with Igor Rojdestvenski's permission:
On Wed, 25 Jan 2006, Igor Rojdestvenski wrote:
> Dear Giuzeppe,
>
> I answer in private as my message limit for this week is over.
>
> 1. I kind of do not understand what is, in your opinion, bottom-up and
> top-down. I always viewes as follows:
>
> Top-down: a system that is built by design
> Bottom-up: a system that develops as a result of creation and recreation
of
> interactions between elements, i.e. self-organizing system.
Top-down: it proceeds by DIFFERENTIATION, from a unity (one cell becomes
several different ones, largely on the grounds of the geometric relative
structuring - and more: embryogenesis)
Bottom-up: put pieces together. Self? I rarely witnessed true
self-organizing, bottom-up. Some active entity usually helps form outside.
The first living entity? first let's try to understand the ongoing living
processes, then try to unify to physical processes, yet to be understood (see
below) and probably not in terms of current physical theories.
> 2. Embryogenesis is also also a combination of top-down and bottom-up. It
> develops in a paradox between hereditary (genetically "planned"
> potentiality) and epigenetic (contextual, environment-ruled actuality).
See
> recent works on drosophila, including very interesting modeling results
of course, the cell is far from equilibrium (it is extended criticality),
thus requires energy fluxes. It also changes in a co-constitutive way with
the environment, but its differentiation has non artificial analog (so
far).
> 3. A single virus might not be autopoietic, but the whole virus world
> certainly is, otherwise we would not be so afraid of mutations of the
> bird's flu :))). In fact, none of the organisms are autonomous, as the
> flows of energy and information from the environment are absolutely
crucual
> for their existense. Same with the virus, just the environment for the
> virus is the cell of another organism.
also a cow, alone in the universe, doesn't survive for long. Yet, there is
a subtle difference between the lack of autonomy of a virus (just genetic
structure, no distinct phenotype, practically) and a cell: autopoieisis
expresses it well, though vaguely.
> 4. If we talk about autonomous theory of living systems, and base it on
the
> concept of the cell as the "primary element", what is wrong
then with the
> modern biology? Cell as a concept is well developed nowadays, and I do
not
> think that anything critically new might arise from such approach, except
> from excluding viruses and such like life forms from living matter
> altogether. I would rather think, and your examples from physics actually
> support this view, that for a new theory we should rething the primary
> concepts and build a new system of axioms.
yes, I agree. yet, I would not use the word axiom: except for Peano
Arithmetic, meaning is needed in mathematics (and even Arithmetic is
incomplete). Let's say, a (informal first) theory of living structures.
> 5. Besides, a single cell is not life, as well as a single organism.
> Evolution exists only in species and populations, there is no evolution
of
> a single organism, a single cell.
yes, of course, the system matters, like for the cow
> 6. And still, what about artificial evolution and A-life in the works of
> the Santa-Fe institute? Is it life or not?
again, was the planetary theory of the atom good or not? Yes, there was
some good in Rutherford, good approximation. We must work at that as well.
But understanding proceeds by separating first, history told us (since
Greek physics was separated from magic: before, gods were handling ethics
and physics and.. much more; but XX century was a splendor for this practice
of separation and left us with fantastic theories and ... major unification
problems). The expected unification is redesigning the very objects of the
theories (strings) or space-time (non-commutative geometry): reduction to
existing theories is not likely to be possible.
Theoretical biology is so poor that people do not dare to claim its
automony as physicists did for they three(!) different phenomenal levels
(by the way, with one fantastic reduction: thermodynamics to statistical
physics, the rare, incomplete, yet remarkable counter-example)
Thank you very much for your attention
g
> Yours, Igor
>
> ----- Original Message -----
> From: "Giuseppe Longo" <Giuseppe.Longo@ens.fr>
> To: "Igor Rojdestvenski"
<igor.rojdestvenski@plantphys.umu.se>
> Cc: <fis@listas.unizar.es>
> Sent: Wednesday, January 25, 2006 1:55 PM
> Subject: Re: Fw: [Fis] Robert Rosen's modeling paradigm
>
> >> > Dear Giuzeppe,
> >> >
> >> > Please allow me making some comments regarding your post.
> >> >
> >> > 1. There are artificial systems which are built bottom-up,
i.e.
> >> > self-organize. One of them is Internet, which is an obvious
example.
> >> > But,
> >> > in fact, any computer system is usually built bottom-up.
> >
> > this is what I said: artificial systems are built bottom-up, in
contrast
> > to embryogenesis, which is top-down.
> > Internet does not "self-" organise: there are humans
typing on keybords.
> > If we were all dead, the net would keep working breifly, as some
daimons
> > would keep exploring the system and create new automatic web pages.
Not
> > for long though, as any computer net as the permanent need of a
system
> > manager fixing it, every few days, every few hours. And this for
deep
> > theoretical reasons, well expressed by concurrency: the dynamic
interface
> > between discrete data types and continuous space-time rapidely leads
to
> > deadlocks and reveals bugs of all sorts.
> >
> >> > 2. I would be careful proclaiming the cell an elementary
component of
> >> > life. How about organelles? Mitochondria and chloroplasts,
which are,
> >> > in
> >> > fact, former separate organisms that were fused into the
plant cells
> >> > in the course of evolution? How about viruses, which are
not cells but
> >> > rather
> >> > "crystals"? Are they alive?
> >
> > They are not autonomous, in the sense of Rosen, they do not have
> > autopiesis, as Varela would put it. A fungus, though a parasite,
does
> > have autopoiesis.
