<<

. 30
( 82 .)



>>

sive structure of the work, the intention is the focus of our interest in and atten-
tion to the artwork. On the external-episode view, authorial intention is a
dispensable, if not distracting, adjunct to the artwork, which adjunct is best
ignored. But on the neo-Wittgensteinian approach, tracking the intention “ the
purposive structure of the work “ is the very point of appreciation.
Given the conception of authorial intention as external to and independent of
the artwork, the anti-intentionalist claim of its irrelevance to the meaning of the
work is eminently comprehensible. But with developments in the philosophies of
action, mind, and language, the neo-Wittgensteinian picture of authorial intention
seems more attractive.The persuasiveness of anti-intentionalism comes to hinge on
which view of intention in general theorists find more plausible. And to the extent
that early anti-intentionalism was based upon a crude view of intention, its conclu-
ART, INTENTION, CONVERSATION 161
AND


sions are questionable.16 Moreover, the more attractive, neo-Wittgensteinian view of
intention not only makes authorial intention relevant to the interpretation of art-
works but implies that in interpreting an artwork, we are attempting to determine
the author™s intentions.Thus, at this point in the debate, if anti-intentionalism is to
remain persuasive, it must do so not only without presupposing a crude view of
intention but also must accommodate the neo-Wittgensteinian picture of intention.
With these dialectical constraints in mind, it seems that two anti-intentionalist
strategies have become popular recently. The first relies on adducing ontological
reasons based on the nature of artworks to deny the relevance of authorial inten-
tion to interpretation. The second argues for the irrelevance of intention by
exploring the aesthetic interests that audiences have in art.That is, the first sort of
argument “ the ontological argument “ advances anti-intentionalism on the
grounds of the nature of the artwork, while the second sort of argument “ the
aesthetic argument “ is grounded on what might be thought of as policy consid-
erations about the best way to regard artworks for aesthetic purposes. Both kinds
of arguments presuppose that artworks, for one reason or another, are to be or
should be interpreted differently from ordinary words and actions.

III

As noted earlier, we ordinarily interpret words and deeds with the cognitive goal
of ascertaining the intentions of authors and agents.As the investigations of histo-
rians reveals, there seems to be no principled difficulty in such practices even
when the agents in question are long dead and the record fragmentary.Thus, the
question arises, Why should matters stand differently when it comes to art?
Should not artworks be interpreted in the way in which we customarily interpret
other words and actions? At this point, the anti-intentionalist may attempt to
argue that artworks are ontologically different from ordinary words and deeds, and
therefore different interpretive practices are appropriate to them; specifically, given
the nature of artworks in general and literature in particular, authorial intent is
irrelevant to interpretation.
This conviction of ontological difference can be found in different and indeed
widely disparate literary theorists. It is, for example, an article of faith of contempo-
rary literary critics who endorse Roland Barthes™s notion of “the death of the
author.”17 And it is, at the same time, a view that underpins the more traditional
approaches of the New Criticism, as that approach was defended by the late Monroe
Beardsley.18 Perhaps this convergence of theorists of different stripes on anti-inten-
tionalism should be less surprising than it seems, for both Barthes and Beardsley
arrived at their positions “ albeit in different decades and in different countries “
while in the process of reacting to what was earlier called biographical criticism.
Though Roland Barthes does not explicitly speak of the issue of intention, he
clearly believes that, with a literary text, the reader™s activity should not be con-
strained by the “myth” that the author is confiding in us. One reason advanced in
support of this view is that
162 INTERPRETATION INTENTION
AND


writing is the destruction of every voice, of every point of origin.Writing
is that neutral, composite, oblique space where our subject slips away, the
negative where all identity is lost, starting with the very identity of the
body of writing.
No doubt it has always been that way. As soon as a fact is narrated no
longer with a view to acting directly on reality but intransitively, that is to
say, finally outside of any function other than that of the very practice of
the symbol itself, this disconnection occurs, the voice loses its origin, the
author enters into his own death, writing begins.19

