Sunday, September 19, 2010

ARTISTIC TASTE CONVERTED TO BINARY CODE

When people talk about computer art, they usually focus on the "supply" side: artists using computers to create and distribute art.

But computers have major consequences for the "demand" side of the equation: what viewers want.

We have already witnessed the first primitive applications of computers to understanding what kind of art viewers like and why:

1. In 1994, artists Vitaly Komar and Alexander Melamid conducted a statistical analysis to calculate the ideal painting for U.S. audiences. They discovered that 60% of the population preferred paintings that are "realistic looking" while 88% preferred outdoor scenes featuring lakes or rivers. 53% preferred paintings to have visible brush strokes. Komar and Melamid "translated the numbers into paint on canvas." Their analysis produced the following picture:



2. Information technology is being used to rank the greatest artworks of the 20th century: Economist David Galenson has proposed quantitative methods to rank art, such as the number of times pictures appear in art history textbooks. Other economists, such as Michael Rushton and Charles Gray feel this approach shows great promise. Says Gray: "We all want to believe that there is something special about the arts but I don't buy that there is any difference between artistic and economic value."

3. Other computer scientists take a different approach, claiming that "with the use of mathematics, computers and massive data bases of attractive faces, we have been able to quantify facial attractiveness in a consistent mathematical computer model...."



Building on historical archetypes of beauty, companies now claim to have calculated the formula for beauty and attractiveness: "it is a mathematical ratio that seems to appear recurrently in nature as well as other things that are seen as Beautiful. The Golden ratio is a mathematical ratio of 1.6180339887:1, and the number 1.6180339887 is called phi." Using computer programs and a trademarked "golden grid," an artist might tailor an image to what viewers would find most attractive.

But these early, sometimes laughable efforts have given way to more sophisticated applications of information technology. Rather than gathering raw data through telephone surveys the way Komar and Melamid did in 1994, science has gained the ability to monitor brain, blood, skin and other biological reactions to art. Until now, these nascent technologies (especially electroencephalography and infrared optical tomography) have found uses in the gaming and neuromarketing industries:
[neuromarketing is] a new field of marketing which studies consumers' sensorimotor, cognitive, and affective response to marketing stimuli. Researchers use technologies such as functional magnetic resonance imaging (fMRI) to measure changes in activity in parts of the brain, electroencephalography (EEG) to measure activity in specific regional spectra of the brain response, and/or sensors to measure changes in one's physiological state (heart rate, respiratory rate, galvanic skin response) to learn why consumers make the decisions they do, and what part of the brain is telling them to do it.
Would a CGI picture evoke a better reaction if the hero's shirt was blue rather than red, or the heroine had longer hair? Would a CGI animated kiss come across as more passionate if it were five seconds longer or shorter?

Computers can identify the electrical activity in the brain that accompanies the thrill of seeing a good work of art. They monitor localized changes of oxy- and deoxy-hemoglobin concentrations in the brain in response to various images. With increasing precision, computers are likely to explain the pharmacological activity that accompanies a diverse range of artistic thrills.

From there, it will become much easier (and more efficient) to stimulate those same reactions by skipping over that obsolete middle man between the work of art and the audience: the artist, who struggled for centuries relying on nothing but highly fallible intuition.


78 comments:

Anonymous said...

"But computers have major consequences for the "demand" side of the equation: what viewers want."

Thomas Kinkade is years ahead of them.

"we have been able to quantify facial attractiveness"

Here is a simple Star of David template I created that seems to be a reliable indicator of female Caucasian beauty:
http://yfrog.com/0fevlij

kev ferrara said...

Wunnerful post. All this stuff makes me laugh... but also makes me sad.

What is it about quantification that impresses people so much? The surety? The association of quantification with physics or computers, or "smart people"? The lack of faith in our own human intuition is pathetic. I guess being told what to think all the time eventually wears away the ability to consider things one's self, replacing it with pavlovian insecurity...

Art made by the quants-at-heart will always feel like math; perfectly frozen. Does this please the afraid?

Surveys are answered by people who think their opinion is either valuable or valued, but who are unaware that they are being subjected to disguised advertising.

In other words, let us not trust anything said by the credulous and narcissistic... which is the exact reason that box office returns and references in the history of art are no measure either... being surveys of a kind as well.

Surveys are what Shepherds make when they look over their flock grazing idly in the hillside. All sheep eat the grass under their nose (life truism).

Paolo Rivera said...

Is that a hippo in the background? Maybe that's what my paintings have been lacking!

Sarsaparilla said...

What an intriguing post! All I know is that when it comes to beauty and art, I'll take infallible intuition any day over statistics and computers.

raphael said...

kev:
Does this please the afraid?

i would guess so. after all, when the currency of your being is how well you are liked and/or respected by similar minds, things like "i think its pretty" is a pretty risky thing to say. after all, you could be completely wrong, and everyone else will snicker at you. not to mention the inherent loss of said currency for you.
being able to point towards the numbers and prove that there is an objective point to your appreciation, is something rather reassuring, and also breaks the whole thing down into easy to grasp bits. theres nothing so mystifying about the whole amazement and beauty thing, after all.

what plagues all these (neurowhateverical, statistical...) endeavours is that they dont talk about beauty. david worded this exactly right when he wrote:
---
...computers are likely to explain the pharmacological activity that accompanies a diverse range of artistic thrills.
---
the key word is "accompanies". that is what science can do: make pretty good guesses about what stuff does often come hand in hand. what this not is, is a justification of subsuming one under the other. no statistical insight makes beauty a result of adherence to a certain geometric face pattern. dito with neurological brainmagic.
that would be like deriving the laws of maths from the laws of mechanics when examining a mechanical calculator.

an aside note, kev: i posted a question for you in the comments of the last post - maybe you would have seen it anyway, i just wanted to make sure you know, with a new post to hold our attention, and all. :)

raphael

Fryewerk said...

I had a physics teacher that expressed his frustration at my sketches in my notebooks and margins of papers that I turned in. Being a left-brained genius, in the past, he had tried to learn to draw and thought that it was something that could be learned through logic and reading. Most of these studies seem to be by similar intelligent, but sadly left-brained who want to make the intangible definition of beauty and attraction definable by math and logic. It just doesn't work, it may approach some reason, but numbers and words will never be able to fully explain the etherial aesthetic attraction and reason for beauty. Sure, they can define the geometry of an attractive face, but can they explain WHY that geometry is attractive?

David Apatoff said...

Kev Ferrara and Sarsaparilla,

Some of what the quants do is easy to ridicule. For example, I think what the economists (Galenson, Rushton and Gray) are proposing is superficial nonsense. But the real scientists seem to be on the trail of something that will be much harder to ridicule, something which will put our assumptions to the test.

From both the supply and the demand side of art, the new powers of information technology are nibbling away at the traditional realm of artistic taste and judgment.

