Back in the early days, before Trump Derangement Syndrome set in, I did a post entitled:
Are the laws of physics mere social conventions? No, they are social conventions.
The basic idea was that theories are beliefs, but not “merely” beliefs. There is nothing more important in the entire universe that mental states such as beliefs. Now, after a delay of 7 years, David Deutsch has left the following comment:
The astronomy example puzzles me; how closely our model reflects objective reality is somewhat orthogonal to whether “objective reality” exists in the first place. Elsewhere, the author asks if a model that is right 99.8% of the time is “false” while one that is right 99.99% of the time is “true” (or something to that effect). Truth is quantitative/probabilistic, not binary; some models are more “true” than others, but their level of “truthiness” is still an objective fact.
If this well known piece by Asimov has not been referenced yet, I’d be surprised, but here he sums it up quite well:
“When people thought the earth was flat, they were wrong. When people thought the earth was spherical, they were wrong. But if you think that thinking the earth is spherical is just as wrong as thinking the earth is flat, then your view is wronger than both of them put together.”
http://chem.tufts.edu/answersinscience/relativityofwrong.htm
I have no idea if it is the David Deutsch, one of the smartest writer that I have ever read, but in the off chance that I have been honored with a comment from him, then I really need to offer some sort of response. First, a bit of context. I was defending Richard Rorty’s views on truth (sometime summarized as “truth is what my peers let me get away with”). Or perhaps one might say, true things are things that are regarded as true. Not surprisingly, this drives lots of people crazy, because we all see things that others regard as true, which clearly don’t seem true, and there are things that we see as being obviously true. Is this Rortian stuff too much relativism?
I don’t see it as relativism at all. I don’t see it as the world of fuzzy post-modern philosophers attacking the virtuous hard sciences. It’s important not to get confused by semantics, and focus on what’s really at stake. In my view, Rorty’s views are most easily seen by considering his denial of the distinction between objective truth and subjective belief. In order to see why he did this, consider Rorty’s claim that, “That which has no practical implications, has no theoretical implications.” Suppose Rorty’s right, and it’s all just belief that we hold with more or less confidence. What then? In contrast, suppose the distinction between subjective belief and objective fact is true. What then? What are the practical implications of each philosophical view? I believe the most useful way of thinking about this is to view all beliefs as subjective, albeit held with more or less confidence.
Let’s suppose it were true that we could divide up statements about the world into two categories, subjective beliefs and objective facts. Now let’s write down all our statements about the world onto slips of paper. Every single one of them, there must be trillions (even if we ignore the field of math, where an infinite number of statements could be constructed.) Now let’s divide these statements up into two big piles, one set is subjective beliefs, and the other pile contains statements that are objective facts. We build a vast Borgesian library, and put all the subjective beliefs (i.e. Trump is an idiot) into one wing, and all the objective facts (Paris is the capital of France) into the other wing.
Now here’s the question for pragmatists like Rorty and me. Is this a useful distinction to make? If it is useful, how is it useful? Here’s the only useful thing I can imagine resulting from this distinction. If we have a category of objective facts, then we can save time by not questioning these facts as new information arises. They are “off limits”. Since they are objective facts, they can never be refuted. If they could be refuted, then they’d be subjective beliefs, not objective facts.
But I don’t want to do that. I don’t want to consider any beliefs to be completely off limits—not at all open to refutation. That reminds me too much of fundamentalist religion. On the other hand, I do want to distinguish between different kinds of beliefs, in a way that I think is more pragmatic than the subjective/objective distinction. Rather I’d like to assign probability values to each belief, which represent my confidence as to whether or not the belief is true. Then I’d like to devote more of my time to entertaining critiques of highly questionable hypotheses, than I do to less plausible hypotheses.
Thus if someone tells me that I really need to read a book showing how 9/11 was a CIA plot, my response is, “No, it’s not worth my time.” It’s possible that it was a CIA plot, but so unlikely I don’t want to waste limited time trying to refute the view that Al Qaeda launched the attack. It’s not that I believe Al Qaeda’s culpability is an objective fact; rather my subjective belief that it was Al Qaeda is so strong that I don’t want to waste time on it. Ditto for my view that 1+1 = 2. On the other hand, at some later date new information on 9/11 may arise and reach the headlines of the New York Times, where I see it. Now I may want to read that book. Similarly, I can imagine a physicist not wanting to read some idiot’s crackpot anti-Newtonian model in 1850, but finding anti-Newtonian models quite plausible after the work of Einstein.
The subjective/objective distinction would only be useful if it put some ideas off limits, not open to questioning. There are certainly some ideas where it’s a waste of time to question them, but I don’t like this as a general category, because I don’t know where the boundary lies between claims that should be beyond questioning, and claims that should be open to question. So it’s simply more pragmatic to regard all statements as being beliefs about the world that are open to question, and then assign probability estimates (guesstimates?) to the chance that these claims will be overturned.
The other point of confusion I see is people conflating “the map and the territory”. Then they want to view “objective facts” as aspects of the territory, the underlying reality, not (just) beliefs about the territory. I don’t think that’s very useful, as it seems to me that statements about the world are always models of the world, not the world itself. Again, if it were not true, then theories could never be revised over time. After all, Einstein didn’t revise reality in 1905; he revised our understanding of reality–our model of reality.
Reagan said “Trust, but verify”. That means it’s OK to believe that certain things are true, but always be open to evidence that these things are not true.
PS. Recall this statement I made above:
Rather I’d like to assign probability values to each belief, which represent my confidence as to whether or not the belief is true.
Rorty was criticized when people pointed out that one often hears something like the following: “Although most people believe X, I believe that Y is actually true.” If there is no objective standard to determine whether X is true, then what can this statement possibly mean? I seem to recall that Rorty said something to the effect that when people claim Y is actually true, despite most people believing X, they are actually predicting that in the future Y will eventually be regarded as true. Or maybe it’s a claim that, “If other people had seen what I saw, then they would also believe Y is true.”
PPS. Back in 2013 I mentioned Deutsch in a post:
David Deutsch likes to sum up his philosophy as:
1. Problems are inevitable.
2. Problems are solvable.
The horrible nationalism sweeping the world was inevitable, and it’s solvable. Good times will return.
PPPS. I had a discussion of Deutsch’s views on quantum mechanics, and well as Eliezer Yudkowsky’s views, back in this 2013 post.