On Hockney, mathematicians, and the cross-cultural fallacy

Via Udell, a link to a paper that attempts to refute David Hockney’s theory that the sudden shift to photographic styles in the Renaissance was due to the use of optical projection.

It’s an interesting paper because it introduces what seems to be a new method of objectively measuring geometric deviations of paintings from “reality”, and applies that geometry to a chandelier that figures prominently in Hockney’s book :

In Sect. I we introduce the basic mathematics of homographies and plane-induced image registration. In Sect. II, we apply digital image registration to detect and measure geometric imperfections in the painted chandelier. The analysis of photographs of a representative sample of surviving 15th-century dinanderie is conducted in Sect. III. Finally, in Sect. IV we judge the abilities of contemporary realist painters in the absence of optical aids by testing the perspective of paintings of elaborate chandeliers done “by eye”. We find the accuracy comparable to that in the Arnolfini painting.

I haven’t sorted through the method yet, and in all likelihood I doubt I’ll be able to add anything to that side of the debate. But the fascinating part of the article is toward the end.

Underlying the arguments of Hockney and Falco is the assumption that good perspective cannot be easily achieved “by eye,” that is, without the help of optical devices.

To test their assumption, as part of our research, British realist painter Nicholas Williams painted two chandeliers entirely “by eye.” Figure 6 shows one of the two chandelier paintings he realized for us. Our perspective analysis applied to this painting resulted in a good but, as expected, imperfect alignment of arms. The average measured deviation was about 8.55% the image width, of the same order of magnitude as that of van Eyck’s chandelier. This experiment confirms that realistic-looking structures can be painted merely by eye, without the help of optical tools of any sort.

This part of the experiment shows more than a little historical naïveté. They took a 21st century painter, who has seen photographs and realistic paintings every day of his life, and found that he can paint photo-realistically.

What they of course miss is that to a person in a pre-photorealist society the process of mapping a three-dimensional reality onto a flat plane is considerably more difficult — that is, if such a notion even occurs to them.

Technology changes us. It changes what is possible to think. Take a top-notch mathematician trained before the computer and ask them to model the growth of lemming populations in a limited resource environment and watch as they drive themselves slowly insane in pursuit of the solution. Take an average Excel user or the most novice programmer of today, and watch how effortlessly they stumble into the algorithmic thinking that is crucial to the solution.

Does this prove it was possible for the mathematician to have come up with the solution on his own? Quite the opposite. The computational thinking that is now taken for granted in our society evolves not from native cognitive abilities, but from a cultural store that finds it’s birth in giving machines iterative instructions where output states become inputs.

Such historical thinking is crucial to Hockney’s premise. In the history of computing, one notes that Turing and Post came up individually with their theories of computing in 1936 — and from there the logical progression is to ask what was happening, technologically, that led two people to individually come up with the same model at the same time.

Hockney has before him a similar conundrum — the sudden explosion of photorealistic technique in Rennaissance art. And he searches for the technology that may have made such photographic thinking possible. He concludes, based on various effects of darkness and light, restricted focus, and orientation that optical projection technology may have been largely responsible for this sea-change in technique, and more importantly, this sea-change in thinking.

The research paper may be right about the irregularities in the painting for all I know. But in structuring their final test they show that they have very little understanding of cultural history. The briefest look at Derrida, the history of fractals, or the development of the slow-motion action shot in film would have demonstrated to them the fallacy taking a modern painter to prove a historical point. Technology, and the cultural store that evolves from it, changes us. My eight year old paints cups from life in a way that was simply not available to the Ancient Egyptians. My four year old concieves of quick events happening in slow motion. Both of them have a conception of language impossible before writing — and my four year old can’t even spell yet.

The ultimate thing to take away from Hockney is not whether the Masters “cheated”. It is to better understand the fluid boundaries between technology, culture, and our own cognition. Hockney may or may not be right about methods or particular paintings. But his impulse is right on target.

The hypocisy of the recent ECAR study

I had intended today to write today about the odd fracture in the recent ECAR Study of Undergraduate Students and Information Technology, a fracture between Chris Dede’s “technology as world changer” intro, and the rather pedestrian “technology as customer service” bias of the questions posed by the actual report.

I say “intended”, because when I went to pull a paragraph out of the document to demonstrate Chris’s line of thought, I found that this freely distributed PDF had been secured against all copy operations.

You want to cite the ECAR report? Start typing.

I know, I know. I’m supposed to be delighted this is free. I’m supposed to be thankful I even get to cite it in my blog.

And I suppose making me retype that quotation will just make me appreciate the report all the more, right?

Give me a break. This is ridiculous. To write a stirring intro about how the free flow of information is revolutionizing the world and then distribute it in a PDF format that disallows copy operations?


But yeah, I won’t be citing it. Mission accomplished.

Dr. Sander Lee on James Madison and Democracy

"I will be arguing, and this is a controversial opinion, that Madison’s support for democracy comes not because of a belief in the innate wisdom of the majority in society, but because he believes that in the absence of objective answers it is better for the honor of the individuals in society to allow everyone to participate…"

We’re having a Symposium on Citizenship here at Keene State, November 6-9. And since we are putting together a blog on the event, we thought it might be a good idea to go out and get some video of some of the speakers talking about their presentations.

