## Friday, January 27, 2017 ... //

### Entropy, freedom, life's purpose, animals' desires, consciousness, and vitalism

First, a real advance in biology. A large team in Japan has developed the Manbearpig, a half-man, half-bear, half-pig (OK, so far without the bear) from pluripotent stem cells. They were obviously not satisfied with the fact that the human himself is already a half-chimp, half-pig. It may be a good source of human organs – but this usage will correspond to morally problematic half-murders or half-abortions, too.

Yesterday I was amused by the article

How Life (and Death) Spring From Disorder
by Philip Ball in the Quanta Magazine. My understanding is that Ball is a British science writer with no research experience so this essay is really what a bright journalist, and not a scientist, thinks but I found it inspiring, often thinking that there must be something terribly deep about it.

However, after some efforts, I always return to my normal thoughts which indicate that this depth is mirage – much like the motion of the photograph above that surely isn't blinking (something must be wrong with your monitor or brain!) – and the excitement is mostly poetry. What's going on?

All of us know how life works. It knows how to reproduce itself (clone DNA, produce proteins according to a DNA code etc.), collect energy from the environment, and other manifestations of life. And then the forms of life are fighting for survival. Natural selection picks the most viable ones in the dynamically changing environment and those reproduce, mutate, and produce additional variations that may perhaps reproduce and survive even more effectively or reliably etc.

The detailed microarchitecture of life on Earth – involving RNA, DNA, and proteins and certain other omnipresent molecules and processes – is shared by everything we have ever considered "life" but much of it seems like a bunch of technicalities, anyway. Ball is basically defining life in some more general way.

Before he mentions a paper by Alex Wissner-Gross (whom I've known well) and Cameron Freer – which also seems cryptic but I discussed the paper three years ago – Ball makes lots of very exciting observations or suggestions, e.g.
Death is the ultimate equilibrium. Life is the effort to escape this equilibrium.
I think it's a very cool slogan. It's similar to a proverb saying that you should live vigorously because you'll have enough peace in the grave. (I am unable to figure out what is the exact form of this proverb. The closest thing I could find with the help of Google was the statement that The peace of the grave is the only peace Muslims want with Israelis occupying the grave.) But when I try to turn it into something like an additional law of physics, I am ultimately led to the belief that the cute slogan above is just a metaphor. In a minute, I will tell you the most general reason why I am skeptical about the "physical truth" of any similar statement.

But before we get there, I must mention that Ball goes much further. He also says that
Living forms need to encourage the out-of-equilibrium processes in their environment to survive, and therefore to seek some correlations that allow them to mine the energy or entropy or negative entropy from the environment. To do so, they must basically become very good prediction machines but they must predict the future effectively (not to spend much energy).
All these things are very interesting thoughts but it's often unclear whether he's really minimizing the energy or maximizing energy or minimizing the entropy or maximizing the entropy. It seems to me that none of these theories that Ball promotes makes much sense and he can't possibly be sure which of them he really wants. Well, at least I think that numerous sentences written by Ball about the entropy have the wrong sign. Also, I think that his insistence on the energy efficiency of the thinking process is largely unjustified and exaggerated. There is also the interesting observation that
Organisms want to maximize their distance from the equilibrium which is equivalent to maximizing their freedom in the future.
I had some deja vu and indeed, in 2013, I did notice that while discussing his "equation of intelligence", Alex Wissner-Gross was conflating the number of options with the entropy in a way that looked strange to me and it still does. In particular, if a life form is capable of avoiding the equilibrium (=death) in the future, it means that the future entropy is smaller than the maximum it could be: the maximum entropy is the entropy of the equilibrium, isn't it? Ball, Wissner-Gross, and probably others seem to constantly say the opposite – that organisms want to maximize the entropy quickly etc. – which seems upside down to me.

But maybe I am missing something and/or these bugs could be fixed. But I don't think so.

Finally, we're getting to the general reason why I am skeptical about the very general opinion that similar thoughts produce "alternative laws of Nature" that govern the living structures. What is the reason? Well, I think that all these thoughts are basically revivals of vitalism which Wikipedia defines as:
Vitalism is a discredited scientific hypothesis that "living organisms are fundamentally different from non-living entities because they contain some non-physical element or are governed by different principles than are inanimate things". Where vitalism explicitly invokes a vital principle, that element is often referred to as the "vital spark", "energy" or "élan vital", which some equate with the soul.
Ball, Wissner-Gross, Freer etc. may talk about entropy or freedom instead of vitality but aside from this modified terminology, the basic spirit seems to be exactly the same. Most importantly, it still suffers from the basic assumption that life contains something fundamentally different than inanimate objects in Nature.

I find it very attractive - and it's very tempting to improve all these vitalist hypothesese and claim that the vitality/soul/entropy... should actually be called "consciousness" and their quantity measuring the freedom in the future actually measures the amount of consciousness. Maybe I could make modest changes so that it would sound nearly equivalent to Tononi's Integrated Information Theory of consciousness. ;-) But the problem is that life ultimately co-exists with no life and we may observe both. And when this system containing both obeys some laws according to an external observer such as ourselves, these laws simply cannot care whether something is life or not.

Tononi's ideas don't break down: Tononi is just saying that he may quantify consciousness and when life dies away, the consciousness just drops close to zero or exactly to zero, too. But Ball et al. do something else than Tononi: They want to say that some laws hold for the living structures. It's strange because death must mean that the laws cease to hold. And because "life is weak" in some situations, it should also mean that the laws optimized for life must be "mostly broken".

