works of art, this type of theorising holds, have a finite form, are delimited in time and structure, the best of them possessing a completeness which can be emotionally satisfying on. a formal plane, although the feelings aroused during our experience of it may live ori after the work is over. Tristan und Isolde produces in me a feeling of loss and waste that persists long after the performance ends; Die Meistersinger arouses a sense of euphoria even before it begins. These personal observations, trivial to everyone else, are meant to make the point that bur experience of a work of art is complicated, involving aesthetic, ethical, psychological, philosophical responses.
On the other side are the exponents of 'radical literary theory', as they still call it, who reject virtually every concept ori category that I have just used. The author as such does not exist, any more than the subject (individual) does. There can be no distinction between works of literature or other written forms since all are texts, all manifest the operation of textuality, which both shapes arid constrains utterances. Discourse: is an
; impersonal circulation of language, subject to contestation and appropriated by - interested power groups. Language cannot refer to reality, works
5 of literature by definition suffer from the same limitation. If mimesis or
^representation is rejected, then the notion of character must go, too.
i As Denis Donoghue defined 'graphireading', his1 name for the school of Derrida, Barthes and de Man, the exemplary stance is to 'put under the
.greatest possible stress such words as these: self, subject, author, imagination, story, history, and reference' (Donoghue 1981, p. 200). • Anyone who has followed these developments in contemporary literary theory is entitled to ask, 'What, then, are we left with?' The answer would
' seem to be: a collection of negativities, of positions or principles negated,
94
critical theories
creator and interpreters
95
evacuated, destroyed. No replacements have been erected, since no one 'at the cutting edge' of radical theory believes that any ground can be discovered on which firm theoretical structures can be built. John Ellis's account of the peculiarity of deconstruction, that it attacks accepted models without attempting to replace them, indeed preserves them as if to validate its own success (Ellis 1989, pp. 70—74) — a David, cast in lead, standing on the heads of a hundred Goliaths -— can be applied to Current Literary Theory in general. It has rejected the extant system but has no replacement for it. In that sense it reminds me of Nietzsche's definition of the philosophical nihilist as one who 'is convinced that all that happens is meaningless and in vain; and that there ought not to be anything meaningless and in vain'.4 So, in this interregnum, seated on the bare ground surrounded by ruins of his own making, the contemporary theorist may experience-.a sense of triumph, but may also be wondering what on earth to do next.
The situation in advanced literary theory today, if one were to accept all the radical proposals put forward by all the interest groups, would be a bit like the seventeenth-century response to Copernicus' destruction of a geocentric universe. As Donne described it in his First Anniversary (1611), the .'new Philosophy calls all in doubt', and men frantically seek new worlds in the universe because
they see that this Is crumbled out againe to his Atomis. 'Tis all in pieces, all cohaerence gone; Air just supply, and all Relation. . . .5
The attacks on established critical theory since the 1960s, however, do not amount to a Copernican revolution, for they offer no alternative model to replace what they have attacked. Theirs is an anti-system, wanting to annihilate every accepted notion, destroying whatever coherence the previous system had. But it does not constitute a true revolution of ideas, as Thomas Kuhn has described it, which depends on maintaining an 'essential tension1 between the accepted consensus and those areas in which innovation is being pursued.6 The successful scientist, Kuhn writes, 'must simultaneously display the characteristics of the traditionalist and of the iconoclast' (p. 227), for it is not violent opposition but.'the continuing attempt to elucidate a currently received tradition' which finally produces 'one of those shifts in fundamental theory' that make a scientific revolution possible. The accepted consensus arises out of a convergence of ideas and attitudes which defines a problem and focusses the energy of thought on it in a concerted way. Paradoxical though it may seem, 'work within a well-defined and deeply ingrained tradition seems more productive of tradition-shattering novelties than work in which no similarly convergent standards are involved' (p. 234). In the purely iconoclastic criticism that has reigned in some areas since the 1960s all traditions have been swept contemptuously
aside, leaving not so much a tabula rasa as a ruin, or a ground so thoroughly devastated that (it is hoped) no one will build there again.
Yet, for all its polemical energy, the iconoclasts' attack on extant literary theory is both merely negative and poorly argued, indeed its success has come from readers being persuaded to accept polemic rather than argument, and to naively trust the polemicist. The uncritical acceptance of their anti-theories has derived from the credulity of their audience, its readiness to take at face value the iconoclast's description of the position that he is attacking. For instance, the reception of Derrida's work, Peter. Dews observes, has been 'marked by an astonishingly casual and unquestioning acceptance of certain extremely condensed -— not to say sloganistic — characterisations of the history of Western thought, as if this history could be dismissed through its reduction to a set of perfunctory dualisms' (Dews 1987, p. xv). The hundreds of articles that unquestioningly reproduce Derrida's polemic manifest a striking (sophisticated and well-informed though it imagines itself to be) credulity, which believes that just because a position has been questioned, the questioner must have a case, the position must have a weakness. So Frank Lentricchia observes that intentionality is 'a category long under attack in modern critical theory' (Lentricchia 1983, p. 115) —but, he might have added, one that has also had some powerful and cogent support from, philosophers, linguists, and literary critics. Janel Mueller, introducing a special issue of Modern Philology, and therefore professionally obliged to demonstrate an awareness of current trends* records her feeling that many 'presuppositions about language use and language users . . . are deeply discredited at present. . . . In the wake of two decades of intense theorising', she feels, an atmosphere of 'skepticism' has been produced, largely by deconstruction, which is 'radically skeptical" with its 'aporias and indeterminacies .. .'.7 But to many people unpersuaded by its programme or methods, deconstruction has displaced nothing, although it has certainly placed its most successful exponents in tenured positions. Universities that ate anxious not to appear outmoded have no option but to give place to those who accuse them of being so. The fact that deconstructionist critics have jobs speaks much for the catholicity (or vulnerability) of the departments that have hired them, but it hardly means that language and its users are now being institutionally discredited, too. ;
In some ways the conquest has been too easy. Those who have been willing to concede crucial philosophical positions to their critics have had little notion of the meaning or value of what they abandoned. Those with a superficial knowledge are easily impressed, and easily converted. The vogue for post-structuralist theory had faded in France by the 1970s, giving way to the rise of 'nouvelle philosophie', a new interest in Kant, and a revival of phenomenology and hermeneutics, so bitterly traduced and .travestied by the structuralists and their successors (Dews 1987, pp. xii— xiii), but it persists in America, especially. In 1986 Julia Kristeva, long-
96
critical theories
creator and interpreters
97
time associate of Philippe Sollers and the Tel Quel group, one of the most intelligent and knowledgeable survivors of that generation, commented on the continuing naive acceptance in America of the polemical successes claimed by deconstruction:
'In America, the so-called deconstructionists think that', because ethics and history belong to metaphysics and because metaphysics is criticized by Heidegger or his French followers, ethics and history no longer exist.'
Lecturing in America, she added, 1 "somebody in the audience always asks why I speak about ethics and history when those notions already have been deconstructed. Not even the most dogmatic French deconstructionists ask such questions"' (cit. Lehman 1991, pp. 55-6).
Lest 1 be accused of anti-Americanism, let me take an example from nearer home to show' that we ought not simply to believe iconoclasts when they claim to have destroyed some monument or other. FiR. Leavis, who disliked the largely unthinking endorsement of Milton in the "1920s as a major English poet, took 'the few critical asides' of T.S. Eliot and 'an acute page or two' by Middleton Murry as having constituted a decisive blow t:o his reputation, and began his chapter on Milton in Revaluations (1936) with the coolly triumphant words: 'Milton's dislodgement, in the past decade, after his two centuries of predominance, was effected with remarkably little fuss.' In the same year T.S. Eliot published a polemical essay on Milton, accusing him of 'having done damage to the English language from which it has not wholly recovered', and the dislodgement seemed complete. However, in 1947 Eliot publicly recanted,8 and the status of Milton as a major English poet — on rather different grounds than those held by Robert Bridges, needless to say — now seems beyond question. Current Literary Theorists who think they have totally demolished character, or representation, or the author, may be surprised if they look again, to see that life goes on undisturbed by their intervention.
DESTRUCTIVE CRITICAL THEORIES
I went to the Garden of Love And saw what I never had seen: A Chapel was built in the midst, Where 1 used to play on the green.
And the gates of this Chapel were shut,
And Thou shalt not, writ over the door —
And Priests in black gowns were walking their rounds,
And binding with briars my joys & desires.
Blake9
The 'intellectual origins' of European literary theory in the 1960s, Edward Said observes, were 'insurrectionary. The traditional university, the
hegemony of determinism and positivism, the reification of ideological bourgeois "humanism", the rigid barriers between academic specialities', these were some of the targets which unified literary theory in attacking the status quo (Said 1983, p. 3). Similarly, Luc Ferry and Alain Renaut's study of La pensée 68, while denying that French philosophy of the 1960s can be summed up under that title, nevertheless presents May 1968 as the catalyst which precipitated many of the issues in that adversarial mode which continues to absorb so-called radical theorists in the 1990s. These include anti-humanism, which pronounced the 'death of man', and claimed that the; autonomy of the subject was illusory; the theme of the end of philosophy; the dissolution of tiie idea of truth; the historicisation and relatiyisation of all categories, and a denial of universal values (Ferry and Renaut 1985, pp. 11-103). The thrust of this reaction against established ideas, established figures, was violently oppositional, not just qualifying but dismissing altogether previously respected figures ;in French thought; Descartes, Sartre, Merleau-Ponty. While historically understandable, precisely its nature as a French reaction ought to have limited its value and relevance in other cultures. If, rejecting their own tradition, and rushing to find new; authority figures in Marx, Nietzsche, Freud, Heidegger, one school of French thinkers reduced itself to 'the hyperbolic repetition of German philosophy' (ibid., p. 46), that is a reaction that English, German, or,Italian philosophers, who start with other traditions, would find it hard to share. A philosopher who has never taken the Cartesian cogito as the centre and circumference of philosophical, enquiry will hardly need to reject the concept of a subject so violently.
We must reconstruct the French cultural context to understand the forms .that this reaction took, forms that are still at work (long after they have disappeared in France) in the English-speaking world. The extraordinary intellectual energies produced in Paris, which made it for so long the goal of; all forward-looking intellectuals, were in many respects autocentric, preoccupied with their own traditions and not well informed of Others. The fact that Derrida, for instance, oriented himself so strongly to domesticating German philosophy in France, meant that he was largely unaware of the developments in linguistics and the. philosophy of language in America and England. As John Ellis has shown* ignorance of. this tradition led to Derridian deconstruction presenting as an original critique (as of linguistic essentialism, for instance) a position which had long been worked out, with more detail and cogency, in the Anglo-Saxon world (Ellis 1989, pp. 37-8, 42-4, 142). The fact that so many English-speaking literary critics have taken Derrida's claims to innovation and iconoclašra at face value, in ignorance of their own native traditions, can only be explained by an uncritical acceptance of 'the new' which ranks in its passivity with the medieval deference to the auctoritas of virtually any written record. Ellis also observes that some of the literary targets that Barthes and others attacked were specifically French :phenomena, such as the notion of an objective literary history as presented by an authority
98 critical theories
figure like Gustave Lanson (ibid., pp. 83-6; Bárthes 1972, pp. 256-7). One could add that a standardised national examination system (the Baccalauréat, Agřégation, GAPES), can only encourage a centralised corpus of authoritiative interpretations. But English and American critics, who have never been confronted with this degree of regimentation, hardly need to attack it so violently.
Understandable within a French context in the 1960s, many of the polemics of radical literary theory as translated into the Anglo-American world of the 1970s and 80s ('Say, heart, what doth the future bring?'), look distinctly odd. Their targets seem unreal, straw men, opinions that have been manufactured only to be attacked, only to have something to attack. One form taken by this polemic was to erect a binary opposition, in which category I represented some absurdly stupid or superficial point of view, to be mocked with withering scorn, and replaced by category II, in all respects superior, and which was recommended without being subjected to critical enquiry. This is an ancient debating technique, already brought to a high peak by Plato, as Cha'im Perelman, restorer of a rhetoric based on philosophical argumentation, showed so well.10 In all such oppositions, as between 'appearance' (Term f) and 'reality' (Term ÍÍ), it is the second term that provides the criterion, the unexamined norm to which value is attached in order to devalue the opposed pole (I), here seen as 'merely illusion and error' (pp. 416-17). But in fact the dichotomy begs a number of questions ('appearance' is in many ways a less problematic concept than 'reality'). All these dichotomies are meant to be accepted at face value, and those who use such pairs introduce them, not as term? open to dis> cussion but as data, 'instruments that make it possible to structure the discourse in a manner that appears objective' (p. 422; my italics). But in fact they are designed to weight the discussion and to put one's interlocutor in the wrong. 'For in argumentation, what one person terms appearance is generally what was reality to someone else, or was confused with reality' (p. 424), so that to introduce the concept of error is 'to assert that there is a rule', which one's opponent is either ignorant of or' has broken knowingly (p. 425). All such binary oppositions ultimately express the simple attitude of'I'm right, you're wrong'.
The most persistent users of these excluding dichotomies have been the .| deconstructionists. Derrida deployed many such oppositions in order to privilege his own ideas and disqualify his opponents'. In Of Grarmnatoiogy, for instance, he set 'linearity' in language in the disfavoured pole i, opposing itself (note the cunning formulation) to 'pluridimensionálity': who would not prefer the latter? What Derrida mockingly described as linear thought (Term I, triumphing in the Enlightenment and in modern science) 'paralyses' and 'reduces' history; its pluridimensional converse (Term ÍI) defeats successivity (Derrida 1976, pp. 85-7). At the end of his 1966 attack on structuralism Derrida placed in the inauspicious, left-hand category Rousseau (and by implication, Lévi-Strauss), nostalgia, guilt,
creator and interpreters
99
origin, presence, man, humanism, metaphysics, ontptheology — the whole Derridian 'hit list'. (This mode of itemising what you hate resembles the tabedcte defixionum or curse-tablets which practitioners of magic in classical antiquity employed to bring disaster on their enemies, inscribing their ill-wish on a sheet of lead and burying it in some gloomy spot, near graves and water.) On the other, right-hand or pre-eminent side, winging their way into the sun, their wings tinged with gold, Derrida placed Nietzsche (and by implication., Derrida too), play, and 'a world of signs , . . without origin' (Derrida 1978, p. 292). As several commentators observe, Derrida's dismissal of logocentrism as an illusion was intended to put all of Western philosophy into the Platonist dustbin, his system alone perceiving the true reality (Pavel 1989, p. 15; Butler 1984, p- 64). These claims to superior certainty can also take the form of crudely branding any system other than deconstruction as 'naive objectivism' (Dews 1987, p. 11). Derrida's whole attack on the referential dimension of language as having only a secondary and derivative status emerged from his strategic handling of 'a simple opposition between identity and difference1. Yet, Peter Dews counters, 'nothing obliges us to accept Derrida's initial dichotomy', which would result in the perceived world being reduced to 'no more than a system of traces' (ibid., p. 115). Dews sets us an example of independence, that crucial position in all serious intellectual exchange, it seems to me, of refusing to accept the terms in which an argument is framed if the effect will be to drive us into a corner. Coercive dichotomies are intended to do just that, trap you in a cul-de-sac from which you can only yearn after the illusory vista on the other side.
I draw attention to this sophistic trick since Current Literary Theory has made great use of it, paradoxical though it may seem, to call in question the possibility of interpretation. As M.H. Abrams was the first to point out, Derrida made absurdly high demands on the terms traditionally used 'to analyze the workings of language — terms such as communication,
■ ■context, intention, meaning1 —• demanding that 'in order to communicate "a •determinate content, an identifiable meaning", each of these words must signify a concept "that is unique, univocal, rigorously controllable"', and
: so forth. Abrams made the essential objection, that ho words can meet 'these criteria of absolute fixity, purity and singularity', since language does
■i not work like that. The 'finite set of words' in language, and the 'finite set
> of regularities' governing them, can, however, generate 'an unlimited variety of utterances adaptable to an unlimited diversity of circumstances. . .' (Abrams 1979, pp. 275-6). Instead of examining,the actual workings of language, 'Derrick's aU-or-none principle admits;of.no alternative: failing
; to meet absolute criteria which language cannot satisfy without ceasing to do.ifs work, all spoken and written utterances... are deconstructable into
(.semantic indeterminacy' (ibid., p. 276). Several critics have commented on ■Derrida's habitual either-or, all-my-way-or-no-way type of argument,11
i which has resulted in one of the strangest developments in modern literary
100
critical theories
creator and interpreters
101
theory, the idea that since no interpretation can be single and un-ambivalent, there can be no interpretation. This claim, if it were true, would mean the collapse of all the varying critical approaches to literature, and indeed of literary theory as a whole. Although it has often been refuted, since the critical avant-garde maintains a protective silence over criticism that devastates its foundations, the counter-arguments may be worth rehearsing again.
Followers of Derrida have made great play with this false opposition, typically collapsing the poles into one shocking, attention-catching formula combining assertion and denial (this is another rhetorical figure, synoetiosis). So Paul de Man declared that 'Interpretation is nothing but the possibility of error' (de Man 1971, p. 141), a statement to which, Wendell Harris replied, 'the appropriate analogue would be that "food is nothing but the possibility of poison'" (Harris 1988, p. 160). In a later book de Man asserted that 'any narrative is primarily the allegory of its own reading' but at once added, 'the allegory of reading narrates the impossibility of reading' (de Man 1979, pp. 76—7), Whatever form the assertion took in de Man, its negation followed hard on its heels: 'allegorical narratives tell the story of the failure to read' (p. 205); 'the allegory of unreadability begins by making its pretext highly readable' (p. 216). In this book dc Man's insistence on the collapse not just of reading but of meaning and reference reached an obsessive degree.12 Less obsessively, but equally categorically, J. Hillis Miller asserted without exception that 'all reading is misreading' (Miller 1976c, p; 98), and that 'any reading can be shown to be a misreading on evidence drawn from,the text itself (Miller 1976b, p. 333). The contradiction within these positions — that we know that all reading is misreading from the evidence of Miller's own text, which claims tobe the true reading — has not bothered this school of theorists. One of the most provocative-seeming statements in this line was Miller's claim that Shelley's Tke Triumph of Life
can never be reduced to any 'univocal' reading. . . The poem, like all texts, is 'unreadable', if by 'readable' one means a single, definitive interpretation. (Miller 1979, p. 226)
To which Gerald Graff rightly objected that 'the concept of a "single" interpretation could not be legitimately maintained by anybody since it is nonsensical: there are no units for segmentary interpretations' into 'one' or 'many'. And since no one claims that a 'definitive' reading is possible, to parody such dogmatism by 'merely placing the prefix un- before the word' gets us nowhere: 'a third possibility is needed' (Graff 1980, p. 421), that vast 'excluded middle' ground so often ignored by contemporary critics.
As for the claim, parroted by so many popularisers of deconstruction, that 'all interpretation is misinterpretation', John Ellis has produced a definitive demolition, which I shall not attempt to summarise (Ellis 1989, pp. 97-112). But I select and endorse one point, that it is precisely the j
'categorical foim of the assertion [that] dooms it', making it either trivial or false:
For example: either interpretation is never final (trivial) or it is always suspect to the same degree (false); either bias is a constant problem to watch in interpretation (trivial) or bias is always damaging to the same degree or in the same way (false). All the important theoretical issues involved in interpretation require. . . making distinctions between different cases....
But any chance of contributing to the necessary process of differentiation is destroyed from the outset by 'the grandiose form of the categorical claim'. For Ellis the appropriate judgement on this proposition 'is not that it is wrong but that it is empfy', having no 'theoretical content! (pp. 107—8). That judgement applies to many of the sweeping assertions made by current literary theory, seemingly exception-free, on closer inspection found to be hollow. Yet, as their reception proves, the confidence conveyed can be rhetorically effective in capturing and hypnotising the unwary reader. As Ernest Gellner has observed of Freudian psychoanalytic theory,
Sheer confident assertion, in a world. . . in which no one of any sense has any grounds to be confident of either certainty or doubt,; makes a certain impact. . . . The fact that the confidence is devoid of visible means of support hardly seems1 to diminish its authority. In fact, were its means of support visible, they could be subject to evaluation; arid the assertion would have. . . conceded its own ordinary, human status —
■ making its own contentions arguable, negotiable. ... But bare, brazen, unnegotiated assertion, if skilfully presented, can have a kind of stark
■ authority. (Gellner 1985, pp. 42-3) '
Despite the deconstructionists' frequent assertions of scepticism and doubt, the mode they use to communicate their critique is just this dogmatic assertiveness. As we know so well from other contexts in life, outside the rarefied discourse of literary theory, not everyone who asserts their; views confidently deserves to be believed. .        . ; ,
THE'DEATH OF THE AUTHOR' THESIS,
Among the inherited notions that the 1960s iconoclasts tried to destroy was the figure of the author. They attacked not the existence of actual authors, such as 'Barthes', 'Foucault', 'Derrida', and 'Lacan'y but the critical practice of invoking an author by working back from a book to the mind and hands that conceived it. They wished to disqualify such a move, stigmatise it as irrelevant to the true experience of reading. Two brief essays by Roland Barthes ('The death of the author') and Michel Foucault
102
critical theories
creator and interpreters
103
{'What is an author?'), published, appropriately enough, in 1968-69, document this attempt to banish the author from critical discourse. Both essays are much cited to this day, and are frequently reprinted in undergraduate anthologies of 'Modern Criticism and Theory' (e.g., Lodge 1988). Both typify, it seems to me, the simplifications and unargued assertions vitiating so much recent literary theory.
