Deconstruction, Programming, and Humility

Editor's note: this post is a continuation of this discussion in Dan Read's blog about postmodernism, science, and software development, in particular this comment about Edsgar Dijkstra, deconstruction, and humility.

Of course, we have to beware of a capsule formulation such as deconstruction is humility, deconstruction = humility, or even in C# deconstruction == humility.

However, Dijkstra was postmodern-in-relation-to "modernist" computing fashions current in big development projects in the 1960s, founded as the latter were on the idea of a computing grand-narrative-in-code such as MIT Multics, and constructed by an older generation of men (who were mostly men and more important informed by midcentury male Modernism; Grace Murray Hopper was a preview of a postmodern feminist success at software development).

These men actually thought that the world is comprehensible in such a matter that a Great Big Mainframe could contain the correct encoding of a Reality. The fashionable, ersatz, "postmodern" pose is that they were damned fools, and under Brezhnev, much of the dissidence that brought the Berlin Wall down in 1989 started in computing centres (Natan Scharansky, currently Israel’s foreign minister, was an early cyber-trouble-maker in Russia). The actual sites, in the Soviet Union and the USA, had unexpected problems in coding "grand" narratives such as PL/I, which was intended to be the LAST programming language, if you can credit that!

It’s more complicated than this. For one thing, the end of the Berlin wall destroyed the livelihood of millions of decent teachers, doctors and computer programmers in the Eastern bloc who are called on to be "postmodern" about their troubles, which for them means to work at McDonald’s. This is bullshit.

Postmodernism is as much, in computing terms, a binary operator and in literary-critical terms the name of an incoherent choreography, a set of stances, gestures and poses. Modernism in art foregrounded, most especially in American abstract expressionism and Dutch de Stijl (Mondrian) the absolute gesture which I think caused us to forget that while it may be necessary to foreground the dance movement, we return in Postmodernism to the idea that "this means something" in the broadest sense: art, and trivially computer programming, bears a propositional relationship to something else.

Postmodernism in dance is a dialectical movement. To forget representation of that which is no more (such as the feminine personality represented by Giselle, the doomed consumptive) we enter the absolute gesture. The synthesis is in recalling its new meaning.

For example, a man in dance training is called upon to forget everything he knows and to make a gesture up out of the blue. The first gesture is usually something out of Kung Fu fighting or the boxing ring for that’s life for our boy in the big city. Then, if he perseveres he asks why not the yielding gesture to arrive at the necessary synthesis of strength and grace you find, for example, in Bruce Lee and Mikhail Baryshnikov.

In art, the referent of what was "abstract" can be an object in the real world, or it may be a psychological state or (as Adorno points out in Aesthetic Theory) necessarily "about" previous works of art; the art student learns, or had better learn, to engage tradition. In programming, I believe, "tradition" is neglected and would constitute starting in machine language.

Dijkstra’s postmodernism, on display in his 1972 Turing Award lecture, "The Humble Programmer" departed from the "new criticism" of taking for granted the crystalline intention of the designers and programmers of a data system and instead as did Derrida about the same time, foregrounded the programming text.

I do not want at all to overdo any parallel here between dance, and art, and programming. For one thing, I may be in a geometry where there are no parallel lines. For another, it is probably the case that constellations are nonpareil while at the same time having a necessary underlying unity imposed by the Kantian structure of the mind. People dance, people paint, people code programs.

While there is no "essence" considered as a lump we can subject to further analysis there is that point beyond which we cannot go, and at that point we find certain unities as well as certain differences. Of course, this is easily confused with what Rorty prematurely dismissed as a "glassy essence" in a markedly ignorant and insensitive book that proposed in a silly fashion to deconstruct what Shakespeare’s divine Isabella was saying in Measure for Measure about that which we violate when we lie.

Rorty thereby sanctioned the popular misconception about Postmodernism: that it is a license to party down and lie like Bush. News fa-lash: every philosophical movement has had in turn this reputation. Socrates was accused of corrupting youth, and Spinoza was used throughout the 18th century (as shown in Jonathan Israel's book Radical Enlightenment) as an excuse to party.

I repeat: while there is no essence considered as a lump we can subject to further analysis there is a limit, and enough, is enough. Into any vacuum rushes something. In programming, to return to programming, this is its fundamental scandal, that of correctness all the way down.

