Noël Carroll
Philosophy of Art: A contemporary introduction
(1999)
[33]
Part II
What is representation?
Pictorial representation
...
...
...
[34]
...
Traditional approaches to
pictorial representation
... The resemblance theory of representation states that x represents y just in case x resembles y. ... the illusion theory of pictorial representation maintains that x represents y just in case x causes the illusion of y in spectators. ...
...
... Plato thought painting to be strictly analogous to holding a mirror toward an object. ... [hence] Plato held what we are calling a resemblance theory of representation. ... Note that this theory claims two things. First that resemblance is a necessary condition for representation—that x represents y only if x
[35]
resembles y. But it also claims something else, namely that if x resembles y, then x represents y. ... The first occurrence of "if" here signals that resemblance is a sufficient condition for representation—... The "only if" portion of the formula states that resemblance is a necessary condition for representation. ...
... if x resembles y, does it follow that x represents y? This seems false; ... Imagine two [identical] automobiles... They roll off the assembly line one after the other, ... These two ... will resemble each other maximally, but neither represents the other. ...
... Resemblance is a
reflexive relation.
... But
representation is not reflexive:
I resemble myself in every respect, but I do not represent myself. ...
... resemblance is a
symmetrical relation.
That is, if x is related to y, then y is related to x in the same way (xRy if and only if yRx). If I am Pat's brother, then Pat is my brother. ... But
representation is not a symmetrical relation.
If a picture of Napoleon resembles Napoleon, it follows that Napoleon resembles his picture, but it does not follow that Napoleon represents his picture. ... Thus,
[36]
resemblance cannot serve as a model for representation, ... there will be many cases of resemblance ... that will not warrant attributions of representation. ...
One might try to get around this objection via amending the resemblance theory by stipulating that x must be a visual design. ... Thus, even if Napoleon resembles his portrait, we will not say that he represents it because Napoleon is not a visual design. But this calls attention to a[nother] problem ...
What most visual representations
resemble most
are other visual representations.
A picture of Richard Nixon looks more like a picture of Bill Clinton than it looks like Richard Nixon. ...
...
...
Resemblance, then, does not appear to be a sufficient condition for representation. But is it a necessary condition? ...
... When we say that
one object represents another object,
we mean, at the very least, that
the first object is a symbol for the second object.
... But what is a symbol? ... Peirce
[37]
defined a symbol as
a sign
"whose special significance or fitness
to represent just what it does represent
lies in nothing but the very fact
of there being a habit, disposition or other effective rule
that it will be so interpreted."
...
Consider a military map. A thumbtack can stand for an armored division, but it does not resemble an armored division. ... In a context like this one, what stands for the armored division is arbitrary. ... But
if the symbol relation (denotation) is the core of representation,
and if denotation can obtain without resemblance,
then resemblance is not a necessary condition for representation.
...
...
[38]
...
... call this the "core argument" against resemblance ... we shall have to return to it shortly in order to see whether it is genuinely successful. ...
... another traditional theory of representation is the illusion theory. ...
x represents y if and only if x causes the illusion of y in spectators.
...
But the illusion theory has even more problems than does the resemblance theory. ... Under standard conditions, who is ever really fooled ... ? ...
... Typically we know that we are looking at a picture, not its referent. ...
[39]
... When viewing pictures, we often appreciate the verisimilitude of the picture. But if we thought the picture was its referent, it would make no sense to appreciate its verisimilitude. ...
Additionally, in order to view pictures properly, we must learn to "see through" their surface distortions.
...
But in order to see through these features, we must know that we are looking at pictures and not their referents "in nature."
...
The conventionalist theory of
pictorial representation
The case for the conventionalist approach begins with the unimpeachable observation that there are different pictorial systems. For example, the ancient Egyptians ... the Italian painters of the High Renaissance. ...
Furthermore, it is often claimed that people from different cultures ... allegedly have difficulty comprehending representations in alternative styles. ... In order to comprehend the pictures of other cultures, the conventionalist argues, the spectator must learn the
[40]
conventions of the relevant pictorial system ...
... [e.g.] In order to understand that the halo around the head of a woman in a painting signifies that she is a saint, one must know what those illuminated circles stand for. The conventionalist contends that all pictorial phenomena are like this. ...
... what denotes what is arbitrary. A representation is a symbol and what fixes the reference of a symbol is some kind of rule or code. Thus, representation is conventional. ... I am calling this approach conventionalist, but it might also be called
"semiotic"
...