On one point we can agree with. Barthes, in his complaint that the image of literature 'in ordinary culture is tyrannically centred on the author, his person, his life, his tastes, his passions', as we see from much literary journalism, interviews, television book-programmes (Barthes 1968a, p. 143). It is true that some journalists (fortunately, perhaps) do not go beyond this superficial concern with writers' lives. But at the same time serious discussion of literature has been going on in Europe and America for centuries, and it is simply false of Barthes to say that 'the explanation of a work is always sought in the man or woman;who produced it'. The remedy, in any case, would be to direct readers away from literary tittle-tattle to more responsible criticism. For Barthes, however > the vulgarisation of critical discourse that he diagnosed became the pretext for abandoning the author altogether. He invoked Mallarme as the first writer 'to substitute language itself in place of the author, and he asserted that 'it is language which speaks, not the author; to write is... to reach that point where only language acts, "performs", and not "me"' (p. 143).
Displacing responsibility from the author to language itself seems now a rather desperate way of countering the journalistic tendency to biographical gossip. Barthes attempted to justify it by briefly invoking linguistics, claiming that 'the whole of the enunciation is an empty process', not needing 'the person of the interlocutors', for '1 is nothing other than the instance saying F, that is, 'a "subject", not a "person"' (p. 145). Barthes was referring to the verbal category known as 'shifters', (or 'indexicals'), the personal pronouns J and you, for instance, which shift between the speakers engaged in conversation. But, as linguists always observe, shifters are essentially markers which serve as 'a function from context to individual by assigning values to variables for speaker, hearer, time and place of utterance, style or register, purpose of speech act', and so forth, a sequence in which the subject is in no sense 'empty'.13 Barthes then switched his linguistic authorities, from Jakobson and Benveniste to Austin and Searle, claiming that in the modem age 'writing can no longer designate an operation of recording, notation, representation', but is limited to what speech-act theorists classify as 'performative, a rare verbal form... in which the enunciation has no other content (contains no other proposition) than the act by which it is uttered. . .' (pp. 145-6). This is a savagely reductive conception of writing, but it is of course a travesty of speech-act theory, which distinguished a whole range of verbal actions, including all kinds of content and performing many different functions in conversation and society.1*
Already we can see how these swiftly produced iconoclastic pronouncements collapsed complex issues into conveniently simple forms,1 skated over vital distinctions. The quickness of the hand is meant to deceive the eye. Barthes then employed the coercive dichotomy, offering a choice between two extreme positions as if these constituted all that could be thought or said, leaving out the many interesting and practicable alternatives in the middle ground. So Barthes declared:
We know now that a text is not a line of words releasing a single 'theological' meaning (the 'message' of the Author-God) but a multidimensional space in which a variety of writings, none of them original, blend and clash, (p. 146)
There we are offered the 'choice' between the Ten Commandments, say, and a kind of soup or wash of texts deprived of originators or indeed originality. The first pole is unreal, to start with, for it is hard to imagine any text having a single meaning. To then parody the notion of an 'Author-God' by labelling such a meaning 'theological' only reminds us that in fact a vast amount of exegesis, quite properly, has been expended on sacred texts in many religions. The second pole; of the opposition is the most drastic conceivable alternative, abolishing authors and: even the notion of a creatively original work. As one critic has observed, 'the assumption that the context of discourse is nothing but other discourse . seals language off from all extralinguistic experience' (Harris 1988, p. 58): but it is only an assumption. In place of reality, authors,, and creative work Barthes offers us 'multi-dimensional space', an impressive-sounding but tautologous concept (space is by definition pluri-dimensionak until delimited). The reader who wishes to preserve the power of (thinking and independent judgement can only say; at this point 'No: I refuse these alternatives. Give me a bigger range of choice.'
But Barthes, instead, pressed on with his preferred option, defining the text -— any text, it would seem, produced anywhere at any time by any person — as 'a tissue of quotations drawn from the innumerable centres of culture'. This is to insinuate a further extreme but unargued claim,; that a text is really made up out of other texts. Barthes then explicitly denied the writer —any writer — the very possibility of creativity: 'the writer can only imitate a gesture that is always anterior, never original. His only power is to mix writings', those produced by earlier writers, all of whom suffered from the same mistaken idea, each thinking he could 'express himself. However, Barthes belittlingly added, the 'inner "thing"' thus expressed 'is itself only a ready-formed dictionary, its words only explainable through other words, and so on indefinitely . . .':{p. 146; my italics)!. Such a ruthlessly reductive form of argument — a variant on Derrida's chain of signs stretching meaninglessly to infinity — would deny the possibility of a work of literature having an autonomous and determinate meaning! at any stage of communication (however trivial). In this way it would lay waste
104
critical theories
creator and interpreters
105
any possibility of individual expression, and ultimately, perhaps, of individual experience. But of course it is ignorant of the simple but fundamental property of language described by M.A.K. Halliday as its-'textual function', language's ability to create a text through which the speaker can 'make what he says operational' in the speech context, as distinct from being 'merely citational, like lists of words in a dictionary' (Halliday 1975, P- IV).
Not really concerned with reflecting how language works, Barthes continued his polemic with an unholy alliance between Oscar Wilde's crisp paradox, that life imitates art, and Derrida's differance, to produce the dismissive statement that 'life never does more than imitate the book, and the book itself is only a tissue of signs, an imitation that is lost, infinitely deferred' (p. 147; my italics). By this mode of argument (if it deserves to be given that name) any book can be reduced to a tissue of quotations from other books — a notion popularised about this time as 'intertextuality' — and writers can be similarly reduced to conduits transmitting each others' ideas. Indeed, Barthes actually brought all writers down to the level of two of the most hopelessly unoriginal characters in the whole of European literature, 'Bouvard et Pecuchet, those eternal copyists', Flaubert's devastating and extreme parody of people utterly lacking invention, originality,, persistence, or judgement. With this massively reductive account of authors and texts, Barthes wished to destroy critics, too, and with them the whole possibility of interpretation:
Once the Author is removed, the claim to decipher a text becomes quite futile. To give a text an Author is to impose a limit on that text, to furnish it with a final signified, to close the writing. Such a conception suits criticism very well... (p. 147)
By abolishing the author Barthes hoped to put the critic out of business, setting in motion a 'truly revolutionary' or 'anti-theological' activity, that of 'refusing to assign a "secret", an ultimate meaning', since 'to refuse to fix meaning is, in the end, to refuse God and his hypostases — reason, science, law'. In 1968 everything could be rejected — verbally, at least. , The iconoclast's mind and pen range so swiftly over his targets in seven short pages that the reader barely has time to voice an objection. But should we feel like taking Barthes seriously as a figure from the culture of a rapidly fading era then we would have to say that 'to give a text an Author" — to pursue his fiction that it doesn't otherwise have one — is in no sense to 'impose a limit' on it, or close it off. This is precisely the point at which the reading experience begins, when the book that has been created by its author — shaped over a period of time by an incomputable series of artistic decisions, each choice involving many rejections — becomes available to its audience. Competence in reading varies enormously, obviously enough, along a spectrum ranging from the brilliantly perceptive to the crassly unaware. To Barthes, however, the reader was simply a focal point in
which the 'multiple writings' that make up the text cohere — in 'the reader', he insisted, 'not, as was hitherto said, the author' ('hitherto' meaning 'in the whole of literary criticism until rhe'). The reader, however (in the orthodoxy of that time), turns out to be not a person but 'the space on which all the quotations that make up a writing are inscribed, a remarkably passive, depersonalised, and incoherent series of metaphors. (It will be difficult to 'inscribe' quotations on a 'space', since you would have nothing solid to write on.) Pursuing his ruthlessly reductive argument to its ultimate position, Barthes issued another prohibition (Thou shalt not, writ over the door'). The writer's relationship with the reader
cannot any longer be personal: the reader is without history, biography, psychology; he is simply that someone who holds together in a single field all the traces by which the written text is constituted, (p. 148)
Barthes ended his devastation of the whole tradition of literary criticism since Aristotle with another facile dichotomy: now 'we know that to give writing its future... the birth of the reader must be at the cost of the death of the Author' (ibid.). But who or what is this reader, slouching towards Paris to be born, 'without history, biography, psychology'? Here the recommended pole of the dichotomy seems to be wholly unreal, another consequence of that violent rejection by French intellectuals of the 1960s of what they derisively dubbed 'humanism'. Readers of a future age will be neither men nor women but androids, holding 'in a single field' (three-dimensional metaphor) the 'traces' or 'quotations' that constitute a text, written, presumably, on air.
Michel Foucaulťs treatment of this topic was more substantial, and included ä historical dimension lacking in Barthes. However, like many of Foucaulťs essays in history, it was a highly personal interpretation which (knowingly or otherwise) simply ignored most of the counter-evidence. As a French historian of philosophy puts it, 'Foucaulťs historical works elude discussion, in that the gist of their argument remains indeterminate', belonging 'properly to the genre of fiction' (Descombes 1981, pp. 116-17). Foucault began by claiming that iii pre-modern times, as in the Middle Ages, 'texts that we today call "literary". . . were accepted, put into circulation, and valorised without any question about the identity of their author; their anonymity caused no difficulties.. .'. Texts that we would call scientific, however, were 'accepted as "true" only when marked with the name of their author' (Foucault 1969, pp. 202-203). Texts 'really began to have authors', he claimed, when 'authors became subject to punishment', their discourses having been 'transgressive', a point which Foucault did not specify but which the surrounding argument would put in the seventeenth and eighteenth centuries (ibid.). Buťthis account is insuf-
106
critical theories
creator and interpreters
107
ficient at every point. Writers in ancient Rome, to go no further back, are regularly described as having published their work at a specific date, and their names were firmly attached to their poems or histories. If we want an instance of a writer being punished for transgressing social, religious, or political norms then we can cite Ovid, banished by Augustus to the Black Sea in 8 AD for.some still unknown offence. As for the Middle Ages, many texts remain anonymous, but they were still identified by their putative place of origin, the poet 'of Paris' or 'of Vienna', say. And there was no such distinction between literary and scientific texts as Foucault claims. Anyone familiar with medieval books of alchemy or astrology will know that equally fictitious or garbled names were attached as authors, especially in the textually hazardous Greco-Arabic-Latin transmission process (such as 'Picatrix').
The history, in other words, was invented in such a way as to support another thesis, involving one of Foucault's gross contrasts between past and present.15 Following Barthes, it seems (although so many of these ideas were common property in 1960s Paris that they can hardly be individually assigned), Foucault celebrated contemporary writing (ecriture) for having 'freed itself from the dimension of expression. Referring only to itself, ecriture is 'an interplay of signs' which 'unfolds like a game', and so on (pp. 198-9). Having erected a completely mythical picture of imaginative literature as breaking with any notion of reference to a human world, real or imaginary, Foucault complained that 'modern literary criticism' still relied on the concept of an author to explain the 'presence of certain events in a work', to provide 'the principle of a certain unity of writing' which ignored contradictions in order to identify 'a particular source of expression' (p. 204). Critics practising such archaic procedures, Foucault objected, 'try to give this intelligible being' of an author 'a realistic status by discerning, in the individual', such qualities as motive, creativity, design. But, he asserted, 'these aspects of an individual which we designate as making him an author are only a projection... of the operations that we [we who? we critics? we readers?] force texts to undergo, the connections that we make, ... the continuities that we recognize. . .' (p. 203). Now, it seems, it is the reader who makes or fashions the author. Foucault wished to deny future critics any such focal point, so — like many, writers of his generation reacting against the phenomenology of Sartre and Merleau-Ponty — he went on to attack the notion of 'an originating subject', the by now (Paris, 1969) sufficiently discredited idea that a 'free subject' could 'activate the rules of a language from within and thus give rise to the designs which are properly its own' (p. 209). This idea being anathema, Foucault expressed the hope that 'as our society changes, ... the author-function will disappear' (p. 210), proposing that the future critic (he.or she still exists, then!) displace attention from the writer or creator to the 'type of discourse', thus 'depriving the subject. ,. of its role as originator' and treating it (him? her? human gender seems to have
disappeared) as 'a variable and complex function of discourse'. This pseudo-mathematical description (like the scientistic metaphors in Barthes, 'a multi-dimensional space', 'a tissue of signs', 'a single field') would replace the human attributes of language, reading and writing, with a de-humanised but completely vague entity or process. But, we must object, where does 'discourse' come from? Who produces or consumes it? Where (or what) are the readers? Who recognises it, even, so allowing it to circulate? This is a non-world, of non-beings, in which some unspecifiable agent(s) produce non-works in a form — 'discourse' — which remains an abstract and privileged category, never defined and never called in question.
In line with the whole oppositional thrust of the iconoclastic literary theory of the 60s, Foucault saw 'the traditional idea of the author' as a concept which needed to be 'entirely reversed'. Where previously the author had been considered 'the genial creator of a work', now 'the truth is quite the contrary', and — like Barthes — Foucault could only express it in negatives:
the author is not an indefinite source of significations which fill a work; the author does not precede the works, he is a certain functional principle by which, in our culture, one limits, excludes, and chooses; in . short, by which one impedes the free circulation, the free.manipulation, the free composition, decomposition, and recomposition of fiction, (p. 209)
Foucault was a much tougher-minded writer than Barthes, who seems intellectually light-weight in comparison (if both more sensitive and more interesting in the long run), but this argument is no better founded. It is to create another unreal pole of a dichotomy to say that the author is 'not an indefinite source of significations', since no one with any sense would ever think he could be. A book is by definition a finite object, the product of choices and exclusions, whose range of meanings may be wide but is still limited. Significations are in any case produced by the reader. Secondly, Foucault's claim that 'the author does not precede the works' (typical, as we have seen, of the structuralists' rejection of language-as-expression for language-as-code), would imply that works of literature were self-aufhoring, acts of'spontaneous creation', as it were, and that 'fiction' could circulate, compose, or even decompose, without any agent intervening. This is obviously absurd. And it is the height of paradox, verging on un-mpaning, to say that the concept of the author 'impedes the free circulation. . . of ■fiction'. Various things, we can imagine, would impede the circulation of fiction or any other form of writing: totalitarian censorship, a paper-shortage, an electricity break-down. But surely the truth is 'quite the contrary': it is authors (and publishers, and readers) who create circulation m the first place/Readers thirty or more years later must judge for them-, selves whether they wish to take over this collection of ideas; to me they seem senseless, the mere product of a wish to reject or negate an
108
critical theories
creator and interpreters
109
existing system of concepts, carried to the absolute limit regardless of the incoherence that results. The whole case is built on a simple adversarial rhetoric, on one side 'impede, limit, exclude', on the other 'free... free . . . free'.
Foucault's treatment of these topics in VArcheohgie du savoir (1969) covered some of the same ground, defining the book as a multiple and fluid entity, 'caught up in a system of references to other books. . . : it is a node within a network', having an undecideable essence (Foucault 1972, pp. 23-5). The only unity that a book or an author's oeuvre can have is 'rhetorical' (pp. 116, 135), a term that Foucault intended as derogatory, and he ascribed to himself 'methodological rigour' in setting aside both these 'traditional unities' (p: 79). In this fuller discussion Foucault seemed at times willing to concede the notion of a subject or writer: 'As we know, there can be no signs without someone, or at least something [sic], to emit them, For a series of signs to exist, there must — in accordance with the system of causality ■— be an "author" or a transmitting authority' (p. 92). Experienced readers of Foucault will at once realise that if this is the case then any 'system of causality' will have to be abandoned, and indeed a few pages later this apparent concession is withdrawn. The speaking subject or author is now 'not in fact the cause, origin, or starting-point' of a sentence or larger discourse. That role 'is a particular, vacant place that may in fact be filled by different individuals', and 'instead of. . . maintaining itself as such throughout a text, a book, or an oeuvre,,this place varies — or rather it is variable enough to be able either to persevere, unchanging, through several sentences, or to alter with each one'.(p. 95).: Yet, in inverting the status quo ante by. a theoretical statement which constitutes itself by negation, Foucault once again lost all sense of reality. Where can we find a book in which the author changes with every sentence, or even after . several sentences? This may happen in that party-game where the players write in turn a sentence or episode of a story on a piece of paper which is folded up before being passed on to the next, but it is hard to imagine any other instance. Theorising in a vacuum risks becoming ridiculous, and inverting a previous system does not necessarily produce a usable alternative.
In their iconoclastic phase Foucault and Barthes were ready to privilege their own highly artificial and synthetic notions of language over and against — so to speak, however unnatural it may seem — works of literature. In S/Z (1970) Barthes could dismissively describe the author as 'that some? what decrepit deity of the old criticism' (Barthes 1974, p. 211), discredited all of two years earlier, and obviously fading fast. In an essay entitled 'From Work to Text' Barthes declared that recently a change had taken place 'in ■ our conception of language and, consequently, of the literary work which owes at least its phenomenal existence to this same language' (Barthes 1971, p. 155). Thanks to the 'sliding or overturning of former categories', we could now contemplate 'a new object,. ... the Text' (p. 156); Barthes described this object by a series of propositions in which he attempted to
ascribe all manner of positive attributes to the Text and to free it of any negative ones. Thus — negative pole —- a mere 'work' (of literature, say) would be 'a fragment of substance, occupying a part of the space of books (m a library for example)', able to be displayed, seen, held in the hand (pp. 156-7). Transcending such mundane attributes, 'the Text' ;— positive pole — would be 'a methodological field', 'a process of demonstration', which 'only exists in the movement of a discourse', or rather, 'is experienced only in an activity of production' (p. 157). This extremely abstract concept seems to be another definition by negatives, its coherence lying only m its systematic contradiction of the dichotomy's other and lower pole. Whereas 'the work' is fathered by the author, aťid modern society has belatedly recognised the laws of copyright (Barthes's debt to Foucault is evident, as are other borrowings from Lacán and Derrida), 'the Text. . . reads without the inscription of the Father', can be 'read without the guarantee of its father' (p. 161). Having rejected God arid his hypostases in the previous essay, Barthes how rejects the father, complementary acts in a culture based on Nietzsche and Fteud. :
Although given specific forms by Barthes and Foucault, with characteristically different emphases, the death of the author thesis was really the extension of another loots communis in French thought deriving from structuralism, the rejection of the subject as a locus of experience and evaluation which persisted in philosophy, in varying forms, from Descartes to phenomenology. Iri that comprehensively iconoclastic reaction led by Levi-Strauss (who dismissed the work of Sartre and Merleau-Ponty as 'shopgirl's philosophy', too concerned with 'private preoccupations',
^whereas the philosopher's true mission was 'to understand being iri relation to itself, and not in relation to oneself1),16 the: standpoint of subjective
i, consciousness was subjected to withering attacks, being travestied arid
, tnvialised in the process (Dews 1987, pp. xii-xiv). Levi-Strauss announced in 1962 — so 'unloosing the slogan of the decade' (Anderson 1983, p. 37) — that 'the ultimate goal of the human sciences I [is] not to constitute but
■ to dissolve man',17 studying instead the 'anonymous thought' found in myth, that type of discourse from which, by definition, the subject of the enunciation has been eliminated. We have'already seen how loften Foucault's Archiofogie du savoir invoked the state of anonymity as the norm for discourse. The widespread nature of what Luc Ferry and Alain Renaut call 'I'antihumanisme contemporain' has been recognised and well documented,18 and there is no need here to rehearse the movement from Foucault's prediction (in the late '60s) of the imminent disappearance of man, to his subsequent turn-round, less than a decade later, in which he fomied the notion of writing a 'history of the subject', starting with a history of sexuality in six volumes (Ferry and Renaut 1985, pp. 148-57).
110
critical theories
creator and interpreters
111
Nor do I need to spell out the parallel move in Bardies, from his doctrinaire rejection of the subject as 'a mere effect of language1 to his rediscovery of the 'enjoying body [which] is also my historical subjecC (Barthes 1976, p. 62). Lacan, similarly, attempted to retain the subject while demolishing consciousness, but as Manfred Frank's, searching critique has shown, this attempt already presupposes consciousness (Frank 1989, pp. 287-316). Derrida, also, uttered the by now conventional diatribe against man and humanism (Ferry and Renaut 1985, pp. 19, 172-4), and in his later work went to flamboyant lengths to deny agency of the texts ; published under his name, a simultaneously sophisticated and naive move, as two shrewd commentators observe. The fact that 'it is always, a subject which decides to efface itself as subject', they add, is a comforting demonstration that 'subjectivity seems strangely capable of resisting the most methodical attempts to liquidate it' (ibid., pp. 192-5). In Althusser too, as Chapter 7 will show, the rejected subject came back like a boomerang.
One result of this euthanasia of the subject might seem to be: the paradoxical survival of thought in the void left by its disappearance (Frank 1989, p. 189). But every time the subject was abolished something had to be erected in its place, usually 'the instatement of the symbolic system . itself, self-enacting and self-perpetuating, as a kind of meta-subject' (Dews 1987, p. 77). We may well ask 'whether the idea of a grounded centre of 1 meaning... is overcome by excluding subjectivity as the unifying centre ( ... or whether it reappears in disguised form' (Frank 1989, p. 53). In Levi-Strauss the alternative to subjectivity was 'objectified thought and its mechanisms', or 'Vesprit human!, a vast abstraction which was postulated by 1 Levi-Strauss as a speaking subject and puzzled over by everyone else. For Foucault's Archiohgte (echoing Derrida), the goal: was to redeploy the ' history of thought 'in an anonymity on which no transcendental constitu- . tion would impose the form of subject', and so 'to cleanse it of all 1 transcendental narcissism', thus freeing history from 'that circle of the lost origin. . .' (Foucault 1972, p. 203). In the Discourse on Language i (1971) Foucault rejected the four notions that, he claimed, had hitherto ^ 'dominated the history of ideas', namely 'signification, originality, unitv, ' creation', preferring to place himself in a newly submissive position in which
the fundamental notions now imposed upon us are no longer, those;, of \ consciousness and continuity (with their correlative problems of liberty i and causality), nor are they those of sign and structure. They are notions, rather, of events and of series, with the group of notions linked ] to these, regularity, randomness, dependance, transformation.... ) (Foucault 1972, p. 230; my italics)
Yet,.despite Foucault's repeated rejection of the, notions of meaning and . sign (not to mention the far more radical rejection here of liberty and 1
causality — this is truly to bind yourself to what he calls 'the rich uncertainty of disorder': ibid., p. 76), it has been conclusively shown that Foucault's system 'appeals constantly to a theory of meaning', and accepts 'the inescapability of the world of signs' (Frank 1989, p. 386). As Schiller wrote, not all are free who mock their chains.
