PDA

View Full Version : Pre-life evolution, ecological niches, complex systems



Colin Robinson
2013-Oct-07, 12:17 AM
from the thread "Did life here, start out there?"


This conversation has gone way off-topic, (which is why I haven't thus far responded), but it also raises many interesting discussion areas. We really need to start another dedicated thread, (methinks) ..



As Pross mentions, the process does involve certain laws: e.g. you don't find two species in exactly the same ecological niche.Did he say that? (Reference please).

OK, here's the reference for you:

From Section 2.1.1. "Natural selection at the chemical level" in Addy Pross' paper Towards a general theory of evolution (http://www.jsystchem.com/content/2/1/1#) in the Journal of Systems Chemistry:


An even more striking expression of natural selection at the chemical level that further highlights the extent of the chemistry-biology nexus has recently been reported by Voytek and Joyce [21]. A key ecological principle, the competitive exclusion principle [22] states: "Complete competitors cannot coexist", or in its more positive expression: "Ecological differentiation is the necessary condition for coexistence". That principle informs us that two non-interbreeding populations that occupy precisely the same ecological niche (i.e., both competing for the same resource) cannot coexist - one will drive the other into extinction. The striking aspect of Voytek and Joyce's study was that it demonstrated that the roots of this quintessential biological principle can be found in chemistry. They reported that two RNA enzymes, when allowed to replicate and evolve in the presence of an essential substrate, were unable to coexist. One of the enzymes drove the other into extinction in line with the prediction of the competitive exclusion principle. More significantly, however, when the two enzymes were simultaneously reacted with five alternative substrates, the two enzymes were able to coexist. Each RNA enzyme evolved so as to optimize its utilization of one of the 5 substrates (a different substrate for each of the two enzymes) so that the system effectively mimicked biological niche behavior, again in accord with the exclusion principle. Darwin's classic finches niche behavior [23] at the chemical level!

A metaphor that comes to mind is the game of musical chairs. A particular species of finch, or a particular RNA enzyme, must either obtain a niche of its own to sit in, or else it will have to leave the game…


There are a couple of problems I have with Addy Pross' theory, which I will elaborate on later, but one aspect that I find problematic is his attempt at reducing biological evolution to chemistry. If I understood correctly, his argument is that we cannot reduce chemistry to biological evolution therefore biological evolution has to be reduced to chemistry.

Pross is working in the area of systems chemistry. That is, his work draws from the discipline that studies complex systems as well as from the discipline that studies chemical compounds and reactions.



The point is, evolution selects for combinations that work — those that do lead to survival. The sort of chemical evolution described by Addy Pross likewise selects for combinations that work. As Pross mentions, the process does involve certain laws: e.g. you don't find two species in exactly the same ecological niche.I don't know if that could be classified as a law. That has been called into question by some, but if it is true then I think it would be a mathematical theorem applicable to a certain class of systems, rather than a strictly chemical or biological or physical law.

Yes, the "competitive exclusion principle" (as it is called in the above quote) may well be due to the mathematics of complex systems rather than to principles of chemistry as such.

The point I want to stress is that the while systems of this class may be less deterministic than simpler systems, they do still seem to exhibit regularities...

Selfsim
2013-Oct-07, 10:33 AM
from the thread "Did life here, start out there?"Thanks for starting the new thread there Colin .. (we were getting way off track in the other thread. :)-default )


OK, here's the reference for you:
...
From Section 2.1.1. "Natural selection at the chemical level" in Addy Pross' paper Towards a general theory of evolution (http://www.jsystchem.com/content/2/1/1#) in the Journal of Systems Chemistry:
...
A metaphor that comes to mind is the game of musical chairs. A particular species of finch, or a particular RNA enzyme, must either obtain a niche of its own to sit in, or else it will have to leave the game…Just a recap here .. I was pointing out that uniqueness exists in natural systems (as does similarity). Science typically seeks out the similarities and dwells on the similarity aspects .. however, uniqueness still presents itself. Ignoring uniqueness, results in the false perception that accurate prediction of 'more of the same' is the default feature of science. As observation scales change, uniqueness can make reappearances however, which then result in similarity reappearing. The scales of observation are therefore crucial. (We dont know the appropriate scales for the recurrence of abiogenesis ...).

Another point (made by me), was that relaxation of certain laws in biology occurs more frequently than it typically does in say, Astrophysics, because the empirical evidence from biological systems frequently 'violates' canonical biological laws.

Ok ... thanks for the Pross quote .. he was drawing analogy from the principle of Ecological Niches from Evolution Theory. Your above point (quoted) is that Laws also underpin this principle, and presumably, (as per to Pross' analogy), these laws also 'work' in biological replicator systems, thereby enabling predictability to map across when it comes to biological replicator precursors to life(?) - (I hope I captured accurately, the relevant points in that summary?)

OK, so Ecological Niches is a working principle. The actual empirical law underpinning it is the law of Competitive Exclusion (or Gause's Law).



Yes, the "competitive exclusion principle" (as it is called in the above quote) may well be due to the mathematics of complex systems rather than to principles of chemistry as such. Yep ... and its based on the Lotka-Volterra models of competition (http://en.wikipedia.org/wiki/Competitive_exclusion_principle#Prediction).


The point I want to stress is that the while systems of this class may be less deterministic than simpler systems, they do still seem to exhibit regularities...And the point I'd like to make is that:
However, for poorly understood reasons, competitive exclusion is rarely observed in natural ecosystems, and many biological communities appear to violate Gause's Law. The best known example is the paradox of the plankton. All plankton species live on a very limited number of resources, primarily solar energy and minerals dissolved in the water. According to the competitive exclusion principle, only a small number of plankton species should be able to coexist on these resources. Nevertheless, large numbers of plankton species coexist within small regions of open sea.However ...
Some communities that uphold competitive exclusion are MacArthur's warblers and Darwin's finches.They go to explain what I was saying originally .. ie: that the paradoxes it raises, can be resolved by revising and redefining the original assumptions (which is along the same lines as my point about reviewing the scales of observation). The Competitive Exclusion law also requires relaxation, thereby allowing for reconsideration of the assumptions.

All of this is interesting ... but I'm more interested in presenting a different model for the abiogenesis process I recently unearthed. It seems to be a newly favoured approach along the lines of what Paul Wally was mentioning. Ie: a global rather than molecular-scaled approach ... I'll see if I can dig up the paper and post it in this thread. (I found it to be quite interesting food for thought ...)

Selfsim
2013-Oct-07, 08:54 PM
All of this is interesting ... but I'm more interested in presenting a different model for the abiogenesis process I recently unearthed. It seems to be a newly favoured approach along the lines of what Paul Wally was mentioning. Ie: a global rather than molecular-scaled approach ... I'll see if I can dig up the paper and post it in this thread. (I found it to be quite interesting food for thought ...)I decided to start a new thread on this topic, here. (http://cosmoquest.org/forum/showthread.php?146601-Global-Abiogenesis-Model&p=2161160#post2161160)

Paul Wally
2013-Oct-07, 09:18 PM
The point I want to stress is that the while systems of this class may be less deterministic than simpler systems, they do still seem to exhibit regularities...

Yes. Although, even if we could find a system with no regularities, then even in that case it doesn't mean there isn't a simple explanation for the irregularities. A law doesn't necessarily have to manifest in the form of a regularity, even a very irregular or even chaotic phenomenon could have a very simple law or principle that generates it. So, I don't quite understand what Selfsim's problem is.



Just a recap here .. I was pointing out that uniqueness exists in natural systems (as does similarity). Science typically seeks out the similarities and dwells on the similarity aspects .. however, uniqueness still presents itself. Ignoring uniqueness, results in the false perception that accurate prediction of 'more of the same' is the default feature of science. As observation scales change, uniqueness can make reappearances however, which then result in similarity reappearing. The scales of observation are therefore crucial. (We dont know the appropriate scales for the recurrence of abiogenesis ...).

I think you're grappling with the dichotomy of uniqueness vs similarity, and this is causing problems for you. For instance, what do you mean by "ignoring uniqueness"? How is it possible to ignore a unique phenomenon and how would we know that the phenomenon is unique if we haven't scanned the whole universe?



Another point (made by me), was that relaxation of certain laws in biology occurs more frequently than it typically does in say, Astrophysics, because the empirical evidence from biological systems frequently 'violates' canonical biological laws.


I don't know whether biology has any fundamental laws of it's own, i.e. not derived from physics, chemistry, mathematics or logic. Could you give an example of a "biological law" that was violated?



And the point I'd like to make is that:However ... They go to explain what I was saying originally .. ie: that the paradoxes it raises, can be resolved by revising and redefining the original assumptions (which is along the same lines as my point about reviewing the scales of observation). The Competitive Exclusion law also requires relaxation, thereby allowing for reconsideration of the assumptions.

"Reconsideration of assumptions" is what science is all about. Why do you think it is a "paradox"?



All of this is interesting ... but I'm more interested in presenting a different model for the abiogenesis process I recently unearthed. It seems to be a newly favoured approach along the lines of what Paul Wally was mentioning. Ie: a global rather than molecular-scaled approach ... I'll see if I can dig up the paper and post it in this thread. (I found it to be quite interesting food for thought ...)

Great, that should be interesting.

marsbug
2013-Oct-07, 09:45 PM
This makes me think of a minor revelation I had from my five year old nephew: While we were walking through the streets of his home town he asked me what the splodges on the pavement were. I told him that was where people had dropped chewing gum, to which he replied "Where do all the other dropped things go?". I explained that the wind blew them away, or people picked them up, or sweepers swept them up. As I was doing so it occurred to me that natural forces often select things to stay in, or go from, a particular environment whether they are living or not. Yet the phrase 'natural selection' is usually only applied to living things in the context of evolution, an unspoken assumption that might limit some peoples thinking (i.e. mine) when discussing abiogenesis.

