Carroll—A contemporary introduction—Chapter 5—Part II

Noël Carroll
Philosophy of Art: A contemporary introduction
(1999)

 

[SK's vitriol]

 

---

[224]

Part II
Two contemporary
definitions of art

The Institutional Theory of Art

...

[225]

...

Institutional Theorists like [George] Dickie were

impressed

by a certain criticism of the family resemblance method ... Suggested by Maurice Mandelbaum,

the objection

scrutinizes

the central metaphor of Neo-Wittgensteinians

—family resemblance—

and finds it wanting.

...   Real family resemblances

 

But what about metaphorical ones?

 

... are not mere resemblances ... Even if my eyes are exactly the same color as Gregory Peck's, they do not bear a family resemblance to Gregory Peck's for the straightforward reason that, no matter how great the similarity between my eyes and Peck's, we are not members of the same family ...

 

Sooo . . .

We simply cannot 'look related' to someone
unless we are actually related to them?

The whole notion
of
looking related
but
being unrelated
is
somehow
incoherent?

What is incoherent is to call something a "metaphor" while demanding that it be taken literally. That is a non sequitur.

Or, if it never was just a "metaphor," then it cannot be called a metaphor.

 

For resemblances to be

genuine family resemblances,

there must be

some underlying mechanism

—such as genetic inheritance.

 

Okay, sure. But . . . what about a "metaphor" whose primary semantic displacement just is this difference between "resemblance" as a real and verifiable product of "some mechanism," and "resemblance" as . . . literal resemblance between two objects, with or without any genetic link?

The Wittgensteinianness of it all is subtler even than this, I think. It is subtler by far than conventional "metaphor."

First and foremost, categorization may be unprincipled in precisely this way. That is to say, in Our Man's idealistic parlance: our practices of 'categorization' may not reflect genuineness. What do they 'reflect'? Something else.

How could that be? Our Man fears that our practices would cave in (208) were there no principle underlying them: a practice which depended upon unprincipled categorization would quickly come to 'reflect' this unprincipledness, namely by "caving in." But the 'family resemblance' trope suggests not only the possibility that categorization can be unprincipled, but also the possibility that practice does not inevitably collapse simply for want of principled conceptual categories. This is the really radical suggestion, at least against the backdrop of Our Man's professional quietism: we can more readily know principle than we can practice in a principled manner.

But then, what is categorization, if not the practice of a principle? It's the practice of . . . something else. There's something else going on here, something which can (so far) only be suggested, hinted at, by such peculiar (metaphorical??) combinations of context (say, games) with turns of phrase ("family resemblance").

This is not my bailywick, but I'll just say that I'm far more impressed by the Wittgensteinian serve than by the Institutionalist volley.

---

In the Peck hypothetical Our Man certainly is not depending on resemblance to evaluate the question of relatedness. What is he depending on? He is depending on the plain fact of his un-relatedness to Peck, knowledge of which he seems already to possess. i.e. He shares with us only the fact of his knowledge and not the basis for it.

If what was really at issue here was the fact of his (un)relatedness to Peck, we would have to say that he assumes the fact he intends to prove. Ostensibly this is not a question of any importance to anyone concerned. Instead, we are concerned with the Neo-Wittgensteinian  central metaphor of family resemblance. The funny thing is: it is Our Man himself who, purportedly along with Mandelbaum, insists upon de-literalizing the "metaphor" such that the bare fact of genetic descent is decisive. To each his own! But then, this just refers the question of "family resemblance" to the question of genetic descent. What's the point of that?

 

If a similarity is to count as a family resemblance, it must have been generated in the appropriate manner. ...

...   once we realize that [the Neo-Wittgensteinians] are not talking about

family resemblances,

but only

mere resemblances,

then ... their so-called method will have the "here-comes-everything" effect. But ... this consequence is obscured [when "mere resemblances" are called "family resemblances"] for the reason that genuine family resemblances really are selective ... Family resemblances are condition-governed ... the similarity must obtain between people who meet the necessary condition that they belong to the same gene pool.

Thus, either Neo-Wittgensteinians were really talking about family resemblances, in which case they were presupposing necessary conditions

[226]

for art status, or they were talking about mere resemblance, in which case their approach was not selective enough, since everything resembles everything else in some respect. ...

...   The Neo-Wittgensteinian advised us to look for family resemblances. He misspoke, because family resemblances

are not something that you can determine

simply by looking.

Family resemblances also depend on something that you cannot see—genetic inheritance. ... This is what the Neo-Wittgensteinian ignored,

 

Knowingly ignored, I think!

 

but, at the same time, it provides a clue for how we identify artworks, namely that

the genesis of a candidate

is crucial for its art status.

We do not classify people as members of the same family because of the way they look.

 

But we DO (often!) find that people 'look related,' whether or not we have any basis for thinking (or not!) that they actually are related.

This is problematic only if we are asking this finding to do something for us which, it is true, it cannot actually do: to confirm genetic inheritance simply by looking. But that is precisely the point: Practice is not kind enough as to be so neat and orderly; Rather, it is (metaphorically, at least!) shaped "simply by looking," and this invites all manner of white lies to enter the fold.

 

Two people may exhibit or manifest the same properties, but we do not say they belong to the same family on that basis.

 

But we often do, on just that basis and nothing more, ask them: 'Are you two related?' And we do not arrive here after desperately searching for some respect in which everything resembles everything else. More likely, we weren't 'looking' for resemblance at all, and yet we did 'see' it.

 

... We will call two people sisters, even if they do not look alike, if they are genetically linked in the right way.

 

This is immaterial.

 

Nonmanifest properties—genetic properties—are what make for family membership, not perceptible resemblances that, as the Neo-Wittgensteinian says, you can detect by looking and seeing (or listening and hearing). That art status might rely as well on a candidate's possession of certain nonmanifest or unexhibited properties is the clue that Institutional Theorists of Art learnt from Mandelbaum's criticism of the Neo-Wittgensteinian's misuse of the notion of family resemblance.

 

It's a pregnant conjecture, pun intended, but it seems to be based on a ghastly misreading of a whole cadre of eminent Philosophers, many of whom were personally acquainted with the OG. They well-and-truly misused this concept of his? That is Somethin' Else!

 

Of course, artworks are not the product of genes. They originate socially, not biologically.

 

🤮 🤮 🤮

 

They are generated within a social context wherein the activities of the artist and the audience are co-ordinated by certain underlying social rules.

 

Well . . . sure, we end up formulating them as (and calling them) rules if we're not the ones occupying the social context thusly implicated. Generally the actual people actually occupying that context are not subjectively experiencing co-ordination by rule. They didn't learn the rules explicitly. I learned what a strikeout and a walk are in a totally different manner from how I learned to size up a pitcher or a hitter. All of my behavior on the diamond could be explained as adherence to rule, but only some of it deserves to be labeled that way.

This matter of the outward appearance of "rules" as against the internal obliviousness to them as such seems very much of a piece with the broader thesis that "our practices" are held together (if they are) by something other than principle. I thought that was what Wittgenstein was concerned to show?

 

Being an artwork is a function of certain social relations. These relations are not something that the artwork wears on its face; they are not something that one can detect by simply looking. They are nonmanifest or unexhibited social relations.

 

Then how do we know that they're real? Because some other social epiphenomena, e.g. an artwork, is itself a function of these relations?

That would mean that the artwork just is itself a/the "manifestation." Nonmanifest would then be a misnomer.

Unexhibited is a more promising descriptor in the present context, but I think that in practice it too proves misleading on a crucial point: The "underlying social relations" are "exhibited" profligately to certain people, but not to everyone; Certain people know everything about the genesis of a candidate while others know nothing.

This time there's nothing 'implicit' or conceptually anarchic: whoever learns of the "unexhibited social relations" learns of them explicitly. These features can be detected by simply looking; it is more that most people are permitted to look only at the artwork and are not permitted to look at the relations.

 

Thus, the Institutional Theorist suggests that if we can isolate those features of the social context of the practice in virtue of which a candidate for art status is deemed an artwork, we will have a way of sorting art from nonart.

 

In order to do this it seems we would need to locate the phantom manifestations, apparently not by simply looking at the work but by . . . doing what, exactly?

We cannot simply count the artwork itself as evidence of the features, since this would mean assuming we already have the answer we are seeking.

I need a timeout to plug some holes.

 

---

 

'Family Resemblance' (Wikipedia)

" things which could be thought to be connected by one essential common feature may in fact be connected by a series of overlapping similarities, where no one feature is common to all of the things. "

Our Man protests:

But THERE IS one essential common feature shared by the members of a 'family,' i.e., some underlying mechanism—such as genetic inheritance.

The Wittgensteinian seems not to be talking about anything "underlying." They seem rather to be honing in (quite specifically!) on everything that one can detect by simply looking.

WHY take that tack?

Why ignore "underlying mechanisms" when trying to understand "our practices"?

Obvs, it's because we are always ignorant of something "underlying" our perceptions and actions. Even the most advanced sciences proffer chains of proximate causes. They do not proffer ultimate causes. The very notion of an ultimate cause actually is nonsensical, an analytic nonstarter. Each proximate explanation gained brings the promise of pragmatic advance, but not of analytic advance. (Bambrough's paper, below, points decisively in this direction, I think.)

The charge that an analyst of some "concept" has failed to consider something "underlying" the practice in which that concept is implicated is a charge to which all analysts are vulnerable. All of us must be overlooking something causal, something which lies beyond our own proximate horizons. Presumably there are myriad conceptual 'standpoints' from which to conduct 'analysis,' each with its own 'promises' and its own 'horizons'; but wherever we go, we draw no closer to ultimacy.

Of course "resemblance" is a rightly infamous example of this problem, not just because everything resembles something else in some respect, but also because things may resemble each other in myriad respects. Hence if any particular axis of resemblance is to be fixated upon, analyzed to death, elevated to salience, etc., this occurs in relation to some already-existing purpose. Here that 'purpose' is something like: 'categorization.'

The distinction between the bleeding edge of the proximate chain and the undifferentiated middle still is an important distinction . . . pragmatically, that is. If you really need to get something done, then recent science, on the whole, is likely to be more useful to you than is distant science. On the other hand, sometimes outmoded paradigms still 'work' just fine for the same mundane purposes around which they initially formed. Newton's mechanics has been given as an example: there do exist, now, myriad purposes for which it is entirely sufficient, alongside plenty of others for which it is insufficient.

This distinction of purposes is the distinction worth making here. The charge that the contemporary Newtonian Mechanist neglects to avail themselves of quantum mechanics can only be referred to that person's manifest purposes qua mechanics. It is not a charge that can be referred to philosopher for a vetting of its conceptual analyticity. It also is a charge to which the professional Quantum Mechanist is every bit as vulnerable as is the naive empiricist: both are ignorant of myriad "underlying mechanisms." That ignorance is not what distinguishes them but rather what they have in common. Both are able to achieve pragmatic gains within the confines of their knowledge and practice. It's not even that the Newtonian, personally, is ignorant of Quantum theory. The point is that he is not using it. We cannot hope to understand his "practice," then, by appeal to concepts which are not operationalized in it.

Not even Our Man Here, in his optimism that "our practices" make sense, has contended that those practices proceed from ultimacy, nor even that they proceed from anything particularly scientific. So, whether any given practice makes sense or not, we can now see why the practice of categorization not merely permits but actually demands to be analyzed as if people did not always know what they were 'really' looking at. For most of human history people certainly did not know about "genetic inheritance"; or at least the proximate chain they were able to string out was only two or three crooked links long. If they had a concept of "family resemblance," that concept must have been severely wanting for moorings in biochemistry and physics; but of course it would still be possible to form the concept around whatever series of overlapping similarities were evidenced by "families" of the time, who would be recognizable as families by the crudest "mechanisms" as well as by the most well-controlled scientific investigation.

And then, if we, now, insist upon drilling down into "underlying mechanisms" until we hit the limits of our scientific understanding, then we are no longer analyzing the "practices" of anyone who lacks this understanding, which (fbfw!) still includes a huge number of our own contemporaries.

" The larger context in which Wittgenstein's philosophy is seen to develop considers his uncompromising opposition to essences, mental entities and other forms of idealism which were accepted as a matter of fact in continental philosophy at the turn of the preceding century. In his view, the main cause for such errors is language and its uncritical use. In the received view, concepts, categories or classes are taken to rely on necessary features common to all items covered by them. Abstraction is the procedure which acknowledges this necessity and derives essences, but in the absence of a single common feature, it is bound to fail. "

And again, though our Institutionalist will have no truck with idealism , he thinks he could get away with it here if he had to, because the concept: family DOES, after all, for these purposes rely on   a single common feature, namely, genetic descent.

If so, then so-called abstraction DOES work, this time; the essence of "family" is thereby derived; and then everyone can go to lunch.

That is certainly an ever-present empirical possibility, but the theorist of "family resemblance" seems to be concerned precisely with those cases which lack an "abstractable essence."

" §67 begins by stating:
I can think of no better expression to characterize these similarities than "family resemblances"; for the various resemblances between members of a family: build, features, colour of eyes, gait, temperament, etc. etc. overlap and criss-cross in the same way. – And I shall say: "games" form a family. "

So, what ultimately justifies the deployment of such a metaphor into the heat of analytic battle?

The key phrase above is overlap and criss-cross. There is an element of 'quantity becoming quality' which is suggested here.

