“I’m so bored. I hate my life.” - Britney Spears

Das Langweilige ist interessant geworden, weil das Interessante angefangen hat langweilig zu werden. – Thomas Mann

"Never for money/always for love" - The Talking Heads

Friday, May 27, 2005

Atlas finally shrugs

Last night LI wrote Paul a semi apology. Starting this series of posts three days ago, we intended to obliquely angle into Paul’s post on ethical individualism. However, we admit the degree of obliquity seems a bit, well, excessive. An unkind critic might call it multitudinously losing the point. Paul wrote back:

“Yeah, I was gonna write you an email from work today, with the subject of "Atlas - start shrugging"! I'm not sure what crazy scheme you have in mind - neither, apparently, do you (though your daimon does!) - but I look forward to reading the resulting opus.”

LI will sniffily ignore the reference to that appalling novel and try to get down to brass tacks in this post. Paul’s post is an enthusiastic appreciation of a book by David L. Norton entitled Personal Destinies: a philosophy of ethical individualism. We thought this was among the best bits we’ve ever read on his site:

“Of especial interest is the fact that Norton understands his account to ground a kind of individualism - an "ism" in disrepute with both Left (collectivism) and Right (communitarianism). Those two poles are often likeminded in taking individualism necessarily to be of the "atomistic" (Hobbesian; sc., merely numerical) variety. Norton's eudaimonism claims to establish "qualitative" individualism: each person, ex hypothesi, is obliged to actualize an excellence uniquely his own; to live in truth to his daimon. The social entailment of this doctrine is the "complementarity of excellences," implying the need for counterparts. Hence an individualism is possible which at once celebrates independence and affirms interdependence and sociality of a kind.”

Our response to this has been to consider a certain set of adventures of the concept of the “ratio,” (oops -- the Germanic amplification of the genitive -- the curse of philosophical class. Sorry) insofar as the human individual is supposed to embody it. If one is to “live in truth” to one’s daimon, it is important to think about the various ways one usually lives – unlike some purists, we like Weber’s term, “lifestyle”, for this. The truth, here, seems to do double duty: it implies, on the one hand, some standard of authenticity to which one can compare one’s lifestyle, and on the other hand, it seems performative – the criteria of authenticity is not prefigured, but is constituted in the living. That doubleness isn't incoherent -- a set of truths can be constituted over time in such a way that future acts can be judged against it -- but it does imply a limit on one's liberty that may, in time, become onerous. No more lighting out for the territory, no more second acts.

One of the perennial philosophical worries is the degree of error inherent in these various lifestyles. This is why we think the match between Gigerenzer vs. Tversky and Kahnman is fascinating, and casts a certain light upon the qualitatively different points of view that are each haunted, in Paul’s view, by a daimon.

Now, this idea of the daimon is interestingly ambiguous in terms of its site. Where, exactly, is it? this parallels the question we have been pursuing – where exactly is the innate tendency to error – if there is one? Where, that is, is its systematic place?

The early moderns were all very anxious about error. However, until Kant, error was conceived as a thing exterior to the subject. Among other of his functions, Descartes malin genie embodied the exteriority of deception. Hume inflected this line of thinking in a way, insofar as he showed that induction was not logically grounded. However, his intent wasn’t to delegitimize induction – rather, it was to estrange us from our mania about the framework of error and falsity. Induction, being on the side of life or habit, couldn’t be turned off, or doubted in any practical way. All of which went into Hume’s project of showing that reason was and should be the slave of the passions. It is important to note that at the same time that the natural philosophers were worried so about the Pseudodoxia Epidemica, the rising merchant/professional class was increasing sensitive to original sin. But let’s bracket that circumstance.

It was Kant, I think, who first interiorized error as an inevitable formation of the reason itself. There’s a famous passage in the Critique of Pure Reason from the section on the transcendental semblence (Schein). That semblence is the idea that one can deduce how the world is (for instance, whether the world has a beginning or not) from what I would call logic – that is, a conceptual analysis of beginning. Kant writes:

“The cause [of the transcendental semblence (Schein)] is this, that in our Reason (perceived, subjectively, as the human capacity to know) lie fundamental rules and maxims of its use, which have the total appearance of objective principles, and through which it appears, that the subjective necessity of a certain conjunction of our concepts, supported by the understanding, can be maintained. This is an unavoidable illusion, as much one as the illusion, that the sea seems higher on the horizon than on the shore, because we see the former through higher beams of light than the latter; or, even more, so little as astronomer can keep the moon from seeming greater in its setting, even if he is not deceived by this appearance.”

