UNFINISHED DRAFT
11 Feb 2021
How to make a competent newly-hatched chick out of chemicals
Some background ideas
Aaron Sloman
Honorary Professor of AI and Cognitive Science
http://www.cs.bham.ac.uk/~axs
School of Computer Science
University of Birmingham, UK
Based partly on correspondence with Vera Matarese
(who cannot be blamed for any errors).
https://sites.google.com/view/veramatarese/home
This document is available at
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/make-chick.html
It is part of the background for my presentation at Sussex University on 16 Feb 2021.
http://www.sussex.ac.uk/cogs/seminars
Metaphysical Causation
I was originally introduced to the phrase "metaphysical causation" by Alastair Wilson https://alastairwilson.org/This document was triggered by a presentation by Vera Matarese https://sites.google.com/view/veramatarese/home at a Framephys workshop on 8th Feb 2021 organised by Alastair (who was then at the University of Birmingham). Vera is not responsible for any errors in this document.
If metaphysical causation happens anywhere it happens inside eggs!
However, although I have been exploring varieties of metaphysical causation for
some time, especially in the context of the Turing-inspired Meta-Morphogenesis
project
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/meta-morphogenesis.html,
until late in 2020 I did not notice the relevance of hatching processes in eggs
(e.g. of chickens, ducks, alligators, turtles, and many more). Those eggs
initially contain a tiny amount of genetic material and relatively
undifferentiated masses of chemical material in the rest of the egg (including
yolk and surrounding matter), which is gradually transformed into an animal
ready to break out of its shell and act in the environment, as shown in this
video:
https://www.poultryhub.org/chick-embryo-development-animation
During her presentation Vera discussed Esfeld's metaphysical "Super-Humeanism" thesis and seemed to me to be raising doubts about its adequacy in the light of biological facts, e.g. facts about development of biological organisms. Those doubts reminded me of doubts I had when I heard him give a presentation in Oxford in 2017 in which he presented a particle-based view of physics in which the particles did not seem to have rich enough properties to be able to form enduring chemical structures that survive potentially damaging influences, including the many different chemical structures found in forms of life.
This is not an attempt to summarise reasons why I think that such a "minimalist" metaphysics is inadequate. I shall not consider possible counter arguments, or defences of Esfeld's claim, since my main aim is not to refute Esfeld but to present some of the metaphysical features of biological development and biological evolution that indicate the enormous creativity of this universe. If his thesis does accommodate such things, that's good!
The phenomena I am referring to are strikingly illustrated in the developmental processes that occur in an egg whose contents are initially relatively undifferentiated yet organise themselves into a richly differentiated living organism, such as a chick, duckling, crocodile, turtle, etc. These egg-processes seem to be highly metaphysically creative, in view of the huge differences between the initial contents of the egg and the contents that emerge during hatching.
At the lowest level, the universe is obviously not made only of Newtonian particles that can move around in space interacting via gravitational (attractive) forces and electrical, magnetic and electromagnetic forces. [Unfortunately only dimly remembered from my ancient 1956 physics degree and a little postgrad study.]
What's missing from that view is that chemical bonds can form and be released, especially in the presence of catalysts: bonds are very different from attractions and repulsions, and have no place in Newtonian physics, as emphasized by Schrödinger in 1944 (What is life?).
[I have some extracts from Schrödinger's book with added comments here, though it's possible that I have misunderstood him because I studied so little physics so long ago:
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/schrodinger-life.html (also PDF)]
Particles that *bond* in the presence of a catalyst don't necessarily *attract* each other when close, like electrically charged particles or ends of magnets. I.e. bonding doesn't use a "standard" attractive force.
Some bonds are very stable, and can be difficult to create or unlock, except in the presence of catalysts.
The single most important point concerning life strongly emphasised by Schrödinger seems to be that QM allows catalysts to play a role in formation and release of strong bonds, for which there is no place in Newtonian physics.
They are also not mentioned in talks I've heard by Esfeld, and another defender of particle-based physics: Fay Dowker (Imperial, London).
Bonds can persist while molecules attached via bonded components change their spatial configuration e.g. DNA twisting or untwisting, strands coming together and separating. Bonds can resist quite strong forces, as required for animal bodies, plant stems, tree trunks, teeth, antlers, etc.
