The Leeds Review - Issue #1How should we interpret a poem? To what extent should we concern ourselves with its author as a person - with their psychology, beliefs, attitudes, experiences... intentions? Do these shape the meaning of a poem? Can a poem be sincere? These are the questions I want to explore here from a philosophical perspective. Along the way it will be necessary to shed light on the special character of poetry as a work, and the role of the poet in its genesis.
Literary work is commonly seen as "the vehicle of a specific act of communication". (1) As a general description this seems harmless enough. It says nothing, however, to distinguish literary work from other forms of speech and writing - the emails, for instance, sent out by the University to students - which facilitate a "specific act of communication". Yet, it might be argued, this is a generic description, correct as far as it goes - just as "team game" applies to but does not distinguish football and basketball - so why quibble with it?
I believe, on the contrary, this description is far from innocuous. Through the terms "vehicle" and "act of communication" it introduces a conceptual framework or model which distorts our whole view of poetry. Two such distortions stand out. Firstly, as vehicle the poem becomes a carrier or transmitter of meaning from sender to recipient - a meaning determined at the outset, i.e. before composition. The poem is thus assimilated, as it were, to an item of information technology transmitting and encoding a pre-determined message. And the way it does this hardly matters - meaning becomes separable from vehicle, content from form. But this picture cannot be right. Poetry of value rarely, if ever, transmits a determinate, unambiguous meaning. Imagery, allusion, and ambiguity abound, and are central to its effect. As Hirsch points out, an author "almost always means more than he is aware of meaning". (2) Of course, since a poem consists of more-or-less recognisable words in some syntactic arrangement we can always attempt a paraphrase, but this is not to disambiguate and fix its meaning once and for all. How, say, might the imagery in Larkin's poem Solar - "Heat is the echo of your / Gold", (3) or Paul Celan's "In rivers north of the future...", (4) be determinately paraphrased?
Moreover, a poem's 'meaning' overflows its semantic content. The way poetry speaks - the manner of its unfolding, its tone, its poise and rhythm, its phonetic qualities such as alliteration and rhyme, and the resonance of its imagery - that is, its whole poetic and aesthetic character - this also constitutes its meaning, or what Heaney calls its "vindicating force". (5) But now this meaning is akin to that of song and music. Speaking of Eliot's poem The Waste Land, Leavis says that: "the organization it achieves as a work of art... may, by analogy, be called musical". (6) All of which should convince us that 'vehicle' is an inappropriate term for poetry.
The second major distortion arises from seeing the poem as an act or action of an agent. This encourages us to interpret it on the basis of the agent's - the author's - intentions. Richard Wollheim follows this approach, drawing support from "the philosophy of action". (7) Simply to observe someone's behaviour, he notes, will not yield adequate understanding of it - we first need to know that it is intentional (and also what the intention is). Correspondingly, adequate understanding of an artwork demands "retrieval" of the "creative process" including the author's designs and intentions. (8) Wollheim is right about actions - we describe and make sense of them in the light of the goals and purposes they serve within an ethical-social framework. For example, I catch a train in order to start work at... in order to further my career aspirations... in order to earn a livelihood, etc. Here, the "in order to" identifies my intentions. Actions of this kind informed by the agent's intentions and goals can be distinguished using the Greek word praxis.
Praxis also includes making things. For example, in the course of a working day I may make sandwiches, write emails, and - let us suppose - execute a sculpture for a commission. A helpful explanation of why the figure I am carving has a defiant stance would refer not to my own goals and intentions ("a contract to fulfil... children to feed"), but to the required character of the work - i.e. what this figure has to be ("a memorial to Arthur Scargill"). The only intentions relevant here are those describable in terms of the work's projected character, and which come to fruition (or not) in it. Such intentions are, of course, likely to be pitched at a high-level, presupposing a practitioner's intimate knowledge and skill, and the 'how' or style of execution. In summary, an account of my behaviour must refer to my intentions and goals, but an account of my work must say what that work is - its function, form and content. Wollheim's parallel between art and action is misguided.