On the supply side, art software such as Corel now provides composition tools such as "the Divine Proportion tool" (
http://apps.corel.com/painterx/us/essence_new_features.html ) which helps artists create an aesthetically pleasing composition.

But today I am more interested in the impact of technology on the demand side. I can foresee a day when the most concentrated, intense aesthetic response may come from (or be assisted by) direct electronic stimulation to key parts of the brain.

T Arthur Smith said...

I'm a bit worried that the most common application of this new science will be in advertising. Twenty years from now, I won't just hate the new jingle for doublemint gum, I'll hate the overstimulation of bombarding imagery, fast zoom ins and outs, curves eclipsing spotlights, jarring flashes of eroticism and sound, etc. If Lady Gaga wants to do that to compensate for whatever she calls music, whatever. But, for scientists to treat everything on the tele the same way, it'll be exhausting! The speedy techno drumming they put in many English commercials is awful enough.

T Arthur Smith said...

And all the film directors lacking self-esteem and dignity will rush to use all the "tricks" learned, primarily making work for kids, because they're an easier audience. It'll never measure up to Pixar.

Anonymous said...

All these scientists do not realize that the beauty or attractiveness in art is not because of perfection. Every work of art guided by fallible human intuition is perfect because of the smallest imperfection or imprecision made by the artist.

Human beings are attractive because of the asymmetry and not because of perfect symmetry.

Jesse Hamm said...

This top-down reasoning, from popularity to practice, isn't new. Hollywood has been trying to operate that way for ages. "People like movies with ______ in them; add more _______ to every movie!"

At best, this dependence on precedent can only ensure that sales are consistent. But sales that are phenomenal, and ongoing, always spring from unprecedented sources. The monster movie was dead when Jaws debuted, the swashbuckler was dead when Star Wars debuted, sword & sorcery flicks were dead when Lord of the Rings debuted, pirate flicks were dead when Pirates of the Caribbean debuted....

I believe beauty and taste can be measured, but it's a mistake to measure them strictly by viewers' responses, which vary. We measure the purity of cocaine by its chemical make-up, not by how good it makes people feel.

David Apatoff said...

JF wrote, "it may approach some reason, but numbers and words will never be able to fully explain the etherial aesthetic attraction and reason for beauty."

It's true that computers are known as "information technology" rather than "knowledge technology" or "wisdom technology." Computers (and science generally) are suited to describe rather than explain.

When Picasso said, "computers are useless, they can only give you answers," I suspect he meant that that the kind of questions he was most interesting in pondering had no answers... at least not the kind of final, clear cut answers that mere information is capable of giving you.

Yet, don't sell technology short. Technology's contributions to art are not just quantitative; they have been qualitative too.

Laurence John said...

"I can foresee a day when the most concentrated, intense aesthetic response may come from (or be assisted by) direct electronic stimulation to key parts of the brain."


David, have you been reading Philip K Dick's 'Valis' by any chance ?

it's one thing to be able to measure the brain's response to viewing a movie. it's quite another to be able to evoke those responses (and images) without the movie, via direct electronic stimulation of the brain. if you're picturing a future where instead of each home having a TV set which shows images and sounds, we each have a black box which fires invisible frequencies at us, and makes us feel elated, concerned, excited.... without any narrative context at all, well i wouldn't lose too much sleep over it, but there's definitely a sci-fi movie in there somewhere.

the Corel 'divine proportion tool'... you have to laugh really. it's basically the familiar old spiral grid that pops up onscreen so you can lay it over your image. that, and those modern computer facial-beauty-grid models are just the latest manifestation of an ongoing human delusion ... that hidden mathematical formulas can explain the mysteries of art / beauty / the 'divine'.

Information Marketing said...

Information marketing is a form of art. When you want to pick a style where money to be made you need to know about information the marketing secrets and strategies.

Don Cox said...

I have never been convinced that the Golden Section is of any use whatever in either painting or design.

It is worth noting that the landscape picture is part of a tongue-in-cheek project. They are really asking the question "Is popular the same as good?"

kev ferrara said...

I'm with Jesse and Laurence.

Our brains have been hardwired to reward the apprehension of life information presented under an aesthetic guise. To skip the information and go right to the reward... that's simply chemical pornography... drug use... the very definition of mindless escapism. Which has nothing to do with Art and never will.

The problem with scientists poking into the arts is that they think a metaphor is merely an electro-chemical reaction.

Related to Jesse's analogy about movies, let's not forget the mass of screenwriting books and classes that have flooded the market with "Myth Structures for Hollywood Success" since it was "revealed" that Star Wars was structured according to principles cribbed from Joseph Campbell's Hero with a Thousand Faces. Even with all these perfect recipes for successful myths the same abysmally low percentage of screenplays are worth fabricating. (Charlatan truism: In a tough market sell schematic dreams of success in that market, don't actually get involved in that market yourself.)

The surety of schematic diagrams is a drug addiction as well.

Structure is only as good as its content, same as it ever was.

Alessandra Kelley said...

Have you heard of the "uncanny valley"? Supposedly it's the way computer-generated faces (or animatronic robots) get creepier and creepier as they get very close -- but not quite -- to human.

There just seems to be something repulsive about not-quite-right computer models, especially when they try to make them move.

Kpakpo Akwei said...

The following is a link to an article on artificial intelligence applied to Art. Even though it is about musical composition, it certainly would inform Mr. Apatoff's post. The article is a bit on the long side but yeilds truly remarkable insights into the topic. I agree that computers and quantitative methods are not to be dismissed out of hand.
Anyway read on and I'd be curious to know what you all think.
http://www.miller-mccune.com/culture-society/triumph-of-the-cyborg-composer-8507/

Kpakpo Akwei said...

If I may, allow me to quote here a post that sums up the danger to artists in a world ruled by markets and corporations.

This is a response to the article mentionned in my previous post by
"anonymous musician · 29 weeks ago

the one thing that concerns me about this program as a musician is that if emily howell was advanced to the point that music creation would be totally autonomous of any human input (and made available for licensing/purchase) is that record companies would completely eliminate composing artists from their rosters.

this situation does make sense from a business perspective, it's why factories use computer controlled equipment instead of tradesmen. why pay a human to do it when it's much cheaper to have computer do it? especially if the results are good enough to be marketable? record companies already have a dim and uninspiring view of artists in the first place. "

Stephen Southerland said...

Yes, we humans are still addicted to control; we would like to think that the universe, human nature, and aesthetic experience only exist to be dominated, subdued, and redefined by us. If Art does not defy this notion, Art is not Art.

António Araújo said...

One doesn't get anywhere by trying to define "beauty".

The idea that we start by defining our terms is a misconception. We define our secondary terms, the real essential ones are defined only implicitly, by their properties with relation to one another.

