Life - F+ S

Chapter 6: What is life?

Hidden Life:
Even things we think arent't live could be (eg. time lapse photography on ocean floor showed 'rocks' moving, waving their arms etc.)
Tardigrades can be dormant, and frozen and still alive (most? every? other earthlife would die in this case
Viruses can remain in powdery state until in a host cell

Depeending on how we define life, we may neglect other living systems because they dont have the properties which we define as necessary (ie. proteins and nuclic acid) and to disregard the environoments in which they occur, to have other forms of life.

More study is needed to see whether life needs components of earthlife ( proteins, nucleic acid) (even on earth)

I quite like evolution as a definition of life. I think it certainly works for homobiota, though it remains to be seen if it can work for extra-terrestrial life. I find it difficult to imagine that there is any life in the universe that does not develop between generations, or has not had to develop to be where it is now, but maybe that is a failure of my imagination informed by my prior knowledge and love of genetics.

F & S define life as the activity of the biosphere. Life as a verb. Catalytic cycles that are dependent on one another through living organisms.

When life is defined as a biosphere, it changes how we can look for other life. We don’t look for individual “alive” things but instead look for a living biosphere. Evidence might be in an atmosphere that isn’t in equilibrium eg. has oxygen and methane at the same time like on earth.

Difference between living and non-living: distinction based on quantitative differences rather than 'special qualities,'; differ in the extent to which they carry out activities (creation and preservation of order) - living things create and order moreso than non-living things


I don't know what to think about Feinberg and Shapiro's claim that we should define life in terms of biospheres. It's probably best to try to decide this only after we've thought about a variety of definitions. I fall back on my standard answer, which is that how we define life is arbitrary and depends solely on what definitions are useful to our purposes; life does not have an ontology. (Edouard Machery agrees with this though - see the paper by him in the optional reading.)

Before I read this chapter; thoughts on viruses (not "virii", because if you try to make plurals conform to the grammar of the original language then what about words borrowed from Japanese or Xhosa? (I mean, is this a problem, really? English already has number of inconsistent pluralisation rules inherited from donor languages; what's one more? I suppose the problem is in intelligibility.) versus prions. Viruses invade foreign cells to make copies of themselves. Can't replicate without a host, but that in itself is not uncommon. They turn a host cell into a virus factory and produce many viruses; there is a mechanism for mutation and adaptation

Prions are harder, because they don't make more prions; they turn existing things into prions - they're zombies. They can't increase the number of prions except in a one for one process. They're an emergent property based on the number of prions in a system, and the number of things that could become prions in a system. I can't see an obvious analogue for mutation, and hence adaptation. There is a possibility that prions could develop a faster way to copy and this way would be selected for and become dominant. (I agree. It's hard to think about, or observe, because prions exist as a particular shape, so their mechanism for mutating would be by sampling different shapes. Shape sampling happens on a femtosecond timescale, so we only ever see the end product where they find the best shape(s) for their particular environment.

This looks more like a self organising system, biasing the local polypeptides towards particular spatial arrangements. ON THE OTHER HAND, they clearly replicate; not being able to directly make more than one copy of yourself does not disqualify a thing from being alive, and a replication strategy being a bit bad (prions eventually run out of stuff to turn into prions) doesn't disqualify one from life, otherwise we wouldn't count as life if we run out of food to eat.

Some people claim that a thing can't count as alive if all it does is turn other things into copies of itself, as opposed to growing first. This makes growth sound more central to the definition of life than it should be. What will we say if we find hugely complicated aliens that say hello to us and display all the other attributes of life, but instead of growing they can only assemble themselves full-grown? (This kind of thought experiment is the master test for definitions of life. We should probably give it a name.)

We'll come back to all this in more detail later when we do Dawkins.

Summary of our oral discussion around this:

Ada notes that virologists sometimes define viruses as an organism with a two stage life cycle; virion and infected cell. Believes this definition arose in response to the definition of life in terms of cells, and the ways in which this definition problematises viruses.