Much of Dijkstra’s output was an "arrogant" attempt, made in genuine humility, to point out that if programmers perform their analytical task of deconstructing the texts they write and those of their coworkers, we find fissures and aporias of meaning as in the case where a false (non-atomic) semaphore fails to mean anything between the time it is set and the time it is "simultaneously" interrogated.

Managers, famously, don’t read code, because this is to engage the so-called "user’s problem" in its horrible forensic glory, where the user may herself be wrong about outlier cases.

Postmodern literary criticism is "humility" insofar as it acknowledges not only the (Kantian) limits of our minds but also their empirical capacity for dealing with complexity. Postmodern programming is simultaneously humility and arrogance for it puts the programmer in the role of questioning not only his preconceptions (his "aporias") but in the most productive venues, those of the end user.

Extreme Programming decenters the grand narrative of the End User, for example, to gain in some cases actual results while the less than extreme guys are skulking in their cubicles trying to guess what the user "wants" (in some cases, as if human desire could trump mathematical correctness: see below concerning the divisibility of 100 by 20 for an example).

The effective development of an MIS system, where MIS systems happen as Dijkstra pointed out to have more complexity than scientific code, is best addressed in my experience not in a fawning, false, Uriah Heep type of humility that foregrounds "corporate needs" or worships "the end user". In calling on us to be humble circa 1972, Weinberg and Dijsktra were in fact calling on programmers to be less Organization Men and to understand the realities of limited resources.

For example, one example of a truly humble programmer in the Psychology of Computer Programming was a young man who rewrote an auto design system to handle option combinations as data and not code. He wasn’t "humble" in corporate terms, because he "offended" the original programmers of the original mess and their managers by his very act. His "humility" was in face of the engineering facts, that selection of options for auto assembly is irreducibly complex.

Here, the End User(s) was (were) the engineers whose perspective is necessarily silent on how you program a computer to combine options in a sensible fashion. Yet the programmer's decision was the best one because at a certain point, logic becomes data, and the engineers were empowered by his choices...which in many contemporary shops would demonstrate a termination level of "insubordination", "lack of [false] humility", and a failure to Listen.

Here’s another simple example, promised above. Shortly after First Chicago Bank merged with Bank One, I tried to withdraw 100.00 at my ATM. However, the withdraw was declined because, the machine "said", 100.00 is not divisible by 20.

The "user" may have insisted upon this "business rule" during the stressful period of merger but to treat her considerably less than grand narrative as crystalline, with a false "programmer humility" is to code a lie.

Programmer "humility" was completely misunderstood, and Dijkstra had in fact a reputation for academic arrogance.

This is because shortly after "structured programming" was unleashed, a confusion occurred between code and reality.

As a gesture, code is always about something, even if it is self-reflexive operating system or compiler code (for the same reason that Kant thought we externalize ourselves when we reflect on the contents of our minds in "self-consciousness"). Dijkstra’s goal remained getting the relation between code (and algorithmic mathematics) and the scientific or administrative reality correct.

But sadly, structured programming confused the map with the territory. The attractive quality of "programmer humility" was answered in the Reagan and Bush era with a management arrogance that is completely out of control; it’s on display this month in Vanity Fair (the one with Paris Hilton on the cover), because while one and the same time being oh, so very, antiwar and "liberal", Vanity Fair celebrates Larry Ellison’s rage for bigger and bigger and yachts, using a fortune that we can safely assume represents the unpaid overtime of thousands of Oracle developers who at this late date know more about relational databases than the Larry man.

[Vanity Fair seems to be in the sack with Ellison, who makes much sexier copy than Bill Gates. The problem is that at the same time, it is highly critical of Bush and the war. Larry Ellison, according to a recent bio, is pro-Bush and pro-war as are most of the heterosexual males in our society who are rich and famous, ultimately because of American economic power, enforced where necessary at gunpoint. This illustrates the most morally serious problem the Islamic world has with the west: our divided soul.]

[As to Paris Hilton, I don’t see the attraction. To me, she is not as cute as Elizabeth Berkeley, the actress now maudit whose career was ruined by Showgirls and who now flogs workout DVDs.]

"Simplification" of data systems became to mean removal or non-implementation of necessary features to the extent that today many data systems, considered as a forensic gesture in relation even to business needs exclusively, are a caricature of those needs which externalize real costs on the people who can least afford them, and in the case of health insurers who demand as a condition of insurability that the applicant/supplicant have an address that matches a preconceived syntax.

