Margolis—The Importance of Being Earnest about the Definition and Metaphysics of Art

---

[215]

The Journal of Aesthetics and Art Criticism, Vol. 68, No. 3 (SUMMER 2010), pp. 215-223

JOSEPH MARGOLIS

---

The Importance of Being Earnest about the Definition and Metaphysics of Art

[SK's comments]

The philosophy of art may be doomed,
again and again
but always once and for all,
to define what it is to be
"a work of art,"
an "artwork,"
"art"
in the sense

best suited to "the fine arts."

Modern efforts seem to end in
exhaustion
or bafflement
or sheer scatter
or

a sort of bad faith

that assures us that it was never worth the bother in the first place.

But many are troubled by the nagging sense of failure that no mere dismissal seems able to dispel.

We are still caught in the puzzles of Morris Weitz's extraordinarily disturbing early essay, ...

published ... only a few years after the appearance of Wittgenstein's Philosophical Investigations (1953),

on which,

however inaccurately,

it claims to depend.

If Weitz's account is regarded
as
the original gauntlet analytic aesthetics took up so eagerly almost at once ...

it is equally plausible
to credit Arthur Danto's immensely influential essay,
"The Artworld," published in 1964,
with
having confirmed
(by the absence of explicit mention,
so to say)
the single most-discussed version
of
the great complication
that
baffled all efforts
at
defining art
in accord with
the new spirit of rigor
that had just taken hold
in
analytic aesthetics.

I mean,
of course,

the trauma

produced by

Duchamp's devilish jokes

(especially the notorious Fountain)

transfigured,

in Danto's hands,
into

the sober metaphysics

of Warhol's Brillo Box.

As always: nobody can make us feel inferior without our consent!

Nothing is either

devilish

or

sober

unless we, so to say, give it permission to that effect.

The (extended) metaphor of a

joke

taken the wrong way
is apt here.

An unfunny joke could be an ungotten joke,
or
it could be perfectly gotten and found (perfectly!) unfunny.

The

transfiguration

from "devilish" to "sober" is indeed emblematic of the problem, for it is a case of the theorist laughing last.

We nontheorists and antitheorists give no such 'permission' and hence suffer no such

trauma

.

So far, so good.
But then, it was not just in

analytic aesthetics

that Duchamp's "jokes" were received with deathly seriousness: in fact they were so funny that a few people kept telling them over and over again, until (very quickly!) they weren't the least bit funny anymore; and this, finally, affords True Believers the tactic of 'compelled speech', whereby the unfunny joke is (re)told and everyone's feigned laughter is gauged for its propriety and earnestness.

That's the "trauma" we really ought to be working through,
whether with our theories or with our work.

A great swath of well-intentioned responses to the double challenge of Weitz's verdict and Danto's new turn fell over the philosophical cliff
in an effort to get things right
in accord with the seeming fashions of the day.

They have engaged a large part of the energies of contemporary aesthetics for the last fifty or so years—and they still do.

Yet
a small caution suggests
that
the puzzles thus confronted
have yielded rather meager gains
(at least of the mettle sought).

Have we perhaps

entrapped ourselves in unguarded readings of the errors we supposed we had uncovered

in
the various now-canonical summaries
of
the work of the entire interval spanning Weitz and Danto
(and beyond)?

Before venturing any new possibility here,
consider only that
Wittgenstein, Weitz, Beardsley, and Danto—and very nearly a hundred other reasonably well-known discussants who have bothered to address the difficulty of defining art—

have never really attempted to define "definition"

as a way of approaching the matter,
though nearly all

have fastened on particularly strenuous kinds of definition

in introducing their own contributions,
and
many have turned from definition to

other sorts of "characterizations" or "theories" of art

thought to escape the unlucky puzzles of definition.

Beardsley and Danto,
it should be noted,

deliberately avoided defining art

—at least initially in their long careers—
and Weitz turned against his own pre-Wittgensteinian effort at definition.

Nearly all of the best-known essays are known to have serious defects.

But the question remains too important to be dropped for reasons of philosophical fatigue.

Can we afford another run-through?

I mean . . . obviously not!

If that's the hook, I'll bite.

---

I don't care half as much about

the question

as I do about this matter of

serious defects

in

Nearly all of the best-known essays

.

All told, that is quite a statement. It implicates the field (the 'philosophical community') as a whole. Arguably, it implicates the authors least of all. Indeed, Margolis is positively Beardsleyan here in attaching the defects to the works rather than to the authors.

Perhaps authors have little to do with their work becoming

best-known

.

The perfuctory ritual of duly crediting peer-reviewers and local colleagues while absolving them of responsibility for errors is quite a tell. The buck has to stop somewhere, and customarily that is with the person whose name is on the marquee (that day).

Really we need the same disclaimer about what happens to the paper after it is published:

". . . The community's propensity to cite and discuss my paper endlessly has made it among the best known essays in the field; responsibility for the propagation and enfranchisement of the paper's many serious defects, however, lies entirely with . . ."

With whom? Not with the author, not unless he is running a malign 'citation cartel' out of his own institutional perch. No, the "best-known essays" are bound to be the ones that tell us what we want to hear, whether we are philosophers or Substack Randos.

The formal and informal review processes are supposed to provide some protection against this kind of 'audience capture'; they are supposed to provide this protection no less for all the chatter which ensues in the wake of a blockbuster than for the seminal paper itself. But if someone of Margolis' stature is willing to make the above remark, then the rest of us are entitled to doubt whether review been of any use at all here; whether absolving reviewers of responsibility has had its intended effect; whether the whole thing isn't just a broken arm society of people patting each other on the back. We've always been told that this epoch of academic philosophy was too adversarial, to the point of toxicity. But then, has it been adversarial enough? Has it been adversarial in the right ways? Obviously not!

The juxtaposition of Weitz and Danto
is
more than ordinarily instructive.

Because,
though Weitz's essay is

smartly up-to-date

in invoking Wittgenstein's compelling remarks about "family resemblances" and "strands of similarity,"

his basic argument largely relies on

purely formal

[216]

considerations

already in place

in the philosophical canon,
collected in the name of

a perfunctory notion of artistic creativity

...

Weitz reminds us,
as his argument unfolds,
of certain well-known artworks that once worried the flexibility of our classificatory schemes
but that
have now been safely reconciled with them: ...

The entire topic is unbelievably tangled—
very hard to lay out fairly
without exposing serious misreadings

that have
dug their heels in
when
confronted by
the doctrinal wounds
suffered by
veteran philosophers
who
were not entirely fortunate
in
their own formulations
(and revisions of formulations)
in
an effort to meet the definitional threat.

In any case,
what is missing in Weitz's original essay
is
any mention or hint of the appearance and philosophical reception of Duchamp's Fountain (in April 1917),

I am ill-equipped to deal in brute facts here, so I won't attempt to.

