Nelson Goodman is an important philosopher in the history of 20^th Century Aesthetics. Whereas aesthetics had largely been considered a branch of philosophy that was concerned with value theory, (e.g. the beautiful, the artistic, the creative, etc.) and what rational basis we have for such judgments, Goodman actually offers a unique take on important aesthetic questions that others had overlooked. Here, he is specifically looking at the nature of pictorial representation and how this is achieved. In his view, even “realistic paintings” succeed in reference by being understood as symbols and part of a symbol system. In the same way that the words of a language achieve reference by being part of the symbol system and interpreted as such, Goodman claims that visual representations must be understood in very much the same way. This is not only intriguing. but it expands aesthetics as a branch of philosophy to the borders of philosophy of language, philosophy of mind, and cognitive science.

Art is not a copy of the real world. One of the damn thing is enough.*

(* Reported as occurring in an essay on Virginia Woolf.)

Denotation

“Whether a picture ought to be a representation or not is a question much less crucial than might appear from current bitter battles among artists, critics, and propagandists. Nevertheless, the nature of representation wants early study in any philosophical examination of the ways symbols function in and out of the arts.

Representation is frequent in some arts.

How pictorial representation as a mode of signification is allied to and distinguished from:

1. Verbal description

2. Facial expression

Representation is a form of Denotation
Understanding the nature of representation crucial to any philosophical examination of the ways symbols function in and out of the arts.

Theories of Pictorial Representation:

1. The Most Naive (Resemblance)

a.) "A represents B if and only if A appreciably resembles B" (would not admit of degrees)

b.) "A represents B to the extent that A resembles B". (would admit of degrees)

Problems Immediately:

1. Representation is NOT symmetric relation, while resemblance is.

2. Further, it papers over the real question: In what ways do paintings “resemble” their referent? He notes that paintings resemble other paintings more than they resemble things in the world.

Overlooks the fact that a picture is a representation of an object means that it is a symbol for it (stands for it, refers to it).

3. No degree of resemblance is sufficient to establish the requisite (symbolic) relationship of reference, thus not sufficient.

4. Nor is resemblance necessary for reference. With regard to symbols, almost anything may stand for almost anything else.

NB: Thus resemblance is neither necessary nor sufficient for denotation. Denotation is the core of representation and is independent of resemblance.

The relation of a painting to its referent is assimilated to the relation between a predicate and what it applies to.

He proposes to examine the characteristics of representation as a special kind of denotation.

But…

Perhaps: resemblance, (while not sufficient condition for representation), distinguishes pictorial representation from denotation of other kinds.

c. If A denotes B, then A (pictorially) represents B just to the extent that A resembles B?

He answers “No.”

This watered‑down and innocuous‑looking version still suffers from a grave misconception.

Resemblance is simply irrelevant to how pictures represent their referents.

Question: What is the mechanism of representation?

2. Representation as Imitation/ Realism

"To make a faithful picture, come as close as possible to copying the object just as it is."

1. But there are numerous “ways” the object “is.” No one way is the way the object is.

2. An artists does not/cannot copy all these.[1]

3. “... and the more nearly I succeeded, the less would 'the result be a realistic picture.”

Maybe:

A realistic Imitation is to copy “the way the object is for (a) perception.” (Copy one of the ways the object is or looks.)

No: Needs more: consider a “realistic” painting of the Duke of Wellington as-he-looks-to-a-drunk-through-a-raindrop.

Maybe:

“To make a faithful picture, come as close as possible to copying the object just as it is for normal perception.”

As the object looks to “the normal eye, at proper range, from a favorable angle, in good light, without instrumentation, unprejudiced by affections or animosities or interests, and unembellished by thought or interpretation.”

In short, the object is to be copied as seen under septic conditions by the free and innocent eye.

But this requires that there BE such a thing as an “innocent eye.”

Ernst Gombrich “no innocent eye”

Gombrich argues that "pure observation" is impossible in either science or art. Rather, all observation is predicated by hypotheses, which in turn, create expectations. Only through testing hypotheses do scientists and artists amend their already perceived picture of reality.