Y2K happens today to people who apply to FEMA for help after New Orleans but cannot provide a sensible address that fits the syntax of street and number and city, state, and zip because their home is under several feet of water. Now, I realize that data has to have SOME format.

The problem is that centering the "end user" designs the system around the convenience, at best, of some corporate drone or suit, who is in charge (let's say) of medical aid for the aged and poor but who doesn't know the difference between Medicaid and Medicare (this happens; Paul O'Neill, former Treasury Secretary in the first Administration relates horror stories about decisions made on behalf of the poor by men who literally did not know this difference).

An ethical postmodern would say that the end users are Legion and must include, if not the wretched of the earth, at least the guy who has 100.00 in his bank account and not 200.00 (which the Bank One system did agree is modulo 20).

This occurs, I think, because as Adorno and Horkheimer knew, modern societies rely on their functionaries to reify and simplify complex phenomena into labels which are then used as if "simple logic" applied to their referents…as in the case of "weapons of mass destruction" and "terrorism" itself.

Real Programmers visit this terrain all the time because they get to see, and are responsible for, the factoring of the reified symbols into bits and bytes which seen correctly are isomorphs of the people with whom in Hong Kong I swarm to work.

"Men and bits of paper, whirled" is T. S. Eliot’s poetic image of our unimportance and isomorphism to single records in a data base, and it has a rather disturbing relation to what fell from the World Trade center on September 11. At the same time, Enlightenment and its recognition of equal rights is what makes us a single record or a single sheet of paper. Before that we were shadowy images of hanged German Catholics in Jacques Callot's etchings of The Thirty Years War.

However, Enlightenment is not the end of history. Adorno was famously concerned that what he called "the administered world" was a second form of nature insofar as we use vernacular expressions to console ourselves when (as programmers often do) we encounter irrationality. We do need to hold in our minds what our grandparents might say, which is that the prison of the administered world was meant in good faith to make us free. In this connection, I love, as an American expat from an America without much of a safety net, to listen as Western Europeans complain about their Nanny state while at the same time taking it for granted.

Consider the programming job-seeker. Adorno wrote that the job-seeker confronts the social abstraction "society" unmediated by a role much in the same way the explorer or hiker confronts a reciprocal first nature. As his position in society becomes concretely untenable through phenomena including homelessness or lack of SIM card time that make his efforts exponentially more difficult, Job’s comforters are apt to use phrases such as "life is tough in the big city" or "you have to learn to adapt to the real world".

In the latter phrase, it’s necessarily ignored that the real world is a collective creation of people and the "real world" becomes massive and sublime, like nature.

As "productive forces develop" they in fact turn against the entrepreneural job seeker who finds, as have programmers, that there is no exit as previous from market relations. The early attraction of programming jobs to many actual programmers was a fairly sordid business. It stemmed from the fact that productive forces and their market don’t arise from nothing but from "primitive accumulation", pre-economic activity.

"Primitive accumulation" can be anything from sacking cities, taking Indian lands, or using Harvard’s mainframe without authorization to simulate the Altair’s runtime environment. Without judging it, we can say it’s what people do before markets. We are trained to think of this as subsistence agriculture, or hunting and gathering, but no path exists from either of these two Arcadian pursuits to markets and if we consider primitive accumulation as a recurrent phenomenon they don’t apply.

Larry Ellison in conditions of primitive accumulation realized he was onto something Big when, almost overnight around 1975, the computing world realized that the Codasyl model of data (a ridiculous contraption, which assumed that hierarchical relations are fixed in the real world) was about to give way to the relational model which reflects the postmodern fact, that "all that is solid melts into air" while enabling us to continue to do business. Larry Ellison was brave enough, and enough of a selfish prick, to get CIA contracts and have other people work ungodly hours under his guidance.

However, a society of The Apprentice, where you are "mentored" to be a selfish prick by a selfish prick and now a convicted felon (Martha Stewart) is under Spinoza's analysis one that contains a real contradiction and isn't ethically sustainable (or, owing to the externalization of environmental costs in global warming, physically sustainable).