The concepts which were most often promoted to executive authority in place of the deposed and discredited subject and author were 'language' and 'textuality'. As historians of modern critical theory have noted (e.g., Fischer 1985, pp. 51-4, 64-6), many of its exponents quote with satisfaction pronouncements by Heidegger, for instance, that '"language itself speaks" and that we are that which is spoken by it', or by Gadamer that '"it is less true that the subject speaks language, than that it is what is spoken by language"' (cit. Frank 1989, pp. 6, 97). For Foucault, as we have seen, our interest in language should be 'at the level of the "it is said". . .' (Foucault 1972, p. 122), a similar preference for the neutral or passive form. De Man, loyally passing on to American readers a decade later the news from Paris, asserted that 'we do not "possess" language in the same way that we can be said to possess natural properties. It would be just'as proper or improper to say that "we" are a property of language. . .' (de Man 1979, p. 162).
This fashion of attributing to language the powers of agency, choice, or even deliberate inconsistency and incoherence, is a form of determinism which writers of many different persuasions have rejected. Wendell Harris, recording the tendency of some theorists to invoke Saussure while giving hngue the completely un-Saussurian connotations of a Platonic ideal, outside daily human experience yet capable of acting on it, objected that 'language cannot act at all: it exists only as understood and used by human minds' (Harris 1988, p. 7). There may indeed be 'constant change within langue', but such change is brought about 'by human activity acting on language', not by 'language somehow acting on itself (p. 8). Commenting on the move in French thought from Levi-Strauss's notorious claim that ' "myths operate in men's minds, without their being aware of the fact"' (a kurinsu), to the wider view that language itself speaks, Harris writes that 'such hyperbolical statements result from speaking of discourse \parok\ as if it were hngue, and of reifying the abstraction of langue. . .' (p.; 58). E.P: Thompson, discussing Althusser, identified behind his pseudo-Marxist categories the fashionable 60s 'notions of men: and women (except, of course, select intellectuals), not thinking or acting, but being thought and being performed' (Thompson 1978, p. 148). Althusser, he objects (as we now know, this was a common tactic of the 60s system-builders), 'offers us a pseudo-choice: either we must say that there are no rules but only a swarm of "individuals", or we must say that the rules game the players', however absurd that formulation. In the jargon of structuralism, which denies that 'men and women remain as subjects of their history', the passive voice rules:
112
critical theories
we are structured by social relations, spoken by pre-given linguistic structures, thought by ideologies, dreamed by myths, gendered by patriarchal sexual norms, bonded by affective obligations, cultured by mentaUtes, and acted by history's script. None of these ideas is, in origin, absurd, and some rest upon substantial additions to knowledge; But all slip, at a certain point, from sense to absurdity, and, in their sum, all arrive at a common terminus of unfreedom. (ibid., pf 153)
For Thompson structuralism was 'the ultimate product of self-alienated reason. . - in which all human projects, endeavours, institutions, and even culture itself, appear to stand outside of men, to stand against men, as objective things, as the "Other" which, in its own turn, moves men around as things' (ibid.). Much of his polemic against Althusser and 'The Poverty of Theory' is directed against such determinism, and to restoring a concept of human beings as actively involved in making their history.
E. P. Thompson is an English Teft-wing historian; Manfred Frank a Gentian philosopher of the hermeneutic school, equally outraged by deterministic systems which attempt to destroy 'the idea that the subject chooses its interpretation of the world sovereignly and by virtue of its own autarchy'. His response is a robust denial: 'On the contrary, language is no fate; it is spoken by us, and we are able to critically "question" it at any time'. Language is a virtuality, not a determinism forcing itself on individual language users (Frank 1989, pp. 6, 97). Indeed, the linguistic form that the determinists choose betrays the need for an alternative agency:
Even the reified statement 'Language speaks itself, or even the systems-theoretical statement about the 'self-reflexivity of systems' has to employ reflexive pronouns that then hypostatize what was earlier considered a characteristic of the speaking subject as a characteristic of language or of the system itself. The subject that is crossed out in the position of the individual recurs in the position of a subject of the universal, (ibid., p- ID ;
Frank shows that the same recoil recurs, d leur insu, in those writers trying to erase the subject, a conclusion reached by many historians of structuralism and after,20
Textuality, the other substitute for the subject or author^ has1 been elevated by some theorists to a kind of universal status, the mode-of-existence for all written communication, a concept intended to level out any distinction between litetature and non-literature, and inevitably resulting in a dehumanising of discourse (Abrams 1979, pp. 269-70). Edward Said has bitterly criticised the post-structuralists' isolation of textuality from 'the circumstances, the events, the physical senses that made it possible and render it intelligible as the result of human work' (Said 1983, p. 4>. Literary theory in the 1960s and 70s, he complains, abandoned the world and 'the existential actualities of human life, politics,
creator and interpreters
113
societies, and events' in its search for aporias and paradoxes (fed., pp. 4-5), promoting a false idea of 'the limitlessness of interpretation' within a 'textual universe which has no connection with actuality'. But texts are in the world, Said rejoins, and 'place themselves . . : by soliciting the world's attention. Moreover, their manner of doing this is to place restraints upon what can be done with them interpretively' (ibid., pp. 39-40). As his discussion of three modern writers shows, each conceives 'the text as supported by a discursive situation involving speaker and audience' (p.. 40). Said's critique receives strong and independent support from the philosopher Peter Lamarque, performing a witty 'analytical autopsy' of'The Death of the Author' thesis. Lamarque takes issue with the ideal shared by Barthes and Foucault, as indeed by so many writers in their milieu, of a notional 'free circulation' of discourse, of an ecriture that is authorless, 'lacking in determinate meaning,.. .free of interpretative constraints'. Barthcs's idea of 'a text as an explosion of unconstrained meaning', Lamarque writes, 'without origin and without purpose, is a theoretician's fiction' bearing no relation to the actual function of language.
Writing, lite speech, or any language 'performed', is inevitably, and properly conceived as purposive. To use language as meaningful discourse is to perform speech acts; to understand discourse is, minimally, to grasp what speech acts are performed. Barthes's view of ecriture and of texts tries to abstract language from the very function that gives it life. (Lamarque 1990, p. 330)
Similarly with Foucault's belief that there is an intrinsic merit in the ['proliferation of meaning' without consideration of context or purpose, to whicb the appropriate vernacular reply might be that here again, 'more means less'. The whole concept of the death of the author has been finally iput to rest by Sean Burke.21
As the iconoclasm of the French 1960s spread to the English-speaking world over the next decade, the pioneering cultural importers spreading ..the news of the author's demise eagerly appliedit to the greatest of authors. J. Hillis Miller, having swallowed Deriida's claim that since the context .surrounding a speech act 'can nowhere be fully identified or fully controlled' (as if that extremely formulated objection automatically implied that meaning could not be communicated), concluded that 'the uncertainty of .context' dissolves other certainties, including
the concept of authorizing authorship, or indeed of selfhood generally in the sense of an ultimate generative source for any act of language. There t , is not any 'Shakespeare himself', 'Shakespeare' is an effect of the text, which depersonalizes, disunities. (Miller 1977, p. 59)
Although Miller jocularly retracted the absolute implications of that state-'ment (saying that the works of Shakespeare 'must have been written by a committee of geniuses'), it was soon taken up by other critics anxious to
114
critical theories
creator and interpreters
115
emulate the iconoclasm, producing a trend in Shakespeare studies that Richard Levin has analysed in a penetrating essay, 'The Poetics and Politics of Bardicide' (Levin 1990). The debts to Barthes and Foucault are embarrassingly obvious in such pronouncements (by Catherine Belsey in 1980) that '"The Death of the Author. . . means the liberation of the text", allowing criticism "to release possible meanings" in the text, as if they were imprisoned there' (cit. Levin 1990, p., 502), or in similar calls by Jean Howard and Terry Eagleton to 'break free' from previous critical practices that 'still inhibit' us, in order to achieve 'the liberation of Shakespeare'. The general influence of the French iconoclasts is seen in Kathleen McLuskie's pronouncement in 1985 that 'she will be "refusing to construct an author behind the plays and paying attention instead to the , , .strategies which construct the plays' meanings'" (p. 493). In effect, as Levin shows, the 'text' is given a 'new status as authorial surrogate', and acquires 'a repertoire of new activities', such as having 'a project' ^ 'the equivalent of the rejected concept of authorial intention, now introjected into the text' (p. 492). Readers familiar with Renaissance Latin tags may recall the much-quoted remark of Horace, Naturam expelles furca, tamen usque recurret (Epistles, I.x.24: 'You may drive out nature with a pitchfork, yet she'll constantly return'). The text now 'has strategies' of its Own, 'displaces' some issues,, 'conceals' others, 'is silent' here, 'offers an imaginary resolution' there. It even 'gets nervous', as Levin wittily puts it, for 'with the Death of the Author the text acquires human emotions, as well as purposes', with The Tempest supposedly expressing guilt at promoting colonialism (p. 496— a claim! discuss in Chapter 7).
The shift of emphasis from Author to Text has, in the end,, achieved very little. The critics analysed by Levin belong mostly to the Marxist, feminist, and neo-Freudian schools, all of whom go about their business as before. The Marxists find the text concealing 'the contradictions in its own ideological project' (p. 493), but they are easily able to validate their own 'conceptions of ideology and class conflict' (p. 501), The neo-Freudian feminists talk less of concealment than; of 'repression or suppression, which is now attributed not to individual characters but to the text itself, often in the form of a 'feminine or maternal "subtext" that conflicts with, and often exposes the true nature of, the dominant patriarchy . . .' (p. 494), In other words, the text reveals 'precisely what the critics want to find there' (p. 497), and even though they have 'personified, reified, mystified, hypostatized, alterized, and demonizeď the text (p. 498), these critics have just reasserted the validity of their own conceptual system. Far from liberating them to new readings, 'the same one meaning' is found over and over within the various schools. As Levin shows, independently endorsing the philosophically-oriented critiques by Manfred Frank and Peter Dews,
The rejection of the Author Function is not just a negation, the removal of a constraint, since it creates a hermeneutic vacuum that must be filled
bv something else that replaces authorial intention as the determinant of meaning, (p. 502)
Each school proposes its own 'universal law. . . that dictates what one must look for, and must find, in every play. The Death of the Author, then, has left these critics not more but less free .. .''(ibid:).
That judgment applies to many facets of iconoclastic literary theory since the 1960s, where a promised liberation has produced rather different effects. In another popular formulation, a 'system' was toppled and replaced by a 'game', which might seem a harmless enough move. Yet, as John Ellis has observed of Derrida's introduction of the wholly alien concept of 'play' into his account of Saussure's notion of difference, the idea of a 'limitless, indefinite play of differences' merely reduces Saussure, and language itself, to.mcaninglessness: ,
If terms can play (a) indiscriminately against all other terms and (b) endlessly and indeterminately rather than specifically, the result would he nothing — ho specific contrasts that generate meaning, no significant differences forming systems, nothing identifiable or recognizable to anyone, hence no communication and no meaning at all. (Ellis 1989,
p. M) ; ■
By returning us 'from the finite system of a language to the infinite — to what there was before a language coalesced "arbitrary" decisions to reduce it to a finite system', Derrida 'has abolished language, not redefined it' (ibid.; also pp. 52-6, 61, 118-19). In opposition to, and perhaps ignorance of, the whole Chomskyan revolution in seeing language as rule-bound, a system of permitted and forbidden combinations that nevertheless allows the speaking subject full creativity, Derrida and his followers'offer us their image of a game without rules, which, if taken literally (none of them ever did, of course), could only result in aimless or purposeless behaviour- The 'repressive' forces of language were attacked, to be replaced by 'the free play of stgnifiers', words without meaning, utterances without purpose,'in theory at least indistinguishable from nonsense or babble. The rhetoric; of liberation produced not an alternative and superior1—more egalitarian, less coercive — system, but nothing, a vacuum. ■
READERS, AUTHORS, INTERPRETATIONS
My work shall not be an utter failure because it has the solid basis of a definite intention — first: and next because it is not an endless analysis of affected sentiments but in its essence it is action ... action observed, felt and interpreted with an absolute truth to my sensations (which are the basis of art in literature). ' Joseph Conrad22
116
critical theories
creator and interpreters
117
In order to rescue the notion of interpretation from the depths to which some iconoclasts wish to consign it, one might begin with a counter-assertion, that 'all interpretations are provisional'. That is, interpretation is an activity performed by the'subject experiencing a work of literature and resulting in a considered; judgement which hopes to possess some good qualities — to be coherent, do justice to as many of the work's facets as it reasonably can within the time or space available, illuminate qualities not immediately visible or not properly seen by other interpreters — but never claims to be definitive. In this respect interpretation resembles perception itself, as Ernest Gellner defines it, which is never 'the innocent encounter of a pure mind with a naked object, and therefore capable of serving as art untainted foundation for an edifice of knowledge'. Perception is, rather, 'the encounter with some given elerhent. .. by a corpus of knowledge acquired up to that time, but open to revision in future' (Gellner 1985, p. 91). An interpretation may contribute to an ongoing discussion but it does not attempt to end that discussion. Interpretation implies choice on two levels, as Christopher Butler has said: the text 'has multiple implications amongst which the interpreter chooses, and... the interpreter can also choose amongst' competing critical approaches (Butler 1984, p. x). The 'object of interpretation', another writer observes, 'is no automatic given, but a task that the interpreter sets himself. He decides what he wants to actualize', and which norm for interpretation he will use, in a 'free social and ethical act'(Hirsch 1967, pp. 25-6).
Interpretation may offer a choice of methods, but as a process it is unavoidable, starting with the simplest verbal exchanges, or reading however elementary texts. Meaning is not something automatically given, conveyed without loss or surplus between transmitter and receiver. As E.D. Hirsch puts it, 'no meaning represented by a verbal sign is manifest; all meanings must be construed. . ..' (Hirsch 1967, p. 61). Understanding is not always certain, interpretation never complete, since both depend, in Manfred Frank's words, on 'the hermeneutical imagination of the recipient'. For this reason 'interpretations of chains of signs or expressions can be motivated, but not. . . "mechanized"'. Abandoning 'the illusion of a primordially given, self4dentical meaning of a text', we must recognise that 'already the signs that are interwoven in the utterance (or in the text) exist, i.e., acquire the status of signs, only by virtue of an interpretation' (Frank 1989, p. 442). The attribution of meaning, then, is not a definite or finite process but depends on, is subject to, 'the life experience, perspective, world-view, and linguistic competence; .of different speakers'. This does not mean that words or utterances have no meaning, however, only that
meaning is not objectively (i.e., extracommunicatively) decidable. In other words, meaning has to prove itself over and over again in social praxis (and without a final judicial authority to which it might appeal).
'To interpret' means nothing other than this. Something whose objective meaning (if there were such a meaning independent of communication) were known would not have to he interpreted. Only the fact that we can never entirely understand each other. .. makes it possible that we understand each other, (ibid., p. 444)
The fact that interpretations remain without a final criterion, equally, does not mean that they 'are arbitrary or cannot be assimilated (nidit nadwollziebar)■ Interpretive hypotheses are always motivated. . .' (ibid., ■ p. 445).
Frank writes as an exponent of hermeneutical philosophy, which resembles phenomenology in basing itself 'on the side of the> speaking subject', seeing 'speech as one means among others of corporal expression'. As several scholars remind us, phenomenology defined speech as 'a gesture, that is, a manner of "being in the world": through the body proper', a world in which the subject's linguistic or other gesture ' "delineates his own meaning'" (Descombes 1981, p. 97); This emphasis on the interplay between speaker and hearer is in profound opposition to structuralist semiology, which, as we know, concerned itself solely with the receiver and with the code as something independent of human users (ibid:, pp. 92-100). While semiology has faded, its limitations all too evident, phenomenology and hermeneutics, continue to develop, and remain of great value to anyone concerned with literature since they also take as axiomatic the existence of author,. text, and reader in a communication cycle, with language as essentially a medium for the expression of meaning. A fruitful link between hermeneutics and literary criticism was: made by E.D. Hirsch in his classic; study of Validity in Interpretation, which I drew on in-Chapter 1 for its analysis of meaning as involving such qualities as sharability, determinacy, and reproducibility. Hirsch made a useful distinction between 'meaning' and 'significance': 'Meaning is that which is represented by a text; it is what the author meant: by his use of a particular sign sequence; it is what the signs represent. Sigm/icarice, on the; other hand, names a relationship between that meaning and a person, or a conception, or a situation, or indeed anything imaginable' (Hirsch; 1967, p. 8). Although an author wills a specific meaning in a conscious act, he ■almost always means, more than he is aware of meaning, since he cannot explicitly pay attention to all the aspects of his:meaning' (p. 48). Hirsch conceives of meaning as 'a willed' or 'shared type', ;an entity that has a boundary, and is therefore determinate and sharable, but which 'can |always be represented by more than one instance', since its 'determining characteristics are common to all instances of the type' (pp. 49-50, 65). The author wills a meaning-type, but since it can contain a bundle of sub-meanings the reader is able to discover other meanings, including; future ■ones. ■
As for significance, that is 'always "meaning-to", inever "meaning-in".
118
critical theories
creator and interpreters
119
Significance always entails a relationship between what is in a man's verbal meaning and what is outside it' (p: 63). Interpretation, then, must begin by understanding a text 'on its own terms', must be 'intrinsic' in that sense (p. 113). Such understanding is 'a construction of meaning', the necessary preliminary to interpretation, here defined as 'an explanation of meaning' (p. 136). Whereas understanding is 'cast only in its own terms', interpretation 'nearly always involves. . . discussing meaning in terms that are not native to the original text' (pp. 135-6), using 'categories and concepts' that are not necessarily found in the literary work (p. 136). The challenge to the interpreter, obviously enough, is to produce an interpretation which develops its own concepts and categories while remaining true to the meaning of the text. Invoking Husserl's concept of'a "horizon", which may be defined as a system of typical expectations and probabilities', so that the 'explicit meanings' of a verbal intention 'are components in a total meaning which is bounded by a horizon' (pi 221), Hirsch defines the interpreter's aim as being 'to posit the author's horizon and carefully exclude his own accidental associations' (p. 222). This is an important point, given the anachronistic nature of much current Shakespeare criticism, from the imposition of modern socio-political expectations down to lexical misunderstandings (e.g. the misreadings of 'grope'and'finger'discussed in Chapter 6 below).
One purpose of Hirsch's book, he writes, is to 'encourage a degree of methodological self-consciousness' (p. 169)i particularly concerning the goal of interpretation. Unlike the natural sciences* with their claims for certainty, the aim of. literary criticism is to achieve nothing higher than 'consensus — the winning of firmly grounded agreement that one set of conclusions is more probable than others' (p. ix). In interpretation 'certainty is always unattainable' (p. 164), and what the interpreter offers is an 'interpretive hypothesis', amounting to 'a probability judgment that is supported by evidence' (pp. 174-5, 180). All the interpretations of Shakespeare that I offer in this book are of this nature, citing evidence to support a hypothesis in the hope of reaching not certainty but a, greater degree of probability than other interpretations discussed here. The interpreter's goal 'is simply this', Hirsch writes, 'to show that a. given reading is more probable than others', and he suggests four criteria that must be met (p. 236). First 'the criterion of legitimacy, that is, a given reading should be possible 'within the public norms of the hngue in which the text was composed'. Secondly, 'correspondence: the reading must account for each linguistic component in the text', and can be judged improbable if it 'arbitrarily ignores ... or inadequately accounts for them'. Thirdly, 'generic appropriateness', not confusing a scientific essay with casual conversation, say. The final and decisive criterion is 'plausibility or coherence', which depends on 'the nature of the total meaning under consideration' (p. 237). Here again the argument concerns the most probable reading, and the interpreter must base his argument on a wide range of evidence: the work's
date or period, its genre, the attitude of 'the speaking subject' (pp. 237— 44). The goal of interpretation, appropriate to the humanities, is not verification but validation, which is always provisional, subject to revision, capable of improvement (pp. 170-1).
To return, strengthened by Hirsch's example, to the iconoclasts' dismissal of all interpretation as misinterpretation, we can now see the movement from interpretation to criticism as a rational activity. Any competent critic ought to be able to justify his behaviour, to give a coherent account of the assumptions about literature which he has brought to bear on a poem, novel, or play. He should be able to account for the kind of method he has used, and the relationship between the initial assumptions, the method, and the judgments that result. We all know that, from the basic nature of perception up to carefully constructed arguments, the assumptions that we make control what we are able to see, and that the type of question we ask will open up one issue but close off others. Some critics cannot or will hot discuss any of these matters, as F.R. Leavis notoriously refused Rene Wellek's challenge to do so, a refusal that I have always taken as a sign of weakness.23 To me it seems reasonable to expect a critic to be aware of both critical principles and practice, just as we expect him to be able to provide evidence for his interpretation. Reference to the text is crucial in arguments which aspire to probability, since a rival interpretation will either cite other passages or submit the passages already cited to a different interpretation. I am not makings the naive suggestion that reference to the text will automatically settle a dispute (as Leavis did so unfortunately in his attack on Bradley's interpretation of Othello),24 merely reiterating the point that interpretation is a two-way activity between text and reader, and that an interpretative construct can win more, or less agreement at various points in history, but will never be final, or definitive.