For instance, in geometry, Euclid tried to define points and lines. It doesn't work. If you define them in terms of something simpler, then that something simpler is stil undefined and so on.

What works is this: you take points and lines as undefined, and define instead what their properties are, how they interact with one another. Points and lines are then whatever holds those properties.

In the same way, as is apparent in our endless discussions here, nobody, certainly not artists, can give a usable definition of beauty. Therefore beauty is a good undefined concept. Instead of trying to define it, investigate its effects and its properties, with regard to other concepts.

That is why scientists (well, the ones who have a clue) don't study "beauty" as such, they study its effects on measurables, like human biochemistry. "Beauty" will be whatever causes the exact same effects. That is how one *starts* to get somewhere.

Donald Pittenger said...

I dunno about this control stuff. Magic buttons to be pushed by eeeevil admen has been a recurring topic of breathless journalists for the better part of a hundred years now. In the 1950s the hot deal was subliminal messages inserted every Xth frame projected at a theatre which supposedly got viewers to leave their seats and scramble to the popcorn stand. Mind control at its finest, that was.

As for the brain-probing, for that to actually work well data would be required for each individual and his response pattern -- and the outside stimulus would therefore also have to be applied on an individual basis.

Absent that, the only results would have to be in the form of statistical relationships whose veracity would be a matter of the law of large numbers. Put another way, such research would yield a set of averages and would not apply to everyone.

Then there is the matter of how one's judgments are formed. I'll use me as an example:

In my first grade class there was a chubby, red-haired, red-faced girl who I didn't like. Since then, I've never been very attracted to redheaded women. And for a reason I can't explain (though it must be based on some other experience), women with dark hair, pale skin and gray eyes tend to fascinate me, all else being equal.

Yet there must be other guys whose appreciation of feminine aesthetics is the exact opposite.

Those brain-measuring tools might someday tease preferences and rejections, but they couldn't tease further to find out why I react as I do. Seems to me that info could be of interest to those future Mad Men.

People are too complicated for "social scientists" to get deep knowledge of. Or so I think.

António Araújo said...

>People are too complicated for >"social scientists"

Yes, but what if real scientists take over the job one day? ;)

You know, knowing the average is all it takes to make a bundle, especially if the standard deviation is small.

And isn't it?

Isn't it predictable, for instance, how people react to sugestions of of "mathematization" of "beauty"? I think it's pretty predictable. Any stage magician or con-man will tell you that most people are pretty much the same: for instance, they ALL think they are special and unique (me included ;D)

Donald Pittenger said...

Antonio said: "but what if real scientists take over the job one day?"

I know this was written in jest, but it raises an interesting issue somewhat related to the main topic of the post.

I believe that it's not the caliber of the scientist so much as the subject of study that matters when the subject happens to relate to humans in one way or another. I recall that, back around 1970 when the aviation industry in California was going through one of its dry cycles, some engineers from that industry were getting involved in urban issues. And so far as I know, their impact was minor.

Many "social scientists" in fact, get decent scientific training. Again, I'll use myself as an example:

One of those little accidents of life resulted in my earning a Ph.D. in Sociology from a spiffy Ivy League university. Between that and a Masters degree from a major West Coast university, I had quite a bit of coursework and seminar time in statistics and research methodology, not to mention a fair amount of on-my-own reading on Philosophy of Science. I suppose this was pretty typical for sociologists trained in my era.

The point being, we had decent training, yet Sociology remained light years in terms of accomplishment behind the physical and even biological sciences -- that before Sociology completed its degeneration into a politicized cesspool.

Sure, sometimes one gets nice tight variances. And I have no doubt at all that fortune tellers, con men and carnie barkers have seriously effective rules of thumb regarding human behavior. But even they are playing the odds and can't always fathom people correctly.

Which was the point of my previous comment.

kev ferrara said...

The symmetry of facial beauty has been "solved" over and over again since antiquity. Then it was geometry using rulers and compasses, now a computer is used. Yes, symmetry and certain proportions are attractive: That's the lesson plan for Day 1 of Basic Aesthetics.

When scientists can replicate a D minor 7th chord in my brain, without going through my senses, then I'll begin to pay attention.

Anonymous said...

When chemists start messing with my watercolors and add a pheromone to them, I'm quitting art altogether. I agree with etc,etc - there is something that TK has figured out that seems to mesmerize his audience and turn them into gooey, pablum-loving plasma. Sigh.

On another note, Flash Modin on the website Physics Buzz, posted digital art created by a radiologist, using a CT scan, that is quite beautiful (at least to this artist and radiologists, perhaps).

Don Cox said...

"what if real scientists take over the job one day?"

There have been a few scientists who became good novelists, for example Nigel Balchin and C P Snow. I can't think offhand of any scientists who were also good artists, but there must be some.

(Biologists often get good at biological illustration.)

Stephen Worth said...

...and if you believe those studies, I can make you a terrific deal on the Brooklyn Bridge

Stephen Worth said...

To Don Cox: Ernst Haeckle

Stephen Worth said...

Sorry, that's "Haeckel"

António Araújo said...

Kev,

>When scientists can replicate a D >minor 7th chord in my brain, without >going through my senses, then I'll >begin to pay attention.

The request is a bit specific, but-... nowadays I seem to recall they can elicit emotions directly - or an orgasm, if you prefer! - and there is even something called the God module in your temporal lobe that you can stimulate with a magnetic field and it gives you a feeling of "religious experience" connected to nothing at all. I don't think it would be too hard - barring problems access (you have to expose inner parts of the brain) and all sorts ethics problems, to stimulate directly the auditory centers, visual centers, emotional centers, etc. Lots of freaky stuff is done with monkeys, but of course they don't tell us much of what they experience. More to the point, lots of freaky stuff happens to people who suffer accidents in certain regions of the brain - stuff including hearing much more than chords and seeing all sorts of things. Now, most of those things could probably be made to happen deliberately right now, except that no sane person will want to do it, but the thing is, it can probably be done, so, by the criteria you stated, the time to pay attention has perhaps already arrived, I think...

D. Pittenger, indeed I was jesting, those problems are hard to grasp with the usual tools. But when you say

> But even they are playing the >odds and can't always fathom >people correctly.

I understood that was your point, but the thing is, I don't think anybody is trying to do more than play the odds. Say a guy finds a method to generate musical hits that get 80% of the public reliably. Or paintings that drive most of the critics wild. What does he care that he can't fathom the rest? What does he even care, economically speaking, about the random superhit that will happen once in a while -unpredictably- on human artistry alone? Economically it doesn't matter. Someone else would probably land that contract anyway. It's like being the house at Las Vegas, all you need is a reliable edge, and an edge is all such studies are trying to get, not a window into the mind of any specific person, much less all specific persons.

António Araújo said...

Kev,

>When scientists can replicate a D >minor 7th chord in my brain, without >going through my senses, then I'll >begin to pay attention.