Cougar
2013-Oct-08, 01:10 PM
"Laws of complexity spontaneously generate much of the order of the natural world. It is only then that selection comes into play, further molding and refining. Such veins... have not been entirely unknown, yet they are just beginning to emerge as powerful new clues to the origins and evolution of life." [Stuart Kauffman, At Home In The Universe]

Selfsim
2013-Oct-08, 08:07 PM
Yes. Although, even if we could find a system with no regularities, then even in that case it doesn't mean there isn't a simple explanation for the irregularities. A law doesn't necessarily have to manifest in the form of a regularity, even a very irregular or even chaotic phenomenon could have a very simple law or principle that generates it. So, I don't quite understand what Selfsim's problem is. I don't have any problems .. I'm quite Ok with uniqueness. Others rule it out as a possibility when it comes to evolving systems however. I have no idea why they do this.

Darwinian Evolution by Natural Selection is specifically one system which evolves through time into different systems (e.g. an aquatic form becomes land dwelling). This is an open-ended form of change. It encompasses the concept of non-cyclic change. It also entails unpredictability. However in the classical sciences, a fundamental requirement of a system model is that it is built to perform a specific function ... a concept which is not possible from an evolutionary standpoint, because an evolutionary outcome has no specific function (or purpose) to begin with.


I think you're grappling with the dichotomy of uniqueness vs similarity, and this is causing problems for you. For instance, what do you mean by "ignoring uniqueness"? How is it possible to ignore a unique phenomenon and how would we know that the phenomenon is unique if we haven't scanned the whole universe? That's exactly the point. Uniqueness in general exists, so it cannot be dismissed from consideration in the general case. Yet, we are unable to say definitively if something like abiogenesis is unique or not, if its definition entails having to search the entire observable universe to establish the fact. Yet the possibility of its uniqueness still exists (and thus cannot be dismissed in theory in the absence of falsifying empirical evidence). This is exactly the same dichotomy encountered when arguing that some 'theory' predicts other instances of abiogenesis (just in reverse).


I don't know whether biology has any fundamental laws of it's own, i.e. not derived from physics, chemistry, mathematics or logic. Could you give an example of a "biological law" that was violated? You should pay more attention Mr Wally! I gave an example of a non-canoncial edit performed by T.brucei. (http://cosmoquest.org/forum/showthread.php?145936-Did-life-here-start-out-there&p=2160789#post2160789) It does not follow the GU-AG rule for RNA splicing. (http://cosmoquest.org/forum/showthread.php?145936-Did-life-here-start-out-there&p=2160932#post2160932)

Colin Robinson
2013-Oct-08, 10:51 PM
Just a recap here .. I was pointing out that uniqueness exists in natural systems (as does similarity).

Please spell out what you mean by "uniqueness": specificity, or incomparability?

"Uniqueness" in the sense of specificity undoubtedly exists in nature. Every biological species has attributes which set it apart from other species. For that matter, every chemical compound has attributes which set it apart from other compounds.

However, a biological species may still be compared with other species, and a chemical compound may still be compared with other compounds. Thus, neither the biological species nor the chemical compound is "unique" in the sense of being incomparable.


Ignoring uniqueness, results in the false perception that accurate prediction of 'more of the same' is the default feature of science.

The models of abiogenesis developed by e.g. Stuart Kauffman and Addy Pross do not imply that life on other worlds will be "more of the same", in the sense of being a chemical duplicate of life on Earth. These models do imply that comparable processes on other worlds will lead to comparable results.


And the point I'd like to make is that:



However, for poorly understood reasons, competitive exclusion is rarely observed in natural ecosystems, and many biological communities appear to violate Gause's Law. The best known example is the paradox of the plankton. All plankton species live on a very limited number of resources, primarily solar energy and minerals dissolved in the water. According to the competitive exclusion principle, only a small number of plankton species should be able to coexist on these resources. Nevertheless, large numbers of plankton species coexist within small regions of open sea.

Yes, but WP also mentions (in its page Paradox of the plankton (https://en.wikipedia.org/wiki/Paradox_of_the_plankton) )


The paradox was originally described in 1961 by limnologist G. Evelyn Hutchinson, who proposed that the paradox could be resolved by factors such as vertical gradients of light or turbulence, symbiosis or commensalism, differential predation, or constantly changing environmental conditions.

Selfsim
2013-Oct-09, 07:22 AM
Please spell out what you mean by "uniqueness": specificity, or incomparability?

"Uniqueness" in the sense of specificity undoubtedly exists in nature. Every biological species has attributes which set it apart from other species. For that matter, every chemical compound has attributes which set it apart from other compounds.

However, a biological species may still be compared with other species, and a chemical compound may still be compared with other compounds. Thus, neither the biological species nor the chemical compound is "unique" in the sense of being incomparable. Well I don't think the definition of 'species' is a particularly good criteria for demonstrating similarity or uniqueness. There have always been major problems with classifications of species (See Species Problem (http://en.wikipedia.org/wiki/Species_problem)).

Coming back to the original point, if we have no other 'lifeforms' other than terrestrial .. then what does 'comparability' (or 'incomparability') mean? 'Specificity' seems to mean whatever we choose it to mean, based on the attributes chosen for the specification .. one goes looking for attributes which match the ones which were arbitrarily chosen in the first place. If we'd chosen the attributes which reveal dissimilarity amongst individuals, we'd conclude that each individual was its own unique species, no?

See, what I'm trying to say is that in the absence of something other than the terrestrial model, we can only stay at the generalised (or holistic) levels. As soon as one tries to 'break it down' to a lower level of detail, comparisons from the terrestrial model come into play, and we attempt to match patterns with some arbitrarily chosen attribute (or set of attributes) taken from what we already know - ie: the terrestrial model. As soon as one does that, all of the dependent variables associated with the terrestrial model come into play, and all those which are dissimilar from the terrestrial ones, are discarded. We'll always find similarity when we do that. What I'm trying to say is that system 'uniqueness' and 'similarity' can be generalised characteristics in their own right at a high level of abstraction, (or generalisation). The behaviours of the system, from that point onwards, can then become the more interesting aspects to study. Correlating 'uniqueness' and 'similarity' with those behaviours, then becomes more relevant. If one goes looking for 'comparability' between systems, one will always find some attribute which matches whatever one is looking for ... in the case of abiogenesis, we have no particular reasons to choose any particular attributes. I mean what if abiogenesis, which might have happened someplace else, resulted in some attribute which we weren't looking for .. would we even recognise that abiogenesis had happened there at all? If on the other hand, we'd taken a high level system behaviour .. like it reaching a state of dynamic kinetic stability, (for eg), we might actually find that the abiogenesis process might be capable of producing 'no-life' whatsoever .. but it does produce something else which could conceivably be unique. Perhaps we could then say that abiogenesis is capable of producing 'uniqueness', (ie: not only 'similarity'), whenever or where-ever, the process 'runs'. This is not possible if we go looking for things which are too specific, which automatically rule out the possible uniqueness of what we call 'life', before we've even gone exploring.

Does that make sense, yet?
(I personally doubt it'll ever make sense for some .. because of an ideologically based stance .. (Note: this is not meant as an insult, but this is where fixed ideological principles generally take thinking on such topics ...)).


The models of abiogenesis developed by e.g. Walter Stuart Kauffman and Addy Pross do not imply that life on other worlds will be "more of the same", in the sense of being a chemical duplicate of life on Earth. These models do imply that comparable processes on other worlds will lead to comparable results.Pross' DKS is far more generalised (has meaning at higher levels of generalisation, or abstraction) than Kauffman's autocatalytic sets. Autocatalytic sets may be completely irrelevant because they are very specific to say, the RNA first hypothesis, than DKS is to any generalised physically dynamic complex system.


Yes, but WP also mentions (in its page Paradox of the plankton (https://en.wikipedia.org/wiki/Paradox_of_the_plankton) )Once again ... "change the variables", (which were arbitrarily chosen in the first place, anyway), "... in order to seek the match". Or, "change the scales of observation of some variable to get the match" ... Similarity will always be found this way .. but uniqueness beyond the Earth, will never be detectable using this method. (In fact, it is disallowed before the search has begun .. and yet, we all know that uniqueness (as a stand-alone generalised, abstracted system characteristic) is real and exists, in physical systems everywhere in nature.

Paul Wally
2013-Oct-09, 10:25 AM
I don't have any problems .. I'm quite Ok with uniqueness. Others rule it out as a possibility when it comes to evolving systems however. I have no idea why they do this.

That you have no idea why "others rule out as a possibility", is entirely your problem. You are using unverifiable terms like "uniqueness" and then you wonder why 'others' supposedly "ignore" it. That's why I say it's your problem, because it is the result of your own scientifically meaningless terms.


Darwinian Evolution by Natural Selection is specifically one system which evolves through time into different systems (e.g. an aquatic form becomes land dwelling). This is an open-ended form of change. It encompasses the concept of non-cyclic change. It also entails unpredictability. However in the classical sciences, a fundamental requirement of a system model is that it is built to perform a specific function ... a concept which is not possible from an evolutionary standpoint, because an evolutionary outcome has [U]no specific function (or purpose) to begin with.

No, evolution is not a 'system', that is just nonsense. It is a principle applicable to biology and it's very simple. All that's needed is variability, reproduction and selection. If we have those three factors then we have evolution.



That's exactly the point. Uniqueness in general exists, so it cannot be dismissed from consideration in the general case. Yet, we are unable to say definitively if something like abiogenesis is unique or not, if its definition entails having to search the entire observable universe to establish the fact. Yet the possibility of its uniqueness still exists (and thus cannot be dismissed in theory in the absence of falsifying empirical evidence). This is exactly the same dichotomy encountered when arguing that some 'theory' predicts other instances of abiogenesis (just in reverse).

You don't have a point. I'm still trying to determine what your point is but you are using unverifiable, meaningless and incoherent terms. You basically acknowledge that your term "uniqueness" is unverifiable, so how exactly do you expect it to be taken into consideration.