Those spatial constructions seem to me not to be allowed for in the kind of particle-based theory I understood Esfeld and Dowker to be defending, to which I took your talk to be referring critically, which was why I started our conversation during the break (joined by Mike).
If an impulse is applied to a cluster of Newtonian particles they will all scatter -- gravitational attraction, the *only* attractive force in a Newtonian universe, cannot prevent such scattering, although if the impact is small the scatter will also be small and may be countered by gravitational attraction, unlike a chemical bond which can keep particles connected when scattered by impact, if the impact is not too large.
To that extent, chemical bonds have an effect that's partly analogous to the nuts and bolts and other connecting devices in construction-kits like Meccano and Fischer technic. [As a child, I played a lot with meccano aged 5 to 11.]
Magnetic forces are partly like that, especially when flat surfaces with opposite poles are together. But from my childhood experience a ball attracted to a flat magnetic surface can fairly easily be rolled along the surface, which is not true of objects bonded chemically, like bits of bone, or plant fibre.
Also, what a particle will bond with in various conditions can depend on what other particles it is already bonded with. There's nothing like such effects, nor such particle type differences, in Newtonian mechanics. Electromagnetic effects added to Newtonian physics don't seem to me to suffice to explain chemical bonds.
As Schrödinger points out, chemical bonds, and processes that form and release such bonds, are crucial for the ability of physics to support life, and especially biological reproduction.
As far as I could tell from Esfeld's presentation, he made no allowance for particles to switch between being bonded to form a larger structure and merely being adjacent.
When I pointed out to him that the system he described, like Newtonian physics, cannot support formation of new structures like molecules formed from atoms he replied that Newtonian physics can explain formation of structures, e.g. galaxies.
But, since component stars and planets in galaxies don't have relative-motion restrictions remotely like chemical bonds, that reply is so inept, that I hope I misunderstood him, or have misremembered. (Al may have a view on that.)
I had the same impression about Fay Dowker (Imperial, London), when I heard her give a talk here to physicists a few years ago emphasising what she calls "causal set dynamics" (partly based on Raphael Sorkin's work):
http://www.imperial.ac.uk/people/f.dowker
As far as I can tell Sorkin also ignores the key phenomena of biology that physics needs to explain.
(Fay and I first met at a conference related to the Turing centenary and we both had papers in a volume following that conference, edited by Cooper and Soskova.)
While preparing this message, I found this fairly recent article, which seems to be relevant, though I've so far only 'peeked' at it:
https://royalsocietypublishing.org/doi/full/10.1098/rspa.2018.0674
The origins of quantum biology
Johnjoe McFadden and Jim Al-Khalili Published:12 December 2018
This is how it ends:
What remains indisputable is that the quantum dynamics that are undoubtedly taking place within living systems have been subject to 3.5 billion years of optimizing evolution. It is likely that, in that time, life has learned to manipulate quantum systems to its advantage in ways that we do not yet fully understand. They may have had to wait many decades, but the quantum pioneers were indeed right to be excited about the future of quantum biology.
NOTES
"Optimising" does not adequately express the *creativity* of evolution,
including its ability to produce major discrete changes. In contrast, optimising
suggests a gradual process of improving something that already exists.
A fact that sometimes goes unnoticed is that evolutionary change is inherently discrete: there can be at most a finite number of individuals between an organism and each of its descendants, or predecessors. Therefore the changes cannot form a continuum.
Not only are they discrete, some biological changes can produce individuals with missing or duplicated parts, or parts that would be described as deformities.
My impression is that Esfeld and Dowker and many other physicists and philosophers of physics have missed important features of the physical universe that I also missed until very recently when thinking about how eggs produce chicks. In particular, the processes of assembly of increasingly complex molecular structures playing different, but collaborative, roles in both the *formation* of highly integrated systems and also in their collaborative *functioning* after formation would be impossible without:
highly sophisticated *assembly mechanisms* performing a variety of different sorts of assembly and coordination tasks, e.g. assembling all the materials and structures required to form a leg, or a wing, or a digestive system, or a heart, or brain, or the nervous system distributed over an animal body, or an integrated functioning system composed of all those...!
