PDA

View Full Version : Chaos theory and the 'Life-outcome'



Paul Wally
2013-Jan-13, 07:29 PM
This post comes from a discussion with Selfsim in the thread Alien genetics (http://cosmoquest.org/forum/showthread.php/141099-Alien-genetics?p=2097911#post2097911).

The question is:
Is the life-outcome a single random outcome of chaotic dynamics on the primordial Earth? In other words if the clock is turned back to the time just before life emerged on Earth, was the emergence of life to be expected from the conditions that prevailed during that time, or was it simply a single improbable random outcome of chaos.

My view is that randomness produces the variety for self-organization to happen, which means that given a sufficiently large 'organic soup' ocean, the chances for emergence of the kind of complexity that we call life are drastically increased. If I understand correctly, Selfsim takes the opposite view, and that is: The sensitivity to initial conditions inherent in chaotic dynamics means that the 'life-outcome' is in the same way sensitive to initial conditions, and therefore by implication, highly unlikely.

Paul Wally
2013-Jan-13, 09:49 PM
A planet can be modeled as an open system, with energy in and energy flowing out. Beyond the Earth's atmosphere there is the vacuum of space, so there exists a natural boundary between the Earth and the rest of the universe. There is nothing artificial about the concept of a planet.



Nonetheless, the model comes from the laws of thermodynamics because Earth is obviously isolated, (at everyday macro scales), when viewed from a thermodynamics perspective.

What do you mean by an isolated system? The Earth is obviously not isolated.

Selfsim
2013-Jan-13, 09:52 PM
I'm pretty sure a popularity 'poll' is not going to resolve your question on this topic.

For the record, I do not take 'the opposite view', (as claimed).

Sensitivity to initial conditions is a fundamental part of Chaos Theory phenomena, and has plenty of empirical evidence supporting it, (from dynamic, organic chemical systems). If life's origins were formed in such non-linear systems, then it would have been, (by definition), sensitive in some way, at some scale, to the precise initial conditions. The outcome of subsequent repeats of that same process elsewhere, at those same scales, would not necessarily have been, (or will be), predictable.

This is about modelling. We can choose different models in order to study different aspects of a system. If a particular class of well-known phenomenon resembles closely, the characteristics of what we're interested in, then it cannot be 'ruled out' from the 'possibility' space. It is in fact, equiprobable with any other model chosen along the same lines, and no 'decisions' about 'rightness' or 'wrongness' are necessary.

Life's origins are 'unknown'.
'Unknown' has consequences, and consistency requires acceptance of the consequences of the paradigm, (or model), which has been chosen on the basis of being consistent with observations of the phenomena, or any of its successor(s).

That is my 'view'.
Its as simple as that.

TooMany
2013-Jan-13, 10:11 PM
I think Wally's point about self-organization shots down the idea that chaos prevails and outcomes are completely unpredictable. Take a cylindrical box half full of ping-pong balls and jiggle it for a while with the axis vertical, then look at the organization of the balls. I haven't tried this but I expect hexagonal close packing would occur. The jiggling process and motions of the balls are quite chaotic, but the result is not. The physics of the situation leads to a predictable state, despite the randomness of the process. The balls tend to find the most stable/lowest energy arrangement. This is just an example, not intended to be directly applied to life chemistry.

Processes that lead to improvements in accuracy of reproduction are selected and thus eventually come to be quite sophisticated (the genetic code). At this point we don't know if quite different types of organic (or inorganic) life are possible or not. E.g. life that uses something unlike DNA to store replication information. However, we do know that this path worked here on Earth so there is every reason to suppose that under similar conditions, the same self organization can occur.

That does not mean that in the much more complicated case of organic life that we expect that exact same forms to emerge from chaos. We do not and for good reason. The randomness of the process results in unpredictable order of adaptations and all of life interacts in this process. So in that sense evolution is somewhat chaotic. On the other hand what is selected depends on environment, what is possible physically and what is useful in competition for survival which thus leads to predictable developments (such as eyes, wings, ears, teeth).