It isn’t surprising that Gigerenzer, too, uses visual illusion as an analogy for cognitive illusion. In Gigerenzer’s work, the necessity he is looking for is ecological – what living function does illusion serve? – rather than metaphysical.

The notion of an error inside (the logical equivalent of Jim Thompson's Killer Inside Me) might seem, at first glance, to have nothing to do with Norton (and Craddick’s) qualitatively different demon. And yet that demon seems inherited from the most famous of all daimons – Socrates. And Socrates is definitely a corrective daimon – a negating spirit. It is not a constructive one:

In the Apology, Socrates says: “…something divine and spiritual comes to me, the very thing which Meletus ridiculed in his indictment. I have had this from my childhood; it is a sort of voice that comes to me, ("some divine (theîon) and spiritual (daimónion) [thing] comes to me...")
and when it comes it always holds me back from what I am thinking of doing, but never urges me forward. This it is which opposes my engaging in politics. And I think this opposition is a very good thing; for you may be quite sure, men of Athens, that if I had undertaken to go into politics, I should have been put to death long ago and should have done no good to you or to myself. And do not be angry with me for speaking the truth; the fact is that no man will save his life who nobly opposes you or any other populace and prevents many unjust and illegal things from happening in the state. A man who really fights for the right, if he is to preserve his life for even a little while, must be a private citizen, not a public man.”

If Socrates is speaking truly, then perhaps the daimon is insufficient to ground Paul’s desire that “an individualism is possible which at once celebrates independence and affirms interdependence and sociality of a kind.”

So -- this is the end of this series of posts. A null-set end? An irony? Not really. LI is neither playing the village explainer or the answer guy, here, but simply responding to an interesting idea with a bunch of his own questions.


k-mort said...

I may lack your wit and eloquence, but I think Kant's account of falsity is utterly mistaken, and like most of the ghost architecture of the Critique, nauseatingly bloated. Do you hold to the analytic-synthetic divide? If so, I think you would agree that the quality of truth and falsity of the analytical realm-- say that trig class you took before pinkos got a hold of you--is quite different than the inductive accounts of truth. That doesn't mean an analytical realm is necessarily platonic (or theological, immaterial etc.); but merely that the means of confirming the errors are not a matter of fact but more of form; perhaps logic and mathematical notation could be altered and so forth but given certain axiomatic systems--a way of ordering sets and relations between sets-- certain things result from it: an error is due to not knowing the axioms (which themselves may be ultimately inductive) and thus not knowing the implications--necessary, at least within the context-- of the given input into a function. I guess thats a bit primitive for postmod or a Kantian but some sort of operational account works for most analytical issues, while remembering, or having Wittgenstein jamming his poker into our heads-- that one could also ostensively define functions, variable constants etc.: the syntax game is arbitrary but not the relationships

I do think the Kantians and idealists are generally wrong in viewing functions as something other ( eternal platonic objects??) than sense abstractions, thus math and logic as well are in a sense all varieties of physics, no? I guess I am arguing, via Quine (but not sure on Quine's somewhat hasty attempt to deny analyticity in regards to artificial languages as well as ordinary) for a thorough nominalism, from which the "analytical" axioms (and then truths and implications of various functions) are derived, obviously over centuries.

As far the status of inductive truths of natural sciences (and history, econ etc.)--since you asked-- I think Popper-Kuhnian views, modified, perhaps radically, by probability (and non-linearity, chaos, complexity, etc.-- those difficult quantum concepts that most philosophers ignore daily) are the most plausible. In many cases truth, therefore, may not be conclusively established, as in evolution which has been repeatedly modified over the last 150 years or so. This contingent aspect of truth is why the neo-cons and biblethumpers hate Kuhn--

k-mort said...