We started with Dr. Sander Lee, who will be presenting on James Madison and Democracy (we’ll get the date and time soon) Renee Staudinger, a student intern of ours, filmed it with our lo-fi guerilla vlogger equipment (and did a splendid job — thanks Renee!).

Here’s the clip:

For more on Dr. Lee, check out his personal page. This will also be available on the Citizenship Symposium site (on KeeneWeb) as soon as we whip that site into shape. I’ve also x-posted this on KeeneFeed, a sporadic blog about Keene State that we’re trying out here.

The answer to the legal objections

The hardest thing to answer when you’re trying to start an institutional blogging community is what the legal ramifications of it are.

It’s not only hard to answer — it’s impossible. It’s a legal question. You’re not supposed to answer it.

The problem is that it’s hard with this sort of thing to start from a default position of strength. Without other institutions doing it, you’re forever the person that has to prove the negative: that there are not legal issues large enough to forgo the venture.

And unless you have a web-savvy lawyer on your side, you’re kind of stuck.

Until now. Because while I follow the good practice of never giving legal advice, I also follow the practice of pointing to the behavior of people smarter than me. So I plan to send this link around a lot:


That’s right. Harvard Law has a blogging area that’s almost an exact policy match to the UMW/KeeneWeb model.

And if the suits down at Harvard are listening to Weinberger (or whoever is at Berkman now), I’m feeling pretty good about our little venture.

Definitely check out Harvard’s Weblogs: not only is it a key bookmark that you are going to want to mail around, it’s a great example of a motley weblog community in action: you can read an analysis of contingency fees, a defense of Britney Spears, admissions talking about their favorite moments from admissions phone calls, and a discussion of Samuel Johnson-themed beer labels.

So… are there any further questions?

Fantasy Congress and the usefulness of wrong models

So, deciding I had not reached my full geek potential, last night I started a league in Fantasy Congress.

Fantasy Congress is like Fantasy Football — you pick a team out of all available members of Congress and the Senate, and during a specified season your team competes — if my Senators “stats” (for legislation passed or co-sponsored) beat all the other Senators stats, I win in that “position”, and so on.

It’s actually a litttle more complicated than that, but you get the idea.

So I started thinking — would this be useful at all in a politics classsroom?

My first thought was no. Because the model is wrong. It’s a lousy model to compute effectiveness.Â

To cite just a few examples: co-sponsoring is rated too highly in the formula relative to authoring legislation. And attendance, which has very little to do with congressional sway, is in the formula, as is level of news coverage, which tends to favor presidential candidates, regardless of their congressional duties. And “Mavericking”, the act of voting against your party, is often a sign of strength, but it also occurs where weak politicians find themselves sitting on districts with rapidly changing red-blue demographics — giving points for bowing to that pressure seems just wrong.

So that was my first thought.Â

My second thought was that any kid who left an American Government class able to articulate what was wrong with Fantasy Congress’s model of congressional power would leave with knowledge superior to that of 99.9% of Americans.

Where the professor in a class has enough knowledge to assist students in critiquing underlying models, a bad model can do you as well as a good one. I still love the Cognitive Arts dream of highly engaging, highly accurate simulations. But failing that, get an inaccurate simulation and have students critique it. They may learn just as much.

Our responsibility

So, like the WordPress junkie I am, I’ve been trying to recruit other Keene Staters here into my fold. Trying to get me some co-bloggers.

And so it was I broke the will of one Jenny Darrow, who leads and implements much of the Academic IT initiatives over here.

Her first posts are up and the third one is music to my ears. Here’s the thumbnail sketch:

The point is that we (anyone over the age of 35) assume that students don’t need support with any kind of technology, that somehow by some miracle they know how to configure their bluetooth access, create a blog, subscribe to syndicated content, create digital presentations, etc. It’s not a wrong assumption — it’s just not entirely accurate.

Agreed. I’ve often railed against the mindset that turns higher education’s abdication of responsibility in this regard into a “blinking VCR clock” joke. Students come to us with certain skills. They always have. Our job is to take those skills and help them refine and focus them. And if we lack the institutional capability to do that — well, it’s really not that funny.

We assume that because kids are blogging, they are blogging effectively. That these are binary skills, stuck either in the on or off position. But it’s not that simple.

But then Jenny goes further, because she goes to the 2007 Horizon Report and pulls the money quote from it:

Although new tools make it increasingly easy to produce multimedia works, students lack essential skills in composition, storytelling, and design. In addition, faculty need curricula that adapt to the pace of change and that teach the skills that will be needed—even though it is not clear what all those skills may be.

This is exactly right. And if we are going to teach them how to tell stories in this new media landscape, we are going to have to see new media as more than a bolt-on to existing courses, and certainly as more than a specialization within a major. We are going to have to see new media as a set of dialects in which all graduates must be fluent.

And if that means we have to set our VCR clock in the process, so be it.

Oh, and fellow rebels, go give Jenny some newbie love. For or against, it doesn’t matter…just enough feedback to nurse that blogging addiction….