What do you think?
Rate this book
The Origins of Order: Self-Organization and Selection in Evolution
Stuart Kauffman here presents a brilliant new paradigm for evolutionary biology, one that extends the basic concepts of Darwinian evolution to accommodate recent findings and perspectives from the fields of biology, physics, chemistry and mathematics. The book drives to the heart of the exciting debate on the origins of life and maintenance of order in complex biological systems. It focuses on the concept of self-organization: the spontaneous emergence of order that is widely observed throughout nature Kauffman argues that self-organization plays an important role in the Darwinian process of natural selection. Yet until now no systematic effort has been made to incorporate the concept of self-organization into evolutionary theory. The construction requirements which permit complex systems to adapt are poorly understood, as is the extent to which selection itself can yield systems able to adapt more successfully. This book explores these themes. It shows how complex systems, contrary to expectations, can spontaneously exhibit stunning degrees of order, and how this order, in turn, is essential for understanding the emergence and development of life on Earth. Topics include the new biotechnology of applied molecular evolution, with its important implications for developing new drugs and vaccines; the balance between order and chaos observed in many naturally occurring systems; new insights concerning the predictive power of statistical mechanics in biology; and other major issues. Indeed, the approaches investigated here may prove to be the new center around which biological science itself will evolve. The work is written for all those interested in the cutting edge of research in the life sciences.
727 pages, Kindle Edition
First published April 11, 1993
52 people are currently reading
2284 people want to read
About the author
Stuart A. Kauffman
16 books184 followersStuart Alan Kauffman (28 September 1939) is an American theoretical biologist and complex systems researcher concerning the origin of life on Earth. He is best known for arguing that the complexity of biological systems and organisms might result as much from self-organization and far-from-equilibrium dynamics as from Darwinian natural selection, as well as for applying models of Boolean networks to simplified genetic circuits.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 13 of 13 reviews
April 11, 2014
Very detailed approach to the self-organization theory.
Basic Self-organizational problems: DNA first or protein first? The standard tools of evolution (mutation + natural selection) have no material to work with here so they don't have any effect. Although Kauffman is very optimistic about laboratory experiments, other experiments have proven that any laboratory-manufactured self-replicating DNA over time tends towards the very minimum complexity necessary (thus shedding any and all unnecessary complexity for replication). In the evolutionary sense (he who reproduces the most is the fittest) simplicity reigns and complexity is a burden. This observational fact is 100% contrary to what Kauffman seeks to demonstrate: that self-organization is an innate quality of matter and complexity is a natural consequence.
Kauffman misses the forest for the trees.
Basic Self-organizational problems: DNA first or protein first? The standard tools of evolution (mutation + natural selection) have no material to work with here so they don't have any effect. Although Kauffman is very optimistic about laboratory experiments, other experiments have proven that any laboratory-manufactured self-replicating DNA over time tends towards the very minimum complexity necessary (thus shedding any and all unnecessary complexity for replication). In the evolutionary sense (he who reproduces the most is the fittest) simplicity reigns and complexity is a burden. This observational fact is 100% contrary to what Kauffman seeks to demonstrate: that self-organization is an innate quality of matter and complexity is a natural consequence.
Kauffman misses the forest for the trees.
June 4, 2020
This is a thick book, but not because of its number of pages. In this book Kauffman goes deep into protein modeling through Boolean networks. This is perhaps a little dated, as the book was written in 1993 (data science has since provided more advanced methods perhaps). But the key is the same. Kauffman looks at modeling different activities to see how likely mutation can work for evolution.
Most of this book is about this kind of modeling. He then ends the book with a chapter on cell differentiation, as it occurs in different zygotes of developing animals. This chapter is very different because it contains a different kind of logic; one that is a far leap from the random chaos of genetic mutation and protein manipulation.
This is an impressive work covering lots of hard biological scientific examinations, of which I am not expert enough to really criticize. The methodology is pretty simple though. Kauffman provides lots of computer models to try and show that mutation can give rise to order. This is something that can be controlled by computers because natural selection criteria is "given" in any examination -- he assumes that it works because any odd selection process could work, we don't need to concern ourselves about that, we only need to see that genetic mutation can accommodate any given selection process... he finds some interesting examples, but ultimately concludes that local valleys/peaks in optimization is sufficient to forestall mutation and that natural selection cannot easily overcome such leaps in development. So that suggests there is a limit on how species can optimize even though niches can slide.
This actually kind of destroys the notion of progress a bit, because there is no "higher" change only that there is change.