Concerning the sheer 'quantity' of perceptually discrete resemblances in a given case, I would distinguish three broad 'quality' regions:

  first, there is the region of perfect naked-eye distinguishability among both the features and among their combination(s);

  next there is the region of 'complex' resemblance, that is, of "family resemblance," wherein the sheer number of distinguishable features afoot makes it impossible to find patterns of combination with the naked eye (one might also say, with the 'naked intellect'), yet which is not so vast so as to forbid the emergence of the gestalt-like quality, "family resemblance";

  finally, there may be such a profusion of features, of combinations of features, and of functionally 'random' "overlap" which is too chaotic even to favor the emergence of a gestalt; this would be something like encountering a large set of 'unrelated' people among whom any resemblance at all is only occasional, accidental, and distinctively not of the "family" (i.e. criss-crossing) type.

(A further conjecture on this: Paradoxically, if two people have exactly the same eyes, this is more likely to be noticed in the third scenario than in the second, because there it stands out against the prevailing non-resemblance, and because no other observable patterns are vying with it for our attention; whereas in a set of people exemplifying "family resemblance" MOST such specific commonalities go unremarked upon or even unnoticed unless we are prompted to notice them. Part of "family resemblance," then, is that we are simply more apt to notice the "resemblance" gestalt than to notice the distinguishable features which comprise it, even though we ARE able to distinguish and enumerate the features GIVEN some prompting and effort. If so, then there is much more to this than the simple problem of being unable to name a common feature of games, or of artworks. Perhaps we could (finally!) name those common features, and still we would have trouble explaining the first thing about "our practices" of naive categorization, because the assumption that we had somehow, the whole time, detected the common property (unconsciously?) but been hitherto unable to articulate it no longer seems like a safe assumption. If it takes a century of tortuous dead-tree serve-and-volleying to hunt down the common property, then as for the thesis that the category was naively formed, the whole time, around JUST THIS common feature, that thesis has been bleeding parsimony the whole long time, until at the century mark it is fully and ultimately un-parsimonious.

" Why do we call something a "number"? Well, perhaps because it has a direct relationship with several things that have hitherto been called number; and this can be said to give it an indirect relationship to other things we call the same name. And we extend our concept of number as in spinning a thread we twist fibre on fibre. And the strength of the thread does not reside in the fact that some one fibre runs through its whole length, but in the overlapping of many fibres. "

Ditto above.

" Wittgenstein's insistence that boundaries do not really exist but can be traced arbitrarily has been described as conventionalism and more generally the acceptance of his conception has been seen to present a refined nominalism. "

I HATE when philosophical -isms name some artifact of analysis without naming the object of analysis!

Hence in nominalism, le nom is the answer to a question that was asked about something other and vaster than mere names: the name was never the decisive question, and to answer that question with: 'names are arbitrary' is first and foremost to be . . . an 'arbitrarian.' It is only secondarily (and distantly) to be a Disciple of The Name; but this is precisely the ad hominem projection of the nominalist's philosophical adversaries, and since they vastly outnumber him, the 'name' sticks. So goes Realism, so-called.

Similarly, the analyst who concludes that convention is merely arbitrary is a conventionalist: he has pulled the rug of essence out from underneath the unwitting follower of convention. But the exemplary practioner of convention ALSO is a conventionalist in the usual sense of that sort of adjective-suffix union. Further, the coinage 'antiessentialist' suffers from the scourge of definition-by-negation: what is the antiessentialist in favor of? Mustn't we all be in favor of something? And perhaps our "conventionalist" is an 'antiessentialist' only with specific regard to matters implicated in/by "convention" writ large. In that case he is, precisely, 'an antiessentialist with regard to convention.' But that label is never going to catch on.

Convention (and naming) cannot survive analysis; this is more because both deeply depend upon remaining unquestioned, and less because of the SPECIFIC details of any analysis which ultimately reveals this GENERAL problem. So, analysts who make this revelation about convention are "conventionalists"; or if they analyze nomenclature then they are "nominalists." But these turn out to be our exemplary SKEPTICS of that-which-is-analyzed, not our exemplary PRACTITIONERS of it.

What shall we call the practitioners? Surely those practitioners did not begin calling themselves 'Realists' until after the real-ness of that-which-they-hold-to-be-real began to be subjected to withering scrutiny?

I presume that the archetypal Nominalist espouses some notion of 'reality'? and that this notion is not less real but simply less expansive than that espoused by his foil?

I am somewhat embarrassed to realize, only recently, just how badly the terms 'realism' and 'realist' have thrown me over the years, for precisely these reasons. If an analyst concludes, e.g., that 'aesthetic properties' are not 'real,' this is THEIR 'reality.' If they are correct, then one tiny link in the larger 'realistic' picture of the world consists of knowledge that 'aesthetic properties' are not real.

I suppose this avoids another potentially harrowing confusion, since it would be untenable for exponents of irreconcilable theories both/all to refer to themselves as THE 'realists' and to their opponents as the 'antirealists' or some such contrivance. But terms like 'antirealist' and 'antihumanist' ARE contrived; in the latter case the term seems not actually to be the perfect inverse of its straight mark. I think this already puts these theories at a tremendous rhetorical disadvantage, for reasons which, yes, demand a "Wittgensteinian" analysis. Human needs and human purposes are afoot!

(Alan Goldman has a paper entitled 'Realism About Aesthetic Properties.' This parlance eases the problem at (great) cost of elegance and conciseness. And I have recently encountered the suggestion, I can't recall where, that Philosophy of X usually means ('really' means): Philosophical Foundations of X. Same old problem solved, same new problem created.)

(What is 'moral realism' all 'about'? What comes after the 'about' in that case? Everybody knows this . . . so why can't I seem to articulate explicitly what it is?)

" See Also ...

---

'Polyphyly'

"A polyphyletic group is an assemblage that includes organisms with mixed evolutionary origin but does not include their most recent common ancestor. "

" For example, the biological characteristic of warm-bloodedness evolved separately in the ancestors of mammals and the ancestors of birds "

" Many taxonomists aim to avoid homoplasies in grouping taxa together, with a goal to identify and eliminate groups that are found to be polyphyletic. ... Researchers concerned more with ecology than with systematics may take polyphyletic groups as legitimate subject matter "

Well okay, who knows how directly this distinction between systematists and ecologists can be appropriated for our purposes here, but it's worth seeing how far this can be pushed.

Presumably Analytic Philosophers are chasing the "systematic." Hence if some art-property indeed evolved separately in, say, Tristram Shandy and Oedipus Rex, then categorizing these works together along that particular axis of similarity does indeed threaten to mislead . . . but it does not INHERENTLY mislead; we are collectively VERY susceptible to the Fallacy rightly called 'Genetic,' but this in itself is no argument against polyphyletic grouping of artworks. More likely, no goodly Ecologist is out to mislead anybody, but there are always plenty of people around who are out to mislead themselves using whatever they find closest to hand.

Perhaps the so-called 'Historical Definition of Art,' to which Our Man claims fealty and on which point he has promised to conclude the book, is just one more instance of this sort of self-deception. Thus family history arrogates to supersede family resemblance.

But this is a non sequitur. We can anticipate that the true genetics of Art History are a motley vis-a-vis appearance; and there is much more to this criticism than the mere sulk that appearances OUGHT to matter more than "underlying mechanisms": again, if everything resembles everything SOMEHOW, that doesn't mean that ANY such resemblance as MIGHT be posited necessarily is implicated in any of "our practices." If we confine ourselves to those resemblances which ARE so implicated, we seem to gain much insight into BOTH appearance and practice (and function and genetics); in the main: There are "resemblances" between Beethoven and Mozart, and between Beethoven and Goethe; there is "descent" also, by most any reasonable standard; but GOOD LUCK teasing out a principled relationship between resemblance and descent if, from the start, you refuse to grant the fact of resemblance without first seeing evidence of descent; and if your support for this orientation is to insist that such a principled relationship simply must exist, and simply must be isolable, if only someone was dogged enough to pursue it, again, GOOD LUCK.

What Our Man calls functional theories seem necessarily to permit works of separate descent to be categorized together, i.e. under their common "functional" umbrella; and this seems less a bug than a feature of the Aesthetic and Formalist theories as Our Man presents them here. In these terms, the failure of such functional theories as Definitions of Art is simply that myriad artworks do not 'function' as posited by the theories; but then, presumably it is an empirical matter (presumably a SIMPLE one!) to determine function post hoc, and then to PERFECTLY draw the category-boundary around the 'function' du jour.

" From a practical perspective, grouping species monophyletically facilitates prediction far more than does polyphyletic grouping. ... [e.g.] Linnaeus' assignment of plants with two stamens to the polyphyletic class Diandria, while practical for identification, turns out to be useless for prediction, since the presence of exactly two stamens has developed convergently in many groups. "

So, . . .

Are we predicting?

Identifying??

Or what???

Evidently there are more reasons than just the ones I hinted at above why art-classing by 'function' is a strictly post hoc endeavor.

The "Neo-Wittgensteinian" says: THERE SIMPLY IS NO grouping of artworks which permits identification in the above sense. Rather, when we do "identify" artworks or games, e.g., as we seem quite often to do, this is based on something even less (or perhaps the real suggestion is: SO MUCH MORE) than what we are doing when we explicitly enumerate the characteristics of some object. When we enumerate properties, we are naming fibres; we are not naming threads. In that case, the true basis for these sorts of categorizations is something which cannot be named. These categories are the human psychic artifacts arising from a very specific situation, namely from the 'intermediate region' of 'complex resemblance.' The categories are ONE kind of artifact arising here; the "fibres" are the (THE) other. If there are only a small handful of fibres, and if they criss-cross always in the same (predictable!) manner, then perhaps we can explicitly articulate a principle, i.e., a "definition" of "necessary and sufficient conditions" for membership. Or, if there are hundreds of fibres and no discernible pattern of "overlap", then no category-artifact can emerge. But there is an intermediate region of number of fibres, and of intermittentness of overlap, within which "our practices" of categorization display the "family resemblance" phenomenon. That is how we arrive intellectually at the 'principle' of a category WITHOUT any 'principle' for drawing its boundaries. In a way it is more remarkable that we CAN do this than that we CANNOT later articulate a principle underlying it. That is cause for optimism, not for pessimism: we only need the principle if someone else challenges us to produce it; and we begin challenging each other to produce it . . . why, exactly? And when? Obvs, it's when the category-dependent "practice" has begun to break down, for whatever reason. But if the practice is working, even if it's only working FOR YOU, you probably aren't going to demand that it offer up a unifying principle; or if you are the speculative type and begin thusly to speculate, there's no emergency created by your failure to turn up the principle, NOT UNLESS something is not working as it should.

The "Institutional Theorist" says: BUT WE CAN predict, even if we cannot identify: we simply have to look at the genesis of the object instead of looking at the object itself. And if we cannot possibly predict how (or if!) artworks will 'function' in the future, then so much the worse for functional theories.

I'm sure I've already pushed this further than it can go, but it could be a good generative frame, if only provisionally, within which to articulate something further which jumps out from this discussion but is difficult to verbalize: By turning away from appearances and towards genetics, our Institutional Theorist seems to precisely to GRANT our Neo-Wittgensteinian's theory of resemblances, not to refute it. The Wittgensteinian admonition to look and see seems to get misprised, somewhere along the way, by the Institutionalist, as if the whole point of this admonition is to insist that the categorizer ought ONLY be able to 'look' at something and not to investigate its "genesis." Perhaps there really is an unstated assumption in the "family resemblance" theory that naive categorizers REALLY DO NOT 'look' at the genesis; or else that the genesis simply does not affect the ultimate outcome of the naive categorization process. That seems right to me, and so if it has simply been assumed as a background condition, I see no problem. I fail to see why the concept: "family resemblance" SIMPLY MUST heed genetics just because the concept: "family" does. Our Man seems headed for suggesting that the Institutionalist has caught the Wittgensteinian with his pants down; but really the Institutionalist grants exactly what the Wittgensteinian says: that there is a form of naive categorization whereby, when we 'look and see,' we FIND a category and some objects but no principles. How to turn up such a principle? We need to sneak a peek behind the curtain. All well and good! Purportedly, before reproduction was fully understood, the family line passed through the mother: anyone could just 'look and see' a particular child literally emerging out of a particular woman; there was no mystery, at least, to that. But the rest (purportedly!) WAS a mystery for a very long time. In such a case, the admonition to consider genetics is a bit like the Onion line (I think it was The Onion?) wherein some inconsiderate celebrity declared that starving children should "eat a sandwich or something." If they could, they would; and if we could know where ANYTHING comes from, with any certainty, then perhaps our naive categorizations, in that isolated domain at least, would finally be principled and would finally be parsable as such. But there is a basic metaphysical obstacle there, which I tried to elaborate above under the terms 'proximate' and 'ultimate.' I am of course wary of wading into 'metaphysics' . . . but if I have this part right, then the Wittgensteinian has been not ignorant but prudent.

" Species have a special status in systematics as being an observable feature of nature itself and as the basic unit of classification. It is usually implicitly assumed that species are monophyletic (or at least paraphyletic). However, hybrid speciation arguably leads to polyphyletic species. Hybrid species are a common phenomenon in nature, particularly in plants where polyploidy allows for rapid speciation. "

Sounds familiar!

 

---

 

[207]

Meeting of the Aristotelian Society at 21, Bedford Square, London, W.C.1, on 8th May, 1961, at 7.30 p.m.