I'll just speculate, having given Weitz the customary Fickle Ears treatment, that he does not strike me as someone who would have been terribly up to date on such matters, in spite even of a half-century's head start. This is pure conjecture, but it is, incidentally, far more parsimonious than a "trauma"-centered account.

...

I am merely guessing

Oh.

Well, good, then I'm not alone after all!

when I suggest
that
Danto "replaces" Duchamp with Warhol
in the "Artworld" paper
partly because

Warhol's Brillo Box is not a readymade

and, for that reason,

serves Danto's argumentative interests

better than Duchamp's specimens—along the lines of the indiscernibility thesis

(which would have had to be presented very differently if Fountain had been preferred).

Why is this?

A 'readymade' Brillo Box would seem actually a better vehicle for Danto: a true indiscernible! And it won't do to say that a maker of indiscernibles has this intent and the Duchampian does not, since, if this were true, each would then do far better to adopt the other's tack.

It is also not unreasonable to suppose
that
Danto wished to avoid the enormous swell of disputes already tripped by the appearance of Duchamp's Fountain—
for instance,
entanglements with Clement Greenberg's slightly earlier statements about modernism and the significance of Duchamp's readymades.

Danto's theory would have had to be viewed in an entirely different light:

its force and originality might actually have been diminished,
since
Danto turned away from the primacy of definition to that of the "metaphysics" of art
(hence, to the puzzle of indiscernibility),
which
might have been raised,
with a very different emphasis,
by preferring Duchamp's Fountain.

So, Fountain was poorly received and Brillo Box well received? That could affect the kind of philosophical reception one could anticipate, but I still don't see how it dictates changes the substance of the argument(s).

Here's a little piece of Intentionalist criticism: it seems Warhol was simply a more artful concealer of his own maliciousness, whereas Duchamp (perhaps also intentionally) was simply more direct.

Duchamp
made the very effort to define art problematic
for reasons entirely missed by Weitz;

and
though the admission of Warhol's Brillo Box as art
surely affected art's definition as well,
it deflected
(and was in part meant to deflect)
that question's priority.

Duchamp

threw

the artworld and the aesthetician's sense of the "artworld"

into disarray;

Warhol

petitioned for admission

to the established artworld.

I prefer:

Duchamp was mounting a blitzkrieg;
Warhol was playing a long con.

Greenberg
saw (and feared) the danger to modernism that Duchamp posed,
and Danto
found in Warhol a simple way to recover a supposedly changeless order beneath the threatening scatter.

. . . 'Theories about art change; art itself is

changeless

' . . .

I suppose that is one kind of 'No-Theory theory'.

To go a step further:
Taking Wittgenstein as giving (esoterically) an account of 'universals', we could venture the guess that 'ideas' are, counterintuitively, the stable ones here, whereas materials are in a slow-but-constant state of change; hence Danto's heartily-avowed 'essentialism' seems also to commit him to an 'idealism' which he is not so anxious to cozy up to.

Whether or not that is right about either D or W . . . let's not overlook everything that is

threatening

in D's historicist Bounce House.

(Again, Our Man Here seems to be indulging in some Intentionalist criticism-of-philosophy.)

The odd fact is that
Greenberg and Danto
separately viewed the answer
to the one question each favored
(of the two just mentioned)—

Greenberg, the definitional question,
Danto, the metaphysical one

—as

dependent on a suitable theory

of the "social context"

in which art appears as art.

In a way,

they were on to

the same relational condition

(which they construed and applied in very different ways),

which
I may as well add
became the unexamined theme

(that is,
the "institution" known as the "art world"
or, more quarrelsomely,
the "artworld")

that George Dickie
allowed to shuttle somewhat aimlessly
between Greenberg's and Danto's questions

(roughly,
between definition and
social history or metaphysics).

All three

neglected

the analysis of

the "social context"

of art,

the "institutional" issue,
which
each approached from a different vantage
and which,
rightly understood,

threatened the autonomy

of the definitional question

and

confirmed the "Hegelian" cast

of the metaphysical one.

Dickie
may have been influenced in this regard
by
speculations about
the philosophical import
of
Duchamp's effect on the artworld
that
appeared in the early 1960s

(notably,
in articles published by Greenberg
and Joseph Kossuth),

when he himself
began to write about the artworld "institution."

Dickie
obviously aligned himself with Danto's theory,
which
he supposed he shared with Danto,
until
Danto made it quite clear
that
that was a mistake.³

---

³ For Dickie's account ...

"The Institutional Theory of Art," in Theories of Art Today, ed. Noël Carroll (University of Wisconsin Press, 2000), pp. 93-108.

See also Thierry de Duve, Kant after Duchamp (MIT Press, 1996), especially "The Monochrome and the Blank Canvas," ...

---

Dickie
viewed the institutional context
as
requiring very little
in the way of analysis
(philosophical or sociological):

it plainly counted
as
the immediate space of
philosophical and art-historical interest,

but

the complexities of its own role

remained

more or less unexamined

in the pursuit of
an acceptable definition.

If so, then notice that

pursuit of

a

definition

evidently does not force a reckoning with these

complexities

.

Evidently it is someone else's responsibility to point you in this direction; and if they point you in the wrong direction, nonetheless,

' . . . all errors which remain
are the author's sole responsibility . . .'

These are the most obvious grounds for thinking that the project of definition can never amount to much, even if it were (somehow) to be deemed 'successful': in the vast majority of art-instances, still we would have 'described' with our definition only the least important 'properties' of those instances; we still would lack any 'description' of everything more "complex" than the successful definition requires for its success.

Until then,
our Great Definers
can simply keep haranguing:

' . . . but you're not doing it right! . . .'

They can simply continue to insist that not enough has been

examined

.

In no other line of work is repeated failure so good for business!

[217]

Weitz
misses the urgency of the definitional question
altogether;

Greenberg
brings it too close to a reductio;

Danto
leads the metaphysical issue to
a related paradox—for,
Danto claimed,
art's status as art
could not be "descried" by perceptual means;

and
Dickie isolates
in an extremely weak way
the contextual condition
on which
both the definitional and metaphysical issues
depend—

since
he emphatically denies
that
the "institutional" matter
plays any substantive role
in
any would-be
realist or essentialist definition of art.

None of these commentators
addresses Duchamp's challenge straight on,
though Greenberg sees the threat.

For Dickie,
the "institution"
marks no more
than

the setting within whose boundaries
a true definition

(one that might even accommodate Duchamp)

seems close to discovery.

Here,
for the moment,
let me simply say
that

the definitional question
probably cannot be plausibly answered
without
bringing
the social-context or institutional issue

into accord

(in logical terms)
with Wittgenstein's discussion of "family resemblances"
(which Weitz misses)

and

with Duchamp's reception in the artworld
(which nearly all recent discussions
of the definition of art
either ignore
or implicitly discount).

    🤔   🤔   🤔    

I don't get this impression at all,
but then, I'm still not caught up on the most

recent discussions

.