I hate to say this because I'm not a believer in a plan, but it's almost like nature has rigged chemistry to promote life as we know it. I'm not a biologist but I can see that the chemistry is just incredible.

Selfsim
2013-Jan-13, 10:32 PM
What do you mean by an isolated system? The Earth is obviously not isolated.We're out of synch here ….

The choice of model is dependent on the perspective of study. From my experiences in thermodynamics discussions, all differences seem to be ultimately resolved by consideration of the immediately surrounding environment, (which is entirely dependent on how big or small one chooses to make the 'scope', or 'scales', of study).

For the purposes of examining Earth-life's chemical origins, I would choose to view the origins as emerging from a pre-biotic chemical 'soup', subject to the immediately surrounding planetary (only) environment (including temperatures, pressures, etc). 'Opening' such a system to external influences, I think, only complicates things, somewhat unnecessarily (which is what science always attempts to avoid, I might add … sometimes to the exclusion of influences which may or may not make a difference to the outcomes). A good baseline starting point would be to agree that all necessary chemical molecules leading to 'life', are already present at 'the start'.

Now, would that 'soup' which has also accumulated on some other similar planetary environment, elsewhere in the universe, lead to the same generalised results (as defined by our hypothesised 'universal' model of life)? - If I were to consider the hypothesis that life could start elsewhere in the universe, I might choose to adopt a different 'system' definition .. eg: by admitting Laws known to be necessary, and by positing their ubiquity, and also, a yet to be defined 'abiogenesis' process which is common to both my Earth-model and to the exo-Earth-like environment. The approach then calls for admission of the consequences of those particular 'choices' … for consistency purposes.

The key differences between the respective emergences, (on Earth and on non-Earth), would then be subject to non-linear behaviours .. (and all that ultimately entails). These behaviours may have no particular significance, if our study focus is singularly the Earth abiogenesis instance.

Selfsim
2013-Jan-13, 11:14 PM
I think Wally's point about self-organization shots down the idea that chaos prevails and outcomes are completely unpredictable. Take a cylindrical box half full of ping-pong balls and jiggle it for a while with the axis vertical, then look at the organization of the balls. I haven't tried this but I expect hexagonal close packing would occur. The jiggling process and motions of the balls are quite chaotic, but the result is not. The physics of the situation leads to a predictable state, despite the randomness of the process. The balls tend to find the most stable/lowest energy arrangement. This is just an example, not intended to be directly applied to life chemistry.What happens when the vibration frequency (or vibration orientation) changes slightly?
How selective are the precise conditions which result in any patterns at all (given the totality of the potential selection space? Can we quantify the relationships and derived a probability estimate of the prevalance of these? What is the basis for any assumptions in coming up with these estimates ..?.. (Ie: is it that we can establish a past history of the phenomena, (from prior multiple cause/effect trials)? If so, does the analogy help us when considering the emergence of life elsewhere .. (in what we all acknowledge, will be a different environment)?


Processes that lead to improvements in accuracy of reproduction are selected and thus eventually come to be quite sophisticated (the genetic code). At this point we don't know if quite different types of organic (or inorganic) life are possible or not. E.g. life that uses something unlike DNA to store replication information. However, we do know that this path worked here on Earth so there is every reason to suppose that under similar conditions, the same self organization can occur.The more I think about where your 'competition for resources, (leading towards particular optimised form)', argument is coming from, the more I think that its ultimate basis lies in the more fundamental Laws of Thermodynamics.

Ie: The system explores all possible ways to reduce conflict for resources. This situation is consistent with the 2nd Law, since in such systems, (dissipative ones), the gradients 'encourage' the system to self-organize to an ordered state, since this actually increases the rate of entropy production and thus, system 'stress reduction'. It can be shown that the greater the energy flows in such systems, the greater the order, (and information), generated, becomes. Some of this 'order', is employed by living organisms to do work, in creating, (temporarily), higher-level 'material' structures. The set of selected states of those 'structures', perhaps being those which maximise entropy production ..