You should pay more attention Mr Wally! I gave an example of a non-canoncial edit performed by T.brucei. (http://cosmoquest.org/forum/showthread.php?145936-Did-life-here-start-out-there&p=2160789#post2160789) It does not follow the GU-AG rule for RNA splicing. (http://cosmoquest.org/forum/showthread.php?145936-Did-life-here-start-out-there&p=2160932#post2160932)

I don't pay attention to it, because there is no reason to think that anything involving RNA should be a fundamental law of biology. What if there are lifeforms that don't have RNA?




Coming back to the original point, if we have no other 'lifeforms' other than terrestrial .. then what does 'comparability' (or 'incomparability') mean?

Why don't you look up the definition of the word 'comparability' in the dictionary?

Selfsim
2013-Oct-09, 08:34 PM
As I said ..


I personally doubt it'll ever make sense for some .. because of an ideologically based stance .. (Note: this is not meant as an insult, but this is where fixed ideological principles generally take thinking on such topics ...)In the absence of relevant data, there is no difference in principle between generalising the concept of 'uniqueness' of life and extending it to universality, and generalising the concept of there being other life 'similar' to our own, and extending that to universality. This is the 'top-down' approach, and a logical consequence of the reality of "we don't know" (when it comes to thinking about other instances of life elsewhere). Skimming over this has major consequences.

In fact, this concept is the entry point to understanding just what non-linear systems complexity is all about also. If one doesn't 'get' this, then one won't 'get' the point of such models, and their implications. There is much more to come which, I think, will be better understood .. namely because the reductionst approach has its place in all this and can be legitimately invoked from hereon, (without resulting in the major consequences it has during this part of the top-down approach - namely deletion of uniqueness from consideration).

It never ceases to amaze me how some latch on to science's philosophical basis, turn it into an ideology, and then use it as a weapon. This basis seems to also be acting as the 'mind-block' here … however reductionism and determinism are just one of science's tools. There are others.

This is about one of them.

Paul Wally
2013-Oct-09, 09:48 PM
In the absence of relevant data, there is no difference in principle between generalizing the concept of 'uniqueness' of life and extending it to universality, and generalising the concept of there being other life 'similar' to our own, and extending that to universality.

There is a huge difference. Your concept of 'uniqueness' is not even empirically verifiable by your own acknowledgment, while any theory making detailed predictions about life on other planets would meet all the criteria for scientific testability. This is not about ideology, it is about what works. Untestable and poorly defined concepts like 'uniqueness' doesn't have any useful role to play in science.



In fact, this concept is the entry point to understanding just what non-linear systems complexity is all about also. If one doesn't 'get' this, then one won't 'get' the point of such models, and their implications. There is much more to come which, I think, will be better understood .. namely because the reductionst approach has its place in all this and can be legitimately invoked from hereon, (without resulting in the major consequences it has during this part of the top-down approach - namely deletion of uniqueness from consideration).

It never ceases to amaze me how some latch on to science's philosophical basis, turn it into an ideology, and then use it as a weapon. This basis seems to also be acting as the 'mind-block' here … however reductionism and determinism are just one of science's tools. There are others.

This is about one of them.

So this is now the part where you convince yourself that you understand what's going on, even after contradictions in your thinking have been pointed out.

Colin Robinson
2013-Oct-09, 10:26 PM
Coming back to the original point, if we have no other 'lifeforms' other than terrestrial .. then what does 'comparability' (or 'incomparability') mean?

For example...


If on the other hand, we'd taken a high level system behaviour .. like it reaching a state of dynamic kinetic stability, (for eg), we might actually find that the abiogenesis process might be capable of producing 'no-life' whatsoever .. but it does produce something else

Given that dynamic stability is an important characteristic of Earth's biosphere, if another abiogenesis-like process elsewhere produced something else with dynamic stability, that would be something we could compare with life on Earth, even if it was not exactly the same or even very similar. It would be comparable to life, even if different enough to persuade the scientific community to categorise it as non-life.

Selfsim
2013-Oct-09, 10:56 PM
There is a huge difference. Your concept of 'uniqueness' is not even empirically verifiable by your own acknowledgment, while any theory making detailed predictions about life on other planets would meet all the criteria for scientific testability. This is not about ideology, it is about what works. Untestable and poorly defined concepts like 'uniqueness' doesn't have any useful role to play in science. Ah .. but you are blinded by 'The Dark Side', Mr Wally .. the 'issues' which stop you, have already been solved in complexity theory!

The complex system, as a functional whole, can be defined such that the parts need not necessarily be broken down to the functions of its piecemeal components. Think of say, a traffic jam. Its behaviours can be described independently from the electrons being transferred in the battery of an individual car. As a matter of fact, the behaviour of a traffic jam doesn't need to call upon the reductionist definition of 'electron' to explain its behaviour! The most important aspect of a complex system is to maintain its description as a functional whole at the higher levels. There is no fixed relationship between 'electron' and 'traffic jam'. 'Traffic jam' consists of interdependent and variable behaviours and the parts need not have fixed relationships, quantities or definitions. The flow attributes of 'stop' and 'start' however, would be of primary importance in describing the behaviours of a traffic jam. One wouldn't just chop off the concept of traffic flow 'start' in a model of a traffic jam would one?
So why would one chop off the concept of 'uniqueness' in a model which studies abiogenesis?


So this is now the part where you convince yourself that you understand what's going on, even after contradictions in your thinking have been pointed out.The only contradictions occur when one imposes reductionism and determinism, both of which originate from the rigid belief that everything in nature follows a determinable trajectory. In other words, all things 'exist' before we've found them .. and they merely await our discovery at some point in an uncertain, and indefinite future … (which has absolutely no scientific basis of rationality behind it).

Selfsim
2013-Oct-09, 11:25 PM
For example...

Given that dynamic stability is an important characteristic of Earth's biosphere, if another abiogenesis-like process elsewhere produced something else with dynamic stability, that would be something we could compare with life on Earth, even if it was not exactly the same or even very similar. It would be comparable to life, even if different enough to persuade the scientific community to categorise it as non-life.Perhaps …
… And the end result of 'comparison' might also be 'unique' (or 'similar') ... with those outcomes depending entirely on the arbitrarily selected definition(s) used in the comparison process. Either way, regardless of 'comparability', (which I think, is more a test model attribute), 'unique' and 'similar' outcomes can still exist in a dynamic system, regardless of 'comparability' criteria assumed by some test. Especially whenever that system is capable of 'exploring' a large degree of freedom accessible to it. 'Comparability' is really meaningless, without its arbitrarily chosen criteria, (or definitions), taken from a reductionist viewpoint (eg: .. like 'metabolism', 'homeostasis', etc). It doesn't impact what the system can produce as a whole, however.

Colin Robinson
2013-Oct-09, 11:57 PM
Perhaps …
… And the end result of 'comparison' might also be 'unique' (or 'similar') ... with those outcomes depending entirely on the arbitrarily selected definition(s) used in the comparison process.

I've asked you for your definition of "unique". I'm still waiting for an answer. I don't care whether your definition is "arbitrarily selected" or not...

Paul Wally
2013-Oct-10, 12:42 AM
The complex system, as a functional whole, can be defined such that the parts need not necessarily be broken down to the functions of its piecemeal components. Think of say, a traffic jam. Its behaviours can be described independently from the electrons being transferred in the battery of an individual car. As a matter of fact, the behaviour of a traffic jam doesn't need to call upon the reductionist definition of 'electron' to explain its behaviour! The most important aspect of a complex system is to maintain its description as a functional whole at the higher levels. There is no fixed relationship between 'electron' and 'traffic jam'. 'Traffic jam' consists of interdependent and variable behaviours and the parts need not have fixed relationships, quantities or definitions. The flow attributes of 'stop' and 'start' however, would be of primary importance in describing the behaviours of a traffic jam. One wouldn't just chop off the concept of traffic flow 'start' in a model of a traffic jam would one?

This example actually supports my position. What it illustrates is the autonomy of the emergent system in relation to underlying material substrate. We could have traffic-like behaviour realized with multiple possible underlying units. What this then implies in the case of life is that if life is such an emergent type of self-organization then it will be multiple realizable in many different possible material substrates.



So why would one chop off the concept of 'uniqueness' in a model which studies abiogenesis?

I don't see what this has to do with the example you provided.



The only contradictions occur when one imposes reductionism and determinism, both of which originate from the rigid belief that everything in nature follows a determinable trajectory. In other words, all things 'exist' before we've found them .. and they merely await our discovery at some point in an uncertain, and indefinite future … (which has absolutely no scientific basis of rationality behind it).

Again, what does this have to do with my objection to unverifiable and vaguely defined concepts like 'uniqueness'? You acknowledged that 'uniqueness' is unverifiable, and yet you expect that we give it "consideration". How do we consider something that will never be verified? What does the consideration entail?

Selfsim
2013-Oct-10, 08:34 PM
This example actually supports my position. What it illustrates is the autonomy of the emergent system in relation to underlying material substrate. We could have traffic-like behaviour realized with multiple possible underlying units. What this then implies in the case of life is that if life is such an emergent type of self-organization then it will be multiple realizable in many different possible material substrates.
...
I don't see what this has to do with the example you provided.(I thought you'd like that one .. :)-default .. Just tryin' to keep the conversation movin' forward ..)

Well, the traffic jam was an example of a Self-Organising Complex System.
However, Evolution through to speciation is an example of an Evolving Complex System.
There are components of both of these types of systems involved in life (and even most current abiogenesis hypothesised 'systems').

Whilst some proteins can be synthesised from non-unique combinations of nucelotide amino acids, some cannot. In this example, we have the phenomenon you mention ('swappability' for the want of a better term), as well as unswappability, (or 'uniqueness'). Ventner would argue that some combinations of sub-cellular machinery, result in incompatibility and subsequent cell death (ie: they are unique combinations). We cannot draw the generalised conclusion that you have here. Both 'uniqueness' and 'sufficient similarity', exist within our genetic coding, simultaneously. We cannot conclude there to be a hidden ultimate purpose behind such sub-cellular 'swappability' if life emerges from the whole (this is a key principle of Evolution theory). We cannot draw a generally universal conclusion about substrate swapping (as you have done). Life includes 'Evolvability' and all it entails, as an equal part of its definition.