Many more programmers simply were able to use force and fraud to hold organizations effectively hostage to several generations of software. As applied "scientists" they effectively were able to reify the scientist’s relation to power, where material power perceives an advantage in applied scientist and the scientist gains some power as a result owing to his knowledge. In the reification, the "knowledge" is the programmer’s relation to code, something more concrete than a PhD in physics which isn’t subject to the vagaries of Congressional committees as were Oppenheimer’s qualifications when in addition to supporting the effort to build a fusion weapon, Oppenheimer questioned whether it made sense.

However, this form of reified hostage taking creates a market. Precisely to the extent that early programmers held companies to ransom using their ultimately trivial knowledge of code, companies created a market for software tools including Oracle which would liberate them from "code". Of course, the liberation was more code, which nonetheless didn’t make the process so much futile (for when a company shares a data base with many others, it gains from shared expertise) as self-repeating over time.

Indeed, Open Source may represent the limit point of this process,where the company can today access not only entire operating systems but also accounting and other business packages which are ironically written by people who have no clue about the (formerly hypostatized) "needs" of the company and its "end users". For one thing, under Sarbanes-Oxley (the law requiring CEOs to sign off on the transparency and quality of their financial reporting), the use of an (unmodified, of course) Open Source general ledger package is a point in the CEO's favor.

As a recurring phenomenon, primitive accumulation is the discovery, or creation through state or military power, of a market, and it was indeed Golconda for some, a Golconda which over time commoditizes what were diamonds and results, over time, in the immiseration of the end producers in the name of the end users at the far end of the supply chain.

Consider, for example, the contrast between a man who discovers diamonds in Africa in the 19th century and a typical African miner today. In the true myth of primitive accumulation the former becomes Bill Gates rich, while through myths and through stories we accept that today’s African miner is paid a subminimum wage…essentially for the same activities (including discovery, albeit under close supervision) as the 19th century hero.

Such a parallel may sound just silly, but B. Traven, the American novelist of the early 19th century, whose novel The Treasure of the Sierra Madre became a famous John Ford and Bogart film, has one of his characters spell it out: the old prospector says that the labor of the unlucky men went into the value of the product of the lucky men. The gold’s value, in the old prospector’s demotic labor theory of value, is the labor of thousands of prospectors.

This means that Windows’ value represents the value of various turkeys including Microsoft "Bob" (an early stab at agent-driven operating systems) and the non-turkey OS of the Commodore Amiga, so famously before its time.

Value is a collective and social product, but we agree under the threat of force to allocate it unevenly, and the result is today the immiseration of Africa, a source not sink of value and the enrichment of sinks of value such as Manhattan. I’d call this "unfair" but it’s also the world we know and as such has the irreducible fairness of a world for which we are responsible.

In 1970, computers were quite common as mainframes but it took primitive accumulation for programmers to see that the secrets they’d learned at large corporations could be transferred to Intel chips.

Thus, between 1973 and 1980, I witnessed a strange reprise of the history of early software in the 1950s, that I’d read about in Saul Rosen’s Programming Systems and Language and other texts, and that I myself had repeated to provide my university with a Fortran compiler and a primitive data base. History does repeat itself, all the time.

The job-seeker, however, has to ignore "that man behind the curtain" and reify without thought a pre-existing second nature, in order to take the interviewer seriously. In programming, the ability to redefine, to create the second nature is fully withheld when (for example) the interviewer sharply distinguishes between C and C++ without the interviewer being able to code a null main procedure or class in either language.

The humility here is a false humility, meant seriously by no-one and in the nature of an agreed-upon scam, because the job seeker is primarily asked in an administered world to show that he subscribes to the proper reification.

Now, in writing the above, I do not mean to have a Bad Attitude. One reason for the fact that I think I’m on-target is that I’ve played the game with as much honesty and enthusiasm as I have been able to muster, for as Bertrand Russell said somewhere, the needs of children impose responsibilities that skepticism does not dare question.

In fact, I’ve used deconstructive theory in applying for jobs. I would like to present my background accurately without making one of those common resumes that is best described as "no more than one page, with your name, address, telephone number and a load of bullshit".

I’ve realized that despite the fact that mathematical and scientific types have nothing but scorn for stories, for narrative versions of the truth and for multiple competing narratives, a man’s professional life is not some simple, ultimately reifiable thing but a massive series of "bits and bytes" which he has to narrate. He realizes that competing narratives exist but sees no dishonesty per se in the corporate game of presenting positive facts.