The resemblance theory and the illusion theory are
naturalistic theories
of pictorial representation in the sense that they presuppose that there is
some universal psychological process
that explains pictorial representation: either the percipient
naturally detects
certain similarities and surmises representation on the basis of this, or, on the other hand, the representation
somehow causes
normal spectators to believe that the referent of the representation is before them. But for the
conventionalist,
pictorial representation is an affair of
acculturation.
...
x pictorially represents y only if x denotes y in accordance with some established system of established conventions.
...
[41]
...
... one objection to the conventionalist theory might be that it fails to explain why we experience certain pictures as more realistic than other pictures. ... this does not appear to square with the conventionalist theory, ... if [pictorial representations] are all arbitrary, none should appear more realistic than others.
... The conventionalist maintains that
the representations that we call realistic are merely the ones with which we are the most familiar.
Once we become habituated to a given style of representation, it seems natural to us.
Interesting to note once again the effect this has on the academic chess board: we would then (i.e. now!) live under a pluralism of unitary convention; and it is then (now!) actually far more difficult to be "correct." In fact, just take the raw number of conventional systems you can be held responsible for (explicitly!) knowing: that's the coefficient of change versus your ancient ancestral environment!
Also, once again, there is, on one hand, the theory and its ability to sink or swim qua theory; and on the other hand, independently of its truth or usefulness, there is the fact that conventionalism too is a spiral-closing maneuver.
...
...
... Giotto's work was once regarded as realistic, but now it seems much less so when compared to the work of David. The reason, the conventionalist maintains, is that the conventions of realism in the West have changed. Perhaps, various works of Cubism
strike us as unrealistic now.
But if Cubism were to persist and become the most
dominant and familiar
form of pictorial representation, then, the conventionalist predicts,
it would come to strike us as realist,
Good one!
since the impression of
[42]
realism is nothing more than habituation within a given symbol system. ...
...
It is hard to imagine that we could ever come to regard Cubist or expressionist portraiture as realistic. But conventionalism is committed to this view,
...
A neo-naturalist theory of
pictorial representation
...
... there is evidence, based on cross-cultural research, that pictorial representations travel far more smoothly from society to society than the conventionalist suggests.
Evidence?!
What is this, social science?
Is this not even philosophy?
/ • :]
... Western cinematography is now employed cross-culturally; people in remote villages in India
understand
Hollywood motion picture images
without taking film courses.
" . . . and without analysing the concepts 'film,' 'image,' 'motion picture,' . . . "
... We [contemporary Westerners] can
recognize the kinds of things
that Assyrian bas-reliefs represent ...
without knowing anything
about Assyrian culture. ...
... Children raised without seeing any
[43]
pictures for the first year and half of their lives are able to recognize what pictures are pictures of on their very first exposure to them. ...
...
... as the preceding examples indicate, ...
pictorial comprehension
is not really very much like
linguistic comprehension,
which is
our best model
for understanding what is involved in coming to learn arbitrary conventions of the relevant sort.
Why tf
is
our best model for
this issue of importance
not really very
good?
...
... the conventionalist is apt to reply: there is also evidence of cross-cultural incomprehension ...
[44]
... However, ... the evidence is not really so clear-cut. There is some question about the quality of the photographs that tribal peoples failed to recognize, and, in some cases, the reasons behind their failures can often be explained by noting that they did not understand what they were being asked to do. ...
Sounds a lot like the TSA recertification process!
...
... the anthropological evidence for conventionalism is at best mixed and surely controversial, while, additionally, there is a great deal of evidence—such as
the smooth dissemination of pictorial comprehension across cultures and generations
—which is difficult to assimilate into the conventionalist model. ...
... The capacity
to recognize what pictures picture
appears to evolve in tandem with the capacity
to recognize objects —
once we are able to
identify
something, such as a horse, "in nature," we have the capacity to
recognize
a picture of a horse, ... [recognition] seems to involve
a natural capacity,
already evident in young children.
... then perhaps what pictures are are objects designed to trigger this natural capacity.
...
[45]
...
... According to the illusion theory, viewers are
deceived
into believing that the referents of the picture are before them. But this theory does not claim that any deception is involved;
it stresses recognition, not deception,
... we may call it a neo-naturalist theory of pictorial representation ...
...
...
[46]
...
... I have already said that [conventionalism] is a counterintuitive doctrine. But that is not an argument. Here is an argument.
Within our own culture,
we have been able to comprehend the emergence of new styles of realism. ... if conventionalism were true, it would be hard to understand how such breakthroughs could be possible, since we will always be more habituated to the existing style than to the breakthrough style. ...