Again, what does this have to do with my objection to unverifiable and vaguely defined concepts like 'uniqueness'? Both yourself & Colin, require a reduced definition in order to accept the concept of uniqueness. As soon as such a reduced definition comes into play, a generalised complex system becomes specific, which is inappropriate at the system level of abstraction.
Unrestricted, generalised 'uniqueness' however, is applicable in some way, shape or form, across multitudes of naturally occurring complex systems. This is sufficient to justify its applicability at this level of generalisation. This concept, I realise, is rejected by you both .. nonetheless it is a ubiquitous phenomenon across the majority of systems consisting of vast numbers of components.


You acknowledged that 'uniqueness' is unverifiable, I did not say 'uniqueness' is unverifiable. It automatically becomes verified when 'similarity' is falsified. And similarity is an indefinite conclusion which draws upon arbitrary definitions. This dependency is an artifact of reductionist thinking, which can be avoided by treating uniqueness and similarity, as equivalent base axioms in complex systems theory … and getting on with it.

Selfsim
2013-Oct-11, 01:38 AM
Ok .. so we have no agreement on one of two global measures, which I say, can apply equally across any (and all) complex systems ie: Similarity and Uniqueness. (Uniqueness being the disagreed item).

Let's try for some more disagreement, eh? :p :)

How about some measures related to: open-ness, unpredictability, causal loops, non-equilibria and non-linearity?

Yep .. I can see all of those applying across complex systems in general, too ...

How do these impact the classical approach of thinking about abiogenesis?
Ie: the 'classical approach' being more or less along the lines of:

Large numbers of inorganic chemicals interacting over large geological timespans + 'Abiogenesis-conducive' environments (external to the system of chemicals) => Complex long chain organics => RNA 'like molecule assembly => self-replicating molecules => metabolism => life (etc). Or, in other words, a formulaic 'shake and bake' type approach, which always produces 'comparable' outcomes (wherever it occurs) ..

Does anyone not see the fundamental incompatibilities between the two different approaches here?

Colin Robinson
2013-Oct-11, 05:20 AM
Both yourself & Colin, require a reduced definition in order to accept the concept of uniqueness. As soon as such a reduced definition comes into play, a generalised complex system becomes specific, which is inappropriate at the system level of abstraction.
Unrestricted, generalised 'uniqueness' however, is applicable in some way, shape or form, across multitudes of naturally occurring complex systems. This is sufficient to justify its applicability at this level of generalisation. This concept, I realise, is rejected by you both .. nonetheless it is a ubiquitous phenomenon across the majority of systems consisting of vast numbers of components.

I haven't "rejected" the concept of uniqueness, I have simply pointed out that this word "unique" has a range of meanings.

There is a broad sense, in which something is unique if it has no exact duplicate. This sort of uniqueness does indeed occur across multitudes of naturally occurring complex systems. Is that what you mean by "unique", Selfsim?

If so, I'd agree that, owing to the role of random mutation in evolution, the results of evolution on Earth may indeed be "unique" in the sense of having no exact duplicate within the observable universe of billions of galaxies...

Colin Robinson
2013-Oct-11, 05:49 AM
Ok .. so we have no agreement on one of two global measures, which I say, can apply equally across any (and all) complex systems ie: Similarity and Uniqueness. (Uniqueness being the disagreed item).

Let's try for some more disagreement, eh? :p :)

How about some measures related to: open-ness, unpredictability, causal loops, non-equilibria and non-linearity?

Yep .. I can see all of those applying across complex systems in general, too ...

How do these impact the classical approach of thinking about abiogenesis?
Ie: the 'classical approach' being more or less along the lines of:

Large numbers of inorganic chemicals interacting over large geological timespans + 'Abiogenesis-conducive' environments (external to the system of chemicals) => Complex long chain organics => RNA 'like molecule assembly => self-replicating molecules => metabolism => life (etc). Or, in other words, a formulaic 'shake and bake' type approach, which always produces 'comparable' outcomes (wherever it occurs) ..

Does anyone not see the fundamental incompatibilities between the two different approaches here?

I think you may be oversimplifying the discussion about abiogenesis.

You seem to be saying that there are two approaches to the topic, one of which considers the process as entirely deterministic and predictable, while the other considers it as an entirely unpredictable process, for which no general principles can be identified.

Paul Wally
2013-Oct-11, 03:29 PM
Well, the traffic jam was an example of a Self-Organising Complex System.
However, Evolution through to speciation is an example of an Evolving Complex System.
There are components of both of these types of systems involved in life (and even most current abiogenesis hypothesised 'systems').


There are just so many category mistakes here. Firstly, speciation is explained by the theory of evolution and is therefore quite well understood on a general level, therefore it is not an example , but it does have examples. Evolution is quite general as a principle, and I can't see it as an example of anything, except being an example of a general principle, which could then find application in general systems theory.

Secondly, I think it is quite vacuous to say Darwinian evolution is an 'evolving complex system', because what do you mean by "a system"? When we speak of a system we usually understand that as meaning some whole consisting of dynamically interacting parts. The evolution of the system could then be modeled as a state evolution in some phase space, with all kinds of a attractors like fixed point, cyclic or chaotic attractors and so forth. This is simply not how Darwinian evolution works. It operates on an entirely different level, that is, on the level of types or species. One type evolves into a different type or it can split into different types (speciation). Darwinian evolution through natural selection is quite a different, more abstract, concept of evolution which you are confusing with dynamic evolution.




Whilst some proteins can be synthesised from non-unique combinations of nucelotide amino acids, some cannot. In this example, we have the phenomenon you mention ('swappability' for the want of a better term), as well as unswappability, (or 'uniqueness'). Ventner would argue that some combinations of sub-cellular machinery, result in incompatibility and subsequent cell death (ie: they are unique combinations). We cannot draw the generalised conclusion that you have here. Both 'uniqueness' and 'sufficient similarity', exist within our genetic coding, simultaneously. We cannot conclude there to be a hidden ultimate purpose behind such sub-cellular 'swappability' if life emerges from the whole (this is a key principle of Evolution theory). We cannot draw a generally universal conclusion about substrate swapping (as you have done). Life includes 'Evolvability' and all it entails, as an equal part of its definition.

You see, when you use terms like "proteins", "nucleotide" and "amino acids" then you are still not thinking on the level of general systems theory. What is a protein but a macromolecule consisting of amino acids performing certain functions within Earth-life. Life on other planets might contain different macromolecules than what we understand by "proteins", but yet such life could still be defined as reproducing, metabolizing and evolving etc. Reproduction, metabolism and evolution are defined as general concepts, therefore they are by definition multi-realizable.



Both yourself & Colin, require a reduced definition in order to accept the concept of uniqueness. As soon as such a reduced definition comes into play, a generalised complex system becomes specific, which is inappropriate at the system level of abstraction.
Unrestricted, generalised 'uniqueness' however, is applicable in some way, shape or form, across multitudes of naturally occurring complex systems. This is sufficient to justify its applicability at this level of generalisation. This concept, I realise, is rejected by you both .. nonetheless it is a ubiquitous phenomenon across the majority of systems consisting of vast numbers of components.

I'm not looking for a 'reduced definition', just your definition. It's not quite accurate to say I reject the concept of uniqueness, because the problem is that I cannot clearly see what the concept is, so I don't know what concept I'm rejecting. All I know is that it's both vague and unverifiable.



I did not say 'uniqueness' is unverifiable. It automatically becomes verified when 'similarity' is falsified. And similarity is an indefinite conclusion which draws upon arbitrary definitions. This dependency is an artifact of reductionist thinking, which can be avoided by treating uniqueness and similarity, as equivalent base axioms in complex systems theory … and getting on with it.

You implied it. What do you mean by 'similarity is falsified', I don't think it's the kind of thing that's falsifiable. If A and B are observed to have similar features then they have similar features. That would just be an empirical observation, so it would be quite definite. Definitions are usually arbitrary, but that doesn't make'em wrong, just useful or not useful. There's no such thing as a wrong definition.




How about some measures related to: open-ness, unpredictability, causal loops, non-equilibria and non-linearity?

Yep .. I can see all of those applying across complex systems in general, too ...

How do these impact the classical approach of thinking about abiogenesis?
Ie: the 'classical approach' being more or less along the lines of:

Large numbers of inorganic chemicals interacting over large geological timespans + 'Abiogenesis-conducive' environments (external to the system of chemicals) => Complex long chain organics => RNA 'like molecule assembly => self-replicating molecules => metabolism => life (etc). Or, in other words, a formulaic 'shake and bake' type approach, which always produces 'comparable' outcomes (wherever it occurs) ..

Does anyone not see the fundamental incompatibilities between the two different approaches here?

First you set up two incompatible views and then you say: "Look, they're incompatible". What you call the 'classical' approach above is certainly not my view. The problem is that you present it as a single process of one system going through a linear sequence of stages, i.e. the one dynamic system approach as I explained above. What you don't understand, and this is my view, is that we're dealing with a massively parallel network of processes occurring simultaneously, not just some small system undergoing state transitions in phase space.

Selfsim
2013-Oct-12, 08:59 AM
I haven't "rejected" the concept of uniqueness, I have simply pointed out that this word "unique" has a range of meanings.

There is a broad sense, in which something is unique if it has no exact duplicate. This sort of uniqueness does indeed occur across multitudes of naturally occurring complex systems. Is that what you mean by "unique", Selfsim?