Greenberg had already considered,
in 1962,
treating a blank canvas as a picture—
accordingly, as a painting;
accordingly, as art.

(Hе was already addressing Duchamp's challenge.)

He had already anticipated the idea
that
an object might be "art"
if duly "experienced" as a "рicture"
(in some unspecified sense,
again weakly read in terms of
the pertinent social context).

If Danto
had invoked Duchamp rather than Warhol,
he would have had to pursue
a close comparison
with
a singularly complex but profoundly problematic
cousin of a theory—Greenberg's.⁴

---

4. See Clement Greenberg, "After Abstract Expressionism," Art International 6 (1962): 24-32.

I rely here on de Duve's excellent account involving Duchamp, Greenberg, Kossuth, and others ...

---

It's a tad strange that Kosuth never gets a proper citation, and that his name is consistently misspelled.

The author is primarily but not solely responsible for this!

That
the philosophical import of Duchamp's readymades
had already been grasped
by the end of the 1950s
and the start of the 1960s

in a way that led directly to
Danto's "first" paper, "The Artworld"

—which introduces
some extremely problematic notions of its own
(the idea of the "artworld" itself,
for instance,
and
the "'is' of artistic identification"
pointedly contrasted with
the "'is' of identity")—

confirms the meaning
of
the great gap
between
Weitz's up-to-date use of Wittgenstein
and
the absence (in Weitz) of
any hint at all of
the deep challenge posed by
the spreading debate regarding Duchamp.

But Danto
views his own 1964 "Artworld" paper
(typographically adjusted, later, as
"The Art World")
as
"an immediate philosophical response to Brillo Bоx."

Warhol's installation
took place in the late spring of 1964,
in the Stable Gallery,
in New York,
and
"The Artworld" piece
was published in
the Journal of Philosophy
in October 1964—
in

a notably short interval
between submission and publication.

Danto was in fact, at this time, a newly-appointed editor of this journal.

Claude locates this corroboration, according to which "prompt publication" was used "as an incentive" following "an act of recklessness" by a prior editor, which had resulted in a terrible backlog of papers.

Along with Danto, the other incoming editors were his Columbia colleagues Sydney Morgenbesser and James J. Walsh.

All of this is

notabl

e indeed!

In any case,
Danto says very clearly
that he was struck
(immediately, it seems),
by
"the somewhat Kantian question"
of
"how it was possible"
that
Brillo Box was a work of art,

"inasmuch as an indiscernible object
could not have been an artwork
at any earlier moment."

A stunning provocation
and
a challenge very different
from that posed by
Duchamp's Fountain.

Is Margolis himself indulging in a bit of Dantoism here?

It certainly is interesting that the same

provocation

did not follow Fountain,
but this is no proof that it could not have.

Or, maybe he is trying to say that this "provocation" should have been possible forty years earlier, yet it was not forthcoming, and this shows either a lack of nerve, an obliviousness, etc.

Danto speaks here about
Jackson Pollock's presence in "the art world,"
which signals

Greenberg's efforts to define painting
(in the essentialist way)
in accord with
his (Greenberg's) modernist theory
(the "flatness" doctrine)
and
confrontation with Duchamp's challenge.⁶

---

⁶ There is a tangential connection, here, with a new flurry of challenges to Paul Oskar Kristeller's "The Modern System of the Arts," ...

See, for instance, James I. Porter, "Is Art Modern? Kristeller's 'Modern System of the Arts' Reconsidered," British Journal of Aesthetics 49 (2009): 1-24.

Porter touches on Greenberg's "Towards a Newer Laocoon," ...

But Porter does not address the Duchamp phenomenon or, indeed, the logic of the argument in favor of essentialist definitions.

(He claims to expose important inaccuracies in Kristeller's well-regarded paper, at the same time he supports, informally, the prospects of an accurate essentialist definition ... )

Porter is not uninteresting in what he says about the arts, but he has no fresh conception of what a definition is;

...

---

Danto also mentions
George Dickie's "institutional theory" of art,

which
he "misrepresents" (Dickie's term)
in
precisely the way Dickie claims he does,

which
Danto cannot have mentioned
without recalling Duchamp
(unmentioned in the "Artworld" paper),

whom
he links to the "artworld" question
and, thus,
to Greenberg and Kossuth,
who
address versions of the question of the artworld context
just prior to the appearance
of Danto's 1964 aсcount.⁷

---

⁷ ...

I cannot defend Dickie's actual theory: it does not work.

But Dickie does make a fair point against Danto and against Richard Wollheim, Steven Davies, and Noël Carroll in his search for the essential definition of art—a definition in terms of necessary and sufficient conditions.

---

Philosophical interest
centers on
the triple fact
that

Danto's theory of art
(in "The Artworld")
is not yet linked to a definition of art
but rather to
a prior "Kantian" question
that sets a condition on
the definition of art;

that
the question of that condition
is essentially the same
(under a different guise)
as the one Greenberg and Kossuth discuss earlier
(for example, regarding the "blank canvas"),
with
potentially more radical consequences,
including
the threatening self-defeat
of
approaching the definitional question
from the vantage each separately favors.

(As it happens,
both Danto and Greenberg are
drawn to strong
—very different—
essentialisms,
which
promise to elude
the vagaries of
the accelerating paradoxes of
the historical practices of avant-garde art.)

For his part,
Dickie does indeed
attempt to establish the relevance
of
the artworld "institution"—

or of

the "context" of

the artworld
—as the

encompassing logical space

within which

the correctly delineated

necessary and sufficient

[218]

conditions

of art
might finally be discovered.

In reading Danto's term
('the artworld')
thus,
Dickie believed he had
correctly grasped Danto's deeper intent:
that is,
to salvage the essence of art
from within the flux of
the history of the artworld.

And, in fact,
Dickie was right,
as
Danto's more recent, otherwise puzzling remarks
make clear.

The trouble is
that
Danto nowhere explains
the metaphysical role of the "artworld" concept,
which is
essential to the argument of the 1964 paper
and to
the invariant "real" definition of art
that Danto favors
in the latest phase of his own account.

Furthermore,
this entire line of speculation
draws Danto
back to Weitz's radical rejection
of the feasibility of defining art.

But, as we know,
Danto actually espouses
a particularly ardent essentialist definition
of both philosophy and art—

a definition,
frankly,
of a "platonist" sort
that is
somehow meant to fit
the profoundly historicized career
of art in our time.

(In this sense, Danto answers Weitz's challenge.)

I am convinced
that
a handful of distinctions
would greatly mitigate
the philosophical disorder
the summary just given
exposes:

we must, I think,
continue to pursue
the strangely important question
of
the definition and metaphysics of art.

I draw some courage here
from
a pair of matched books
(by Steven Davies and Noël Carroll)
that
span the run of analytic aesthetics
down to
the last decade of the twentieth century.