Apologies, yes, apologies are in order for rushing in and not reading the earlier posts regarding the Tversky issue; yet someone like Kuhn would say that inductive research of this type (perhaps also illustrated with Milgrams experiments) must be repeated often and with very large sample sizes (to wax a bit bogusly statistical). There's no reason that another experiment using the same prompts and language in other areas might not result in quite different responses. But I see where you are going: the populace IS manipulated into contrary or at least inconsistent viewpoints quite easily merely by how ideas are presented or "marketed"--but we might want to know about the class, education, racial makeup etc. of the people responding.I suspect a segment of the population from say Ann Arbor are going to respond quite differently than from Vegas. If the Ann Arbor segment is substantially lower in irrationality (just syntactically defined, by choosing the more probabile single predicate rather than the conjunction) than is the Vegas segment, the Ann Arbor humans should get to vote in national election and the Vegas people not: or maybe the test, or a more developed one, is administered at the poll....

roger said...

Damn. K-mort, I had a reply for ya, but the comment machine didn't take it for unknown reasons.
Anyway, I liked some of your points. Others I have problems with.

a. Basically, I sympathize with the nominalistic inclination, but I'm wary of the way nominalism often leads to logic chopping at the expense of knowledge. When I read, for instance, of some philosopher who doesn't know a coelacanth from a butterfly deliver a "crushing blow" to evolutionary theory from the point of view of "falsifiability", it makes me want to scream. However, I do think the nominalism as the instinct for runing testing procedures over propositions is a beautiful thing and essential to civilization.

b. I don't think K. & T.'s psychological tests are so easily criticized by referencing the narrow set of subjects. From what I understand about Prospect theory, tests in other cultures -- Japan, the U.K., Germany -- have come back with similar results. Of course, you have to make changes in the wording, but the structure of the problem -- the recognition of equivalences between sets, translated into probability talk -- remains constant.

c. Kant. Well, Kant is worth a whole post, or book, or library. Myself, I disagree with you about Kant's architecture being bloated. I think Leibniz's Theodicy is. I think Descartes Meditations are elegant, but that he sneaks certain concepts he needs in the back door -- it is systematically lacking. I think Hegel created texts that are like great rituals for some obscure, lost cult -- you have to absorb the symbols and vocabularies to get it, and even then you don't necessarily get it, But Kant seems to lay out his project without any excess that I can see, and make his points within a well organized sequence. To me, he is as compact as a twelve inch ruler. This doesn't mean he's right -- you could say we don't need a ruler, we need a thermometer, or a geiger counter. You might say that it is a mistake to go from experience to the world, and that that direction screws up everything. But his power comes from his systematic integrity. I reread Quine's Two Dogmas after reading your comments, and it seems to me that, as much as Quine destroys Kant's vaunted analytic/synthetic distinction, he still ends the essay on a Kantian note. Which is another way of saying that Kant is a very big guy to reject in toto.

k-mort said...

I am a bit out of my league on this, but it seems Quine is in some sense more concerned with the synthetic nature of definition and reference and not so much with the relationships: he addresses the constituents but not the relationships holding between the constituents. I will read it again.

Consider just a basic valid form such as Modus Tollens: p->q, -q /
-p. Is the conclusion synthetic? Defining p and q is synthetic, sure: what they refer to, but the conditional itself is not defined by reference to a fact; or the implication is not a fact in the same way that the variables are. Maybe that is sort of similiar to Kant's synthetic a priori, but at the very least putting the deductive implication on the same level as induction doesn't seem plausible.

There's also something about identity which might be grounds for analyticity that I think Quine slides over (tho I'm pretty sure he was aware of it): the copula is used in different senses; is "is" = in some contexts, but in others (any normal calc class), the = is more of a operator, as in "this results from this"--it's not the same as a semantic tautology, it it...I think Russell, however much he is now the detested hack, wrote some on this.

In Two Dogmas Quine does not really say much about set theory or the ontological status of tautologies other than the discussion usual "bachelor is an unmarried male."

k-mort said...

to be honest im really opposed in spirit to epistemological circle-jerking, even of the synthetic variety, and think most of us who have fallen among philosophers and so forth would do well simply to embrace a nominal materialism and genetic determinism and try to rise above being male nurses....the idealist-quine-kant- "what is the status of various forms of truths " does little to solve real economic and social problems: what Tversky has done is more valuable, though I think the issues are still quite similiar to what the psych people were doing in the 50s: what is pathopsychology--how is it manifested, how widespread, how to deal with it: I think widespread psychosis is quite a possible reality, as I am reminded every 405 traffic jam (or when I turn on the TV or listen to radio). When the war started in 2003 it seemed as if large numbers of neighbors and associates had turned into jingoistic nazis.