“The eye comes always ancient to its work, obsessed by its own past and by old and new insinuations of the ear, nose, tongue, fingers, heart, and brain.”

The mind does not so much mirror as take and make.

The myths of the innocent eye and of the absolute given are unholy accomplices. (Similar to Kant’s criticism of Empiricism generally.) Knowing is not merely a processing of raw material received from the senses. Rather perception and interpretation are not separable operations; they are thoroughly interdependent. If so, then there can be no separation of the “raw” from the “refined.” Striving for an “innocence eye” may be productive he concedes (fresh insight), but it is never realizable. And the opposite effort can be equally tonic.

NB: The sober portrait and the vitriolic caricature, differ not in how much, but only in how they interpret.

Thus the Copy Theory unable to specify what is to be copied.

Further:

An aspect is not just the object, from‑a‑given‑distance‑and‑angle‑and‑in‑a‑given‑light;

It is the object “as we look upon or conceive it, a version or construal of the object.”

“In representing an object, we do not copy such a construal or interpretation, we achieve it.”

A picture represents x as a man or presents x to be a mountain, or represents the fact that x is a melon.

(And this is no less true when the instrument we use is a camera rather than a pen or brush. The choice and handling of the instrument participate in the construal. A photographer's work, like a painter's can evince a personal style.)

Goodman is Arguing for:

1. Relativity of Vision

2. Relativity of Representation

Gombrich amassed evidence to show how the way we see and depict depends upon and varies with experience, practice, interests, and attitudes.

Possible Objection to Radical Relativism of Realism (Realistic Depiction) : Perspective

Possible Counter-example: Perspectival Pictures are (Objectively) Realistic and Not Merely Conventionally So

1. To represent space correctly, an artist must obey the laws of perspective.

2. The laws of perspective are supposed to provide absolute standards of fidelity that override differences in style of seeing and picturing.

Therefore:

There are absolute standards of what constitutes a (some) realist depiction.

Gombrich himself seems to go along with this.

"One cannot insist enough that the art of perspective aims at a correct equation: It wants the image to appear like the object and the object like the image."

James J. Gibson writes:

“...it does not seem reasonable to assert that the use of perspective in paintings is merely a convention, to be used or discarded by the painter as he chooses, . . . When the artist transcribes what he sees upon a two dimensional surface, he uses perspective geometry, of necessity."[2]

Suggestion seems, initially, plausible:

Quasi-scientific theory of Arealistic perception:

Matching of bundles of light waves is a purely objective matter, measurable by instruments. And such matching constitutes fidelity of representation; for since light rays are all that the eye can receive from either picture or object identity in pattern of light rays must constitute identity of appearance.

Problems: Not sufficient to establish representation:

Of course, the rays yielded by the picture under the specified conditions match not only those yielded by the object in question from a given distance and angle, but also those yielded by any of a multitude of other objects from other distances and angles.

Identity in pattern of light rays is clearly no sufficient condition for representation. (While identical to many, it represents one and not others- similar to the objection that resemblance in symmetric but resemblance is not.)

But not even satisfactory as an account of fidelity of correct pictorial representation –Realism (where denotation/representation is otherwise established).

Problems: Weird as an Account of What is Going on in (Realistic) Pictorial Depiction:

1. Consider Required Restrictive Conditions of Observation:

The restrictive conditions of observation that are prescribed (similar to those used in Aoptimum color perception conditions) are blinding.

1. through a peephole

2. face on

3. from a certain distance

4. with one eye closed

5. the other eye motionless

6. the object must likewise be observed through a peephole

7. from a given (but not usually the same) angle

8. from a given (but usually not the same) distance

9. with a single unmoving eye.

Under these conditions, what we are looking at tends to disappear rather promptly. Experiment has shown that the eye cannot see normally without moving relative to what it sees.

The fixed eye is almost as blind as the innocent one.

2. Consider The Abnormality of the Required Conditions:

The specified conditions of observation are grossly abnormal.

(Yet such Artificial conditions are supposed to yield “realistic” viewing?