They effectively
begin with Weitz's challenge,
continue with Dickie's and Danto's
reactions to Weitz (and allied thinkers),
feature the import of Duchamp's intervention
more than Warhol's,
and finally
come to rely on
the instruction of an inquiry
that
keeps continuous track of its own evolving arguments.

The first decade of the new century
confirms
the strong persistence of Weitz's original provocation.

But
there is still
too much scatter
and
too little in the way of resolution
in Davies's and Carroll's accounts,
and also
in more recent contributions.

We must look again at

Weitz

and

Danto,

who,
after all,
raise

questions

that have yet to be answered.

Weitz has nothing to say
in "The Role of Theory in Aesthetics"
about

the metaphysics of art,

unless

you take him to have signaled
that he

subsumes such questions

under the terms of

the definitional issue.

Well,
how else to

take

him?

For Weitz believes (and says he believes)
that
a definition of art
that is
essentialist, realist, and
formulated in terms of necessary and sufficient conditions
is
quite impossible.¹⁰

---

¹⁰ The trouble is he never says exactly why this is so, beyond what may be drawn from the following claim:

aesthetic theory is a logically vain attempt to define what cannot be defined, to state the necessary and sufficient properties of that which has no necessary and sufficient properties, and to conceive the concept of art as closed when its very use reveals and demands its openness

(Weitz, "The Role of Theory in Aesthetics," p. 147).

---

It is on the basis
of
this rather isolated, relatively undefended dictum
that
Weitz proposes that

the definitional problem
does not begin with

the question,

"What is art?"

but rather with

the question,

"What sort of concept is 'art'?"

Apparently,
the second question
leads Weitz directly
to
Wittgenstein's notion of

things that
"form a family
with family resemblances
and no common trait."

I put it to you
that

Weitz has completely misunderstood

Wittgenstein's distinction
and
that
he has somehow managed
to

mislead the entire labor

of analytic aesthetics
thereby:

witness
the work of Davies and Carroll
and of more recent discussants
who
have followed the official verdicts
to the letter.

Of course,

it is not in the least clear

that "art" is

a concept that,
like "game,"

lacks necessary and sufficient properties,
or
exhibits no more than
"similarities overlapping and crisscrossing,"
or
has "no common trait"—

according to
what may be inferred from
Wittgenstein's own remarks,
beyond Weitz's reading.

Yes,
we have no

clear

ness

here.

We cannot.

. . . 'we'll never find a definition if we don't look' . . .

It can never be
perfectly "clear"
that

a

concept

is open,
only that it is not.

If and when a definition is found, the matter is settled; but no sheer quantity of resistance to definition, measured in time elapsed or effort spent, can settle the matter in the anti-theorist's favor.

'If there are necessary and sufficient conditions,
why can't we seem to say what they are?'

This is a mere contingency
which
proves ("clarifies") nothing.
Hence it is,
regrettably but necessarily,
an utterly inviable argument.

This is the asymmetry.

Did Wittegenstein say anything specifically about this? He identifies a 'craving for generality', but I presume that remark to be merely a means to an end. An evidential asymmetry is a different problem, and a much thornier one.

It is also not clear
that

art's being an "open" concept

(a concept
open to aрplication
to new kinds of instances,
as a result of artistic creativity)

logically precludes

art's eхhibiting any "common trait."

(Many have made the same point.)

And it is certainly not clear
—in fact the claim is plainly false—
that
there is (let us say)

a "game-world"

that
invites definitional and metaphysical questions

(about games)

in
the same straightforwardly cumulative way
in which
our acknowledging the complex "art world"
that

we all pretty well recognize

supports:

the scatter effected
by
merely admitting
solitaire,
ring o'roses, and
soccer
as different kinds of games
in common discourse

(that

plainly do not belong

to any

"gameworld" or "game world"

comparable to the "artworld" or "art world")

defeats the point
of
recognizing

the convergent sense

in which

(with Dickie and Danto,
but against Weitz)

we admit the "art world"
as

a palpable "institution"

(without prejudice
to
its own informalities and vaguenesses).

One or both of two things has happened here:

(1) Perhaps Our Guy Joe merely spoke too soon,
again (mis)taking contingencies for laws:

The

scatter

ed

art-phenomena were drawn into

cumulative

assemblage by contigent rather than essential forces;

but then, nowadays, belatedly as it were

(or
perhaps for Danto and Wölfflin,
right on schedule)

something similar is happening with

games

.

What is this 'something' that I claim has 'happened'? Games have become an immortality project (Rank) and a hero-system (Becker) for a large enough plurality of various 'societies'.

Gaming is not the least bit important . . . ,

" Except while it is thought to be; like the doctor's highly coloured medicine or the reserve ratio of the Bank of England. People believe it is important that the reserve ratio should not fall below 9 to 1, say; and because they believe this, it is important. "

(Wisdom, 'Metaphysics and Verification', p. 55)

The reasons for the varying 'importance' of gaming in different times and places is entirely contingent.

(2) Alternatively, perhaps when Joe contrasts the

metaphysical

scatter

of the

gameworld

with the

straightforwardly cumulative

metaphysics of the

artworld

he commits a Fallacy of Incomplete Evidence.

Furthermore,
Wittgenstein offers "game" as
an example of what he means by

the peculiar tolerance of actual usage

that

functions smoothly enough without
the presence of
a strong sense of encompassing context—

indeed,
without any effort at definition
at all;
he is not speaking of definition there.

Wittgenstein nowhere generalizes
in the way Weitz does.

On the contrary,
Wittgenstein is plainly open to
admitting a great many

[219]

different kinds of definitions
(which Weitz does not acknowledge)
within the scope of ordinary usage,
which
he does not collect in any systematic way;

Of course this proves nothing.

he is not even opposed to
imposing limitations
("for a special purpose")
on concepts as open as "game."

That would allow for definitions of many kinds—
even essentialist definitions
(if viable).

Who am I to dissent from such an established authority as Joe on a matter of such basic importance as this?

I think this is quite confused.

S

pecial purpose

is here distinguished from

ordinary usage

,

not

allow

ed for

within it.

To enter one is to exit the other.

Shaw nuff,
we do have a "special purpose" here,
namely,
to hammer out

the definition and metaphysics of art

.

Challenge met!

Weitz
and
nearly everyone involved with
the definition of art post-Weitz
have overlooked that option.

Here is the argument from silence again.

Alternatively, it could be that they simply did not want to be (did not consider themselves to be) pursuing a

limit

ed

special purpose

.

Charity and philosophical sense alike require that speculation about motives be firmly bracketed.

Within those brackets, I would speculate that no one walks around thinking that their "purpose" is merely "special", in the sense of being 'something less than universal'.

Articulating this kind of definition, then, requires the preliminary articulation of the something-no-one-admits: that their "purpose" is a mere subplot and they themselves a mere background character.

Clearly the principals here have, by now, shown themselves unable to rest content with just that.