What Barthes seems to be getting at here is that once writing is divorced from
ordinary usage “ that is, when language does not serve the purpose of acting on
reality “ the relevance of an author™s intention in writing drops out, and the word
sequence is attended to in terms of its play of potential meaning (“the very prac-
tice of the symbol itself ”).This is a feature of poetry explicitly recognized in mod-
ernist writing following Mallarm©, but it implicitly has been a feature of literature
all along (“No doubt it has always been that way.”)20
Ordinary language is tied to acting on reality, and that is the grounds for our pre-
occupation with authorial intent. But when language is detached from that purpose
“ when language is aesthetized? “ the cognitive goal of fixing authorial intent
becomes feckless.That literary language is not practical severs its conceptual connec-
tion to authorial intention.As soon as language is employed (“narrated¦”) in what
theorists of a more traditional bent than Barthes would call an aesthetic way, the con-
ceptual pressure to make sense of it in the light of authorial intent dissolves, and the
reader can explore it for all its potential meanings and associations.
In his “Intentions and Interpretations: A Fallacy Revived,” Monroe Beardsley,
deploying the machinery of speech-act theory, independently evolves an argu-
ment that, though different from Barthes™s, also parallels it in pertinent respects.
The argument begins by drawing a distinction between performing an illocution-
ary action and representing one. When a pickpocket takes my wallet and I say,
“You stole my wallet,” I perform the illocutionary act of accusation. An illocu-
tionary action is generated (according to Beardsley, following Alvin Goldman) by
the production of a text under certain conditions, and according to certain lan-
guage conventions.21 In contrast, when a stage actor, playing a character, says,“You
stole my wallet,” to another actor, playing another character, she is not performing
an illocutionary action; she is representing one.
The relation between performing illocutionary actions and representing them
is to be understood on the model of pictorial representation. Just as Beardsley
argues that the relation of a pictorial depiction to its referent is that of selective
similarity, he maintains that the representation of an illocutionary action resembles
the performance of illocutionary action in certain, selected respects (i.e., repro-
duces certain, but not all, of the conditions requisite for the performance of the
illocutionary action). For example, when I accuse a culprit of filching my wallet, I
believe that he has taken my wallet; an actor, though repeating much of the for-
ART, INTENTION, CONVERSATION 163
AND


mula for accusation, does not believe her fellow actor has stolen anything.Thus, a
representation of accusation resembles it in many respects, but not in every respect
“ for instance, it fails to fulfill the condition of conviction in the culprit™s guilt.
Most ordinary discourse is preoccupied with the performance of a multitude
of illocutionary actions. Literature, in contrast, specializes in the representation of
illocutionary actions. In this respect, once the author™s intent to represent illocu-
tionary actions is recognized, thereby acknowledging the neo-Wittgensteinian
claim of a conceptual relation between an act and its animating intention, the rep-
resentation of the illocutionary action is regarded as a selective imitation of the
performance of a fictional character “ either the literal characters in the text or
what has sometimes been called an implied narrator or an implied speaker or
dramatis persona.
So when Wordsworth writes about England that “she is a fen,” this is not
Wordsworth directly performing an illocutionary act of accusation. Wordsworth,
in writing poetry, signals his intent to represent the illocutionary act of accusation,
which, in this case, is the imitation of an implied speaker™s disparaging of England.
The language in the poem is not a performance of an illocutionary act of accusa-
tion by Wordsworth. It is a representation of such an action by an implied speaker.
Thus, the meaning of the language token is not tied to Wordsworth™s intention, nor
need it be understood in the context of Wordsworth™s biography. It is a representation
that can be comprehended solely in terms of the conventions of language.
The author of the performance in the text, so to speak, is the implied speaker;
since all we know of the implied speaker are the words in the text “ since the
implied speaker, a fictional entity, has no existence outside the text “ there can be
no question of his extratextual intentions.There is no extratextual author, so there
are no governing, extratextual intentions. Just as the issue of the number of chil-
dren Lady Macbeth has is underdetermined by the fiction, so there is no access to
implied authorial intent beyond the page.
Beardsley agrees that in ordinary language the cognitive goal of interpretation
is the discernment of the speaker™s intentions. But the language in literature is not
a matter of the author™s performance of an illocutionary act. It is a representation
of the illocutionary acts of characters and implied speakers. And such fictional
speakers have no intentions beyond the words on the page, which must, in conse-
quence, be understood solely in terms of the conventions of language (and with-
out recourse to the intentions of actual authors). It is as if in creating fictional
characters, through illocutionary-act representation, actual authors™ intentions are
ontologically detached from the language sequence in favor of the meanings of
characters, both literal and implied, which in turn can, for metaphysical reasons,22
only be a matter of grasping of linguistic conventions (the literal sense of the
words, and the conventions or established strategies for comprehending the sense
of verbal contexts and metaphors).
The language in a literary text in being represented language “ perhaps, this is
what Barthes intends by “narrated ¦ intransitively” “ becomes the linguistic
“performance” of the characters “ implied and literal “ and thereby is discon-
164 INTERPRETATION INTENTION
AND