...
Conventionalism also claims that any pictorial style will be thought of as realistic, once it becomes familiar. This seems unlikely; ... But neo-naturalism can explain this too. ...
... perhaps the conventionalist can use what we previously called the "core argument" against the resemblance theory in order to defeat neo-naturalism.
... is recognition a necessary condition of representation? ... clearly x may represent y—x may denote y-without our being able to recognize y in x. ... [e.g.] the thumbtack ... standing for an armored division. Consequently, can't we run the core argument against the resemblance theory with equal force against neo-naturalism, replacing the notion of
[47]
resemblance in the earlier argument with the notion of recognition in a new argument?
...
...
The first premise states that "x represents y if and only [if] x denotes y." ... from our perspective, this premise is ambiguous. If we are talking about
representation simpliciter,
the premise is true. A thumbtack can represent an armored division when we stipulate that it stands for one. But if we mean by "representation" a
"pictorial representation,"
then the premise is false. That
denotation alone marks representation simpliciter
does not entail that
it defines every particular subcategory
of representation. It does not define pictorial representation entirely, for example.
Though a thumbtack
can represent
an armored division, a thumbtack
is not a pictorial representation
of one. ... inasmuch as the core argument is
[48]
supposed to pertain to pictorial representations, but implicitly rides on the notion of representation simpliciter ... [it] appears to commit the fallacy of equivocation.
...
...
...
...
...
[49]
...
... the neo-naturalist can still learn something from conventionalism. For even if recognition is a necessary condition for pictorial representation, it is not a sufficient condition, since, though
we may recognize faces in clouds,
clouds
are not representations of the faces
we see in them. Why not? Let us exploit the conventionalist proposal that denotation is also a necessary condition of representation [even though it is not sufficient] ...
This results in a more complex statement of the neo-naturalist theory of pictorial representation, namely:
A visual design x pictorially represents y (an object, person, place, action, event or another visual design) if and only if (1) x has the
intended
capacity to cause a normal percipient to recognize y in x simply by looking; (2) the relevant percipients recognize y in x simply by looking; (3) x is
intended
to denote y; and (4) the relevant percipients realize that x is intended to denote y.
...
... neo-naturalism and the resemblance theory may be combinable, since it may turn out that the psychological mechanism that is a cue for percipients to recognize the referents of pictures is resemblance. ... Whatever the mechanism is that secures pictorial recognition is a job for psychologists to discover. ... philosophy attempts to say
what pictorial representation is,
not
how it works psychologically.
[50]
Representation across the arts
...
... we may say that by "represent" we mean that x represents y ... if and only if (1) a sender
intends
x (e.g., a picture) to stand for y (e.g., a person), and (2) an audience
recognizes
that x is intended to stand for y.
In case you missed it . . .
Pity the
audience
who happens to
recognize
some
x
as
standing
for y
when nothing of the sort
was
intended
by any
sender!
That is not even representation!
i.e.
This
just is not
what pictorial representation is.
Care to say what it is?
...
... consider four types of representation in order to characterize the ways in which representational practices differ with respect to different artforms. ...
...
1. Unconditional representation.
... In [such] cases ... we can recognize that x stands for y on the basis of
the same recognitional powers
that enable us to recognize y's "in nature."
If we can recognize women in the real world simply by looking, then we can recognize that the Mona Lisa pictures a woman ...
...
[51]
...
2. Lexical representation.
... [some] forms of representation are
coded, or lexicographic, or semiotic.
In these cases,
in order to realize
that x stands for y, a spectator
must know
the relevant codes. In dance, certain gestures and movements are
correlated to definite meanings
in a dictionary-like fashion.
In the Romantic Ballet, ... when a character draws a circle around her head, that means "I am pretty." One cannot realize this simply by looking; ...
...
the boundary between unconditional and lexical representation often yields mixed cases,
since what we sometimes recognize in an unmediated way is a socially (rather than an artistically) coded signal
(e.g.. we unconditionally recognize the fire truck in part because it is red, but its being red is the result of an antecedent social code).
3. Conditional specific representation.
Sometimes
we recognize
what is being represented only on condition that
we already know
what is being represented. One is unlikely to realize that poison is being put in the king's ear in the play within the play in Hamlet, unless one already knows ...
we wouldn't have a clue as to what is going on without being told (by Shakespeare). ...