Jason raises the issue of wasps that infect plant tissue, is the plant tissue part of the wasp? Implies clearly not.

Ada disagrees; Especially in cases where the wasp larva exerts control over the host tissue to provide nutrients or even to grow a canker, one can absolutely consider the larva and surrounding canker as one organism. I think this problem relates to how we define the boundaries of an organism, specifically as to which cells count as one particular organism, which in some cases could be unclear. Like, the cells have plant origin, but are being coopted and controlled by the wasp. How is this different to a virus co-opting and controlling the machinery of a bacterial cell I think there is a fuzzy boundary where those cells are both wasp and plant.

Consider questions like: Are our mitochondria distinct from ourselves? They have their own DNA that is not part of our nuclear genome. What about our gut bacteria? They are symbiotes, but we could consider them part of ourselves. What about bacterial infections?

Jason says that if you have multiple infections you would be multiple organisms, suggests this is a problem. Ada thinks that's fine, your body is a collection of multiple organisms with fuzzy boundaries where you and your parasites interact. On the other hand, at this extreme our definition of life has become a bit useless.

Ada thinks this goes back to the fundamental problem that life doesn't have an ontology, hence our definitions are arbitrary and based on our needs, and any strict definition is going to give you weird edge cases that don't make sense because life is an emergent property not a real thing.

Jason & Esther note case of a photocopier, which has both replication and mutation. Jason says Dawkins gets into this more; notes that Dawkins resolves the question in terms of scale.

Finally, as an aside, I want to point out that taxonomists constantly bicker over ontology of taxa, and while some of the discourse recognises that this is arbitrary some of it seems to argue as though taxa are real.

There's a chapter about whether species are real in the philosophy of biology textbook "Sex And Death" (Sterelney and Griffiths), and it has all these really good arguments, and towards the end of the chapter I thought, "Those are good arguments. They've shown that species are not real." And then I read the last paragraph of the chapter, which says something like, "And so, as you can see, we've shown that species are real."

CHAPTER 7- Conditions for Life

Restates conditions:

7.1. Free energy sources for life

Need a source and a sink, a place for energy to come from, and a place for degraded, less availible energy to go. (Which is an illustration of my point that they're not really talking about energy at all, but about entropy or something like that. Even if we need a simpler concept than entropy, we should be talking about energy gradients, not energy itself. I find F&S surprisingly sloppy on this point.)

Heat is an obvious energy sink, as are lower energy chemical compounds (eg: through the oxidation of high energy compounds like carbohydrates to lower energy compounds like water and CO2)
In the case of the earth, space is a heat sink.

Examples of energy sources
Energy flows can be stored for later use; they use the example of sunlight in an ear of corn.

They note the energy must be accessible to the matter (chemicals / energy/whatever) composing our life.
Earthlife likes light and chemical energy, as these are easy for it to use.
X-rays are not as accessible to earthlife, no good, too high energy. But other systems could use them.
But a life form that stored order through the capture and loss of tightly bound electrons in heavy atoms might find X-rays very useful as a source of free energy, since X-rays are effective at detaching tightly bound electrons from atoms.

Energy source and order storing mechanism may be well-coupled, otherwise need even more time to evolve.

If the energy absorbtion process is not directly linked to storing order process, can still be harvested through a series of steps. EG: If energy is too large, maybe store it in smaller packages, or if too small, build it up. This seems obvious and also it is what happens in processes like photosynthesis. They note this is more sophisticated than just using the primary energy and so we assume it must arise later.

(Do we tho? Such a system could emerge in the pre-replicator phase)

They note useful energy sources occur at interfaces.

7.1.1 How much energy?

The rate at which free energy must be supplied will always depend on how much order is involved and on how quickly the order would decay in the absence of an energy supply.

In general, we can think of this in terms of the expense to create order and the expense to maintain order. If free energy comes in packets (eg: photons) we can quantify this directly.

