Great question. Indeed I think that there is more to life than just being another mode in a non-equilibrium process such as a convection pattern in a Rayleigh Benard experiment. Nailing down those differences may indeed require a definition of life we don't have. I just through in one intuitive thought. A non-equilibrium relaxation process in a non-living system may be completely stopped for an arbitrary long time and then activated again, e.g. via temperature control. I would argue that it is not possible to completely bring life to a hold (I am not talking about slowing it only dramatically down like a frozen frog or a waterbear) for an arbitrary long period of time and then starting it again as if this interruption never happened using only a single or a few stimulating parameters. However, on writing this, I can already tell how I am getting increasingly vague.

Hi Christian. Are you saying that one difference between living and non-living non-equilibrium processes is that living processes manage to survive even after their external potential that they are dissipating is removed? Like when night arrives and living systems survive until the morning? If this is your point, I'm not sure that this does not happen in non-living processes. For example, if we could take the temperature gradient away for some instance of time from a Benard cell, I'm sure that the cell would survive if the instance were short enough because the cell has some angular momentum which would take some time to dissipate (natural relaxation time). Maybe we could say that life has a greater natural relaxation time than most non-living irreversible processes, but we would have to look at more examples to be sure?

Actually my point was precisely the opposite. What I tried to say is something like this: If something living is dead for an extended period of time you can't just flip a switch and continue life where it stopped. In the case of the Rayleigh Benard experiment you can do just this. Switch off the temperature gradient and the convection mode is gone. Go away for a few years, come back, and switch the temperature gradient on and the convection mode is there again as if nothing happened. In contrast, the living entity seems to need virtually permanent exposure to an energy flow allowing to keep entropy low otherwise it changes irreversibly into something of different quality which we would call irreversibly dead. I am not saying it is not possible with technological advances to shift the life/dead boundary. However, the further you shift it, the more effort it takes with a non-linear growth until it gets completely impractical. In the case of the convection mode it is just one parameter to bring the mode back independent on how long it was absent "dead".

Hi Christian and Karo, for many years I have been researching the questions you discuss about the definition of life. My point of view was that of system science. Instead of giving a long answer, I would like to invite you to take a look at some of the following resources where you may find iinnovative answers in relation to your questions. A detailed discussion in three steps of how to arrive at a definition of life (which I needed as a basis for defining biodiversity) can be found at Pages 27, 87 and 94 of my book "The pursuit of complexity. The utility of biodiversity from an evolutionary perspective". Another source is my paper (in the list of researchgate) about a hierarchical definition of life. Also interesting for you may be the short text on my website at http://the-operator-theory.wikispaces.com/Definition+of+life%2C+the+organism%2C+and+death

In addition you may like to read the intersting historical review of seminal works about the definition of life in the book: The Nature of Life: Classical and Contemporary Perspectives from Philosophy and Science by Mark A. Bedau, Carol E. Cleland

Kind regards, dr, mult. Gerard Jagers op Akkerhuis

Ok Christian, now I understand what you are getting at. Let me raise two points related to this. Life, as we know it today, has had a long history of evolutionary changes, and, by all accounts, increases in complexity. A Bernard cell is comparatively simple and doesn't seem to evolve. The difference, according to me, is that life dissipates very high energy photons (with respect to those emitted by the black body spectrum of earth at 300k). With respect to flows and forces then, life is a very non-linear process compared to the Benard cell. Therefore, there are many stationary states available to life, with greater or lesser entropy production, and therefore life can evolve through bifurcations through these different states (greater entropy production being generally selected). After 3.8 billion years we have a very complex life which has taken a particular path through many bifurcations. If we were to now take away the external photon potential, i.e. the sun and wait long enough for all life on earth to extinguish (except perhaps at hydrothermal vents) and then bring back the sun, then we would confront ourselves with two problems i) maybe life would never start up again because its origin may have depended exquisitely on earths initial conditions (for example, intense UV radiation and high surface temperature), and ii) it would be extremely improbable that life would pass through exactly the same bifurcations during its new evolution. Therefore, we should not expect to "bring back the living from the dead" so to speak.

Benard cells have only a few different stationary states available to them; hot fluid coming up, or going down in their center, plus the different directions of rotation. We also know that we can start Benard cells with todays given conditions on the surface of the earth. Therefore, it should not be surprising that we have a probability to bring Benard cells back from the dead. I emphasize that it is only a probability, not a forgone conclusion, since even Benard cells have a number of different stationary states available to them. Regards, Karo.

Sounds good to me. We may also want to keep in mind that we still lack the instrumentation to really describe what is going on. For example, all terms we borrow from thermodynamics and in particular entropy are strictly speaking defined for equilibrium. (Shannon entropy e,g, can be generalized but ...). The power of equilibrium thermodynamics is the possibility to describe something that is microscopically extremely complex with just a few variables. This powerful tool weakens substantially in non-equilibrium where additional variables are needed. For something as complex and extreme non-equilibrium as life we should not be surprised that those global terms such as energy and entropy are no longer appropriate to get the complete picture. They still remain useful when allowing for flows, but it is far away from the full story.

