Life’s complexity – even if we just stick to its basic core, the prokaryotic world – is overwhelming. This is not only due to the diversity of biomolecules and their amazing structural properties, or to the multiple and sophisticated transformations they undergo, but also to their dynamic organization and functional integration into cells. Let alone what these cells do and have been doing, together with and against other cells – and viruses, viroids, etc. – over thousands of millions of years, on the surface of a planet that they have completely remodelled. No big news, really: a ‘take-home message’ that I have often repeated, in papers and talks. For most of you who are starting to read these lines, it all sounds familiar, I assume. Yet, I will not get tired of putting the message across, because its implications are huge, and I will discuss some of them below, in relation to the process of primordial biogenesis. For starters, though, it is important to focus back on cells, the main actors in the biological scene, and ruminate a bit more on what happens at that level. Each living cell, in its continuous performance of metabolism, constitutes an ‘ontological surprise’, as Jonas [1966] so nicely and wisely captured. Indeed, nothing in human’s extensive knowledge about physical and chemical systems (including rather intricate, non-equilibrium, ‘self-organizing’ processes [Prigogine 1980]) suggests, even remotely, the feasibility of cellular physiology, as we observe it in nature. Nothing in principle excludes it; but nothing hints at it, either.
The metabolic activity that cells constantly carry out is thermodynamically uphill: it implies an enormous material and energetic cost (and that’s why cells are not merely open systems, but agents that actively extract resources from their local environments and transform them into their own means). This activity, precisely, is critical to determine their intrinsic nature and the relationship they maintain – and modulate – with their surrounding medium. In some way, metabolism affords cells the very possibility to be themselves: i.e., the possibility to be free, to be autonomous in a primary, constitutional sense. Although it could be argued that any extant individual cell derives from a previously existing one, and that it heavily relies on the functioning of many others, in parallel (as well as on those matter/energy resources just mentioned, to make it viable), the naturalization of theoretical constructs like ‘freedom’ (again, [Jonas 1966]) or ‘autonomy’ [Varela 1979], grounding them specifically at the cellular level, has been truly insightful for many of us.[1] So to speak, biology takes place within but, at the same time, beyond physics and chemistry: it belongs to the latter, no doubts about that; but it has ‘self-emancipated’ to a remarkable extent.
There is no need to get wet in a philosophical debate about the different types of causes that govern biological phenomena, when science itself is not providing a complete nor coherent causal picture of the world – and less so of the living world. I prefer to leave things open in that regard and simply claim that, indeed, ‘multiple and parallel causality’ (which is still to be adequately explored and characterized) lies at the heart of life’s irreducibility. In line with Polanyi [1968], I would propose that the potential of physics and chemistry to generate biological ‘hyper-complexity’ is realized, in practice, when matter implements constraining mechanisms leading to ‘dual-control’ systems. According to that classical idea, living cells, on top of obeying the universal laws of physics and chemistry, like any other natural system, are characterized by their capacity to generate and maintain their own boundary conditions, their local rules of behaviour. At this fundamental, constitutive level, such rules are essentially molecular and supra-molecular mechanisms that both constrain and enable the dynamics of other components and transformation processes pertaining to the system. As a result, cellular organization, articulated around its metabolic activity by means of a suite of diverse self-constraining mechanisms (spatial, kinetic and energetic control mechanisms, among others), is to be regarded as the kernel of life, the main axis of all biological phenomenology, and surely the key to decipher its emergence [Lauber et al. 2021].