What is interesting in this book is that Kauffman seems to acknowledge implicitly that cell differentiation and genetic mutation are potentially two different logics; how they coordinate is perhaps largely unknown because early cell differentiation in early (pre)fetal stages is completely unknown. He nonetheless provides a chapter in which he speculates based on experimentation, how that differentiation could occur but there seems to need to be some kind of coordination between cell signalling for this to happen. Kauffman is caught up in the mechanization for organization -- tiny incremental changes however seem completely indeterminate. What might be interesting is to model even the cell differentiation mechanism in species development, as if a computer could trace various mutations in a population as it impacts unborn animals. That in itself would probably eliminate the differentiation as a separate mechanism because everything remains controlled by genetics.
In some sense though, Kauffman perhaps lacks the appropriate language for this but I don't know. There seems to be too many unknown and moving parts because the context of the cell itself is self-created via the genetics. As he argues, it is not impossible for every possible cell to be viable as many states of the cell available through genetics won't be functional or even worse -- be completely cancerous.
How this is managed and sorted out is still a mystery, although Schrodinger provides a very interesting view, one in which thermodynamic impulses drive evolution as well, as energetic impulses continually look for more degrees of freedom to be expressed, something that happens regardless of what other constraints (environmental, or otherwise) may have additional influences.
Most of this book is about this kind of modeling. He then ends the book with a chapter on cell differentiation, as it occurs in different zygotes of developing animals. This chapter is very different because it contains a different kind of logic; one that is a far leap from the random chaos of genetic mutation and protein manipulation.
This is an impressive work covering lots of hard biological scientific examinations, of which I am not expert enough to really criticize. The methodology is pretty simple though. Kauffman provides lots of computer models to try and show that mutation can give rise to order. This is something that can be controlled by computers because natural selection criteria is "given" in any examination -- he assumes that it works because any odd selection process could work, we don't need to concern ourselves about that, we only need to see that genetic mutation can accommodate any given selection process... he finds some interesting examples, but ultimately concludes that local valleys/peaks in optimization is sufficient to forestall mutation and that natural selection cannot easily overcome such leaps in development. So that suggests there is a limit on how species can optimize even though niches can slide.
This actually kind of destroys the notion of progress a bit, because there is no "higher" change only that there is change.
What is interesting in this book is that Kauffman seems to acknowledge implicitly that cell differentiation and genetic mutation are potentially two different logics; how they coordinate is perhaps largely unknown because early cell differentiation in early (pre)fetal stages is completely unknown. He nonetheless provides a chapter in which he speculates based on experimentation, how that differentiation could occur but there seems to need to be some kind of coordination between cell signalling for this to happen. Kauffman is caught up in the mechanization for organization -- tiny incremental changes however seem completely indeterminate. What might be interesting is to model even the cell differentiation mechanism in species development, as if a computer could trace various mutations in a population as it impacts unborn animals. That in itself would probably eliminate the differentiation as a separate mechanism because everything remains controlled by genetics.
In some sense though, Kauffman perhaps lacks the appropriate language for this but I don't know. There seems to be too many unknown and moving parts because the context of the cell itself is self-created via the genetics. As he argues, it is not impossible for every possible cell to be viable as many states of the cell available through genetics won't be functional or even worse -- be completely cancerous.
How this is managed and sorted out is still a mystery, although Schrodinger provides a very interesting view, one in which thermodynamic impulses drive evolution as well, as energetic impulses continually look for more degrees of freedom to be expressed, something that happens regardless of what other constraints (environmental, or otherwise) may have additional influences.
May 14, 2016
Way over my level. Scientific work.
January 3, 2020
Part of the book is too difficult for me. The chapters I was interested in and I could understand were worth it.
July 29, 2016
Can I give this 6 stars? It's tough going, but revolutionary.
Read
May 16, 2025This was a very challenging book and was outside my comfort zone. However, the below quoted section was worth writing out and I think will be handy later. Also, hopefully useful for anyone looking for a cursory sketch of some of the ideas.
The jump into fitness 'spaces' that were uninhabited (and unreachable due to x/y/z lack of expressions) was novel to me and applies to newer forms of matching down the line. The comparison between extinctions (climate driven) and explosions (info expression) gave me something to chew on.
One really big aha was this: comparing the 'travelling salesman' problem to evolution. In one modality, there is a constant re-calibration of the next fastest path, with the salesman model moving to a new area and then correcting for its inefficiencies due to the leaps; VS. slow exploration along current-fitness paths in less rugged (a.k.a. better known) landscapes but without as much correction was an interesting metaphor I had not seen before.