The request is a bit specific, but-... nowadays I seem to recall they can elicit emotions directly - or an orgasm, if you prefer! - and there is even something called the God module in your temporal lobe that you can stimulate with a magnetic field and it gives you a feeling of "religious experience" connected to nothing at all. I don't think it would be too hard - barring problems access (you have to expose inner parts of the brain) and all sorts ethics problems, to stimulate directly the auditory centers, visual centers, emotional centers, etc. Lots of freaky stuff is done with monkeys, but of course they don't tell us much of what they experience. More to the point, lots of freaky stuff happens to people who suffer accidents in certain regions of the brain - stuff including hearing much more than chords and seeing all sorts of things. Now, most of those things could probably be made to happen deliberately right now, except that no sane person will want to do it, but the thing is, it can probably be done, so, by the criteria you stated, the time to pay attention has perhaps already arrived, I think...

D. Pittenger, indeed I was jesting, those problems are hard to grasp with the usual tools. But when you say

> But even they are playing the >odds and can't always fathom >people correctly.

I understood that was your point, but the thing is, I don't think anybody is trying to do more than play the odds. Say a guy finds a method to generate musical hits that get 80% of the public reliably. Or paintings that drive most of the critics wild. What does he care that he can't fathom the rest? What does he even care, economically speaking, about the random superhit that will happen once in a while -unpredictably- on human artistry alone? Economically it doesn't matter. Someone else would probably land that contract anyway. It's like being the house at Las Vegas, all you need is a reliable edge, and an edge is all such studies are trying to get, not a window into the mind of any specific person, much less all specific persons.

raphael said...

antonio: "I understood that was your point, but the thing is, I don't think anybody is trying to do more than play the odds."

nobody in the field of empirical science anyway. no matter whether we are talking physics or sociology, its all about making general inductions from specific observation. the price you pay for that is that at no point can there be a definitve answer because you are not guarded from the next specific observation contradicting the theory you had until that point. the other price that is paid is that the first thing any science does is define its field - and afterwards, it isnt concerned with anything outside this scope.
but what science is, is a great tool for making useful predictions that usually work, no matter whether we are talking about the flight path of an asteroid, a pop song that will be a hit with a large enough percentage of listeners or constructing cars.

when we arent concerned with making predictions that are useful, but are interested with what a thing actually is, the empirical scientific method just is the wrong tool, because the question is a philosophical one.

kev ferrara said...

Yo Antoniyo,

So, I am aware of the research you are discussing, the frontal lobe-religious connection, etc. Which is exactly the reason that I specifically mentioned a Dm7 chord. It seems like it should be a simple procedure to cause the chord to apparate in the mind sans senses, but it is actually a phenomenally complex problem that touches on how specific symbolic thoughts are processed in the brain, mechanistically. I would be quite interested if you could produce an article about such an experiment, or a similarly specific production in a subject's brain by scientific means. Inducing one steady tone, even.

Jack-Wheatley said...

It is the artists and there mistakes that create what people find beautiful, i didn't used to like Picasso now i do and i bet my brain would light up with awe

António Araújo said...

Kev,

I have no reference on that, sorry :(

Any neurologists in the house?

ps: I am very ignorant on the auditory cortex. I remember that there are specific zones targeted for pitch, so I suspect that you could induce tones by targeting them? I don't know. I think the details are still unclear in humans, and that, as you said, the various aspects of sound seem hard to isolate from on another. It doesn't help that most deliberate action can only be taken on animals, and you can't ask them about it (directly), and anyway even if you could, the structures are different in humans. For the rest you have to rely on observing people with damage to the ares of interest, or sometimes there are cases where you take advantage of the fact that you have to do surgery on an area to poke around a bit. Investigating any area not near the surface suffers from such problems of access. The so-called god module is a nice exception where you can stimulate from outside the skull by shooting it with a magnetic field. I always found it funny that you have to go deeper in the brain to stimulate an orgasm than a religious experience. I like to think that correlates with their relative philosophical depth. :)

Well, this does tell us that in practice science won't be able to do much in the way of direct elicitation. The fact that it could do so if we were monsters and lunatics simply means that it actually won't. Hey, I'm being an optimist! :D

But I think it is interesting to separate what could conceivably be done from what can reasonably be done. For instance, science *can't* predict accurate long-term evolution of chaotic systems; science *won't* mess around inside people's brain until it learns how to produce music inside them by direct induction...

Thomas said...

I'm highly skeptical of that. Science is just very "stupid" in some areas. And art and beauty is one of them.
The whole thing is just extremely complicated and most of all unpredictable. Simple example: If we all use the same face in our art now (since its the mathematical most beautiful), then it would soon seem boring.

Sure - neuroscience can give us some insight and such. I saw one example about landscapes in an exhibition in Germany ( http://www.oxpal.com/?p=710 ). But in the end it won't come down to a simple number (let alone using 20 decimal digits).

Already the golden mean didn't work out. I don't buy it at all. Sir Lawrence Alma-Tadema painted to the extreme according to the golden mean - and I don't see his compositions as superior at all.

Anonymous said...

Kpakpo Akwei said...
"The following is a link to an article on artificial intelligence applied to Art."

Kpakpo,
I posted that link a few weeks ago but it did not generate discussion. I think the first track is very interesting; reminds me of Rachmaninoff.

The Western musical system is organized into 12 basic notes and there has been for centuries a quantifiable annotation system that can describe pretty much everything, and compositional theory is at least academically well understood by many (not that it makes one a good composer).

There are mathematical annotation systems for visual data such as geometry and color (Munsell, for example); however there is a tremendous
amount of discernable variation in contrast to 12 basic musical notes. Furthermore, in traditional representational art there are additionally compositional frameworks and styles, some not even understood, that organize the
data into patterns, and a huge array of subject matter as well.

In short, a computer art composer similar to a computer music composer would be a far more
complicated undertaking.

António Araújo said...

>a computer art composer similar to a >computer music composer would be a >far more
>complicated undertaking

but maybe not so hard for abstract art.

It is funny, my first take on this post was precisely that I firmly believe that we will have computer produced artworks that are impossible to distinguish from good human-made artworks. Note I was careful to frame it in objective terms, otherwise someone would claim "but it's not REAL Art", "real Art" being whatever circular definition ensures that only humans are allowed to do it.

I had no idea that this had been achieved already in music. Funny enough, the reaction is the one I expected: a lot of anger. People say they care about art, yet I think they care much more about two things: feeling special because they can make art, and making a living out of art. If they really cared about art they'd be happy about both the availability of more art and about what can be learned from a computer that does it at least up to a point - once you such an algorithm written down, you get to peek at how art works, and at how you yourself work.

But love of knowledge, as love of art, stands a long way behind ego and job security.