This troubles programmers who on and off the job want the narrative of professional success (failure being unmentionable owing to lower middle class omerta to programmers who cannot, typically, believe that the author of two books on Java is unemployed) to be a single (grand) narrative.

But it doesn’t take a Dijkstra to see that the success or failure of a biography, or of a data system, consists of multiple competing narratives (consider the religious wars about programming languages) because the thing in question is a constructed entity with thousands, if not millions, of bits and bytes.

Reification in the large without deconstructive reflection is in fact the Lie made flesh…whether it is the reduction of Marx and socialist hopes to a tinpot Leninism, the definition of "terrorism" to mean "what scares rich people", or "a simple matter of programming".

For Dijsktra, "managing complexity" didn’t mean ignoring complexity. It didn’t mean giving the US Army Corps of Engineers 20% of what they needed to protect New Orleans from a flood that was predicted (with eerily prescient illustrations) in National Geographic. It didn’t mean "thinking like a manager and not an engineer" about applied alloy physics on the eve of Challenger’s launch in 1986, or foam chunk size probabilities on the eve of Columbia’s launch in 2003.

It didn’t mean bitch-slapping programmers in meetings because they had an honest concern about error handling or the simple fact that 20 goes very nicely into 100. Nor does it mean agenda setting to the degree where a serious concern about pre-existing bugs is equated to goofing off.

Instead, it mean adequacy-to-the-object in Adorno’s sense.

Much of Modernism can now be understood as the use of inadequacy. I now understand that the Impressionists painted "alla prima", with a single layer of color on top of a white ground, in order to get to market quickly, while Salon-subsidized artists of the traditional French academy were able to continue to use the highly indirect (structured) methods of first painting grisaille (black and white) and then glazing colors.

[John Berger, in Ways of Seeing and The Success and Failure of Picasso, is about the only art critic I know who realize that much art practice can be explained by the fact that artists gotta eat. Berger was a Marxist. Too bad.]

Adorno saw correctly that while the deliberate crudity of Expressionism and Fauvism was adequacy (what he also called truth) its very lack of complexity meant that the next step, in music, and after Scriabin and Mahler, was a return to prettified compositions such as are found in Stravinsky, a sort of pandering to the bourgeois that had survived the horrors of the First World War and whose "return" to "eternal values" horrified Adorno because you don’t "return" after the Somme, much less the Shoah.

Adorno was horrified by the fact that audiences welcomed Stravinsky’s "classical" symphony, which as a musician Adorno savaged as feeble even on its own terms, necessarily so after Verdun (Adorno had the personal decency not to ignore the German veterans of horror begging, as Vietnam veterans beg, in front of the concert hall). He also saw that the barbarism of what in America was "the return to normalcy" was prefigured in Stravinsky’s Sacre du Printemps, in which Mahler’s hope is replaced by the Chosen One who dances herself to death.

Adorno saw this as a dialectical response, but unlike Hegel he didn’t find the dialectic a necessary ascent.

For this reason, Adorno preferred a form of "structured programming" in music, the famous twelve-tone row. It is of course silly to ask if the musical compositions (such as those of Berg and Schoenberg) were "correct" in the way a computer program is correct, but "structured music" did produce an ongoing backlash.

My father was a classical music fan who hated all "classical" music after a point about which my father was very precise: he said it was "the end of the first act of Parsifal". This is of course where Parsifal must leave the scene of innocence and encounter Kundry and my father was being honest in his way about his limitations.

Nonetheless, I noticed even as a little kid that while my father would snap off the radio if the local station so much as played Mahler’s Resurrection symphony, he would sit through movies and TV shows in which the sound-track was distinctively Modern. In fact, I was dragged rather inappropriately along with my brothers and sisters and Mom to rather inappropriately adult films including The Bridge over the River Kwai in which I found the sweaty Asian sexuality disturbing and whose sound-tracks were demotic Schoenberg and unprettified Stravinsky.

However, my father’s opinion is shared by many classical listeners, especially those without training in the actual production of music; just as "computer literacy" avoids actually teaching machine language programming, and just as "art appreciation" is silent on the very interesting story (revealed recently by David Hockney in his book Secret Knowledge) of art technique, music appreciation is taught as a set of correct attitudes, from open-mouthed respect for war horses to the equally shallow rejection of "Beethoven and Brahms" in favor of "original instruments".