(The situation is actually much worse even than all of this, because as soon as

the peculiar tolerance of actual usage

is established, the definers are thus deprived of a huge swath of "purpose" that they might otherwise have claimed. Has Joe yet said a single thing towards establishing the "importance" of his project that is not vitiated by his own reading of Wittgenstein?)

To put the lesson provocatively:
they have overlooked
the sense in which

Aristotle's definition of tragedy

easily accords

with Wittgenstein's remarks.

We cannot possibly know a priori
whether
"art" lends itself to essential definition.

In fact,

there is no essential definition of "definition," unless (once again) for a "special purpose"!

You cannot find
a clear account of this
in
Weitz,
Danto,
Dickie,
Carroll,
Davies, or
—most recently—Berys Gaut
(whom I shall come to shortly).

Also,

"essential" definitions

may be

merely nominal

as distinct from

"real" (or realist);

and
what Weitz might regard as
failed essential definitions of the realist kind
may,
in Wittgenstein's tolerant sense,
actually be successful in their own way
(that is, "for a special purpose").

Do we ever get to know what that

purpose

is?

This passage only feeds suspicion that the definition has come before the search:

the search, actually,
is a search for the "purpose" that would
save the definition;

a definition can

fail

fast,
but it can always be bailed out
by revisionist burnishing.

Here is one way to give the lie
to the we'll never know if we don't look argument:
always demand preregistration!
Such that the theorist
cannot look just anywhere
and then go to market
with whatever the cat drags in.

I am quite prepared to say
that Aristotle's definition of tragedy
is precisely of this kind.

Hence,
when, in a very recent introductory book of his
in aesthetics,
after making an excellent beginning
on
the distinction between definitions
and
theories
of art,

Stephen Davies insists that,
as against

a "theory of art"

or, perhaps better,

"a theory of art's value"

(which "should tell us why
the production and consumption of art
are

significant in human affairs")

,

"a definition

must be

exhaustive of all art

and

exclusive of all that is not art"

(including art that is poor in value),

I would say
he (Davies) makes a rather serious mistake.¹²

---

¹² Stephen Davies, The Philosophy of Art (Oxford: Blackwell, 2006), p. 45.

Davies offers the pretty point, in passing,
that

"there can be

theories of art that are not also definitions.

There is, after all,

the theory that art cannot be defined"

...

---

Davies agrees with Weitz about
what a proper definition requires
but does not agree with Weitz's verdict
about the futility of such efforts.

I'm sure Joe knows the literature better than anyone, but let's just point out, as politely as possible, that this sentence is so terribly unclear as to be highly misleading.

The

verdict

about the futility of such efforts

is that the hope of actually finding and settling upon a definition is "futile". But then there is Weitz's reclamation of the results of said "efforts", whereby he can hardly be said to conclude that these "efforts" have been entirely wasted.

He [Davies] also does not come to terms with
Wittgenstein's tolerance of (many) different kinds of definitions,
some decidedly unorthodox.

In effect,
he permits Weitz
to
define the very sense of the definitional question,
and
he does not see that
Aristotle's definition of tragedy
cannot be squared with his own dictum.

I can't figure out which

dictum

this is supposed to refer to.

Hence,
he fails to grasp the sense in which

a "defective" definition of the essentialist sort

(that he and Weitz would instantly recognize)
may, nevertheless, be

a complete success

in terms of

a reasonable reading of

its philosophical contribution.

Either this argument is extremely weak or else I have completely failed to understand it.

Why don't we ever get to know what the

contribution

is to be until after it has been made? We should have to say that part first.

This hints at
the skewed influence of
Weitz's misguided "Wittgensteinian" correction
down to our own day.

Davies is unguarded in insisting on
his "exhaustive and exclusive" conditions;

but the same extravagance
is implicit in Weitz and
a small army of others who oppose
certain kinds of definitions of art
or the definition of art tout court.

Hmm.

Weitz was responding to existing essentialist attempts.

The

extravagance

had already gotten started and run amok for quite a while.

Joe has explicitly conceded the "extravagance" of the project Weitz was responding to, so he cannot mean to defend Weitz's targets.

The crux of the matter seems to be

the skewed influence

of Weitz's argument, i.e., its deployment against later definers whose ends are (for Joe) more modest, reasonable, and constructive than are any of Weitz's original targets. That could be a convincing rejoinder if we yet had any earthly idea what this more recent cohort is actually trying to do. But we seem to come up short of that each time it arises. It seems we just have to wait for them to publish, and then we have to scramble (reactively) to try to figure out what their ends are and have been, and whether their means can serve.

This leads directly to "disjunctive" accounts,
whether definitional or "cluster context construals"
(to speak with Berys Gaut
and Robert Stecker and Davies himself).

I get the sense (not for the first time) that Joe thinks we should land

in con

juncti

on

anywhere

a reasonable reading

has been made; never mind what is being "read"!

This is vaguely consonant with W's 'Nothing is hidden', and 'not solve but dissolve'.

On the other hand, those of W's arguments which culminate in positing a 'form of life' lay the groundwork for an arch-skepticial line against all such phenomenal consensus and for understanding all such mentalistic artifacts as spurious.

Agreement does not exist prepredicatively or preconceptually. Rather, it is a 'pattern', constructed by way of that unholy collaboration between automatic perceptual filtering and intellectually-motivated interest, whereby an exceedingly tiny portion of everything that could

reasonabl

y

be noticed actually is noticed . . . including the (social) fact of 'agreement' itself!

Maybe I should be more hesitant to class written language as 'phenomenal' . . . or maybe the situation there is ten times worse!

Weitz has misread his Wittgenstein;
worse still,
what he says about "family resemblances" and "open concepts"

finds no explicit support

at all in Wittgenstein.

C'mon Joe,

e

xplicit

is quite the weasel-word when we're dealing with a writer whose intent is as hotly and perpetually disputed as W.

He has somehow managed to transform his rejection of "realist" and "essentialist" definitions (of "art")
into
a generalized, off-the-wall summary of Wittgenstein's more modest observation
that
some concepts ("game," for instance)
work well enough
in terms of some version of family resemblances.

(This is actually a very subtle observation,
and very widely misunderstood.)

It may well be too

subtle

for me.

The hodgepodge and self-contradiction
of Weitz's account
are clear enough
from the following claim and concession:

A concept is open if its conditions of application are emendable and corrigible; i.e., if a situation or case can be imagined or secured which would call for some sort of decision on our part to extend the use of the concept to cover this, or to close the concept and invent a new one to deal with the new case and its new property. If necessary and sufficient conditions for application of a concept can be stated, the concept is a closed one. But this can happen only in logic or mathematics where empirically descriptive and normative concepts are constructed and completely defined. It cannot occur with empirically descriptive and normative concepts unless we arbitrarily close them by stipulating the ranges of their uses. . . . Of course there are legitimate and serviceable closed concepts in art. But these are always those whose boundaries of conditions make it logically impossible to ensure any set of defining properties.

