What is a "better explanation"
David Deutsch argues that the project of rational inquiry (including, but not limited to, science) is finding “better explanations.”
An explanation is a form of knowledge.
Knowledge is a form of information.
Information is something with the following properties:
It requires a physical substrate. There is no way to have information without embedding it in a physical medium.
It is substrate independent. The same information can in any medium. It can be written on paper, encoded in magnetic domains, transformed into sound waves, resident in a human brain.
It is not observable. There’s no way to see information directly. You can see the medium or substrate. You can extract the information from the attributes of the substrate. But you cannot see the information itself.
Yet, information exists.
Knowledge is a particular kind of information: it’s information that an environment tends to preserve.
If you record bits of information on a bunch of pieces of paper, the ones that contain knowledge are the ones you’ll tend to preserve–or move to another medium so they’ll be preserved.
DNA contains information–and knowledge since it tends to be preserved.
Complex physical objects–like the computer in front of me are embodiments of information, and because they contain knowledge, they tend to be preserved.
And I am adding information (and knowledge, I hope) though this computer.
Knowledge can exist without a knower.
If I write something that contains knowledge, it remains knowledge whether I am alive or dead, whether someone remembers it or not.
As long as the environment tends to preserve it, it qualifies.
Kinds of knowledge
There are two kinds of knowledge.
Ordinary knowledge has use and causes itself to persist.
Explanatory knowledge can be used to answer questions: how, why, in what way?
The DNA of each living organism contains knowledge vital to the survival and reproduction of that organism’s ancestors, and to that organism itself.
But it is not explanatory knowledge.
Until recently, there was no knower of that knowledge.
A bacterium’s DNA contains knowledge sufficient to synthesize the proteins needed to create a copy of that bacterium.
But the bacterium does not know any of it, nor can it explain it.
Where does explanatory knowledge come from?
Deutsch says that explanations are created by people–a term that includes humans, intelligent creatures on other planets (if they exist), and possibly, someday, computers.
Non-human animals carry and use knowledge; some even create new knowledge; a few can pass their knowledge to others of their kind by imitation.
But none, as far as we can see, can create knowledge of the explanatory kind, or pass knowledge by explanation.
Right now, as far as we know, the creation of explanations is uniquely human, and the laws governing explanations are a lot like those that govern biological evolution.
Biological creatures adapt by embodying knowledge about their environment: what’s good to eat, what material to use for a nest, how to do a mating dance.
Most of that knowledge is embedded in the creature’s DNA. Some knowledge is discovered by an individual using sensory apparatus and rules of inference embedded in their DNA.
Some is conveyed by imitating others of their kind–because of DNA tells those with knowledge how to exhibit it, and those who need the knowledge how to imitate.
Our DNA contains knowledge about the structure of human languages (cf “The Language Instinct” by Steven Pinker) and knowledge of how to acquire language without instruction–if you happen to be a child.
Our DNA also contains knowledge about how to evaluate other knowledge. That knowledge has gotten us to this stage of existence, but some of us have discovered that it contains systematic errors. Some of us work hard to correct those errors and remove the knowledge that we acquired before we realized the errors. Others of us don’t give a shit.
I’ve come to realize that my goal is not “finding the truth,” but rather “finding better explanations.” The difference is subtle but important.
If something has an existing explanation, I can determine whether a new explanation is better or worse. If it explains everything that the old explanation explained, and explains more, or is more general, or simpler, then it’s comparable and better. If it explains only some of what the old explanation explained and nothing else, then it’s comparable and worse. Einsteinian Relativity, for example, is better than the Newtonian motion and gravitation. Relativity explains everything that Newton’s laws do and more.
But often, two explanations may not be strictly comparable. One may explain some things and leave others completely unexplained. Another may explain that which is unexplained by the first, but fail to explain everything that the first explains. If both are “good” explanations, then neither may be described as strictly better than the other. So Relativity explains one set of phenomena and Quantum Physics another non-overlapping set. Both are good explanations, and both explanations are continually being improved. At some point, we may have a theory of quantum gravity that is better than both. But we are not there yet.
Click here to subscribe to 70 Years Old. WTF! by Email