Societies-countries-corporations are being run by humans that should be on psych meds or in mental health facilities. What the F are shiite muslims if not psychotics. Yet I think say a westside dyke bar on friday night might be chock full of psychotics as well.

Paul craddick said...


I'm grateful that you saw fit to consider my post with a penetrating posting of your own. I hope I'm not intruding into the interlocution between you and kimort ... but I will offer a couple initial thoughts and questions of my own. My exposition isn't going to proceed systematically, I'm afraid.

It was insightful of you to reframe some of the entailments of the daimonic conception in terms of the ratio. I wonder, though, why you confined your survey of the endoxa to moderns; I see your point about the preoccupation with error being characteristic of titans like Hume and Kant, but the limning of fallibility (both practical and congenital) has a long provenance. Hume explicitly adverts to the Pyrhonnism of Sextus Empiricus; and one might recall the latent skepticism in Platonism, which flowered in the New Academy.
These examples, I believe, belie your contention that "until Kant, error was conceived as a thing exterior to the subject."

Now, it seems to me that viewing as problematic the notion of "living in truth" to a ratio can suggest at least 2 other questions: (1) How does one come reliably to know the putative ratio? (problem of moral principles, knowledge) (2) If one indeed "knows" the truth, how do we account for missteps, deviations, perversions, etc., along the way? (problem of practicability).

I wonder why Aristotle's reflections aren't germane? - if not to Hume and Kant, then certainly to the daimon. Roughly, in response to (1) Aristotle focuses on the role of habituation, especially through having good examples and upbringing; one "has" the principles viscerally before he has (if ever he has) them reflexively. To (2), the role of emotion, passion, and, generally, "moral weakness" is underscored.

To speak of upbringing and the like entails a concern with the wider social milieu; as the conditions for the realization for the best polis are rare, so will they be for the cultivation of general excellence. This, I think, is the way to harmonize Socrates' grim valediction in the Apology with Norton's apparently more optimistic view. Norton is perfectly willing to concede that most persons are morally frivolous, most of the time. However, he's enough of a modern to hold out hope that a new educational program, animated by salutary poesis and "myth" even, can do the job - evoking a more consistent, and widespread, excellence. As I wrote - the complementarity of excellences is "possible" (imaginable, envision-able); that doesn't mean it's likely. The best are always rare.

I'll have to put my quill back in the ink-pot for now, and hopefully weigh in with some more thoughts tomorrow. A parting thought: the "where" of the daimon is exactly where any final cause is - as an immanent metaphysical determination of an organic thing's existence, which, like all possibility, can most surely be reasoned to retroactively. That is, it is the ground of a thing's being.

roger said...

Paul, I thought you had given up on me, man. Thanks for the comments.

I guess one of the things I didn't make plain was that the kind of cognitive error that T and K are talking about -- and the geneology back to Kant -- reference the errors of perception (which were meat for Sextus Empericus, and figure, definitely, in Plato) as analogies to a deeper inability of the reason. But you are right -- Kant, as always, can be looked at as restating Plato.

I was leading you up several primrose trails to lead you, actually, to a Platonic point -- which is that Socrates, at least, sees his daimon as a form of negation -- a form of blocking error -- without it being constitutive. I suppose the latter role could be filled by a muse. In the Phaedrus, there does seem to be that latent conflict, and (as I remember) -- doesn't Socrates say that he is visited by a creative, rather than a purely negative, voice?

What I was suggesting is that the interdependence is forged by the spirit of negation -- by what we all can't do, in common, ethically -- whereas what we can do in common is not about the daimon. In other words, Socrates has good reason to say that the daimon merely negates. But this is the classic liberal in me -- as you will have spotted -- struggling to get free.

k-mort said...

Herr Craddick, though really not advancing any practical argument, puts forth some impression erudition and platonic terms, but I think he is wrong in trying to place Hume in with the ancients. Hume, whose thoughts are often really a refinement of Hobbes, held to a material causality though with reservations. His inductivism is far more relevant to biology and economics --any sort of empirical endeavor--than it is to philosophy.

The infallibilism has been attacked by numerous people, perhaps by none more capable than David Stove, even if we do not agree to all of his somewhat nihilistic conservatism.

