Re: Closer To Truth interview

--- In [email protected], "hibbsa" <hibbsa@...> wrote:
>
> 
> 
> --- In [email protected], David Deutsch <david.deutsch@> wrote:
> >
> > On 17 May 2013, at 00:50, hibbsa <hibbsa@> wrote:
> > 
> > > --- In [email protected], David Deutsch <david.deutsch@> wrote:
> > >> 
> > >> On 15 May 2013, at 20:49, hibbsa <hibbsa@> wrote:
> > >> 
> > >>> It seems to me you need conscious decisions for multiverse histories to see 
> > >> 
> > >> Does that mean "you need conscious decisions for multiverse histories to be conscious of..." 
> > >> 
> > >> If so, then yes: consciousness can't exist at all in the absence of consciousness.
> > >> 
> > >>> macroscopic divergence of a significant kind. 
> > >> 
> > >> 
> > >> What does 'seeing macroscopic divergence' mean? Presumably it doesn't mean 'being conscious of more than one universe'. Perhaps it means the existence of apparent randomness affecting macroscopic observables. But you don't need consciousness for that. All you need is amplification, which happens all the time and doesn't need any specially designed equipment.
> > >>> 
> > >>> Which sort of puts MWI in the dock for one of the problems it claims to solve: putting conscious observers at the heart of things.
> > >> 
> > >> It sort of puts MWI in the dock for *your* having put conscious observers at the heart of things, contrary to the predictions of quantum theory. Those are that multiple, approximately decoherent histories are a generic phenomenon in the multiverse.
> > >> 
> > >> -- David Deutsch
> > >> 
> > > 
> > > DD...I'm finding it hard to understand what you are saying in terms of what I am saying. 
> > > 
> > > I think it would help (me) if you answered my first question. Which I'll restate here, a bit clearer hopefully. 
> > > 
> > > Let's say 10^23 instances of approximately the same mole existed in this universe.
> > 
> > That sentence means that we are to consider 10^23 instances of a sample of, say, a gas, each of which would occupy 22.4 litres at standard temperature and pressure. You're right that I didn't realise that that is what you were referring to before.
> > 
> > > I know you'd need more,
> > 
> > I don't understand the thought-experiment you're envisaging, so I don't know what you'd need more of them for. 
> > 
> > > but let's say each of the 10^23 multiverse histories
> > 
> > I can't tell whether it's a coincidence that you're now envisaging the same number of multiverse histories as the number of separate instances in one universe that that you have just envisaged above, or whether this indicates that I have some misconception about what thought experiment you have in mind. But I assume from the following clause that you do mean both numbers:
> > 
> > > was represented by each of those moles in this universe. 
> > > 
> > OK, we have 10^23 multiverse histories of 10^23 samples, so we are considering 10^46 instances altogether.
> > 
> > > What extent would all 10^23 (or however many) moles react to exhibit  the same emergent properties? 
> > 
> > Unfortunately I can't parse that sentence. But, first of all, all 10^23 instances in the same universe have *different* emergent properties, such as centre-of-mass position. Also, as I understand the thought-experiment you have set up, none of the 10^46 instances are interacting with each other. So none of them "react to exhibit" anything.
> > > 
> > > Or said differently, however many instances necessary to represent every multiverse history, of a litre H20 would boil at the same temperature?  
> > 
> > Unfortunately I can't parse that sentence either. But if you're asking: what proportion of the multiversal instances of a litre of water boil at the same temperature -- well, that question mixes two levels of explanation. The literal answer is zero: whatever exact microscopic meaning we might artificially attach to the approximate, emergent concepts 'boil' and 'temperature', the probability of a litre of water 'boiling' at exactly that 'temperature' is zero. 
> > 
> > -- David Deutsch
> >
> 
> DD - I'm still thinking about your reply here. 
> 
> There's a communication issue...we aren't talking about the same thing. 
> 
> I think I'm talking about the reality of abstractions as you describe them in BoI. 
> 
> It could be you are thinking in terms of there having to be fine tuning for these abstractions to emerge similarly across different multiverse histories of the same object. 
> 
> I don't understand why this is. Does water boil with similar characteristics in all multiverse histories of the same can of water heated to boiling, with the same sort of diversity of outcome as 10^30 or whatever cans of boiling water in this universe? 
> 
> I'm sure I'm confused about something here...I just want you to be able to see what I'm confused about, if that's the way things are.
>

DD - I thought it might help things along if I try quoting directly from BoI what it is I am trying to ask you about. 