The starting point for interpretation until very recently has been the assumption that a literary work constitutes, or at least aspires to be a unity, a coherent whole. This idea, too, fell into disfavour among Paris iconoclasts. Roland Barthes, restlessness embodied, was one of the first to abandon the notion of literature as a coherent activity. In an essay; of 1955 he praised a novel by Robbe-Grillet as existing 'in that very; narrow zone . . . where literature unavailingly tries to destroy itself, and apprehends itself in one and the same movement, destroying and destroyed' (Barthes 1972, p. 57). Never slow to widen a judgment on one writer to embrace the whole of a period (a trick imitated by de Man), Barthes pronounced that 'in our present social circumstances . . . literature can exist 'only as the figure of its own problem, self-pursuing, self-scourging' (ibid:, p^ 58). Over the last hundred years, Barthes repeated in 1959, from Flaubert to Robbe-Grillet, an 'interrogation' had been: going on 'in that asymptotic zone where literature appears to destroy itself as a language object without destroying itself as a metalanguage, and where the metalanguage's quest is
120
critical theories
defined at the last possible moment as a.new language object' (p. 98). Here, as so. often, Barthes reduces literature to the level of language, 'a language, subject, like any other, to logical distinction' (p. 97),, before disabling it in one or other key function.
In his essays of the early 1960s the main literary function that Barthes attacked was meaning. Literature, he declared in 1960, 'is at bottom a tautological activity') a 'narcissistic activity', that starts by interrogating the world but 'becomes its own end': 'the author rediscovers the world' in his writing but in a process of'perpetual inconclusiveness', an 'alien world' which literature represents ,'as a question -—never, finally, as an answer' (pp. 14-4—5). The influence of structuralism, and its imperialistic aggression towards other theories of language, is evident in the generalising deduction that follows: 'Language is neither an instrument nor a vehicle: it is a structure' (p. 145). Levi-Sttauss, with his penchant for binary categories, may have inspired the dichotomy Barthes,then made between author and writer. The non-creative 'writer' (Term I) 'is a "transitive" man, he posits a goal (to give evidence, to explain, to instruct), of which language is merely a means. . . an instrument of communication'. The 'writer' uses language in institutions which are not primarily ■ concerned with language: 'the university, scientific and scholarly research, politics, etc.', using language naively to resolve an ambiguity, making it 'a simple vehicle, its nature as. . . merchandise transferred to the project of which it is. instrument'. The 'author', by contrast (Term. II),. novelist, poet or dramatist, 'knows that his language, intransitive by choice and by labor, inaugurates an ambiguity', offering itself as a question, 'a monumental silence to be deciphered' (pp. 147-8).. Transitivity^ that fundamental, property of language (as Halliday defined it) to convey 'the experience .of process in the external world', with its basic categories- of actor, process, goal, is thus granted the critic but denied the creative writer: 'for; the author, to write is an intransitive verb' (p. 145); or again, 'to write is,either to project or to terminate, but never to "express"; between beginning and end, a link is missing, which may yet pass for essential; the link of the work: itself (p. xiii). True to type as iconoclast (Derrick, learned from him), Barthes only usedthe metaphor of a communicative link in order to subvert it: The texts which follow are like the links of a chain of meaning; but this chain is unattached' (p. xi).
Developing his hostility towards the notions of literature as either a coherent and meaningful utterance or as a representation of reality, its language never able to function as 'a vehicle of "thought"', doctrine or evidence ('by identifying himself with language, the author loses alb claim to truth': ibid.), Barthes increasingly saw literature as a .non-meaning, noncommunication. The 'entire art' of Robbe-Grillet,. he declared in ,1-962, consists in 'disappointing meaning precisely when he makes it possible' (p. 200). Jumping to generalisation again, Barthes affirmed that 'all literature'. posed the question, 'what does the world signify?', but it merely.recorded
creator and interpreters
121
'tins question minus its answer' (p. 202)V and he looked forward to the day when it would 'be possible to describe all literature as the art of disappointment, of frustration' (p. 203). In formulae that de Man was to take over, Barthes stated that literature 'at one and the same time designates and keeps silent' (p. 204); the literary work 'offers itself to the reader as an avowed signifying system yet withholds itself from him as a signified object' (p. 259); Diagnosing,in 1963 this fundamental split between purpose and effect — denying transitivity — Barthes saw the functioning of literature to be a 'disappointment of meaning', 'undermining. . .assured meanings' yet providing no alternatives, 'simultaneously an insistent proposition of meaning and a stubbornly fugitive meaning' (if»d::). All literature, from Horace to Prevett, is both 'the institution and the "disappointment" of meaning,', i; .a system of meaning at once posited and unfulfilled' (p. 269). ■; Basing himself on the nouveau roman of Robbe-Grillet -— his Parisian parochialism. nowhere more glaring when we consider what; a powerful instrument the novel became at this time in Latin America for the representation and criticism of social and political reality — Barthes developed his theory of the> dysfunctioning of literature as a source of meaning by such categorical assertions, not attempting anythingTike detailed analysis, relying on formulae and slogans. Whether, in1 wishing to frustrate one of the:■ traditional attributes of literature, Barthes was original or merely a symptom of a climatic shift, must await a more detailed history.: But his image of literature as in some basic way divided against itself was embodied by others at this time in methods of interpretation designed ito demonstrate the incoherence of all texts. One such was Derrida's technique. of decon-struction (to be considered in Chapter 3). Another was the Freudian fashion in Paris of the 50s and 60s, from Lacan and Derrida to the Althusserian Macherey (see. Chapter 7), of applying the psychoanalytic method of 'symptomatic reading', which looks for absences or contradictions that undermine a patient's utterance or a literary work's apparent meaning. Yet, as several commentators observe, the soi-disaiit avant-garde's search for incoherence is an a priori .expectationwhich is somewhat perverse. AsJohn.Reichert has judged, the search for incoherence iti texts is as odd as if one were to choose 'one hypothesis over another on the ground of its greater inconsistency', or because it accounted for 'fewer of the facts that we want to explain', or was 'unnecessarily complicated' (rit; Harris .1988, p. 70). This is to deny the possibility: of interpretation from the outset. •■
More damagingly, even, the expectation of incoherence runs counter to literary theory historically considered,! which regularly emphasised die need for a writer to unify his work in'style, 'subject-matter, and construction; To give only one example, the opening of Horace's Ars poetica, a key text in European literary theory from the fifteenth to the eighteenth centuries, considered the ridiculous idea of a painter producing a picture which combined 'a horse's neck with a human head', clothing 'a miscellaneous
122
critical theories
creator and interpreters
123
collection with various kinds of feathers, so that what started out at the top as a beautiful woman ended in a hideously ugly fish'. Horace argued from this analogy that 'a book whose different features are made up at random like a sick man's dreams, with no unified form to have a head or tail, is exactly like that picture' (1-8). In Renaissance literary theory the demand for unity was met by the complementary demand for variety: but no-one before I960 expected incoherence! This is another iconoclastic attitude whose date is out.
The process of reading resembles the general mental operation of trying to make sense of the reality perceived by our senses. As several studies of reading have shown, in this activity the mind tries to find cohesion in a text, is prepared to make numerous redefinitions of what it has read in order to achieve that goal, and only when frustrated recognises incoherence (Butler 1984, pp. 4-6; Harris 1988, pp. 50-51, 68-70). The linguistic criteria for cohesiveness, syntactical and lexical, are not enough, as Wendell Harris has shown in his book. Interpretive Acts. In Search of Meaning. There must also be an internal coherence relating 'the content of the sentences which make up the discourse' to each other, according to conventional patterns of organisation. However, that internal coherence turns out to be* 'inseparable from intelligible relationships to external patterns of organization and sets of experientially understood relationships' (Harris 1988, p. 53).
Harris draws on discourse analysis, which began to develop in the 1970s and 80s (another new development in the humanities ignored by Current Literary Theory, clinging to its relics of 60s Paris), for the insight that understanding an utterance presupposes 'knowledge and a process that is interactional, situational, and strategic'. The.hearer or reader takes.into consideration the situation in which discourse appears, assumes intentions and purposes in the utterer, and reaches understanding 'by means of tentative interpretations that are matched and rematched against experiential knowledge' (ibid., pp. 53-4). Harris's discussion extends the account of implicature and inference that I gave in Chapter 1 into the area of literary interpretation, The author's, intentions, he argues, are important, whatever the post-structuralist dogma about their absence from the text. 'Of course authors are not present in their texts, but all sorts of clues as to their intentions are'* so that 'to try to regard a text as embodying no authorial intent requires us to regard it as having no author at all', a pretence that cannot be maintained for long (p. 60). The criteria which the reader uses for judging the probably applicable contexts which the author intends are 'consistency' and 'unity', which Harris subsumes under the term 'consonance' (p. 68). Both speech-act theory and discourse analysis, he shows, converge in agreement on some fundamental principles, namely that
interpretation requires die assumption of (1) knowledge, both immediate and general, shared between the formulator of the: discourse and the
interpreter, (2) that the formulator intends a meaning that is to be understood within that shared context, and (3) that the pursuit of coherence (consonance) is essential to the pursuit of meaning, (pp. 54-5)
Drawing on the well-established fact that 'readers will always seek some explanation for the total pattern of a work, or indeed of any set of words that they believe to be intended as a connected utterance' (p. 69), Harris invokes Grice's Co-operative Principle to suggest that in reading 'we seek a consistent interpretation because we assume that we and the author share certain assumptions about the communication of meaning' (p. 70). Harris agrees with Hirsch that all interpretation of the author's intention rests on probability, not certainty (pp. 91, 160), and concedes, with Grice, that we may fail 'to recognize a crucial instance of implicature because we do not possess certain elements of knowledge or awareness that the author expected'. However, he points out, 'the direction in which implicatures are resolved is often indicated by several elements of the surrounding context', and a misinterpretation at one stage can be corrected in the light of further reading (p. 71) It is a common experience that a later passage in a book helps us to grasp something that we had not properly understood before.
Developing the principle that 'ail uses of language require interpretation, and interpretation is possible only within shared contexts' (p. 72), Wendell Harris argues that 'the most pervasive dimension of context' is a 'knowledge of reality', which interpreters often take for granted (p; 78). The most 'pedestrian of situations', as he puts it, such as cashing a cheque or filling up a car, 'carries with it a congeries of presuppositions arid related bits of knowledge' (p. 110). We bring all our knowledge of the operation of the world with us to the reading of literature, arid 'unless we arc told otherwise, we interpret discourse on the assumption that the world is the world we know' (ibid.). Attacking eleven currently fashionable fallacies in literary theory (pp. 162-3), Harris lays down as a first interpretive principle that all discourse assumes the existence of 'a "brute" reality', even! though 'we can never get outside or beyond that interaction of sense reports, language, and experience we call thought' (p. 157). It is true that 'our conceptualization of reality is a fabric of conventions, but these conventions have been called into being by the interaction of human purpose and an insistent physical reality' (ibid.). The conventionalities so developed cannot simply be detached as such from reality, or used to deny the .existence of reality. •
Another important principle, arising from this (and one which recalls Halliday's account of a child's language-learning), is that 'words are always used (uttered or written) in relation to a set of contexts', which can be described in or influenced by language, 'but most of which require the existence of extralinguistic objects, feelings, and institutions' (p. 158). Since discourse 'occurs in a complicated and interactive environment or context', then, as we know from speech-act theory and pragmatics,
124
critical theories
creator and interpreters
125
meaning depends on a context assumed by the author and reconstructed by the reader. The expectation of consonance turns out to be functional in the whole communication of meaning (which means that the iconoclasts' preference for incoherence is actually parasitic on meaning as communicated). As Harris puts it,
That authors intend consonance of discourse as the means of expressing their intended meanings and that readers will attempt to discover the intended meaning by seeking consonance in their reconstructions can only be assumptions, but they are constitutive assumptions: they constitute the possibility of meaning by defining it. (ibid.) -
Harris deduces from his argument two basic principles for interpretation. First, 'that an author's intended meaning can be understood with reasonable probability and accuracy, but only through knowledge of as much of the context assumed by the author as possible'which means that 'interpretation requires knowledge of historical events, cultural attitudes, linguistic usages, and biographical fact' (p. 168). The. unthinking anachronism of much current Shakespeare criticism stands condemned. Secondly, 'that multiple analyses of any discourse are not only possible but may be of great value' (p. 169). There can be no single, correct interpretation, however vociferous its proponents may be.
The way in which we seek for, and discover, cohesion in literary texts has been well described by the Danish scholar S.H. Olsen in a book called The Structure of Literary Understanding. (1 am deliberately quoting a wide range of contemporary literary theories to show that Current Literary Theory, as commonly defined by some of its exponents, and by some of their publishers, is a limited, derivative undertaking which has already closed down most of the interesting and complex issues.) Olseh argues that reading a literary work involves the ability to recognise formal features — episodes involving contrasts or parallels — and to judge how they affect the work's overall meaning and our response to it (Olsen 1978, p. 14), The experienced reader tries to discover the structure of a complex work by dividing it into parts and interrelating these parts or segments into a pattern. His aim in singling out a segment is 'to determine the artistic significance or purpose he believes this passage to have in the context of the work', (ibid., p. 82). Interpreting a text, we place each segment into one or more descriptive categories, gradually interrelating the descriptions by a 'process of redescription' which ultimately produces an 'explanatory grid of concepts', covering the whole work and relating different parts to each other (ibid., p. 83). Although that may sound forbiddingly technical, redescription is in fact the activity all readers perform.to make sense of a literary work, from a brief lyric to a large novel: 'the construction of a web of concepts which the reader uses to interpret the work is a necessary condition of getting aesthetic satisfaction from it' (ibid., p. 89) — or not, as the case may be.
Br ; Olsen suggests five criteria (overlapping with Hirsch's four) by which we K      can 'judge interpretations: completeness, correctness, comprehensiveness,
■ consistency, and discrimination' (p. i26). Anyone who compares different jfif critical accounts of a novel of play will see that some can be remarkably ■IV comprehensive, leaving out relatively little of the text, showing how ■J the parts cohere; others deal with a small segment, or from a limited Bis, viewpoint. Although Olsen speaks of evaluating interpretations he believes H that evaluating a literary work is a separate activity, a 'second-order judge-Ks ment'c made only as a recommendation (p. 168). To me, however, the act Hfc of description already implies a process of evaluation, since the very terms Bv: that we use to analyse works of literature are not value-free (and are B none the worse for that). Still, I welcome his emphasis in a later book K (optimistically entitled The End of Literary Theory), in opposition to the B- dehumanising systems of Foucault and others, that 'literary practice is a H! human practice defining and serving human goals and purposes. Evaluation, |p      both in the form of appreciation and. . . value-judgements, is at the heart
■ of this practice'(Olsen 1987, p. 155).
f In defending the possibility of interpretation Olsen joins an increasing
number of contemporary thinkers who accept the notion of human inten-■      tionality as fundamental. In his version of speech-act theory a work of l      literature is the result of 'intentional behaviour.,. directed at some ^      response in the receiver', who attributes an intention to the producer :      (Olsen 1978, p. 5). An intentional (or 'purposive') theory stresses 'the I;      central role of the reader's/author's attitude, feelings or intentions in a i       description of the aesthetic properties of a literary work' (ibid., pp. 6-7). Conceiving a novel or play as the product of intentional action does not involve us in trying to discover the writer's intentions other than those discernible by the very existence of the work (p. 49). Rather, the concept ;      of intentioriality brings with it the notion of literature as an institution — *       an ongoing social activity — depending on 'a framework of assumptions, | ,      concepts and practices' which the reader applies to literary texts, whether aware of them or not. Where some people today think of theory as a self-sustaining activity unrelated to the experience of reading, Olsen seesthe function of literary theory as being 'to provide an account of the'logic of these practices and judgements which the reader makes about the! literary work' (p. 23). Since, according to speech-act theory, utterances are i-     purposive, intended to have consequences (pp. 50-4), the author of a '       literary work commits himself to having certain aesthetic intentions, as T       realised in the literary work,25 which 'differs from other texts in its capacity to fill a certain role in a community of readers: a role defined by concepts and practices which the readers master' (p. 82). The role played by literature is to be 'a cultural value', experienced for its own sake (Olsen 1987, p. 26). Whereas a speech-utterance aims at the primary effect of securing, uptake only as a means to a further effect, a literary work is 'a text where the interest is only in interpretation and not in use' (ibid., p. 39).
126
critical theories
While I agree that works of literature are meant to be enjoyed, are different in kind from factual hand-books, say, and have uses prescribed by the community (a point well argued in Ellis 1974), I shall argue in the next section that their primary function is to make possible the games of make-believe that we perform as readers.
Olsen's account of interpretation, like those cited earlier by Manfred Frank, E.D. Hirsch, and Wendell Harris, preserves the separate categories of author, reader, and text. Indeed, it is hard to see how the discussion of literature, or any human product could take place if we did not distinguish the artificer from the artefact, the artist from his public. Olsen even uses at one point the word 'aesthetic', which, like several concepts deemed undesirable, old-fashioned {'bourgeois', 'humanist', or whatever), has been stigmatised and rejected.26 But unless you wish to deny that King Lear or the Missa Sohmrds are works of art created by their (paid) composers to be seen and heard (by a paying audience), and that they provide experiences differing in kind (and degree) from those conveyed by a view of the Mont Blanc or the Grand Canyon, the term 'aesthetic' is unavoidable and harmless. It can become harmful when barriers are erected around works of art in order to deny, for instance, the continuity between our cognitive and emotional responses to them and to life. But the theory I shall be outlining here refuses any such separation. The term aesthetic, as I use it, describes some specific properties of works of art and our experience of them, but does not exclude consideration of their ethical, rhetorical, political or any other properties. Rather, it interrelates with, and complements those considerations. I share Olsen's view that the reader's response to a literary work involves an imaginative reconstruction of it, expressed in a vocabulary which is not specialised (as are those of law, economics, chess) but open, continuous with life: 'literary aesthetic argument' uses 'terms which have an established use in other spheres of life'. We describe Hamlet with the same terms we 'would iise to describe a real human being', and we impute 'motives, mental states, and emotions to characters' in literature as in real life (Olsen 1987, pp. 16-17). There cannot be a separate, non-human conceptual system to describe literature.
Aesthetic theories need not erect a barrier between art and life, nor retreat behind it into self-absorption. Nelson Goodman, a distinguished philosopher of the arts, argues that aesthetic theories must include both the art-work and its appreciators. As he puts it in Languages of Art, 'since the exercise, training and development of our powers of discriminating among works of art are plainly aesthetic activities, the aesthetic properties of a picture include not only those found by looking at it but also those that determine how it is to be looked at' (Goodman 1976, pp. 111-12). Subsequently Goodman attacks (in similar terms to Gellner's rejection of the naive idea of perception) the erroneous tradition which sees the aesthetic attitude as 'passive contemplation of the immediately given, direct apprehension of what is presented, uncontaminated by any conceptualization'. Goodman's whole work argues the contrary view,
if
1
i
creator and interpreters
127
that we have to read the painting aswell as the poem, and that aesthetic experience is dynamic rather than static. It involves making delicate discriminations and discerning subtle relationships, identifying symbol systems and characters within these systems and what these characters denote and exemplify, interpreting works and reorganizing the world in terms of works and works in terms of the world. Much of our experience and many of our skills are brought to bear and may be transformed by the encounter. The aesthetic 'attitude' is restless, searching, testing — is less attitude than action: creation and re-creation, (pp. 241-2)
Goodman's theory of art emphasises the dynamic inter-relation of the artwork and the person experiencing, an exchange that takes place in, uses the same terms as, normal life. The idea that 'aesthetic experience is distinguished... by a special emotion', he comments, is one that 'can be dropped on the waste-pile of "dormitive virtue" explanations' (p. 243). The spectator or viewer's response involves both the mind and the feelings. Rejecting attempts (such as those of Levi-Strauss) to distinguish science ffoni art in terms of cognitive as opposed to emotive satisfactions, as if knowing and feeling: were separate activities, Goodman !argues that both aesthetic and scientific experience are 'fundamentally cognitive' (pp. 243—5). Certainly works of art offer an emotional experience in a way that scientific discourse and practice rarely do, but aesthetic experience is not 'a sort of emotional bath or orgy'. The continuity he has been arguing between art and life here undergoes a reversal, in his claim that the emotions involved in aesthetic experience are 'reversed in polarity. We welcome some works that arouse emotions we normally shun. Negative emotions of fear, hatred, disgust may become positive when occasioned by a play or painting' (pp. 245-6). Goodman's crucial argument, I believe, is that 'in aesthetic experience the emotions function cognitively' (p. 248). We apprehend a work of art through the feelings as well as through the senses, and using our emotions cognitively involves, 'discriminating and relating them in order to gauge and grasp the work and integrate it with the rest of our experience and the world' (ibid.). Our emotions combine 'with one another and with other means of knowing': (p. 249), for 'in ;aesthetic experience, emotion positive or negative is a mode of sensitivity to a work' (p. 250). Knowledge and feeling work together.