XII-UNIVERSALS AND FAMILY RESEMBLANCES

By RENFORD BAMBROUGH

I believe that Wittgenstein solved what is known as "the problem of universals", ...

... I naturally do not claim to be making an original contribution ... My purpose is to try to make clear what Wittgenstein's solution is and to try to make clear that it is a solution.

...

In seeking for Wittgenstein's solution we must look mainly to his remarks about "family resemblances" and to his use of the example of games. In the Blue Book he speaks of

"our craving for generality"

and tries to trace this craving to its sources:

... There is—

(a) The tendency to look for something in common to all the entities which we commonly subsume under a general term.—We are inclined to think that there must be something in common to all games, say, and that this common property is the justification for applying the general term

[208]

"game" to the various games; whereas games form a family the members of which have family likenesses.

Generated likenesses?

Or mere "likenesses," i.e. as from the standpoint of naive intention?

Some of them have the same nose, others the same eyebrows and others again the same way of walking; and these likenesses overlap.

Indeed.

But what has generated them?

Or is that part unimportant?

The idea of a general concept being a common property of its particular instances connects up with other primitive, too simple, ideas of the structure of language. It is comparable to the idea that properties are ingredients of the things which have the properties; e.g., that beauty is an ingredient of all beautiful things as alcohol is of beer and wine, and that we therefore could have pure beauty, unadulterated by anything that is beautiful.

So, it seems that we ourselves, the naive onlookers, are the source of our own general concepts; moreover, we are the source of the conceit to elevate such "concepts" to the status of common properties of objects; those objects, that is, that we find ourselves (naively!) categorizing together.

Perhaps, then, the phenomenon of "family resemblance" simply describes the aftermath of this naive process?

(b) There is a tendency rooted in our usual forms of expression, to think that the man who has learnt to understand a general term, say, the term "leaf", has thereby come to possess a kind of

general

picture of a leaf, as opposed to pictures of

particular

leaves. He was shown different leaves when he learnt the meaning of the word "leaf"; and showing him the particular leaves was only a means to the end of producing "in him" an idea which we imagine to be some kind of general image. We say that he sees what is in common to all these leaves; and this is true if we mean that he can on being asked tell us certain features or properties which they have in common. But we are inclined to think that the general idea of a leaf is something like a visual image, but one which only contains what is common to all leaves. ... This again is connected with the idea that the meaning of a word is an image, or a thing correlated to the word. (This roughly means, we are looking at words

as though they all were
proper names,

and we then confuse

the bearer

of a name with

the meaning

of the name.)

In the Philosophical Investigations ...

66. Consider for example the proceedings that we call "games". I mean board-games, card-games, ball-games, Olympic games, and so on. What is common to them all?—Don't say: "there must be something common, or they would not be called 'games'"—but look and see whether there

[209]

is anything common to all.—For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat: don't think, but look!—Look for example at board-games, with their multifarious relationships. Now pass to cardgames; here you find many correspondences with the first group, but many common features drop out, and others appear. When we pass next to ball-games, much that is common is retained, but much is lost.—Are they all "amusing"? Compare chess with noughts and crosses. Or is there always winning and losing, or competition between players? Think of patience. In ball games there is winning and losing; but when a child throws his ball at the wall and catches it again, this feature has disappeared. Look at the parts played by skill and luck; and at the difference between skill in chess and skill in tennis. Think now of games like ring-a-ring-a-roses; here is the element of amusement, but how many other characteristic features have disappeared! And we can go through the many, many other groups of games in the same way; can see how similarities crop up and disappear.

And the result of this examination is: we see a complicated network of similarities overlapping and criss-crossing: sometimes overall similarities, sometimes similarities of detail.

67. I can think of no better expression to characterise these similarities than "family resemblances"; for the various resemblances between the members of a family: build, features, colour of eyes, gait, temperament, etc. etc. overlap and criss-cross in the same way.—And I shall say: "games" form a family.

Wittgenstein expounds his analogy informally, and with great economy. Its power can be displayed in an equally simple but more formal way ... We may classify a set of objects by

[210]

reference to the presence or absence of features ABCDE. It may well happen that five objects edcba are such that each of them has four of these properties and lacks the fifth, and that the missing feature is different in each of the five cases. ...

e ABCD

d ABCE

c ABDE

b ACDE

a BCDE

Here we can already see how natural and how proper it might be to apply the same word to a number of objects between which there is no common feature.

And perhaps we can see that the naturalness and the propriety alike grow exponentially with every mere arithmetic gain in complexity.

e.g.,

f ABCDE

e ABCDF

d ABCEF

c ABDEF

b ACDEF

a BCDEF

If we let 'similarity' and 'difference' be defined entirely by presence or absence of features, so to speak, then it's obvious that when we re-iterate Bambrough's taxonomy using five "features" instead of four, we thereby necessarily create greater 'similarity' (and less 'difference') among the "objects."

Is this 'really' what happens?

Presumably this consistent exclusion of exactly one "feature" at each and every level of complexity is a rhetorical idealization. To be as skeptical as possible about this detail, but without contravening any of the classic Wittgensteinian observations above, I would prefer to assume that the arithmetic is just barely robust enough to sustain the truth of the phenomenon Bambrough identifies here: That is, the degree of similarity among the "objects" approaches some constant as the number of discrete "features" approaches infinity. Befitting the lessons of evolutionary theory, I would assume that this constant, so construed, represents a minimum rather than an optimum 'degree of similarity' vis-a-vis the possibility of naive category formation.

I'm sufficiently mathematically inclined to think up this fudge but not mathematically learned enough to express or calculate it properly, nor to care to do so. I'm just going to guess that it turns out to be the Golden Mean, i.e., that the ratio between 'the total number of features' and 'the average number of shared features' equals the ratio between 'the average number of shared features' and 'the average number of unshared features', and that this ratio is indeed Golden.

By this standard, already in Bambrough's example we are indeed dealing with an unusually high degree of similarity. We can bring the example closer to our theorized 'reality' simply by ratcheting down the 'similarity' by one coarse degree:

e ABC

d ABE

c ADE

b CDE

a BCD

Now the ratio is Golden-er, but there are not enough total "features" to sustain the conceit that e and c, say, will readily and naively be categorized together, since by this latest logic they are quite literally 'more different' than they are 'similar.'

What if we take the same ratchet to the six-feature version?

f ABCD

e ABCF

d ABEF

c ADEF

b CDEF

a BCDE

Now we're getting very close to what I have in mind.

What else? Suddenly the locution no common feature is seen to be itself too coarse: It actually understates the case; it understates the 'degree of unsharedness' of any given property, or at least the possibilities for unsharedness within a still-robust category.

Each of our latest "objects" now lacks two of the six properties; a different two in each case. That is something much more complex and difficult to verbalize than mere absence of a common property; it is an even patchier principle which nonetheless, I would say, is totally believable as a theory of naive categorization.

And if we confine our attention to any arbitrarily selected four of these objects, say edca, then although they all happen to have B in common, it is clear that they are all rightly called by the same name. ... [but this] would not be in virtue of the presence of the common feature or features ... since the name also applies to possible instances that lack the feature or features.

... if we set it out more fully and formally in terms of a particular family than Wittgenstein himself ever did. ... "the Churchill face" ... when a family group photograph is set before us it is unmistakable that these ten people all belong to the same family. It may be that there are ten features ... It is obvious that the unmistakable presence of the family face ... is compatible with the absence ... of

  one  

of the

  ten  

constituent features ...

How about . . .

two of the ten?

three . . . ?

. . . ?

Six presences and four absences gets us even closer to my magic number. Is this a reasonable hypothesis? I think it's very reasonable provided that some further granular details shake out in a certain way.

If the distribution of properties is truly random, i.e., if no patterns (read: 'predictions') of any kind can be coaxed out of the raw data, then it becomes harder to believe that a category could emerge from a feature-array wherein close to half of the features were absent from most or all of the members;

but if, say, we find three or four of the ten properties to be in evidence in the vast majority of cases (say, in 80 percent or more of all cases), then I find it reasonable to expect that the behavior of the other six or seven properties hardly matters; all that matters is that their number is finite and their individual qualities qua "properties" are just consistent enough to justify distinguishing and designating them as properties in the first place. (See Bambrough below, re: cleft chins.)

Perhaps this is how we end up with the otherwise vexing result that categories can form without any common property among the members: In a sense, the 'common property' that Naive Realists expect to find does exist, but it's not literally held in common, 100 percent of the time, and it's not literally the same property each time; rather, it's actually just the common-est property among certain empirically propitious subgroups of exemplars, exceeding a certain threshhold of common-ness, but able to vary widely in the entire region above that threshhold; or perhaps in the case of a category with ten properties we need for several of the arrayed properties to be robust in this way, but we do not need more than half of them to be this robust.

... also obvious ... it does not matter if it happens that the [one] feature which is absent from the face of each individual member of the family [N.B. a different feature in each case] is present in every one of the others. ...

[211]

This example is very artificial, ... But ... the more natural the example is made the more it suits my purpose. If we remember that a family face does not

divide neatly

into ten separate features, we

widen

rather than

reduce

the scope for large numbers of instances of the family face to lack a single common feature. And if we remember that ... all cleft chins have

nothing in common

except that they are

cleft chins, ... there could in principle be

an infinite number

of unmistakable Churchill faces which had no feature in common.

I'm not sure the last part works. It seems possible for the range of variation among cleft chins per se to be vast enough for at least some of them to work against the "family resemblance." In other words, it seems that "cleft chin" here is itself a category, subject to the same vicissitudes as are any "faces" which supervene upon it. If so, then its supervenient categories cannot be infinite in number.

Certainly this number must be huge in relation to the number of the uber-category's constituent properties; and perhaps that means that this number is functionally or pragmatically infinite, seeing that there are well-understood limits to naive perception and cognition; but I fail to see how this 'infiniteness' could be analytic to the concept of a 'supervenient category.' (Of course we concurrently hope, pray and assume that this doesn't matter at all, so that we don't have to deal with it in any detail.)

(And yes, if the above is correct then there must be some f(x) expression which describes the relationship between the ranginess of the subvenient 'categories' on one hand, and the emergent supervenient categories on the other. And again, let's just hope, pray and assume!)

...

The passages that I have quoted contain the essence of Wittgenstein's solution of the problem of universals, but they are far from exhausting his account of the topic. ... most of his philosophical remarks in The Blue and Brown Books and in the Philosophical Investigations are concerned with such questions as

"What is the meaning of a word?"

"What is language?"

"What is thinking?"

"What is understanding?"

And

these questions

are

various forms of the question

to which

theories of universals,

including Wittgenstein's theory of universals, are

meant to be answers.

...

... the point Wittgenstein made with the example of games has a much wider range of application than that example itself. ... exactly how wide ... ? ... In his striving to find a cure for "our craving for generality," in his polemic against "the

[212]

contemptuous attitude towards the particular case," he was understandably wary of expressing his own conclusions in general terms.

i.e.

"there is no important idea that stupidity does not know how to make use of, for it can move in all directions and is able to wear all garment of truth. Truth, on the other hand, has only one garment and one road and is always at a disadvantage."

(Thanks, Milo.)

i.e., The greater the discovery, the greater the danger of midwit misprision, and the greater the discoverer's wariness.

... [hence we] are consequently impelled to make use of glosses and paraphrases and interpretations [in order] to relate his work ... to most other philosophical writings and doctrines.

I believe that this is why Wittgenstein's solution of the problem of universals ... has not been widely seen to be a solution.² ... He talks about

games
and families
and colours,

... but not about

"the problem of universals."

... He did not, for example, plot the relation between his remarks ... and the doctrines of those philosophers who had been called

Nominalists

and

Realists.

... I am claiming that his remarks can be paraphrased into a doctrine which can be set out in general terms and can be related to the traditional theories, ...

... the range of application of the point that is made by the example of games, ... it is at this crucial first stage that most readers of Wittgenstein go wrong. ... we ... ask ourselves, "With what kinds of concepts is Wittgenstein contrasting the concepts

---

² Of recent writings on this topic I believe that only Professor Wisdom's Metaphysics and Verification (reprinted in Philosophy and Psycho-analysis) and Mr. D. F. Pears' Universals (reprinted in Flew, Logic and Language, Second Series) show a complete understanding of the nature and importance of Wittgenstein's contribution.

[213]

of game, language, ... ?" ...

The first [possible] answer ... Ayer contrasts the word "game" with the word " red", on the ground that the former does not, while the latter does, mark "a simple and straightforward resemblance" between the things to which the word is applied. ... " ...

the resemblance

between the things to which the same word applies may be

of different degrees.

... " Now this contrast ... is important, ... but I am sure that this is not the point that Wittgenstein was making with his example. ... Wittgenstein could as easily have used the example of red things as the example of games to illustrate "the tendency to look for something in common to all the entities which we commonly subsume under a general term." ...

A second possible answer ... Mr. P. F. Strawson ... The contrast is not now between simple and complex concepts, but between

two kinds of complex concepts:

those which

are definable

by the statement of

necessary and sufficient conditions

and those which

are not.

But once again

[214]

the contrast, although it is important, ... is not the point that Wittgenstein is concerned with. In the sense in which ... games have nothing in common except that they are games, and red things have nothing in common except that they are red, brothers have nothing in common except that they are brothers.

. . . that they are brothers . . .

i.e., that they just are "brothers" according to our practices.

Must things be this way? Can things be no other way?