It's like with A.I., in general. Once, playing chess was considered to require intelligence. Once computers could do it, it wasn't thought so anymore. Same with face recognition, and now musical composition. As more and more tasks are achieved by computers, they are considered not to need intelligence anymore. The problem is that if we keep retreating that dividing line so much we'll get to the point where neither computers nor humans will have anything left to call inteligence. Oh, wait, intelligence will be "whatever cognitive quality humans have and computers don't". And when you ask a future cumputer what that is it will mutter under his breath "ego and stupidity!" :)

Ever notice how many discussions on what is True Art are just beating around the bush of job security? Like, "photography killed realistic art" (as if it could take away the pleasure and knowledge that comes from drawing realistically!) really means "killed the realistic art market"? Or "Modern art is fraud" really means "we can draw and they can't and we should be getting that money"? Art is never in danger, only artist's jobs are. And I agree that this is a problem, but should be called by its own name.

António Araújo said...

ps: I am now Cope's nº1 fan. I love his crude language and myth-bashing. Programming is still in the realm of art, and Cope is a grand artist in that field. Soon, of course, computers will be making their own programs much more effectively. ;)

I am not worried. I'll be on the beach, making my inferior realistic drawings, slowly proving theorems that a computer could prove in a milisecond, and that will give me exactly the same pleasure as now. Both tasks will of course have no financial value.

Or maybe not. Notice how musicians refused to play the machine works, and how publishers refused to publish them.

This reminds me of those artifical diamonds. "Real" diamonds are very expensive, and artifical ones are cheap. Funny enough, the only way to separate the two is that the real ones have more imperfection, including "watermarks" that the companies that dig them out and polish them put in there for identification. So you get an inferior product for more money, because you want to claim a scarcity that belongs to the past!

Maybe future artists will be spared joblessness by an apartheid system. I recall reading in Vasari's "lives" that slaves were barred from painting,so why not computers?... :p

The future's so bright I gotta wear shades.

Laurence John said...

David Cope's computer doesn't 'compose' anything. it simply makes composing easier for him, by following sets of pre-programmed algorithms. HE had to feed everything he knew about Bach and Mozart into the computer (lucky those old guys did most of the hard work for him eh ?) HE still has to decide if the resultant computer-music 'works' before he puts is out there. the computer doesn't know.

Antonio, computers are already making abstract art... you must have seen those endlessly flowing trippy shiny fractal patterns that can multiply in the most complex patterns indefinitely. any good ? don't ask the computer, it doesn't have an opinion. it can't conceptualize an idea. only humans can do that.

António Araújo said...

> endlessly flowing trippy shiny >fractal patterns that can multiply >in the most complex patterns >indefinitely. any good ?

Very good, in fact, but not as art :). I never had a fractal on my wall, but I had there some equations that generate them :). It's hard to call it art as it is exactly like plotting a circle (and just as beautiful - which it is, as maths is just as beautiful as art, i.m.o.). A fractal is just the graph of a mathematical object, so you can argue that you don't create those images, you just spot them and take a picture; on the other hand I guess plotting a fractal is art if wildlife photography is art (is it? I doubt anybody knows what art is, we discussed this above)...but if you want to pass printed fractals for paintings, then I guess they cannot compete with good art. I don't feel the need to put them on my walls, as they'd be at the level of the Tate modern, say, but not any better than that ;)

I don't care what art is, art is what art does. Actually fractals would probably make a decent career for an artist if people thought they were paintings from imagination. Yes, at least at the level of the crap at the Tate. More probably deserving a little museum somewhere, at the level of Escher, say...

Now, Cope's work is not a mere plotting of a graph. And it spews out something that is recognizable as previously unavailable examples of a previously accepted mode of art, in previously accepted form. So it allows for direct comparison.

As for Cope having to guide the computer, please notice he did this out of choice. The previous version of the program did not need it, and it did pretty well. Cope wants to use this version as a tool to help him produce his music, because of his interest as a composer rather than as a scientist, so interactivity was by choice. Also, you could use this interactivity to guide the computer, by example, on the musical tastes of humans: a bit like we learn by example what the public likes (most artists tend to think highly of their work until public reaction teaches them their place). The computer is a martian, it needs to know what humans enjoy :)

As for the computer not composing anything because it just follows rules, you are arguing definitions of "composing" (see what I said about "intelligence" above). Objectively, the computer spews out compositions that didn't exit before and that are mistaken, by specialists, for Bach's compositions. You are arguing that it follows rules (including, mind you, the deliberate braking of rules), so it doesn't actually compose. But you are defining "compose" by opposition to what the computer does, it is just what I talked about above. You are not telling me what "compose" is, or how human brains do it (how do you know brains aren't algorithmic and if they aren't, what are they?), you just ensure that if a computer can do it then it is not called "composing" anymore. How do you know that "following rules" is not what a human composer does without realizing? What is a "style" if not that? Read Cope's argument about plagiarism.

(continues)

António Araújo said...

Now, your other argument is more concrete: if it can only compose in the style of Bach then it is derivative. But Cope has answered that argument by making a new program that creates music that doesn't follow a specific musician's style. Well, perhaps in the end it follows Cope's style in some way, or a mix of many styles, so it isn't much different.But anyone has a style, where does it come from? How do you know a man's style is not also a mix of what styles he observed (all the way back to copying natural sounds of birds, water, etc). The point is, we don't know, because we don't know how we work inside our brains!(in fact these studies are interesting because success in objective results sugests one possible mechanism we may follow ourselves)

Those are unobservables and we can argue endlessly. But this is what is objective:

Cope makes it reasonably to expect that we will have machines that produce music that has not been explicitely composed by humans, and will be able to comply to human taste enough to compete in the market, as long as people aren't told it was done by a dirty slave (sorry, computer). *That* is an objective fact. I'd say if it is objective enough to put a lot of artists out of a job, it is objective enough to be important.

And please see the reactions to those early compositions. Many people thought the machine was actually a *good* composer before they were told it was a machine, so for some people we have already arrived! But suppose you think not. Well, Big Blue can beat Kasparov. Can you guess what computers will be able to do with music say 100 years from now? I bet that in a 100 years the most successeful commercial music will be routinely made by computers and you'll have computers spewing out oil paintings in the style of Vermeer. If they'll be art, I won't argue, that argument has no end, and I don't argue over definitions: I just argue that people will think it is art unless they know the author.

Laurence John said...

Antonio, i think you've missed my point. let's say Cope started his programme running with as little intervention as possible (notice that he had to add the 'randomize' factor because the early musical results were too bland to US, but not to the computer). let's say he didn't influence the computer at all with the structures found in Bach and just pressed 'go' on a programme that started simply but had built into it the possibility of endless complexity. the results might be interesting, but they'd probably be for the most part, machine-like, irritating and maddeningly complex. why ? because the computer has to be told what is acceptable and what is not. it is not 'composing' because it cannot make conscious decisions by itself regarding what sounds good or bad. if 'compose' to you means simply 'create a musical sequence' then yes, the computer is composing. i suppose a garden-hose is 'painting' too if you pump red paint through it and let it flail about on a gallery floor ?

human decisions put shape to sequences of notes / lines and colours and that makes them interesting to us, not those things randomly in themselves.