I have no problems in adopting this particular view .. its a choice, y'know .. and that choice is dependent on what I'm keen to snoop out, at any particular time of my choosing. :)


That does not mean that in the much more complicated case of organic life that we expect that exact same forms to emerge from chaos. We do not and for good reason. The randomness of the process results in unpredictable order of adaptations and all of life interacts in this process. So in that sense evolution is somewhat chaotic. On the other hand what is selected depends on environment, what is possible physically and what is useful in competition for survival which thus leads to predictable developments (such as eyes, wings, ears, teeth).I think the examples your cite are overly specific, and exclude considerations of other equally plausible outcomes, which are easily permissible, (under the principles I cite above). For instance, some selected states, (or structures), which maximise entropy, may not be sufficient to result in a set of functions which we choose to have mean: 'intelligence', (via our definitions for it). So far, intelligence on Earth is integrally related to 'a brain' ('eyes', 'wings', 'ears', 'teeth', etc). The two concepts are separable in an abstracted model and once again, there is choice involved. I think you are making those choices .. but there are others to make (given the open-ness of the possibility space).


I hate to say this because I'm not a believer in a plan, but it's almost like nature has rigged chemistry to promote life as we know it. I'm not a biologist but I can see that the chemistry is just incredible.Well, you may take some solace by adopting the Anthropic Principle here … ie: that we notice what you observe, because we are here, (and adapted, (or optimised), to notice exactly that). In other words, its Ok to say what you say .. and it is even 'expected' .. but that isn't the only thing to expect …
:)

TooMany
2013-Jan-14, 12:31 AM
What happens when the vibration frequency (or vibration orientation) changes slightly?


Same thing happens over a large range of frequency and amplitude, but there are limits. If you shake very slowly, the balls will not move. If you shake too violently they will start falling out of the box. No it is not highly sensitive to conditions and that is precisely the point of self-organizing phenomena, they can be robust.



... does the analogy help us when considering the emergence of life elsewhere .. (in what we all acknowledge, will be a different environment)?


Different how? I thought we've been taking about Earth-like conditions, not planets with frozen or boiling oceans.


The more I think about where your 'competition for resources, (leading towards particular optimised form)', argument is coming from, the more I think that its ultimate basis lies in the more fundamental Laws of Thermodynamics.

Ie: The system explores all possible ways to reduce conflict for resources. This situation is consistent with the 2nd Law, since in such systems, (dissipative ones), the gradients 'encourage' the system to self-organize to an ordered state, since this actually increases the rate of entropy production and thus, system 'stress reduction'. It can be shown that the greater the energy flows in such systems, the greater the order, (and information), generated, becomes. Some of this 'order', is employed by living organisms to do work, in creating, (temporarily), higher-level 'material' structures. The set of selected states of those 'structures', perhaps being those which maximise entropy production ..


You may well have something there. I'm not sure if that alone is enough. Consider the complexity of organic chemistry. That sort of complexity may be required for life to emerge, regardless of thermodynamic considerations.



I think the examples your cite are overly specific, and exclude considerations of other equally plausible outcomes, which are easily permissible, (under the principles I cite above). For instance, some selected states, (or structures), which maximise entropy, may not be sufficient to result in a set of functions which we choose to have mean: 'intelligence', (via our definitions for it). So far, intelligence on Earth is integrally related to 'a brain' ('eyes', 'wings', 'ears', 'teeth', etc). The two concepts are separable in an abstracted model and once again, there is choice involved. I think you are making those choices .. but there are others to make (given the open-ness of the possibility space).


I mentioned those because I think that have evolved independently multiple times, providing us with an data point.



Well, you may take some solace by adopting the Anthropic Principle here … ie: that we notice what you observe, because we are here, (and adapted, (or optimised), to notice exactly that). In other words, its Ok to say what you say .. and it is even 'expected' .. but that isn't the only thing to expect …
:)

I guess so. I'm not fond of the multiverse idea but it does allow me to have an explanation that does not involve an infinite regress of planers.