3. Consider The Effect of Previous Experience:

Even such artificial and blinding conditions overlook the role of previous experience on visual perception. There is more to vision than meets the eye; the same stimulus gives rise to different visual experience under different circumstances. The preceding train of visual experience, together with information gathered from all sources, can make a vast difference in what is seen.

4. Consider the Infrequency of Such Contrived Viewing:

The artist couldn’t (wouldn’t) paint a picture which is to be viewed one way, according to how it might appear under radically different conditions.

All this suggests that the artist's task is conveying what he sees.

How this is best carried out depends upon countless and variable factors, not least among them the particular habits of seeing and representing that are ingrained in the viewers.

Conclusion: Pictures in perspective, like any others, have to be read: Goodman’s Big Point

Possible Objections to GBP:

Photos:

But don’t photos give us objectively realistic renderings of visual reality which are innocent of interpretation and need not me “read” but merely (passively) viewed?

Goodman counters with the old saying,

“There is nothing like a camera to make a molehill out of a mountain.”

Note: Consider the social, political upshot of this point, particularly now that we are living in a “media” age.

2 Grandmothers in the park....

#1: Your granddaughter is so beautiful!

#2: Oh that’s nothing; you should see the photographs.

Consider Further the Pliancy of Human (depth) Perception:

Adaptation to spectacles of various kinds has been the subject of extensive experimentation. See, for example,

Melville J. Herskovits writes in Man and His Works (New York, Alfred A. Knopf, 1948), p. 381: "More than one ethnographer has reported the experience of showing a clear photograph of a house, a person, a familiar landscape to people living in a culture innocent of any knowledge of photography and to have the picture held at all possible angles, or turned over and over for an inspection of its blank back, as the native tried to interpret this meaningless arrangement of varying shades of grey on a piece of paper. For even the clearest photograph is only an interpretation of what the camera sees."

The rules of pictorial perspective no more follow from the laws of optics than would rules calling for drawing the tracks parallel and the poles converging.

The artist who wants to produce a spatial representation that the present‑day Western eye will accept as faithful must defy the 'laws of geometry'.

(Klee’s point)

All parallels in the plane of the facade project geometrically as parallels onto the parallel plane of the picture.

With eye and picture in normal position, the bundle of light rays delivered to the eye by the picture drawn in standard perspective is very different from the bundle delivered by the facade.

The behavior of light sanctions neither our usual nor any other way of rendering space; and perspective provides no absolute or independent standard of fidelity.

Sculpture:

Imitation is no better gauge of realism in sculpture than in painting.

“To portray faithfully is to convey a person known and distilled from a variety of experiences, not a momentary phase, embalmed.”

The sculptor undertakes, the problem of translation.

The distant or colossal sculpture has also to be shaped very differently from what it depicts in order to be realistic, in order to 'look right'. (Consider “The David” or Billboard Images.) These deviations are not reducible to fixed and universal rules. How an object looks depends not only upon its orientation, distance and lighting, but also on our training, habits, and concerns.

Fictions

A picture (like a predicate), may denote a single item. (Unit class)

A picture (like a predicate), may denote severally the members of a given class.

Example: a picture in dictionary is not denoting 'uniquely some one eagle, but distributively eagles in general.” Thus the picture has multiple denotations.

Note: he is NOT claiming that it denotes an abstract object Aform or eagle.@

A picture (like a predicate) may have neither unique nor multiple denotation. Pictures depicting Imaginary creatures or people have null denotation. The denote an empty class.

Note: He is NOT claiming that they denote possible or fictional objects. Like abstract objects, he wants to account for these types of pictures WITHOUT claiming that such things exist in ANY sense. But How? Read on….

Locutions as "picture of' and "represents" have the appearance of two‑place predicates. (misleading)

X is a picture of Y suggests (misleadingly) that, if true, one is ontologically committed to an X and a Y

From the fact that P is a picture of, or represents, a unicorn we cannot infer that Athere is something@ that P is a picture of or represents. (i.e. That there exists a Unicorn –in ANY sense.)

Goodman the puzzle is cleared up by pointing to the ambiguity between saying what the picture denotes and saying what kind of picture it is.