Anonymous said...

António Araújo said...
"but maybe not so hard for abstract art."

Agreed Antonio; I had written that in a draft of my post but later pared it out. I was merely trying to argue that one must have an "authentic" intelligence of art before an "artificial" intelligence can exist. I suppose it is conceivable that a computer could actually make the "authentic" discovery, though.

António Araújo said...

> i suppose a garden-hose is >'painting' too if you pump red paint >through it and let it flail about on >a gallery floor ?

It is, if you write an artist's statement to go with it! :))
You just described about 80% of modern art! :)

>let's say Cope started his >programme running with as little >intervention as possible

I do get your point.
But suppose you start a human with as little intervention as possible! What will happen then?

Without any teaching, what will happen to a child? Will he compose anything? What sort of music? How many generations from a few feral children shouting like monkeys to Bach? I think you are not giving the computer a fair chance.


>human decisions put shape to >sequences of notes / lines and >colours and that makes them >interesting to us,

The feral child probably starts with: a pre-programmed sense of appreciation for some sounds and distaste for some others. That was put there by evolution (possibly a side-effect of programs destined to detect dangerous sounds, or to compete in mating rituals, etc). I think it is fair to give the machine that sense, as it is mostly accidental and survival- related and the computer (like a martian) might be given any other sensibility that we indeed would not appreciate. Because we don't know what that sense *is* in itself (our inner program), we give the computer instead a sample of what we enjoy and ask it to extrapolate. I think that this added task is already putting the computer at a huge disavantage, and to ask for more is simply unfair. If you insist that the computer has to make it interestinf *for us* then you cannot deny him the programming of taste that your genes and culture gave you. Hey, for many people japanese music sounds like noise! That is how specific humans are!

Notice that a human child also is given examples and asked to extrapolate! The feral chils will immitate natural sounds, or create simple rithms that resonate on his neurons (influenced by his beating heart? By counting time? By natural rithms? Who knows). The 20th century child begins by learning the music of generations past. How many musicians in a thousand do anything actually new? Even Bach was an extremely able composer working within the framework of his time! He is not generally seen as a great inovator, except in complexity, I think; mostly he is seen as the great final genius of a mostly declining style.

There are different musical trends in different (isolated) cultures (take japanese music! Why didn't they develop the baroque style? Why didn't we develop japanese style-music?). You can see that for generations whole peoples would not stray from certain styles! Why? They are all, like the computer, building on what they were taught, and on initial accidents. Only a few actually make great jumps from the trend. Are you sure the computer can't do that? How many people really make those jumps? How many computers have produced music? One. Wait for the first thousand and you'll have enough of a sample. Allow for a good random generator in there and you may find some great musical inventions. Allow furthermore for our tastes to be taught to the computer, and those inventions may even appeal to us.

Be that as it may, I have simply argued for what I've argued: I bet the machine will do what I stated it will do, that's all. I'm not calling it art. I say we'll call it art if we don't know who did it. I say we will be fooled, if you prefer to put it that way. If we agree as far as that, then we have no disagreement.

António Araújo said...

etc, etc:

representational art is much harder because the computer has to know so many things about reality! Drawing a man pouring over a chart, with a compass in hand and a telescope near him has so many layers of meaning beyond the merely visual that the computer has to have perhaps a full AI to deal with the parts that have nothing to do with composition, lighting, brushstrokes, etc. Without a full AI it will me limited to interesting rendering and composition (not full composition, mostly just cropping or viewpoint placement, nothing "semantical") of a previously human-determined scene.

But abstract art is like music. It doesn't need to have any meaning outside itself. It just has to massage our neurons in the right way.

Actually, bombarding our senses and watching for the reaction is the way to do reverse engineering of our brains. That is why teaching the computer compositions that are known to work on humans is a legitimate way to try to give it a working human "taste". That is how you make a copy of a "black box". There's no other way I can imagine, except by subjecting it to some simulation of human evolutionary pressures themselves, but that is asking way too much.

António Araújo said...

etc,etc:

an exception: nudes, ARC style. Basically no meaning, just posing and rendering. Teach the computer about interesting poses, teach it the rendering styles, allow for some variation. Computer directs the model to pose by giving it directions, lights it according to standards, takes a photo, renders the photo by ARC standards. The only hard part is the robotic device to actually handle the charcoal or oil application on physical support. If you only ask the computer to fool a viewer of a jpeg "copy" of a non-existing origin, then probably it could be done right now, given the right incentives, for much less than the manhattan project. :)

Tom said...

Just playing the flip side, it seems to me the best art always arises spontaneously. Like life, it is always new. No one nor no thing "thought" a flower or the human skeleton they emerge from nothing with no thinking. Or as Ramana Maharshi said, "in due course, we will know that our glory lies where we cease to exist."

It is also interesting how much of nature's beauty or patterns are forced, if you read books like," On Growth and Form" or any Philip Ball book.

Kpakpo Akwei said...

@ Tom: What does spontaneously mean? Impressionism was groundbreaking for it's time but one could easily argue that it was in fact a culmination of all the Art prior. Monet did not "spontaneously" discover a new way to paint he distilled the influences and experiences he had and channeled it in a new way yes but spontaneously, I'm not so sure. All the major arts movements in the West seem to have evolved in opposition to each other rather than being just created.
Again can we agree on what spontaneity means?

At any rate, if the output (man vs machine)theoretically is the same then does it make a difference what we call it or by what process it was obtained? is Art/should Art be, by definition, exclusively human?

This guy makes Art that could easily be "taught" to an artistic AI. Would you agree?
http://www.ted.com/talks/tom_shannon_the_painter_and_the_pendulum.html

Anonymous said...

António Araújo said...
"But abstract art is like music. It doesn't need to have any meaning outside itself."

I'll have to disagree with you there; music expresses an underlying rational mathematical order, whereas abstract art can be (and often desires to be) utterly random.

Mellie said...

Any attempt to come up with formulae for beauty is misguided. Beauty is literally "in the eye of the beholder" - it is a human response to our external world. It is not based upon any innate properties of matter, in any particular material or shape or indeed the golden ratio. Human beings can find literally anything beautiful.

If a computer created art via programming we'd still respond to it as if it was done by a human, just as we respond to paintings done by chimps. We cannot perceive anything except via humanised senses.

What we perceive as beauty is our own humanity affirmed in the things we make and in the objects we perceive.

António Araújo said...

