Galileo’s greatest book, perhaps one of the greatest ever addressing the scientific method, was his 1632 Dialogue On Two World Systems, where he invented a dumb character named Simplicio and two smart ones to argue with him, including one “who’s read a lot but just repeats whatever Aristotle says. He’s erudite and ignorant, [or, as Gopnik says a bit later], “he could be very subtle and very silly all at the same time.” (Beginning to sound familiar?)
His other alter ego, Salviati, actually gets it, and here, as Gopnik says, is one of the nicest capsule summaries of the scientific method on record:
“Therefore Simplicio, come either with arguments and demonstrations and bring us no more Texts and authorities, for our disputes are about the Sensible World, and not one of Paper.”
Indeed.
The scientific method elevates debate over dogma, experience over authority.
Famously, Aristotle had written that small and large objects would fall under the force of gravity at different speeds. Galileo took small and large cannonballs to the top of the Tower of Pisa to check it out. Sorry, Aristotle.
Now, when it comes to how we run our firms—how we compensate people, the contribution of “non-lawyers,” the types of career paths our firms can and should house, indeed how we recruit talent to begin with (our lifeblood, need I point out?), we could take a lesson from Artemus Ward, the pseudonym of Charles Brown [1834—1867], a popular humorist andcontemporary of Mark Twain:
“It ain’t so much the things we don’t know that get us into trouble. It’s the things we do know that just ain’t so.”
Modern science has of course progressed so incomprehensibly far from Galileo’s day that now only a very few of us can understand its particular theories in all their subtlety and power. But all of us can understand its peculiar approach.
If you have a theory about how the world works (“our firm exclusively hires the top law students from the top law schools—no other information about these individuals really matters,” “non-lawyers can’t [really, truly] do anything as well as lawyers”), you might give it a brief go and actually try a small experiment or two. (a) What have you got to lose?; and (b) Who knows what you might find out?
Gopnik ends:
There’s supposed to be a sign up on the Tower Of Pisa: “Please don’t throw things from this tower”. That sign is the best memorial that Galileo could ever have.
Of course, I’m not sure that it’s actually there. I’ll have to go and look.
The key thing about experiments is that good ones almost always take thought and effort. Firstly, we have to establish our hypothesis in a form that is amenable to experimentation. If we think of our school-boy version of Galileo at Pisa, this may seem simple, but if even that were so simple, why did it take so long to happen in a form that could be analyzed? Then we have to undertake a step that is much harder than it seems: we need to determine what will count as an answer to the question. Then we need to work out how we will measure the outcomes, including whether we think this is a strictly deterministic system, or (surely, more likely) whtether the experiment will be present a range of outcomes. This has deep implications for experimental design, such as the need for replication and selecting methods for analyzing results. Finally, we assemble the plan for the experiment. And almost always we find that the first experimental approach did not go as well as we would really have liked, so we probably will need to tinker with our design a bit.
Does that mean it is all too hard? Not at all. As many people have pointed out, the greatest improvement in understanding comes with the first increment of new information. Provided of course that the information is well posed with respect to the decision you need to make. In physical science and engineering, we believe strongly in experimentation as a basis for generating knowledge. We also believe, as my Dad had it: measure twice and cut once when it comes to designing and executing experiments.
Mark