It is also at odds with the facts of poetic composition. Seamus Heaney, recalling the ancient idea - voiced by Plato in The Ion (9) - that "when a lyric poet gives voice, 'it is a god that speaks'", (10) suggests that poetic form "is achieved... by the self-validating operations of what we call inspiration". (11) The poem has a life and character of its own distinct from that of its author. For Paul Celan it is an "old" view worth reiterating - "that once the poem is really there, the poet is dismissed, is no longer privy". (12) Likewise, the Orkney poet George Mackay Brown attests:
Strange the relationship of poet and poem. The poet is not really the master at all. The poem is the real shaper; it will not stand being coerced or driven. In the end its says, coldly, Destroy me , don't let me go through to the market place misshapen and hirpling. And the poet who has spent so much blood and tears pleads, Truly you are beautiful, you will be beautiful soon. If he's wise he does what the poem wants: tears it up and gives it to the wind, burns it. A poem exists in its purity, as the perfect statue is hidden in the rock. (13)
Without doubt poems have human authors. But to look to authorial intentions in evaluating a work is to fall prey to what, following Wimsatt and Beardsley, is called the Intentional Fallacy. The fact that a work originates from a "designing intellect", they argue,
is not to grant the design or intention as a standard by which to judge the worth of the poet's performance. (14)
If the poet's intentions are fulfilled the poem itself shows this; if unfulfilled, they are irrelevant. Intentions can only be of incidental interest in interpretation. But isn't this counter-intuitive, you may ask? A poem is not an artefact like a shopping trolley whose function, not its meaning, is paramount. Surely a poem, like anything written or spoken, conveys its author's intended meaning? The author cannot be dismissed! The supposition implied here must, I suggest, be handled the utmost caution. Firstly, we do not assign meanings by fiat to the words we use, in the manner of Lewis Carroll's Humpty Dumpty: "When I use a word it means just what I want it to mean". (15) Humpty Dumpty's principle would ensure linguistic anarchy - just to understand Humpty Dumpty's declaration we must first suspend the principle! Secondly, and more insidious is the equivocation in "meaning". The question 'What did you mean by that email?' could be taken as enquiring about its purpose and intent ("I meant to annoy the boss"), or as asking for a précis of it ("an account of our failings"). The former concerns a human action, the latter concerns a text and its semantic content. In our daily lives these two senses often overlap - we produce certain texts in order to achieve certain actions. Language here serves an "act of communication" relying upon a shared culture, and a currency of words, utterances, sentence-structures and styles we draw upon according to context and audience - and which we can appropriate according to our purposes.
Thirdly, we must distinguish poetic utterance from the commonplace or prosaic. In poetry, as already seen, language takes on a special character - in some respects like that of song and music. Through its qualities of enunciation, its rhythm, evocation and allusion, and through its imagery, poetry acquires the energy and "vindicating force" of which Heaney speaks. It loses the transparency and familiarity of everyday language; it ceases to reflect "ready-made meaning"; (16) it establishes its own meaning. It does not just re-cycle our linguistic currency, but adds to it. As Robert Frost says of the true poem:
Read it a hundred times ...It can never lose its sense of a meaning that once unfolded by surprise as it went ...It begins in delight... it runs a course of lucky events and ends in a clarification of life... (17)
So to characterise poetry as an expression of intended meaning is simply to revive the discredited view of poetry as a 'vehicle' for pre-determined meanings. The only "expression" appropriate here is that possessing "legislative and executive powers". (18) All of which puts into question the author as agency of the work's meaning. The author's role - I would now suggest - is one of orchestrating and enabling the work by putting their knowledge and feeling and craft at its service.
Roland Barthes, in his essay The Death of the Author, goes further in dismissing all links between author and work. The fiction of the author as a controlling agency, he claims, is a "culmination of capitalist ideology" (19) - presumably because it establishes the author as a producer with property rights over a tradable commodity - literature. But writing itself resists such a proprietorial relationship.