If so, I'd agree that, owing to the role of random mutation in evolution, the results of evolution on Earth may indeed be "unique" in the sense of having no exact duplicate within the observable universe of billions of galaxies...The property of 'uniqueness' cannot be measured to exact precision. Therefore a 'uniqueness' which requires such precision, has little/no value in science. I think you're wrestling with how to measure simularity and uniqueness(?) How the measurement is done, and the decision making process following the measurement, are both dependent on many things, not the least of which is the nature of the thing actually being measured.

Of the vast combinations possible in all of organic molecular chemistry, how 'unique' is the combination we know of, which happens to result in system functions which we define as replication, heritability, catalysis, and metabolism? Are these functions comparable or incomparable across the possible resultant space?

Colin Robinson
2013-Oct-13, 02:32 AM
Of the vast combinations possible in all of organic molecular chemistry, how 'unique' is the combination we know of, which happens to result in system functions which we define as replication, heritability, catalysis, and metabolism? Are these functions comparable or incomparable across the possible resultant space?

An important question.

All of the system functions you've mentioned (replication, heritability, catalysis, and metabolism) can be regarded as forms of catalysis. Metabolism is about catalysts breaking down certain complex molecules and building up new ones. Replication and heritability are where an autocatalyst, or an autocatalytic system, is able to undergo small variations and pass the variations to the copies of itself whose formation it catalyses.

The possible number of combinations in organic molecular chemistry is, as you say, vast. Among them, the number with catalytic properties is certainly much greater than the subset which play a part in known terrestrial biochemistry. This is one of the points Stuart Kauffman has looked at in his work on autocatalytic sets.

Selfsim
2013-Oct-13, 06:55 AM
There are just so many category mistakes here. They are not mistakes .. they are deliberate.
Firstly, speciation is explained by the theory of evolution and is therefore quite well understood on a general level, therefore it is not an example , but it does have examples. Evolution is quite general as a principle, and I can't see it as an example of anything, except being an example of a general principle, which could then find application in general systems theory.And principles can't be modelled as a complex system?


Secondly, I think it is quite vacuous to say Darwinian evolution is an 'evolving complex system', because what do you mean by "a system"? When we speak of a system we usually understand that as meaning some whole consisting of dynamically interacting parts. That, and ... if patterns appear and disappear over time, then that would be an instance of a Dynamic Complex System .. there are other types such as Static Complex Systems.
The evolution of the system could then be modeled as a state evolution in some phase space, with all kinds of a attractors like fixed point, cyclic or chaotic attractors and so forth. This is simply not how Darwinian evolution works. It operates on an entirely different level, that is, on the level of types or species. One type evolves into a different type or it can split into different types (speciation). Darwinian evolution through natural selection is quite a different, more abstract, concept of evolution which you are confusing with dynamic evolution.On the contrary, I have distinguished Evolving Complex Systems as having ongoing open-ended outcomes (like species)… not repeated cyclical patterns .. the latter being distinguished as a system exhibiting Dynamic (cyclic) Complexity. (There's no reason that systems cannot interact either).


You see, when you use terms like "proteins", "nucleotide" and "amino acids" then you are still not thinking on the level of general systems theory. What is a protein but a macromolecule consisting of amino acids performing certain functions within Earth-life. Life on other planets might contain different macromolecules than what we understand by "proteins", but yet such life could still be defined as reproducing, metabolizing and evolving etc. Reproduction, metabolism and evolution are defined as general concepts, therefore they are by definition multi-realizable.They are specific emergent functions observed from our terrestrial type of bio-chemistry. They appear nowhere else in nature outside of this particular realm, as far as I'm aware (other than perhaps, in engineered simulations). Perhaps you can show otherwise(??)


I'm not looking for a 'reduced definition', just your definition. It's not quite accurate to say I reject the concept of uniqueness, because the problem is that I cannot clearly see what the concept is, so I don't know what concept I'm rejecting. All I know is that it's both vague and unverifiable.So, pick some other 'vague and unverifiable' concept ... say, 'consciousness'. Can you accept that concept? How about 'existence' or 'reality'? What about 'sentience'?
If you cannot reject these as concepts due to a lack of definition, then there must be something about them which allows you to accept them (I'll assume you do accept these, just for the sake of argument) ... so what is that? There is, I suppose, another alternative which may be (for eg): "I neither accept nor reject them" .. (but this would seem to be a rather strange stance to take) (??)


You implied it.Hmm .. I seem to recall you were saying that a uniqueness which required a search of the entire universe was not verifiable ... I think I could probably concur with that statement but it has been limited in scope by the definition you applied to it. This is what I've tried to point out as being a fundamental difference in starting positions between Complexity and Classical Science.
What do you mean by 'similarity is falsified', I don't think it's the kind of thing that's falsifiable. If A and B are observed to have similar features then they have similar features. That would just be an empirical observation, so it would be quite definite. Definitions are usually arbitrary, but that doesn't make'em wrong, just useful or not useful. There's no such thing as a wrong definition. What we have there is a testable hypothesis: "Might A be similar to B?"
Followed by some test which invokes some measurable thing about B as a reference, which can be used to produce results. Comparative results are then subject to some form of decision making process, having its own set of criteria. The hypothesis can then be decided as being verified ('consistent with'), falsified ('not consistent with'), or indeterminate (a non-result).
From this, I'm not the one attempting to define "uniqueness" at all .. so its not me who is saying there is 'a wrong definition'!


First you set up two incompatible views and then you say: "Look, they're incompatible". What you call the 'classical' approach above is certainly not my view. The problem is that you present it as a single process of one system going through a linear sequence of stages, i.e. the one dynamic system approach as I explained above. What you don't understand, and this is my view, is that we're dealing with a massively parallel network of processes occurring simultaneously, not just some small system undergoing state transitions in phase space.Ok ... I can accept that as your 'view of it', (if you like) ...

Let me ask: "Does it always produces 'comparable' outcomes, wherever it occurs?" ... (In your view)?

Paul Wally
2013-Oct-13, 04:33 PM
The possible number of combinations in organic molecular chemistry is, as you say, vast. Among them, the number with catalytic properties is certainly much greater than the subset which play a part in known terrestrial biochemistry. This is one of the points Stuart Kauffman has looked at in his work on autocatalytic sets.

This is also part of the notion of what I referred to earlier as the multi-realizability of the same type or class of things. You're just looking at it
from the point of view of combinatorics.


They are not mistakes .. they are deliberate. And principles can't be modelled as a complex system?

Well, just because it's deliberate doesn't mean it's all right. What I mean by a category mistake is that you are grouping concepts together as if they are
comparable on the same level, one might call it comparing "apples and oranges", just worse. Saying that evolution is a system is a category mistake. It's like saying electricity, gravity or entropy are systems. Principles can be used to model systems, but they are not in themselves systems.


That, and ... if patterns appear and disappear over time, then that would be an instance of a Dynamic Complex System ...

Patterns appearing and disappearing do not capture the notion of a dynamic system, we're not talking about some magic show here. A dynamic system has physically interacting parts. If there is no interaction there is no dynamic system. For example, the solar system is a dynamic system because all the parts are gravitationally coupled to form one physical system. Darwinian evolution is quite different and should be understood on a more abstract level than that of a dynamic system of interacting physical parts. As I said, it is described on the level of types (species). Types are not interacting parts in a physical system, that would be a category mistake.



They are specific emergent functions observed from our terrestrial type of bio-chemistry. They appear nowhere else in nature outside of this particular realm, as far as I'm aware (other than perhaps, in engineered simulations). Perhaps you can show otherwise(??)

With "specific emergent functions", I take that you mean general characteristics of life, like evolution, metabolism, replication etc. Correction: They appear nowhere else we've looked and we've hardly looked anywhere else, so that's just yet another meaningless statement. Perhaps you could explain why something like metabolism would exist only in the form of one particular type of bio-chemistry? It's quite possible and scientifically meaningful to look for other bio-chemical instances of "metabolism" on Titan for example, even though we wouldn't expect the underlying bio-chemistry to be anything like we have here on Earth, but we know what the word metabolism means and that makes it possible to look for other instances of it.



So, pick some other 'vague and unverifiable' concept ... say, 'consciousness'. Can you accept that concept? How about 'existence' or 'reality'? What about 'sentience'?
If you cannot reject these as concepts due to a lack of definition, then there must be something about them which allows you to accept them (I'll assume you do accept these, just for the sake of argument) ... so what is that? There is, I suppose, another alternative which may be (for eg): "I neither accept nor reject them" .. (but this would seem to be a rather strange stance to take) (??)

I don't have a problem with any of those concepts, including the concept of uniqueness, as long as these concepts are clearly defined such as to make scientific discussion possible. If you define your concept of uniqueness in such a way that we could all have a meaningful scientific discussion about it then I have no problem. But you refuse to clarify what you mean by 'uniqueness', while at the same time using it in a way that appears to be unverifiable in principle. I think it is then only reasonable to request what it is that you're talking about when you say we have to 'consider uniqueness'. How are we supposed to consider uniqueness if you don't clarify what you mean by it?



Hmm .. I seem to recall you were saying that a uniqueness which required a search of the entire universe was not verifiable ... I think I could probably concur with that statement but it has been limited in scope by the definition you applied to it.

Then please broaden the 'scope' to something we could recognize as meaningful.



From this, I'm not the one attempting to define "uniqueness" at all .. so its not me who is saying there is 'a wrong definition'!

You are the one using the term, so it's only reasonable that you define it.



Ok ... I can accept that as your 'view of it', (if you like) ...

Let me ask: "Does it always produces 'comparable' outcomes, wherever it occurs?" ... (In your view)?

What is this?? You presented the following two 'incompatible' views:


How about some measures related to: open-ness, unpredictability, causal loops, non-equilibria and non-linearity?

Yep .. I can see all of those applying across complex systems in general, too ...

How do these impact the classical approach of thinking about abiogenesis?
Ie: the 'classical approach' being more or less along the lines of:

Large numbers of inorganic chemicals interacting over large geological timespans + 'Abiogenesis-conducive' environments (external to the system of chemicals) => Complex long chain organics => RNA 'like molecule assembly => self-replicating molecules => metabolism => life (etc). Or, in other words, a formulaic 'shake and bake' type approach, which always produces 'comparable' outcomes (wherever it occurs) ..

