<<

. 24
( 82 .)



>>

of the form: x at t3 because of y at t2 because of z at t1.
“The battle was lost for want of a horse and the horse was wanting for lack of
a horseshoe” is not a narrative, though “King Philip could find no shoe for his
horse and could not ride into battle and as a result the battle was lost” is a narra-
tive. The forward-looking aspect of narrative here is what might be called a dis-
cursive, rather than a logical requirement, of this particular story form.11
Summarizing my arguments, then, a narrative connection obtains when (1)
the discourse represents at least two events and/or states of affairs (2) in a globally
forward-looking manner (3) concerning the career of at least one unified subject
(4) where the temporal relations between the events and/or states of affairs are
perspicuously ordered, and (5) where the earlier events in the sequence are at least
causally necessary conditions for the causation of later events and/or states of
affairs (or are contributions thereto). I say that the earlier events must be at least
causally necessary conditions in order to allow that they may also be sufficient or
jointly sufficient conditions for the later events. Basically, what I am saying is
required is that the earlier events fall into the causal network that gives rise to the
later events where the weakest, but perhaps the most frequent, way of figuring in
that causal network is as a causally necessary condition (or a contribution thereto)
for the causation of later events.That is, the earlier event in the narrative connec-
tion must be causally relevant to the effect event.
Moreover, I have added parenthetically that the earlier events and states of
affairs may merely be contributions to a causally necessary condition. This last
qualification is meant to accommodate the fact that many of the descriptions and
depictions of states of affairs in narratives may not themselves present causally nec-
essary conditions for what is to come, but only elements of a causally necessary
condition. For example, we may be told that a character was born in a humble
ON THE NARRATIVE CONNECTION 127

neighborhood in Arkansas. Later we are told that he became President of the
United States. Here the humble birthplace of the president is not itself a causally
necessary condition to his presidency, but only a contribution to setting forth such
a condition, since it is a requirement for the presidency that the candidate be an
American citizen, which condition is satisfied by being born in Arkansas, however
humbly. A great many of the details we encounter in narratives are of this sort;
they do not constitute causally necessary conditions for later events, but are con-
tributions to the characterization of such conditions.
In passing, I mentioned that a typical effect of a narrative is retrospective sig-
nificance. That is, later events in a narrative disclose the significance of earlier
events. For example, in the historical narrative “The Allies and the Central Powers
had fought themselves to a standstill, but then the Americans entered the war and,
as a result, Germany was defeated,” the significance of America™s entry into the
war is made clear in terms of its contribution to the defeat of Germany. In this
narrative, the significance of American™s entry into World War I is foregrounded as
a turning point in that conflict. The end of the story, so to speak, makes prior
mention of America™s entry into the war pertinent. It discloses the significance of
that event; it makes the relevance of the inclusion of that event in the story clear.
Likewise, a later event in the fabula of Oedipus Rex, the plague, makes an earlier
event in the story, the murder of the stranger at the crossroads, retrospectively sig-
nificant. Moreover, as this example perhaps suggests, it is the fact that narratives
track causal networks that most frequently enables us to identify their subjects as
unified, since we usually colligate or collect elements of a narrative under the
overarching umbrella of some causal network. In narrative, causal relations are
standardly the cement that unifies the subject of the story.
The account that I have offered of the narrative connection should help
explain the phenomenon of retrospective significance. The earlier events in the
narrative connection are at least causally necessary conditions of later events in the
story. Thus, when later episodes are added to the story, they reveal the relevance
and importance of earlier events in terms of the causally necessary roles they play
“ something that may not be evident when these events and/or states of affairs are
first mentioned. Consequently, the characterization of the narrative connection I
have offered can explain the phenomenon of retrospective significance that is typ-
ically observed to attend reading, viewing, or listening to narratives. On the other
hand, retrospective significance, though a typically recurring and explicable fea-
ture of narrative, should not be mistaken as the mark of narrative.12 For the tem-
porally ordered discourse “Aristarchus hypothesized the heliocentric system and
then centuries later Copernicus discovered it again” affords the apprehension of
retrospective significance “ it indicates the point of mentioning Aristarchus™ dis-
covery in light of Copernicus™ “ but it is not, as I have argued, a narrative proper
inasmuch as it lacks a narrative connection.
Earlier I justified the taxonomy involving annals, chronicles, and narrative
connections proper, rather than accepting any time-ordered series of events as
constituting a narrative, on the grounds that it provides us with more theoretical
128 ART, HISTORY, NARRATIVE
AND