The scope for that kind of variation lies in the arena of subvenient "features," not in the arena of any supervening classification.

Bored? Seeking excitement? You'll have to settle for the 'variety' and 'arraying' of "features" coexisting strictly within some supervening class.

Anything exciting here, obvs, is in relation to the fact (call it perplexing if you insist) that we do practice   brother-ness in a dizzying 'variety' of ways; that we can in principle imagine an infinite number of "brothers," yet there remains only one concept: brother.

So, of course things may be other than they are, and "infinitely" so. There is infinite scope for that.

What is there no scope for? For inclusion of non-brothers in the category: brothers.


. . . brothers have nothing in common except that . . .

i.e., If we try to say ('really' say) what "brothers have in common," we end up simply reiterating the concept as we've found it rather than coaxing something further out of it.

It is true that brothers have in common that they are male siblings, but their having in common that they are male siblings

is

their having in common

that they are

brothers, and

not

their having in common

something in addition

to their being brothers. Even a concept which

can be explained

in terms of necessary and sufficient conditions

cannot be
ultimately
explained

in such terms.

I landed on something similar above, re: 'proximate' and 'ultimate' causality.

To satisfy the craving for an ultimate explanation of "brother" in such terms it would be necessary to define "male" and "sibling", and the words in which "male" and "sibling" were defined, and so on ad infinitum and ad impossibile.

What then is the contrast that Wittgenstein meant to draw? I suggest that he did not mean to draw a contrast at all.

i.e.,

"With what kinds of concepts is Wittgenstein contrasting the concepts of game, language, ... ?"
(212-213)

Rather, these exemplify the concept of a concept.

(If there really is no such "contrast" drawn, then the concept: art cannot be any 'contrasting' exception to the general working of "concepts" . . . and if not, then we have to return to the most basic and supremely unexciting aspect of the art-definition problem, which is that it has been possible for a long time for no two people anywhere to possess quite enough of the same exemplars, and for only a few people to "possess" exemplars which meet or exceed the minimum threshhold for typification.)

Professor Wisdom has remarked that the peculiar difficulty of giving a philosophical account of universals lies in this: that philosophers are usually engaged in implicitly or explicitly

comparing and contrasting

one type of proposition

with another type of proposition

... whereas propositions involving universals cannot be compared or contrasted with propositions that do not involve universals, since

all propositions involve universals.

... [thus] philosophers are led to offer accounts of [Wittgenstein's] doctrine which restrict the range of its application. They recognise ... that at least some general terms can justifiably be applied to their instances although those instances have nothing in common.

[215]

But they are so deeply attached to the idea that there must be something in common to the instances that fall under a general term that they treat Wittgenstein's examples as special cases, ...

Here we come across an ambiguity which is another obstacle ... Ayer remarks ... "It is correct, though not at all enlightening, to say that what games have in common is their being games." It is certainly correct, but I strongly deny that it is unenlightening. ... trivialities and platitudes deserve emphatic affirmation when ... they are explicitly or implicitly denied, or forgotten, or overlooked. Now the platitude that all games have in common

that they are

games is

denied by
the nominalist,

who says that all games have nothing in common

except that
they are called
games.

Hmm.

I was expecting the nominalist to say:

  "to BE A game
  just is
  to BE CALLED A
  game."

??

And it is

not only the nominalist,

but also his opponent,

who misunderstands the central importance of the platitude that all games have in common that they are games. When he ... rightly wishes to insist that games have something more in common than simply that they are called games, he feels that he must look for

something that games have in common

apart from being

games.

This feeling is entirely misplaced. The very terms of the nominalist's challenge require only that the realist should point out

something that games have in common

apart from being called

games,

and this onus is fully discharged by saying that

they are games.

Hmm.

They are, like, in some ultimate way?

Although the feeling is misplaced, it is a very natural feeling, ... If I ask you what these three books have in common, or what those four chairs have in common, you will look to see if the books are all on the same subject or by the same author ... ; to see if the chairs are all Chippendale or all three-legged ... It will never occur to you to say that the books have

[216]

in common that they are books ... And ... if you cannot see any other specific feature that they have in common, you will say that as far as you can see they have nothing in common. ... I may then tell you that all the books once belonged to John Locke ... But it would be a poor sort of joke for me to say that the chairs were all chairs or that the books were all books.

Well . . . the poor-est sorts of all really do come in the form:

. . . I may then tell you . . .

Why?

Because

. . . I asked you . . .

in the first place!

That is not very nice!

I suppose there is a form of skepticism, readily available here, whereby the existence of the "common feature" can never be ruled out conclusively. It's not unreasonable to think so, since "features" get overlooked or concealed all the time.

Surely this is not the intended point of the above hypothetical, but it makes this point all the same: If you have, in whatever manner and for whatever reasons, managed to avoid learning of Locke's ownership of the books, then you cannot classify them together under that heading; not even if you do in some sense 'possess' the concept: books owned by Locke.

This is where I land every time with the Institutional Theory. In art-philosophy it functions something like the Theory of Lockean Ownership functions above: Many of us possess the concept of an 'institution,' but how could we ever classify institutional products under that heading without simply knowing from the start which are the institutional products and which are not? The point is not so much that we would not or could not do this, but that we do not. We cannot argue that this is our practice, not even if the Institutional Theory succeeds, in spite of itself, in drawing the art-boundaries with perfect accuracy. In that case it is not 'practice' that has gone into the definition and it is not "practice" which comes out of it either. Perhaps then it is the Institutional Theory which is the 'polyphyletic grouping' and naive 'practice' that is the 'genetic' art-engine. What's stopping us from flipping the script this way? It seems to depend rather straightforwardly on our standpoint, on our view of the 'object.' I mean nothing coded or deconstructionist by 'standpoint.' If you have no view of the 'genesis' of two stamens, then of course two stamens can only appear distinct from one or three stamens. Whether you are obligated to try to figure out the 'genesis' depends entirely on what you're trying to do.

If I ask you what all chairs have in common, or what all books have in common, you may again try to find a feature like those you would look for in the case of these three books ... ; and you may again think that it is a poor sort of joke for me to say that what all books have in common is that they are books ... And yet this time it is not a joke but an important philosophical truth.

Because

the normal case

where we ask

"What have

all these

chairs, books or games in common?" is one in which

we are not concerned

with

their all being

chairs, books or games, we are

liable to overlook

the extreme peculiarity

of the philosophical question that is asked with the words

"What do

all chairs,
all books,
all games

have in common?"

More concisely:

  all these blanketies  ≠  all blanketies

For of course games do have something in common. ... and yet when we look for what they have in common we cannot find it. ... And this is not ... because what we are trying to say is too subtle and complicated to be said, but because it is too easy and too simple to be worth saying: and so we say something more dramatic, but something false, instead. The simple truth is that what games

have in common

is

that they are

games. The nominalist is obscurely aware of this, and by rejecting the realist's talk of transcendent, immanent or subsistent forms or universals he shows his awareness. But by his own insistence that games have nothing in common except that they are called games he shows the obscurity of his awareness.

With the theory of "family resemblance" we seem to be involved precisely in the task of giving the reason why certain things are called by the same name. And since the nominalist grants precisely that this is what we do (not to say that it is all that we do), I def wasn't expecting the call-ing to be quite so fastidiously distinguished from the be-ing. To me it doesn't seem worth insisting on such a thoroughgoing conception of 'being,' not even if I tend to fall on the nominalist's side of this question. But I suppose I too show my dim awareness of matters ontological in arguing that the most pragmatic (practical) course is simply to avoid pressing the issue, i.e., to avoid staking anything important on the presumption of correspondence between names and reality. The assumption underlying that argument is that the question of ultimate being can be made to matter; that is, we can force it to matter: first, by betting on its correspondence to our nomenclature; and subsequently, by watching the contraindications roll in, one by one, until inevitably one of these contraindications literally kills not just our bet but us ourselves. If anyone survives this gratuitous revelation of ultimate ontology, then they will get to enjoy possession of that rarest sort of knowledge; but here is one case where we should indeed always think first of the victims.

The realist too ...

[217]

By his talk of transcendent, immanent or subsistent forms or universals he shows the obscurity of his awareness. But by his hostility to the nominalist's insistence that games have nothing in common except that they are called games he shows his awareness.

All this can be more fully explained by the application of what I will call "Ramsey's Maxim." ...

... it is a heuristic maxim that the truth lies not in one of the two disputed views but in some third possibility which has not yet been thought of, which we can only discover by rejecting something assumed as obvious by both the disputants. ...

It is assumed as obvious by both the nominalist and the realist that there can be no objective justification for the application of a general term to its instances unless its instances have something in common over and above their having in common that they are its instances.

The nominalist rightly holds that there is no such additional common element,

and

he therefore wrongly concludes

that there is no objective justification for the application of any general term.

The realist rightly holds that there is an objective justification for the application of general terms,

and

he therefore wrongly concludes

that there must be some additional common element.

Wittgenstein

denied the assumption
that is common
to nominalism and realism,

and that is why I say that he solved the problem of universals. ...

...

[218]

...

Wittgenstein thus denies at one and the same time the nominalist's claim that games have nothing in common except that they are called games and the realist's claim that games have something in common other than that they are games. He asserts at one and the same time the realist's claim that there is an objective justification for the application of the word "game" to games and the nominalist's claim that there is no element that is common to all games. And he is able to do all this because

he denies
the joint claim
of the nominalist and the realist

that

there cannot be an objective justification
for the application of the word "game"
to games

unless

there is an element that is common
to all games

(universalia in rebus)

or
a common relation
that all games bear
to something that is not a game

(universalia ante res).

Wittgenstein is easily confused with the nominalist because he denies what the realist asserts: ...

When we see that Wittgenstein is not a nominalist we may easily confuse him with the realist because he denies what the nominalist asserts: ...

But we can now see that Wittgenstein is neither a realist nor a nominalist: ...

I will now try to put some flesh on to these bare bones.

The

value and the limitations

of the nominalist's claim ... can be seen if we look at a case where

a set of objects
literally and undeniably
have nothing in common

except that

they are called by the same name.

If I choose to give the name "alpha" to each of a number of miscellaneous objects ... then I may well succeed in choosing the objects so arbitrarily that I shall succeed in preventing them from having any feature in common ... But

[219]

... In the first place,

the arbitrariness of my selection

of alphas is not paralleled in

the [typical] case

in which I apply the word "chair" successively ... The second point is that the class of alphas is

a closed class.

Once I have given my list I have referred to every single alpha in the universe, actual and possible. ... if I later add an object that I excluded from my list, or remove an object that I included in it, then I am making a different use of the word "alpha." With the word "chair" the position is quite different. There are

an infinite number
of
actual and possible

chairs. I cannot aspire to complete the enumeration of all chairs, ... and

the word "chair,"

unlike the word "alpha",

can be applied
to an infinite number of instances

without

suffering

any change of use.

These two points lead to a third and decisive point. ...

The use

of the word "alpha"

cannot be learned or taught

as

the use

of a general word

can be learned or taught.

...

[220]

... The

reference that we make

to

a finite number of objects

to which the word applies, and to a finite number of objects to which the word does not apply,

is
capable of equipping

the pupil with a capacity for correctly applying or withholding the word to or from

an infinite number of objects
to which we have made no reference.

... Even

if everybody always called a particular set of objects by the same name,

that would be insufficient

to ensure that the name was

a general name,

and

the claim of the name

to be a general name

would be

defeated

by

just that necessity for reference

to

the arbitrary choices

of the users of the name that the nominalist

mistakenly claims to find

in the case of

a genuinely general name.

For the nominalist is right in thinking that

if we always had to make such a reference

then
there would be no

general names

as they are understood by the realist.

The nominalist is also right in the stress that he puts on

the role of
human interests and human purposes

in

determining our choice

of principles of classification.

Whose purposes?

When is purpose??

Hat tip to McLuhan, of course . . . but this time it's just a literal question:

Do the purposes REALLY come BEFORE the principles?

Perhaps they do. But . . . then there comes . . . a CHOICE? That seems like the wrong word. We don't (entirely) "choose" even our "interests and purposes," hence we don't (quite) "choose" any downstream "principles."

It becomes much harder, not easier, to explain how "universals" could exist once "choice" enters the picture.

How this insistence on the rôle of human purposes may be reconciled with the realist's proper insistence on the objectivity of the similarities and dissimilarities on which any genuine classification is based can be seen by considering an imaginary tribe of South Sea Islanders.

Ok, but BE NICE!

Let us suppose that trees are of great importance ... , and that [these people] have a rich and highly developed language in which they speak of the trees ... But they do not have names for the species and genera of trees as they are recognised by our botanists. ... Our hosts are puzzled that we should call by the same name trees which appear to them to have nothing in common. They in turn surprise us by giving the same name to each of the trees in what is from our point of view a very mixed plantation. They point out to us what they called a mixed plantation, and we see that it is in our terms a clump of trees of the same species.

[221]

...

This looks like the sort of situation that gives aid and comfort to the nominalist ... But ... [actually] it cannot help him. We know already that our own classification is based on similarities and differences between the trees, similarities and differences

which we can point out

to the islanders

in an attempt to teach

them our language. Of course

we may fail,

but if we do it will not be because

we must fail.