Selfsim
2013-Jan-14, 03:47 AM
Same thing happens over a large range of frequency and amplitude, but there are limits. If you shake very slowly, the balls will not move. If you shake too violently they will start falling out of the box. No it is not highly sensitive to conditions and that is precisely the point of self-organizing phenomena, they can be robust.Well, I wouldn't say: "that this is precisely the point of self-organizing phenomena, they can be robust."
They can also be not robust, which is another way of saying they can also be very sensitive to perturbations. (It depends on the nature of the system under study). For instance, 1 -10 nanograms of botulinum toxin ingested, inhaled or injected, is sufficient to bring 50% of any tested population of humans, to a screaming halt.

A brick balanced on an inclined board with a sandpaper surface, will never produce precisely the same movements over the time of its falling, and one cannot produce predictions about this. (One cannot also predict precisely, the inclination angle, needed to trigger the fall, either .. no matter how many times the experiment is carried out).

The unpredictability of avalanches typifies this same phenomenon, also.

The unpredictability of the weather, (beyond a week or two), is another .. the apparently same starting conditions don't result in the same weather phenomena each time we observe those conditions.

The list of analogies is virtually endless .. and they are all purely a matter of choices made .. (ie: 'horses for courses').


Different how? I thought we've been taking about Earth-like conditions, not planets with frozen or boiling oceans.Hmm .. I thought we were talking about variations of 'Earth-like' conditions .. if not, then why not? And even if we weren't talking about this .. our interest would surely be different, when considering pre-biotic Earth-like conditions too, wouldn't it?

The point of discussing in these analogies, is to distinguish the large variety of everyday phemomena, which are perfectly capable of exhibiting chaotic, and unpredictable behaviours, under slightly different conditions … so why should we expect life emergence to be exclusive from those, or unusual (or 'special'), in any way, from what effectively constitutes the majority of natural system behaviours? Especially as the biology which it seems to leaves us with, exhibits these very same characteristics?

You may well have something there. I'm not sure if that alone is enough. Consider the complexity of organic chemistry. That sort of complexity may be required for life to emerge, regardless of thermodynamic considerations.Well, excluding Thermodynamics Law, would certainly be 'ATM', and I see no pressing need for going there …
See, the mechanistic explanations you've put forward, I think, represent a move in the opposite direction, ie: away from attempting to define laws of behaviours at a more universal, (ubiquitously applicable), level … which is where I think a universal law of life sits … not down at a specific level of:
"This is how it worked on Earth, therefore … "
… (y'know…(??))

I agree that complexity, in terms of the shear numbers of constraints 'shaping' already interdepending variables changes the fundamental picture … but we still don't have sufficient 'beyond Earth' experiences, at the levels of detail, (scales), necessary, to create a universal law. That's what science needs returned from 'the search for exo-life' .. ie: more data, in order to develop that missing law.


I mentioned those because I think that have evolved independently multiple times, providing us with an data point.Sure .. and I don't think I've disagreed with the examples drawn from Earth's heritage .. its just that when we look to these as a way of taking us more towards a more generalised, universal phenomenon, we need more generality, which then forces us into having to acknowledge other things, which broaden our understanding of that 'possibility space'. This also helps in taking us away from the more parochial perspective of an Earth-centric focus, (with which we're all particularly familiar), y'know(??)

Selfsim
2013-Jan-14, 04:16 AM
Same thing happens over a large range of frequency and amplitude, but there are limits. If you shake very slowly, the balls will not move. If you shake too violently they will start falling out of the box. No it is not highly sensitive to conditions and that is precisely the point of self-organizing phenomena, they can be robust.The 'tuning' requirements of a non-linear system can be modelled along the basis of inputs, outputs and system transfer functions. In this way, the responses of the system to perturbations, can be characterised and generalised. In the case of the organic chemical systems I mentioned (ie: the Belousov-Zhabotinski autocatalytic reactions), they can be set up to exhibit underdamped, overdamped, critically damped, or oscillating responses to perturbations. (The analogy is an oscillator circuit in a radio).