“W. V. Quine had sharpened the distinction between syncategorematic[3] and other expressions, and had shown that careful observance of this distinction could dispel many philosophical problems. I use the device of hyphenation (e.g., in "man‑picture") as an aid in technical discourse only, not as a reform of everyday usage, where the context normally prevents confusion and where the impetus to fallacious existential inference is less compulsive, if not less consequential, than in philosophy. In what follows, "man‑picture" will always be an abbreviation for the longer and more usual "picture representing a man", taken as an unbreakable one‑place predicate that need not apply to all or only pictures that represent an actual man.”

Thus “man-picture” describes the KIND of picture it is. Not what it represents if anything at all.

A picture may be of a certain kind (be a Pickwick‑picture or a man-picture) without representing anything.

Close parallel in the difference between a man description (or man‑term) and a description of (or term for) a man.

(For example “A jazz pianist who lived in Pennsylvania and died of lung cancer” is a description of a man, i.e. refers to a man (my father) and also a man description, while “the present King of France” is a man description, but not a description of (any) man -refers to an empty set.

1. Some pictures denote a particular man (singular reference)

2. Some pictures denote each of many men (multiple reference)

3. Some pictures denote nothing. (null reference)

The names and noun phrase "Pickwick," "the three‑headed man" and "Pegasus" have null extension i.e. they all fail to refer to any single thing. The second differs from the first in being, for example, a many‑headed man‑description, while the last differs from the other two in being a winged‑horse‑description. They DO NOT differ in terms of what they denote.

We can learn to apply "corncob pipe" or "staghorn" without first understanding, or knowing how to apply, "corn" or "cob" or "corncob" or "pipe" or "stag" or "horn" as separate terms.

Note: It is by learning what are unicorn‑pictures and unicorn descriptions are that we come to understand the word "unicorn."

Understanding a term is not a precondition, and may often be a result, of learning how to apply the term and its compounds.

Claims that denotation is a necessary condition for representation (the verb), but that sometimes the thing denoted is an empty set.

A picture must denote a man to represent him (accomplish the >act=), but need not denote anything particular man to be a man‑representation (picture belonging to a certain class) .

Note: Where a representation does not represent anything there can be no question of resemblance to what it represents.

Goodman concedes that understanding a term is not exclusively a matter of knowing how to apply it and its compounds; such other factors enter as knowing what inferences can be drawn from and to statements containing the term. (Thus involves a host of functional cognitive capacities)

The Repercussions of this for Art Appreciation and Criticism:

Because the world of pictures teems with anonymous fictional persons, places, and things, and what, if anything, is denoted is often irrelevant and inaccessible, we must proceed as if it did not represent anything. Rather we should only considering what kind of picture it is.

Indeterminate denotation treated as cases of null denotation.

Not every man‑picture represents a man, and conversely not every picture that represents a man is a man-picture.

Another observation:

"represents . . . as" has two quite different uses.

Take for instance “This painting represents the Duke of Wellington as an infant.”

a. a certain (long or short, continuous or broken) temporal part or 'time‑slice' of him. (replaced by represents "the infant Duke of Wellington")

b. genuine cases of representation‑as (represents the adult Wellington as an infant -infantile).

1. A picture that represents a fictional man is a man‑picture (denoting an empty set);

2. A picture that represents a(n actual) man as a man is a man‑picture denoting a man (him).

3. A picture that represents an actual man as a baby or a carrot stick is not a man-picture but denotes a man.

A picture may denote as a whole or by a part of it. (Duke and Duchess of Wellington)

1. denotes the couple, (whole)

2. denotes the Duke. (part)

Further it can classified in whole or part:

3. a two‑person‑picture (whole)

4. a man‑picture (part)

Does not represent the Duke as two persons;

Everyday usage is often careless about the distinction between representation and representation‑as.

Sometimes in saying that a picture represents a so and so we do not mean that it denotes a so and so but that it is a so and so picture.

In other cases, we may mean, both

A picture of a certain black horse may not be a black horse picture.

Conceivably, saying a picture represents the black horse might on other occasions mean that it represents the horse as black (i.e., that it is a black‑thing‑picture representing the horse) or that it represents the black thing in question as a horse (i.e., that it is a horse‑picture representing the black thing).