7.1.2 Noteworthy energy flows

While X-rays and UV are smaller in number, they are sufficiently high energy to play silly buggers with electrons (forming positive ions) and chemistry, which makes them interesting. This can make it hard to form molecules because the high energy photons break them up, but it's not a problem if you have enough concentrated matter so that the external matter shields the rest (eg: the ozone layer). We can imagine this happening in dust clouds.

The average atom in space will be ionized by a cosmic-ray particle only about once in three billion years, or just a few times during the life of the Universe.

But that's an average, and there are places they may be interesting. Eg: galactic dust clouds, concentrating of cosmic rays via gravitation or magnetism eg in van allen belt. And now that we know that planets very close to stars are common, and also thinking about the density of stars in the centres of galaxies. Solar wind is a huge stream of charged particles.

Stored energy flows could also serve as a basis for life, tho these might eventually be depleted depending on whether and how quickly they are replenished. Relevant especially to the stored heat in gas giants not close to stars, which is my personal favourite source of an entropy gradient.

They say that extracting energy from stored flows more expensive because you have to move from place to place. I don't know if I agree, on small scales the energy flow of a geothermal vent could be more than sufficient for a very long time, although in the long term life might need to be able to move to other vents due to eventual flux. Depends how much is stored, and a planet stores a LOT of heat.

Eg: temperature differences within the earth, fueled by radioactive decay. Ocean surface warmer than ocean floor bc sunlight.
To use temperature, a thing needs to be able to interact with both the hot and cold places, either by moving between them or by spanning both of them (like a steam engine!).

They term life that can extract energy from a thermal gradient "thermophages".

They mostly talk about hydrocarbons and carbohyrdrates here, but there are many other sources (eg: metal ions and sulphur compounds can be oxidised).

As we have stated, human beings are on the verge of using fusion energy industrially
Ha ha ha ha ha ha ha!

They say fusion is not well matched to nucleic acid mechanism for storing order. It's interesting that they only discuss fusion as an energy source, and not radioactive decay. Is this because radioactive decay is often so slow? Life that operated on a much slower timescale might not mind sitting wrapped around a pile of heavy isotopes slowly eating radioactive decay (and anyway, some things decay considerably faster, like potassium). Are they excluding radioactive decay because they've discussed charged particles earlier? If so I wish they would mention that.

*Most useful form of stored energy is one that is replenished by an energy flow. They make the example of sunlight captured by photosynthesis, but note energy need not be captured by an external process. EG: high energy radiation creating a stock of high energy molecules.

They consider directly harvesting the energy of atoms at an excited state, rather than chemical compounds, and note that this depends on life managing to harvest a sufficient quantity of excited atoms. I note that this is exactly how photosynthesis works, just that life holds its atoms and waits for them to get excited rather than going around colelcting them. I suppose that's not stored energy tho.

There are problems with directly collecting energy from a store of excited atoms because excitation relaxes on a very quick timescale, so you need to catch them in the very short moment where they are excited, but maybe some very fast life would be fine with that.

F&S conclude:

We can summarize the energy prospects for life in the Universe as follows: In almost all the regions where there is sufficient matter to make life imaginable, there is enough free energy in one form or another to act as an ordering agent for life processes.

They really do seem to be biased towards matter-based life.

7.2. Mechanisms for storing order

They state that a supply of free energy is not sufficient for life, there must also be an appropriate form of matter to interact with it. I again object to their use of matter specifically, but there does need to be a something that can interact with the energy. It can't interact with itself? Why not? Why not indeed, energy could be the something.

They state that to produce order, first there must be non-equilibirum states that can be reached with the availible energy (ie: to climb up the order ladder out of equilibrium, you need to be able to reach the rungs). The example used in earthlife is stepwise assembly of more ordered chemicals.

