Comment by barrkel | Hacker News Reader

Comment by barrkel | Hacker News Reader

Parent Full thread

There's a real risk of storytelling with 'why'. You can construct a narrative that explains anything - this is probably the primary thing our consciousness does when communicating to other people - but there's no necessary relationship between the story and the world, and humans are great at believing the stories they tell themselves.

Interesting events outside of abstract, simplified, theoretical scenarios normally have multiple probabilistic causes, and while the causes could be analyzed at multiple levels - atomic, mechanical, emotional, symbolic - only the atomic is real. The others are complex webs of stories we tell ourselves, stories using symbols that hide details because we cannot understand the world as it is, we can only cope with it in a simplified form. Only as a story.

Our choice of symbols and abstraction level to tell our story creates artificial entities and forces to explain things. But the boundaries of these entities are a fiction. The world has no ontology; it's a giant bucket of interacting particles. The problem with getting serious with this kind of a principle is that it will break down. When you try and zoom in on the referents of the symbols, you'll find that they're a lot blurrier than your stories told you.

The different levels emerge from each other. And I don't think that we can be sure to have found the most basic one, or even to assume that there is a most basic level. But the emergent levels are not less 'real' than the more basic ones. There is more about emergent levels than just to hide details.

Different abstraction levels are normally different ways of talking about the same physical phenomena. That is, they are parallel; they don't stack. The most basic one is the one that doesn't exist in our heads and doesn't have a model with predictions to violate.

What does "real" mean here? I'm using it as a proxy for truth, or perhaps better, accuracy in a model. That is, if a model of the world makes predictions about the world, the model is "real" to the degree that the predictions are correct. But all human models of the world have their limits, usually forced by our tiny little brains and meagre data input senses.

We take something that we've decided to categorize, and create symbols to represent commonalities, and theorize relations between the symbols, and test these theories, and gradually refine a calculus of symbols that we can use to apply to the world, and the system works, in so far as the commonalities stay common, and the symbols aim squarely at the average of the referents. But it's not reality; it's all made up, like mathematics. It's just a story, it's all in our heads and in narratives we communicate with one another. The symbols don't stick; the world moves on. People change; humans are animals that evolve over time, and continue to evolve. We can't even decide what a human is - the evergreen abortion debate - and other fictions like races, species and planets etc. shift as we change our mind.

So when you want to get serious about causes, you're going to be using a system. You're going to be using symbols and relations between symbols; you're going to define your symbols in terms of commonalities between events and matter that appear to repeat reliably. But it only works because we have so little thinking matter, so little data, so little input; it's the only way we can squeeze the enormity of the world into our tiny heads.

I think this is a consequence of the fundamental problem with a materialist rationalist view of reality; namely, that when you consider reality to be only physical matter which can broken down into constituent matter and exists as a result of cause and effect, all meaning is stripped from the universe and you're forced to accept the first cause argument.

The real problem comes down to what we mean by "truth". Truth can't simply be the accuracy of our models of the material world, because the material world can't tell us anything about how we should act.

Atoms are an abstraction, just like the higher level abstractions you mention. They happen to be a very good/stable abstraction over the time/energy scales relevant to day-to-day life, but they too are made of pieces that can disintegrate, recombine, etc.

It's a mistake to assume that anything that can be analyzed in terms of smaller pieces is less real than the pieces. Neither is real, or both are.

Sure; I'm using atoms to refer to physical reality rather than abstractions people are applying to the top; atoms more in the indivisible sense, than what you think of in the physics lexicon.

I think your dichotomy is false. Human simplifications of the world are "real" in so far as they correspond with what they model (or else we'll get into a debate about what "real" means). The higher up in the abstraction stack you go, the bigger things (generally) you can model, but the details vary more. This may be more "real", or less "real", depending on what you're trying to model and how well the conceptual framework fits; e.g. you won't get far with modelling fluid as Newtonian marbles. The point being that it's not a binary proposition, the models are all approximations, and any "cause" you find is only with respect to a story people have made up about the way the world works.

It all depends on your ability to model things. In the case where that ability is perfect (which may not exist,) all the "macroscopic" elements of reality you recognize are as true as their "microscopic" elements. That is, if your model of a fluid is perfectly correct in respect to individual atoms, then proving something about atoms automatically proves it to a fluid. If you prove that atoms exist then fluids exist (if your model is perfect.) This problem is already tackled by science. In physics, models are known to be limited and target only a simplification of reality, and they become greatly wrong in special cases and thus unacceptable. For example, modelling a ball thrown in the air neglecting air resistance may be "correct" enough when it is a lead ball, but will become apparantly unacceptably "incorrect" when applied to a ping pong ball.

But even "physical reality" is hard to pin down. Plenty of evidence from QM suggests that we should not think of (any) physical properties as objectively real. Trying to pin reasons on them seems troublesome as well.

This exactly.

Certain shapes and patterns of particles become larger and larger semantic entities. Particle patterns that repeat or recur become self-referential and correlated. With systems that are complex, chaotic and highly sensitive to initial conditions -- there is no "reason" or the "reason" can be everything -- there is no separability of part from whole.

My wariness with respect to the PSR are these two features:

1) The PSR is typically defended empirically. That is, the reason for believing in the PSR is usually some form of "we observe that things happen for reasons."

2) No process is ever provided for determining whether or not a particular reason is actually sufficient.

Taken together, 2) says that we have no justification for believing that any of the reasons we observe in 1) are actually evidence for the universality of sufficient reasons posited in the PSR.

Science is as much a study of limitations of the human mind as it is of anything else.

Reply on news.ycombinator.com