John Horgan recently wrote a column which has received a lot of attention. Horgan’s thesis is that when it comes to three topics: the existence of God, the mind-body problem, and the measurement problem in quantum mechanics, science can’t provide the answers and may never be able to. Horgan advises that the only responsible position for these questions is agnosticism, an acknowledgement that knowledge in these areas may be impossible.
Those of you who’ve known me for a while can probably guess my reaction. To be fair, I do think Horgan makes some important points about the current limitations of science. But arguably two assumptions he makes in his piece are wrong.
The first is that the current limitations on knowledge is a static condition and will never change. The history of science isn’t kind to people who said we’d never know the answer to a particular question, at least to any precise concrete one. Ideas that are untestable today may well be testable tomorrow.
But the second is what I want to address in this post, the idea that there’s some sharp boundary between belief and knowledge, that a belief requires absolute certitude to be considered knowledge. Since we never have absolute certitude about anything, I don’t find that to be a productive standard. It effectively renders words like “knowledge” or “know” unusable. I think a more productive outlook is that knowledge is always probabilistic. In this view, knowledge is a belief with which we are justified in feeling a higher degree of certitude, with the idea that we never reach 100%.
So simply saying we’re agnostic on a certain question is often an ambiguous statement. It leaves a lot unsaid about what credence we do have for the possible answers. Of course, for these questions, it’s not possible to give a precise probability, like 78.9%. But a book I read a few years ago, How to Measure Anything, pointed out that often when a precise statement isn’t possible, a broad approximate one might be, and that can still be useful for many purposes.
With similar reasoning, Richard Dawkins, in his book, The God Delusion, in trying to add some clarity to the question of agnosticism about God, came up with a seven point spectrum of probability to indicate where someone might be in terms of certitude about their conclusion. I think that scale has a lot going for it, but sticking with percentages, even imprecise ones, seems to give a better idea of the magnitude of credence we put in something.
In terms of God, it’s worth pointing out how culturally specific this question is. Very few people today are agnostic about the existence of the Sumerian, Norse, or Aztec gods. Of course, we can come up with increasingly generic versions of “God” that are progressively less culturally specific. Or we can use the word “God” to refer to the universe or the laws of nature.
However you feel about these moves, my credence for any culturally specific version is south of 1%. The other versions get increasingly higher until we get to a deistic intelligence that created the universe but doesn’t interfere with it, where I’m probably around 30%. When it comes to referring to the universe or the laws of nature, it becomes equal to my credence in the external world, well north of 90%.
I long ago lost interest in the agnostic vs atheist debates, so I’ll leave it to others to figure out which label is appropriate here.
When it comes to consciousness, my credence for reductionist physicalist approaches is north of 90%. Which doesn’t leave much room for the competing views: non-reductive physicalism, dualism, panpsychism, idealism, etc.
In terms of any specific physicalist theory, that credence is much lower. I think they’re all wrong, but some seem less wrong than others. My deep suspicion is the answer will involve numerous theories, involving a mix of biology, neuroscience, and computation. But these distinctions are tractable scientific questions involving no deep metaphysical mystery.
And then there are interpretations of quantum mechanics. I’m going to set aside anti-real approaches (weak Copenhagen, QBism, quantum information, etc), since they’re not really making a statement about reality. (Except possibly an assertion that an anti-real explanation is all we’ll ever get, in which case see the first point above.)
My credence in straight quantum theory with no physical collapse (aka Everett many-worlds) is around 50%. That leaves a 50% credence that new discoveries significantly change the picture. However, my credence is pretty low (10% at best) for any specific realist interpretation with additional assumptions. Until we have experimental results pushing in a particular direction, they all seem like speculative guessing. Some, like objective collapse and pilot-wave theories, seem too much about saving appearances.
In summary, for the most common conceptions of God and mind, I think Horgan overstates the mystery. There are straightforward answers if we’re willing to accept them. That could also be true for quantum mechanics, but the uncertainty there seems higher. For now.
What do you think? On any of the three topics that you’re willing to share, what do your numbers look like? Do rough credence percentages work for you? If not, is there another framework that’s better?