Quine himself, however much the reductionist (or irritant to those who think Plato still has the answers to current problems) does in Two Dogmas put forth a pretty convincing arguments for an empirical epistemology. But he doesn't have the big stamp of Tradition on the frontipieces of his slim texts.

Infallibilism, verification, issues related to confirmation: those are the issues, not the status of truths or the attempt to justify a res cogitans or synthetic a priori. Germans may have spewed hundreds of pages delineating the "noesis" and then went to eat schnitzel and swill their Pils dinner each evening: yes obvious as phuck but something the German mind (and the people interested in keeping the Philosophy machine running) continually overlook. Whatver form the Ghost takes--gallic, teuton or classic greek-- he likes to eat and screw.

Viva Roger! (y comed mierda, Cradickito)

k-mort said...

Low See-into Señor Rogerio, if that is a trifle close to the grunt of the philistinus Americanus.

Here's a big issue with Kant, that I, low-rent Voltairean inductivist, discover when skimming through the 1st Critique: apart from the use of Aristotle's Physics categories (perhaps themselves sufficient grounds for Kant's inclusion into the Historical Interest Only category), the synthetic a priori argument is, as men more qualified than I have pointed out, resting on some rather flimsy support.

This is synthetic a priori according to Kant: "Every event must have a cause." Kant seems to be saying that the subject "Event" has within it the necessary predicate of causality, and that this is not something we learn by looking at the world. Yet as he says knowledge starts in experience. How do children learn this fact? THey learn the word and understand causality? It would seem the child does not undertand causality until he perceives it working in the world: he sticks a knife into the outlet and gets a nasty shock. Ouch--avoid it. And so forth. Not all learning is like this perhaps but most is. And learning the meaning of the word "event" means looking at something --actions in the world as well as definitions, synonyms. Where is the a-priori-icity to this?? Mathematical terminology does not just appear in our brain, and it is not unrelated to physical events. At the very least Kant's definition seems as much linguistic as ontological. TAke any basic truth of classical physics--say Newton's equation stating weight as inversely proportional to the square of its distance from the center of the earth--it is synthetic a posteriori ; at least it is not necessarily a priori. And if this is the case it seems the edifice of the Critique is substantially ruined.

Paul craddick said...


Thanks for the clarifications. I don't want to quibble - since now I see what your main concern is - but Sextus Empiricus doesn't merely entertain puzzles about the alleged betrayals of/by the senses. He would locate the grounds for skepticism in "the reason" itself (as arguably would the later Platonists, who followed certain aporia re: the forms to their logical conclusion, entailing that within the frame of the physico-somatic hypostase, no knowledge is possible).

The citation you gave from the end of the Apology is a good piece of evidence, but it needn't bear the interpretation you give, it seems to me. First, you're right that in other dialogues (Phaedrus and the Symposium, e.g.) the daimon isn't described thus. Second, the Apology statement is compatible with the view that Socrates' life exemplifed the tension between the two horses of the Phaedrus, so that his daimon was a constant corrective to the inclinations of his soul's rebellious steed. In Norton-ian language, his "empirical self" was constantly moving towards a kind of entropy (Socrates found self-unity extremely difficult).

And, more to the point, Socrates says that "if I had undertaken to go into politics [which his daimon prevented], I should have been put to death long ago and should have done no good to you or to myself ..." So, the daimon's restraint allowed Socrates to benefit himself and his fellows. That seems to undercut directly your view that "the daimon merely negates." For, by negating, it makes room for the philosopher's true beneficence - which seems rather "constructive" to me.

Yet at the same time this does underscore the danger of philosophy to the city, and the city to philosophy. So, I'd say Socrates' testimony has a more mixed character - vis-a-vis Norton's claims - than you seem to think.


I haven't read Hume's Treatise or Enquiry in about 17 years, but I'll wager that he explicitly aligns his skeptical conclusions with those of Pyrrho/Sextus. Of course Hume doesn't simply restate the ancient views - but there is a core similitude. Just out of curiousity (and if you've given it any thought), what's your intepretation of Plato's Cratylus?

k-mort said...

Professor Craddick--

I've not read Cratylus--I will if I find the time. Hume's Treatise I need to reread as well. Nonetheless I think the semantic turn--of the Tractatus, Russell's On Denoting, Quine's writings on reference and meaning, Tarski--provide more useful information on this issue than do the ancients or Hume.