Yes, you are right Christian, entropy, traditionally defined is only valid for equilibrium situations. However there are some valid approximations that can be made, such as "local equilibrium" in which, even though the system as a whole is out-of-equilibrium, small, but still macroscopic, regions can be considered in equilibrium and all the usual thermodynamic variables retain their meaning and even the Gibbs equation is satisfied locally, but now the variables taken on position dependent values throughout the system. On the local equilibrium hypothesis is based the whole of Prigogines "Thermodynamics of Irreversible Processes". It is surprisingly valid for most situations on earth. It only becomes invalid for very rarefied gases and extremely fast chemical reactions, but we always have to keep this in mind when we talk about entropy and entropy production.

…. But virus may crystallize and then remain in a state of equilibrium.

“A non-equilibrium … process in … (this living) system may be completely stopped for an (arbitrary) long time and then activated again”.

This may be why some people argue that a virus qualifies as a living entity but is rather a large molecule.

Virus seems to be something between living and nonliving, and in this opinion, the evolution from nonliving to living may be a continuous process, there is no an either-or boundary between the two.

I agree, but that doesn't help too much. Almost nothing in reality is without some fuzziness. That even holds for phase transitions where true discontinuities or singularities only appear in the thermodynamic limit of infinite systems. That doesn't stop us from clearly distinguishing between phases even in finite systems (all systems are finite) and I think it is meaningful to do so.

A virus alone is certainly not an irreversible thermodynamic process dissipating a generalized thermodynamic potential. However, a virus may interact, or interfere, wth such a process, for example a living cell, and change somewhat the nature of the dissipation. But what could be gained from a thermodynamic point of view through this interaction?

I guess we arrived at this point because we were struggling (and we are in good company with this struggle) to come up with a decent intuitive understanding of what in essence life is. This questions seems to be entangled with your original question. One may argue that one has to know/define life to proceed and utilize this insight to advance thermodynamics, but yes I agree we lost somewhat focus.

Viruses and disease help spread nutrients around the globe for disspating plants and cyanobacteria. Maybe that is the thermodynamic function that they perform?

Yes or perhaps one can say it is why the laws of physics and here in particular thermodynamics allows that they evolved. Probably other scenarios are also compatible with the fundamental laws. They don't need viruses I would assume. But yes overall life is there because it is an extremely efficient way to increase the entropy of the universe to quote Clausius. Nature "pays the price" to create low entropy entities or living beings such as humans, but look how good we are at making a complete mess which means increasing entropy. If one would just quantify the entropy increase by burning all these fossil fuels from trees over coal to oil and gas and our conversion of low entropy food into high entropy waste, I would guess that we are a very efficient mode to increase entropy.

Right Chrstian, but it seems that we humans still only produce directly a small fraction of one percent of the total entropy production of the biosphere. Surely, however, this fraction is increasing exponentially and one day there maybe a kind of phase change and the biosphere will be superceded by the humansphere. It is probable that the stars of highly advanced extraterrestrial civilizations will be invisible because the habitants are dissipating to the far infrared all the light of their star. There seems to be two relevant aspects to entropy production in nature, 1) rate, 2) stability (duration in time). Therefore, maybe the biosphere could never be superceded by the humansphere. Our belief in our apparent stability maybe only an illusion, the "complete mess that you talk about".

High Cristian and Karo, I have the impression that your discussion about entropy production shows a close association with the work of Chaissons about increase in the rate of entropy production per kg material, when comparing entities early in the universe with those in the older universe. The topic of maximum entropy production links closely to the work of Volterra, Swenson, Schneider, Salthe, Kleidon, Bejan, Anilla and others. Similar topics are discussed in The Pursuit of Complexity, where entropy production is related to the acquisition of degrees of freedom (both chaotic degrees and organised degrees). The above question of whether a virus is living or non living is also discussed there. To decide whether a virus is life, it is essential to have acces to a definition of life. Of course there are scientists which say 'science has no definition of life'. But this opinion may be considered a bit fatalistic. I have offered a link to a fundamental approach to the definition of life (a few mails ago), but it seems that this information has not yet played a role in this discussion so far? I look forward to your response.

Hi Gerard,

Sorry for the delay in replying. It wasn't due to lack of interest but rather to a lack of understanding on my part of what exactly you are getting at with your "operator heirarchy". I have read some of your papers. That there are many "levels" and coupling between these levels in life, I completely agree with you. That there is an emergence or evolution of complexity or coupling in living systems, I again completely agree with you. You state that as long as the "level-defining" organization is present, then "that is life", and that this definition avoids some of the problems that other definitions suffer. However, your definition of life appears to be just another description, not a theory, even though, as you state, more encompassing than other definition. It dosen't seem to help me understand life, for example, it would be important to know, for example, where life is headed and how life began. Maybe you could explain us in more detail your view, perhaps I am missing something?.