However, minimal living cells (bacteria/archaea) are way more complex than what the principles of autonomy seem to require for an elementary metabolism. ‘Genetically-instructed’ cellular metabolisms (involving the materially and energetically extravagant synthesis of a whole suite of interdependent macromolecules: DNA, RNA and proteins) appear to be very elaborate instances of ‘autonomous molecular agents’ (as coined by Kauffman [2003]). Despite the lack of empirical evidence on prebiotic (or infra-biological) autonomous systems, a number of theoretical results point clearly in this direction, including our own protocell modelling work over the years (e.g.: [Ruiz-Mirazo & Mavelli 2008]; [Piedrafita et al. 2017]). Anyhow, such a ‘complexity gap’ is too evident to be ignored, on bare conceptual grounds, and those authors who do not consider genetic mechanisms as relevant for minimal biological organization (following Varela’s or Rosen’s premises, on this matter) might be overlooking something big there. While we wait for the field of origins-of-life (or synthetic biology) to solve the question about autonomous systems below the ‘minimal-cell threshold’, two alternative positions can be adopted: (i) take this apparent gap as a historical/evolutionary accident (i.e., all living cells just happen to be at least as complex as current terrestrial prokaryotes); or (ii) take it as a necessary historical/evolutionary outcome (i.e., all living cells need to be, here and anywhere else in the universe, as complex as current terrestrial prokaryotes).[2]
I imagine Rosen and Varela (and their close advocates) would still opt for (i); I definitely favour (ii). The reasons behind are better explained in articles on the definition of life [Ruiz-Mirazo et al. 2010], and on how to approach the problem of biogenesis [Ruiz-Mirazo et al. 2020], which highlight the collective-ecological and the historical-evolutionary dimensions of both problems (i.e., understanding the nature of life and its emergence). Here, in the remaining of this brief entry, I would simply like to focus on the implications that this has for biological autonomy, and how these implications can be interpreted in dialectic terms. So let us take as working hypotheses the following two ideas. First, that autonomy is an adequate theoretical construct to capture biological individuality, realized – in its minimal and most universal expression – as unicellular (prokaryotic) organisms. Second, that autonomy is also bound to illuminate fundamental transitions in the process of origins: in other words, that different forms of autonomy, simpler than unicellular (prokaryotic) organisms, are not only feasible in the world of physics and chemistry, but necessary to pave the way towards biology. Under these two general assumptions, it seems more than reasonable to put forward a prebiotic scenario in which autonomy unfolds, from relatively simpler instances, towards more and more complex realizations of it.

Now, this unfolding process involves a complex historical process (primordial biogenesis) in which protocells, in order to overcome major bottlenecks and look increasingly similar to full-fledged biological cells, must engage in interactive dynamics in – at least – 3 different spheres: [1] first, as far-from-equilibrium systems, in their continuous coupling to the environment (by means of a self-made and self-regulated active boundary); [2] second, with other protocells in the population, with which they establish both cooperative and competitive relationships (leading to ‘proto-ecological’ networks); and [3] third, with their own progeny, their clan, through precursor but sufficiently reliable reproduction and hereditary mechanisms (implementing the first ‘proto-phylogenies’). In [Ruiz-Mirazo et al. 2020] we give a more detailed explanation about the relevance of these interactive dynamics, and how they transform the identity of the protocells involved in the process of biogenesis. So much so that, by the end of the process, it is difficult to speak about identities, in singular, anymore. Rather, we argue there for a change of perspective that conceives the identity of the resulting entities, more comprehensively, as a collectively and historically constructed ‘inter-identity’.
Well, I guess that all of you, intelligent and philosophically well-educated readers of this blog, will have started intuiting the parallelism I am about to draw. These interactive relationships can naturally – and, in fact, quite meaningfully – be interpreted as dialectical exchanges through which protocells continuously resolve the tension between maintaining and transforming their identities (their autonomy) as individuals. The triggers of such an internal tension are the different ways in which protocell integrity is challenged (thermodynamic, ecological, reproductive…) and, in return, protocell active responses to those challenges bring forward a self-determination process that never gets fully accomplished: namely, a constant being-through-becoming process. The trickiest part of the problem, in any case, is to understand how this process expands, both spatially and temporally, far beyond the individual activity of forerunner protocells, with deep consequences on the identity/autonomy of the resulting (biological) cells. Here lies the key issue, in my view, and solving it requires a synthesis between physiological, ecological and evolutionary principles, to be applied to the origins of life – and, possibly, to start constructing a general theory of biology. Not a trivial task, for sure, but some of us keep our minds busy trying to figure out how that could be tackled. In the meantime, let me just highlight that, in this richer context, the ‘autonomy account’ for biogenesis turns definitely more Hegelian than – merely – Kantian!