Now either (a) The islanders have means of teaching us their classifications, by pointing out similarities and differences which we had not noticed, ... in which case both classifications are genuine,

Hmm. Helluva time to introduce the concept: genuineness into the mix!

and no rivalry between them, of a kind that can help the nominalist, could ever arise;

or (b) Their classification is arbitrary in the sense in which my use of the word "alpha" was arbitrary, in which case it is not a genuine classification.

Seems to me that if a real live case of (b) were located, this entire paper would be thrown into grave doubt.

If the Wittgensteinian arguments here are correct then (b) should not be possible at all. Of course we can enumerate all manner of Rankian-Beckerian irrationalisms which could serve as the "purpose" underlying an "arbitrary classification," but all that this proves is that, in that case, there really is no arbitrariness, because there is a purpose which we (outsiders, even) can readily comprehend upon explanation; all the same if it is a 'rational' explanation following from empirical or pragmatic inquiry (e.g. "botany" or boat-building) as if it is an 'irrational' explanation following from mystical tradition or faulty observation (e.g. belief in human reincarnation as trees).

In fact Bambrough did have a purpose in his "alpha" hypothetical, and it is not quite right to say that there was no principle underlying his choices: the principle, namely, was 'difference' rather than 'similarity' vis-a-vis his culture's already-prevailing classifications. What thus distinguishes his "alpha" process from "genuine" classification is not the absence of principle or of purpose but rather its psychic explicitness and its unilateral stipulation. Hence the use of the word "alpha" cannot be learned or taught as the use of a general word can be. But now this cuts against the mere possibility of (b) above, since (b) simply cannot arise from the "alpha" backstory: the island would have no rich and highly developed tree- language in the first place if only the "arbitrary" kind of classification had been attempted.

Rather, (a) is the only scenario that can generate this backstory. Why? Because "similarities and differences" can be invented as well as discovered. Moreover, the 'invented' kind can be just as readily taught and learned as the 'discovered' kind; much more readily, perhaps! And of course "learning and teaching" too can be either implicit or explicit. But all above suggests that the genuine-arbitrary distinction is supervenient upon the implicit-explicit distinction. Why?

I think we are once again in crying need of a Rankian-Beckerian corollary to fully explain why "alpha"-like contrivances never quite get off the ground. Intersubjective convergence upon "classifications" seems actually to require the bracketing of desire. Whoever it was that overturned Linneaus' two-stamen grouping acted on the world in a certain sense; and because their purpose was both already-explicit and already-shared, if only among a small group of elite scientists, any superior classification for that purpose was bound to be accepted by that group, if only it could be found. This is a highly peculiar human situation; it's no wonder that it took so long for us to engineer it, nor that it is extremely fragile. The situation elsewhere, i.e. 'naive classification,' is different in every detail. We can't really 'seek' anything in that domain but can only 'find' it. The shared 'purpose' even of a small community is nowhere near as unified as is the scientific mandate. The Rankian-Beckerian part is: We have a deep need (in some sense it is a dual need) for individual validation within a collective structure; this is both the glue and the dynamite taken to those structures. When we go 'seeking' for anything at all, we are in immediate danger of self-seeking; but in most material contexts we cannot literally 'let it come to us' either. So, the most efficient and effective approach turns out to be: to notice-without-noticing-that-we-noticed. If there is anything like 'universals' out there to be had, this is the only way we ever catch a glimpse of it: by the world acting on us. This is totally contrary to the Scientific Method, yet it is highly productive of knowledge. This knowledge, however, cannot just be 'applied', because to apply knowledge to a problem is to unleash the desire that once was bracketed. So, we had better 'find' some other arena(s) into which that desire can be 'unleashed.' Otherwise it will destroy not only our "classifications" but us ourselves.

(To believe in "universals" I suppose you must believe that the appearance of 'invention' is only an appearance, and that really it's 'discovery' all the way? It appears that we can 'invent' similarities and differences only because we truly perceive so little of the world at a time, such that whenever we happen to grab something new-to-us it is so new that it seems we must have 'invented' it; but really it was there the whole time. I don't mind this either, because I'm really, really hoping that this kind of thing just doesn't matter much at all; or, not for my "purposes.")

It may be that the islanders classify trees as "boat-building trees", "house-building trees," etc., and that they are more concerned with the height, thickness and maturity of the trees than they are with the distinctions of species that interest us.

In a particular case of prima facie conflict of classifications, we may not in fact be able to discover whether what appears to be a rival classification really is a classification. But we can be sure that

if it is
a classification

then

it is backed by

objective similarities and differences,

and that

if it is not

backed by objective similarities and differences then it is

merely an arbitrary

system of names.

Hmm. I fail to see why arbitrary classifications are not also backed by objective similarities and differences.

I think the distinction lies elsewhere than where Bambrough finds it: In the "arbitrary" case of "alphas," the "similarity" just is created by the "classification." This is 'subjective' only in a euphemistic sense. If it were truly subjective in the sense of presenting itself differently to different observers, then we would have to be equally weary of the possibility of verbally communicating it intact to those observers; But presumably there is no such difficulty: Presumably, when Bambrough tells us what he has done in constructing this class, "arbitrary" or not, there is nothing further hidden or ineffable for him to keep to himself.

Moreover, we have just seen that two classifications can both be objective without being shared intersubjectively. If so, then so shared-ness cannot be any criterion of objectiveness.

What has happened in each case, rather, is that classification has effected an extrinsic change in the objects: their relationship to each other and to the subject has changed; whereas there is no change intrinsic in any one object's so-called "properties" or "features." (So long as we grant that 'intrinsic' is a sensible concept.)

This means that the formation of the class has not been 'principled' or 'analytic.' But analyticity is something else entirely from objectivity; dare I say, it is something much less.

Put the other way 'round: there really is nothing arbitrary about this hypothetical classification-by-build. How can we reasonably apply the a-word here when we have said, as part of the background to the exercise, precisely what is not "arbitrary" about this classification? What he means, I suppose, is something like: 'analytically arbitrary.' Or maybe that's not at all what he means, but it is the only way I can see to read this passage coherently. Maybe I am a bit more of a Pragmatist that I once thought . . .

In no case will it appear that we must choose between rival systems of genuine classification of a set of objects in such a sense that one of them is to be recognised as the classification for all purposes.

There is

no limit

to the number of possible classifications of objects. (The nominalist is right about this.)⁴

---

⁴ Here one may think of Wittgenstein's remark that "Every application of every word is arbitrary," which emphasises that we can always find some distinction between any pair of objects, however closely similar they may be. What might be called the principle of the diversity of discernibles guarantees that we can never be forced to apply the same word to two different things.

[222]

There is no classification of any set of objects which is not

objectively based on
genuine similarities and differences.

(The realist is right about this.)

" . . . intrinsic similarities and differences . . . "

??

The nominalist is so impressed by

the infinite diversity of possible classifications

that he is blinded to

their objectivity.

The realist is so impressed by

the objectivity of all genuine classifications

that he underestimates

their diversity.

So, if I'm understanding correctly . . .

  the reason why
  infinite diversity and objectivity
  do not preclude each other
  is:

  because we can always
  find some similarities
  and/or differences.

Then, in addition to objective-subjective and intrinsic-extrinsic, eventually we will need a strictly empirical concept of convergent-divergent so as to take note of what is actually "found" in the way of similarity and difference.

(Suddenly 'look and see' has distinctive echoes of 'I do not seek. I find.' A painter could be so fortunate; but truly we do not 'seek' our classifications the way Bambrough seeks the arbitrary class: "alphas" as a philosophical foil. The seek-find distinction is entirely oblique to obj-subj, int-ext, conv-div, etc. This is important at all only to say, once again: Family Resemblance seems to be a theory of finding, not a theory of seeking; but Dickie, Levinson, Carroll, et al are nothing if not seekers and cravers-after-generality.)

Of course we may if we like say that there is

one complete system

of classification which

marks all

the similarities and all the differences. ...

That is, we can triangulate these coordinates only to find that we cannot possibly travel there? This is the epitome of a useless fact.

Or we may say that

there are only

similarities and differences,

from which we may choose

according to our purposes and interests. (This is the nominalist's summing up.)

This is what I would say, except for the part about choosing. Choosing is too much like seeking.

If nothing else, there is a certain metaphysical absurdity to the notion of "choosing" from among a literally "infinite" array of choices. Of course if all above is correct then in one sense this really is possible, it does happen, and most of the time it is utterly unremarkable; but the concurrent fact is that you can never choose optimally in such a scenario; and if you really are "choosing" wilfully rather than simply rolling downhill, if there really are "purposes and interests" explicitly driving your choice, then this is no mere academic question of Game Theory applied to the Game of Life; rather, it is a fully existential question of the first importance.

Rather than issuing an empty metaphysical proscription, I would prefer to think that the matters at hand don't actually implicate "choice" in the conventional sense, but rather (if anything) in the Sartrean sense, whereby the proscription-to-be-issued is perfectly tractable and earthly: it is issued against all of life's could-haves and should-haves; it is issued against nothing which comes before and everything which comes after the "choice."

i.e. If there was no talk of choice before or during formation of the classification, then there may be no such talk after the fact.

Certainly if we refuse to consider all that might have been done and concentrate instead on what has been done, then it becomes more natural (perhaps unavoidable) to broach the topic of "universals."

Nothing I've yet consulted here suggests in any way that classification results from "choices," except that the latter c-word itself keeps popping up in unfortunate places. Hopefully that's just a semantic crutch.

In talking of genuine or objective similarities and differences

Yes, more about that please . . .

we must not forget that we are concerned with

similarities and differences

between possible cases

as well as

between actual cases,

and indeed that we are concerned with the actual cases only because they are themselves

a selection

of the possible cases.

How un-existentialist! But perhaps the point is to consider just how much more is possible than is actual. The could-haves and should-haves multiply infinitely while the "actual" event remains (intrinsically) fixed. Or, more mundanely: As difficult as it is to name common properties for certain classes, it is no more difficult in those cases to (mentally) construct new "possible" cases which are far as can be from "actual"; the lack of a graspable common property impacts the compositionality of the possible not in the least.

The Existentialist may still carry the day, then, simply in being able (willing?!) to point out that mere possibility can never prove actuality, not even as a drop in some leaky bucket of evidence. Mere possibility points in no evidential direction. It is evidence only of what we're able to think up. There is no principled relationship that runs in that direction. Rather, the principle runs the other direction: What is actual must necessarily be considered possible too.

Given all of this, genuine or objective again seem like misdirected distinctions. "Possible cases" are not created the way "actual cases" are created. In fact the two processes are inverses in something like the manner of deduction and induction (though not, I suspect, only in that respect).

More importantly, I can't imagine that our interest in empirical actuality rests entirely in its being our sole glimpse at unglimpsable possibility, the bare ankle of reality, carelessly exposed. Instead, I would guess that the actual-possible distinction is principled, that this principle derives from "similarities and differences" that are 'intrinsic' to the objects, and that we do best here to limit ourselves to 'principled' thought and action.

Finally, this would mean that the 'Existentialist' irreverence for "possibility" is much more than mere bluster: in fact, as soon as we distinguish the actual and the possible, then there must be reasons (principles) underlying any fact of actualization; or else, if we bluster hard and refuse to make the distinction, then of course there can be no underlying principle because there is nothing for it to underlie.

Because the nominalist and the realist are both right and both wrong, each is driven into the other's arms ... The nominalist

talks of resemblances

until ... he must acknowledge that resemblance is unintelligible except as

resemblance in a respect,

and

to specify the respect

in which objects resemble one another

is

to indicate a quality or property.

The realist

talks of properties and qualities

until, when properties and qualities have been

explained in terms of other

properties and other qualities, he can at last do nothing but

point to the resemblances

between the objects that are said to be characterised by such and such a property or quality.

The question

"Are resemblances ultimate

or

are properties ultimate?"

is a perverse question if it is meant as one to which there must be a simple, single answer. They are both ultimate, or neither is ultimate. The craving for a single answer is the logically unsatisfiable craving for something that will be the ultimate terminus of explanation and will yet itself be explained.

 

---

 

[35]

III*—UNIVERSALS: BAMBROUGH ON WITTGENSTEIN
by Ilham Dilman

What is the problem of universals?
It concerns

our use of words
in general

and

general nouns

in particular
and asks

what in reality corresponds

to
our use of words
and
our principles of classification.
It seems that

either
some reality
independent of our language
does correspond

to the meaning of our words
and

justifies their use

in many particular cases,

or otherwise
the way we use them
is arbitrary.

That

in reality

which is

supposed
to correspond

to the meaning of a word
is
a universal.
It

is itself one thing

and justifies the use of the word

in many cases.

Because the one word
is
correctly used in many cases,
applied to many things,
we can describe the word
as

a general term.

Since

all words,

including demonstratives and proper names,
have

more than one application,

their meaning must have

some form of generality

in every case. So
the problem of universals
asks

what it is
that gives words
the generality inherent in their meaning.

1. Universals as ingredients

... Wittgenstein says that when we ask

" what are signs?",

instead of trying to find a general answer we would do well "to

look closely at particular cases

which we should call

'operating with signs'"

...
when we look at such examples
" ... We see

activities,
reactions,

which are clear-cut and transparent."
However
we tend to dismiss these
as peripheral and inessential to
what we are trying to understand
and [to]
treat the particular case with contempt.
...

---

*Meeting of the Aristotelian Society at 5/7 Tavistock Place, London WCI on Monday 6 November 1978 at 7.30 p.m.

[36]

...

...