They note that the nonequilibrium situations that are produced must exist for sufficient time to allow order to accumulate; they can't immediately decay. Here they talk about reaction rates needing to be fast but not too fast, but I feel weird about this because the reaction rate for assembly and decay may be distinct; what you need is for the rate of assembly reactions (which has a component of the reaction rate, and a component relating to the available energy), to be greater than that of the reaction rate of the decay (which is mostly a factor of the reaction rate of decay, but if the decay process also requires energy (as is likely) may depend on the availible energy)

They note that environmental disruptions must not be sufficient to wreck the nascent order.

They note that life need not use the most abundant elements in a particular environment, because it needs to be able to do chemistry which means it needs to be sufficiently unstable that you can do things with it. There's a lot of silicates on the earth's surface, but they are quite stable under these conditions so the natural can't do things with it. (We can do lots of interesting things with it under artificial conditions, but one does not find weird naturally occuring integrated circuits on the earth's surface as far as I know)

They raise the possibility of change in temperature influencing the viability of reaction cycles, although they don't seem to get to the point discussed (for example) in Ball and Brindley that temperature fluctuations could be part of reaction cycles, which is now a very common idea. The Polymerase Chain Reaction, which one could see as a sort of platonic ideal of the notion that temperature cycles can form part of a reaction cycle (it's certainly the most commonly used such thing by far), didn't really get noticed until 1983, which is after the book was published so that's forgivable.

They note that the temperature, elements, and media characteristic of earthlife should not be seen as essential, and indeed we keep finding earthlife that pushes the boundary of these conditions.

They note that you need your elements used to assemble order in sufficient quantities, but that being said some of the ones we use are essentailly trace elements in the environments we inhabit, so the boundary on that is presumably quite low. Like, perhaps a palladium based life might have some problems here, but on the other hand a lot of earthlife needs selenium for catalysis and that is really very uncommon.

They note that a selective concentration mechanism might create an environment enriched in weird rare heavy elements, which is great if you want lanthanide based life, or even just life based on more mundane heavy elements.

I wonder why earthlife is so big. Greg Egan (of course) has a novel in which there are organisms about 10^-30 metres long. If Greg Egan can do it, why can't the real world? Egan has to invent new physics to make it work, but since our current physics at that scale is broken maybe it's not so bad that he does that.

They discuss weirder ways to store order.

exists for natural lasers in certain interstellar regions, where it is thought that microwave radiation acts on molecules to keep a large fraction of them in excited states (p206)

They note that this constant excitation might work better in low density regions, where atoms are not shielded by other nearby atoms.

They note the existence of weird forms of matter incompatible with chemistry, such as plasma, or neutron stars. Note that hardly any of the matter in the universe can support chemistry - it's almost all much too hot (and therefore plasma = fully ionised) or much too cold.

7.2.3 Amount of order.

In general, they argue, we can come up with a way to store order in any disordered environment of which we can conceive.

They suggest that a significant fraction of orderable things need to be ordered to support life, as otherwise the system is too easilly disrupted by random environmental flux.

They suggest that a larger number of ordered things allows for more diverse ordering, with different ordered subunits performing different functions. This makes more ordered things useful, but doesn't seem to make it necessary? There would be a minimum number of different types of order based on the number of processes required to maintain order, but one can conceive of a solitary replicator that does it all on its own, so I suppose that minimum number could be "one".

They consider the question of a minimum size of a biosphere. The consider the disordering effect of the environment outside the biosphere, and note that the minimum viable size may be related to the proportion of biosphere to non-biosphere. It could also be seen as a surface area to volume problem, as only the parts of the biosphere that interact with the nonbiosphere are degraded, and if the interior of the biosphere (however we define interior) is sufficient to replenish the boundaries, it's fine. On the other hand, they note that free energy needs to be obtained in sufficient quantities, which poses another constraint on the surface area to volume relationsip.

In this discussion, I am using surface area to refer to the degree of interaction between the biosphere and the nonbiosphere, and area to refer to the total amount of biosphere. These need not be defined spatially.

