Dear Andrew,

Because most film semioticians have adopted the continental (or
structuralist) perspective, much of the work done in this field has reduced
interpretation to a secondary role. This of course need not be the case,
especially if one adopts a Peircean point of view on representation and
interpretation. According to Peirce, interpretation is that which ensures
the relationship between a sign (representamen) and its object (immediate
object).  And in this respect, Eco sums up very well Peirce's thinking when
he writes:

"According to the process of unlimited semiosis founded and described by
Peirce, it is impossible to establish the signified [sic] of a term, i.e.,
to interpret this term, without translating it into other signs (whether or
not they belong to the same semiotic system) in such a way as for the
interpretant to account for the interpreted in some respect all the while
enabling a better knowledge of it." (Eco, _Semiotique et philosophie du
langage_, p. 109. my translation).

Interpretation thus constitutes an essential aspect of all semiosis
understood more or less as an infinite process whereby each interpreted
sign is in turn translated (i.e., interpreted) into another system (or
network) of signs and so on indefinitely. For Eco, Peirce's unlimited
semiosis, justifies the development of an encyclopedia-like model of
meaning. Yet, while readily accepting the fundamental semiotic principle of
such a model, Eco also raises the specter of overinterpretation. Hence, his
well known attempts to regulate unlimited semiosis and avoid any "cancerous
type of connotative growth" (Eco, _Les limites de l'interpretation_, p.
372, my translation -- the French text differs considerably from the
English edition)  and thus to distance himself from a "usage" of the text
akin to that of deconstruction and reader-response fundamentalism, and
whose roots the Italian semiotician sees as going back to medieval
hermetism (ars memoriae, alchemy, mysticism, doctrine of signatures, etc.).


According to Eco, overinterpretation occurs when the interpreter "uses" a
text, i.e., when the reader distances him/herself from the Model Reader
elaborated by the text and which constitutes the text's intention. In other
words, (proper) interpretation, for Eco, is a process whereby the reader is
constantly guided by the text in the construction of a plausible
explanation. There is no doubt that the epistemological framework for such
a conception is found in communication theory: Eco's interpreter (or Model
Reader) is someone who wonders what the text wants to communicate to him.

However, when we "use" a text, notes Eco, we conceive of it as a "stimulus
for the imagination." (Eco, _Lector in Fabula_, p.76 my translation)   For
example, one can read _The Trial_ as a detective novel or do like Proust
and "read a train schedule [to find] in the names of localities from the
Valois the soft and labyrinthine echoing of Nerval's journey in search for
Sylvie." (p. 78) But, according to Eco, such "readings" have little to do
with interpretation altogether.  Reading _The Trial_ as though it were a
detective novel is of course "legally allowed," he explains, "but textually
speaking it produces poor results." (p. 78).  As for the "Proustian
reading" of the train schedule, "it did not correspond to an interpretation
of the schedule, it was rather one of its legitimate uses, almost
psychedelic" (p. 78).

There is no doubt that Eco's conception of "limits of interpretation" can
be used in the context of film (or any other semiotic context). Its
epistemology is still very much structuralist, however (even though he
refers to Peirce). The initial problem is one that concerns semantics, but
philosophically, the core of the issue concerns the theory of knowledge.
Eco uses Peirce to substitute a dictionary-type (semantic) model for an
encyclopedic one. Yet he feels he must be very careful with this model: for
it can lead to forms of "hermetic" semiosis (i.e.,deconstruction, or even
pre-modern forms of knowledge -- such as those described by Foucault in
_The Order of Things_).In short, Eco is afraid of fraying to far from the
theory of codes. Yet if one considers the examples mentioned above, it
seems to me that as long as one avoids believing that such "readings"
uncover the indirect, yet "truthful", meaning of Kafka's novel or of the
train schedule, as long as one avoids seeing in them the "intent" of the
text or of the author, one also avoids the charge of symbolism (or
hermetism). What one does instead is interpret (an act which can -- and
should -- make use of the faculty of imagination). I don't have enough time
at my disposal to go into a full critique of Eco (and structuralist
semiotics) from the perspective of a non-structuralist semiotics, but it is
interesting to note that, as a novelist, Eco clearly enjoys
"overinterpretation" (this is, after all, the point of _Foucault's
Pendulum_). The fact is that art (literary, filmic, musical, etc.) lies
more on the side of unlimited/unrestricted semiosis than does (modern)
science. The failure of structuralist semiotics in the Fine Arts (film is a
good example -- but music is probably better still) is due in no small part
to the fact that the "discipline" of semiotics has been conceived of in
terms similar to those of modern science (i.e., it has been understood
solely in terms of a theory of knowledge). A different semiotic perspective
must be devised. One willing to consider singularity and work with
hypothesis rather than laws (or codes); one alluded to by Barthes (_Camera
Lucida_) in the distinction he made between the studium and the punctum. It
is no small surprise that, throughout the debate over iconicity, Eco has
sometimes been labeled an "iconoclast." Iconicity, after all (and we can
use Peirce on this), relies on ampliative rather than explicative
inferences and is based on resemblance. Semiotics can explain how one can
use resemblance semiotically (in representation), but it cannot explain the
experience of resemblance at the core of the iconic use (sign function) of
an object or thing through steadfast laws (or codes). To think otherwise is
to be an iconoclast.

hope this is of some use.


Martin Lefebvre

********************
Martin Lefebvre
Associate Professor

Editor RECHERCHES SÉMIOTIQUES/
SEMIOTIC INQUIRY

Mel Hoppenheim School of Cinema
Concordia University
FB 319
1455 de Maisonneuve, West
Montreal (Quebec), Canada
H3G 1M8

tel. (514) 848-4676/FAX. (514) 848-4255





%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%