Recognising Top-Down Causation by George F. R. Ellis

Frank,

no one but you thinks your essay is fundamental to science. Your claim that dreams can be used as a basis of understanding physics takes you outside the mode of operation of the scientific community, because dreams are not replicable evidence.

You are not willing to concede this point, so we have such different standpoints that no profitable debate between us is possible. I will not be responding to any further posts from you.

George

"There is nothing new under the sun": one open question and two failed challenges.

The one respondent on this thread who has seriously challenged the scientific content of my essay is an anonymous physicist operating under the pseudonym "There is nothing new under the sun", claiming I have not given a single genuine instance of top-down causation that could not be explained in a purely bottom up way. I have responded to him in various posts, particularly one on Sept 14:2012@22:51 GMT, and in a summary response on Sept 17th:2012 @18h15GMT give a series of counter examples to his claims.

I have conceded that one point in my argument can be queried. He has failed two challenges I have set him.

The point that remains open is the validity of my arguments regarding the Caldeira-Leggett model, which I claim is a top-down effect; he claims it can be explained in a purely bottom up way via the renormalisation group. I still believe that my argument, set out in detail here , is valid, but I have to look into the link with the renormalisation group when I have time. That study might make me withdraw my claim about the Caldeira-Leggett model, or it might lead me to claim that renormalisation group descriptions, like superconductivity theory, embody an essential top-down element. This is work in progress; but I acknowledge that there is a legitimate query to be answered.

The first challenge he has failed is as follows: on Sept 14:2012@22:51 GMT, I wrote the following: "Here is a challenge for you. Explain to me in a purely bottom up way how state vector preparation is possible, as for example in the Stern Gerlach experiment. Quantum physics is unitary, as we all know: how does the non-unitary behaviour of state vector preparation emerge in a purely bottom up way from that unitary dynamics? You won't be able to explain this action without invoking the effect of the apparatus on the particles - which is a form of top down action from the apparatus to the particles." He has not responded to this in any way. He has failed that challenge.

The second challenge he has failed is as regards the arrow of time. He has strongly insisted in various posts that Weinberg's quantum field theory derivation of the H-Theorem resolves the arrow of time issue in a purely bottom up way, because it shows that entropy always increases. As well as referring him to other sources that support my view, I have responded to this claim with a step by step demonstration that this is not the case: see my posting of Sept.16:2012@14:06 GMT, which definitively shows the arrow of time issue cannot be resolved in a purely bottom up way, because Weinberg's derivation -- just like Boltzmann's -- works in both directions of time.

In my follow up posting on Sept19:2012 @ 05:37 GMT, I said the following:

" So come on. Which is it?

* Do you have a counter argument showing I'm wrong? If so what is it? Where is the mistake in this elementary logic?

or

* Do you have the stature to concede you and your Santa Cruz experts are simply wrong? - you did not grasp this elementary logic?

or

* will you lurk in the shadows, unable to answer and unable to admit you were wrong? -- proving you don't have the capacity to admit that you are wrong, nor the stature required to apologise for the insulting nature of your comments.

If you give no reply, you choose the last option. Wheeler, Feynman, Sciama, Davies, Zeh, Penrose, Carroll, and others including myself are vindicated, and your condescending comments are discredited."

He has chosen the third course. He has failed that challenge as well.

I only need one example to prove that top-down processes do indeed occur in physics, just as they do in many other contexts such as in digital computers and in the human brain and in evolutionary theory . My case (elaborated here ) stands undefeated.

George Ellis

Dear Fred,

that is an interesting question.

There are two kinds of duality in physics. The one is such cases as the Ads/CFT duality and the duality between Lagrangian and Hamiltonian formalisms, where they are different descriptions of the same physics. The other is the kind of duality that occurs in quantum physics as expressed in Dirac's bra and ket notation, based in the duality between vectors and co-vectors, where these are complementary aspects of the physics that work together to give the final outcome.