>I'm a bit worried that the most >common application of this new >science will be in advertising

then again, the most common application for art has often been advertising. Arrow collars anyone? :)

>but can they explain WHY that >geometry is attractive?

A good theory should predict, not explain. If it predicts reliably that something will be found attractive, then you are on to something. If it merely explains, you can never know it is not just wishful thinking and nice sounding words.

>All these scientists do not >realize that the beauty or >attractiveness in art is not >because of perfection.

They don't? Citation needed. :)

>Human beings are attractive >because of the asymmetry and not >because of perfect symmetry.

Surely that is taken into account. You know who really works on these matters? Plastic surgeons who have to reconstruct horribly damaged faces. There is endless discussion in plastic surgery books about measures of beauty, including taking care to avoid "synthetic" perfection like you mention, or the "uncanny valley". Of course, things are still in a primitive stage, but everyone is aware of such symmetry considerations, its handling the "materials" that is harder. I really admire the efforts of these guys, and they need all the help they can from both artists and scientists. Optional cosmetic surgery, however, is still a joke today, even rich actresses can't get a nose job that doesn't look like it came from a factory (I wonder why they keep doing it!). BTW, plastic surgery books, and odontology books are the place to look for antropometric discussions related to human aesthetics. Much better than the rehashed nonsense found on most art books (like rules of thirds and golden ratios ad nauseum).

There is a long tradition of science and art walking hand in hand: Paolo Ucello's obsession with perspective, Leonardo with anatomy (and just about anything else), Durer with geometry and (the first book on) anthropometry, and all the contributions of science to color theory, from Newton to Maxwell and so on. It is amazing to me that today there is such separation and even hostility between the fields. I can never forget that when I tried to learn color theory my art teacher recommended Goethe's nonsense and that fraud Johannes Itten, this in the days of CIE diagrams and Munsell (re)notation. Pure fear of concrete knowledge, refuge in high-sounding meaningless words and empty, circular definitions like "hue is the color itself"!

Well, I rant... the point is, science and art should feed on each other, not byte at each other. There never was any left brain/right brain incompatibility until people insisted upon it. There surely was none in the renaissance. Geometry fed art which fed geometry. Only since Romanticism started to spew out nonsense did people start to take pride in walking around using only half a brain (how people love choosing sides, especially in any right/left divide!). I think excessive specialization is the problem, we become balkanized, hostile, protective of our limited territory, fearful rather than curious, eager to believe that our orange is somehow better than another guys apple. Life is too short for that sort of nonsense, I think...specialization is for insects, we should revel like eager swine in the multiplicity available to us. (oink oink) :)

ok, I promise no more rants tonight. Back to work (coincidence or not, I should be writing course notes in statistics :p)

HiFi said...

Very interesting post. I suppose all these studies would go hand in hand with AI and the future of robotics technology.

Anonymous said...

Mellie said...
"Any attempt to come up with formulae for beauty is misguided."

Wow, Mellie; that seems like an incredibly broad statement. In the past couple of years I've been reading German and Austrian scholars loosely centered around the Vienna School of Art...Wolfflin, Riegl, Gombrich, Semper...are you aware of their work yet still flippantly dismissing them as "misguided"?

Tom said...

Hi Kpakpo Akwe

What I mean by spontaneously is, arising without thought. I am not talking about training. One trains hard and then something else takes over and does the work. Sorry if that sounds vague.

emikk said...

I go with the fabinacci ratio myself....it works every time!

अर्जुन said...

Testing, 1, 2, 3… Am I spam?

अर्जुन said...

""and you'll have computers spewing out oil paintings in the style of Vermeer""

If they are good, I just might click "paint".

What is with the dreams of machines & Munich!

David Apatoff said...

T Arthur Smith-- you're right, but it's our fault for developing what's called "psychical distance" between us and the previous generation of advertisements. We have dulled ourselves to their message so they are forced to use ever more sophisticated neuromarketing techniques to remain at the forefront of our consciousness.

Laurence John wrote: "an ongoing human delusion ... that hidden mathematical formulas can explain the mysteries of art / beauty / the 'divine'."

Laurence, perhaps they were on the right path but just lacked today's winning combination of information technology and pharmacology.

Don Cox wrote: "It is worth noting that the landscape picture is part of a tongue-in-cheek project."

Agreed, Don; the landscape project was an interesting social commentary even if it was a second rate conceptual artwork. Its meaning derives from the fact that it makes explicit what goes on implicitly all the time.

David Apatoff said...

Kev Ferrara wrote: "To skip the information and go right to the reward... that's simply chemical pornography... drug use... the very definition of mindless escapism. Which has nothing to do with Art and never will."

Kev, I'd like to think this is true, but it is at least conceivable that by going directly to what you call the "electrochemical reactions" of the brain, we can circumvent the limited bandwith of the five senses (which let's face it, are a narrow pipeline) and stimulate artistic reactions that are more pure and intense than the ones we currently get. Such feelings might not be as glorious or as hard-earned as traditional aesthetic responses but they might, with experimentation, still beat what we have now. Would you say they are illegitimate just because they are "chemical pornography"?

Allessandra Kelley wrote, "Have you heard of the 'uncanny valley'?"

I find the notion of the uncanny valley absolutely fascinating, and I have discussed it at length with animators and technical types at Disney. (Of course, I have known some real live humans who sometimes get that look on their face.) That is the subject of a future blog post.

David Apatoff said...

Stephen Southerland wrote, "we would like to think that the universe, human nature, and aesthetic experience only exist to be dominated, subdued, and redefined by us. If Art does not defy this notion, Art is not Art."

Steve-- I don't know, I've met a lot of artists who are genuine control freaks and construct meticulous little worlds of great precision. Not my taste, but still...

Antonio-- I am always especially appreciative when mathematicians and scientists weigh in here.

David Apatoff said...

Donald Pittenger wrote, "such research would yield a set of averages and would not apply to everyone."

I agree that would certainly be the starting platform-- a general approximation of taste or feeling that might resonate with perhaps a third of the population. That would still be a good starting point and a huge advantage over a competitor who doesn't know if their work will find favor with even one percent of the population.

But that would just be the starting point. Digital media make it relatively painless to fine tune a product for different populations, changing that blonde hair to black, or lengthening that nose. Surely one could increase one's chances that way. And if the stakes were high enough and we became a little better at implementing data, who knows? In the future they might be able to adapt that "average" platform to tailor a product that would please Don Pittenger!

Antonio Araujo wrote, "You know, knowing the average is all it takes to make a bundle, especially if the standard deviation is small."

I see you and Don were ahead of me on this point. I agree with both of you, we are like Carnival barkers playing the odds, only applying the latest informational technology and and pharmacological science as a replacement for a seasoned Carny barker's practiced eye. Hard to say which one would lead to a more reliable result.

David Apatoff said...

