Jason Crawford

He thought, aha, what if a similar compound selectively kills the bacteria?

He had this theory, which is correct, that these molecules were fitting into each other in a very precise way, like a key and a lock.

And so some of the early drug experiments were based on modifying dyes.

And I believe that's how they found some of the early antibiotics like the 1930s.

So why do we need an epistemic base?

Well, a key reason is that the search space is just absolutely enormous.

Driving this home, Richard Nelson wrote a classic paper in the 1950s when he was at the Rand Corporation called The Simple Economics of Basic Scientific Research.

And he gave these compelling examples.

One was, you know, what if you got some scientists at the turn of the century, the last century, to examine the, say, come up with a better way to examine the inner organs of the body?

They would never have come up with x-ray analysis if they hadn't already identified that, you know, x-rays existed and what they could do.

Similarly, if a bunch of people in the mid 19th century had tried to come up with a better method of long range communication, just starting from that goal without the science, they would never have figured out Maxwell's equations or developed radio.

The other reason, so the search space is enormous, but the other thing is that the devil is in the details.

Tiny, seemingly insignificant changes can make the difference between success and failure.

And so not only is the search space huge, but you can be right next to the right thing and miss it completely.

You know, what if you tuned in just the wrong frequency or what if, you know, you boiled the water for just you needed to boil the water for just like one minute longer, you know, and you didn't know that you can fail completely.

So you don't always know when you're close.

And so this is what we need, you know, theory to guide us for.

So science, to wrap up this part, forms the epistemic base for invention in at least three different ways.

It comes in before, during, and after the invention process.

Before, it identifies phenomena that can be exploited, such as the vacuum or germs or semiconductors.