I suggest that the case of bottom up and top down causation is a duality of the second kind. This dual working allows interlevel feedback loops, which are the underlying enabling factor for the emergence of true complexity in biology.

George

Dear Eugene

I have carefully explained in my essay that I do not claim there is an identifiable topmost or bottom most layer:my comments relate to relations between any two neighbouring levels (and hence to any further levels related via neighbouring levels).

Is the idea of levels vacuous or arbitrary, as suggested by Pentcho Valev? Well please tell me what is wrong with Tables 1 and 2 in my essay. The very existence of different academic subjects such as physics, chemistry, and biology is testimony to the reality of the hierarchy. I give a much fuller exposition of its nature here . In my reply to Pentcho on Sep. 9, 2012 @ 17:56 GMT, I give references that show the reality of the hierarchy in many different contexts.

I agree there are complications in determining its nature in very complex contexts; there is a whole literature on how to determine it in complex networks. Remember I am only claiming existence of local hierarchies; and there is abundant evidence there are a great many such local hierarchies in physics, biology, computers, and the brain. Please see for example the book by Campbell and Reece on biology and the book by Tanenbaum on computers (the references are in my essay).

A note to readers of this thread: I am no longer prepared to stand for the personal insults that Pentcho Valev insists on including in his postings to my thread. The administrator of this competition, who decides whether to accept or reject requests for deletions of posts, agrees with me that the tone of his postings has been intolerable.

If further such personal attacks are made in postings to my thread by anyone at all, I may well cease reading what is posted here and stop answering all postings. Why on earth should I put up with this kind of behaviour?

George Ellis

The reference was this

GE

Hi Tom

thanks for that, that is an interesting paper.

You quote "No real experiment, having finite precision, can demonstrate the presence or absence of continuity, which is defined in terms of limits." I agree. In fact I have a closely related strong position, based on a statement by David Hilbert:

(a) no physics theory or proof that relies on infinity in an *essential* way describes the real world;

one implication is that spacetime must be quantised at a small enough scale (which is supported by many other arguments);

(b) the claimed existence of infinities of anything whatever in any physical theory is not a scientific statement, as there is no way that this claim could ever be observationally or experimentally proved.

However I did not see any close link of Lamport's paper to state vector preparation. Did you mean another paper? Or is it related to the fact that state vector preparation is never perfect? - even if so, it's still non-unitary (consider a wire polarizer).

Best,

George

So I perused Lamport's very interesting set of papers, and would like to comment on #31: "On-the-fly Garbage Collection: an Exercise in Cooperation" (with Edsger Dijkstra et al.)

The point here is that garbage collection is a crucial part of computing, and is an example of the top down process of adaptive selection: that is of selecting a set of elements to be deleted, leaving behind the ones that are meaningful. This depends on a selection criterion, which (in my terms) lives at a higher level of abstraction than the elements to be selected. That is made explicit here: " starting from the roots, all reachable nodes are marked; upon completion of this marking cycle all unmarked nodes can be concluded to be garbage, and are appended to the free list". The selection criterion is reachability.

This is the key process by which meaningful information is garnered: you delete the stuff that is not meaningful, thereby creating a smaller set of stuff which has meaning. This is related for example to deleting emails and unwanted files n your computer, as well as to the fact that this process (it's essentially clearing memory) is where entropy is generated - because it's an irreversible process (assuming that deleted stuff is gone). Just like biology (deleted animals are past history) -- and like state vector preparation.

Best,

George

Ok I'm very slow. Now I see your relation of Buridan's Ass to selection, which is of course a binary decision.

Yes it's very nice. The idea of quantisation of the decision process is needed: it's based in discrete rather than continuous variables - which fits in well with Hilbert's maxim: "the infinite is nowhere to be found in reality, no matter what experiences, observations, and knowledge are appealed to." There is no continuum of any variables in the real world. Another example of the disjunction between mathematical models and reality.

George

No I did not.

GE