Kevin Ferrara wrote: "Yes, symmetry and certain proportions are attractive: That's the lesson plan for Day 1 of Basic Aesthetics."

Kev, obviously your aesthetics teacher did not see Ellen Barkin's smile in "Sea of Love."

Snoringdogstudio, thanx for the reference to Physics Buzz. Very interesting.

Don Cox wrote: "I can't think offhand of any scientists who were also good artists, but there must be some."

I would offer Leonardo as exhibit A.

António Araújo said...

>I would offer Leonardo as exhibit A.

I suppose it depends on what you call a scientist. People can get very specific about that and call Leonardo pre-scientific or something. Also, a mathematician is not a (natural) scientist. Words, again.

But if we take the common usage of the word, I'd say Leonardo must qualify, and also Durer (certainly a scientist as he practically invented anthropometry), and I'd include good old Paolo of the little birds (Ucello) and all the rest who developed perspective. According to Vasari, Ucello ruined himself because he kept working on difficult problems of perspective instead of painting the pretty pictures clients wanted. Apparently his wife kept calling him to bed while he raved on and on about the beauty of perspective! If that isn't a scientist, what is? :)

kev ferrara said...

David,

I didn't say only symmetry and proportions are attractive. (foul, I cry.) But symmetry and proportions are the most obvious and basic characteristics associated with attractiveness. Ellen Barkin's twinkle is a much more complicated matter... which has nothing to do with the computer software we were discussing (foul again, I cry.)

Would you say they are illegitimate just because they are "chemical pornography"?

Here we are running into the same problem again. That you seem to think any human endeavor that causes emotion is art. Provided it is labeled art and you agree with the label.

Aesthetic metaphoric epiphany (to coin a phrase) provides a rush of dopamine and endorphins in the brain... To think we can skip the aesthetic metaphoric epiphany and go directly to the dopamine and endorphin rush... and somehow we aren't simply abusing a drug? Maybe we should declare Albert Hoffmann another Rembrandt? Maybe the New York City Marathon runner who took the subway to the front of the pack wasn't cheating after all?

Believing that art neither conforms to any particular set of principles nor has any inhering characteristics... leaves many speculations open for debate, including the idea that art can dispense with content and form altogether and still be art.

I'm beginning to think I should start a blog on the law as retaliation. My first order of business would be to declare that trials are unnecessary as long as somebody wins the case. And that nobody really knows what the law is, except that it causes verdicts.

अर्जुन said...

""I'm beginning to think I should start a blog on the law as retaliation.""

Until then, those interested can start here~ http://lawschoolscam.blogspot.com/

Laurence John said...

"In the past couple of years I've been reading German and Austrian scholars loosely centered around the Vienna School of Art...Wolfflin, Riegl, Gombrich, Semper..."


etc, etc i assumed your Star of David template in the first post was a joke. can you share any insights into 'formulae for beauty' gleaned from the above reading list ?

Anonymous said...

Laurence,
I was being utterly sincere; as I said the star pattern is something I have found to be a reliable indicator of female facial attractiveness, obviously for me personally and I'm not suggesting that everyone should find it to be so. I really don't know what to say beyond that; if someone finds it absurd I'm perfectly o.k. with that.

Regarding the Vienna School of Art authors, I definitely could not do them justice in one or even several blog comments; I suggest you investigate them for yourself via the English translations as I have. I have found them to be very worthwhile. Dover Publications' Principles of Art History by Wolfflin is a fairly inexpensive way to start; here's a website with supplementary illustrations:
http://tinyurl.com/2abzdrg

Mellie said...

"Wolfflin, Riegl, Gombrich, Semper...are you aware of their work yet still flippantly dismissing them as "misguided"?"

There's nothing "flippant" in my view. I don't believe a formula will ever "explain" beauty, because human beings can find anything beautiful. Someone comes up with some set of "ideal" facial proportions - well, there are millions of people out there who do not fit them, yet may be considered beautiful.

Or do symmetrical faces really indicate "better" genes? A pretty tough verdict on the healthy, gifted, successful people who don't have symmetrical faces.

Beauty is ultimately not an innate quality of matter, true and absolute across eternity, but a human response. Now it may be that we respond to certain qualities more, on average, than others, e.g. for reasons based in our evolutionary history. This will be a highly complex relationship. My argument is that beauty cannot be reduced to a formula.

Anonymous said...

http://tinyurl.com/2497tug

Laurence John said...

alluding to a weighty sounding reading list doesn't prove much unless you're actually prepared to share some of the insights you've gleaned from that reading.

Anonymous said...

Laurence John said...
"alluding to a weighty sounding reading list doesn't prove much unless you're actually prepared to share some of the insights you've gleaned from that reading."

Well, in that case...nevermind; I was just bluffing.

Laurence John said...

aw, at least give me a quote.

lj said...

what an interesting post. I liked the geometry part with the woman's face. I just wouldn't want to be that woman and have viewers say "She's not that beautiful!?"

I just joined your page. I got lots to catch up on. wink.

David Apatoff said...

Antonio wrote: "if we take the common usage of the word, I'd say Leonardo must qualify, and also Durer (certainly a scientist as he practically invented anthropometry), and I'd include good old Paolo of the little birds (Ucello) and all the rest who developed perspective."

Antonio, back in the era when art was more closely related to empiricism there were a number of hardworking artists such as the ones you describe who I think qualified as at least first cousins to scientists (just as mathematicians were first cousins). Keep in mind that the definition of "science" in that era was broad enough to include astrology and alchemy. I think your point about perspective is right on the money-- it was one of the first truly effective and beautiful efforts to apply the quantitative tools of math and science toward enhancing the qualitative image. I can see how Ucello thought perspective was breathtakingly beautiful. I have no idea what Ucello's wife was like.

David Apatoff said...

अर्जुन -- I love the painting machine but I could have lived a full and happy life never hearing the disco version of "whiter shade of pale."

Mellie wrote, "I don't believe a formula will ever "explain" beauty."

I agree, and yet I do believe we can, with words and thoughts (perhaps even aided by numbers) identify things that are usually not beautiful and things that ususally are beautiful and search in good faith for the distinguishing characteristics that make them so, at least for a majority of the viewers. The fact that we can never seem to come up with an airtight formula doesn't mean that we are helplessly cast adrift in the face of the inscrutability of beauty.

Returning to the fruitful metaphor of math, I think it is an honorable and fitting pastime for us to chase that asymptote; we gain meaning from the fact that our curve is tending to zero as our distance from the origin increases to infinity. We recognize that the meaning we derive is diminished by the fact that we will never reach zero, but it is not obliterated altogether by this fact.

I suppose the hard work of approaching that asymptote is much of what this blog is about.

अर्जुन said...

"…but I could have lived a full and happy life never hearing the disco version of "whiter shade of pale.""

Soulless bastard.