Prebiotic evolution is quite idiosyncratic, compared to what happens later (throughout proper biological evolution), because it culminates in a singularity: life’s minimal complexity threshold. There are good reasons to believe that any process of origins of life must have the shape of a funnel (see Fig. 2 of [Ruiz-Mirazo et al. 2020]), no matter the starting conditions, or the planetary environments where it might develop. Von Neumann already gave us some general insights in that direction, before the discovery of the genetic code, when he reflected on how a universal constructor should be designed in order to avoid eventual decay across an indefinite sequence of self-reproductive steps (see McMullin [2000] for a nice interpretation of von Neumann’s legacy). If one follows that line of reasoning (and adds material and thermodynamic requirements to the equation), the conditions for autonomous systems to acquire ‘open-ended evolution capacities’ (and, thereby, also to complete their ‘self-emancipation’ pathway) come out to be quite stringent [Ruiz-Mirazo et al. 2008]. Moreover, from a top-down (or backward-looking) perspective, there is a wide range of basic biochemical features, shared by all living cells (in particular: a common genetic code, universal energy currencies, the homochirality of some biomolecules), which also indicate that – at least several – prebiotic transitions implied narrow bottlenecks. It is always risky to generalize from single examples (in this case: life’s emergence and properties on planet Earth). Similarly, the important roles of random effects and stochasticity in some of these processes introduce difficulties to reconstruct those prebiotic transitions as a neat ‘causal-chain of events’. However, even if the historical dimension of primordial biogenesis involved contingencies and ‘frozen accidents’ (for instance, the specific genetic code that we know for terrestrial life) some other features may still be ‘general truths’ for any biology (for instance, the need for a translation code through which cells can functionally interpret informational records [Pattee 1977]).
Thus, back to the central idea, to conclude: if all this story was true, the historical unfolding of autonomy throughout prebiotic evolution would look dreadfully Hegelian, in so far as it involves a collective effort taking place across many protocell generations, developing in a sort of ‘constant transformative tension’ and leading to higher levels of freedom/emancipation from physics and chemistry – thanks, essentially, to the implementation of more and more robust forms of metabolism. Along that process, diverse expressions of autonomy are explored – the majority of which eventually decay, by themselves or in interaction with others. Nevertheless, all of them important, as long as they contribute to keep open the pathway towards a population of ‘genetically-instructed’ cellular metabolisms (roughly analogous to LUCA, the so-called ‘last universal common ancestor’). There will never be a way to demonstrate it (and less so if biology is not fully formalizable, as I believe – in that sense, perhaps not everything is so Hegelian, after all: universality does not imply a single/unified logic); but I bet all my chips that any other material excursion into the ‘hyper-complexity’ of a living domain should involve a similar collective and historical effort: a truly dialectical effort performed by precursor autonomous entities, in their becoming sufficiently free from the physical/chemical burdens of the natural world to be able to begin an emancipated pathway, within it.
To my knowledge, Hegel never met Kant, a big figure by the time they could have exchanged ideas and perspective. What I know, for certain, is that Moreno did encounter Varela, and I am sure they debated on the relevance of genetics and informational records for biological organization… so, at least, he had the opportunity to tell the preceding figure that ‘the autonomy account’ he was offering for biology could benefit from a significantly more complex reading (see: [Moreno & Mossio 2015] and, of course, more to come!).
References
Jonas, H. (1966): The Phenomenon of Life: Toward a Philosophical Biology. Harper & Row. New York.
Kauffman S. (2003). Molecular autonomous agents. Philosophical transactions. Series A, Mathematical, physical, and engineering sciences, 361(1807), 1089–1099.
Lauber, N., Flamm, C. & Ruiz-Mirazo, K. (2021): ‘Minimal metabolism’: a key concept to investigate the origins and nature of biological systems. BioEssays 43: 2100103.
McMullin, B. (2000): John von Neumann and the evolutionary growth of complexity: looking backward, looking forward. Artificial Life 6, 347-361.
Moreno, A. & Mossio, M. (2015): Biological Autonomy. A Philosophical and Theoretical Enquiry. Springer. New York.
Pattee, H. H. (1977): Dynamic and linguistic modes of complex systems. International Journal of General Systems 3, 259-266.
Piedrafita, G., Monnard, P.-A., Mavelli, F. & Ruiz-Mirazo, K. (2017): Permeability-driven selection in a semi-empirical protocell model: the roots of prebiotic systems evolution. Scientific Reports 7: 3141.
Polanyi, M. (1968): Life’s irreducible structure. Science 160, 1308-1312.