"Analysis of the generic properties of adaptation on rugged but correlated fitness landscapes suggests an alternative hypothesis: the Cambrian-Permian asymmetry may be a direct expression of adaptation on such landscapes. There are three basic ideas in such a framework:
1. Adaptive evolution occurs on rugged fitness landscapes.
2. The rape of finding successively fitter variants decreases very rapidly both on very rugged landscapes and on smoother landscapes explored by long jumps; in contrast, the rate of finding successively fitter nearby variants decreases gradually on smoother landscapes.
3. Mutants affecting early ontogeny typically have more profound consequences than mutants affecting late ontogeny [a.k.a. biological complexity...]. Thus the rate of finding fitter mutants altering early development and fundamental baupläne decreases more rapidly than the rate of finding fitter mutants affecting late ontogeny.
Together these ideas imply that early, poorly fit multicellular organisms could rapidly explore a large diversity of improved alternative basic morphologies [a.k.a. the Cambrian explosion], thereby establishing phyla. As the rate of finding fitter mutants altering early ontogeny deceases but fitter variants affecing later ontogeny were still readily found, variant species founding classes, orders, and the lower taxa became established. Taxa filled in from the top down. In contrast, by the Permian extinction, early ontogeny was large frozen, and so fitter variants altering fundamental baupläne became very hard to find. But mutants altering late ontogency remained easier to find. New genera and families arose. Taxa replenished from the bottom up.
(Kauffman 76)
And later:
"Thus again I suggest that life may have started with a critical complexity, then simplified under selection. Afters its simplification, it appears to us as a marvelously improbably system, a whole whose chicken-and-egg self-referencing necessity mistakenly mystifies us." (242)
And lastly:
"There is a zome of tolerance for moderate errors of [RNA] translation, which tend to die out rahter than amplify. Outside that zone, or threshold fraction of erorrs, the translation system would indeed collapse to disorder. Conversely, the onset of coding would appear to require crossing the threshhold from below. This is what Bedian supplies." (365)
Bubbling up from below, slowly surpassing the watershed where errors impact the ability to consistently re-write (using the same aminoacyl synthetases) and capture the fitness data that came before it.
A language can be settled upon over time with sufficient error correction, so it is the monkeys at type-writers scenario. And the universe is very, very old, and yet still very, very young. And the time horizon shrinks when system re-write faster in new paradigms. Bingo.
The jump into fitness 'spaces' that were uninhabited (and unreachable due to x/y/z lack of expressions) was novel to me and applies to newer forms of matching down the line. The comparison between extinctions (climate driven) and explosions (info expression) gave me something to chew on.
One really big aha was this: comparing the 'travelling salesman' problem to evolution. In one modality, there is a constant re-calibration of the next fastest path, with the salesman model moving to a new area and then correcting for its inefficiencies due to the leaps; VS. slow exploration along current-fitness paths in less rugged (a.k.a. better known) landscapes but without as much correction was an interesting metaphor I had not seen before.
"Analysis of the generic properties of adaptation on rugged but correlated fitness landscapes suggests an alternative hypothesis: the Cambrian-Permian asymmetry may be a direct expression of adaptation on such landscapes. There are three basic ideas in such a framework:
1. Adaptive evolution occurs on rugged fitness landscapes.
2. The rape of finding successively fitter variants decreases very rapidly both on very rugged landscapes and on smoother landscapes explored by long jumps; in contrast, the rate of finding successively fitter nearby variants decreases gradually on smoother landscapes.
3. Mutants affecting early ontogeny typically have more profound consequences than mutants affecting late ontogeny [a.k.a. biological complexity...]. Thus the rate of finding fitter mutants altering early development and fundamental baupläne decreases more rapidly than the rate of finding fitter mutants affecting late ontogeny.
Together these ideas imply that early, poorly fit multicellular organisms could rapidly explore a large diversity of improved alternative basic morphologies [a.k.a. the Cambrian explosion], thereby establishing phyla. As the rate of finding fitter mutants altering early ontogeny deceases but fitter variants affecing later ontogeny were still readily found, variant species founding classes, orders, and the lower taxa became established. Taxa filled in from the top down. In contrast, by the Permian extinction, early ontogeny was large frozen, and so fitter variants altering fundamental baupläne became very hard to find. But mutants altering late ontogency remained easier to find. New genera and families arose. Taxa replenished from the bottom up.
(Kauffman 76)
And later:
"Thus again I suggest that life may have started with a critical complexity, then simplified under selection. Afters its simplification, it appears to us as a marvelously improbably system, a whole whose chicken-and-egg self-referencing necessity mistakenly mystifies us." (242)
And lastly:
"There is a zome of tolerance for moderate errors of [RNA] translation, which tend to die out rahter than amplify. Outside that zone, or threshold fraction of erorrs, the translation system would indeed collapse to disorder. Conversely, the onset of coding would appear to require crossing the threshhold from below. This is what Bedian supplies." (365)
Bubbling up from below, slowly surpassing the watershed where errors impact the ability to consistently re-write (using the same aminoacyl synthetases) and capture the fitness data that came before it.