So Ball and others may give us wonderful definitions boasting to clarify "what is life" and "how much life there is somewhere" and "what life is trying to maximize". However, despite all these wonderful attributes and dreams of life forms, life isn't guaranteed to survive and win. Even if life prolongs the time towards the equilibrium, dreams about maximizing its freedom, or predicts the future while spending just a little bit of energy, it has an Achilles' heel. What is life's most universal Achilles' heel? Well, the main weakness of life is that it may die, go away, and be replaced by death.

While you may have invented lots of great alternative laws about the quantities optimized by life and the purpose of it, these laws sometimes break down because the life forms – plants, animals, people, and others – sometimes die. Whatever they were doing ceases to hold once they're dead. Death is an unfortunate career move and when it becomes the next mission in someone's life, all the laws about the life's wonderful optimization get invalidated.

They may probably be invalidated well before the life forms die. Ball's whole article is full of vitalism of more conventional forms, too.

You know, one of the genuine processes that are needed for life on Earth is the conversion of the highly concentrated energy coming from the Sun in the form of the visible light to the not so concentrated thermal leaving the Earth in the form of the infrared photons. (Life can locally live without the Sun, of course: I am just saying where the energy ultimately comes from in all realistic setups we know on Earth.)

Recall that the most likely energy of a photon in thermal radiation at temperature $T$ scales like $E\sim kT$. The photons coming from the Sun are emitted from the solar surface whose temperature is around 6,000 kelvins. The temperature of the surface of the Earth is some 300 kelvins instead. The ratio is 20 – and it's also the ratio of the number of typical incoming and outgoing photons to/from the Earth.

(Note that the solar radius is about 2.3 light seconds while the Earth-Sun distance is 500 seconds. The ratio of these two is 217 or so and it's no accident that it's of the same order as 20 squared. I hope that you know how to write this equation, even a more accurate version of it. You will need the fact that the radiation flux scales like $T^4/r^2$ with the absolute temperature and the distance.)

OK, to conserve the energy, the incoming solar photon may give rise to 20 outgoing terrestrial infrared photons. Each photon has approximately the same entropy of order "one bit", so the entropy carried by the photons increases 20-fold which is more than enough to allow some decrease of the entropy of the life forms that live "in the middle" of the gadget that converts the solar visible photons to the terrestrial infrared ones. The total entropy will go up even if the organisms manage to reduce their entropy.

Note that the egg is very regular or organized or symmetric or non-uniform which means that it has a lower entropy than the same matter at the same temperature in a more messy state (e.g. the mixed egg). It sounds wonderful and vitalist that organisms are capable of achieving this task – to reduce their own entropy while increasing the entropy of the environment.

But when you decide not to forget about science and think rationally, you will see that there's really nothing mysterious or "characteristically life-like" or "vitalist" about this reduction of entropy. A fridge is doing the same thing. You may produce a simple fridge as a gadget in which some body of gas expands at place A and shrinks at place B to the original volume. This simple heat engine will cool the place A and heat up the place B. The place A may be identified with the "head of a very smart organism" while B is the "external world" except that by construction, there is nothing mysterious about the simple heat engine or fridge.

Organisms are doing much more complicated things but the goal of reducing the entropy or temperature of A while increasing the entropy or temperature of B is purely physical, fully understood, and doesn't need any special deep insights.

My broader point is that the known laws of physics – I mean string theory (ST) but feel free to approximate it by the Standard Model and general relativity if you suffer from a psychological problem with string theory – describe all phenomena we observe, whether or not we would describe the given situation as containing life or not. These laws of physics don't distinguish life from non-life which is why they also can't assign a purpose to life.

The most general feature of Ball's approach – and that of similar folks – is that there also exist some alternative laws of Nature that say that there are sometimes some living agents somewhere and they try to maximize one thing or another but the formula for this thing is universal. Let me call the best possible laws of this type modern vitalism (MV).

Now the question is whether ST and MV are equivalent and whether they may hold at the same moment. I think that they just can't be equivalent because MV has nothing to say about some dull situations that seem to contain no life, no "deliberate" efforts to survive, no "conscious" predictions of the future etc. while ST says a lot about these things.

Also, I have some trouble imagining that ST and MV hold at the same moment. ST fully – albeit probabilistically – determines what happens with any physical system, whether it's living or not. So some inequivalent conditions are unlikely to be obeyed at the same moment.

Even though I may sometimes find myself spiritually thrilled about some additional maximization laws of physics that are centered around the organisms or living agents (and could therefore serve as a "very different dual description" of Nature), I ultimately think that all laws for life must be just effective laws deduced from ST, i.e. from conventional laws of physics. And one of the general properties of the ensemble of effective theories is that they heavily depend on the environment – there is no one omnipotent effective theory (the only unique theory is ST, the fundamental theory).

So all the efforts to say that in any living system, one quantity or another is always maximized, must be wrong simply because that claim would be equivalent to the claim that the number of effective theories is one, or to the statement that "everything is life" and death is impossible. For this reason, I think that all the proposed deep entropy-based etc. alternative laws of Nature describing everything about life must be impossible to formulate quantitatively and all their shared, precisely formulated properties have a limited range of validity. I would bet that they will always be either vague or a system of inequivalent laws whose details must be adjusted according to the context or the species.

By the way, I am annoyed that Google Chrome's spellchecker underlines all appearances of the word "inequivalent" as if it were a typo. The word may sound archaic to some folks but it is completely correct and popular among some very deep thinkers. ;-) The word is constructed just like an "inequality" which is unassailable so I would always defend "inequivalence" as if I were a double-tailed lion.