Nor is this a case of lexical representation, since there is no preestablished code for ear poisonings in drama. ... not all conditional specific representation need be lexical, since in many cases it
[52]
can operate by engaging natural recognitional capacities,
once those have been alerted or cued
in terms of
what to expect
(i.e., what to be looking for).
Sounds like a job for experts about art who are essentially reformers! Here is a chance for them, finally, to nondestructively "reform" some people who don't already know what to expect, but who really can come to know as much and as well as anyone ever has, simply by being told.
We finally have located this prima facie legitimacy for the Appreciation Racket . . . where, exactly? Under the third of four sub-sub headings in a brief concluding section, marking just one of four types of representation, none of which are fully applicable across each and every artform.
Don't call us, we'll call you.
NB/TBC: Conventionally, a "racket" in Thomson's sense denotes either
(1) a two-sided operation, befitting the metaphor of a sportsball being volleyed back and forth openendedly;
or perhaps more generally
(2) mere "deceptive practices to manipulate a market or defraud customers," as a cursory Google search suggests.
Thus, if we were malignly interested in extending the metaphor beyond its sparser manifest content, we might point out that these reformer-experts are dependent on some artist(s) to take the other side of the racket: somebody somewhere must be engaging in profligate conditional specific representation.
If not, well . . . we'll just have to insist (i.e. "deceptive"-ly) that this just is what some artist(s), somewhere, are doing or have done. We must hope against hope that these artists somehow become important enough to enough people but, somehow, without these artists just telling those people what is being represented, and without those people being smart or resourceful enough (despite their interest) to find out for themselves.
And if they have found out, we can just lie to them: we can hint at all the other things they haven't learned and don't know. How would you know that something is made up if the first premise of the person informing you is precisely that you don't know about it?
in predominant advertising theory [...] It was recognized that in order to get people to consume and, more importantly, to keep them consuming, it was more efficient to endow them with a critical self-consciousness in tune with the "solutions" of the marketplace than to fragmentarily argue for products on their own merit."
...
Linking the theories of "self-consciousness" to the exigencies of capitalism, one writer in Printer's Ink commented that "advertising helps to keep the masses dissatisfied with their mode of life, ... Satisfied customers are not as profitable as discontented ones."
(Ewen, Captains of Consciousness, pp. 38-39)
In other words, a "racket" becomes self-sustaining when the consumer takes both sides against themselves. Once the consumer engages in "deceptive practices" against themselves, they're in for a long rally.
Deciphering a case of conditional specific representation can involve a complex interplay of cognitive abilities ... However, this category still marks an important difference, since
in order to mobilize
the relevant cognitive capacities,
we need
the clue
that something
is being represented.
...
4. Conditional generic representation.
Here the spectator is able to detect or to recognize that x stands for y on condition that she
knows that something is being represented.
... unless you know that I am trying to represent something, you might not take my rolling arm movements to stand for waves. ... if we know that a piece of music is a tone poem, then we are likely to interpret certain "rushing" or "flowing" phrases as water.
And if we do know but do not interpret, . . . ?
If the audience too has intent, and it is not this, then . . . ??
Simply knowing
that an artistic signal is
meant to be representational,
even if we are not
told
exactly
what it represents, leads us to
mobilize
our natural recognitional capacities,
...
without being told its specific, intended meaning.
...
you might try to get the hang of the distinction by playing charades.
...
Team A gives a player on Team B a saying
...
...
[53]
...
For Team B, their teammate's gestures are regarded as conditional generic representations at the same time that for observers from Team A, they are conditional specific representations.
...
Clearly, the players on Team A and the players on Team B are engaged in different cognitive tasks,
...
artworks can employ either sort of representation as well.
Honegger's
Pacific 231
and Berlioz's
Symphonie Fantastique
are examples of conditional specific representation, while the thunder in Beethoven's
Pastoral
is more of the nature of conditional generic representation
(one wouldn't
hear it as
thunder
What if the
knowledge
(I mean
collectively
forgotten, not individually.)
This kind of knowledge is constantly passing out of existence. Most
conditional specificity
is very fragile. So, it takes constant
expert
intervention just to maintain it.
Alternatively, we could stop trying so hard to maintain it, and thereby put any given
representation
to the
unconditional
test.
And if it fails? Then the work has become
decorative.
If the work itself cannot
speak for itself,
not even in this purely procedural, mechanistic sense, then what is the sense in
mobilizing
to speak on its behalf?
Flux and forgetting are not quite what the self-regarding artist
intends! But this is what happens! This too is a
naturalistic theory, because forgetting too is an eminently natural phenomenon.