, Our experience of OtheEo or King Lear, then, is not the palliative-therapeutic notion of catharsis (p. 246) — in any case a doubtful interpretation of Aristotle's Poetics27 — but a heightening of feeling which makes us perceive both the play and the human actions it represents, more intensely. So it is perfectly legitimate for an interpreter to report on his feelings, as Dr. Johnson did so memorably as he completed annotating the scene (5.2) where Othello kills Desdemona;— T am glad that I have ended my revisal of this dreadful scene.: It is not to be endured'.28 Inded, his emotional reactions have shaped the interpreter's experience of the play throughout. If Johnson's reaction is normal, then we may be surprised how
128 critical theories j
few contemporary critics report their feelings, or even acknowledge that •
they have any. Nor will we find them discussing the fact that, over time, i
an interpreter's or appreciator's response can change. Critics of other j
persuasions regularly observe that the appreciation of the arts is an ability j
that has to be acquired, and that readers (can, at least) learn from gritics \
(Butler 1984, p. 33; Olsen 1978, p. 21). In learning, in reading and re- j
reading complex literary Works, we come to see and feel some things more 1
clearly, more strongly. As Goodman puts it, j
where there is density in the symbol system, familiarity is never com- i plete and final; another look may always disclose significant new ! subtleties. Moreover, what we read from and learn through a symbol I varies with what we bring to it. Not only do we discover through our symbols but we understand and reappraise our symbols progressively in the light of our growing experience, (p. 260)
Awareness of this dimension is wholly lacking in Current Literary Theory.
Goodman's argument makes the appreciator's response to a picture of play an integral part of that work's existence. While welcoming that | emphasis, the intentionalist notion of interpretation, as I understand it, 3 will not lose sight of the artist's prime responsibility'"for the work that makes such experience possible. Another distinguished philosopher of aesthetics, Richard Wollheim, in Art and its Objects (1968; rev. 1980), has ' attacked the tendency to give primacy to the spectator and to ignore 'the aims or intentions of the artist'. This has the damaging consequence that ' 'once the artist is not accorded at least equality with the spectator, he ends ( up by dropping out of the picture altogether' (Wollheim 1980, p. 228). ] Wollheim insists on keeping a balance between artist and spectator, arguing that 'it is always the artist who.shapes the forms that bear his name'. These forms are not created out of nothing but from 'inside a continuing activity or enterprise', with its own repertoire, offering both constraints and opportunities, and thereby providing within the framework ! of an agreed system 'occasions, inconceivable outside it, for 1 invention and audacity' (pp. 124-5). The artist's intentions, Wollheim argues, are J realised in the art-work, but there can never be a complete correspondence between the artist's activity and the spectator's reaction. In a remark which links up with E.D. Hirsch's notion of verbal meaning as a willed type which the author intends, but whose sub-meanings he cannot control, i Wollheim observes that 'the spectator will always understand more than : the artist intended, and the artist will always have intended more than any single spectator understands . . .' (p. 119). Both roles must be emphasised equally, then, for creation and interpretation are reciprocal activities. j Quoting.Valery, Wollheim writes that 'the artist is active, but so also is the spectator, and the spectator's activity consists in interpretation. "A creator", Valery puts it, "is one who makes others create"' (p. 87).
Yet the interpreter's 'creation' is not free and indeterminate, as Current ■-/
creator and interpreters 129
Literary Theory would wish, but is subject to constraints, limits, as is that of the artist himself. These limits are in fact set by the artist, in the complex process of selection and rejection which has given his work its distinctive shape* content, style. As Christopher Butler has argued, all texts 'construct situations which control a great deal of what it. is plausible to say in interpretations', for they all 'implicitly select an area of experience with which to deal' (Butler 1984, p. 46). That proposition would be anathema to exponents of rule-free play, as would its corollary, that the situations constructed by an author 'are part of the mimetic commitments of the text., They thus relate to the ways in which we think about the world' (ibid.), the process Nelson Goodman described as integrating the work of art 'with the rest of our experience and the world', by interpretation 'reorganizing the world in terms of works and works in terms of the world'. One of many vital elements ignored in recent literary theory is that shared knowledge of life as lived in specific places at specific times which an author depends on his reader having, or being willing to acquire. If you once reflect how much you need to know about human society in order to understand Measure for Measure — what justice is, or authority, or power, or corruption, the responsibility of the magistrate, the vulnerability of the woman in an informal marriage-contract sworn de futuro, the professions of whore, criminal, executioner — then you may agree that that play is as it iSi to begin with, because Shakespeare has chosen these situations, and the issues they raise, and rejected everything else. As that small example shows, more needs to be said about mimesis, and how the author shapes his text.     . ■
REFERENCE, REPRESENTATION, MAKE-BELIEVE
And let us, ciphers to this great account,
On your imaginary forces work. . . .
Piece out our imperfections with youf thoughts;
Into a thousand parts divide one man,
And make imaginary puissance.
Think, when we talk of horses, that you see them,
Printing their proud hoofs i'th'receiving earth.
For 'tis your thoughts that.npw must deck pur kings...
Henry V. Prol. Act 1, I5lf.
It was only to be expected that the iconoclastic movement of the mid-1960s, having, denied that language could reliably convey meaning about or reference to the world we live ;in, would sooner or later extend that veto, embargo, prohibition — since this, too, is an attempt to deter people from thinking differently — to literature. A convenient starting-point for
130
critical theories
this polemic might be 1968, again, and another essay by Roland Barthes, called 'L'effet du reel', which argued that description in literary texts was included purely for aestheticeends and could not denote reality. To think otherwise was to indulge 'the referential illusion'.29 From that time on, we can say, and for all would-be avant-garde or post-whatever critics, any notion that a drama, novel, or poem could or should be seen as (among other things) representing human life was to be rejected with scorn. With another of those easy but empty'polemical dichotomies, the literary text was said to be not at all a representation of reality but instead a self-contained system of signs or codes. As structuralism-gave way to post-structuralism, the positive, privileged pole of that dichotomy was replaced by a new formula, a self-contained system calling its own existence in question by its unfulfillable desire to refer outside itself. I shall be discussing deconstructive literary theory, and its catastrophic effects on the interpretation of Shakespeare — deliberately catastrophic, demonstrating the workings of human inten-tionality at its most intransigent — in the next chapter. Here I want to review some arguments against representation, and some of those for it, in order to outline a qew theory which moves the discussion in a different and more fruitful direction.
In the 1960s attack on mimesis we find the same sloppy, ■ self-serving modes of argument that underwrite so much of Current Literary Theory, unjustified assertions, facile dichotomies that privilege one pole and dismiss the other, mockery of competing views. One exponent of post-structuralism, as A.D. Nuttall points out, 'proposes without a tremor that we cease to pretend that the ordinary world is real: "First, there is the , socially given text, that which is taken as 'the real world'". Notice that [he] places inverted commas around the real world but none around text. The traditional relation of object and representative is silently reversed* (Nuttall 1983, p. 10). Robert Alter, discussing 'Mimesis and the motive for fiction', noted that in the iconoclastic theories of the '60s 'the attack on mimesis ultimately depends on defining experience out of existence' (Alter 1978, p. 233). Where 'previous epistemologies talked about data of experience or objects of knowledge'; the structuralists and their successors only discuss signs, codes, texts, and intertextuality, in a deliberate exclusion (as we saw in Chapter 1) of language as a medium for expression. And whereas 'intertextuality' originally referred to the practice of literary allusion, that secondary process of imitatio (as it was called in the Renaissance to distinguish it from the primary process of mimesis, which drew on human life not on books), and to the conscious use of models implied by the whole notion of genre, now, Alter suggests, 'intertextuality' replaces 'the old-fashioned notion of verisimilitude'. In the once^new Paris theories, since the real world is itself only 'something we construe, a shifting constellation of texts we decode', then intertextuality can serve as an all-purpose, freshly-minted, but 'ultimately casuistic' term {ibid., pp. 233^4). This attempt 'to exorcise the other-than-literary presence of the real
creator and interpreters
131
i
1 :
world by reducing everything to text' is one of two 'complementary strategies' that Alter diagnoses in the iconoclasts' attack on mimesis (ibid,, p. 234). The other — building yet again, we see, on their travesty of Saussure—is to present the literary text as
a collocation of arbitrary signs that can be joined together only on the basis of internal principles of coherence even as they pretend to be determined by objects outside themselves to which they supposedly refer. In this view, reality, whatever it may be, is inaccessible to the literary text because of the text's very constitution, (ibid.)
Here the pseudo-Saussurian inheritance, even in such an intelligent critic as Gerard Genette,30 appears in the claim that since the linguistic sign is arbitrary then ' "the motivation of the sign, and particularly of the 'word', is a typical case of realist illusion'". With the term 'motivation' Genette was referring to the Russian Formalists' analysis of narrative elements and their internal functioning, which they used to rule out any discussion of a novel's relation to the society it portrays. Alter is disturbed by the 'either-or rigidity' of Genette's formulation, but as we now know, this technique of privileging your own case and disqualifying every other is the foundation method of the so-called literary theories that we have inherited from the 60s.
Alter's reply to 'the linguistic fallacy' of post-structuralist theory, which assumes a total coherence between its view of language and its view of literature, is to argue that the word 'tiger', for instance, may be a conventional and arbitrary sign, but that we know the difference between a tiger in a poem by Borges, one in a painting, and one in reality. The fact that the same perceptual apparatus and cognitive processes are involved in our recognition both of the image and the real thing does not mean that we can't tell them apart (pp. 235-7). This may seem an elementary reply, but after all, the mistakes made by Current Literary Theory are often elementary.31 As I showed in Chapter 1, philosophers and linguists have, independently of each other, built up a model of reality to serve as the framework for analysing how language can reliably refer to the world of persons, animals, objects, processes, and abstract categories i such as mimesis. Since literature uses language there is no; reason why it cannot refer to imagined worlds, animals, objects and categories, which may behave like those with which we are familiar, or may not. In refuting the iconoclasts' notion of language we must also refute their view of literature. As Alter says, this 'tendency to see literature as closed-circuit poesis rather than mimesis' would have disastrous effects on the novel, to which an untold number of readers since the seventeenth century have turned in order to 'enter into' the experience of fictional characters, 'testing our perceptions of reality against the absorbing details of their invented worlds' (ibid., p. 237).
Against the Literary Theorists' coercive and reductive 'not mimesis but
132
critical theories
creator and interpreters
133
poesis' we must assert a noncoercive 'bothY'and', truly 'freeing' ourselves from a new and oppressive critical system. Christopher Butler, rejecting 'useless discussion' concerning the' "pseudo-statement" nature of literature', argues that interpreters of a text can either relate it to the world or concentrate on its'internal thematics':
There is ultimately no.logically coherent or reasonable way of legislating in favour of one of these modes of interpretation: to the peraianent detriment of the other. What is at issue is the pragmatic ends which interpretation may serve. . . (Butler 1984, p.: 54)
And in fact, nearly all forms of interpretation outside the self-sufficient post-structuralist camp 'inevitably move to and fro between the two' (ibid.), Thomas Pavel similarly attacks these excluding dichotomies: 'Madame Bovary is both a conventional narrative structure textualized with the help of a set of equally conventional discursive procedures and a description of provincial boredom in mid-nineteenth-century France' (Pavel 1989, p. 119). Since Formalists have tried to 'conjure away the referential properties of realist texts', hoping to reduce them to 'a combination of purely formal characteristics', Pavel finds it necessary to restate the legitimacy of mimesis as a critical category for discussing literature, and to make the challenging counter-assertion that we should 'start out from the primacy of reference and representation to which narrative structures and discourse techniques are subordinate' (ibid., p. 111).
The concept of mimesis can be presented crudely* as the iconoclasts have shown all too easily, but it can also form the basis of a flexible and sensitive theory of Literature. Those wishing to travesty it frequently pick out some previous exponent (Aristotle, Sidney) or period (the neoclassic critics), and summarise their views inaccurately and sarcastically. Aristotle is taken to task for implying that literature imitates reality as if that were a given and the writer passively transcribed it.32 For late twentieth-century critics aware of the degree of interpretation involved in perception these are obviously primitive notions. But Aristotle in fact said that a drama was the mimisis of a human action involving words, music, and all the other resources of theatrical presentation. According to the Poetics,33 'the objects of this mimesis are people doing things' (48al), interacting, as we might put it, in the pursuit of shared or conflicting goals. The word translated there as 'doing', prattontas, means for Aristotle 'people performing responsible and characterizable actions', for, as he puts it. later, 'tragedy is a mimesis of an action; action implies people engaged in it; these people must have some definite moral and intellectual values' (49b36ff). Thus the dramatist's mimisis of character inevitably 'makes plain the nature of the moral choices the personages make' (50b8ff). Aristotle firmly places the dramatist's mimetic act in the realm of recognisable human behaviour having both an intellectual and an ethical dimension: a definition that Can still stand. As Aristotle knew better than anyone (see his two books on ethics, and the
I
masterly treatise on rhetoric), 'moral choices' involve disagreement and discussion, so that there can be nothing passive about the literary presentation of human behaviour. The dramatist, involved in the story being dramatised, cannot help making judgments about the issues concerned. And far from literature having the purely mechanical role of transcribing events as they had happened, that was the nature and limitation, according to Aristotle (deploying one of the original Greek evaluative dichotomies), of history rather than poetry: 'the one tells us what happened and the other the sort of thing that would happen'. History tells us 'what Alcibiades did or what happened to him', poetry tells us 'what sort of man would, probably or necessarily, say or do;what sort of thing' (51b3ff),- the picture it presents being 'an investigation of moral possibilities' (as Margaret Hubbard notes in her admirable translation).
This distinction between poetry and history on the grounds of generality had a vast influence in the Renaissance,34 and is one of the main supports of'Sidney's Apology for Poetry, accounting for the poet's superiority over both historians and philosophers. All human arts are based on nature, Sidney writes,35 but 'Only the poet, . . .lifted up with the vigour of his own invention, doth grow^ in effect another nature, in making things either better than Nature bringeth forth, or quite anew. ..' (p. 100). In a memorable image, Sidney describes how the poet 'goeth hand in hand with Nature, not enclosed within the narrow warrant of her gifts, but freely ranging only within the zodiac of his own wit' (ibid.). These famous lines have recently been interpreted by Howard Feiperin, a horn-again decon-structionist, who comments on that final image (the zodiac as the circle-through which the world moves) in purely Derridian terms:
poetry, that is, is an autonomous, generative, and self-enclosed system. To the extent that the two, poetry and nature, are heterogeneous, they are clearly incommensurate. If poetic language is orphic [a misplaced late twentieth-century concept, never used by Sidney] and autonomous, it can never quite be referential; if it is referential, it cannot be completely autonomous and orphic. (Feiperin 1990, p. 48)
Reading through Derridian spectacles;is not necessarily good for the eyesight. Sidney in fact states that although the poet is hot bound by mimesis, can freely invent, he is still walking hand in hand with' nature, and: his subject-matter is still the prattontas, people doing things. Any gulf between poetry and nature is created by Feiperin, who predictably goes on to accuse Sidney of 'self-contradiction', for 'to make something "better" than nature is not the same thing as. . . "other"; than nature', thus making Sidney prove a deconstructionist thesis, namely 'the potential frustration of representation by the self-enclosure of the formal systems that mediate it'
(p. 49).     ■ ■ ;        >■ .
But Sidney wrote riot 'other than nature' but 'another nature', that is, a new nature, having all the properties of the world and human life as we
134
critical theories
creator and interpreters
135
know them. Representation is not frustrated in Sidney's theory of poetry but enabled, gloriously so, above all other forms of writing. Felperin's final attempt to insert a deconstruetionist fissure into Sidney's concept of 'a paradise lost than can be represented but not regained through poetic language, "figured forth" but not literally delivered' (ibid.), reveals the same elementary error as his master's voice. The most that any linguistic utterance can do is to 'figure forth'. Only magicians, and. frustrated Derridians, believe that language could 'literally deliver' an idea or state, as if it could arise from off this page and we could enter into it. Such a confusion between the actual and the represented is amusing when we find characters in films (Buster Keaton's Spite Marriage, or Woody Allen's The Purple Rose of Cairo) who can walk into and out of the screen. But such a confusion coming from professional philosophers and literary critics, and then being used to discredit language and literature, is absurd and debilitating. As A.D. Nuttall has observed, the effect of Derrida's 'complaisant nihilism' on his followers is 'strangely enervating' (Nuttall 1983, p. 28). It removes personal initiative, encloses criticism within its wholly determined 'zodiac', where, as Alter, puts it, the effect of'insisting, for book after book, that the novel is a self-referential system of signs in which the real interest is in the way language calls attention to its own predicament', is to 'make all novels, from Eugenie Grandet to A Hundred Years of Solitude, sound tediously alike' (Alter 1978, pp. 237-8).
Mimisis, like any other concept or category, can be trivialised. Some neo-classical critics succeeded in reducing it to a mechanical criterion. In the seventeenth century Aristotle's system became broken up into rules, which then took on a prescriptive nature, intended to legislate how drama should be written now and in future. Plays were evaluated under separate Aristotelian categories: plot (or fable), characters, thought, style, in each Case appealing to the criteria of probability or correctness that then applied. Throughout the dominant period of neoclassicism (well into the nineteenth century), many critics also applied this prescriptive system retrospectively, without any awareness of anachronism, and Shakespeare in particular was found guilty of having violated rules that he could not possibly have known, including those involved with mimisis.36 The neo-classic system, like many others, subordinated individual aesthetic criteria to general notions of decorum and convention, as defined by polite society. The nature to be represented in literature was always selective, with clearly defined rules and exclusions (as it still is today); These governed both specific genres and the forms of life that could be represented, so that kings should behave in a king-like manner, comedy should not be mingled with tragedy, and so forth.
These criteria derived jointly from life and literature, but in his claim that reality is a mere 'effect' produced by literary means Barthes asserted that neoclassical literary theory opposed 'the "real". . . to the "vraisemblable"', and forbade any 'contamination of the "vraisemblable"
i      by the real' (Barthes 1982, p. 15). This is a specious claim, which derives, <      as A.D. Nuttall has shown, from a misreading of the single seventeenth-century text cited to prove it (Nuttall. 1983, pp. 57-9). For that period, as 1      for the following century at least, J.E. Spingarn's description of Hobbes's !      literary theory remains the norm: ' "the subject-matter of poetry is, then, the manners of men; its method is that of verisimilitude, or resemblance to the actual conditions of life" ' (cit. ibid., p. 59). That neo-classic criteria of probability derived from their own 'conditions of life' can be seen very !       clearly in Thomas Rymer's vicious attack on Othello in A Short View of Tragedy (1693). Rymer invoked not only the literary criterion of genre for defining tragedy and judging Shakespeare incapable of it (in the coarsest possible manner), but he also appealed to social and indeed racist attitudes. With a fine disregard for history (remarkable considering that he became Historiographer Royal in 1693), Rymer found Shakespeare guilty of improbability in his fable by fancying that the Venetians would 'set a Negro to be their General, or trust a Moor to defend them against the Turk. With .      us a Black-amoor might rise to be a Trumpeteer. . . . With us a Moor might |      marry some little drab, or Small-coal Wench. . .', and so forth.37 Rymer ■      could indict Shakespeare for violating probability by reference to the actual
I'j      conditions of life, but other critics writing at the same time and using the i      same system could defend him from the charge.38 This disagreement shows that no critical category (before post-structuralism, at least) is so rigid that it denies individual variations in its use, but it also shows that the concept [•      of mimesis is always governed by conventions, social and/or artistic.
Any attempt to represent reality must submit itself to the conventions of n      the medium it uses. As E.H. Gombrich showed in Art and Illusion,39 f      spectators of a painting bring with them a set of conventional expectations: I      there is no innocent eye; Yet, Gombrich argued, artistic conventions are far from arbitrary, being determined by the psychological constraints of visual perception. Resemblance constitutes a condition of representation, and artistic conventions have been introduced or regularly challenged in ,      the name of a more accurate resemblance. In the same way mimesis in
■ literature submits itself to the conventions of the age and culture in which K      the writer works. In the novel, as Robert Alter puts it, 'mimesis is never a
■ direct reproduction of reality but rather a way! °f eliciting in the mind of
■ the reader — through complex chains of verbal: indicators — the illusion of K persons, places, situations, events, and institutions convincingly like the K ones we encounter outside the sphere of reading' (Alter 1978, p. 238). At B the same time we accept, often unquestioningly, the conventions of the K, novel genre. We accept that the chapters of a novel can be numbered in E sequence, and that between two successive chapters the action can stay in E the same place and time, or shift to a new place, at the same or any other :» time, involving the same or (up to a point) any other characters. We K accept that the narrative can be told by one omniscient narrator, by a first-H!      person actor in the narrative, or even by two first-person actors alternately
136 critical theories
(as in Bleak House). We accept that the main narrator can draw on eyewitnesses and other agents to fill out the narrative, as Conrad's Marlow does in Lord Jim, and also that the result may be to shed not more light but more uncertainty on the main character and his actions. We even accept the so-called 'self-conscious novel', which keeps the reader constantly aware of being engaged in reading an act of representation. This may seem an extreme case where literary conventions push against the process of representation, but I agree with Alter that such works involve not so much the abandonment of mimesis as 'an enormous complication and sophistication of it: mimesis is enacted as its problematics are explored' (ibid., p. 239).40
The need to recognise that specific conventions govern the representation of reality in a given genre at a given time is particularly pressing for Elizabethan and Jacobean drama. The fact that there has been a continuous tradition of theatrical performance in England since the Middle Ages (evert during the Civil War) means that the, theatregoer:— and, until recently, the critic —- may be influenced in his attitude to older drama by the conventions and production practices of the contemporary theatre. The more remote the drama, the more likely we are to misunderstand it. (The 'theatre of cruelty' already seems remote.) One of the great breakthroughs in modern drama criticism was the conscious and constructive attack, in the 1920s and 30s, on Victorian attitudes to Elizabethan drama, attitudes formed by an alien, naturalistic tradition, as seen in such a typical figure as William Archer, It was constructive inasmuch as the scholar-critics involved, notably E.E. Stoll and Muriel Bradbrook, went back to Elizabethan drama to define the conventions that then applied. In Bradbroqk's suggestive definition, convention is 'an agreement between writers and readers, whereby the artist is allowed to limit and simplify his material in order to secure greater concentration through a control of the distribution of emphasis1.'*1 Conventions vary from genre to genre, each arousing and fulfilling the expectations proper to itself, a process which, as Wendell Harris has observed, 'implies participation by the reader', who is invited to read a text in a certain: way. The 'recognition that we are reading a poem leads us to attend more carefully to the sounds of words, the relations between them, the associations that accompany them, and the precision of their choice' (Harris 1988, p. 96). Since, in David Lewis's helpful formulation, a convention is 'a regularity . . .which benefits each member of a group' (cit. Blackburn 1984, p. 119), it is essential to discover what agreements are made by the group participating in the experience of drama, a genre more dependent than any other on those involved — actors and audience — simultaneously observing a whole range of conventions. And since, in any medium, conventions are usually silently observed, it needs an act of historical reconstruction to recall, those1 operating in the past. It is important not to lose sight of the conventions of Renaissance drama as scholars have retrieved them, especially at the present time,
creator and interpreters 137
when any piece of scholarship or criticism more than twenty years old risks being automatically and unthinkingly treated as ■ obsolete. Erroneously naturalistic expectations continually crop up in Shakespeare criticism, as with Greenblatt on King Lear (Chapter 4), the psychocritics Nevo and Stockholder (Chapter 5), or any number of feminists (Chapter 6), who express disapproval of characters on the simple everyday grounds that 'people ought not to behave like that'. These are wholly inappropriate criteria.