For an amusing essay on the consistent shallowness of this constellation of attitudes, read Adorno on "Bach Rescued from his Devotees".

To "appreciate art" learn how paintings are made.

Like Yogi Berra, you may already appreciate music, for the gnomic catcher of the New York Yankees said, "I would like to go back to college and study, but I would not study music appreciation because I already like music". If not, learn how to produce music.

To become "computer literate", learn how to program, preferably in binary machine language, and to use compilers with maximal effectiveness buy my book and Aho, Sethi and Ullman.

Absent this appreciation of creative labor, a dialectical backlash sets in in music, where the (uneducated) "end user" or music audience is enraged at the very idea that producers of music might take counsel in secret, in the academy, and conspire to produce wrong notes.

Keywords in the dialectical backlash are eerily redolent of the dialectical response to structured programming, for just as programmers (like Dijkstra) who insisted that they had a novum organum were coded as arrogant, a theme in music criticism of the middlebrow sort is that modern composers form a precious academic guild who don’t realize the needs of the User, that is the wealthy and middle class public who ignorantly listen to "high class" music (about whom Adorno was most amusing).

This parallel can be stretched too far, but in many ways the reaction to Dijkstra’s insistency on adequacy paralleled the resentment to modern music. And, end users seek refuge in nostrums and the deprofessionalization of the programmer (whether proud or humble) for socially the same reasons that, as Time magazine reported in 1988, an audience of "classical" listeners reacted to Phillip Glass’ "postmodern" return to pretty harmonies and a structure as easy to comprehend as be bop a lula (she’s my baby) with happy relief that they were spared Moses and Aaron or Transfigured Night, those same people on the job are reassured by promises that new paradigms will mean they don’t have to think anymore.

I am writing at the crisis, probably, of a United States whose elite asked its midlevel, its intelligence analysts, journalists and programmers, to ignore that man behind the curtain (or that Class 5 hurricane). I logged onto Yahoo on August 29th and was astonished to see, for the first time, a tightly wound monster hurricane with a perfect eye because my daily reading of several newspapers had found no mention of Katrina.

Bush, who was probably in the bag at his ranch, had not made preparedness front page news.

In the words of the September 11 official report, "the system was blinking red" AGAIN and again Bush was out to lunch and/or back on the sauce. But because the administered world had no cheap answer to a profoundly exogenous event (that was created by its externalization of environmental costs in global warming) the signals were ignored. I am certain many analysts at NOAA wanted to get on the horn but today, just as programmers are silenced by layers of bureaucracy, these guys "thought like a manager and not an engineer", the analysts’ concerns were not transmitted to FEMA or the local authorities.

A pre-Aug 29th narrative was agreed upon in the same way independent epistemologies about data systems are discouraged and programmer lesson one is that "be humble because everything you know is wrong".

The special case of programming reflects a constellation in which "subjectivity" no longer means what Descartes meant by the French or Latin cognate. Today, to the extent you claim knowledge unshared with the team in the administered world, your subjectivity (your capacity for judgement) is contrasted to its disadvantage with a reified, brain in a vat "objectivity" consisting in a vast amount of mathematical and technical deduction from a set of reified and unquestionable axioms (such as "savings is investment" in neoclassical economics).

I am not saying that we should all become "subjective" because the administered world would not have evolved unless it was necessary alongside "the development of productive forces". The problem in the key decade of the 1970s (in programming groups and elsewhere) was that as productive forces evolved, humans of that era (whose "nature" was probably different from today) thought they knew in more ways and in qualitatively different ways than these crazy kids today, and warred with each other while snorting coke and having sex in the restroom (I was there, and I remember because as a geek I didn’t have access to good drugs). The system was near breakdown as a result with Britain enduring a "winter of discontent" in Britain in 1979 because trades union members "knew" they deserved a higher pay packet, therefore, to survive, it necessarily evolved the Eighties persona.

This evolution was visible on the plane of art, as in the case where popular music fissured into punk versus techno, with punk expressing pure outrage at a social operation being carried out on bodies without anesthetic and techno being a sort of acceptance that could be coded in a few lines of Basic on midi.

We are showroom dummies – Kraftwerk
Are we not men? We are devo – Devo
We live in a material world, and I’m a Material Girl – Madonna
And I was gripped…by that deadly phantom – The Clash

In other words, when I was passing out on Tsing Tao and the Clash in Mountain View in 1982, human nature was changing. When I stopped drinking and moved thereafter to Princeton, I started to read great whacking books on deconstruction from the Princeton University Store in part, I realize now, because I needed to restore analytic powers to perceive, under the rapidly chaging shape of "human nature" a continuous biography.