BTW--do you feel there is no claim made here agaisnt Kant's synthetic a priori worth considering, and if not what is the argument to be used? The issue is not that some knowledge is synthetic but why is that example--every event must have a cause--a priori. The example is a truth about the world, about matter, knowledge of which would depend on inferences (correct ones). I think its wrong for a few reasons.

roger said...

Paul -- maybe the "merely" makes it seem that the negation is trivial. Sorry. However, your point that negation can lead to the seizing of one opportunity among a set -- instead of Epicurian apathy -- is, sorta, what I meant by the community of negation. To use the example most at hand and most noted, the ten commandments are predominantly negative -- don't do this and don't do that. The exceptions are interesting -- they are about 'honoring' parents and worshiping God -- positive counsels, but peculiarly empty ones.
It is my impression that the daimon always says no, or don't go that way, etc. While poetic rapture, the muse, the spirit of the hour (well known to bloggers) say yes, do this. I'll have to check on that. But the larger point is the doubtfulness attending any project to create a positive ethical subject.
Which brings us back, Kmort, to your subject -- Kant, with his notorious noumenal subject. When you say "Take any basic truth of classical physics--say Newton's equation stating weight as inversely proportional to the square of its distance from the center of the earth--it is synthetic a posteriori ; at least it is not necessarily a priori" -- I don't think Kant would say that it was a priori. What he would say is that time and space are the apriori forms of experience. That isn't the same as saying that everything about time and space is known apriori. The steps to Newton's claims can be retraced, and all seem to fall within the domain of understanding. What Kant does claim is that meta-claims about time and space,however, transgress against the method of understanding by erecting claims that the understanding is conceptually unequipped to judge -- hence, the antinomies of reason. I'll quote a chunka Kant, here -- Kant comes in big chunks, like peanut brittle:

For the sceptical
method aims at certainty. It seeks to discover the point of B452
misunderstanding in the case of disputes which are sincerely
and competently conducted by both sides, just as from the
embarrassment of judges in cases of litigation wise legislators
contrive to obtain instruction regarding the defects and
ambiguities of their laws. The antinomy which discloses itself in
the application of laws is for our limited wisdom the best
criterion of the legislation that has given rise to them. Reason,
which does not in abstract speculation easily become aware
of its errors, is hereby awakened to consciousness of the
factors [that have to be reckoned with] in the determination
of its principles
P 396
But it is only for transcendental philosophy that this sceptical
method is essential. Though in all other fields of enquiry
it can, perhaps, be dispensed with, it is not so in this field.
In mathematics its employment would, indeed, be absurd; for
in mathematics no false assertions can be concealed and
rendered invisible, inasmuch as the proofs must always proceed A425
under the guidance of pure intuition and by means of a synthesis
that is always evident. In experimental philosophy the
delay caused by doubt may indeed be useful; no misunderstanding
is, however, possible which cannot easily be removed;
and the final means of deciding the dispute, whether
found early or late, must in the end be supplied by experience.
(I haven't looked at the Germany. I gotta do some work today!)

roger said...

Oops. Make that -- 'at the German."

kmort said...

Yes there is something quite different assessing how humans perceive time as opposed to say how they perceive the trajectory of a body in space; I just don't think Kant's arguments prove much of anything. And if this is about perception as well as epistemology (not separate methods really), Kant's primitive, pre-psychological speculations may very well be outmoded. What about say how our eye functions in regards to out nervous system? Where is the syntax of this text realized biochemically and converted into thought? Kant is trying to conjure up this strange noumenal reality--and I think justify his theology-- simply by means of logic--aristotelean logic at that, yet the logic and the equations adn arguments routinely to refer to mental processes--ways of perceiving-- that are in no way instantiable.

I will go through some more of the Critique (though there are quite a few more or less insensible arguments such as the one above implying that knowledge of causality is a priori).

The entire Kantian noumenal concept that we cannot know the "ding an sich" has been repeatedly attacked by scientists, including Heisenberg and crew following the Copenhagen position on quantum physics.

kmort said...

'Scuzi pisspoor editing. I'll give up on the rants for a while, read up, work on my anti-metaphysical invective with assistance from David Stove, who tells us that reading the philosopher Titans from Plato to Kant and Hegel to postmodernism should fill us "not simply with pity, but with horror."