To say that Pickwick is represented as a clown can not mean that the picture is a clown‑picture representing Pickwick because there is no such thing as Pickwick;

The picture belongs to a certain rather narrow class of pictures that described as Pickwick‑as‑clown‑pictures.

Being a matter of monadic (one-place predicate) classification, “representation‑as” differs; drastically from dyadic (two-place predicate) denotative representation.

If a picture represents k as a (or the) so and so, then it denotes k' and is a so and so‑picture.

If k is (numerically) identical with h, the picture; also denotes and represents h.

And if k is a such and such, the picture also represents a (or the) such and such, but not necessarily as a (or the) such and such.

Thus with a picture as with any other label, there are always two questions:

1. what it represents (or describes)- (what objects, if any, it applies to as a label)-reference

2. what sort of representation (or description) it is.- (which among certain labels apply to it)

NOTE: If representing is a matter of classifying objects rather than of imitating them, of characterizing rather than of copying, it is not a matter of passive reporting.

Classification involves preferment; and application of a label (pictorial, verbal, etc.) as often effects as it records a classification.

Goodman’s Nominalism

When we classify an object, we merely place it is a set of object with which it has certain relations or shared properties. But any two object can be classified together, exemplifying some relationship or another.

The 'natural' kinds are simply those we are in the habit of picking out for and by labeling. Moreover, the object itself is not ready‑made but results from a way of taking the world.

NB: The making of a picture commonly participates in making what is to be pictured. A representation or description, by virtue of how it classifies and is classified, may make or mark connections, analyze objects, and organize the world.

A label associates together such objects as it applies to, and is associated with the other labels of a kind or kinds. Less directly, it associates its referents with these other labels and with their referents, and so on.

Standard sorting is often serviceable, even if humdrum.

“Grasps fresh and significant relationships and devises.”

Novel uses of old categories bring out neglected likenesses and differences, force unaccustomed associations, and in some measure remake our world.

And if the point of the picture is not only successfully made but is also well‑taken, if the realignments it directly and indirectly effects are interesting and important, the picture, like a crucial experiment makes a genuine contribution to knowledge.

To a complaint that his portrait of Gertrude Stein did not look like her, Picasso is said to have answered, "No matter; it will."

NB: “Nature” is a product of art and discourse.

So then…. “What constitutes realism of representation?”

Possible Remaining Answer #1:

A picture is realistic just to the extent that it is a successful illusion, leading the viewer to suppose that it is, or has the characteristics of, what it represents.

Improvement over the copy theory for what counts here is not how closely the picture duplicates an object but how far the picture and object, under conditions of observation appropriate to each, give rise to the same responses and expectations.

Not confounded by the fact that fictive representations differ in degree of realism; for even though there are no centaurs, a realistic picture might deceive me into taking it for a centaur.

Problems:

1. What deceives depends upon what is observed, and what is observed varies with interests and habits.

2. If the probability of actual confusion is 1, we no longer have representation; we have identity.

3. Probability of real confusion seldom rises above zero. Seeing a picture as a picture precludes mistaking it for anything else.

4. Appropriate conditions of observation defeat deception.

When viewing a representational painting I recognize the images as signs for the objects and characteristics represented signs that work instantly and unequivocally without being confused with what they denote.

What will deceive me into supposing that an object of a given kind is before me depends upon what I have noticed about such objects, and this in turn is affected by the way I am used to seeing them depicted.

(Consider the various attempts to mask the signs of aging. What would have worked in the 1700=s no longer does because we have come to notice other signs.)

Possible Remaining Answer #2:

The most realistic picture is the one that provides the greatest amount of pertinent information.

Problem:

Reversed perspective and colors, appropriately interpreted, yields exactly the same information. Realistic and unrealistic pictures may be equally informative. Since it provides the same true information it is Afaithful@ though not realistic. Thus correctness or truth is not a sufficient condition for literalism or realism.

The Atouchstone of realism@ is in how easily the information contained issues.

But this depends upon how stereotyped the mode of representation is (upon how commonplace the labels and their uses have become).

