There's a generative aspect to the scientific and artistic processes, and also an evaluative one. The generative aspect involves how new theories or works are conceived -- spark of insight, analogy, accident, trial and error, careful observation, statistical analysis, etc. The evaluative aspect applies aesthetic judgments to theories or works in attempt to understand their applicability, importance or general goodness. Aesthetics judgments might be pragmatic, relating to function, performance, ability to describe or summarize, optimizations. They might also be cultural, relating to preference, social norms (or their violation), familiarity or tradition.
It seems to me that what goes on is mostly evaluative, perhaps because the generative is harder to invoke or control. Perhaps because it's human nature to fit observations into a pattern, and these patterns require everyone to develop evaluative rules for what fits and what doesn't. There have been techniques for the generative though-- at least in the arts, e.g. automatic writing, oblique strategies, etc.
Perhaps the generative and evaluative aspects of the creative process are simply inductive and deductive reasoning, respectively. Where does "reductive" reasoning fit, i.e. refining broader rules into more specific ones as new data comes along? Is there a fundamental distinction between the rules themselves, and the domain objects being reasoned about when it comes to the nature of creative processes? Scientific theories seem closer to rules in the sense that they are conditional on their assumptions, yet we apply aesthetic judgments to them as if they were the objects of the creative process.
Are there universally valued aesthetic judgments? Is there a scientific basis to the sense that a simpler theory or expression is the more "correct" one? Is there any intrinsic value to harmony, symmetry, or balance? How is the coherence of a group of judgments related to truth of the underlying theory or work?
Are there fundamental distinctions between a creative process with the objective of exploring an unknown space (e.g. string theory) vs. one that is purely constructive (e.g. Pollock slithering on a canvas)? In the exploration case, the theories or works can be judged against an external measure or reality ("NP complete"?). Does the purely constructive creative process even exist since the act of putting it out there invites evaluation which in turn defines a context for its existence?
I know I haven't related this at all to the creative process of software, painting, writing, etc., but I think at the level I'm describing, it's all the same. (Forgive me if this is all covered by Kuhn -- I still haven't read him.)
There are several interesting points here. Generation and evaluation are perhaps only 2 of several steps in creating anything-whether scientific or artistic. Another step is revision and tinkering. The drafts from Mendeleev and Blake show that. But within tinkering and revision, generation and evaluation continues.
I was talking to a mathematician yesterday about how he works. He told me that he starts by exploring a space of ideas, usually triggered by a real-world problem. He tries to visualize the context, even when it is fairly abstract like groups. He says he sees elements being mapped to others through automorphisms, or he pictures spaces when he's doing analysis (the type of mathematics that is in contrast to algebra). From these pictures he tries to see patterns-things that re-occur are which are repeated in different places. Then he tries to characterize what about the space he's picturing makes statements about these patterns be true. Then he tries to prove some of these truths, but this sometimes requires him to reformulate the necessary parts of the context. These correspond to other axioms or assumptions in the theorem. Sometimes these revisions of the theorem prove too unpleasant and the truths become less interesting.
This sounds like both art and science to me.
In creative writing, a common worry is that the "inner critic" will overtake the generation phase, causing more fanciful ideas or gestures to be first edited and finally edited out, which can cause the generative process to shut down-this is called "writer's block." But it is possible-at least it has been for me-to practice generation. It's a kind of letting go. Sometimes I call it "shallow thinking," harking back to PaulFeyerabend.
On simplicity and aesthetic judgments, I believe we (as scientists) have gone a little too far in that direction. We need to embrace complexity more. One could even argue that it was our infatuation with simplicity that caused us to miss all the phenomena that we now characterize as being the subject of complexity science-phenomena that we didn't really pay attention to for millenia. And, it took the art of fractals and Mandelbrot sets to spark our curiosity enough to take it all seriously-after a period of thinking of fractals and Mandelbrot sets as games or harmless diversions.
Ignoring complexity seems like gain is up too high on the evaluative parts of us.
Finally, is there much of a difference between creation and discovery? When a physicist is looking at string theory, is that a discovery or a creation? And what does it mean that almost all of the physicists we revere are/were theoreticians?