Does anyone not see the fundamental incompatibilities between the two different approaches here?

Now, what I'm saying is that if this so called 'classical approach' is intended to be a representation of my view then it is a misrepresentation. My view
is that there is a whole network of simultaneous processes and not a single linear sequence of events. There is then no incompatibility between things like unpredictability and nonlinearity with the possibility of high likelihood of abiogenesis, i.e. unpredictability and nonlinearity don't imply that abiogenesis should be unlikely. Quite the contrary. The randomness associated with unpredictability leads to massively parallel 'exploration' of the possibility space and should therefore increase the likelihood of abiogenesis!

Selfsim
2013-Oct-13, 10:37 PM
So, just summarising the points made in this thread about Complexity Modelling:

i) Complexity addresses things normally taken by classical reductionist approaches to science, as being too vague and ill-defined. One illustrated example is 'Uniqueness' (and its couterpart, 'Similarity);

ii) One can distinguish several types of complex systems. Examples given thus far are:
(a) Static Complexity, (b) Dynamic Complexity, (c) Evolving Complexity and (d) Self Organising Complexity.

iii) The basis for globally quantifying Complex Systems which encompass:
- interacting variable parameters (which can also swap places over the lifetime of a given system when crossing boundaries such as the edge of chaos) and;
- time varying numbers of variables, have been suggested as:

- Unpredictability, non-equilibria, causal loops, nonlinearity and open-ness.

iv) The aims of distinguishing all of this, might be to explain emergent structures, measure (relative) complexity, distinguish common behaviours and conditions towards which certain systems 'gravitate', construct useful models for study, attempt to distinguish any statistical predictors which may be present, and distinguish any physical laws (where possible).

So, the question now is: "Can this approach be applied to Pre-Evolution and Evolution?"

(I think it already has ..)

Colin Robinson
2013-Oct-14, 01:20 AM
So, just summarising the points made in this thread about Complexity Modelling:

i) Complexity addresses things normally taken by classical reductionist approaches to science, as being too vague and ill-defined. One illustrated example is 'Uniqueness' (and its couterpart, 'Similarity);

In other words, your conception of "complexity modelling" allows you to be as vague as you like? And anyone who requests clarification of a term like "uniqueness" can be labelled as a "classical reductionist"?


ii) One can distinguish several types of complex systems. Examples given thus far are:
(a) Static Complexity, (b) Dynamic Complexity, (c) Evolving Complexity and (d) Self Organising Complexity.

iii) The basis for globally quantifying Complex Systems which encompass:
- interacting variable parameters (which can also swap places over the lifetime of a given system when crossing boundaries such as the edge of chaos) and;
- time varying numbers of variables, have been suggested as:

- Unpredictability, non-equilibria, causal loops, nonlinearity and open-ness.

iv) The aims of distinguishing all of this, might be to explain emergent structures, measure (relative) complexity, distinguish common behaviours and conditions towards which certain systems 'gravitate', construct useful models for study, attempt to distinguish any statistical predictors which may be present, and distinguish any physical laws (where possible).

So, the question now is: "Can this approach be applied to Pre-Evolution and Evolution?"

(I think it already has ..)

Yes, theorists including Stuart Kauffman have applied theories about complex systems to abiogenesis and evolution… You keep telling us that complexity theory addresses "uniqueness". Well, have you looked at whether and how Kauffman (for instance) uses that word "uniqueness"? Perhaps you could come up with some quotes?

Selfsim
2013-Oct-14, 07:37 AM
In other words, your conception of "complexity modelling" allows you to be as vague as you like? And anyone who requests clarification of a term like "uniqueness" can be labelled as a "classical reductionist"?I find this comment to be rather amusing.

The opportunity exists to include as an axiom, for a theory of nature, a placeholder for a generally undefinable-in-advance, but abundantly evident phenomenon. (For which specific measures and decision making criteria do exist ... and are used all the time).

It is recognised that uniqueness is an artifact of scale, and can also become apparent by imposing specifically defined constraints on a generalised system ... both of which, by necessity, end up being scoped from what is already known (ie: past knowledge).

The approach is different from the classical science approach. It has to be because of the nature of the systems under consideration. This happens all the time in science ... (For example: Newton's/Liebnitz's calculus, Poincaré's topology, Cantor's Set Theory, etc, etc).


Yes, theorists including Stuart Kauffman have applied theories about complex systems to abiogenesis and evolution… You keep telling us that complexity theory addresses "uniqueness". Well, have you looked at whether and how Kauffman (for instance) uses that word "uniqueness"? Perhaps you could come up with some quotes?Goodness me ..

I addressed one of the challenges confronting Kauffman, which calls on him to justify his implied assumptions ... here (http://cosmoquest.org/forum/showthread.php?146161-Models-of-abiogenesis-Stuart-Kauffman-and-Addy-Pross&p=2156603#post2156603). (Neither yourself nor Mr Wally, chose to respond). Whilst the post was admittedly about the probability of the appearance of autocatalytic sets, the matter is related to pattern measures and consequent probability estimation.

To the best of my knowledge, he has never fully addressed the effects of inhibitor products typically produced in autocatalysis, either.

I'm sorry if it disappoints, but Kauffman appears to be no authority figure, when it comes to statistical interpretation of pattern measures, or in his use of concise terminology either, (IMO).

Paul Wally
2013-Oct-14, 10:14 AM
So, just summarising the points made in this thread about Complexity Modelling:

i) Complexity addresses things normally taken by classical reductionist approaches to science, as being too vague and ill-defined. One illustrated example is 'Uniqueness' (and its couterpart, 'Similarity);

ii) One can distinguish several types of complex systems. Examples given thus far are:
(a) Static Complexity, (b) Dynamic Complexity, (c) Evolving Complexity and (d) Self Organising Complexity.

iii) The basis for globally quantifying Complex Systems which encompass:
- interacting variable parameters (which can also swap places over the lifetime of a given system when crossing boundaries such as the edge of chaos) and;
- time varying numbers of variables, have been suggested as:

- Unpredictability, non-equilibria, causal loops, nonlinearity and open-ness.

iv) The aims of distinguishing all of this, might be to explain emergent structures, measure (relative) complexity, distinguish common behaviours and conditions towards which certain systems 'gravitate', construct useful models for study, attempt to distinguish any statistical predictors which may be present, and distinguish any physical laws (where possible).

So, the question now is: "Can this approach be applied to Pre-Evolution and Evolution?"

(I think it already has ..)

Argg ... this is just more of the same old post-modernist non-sense that we usually find under the guise of "complexity theory".

Selfsim
2013-Oct-14, 10:21 AM
Argg ... this is just more of the same old post-modernist non-sense that we usually find under the guise of "complexity theory".Please explain ...

Colin Robinson
2013-Oct-14, 08:03 PM
I find this comment to be rather amusing.

I raised a serious question about a key term used by yourself — the term "uniqueness". If you took such questions seriously, you might perhaps develop a more convincing argument...


This happens all the time in science ... (For example: Newton's/Liebnitz's calculus, Poincaré's topology, Cantor's Set Theory, etc, etc).

Are you saying that Newton, Poincaré and Cantor didn't need to define their terms or answer questions about their terms?


I addressed one of the challenges confronting Kauffman, which calls on him to justify his implied assumptions ... here (http://cosmoquest.org/forum/showthread.php?146161-Models-of-abiogenesis-Stuart-Kauffman-and-Addy-Pross&p=2156603#post2156603). (Neither yourself nor Mr Wally, chose to respond).

Sorry if I haven't always given your postings the attention you think they deserve…

In the posting you mentioned, you referred to questions about Kauffman's mathematical model of autocatalytic set formation which were raised by Elchanan Mossel and Mike Steel.

Well then… do Mossel and Steel use the word "uniqueness"? If so, how do they use it? What do they mean?

Selfsim
2013-Oct-14, 10:29 PM
I raised a serious question about a key term used by yourself — the term "uniqueness". If you took such questions seriously, you might perhaps develop a more convincing argument…I have answered your question seriously (several times now). That the answer isn't recognisable to you, or that it doesn't fit into your ideology, is not something which particularly concerns me because he conversation about pre-evolution and complex systems, can still continue independently from this philosophical disconnect.


And it is a philosophical point of contention here .. not a scientific one. I have suggested moving forward, but this issue seems to be a show-stopper from both yourself's and Paul's shared ideological belief system. Such dogmatic attachment highlights for me, the major downside of imposing definitions, as a way of deliberately excluding abundantly evident phenomenon from consideration, at a generalised systems level.


The hint (I provided) was that uniqueness could be treated axiomatically. One of the few logical reasons I can see as to why it would be rejected as an axiom, by both Paul and yourself, is that neither of you accept that uniqueness is self-evident.



Are you saying that Newton, Poincaré and Cantor didn't need to define their terms or answer questions about their terms?No.
And I have distinguished many other 'terms' (concepts really) throughout my commentary here. 'Tis exactly the same process used by those mentioned above.



Sorry if I haven't always given your postings the attention you think they deserve… Oh, I wasn't complaining .. I was just pointing out that I had addressed Kauffman's oversights elsewhere ..



In the posting you mentioned, you referred to questions about Kauffman's mathematical model of autocatalytic set formation which were raised by Elchanan Mossel and Mike Steel.


Well then… do Mossel and Steel use the word "uniqueness"? If so, how do they use it? What do they mean?Would you explain the relevance of this question?


I use lots of words and concepts which I'm sure they didn't explicitly single out in their paper. The arbitrary assignment of probability estimates, is one way of defining uniqueness or similarity ... which is exactly what Kauffman has done. He chose similarity to the extent which determines his conclusion that: 'the emergence of autocatalytic sets is inevitable' (or non-unique, from a temporal perspective).

