[FRIAM] better simulating actual FriAM

[David Eric Smith]

Hi All,

Two quick comments (two emails).

I think the way Glen has this set up below does all the main work Nick wants done.

The idea is that state spaces of different dimensions are different kinds of settings.  A state space for one ladder is one kind of setting.  A state space for two ladders is a higher-dimensional and different kind of setting.  2-sticks 3-sticks 4-sticks 5-sticks likewise.  There can be regions in these state spaces where properties are found that simply are not found in other regions.

Here I think the word “irreducible” often does better service to the idea Nick wants than the word “emergent” that he has used historically in reference to the sticks.  The property of two ladders’ holding each other up is an irreducible property of some configurations, which is not a property of other configurations.  Consider that even being in the off-diagonal quadrant of a square is a property of the square that cannot be “reduced” to a property of either of its edge-axes, because “off-diagonalness” is defined through a relation.

The reducible/irreducible dichotomy that we are using here in a somewhat informal way shares an umbrella with a variety of more technical uses that are very similar in spirit, such as reducibility or irreducibility of representations of groups in algebra, and those kinds of ideas.  I have on this list referred to Tononi’s Phi as being mainly an attempt to quantify the irreducibility of an information-transmitting pipeline made of events on some network.  

Best,

Eric




> On Jul 21, 2020, at 3:05 AM, uǝlƃ ↙↙↙ <gepropella at gmail.com> wrote:
> 
> In the 2-ladder system, if there's downward causation, I would not say "each causes the other not to fall". But I would say something like "the attribute of non-falling constrains the valid arrangements of the ladders". The point of rewording it like that is to remove the emergentism woo and talk more closely about the freedom of 2-ladder arrangements. As with the sticks, it's easier to place 1 ladder than it is 2 ladders. The requirements to be met  circumscribe a space of possible arrangements. And if you cherry-pick some special set of constraints, then it can *seem* magical that, in some arrangements, some requirements can be met that no/few other arrangements meet.
> 
> But objectively, all we're talking about is the space of possible ways to place ladders, given some set of requirements. Obviously, if you inscribe the conclusion into the premises by setting your requirement to be "the ladders must extend by their lengths up into the air", then the set of arrangements of 1 ladder that meet that requirement will  be smaller than the set of arrangements of 2 ladders that will meet it. But if you pick *another* requirement, say, "all ladders must be perpendicular to all other ladders", then laying the 1st ladder is trivial and the 2nd becomes more difficult.
> 
> A minimal conception of downward causation is *only* that the collective constrains the space of arrangements of the parts.
> 
> There is a debate we could have whether *some* systems (parts and the ways they compose) about whether or not collectives can *facilitate* (enlarge) the space of possible arrangements. I call that concept "scaffolding". Your ladder example is a good foil because it allows us to argue that the *size* of the valid 1-ladder arrangements that meet the criterion is 0. But the size of the 2-ladder arrangements that meet it is larger (2 or 3 ways to arrange the ladders such that they stick up into the air). But I'd argue it's an imputation. That cherry-picked arrangement (so that they don't fall) is NOT downward causation because that requirement was installed from the outside, imputed, not an inherent property of ladders and their possible arrangements. (I.e. 2 ladders leaning against each other is not a special state of 2-ladder arrangements.)
> 
> We might be able to argue that EricS' and Morowitz' hierarchy of matter phases might qualify as scaffolding, too ... a little bit of freezing might facilitate regions of the space so that we can get weirdo things like fish or plants. Remove any of the prior freezing layers and life may not "emerge" at all.
> 
> But none of this extra conversation is necessary to get that minimal conception of downward causation.
> 
> On 7/20/20 10:32 AM, Frank Wimberly wrote:
>> There are many systems with causal graphs with feedback loops.  In genetic regulatory networks, for example.  Is that downward causation?
>> 
>> A classic example is the case if two ladders leaning against each other so that neither one falls.  Each causes the other not to fall.
> 
> -- 
> ↙↙↙ uǝlƃ
> 
> - .... . -..-. . -. -.. -..-. .. ... -..-. .... . .-. .
> FRIAM Applied Complexity Group listserv
> Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
> un/subscribe https://linkprotect.cudasvc.com/url?a=http%3a%2f%2fredfish.com%2fmailman%2flistinfo%2ffriam_redfish.com&c=E,1,CxfH1wKwKGMOeiKFJzWCU0K1rWmo8s0JRhJfdpW9zTIXeC2kJ6F2zBDBl8TwABstatV3tls8yk7vsoo1K9PfsrfFID_AkXRK-qRjryL5PcunxTlQYpxSxMw,&typo=1
> archives: http://friam.471366.n2.nabble.com/
> FRIAM-COMIC https://linkprotect.cudasvc.com/url?a=http%3a%2f%2ffriam-comic.blogspot.com%2f&c=E,1,QWJWueFEylVLFHMUanuYhE1Bz9SvaDuRXU-g6bY0xqiVXSGd-wIAZU-u06O80HFeUDpRasLWacMBBLlivDmQxqTtFSlAE82cwwvojVgM&typo=1