> >
> >> > And my main question is: Why do we always seek distinction
along
> >> > bio-nonbio? Maybe the correct question is self-organizing
and
> >> > self-evolving versus rigidly designed?
> >
> > Physicists have learned that in order to understand one has first to
> > distinguish phenomenal levels. So, from Reimann to Einstein they
said:
> > the geometry of senses has nothing to do with the geometry of
> > astrophysics: reimannian geometry for relativistic spaces is not
closed
> > under
> > homotheties (we cannot transfer Eucild's theorems to outerspaces).
Even
> > more so in quantum physics: the quantum field has nothing to do with
> > relativistic and classical fields; there are no trajectories, in
> > space-time, after 2000 years of a physics of trajectories (from
> > Aristotle, to Newton and Einstein). They had the courage to separate
> > theories: forget atoms as little planets (a fake unit), propose
> > non-locality,
> > non-separability (classical non-senses). Then, of course, once a
robust
> > and effective microphysics, QM, has been built, look for unification
> > (unification, not reduction!).
> > Darwin did the same: at the time when iatro-mechanics still
prevailed
> > (something like those who claim that a metazoan is like a computer
net)
> > he proposed a theory totally unrelated to the physics of his time.
We
> > need an autonomous theory of living matter, along the lines of Rosen
and
> > Varela, starting with the least autonomous living entity: a cell,
with
> > its complexity (also in QM theoretical elements are very complex:
> > strings, classically absurd). Later we may try to unify to suitably
> > developped physical theories (of macromolecules, for example). Our
modest
> > attempt is called "extended criticality" and can be
downloaded.
> > giuseppe longo
> >
> >> > ----- Original Message -----
> >> > From: "Giuseppe Longo"
<Giuseppe.Longo@ens.fr>
> >> > To: "James N Rose"
<integrity@ceptualinstitute.com>
> >> > Cc: <fis@listas.unizar.es>;
<igor.rojdestvenski@plantphys.umu.se>
> >> > Sent: Wednesday, January 25, 2006 11:19 AM
> >> > Subject: Re: [Fis] Robert Rosen's modeling paradigm
> >> >
> >> >> There are at least two crucial differences between a
biological
> >> >> system and
> >> >> an artificial one.
> >> >>
> >> >> 1- The first is built top-down, the second bottom-up.
Embryogenesis
> >> >> begins by a unique cell, which differentiates into
tissues and
> >> >> organs. This espablishes a peculiar causal regimes,
where various
> >> >> levels of organisation are causally entangled since the
beginning
> >> >> (regulation, integration fluxes). So far, we are only
able to
> >> >> construct artificial systems by assembling parts, from
bottom. And
> >> >> this is probably inherent
> >> >> to
> >> >> the very notion of (human) designed system.
> >> >>
> >> >> 2- All our machines are cartesian: they are made
possible (and
> >> >> intelligible) by assemblying simple parts,
> >> >> which may lead to extremely complicated artefacts. Then
the
> >> >> elementary is
> >> >> simple, as Descartes wanted us to decompose beings (and
reasoning).
> >> >> Complexity, in natural systems, begins with the
elementary
> >> >> components, which may be very complex. A single living
cell is
> >> >> elementary (if decomposed, it is dead), but it is
extremely complex.
> >> >>
> >> >> One could also observe that it is exactly this
elementary component
> >> >> of live that in no way we can reproduce.
> >> >> We are able to construct organs of all sorts, and even
assembly
> >> >> them, but not a single cell, with its top-down
generating process.
> >> >> I think that this is compatible with the great
refections by Robert
> >> >> Rosen.
> >> >> More may be downloaded from my web page.
> >> >>
> >> >> Giuseppe Longo
> >> >>
> >> >> Laboratoire et Departement d'Informatique
> >> >> CNRS et Ecole Normale Superieure
> >> >> et CREA, Ecole Polytechnique
> >> >> (Postal addr.: LIENS
> >> >> 45, Rue D'Ulm
> >> >> 75005 Paris (France) )
> >> >> http://www.di.ens.fr/users/longo
> >> >> et :
> >> >> CENtre d'Etude des systemes Complexes et de la
Cognition (CENECC)
> >> >> http://www.cenecc.ens.fr/
> >> >>
> >> >> e-mail: longo@di.ens.fr
> >> >> (tel. ++33-1-4432-3328, FAX -2156, secr. -2059)
> >> >>
> >> >>
> >> >> Upon kind permission of the M.I.T. Press, the book
below is
> >> >> currently downloadable from Longo's web page above (its
n-th
> >> >> edition is out of print...):
> >> >>
> >> >> Andrea Asperti and Giuseppe Longo. Categories, Types
and
> >> >> Structures: an introduction to Category Theory for the
working
> >> >> computer scientist. M.I.T.- Press, 1991. (pp. 1--300).
> >> >>
> >> >>
> >> >>
> >> >>
> >> >> _______________________________________________
> >> >> fis mailing list
> >> >> fis@listas.unizar.es
> >> >> http://webmail.unizar.es/mailman/listinfo/fis
> >>
> >> _______________________________________________
> >> fis mailing list
> >> fis@listas.unizar.es
> >> http://webmail.unizar.es/mailman/listinfo/fis
> >
> > _______________________________________________
> > fis mailing list
> > fis@listas.unizar.es
> > http://webmail.unizar.es/mailman/listinfo/fis
-------------------------------------------------------
_______________________________________________
fis mailing list
fis@listas.unizar.es
http://webmail.unizar.es/mailman/listinfo/fis
Received on Thu Jan 26 19:34:50 2006