precision. Now that I have developed my account of the narrative connection, I
can perhaps offer a more powerful consideration in behalf of this taxonomy. It is
this: when confronted by specimens of annals, chronicles, and narrative connec-
tions proper, and asked which most accords with what one expects of a full-
blooded narrative, I predict that most respondents will choose the narrative
connections proper. Narrative connections proper correspond best to our intu-
itions about what a narrative is. Moreover, this is because our intuitions are rooted
in a vague sense that narratives have something centrally to do with changes of
state and, thus, with causation. Other theorists may suggest that we deal with the
preference for narrative connections proper by saying that they possess greater
degrees of narrativity. But reverting to my earlier argument, I think that it is more
enabling theoretically to talk about different structural variations, notably the
structural variations between annals, chronicles, and narrative connections proper.
Perhaps a related consideration in favor of my view of narrative is that narrative is
a common form of explanation. In ordinary speech, we use narratives to explain how
things happened and why certain standing conditions were important. Narrative is
capable of performing this role because it tracks causal networks. The rationale for
citing earlier events in the course of an explanatory narrative is that they play some
role in the etiology of the events we wish to explain.To perform that role they must
minimally belong to the causal network, a requirement that can be satisfied by their
being a causally necessary ingredient (or a contribution thereto).Thus, insofar as what
we call narratives are explanatory, it seems advisable to regard narrative properly so
called as connected to causation and not merely temporal succession.

II. SOME OBJECTIONS

This account of the narrative connection is likely to draw forth a number of crit-
icisms.Though I cannot anticipate them all, let me mention a few. In literary stud-
ies, it is often customary to talk about narratives not only in terms of temporal and
causal relations, but also spatial relations. However, I have not included spatial rela-
tions in my characterization of the narrative connection. Is this a problem?
Most narratives do involve spatial relations. However, since I am attempting to
get at the essential features of our concept of narration, I do not invoke spatial
relations, because it seems to me that there can be narratives bereft of spatial rela-
tions, though, on the other hand, narratives bereft of temporal and causal relations
seem unfathomable to me. Consider this example: “This morning I was upset
because I thought that I had forgotten how to add, but then I remembered that 2
+ 2 = 4 and now I am so very happy.” This seems to be a narrative to me, or,
rather, to possess a narrative connection, but it doesn™t involve spatial relations,
since, of course,“I” might be a disembodied spirit. Nor is this simply an invented
example. In various religious narratives we are frequently told of spirits and spiri-
tual forces who undergo changes of state.
Another objection to my view involves a counterexample. Suppose we hear a
story of the sort we are all likely to categorize as a full-blooded narrative “ at least
ON THE NARRATIVE CONNECTION 129