I feel that otherwise my fate would have been what the actor Nicholas Cage (an alarmingly intelligent fellow) portrayed in films like Matchstick Men and Leaving Las Vegas: a very "postmodern" disintegration of personality constituted by the loss of emotional memory.

In Leaving Las Vegas, an earlier film, Cage portrays a screenwriter (whose relation to the tools of movie production is isomorphic to the programmer’s relation to information production) who is blasting out memory with booze. In the much more evolved recent film Matchstick Men, his portrayal of a con artist who is addicted to a drug whose name he cannot recall is disturbingly true to the actual America I know.

All around me, even at Princeton (and in ways especially at Princeton) intellectuals seemed stuck in the Fifties groove of "simplicity and clarity" and a "new" criticism of author worship that in Harold Bloom’s and Camille Paglia’s cases was devolving to intellectual thuggishness. But reading, trying to parse, Adorno made sense to me even at the very basic level of restoring circuits I’d blasted out with alcohol.

In certain punk phenomena, a portion of the human personality, divided already from sweetness and from light by fissures traceable to the divided theology of Trinitarian Christianity (its god, crucified by its god when you get down to brass tacks) physically broke away in the manner of an Antarctic ice shelf about this time.

This was the genesis of the programming persona of today who can on the job completely conform to corporate expectations while hanging loose without moral seriousness or even basic honesty elsewhere, and who on the job only displays a partial honesty. Of course, as is the case with most human phenomena, the split is incomplete, because the persona I am thinking about, having no global honesty or moral seriousness worth mentioning, is able to transform his subjectivity, and capacity for honesty, into a cash nexus, which makes him a convenient tool.

As writers for The Baffler and The Onion have seen, this makes today’s rebel the most convenient corporate employee.

I witnessed the dawn of this phenomenon at Bell Northern Research. A younger guy who was the head of the firmware and software team which needed my compiler was an honest deviant punk who dragged me to clubs and the symphony when Berg was being performed. He was a natural leader and our team delivered a successful product. I was late with the compiler because I wanted to fix pre-existing bugs and did so, much to his scorn, anger and dismay, which he was able to express openly. We’d go out into the parking lot, I’d smoke, and we’d almost come to blows, but in the end I delivered a working compiler.

But, I noticed at the same time that the corporation sponsored a completely different false deviant as the leader of a different team, who on the surface resembled my project lead but was in fact a complete moron, whose deviance off the job made her completely ineffective on the job, almost as if the company could see that my project lead was successful owing to his truth to himself, but at the same time could not abide moral seriousness, independence, and sincerity.

My guy took things seriously, whereas the punkette treated everything except her needs as a joke.

I am sad to say that my project lead is an ex-friend. I started receiving abusive emails from him when I stopped being serious about wanting to work any old development job and entered changes I am still am experiencing, where I realize that I can’t work for managers who are twenty years younger than me.

Postmodernism by no means excuses disorganization, and this has been a rather disorganized essay. Again, see my fat and prolix pal Adorno, this time in "The Essay as Form".

Outlines are great but also are a prison. "Postmodernism" may seem like lack of form but in fact the work of art, or computer program, may be the simultaneous creation of a single work of art AND a new form.

A former manager complained to me recently that one of his guys had coded a data base system overusing Microsoft GUIDs and stored procedures in such a way that the system couldn't be converted, "easily", to Oracle.

I don't know enough to say whether this was an actual mistake. However, note that we consistently narrate such choices in one way that almost always puts the developer at fault, and one responsible use of postmodernism is understanding that when we describe systems that are, in fact, complex, more than one competing narrative may apply.

Of course, good managers know this and are always instinctively balancing narratives, but I have noticed a decline in this ability in an increasingly Fundamentalist America.

Postmodernism in insisting on a democracy of forms is confused with formlessness. But, I find when writing that there is an underlying form. The problem is communicating the new form along with new content. For me this is not easy and demands continual revision.

Dan: "grand narrative" is perjorative. I’m not looking for a grand narrative, only a best or even first fit. We can reject "grand" narratives in favor of something simple such as the very idea that a professional does due diligence in the service of truth over and above the short-term "needs" of the user.