, The scholars who recovered Elizabethan theatrical conventions were able to show that Renaissance drama was symbolic rather than naturalistic, that it relied on broad imaginative effectSj not minutely observed detail. Frank appeals to the audience's cooperation were intended to overcome the deficiencies of representation, and a study of the theatrical props owned by the impresario Philip Henslowe—'one rock', 'one cage', 'one comb', 'one
IHell mouth', 'one bedstead', 'two steeples &. one chime of bells', 'one golden fleece', 'one lion's skin', 'a robe for to go invisible'42 —? reveals an odd mixture of the schematically suggestive and the faithfully realistic. The actors' costumes were gorgeous, costing the company much more than the play text. Dramatists received between £3 and £6, while Henslowe paid 'over £38 for taffetas, velvets and gowns for Cardinal Wokey' alone (Chillington 1980, pp. 448-9, 452). Costumes were also much sought after, as we see from the list of items that Henslowe records as.'Gone and lost': 'one orange tawny satin doublet; laid thick with gold lace'; 'one blue taffeta suit'; 'one pair of carnation satin Venetians [breeches], laid with gold lace'; 'Harry the Fifth's velvet gown'. In the Elizabethan playhouse real swords were used, pistols and cannons were set off, bells rung, candles lit, food: and drink consumed, all devices to aid the faculty of make-believe on which all representational works of fiction depend. Not that every invitation to 'pretend' needed elaborate helps. In the scene where Falstaff and Hal act out an imaginary confrontation between Hal and his father, the King, they seize whatever props are to hand: ; , FALSTAFF. This chair shall be my state, this dagger my sceptre, and this cushion my crown. . . . Give me a cup of sack to make my eyes look red, that it may be thought I have wept. . . . (1 Henry N, 2.4.374ff) And when the artisans in Midsummer, Night's Dream prepare their production outdoots, they transform the natural setting into a theatre: 'here's a marvellous convenient place for our rehearsal. This green plot shall be our stage,.this hawthorn brake our tiring-house ...'(3.1.2ff). . [Conventions applied not only to costumes-and'settings, to time and place (where the flexibility they used seemed scandalously irregular to neo-classical critics), but also; to the representation of human behaviour. Characterisation was often a question of types and roles: the melancholic lover, the quarrelsome braggart, the dignified king. Two obvious influences here were the classification of human types in ethics, rhetoric, and medicine
138
critical theories
creator and interpreters
139
(physiognomy, humours theory),43 and the sheer volume of plays that companies produced and performed. In the forty weeks between October 1596 and July 1597, for instance, 'the Admiral's Men at the Rose gave 184 performances of thirty different plays', fifteen of which were marked by Henslowe as 'new', a frequency of about two new plays per month (Chillington 1980, p. 446). Actors learning a new part every two weeks, while continuing to perform a dozen old ones, do not have time to construct elaborately naturalistic interpretations, any more than the dramatists involved in joint authorship could provide them with psychologically detailed roles. For the period 1590 to 1642, in which we know the authorship details of about 1,100 plays, anything from a third to a half of them 'incorporated the writing at some date of more than one man. In the case of the 282 plays mentioned in Henslowe's diary. . .nearly two-thirds' were joint-authored (Bentley 1986, p. 199). In some cases, as Carol Chillington has argued for Sir Thomas More, collaborators 'distributed the work by looking at a plot outline and selecting episodes that appealed to them'. Her breakdown of its eighteen scenes suggests a collaboration between Anthony Munday, Henry Chettle, Thomas Heywood, Thomas Dekker, and John Webster, with an almost communal revision process (Chillington 1980, pp. 461-9). That play was unusual in several respects, and ran into trouble with the censor Edmund Tilney. With other companies, it seems, a more common division of .labour was for writers to compose whole acts (Bentley 1986, pp. 227-34), even though their styles were noticeably different. In at least a dozen plays on which Massinger and Fletcher collaborated 'they obviously divided the first two acts between them and the style of neither appears in the other's act' (ibid,, p. 232 note). Whichever form collaboration took, different authors would necessarily create parts of the same personage, so that characterisation could only have been carried out using conventions as a kind of short-hand.
It is a well-known convention in Elizabethan drama, for instance, that a slanderer is usually believed. This may be thought to derive from the highly-compressed nature of the performing-time, but it is just as much a convention in early prose-fiction (such as the romance), and indeed the novel. It is a way of making things happen, and. the psychology concerned is not so much unlike life — do official denials always dispel newspaper rumours or malicious report? —for us to reject it as 'artificial'. Conventions were often subsumed under the criterion of verisimilitude in Renaissance literary theory, but this drew, in turn, on attitudes applying in the real world. To an age that conceived women as the weaker sex on medical, ethical, and legal grounds,44 it was inevitable that a literary theorist like Giraldi Cinthio should lay down a rule that 'young girls should be shown as shamefast and timidi matrons as chaste and solicitous' (Doran 1954, p- 221), The convention of women's weakness, as Madeleine Doran notes, was 'observed in at least one way in English drama. Women, even good women, generally yield quickly and easily to persuasion'. Queen
'. Anne in Richard III- (in an episode invented by Shakespeare) would be a familiar instance of this rule. Yet against it we must set some 'striking exceptions', such as Isabella in Measure for Measure, which 'gain force through our awareness of the convention. Isabella's resistance to Angelo is a sign of superior strength and nobility of character, not of prudishness, as
i modern readers are likely to feel' (ibid.). Another noteworthy exception is ' Marina in Pericles, sold as a prostitute but having so much power in S      eloquence that she can shame her customer Lysimachus, and turn him
into a husband. This transformation, which is all the more remarkable given the wide-spread Renaissance belief that women need not be taught rhetoric, since they také no part in public life,45 shows that conventions may be a type of shortcut that enable dramatists to achieve many effects quickly, but do not constrain them.
The relationship between convention and representation is variable, flexible. Yet the iconoclastic, formalistic criticism since the 1960s has produced what A.D. Nuttall describes as the 'curiously uncritical yet metaphysically radical' theory by which realism itself is said to be fa. tissue ; of conventions and therefore has nothing to do with reality'. Striking though this idea may seem on first reading,
The truth is almost exactly converse. All reference to reality (including Í:        pointing with the finger) is conventionally ordered. Language is an immensely rich system of conventions and is the best means we have of referring to the real. (Ňuttall 1983; p. 53)
IIn rejecting previous theories the radical critics have again used the yes/no, all-or-nothing claim. But no one of any sense is claiming that representational works of art present life 'exactly as it is'. The writer of realistic
Si      fiction, such as George Eliot, does not wish 'to dupe her readers into supposing that her books are factually true. . . She does however implicitly ;      claim probability for her stories', suggesting no more than that 'such events do occur in real life' (ibid., pp. 54-5). Representational art, Nuttall writes, does not 'transcribe actualities; it offers variously quickening hypothetical
ii cases. And these must be answerable.to reality or they will not succeed' (ibid., p. 77). The reality of people and things to which the writer refers is
;      conventionally: represented, and thus 'it is never possible in a finite work to i      exhaust reality; always we receive a selection only. ..' (ibid., p. T82). That restatement of a fundamental truth can be linked with what we must now probably call the 'rehabilitation' of the author, who made that selection from reality in the first place. A novelist's control over his material is total, r      including the power to falsify history, as Zola did in Germinal, making the mining industry look far worse than it was at: that: time (Butler 1984, p. ;      57).1" Yet no text guarantees a relationship between itself and the world.
That relationship is always hypothetical, and the estimate of the degree to jí which a text may or may not resemble the external world is made by the ■      interpreter, drawing according to his lights on the interpretative conven-
140
critical theories
tions that govern the existing conceptions of the 'vraisemblance' (ibid., pp. 53-4). What is involved is not merely the meaning of the text so much as 'its significance and hence use to us'(p. 57).
What use are texts? What do we do with novels and plays while reading or watching them? What role is played by their representation of reality in our experience of them? An attractive answer to these questions has recently been provided by Kendall Walton: Mimesis as Make-Believe^1 This thoughtful and wide-ranging study links up with a number of positions adopted here and in other recent literary theory, starting from the inadequacy of a purely linguistic model, which has produced so many 'distortions' when applied to novels, paintings, theatre and film (Walton 1990, p. 5). The approach through language has too often focussed on literal, non-fictional discourse, with works of fiction then seen as 'deviant' or 'parasitic'. However, Walton argues, fiction is something positive, special, not just language stripped of its normative functions (p. 78). Works of fiction 'are not necessarily vehicles of assertion or of any other illocutionary acts' (pp. 76, 81), so that speech-act theories alone cannot explain them, nor is their goal 'communication' in the sense of imparting information (p. 89). Walton expresses the view of every philosophically-aware person in not believing that 'reality is a realm of things-in-themselves independent of sentient observers, nor that to be true is somehow to picture or mirror this objective reality', since notions of truth and falsehood are notoriously culture-dependent (p. 99). But equally he rejects the extreme deduction, put about by the 1960s iconoclasts, that discourse creates reality: 'the insight that facts are not "brute" . . . is a far cry from collapsing the distinction' and. arguing that everything is fiction (p. 100). Walton judges the notions of truth and reality 'so central to our thinking' as to be 'inseparable from the subject matter of any investigation of human institutions. An investigator cannot dispense with the very thing he is investigating' (p. 101). Representation concerns precisely 'the difference between truth and fictionality, the possibility of propositions' being true but not fictional, or fictional but not true', and how these differences affect 'our personal and social experience' (ibid.)'. Actual objects of representation may indeed exist 'independently of the works that represent them' (p. 127), but the key issue is not the accuracy of the representation.
The core of Walton's theory is that all representations have in common 'a role in make-believe', which can be explained in terms:of imagination and participation (p. 4). Whereas some theories of fiction have made an unfortunate separation between aesthetic and metaphysical approaches, his theory unifies them by basing itself on make-believe (p. 6). 'There is nothing distinctively "aesthetic" about make-believe itself, Walton argues, 'and works of art are neither the sole nor the primary instances of
1
creator and interpreters
141
representation in our sense', for children's games of pretending will do just as well. Since I have been arguing that the primacy of human behaviour as represented in fictional works means that our response to them cannot take some special aesthetic form, but must involve the same feelings and the same language that we would use in real life, I welcome Walton's theory as allowing us 'to see representationality in the arts as continuous with other familiar human institutions and activities rather than something unique requiring its own special explanations' (p. 7). To understand 'paintings, plays, films, and novels', he writes, in words that will horrify the critical avant-garde, 'we must look first at dolls, hobbyhorses, toy trucks, and teddy bears' (p. 11). Just as these toys function as 'props' in games of make-believe, so do representational works of art, which permit 'indefinitely many' games of make-believe, 'played by different appreciators on different occasions' (p. 51)'. Imaginings resemble day-dreams, being mental events which are not disconnected but woven together into a continuous sequence (pp. 13-17). Yet neither are they 'free-floating fantasies': 'most imaginings are in one way or another dependent on or aimed at or anchored in the real world' (p. 21), and many imaginings are about real things (p. 25).
One strength of Walton's theory is that unlike those who invoke 'jeu' as a desirable state freed from all constraints — it takes the notion of game seriously, and follows out the implications of the model, to which certain conditions apply. When something is true in a game of make-believe, Walton suggests> the proposition is 'fictional, and the fact that it is fictional is a fictional truth', true in the world of fiction, not necessarily outside it (p. 35). The props provided by representational works of art, created by and for human beings, 'function only in a social, or at least human, setting' (p. 38). As in any game of make-believe, some rules must apply. The work of art needs 'a certain convention, understanding, agreement' from, its players (ibid.), most importantly 'the acceptance rule' by which we agree to ■take part in the game world (p. 44). Imaginings (like language, I add, and like the creation and interpretation of art works) are not 'free, unregulated', but 'constrained also; some are proper, appropriate in certain contexts, and others not'. So, 'a fictional truth consists in there being a prescription or mandate in some context to imagine something' (p. 39). The person participating, while bound by some rules of context, is free both to share and enlarge the game or dream: 'the body of propositions fictional in the dream is to be filled out in certain natural or obvious ways, preserving the coherence of the whole' (p. 46)., The artist need not define everything.
Walton's theory is based on the fundamental point, often lost sight of in iconoclastic anti^theories, that works of art exist in order to be appreciated; and that the appreciator's experience involves him or her in participation. 'Appreciation of representational:works of art is primarily a matter of participation' (p. 213), for it is chiefly by imagining ourselves facing certain situations that we come to terms with our own feelings, or
142 critical theories
can understand other people's (p. 34). By taking part in a game of make- \ believe we consider its 'rules or principles of generation' as applying to us, too (p. 209), so that participants in games of make-believe 'are thus props, i objects, and imaginers all three'. They prescribe imaginings, which are in , some way both about themselves and addressed to themselves, even though I they concern a character on a stage or in a novel. The participant takes j part in the game by imagining about himself 'in a first-person manner... I he imagines this from the inside' (p. 212), and so he achieves 'a kind of . empathy with the characters, an ability to look at things more purely from their points of view' {p. 237). This notion of first-person participation, as if we were on speaking terms with the characters, helps account for the fact ' that we often .
seem to be in psychological contact with characters, sometimes even intimate with them. We have epistemological access to fictional worlds. . . . Often we are privy to characters* most private thoughts and feelings. And we respond to what we know, apparently, in many of the ways in which we respond to what we know about the real world, (pp. 191-2) '1
Our involvement can take quite intense forms: 'fictional characters cause real people to shed tears, lose sleep, laugh, and scream' (p. 192). 'Fiction- ; ally one may fear for someone else, if not for oneself (p. 250)— both, I would suggest.* It is not so much that 'appreciators lose touch with reality tf when they are immersed in a work of fiction' (p. .241), as that most i of us (Don Quixote excepted) have the ability to live in two worlds ] simultaneously, and can tell the difference between them. As in life, our responses to characters in literature can be complex, a mixture of attraction and resistance, anger and respect, sympathy and resentment (p. 253).
While applauding this whole direction in Walton's argument, I would add the qualification that although our involvement with fictional char-actets resembles similar feelings for people in real life, it is both more pronounced and more explicit. It may only be once or twice ih real life (if we're lucky) that we are ever confronted with the totality of devastation that we feel when participating in The Trojan Women or King Lear. It is very seldom in life that we see a crucial issue facing us as unavoidably as Cordelia does, or experience a sense of uncertainty as totally as Othello, undermined by Iago. When I read Middlemarch I can share the crucial stages in the lives of Dorothea or Lydgate, the decisive decisions or encounters that will shape their whole beings, with a clarity and coherence that my own life seldom brings. In life we pass through crises sometimes without knowing it: never in literature. In that sense the games of make-believe that we play are unrealistic. Literature makes an artful selection from life, speeded up, brought to a crisis and resolution, within the bounds of the time it takes to watch a play or read a novel, the whole represented experience being subjected to rational explanation. (Even works that do not clarify the main characters' situation or life in general make it clear
creator and interpreters 143
that they have reasons for not doing so.) We understand works of literature far better than we understand our own lives, and they form satisfying wholes, aesthetic and ethical and intellectual unities, in a way that life seldom does. (This is a potential source of danger, too.) But I agree with Walton that the illusion of life is created for our participation, and that 'we don't just observe fictional worlds from without. We live in them', experiencing what the characters experience. We know that these worlds are fictional, but 'from inside they seem actual. . . and our presence in them .    gives us a sense of intimacy' (p. 273).
That participation, that sense of intimacy, is shaped and controlled by the artists who create and/or perform the art work, the author, actors, producer, lighting-engineer, camera-man, editor, all with their experience of guiding make-believe. As Walton puts it, 'creators of props can predict how their creations will be used, and so can direct people's imaginings by designing props appropriately' (p. 53). If fiction-making is the activity of constructing the props to be used in games of make-believe (p. 88), then, I suggest, study of that shaping activity may help us to understand better the work that has been constructed, why it takes one form rather than another, what modes of imagining it directs, and why. If it is true that 'representations do not just present us with a collection of fictional truths; they order and arrange them for us, focusing on some more than others' (p. 149), then the selection and forms are the author's doing, and to a lesser extent the performers'. Walton concedes that 'much of what is fictional in appreciators' games is determined by the work appreciated and by the artist responsible for it' (p. 228), just as with a painting 'the artist's choice' has a decisive effect on the viewer's visual games of make-believe (pp. 318—27). We 'are not free to play any game we like with a given prop' (p. 303). While make-believe functions: equally well in literature and the representational arts, a particular strength of literature is that 'words are well suited for use in make-believe' since they come 'with built-in semantic and syntactic properties', so an author can combine them 'to prescribe imaginings' in quite precise, or delimited ways (p. 353). The: Derridian 'indeterminacy of language' thesis receives no support here, either.
Kendall Walton's study of mimesis as make-believe made me concede at once its main argument, as if it were something I had always thought, but never got round to formulating. It is perfectly symptomatic, as he points out, that many discussions of fiction, visual or verbal, never mention make-believe, pretence or imagining, so deeply are we immersed in it (pp. 388, 390). Yet, while retaining our awareness of the difference between it and reality, we must remain 'sensitive to the fictional world':
Appreciation and criticism, participation arid observation, are not very separate. . . .The critic usually cannot get very far in describing the world of a work unless she allows herself to bfe caught up in the spirit of pretense to some extent, as appreciators are. i . . (p. 394)
144
critical theories
Interpretation presupposes participation, for understanding a work of art is not possible unless we have first taken it in, taken part in it as an organised whole. The main weakness of many Shakespeare critics writing today is that, intent on practising the type of interpretation of the school to which they belong, they tend to lose all notion of experiencing or interpreting a play as a whole.
While Walton's book obviously does not solve all the problems concerning mimesis in literature, his view of representation as a collaborative activity between author, actor, and audience seems tp me a great improvement on anything yet produced.49 His account of the relation between the art-work and the audience's involvement with it needs to be. supplemented with the theory 1 have outlined abpve recognising the relevance of. the author's intentions towards both the art-work and the, audience. The two approaches are complementary, and if properly integrated open thfcway for a coherent approach to literature as a totality. Given the element of convention involved in representation and make-believe, a suitable conclusion to this section might be a passage from Borges' story 'Everything and Nothing', with its tribute to Shakespeare. Here Borges describes the dramatist coming to London as a young man and finding
the profession to which he was predestined, that of the actor, who on a stage plays at being another before a gathering of people who play at taking him for that other person.50
SHAKESPEARE AT WORK: THE AUTHOR TRANSFORMS HIS SOURCES
. . it is pleasant to see great works in their seminal' state, pregnant with latent possibilities of excellence;. i. to trace their gradual growth and expansion, and to observe how' they are sometimes suddenly advanced by accidental hints, and sometimes slowly improved by steady meditation. Samuel Johnson51
Dr. Johnson's comments on Milton's sketches.for Paradise Lost come from a work which the proponents of Current Literary Theory would deter anyone today from emulating, TJie Lives of the Poets. Johnson was himself a poet, a professional writer in several genres, and his remarks on the unpredictable growth of literary works doubtless derive from his own experience and from studying many authors and many books over a long literary life. Some of our contemporaries, content to discuss literary theory without reference to concrete examples of authors creating and revising their works, have been able to take the Barthes-Foucault attacks on the concept of an author as constituting an unquestionable principle of post-modem aesthetics. Having
creator and interpreters
145
accepted all its implications for a dehumanising of the reader (a mere focal point, a three-dimensional space on which traces of the passage of a text have been left), a reduction of all written works to an undifferentiatable mess of ecriture and intertextuality, and a depersonalisation of language to a discourse lacking agents — this fortunate few must find the literary criticism of a previous age:, or of any school other than their own, unreadable, incomprehensible. What is this thing called an author? What ate these 'works', let alone 'great works'? How can they be 'advanced' or 'improved' by human agency? Presumably such readers imagine intertextuality washing, over texts at various intervals in time, adding a character here, deleting one there, changing motives, altering beginnings and endings, filling out the background of the representation of a given society at agiven time with the relevant historical, geographical; social, ecpnomic, and cultural detail. Can discourse do all these things? How remarkable!
Such might well be the reflections of a very naive reader brought up on a curriculum limited entirely to the iconoclasts of the 60s and their imitators of the 70s and 80s. It may seem only a pleasantry to imagine the effects of an education confined to such a canon, but a growing number of commentators disenchanted with iconoclasm are beginning to realize the effects such a School of Criticism and Theory would have on its pupils. 1 do not raise this as a real possibility, since 1 find it hard to believe that any teacher could so effectively close off all alternative approaches, and I dislike the sensationalising of intellectual issues, that arousal of anxiety or fear of the enemy as a means of rallying the forces of good sense against the threat from outside. Such polarisation and demohising are in fact the techniques used by the ideological schools of criticism which I shall be discussing shortly, and to emulate them would be to risk reducing literary criticism to a no-mah's-Iand.