As soon as a fact is narrated no longer with a view to acting directly on reality but intransitively... disconnection occurs, the voice loses its origin, the author enters into his own death. (20)
And he appeals to "linguistics": -
Linguistically, the author is never more than the instance writing, just as I is nothing other than the instance saying I: language knows a 'subject', not a 'person', and this subject, empty outside of the very enunciation which defines it, suffices to make language 'hold together'... (21)
The written text should be seen not as
a line of words releasing a single 'theological' meaning (the 'message' of the Author-God) but a multi-dimensional space in which a variety of writings, none of them original, blend and clash. (22)
Barthes is right to expose the emptiness of author-centred literary culture and criticism, with its attendant focus on celebrity, which encourages us to seek an explanation of the work "in the man or woman who produced it". As if Van Gogh's "madness" could explain Van Gogh's art. (23) Or - we might add - as if Coleridge's opium could explain his poetry, or Sylvia Plath's suicide, her poetry. (24)
Yet Barthes, I sense, protests too much. And his thesis leaves a mystery unsolved. If all writing slips free from authorial control how does Barthes present his arguments? Digressions into French cuisine or Euclidean geometry are somehow avoided. Does this not indicate some editorial and thematic control over the content of Barthes's essay - an endeavour to release "a single 'theological' meaning"? Barthes would surely say that his thesis applies not to discursive or philosophic writing such as his own, but to literature, i.e. what is written "no longer with a view to acting directly on reality but intransitively". This locates literature outside the scope of practical communications - consistent with Auden's "Poetry makes nothing happen" [Auden: 'In Memory of W.B.Yeats']. Yet it is neither a necessary nor sufficient criterion for literature. I can imagine something written "intransitively" but lacking literary or poetic character (perfunctory diary notes?). Conversely, wartime speeches and passages from the Bible can convey practical and moral imperatives in poetic form. Again, Ted Hughes's poem Crow has reportedly provoked laughter. (25) Either we deny these are "acting directly on reality" or we deny them a foothold in literature - neither of which seems an attractive option.
Disregarding the difficulties of definition, it is unclear how the literary text retains its thematic direction and character without the organising intelligence of an author. Or are we to attribute "the voice" of a text to language itself? This claim was indeed made by Peter Ackroyd regarding his work Albion - that "the language" took over from authorial contrivance and intention.
As I wrote, after a while the sentences formed themselves. If I was superstitious I would believe it was the language doing it for me. (26)
But to leave it all to "language" may simply result in a rambling work replete with undigested borrowings and deficient in critical awareness - or in what Leavis calls "thought" and "the discipline of intelligence" (27) - a criticism perhaps relevant to Ackroyd's Albion.
Barthes's critique of authorship rides a wider philosophical assault on the notion of a substantial, knowable self - epitomised as the res cogitans or 'thinking substance' of Descartes. The grounds for such a concept had already been questioned by Kant:
Consciousness is, indeed, that which alone makes all representations to be thoughts, and in it ... all our perceptions must be found; but beyond this logical meaning of the 'I', we have no knowledge of the subject in itself. (28)
Sartre argues that the substantial self believed to underlie our thoughts and actions is a construct, a "false hypostatisation", (29) which can serve no purpose in unifying the activity of consciousness. We mistake it for a cause when in reality it is an effect of consciousness and action. Within the translucency of consciousness there can be no room for this thingly "inhabitant" - invasive and opaque it would only congeal consciousness. (30) For Sartre, the subject or pour-soi is "always in abeyance, because its being is a perpetual deferring". (31) Sartre's later view - not dissimilar to structuralism - is that we are "born into a language that precedes us, alienates us, and determines us in ways of which we are often unaware". (32) Lacan's more extreme position - "Man... is a pawn in the play of the signifier" (33) - risks self-contradiction in the way already noted with Barthes. Derrida would retain a role for human subject and agent, but...
the subject... no longer has the form of self-mastery, of self-adequation, centre and origin of the world... [but is] the finite experience of non-self-identity, of the inderivable interpellation that comes from ...the trace of the other. (34)
Even if we reject the more virulent of these attacks on the classical, humanist conceptions of self and agency we are still faced with this conclusion: The author - as presence in the work, agent of its meaning, and a resource for its interpretation - must be exorcised.
» Read the second part of Poem and Poet