Here, Weitz
gives back with one hand
what he takes away with the other.

Hmm.

Ya gotta say, tho,
he gets quite a lot done, quite efficiently,
just by invoking

uses

.

Use is not to be

stipulate

d.

That's the whole thing right there!

Wittgenstein
says nothing about any of this
that could possibly support
what Weitz ventures;

How's this?

Life is

open

;

only our own

construct

ions are

closed

.

and what Weitz offers is nowhere defended.

It all hangs on
Weitz's animus against "real" definitions.

But
"real" definitions need not be "essentialist,"
need not be cast in terms of
"necessary and

[220]

sufficient" conditions,
and
need not be defeated by
a concept's functioning in terms of
"family resemblances" or anything of the kind.

(Look more closely at vWittgenstein's discussion in Philosophical Investigations,
roughly at §§60-77.)

I would say that
a definition of any ordinarily admissible kind
(Wittgenstein is open to
definitions of almost any degree of informality)
would be "realist"
if
it claimed to

address the "nature" of anything that could be found in the world.

When Wittgenstein introduces the concept "game,"
he makes it clear that
he is speaking of the various "proceedings"
(Vorgänge)
"that we call 'games.'"

He is thinking of our ability to

use terms informally,

without fixed boundaries,

and

without fixing the context

of their use.

He actually considers
the bearing of all this
on the definition of terms,
even of "game."

And he does not say, as Weitz says,
that
where there are "family resemblances"
there are no "common features";

he says, rather, that
when we pass from board games to card games,

"you find
many correspondences with the first group,
but
many common features droр out,
and others appear."

He also does not say that,
for such concepts as "games"
(and, I should add, "works of art"),
there

cannot be

any boundaries;

he says only that

"we do not know the boundaries

[of such concepts]

because none have been drawn.

To repeat, we can draw a boundary—

for a special purpose.

Does it take that

to make the concept usable?

Not at all!

(Except for that special purpose.)"

Clearly,
Wittgenstein does not support Weitz in any of his severe claims against "real" definitions.

Wittgenstein has no a priori objection against specialized definitions at all.

His main concern
is to explain
the sense in which ordinary usage
is not an "approximation"
to the perfection of any "ideal" language
(à la Frege),

only that

"where there is sense

there must be perfect order.

So

there must be perfect order

in the vaguest sentence."

Following the logic of Wittgenstein's argument,
it can be seen that
a "real" definition need not be
exceptionless,
essentialist,
cast in necessary and sufficient terms,
free of vagueness,
"exhaustive and exclusive" while ranging over "all" cases,
or bounded in any ideal way.

It needs only to be "usable" in the way ordinary usage tolerates, even where it is introduced "for a special purpose."

Well, this part is clearly bunk

Evidently here we have kept "use" as the standard;
but now allowing for

terms

to be

introduced

with impunity,
and with only

sense

to guide us as to what the result actually has been.

But it seems that if we continue to use the same word then "sense" cannot fail to materialize. It is as if I were to define art as a certain concealing maneuver, based on the "special purpose" I once had as a TSA screener and the "stipulated definition" of artful concealment in which I was instructed during my training. I could then refer to the "art" of certain musicians in managing to fool others into thinking they're much better than they are, thereby 'concealing' their deficiencies; and because this is bound to involve deploying "art" in an entirely grammatical way, my stipulative re-definition has won the game, to the extent permitted by law. I could then turn around and define one's "art" as one's left buttock, and everything would continue to turn out alright for me: I am still deploying the word "art" grammatically and with the same syntax as always before; yet apparently I mean something entirely different by it.

Furthermore,
nothing that Wittgenstein says in these regards

settles the determinacy

of the predicates we use in any of these ways.

It is a mistake to think, for example,
that
where we have only "strands of similarity,"
the "strands" are either

more or less determinate

than they are when we have
necessary, sufficient, or necessary and sufficient conditions.

It is also a mistake,
if I read Wittgenstein correctly

—and it

remains

a profound mistake,

apart

from Wittgenstein—

to speak of
the determinacy of predicates
when separated from
their pertinent, particular exemplars,
which
may be changeably and variously selected.

Who are the

exemplars

here?

Artworks?

But then, the current class of 'artworks' simply ceases to be

pertinent

once I define one's buttock as one's 'art', for the "special purpose" of microaggressing against fat people?

Then I remark that Mary spun her true masterpiece in late middle age, right before her husband left her for a 20 year-old fencepost, and after which her work was never quite the same?

Broadly speaking, predicates
applied to the real world
in ordinary usage
are of the same gauge everywhere
or
are capable of being made
more or less precise in the same way.¹⁶

---

¹⁶ This, I should add, is the fatal error of Renford Bambrough, "Universals and Family Resemblances," ...

---

The generality of predicates
is a matter very different
from that of
the variety of kinds of definition.

(Predicative similarity,
I would say,
is not a matter of
discerning "real generals"
as opposed to
nominalist convention;

it is tethered
rather
to the "general" use of a given predicate
ranging over a run of would-be similarities.

This is an account shared by George Berkeley and the pragmatists—and is very possibly

the best we have.

🤨   🤨   🤨

But its defense would take us much too far afield.)

The point of all this
is that
Wittgenstein is attempting to free us
from
the utopian constraints of early analytic philosophy favored by Frege, Russell, Wittgenstein himself
(in the Tractatus),
and
the self-entrapping mistakes of dependent movements like those of logical positivism—

false forms of rigor that deform philosophy.

(He could easily have considered
the use of
freeing us from Weitz's late addition
and

the dishearteningly easy slide of analytic aesthetics from then till now,

whether in agreement with or in opposition to Weitz.)

That's a clever juxtaposition,
but the charge doesn't stick.

Weitz too is encountering

false forms of rigor

and staking out a contrary position, and you can't say that his solution doesn't also

free

us from

the Moloch du jour.

Joe seems to be saying: it "frees" us from too much: from "false rigor" but also from the 'true' kind. But still he has not said what his "purpose" is, only that certain prevailing assumptions about it can be scrutinized.

The more talk without action, the more all of this indeed "skews" toward the "false".

I venture to say
that
the key to understanding this part of Wittgenstein's lesson appears at Investigations §77,
where
Wittgenstein mentions one's "look[ing] for definitions corresponding to our concepts in aesthetics or ethics."

There, he says,

"Anything-and nothing-is right":

"In such a difficulty
always ask yourself:
How did we learn the meaning of this word
('good' for instance):
From what sort of examples?
In what language-games?
Then it will be easier for you to see that the word must have a family of meanings."

I think we are being led here
to see that
to understand the notion of "family resemblances"
linked to ordinary usage,

we must return to Wittgenstein's own beginning,
reflecting (in reviewing Augustine) on
how children learn a first language.

(This is what I was flagging earlier by noting Wittgenstein's deliberately informal term Vorgänge ("proceedings").