The point of interest here, is that such systems can put themselves into any or all of these states, at any time, particularly if they are left to evolve in a 'natural' way. (Radio oscillators however, are specifically engineered to behave in typically, only one of these ways, unless some component behaviour is made deliberately variable).

The different damping responses represent generalised behaviours achievable from the exact same system, so there are typical behaviours which might be predictable, provided sufficient prior information is known about the characteristics, before they start up. If those prior specifics are unknown, then any of these behaviours is possible, and expected from such systems.

One steady-state behaviour which is known in advance however, is that the initial starting conditions are critical to predicting their future behaviours.

Paul Wally
2013-Jan-14, 10:03 AM
The choice of model is dependent on the perspective of study. From my experiences in thermodynamics discussions, all differences seem to be ultimately resolved by consideration of the immediately surrounding environment, (which is entirely dependent on how big or small one chooses to make the 'scope', or 'scales', of study).

For the purposes of examining Earth-life's chemical origins, I would choose to view the origins as emerging from a pre-biotic chemical 'soup', subject to the immediately surrounding planetary (only) environment (including temperatures, pressures, etc). 'Opening' such a system to external influences, I think, only complicates things, somewhat unnecessarily (which is what science always attempts to avoid, I might add … sometimes to the exclusion of influences which may or may not make a difference to the outcomes). A good baseline starting point would be to agree that all necessary chemical molecules leading to 'life', are already present at 'the start'.



There is one important external influence that we cannot leave out of the model, and that is the sun. Life on Earth is inextricably linked to radiant energy entering the system from the sun. Because the Earth is spinning, this also leads to a day-night cycle, and as the tilted Earth revolves around the sun there are also seasonal cycles. Life processes have adapted and synchronized with these natural rhythms to form biorhythms. This is just to show how important external influences are to life on Earth. The Earth simply is not an isolated system.

Paul Wally
2013-Jan-14, 02:40 PM
Well, I wouldn't say: "that this is precisely the point of self-organizing phenomena, they can be robust."
They can also be not robust, which is another way of saying they can also be very sensitive to perturbations. (It depends on the nature of the system under study). For instance, 1 -10 nanograms of botulinum toxin ingested, inhaled or injected, is sufficient to bring 50% of any tested population of humans, to a screaming halt.



I think the robustness of which TooMany speaks is that of the existence of life. Once life got started on Earth it continued for 3.5 Billion plus years, in spite of all the perturbations like meteorite impacts, multiple occasions of radical climate change, other natural disasters etc.

Allow me to clarify what I mean by 'life-outcome', because I think you're getting stuck on the notion Earth-like life again. A 'life-outcome' means any life whatsoever. So the OP question is then: "Is the life-outcome a single random outcome of chaotic dynamics on the primordial Earth? ". If it's a single random outcome then that means that there are no other possible life-outcomes.

Selfsim
2013-Jan-14, 10:22 PM
I think the robustness of which TooMany speaks is that of the existence of life. Once life got started on Earth it continued for 3.5 Billion plus years, in spite of all the perturbations like meteorite impacts, multiple occasions of radical climate change, other natural disasters etc.I did a post on this very topic (Dec 2011). (http://cosmoquest.org/forum/showthread.php/135808-EXPOSE-E(d)-life-in-space!?p=2036949#post2036949)

The point, (at the time), was that there is well considered evidence in support of biological systems being poised at, or near, a critical point, and that it is this state, which enables them to exploit energy gradients induced by perturbative transients.

In non-equilibirum systems poised at criticality, it is the transients that dominate the actual behaviours .. the 'steady-state' becomes irrelevant.

The Mora etal paper, in the linked thread, goes into much deeper detail). The models discussed in it, once again, are more generalised than examples taken from specifically, Earth's biological instance, (which may or may not be 'typical' in the overall scheme of things).