What produces those assembly mechanisms? The only possibility seems to be that the construction of the assembly (and maintenance) mechanisms must also use chemically implemented assembly and maintenance mechanisms. I.e. biology uses what computer systems engineers call boot-strapping.
The simplest mechanisms must be semi-automatic so that initial stages of DNA-triggered assembly are triggered by fairly basic physical structures and relationships, e.g. forces of attraction and bonding triggered by proximity of parts.
But the assembly of more complex structures whose parts have to be assembled and then brought together while retaining their internal structure requires the presence of assembly mechanisms that can bring required previously assembled components together in the required spatial relationships: whether rigid, as in bones and shells, or flexible, as in muscles and nerves.
For more complex structures, mere random motion (e.g. thermal motion) cannot suffice, e.g. if several components have to be spatially aligned and then moved together.
(Compare the inability of random physical processes to do the following:
align two holes in two metal strips
get a screw-threaded bolt through the holes
bring a nut in the appropriate orientation to the end of the bolt with the hole of the nut appropriately related to the axis of the bolt,
rotate the nut in a plane perpendicular to the axis of the bolt, while pushing the nut in the direction of the axis
all of which are required to form a new enduring structure made of the two strips held firmly together by the nut and the bolt. Are biological assembly processes simpler than that??)
This needs to be unpacked
If the construction of more complex components of an organism requires prior assembly of required parts followed by final assembly steps, and the required parts also need to be constructed from pre-existing, previously constructed, required parts, except for the smallest parts assembled from readily available atoms or molecules, and if the more complex assembly processes require previously constructed manipulators or assembly frameworks, then the construction of the whole organism and its parts will also require construction and control of construction machinery of various kinds (including branching *types* of machinery for later stages as diversity and complexity of newly assembled products increases).
Flowing water does not require intelligence to make a meandering river bed, but construction and assembly of parts as different as muscle fibres, tendons, blood vessels, nerve fibres, neurons, bones, skin, fur, feathers, eyeballs, beaks, teeth, etc.; and the prior assembly of increasingly complex molecules used in those new structures, would be impossible without informed control, i.e. mechanisms using information about locations, orientations, and relations between previously assembled parts -- at a late stage creating connections between bones, tendons and muscles, along with neurones and appropriately routed nerve fibres, so that the nerves can transfer (push? pull?) information from sources to various information-using mechanisms.
There may be some cases where existing gradients, e.g. electrical or gravitational gradients, suffice to control new movements of parts, but the rapid and reliable multi-stage, multi-branch assembly of something as complex as a newly hatched chick will need some very specific assembly machinery, including highly competent control mechanisms, whose complexity of function will have to be much greater at later stages of development of the organism.
Whereas evolution can spend millions/billions of years searching for good solutions, individual development cannot: mechanisms and strategies are needed that work very rapidly, as required for their tasks, though limited by the speeds of the 'food transport' mechanisms and the speeds of various types of information processing (control) machinery.
The machinery used to construct nerves and brains cannot all depend on pre-existing nerves and brains. Instead there must be bootstrapping, possibly multi-stage bootstrapping of increasingly intelligent assembly mechanisms, with some understanding of spatial structures and relationships. If possible, simpler forms of assembly apparatus should be used in construction of more complex forms, so that the latter do not have to be assembled from scratch using basic physical particles.
The only possible options are chemical control mechanisms that exist before brains do.
Moreover, whereas evolution of life on a planet can explore multiple branching routes in parallel, using many different coexisting individuals of different kinds, developments within an individual organism cannot afford so much branching (individual lives are too short) and must therefore reuse information about what has been found to work in previous generations: this requires mechanisms for "compiling" information found by evolution into information that later on can be used (far nore quickly) in individual development.
Is it possible that after producing such mechanisms in the distant past evolution found new post-hatching uses for the spatially informed chemical control mechanisms that initially evolved for use in building key components of the organism?
Those new uses could include controlling actions of the whole organism in relation to its environment after hatching (or birth).
Once started, such a process could go on evolving, e.g. producing new mechanisms for coordinating the information used by earlier control mechanisms. Could that be how brains evolved, while retaining chemical control mechanisms for use internally during construction and maintenance, along with new copies for controlling external actions, such as actions that require information to be acquired by new sensors?