For,
if I understand

[221]

him aright,
Wittgenstein is drawing attention
to
the fact that children learn the "correct" use of terms
(according to the usage they are being introduced to)
in a way in which

context is never really spelled out

and in which

terms never quite amount to concepts.

This is the key, for instance, to
the fatal weakness of Berys Gaut's bold attempt
to replace definition
with
what he calls the "cluster concept" account.

It would not be unreasonable
to view Gaut's effort as
the most salient innovation in recent Anglo-American efforts
to apply Weitz's (mistaken) reading of Wittgenstein's remarks about definition
in support of an "antiessentialist" treatment of the concept of a work of art.¹⁷

---

17. See Berys Gaut, "'Art' as a Cluster Concept," in Noël Carroll, ed., Theories of Art Today, ...

...

Gaut's "cluster concept" account closely resembles views advanced by P. F. Strawson and John Searle in other settings, which are convincingly rejected by Saul Kripke in Naming and Necessity (Harvard University Press, 1980), Lecture I.

...

---

Gaut views his own effort
as
building on Weitz's reading of Wittgenstein.

But
he scants Wittgenstein's argument
(as does Weitz)
in several regards:

first, because
he makes no allowance for Wittgenstein's accommodation of a great many different forms of definition;

second, because
he nowhere explains the sense in which his cluster concept is genuinely serviceable as a replacement for a definition;

and, third, because
in accord with the conditions adduced,
the cluster concept
cannot exceed the sense of
that use of words
that Wittgenstein explains
in reviewing Augustine's account
of how children are
just "trained" to grasp the "right" use of words
before they are introduced to
"criteria,"
"context," or
"definition."

In short, the cluster concept
cannot invoke criteria of usage
without itself becoming
a candidate definition
or
counting as an insufficient step
in proposing a definition or criterion of use;

or
where it foregoes criteria
(in Wittgenstein's sense),
it could not be put to the use Gaut intends.

Since it would lack a "grammar,"
it would count as "usage"
only in the weak sense suited to "training" children.

Gaut has trapped himself in an insuperable paradox.

To understand the point of this
is
to begin to see
where the whole definitional effort
(regarding art)
has gone wrong.

When Wittgenstein speaks of а family of meanings,"
he means (I suggest) that
we must try to guess or invent
(as children do)
what the putative "family"
(or the common referential "context") is
regarding
which of the scattered application of terms
might lead us to suppose that
they also function as concepts
that yield "criteria" of use.

As far as I know,
Wittgenstein does not draw the obvious conclusion,
though
it would fit his characteristic informalities
to a T:

that is,
that

the "realist" standing

of a definition
is a question entirely separate
from
the question of

its logical form.

There need be
no uniquely correct logical form
fitting all would-be definitions,

there need be
no minimal logical form
that
all proper realist definitions must satisfy,

and
the realist standing of any claim
need not ever be
a question of mere logical form.

Apply all this to the fluencies of ordinary usage:
none of Weitz's warnings about
what cannot be defined
will stand.

For example,
"open concepts" (art, let us say)
can be defined
under closure

("for a special purpose"),

can be defined by invoking
necessary and sufficient conditions

(in all sorts of

deliberately restricted ways),

can be defined

essentially,

and
can be defined

in realist terms—

where
doing any of these things
need not entail
doing any of the other things mentioned.

When Aristotle
defines tragedy

in accord with his own preference

for
Sophocles' Antigone and Oedipus,

he is aware

that
his definition

cannot fit

specimens drawn from
Aeschylus and Euripides;

and when
he casts his definition
in an "essentialist" mode
(as he surely does),
he is

under no obligation

to treat the conditions he supplies
as

realist

(or exclusionary)
rather than

nominal

(and aptly invoked among possible options),

regardless of his own philosophical commitment.

Indeed,
Aristotle is not obliged to treat
tragedies that do not fit his definition
as not genuine
in the way of being tragedies at all.

His definition

serves to isolate

a particularly

attractive and instructive model

of Greek tragedy

—which has been favored ever since—

not so much because
he has found the natural joints
(the logically essential distinctions)
of the concept
but because
its formulation

illuminates

in a notably arresting way

the salient history

of
what has come to be regarded as
the continuation of
the inclusive genre
among newer species.

favored ever since

As if this proved anything!

salient history

More terrible weasel words!

Once again, by the end of the paragraph Weitz has been reaffirmed rather than disaffirmed. Why does that seem to keep happening?

Weitz:

"If we take the aesthetic theories literally, as we have seen, they all fail; but if we reconstrue them, in terms of their function and point, as serious and argued-for recommendations to concentrate on certain criteria of excellence in art, we shall see that aesthetic theory is far from worthless. Indeed, it becomes as central as anything in aesthetics, in our understanding of art, for it teaches us what to look for and how to look at it in art."

Perhaps Joe is at pains to thread the needle
between
having a

preference

and
making

recommendations.

This is a philosopher's curse for which there is no curative elixir.

Evidently Kant at least tried to exorcise it:

"Many things may have for him charm and agreeableness—no one troubles himself about that; but if he declares something to be beautiful, then he expects the very same satisfaction of others; he judges not merely for himself, but for everyone, and speaks of beauty as if it were a property of things.

Short of having any curative effect (aside from eliciting a chorus of dissent, if that might be counted as therapeutic on grounds of the hydraulic metaphor of the psyche), this at least marks out the ways in which the matter is susceptible to empirical inquiry.

The Nietzschean streak in Becker's books makes a start towards building such an empirical case: morality is born in

satisfaction

and reared to maturity on grievance against its enemies.

Per Becker, we might actually have to reverse Kant's order of operations: we so expect the same 'satisfaction' of others as of ourselves that our very choice of idols must be made with an eye toward this from the start. Nothing offends the self-esteem quite like the presence of Others who gain their satisfaction differently. But if we simply give ourselves over to whatever the going thing happens to be, then we have avoided the trouble before it can get going.

If so, then the phenomenon Kant describes, in the order he describes it, is best understood as the aftermath; that is, of being 'arrested' by beauty, rather than of (first) 'judging' an object and (only then) 'finding' it to be beautiful.

That is, by 'arrested' I mean to suggest not only that 'the will' is in some sense preempted, bypassed, or at least concurrently challenged by 'prepredicative' or 'preconceptual' input, but so too (perhaps as one part of this) is the deeply transactional and performative problem of self-other identification elided. Some beauty 'grabs' us before our Mind's Eye has had a chance to locate it on the social chess board. And then we're in a tough spot! Now we have a problem, and the problem has to be either with others or with us ourselves; and then, there's no real mystery about how that sort of inner conflict will eventually be resolved. The only question is just how far each person will attempt to carry their own 'resolution'.