7.3. Is there enough time for life? I often wonder.

Need sufficient time for order to accumulate before our energy source and orderable medium are disrupted.

They say that insufficient time may be due to a process being too very slow, or to drastic change in the environment. I think these are two sides of the same coin; a process is too slow if it isn't fast enough to happen in the timeframe alloted to it.

They note the possibility of an environment that undergoes rapid cyclic change, such as the massive temperature flux between day and night on the moon. They think it would be hard for an organism to survive in this condition, but to be honest the first thing that occurs to me is that it could serve as a temperature gradient. Soak up heat in the day, release it during the night. And it could use something inorganic as a heat shield and heat sink.

They consider the difficulty of quantiphying the rate of evolution in earthlife. developments in Phylogenetics have helped here, but that only really allows us to make gueses about the rate of genetic mutation and change in DNA based cellular life under modern conditions, and gets increasingly speculative the further back you go. I don't think it's useful for coming up with an age of earthlife because the early stage is too fuzzy, and it only takes you back to the LUCA anyway. I guess it could give you a very messy lower bound.

They note the timescale of evolution is likely to be linked to the timescale of the physical processes of a form of life, which allows for emergence of life on radiacally different timescales to what we think is reasonable for earthlife. Constrained by rate at which energy supplied, and rate of ordering processes. They then examine bottlenecks on the physical processes of life

7.3.1. Chemical life bottlenecks

Very general point: we still don't know much about how chemistry works at temperatures and pressures much different from the surface of the Earth. (Agreed, although we are getting better and better at this) We do know -
* Cold: Chemistry slows down in the cold, and so pretty much doesn't happen close to absolute zero.
* Heat: Chemistry doesn't work at all (by definition, even) when it's hot enough for all the electrons to be stripped off all the molecules ... almost all (interestingly, not quite all) stars are hot enough that they can't have any chemistry for this reason. They have interesting processes involving elements, but those elements are bare nuclei, not whole atoms or molecules.
* Pressure: Recently we've started learning just a bit about chemistry at high pressures, and it's totes different from chemistry at our pressure, which is awesome if you want there to be possibilities for chemical life that we haven't thought of yet. Most planetary environments (if you weight them for size) have huge huge high-pressure regions.

Chemicals need to be able to interact to do chemistry. They note that reactions in solid phase are much slower than gas or liquid phase, bc reagents are more mobile and hence interact with a greater number of potential reaction partners, in a greater number of spatial configurations. Similarly, liquid phase reactions are often faster than gas phase because the reagents much more concentrated and therefore more likely to bump into each other (though this depends on the pressure). Another consideration is that chemistry generally happens faster at higher temperatures, and for a given medium and pressure the gas phase is hotter than the liquid, and the liquid is hotter than the solid. Hotter media means higher energy molecules means more chemistry.

They also note that reactions occuring in liquids can have solvent effects, where the medium in which the reaction takes place participates directly or indirectly in the reaction. For example, water molecules or dissolved salt ions can interact with positive and negative charges in a reagent, reaction intermediate or transition state, stabilising the reaction and making it more likely to happen. Reactions between positive and negative reagents tend to happen faster than those between neutral reagents, and faster still than those between reagents with the same charge. They note that it can be very hard to predict this kind of thing.

They note that dense gases at high temperature might make an excellent medium for exotic chemistry, which would mean you don't need to have liquid solvents for life.

I think you could have life in the solid phase, too, but its chemistry would be much slower. There are many conceptual examples of immobilised life that exists, for example, only at the edge of where it is growing, leaving its dead descendants behind. This is basically how a tree trunk gets wider, (mostly only the surface is alive, much of the tissue deep in the trunk is long since dead) so it's not such an exotic idea. This is basically what Wang's Carpets are, iirc. Or even better think of a crystal growing inside a solid rock or metal. This has no fluid at all.

