I might add that I: (i) am not a determinist (ii) believe in libertarian free will (iii) deny the existence of intrinsic (non-statistical) randomness (iv) deny that evolution or abiogenesis can be adequately simulated on a machine designed to be deterministic. (v) claim that property dualism of some sort (including what might vaguely be described as micro-teleological or pan-experientialist) is necessary to make an adequate account of evolution (vi) think that the idea that human civilization can arise in a strictly deterministic or stochasto-deterministic fashion from a random ball of gas is absurd, notwithstanding that to an outside observer that may superficially appear to be precisely what happened.

I guess I’m just really confused about what relevance the whole reductionism thing has to the conversation. I’ve traced it back to your comment in #90 directed at Rob, but I can’t for the life of me figure out where you are coming from and where you are going with it.

I took you to be using Dennett as a kind of poster boy for bad evolution: Evolution entails reductionism, and reductionism entails a meaningless existence.

Not let’s get a few questions in the open:

1. How do you feel about evolution?
2. How do you feel about reductionism?
3. Do you think evolution entails reductionism?
4. Do you think reductionism entails nihilism?

First, why require random initial configurations?

Because it is conventional in science not to assume the Platonic existence of that which you are trying to prove. That means starting with anything other than a ball of gas is an unjustifiable assumption. After all, we could just assume the existence of human civilization and be done with it, right?

Second, if the supercomputers are capable of accurately simulating the analogue of whole DNA strings and organisms

They are not. In fact at present they are incapable of accurately simulating the operation of a single cell, even with football fields full of computers. How many years will have to pass before a computer can simulate the advent and propagation of the first cell from some proteanaceous ooze? Personally, I would be impressed with any physical theory of abiogenesis that did not rely on a cosmic accident.

Explaining consciousness, for example, in terms of evolution is not reductionism unless one assumes that consciousness is a higher-level phenomenon than evolution

This whole thing about “explaining” is a bit of a distraction, in fact it is one of the weaker claims that reductionism entails. Reductionism from consciousness to evolution entails that knowing everything about the evolution of a system entails knowing everything about the consciousness of a system. A difference of level is not required. If consciousness and evolution are at the same level, it merely implies that the converse is also true, that knowing everything about the consciousness of a system entails knowing everything about the evolution of a system.

Most people find the latter proposition dubious, hence the assertion that reductionism generally only works in the direction of consciousness to evolution, to the extent that they maintain that consciousness reduces to evolution at all.

On the contrary, what Godel demonstrated was that it was impossible for a logical system to be used to construct a proof of its own self consistency. I am making a higher level argument, that the uncontroversial proposition that the world is real implies (by definition) that it is not inconsistent with itself, therefore any mathematical relationship that is consistent with the real world is consistent. The only way out is to claim that the world isn’t real, i.e. that there is no such thing as objectivity.

Mark, really no — scientists tend to understand their mathematical models as models, not reality.

I could hardly agree more (I majored in physics, by the way). However, there is a difference between a mathematical model of the natural world, fully limited by imprecision and epistemological considerations and the mathematics (and natural laws) that actually obtain. In other words, you are confusing epistemology and metaphysics.

The existence of a single, time-independent natural law is adequate to establish the instantiation of some form of mathematics in the real world. It doesn’t matter if we know what that law is, or in what form it takes. All that matters is that there is at least one natural law. The reality of the number of apples in my basket is a trivial example, one that directly corresponds to the law of non-contradiction.

You imply that the law of non-contradiction with regard to the properties and number of real world entities is a figment of the imagination, as if a tree fell in a forest and didn’t make a sound. If the tree inevitably makes some kind of sound, some kind of natural law applies, completely independent of whether we know about it, what we believe about it, or anything else.

Either trees that fall in a forest reliably make some sort of sound, or there are no natural laws. Real natural laws, ones completely independent of what anyone thinks or knows about them. That is what “real” means. You claim that mathematics isn’t real in any and all conceivable forms. I beg to differ, and my position is hardly an uncommon one, even among those so commonly divorced from natural reality as mathematicians themselves.

Kurt Godel was a mathematical realist, by the way, i.e. one who holds that mathematical entities (like numbers) exist independently of what we think or know about them. Paul Erdos is another.

Some problems. First, why require random initial configurations? That doesn’t seem to be part of evolutionary theory. Second, if the supercomputers are capable of accurately simulating the analogue of whole DNA strings and organisms, then there should be no problem. Just allow the simulated equivalent of a couple billion years to pass and all should be well.

But you seem not to believe that — which is because you don’t believe consciousness evolved. So your comment doesn’t really get you anywhere; your skepticism is grounded in begging the question.

I hereby declare Mark D. a troll and request that he not be fed.

Explaining consciousness, for example, in terms of evolution is not reductionism unless one assumes that consciousness is a higher-level phenomenon than evolution. That assumption, among others, seems uncertain. Mark, are you prepared to defend it? If not, your position is hard to decipher.

There is absolutely nothing about reductionism that implies that the structure (or information content) of the relationship between higher and lower level properties is somehow irrelevant, to say nothing of the structure or information content of the relationship between the lower level properties themselves.

A molecular level model of a microprocessor has a rather high information content, does it not? And reducing an engineering level model of a microprocessor to a much more detailed molecular level model of the same is a classic example of the reductionist approach. Information isn’t going away, if anything it is being introduced – due to real world physical considerations. No one suggests that we can derive the properties of a modern computer by studying the properties of a single atom. That is absurd.

The emergent behavior is an algebra, the material building blocks are (merely) the carrier. All intelligence needs is enough stuff to form basic spatially and temporally recursive control structures. Then natural selection takes over, literally inventing (through massive pruning) a highly evolved attribute grammar. The carrier is unimportant.

This is an excellent analogy. The first problem is, no one has *ever* done it from neutral initial conditions and neutral transition rules, not even in computer simulation. Take the game of LIFE for example. If we turned a thousand supercomputers to simulating the structures that evolved from random initial configurations, do you think we would ever see something even plausibly described as intelligent, within a thousand lifetimes?

In any case, the mathematics of integers has proven, in the hands of Godel, that it is impossible to know that arithmetic is logically coherent. So you claim to know that which cannot be known.

As far as I can tell the only meaningful dispute here is that you claim that Sober’s anti-reductionism doesn’t include opposition to one of the very things Sober claims reductionism does include. Well, that is a rather peculiar perspective, and there is nothing I can say to make you change you mind, if Sober’s own statement on the subject isn’t sufficient to convince you. I have no desire to belabor the point any further.

JNS: This is nonsense, isn’t it? As I’ve pointed out above, mathematics is not instantiated in the real world.

So you say. I seriously doubt you could find more than a tiny minority of scientists who would agree with you on that point. I have three apples in one basket and two apples in the other basket. Together, I have five apples. Sounds like mathematics is instantiated in the real world to me. Is there any risk that a mathematician is going to come along and prove that I actually have seven?

Here again I do not have anything new to add to the subject, and if elementary arithmetic with regard to concrete objects doesn’t reliably obtain in the world as you see it, I can hardly say more.