If we're attempting to generalise about what influences life elsewhere, we need to focus on the generalised behaviours of analogous and diverse systems, and their observed responses to varied environments throughout the Universe, (ie: beyond Earth's particular flavour of past environment(s)).


Allow me to clarify what I mean by 'life-outcome', because I think you're getting stuck on the notion Earth-like life again. A 'life-outcome' means any life whatsoever. So the OP question is then: "Is the life-outcome a single random outcome of chaotic dynamics on the primordial Earth? ". If it's a single random outcome then that means that there are no other possible life-outcomes.Well, its a slow process, but I'm heading in the same general direction, (given that its pretty clear that your question will find no answers … other than pure opinions).

There is a cause/effect relationship between abiogenesis and biology (by definition). We cannot reconstruct with sufficient precision, the conditions under which life emerged on Earth, so as to address the sensitivity issue.
We can figure out generalised system behaviours not only from biological systems, but analogous systems in nature and notice the similarities. (This is in fact, a way of getting away from the 'Earth-centric' bias …).

As mentioned previously, the phenomena addressed in Chaos (and Complex Systems) Theory, are not the same as those described by pure randomness models. Those are Stochastic Systems, and are the subject of Statistical Mechanics.

Paul Wally
2013-Jan-15, 09:38 AM
I did a post on this very topic (Dec 2011). (http://cosmoquest.org/forum/showthread.php/135808-EXPOSE-E(d)-life-in-space!?p=2036949#post2036949)

The point, (at the time), was that there is well considered evidence in support of biological systems being poised at, or near, a critical point, and that it is this state, which enables them to exploit energy gradients induced by perturbative transients.

In non-equilibirum systems poised at criticality, it is the transients that dominate the actual behaviours .. the 'steady-state' becomes irrelevant.

The Mora etal paper, in the linked thread, goes into much deeper detail). The models discussed in it, once again, are more generalised than examples taken from specifically, Earth's biological instance, (which may or may not be 'typical' in the overall scheme of things).

If we're attempting to generalise about what influences life elsewhere, we need to focus on the generalised behaviours of analogous and diverse systems, and their observed responses to varied environments throughout the Universe, (ie: beyond Earth's particular flavour of past environment(s)).

While I'm sure there are various mechanisms and principles involved in biological complexity, I think the stability of living systems is best understood in terms of negative feedback, as is found in homeostasis. It is this adaptive nature of life that explains why life has existed for so long and continues to exist on this planet. But how does the highly complex behavior of present day biological systems relate to how the much simpler original proto-life got started? Logically, 'biological systems' did not exist prior to abiogenesis.


Well, its a slow process, but I'm heading in the same general direction, (given that its pretty clear that your question will find no answers … other than pure opinions).

It's not just about answering the question, but also about clarifying the problem. More in particular, I'm trying to understand what it would mean for abiogenesis to be sensitive to initial conditions. If someone tells me that abiogenesis could be sensitive to initial conditions, I would like to know what that means. Surely if something could be the case, then it must make sense that it could be the case.



There is a cause/effect relationship between abiogenesis and biology (by definition). We cannot reconstruct with sufficient precision, the conditions under which life emerged on Earth, so as to address the sensitivity issue.

That assumes that there is a 'sensitivity issue', which is what I think this discussion is (supposed to be) all about, i.e. whether there actually is a sensitivity issue as far as abiogenesis is concerned.



We can figure out generalised system behaviours not only from biological systems, but analogous systems in nature and notice the similarities. (This is in fact, a way of getting away from the 'Earth-centric' bias …).

One way to cure your 'Earth-centric bias' is to define life in a simple universal way, e.g. define life as self-replicating and evolvable units. The problem of abiogenesis becomes then: How do self-replicating evolvable units emerge through natural process. The chaos theory problem becomes: Is the emergence of self-replicating and evolvable units sensitive to initial conditions in the primordial ocean?