[It might seem completely opportunistic, but this is a repost of yesterday’s post. I wrote it hurriedly as i was about to go out, and today by morning, as i woke up, i realized the ordering of my arguments could be lots improved. The original form might probably be gotten from feed-readers or some of those “web-memory” sites. Anyway, this is a commentary on “A Simple Truth“, a text by a certain Eliezer Yudkowsky from Overcoming Bias blog.]
To oversimplify, my argument is somewhat like this — Eliezer tells this tale:
Someone says to you: “My miracle snake oil can rid you of lung cancer in just three weeks.” You reply: “Didn’t a clinical study show this claim to be untrue?” The one returns: “This notion of ‘truth’ is quite naive; what do you mean by ‘true’?”
And i say: i can’t really conceive anyone that, spoken like this, would answer anything except: “My notion of ‘truth’ might be naive but you are still not getting my money for your whatever-it-is-oil!”
If you didn’t read the Simple Truth article, let me present an abstract. It is all a farcical fable. A shepherd “invents” a system of using pebbles on a bucket to count his sheep to help management, and his method gets gossiped as a “magic bucket”. Then a sophisticate snob called Mark appears to see what’s the magic. They begin arguing, and Mark ends up saying that the bucket can’t work (despite seeing it working), because everything is subjective. After that, he transforms into a litcrit caricature, dropping all sort of ridiculous arguments. Finally, a guy named (not so unbiasedly) Darwin comes, proposes that if Mark sustains that everything is belief then he should really believe to be able to fly and jump off of a cliff.
Eliezer (and he’s definitely not the first to try to sell me this one) is arguing that despite being intellectually indefensible, we can still “use” the idea of truth “naively” without problems. Even without complicated, elaborate, detailed, complete “definitions of truth”, the basic idea is still useful. What is true is, and that’s that.
Unhappily, every time in his fable that a credible argument against — well, against whatever it is that he defends, i’ll call it realism here for short, and the opposition relativism — anytime a valuable argument against realism is posed, by Mark, it is posed as a mere caricature of the original arguments, completely out of their contexts and over-expanded. It’s just eristik‘s first stratagem.
Mark is just a caricature. When, for example, he says:
Look, you’re taking the wrong attitude by treating my statements as hypotheses, and carefully deriving their consequences. You need to think of them as fully general excuses, which I apply when anyone says something I don’t like. It’s not so much a model of how the universe works, as a “Get Out of Jail Free” card.
he is not exactly saying what relativists actually defend, he is saying what realists prejudicedly think they should defend. Eliezer is saying: to defend that there is no truth should lead you to defend such stupidity. But why so?
Imagine that the whole universe was a simple dualistic ensemble of one brain and one TV-screen. The TV would be the matterless subjective non-true phenomena of the brain-subject. Nothing that the subject sees or feels is “real” or “true”, it is just something in a screen. But even then, why should this subject be unable to observe correlations between different stimuli on the screen? For example, if every time a saber-tooth tiger appears on the screen he also feels a bite on the feet. Or every time he sees a green mushroom he feels better. Why couldn’t he learn to run from the tiger and run towards the mushroom?
In our case, why should Mark not fear jumping from the cliff even if he believed that the cliff was a belief? If everything in the universe is belief, that should definitely mean that beliefs can hurt!
When you put aside all the flowering language and so called “unbiased” methodology of the realists, what you find is that they are giving — for free, mind you — giving to truth a whole lot of properties that they simply cannot prove truth to have — or for that matter disprove, or even discuss their motives for so believing.
The obvious example is effectiveness. Case in point, the shepherd’s truth is more powerful than Mark’s relativism, so Mark dies. Mark’s is (supposedly) forced to jump from a cliff because his beliefs “do not work”.
Nevertheless, for example, Aristotelian physics did work on his time, and in day-to-day life it actually works much better than Newtonian mechanics! Almost nothing we experience directly obeys, for example, inertia, and i certainly do not feel other people’s gravity.
I know that the discrepancies were explained — but i also know that in Galileo’s time this very same “simple truth” argument was used against the now established “scientific” idea of heliocentrism. (For anyone interested in this debate, i heartily recommend Feyerabend’s Against Method. Really. It’s simply wonderful.)
The development of our opinions and understandings happens through a delicate process, in which we compare our experiences and seek new meaningful ones, but in no part of this process we do need truth.