Prigogine, I. (1980): From being to becoming: time and complexity in the physical sciences. Freeman, New York.
Ruiz-Mirazo, K. & Mavelli, F. (2008): On the way towards ‘basic autonomous agents’: stochastic simulations of minimal lipid-peptide cells. BioSystems 91(2): 374-387.
Ruiz-Mirazo, K., Peretó, J. & Moreno, A. (2010): Defining life or bringing biology to life. Origins of Life and Evolution of the Biosphere 40: 203-213.
Ruiz-Mirazo, K., Shirt-Ediss B., Escribano-Cabeza M. & Moreno, A. (2020): The construction of biological ‘inter-identity’ as the outcome of a complex process of protocell development in prebiotic evolution. Frontiers in Physiology — Systems Biology 11: 530.
Ruiz-Mirazo, K., Umerez, J. & Moreno, A. (2008): Enabling conditions for ‘open-ended evolution’. Biology & Philosophy 23(1): 67-85.
Varela, F. J. (1979): Principles of Biological Autonomy. Elsevier, New York.
Weber, A. & Varela, F.J. (2002): Life after Kant: Natural purposes and the autopoietic foundations of biological individuality. Phenomenology and the Cognitive Sciences 1, 97–125.
[1] See [Weber & Varela 2002] to appreciate how closely related these ideas are — while also connected historically to a Kantian understanding of natural purpose and intrinsic teleology.
[2] One could also consider, of course, that the complexity of DNA-RNA-protein-based life is grounded on strict physiological principles, without appealing to history, or evolution to account for it (for instance, arguing that it is required, from the very first prebiotic steps, to put together minimally efficient metabolisms). However, that scenario would make the emergence of prokaryotic life implausible, in practice (too sudden, almost miraculous). In addition, like option (i) above, it would mean an obvious ‘lost opportunity’ epistemologically speaking, since it would restrict, for no good reason, the ‘explanatory toolkit’ available to make sense of biological phenomena.
Yordan Yordanov says:
Running the risk to disturb the tone of this blog I shall put forward the hypothesis that there was NO PRE-biotic evolution. There are only chemical processes understood well within both equilibrium and non-equilibrium thermodynamics and Life-an attractor capable of self-maintanence in the face of the laws of thermodynamics mentioned above. Therefore, the entire thesis of a “dialectical turns” of something which was already within its full functionality ever since its beginning lacks any base in physical reality.
Life exists as a negation to the Second law of thermodynamics and the principles of maximazing the dissipation of free-energy which it implores. These are facts known ever since Schrodinger. This entails a beginning of the phenomenon as a sharp distinction from the inorganic world meaning the first, even the simplest, forms of life came about with all the defining properties of Life CO-emerging out of each other-Life comes as a “package”, not as a gradient. All characteristics attributable to it-be it metabolism, the genetic code, agency and so on are CO-emergent. You have a supramolecular entity having all those at once as a result of its active resistance to the Second law of thermodynamics and the maximazing entropy it emplores. You don’t have a gradient of “pre”-biotical systems with “degrees” of autonomy, complexity or anything else. Of course, the most primitive form of life is far simpler than what we can encounter today but even it-a simple supramolecular complex capable of coded self-replication-has the defining properties of Life in the form of negentropy, evolution, genetic “memory”, agency and so on. Life emerges far earlier down the path than minimal complexity threshold would imply. It’s just that the definition of Life implored here isn’t sufficiently advanced to capture it.
Lifeless chemistry is bound to search for paths to maximize the dissipation of free energy. That’s the very definition of dissipative structure. However, contrary to what the majority of current scientists believe, Life isn’t a dissipative structure. At least, not in the classical sense of the term because the living matter strives to conserve, rather than dissipate said energy. That’s the very difference between Life and non-life. It’s, however, quite rare to see someone put it as straightforward as I do now. In order to conserve said energy Life must adapt its structure in ways contrary to the path of minimum resistance to the dissipation of energy followed by all other matter. That means it must acquire agency and individuality from its very start. It is born “perfect” in a sense. Everything else before it is dead matter blindly following the laws of physics and chemistry. It, however, transcends those laws precisely by utilizing them against themselves-it’s a low probability structure which becomes dominant by exhausting energy for its self-replication. As mentioned above it resists these laws by imploring them. But in order to resist it needs an unique structure and a mechanism to maintain it-that’s what its genetic “memory” and code, in any form they may ensue. are for. In essence the start of the code is the start of Life and only a single supramolecular structure is required for it, not an entire protocell and even less a population of them. Those things are secondary to life, not primordial to it.