To reject the simple correspondence of word and action, promise and result because "I yam a hip postmodern dude who don’t like them grand narratives" is to throw the baby, poor mite, out with the bathwater AS IF Kant was responsible for Nietzche or Brezhnev.

Deconstruction helps us to understand that a necessary feature of a text in context is misreading. If our fathers or your grandfathers, like Arthur Miller’s Willy Loman, reified and put their trust where it did not belong (whether in Stalin or a smile or a shoeshine) then it still remains the case that as Kant said the only absolutely knowable good is purity of the will.

Amy Tan, in the Joy Luck Club (a fictionalized weaving-together of stories about Chinese immigrant mothers of the wartime generation) and Miller, in Death of a Salesman, illustrate the idea that the default life in an administered world is one of necessary and unavoidable aporias and compromise, and that in one sense the well-lived life cannot be attained, as Adorno seems to have believed, within the administered world.

One Chinese mother in The Joy Luck Club had killed her baby in Shanghai in post-partum depression like that woman in Texas (whose husband was a programmer, caught it seems in the reified headlights of a single narrative of what constituted a happy family). Willy Loman sold his soul to the company with the best of intentions, like Homer Simpson at the nuclear reactor.

Neither the China mothers nor Willy had what Aristotle called a modicum of moral luck. Unfortunately (as I tried in my way to point out with regards to Plato’s scorn for reference manuals in Build Your Own) we imbibe at university and in the yellow press the ancient philosophy of a slave-owning class whose philosophers simply did not believe that without moral luck (wealth and property) the question as to whether a woman or slave or artisan (computer programmer: salesman: immigrant housewife) was a good person was meaningless, unaskable, mere words.

Plato's dismissal, which I quote in Build Your Own, of the reference manual is an owner's trope, a slave society's need to ensure a static base without a written memory. Aristotle may have been a slave of Phillip of Macedon and according to Copleston, Plato himself may have endured a period as a slave.

A slave among other things cannot have a history: he cannot write (which is why programmers in the worst environments don't document a thing). At the same time, he has to write. The slave-owner wants him to be a transcription machine, but then writing escapes from the forced "focus".

The paradox in Western philosophy is that it was created by slaves in an originary scene where it was "knowledge" that only slave-owning males could philosophize with a pure heart, and if this is what Derrida meant, I'm a happy man.

Consider the moral climate of the major corporation, wherein information is withheld by default even from let us say programmers who sort of need all the business rules.

Influenced by common-law traditions encapsulated in phrases like "the employee is necessarily worthy of his hire", white collar functionaries have in a literally self-contradictory fashion to be told, in the performance of jobs which are legally exempt from overtime because in law they are said to involve decision-making, not to ask too many questions or to question.

Consider that recent ethnographic research in the sociology not of "stodgy" corporations, but of "cutting edge", New Era firms where speakers blast the latest ragtime tunes, the ethnographers have consistently discovered the return of repressed control, and indeed the absence of key information flows and decision-making power whenever this would threaten the position of vulture capitalists.

Watch the 2000 movie dot.com, a "reality show" style presentation of the fortunes of a "cutting edge" firm. One of the founders wants just to code and design and is pushed out for this reason because the other founders and, far more important, the venture guys need exponential growth in potential customers, and thereby stock price. Despite the fact that the other partner is a better salesman by far, the desire to "design and code" (to write) makes the loser partner a marked man.

This means an absence of moral luck and it places the employee, whether of the traditional corporation (a dying breed because in almost direct proportion to their size, such corporations as IBM insist that they are dancing elephants) or of the startup, in a position of not being able to make informed moral choices.

Which is and was the intent of the admnistered world as it morphs in a Matrix while preserving its necessary features.

Miller’s accomplishment, however, was to make us understand Willy’s brother when he says at the end of Death of a Salesman, "nobody dast blame this man".

Amy Tan’s accomplishment is in the lines "this swan feather I bring from afar and in it are all my good intentions", because the only thing we can know to be good without doubt is purity of the will.

Subjectivity and moral choice in other words persist, and a postmodern ethics as well.

Deconstruction is a global humility and not some sort of willingness to suck up to the global corporation, and it is a humility not exclusively about one’s limitations (which may not exist after all) but about those of society and its shibboleths, its ears of corn.