nected from the intentions of actual authors by means of a fictional frame
(Barthes™s notion that language is detached from acting on reality). Moreover, the
“intentions” of characters have no existence beyond the page and are available
solely in terms of linguistic conventions. Stated formally, Beardsley™s argument
seems to be as follows:
1. If x is a literary work, then x is only a representation of an illocutionary act.
2. Though actual authorial intentions are relevant to whether x is a representa-
tion of an illocutionary act, what x is a representation of (its meaning) is
solely a matter of the relevant linguistic conventions (the literal sense of
words and the conventions or established strategies for grasping the sense of
a verbal context and metaphors) and not a matter of fixing authorial intent.
3. Therefore, if x is a literary work, then what x is a representation of is solely
a matter of the relevant conventions.
Thus, in interpreting the language in a literary text, we will be concerned with
the meanings of characters “ literal ones, implied authors, or dramatis personae.And
since these characters have no existence outside the words in the text, interpreting
their meanings is exclusively a matter of convention.The actual author, metaphori-
cally speaking, banishes himself from the text in the process of representing illocu-
tionary actions. This argument grants some role to authorial intention as an
ingredient in identifying the author™s act as one of representing. But once the repre-
sentational frame is in place, so to speak, the author™s intentions are outside it. And
given the ontological status of the representational frame, it is a category mistake to
be preoccupied with authorial intent; it is metaphysically irrelevant.
(Moreover, though this argument is stated in terms of literature, one supposes
that it can be extended to other art forms, given, for example, Beardsley™s analogies
between pictorial representation and illocutionary representation “ perhaps land-
scapes are to be understood as vistas seen by implied observers.)
It is absolutely central in this argument that literary language and ordinary lan-
guage be ontologically distinct. Literary language is a special zone, so, even if in
ordinary language authorial intent is a guide to meaning, it is not relevant in liter-
ature because literature is not a performance but a representation. In ordinary lan-
guage, we are prone to say that when a speaker disambiguates her earlier
utterance, she has told us the meaning of the utterance.With literature, however,
there is no comparable resort to the author™s intent, for the relevant speaker is not
the living author but various dramatis personae who are ontologically unavailable
for comment. If their words are ambiguous, one suspects that Beardsley would be
prone to say that the dramatic speaker is being represented as ambiguous.
The crux of Beardsley™s argument is, given the distinction between perform-
ing and representing, the claim that literature is by definition a matter of repre-
senting illocutionary acts.23 This effectively boils down to the assertion that all
literature is essentially fictional. For even if a literary text does not deploy imagi-
nary characters and places, it is involved in presenting its persons, places, and
events through the fictional medium of an implied speaker or narrator. Such
ART, INTENTION, CONVERSATION 165
AND


