Purple Swans

Scientific research works like this: you can not prove a statement about an entire group of things is true (unless you have access to the entire group, which tends not to be the case), but you can prove it is false with just one counter-example.

Scientific research then involves coming up with a statement about a group of things, a hypothesis, and then actively trying to disprove it. If you try really hard to disprove it, and if others too, and you fail to disprove it, it becomes more and more likely that your hypothesis is true. You haven't proven it, but you can make it arbitrarily likely to be true.

For example, a hypothesis can be that all swans are white. I go out to lots of cities, rural areas, and countries, and all I see are white swans. You go out and do the same. You go out to different cities and countries, even, and you, too, only see white swans. It is becoming more and more likely that all swans are, indeed, white.

We even come up with a Latin expression for it: “rara avis in terris nigroque simillima cygno” (“a bird as rare upon the earth as a black swan”).

But then, one day, in somewhere in 1697, a Dutchman wanders around in Australia, as one does, and, lo and behold, he sees a black swan!

The hypothesis that all swans are white, a statement that had become increasingly likely, had now been proven to be false!

You can read more about it here.

Facts. Fact-finding.

I am reading the book “make it stick”. It presents scientific research into educational psychology. It's tremendously interesting to read. Psychologists have discovered a lot about what works and what doesn't when it comes to learning. I have been trying to map these theories to learning to draw, and I realized that regular school learning is about learning facts. The above is an example: it is documented that people believed all swans were white until someone found black swans. These are factual details.

With drawing, it doesn't work that way! What is “true”? What is “a fact” when drawing? I can draw a hyper-realistic portrait, which is one representation of a person. I can also draw their caricature, one that deviates significantly from their true appearance, but that instead conveys their character, their personality. Which one of these two is more “truthful”?

This is important because the book “make it stick” convincingly shows that test-based learning, where you test yourself immediately after reading a text, is far more effective than re-reading for memorization. Actively recalling the information is harder, but it also makes it more effective.

An important step is to correct your results after that, because otherwise, you will memorize the wrong answers you came up with.

On a side note, it works the same way with my memory-drawing routine. You draw the thing from observation, then try to recall it from memory, and then correct mistakes. If the reference isn't too hard, you can draw it from memory after that! You retained it! You “stopped forgetting”, as the book calls it. I found out for myself that this works for drawing, and I'm elated that there is even scientific research to back it up! The research dates back to 1939. What other useful information are scientists hiding from us? They tend not to bother with popularizing their findings, unfortunately.

But here is the rub: a text can present facts, and you can be quizzed on those facts, and a teacher can then correct your answers.

But when it comes to art, what is “correct”? Is the precise copy of a portrait “correct”? Is a caricature not “correct”?

If I forget details of something I am drawing from memory, it can be fun to imagine what it could look like, change things, and add details. Is that “wrong”? Kimon Nicolaides even proposes to do the memory drawing the day after you saw the thing. If you try that, you will notice you've forgotten even more, and you're free to imagine lots of details to fill in the gaps.

While in regular learning we learn the facts a textbook presents, for example, in art, it is less clear-cut because there is a creative component. This creative component is crucial, allowing artists to deviate from surface truths to express deeper truths.

The whole point of making art is to respond to things and express your thoughts and feelings about them.

If you want to check that you got the proportions right, you can use the AR Drawing Tutor on “Practice Drawing This” to overlay the reference image over your drawing using your mobile camera. You can stick to the facts and draw swans white or black, so to speak.

But we artists also have this thing called Artistic License. We don't have to stick to the facts.

We can paint a swan *purple* if the situation asks for it.

But how then do we “test” our art? How do we check if a drawing is “wrong”?

As I argue in this YouTube video , art is a personal experience; it is an interaction between the art and the viewer. The viewer can decide what the piece of art means to them, how it makes them feel, and what it makes them think. The artist did have an intent for the piece, but the viewer might engage with it in a different, unintended way.

The usual approach to feedback is for one artist to critique another's work. This often also involves advice, which can be problematic because it assumes the critique-giver knows the artist's goals for the piece.

The Critique Server on the “Practice Drawing This” website suggests you look at a work carefully and try to write down five things you notice about the piece. This makes it objective: it is a fact that you noticed that, so it becomes factual again. The artist can then figure out what to do with that information. Are they okay with the viewer having experienced it that way? Or were they going for a different effect? The artist can then begin formulating and implementing changes to their process, if they so desire.

The Critique Server allows the artist to formulate questions about parts of their piece they want feedback on. Still, there is often a deeper reason behind that, so giving appropriate advice can be hard without the ability to interrogate the artist. Plus, there may be aspects the artist is not seeing themselves, and a report on what a viewer sees might point them out. The artist is then free to decide what to do with these observations by a stranger, of course!

Yours sincerely,