How many millions of years would it take to produce big-brained primates with rich languages for encoding and using or storing information, initially only information required for assembling the organism, but later on also mechanisms for acquiring information about the environment that can be used for acting in the environment. (This might later have led to ancient discoveries in geometry and topology, long before Euclid.)
Despite their differences, the interior of the organism and its environment would both be spatio-temporal, so that previously evolved internal spatio-temporal control mechanisms might later prove useful for control of actions in/on an external environment: using spatial and temporal control - e.g. moving toward and consuming sensed nutrients and moving away from sensed dangers, then later producing newly evolved abilities to find good control processes for combining such actions.
Since both the interior of the organism in which control mechanisms are needed and the external environment are spatio-temporal structures, some of the sensory, reasoning, and control mechanisms that initially evolved for self-assembly might later be used (with increasingly complex modifications) for sensing and acting on objects in the environment.
Ancient human discoveries in geometry and topology could then turn out to be products of modified copies of earlier control mechanisms that used spatial intelligence during construction of individual humans (and their predecessors).
If all this is at least approximately on the right lines, then spatial intelligence in humans and other animals may be implemented in recently evolved, sub-neural chemical mechanisms using structural manipulations, rather than in currently fashionable neural networks collecting statistics and deriving probabilities, which could not provide a basis for the ancient mathematical discoveries of spatial necessities and impossibilities in geometry and topology.
If these ideas are not all wildly mistaken, then alongside the history of production of an animal (chick, duckling, alligator, .... human) and all its well-studied parts there must also be a history of production of hitherto unnoticed (increasingly "competent") chemically implemented *assembly/construction mechanisms* that perform increasingly complex assembly tasks during individual development, including *maintenance* of previously assembled physiological components in parallel with construction and assembly of new components combining results of previous assembly processes.
Re-deployed variants of those ancient assembly mechanisms could be linked to external sensors and manipulators.
So, as illustrated in this
Chick-development video,
https://www.youtube.com/watch?v=PedajVADLGw
construction of a chicken within an egg requires a cascade of constructions
of new, increasingly complex construction kits/tools, including construction
kits for building brains with chemically implemented spatial intelligence that I
suspect supported the ancient discoveries in geometry and topology by human
mathematicians, both before and after Euclid collected a subset of those
discoveries in his Elements.
If all that is correct then current artificial neural nets are not replicating the functions of brains, but doing something simpler, based on collecting statistics and deriving probabilities, rather than producing the understanding of spatial necessity and impossibility required for the discoveries in geometry and topology. Kant understood that something deeper and richer was required (that may "forever lie concealed in the depths of the human soul").
Since around 2011, I've been working on a Turing-inspired theory of evolved construction kits, including construction kits for producing construction-kits, as reported in this messy, still evolving, paper derived from an old version published in a collection of papers for Turing's centenary
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/construction-kits.html
(also PDF)
-- but it did not occur to me until a few weeks ago that some of the history of construction kits used in a species must be replicated *in each individual organism* during its construction from an egg or seed.
This seems to be a novel version of the slogan: "ontogeny recapitulates phylogeny", which is widely regarded to be an out of date myth (e.g. relegated to "biological mythology") as stated in
https://en.wikipedia.org/wiki/Recapitulation_theory
I think these ideas can also be shown to illustrate Alastair Wilson's ideas about "grounding as metaphysical causation" (G = MC !), insofar as the processes I refer to are metaphysically creative: new *types* of material, structure, process, etc. come into existence both during biological evolution and during individual development.
As far as I can see, life, including evolution, reproduction and development, would be impossible in a particulate physical universe without any chemical bonding that significantly alters (reduces) the degrees of freedom of particles and thereby produces new types of entity with different sorts of degrees of freedom.
Similar things happen when a child uses a construction kit (e.g. tinkertoy, meccano, lego, fischer-technic) to create a new complex object O composed of less complex parts that persist through configuration changes as O becomes more complex and also capable of increasingly complex behaviours that depend on restrictions of degrees of freedom of some of the parts -- e.g. parts that must remain rigid, or parts whose non-rigidity is restricted to operation of a hinge, or to rotation around an axle, two examples of one-dimensional non-rigidity, which can themselves be composed to form new types of non-rigidity.