Paul Wally
2013-Oct-14, 11:30 PM
The hint (I provided) was that uniqueness could be treated axiomatically. One of the few logical reasons I can see as to why it would be rejected as an axiom, by both Paul and yourself, is that neither of you accept that uniqueness is self-evident.



Ok, now apply your concept of 'uniqueness' as an axiom. Let's see of what scientific use the concept is.
As an example let's propose the hypothesis: "Life is unique to Earth". How do we test this hypothesis?

Selfsim
2013-Oct-15, 03:22 AM
Ok, now apply your concept of 'uniqueness' as an axiom. Let's see of what scientific use the concept is.
As an example let's propose the hypothesis: "Life is unique to Earth". How do we test this hypothesis?Firstly, that statement fails what I'd call a scientific hypothesis, regardless of its subject.

A valid hypothesis is a tentative statement, so rewording it would read:
"Life might be unique to Earth".

Secondly, we have only one suite of tests capable of eliminating uncertainty, and that suite of tests is specific to carbon-based life. So, rewording again, we get:
"Carbon-based life might be unique to Earth".

Ok so following convention, we devise an appropriate test able to falsify the hypothesis. So we look for carbon-based life, using our best diagnostic tests. Three results are possible:
(i) null - we didn't find carbon-based life (.. the result is of no significant consequence);
(ii) we found carbon based life - uniqueness to Earth is falsified or;
(iii) a non-result was returned - the test and/or conditions were shown to result in uncertainties exceeding the precision of the measurement.

So there ya go. A perfectly testable scientific hypothesis having an axiomatic basis of uniqueness.

Colin Robinson
2013-Oct-15, 08:36 AM
IOne of the few logical reasons I can see as to why it would be rejected as an axiom, by both Paul and yourself, is that neither of you accept that uniqueness is self-evident.

It is by no means self-evident to me that the word "uniqueness" has a single, clear meaning. On the contrary, I think it is evident that the word has at least two distinct senses, and has indeed been used in both these senses recently by you.

Sense 1:

A couple of weeks back, in the thread from which this thread branched off, you gave the following examples of (one sort of) "uniqueness":


So you deny 'the availability' of the examples I cite of unpredictability and uniqueness of (in natural environments and at certain scales): a brick covered in emery cloth sliding down an incline (avalanche model), weather a couple of weeks out, A.DIM's apparent uniqueness, Paul Wally's apparent uniqueness, fingerprints, an individual's retinal pattern, an individual's brain morphology, Earth's uniqueness within the Solar System, snowflake patterns, etc, etc?… (The list is virtually endless).

A human individual appears unique in the sense of having no exact duplicate, but is nonetheless a member of a larger class of entities with much in common: the class of all humans. Likewise, a fingerprint appears unique in the sense of having no exact duplicate, however it too is a member of a larger class of things with much in common — it is a member of the class of all fingerprints.

Sense 2:

In your latest posting you've written:


A valid hypothesis is a tentative statement, so rewording it would read:
"Life might be unique to Earth".

Secondly, we have only one suite of tests capable of eliminating uncertainty, and that suite of tests is specific to carbon-based life. So, rewording again, we get:
"Carbon-based life might be unique to Earth".

Ok so following convention, we devise an appropriate test able to falsify the hypothesis. So we look for carbon-based life, using our best diagnostic tests. Three results are possible:
(i) null - we didn't find carbon-based life (.. the result is of no significant consequence);
(ii) we found carbon based life - uniqueness to Earth is falsified or;
(iii) a non-result was returned - the test and/or conditions were shown to result in uncertainties exceeding the precision of the measurement.

So there ya go. A perfectly testable scientific hypothesis having an axiomatic basis of uniqueness.

Here (it seems to me) you are talking about a sort of (hypothetical) uniqueness which is very different from the apparent uniqueness of a fingerprint.

It is different because the apparent uniqueness of a fingerprint is not falsified by discovery of another fingerprint from another person, unless the other fingerprint is exactly the same.

Carbon-based life on Earth could still be unique in Sense 1 even if carbon-based life were found on Titan (for instance), as long as the life found on Titan was not exactly the same as life on Earth.

There are, indeed, many apparent examples of "uniqueness" in Sense 1. But do any of these examples add to the plausibility of hypothesized "uniqueness", in Sense 2, of carbon-based life on Earth?

Paul Wally
2013-Oct-15, 09:11 AM
Firstly, that statement fails what I'd call a scientific hypothesis, regardless of its subject.

A valid hypothesis is a tentative statement, so rewording it would read:
"Life might be unique to Earth".

Secondly, we have only one suite of tests capable of eliminating uncertainty, and that suite of tests is specific to carbon-based life. So, rewording again, we get:
"Carbon-based life might be unique to Earth".

Ok so following convention, we devise an appropriate test able to falsify the hypothesis. So we look for carbon-based life, using our best diagnostic tests. Three results are possible:
(i) null - we didn't find carbon-based life (.. the result is of no significant consequence);
(ii) we found carbon based life - uniqueness to Earth is falsified or;
(iii) a non-result was returned - the test and/or conditions were shown to result in uncertainties exceeding the precision of the measurement.

So there ya go. A perfectly testable scientific hypothesis having an axiomatic basis of uniqueness.

Just more nonsense. I asked a very simple question, how do you test the hypothesis of uniqueness, now it's being changed to "possible uniqueness" by introducing the word "might". This is then even less testable and meaningful a hypothesis. Firstly, life or carbon based life (if you like) might already be unique, so there is nothing to test. It might be unique, period.

What do we have to observe to come to the conclusion: "Life might be unique to Earth"?
This is just nonsense ... absolute nonsense.

Paul Wally
2013-Oct-15, 10:29 AM
Just to clarify the concept of testability, a good example is the test for the prediction made by general relativity that light bends in a gravitational field because of the curvature of space. A very specific test was then proposed and that was to observe whether the background stars shift their apparent positions during a solar eclipse by a certain amount. No such scientific test is possible for "Life might be unique to Earth" nor "Life is unique to Earth", simply because it can never be positively confirmed, it's simply not verifiable. So it's really of no scientific use.

Selfsim
2013-Oct-15, 11:11 PM
… No such scientific test is possible for "Life might be unique to Earth" nor "Life is unique to Earth", simply because it can never be positively confirmed, it's simply not verifiable. So it's really of no scientific use.My dear Mr Wally … you have this sooo confused ...

The only thing of importance to the scientific process, is that a hypothesis be testable in principle. It doesn't matter whether we think (in advance of the test), whether something can be verified or falsified .. that is the whole reason for executing the test in the first place … (ie: to ascertain that). Theoretical science doesn't care about the impracticalities associated with conducting a test .. all that matters is that the test can be shown to be:

i) capable of returning a result by eliminating inherent uncertainties and;
ii) that it does not require the violation of any physical principles, (or scientific laws).

You have also fallen into the trap of believing that human contrived definitions can be substituted in place of physical reality, and can therefore be used to eliminate things which may, or may not, become known to science. They are not, and cannot. They can be changed at will, and frequently are, when properly formulated tests produce their repeatable results and challenge the assumptions behind these definitions.

Categorisation (taxonomy) in the biological sciences, provides useful general guidelines, but it does not determine predictability.

Uniqueness and similarity are two sides of the same coin. They are distinguished only by measures produced by a test, which has been specifically designed to facilitate a decision.

sapcealien
2013-Oct-16, 06:58 AM
Hi Selfsim, I think Evolution by Natural Selection is best described as a process or perhaps a mechanism. I can't really see how it is a system?

"Darwinian Evolution by Natural Selection is specifically one system which evolves through time into different systems (e.g. an aquatic form becomes land dwelling). This is an open-ended form of change. It encompasses the concept of non-cyclic change. It also entails unpredictability."

Noclevername
2013-Oct-16, 12:51 PM
Welcome, sapcealien! (I don't know if that spelling is deliberate, if it's not you can change it by using the Settings button at the top of the page; if it is, please disregard this sentence)

I agree with you about evolution being a process. I don't think Selfsim was saying that evolution is a system, I just think his statement was worded in a slightly confusing way or skipped a word; Maybe "evolution is when one system evolves into different systems" would be more accurate.

Paul Wally
2013-Oct-16, 01:27 PM
The only thing of importance to the scientific process, is that a hypothesis be testable in principle.

Exactly, an unverifiable hypothesis is not testable in principle, because it's not verifiable in principle. Uniqueness of life on Earth is not verifiable
in principle because there is no possible test that could be conducted that would verify whether life is unique to Earth.




It doesn't matter whether we think (in advance of the test), whether something can be verified or falsified .. that is the whole reason for executing the test in the first place … (ie: to ascertain that). Theoretical science doesn't care about the impracticalities associated with conducting a test .. all that matters is that the test can be shown to be:


My goodness! You think it's an impracticality? Now tell me, what are the "impracticalities" that we have to overcome in order to verify whether life is unique to Earth?



i) capable of returning a result by eliminating inherent uncertainties and;
ii) that it does not require the violation of any physical principles, (or scientific laws).

Your first point is just nonsense. Any act of observation or measurement is capable of returning a result and eliminate some uncertainty, and that is
independent of any nonsense "hypothesis" that you may hold. I can hypothesise that there is an amusement park on Pluto, and when we go look and find no amusement park then some results are being returned and some uncertainties are being eliminated.



You have also fallen into the trap of believing that human contrived definitions can be substituted in place of physical reality, and can therefore be used to eliminate things which may, or may not, become known to science. They are not, and cannot. They can be changed at will, and frequently are, when properly formulated tests produce their repeatable results and challenge the assumptions behind these definitions.


I'm not falling into any trap. It is you who continually misunderstand and misrepresent. We need definitions to talk about physical reality. Without definitions we would be unable to communicate with clarity about physical reality, which is exactly what the problem is here. You are
unable or unwilling to present clearly defined and verifiable concepts, probably to immunize them against critical scrutiny.