The wasting-away of all
lexical,
conditional,
historical
baggage
is
a natural process.
What is "artificial"? It is the desperate lurch to keep the corpse of context alive that is artificial, faux-heroic, evil.
TBC #1: Here the parallels with Taleb's concept of
"fragility"
are unmistakable. Interested intervention to subvert "natural" processes-in-progress
is
the
fragilista
archetype.
(Of course saying concretely what is natural and what is not is not so easy; that is one huge piece of work left to be done in making this connection.)
(Another would be in taking account of any differences between Taleb's domains and ours here. This is a base to be covered, certainly, but I expect it is not nearly so obstinate as the "nature" issue; it may even point towards a simple empirical case that does not rely on any contested concepts.)
TBC #2: David Rieff has written a book,
In Praise of Forgetting: Historical Memory and Its Ironies
which argues, in an unforgivably pompous manner but effectively nonetheless, that it would be better if more were forgotten. Perhaps there's no sense in arguing that the elapsing of conditions of representation actually is a "better" outcome, but I see no more sense in arguing that it is worse. What we can say is that powerful, endemic forces (can we just call them "natural" even though that's not the
absolutely perfect
word?) are pushing very hard in the direction of forgetting; so, to countervail those forces, we have to push even harder against them; this creates fragility, and as the saying goes,
the cracks are starting to show.
...
though
[these "four types of representation"]
can be combined and melded in particular cases in very complex ways, they nevertheless are helpful in characterizing the typical ways
[of]
particular artforms
...
[which]
tend to emphasize or to rely upon certain of these categories more than others.
...
...
[e.g.]
dance representation can fall into all of the four categories,
...
And yet it does seem that dance relies on some of these categories more than the other arts do—
...
[54]
...
...
theatrical dance as we know it does employ forms of conditional representation more than the dominant forms of
[e.g.]
theater, film and TV do
...
Call this difference in choice of representational strategies a
proportionate difference
in the representational means
...
...
in the main (the mainstream),
theater, film and TV use categories 3 and 4 to a lesser extent than they use category 1,
whereas,
while dance uses category 1,
it also relies very heavily on using category 2 and, especially, on categories 3 and 4
...
This should be clear from the degree to which
even mainstream
dance representations
depend on
for intelligibility,
whereas mainstream theater, film and TV are
generally accessible without
such
enabling texts.
...
what of music?
...
when there is representation in music, doesn't it rely heavily on conditional representation too (think of the
1812 Overture)?
...
...
dance uses unconditional representation more than music does.
So where dance differs from theater, film and TV by its emphasis. on forms of conditional representation,
it differs from music in its far more frequent use of unconditional representation.
Of course, it is also in virtue of their primary reliance on unconditional representation that the dominant practices of theater, film and TV distinguish themselves from music.
...
[55]
...
representational painting, drawing, and sculpture depend primarily on unconditional representation
...
[but]
also
...
on conditional representation, deploying
titles, captions, and backstories
...
these artforms appear to fall somewhere between dance, on the one hand, and film, theater and TV, on the other hand,
...
I gather
that the "objective" musical
program note
is only about two hundred years old. What about these other
enabling texts?
Isn't this all about chasing the wolf of pluralism off of our doorstep?
...
...
each artform employs the same types of representation that the other artforms do.
Where they differ, when they differ, is in virtue of the differential, proportionate emphasis they place on the different types of representation available to them.
Chapter summary
...
[56]
...
the development of nonrepresentational art in the nineteenth and twentieth century rendered the representational theory of art obsolete, while also alerting theorists to the fact that it had never really been fully comprehensive.
However, even if the representational theory of art is
false as a general theory of art,
it is still the case that
a large amount
of the art with which we are familiar is representational.
...
Thus,
a theory of representation is
still an urgent task
for the philosophy of art.
Ostensibly this is the destiny of all future chapter-concepts too. But why not just devote the chapters to analyzing each concept? What there is of that here is very helpful. The search for
a general theory of art
has been, so far, not nearly as productive, and it has been quite awkward and choppy, perhaps in part because its concepts are introduced with insufficient prior analysis.
The "What Is Art, Anyway?" stuff should be all together at the end, in a concluding chapter, after we
know what we're talking about.
(Do people actually teach this book? Tbh that's a little scary.)
...
...
Annotated reading
...
unless
one knew
that the piece was
illustrative).
that such a piece
was illustrative
is forgotten?
accompanying descriptive texts
—such as program notes—
Oh?
No comments:
Post a Comment