My aim in outlining the possible consequences of an exclusive diet of iconoclastic literary theory'is to show that its rejection of so many fundamental concepts and categories would leave us with no way of discussing individual writers, or individual works. It would also make it impossible to think about a writer's development,, from The Comedy of Errors, to The Winter's Tale, say, or from Titus Andronicus to King Lear. Shakespeare's output covers a period of twenty to twenty-five years at the most, which would amount to a single 'episteme' of Foucault (since only one episteme is supposed to operate in any given period). Therefore, unless they are really going to ascribe everything to intertextuality, exponents of a depersonalised theory of literature would have no way of accounting for the differences between Shakespeare's early attempts -—by no means negligible, on any criteria — and the masterpieces of his maturity. But since their literary theory couldn't detect such differences in the first place, having abandoned the concept of character, and rejected as 'formalism' a critical concern with plot, structure, poetry, prose, rhetoric or any approach other than that approved by their ideological carnp^ then, presumably they would not be
I
146
critical theories
disturbed at the inadequacy of their theory here; indeed they could hardly recognise it.
To readers of another persuasion, however, it makes a lot more sense to think of the author as the controlling consciousness that draws on a lot of varied material, his own or other people's experience, often recorded in books (histories, biographies, other poems, other novels, other plays), organising it into a new literary work by the processes of selection, omission, amplification, and invention. To use sources as an aid or framework for individual creation has been the common practice of many writers in many cultures at many times. For some the recollection of their reading may be unconscious, as J. Livingston Lowes claimed for Kubla Khan.*1 For others we have abundant evidence of what they read and how.they used it. Modern scholars have reconstructed George Eliot's 'Quarry for Middlernarch' and other notebooks.53 We are extremely well informed as to'what Joyce tead and how he used it in Ulysses and Finnegans Wake,5* or just what Wagner did to his sources for Der Ring des Nibelungen.55 Joseph Conrad, accused in 1898 by a reader in Singapore of displaying in Almayer's Folly a 'complete ignorance of Malays and their habits and customs', defended himself in a letter to his publisher, William Blackwood:
Well I never did set up as an authority on Malaysia. I looked for a medium in which to express myself. I am inexact and ignorant no doubt (most of us are) but I don't think I sinned so recklessly. Curiously enough all the details about the little characteristic acts and customs which they hold up as proof [of ignorance] 1 have taken out (to be safe) from undoubted sources — dull, wise books. It is rather staggering to find myself so far astray.56
One writer who specialised in reading 'dull, stupid books' was Flaubert, who for some thirty years kept a notebook (which he called 'un sottisier') in which he entered specimens of human stupidity. He then decided to use this material for a novel, originally to be called Les Deux Chportes (The Two Woodhce), which became Bouvard et Pecuchet, unfinished at his death in 1880, and which bears the evidence of his having read, and taken notes from, some 1,500 books.37 Flaubert may be a special case in this hovel, but it remains quite normal for writers to research their works. The contemporary Peruvian author, Mario Vargas Llosa, based his novel La Guerra del Fin del Mundo (1981), set in late nineteenth-century Brazil during a millenarian uprising, on a novel, Os sertoes by the Brazilian Euclides. da Cunha, and- on several historical' accounts. Comparison between the sources and the novel reveals all kinds of transformations.58
Shakespeare is no different from other writers in using sources and reshaping them to express his own conception of a story and-the issues raised
creator and interpreters
147
by it. We know a great deal about his life, and although documentary evidence is lacking on this point it has long been evident that the range of his knowledge of classical authors, myths, and customs can only be accounted for by assuming that he had a normal grammar-school education. The sixteenth-century grammar school had a limited and essentially literary curriculum, based on acquiring a working knowledge of Latin grammar, rhetoric, and compositional forms, but it instilled its teaching with formidable thoroughness.59 There can be no doubt that Shakespeare had been through the normal training in verse and prose composition, and could read Terence, Caesar, Ovid, and some Virgil in the original. When it came to collecting materials for a play, Shakespeare used two major sources in English, Holinshed's Chronicles of England, Scotlande andlrelande, and North's English version (via Amyot's French) of Plutarch's Parallel Lives of the Greeks and Romans. Other translations he is known to have used include Chapman's Homer, Holland's Pliny, and Florio's Montaigne. With Ovid's Metamorphoses, a popular school text that he had studied in Latin, he regularly used Arthur Golding's English version. Yet at some points he went back to the Latin, and translated more exactly (Muir 1961, pp. 3-4). There is ample evidence that Shakespeare had 'read Latin works of which there was no translation — two plays by Plautus, Buchanan, Leslie, some of Livy. . .' (ibid., :p. 4), and he certainly knew enough French60 and Italian61 to be able to refer to sources in those languages if a story interested him. The full extent of his reading can now be studied in Geoffrey Bullough's magnificent collection, Narrative and Dramatic Sources of Shakespeare, which reprints sources and analogues, both definite and possible, with full critical discussion.62
Source-study can help us to reconstruct the process by which Shakespeare formed a play. Given that any literary work is an intentional product, by comparing the finished play to its sources we can deduce something of the author's intentions towards that work by considering what, he included, omitted, changed. Throughout, the time he spent reading, planning, and writing a play, usually between three and nine months (given his responsibility for at least thirty-seven plays over a period of about twenty years), we can think of Shakespeare performing a sustained exercise of discrimination. As Susanne Langer wrote,
Art, and especially dramatic art, is full of compromises, for one possible effect is usually bought at the expense of another;, not all ideas and '  devices that occur to the poet are cd-possible. Every decision involves a rejection.63 '
This tissue of choices and rejections is governed by an aesthetic, a conception of drama which, consciously or otherwise, guided the process of invention, transformation, and integration by which a play became unified and. complete in itself. Like other Elizabethan dramatists, Shakespeare had .a > working notion of plot in drama as implying a completed unit of action.
1
148
critical theories
creator and interpreters
149
Perhaps they knew some of the commonplaces of classical literary theory, which itself derived from a discipline they certainly knew at first hand, rhetoric (the major document being Horace's Ars poetica);64 The period in which they worked had in any case seen an enormous-explosion of literary theory, most of it written in the universal language of Latin.65 Another influence was from Latin drama, for the plays of Plautus and Terence were staples of every grammar school throughout Europe, both for their Latin and their improving doctrine, while Seneca's tragedies attracted many would-be actors and dramatists.66 Schools and colleges regularly performed their own plays in Latin and English, pieces which, whatever their other faults, did not lack knowledge of classical models and the appropriate decorum.67 Despite some pseudo-Marxist attempts to fence Elizabethan drama within the confines of 'popular culture', the dramatists were — by the standard of their contemporaries, and even more by ours — well-educated men, often keen to display their knowledge. Some had a University education (Marlowe, Lyly, Greene, Peele), but the output of those who had 'only' been to grammar school (the,'only' typifies the unconscious imposition of our social experience.on to theirs) proves that they could cope with quite demanding Latin texts, while autodidacts like Jonson and Chapman were as knowledgeable as any professional writer in Europe. The.English Renaissance, as J.W. Binris has recently shown, was open to international neo-Latin culture in ways that many modern scholars could not emulate.68
Elizabethan and Jacobean drama constitutes an almost unique phenomenon, a medium in which highly and lowly educated people could take part in the invention, performance, and enjoyment of works that drew on a great deal of literary and historical knowledge while including an immediately comprehensible, and often enthralling presentation of human behaviour. As the output of plays and the number of dramatists increased, actors and theatregoers grew in proportion, and conventions governing the writing of tragedy and. comedy, already familiar from classical. models, became widely accepted (Doran 1954, pp. 55-7, 85-8, 105-11, 217-20). The conventions of such newer genres as the chronicle play and tragicomedy had to be worked out, often in the face of disapproval from the more classically-minded, such as Sidney and Jonson. Without digressing into literary or theatrical history, we can simply affirm that all writers faced the problem of synthesisirig their literary and historical sources, which could be vast and diffuse, into the 'two-hour traffic of the stage'. They could all say, with the chorus in Henry V, 'we'll digest / Th'abuse of distance, force a play' (2, Prol. 31-2). They must all have had some notion of how much 'stuff', copia, was needed to .fill a play, give it a satisfactory body and extent. Since, as we have seen, dramatists regularly collaborated on plays, with as many as five involved in an agreed division of labour66' (something like 'story conferences' must have been regular events at the Globe or Swan), then they must have had some shared
f aesthetic principles for drama.70 When Hamlet asks the First Player to
* recite the speech of 'Priam's slaughter' from 'an excellent play, well
jf digested in the scenes, set down with as much modesty as cunning'
\ (2.2.439-40), he is drawing on criteria for both structure arid style, and
r when he rebukes those clowns who speak 'more than is set down for them'
- and 'themselves laugh too, though in the mean time some necessary question of the play be then to be consider'ď (3.2.38-43), he is appealing to a concept of dramatic focus as against digression and dissipation. The
I Elizabethan dramatists' working criteria for the composition of a play must
í have concerned the form, extent, and above all finality of the fable.
1 'Tragedies end with deaths, comedies with weddings, history plays with
' coronations or peace-treaties' — such a rule of thumb might be a starting
f point. As Thomas Heywood put it, in his Apology for Actors (1612:
I paraphrasing the fourth-century grammarian Evanthius in his commentary on Terence), 'Comedies begin in ttouble, and end in peace; tragedies begin
I in calms, and end in tempest' (Doran 1954, pp^ 106-107). While comedy
I dealt with the private affairs of citizens, especially love, ending with a
r happy solution to all problems (ibid., pp. 107-108), 'ruin in tragedy meant'
I' death' (p. 118);
I Shakespeare clearly had a concept of final unity. All his comedies end
|i with weddings, sometimes multiple, having overcome any number of
I obstacles. The one exception is called, precisely, Love's Labour's Lost, in
I, which, as Berowne complains of the year's penitence imposed on the lovers
K before they can renew their suit,
ť Our wooing doth not end like an old play;
I Jack hath not Jill. These ladies'courtesy
I Might well have made our sport a comedy. I
I; King.   Come, sir, it wants a twelvemonth and a day,
I And then'twill end.
I BEROWNE.     .   That's too long for a play. (5,2.870ff)
i    The chronicle material that Shakespeare used was all 'too long for a play', ft    and. anyone interested can see for themselves how he went about his task of I    cutting and re-shaping. Not that he read slowly, word-by-word: Matthew ■    W. Black, looking at his shaping of Richard II from Holinshed, thinks he i    can trace Shakespeare's eye skipping as it read, first sifting the summaries in the marginal notes.71
If the chronicles are inherently shapeless, by definition inclusive rather :    than selective, much of the narrative rriaterial, whether verse or prose, that Shakespeare used for the histories and: comedies is too drawn-out for the ;    stage, being designed for sixteenth-century readers, who either read slowly j;   or.liked their money's worth in bulk at least. Adaptation of prose-fiction for drama carried its own problems, as has been shown.72 Verse sources needed the same drastic handling. The events narrated, in Arthur Brooke's fiwnews and Juliet take nine months, in Shakespeare's play only a day or two,
iso critical theories í
a reduction of time with a corresponding growth of intensity. Adapting 3
Chaucer's Troilus and Criseyde, with its long-drawn and marvellously subtle ]
account of a love affair building from desire to fruition to betrayal, \
Shakespeare starts late in the sequence and compresses the lovers' whole j
union and separation into half-a-dozen scenes, interwoven with the war |
material derived from the medieval chronicles of Troy and Chapman's »
Iliad.73 The sudden parting forced on Troilus and Cressida is as much a j
shock to the audience as it is to them. í
Dramatic unity, as Shakespeare practised it, referred not to the two ;
lesser unities of place and time (although already formulated as prescriptive j
literary theory in the neoclassicism of Castelvetro), but to the major ]
criterion of unified action. Two principles that he seems to have observed ]
were that the deaths of characters, when relevant to the action, had to be i
contained within it; and that all characters should be integrally: related -j
either to the main or subplot, and ideally to both. Not all deaths were j
relevant to the action as Shakespeare conceived it. In Greene's Pandosto, ]
his source for The Winter's Tale, the main character Pandosto (Leontes), !
having failed to recognise his sixteen-year old daughter Fawnia (Perdita), j
lusts after her, attempting her chastity with threats of force. This is one ôí 1 several titillating scenes in his sources that, as Alfred Harbage showed,
Shakespeare (unlike some dramatists) deliberately omitted.74 Greene ends S
his novel, after the reunion of Leontes and Polixenes, and the wedding of ]
Florizd and Perdita, by making Leontes commit suicide in self-disgust, j
Shakespeare keeps Leontes alive since he has kept the slandered wife alive í
(in Greene Bellaria really dies), in order to end his play with the recog- j
nition and reunion scenes so essential to the genre of Greek romance, on I which he had long drawn. The continuity of life and love are proper to
comedy, so Leontes lives. 1
In the sources Shakespeare used for his tragedies the main characters
often live on in the rhythms of real life. Fúr Hamlet he used the Histotres ;
TragiijMes of Belleforest (1576, 1582; not yet translated), in which the hero I
successfully revenges himself on his uncle, murderer of his father, and is < crowned King. Subsequently he evades being killed by Hermutrude, Queen of Scotland, who has slain all her previous suitors but falls in love with Hamlet,and 'marries' him. (He is already married, but his first wife doesn't object.) A wishfulfilling hero, not unlike the Trickster in many folktales and myths, this Hamlet is only killed a long time later, by another man, in
a different quarrel (Bullough VII: 6-9). His death is 'outside' the events of ; the main plot, and unrelated to it, typical enough for chronicle material but of no use to Shakespeare's conception of tragedy. The source for King
Lear, Geoffrey of Monmouth's História regiumBritanniae, also has several j
unrelated climaxes. Lear divides his kingdom among his three daughters :
but disinherits Cordelia for not flattering him. Discovering the ingratitude .
of Goneril and Regan, Lear goes to Cordelia (who has married the King of t
Gaul), is well received, and is helped to regain his throne.1 Three years '
creator and interpreters 151
later he dies. Cordelia then becomes Queen and rules for another five years. Then her late sisters' sons depose her and put her in prison, where she kills herself (Bullough VII: 272-3, 311-16). This is the version retold in the Mirror for Magistrates and The Faerie Queene, both known to Shakespeare (ibid.: 274—6, 323-34), but he also knew The True Chronicle Htstorie of King heir and his Three Daughters (1594; publ. 1605). The anonymous author of this play made the obvious decision over dramatising the chronicle, cutting it off at the point where Lear is restored to his throne and apologises to Cordelia and to his good counsellor for the wrongs he has done them (ibid.: 276-83, 287-98, 337-402), a happy end that Nahum Tate was to revive so successfully.
In Hamlet and King Lear Shakespeare reshaped his source material to give a tragic ending, and in both cases he did so by a remarkable demonstration of an author's shaping power, which Current Literary Theory would be wholly unable to describe. He attached a second level of plot to the main action, then intertwining the two to create many significant parallels and contrasts before making them collide destructively. Multiple plots are a well-known feature of Renaissance drama,75 and the problem that English dramatists set themselves was not 'just the conversion of one story irito a play, but the combining of two or three stories' into a coherent whole (Doran 1954, p. 298). In some plays the levels of plot run in parallel, without impinging on each other, in others unity is attempted by a contrast of tones, with the comic action parodying the serious one (ibid., pp. 289-91). The great classical model for the skilful interrelation of double plots was Terence (p. 277), and contemporary Italian critics introduced the concept of a '"favola intrecciata", in which all lines... lead into and out of a central knot' (p. 301). As a budding author Shakespeare had outdone Plautus' Menaechmi in The Comedy of Errors, involving not one but two pairs of identical twins, developing the master-rservant convention in ways that contemporary Italian comedians might have envied,76 and inserting an episode from Plautus' Amphitruo for good measure. In the later comedies we can see him learning from Lyly,i as G.K. Hunter showed in a classic study,77 setting up several parallel strands of action: before intertwining them, achieving deliberate effects of symmetry and repetition (Bullough VIII: 355-61). ;
The ability to intertwine plots presupposes that they have something in common, similarities of subject or theme which allow the dramatist to create meaningful juxtapositions. To the main plot of Hamlet, involving a father murdered and a son seeking revenge, Shakespeare added a subplot where another father is murdered. Hamlet, having killed Polonius (a deed which drives Ophelia to madness and suicide), becomes in turn the object of Laertes' revenge. That parallelism in the private world is echoed by one in the public realm, with Fortinbras: trying to achieve compensation for the death of his father and the loss of his heritage not by covert revenge but by open military action. Shakespeare makes the parallels and contrasts
152
critical theories
creator and interpreters
153
between these three avenging sons explicit, in Hamlet's comparison between himself and Fortinbras (4-4.32-66), and in his resemblance to Laertes; 'by the image of my cause I see / The portraiture of his' (5.2-75-8), adding a fourth one in the Player's speech describing Pyrrhus killing Priam (2:2.448-518). The two major plot-levels are then made to cross destructively as Laertes, now the instrument of Claudius, kills Hamlet and is killed in turn by him- Having finally refused covert revenge Hamlet is able openly to kill Claudius in retribution, and without planning. Significantly, it is Fortinbras, the only character remaining outside the revenge action, who survives. In King Lear the sub-plot again parallels situation and theme. Shakespeare took a striking instance of filial ingratitude from Sidney's Arcadia, the story of the Paphlagonian king, who rejected his good son, Leonatus, trusted his bastard son Plexirtus, only to be blinded by him and cast out. (In the source the good son returns to nurse his father, defeating the bastard's murder-attempt, yet then pardons the evil son, who lives on to wreak more evil.) Shakespeare, having set up a remarkable resonance between the mad Lear and blinded Gloucester, brings the two plots together, once again with a destructive effect. Edmund is linked to GoneriL and Regan by a web of lust and rivalry which makes one daughter kill the other and then herself, while he has the third hartged, so causing Lear's death. Edmund is himself killed by'Edgar, a justified revenge and the logical consequence of Shakespeare's conception of tragedy as a disaster deriving from human conflicts in a strict pattern of cause and effect. In this play he not only makes the two plots echo and cross each other destructively, but also allows the blinding of Gloucester to create a whole sequence of imagery relating to sight and blindness in the main plot (Bullough VIII: 395-6).
The central problems facing the dramatist, Madeleine Doran observed in her pioneering study of form in Elizabethan drama, were 'how to get concentration, and how to achieve organic structure', requiring a successful 'fusion between story and characters' (Doran 1954, pp. 265, 250). The last play in which I want to explore, in rather more detail, Shakespeare's deliberate and controlled transformation of his sources is Othello, usually regarded as having a single plot and a single source, the Hecatommithi or 'Hundred Tales' of Giraldi Cinthio (1565; not translated). In fact, there is evidence that Shakespeare did some well-directed background reading for a plot-element that he added to the play, the Turkish military threat. Cinthio has no mention of the Turks, but Shakespeare was well aware of the danger they represented in the Mediterranean in the sixteenth century. Just as in King Lear, where he drew on a recent exposure of a Jesuit pretence to exorcise, demons, Samuel Harsnett?s Declaration of Egregious Popish Impostures (1603), to give Edgat in his feigned madness the authentic
jargon of devils,78 for information about the Venetian-Turkish wars Shakespeare consulted a recent book (the play was written in 1603-04), Richard Knolles's History of the Turks (1603), alongside two older authorities. Having assembled these background sources (Muir 1961, pp. 128-9; Bullough VII: 211—14* 262-5), Shakespeare then wholly integrated them into his plot. (It is always a surprise to start from a play you know well, then to compare.the sources and find borrowings which you could never have suspected, so seamlessly are they absorbed.) News of the Turkish fleet sailing towards Cyprus reaches Othello just after Iago has told him of Brabantio's anger at Desdemona's elopement, beginning that interplay between private and public life that is to run through the whole play. Cassio reports'that the Council of Venice is in emergency session) having 'hotly called for' Othello, and sent three search-parties to find him (1.2.34ff). Othello is prevented from going there at once by Brabantio's angry arrival, but the news of this urgent business of state in turn means that Brabantio has to accompany Othello to the council-chamber in order to lay a formal accusation of seducing his daughter by magical arts. In this invented episode of a man defending himself against the charge of witchcraft (for which Shakespeare drew on classical models),79 :Othello is allowed to demonstrate that dignity and imperturbability in the face of danger or misfortune which belonged to the classical virtue of fortitude, and which was his settled characteristic before Iago destroyed him. ('Is this the nature / Whom passion could not shake?', as Lodovico subsequently asks in amazement: 4.1.265ff.) Shakespeare invented not only the accusation and self-defence, but added several details in this scene to show Othello's high standing as a servant of the state: the fact that the Duke welcomes him so warmly — 'Valiant Othello, we must straight employ you / Against the general enemy Ottoman' (1.3.48) — and justifies appointing him governor of Cyprus in place of Montano as having a superior reputation (221ff); the readiness with which the Duke accepts Othello's request to allow Desderhoha to go with him (275ff). The picture that Shakespeare creates of Othello in Venice is of a man who has_ triumphantly overcome all obstacles in both public and private life.
