Underdetermination in Science
Underdetermination is the idea that evidence and data alone aren't enough to justify scientific conclusions. That is: The data doesn't determine the conclusion.
Gasp! Perhaps the solipsists were right since there is no way that evidence can prove that they were wrong.
But one needn't be surprised.
I understand science to be an intellectual tradition that learns about reality by observation and reason. Science gathers facts, thinks a bit, does experiments to gather more facts, thinks a bit and then eventually comes to a conclusion. The 'thinking' part of that process pulls in lots of knowledge from experience and culture that is not apparent from the data alone.
One of the ways we define things is by pointing - Say I'm out having a stroll with a person who speaks a language I don't know and who speaks no English. A rabbit hops by. My companion says "Gavagi" . How should I interpret that? Does gavagi mean rabbit or food or cute or . . . ?
Quine used that example to show that we can never be sure if the meaning we get from a translation is the same as the meaning taken by a native speaker of the translated speech.
Quine also talked about how there can be many explanations for any set of facts and the facts alone can't tell us which explanation is true.
One characteristic of culture is that groups of people can share the same explanation for a set of facts. But different groups can have very different explanations for the facts. Historically and at present that fact is causing a lot of strife and misery (but that's just my explanation).
I wonder if the underlying assumption of underdetermination is true - that evidence alone is not enough to justify scientific conclusions. I'm like: Wait a minute, Not so fast.
For one thing, science isn't just a catalog of facts. It's a structure of knowledge into which facts fit. EO Wilson called that structure 'consilience'. It's the idea that knowledge gained from biology is complementary with knowledge gained from physics. That structure would be impossible without all the facts connected within it.
For instance it's a fact that many people have experienced ghosts but the idea of immaterial beings that can float through walls and scare people doesn't fit with science's ideas like the conservation of matter and energy so it can't find a place in the structure of knowledge we call science. And heck - perhaps one day we'll learn where ghosts fit in the structure of consilience.
Science has data and then metadata - that is data about the data. The meta can be taken out to many levels. I can interact with the meta-thing I call a coffee mug without having to know all the details down to a sub-atomic level.
But also, as a metathinking thing I believe that given time and study I could learn about those levels to astonishing detail.
An important meta-fact about science is that it works. Airplanes don't fly by magic. Most places use modern medicine. And acupuncture is becoming mainstream.
I must add that just as not every explanation fits into science that science can't inform all our thinking. I asked an AI what science says about beauty and I got a long explanation about how our sense of beauty evolved. None of that could explain why one young woman in a crowded room would grab my attention so strongly.
Evolution by Natural Selection has a lot in common with the stringent selection process of science. Natural Selection does not intend. Science is an artifact of people interacting with each other and people certainly do intend. But science is an aggregate of the interaction of many people - it isn't something that can intend.
What do you think?
I present regular philosophy discussions in a virtual reality called Second Life.
I set a topic and people come as avatars and sit around a virtual table to discuss it.
Each week I write a short essay to set the topic.
I show a selection of them here.