They note that slow chemical processes can be sped up by catalysis and/or intermediate steps. Catalysis is particularly interesting as some catalysts can be very selective, which biases your system towards a particular order. For example, all of our proteins have particular stereochemistry (I shall say they're all left handed, because lefties deserve more representation), and those proteins that catalyse reactions related to polypeptides and other proteins are necessarily usually biased towards left handed reagents and products, forming a self perpetuating selective system. Many simpler catalysts (such as metal ions) can also be quite selective.

They note that at different temperatures we might expect different chemistry or timescales, but note that it's not linear. Chemistry happens faster at higher temperatures, but you also get different chemistry at higher temperatures as new reactions become availible. Remember I drew the diagram with the hill you have to get over in a chemical reaction? At higher temperatures, you can climb different hills, and you get different results.

They use the example of free radicals, which react very quickly at earth temperatures but would be more stable at lower temperatures.

Harold Urey has remarked that "... the apparently unusual conditions under which chemistry occurs in space are really the ordinary conditions for chemical phenomena in nature." p214 ;
Jason this relates to what you said before about what I called weird chemistry not actually being weird.

They note that chemical cycles must be balanced, the products of cycle X must be able to be used by the next cycle before they decompose. This is particularly true at the early stages where life is still developing, when the cycles haven't been fine tuned by selective processes.

One possible solution is to have each reaction produce a huge excess of product, so cycle x produces far more products than cycle y could ever consume, and there is always copious x availible. This kind of reaction will quickly reach an equilibrium, as cycle x has produced so much x that it can't rarely encounters starting materials to make more. If x decomposes back into starting materials, eventually there will be a steady state at which x is made as fast as it decomposes, so there is a constant quantity of x in the environment. Exotic density chemistry

They discuss the fact that much of the matter in the universe is at a much lower density than what we consider normal, quoting a figure of 1 atom per cubic centimeter in interstellar and intergalactic space (is this still accurate?) (Yes, although it varies a lot - there are interesting regions in interstellar space and even in intergalactic space, so you can pretty much take your pick of a density and find an utterly enormous region of the universe that has that density.) At this density, they argue, it would take millions of years for a bacteria sized object to interact with enough stuff to double in size. They say this is not a promising timescale for a chemical life form, but I don't know if I agree; while the chemistry would happen slower, there are other factors; for example you can use less stable chemistry because you don't have to worry as much about bumping into things that will wreck you, and your environment may be less subject to disruption from things like an asteroid wrecking your atmosphere.
And if you're only trying to make any life, not the most complex life, then there's been plenty of time. Also, because you are interacting with less stuff maybe you don't need to be as big because you might not need to encapsulate yourself if you can stick together in other ways.
Ooh, good point.

They note that this density is variable, some denser clouds would have interactions at rate 1/y, however most of the stuff you're going to hit is hydrogen and helium, and those two in isolation at low pressure are maybe not so useful for chemistry.

They note that while higher density regions can exist, they often collaps quite quickly into stars. Table 7-4 gives some maximum limits on size and density before a cloud would collapse. I wonder how dense the media is near the center of the galaxy, but I suspect the maximum density before collapse constraint still holds.

Under high density, they say that the main bottleneck is the reduced mobility of potential reagents, particularly if the medium is solid. They suggest that one alternative would be to change the internal qualities of atoms rather than engage in chemical reactions.

They also note that high pressure media need not be solid, high pressure media remains liquid at sufficient temperatures. In such a situation reaction rates could be quite high (lots of collisions and high temperature), and might arise more quickly than it does on earth.

They note that this is important because some high pressure environments might not have energy for very long; a neutron star might only have free energy for something of the order of 10^3 years. If the physical processes are fast enough, even a system like this could develop life.

While we can understand that the availability of sufficient time is important to the development of life, F&S argue that it is difficult to quantify. What is "sufficient time" in a given medium? We aren't sure how long it took to develop the first replicators on Earth.