Here
we think of properties
as ingredients of things—
of "beauty as an ingredient of all beautiful things
as alcohol is of beer and wine" ...
Wittgenstein turns to a different example, ...
... what happens when a man can be correctly described as expecting a visitor to tea: ...
He gives an example and in it mentions
some of the things

that may happen

... Not

what must happen.

The details he mentions
are

not necessary

for the example
to be one of expectation.
Yet

they are
precisely what

makes us talk of the man as expecting a visitor
in this case.
... we could have a case of expectation
which lacked those features.
There it would be

something different

that made us use

this same description.

So

what makes us
talk of the same thing
in different cases
need not be
the same thing.

...

But we find this
difficult to accept ...
We think:
"The symptoms may be different,
but if we are justified in using the same word,
the thing itself must be the same."

In one sense
this is perfectly true:
If we are justified in using the same word
"expectation",
these different cases
must all be cases of
expectation.
It is because of this ... that we are justified
in calling them by the same name.
But what makes this so is not some one thing
or set of things (properties) present in all, ...

...

[37]

...

We may still say:
"True we find no common feature,
but there must be."
Wittgenstein replies:
"Don't think, but look” ...
He means:
There is no must about it.
Do not trust

your abstract thinking,

but

your philosophically unadulterated vision

in particular cases.

How Phenomenological!

So . . .

is the Institutional Theorist
using
his abstract thinking
or
his unadulterated vision
?

Given only this passage and only Carroll's exposition so far, one certainly would have to think that the Institutional Theorist merely talks past Wittgenstein rather than engaging with him.

That aside, a more apt point is that there are many different kinds of "features" and the kind that the Institutional Theorist pursues just is the kind that is thought up, and not the kind that can be looked at. That is one way to keep the faith that there must be, since all concerned seem to agree that the thinking-up project can never fail.

Both the Institutional Theorist and the so-called Neo-Formalist are quite impressed with themselves for this leap, but it seems not to be an account of any practice. Probably that's because even naive 'practices' don't exemplify (don't permit) this kind of thinking-up; if they did, then they would not be recognizable as 'practices' but only as a series of discrete 'subpractical' actions or thoughts: these would present more like the objects comprising Bambrough's class of "alphas" would present to someone who had no inkling of Bambrough's intent.

Come to think of it . . . what really is the difference between Bambrough's "alphas" and Duchamp's readymades?

... "you will not see something that is
common to all" ... But even if you did,
why should
that
be what makes
each
a case of expectation?
... any such common element that you can think of, ...
will justify the claim that he is in a state of expectation
only in certain surroundings—
the very thing we ignore in our search
for a common element. ...
Can you not imagine surroundings
in which it would be ludicrous
to describe the man as
expecting someone
despite the presence of this common element—
whatever it may be?
[e.g., pacing]
...

The last point that Wittgenstein makes
is this:
suppose that

the reason why
we do not

in fact find a common element
is
that

it is hidden from us.

Sounds familiar!

But if so,
how is it that

we nevertheless do use the word

in them all—
that is
correctly?
How are we supposed to know that
they are all cases of expectation?
And
if we don't know this,
how are we supposed to know
where to look
for the common element before we fail to find it?

...

"In order to find the real artichoke,
we divested it of its leaves" ...
No wonder we cannot find what we look for.
Wittgenstein points out that
what makes a case one of [a certain] kind
is not

something beneath the surface

...

[38]

... , but

the "surface" itself.

In different circumstances different things
make us speak of a man as expecting someone, ...
What thus varies with the circumstances
("the surface")
is what Wittgenstein characterizes as
a criterion:
"In different circumstances
we apply different criteria”
...
[hence]
"the aspects of things
that are the most important for us
are hidden
because of their simplicity and familiarity" ...

Ya gotta admit, this pegs the definers-of-art pretty perfectly.

Even the paradigm examples come from such vastly different circumstances (i.e. from different art-forms) that the mere conceit to have applied the same criterion in all cases cannot be accepted at face value;

  and/or,
the definers have kept 'zooming out' until they reach a distance from which the blurry outlines of a symphony and novel begin to look the same;

  and/or,
there really is no philosophically unadulterated vision which registers a symphony and a novel as tokens of the same type; to turn that trick, it's abstract thinking all the way. And this is perfectly fine for entertainment purposes only, as used to be said by sportscasters when gambling was mentioned; but it hasn't anything to do either with the practice of art or with its ontology.

(If it's true that we merely apply different criteria in different circumstances, then maybe the polemic of Objectivism and Subjectivism is resolved by way of a Ramsey's Maxim solution.)

[But] We ignore ["the surface"]
and search for what is the same

in the many cases where we use the same word
because

we assimilate

the criteria for

the identity of kinds

to those for

the identity of individuals.

...
when we say that
in all the places he visited he found kindness,
though

in different forms,

we are speaking differently from the way we do
when we say that
in the different places he went to
he was followed by the same man, though

in different guises or disguises.

...

The mask, the clothes, the false moustache
are one thing,
the man in question another.
But this is not so with
the gestures, the words, the actions
in which, as we say,
kindness finds expression.
...

Kindness is its expressions;
it is not something behind them.

How Existentialist!

Of course,
one could say that
kindness and its expressions
in any particular case
are not the same thing.
But
only in the sense that
what are
expressions of kindness in some circumstances
may not be
expressions of kindness in different circumstances.
...

...

[39]

...

If Wittgenstein rejects this idea,

["that for a thing to be of a certain kind it must possess a set of characteristics which it shares with all other things of that kind"]

this does not make him a nominalist.
[It is] ... not to say that

we just happen

to use the same word in all these cases.
Of course

it is not for nothing

that we use the same word. ...

...

...

[40]

...

Seeing a face

under a certain aspect,

or a thing

as an instance

of a certain kind,
and seeing it

in relation to

other faces or other things:
these are

the two sides of the same coin.

For
seeing something under a certain aspect
is
seeing it as related to other things,
seeing it as

being like other things

in different ways.

When Wittgenstein
speaks of "family resemblances"

he is

offering a comparison,

using a simile,

which is

meant to dislodge

our idea

of how

things we call by the same name,

cases in which we use the same words,

are related

to each other. It is also meant to throw light on

the ways

in which they are related and on how

grasping these relations

affects the way

we see and understand them.

And in this capacity he does not seem to be talking about the genesis of any things but rather about the 'genesis' of relations, of same names, and the like.

"Family resemblance" is a simile, after all!

Here Wittgenstein talks of

relations,

resemblances

and also of

common characteristics.

But

what is it

for two things
to have a common property? ... for them
to resemble each other?
...

2. An aspect of Wittgenstein's attack on essentialism

So Wittgenstein opposes the idea that
"wherever we make a predicative statement
we state that the subject has a certain ingredient ... "
The alcohol ... can be separated, distilled,
and then one has pure or neat alcohol.
It is the presence of it
in these different drinks
that makes them alcoholic ...

The model for such neat ingredients
is to be found in mathematical concepts
and what it is they are supposed to mean.
Thus
in the Phaedo
Socrates rightly contrasts

the equality of two logs

and

the equality
of the sums
of two strings of number:

... it seems to him
that in the mathematical case
we have

equality pure or neat,

while
in the physical case
we have

equality adulterated:

"No two logs are perfectly equal."
...

[41]

...

One could say
that there is no such thing
as an absolutely accurate measurement; ...
...

... the term "equals"
is not governed here by strict criteria, ...
It seems that
where there are no such criteria,
where a word cannot be defined in terms of the
necessary and sufficient conditions of its employment,
its meaning has been only imperfectly determined
or not determined at all:
...

It seems that
unless we can extract the essence of gamehood
we do not know what a game is
and
neither can we explain it
to someone who wants to know.
...

Thus Plato hunted for the essence ... ,
and it seemed that unless he could capture it
he would remain in ignorance.
We have seen that Wittgenstein
argues this supposed essence
to be a myth.
I am now suggesting that it is

a shadow
cast on our language
by mathematics

and that
in rejecting the idea
of universals as common properties
Wittgenstein is combatting
the tendency

to think of language

as

an exact calculus.

He does not deny
that in some cases it is possible ...
It is not true,
however,
in a great many other cases, ...

[42]

But
is this all ... ?
I agree with Bambrough
that [Wittgenstein] is saying something deeper
...

If asked to explain what a game is,
we should describe games
and add:
"This and similar things
are called 'games'" ...
What is it we feel is missing here?
We feel
that we have not been given

sufficient guidance:

... in what other cases
[can I] correctly apply this word [?]
You have not

prevented me from going wrong,

...

I have been

given examples

but I have not been told

the mark by which I can identify them.

...

Is this not how Socrates complains
in the Theaetetus ... ?
He says:
We wanted "to find out
what the thing itself—
knowledge
—is . . . You do not suppose
a man can understand the name of a thing,
when he does not know what the thing is?"
Theaetetus agrees
and goes on to tell Socrates about
a mathematical problem
on which Theodorus was working with him.
...

...

... the mathematician proves
that something holds of all numbers
whose roots are irrational,
...

[43]

... He presents a formula ...

...

...

... [But] Wittgenstein asks us ...

how I am guided

by [such a formula] in mathematics [?]

Is it really different

from the way I am

guided by examples

... ?
Wittgenstein's answer is : No.
He argues that

rules and formulae,

and also

definitions, examples and illustrations,

can always be

meant
and so taken or understood
in a variety of ways.

Hence
the statement of a rule,
a definition,
the essence we have grasped in geometry,
cannot itself give us a criterion
of the right way to continue,
the correct way to use a word ...

...

Wittgenstein
argues that

we are in the same position

even where we can
extract an essence,

that it does not give us
the support we had hoped for.
...

...

"There is no essence or common structure
that can guide us infallibly
in our application of general words."
"There are no ethereal rails or tracks
that can compel us
in our deductive reasonings."
...

The virtue of Wittgenstein's contribution here
is
to reject these myths
without embracing the

[44]

Nominalist's and the Conventionalist's mistakes.
...

It is true that
where, in a particular case, ...
the essence is exemplified,
it follows deductively
that it is correct to use the term.
...

What we overlook, however, is that

we still have to
decide or judge

whether or not

the conditions in question
are satisfied anew
in each particular case.

Thus what we thought had been
eliminated
reappears.

To put it differently.
We

support
the application of the term
deductively

and

this pleases us.

We then notice
that
this justification

presupposes

that the conditions

from which it follows deductively
that
it is correct to apply the term

are satisfied

in the particular case.

But

the claim that they are satisfied

cannot itself be supported

deductively.
Thus in geometry, for instance,
we can say that

  if  

this figure satisfies condition E

  then  

necessarily it is a triangle.
But

  does it  

satisfy them
?
Geometry itself cannot decide this question for us.
This is what Bamborough means
when he says
that no concept can ultimately be explained
in terms of necessary and sufficient conditions,
that
in no case
can the application of a term
be ultimately justified deductively.

...

... Given that any inference
that accords with such-and-such rules
is a valid deductive inference,

does this inference
accord with
those rules ... ?

... even were we to admit that
all valid deductive inferences
have a common structure or essence, ...
we would still be faced with the question whether
this particular inference here
has the required structure.
And the question,
"what is it for an inference to have this structure?"
can in the end only be answered
in terms of
examples and their similarities to one another.
Since it is the instances that have to be mentioned
in explaining what it is for an inference to have
the structure necessary for its validity
Professor Wisdom speaks of this

[45]

procedure
(which he sometimes calls "mothers method")
as being more fundamental :
All deductive proof and justification
(he says)
in the end comes down to "proof by parallel cases".
This is directly connected with
what Wittgenstein has in mind
when he says that
for people to be

able to communicate
they must agree

not only
in

their definitions

of the words they use
but
in

their application

of these words
in particular cases ...
Since
it is only

the latter agreement

that
would show whether or not they agreed
in their definitions
[then] it
["the latter agreement", i.e.
agreement on the "application in particular cases"]
is more fundamental:
"Our rules (and definitions) leave loop-holes open,
and the practice has to speak for itself" ...

"Practice"

here means

what we do
in particular cases.

The difficulty of the Art case,
then,
is that
we do so many different things
with the term;
'different' across time
as well as space.

Danto takes a Hegelian stab at resolving space into time in Chapter 3 of After the End of Art, evidently with the aim of showing that such a synchronic lack of agreement simply masks the inevitable emergence, over 'time', of the diachronic essence which so pleases us. He looks specifically at the rudeness with which early modernism was greeted, and at the process of its eventual absorption into the Grand Narrative.

I think that no Wittgensteinian worth his salt could fail to see the basic significance of the response Danto finds to be typical of this reactionary moment:

  . . . "I don't know what it is,
  but it is not art" . . .

    (" . . . not painting" . . . )

What clearer example could there be of the simple absence of agreement?

But it seems that Danto's Hegelianism (or whatever it is) forces him towards an irreverence for what we do in particular cases and an over-reverence for essence. I hesitate to say 'an irreverence for practice', though that's also a fair assessment, and, for Dilman here, this literally means the same thing; But Danto forces us to read (so as to deploy) these two locutions noninterchangeably: 'Practice' means: 'particular cases' aggregated at scale so as to invoke the quantity-into-quality corollary.