up to the last line of the story. But, then, the last line of the story says that every-
thing has transpired in a logically possible world where there are no causes, just
coincidences. In the world of the story, there are no causally necessary conditions.
So, according to my view, there are no narrative connections in the story, and,
hence, it is not a narrative, properly so called.13
This is an ingenious example. However, my temptation is to bite the bullet and to
say that it is not a narrative. It looked like a narrative up until the very end, but then
it turned out not to be. Is this simply arbitrary or questioning begging? I think not.
If one is attracted to this counterexample, it may be because one thinks that
the tale being told is some kind of story, and that my claiming that it is not is sim-
ply dogmatic. However, I too can agree that it is a story “ it is a chronicle on my
accounting. It is not a narrative properly so called, but it still a representative of a
story form. We thought at first that it was a narrative, but by its conclusion we
realized it is a chronicle.Thus, denying it the status of narrative is not as drastic or
counterintuitive as it may sound at first, since I am not claiming that the example
fails to be any kind of story; it just fails to be a full-fledged narrative. Indeed, slot-
ting it under the category of the chronicle may even be attractive, since it says pre-
cisely what story form it belongs to.
A related counterexample might be to imagine a race of alien peoples who
have no concept of causation. They string together event-ordered series unteth-
ered by causal connections. On my account, these people would not have narra-
tives even though they might spend most of their time recounting past events to
each other. But once again, this sort of counterexample does not seem problem-
atic to me. On the one hand, it seems to me clear that these people would not
have our concept of narrative, and that is the concept I am trying to unravel.And,
on the other hand, if we knew more about the kind of stories they tell, we might
be able to identify them in terms of story forms other than narrative.
Literary and cinema scholars are apt to be distrustful of my characterizations
because they are often concerned with modernist or avant-garde narratives.These
are often antinarratives, like Kathy Acker™s Pussy King of the Pirates, that are designed
to disrupt or subvert ordinary narrative connections.These sorts of practices do not
accord with my characterizations, since they are expressly intended to violate the
standard conception of narrative. But, in that sense, these practices themselves must
presuppose something like the ordinary concept of narrative in order to negate it.
Such practices are parasitic on the ordinary concept of narrative. But so much con-
cedes that there must be an ordinary concept that functions as their, so to speak, foil.
I would argue that that is what my conception of the narrative connection models.
Thus, practitioners and critics of the avant-garde antinarrative, oddly enough, need
something like my concept of the narrative connection as a condition of possibility
for their own practices and, in that way, they will confirm my theory, or something
very much like it, if only backhandedly.
The picaresque is another genre that literary critics advance as a counterexam-
ple to theories like mine. However, this is to enter the debate over the identifica-
tion of what I earlier called large-scale discourses. Picaresques, like Tom Jones, may
130 ART, HISTORY, NARRATIVE
AND


recount the diverse adventures of a character over a long period of time and many
of these events may not cohere in a connected causal network. That is certainly
true. However, none of the best known picaresques are totally bereft of narrative
connections. Actions and events within given episodes standardly evince what I
have called narrative connections, not to mention the fact that there are often,
even if not always, narrative connections between episodes.Thus, at least the best
known picaresques will not be discounted as narratives proper on my characteri-
zation of the narrative connection. They will have some narrative connections,
even if they contain fewer narrative connections than other genres like the classi-
cal detective mystery or the thriller.

I I I . N A R R AT I V E C O M P R E H E N S I O N

Having outlined a theory of the narrative connection proper and having consid-
ered some objections to it, the question arises about whether this theory can con-
tribute anything to our understanding of narrative comprehension. I think that it
can. Comprehending a narrative involves following it. But following a narrative
involves anticipation. It involves having a sense of where the narrative is headed.
But what is the nature of this anticipation? In what sense does the reader, viewer,
or listener intuit the direction in which the narrative is headed?
Where ideas about outcomes are at issue, it is natural to think in terms of pre-
diction. And sometimes when reading, viewing, or listening to a narrative, we do
predict what will come next. But this is not our typical posture toward future
episodes in a narrative. For, as I have already noted, generally the earlier events in
a narrative underdetermine the later events. Prediction would be a feasible mental
state for readers, listeners, and viewers typically to be in if deterministic causal
models of narration were persuasive.That is, presented with causes, we would pre-
dict the effects. But I have suggested that very frequently the connection between
earlier events and later events in narratives is not that of a strict causal chain.
Indeed, typically, earlier events in a narrative underdetermine later events. So pre-
diction, in any strong sense of the term, seems an unlikely characterization of the
sort of anticipation that characteristically accompanies narrative.
But if prediction is not the core of narrative anticipation, what is? On my
account of the narrative connection, earlier events raise certain possibilities, rather
than others.They, so to speak, open a range of possibilities about what might hap-
pen next. From the point of view of the reader, viewer, or listener, this is a matter
of encouraging a range of expectations. Following the story involves having a
broad sense of where it is headed “ a broad sense of what might happen next “
rather than having a definite sense of what will happen next (as would be the case
with prediction).Thus, narrative anticipation is structured in terms of the reader™s,
listener™s, or viewer™s possession of a range of possibilities. When later events are
then entered into the narrative, the audience finds them intelligible just because
they accord with or fall into the range of possibilities suggested by the earlier
events in the story.
ON THE NARRATIVE CONNECTION 131