However, no child's construction kit comes close to matching the construction kits biological evolution plays with, including meta-construction-kits, all based on chemistry.
One of my long term motivations in doing all this is to defend Kant's philosophy of mathematics against fashionable criticisms. I started that defence in my DPhil thesis (1962) http://www.cs.bham.ac.uk/research/projects/cogaff/sloman-1962, then tried to elaborate it when I discovered AI and learnt to program (from about 1970) but eventually around 2011, I concluded that the tasks require forms of chemical processing that are richer than the discrete forms of processing in Turing machines, electronic computers, etc. I.e. universal Turing machines are not really universal.
(Turing himself distinguished mathematical intuition from mathematical ingenuity, claiming (in 1936) that computers were capable of ingenuity, but not intuition. But he did not explain what he meant by "intuition" and why computers are not capable of intuition. I don't know whether he ever read 'What is life?' or related Schrödinger's ideas to his distinction.)
My impression is that it would be hard, or impossible, for Esfeld, Dowker, and similar particle metaphysicians, to accommodate all the above.
I gave an invited Zoom talk closely related to this at Sussex university
on Tuesday 16 Feb 2021 4pm GMT, as advertised here:
http://www.sussex.ac.uk/cogs/seminars
Note added 2 Feb 2024:
A far more complex and ambitious discussion is now available at
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/metamorphosis.html
NOTE FOR PHILOSOPHY TEACHERS
I suggest that, in view of what we now know about life, and the rate at which
such knowledge is being extended, teaching philosophy of mind and philosophy of
mathematics without teaching any evolutionary and developmental (micro)biology,
and associated biochemistry, is educationally misguided.
References
Chicken embryo developmenthttps://www.poultryhub.org/chick-embryo-development-animation
Photographs of chick embryo stages (jpg):
https://www.cs.bham.ac.uk/~axs/fig/chicken-egg-devel.jpg
apa-stuff.d/Poster_Chick_Embryo_Dev_English.pdf
DAY 1: Appearance of embryonic tissue.
DAY 2: Tissue development very visible. Appearance of blood vessels.
DAY 3: Heart beats. Blood vessels very visible.
DAY 4: Eye pigmented.
DAY 5: Appearance of elbows and knees.
DAY 6: Appearance of beak. Voluntary movements begin.
DAY 7: Comb growth begins. Egg tooth begins to appear.
DAY 8: Feather tracts seen. Upper and lower beak equal in length.
DAY 9: Embryo starts to look bird-like. Mouth opening occurs.
DAY 10: Egg tooth prominent. Toe nails visible.
DAY 11: Cob serrated. Tail feathers apparent.
DAY 12: Toes fully formed. First few visible feathers.
DAY 13: Appearance of scales. Body covered lightly with feathers.
DAY 14: Embryo turns head towards large end of egg.
DAY 15: Gut is drawn into abdominal cavity.
DAY 16: Feathers cover complete body. Albumen nearly gone.
DAY 17: Amniotic fluid decreases. Head is between legs.
DAY 18: Growth of embryo nearly complete. Yolk sac remains outside of embryo. Head is under right wing.
DAY 19: Yolk sac draws into body cavity. Amniotic fluid gone. Embryo occupies most of space within egg (not in the air cell).
DAY 20: Yolk sac drawn completely into body. Embryo becomes a chick (breathing air with its lungs). Internal and external pipping occurs.
https://www.youtube.com/watch?v=PhOqP_GasVs
Baby Crocs Hone Hunting Skills -- National Geographic
https://www.youtube.com/watch?v=nOkq69T6j7E
Ducklings first feed after hatching. First Swimming baby ducks.
Hatched without mother. (Incubated??)
https://www.youtube.com/watch?v=9jRSgZVhWvw
Baby chicks with hen.
https://www.youtube.com/watch?v=OsoNKlyFtpI
Chimpanzees React to Their Reflections in a Mirror
CenterForGreatApes
https://video.nationalgeographic.com/video/00000144-0a34-d3cb-a96c-7b3dd2970000
Mother crocodile takes babies swimming, to hunt for food.