As I read this work of Danto's, the analysis is much the same as that which Carroll gives in defense of Aesthetic Realism: we really "agree," no matter if we really evince verbal agreement. But there's something daft about putting all the blame on practitioners, on language, circumstance, bad faith, and so on, leaving all the credit lying around to be scooped up by whatever idle observer happens to be passing by. Hence both the reactionaries and the hagiographers of early modernism were "at some level . . . responding pretheoretically, so to speak." (p. 56) But this just is what we do in particular cases: In precisely those cases such as Danto examines here, where there as yet exists no 'theory' that would 'enfranchise' the radical art, this is indeed all that we can do. This is "practice." It cannot be idealized away, not even if the people and the statements in question, qualitatively, all leave much to be desired.

One thing to be sure to appreciate about the Wittgensteinian tack, then: The mere fact that there are different conditions doesn't lead to the selective bracketing of inconvenient cases as unideal in one way or another.

The further we go down the road of idealization, the more we simply beg the question of what counts as 'practice' in the first place. e.g. We can decide, as I probably would if challenged, that the "but it's not art" response is a simple projection of the anger that so often accompanies frustrated expectations: I am deprived of the simple pleasure I was expecting, so I will deprive the artist responsible of the thing their action most suggests they desire: recognition, feeling seen, etc. If we decide that's what's 'really' going on, or 'all' that's going on, in such cases, then it makes a certain amount of sense to decide, also, that this is no instance of 'practice' at all; that is, it is no 'condition' at all for deductive application of a term. But 'deductive application' names an ideal practice which is seldom attainable, and which is not often attained even when "conditions" are favorable. If this maneuver is permitted once, we'll never again find a single instance of the practice anywhere, because everywhere we look we will be stalked by base motivations, incentivizations, idealizations, and on down the line of familiar human cognitive and emotional frailties. Who is to say that the rage for deductivity is not similarly motivated, and that these motives are not similarly base?

Of course Danto is far too adept a philosopher to be caught stooping to the level of armchair psychology, but he could benefit from an infusion Armchair Behaviorism!

So
where a common essence or structure
can be really extracted
from the cases in which a general term is applicable,
this does not place us at a greater advantage.
In

this essence

we do not have

something superior

to

the instances themselves.

...

3. Classification and Reality

Bambrough rightly points out that
Wittgenstein rejects the essentialist form of Realism
without embracing Nominalism.
But
the positive view he attributes to Wittgenstein
of
what makes our classifications
and our use of general terms
non-arbitrary
is one which Wittgenstein criticizes.
... the way in which in the end he

separates language and reality

is antithetical to the whole tenor of Wittgenstein's thought about language ...

Bambrough's view is that while Wittgenstein
rejected the essentialist's view ... ,
he nevertheless retained
a place for "objective” justification.
With the nominalist he affirmed

the relativity of our classifications,

and
with the realist he also affirmed

their "objectivity".

...

Any

[46]

genuine classification
is and must be based on
["objective"] similarities and differences.
These provide

a common measure

between
different classifications belonging to different cultures
and, therefore, also
between the different uses of words
in their languages.
The

classifications and uses

of words are

relative

... ; but

not
the similarities and dissimilarities

...

...

[All of this] is very similar
to the way
in the Tractatus
"objects"
were meant to provide
a common measure between different propositions.
...

But it is well known
that Wittgenstein rejected his early views
on the commensurability of propositions.
His

simile of "family resemblances"

is

a rejection of just this.

Thus
if

what we call
"difference"

varies

from one language-game to another
and, therefore, also
from one system of classification to another,
how can similarities and differences
constitute the common basis
of different classifications?
... Bambrough ... seems to hold that
similarities and differences
exist apart from
our systems of classifications; ...

[47]

...

Applying "Ramsey's Maxim" to Bambrough
one could point out

an assumption he shares

with the Platonic realist:
Unless our application of general terms
has

a basis

in

something that is
independent of language,

it

must be arbitrary.

[For Bambrough] This basis is provided
by
the "similarities and differences
from which we may choose
according to our purposes and interests".

... Wittgenstein does not deny
that things we call by the same name
may have certain common properties.
Nor does he deny
that they may be similar to each other
in various respects.
Bambrough appreciates this ...
... The Brown Book Part II ... two important points ...
... first ... just as we cannot
ultimately explain our use ... in terms of
noticing [common] properties ... ,
we cannot do so
in terms of similarities striking us either:
"...   why shouldn't
what we call 'the similarity striking us'
consist partially or wholly
in
our being prompted to use the same phrase?"
This shared natural inclination or reaction
may

be what is

fundamental
and not susceptible of

a further explanation.

...

Bambrough wants to say:
"A man must know that these cases are similar
in order to be able to use the same word.
It is

this similarity

that

justifies him

in using the same word."
Compare with ... :
"А man must understand the order

in order to

obey it."

"He must know where the pain is

before he can

point to it."
...

Wittgenstein argues
that

understanding

is not

a source
from which the man's actions flow;

it is not

something
over and above

what he does.

Rather

what the man does

in certain circumstances

shows

that
he has understood the order;

his actions
are
a criterion

of his having understood ...

How Behaviorist!

N.B.:
  our purposes and interests
  re-appear
  with every appearance of a criterion.

We might apply the Principle of Least Effort here:
  it is much more work
  to demand criteria
  and (doubly so)
  to determine if they have been met
  than it is
  for the social water
  to run downhill.

And then,
  understanding is
  not
  something over and above
  what
  one
  does.

So,
  the rage for criteria
  just is
  good ol' purposes and interests
  making a curtain call
  over and above
  just as everyone is getting up to leave.

As for his justification for doing what he

[48]

does in reponse to the order,

N.B. A criterion here seems to belong to the observer, while a justification belongs to the person thusly observed.

Wittgenstein says:
"When someone whom I am

afraid of

orders me to continue the series,

I act quickly,
with perfect certainty,

and

the lack of reasons does not trouble me" ...

"If I have

exhausted the justifications

I have reached bedrock,
and

my spade is turned.

  Then
I am inclined
    to say:

'This is simply what I do'" ...

... these remarks ... apply equally to

the noticing of similarities

and

being guided by them:

"We are treating here of cases
in which . . . the grammar of a word
seems to suggest the 'necessity'
of

a certain intermediary step,

although
in fact the word

is used

in cases
in which
there is no such intermediary step" ...
Thus
Wittgenstein rejects as

an "unnecessary shuffle"

not only

"intuition",

"grasping the essence"

cast in such a role,
but also

"noticing similarities".

Unnecessary shuffle:
this means that

sometimes there is

an intermediary step
and

sometimes there is not.

But it is not necessary,
and it is an evasion,
for

when it is present

it solves nothing,

the same problem reappears.

The writing is a bit choppy here even without me chopping it up. I presume this means: The noticing of similarities cannot be principled; rather, it is motivated and ephemeral. Therefore, we have the same problem with classes that do share a common property as with those that do not. Namely: We simply do not ever "notice" very much at all, and when we do notice something, this really is a fact about us ourselves, and it is no fact at all about the object of our interest.

The second point ...
the contrast similar/dissimilar,
like the contrast simple/complex,

has no content
in separation from
the language-games in which we use it:
"What do these colours
[e.g., "light blue" and "dark blue"]
I am pointing to have in common?' ...
The answer to this
really ought to be
'I don't know what game you are playing.'
And

it depends upon this game

whether I should say

they had anything in common,
and

what I should say

they had in common"
...

We could say that terms like
"simple",
"uniform",
"consistent",
"similar",
"same",
"exact"

are relative terms,
since
what may truly be called [any of these]
in one context
may count as [some contrary term]
in another.
...

Bambrough well appreciates that

different likenesses or resemblances

may be

made into criteria

of classification,
so that

our classifications vary

with

[49]

which likeness

we choose to make

into such a basis.
He recognizes

the relativity
of
our principles of classification.

What he does not recognize is

the relativity
of
our criteria
of
similarity.

These are relative

to
what it is

we are classifying
and, therefore,

to
the grammar in which

we classify things. They are also relative

to
our interests and purposes.

...

...

The idea that there are
(at least, or only)
similarities and differences ...
Wittgenstein has shown that
such "ontological" claims come from confusion.
...

Wittgenstein's reply
to Price's
"There are recurrent characteristics"
is:
"We,
in our conceptual world,
keep on seeing the same, recurring with variations.
This is how our concepts take it" ...

So

what we count as a similarity

is
relative
to

the grammar in which we make comparisons.

Here
similarities can neither give us

a common measure

for our different uses of words
and systems of classification;
nor can they
provide

an ultimate support

for these.
But

this doesn't mean

that
our use of words
and
our classifications

are arbitrary,

and that we have to abandon
all hopes ... of being objective ...

[50]

When I
judge this
to be
similar to that
in such-and-such a respect,

my judgement is

objective

if
it is

responsible to criteria

that are

independent of my thoughts.

✨ ✨ ☝ ☝ ☝ ✨ ✨

It is
not

my thinking that

they are similar
that

makes them so;

I could be wrong or mistaken.
...

But
what is it for them
to be similar?
...

Unless there were such a thing
as the correct use of words
we could have no concерtion
of an independent reality
by means of which
we try to be guided
in the judgements we make, ...
Therefore,

if anything could

be said to

underline the non-arbitrary use of words

and to

make genuine classifications possible,

it is what Wittgenstein calls

"agreement in judgements",

and
not

"objective similarities".

...

"Some people
when asked to sing a note
which we strike for them on the piano,
regularly sing the fifth of that note.
That makes it easy to imagine that a language

might

have one name only
for
a certain note and its fifth.
On the other hand
we should be embarrassed to answer the question:
What do a note and its fifth have in common?
For of course
it is no answer to say:
"They have a certain affinity" ...

The kind of thing that Bambrough wants to say
in general
doesn't fit this case well at all:

All notes
resemble each other in certain respects
and
differ from each other in others.
This is what makes them notes.
Our language picks out
some of these similarities
at the expense of others
and
marks them in the diatonic scale of seven notes.

UGH.

Perhaps this really is correct, but the example is poorly chosen and the interpretation is poorly worded.

I haven't read Wittgenstein directly or in full . . . but indeed, the above seems to makes a point about language and about names but without suggesting anything about universals: Namely, here is a peculiar domain (don't call it 'music'!) wherein some 'object' really is pervasively identified in two imcommensurable ways, without giving any clue whatsoever as to the reason why. Because we are so conspicuously lacking 'reasons' for the divergence in identification, therefore we are just as lacking in reasons why our language does (or does not) reflect that divergence.

It seems a poor example, since most notes on the piano produce an exceptionally rich overtone spectrum, and this threatens to serve as the principle underlying the divergence. Admittedly, it's almost impossible to convince the Craft Idiots who populate the ranks of professional music that such things as spectra and overtones exist at all. How could the most astute and experienced observers be (moreover, how could they remain) the most ignorant here? It's probably because 'professional music' makes an intense pragmatic demand in precisely this direction: that is, it demands the reduction and systematization of three-dimensional 'sound' into a two-dimenstional mental map of 'notes', along with (arguably) a further one-dimensional reduction of 'notes' to 'notation.' And of course no one is more enamored of the products of this system than I am, but it is substantially reductive in comparison with most forms of naive experience; and moreoever, it is learned, more or less readily (and usually less).

All to say: I'm the Craft Idiot here, so, sure enough, I've never quite been able to believe that aural 'misidentifications' of pitch could be sincere and principled reports of naive experience. One grad school classmate even told me that his 'synesthesia' was the reason why he couldn't pass remedial ear training. I have no idea how he could play his instrument so well if his 'ear' was so debilitated by this condition; I flew through ear training with very little effort, and still, as a player, I find my ear constantly letting me down. But then, it seems he really could not pass remedial ear training. We know at least that part of the story is true; and we know that he can play. Non-understanding also is not something over and above . . .

...

But the question is:
What are these similarities,
affinities which we ignore because
we don't hear them?
Can they be said to exist "objectively"
and
can we be said
to be blind to them in our reactions?

When we hear the diatonic scale we are inclined to say that after every seven notes the same note recurs, and, asked why we call it the same note again one might well answer "Well, it's a C again". But this isn't the explanation I want,

[51]

for I should ask "What made one call it a C again?" And the answer to this would seem to be "Well, don't you hear that it's the same note only an octave higher?"—Here, too, we could imagine that a man had been taught our use of the word "the same" when applied to colours, lengths, directions, etc., and that we now played the diatonic scale for him and asked him whether he'd say that he heard the same notes again and again at certain intervals, and we could easily imagine several answers, in particular for instance, this, that he heard the same note alternately after every four or three notes (he calls the tonic, the dominant, and the octave the same note)

(Brown Book, pp. 140-1).

The point ... is the same
as the one ... in §§ 143 and 185
of the Investigations ...

of someone being taught
to develop arithmetic series (e.g. + 2) ...
[he] is left to go on by himself and he goes on "1000, 1004, 1008, etc.".
His reactions differ from ours.
But

those reactions,

if they were shared,

could be the basis

of a different language—
a different arithmetic,
a different musical tradition.

Well then . . .

WHY
do you suppose
we share +1
and not,
in fact, +2
?

It cannot be entirely a matter of circumstance,
I think Dilman has already said why:

We
support the application of the term
deductively ... [but] this ... presupposes
that
the conditions
from which it follows deductively
that
it is correct to apply the term
are satisfied
in the particular case.
But
the claim that they are satisfied
cannot itself be supported
deductively.