For example, the Shepherd did not first thought “hey, there is truth!” and after proceeded to devising his pebble-and-bucket counting system. He invented the system, and he could see how it worked, and the idea of “truth” only passed through his head when questioned by Mark.
To suppose that he had a “naive idea of truth” that enabled him to do mathematics is completely stupid (or, if you prefer, Occam-vulnerable). The shepherd didn’t think about truth. First you invent maths and then you question whether or not it is truth. How could it be the other way around?
Again, realists are making donations to truth: that everything that is obvious has to be truth. But the sun obviously revolves around the earth, which is obviously not moving. Nevertheless, those claims are simply mistaken.
I can’t figure out where does it come from, this conviction that a complicated notion of truth exists in exclusion of the naive one. I do not think that anyone who ever exposed the idea that “there is no truth” would really object to (as an example) saying that “today is truly Friday” or that “5 is the true result of the sum of 2 and 3”. No one is arguing against the word, or against simple, basic, commonplace statements or notions. Nor am i arguing that we can’t ever accept anything for obvious.
When i say there is no truth i do not imply that we should stop using mathematics. I do not propose acting as every one of our experiences was only an transient hallucination. Why should we? What is the connection?
If some business can be solved by simple counting, use counting. But there are quite some issues in my own life that are way beyond counting. Interpersonal relationships can’t be dealt with in a purely “factual” basis. Politics can’t be decided (let alone understood) on purely objective considerations. Hell, i am trying to rent an apartment and even this can’t be decided by the simple figure in my bank account — they need all kinds of proof that i intend to pay, even if i have the money, and this is subjective and therefore basically undecidable.
Realists accuse relativists of bad academicism — of not being honest about their intellectual endeavors. Specifically, they accuse the “there is no truth” slogan of being an excuse for not accepting theories that do not fit with the relativist’s own convictions.
And it is beyond obvious that relativism CAN be used this way. But so can realism. Religion, for one, has been using this truth business for 2 thousand years now — “I Am the Way, the Truth, and the Life” (John 14:6). Even Umberto Eco had to admit that Derrida is much better than his followers.
Now serious relativists only like different idea-evaluation procedures. They think that truth-evaluation is real cool, but there are other evaluations that are even cooler. We can, and indeed we do, use lots more. If we didn’t we would be reduced to simplistic problems.
Sometimes it is enough to check whether or not something is “true”. For example, when we need to know if there are pebbles in the bucket. This is so simple that we can just go there and look. The situation is so simple, in fact, that a crude binary taxonomy of ideas [TRUE|FALSE] actually makes us work faster. In other situations, we need “otherness revealing methods”. It depends.
As the bottom line, i guess if you can’t really point with your finger at a discrepancy, then the argument in fact involves so much interpretation that the simple recourse to “naive truth” is not guarantee enough. If you can’t rephrase your “request for truth” in terms so obvious that the relativist cannot evade, than the subject at hand is more complex than simple truth-evaluations can handle. I do not mean the relativist is right, mind you, i only mean the problem is hairy. Let me give an example.
In an architecture master’s thesis presentation the graduating student had as a central argument that the street vendors would distribute themselves spontaneously. An anthropologist in the examination committee said this was not the case, that they usually did distribute according to family patterns. Nevertheless, the student not only got his degree, he had high marks.
This seems a BAD thing™, doesn’t it? (And the example was given by a realist). But do the family structures really concern architecture? It might even be that the student would benefit from knowing it, but on the other hand maybe his designs are just good, despite his “research” being “untrue” and therefore completely invalid. What i mean is, a big enough ensemble of street sellers is so complicated that any account of their distribution will be full of interpretations, ones that we can’t simply go and check.
It is like the “magic bucket ritual” involved trowing a pebble for each “beautiful” sheep — that simply wouldn’t work. Who is to say that the one-eyed brown sheep is not beautiful? But on the other hand, there are times we simply can’t avoid interpretation. We simply can’t expect (for example) that the beauty of the candidates will not have an effect in their campaign. The world is messy.
I think anyone defending a non-naive idea of truth really thinks the naive one is, well, naive, but OK. I do not feel like tricking naive-realists into jumping off of a cliff.
Now there is another idea of truth. One that asks: what makes you so certain? And asking this question is productive in many situations, as i doubt Eliezer would dispute. Why can’t both ideas co-exist? I really do not know.
To assume that the “naive notion of truth” is a good enough basis for all of our world-understanding is a bet. So is believing that there is no truth. But, somehow, i do really think the realists are not on the safe side.