Having introduced the Turkish threat to motivate Othello's despatch to Cyprus,, Shakespeare dispels it with the same creative freedom by calling up a storm to destroy their fleet (2.1.1-32). But he then makes the storm serve several dramatic functions. Not only Cassio but the whole population of Cyprus expresses anxiety for Othello's safety (2.1.53-4), showing their dependence on him. Shakespeare's placing of Desdemona and the other Venetians in a second ship, which arrives first, puts Desdemona in the temporary care of Cassio and Iago (2.1.83-179), a situation immediately used to display Iago's sinister abilities in improvising'witty (and misogynist) verses, and to reveal the true depth of his hatred of Cassio ('Yet1 again your fingers to your lips? Would they were clyster-pipes: for your sake!').'The foregrounding of Iago for the audience in asides and soliloquies sets up a
154
critical theories
creator and interpreters
155
double action by which Shakespeare can both celebrate Othello's reunion with Desdemona and undermine it. Othello's sense of having attained the highest degree of human happiness seems immediately vulnerable given Iago's aside to us: 'O, you are well tuned now! But I'll set down the pegs that.make this music, as honest as I am'. On the outer level of events, the public rejoicing ordered by Othello to mark the 'perdition of the Turkish fleet' (2.2) allows a scene of drinking and revelry, which Iago can exploit to make Cassio drunk and to foment the fatal quarrel that has him cashiered (2.3.12-145), which in turn. . . and so on. Shakespeare not only invented the Turkish threat and the storm, he tied them into the plot at several levels in order to define and display the human beings involved and the relations between them, so stable now, so suddenly changed. If we look at his source, we find Cinthio reporting only that'the Venetian lords made a change in the forces that they used-to maintain in Cyprus; and they chose the Moor as Commandant of the soldiers whom they sent there' (Bullough VII: 242). There is no crisis, no Turkish threat, and no storm. Having been persuaded by Disdemona to-let her accompany him, Cinthio's moor 'embarked in the galley with his lady and all his train,... and with a sea of the utmost tranquillity arrived safely in Cyprus' (243).
Already by this point in the play Shakespeare has articulated many of the important issues, matters never touched on in Cinthio's narrative. Where the source gives Disdemona some 'relatives' who try to dissuade her, Shakespeare invented the disapproving father, ready to welcome Othello into his house to entertain his daughter, but not to have 'the sooty bosom / Of such a thing as you' for a son-in-law; Either the awareness that her father will not give his permission, or Othello's impending despatch to the Cyprus wars, makes Desdemona elope rather than try to, marry him openly. But while proving her love for Othello, their tricking Brabantio with a secret mamage gives Iago one of his most vicious insinuations about her:
She that so young could give out such a seeming.
To seal her father's eyes up close as oak,
He thought'twas witchcraft. .. . (i. i.208ft)
Brabantio in fact accused Othello of witchcraft, but Iago cunningly leaves the imputation open, as if referring to Desdemona. Iago can make such insinuations without being refuted because Othello obviously does not know Desdemona well. Whereas the Moor in the source has served Venice for an unspecified period, living a happy and stable marriage with Disdemona for some time before leaving Venice, Shakespeare's Othello has only been in Venice for nine months, having seen uninterrupted military service until then. He is truly a stranger in 'this great world' (1.3.83ff), and only just married. Indeed, the urgency of his, despatch to fight the Turks means that their marriage cannot be consummated on their wedding night (1.3.221-8, 298-300), but has to await their safe arrival on Cyprus (2.2.6-7; 2.3.8-11, 1-5-18). Even then, the consummation is interrupted
by the drunken brawl that Iago has instigated (2.3.246-53). The fact that Othello is unfamiliar with Venetian society, and can have no deep knowledge of Desdemona, these are two crucial changes to the source, making Othello unusually vulnerable to Iago's attack. As so often in the tragedies, Shakespeare takes great pains to place his characters in a context where their personal qualities can only bring them greater sufferings.
In adapting Cinthio Shakespeare is pursuing several goals simultaneously, the most important being to give Iago greater prominence. Having argued in Chapter 1 that Iago's destructive intentionality is the decisive element energising the plot, I find it fascinating to see the ways in which Shakespeare adapted the source to establish Iago's supremacy. Cinthio gave a concentrated account of Iago's hypocrisy which provided a useful character-outline:
The Moor had in his company an Ensign of handsome presence but the most scoundrelly nature in the world. He was in high favour with the Moor, who had no suspicion of his wickedness; for although he had the basest of minds, he. . . cloaked the vileness hidden in his heart with high-sounding and noble words. . . . (243)
That passage gives us the 'honest' Iago, the hypocrite whose true nature is never suspected.- Shakespeare then had to dramatise his powers of dissimulation, make them plausible by showing how Iago transmits a positive image of himself of everyone he meets, in quick succession: Othello, the Venetian Council, Cassio, Montano, Desdemona, Lodovico. The only people who know something of his true nature are Roderigo and Emilia, but neither can disclose it, his wife out of loyalty, Roderigo since he is Iago's accomplice or dupe. One decisive change made by Shakespeare which at first seems to reduce Iago's importance concerns the 'love-triangle'. In the source we read how
The wicked Ensign, taking no account of the faith he had pledged to his wife, and of the friendship, loyalty; and obligations he owed the Moor, fell ardently in love with Disdemona, and bent all his thoughts to see if he cduld manage to enjoy her. ... (243-4)
Afraid that the Moor 'might straightway kill him? if he wooed her openly, the Ensign attempted to communicate his desires to her 'deviously', but she did not respond. The Ensign then thinks that he has failed because Disdemona is 'in love with the Corporal' (Cassio), a good friend and frequent visitor to the Moor's house. His desire frustrated,
the love which he had felt for the Lady now changed to the bitterest hate, and he gave himself up to studying how to bring it about that, once the Corporal were killed, if he himself could not enjoy the Lady, then the Moor should not have her either. (244)
So he decides to accuse her to! the Moor of adultery with the Corporal.
156
critical theories
creator and interpreters
157
The Iago of the source primarily hates Desdemona, because she has not reciprocated his desire; in the second place he hates Cassio, whom he suspects Desdemona of loving; and only as an afterthought does he attempt to harm Othello, the one man for sure who enjoys her favour. Shakespeare eliminates the motif of Iago loving Desdemona, apart from one reference — 'The Moor [will] prove to Desdemona / A most dear husband. Now, I do love her too. . .' (2,2.282ff) — a remark which, in isolation, and considering all his other stated motives, can only seem gratuitous. Given that this Iago no longer loves Desdemona; there is no occasion for hatred, indeed he expresses (in his own person) unforced admiration for her beauty and virtue. As for lago's feelings towards Cassio, Shakespeare makes two crucial changes. First, Iago resents Cassio not because Desdemona loves him (Shakespeare transferred that motive to Roderigo), but because Cassio has been promoted over lago's head-— or so he claims (1.1.8-40). Yet the terms in which Iago vilifies Cassio ('a great arithmetician . . . bookish theoric') make it clear that Cassio is one of that new breed of soldiers in the sixteenth century with a knowledge of military science and technology, while Iago can only claim the old sweat's seniority as grounds for promotion. Further, the trust in Cassio shown by Othello, whose professional judgment is otherwise endorsed by everybody (including Iago), and by the Venetians who appoint Cassio to govern Cyprus (4.1.235ff), proves that lago's criticisms of Cassio's soldiership are not justified — but therefore all the more bitter. Secondly, Shakespeare makes Iago jealous of Cassio as a lover, not with Desdemona but (improbable though it seems) with Emilia ('For I fear Cassio with my night-cap too': 2.1.301),; Both changes give Iago more powerful and (presumably) longer-standing motives, in public as well as private life, to hate and harm Cassio than his putative success with Desdemona.
The biggest change of all concerns lago's attitude towards Othello. In the source the Ensign accuses Disdemona of adultery in order to harm her and .the Corporal, using the Moor to bring about their destruction. Shakespeare reverses this motivation. From the first few moments of the play, as we saw in Chapter 1, Iago expresses his consuming hatred of Othello: 'Despise me if I do not' hate him, he tells Roderigo (1.1.8), making a real litany of his malice: '1 have told thee often, and I re-tell thee again and again, I hate the Moor' (1.3.362ff). To Roderigo, we notice, Iago justifies his hatred purely with motives from his public life, the career-soldier passed over for an upstart. To the audience 'in private', however, he admits another motive, from his private life:
I hate the Moor; And it is thought abroad that 'twixt my sheets He's done my office.
(1.3.384ff)
Again he tells us, 'I do suspect the lusty Moor / Hath leaped into my seat', so that he must have 'revenge', be 'evened with him, wife for wife'
(2.1.289ff). This is a completely new, obviously irrational but therefore all the more powerful motive for hatred. These several changes intertwine public and private domains to achieve a maximum friction. Where the source gave the Ensign a purely sexual motivation for hating the Corporal, Shakespeare has invented the fact of Cassio being promoted at lago's expense, so giving an additional reason for resentment but then redirecting it away from the woman towards the two men. Uninterested in Desdemona, Iago evidently believes that Cassio and Othello have destroyed both his public career and his marriage, by seducing Emilia. (As critics have long observed, this makes Iago the truly jealous man, displaying both misogyny and paranoia.) To achieve revenge on Cassio Iago gets him dismissed, just fails to murder him, and has Bianca arrested as an accomplice, nearly ruining both Cassio's career and his sole personal relationship. By destroying Othello's private life Iago destroys his public one, too, as Othello recognises in that anguished word-play concluding his formal renunciation: 'Othello's occupation's gone' (3.3.359). The desire to härm Othello, an afterthought in Cinthio,.becomes for Shakespeare's Iago the driving force in all his deeds, but always covert: hiding in the dark while slandering him to Brabantio (1.1.68—160), producing those destructive predictions in his asides when Othello greets Desdemona (2.1.166-198), and when he manipulates Cassio's appearance to Othello (4.1.93-172). The hatredis expressed most of all, as we alone know, in the many soliloquies, only equalled in number among Shakespeare's characters by Richard III and Hamlet; but distinguished from them by a single-minded focus on hatred and destruction. Our view of Iago is so complete that it is'easy to forget how little anyone else knows about him.
In Acts I and II Shakespeare created a dramatic structure of his own, only beginning to draw on Cinthio closely in Act III, for lago's insinuation of Desdemona's adultery.80 By then Shakespeare has so transformed'the nature of lago's intentionality that his manipulation of Othello is put to a wholly different purpose, and is in any case carried through with a psychological intensity that Cinthio never approaches. Given this fundamental change of motive it was only to be expected that Iago should now play a larger role. But what is surprising when we read Cinthio for the first time is to see how many times Shakespeare invented new, or adapted extant material in order to link it to lago's controlling design. The mošt striking invention in connection with Iago is i Roderigo. In Cinthio the Ensign performs his slanders to the Moor single-handed, and he; himself attacks the Corporal in the dark,:managing to!cut off his leg but not kill him (Bullough V1I:249). Iago has fár more ambitious plans, involving the destruction of Cassio, Othello, and Desdemona ('out of her own goodness make the net / That shall enmesh them all': 2.3.353f). So Shakespeare invents Roderigo as an accomplice to do the visible dirty work, slandering Othello to Brabantio, picking a quarrel with the drunken Cassio, and undertaking his murder. Since this 'gull'd Gentleman', as the Folio dramatis
158
critical theories
creator and interpreters
159
personae describes him, would hardly agree to such evil deeds unless he j I stood to benefit from them, Shakespeare makes Roderigo hopelessly in love with Desdemona. Roderigo the suitor has for some time been the familiar I gull for Iago, who has pretended to act as go-between, demanding: money \ to advance his affairs. (As Kenneth Muir noted, this relationship echoes -that of Toby Belch in Twelfth Night, exploiting Aguecheek while purportedly furthering his suit to Olivia: Muir 1961, p. 131.) *
Roderigo is more than a cipher, however, for Shakespeare uses him to ■
reveal more facets of Iago's character. The oscillation of hope and despair ,
that the unsuccessful suitor experiences means that Iago must periodically \ cheer him up, encouraging his aspirations by describing both Othello and
Desdemona as attracted by sensual appetite only, a denigration so far from ■
the truth as to show Roderigo's pathetic gullibility, as well as Iago's talents j
for defamation (1.3.341-59; 2.1.215-57; 2.3.362-74; 4.2.174-245). !
Shakespeare creates Roderigo and lets him act to fulfil his desires; but what |
seems to him freedom is, we see, a terrible dependence on a ruthless 1
exploiter. 'Thus do I ever make my fool my purse' (1.3.381) is Iago's first 1
comment on Roderigo, and he punctuates his manipulation with equally | contemptuous metaphors — 'a snipe' (1.3.383), 'this poor trash of Venice, whom I leash / For his quick hunting' (2.1.297—8), 'this young quat' (5.1.11). 'O inhuman dog!' are Roderigo's dying words (5-1.62), yet we
wonder at his self-deception in thinking that Iago would treat him more :
kindly than he. does other people. |
The scenes with Roderigo were designed by Shakespeare to have another important function, that of showing us how Iago manages to convince his dupes that he loves them and acts only on their behalf. Iago does the same thing with Cassio, first in public, that cunning pretence of: loyalty to his
comrade in which he seems to be trying to suppress the truth about his Hi
drunken quarrel (2.3.173-245) — Shakespeare invented the drunkenness JK
— then in private, when he offers Cassio sympathy and advice (2.3.254- B
328). In this sequence, too, Shakespeare adapts his source to show the H|
action being controlled by Iago, as if everything that happens takes place IB
at his bidding. In Cinthio Cassio's dismissal is a brief incident with which K
Iago has no connection: 'Not long after, the Moor deprived the Corporal B
of his rank for having drawn his sword and wounded a soldier while on «J
guard-duty' (Bullough VII: 244). In Shakespeare the plot is:devised, partly JB
controlled, and finally exploited by Iago in a development that, for once, K
he had not forecast in the regular briefings he gives us, All that he told us Mm
was: 'I'll have our Michael Cassio on the hip, /.. . . Make the Moor thank jjB
me, love me, and reward me' (2.1.299ff). Having achieved both goals Iago H!
then deftly improvises a new stage of pretence, consoling Cassio and giving H
him probal' advice: 'I'll tell you what you shall do. Our general's wife is IB
now the general. . . . Confess yourself freely to her; importune her help to J8|
put you in your place again' (2.3.307ff)- This invention, too, shows Mm
Iago both directing events and predicting their outcome. In the sources Bf
Disdemona, grieved by the Corporal's loss of rank, on her own initiative 'tried many times to reconcile the Moor with him' (244), without the Corporal asking her to intercede. By persuading Cassio to petition Desdemona, knowing that she will gladly act in a good cause, Iago sees that he can set up a chain-reaction in which, while
she for him pleads strongly to the Moor, I'll pour this.pestilence into his ear, That she repeals him for her body's lust; And by how much she strives to do him good, She shall undo her credit with the Moor. So will I turn her virtue into pitch...
(2.3.346ff)
In this way Iago establishes a linked intercession, Cassio —» Desdemona —» Othello, in the realm of Cassio's public life, his military career, which he can then present to Othello as displaying the relationship Cassio Desdemona, in the private, erotic sphere. Neither Cassio nor Desdemona ever realise how Iago misrepresents their innocent concerns.
Having carefully constructed this series of plots, all coined by Iago, Shakespeare inserts them into the next stages of Cinthio's narration, which is a straightforward sequence of action and reaction. In the source it is the Moor who reports to the Ensign that 'his wife importuned him so much for the Corporal', and only then does the Ensign, in response, insinuate that she does so out of love for the Corporal. The Moor then openly confronts Disdemoha with his suspicion, and she affirms! her innocence while not managing to remove the Moor's suspicion (Bullough VII: 244-6). Shakespeare takes over several details in Iago's slandering of Desdemona and Cassio but adapts them to a narrative quite different in kind, since all the suspicions now emanate from Iago. By his devious insinuations Iago has created so much distrust that Othello cannot ask his wife direct questions, since he no longer believes that she: would reply truthfully. Iago's inten-tionality in manipulation and destruction is given its climactic expression with the handkerchief. In the source the Corporal, having found Disdemona's handkerchief in his lodgings (where the Ensign had placed it), goes to their house to return it. He knocks on the door, but the Moor has unexpectedly returned, and the Corporal, 'fearing that he might come down and attack him, fled without answering'. The Moor opens'the door, finds nobody there, and angrily asks his wife who it was. She does not know, but the Moor is convinced that it was the Corporal (247). Shakespeare uses this incident, yet brings it forward to the beginning of the insinuation scene, and allows Iago to initiate mistrust that Cassio 'would steal away so guilty-like, / Seeing you coming' (3.3.35ff). That gives Iago the all-important opening move for his duping and destroying of Othello. These scenes (3.3.93-481; 4.1), which I have already discussed in terms of Iago's perversion of the principles governing human intercourse, seem
160
critical theories
creator and interpreters
161
completely convincing within the highly concentrated form of drama, since Shakespeare has invented virtually the whole of Acts I and II in order to show Iago successfully manipulating, one after the other, every one to whom he addressed himself. In Cinthio, given the normal rhythms of prose narrative, the arousing of the Moor's jealousy to the point where he wants to kill both Disdemona and the Corporal takes weeks. In reducing the time-scale of his source Shakespeare vastly increased both its intensity and the plausibility of Iago as the supreme deceiver.
While consistently tying events to Iago, Shakespeare makes another change which apparently reduces Iago's involvement, this time concerning his acquisition of the handkerchief. In the source the Ensign, using his three-year old daughter as a decoy, himself steals the handkerchief from Disdemona, and plants it in the Corporal's lodgings (246-7). In Shakespeare Desdemona drops it, Emilia picks it up and, all the loyal wife — 'I nothing but to please his fantasy' — gives it to her husband,(3.3.292-322). Although Iago's plot prospers here by accident rather than design, as every reader of Machiavelli knows, fortwna smiles on those who display virtu. In any case, the more important result achieved by this change is that Emilia unknowingly becomes the accomplice of Iago in his whole plan to destroy Othello and Desdemona. By humouring her husband's wishes here she provides Othello with the 'ocular proof' he has desired, the clinching detail to justify him executing his wife. Although that is. a terrible service she performed, the fact that Emilia can reveal to Othello after the murder that it was she who found and gave the handkerchief to Iago provides undeniable proof, now, of Desdemona's innocence and Iago's guilt (5.2.216-35). Critics who object to this trivial piece of linen have seldom noticed that Shakespeare's plotting involved Emilia with the handkerchief for the crucial evidence she could supply later. Emilia is the mute but surely uncomfortable witness in three scenes to the sufferings Desdemona endures for having lost this all-precious love token: Othello's cross-questioning her (3.4), the scene where he calls her a whore (4.2)., and the scene following, where she helps to get Desdemona ready for bed. Emilia's crime is still venial, for she cannot know what Iago and Othello intend. Wifely obedience is another virtue that Iago can exploit. In the source the Ensign's wife was not involved in stealing the handkerchief, but otherwise she 'knew everything (for her husband had wished to use her-as an instrument in causing the Lady's death, but she had never been willing to consent)...' (248). This detail makes her a much more serious accomplice to the crime, but she escapes punishment. ■
The tragic catastrophe in Cinthio is sensational, sordid, and anti-climactic. The Moor conceals the Ensign in a closet in his bedroom, and when Disdemona gets out of bed in the night to investigate a noise the Ensign kills her with three blows from a sand-filled sock. 'Then, placing her in the bed, and breaking her skull, he and the Moor made the ceiling
fall', so that Disdemona's death is put down to accident (250-51). The Moor, 'who had loved the Lady more than his life', is inconsolable after her death, and begins to loathe the Ensign, openly expressing his hostility. The Ensign then goes to the Corporal, who now has a wooden leg, and persuades him to return to Venice, where he alleges that it was the Moor who cut off his leg and murdered Disdemona. The Corporal duly denounces the Moor to the Signoria, who bring him back from Cyprus for trial, but despite torture 'he denied everything so firmly' that no admission could be extorted from him. The Moor was released from prison but condemned to exile, 'in which he was finally slain by Disdemona's relatives, as he richly deserved'. The Ensign returned to his country and to his evil ways, made another false accusation, was arrested and tortured on evidence 'so fiercely that his inner organs were ruptured', and he died miserably. 'Thus did God avenge the innocence of Disdemona. And all these events were told' by the Ensign's wife, the sole survivor (251-2), the accomplice kept alive to be their narrator.
That sequence is typical of much of the source material Shakespeare used, in that a series of disasters is spread over a great period of time and only loosely related to the main events of the story. By having the Ensign murder Disdemona Cinthio allows him to fulfil his revenge on her for not reciprocating his love — a motive of which both Disdemona and the Moot remain ignorant to the end, so denying them recognition of the cause that destroys them. The Moor's attempt to deny responsibility for Disdemona's murder deprives him of any ethical dimension, and his being killed by Disdemona's relatives introduces an outside element into the plot, quite unprepared for. The Ensign's death 'in another country', for an unrelated quarrel with an unnamed figure, completes the dislocation in Cinthio of action from consequence. In Orfteflo the fates of all these characters are interrelated, derive organically from the destructive designs of Iago and their virtual realisation, were it not for the intervention of Emilia. Most important for the tragedy, Shakespeare gives Othello the feeling of responsibility for his deluded actions which makes him accuse and destroy himself in a mixture of disgust and remorse. His ethical sense finally gives him tragic stature.
The credit for this coherent and meaningful transformation of the diffuse source lies neither with intertextuahty nor discourse, but with the author. In order to appreciate Shakespeare's decisive intervention in this story we have simply to reverse Foucault's scheme and reclaim all the things he discarded. The author is indeed 'the unifying principle in a particular group of writings . . ., lying at the origins of their significance, as the seat of their coherence'. It is he. 'who implants, into the troublesome language of fiction, its unities, its coherences, its links with reality' (Foucault 1972, pp. 221-2). Far from rejecting them, we shall continue to use the four notions that have dominated the history of ideas, 'signification, originality, unity,
ft
162
critical theories
creation', and we shall continue to concern ourselves with 'the point of creation, the unity of a work,. . . the mark of individual originality and the infinite wealth of hidden meanings' (p. 230). The meanings are indeed multiple, but not hidden: they are open to all who read or see these plays performed.
PART II CRITICAL PRACTICES