Thus, I am afraid I have to disagree with the article above on the base of a false premise-there is no pre-biotic evolution because life has already well started by the time first cells appear. Molecules, or to be precise-supramolecular complexes with an established structure defying the most efficient way to dissipate energy, catalyzing a web of reactions leading to their own synthesis and the synthesis of their precursors, heredity and a specific mechanisms for its transmission to a progeny already existed. The problem is our definition of Life should be corrected in order to accommodate such an early beginnings of the phenomenon.
Kepa Ruiz-Mirazo says:
All feedback is welcome here, Yordan… even critical comments, like yours, that express the view that a complete amendment of the conjecture is necessary 😉 Excuses for the delay in my answer — I did not check the entry for more than a year! 🙁
Well, after reading your lines, I am just looking forward to a more complete explanation (perhaps better in a paper, or in a book — not in this blog, devoted to “dialectical systems”) about: (i) the first sudden transition you speak about, from physics and chemistry, to those supramolecular complexes that you already consider “alive”; (ii) the subsequent transitions towards prokaryotic organisms, mentioning (if possible) the steps and reasons behind cellularization. In this way, you will demonstrate that such a scheme is more feasible than the protocell scenario I support. By the way, don’t forget to include your own definition of life in the explanation, because it will add relevant information about your conception of the process, I am sure!
Alvaro Moreno says:
Well, I certainly met Varela and had the opportunity to learn a lot from him. At that time, to be frank, I did not have a great opinion of Kant’s and Hegel’s conception of the Natural world. Partly because of my ignorance about what those great minds had intuited but I guess it was more due to my youthful enthusiasm for the importance and scope of the new idea: biological autonomy. Nevertheless, as one gets older and gains perspective, a more critical and complete picture develops about the breakthroughs that we were witnessing and embracing at the time. So I came to realize that Varela’s theory on biological autonomy was neither as original as I initially thought (since Kant had already advanced it), nor was it that simple or self-sufficient. This added complexity is wonderfully explained by Kepa. And the rest, my friends, is new continent waiting for its discoverers.
Kepa Ruiz-Mirazo says:
Thanks, Alvaro, for your nice comment… but, more importantly, for opening my eyes to that vast continent of complexity. Let’s keep exploring it! 😉
Matt Segall says:
Lovely thoughts here. I wonder, though, if Whitehead might be the better dialectician here, given Hegel’s strong predilection for logical rather than actual evolution? Didn’t Hegel reject the idea of a temporal evolution of life? (on this latter question, this old blog post comparing Hegel and Whitehead on this question may be of interest: https://footnotes2plato.com/2011/07/12/platonic-realism-in-whitehead-hegel-and-schelling/)
I am also curious to know if you are familiar with Deamer and Damer’s “hot spring hypothesis” for the origin of life? Sounds to me like they may have a probable empirical scenario for the sort of funnel you are describing. Damer and I coauthored a chapter (in press) trying to make the case that Whitehead provides the metaphysics required to grasp the complex dialectics of abiogenesis.
Thanks for this post.
-Matt
Kepa Ruiz-Mirazo says:
Thanks to you, Matt, for the kind and pertinent comment.
I take note of your suggestion to look into Whitehead (leaving it to the experts on Hegel whether he rejected a “temporal evolution of life” or not — but agreeing, in any case, that his main focus was on “logical evolution”, understanding logic in a strongly naturalized way).
Regarding Deamer & Damer’s “hot spring hypothesis”, I think it does fit within our general theoretical framework — although other experimental scenarios proposed in the literature do, as well. In order to argue for the “funnel shape” of our figure, one needs to reflect deeply on the process of biogenesis as an evolutionary development of protocell populations (especially during late stages/transitions of the process, which are still very far from empirical testing). In their 2020 Astrobiology article, if I remember well, they do not reason so much on these lines (like we do — check the various references included in the post, but [Ruiz-Mirazo et al. 2008; 2020], in particular).
Cheers!