A description of life in terms of non-equilibrium thermodynamics tells me much more since it is based on fundamental laws of nature. For example, I suggested that life originated to dissipate UV photons of around 260nm. Both RNA and DNA dissipate very strongly around this wavelength. I also suggested that life has incorporated more and more pigments over the eons to dissipate more and more of the solar spectrum, and that life has proliferated to cover the entire surface of the earth and in this way dissipate still more photons. Please have a look at my papers on ResearchGate, "Thermodynamic Dissipation Theory for the Origin of Life" and "Biological Catalysis of the Hydrological Cycle". In this manner, life has a "reason" for existence (to dissipate the solar photon potential) just as hurricanes have a "reason" for existence (to dissipate the temperature gradient between the surface of the oceans and the upper atmosphere).

It is always a challenge to define very common things in the world, including the life. It might be the reasons we need some basic self-evident truths. Carbon world or the organic world ( C, H, O, N, S, P ..and trace metals Fe, Zn, Mo,,,) is recongized for the life in the current science concept. Protein and DNA ( or RNA) are two type of basic moleculars. Which come first become one of the key questions of life evolutions.

Recently my work on the catalysis of phosphate ester hydrolysis in the aged inorganic iron solution indicated that the inorganic nano-particles of iron oxide have the same functions of enzymein the process of the organic phosphorus transformation, which is essential for the supply of all life ( e.g. ATP). This observation demonstrates that ‘‘the chain of life is of necessity a continuous one, from the mineral at one end to the most complicated organism at the other’’, as proposed by Leduc (The Mechanism Of Life William Heinemann, Londan,1911). The hydrolysed iron solutions (nano-particles of iron oxide) might be just one of ubiquitous sets of undiscovered inorganic biocatalysts (enzymes) in nature. Such inorganic enzymes might act as a bridge between the inorganic and organic worlds and would have played important roles in the origin of life. Discovery of further inorganic enzymes might provide clues on the emergence of life and a potential solution to the puzzle of ‘‘chicken and egg’’ in life’s evolution. Please read my paper"Hydrolysis of Glucose-6-phosphate in Aged, Acid-Forced Hydrolysed Nanomolar Inorganic Iron Solutions — An Inorganic Biocatalyst?" for the details.

RSC Advance, 2:199-208. doi: 10.1039/c1ra00353d (http://pubs.rsc.org/en/content/articlelanding/2012/ra/c1ra00353d)

Hi Xiaolan, thanks for the interesting paper. We are looking for an inorganic substitute for polymerase. Something with Mg++ Do you have a candidate? Please have a look at my paper "Thermodynamic dissipation theory for the origin of life" in Earth Systems Dynamics available on ResearchGate.

Dear Karo,

I like to answer you by starting with a short citation of your above mail: "..., your definition of life appears to be just another description, not a theory, even though, as you state, more encompassing than other definitions. It doesn't seem to help me understand life, for example, it would be important to know, for example, where life is headed and how life began. Maybe you could explain us in more detail your view, perhaps I am missing something?"

First: It is not always easy to separate definition and theory. The reason is that a definition generally is linked to a theory. In this case, I use the operator theory, which shows how by means of subsequent 'closures' low complexity operators construct high complexity operators. This may seem a 'description' but at the same time, every closure step represents a thermodynamically favourable process that leads to a change in system configuration. Process and form go hand in hand.

Second: you ask how life began and where it is heading.These are not easy questions ;-) Interestingly I would think that an aswer to this question requires an a priori definition of life. Only if we know how we define life, can we tell when a certain phenomenon fits to the criteria, and where we can observe it.

As criteria for the 'start' of life, the operator theory suggests the two closures of the cell, autocatalysis and a membrane, and these should maintain eachother. If the dynamics and form of a system comply with these criteria I would consider the system the simplest form of a living being. But the simplest form of life does not yet offer a definition of life in a general form.

The reason is that higher complexity forms of living beings can be found on higer steps on the complexity ladder of the operator hierarchy, for example edosymbiont cells, multicellulars, etc.

Each time, the step towards any next higher level would be possible only because it is thermodynamically favorable. And the closures actually offer the mechanism how this can be done.

The result is that one can use the ladder of the operators, to define all the operators from the (bacterial) cell and higher as organisms.

With respect to a definition of life, one may now define life as the concept (an abstraction) which indicates a group property, namely the fact that all organisms posses functional and structural closure. If an organism looses these closures it is no longer an organism and does no longer represent life. This definition makes use of the concept of organism, but it is not circular, because the organism concept is defined by using the ladder of the operator hierarchy.

Finally, you ask, where life is heading. Well, life as such is not heading anywher ;-). For a prediction I would take the operator ladder as a start. Then I would extrapolate it towards future operator. And because I defined as organisms all the operators with a complexity which is at least of the level of the cell, these future operators would represent living beings. Using this methodology, my prediction would be that life is heading towards technical neural network organisms (e.g. 'robots').

I wonder how you look at this.

Kind regards, Gerard Jagers op Akkerhuis