Scientific experiments are expected to be reproducible and repeatable. First you design an experiment. On paper. Then you try it. Then, as you do it, you might recognize that you need to do this and that in a particular fashion. The experiment, and the result, progresses as you repeat it.
Let’s take a concrete example: preparing a suspension of particles. You put 3 g of water in a pill glass. You use a pipet, as it is the most convenient way to transfer a liquid. Then you add a dispersant, an additive that helps mixing and separate them. One drop of the smallest plastic pipet works fine. Two drops, same. You weigh out 3 g of particles in a plastic cup, and transfer them into the pill glass. Some of the particles stay stuck to the plastic cup but you can assume it will always be the same quantity. Then close and vortex to start mixing. You know if you see agglomerates that are really big, it is better to break them now with a spatula. Then you put it under the ultrasound probe. You can tell by the sound how it the viscosity of the suspension. Leave it 5 minutes. If too many particles are stuck to the probe, do it longer. If it gets too warm, cool it down. Vortex one time in between just to help. Do not let it open too long under the ultrasound probe as you don’t want water to evaporate.
Art pieces are supposed to be grabbing the instant, to be emotional, to be unique. But usually it starts by a draft, an idea. On paper. Then you try it. Then, as you do, you might recognise that you need to do this or that differently. The artwork, the emotion and message it conveys, progresses as you repeat it.
Let’s take a concrete example: drawing the character of peace in calligraphy. You take a piece of paper, the paintbrush, some ink. Dip the brush in the ink. Tap it to remove the excess of color. Add water to dilute a bit. Reload the brush as you know it should be enough for the entire symbol. Start with the brush at 45° then turn it to 0°. Press on the brush at the middle of the stroke as you know it will release more ink. Breath at the same time. At the end of the first stroke, end at 45° again. Slowly lift the brush but still firmly. Start the second stroke at 45° again. Do not go to fast as you might have white windows in the stroke. Press a little bit firmly on the paper, but not too much as you don’t want to use all the ink. Use the end of the brush to make a thinner end of the stroke. Lift, go to the next one. This one you know you manage the curvature better when you go faster. This will give a sensation of movement that you like. Again, 45°, 0°, press, lift, 45°, slow down, breath.
You had to prepare at least 10 suspensions before finding the right procedure. You can tell by eye if it’s good or not and if you should try again.
You had to draw at least 10 times the character before finding the right gesture. Your whole body is controlling the result and if you don’t like it, you try again.
The similarity between the two processes is striking.
Scientific experiments can be reproduced after training of a researcher by the person who set up the experiment. Same for the calligraphy. Does it make the experimental result less scientific? Does it make the art piece less artistic? Maybe one way to slow down the reproducibility crisis in science could be to train, as a master in art would do, other younger researchers. Maybe the issue has more to do with knowledge and practice transmission than with the true reproducibility of data.