To pile words upon words:

He seems to be saying that "circumstances" per se cannot be evaluated for their "conditions." 'Metadeduction' is not a thing.

i.e. Assume that any given "conditions" must permit some "deductive" application of some "terms." This would explain why-and-where we occasionally glimpse terminology that is "objective" in the colloquial sense. But even this overgenerous hypthetical can't guarantee that this objectivity 'travels', nor can it help us to theorize (or even try to guess!) just what it is that makes for a match between "circumstances" and "conditions." Rather, 'we know it when we see it,' but if we don't just happen to see it, we are out of luck: We can 'find' but cannot 'seek' it.

What can we 'seek'? There is no difficulty whatsoever in enumerating examples of practices which quite well could be different if only their peculiar bases (which we also have no trouble compiling!) happened to be shared.

Moreover, it is impossible to argue that there are no needs or purposes for such alternatives! In some cases the human race is left literally begging for these alternatives to materialize!

So,
either
"needs and purposes" are not meaningfully determinative of 'sharedness',
or
we really haven't any better understanding of our own basic needs than we do of the most abstruse philosophy.
('Either' or 'both'.)

Whither sharedness?? It does not seem to be up to us! Against all appearances, sharedness turns out to be responsible to criteria that are independent of our thoughts. Against all appearances, these latter "criteria" do come from us ourselves; they are largely if not entirely 'mental'; but they are not "thoughts." They are non-cognitive, pre-theoretical, pre-conceptual, pre-rational, etc. That's what I 'think', anyway.

... we speak of people
who do not hear the differences we hear
between notes ...

as being deaf to music—"aspect deaf",
i.e.
deaf to

the aspects of sounds

that

we hear.

But

the reality of this aspect

depends on

this reaction being shared
and on much else ...
Thus
if we can describe a person like Wittgenstein's pupil
as “stupid"
or ... "deaf to music",
implying

the existence
of

something
to which

he is blind or deaf,
that is only because of

the overwhelming agreement

of reactions

from which he is separated.

In Zettel
Wittgenstein asks us to imagine
a society of the feeble-minded
and
comments on our inclination to think of its people
as

"essentially incomplete"—

that is
as living in a world
many aspects of which
are a closed book to them.
He says that we imagine them "under

the aspect of disorder

instead of

a more primitive order

(which would be
a far more fruitful way of looking at them)" ...
His point is that
if there were a people
who
in their reactions resembled those among us we call
feeble-minded
and if
these people

evolved a life
and a language

of their own,
then in some

[52]

respects

their world would be different

from

the world we live in.

🤐 🤐 🤐

Some of the features

of our world
would
not be just

unmarked
in their language;

they

would not be features

to be marked at all.

I think there is something to this, and it has nothing at all to do with compassion or disdain for the feeble-minded. Reference to that group certainly vivifies the example, but the point of the example is not dependent upon 'intelligence' to serve as the differentiator among people. Rather, given the laundry list of psychic differentiators we can now produce, it's more amazing than ever that we can share anything at all. Perhaps most of those purported differentiators, then, are just bunk qua differentiators; but even so, presumably, 'strength of mind' is not the only robust one left standing.

The disorder-primitive order distinction seems like yet another important puzzle piece falling into place, re: 'objectivism' and 'subjectivism'. I presume that the variation of marking in language is plenty wide enough to create some surprising divergences? But not wide enough to proscribe some predictable convergences?

I said that
if
my judgement of similarity
is objective
then

there are criteria independent of my thought
to which it is responsible.
Independent of

my judgement,

but not of

the language I speak.

They come from that language.
They cannot, therefore,
give foundation to our use of words— ...

These criteria,

which can be

gathered from
our actual use of words
in particular cases,

are not themselves responsible
to anything more fundamental.
Bambrough thinks they must be,
or they would be arbitrary,
and
what they are responsible to
are

"objective similarities and dissimilarities".

But

what does "objective" mean here?

If it means that
we can
make objective judgements about them,
we are back to where we started.
For

what makes it possible

to judge them objectively
are

the criteria that belong to language.

If,
on the other hand,
it means that they are independent of language,
then
this is precisely what Wittgenstein denies.
...

Does this ... make
our criteria of similarity
arbitrary?

If Bambrough's example
of the way his class of alphas is constituted
is to serve as a model,
then
the answer is that
they are not arbitrary.
As Wittgenstein puts it of "counting",
an example of a methodical procedure:
"What we call 'counting'
is an important part
of our life's activities . . . (It) is not
simply a pastime . . . (It) is
employed daily in the most various operations
of our lives” ...
Не would say that
the particular ways in which we use words,
the routes we follow in our thinking,
our comparisons,
and our inferences—
these
"are of course bounded for us,
not

by an arbitrary definition,

but

by natural limits

corresponding to
the body of what can be called
the role of thinking and inferring
(of using words and carrying out comparisons)
in our life" ...

So
if we wish to understand
what makes the grammars within which
we make comparisons and classify things
"akin to what is non-arbitrary" ... ,

we have to turn our attention
not to

something independent of language,

but to

what surrounds our use of words.

I mean

the surroundings of our life

in which

a particular use of words
has a firm place

and also

its particular

[53]

identity.

This place
is

fixed by
the spontaneous reactions

we exhibit and develop
in the course of learning to speak.
It is thus that
the tracks we follow in our use of words,
and in the many comparisons we make,
do not hang in mid-air
but are "held fast by what lies around them" ...
It is

"the movement around them",

as well as

the way they are rooted in
our natural reactions,

that distinguish them from

arbitrary procedures.

4. Kinds and Formal Categories

...

... Wittgenstein ...
was

concerned to combat

a particular way

in which we are strongly tempted
to think about

what justifies

the general descriptions we use
in particular cases.

... He was not himeslf

putting forward a positive theory.

For I have suggested that such a positive theory
is open to some of the same criticisms
that apply to the theory
which it is meant to
replace.

If it were said
that family resemblances determine kinds,
this would leave out ...

that

what we count as resemblances

is

in turn determined by our "formal concepts".

And these formal concepts
do not themselves indicate kinds.
Wittgenstein ... already ... in the Tractatus ...
the distinction ... between

"what can be said"

and

"what can only be shown" ...

In The Blue Book ...
"The attitude towards

the more general

and

the more special

in logic
is connected with the usage of the word 'kind'
which is liable to cause confusion.
We talk of kinds of numbers, kinds of propositions,
... kinds of apples, kinds of paper, etc.

In one sense

what defines the kind are

properties,

like sweetness, hardness, etc.

In the other

the different kinds are

different grammatical structures" ...

Thus ... we classify plants
into trees and shrubs, and

[54]

trees into elms and oaks, ...

and also ... these things
can be classified in other ways.
We can ask of a particular tree
what kind of tree it is ...
Similarly for an animal: ...
What kind of animal it is
depends on what outstanding properties it possesses—
outstanding

from the point of view of

the zoological taxonomist,
or that of someone who is familiar with animals ...
Here we need to notice that

what is in question
is
an animal.

Before we can ask "what kind?"
we must be certain about this much.
What kind of feature or similarities and differences
will be relevant to its being
this or that kind of animal
depends on

"the kind of thing" we take "it" to be
at the start

of our inquiry.
We do not

start with "bare particulars"
and compare and contrast them

to group them into kinds
on the basis of those of their
"objective similarities and dissimilarities"
relevant to our purposes and interests.

As infants we learn movements
and the co-ordination of movements.

We learn to follow what we see with our eyes,
to reach for it, to touch, ...
We learn to see,
to discriminate between sounds,
in harness with all this.
By the time we come

to learn to speak

we will already have

come a certain way.

...

What we thus

acquire indirectly

may be described as

formal categories—

those within which
we begin to be able to think
and to follow other people's
reasonings and explanations.
Thus
by the time we can be said to have
an awareness of our surroundings
we can already be said to see things
under certain aspects.
...

We can ask, for instance,

"What was it I heard?"

and answer
that it was a clap of thunder.
But we cannot ask:

"Was that a sound—or noise?".

...

[55]

A clap of thunder

is a kind

of sound or noise.

Though this is not something we find out

by observation.

What we do find out is that
what we heard
was a clap of thunder
and not an explosion.
But

a sound is

not itself a kind of thing.

It is not an instance of something
I have learned to distinguish from other things
or to sort into kinds.
It is not
"a network of similarities,
overlapping and criss-crossing"
that

makes

a clap of thunder,
a screeching of brakes
and the bark of a dog
into sounds.

Wittgenstein would say that
"That is a sound",
"That is a colour",
"This is a proposition",
"This is a thought",
"This is a human being"
are not things that can be said,

except in special circumstances—
perhaps in teaching a foreigner
the meaning of these English words.

You cannot point to a colour:

"This is a colour".
For here there is no distinction
between

how the pointing is to be understood

and

what it points to.

Whereas, in contrast,
you may know that I am pointing to
one of the objects on the table,
but you may not know which one.
Here you understand

how my pointing is meant,

how it is to be taken,
but not

what I am pointing to.

...

It is this distinction that is not possible
where I say that I am
pointing to the object's colour,
not its size.
...

If we can be said

to possess the concept

of colour,
then we have certainly

acquired it indirectly,

by learning to use such words as
"red",
"blue",
"orange" ... ,
to compare and contrast
what we use these words to describe, ...

"Red is a colour"

cannot be said informatively.

Its

being a colour
appears in the use

of the word "red".
It is not

something that you know

in the sense that

you might have been ignorant of it.

Since
in order not to know it
you would have to

know what "red" means

and

not know that it is a colour.

...

[56]

... A word like "colour",

a formal concept,

signifies

something that belongs to our way of talking,

not to

what we talk about.

One could say that
it refers to

a dimension of comparison

rather than
to

an object of comparison.

Wittgenstein would also say
that
you cannot

define formal concepts

without

circularity.

You cannot say what
a number
or colour
or proposition
is,
as you can say what
osmosis or photosynthesis
is, ...
Here the point is not that you cannot capture
the essence of colour or of proposition,
or that there is no essence to capture.
In the Tractatus, ... Wittgenstein spoke of

"elucidations"—

in contrast with

"definitions".

These necessarily

presuppose an understanding

of

the very thing
one is concerned to understand,

namely
features of the logic of our language ...
So
you cannot give an account
of anything pertaining to logic
because if, in order to achieve this,

you step outside it
you are speaking nonsense,

and
if

you remain within it
you are presupposing an understanding

of the very thing you are trying to explain.
This is one reason why Wittgenstein said that he

advanced no theories

in philosophy.
I hope I have made the reason for this obvious
in the case of a "family resemblance theory"
of formal concepts:

what we count as resemblances
is
determined by our formal concepts.

This is what Bambrough's theory of universals
presupposes but fails to recognize.
Yet
there can be no ultimate account
of the generality of any word
that leaves out the generality
which belongs to our formal concepts.
And

the peculiarity of their generality

... is that it

cuts across different grammars.

Wittgenstein was especially concerned with
the generality of these words or concepts
when he coined the phrase
"family resemblances",

and in particular
the generality of the concept
proposition.
In the Tractatus he had given
a general account ("elucidation") of "proposition",
the general form of propositions.
Now
he was
combatting the misunderstandings involved in it,
emphasizing
the incommensurability of different grammars.
Here to deny that cases where we use the same word
have a common essence
is not only to deny that there is

a set of properties

necessary and sufficient for the correct use of the word.
It is also to

[57]

deny that

these cases are commensurable.

This is another aspect
of Wittgenstein's attack on essentialism.

So here
we have words
which

we use in different grammars,

yet

they belong to the same language.

This raises new questions, ... :
In what sense
do words like "proposition", "truth", or "between"
have the same meaning in different grammars?
Well,
with a word like "red" or "table" or "dog",
once we have gone through the initial stage of training

we go on by ourselves

and

apply the word in new cases.

We could be said

now to recognize
a new variety

of table or dog as a table or dog.
It is not like this
with words like "truth" or "between".
Here
we learn

new uses

of those words.
...

One can recognize, for instance,
that 6 is between 5 and 7, or
that orange is between red and yellow,
and spontaneously use this word in these cases
once one has learnt arithmetic and colour-language.
But until then,
nothing that we have learnt
in learning the use of the word "between"
in connection with the language of spatial location
can enable us to use this word in these new cases.
Until then,
the new cases could not even exist for us.
Here
"new uses of the same word",
"new cases in which it is correct to use the same word"

doesn't mean

what it means
with such words as "dog", "table" and "game".
...

When Wittgenstein speaks of
a family of cases
and of family resemblances
in connection with language-games

he is no longer concerned with

the different things and activities
we use language to describe and compare,

but with

the measures of comparison we use
in those descriptions.

It is, therefore, a mistake to think of
the problems with which he is concerned here
as belonging with the traditional problem of universals.

5. Conclusion

...

I have

[58]

argued that there is no ultimate explanation
of our use of words
in terms of the features of these cases
or their similarities:
...

What we count as a common property or similarity
(not what similarity we select out of a common pool)
varies from one context to another and depends on
features of the culture to which we belong
and the language we speak.
Thus

the generality of our words

does not mirror

any feature of
a reality independent of our language.
But this does not mean that
our use of words and our classifications
are

arbitrary.

Bambrough rightly rejects
the idea of "natural kinds"

and appreciates an important aspect
of Wittgenstein's attack on "essentialism".
But I have argued that
his idea of "objective similarities"
keeps him wed to a form of realism
which Wittgenstein equally rejected.
I have also suggested
that Wittgenstein's attack on “essentialism"
has other aspects which go beyond
the realm of the problem of universals
and raises new questions.
...

NOTES

...

 

---