claims are not unfamiliar.24 If anti-intentionalism depends on this generalization,
however, it is surely in trouble.25
Pretheoretically, many works of what we classify as literature fall into the
category of nonfiction. Lucretius™s Concerning the Nature of Things is one exam-
ple; The Mahabharata is another. Both appear to be illocutionary acts of asser-
tion, even if what they assert turns out to be false. It does not seem correct to
attribute to Lucretius the intention of representing the illocutionary acts of an
Epicurean philosopher “ he was an Epicurean philosopher philosophizing.
Similarly, the authors of The Mahabharata were not imitating the telling of the
history of their race; they were telling it. Nor do we need, I think, to travel to
the distant past for our counterexamples. When in “Howl,” Allen Ginsberg
wrote “I saw the best minds of my generation destroyed by madness,” there is
every indication that, however hyperbolically, he is speaking in his own voice
and not representing the illocutionary act of accusation of some “angel-headed
hipster.”The notion of implied narrators and dramatic speakers, no matter how
useful in explicating a great deal of literature, does not afford a necessary con-
dition for being a literary text.26
Thus, a literary text is not necessarily a representation of an illocutionary act;
it may be a performance of an illocutionary act of assertion, accusation, and so
forth.Therefore, the fact that many literary texts involve representations of illocu-
tionary acts does not entail that every literary text must be interpreted without
concern for authorial intent in contradistinction to ordinary language.
Of course, it would be a mistake to conflate the representations of illocution-
ary acts presented through fictional characters with the performance of illocu-
tionary acts by actual authors. It would be an error to identify Emily Bront« with
the narrator of Wuthering Heights. But that distinction can be readily marked with-
out resorting to the extreme theoretical concession that the literary speaker is
always fictional.
Not only are there entire literary works that it seems ill advised to regard as
representations of illocutionary acts. There are also many parts of literary works
that do not appear to be representations of illocutionary acts: the discourse on
whales in Melville™s Moby Dick, the history of symbols in Hugo™s Hunchback of
Notre Dame, and the philosophy of history in Tolstoy™s War and Peace. Though
housed in fiction, where they undeniably perform a literary function, they are also
essays whose authors produced them in order to make assertions. In interpreting
these interludes, one needs to approach them as one would any other form of
cognitive discourse. Some may be tempted to prefer to read them as representa-
tions of illocutionary acts when one finds a particular author™s ideas rather hare-
brained. But such considerations “ however cosmetically well intended “ are, in
fact, irrelevant to the issue of whether the passages in question are performances
of illocutionary acts rather than representations thereof. Furthermore, if, as I
argue, these are performances of illocutionary acts of assertion, then in such
instances, it will be appropriate, as Beardsley would appear compelled to admit, to
interpret them with the cognitive goal of discerning what the authors intended.
166 INTERPRETATION INTENTION
AND


So far, we have been whittling away at the first premise of Beardsley™s argu-
ments by finding poems and passages to which the generalization does not apply
and by arguing that in these instances, given Beardsley™s own views, interpreting
with respect to authorial intention is as appropriate as it is in the case of ordinary
illocutionary acts of assertion. But the million-dollar question is: How extensive a
problem does this pose for the anti-intentionalist?
My own hunch is that the problem will be very extensive. For once we admit
that there can be explicit nonfictional passages (which may range in scale from
clauses and sentences to chapters and beyond) housed in fiction “ and which are
best construed as performances of illocutionary actions “ the door is opened to
the recognition that there are many implicit or implied propositions in literary
works as well, which are also best conceived in terms of performances. Brave New
World expresses a point of view about what Huxley sees as the prospect of utili-
atarian social control. I see no particular advantage in rephrasing this observation
in terms of the point of view of a fictional dramatic speaker.And, of course, if it is

<<

. 30
( 82 .)



>>