A language can be settled upon over time with sufficient error correction, so it is the monkeys at type-writers scenario. And the universe is very, very old, and yet still very, very young. And the time horizon shrinks when system re-write faster in new paradigms. Bingo.
April 13, 2022
This work explores the interplay of natural selection and self-organization in evolution. Essentially: can order be achieved solely through selection on random mutations, or must there be some additional scaffolding, whether dynamical or structural, that anchors and helps direct the selective process? An important and tantalizing topic to be sure. There are some deep and transformative ideas in these pages, and for these the book has value. Unfortunately, the text is needlessly verbose, unnecessarily in the weeds, and has wildly inconsistent levels of sophistication of presentation.
The first part of the book, on numerical models of evolution and self-organization, requires some mathematical sophistication on the part of the reader, and they must know the very basics of biology and evolutionary theory. But besides that, it's self-contained and well-presented. The problem with this portion is that Kauffman reproduces virtually every plot and conclusion from his long series of academic works on these subjects. While these results are for the most part accessible to the reader (who has the pre-reqs listed above), most are really quite tangential to his main suite of points and merely serve to clutter and distract from the exposition. So while the text is geared towards your lay reader, the results and analysis are at the granular level of a conference proceeding or journal article. It strikes a dissonant chord and takes the wind out of one's sails.
The next major section of the book is about abiogenesis, or the origin of life processes on Earth. Also a topic of awesome import. For this portion, sadly, one needs a PhD in biochemistry to even get started. Kauffman seems entirely unaware of who is audience is or should be: after working through the first half of the book, Kauffman basically throws you out of the speeding car and drives off cackling madly.
While the first part of this book was thought-provoking and entertaining, it was not worth the cuts and bruises.
The first part of the book, on numerical models of evolution and self-organization, requires some mathematical sophistication on the part of the reader, and they must know the very basics of biology and evolutionary theory. But besides that, it's self-contained and well-presented. The problem with this portion is that Kauffman reproduces virtually every plot and conclusion from his long series of academic works on these subjects. While these results are for the most part accessible to the reader (who has the pre-reqs listed above), most are really quite tangential to his main suite of points and merely serve to clutter and distract from the exposition. So while the text is geared towards your lay reader, the results and analysis are at the granular level of a conference proceeding or journal article. It strikes a dissonant chord and takes the wind out of one's sails.
The next major section of the book is about abiogenesis, or the origin of life processes on Earth. Also a topic of awesome import. For this portion, sadly, one needs a PhD in biochemistry to even get started. Kauffman seems entirely unaware of who is audience is or should be: after working through the first half of the book, Kauffman basically throws you out of the speeding car and drives off cackling madly.
While the first part of this book was thought-provoking and entertaining, it was not worth the cuts and bruises.
July 15, 2025
This is a hefty book. My background in cognitive science provides a basic understanding of evolutionary biology; however, I found myself wishing I had a deeper knowledge of genetics and theoretical biology while reading this. Still, I find this a breathtaking feat. The book covers an incredible range of topics on self-organising complex systems, from Boolean networks and neural network models to morphogenesis and the dynamics of co-evolving systems. Kauffman has a visionary mind. He was among the pioneers in studying simplified computer models of genetic regulatory networks. His model, known as a Random Boolean Network (RBN), extends cellular automata. Through his research and simulations of RBN, Kauffman concluded that for an organism to be both alive and stable, its RBN must operate in a' liquid” regime, not too gaseous or frozen. His theory states that' Life exists on the edge of chaos.’ His main point is that natural selection isn't necessary to develop a complex organism. He believes that once a network reaches sufficient complexity, self-organising behaviour will emerge.
Want to read
May 23, 2023wisdom unlimited bibliography
Not in Libby or cv library
Not in Libby or cv library
October 6, 2022
this is a classic but very math-heavy that looks at the landscape of fitness upon which evolution works and shows that factors beyond natural selection (the mathematical landscape it plays out on) shape the forces of evolution. It uses the NK model for the landscape with N components to work on in any kind of fitness machine and how much its parts depend on each other or are coupled in fitness. If the K is low like zero or one it creates a smooth hill over the landscape where it is easy for a random walk of mutation to reach high peaks of fitness. If K is high or higher than two we get a rugged landscape with many peaks and valleys on this landscape a random walk is overwhelming likely to get stuck on a suboptimal local peak and not reach the heights of fitness. But if the coupling is kept down to 2 it hits a sweet spot where high peaks.