There is a nice book-intended for writers but maybe it's relevant to scientists-coming out soon called "Maps of the Imagination," by Peter Turchi (TurchiMaps) who is the director of the MFA program that JanetHolmes and I graduated from. It is about how writing-an act we think of as creative-is very much like discovery, and the link is that the creative part is like map making. The generation is like exploring an unknown land, but even so there are limitations on what can be explored, and often the material or landscape leads the explorer (and writer) in a particular direction. There is evaluation of which ways to go, both literally and metaphorically, but what we put on our maps is determined by revision and tinkering. Moreover, what's on the map is guided by its purpose-what we are trying to show. This map-the London Tube-is considered one of the best maps of anything, but as you can see, it is highly abstract. One could imagine this map being produced by a process of discovery, but is it also an act of creation?
http:/www.dreamsongs.comExtravagaria/Tube.gif
-rpg
I think of tinkering and revision as a way to generate ideas.
Perhaps it is really a different step.I'm got a project that is reusing two pieces of Java software; Eclipse and a Fortran parser. The three students who have been working on this project have either graduated or are graduating this summer, and there will be a new team working on it in the fall. While the students are still here, I am working to learn the system. I had the big picture already because of working with the students, but I was always "looking over their shoulder" instead of actually doing the work. Now I am doing the work. I'm fixing bugs and making failing tests work.
Working with code always gives me idea for improving it.
In fact, I think of tinkering with code as the main way tolearn it and to generate ideas to improve it. Before i tinker with something, I'll see whether there are some documents that describe it, and I'll scan the code to try to figure out its major structures, but I never feel comfortable with it until I have made some changes to it. So, I think of tinkering as a tool for generation. But it is certainly different from other tools, and is interesting in its own right.
Letting the unconscious in (Kekule's dream of a snake biting its tail leading to his envisioning the Benzene atom as a circular structure) doesn't happen often in science in my experience. I do think this is where Oulipian exercises and such like can be of help.
Something seems to happen when you get something down on paper or on a canvas (or into a code file) that can't happen so easily in the head. I've heard wild-ass theories about how people whose corpus callosum is compromised (possibly through surgery) communicate between the two halves of the brain by using their eyes-write something down, and it's read by both halves of the brain independently. This reminds me of a biological phenomenon called stigmergy.
Seeing something on the page, so to speak, seems to trigger some other parts of the creative process. -rpg
WarrenHarris questioned whether "Are there universally valued aesthetic judgments? Is there a scientific basis to the sense that a simpler theory or expression is the more "correct" one? Is there any intrinsic value to harmony, symmetry, or balance?"
IMHO, the answer to all these questions is NO. So, I'm very cautious about talking about a "truth". Truth is relative, it varies across time, across space, and across cultures. So, I don't see my work as a musician or computer scientist as a search for truth. - FabioKon.
Ah, but if truth is relative, then the search is ongoing. Let's suppose that the following statement is true: Both scientists and artists seek the truth. What you're saying is that, for art, truth is relative and changes across time, space, and cultures. Then this must be the case for science as well. And the history of science doesn't provide much evidence to contradict this. In my view, "truth" is a slippery concept. -rpg
I'm still thinking about this question about the universality of concepts like harmony, symmetry, balance and simplicity. I know it's easy to argue that these are not universally held values/truths, and indeed it's always possible to find someone who doesn't value them. But something in my gut tells me there's something "more correct" about them -- perhaps something that transcends the relative truths of people and history. The universe seems to resonate when concepts are symmetric; the law seems more fair when the principles are balanced; music is more soothing when the frequencies are harmonic; theorems seem less refutable when they are simpler.
You can call these subjective judgments, but there's something that feels deeply objective about the fact that physics and mathematics are as simple as possible, but no simpler. Fractals and chaos seem more complex at one level, yet it is our newfound ability to characterize, model and understand those domains that brings a new kind of simplicity/order to the pandemonium.
Sorry, but I seem to be off on a tangent triggered by rpg's idea that creativity is a search for the truth. So why is it that I seem to gravitate toward art that is dissonant, asymmetric, unbalanced, complex and in general, unexpected?