Conclusion:

“Realism is relative.” (i.e. determined by the system of representation standard for a given culture or person at a given time).

Egocentric ellipsis must not tempt us to infer that such pictures (or any others) are absolutely realistic. Whether a picture is realistic depends at any time entirely upon what frame or mode is then standard.

Realistic representation, in brief, depends not upon imitation or illusion or information but upon inculcation.

Representation is a matter of choice.

Faithfulness is a matter of information.

Realism is a matter of habit.

Our (Western) representational customs which govern realism also tend to generate Resemblance. This explains our (western) tendency to link representation and realism and resemblance.

Individual judgements of similarity are more or less objective and categorical, but the assessment of total resemblance is subject to influences galore, and our representational customs are not least among these.

The pictorial properties might be roughly delimited by a loose recursive specification. (more or less complete)

A picture=s denotation is dependent upon pictorial properties.

Formula can easily be broadened a little but resists generalization.

Not the crucial difference between pictorial and verbal properties, between nonlinguistic and linguistic symbols or systems, that makes the difference between representation in general and description.

Goodman subsumes pictorial representation under the category of “description.” He posits an analogy between pictorial representation and verbal description.

Reference to an object is a necessary condition for depiction or description of it, but no degree of resemblance is a necessary nor a sufficient condition for either.

Both depiction and description participate in the formation and characterization of the world; and "they interact with each other and with perception and knowledge. They are ways of classifying by means of labels having singular or multiple or null reference.”

“Application and classification of a label are relative to a system; and there are countless alternative systems of representation and description. Such systems are the products of stipulation and habituation in varying proportions. The choice among systems is free; but given a system, the question whether a newly encountered object is a desk or a unicorn‑picture or is represented by a certain painting is a question of the propriety, under that system, of projecting the predicate "desk" or the predicate "unicorn-picture" or the painting over the thing in question, and the decision both is guided by and guides usage for that system.”

To represent, a picture must function as a pictorial symbol; that is, function in a system such that what is denoted depends solely upon the pictorial properties of the symbol.

“The pictorial properties might be roughly delimited by a loose recursive specification. An elementary pictorial characterization states what color a picture has at a given place on its face; Other pictorial characterizations in effect combine many such elementary ones by conjunction, alternation, quantification, etc. “

Representation is thus disengaged from perverted ideas of it as an idiosyncratic physical process like mirroring, and is recognized as a Symbolic relationship that is relative and variable.

[1] In "The Way the World Is". Review of Metaphysics. voLj4 (1960), 1 pp. 48‑56, (Goodman has) argued that the world is as many ways as it can be truly described, seen, pictured, etc., and that there is no such thing as the way the world is. Gilbert Ryle (1900 – 1976) takes a somewhat similar position (Dilemmas [Cambridge, England, Cambridge University Press, 1954], pp. 75‑77) in comparing the relation between a table as a perceived solid object and the table as a swarm of atoms with the relation between a college library according to the catalogue and according to the accountant. Some have proposed that the way the world is could be arrived at by conjoining all the several ways. This overlooks the fact that conjunction itself is peculiar to certain systems; for example, we cannot conjoin a paragraph and a picture. And any attempted combination of all the ways would be itself only one and a peculiarly indigestible one of the ways the world is. But what is the world that is in so many ways? To speak of ways the world is, or ways of describing or picturing the world, is to speak of world‑descriptions or world‑pictures, and does not imply there is a unique thing or indeed anything that is described or pictured. Of course, none of this implies either that nothing is described or pictured.

[2] From "Pictures, Perspective, and Perception", Daedalus (Winter 1960), p. 227. Gibson does not appear to have explicitly retracted these statements, though his interesting book, The Senses Considered as Perceptual Systems (Boston, Houghton Mifflin Co., 1966), deals at length with related problems. ''

[3] Incapable of serving as a categorical term, that is, word or phrase that designates a class. A categorical term divides the world into two parts: the original class and its complement. Hence, a syncategorematic term any linguistic expression that does not refer to anything else. Thus, "if," "while," and "and," are all syncategorematic terms, i.e. terms with no reference.