Uniqueness and similarity are two sides of the same coin. They are distinguished only by measures produced by a test, which has been specifically designed to facilitate a decision.

How about difference and similarity. They look like they make a better couple ... and hey they're both verifiable.

Selfsim
2013-Oct-16, 09:01 PM
Hi Selfsim, I think Evolution by Natural Selection is best described as a process or perhaps a mechanism. I can't really see how it is a system?

"Darwinian Evolution by Natural Selection is specifically one system which evolves through time into different systems (e.g. an aquatic form becomes land dwelling). This is an open-ended form of change. It encompasses the concept of non-cyclic change. It also entails unpredictability."From Wiki, "System (http://en.wikipedia.org/wiki/System)"
A system is a set of interacting or interdependent components forming an integrated whole or a set of elements (often called 'components') and relationships which are different from relationships of the set or its elements to other elements or sets.We are discussing modelling of complex systems. I can model Evolution as a theory, abstract it then simulate its development over time in a computing environment, and refer to that as a "System".

This is yet another example of taking a word meaning and assuming it to be universal. A "System" can have many meanings .. it does not apply solely to physical-'type' systems (or for example, 'species'). Natural selection could be abstracted as a set of system rule class types. 'Species' would be a system component, etc, etc.

Selfsim
2013-Oct-16, 09:06 PM
I agree with you about evolution being a process. I don't think Selfsim was saying that evolution is a system, I just think his statement was worded in a slightly confusing way or skipped a word; Maybe "evolution is when one system evolves into different systems" would be more accurate.No .. the issue in this thread is about abstraction to generalised models. "System" can have many meanings at the highest level of abstraction.
The system I'm talking about here (if I could get it off the starting blocks), would also encompass the environment as another (physical) system component (for example).

Paul Wally
2013-Oct-17, 02:20 PM
No .. the issue in this thread is about abstraction to generalised models. "System" can have many meanings at the highest level of abstraction.
The system I'm talking about here (if I could get it off the starting blocks), would also encompass the environment as another (physical) system component (for example).

However ... you have to be very careful not to conflate the different levels of system. If you want to model evolution as an abstract system with
species as a basic component, then you shouldn't confuse that for a real-time physical dynamic system.

Colin Robinson
2013-Oct-20, 10:25 PM
Just to clarify the concept of testability, a good example is the test for the prediction made by general relativity that light bends in a gravitational field because of the curvature of space. A very specific test was then proposed and that was to observe whether the background stars shift their apparent positions during a solar eclipse by a certain amount.

Even if a hypothesis (such as general relativity) makes a testable prediction, and the prediction is confirmed by observation, does that prove that the hypothesis is true? Isn't it still possible that other predictions that derive from the same hypothesis will be falsified by later observations?


No such scientific test is possible for "Life might be unique to Earth" nor "Life is unique to Earth", simply because it can never be positively confirmed, it's simply not verifiable. So it's really of no scientific use.

I've criticised Selfsim's use of the word "unique", on the grounds that he shifts from one meaning of that word to another. But this doesn't mean we can dismiss the idea that Earth may be the only world with life. Actually, there are testable predictions inherent in such a conception. For instance, it predicts that an experiment in radio SETI will have a negative result: no extra-terrestrial radio signals will be detected.

Paul Wally
2013-Oct-20, 11:32 PM
Even if a hypothesis (such as general relativity) makes a testable prediction, and the prediction is confirmed by observation, does that prove that the hypothesis is true? Isn't it still possible that other predictions that derive from the same hypothesis will be falsified by later observations?

No, it doesn't prove that GR is true, but the derived prediction is proven true (within limits of accuracy). These derived predictions are verifiable and they corroborate GR. I see what you getting at, but in the case of GR we have actual verification of very specific results; results not predicted by any other theories, as far as I know. Selfsim's "uniqueness hypothesis" is quite different. It doesn't make specific verifiable predictions that cannot be explained by a large number of other hypotheses, i.e it doesn't make verifiable predictions that distinguishes it from other hypotheses such that we could then say it is corroborated in the same way as GR is.




I've criticised Selfsim's use of the word "unique", on the grounds that he shifts from one meaning of that word to another. But this doesn't mean we can dismiss the idea that Earth may be the only world with life. Actually, there are testable predictions inherent in such a conception. For instance, it predicts that an experiment in radio SETI will have a negative result: no extra-terrestrial radio signals will be detected.

I'm not dismissing the possibility, but the problem is that if Earth is the only world with life how would we test it? What I mean by testability is: What observation would imply that life is unique to Earth? The shifting of apparent stellar positions during a solar eclipse implies that space is curved by massive objects as GR says it is. Does "uniqueness" make a similar kind of prediction?

A SETI negative result is physically impossible. It means SETI has scanned the entire ontological universe, i.e. that is an untestable prediction. But if you mean that SETI has found nothing thus far then there is again the problem that it can be explained by a large number of valid scientific hypotheses, e.g. that intelligent technological civilizations are rare. So a null finding in this latter sense provides no reason to accept the uniqueness hypothesis above other hypotheses that actually do make testable predictions.

Colin Robinson
2013-Oct-21, 02:20 AM
No, it doesn't prove that GR is true, but the derived prediction is proven true (within limits of accuracy). These derived predictions are verifiable and they corroborate GR. I see what you getting at, but in the case of GR we have actual verification of very specific results; results not predicted by any other theories, as far as I know. Selfsim's "uniqueness hypothesis" is quite different.

It is not only Selfsim's hypothesis. Paul Davies has put forward as a possible resolution to Fermi's paradox the conjecture that Earth is the only place in the universe with life on it. I think Michael Hart also has addressed the Fermi paradox along the same lines.


It doesn't make specific verifiable predictions that cannot be explained by a large number of other hypotheses, i.e it doesn't make verifiable predictions that distinguishes it from other hypotheses such that we could then say it is corroborated in the same way as GR is.

I'm not dismissing the possibility, but the problem is that if Earth is the only world with life how would we test it? What I mean by testability is: What observation would imply that life is unique to Earth? The shifting of apparent stellar positions during a solar eclipse implies that space is curved by massive objects as GR says it is. Does "uniqueness" make a similar kind of prediction?

I think it doesn't — there is no single observation that would confirm it in the way the eclipse observation confirmed GR.

On the other hand, I can imagine a scenario where we might at least become convinced that the probability of life emerging on any given world is extremely low.

For instance, studies of other planets might show that processes similar to those which led to life here can also lead to something other than life, and further, that the non-life outcome actually has a hugely high probability. (Right now, we don't know whether abiogenesis-like processes ever lead to an outcome other than life.)

In that case we would still not be certain that Earth was the only place where life happened. However, the conjecture that life is unique to Earth would then become more plausible than it is now.

Paul Wally
2013-Oct-21, 09:41 PM
It is not only Selfsim's hypothesis. Paul Davies has put forward as a possible resolution to Fermi's paradox the conjecture that Earth is the only place in the universe with life on it. I think Michael Hart also has addressed the Fermi paradox along the same lines.

I think it doesn't — there is no single observation that would confirm it in the way the eclipse observation confirmed GR.

On the other hand, I can imagine a scenario where we might at least become convinced that the probability of life emerging on any given world is extremely low.

For instance, studies of other planets might show that processes similar to those which led to life here can also lead to something other than life, and further, that the non-life outcome actually has a hugely high probability. (Right now, we don't know whether abiogenesis-like processes ever lead to an outcome other than life.)

In that case we would still not be certain that Earth was the only place where life happened. However, the conjecture that life is unique to Earth would then become more plausible than it is now.

I'm not arguing about plausibility. My point is that I don't see any use in holding such a hypothesis. I agree with what you say that
"studies of other planets might show that processes similar to those which led to life here can also lead to something other than life, and further, that the non-life outcome actually has a hugely high probability..." , but I don't see how the hypothesis that life is unique to Earth would play any role in making such a discovery.

Colin Robinson
2013-Oct-21, 11:27 PM
I'm not arguing about plausibility. My point is that I don't see any use in holding such a hypothesis.

Perhaps it depends what you mean by "holding". I think it is useful to consider a range of hypotheses and speculations, including this one.


I agree with what you say that
"studies of other planets might show that processes similar to those which led to life here can also lead to something other than life, and further, that the non-life outcome actually has a hugely high probability..." , but I don't see how the hypothesis that life is unique to Earth would play any role in making such a discovery.

Whatever is out there, I think we are more likely to discover it if we do some conceptual modelling first. One way that idea that life may be unique to Earth could play a role, is that it could prompt someone to develop a detailed model where a planet or moon with complex organic molecules and liquid solvents develops not into a biosphere but into something else.

marsbug
2013-Oct-21, 11:30 PM
Re Pauls last post:

Is the hypothesis that life is unique to Earth a special case of a hypothesis that worlds with prebiotic conditions similar to early Earths can produce a range of unique complex systems with unique behaviours? If so then the idea that life (as opposed to some other system) is unique to Earth is pointing towards the more general possibility, of other Earth like worlds giving rise to different complex systems. Phenomena like Prontienoid globules (http://en.wikipedia.org/wiki/Proteinoids) that show some of the characteristics of life, but not all, might be partial evidence in favour of such a general hypothesis. If it was shown to be a valid hypothesis, given current evidence, then it can be arrived at through considering the possibility that Earth is the only world that harbours 'life'. So considering that life is unique to Earth might have utility in that sense?

Paul Wally
2013-Oct-22, 12:32 AM
Whatever is out there, I think we are more likely to discover it if we do some conceptual modelling first. One way that idea that life may be unique to Earth could play a role, is that it could prompt someone to develop a detailed model where a planet or moon with complex organic molecules and liquid solvents develops not into a biosphere but into something else.

Well ... I think we're finally getting somewhere. So such a model could then be based on various principles of self-organizing complexity, laws of physics and chemistry and so on. However, I think if life turns out to be some sort of unique outcome of a complex system then it must come out as a consequence of the model, not as something we start out with as basic postulate.