Beyond all of that . . . I would hate to be the one to resort to eco-intentionalism here, but it's hard to resist speculating that perhaps the mere gesture of writing down a definition, refining it, publishing it, and subsequently defending it is a gesture which is quite readily understood and which really needs no interpretation. The notion that one would nevertheless continue to accept any and all negative exemplars as

genuine

members of the category is a notion that doesn't pass the smell test; as if the Goth or the Hippie really was in the bag for alternative lifestyles in general.

Nor does passing the buck to

the continuation

testify to any such broad disinterestedness; if anything it further gives the game away by raising the stakes of the initial exclusion, i.e. by carrying them forward openendedly. Neither plebeian 'nostalgia' nor 'pedantic necrophilia' are particularly well-rewarded. The future-oriented 'resolutions' are the really ambitious ones.

I see no reason
why one must choose,
disjunctively,
between Aristotle and Nietzsche,
for instance.

Nietzsche
is more attracted to
the possibilities of psychological pathology,
hence to
the threat of irrationality
at the heart of the rational,
hence to
Euripides
as a kind of companion to Socrates.

The accounts of both
are

"realist" and "essentialist,"

yet

profoundly provisional,

"open" to revision,
supplementation,

endless reinterpretation,

despite

the historical (extensional) closure

of authentic Greek tragedy.

Both succeed
and
we embrace both.

How bizarre that all along this paper has been building towards a paean to

endless reinterpretation

.

Who is clamoring for more of that?

Everything I read in this field I read first and foremost through the lens of whether it makes the world more or less safe for critics and criticism. Add to the safe-making maneuvers: the notion that

accounts

may be both

essentialist

and

provisional

.

That's a new one to me!

Importantly, the version given above seems not to partake of the (Dantoist) premise that the institution's "essential" nature is revealed in and through 'history'. It could be read, actually, as asserting the wholly contrary notion that each historical art-artifact is, in and of itself, "endlessly" revealing of art's essence, awaiting only the never-ending efforts of theorists and interpreters to make endless revelations, always moving halfway toward

closure

without ever arriving there.

Against Weitz,
we might put the lesson
in Wittgensteinian terms:

the rigors of definition
are,

[222]

finally,

ordinary liberties taken within the flux of linguistic usage.

But if so, then
Weitz utterly misreads Wittgenstein
and disallows
(for no compelling reason)
definition's options,
when he says:

The primary task of aesthetics is not to seek a theory but to elucidate the concept of art. Specifically, it is to describe the condition under which we employ the concept correctly. Definition, reconstruction, patterns of analysis are out of place here since they distort and add nothing to our understanding of art. What, then, is the logic of "X is a work of art"?

The answer to Weitz's question stares you in the face:

Hmm.

I have barely scratched the surface of Wittgenstein exegesis, but tbh the admonition

"to describe the condition
under which
we employ the concept correctly"

does strike me as an accurate reading. We might want to change "correctly" to "grammatically". Either way, I have been assuming that this is the answer to the question: 'What was Wittgenstein's view of (so-called) concepts?' If I've

misread

something here, then I suppose that would explain why I don't understand Joe's harshest criticisms.

Anyway . . .

preeminently,
the philosophical definition of art
(in all its informal diversity)
provides
a memorable sense of
a kind of
"open-ended convergence" on the concept,
deliberately fitted to a set of
important,
strategic,
relatively systematic
claims about the arts writ large,
meant (dialectically) to test
the relative strength and adequacy
(in a variety of senses)
of all contending alternative conceptions
in
that context of reference
that
we signal as relevant
to
the arguments we provisionally invite.

. . . "What, that was not

star

ing you

right

in the face

?

What's wrong with you? . . .

The convergence and divergence of definitions
makes sense only in some such setting;

in fact,
it is the same sort of setting
(minus the insistence on "correctness")
that Weitz himself envisages,
except that (for some reason)
Weitz refuses to admit

the profound informality of ordinary usage—in Wittgenstein's sense.

(It is only that
Weitz finds fixity
in
another part of the same woods.)

Okay, so the word "correctly" is in fact doing all the work, for better and worse.

I'm just not seeing any such

refus

al

here. 'Informal' does not mean 'boundless' or 'anarchic'. If Joe means either of these, he should say so.

Failing that, the informality is a process, not a result.
The point is: look at the results.

There is some

fixity

in the results!

A theorist who is

under no obligation
to treat the conditions he supplies
as realist

is not to be trusted.

How else to take Wittgenstein?

The essential key lies elsewhere,
in
the articulation and championing
of
a particular theory of art:

the appropriate logical constraints

on definitions we recommend
are no more than

a function of the nature of the claims being advanced.

Well okay, wha bout:
the accusation that "we gibber"
in absence of
an essential, timeless definition?

The

function of

that

claim

just is
to impose
'necessary and jointly sufficient conditions'
as the topline

constraint

.

If we eagerly agree that such enterprises (such constraints) are fruitless, then once again it seems we are actually reaffirming Weitz.

There are no a priori requirements on definition,
except what we choose to confer.

There is the point of Wittgenstein's speaking of games exhibiting "family resemblances":

"'games' form a family,"
he says.

He does not mean
that
we can (first)
identify the family
that includes different kinds of games

and then
identify the "strands of similarity"
that
that family manifests.

No,
he means that,

in the entire range of ordinary usage,
we do not have to
rely on
any determinate context
defined in terms of actual families;

but
we can still
usefully draw an analogy
to
looser sorts of "strands of similarity"
that
we would inevitably have to consider
if
we sought to impose
any logically stricter sort of definition
on 'game'²⁰

---

²⁰ This may be a helpful way of accommodating the point Maurice Mandelbaum makes in his "Family Resemblances and Generalization Concerning the Arts," ... Mandelbaum pursues the matter in an entirely different way.

---

How else
could one possibly understand
what Wittgenstein means by "definition"
in the following remark
(concerning "seeing what is common"):

Suppose I show someone various multi-colored pictures, and say: "The color you see in all these is called 'yellow ochre."—This is a definition, and the other will get to understand it by looking for and seeing what is common to the pictures. Then he can look at, can point to, the common thing.

Here, Wittgenstein settles
the informality and diversity of
the use of the term
'definition';

but then
he replaces
the problem of definition
with
the problem of the determinability of what is common among the instantiations of a predicate
("similarity in general"),
which
cannot be determinately specified
among the instances of 'yellow ochre'—

that is,
"looked at,"
"pointed to,"
in an entirely different sense
that,
let it be noted,
defeats Frege and nearly the whole chute of analytic definitions of 'art.'²²

---

²² This line of reasoning poses a serious question, for instance, about Danto's sources of assurance supporting his own well-known essentialist definition of art. See Arthur C. Danto, "From Aesthetics to Art Criticism," in p. 95; "Responses and Replies," in Danto and His Critics, ed. Mark Rollins (Oxford: Blackwell, 1993), pp. 193-216, p. 194.

---

JOSEPH MARGOLIS
Department of Philosophy
Temple University
Philadelphia, Pennsylvania 19122

---