In the Reality of Abstractions chapter on page 108 you speak of boiling water for the sake of a cup of tea, in terms of a "class of high level phenomena - including the liquidity of water and the relationship between containers, heating elements, boiling and bubbles - that can be well explained in terms of each other aline, with no direct reference to anything at the atomic level or below. In other words the behaviour of that whole class of high level phenomena is quasi-automonomous - almost self-contained. This resolution into explicability at a higher , quasi-autonomous level is known as emergence" 

You go on to explain emergent phenomena are a tiny minority. You explain the reality of the abstract layering in terms of various conceptions of independence (e.g. in terms of predictability of the emergent properties vs predictability of lower level properties. You continue to develop the concept some way forward which I won't try to type given I should think everyone here has the book to hand. 

You consistently refer to the emergent properties in context of a purpose or problem (tea making), and by your profile (for me) I'd normally assume you intend significance to be conveyed by that. It does seem to make sense that the specific emergent properties would be goal or purpose oriented. But objective, non human, ways to represent all possible goals/problems by emergent properties seems a readily soluable problem in itself, even if infinite sets were necessary. 

So it's exclusively this general set of emergent property I am communicating about. Namely for the same transformation of the same object for a given set of emergent properties arising, is it logically possible for the summation of all the possible multiverse histories associated with that object and that transformation in terms of those emergent properties to be significantly different to...the single-universe 'statistical' description of the object in just this universe? 

The best way I can think of to nail this comparison down is by starting with all possible multiverse histories, and comparing that to whatever multitude of that would be necessary for enough objects in this universe to be transformed such that each of the multiverse histories was represented at least once. 

I suppose another possible communication problem would be in terms of how same-universe duplicate objects were defined. The best way to kick that aside, I should think, is to regard the definition of the object as a set of emergent properties in their own right. That way, if you wish, it would seem feasible to tie the transformation of the object to those emergent properties (say...if you required the object to be defined in terms of emergent properties resulting from a transformation of itself....there's probably a fine way to do it). 

So that's the basic question. 

But this is the sort of stream of knock-on questions that seems to fall out of it:  

 Is it even possible for a multiverse history to break free of its emergent properties, in a way that couldn't be described by the relevant single universe statistics of that same object for those same emergent properties?

If it is possible, then is it possible for a multiverse history to be a physically impossible occurrence in this universe? 

If that is possible, is this because a law of physics would be broken? 

If that is the reason, is the law of physics broken in the multiverse history it does occur in? 

If not then do the laws of physics vary in the multiverse, *differently* to their variance across this universe? 

if so, are there variations of the laws of physics in the multiverse that are impossible in this universe? 

if so, is that because a higher law of physics would be broken in this universe? 

and so on and so forth....as what seems to be an endless loop that goes up through each level of abstraction. 

IMPLICATIONS: If these emergent properties are abstract realities, and the possible multiverse histories for the same object/transformation cannot break free of the associated emergent properties, then that seems to place an absolute limit on the total relative divergence possible for any set of multiverse histories in terms of the same common ancestor universe. 

Very tightly constrained indeed. 

And I do completely recognize and anticipate this is all explained by some hopefully embarrassing misconception myself. But I really need to be told what it is....I don't think I can get past this on my own. I'm too looked into being convinced. I've even halfway finished a mathematical formalism using a mathematical induction variant.