The introductory page suggests that artists and scientists share the important goal of searching for truth. I'd like to agree with that but I'm not sure I can. When I started writing, I thought I was going to explain the world to myself and then figure out my place in it. But I was apparently afflicted by the grand narrative dis-ease where a "truck" is just a black scrawl on a page. That it brings up a signified to me doesn't mean it brings up the same thing with wheels to the Malaysian rice farmer who doesn't speak the same speak or the indigenous person in the rain forest who's never seen one. At some point in the proceedings I was cast into the blur where language only points to other language and of all that meaning is temporary, instantaneous and arbitrary at best. Does the reader want to imagine April going wild? Some readers but not the atholes. We don't. We're interested in the way it helps itself to another vodka martini, the way himself never becomes itself without a nod from the spirit guides. You can swing the big axe but you still won't hear the letters rearranging themselves. What started as I finished as I, the I small with fist, then someone's waiting museum.
So how then do I generate anything? And is it any "good"? To the former, I gather up what Dean Young calls all the "strange night barkings" and put them on a page. This drives some other poets crazy because I think of it as just a collection of lines. The lyricists want me to follow an image and see where it goes. But for me, it only goes to the next image or concept. The next is is only around the corner, with it beside it. Mostly I'm thinking about the not I, the I bar, the bright I, the I of our forefathers, the I liking pie. If sense appears to be vacationing up the coast then you begin to get the idea.
As for the good, well the boss is calling right now so I'm going to have come back to that.
How have you not enacted the statement of this workshop?
Your observation that "truck" is just pixels on a screen could apply, should, and does to scientific marks like "superstrings" or an equation. -rpg
Communication is two-way. The way that an author finds out
whether the communication is working is by audience communicatingback. Communication is always relative to a particular audience.
Sometimes I write to the practicing programmer, sometimes to thecomputer science academic, sometimes to the beginning computer
science student, sometimes to my wife, sometimes to myself.A document that is good for one is not good for another.
The fact that a document communicates to one person and not another
does not mean the document is defective or that what it says isinvalid. Whether it is defective depends on the intend of the author.
If the author intended for it to communicate with someone and itdoes not then it did not achieve its intended effect, i.e. it is
defective. But if you are writing for an educated English speakerthen there is nothing wrong with not being able to communcate with
a Malaysian rice farmer. -RalphJohnsonI believe art and science are always a search for truth - the truth that the artist and scientist believes or feels or just "knows"; a truth they feel has not been yet revealed by others; a truth which may change the world - or a truth which is simply ridiculous to the world. Many deep truths are both (relativity, maybe string theory, maybe the Gaia hypothesis).
Software for me has always been an act of creation, and an immensely fulfilling one. As I tackle the problem I am facing, as solutions are arrived at, refined, discarded (either in my head over a few seconds, or as physical code over minutes, hours, days, weeks), I am always aware that what I am doing is exploring. I am exploring ideas; I am exploring the problem from different angles; I may be exploring new tools, new languages; I may be exploring other creative disciplines to gain insight into this problem in the software discipline. Most importantly for me, I talk about the problem with others; all of it points to the same theme; something is being created by the connection of thoughts and ideas, and refined by trying it out. The more widespread the connection of ideas, and especially when receiving the ideas from other people, the more satisfying and beautiful the moment of "truth".
A line of poetry I have never forgotten, by Keats from one of the odes - "Beauty is Truth, Truth Beauty - that is all ye know on Earth, and all ye need to know." -gkc
I disagree that the search for truth is the 'most important human goal'. Words such as 'truth' and 'reality' are, for me, only used for convenience. They are useful in certain situations but loose all their meaning when enough context is removed. Individuals, whether they are artists or scientists, have various and changing goals driving their activities. Presenting these goals at the uppermost level of abstraction might make it appear that we share goals but, in fact, we do nothing of the sort.
Having said all that, I agree that there is a role for art in science and software. I agree, not because we have shared aims, but quite the opposite. I think that it is diversity that forces out new and exciting ideas. Indeed I would go further and say that historians, poets and comedians also have a great deal to offer. Whether or not this wealth of ideas can be teased out in a systematic manner is quite another question. –ChrisSteinbach