This is not to say that the reader, listener, or viewer has an array of concrete
possibilities before her mind as she processes the narrative. Rather, earlier events
open a range of possibilities and when later events arrive, we recognize that they
fall into that range. If ensuing events fail to fall into that range and are not recu-
perated by the addition of causal inputs, the narrative will appear incoherent and
unintelligible. Following a narrative with understanding is a matter of seeing how
the states of affairs that obtain in the narrative are possible, given the earlier events
in the narrative. This involves, in short, recognizing that the earlier events pre-
sented conditions for the realization of the later events. Specifically, it involves rec-
ognizing that the earlier events were causally relevant conditions (or contributions
thereto) for the occurrence of the pertinent later events in the story at hand.
These conditions make later events possible. Our sense of the direction of the
narrative “ of its intelligible unfolding “ is rooted in our sense of what kinds of
events are possible, given other events. Of course, causally necessary conditions
can branch in many different directions. Many subsequent itineraries are possible.
One causally necessary condition will support an indeterminate range of conse-
quences. But, of course, this is consistent with the fact that our sense of the direc-
tion of a narrative may also be very indeterminate. Remember that following a
narrative is not typically a matter of prediction.
On the other hand, that the direction of the narrative is indeterminate does
not mean that it is wide open “ that just anything can follow anything else. For
the causally necessary conditions that access some lines of development will
exclude other lines of development. So while earlier events open a range of possi-
bilities, they also preclude (save the addition of extenuating causal inputs) other
possibilities. Our sense of the direction of the narrative is not typically precise “
we do not typically predict future narrative events with exactitude. But our sense
of the direction of the narrative is not altogether without shape, even in the initial
stages of narration. Presented with a boy and a girl, we entertain the possibility
that they may or may not subsequently become lovers, but we do not, without the
addition of more information, anticipate that Mars will explode.The presentation
of a boy and a girl is not a causally necessary condition for a planetary eruption.
That our sense of the direction of a narrative is typically indeterminate fits
nicely with our hypothesis about causally necessary conditions.That is, an indefi-
nite though not wide-open sense of direction is what one would expect if the
basis of the narrative connection is generally only a matter of causally necessary
conditions (in the sense of INUS conditions), or contributions thereto. Or, alter-
natively, that the narrative connection depends primarily on such causally neces-
sary conditions explains why our sense of the direction of a narrative is generally
indeterminate (though not unbounded). Of course, as narratives proceed, piling
up more and more causally necessary conditions, the range of possible subsequent
events shrinks. We expect subsequent events to fall in the intersection of those
events made possible by the combination of the antecedent causally necessary
conditions. Indeed, the compilation of causally necessary conditions may even get
so exhaustive that we treat them as jointly sufficient and predict subsequent
132 ART, HISTORY, NARRATIVE
AND


events. But even when prediction becomes feasible, it is the special case and not
the norm of narrative anticipation.That is, it is not what we are generally involved
in doing when we follow a narrative, even if it is what we do at certain points in
a narrative, especially with respect to penultimate episodes.
Rather, narrative anticipation is a matter of forming expectations on the basis
of what events are possible, given earlier events in the story. Where do these

<<

. 24
( 82 .)



>>