Some References
Aaron Sloman, 2013--2018, Jane Austen's concept of information (Not Claude Shannon's)Online technical report, University of Birmingham,
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/austen-info.html
http://www.cs.bham.ac.uk/research/projects/cogaff/misc/austen-info.pdf
William Bechtel, Adele Abrahamsen and Benjamin Sheredos, (2018), Using diagrams to reason about biological mechanisms, in Diagrammatic representation and inference, Eds. P. Chapman, G. Stapleton, A. Moktefi, S. Perez-Kriz and F. Bellucci, Springer, https://doi.org/10.1007/978-3-319-91376-6_26
Erkurt M. (2018) Emergence of form in embryogenesis. in Interface, 15: 20180454. Journal of the Royal Soc. http://dx.doi.org/10.1098/rsif.2018.0454
Godfrey-Smith, P. (2007). Innateness and Genetic Information. In P. Carruthers,
S. Laurence, & S. Stich (Eds.),
The Innate Mind Volume 3: Foundations and the Future
(pp. 55-105). OUP.
https://petergodfreysmith.com/PGS-InfoAndInnate.pdf
Godfrey-Smith, P. (2017). Other Minds: The Octopus and the Evolution of Intelligent Life, William Collins.
Carl G. Hempel (1945), Geometry and Empirical Science, in American Mathematical Monthly, 52, 1945, also in Readings in Philosophical Analysis eds. H. Feigl and W. Sellars, New York: Appleton-Century-Crofts, 1949, http://www.ditext.com/hempel/geo.html
Kant, Immanuel (1781). Critique of pure reason, (Translated (1929) by Norman Kemp Smith), London: Macmillan. Retrieved from https://archive.org/details/immanuelkantscri032379mbp/page/n10/mode/2up
Piaget, J. (1952). The Child's Conception of Number. London: Routledge & Kegan Paul.
J. Schmidhuber, 2014, Deep Learning in Neural Networks: An Overview, Technical Report IDSIA-03-14, IDSIA, http://arxiv.org/abs/1404.7828
Schrödinger, E. (1944). What is life? Cambridge: CUP.
A. Sloman, (1965) `Necessary', `A Priori' and `Analytic', in Analysis 26, pp. 12--16, http://www.cs.bham.ac.uk/research/projects/cogaff/62-80.html#1965-02
Sloman, A. (2013). Virtual machinery and evolution of mind (part 3) Meta-morphogenesis: Evolution of information-processing machinery. In S. B. Cooper & J. van Leeuwen (Eds.), Alan Turing - His Work and Impact (p. 849-856), Amsterdam: Elsevier. http://www.cs.bham.ac.uk/research/projects/cogaff/11.html#1106d
A. M. Turing, 1952, The Chemical Basis Of Morphogenesis,
Phil. Trans. R. Soc. London B 237, 237, pp. 37--72,
Alastair Wilson, 2017,
Metaphysical Causation,
Nous
https://doi.org/10.1111/nous.12190
(He also has a book published later.)
MORE REFS
A more recent document about varieties of causation and control in biological processes of evolution, reproduction, development, metamorphosis, etc.:http://www.cs.bham.ac.uk/research/projects/cogaff/misc/metamorphosis.html
https://www.youtube.com/watch?v=hFZFjoX2cGg
Building the Perfect Squirrel Proof Bird Feeder (Failed?)
Also some other animals.
https://www.cs.bham.ac.uk/research/projects/cogaff/movies/apa/videos.txt
https://www.youtube.com/watch?v=9jRSgZVhWvw
Hens and chicks
MURGI Hen Harvesting Eggs to Chicks new "BORN" Roosters and Hens Small Birds
https://www.youtube.com/watch?v=QPqcSKhtxKk
Ducklings around the lake. (4.24 starting to paddle)
https://www.youtube.com/watch?v=KBm698UoROs
Newly Hatched Ducklings [2008] --
All waiting for last eggs to hatch.
'Peak hype': why the driverless car revolution has stalled
https://www.youtube.com/watch?v=QPqcSKhtxKk
PYTHAGORAS wikipedia
https://en.wikipedia.org/wiki/Pythagorean_theorem
The above is a small sample of references relevant to this document. More will be added later.
Maintained by
Aaron Sloman
School of Computer Science
The University of Birmingham