The latter part of the book explores the amount of randomness and how much of it is needed for good outcomes. It seems there is a critical level of randomness and staying close to that edge sees more development and variegated outcomes just at the edge of boiling chaos. Not freezing like a solid or too chaotic like gas but more akin to a liquid state some order but with enough randomness to make for variety and diversity.
the book then covers attractive basins in the landscape and attractors systems can fall into. On such attractor is a stable point but others are cycles of attraction a cycle of states that processes circle around. Again Kaufman reveals sweet spots if the landscape has the right parameters. Showing the mathematical background of the landscape inherent in the processes (boolean networks with random connections but again with different numbers of connections between nodes where again K=2 seems to bring about remarkable limit cycles as basins of attraction) makes for a fruitful natural selection. Namely, there is more to life's ingredients (namely the right mathematical form from Plato's Plenum) than simply Darwin. Interesting and 2thought-provoking.
Update 10/4/2022 I am tired today I've been up all night. I will write a review later when I am more awake.
Update 10/5/2022 This is a book by Stuart Kauffman is highly influenced by the Santa Fe Institute. This was a complexity research outfit in the heyday of optimism around computers and free markets. I mean it is a science research laboratory but its fascination with Self-organizing systems fit like a glove on invisible hands and the magic of markets in the 1990s. I think there is something to spontaneous order arising out of less complex things like a conglomeration of independent entities interacting with each other but frankly, it chimed in well in the neoliberal nineties free-market fantasies. The material is interesting but it is used to bolster only one side of the story.
The latter part of the book explores the amount of randomness and how much of it is needed for good outcomes. It seems there is a critical level of randomness and staying close to that edge sees more development and variegated outcomes just at the edge of boiling chaos. Not freezing like a solid or too chaotic like gas but more akin to a liquid state some order but with enough randomness to make for variety and diversity.
the book then covers attractive basins in the landscape and attractors systems can fall into. On such attractor is a stable point but others are cycles of attraction a cycle of states that processes circle around. Again Kaufman reveals sweet spots if the landscape has the right parameters. Showing the mathematical background of the landscape inherent in the processes (boolean networks with random connections but again with different numbers of connections between nodes where again K=2 seems to bring about remarkable limit cycles as basins of attraction) makes for a fruitful natural selection. Namely, there is more to life's ingredients (namely the right mathematical form from Plato's Plenum) than simply Darwin. Interesting and 2thought-provoking.
Update 10/4/2022 I am tired today I've been up all night. I will write a review later when I am more awake.
Update 10/5/2022 This is a book by Stuart Kauffman is highly influenced by the Santa Fe Institute. This was a complexity research outfit in the heyday of optimism around computers and free markets. I mean it is a science research laboratory but its fascination with Self-organizing systems fit like a glove on invisible hands and the magic of markets in the 1990s. I think there is something to spontaneous order arising out of less complex things like a conglomeration of independent entities interacting with each other but frankly, it chimed in well in the neoliberal nineties free-market fantasies. The material is interesting but it is used to bolster only one side of the story.
Want to read
November 1, 2007Well after busting through the first 15 pages of very dry and condensed summary of the main theories being debated today in evolution, I've moved the book from my current read to my to-read shelf. I was curious about the book based on a cite from one of Dawkins books, and am interested in reading more about matter's inclination to organizing itself, and what this could imply not only in regards to additional factors in evolution (besides the most important natural selection) but also what this could imply regarding the very first forms of life-how life could have evolved from matter. The book also raises what could be constraints on evolution and also what are the patterns we can expect to evolve. Just to note not recommended to those intimidated by math.
Want to read
May 23, 2020recommendation: http://www.ynotds.com/
rec0mmendation: [Transcript of Episode 9 - Joe Norman - The Jim Rutt Show](https://jimruttshow.blubrry.net/the-j...)
recommendation: [EP18 Stuart Kauffman on Complexity, Biology & T.A.P. - The Jim Rutt Show](https://jimruttshow.blubrry.net/stuar...)
rec0mmendation: [Transcript of Episode 9 - Joe Norman - The Jim Rutt Show](https://jimruttshow.blubrry.net/the-j...)
recommendation: [EP18 Stuart Kauffman on Complexity, Biology & T.A.P. - The Jim Rutt Show](https://jimruttshow.blubrry.net/stuar...)
January 1, 2016
Thought provoking.
Displaying 1 - 13 of 13 reviews