Review | The Nature of Organization in Living Systems

Introduction

The question of what constitutes an organism has occupied thinkers from various traditions over centuries. Immanuel Kant was among the first to formulate a coherent philosophical inquiry into the nature of organisms. In his Critique of Judgment (1790), Kant described living systems as self-organized entities, where each part exists both for and by the system. This duality, where the parts serve structural and functional roles simultaneously, is a recurring theme in the history of theoretical biology. Kant’s work laid a conceptual foundation upon which later theorists such as Rashevsky, Rosen, Maturana, and Varela would build.

Rashevsky and Rosen developed relational biology, which sought to describe organisms by focusing on their organizational structure rather than their specific material components. Robert Rosen’s (M,R)-systems, introduced in 1958, modeled biological systems using categories and natural equivalence. His work emphasized the importance of capturing the causal structure of organisms rather than merely their material composition, becoming an epistemological infrastructure for modeling life. Simultaneously, Maturana and Varela introduced the concept of autopoiesis in the 1970s, which characterized living systems as autonomous units that produce and maintain themselves through a network of processes. Although they emphasized productivity, their approach was also criticized for neglecting the consumptive aspects of biological organization.

Currently, in a recent paper entitled The Nature of Organization in Living Systems, the authors aim to revisit and expand upon these ideas by emphasizing the often-overlooked role of consumptive processes—what they term autoanalysis. This contrasts with the traditional focus on production and synthesis, and it is essential for maintaining self-organization. Today I am going to review such a paper, pointing out some gaps I could detect and comparing it with other approaches in relational biology. For a further discussion on this topic you could visit our recent trilogy. Click here to see Chapter 1, 2 and 3.

How to formalize life?

The authors propose a formal framework to study organisms that is both conceptual and mathematical. Their approach focuses on the dynamics of transformation processes that occur within living systems. Unlike many traditional models, which impose physical and thermodynamic constraints, their formalism is general enough, intended to be broadly applicable across various levels of biological organization, from molecules to multicellular organisms.

In their approach, entities and their transformations are represented as directed bipartite multigraphs, where nodes can correspond to entities or transformations. This approach is reminiscent of Rosen’s (M,R)-systems, which before using categorical structures originally used graphs to represent processes and their interactions. Rosen’s approach was grounded in the use of categories and natural equivalence to describe the relationship between entities and processes. His aim was to formalize the causal structure of organisms through a categorical framework, emphasizing the mapping between different organizational levels and their transformations.

The authors emphasize that their framework avoids privileging any particular scale of biological organization. Processes are represented in such a way that they can be applied to various levels of biological complexity. This is compatible with Noble’s biological relativity, which is the principle that no single level of biological organization (e.g., genes, cells, organs) is inherently privileged in terms of causation, and all levels are essential for the functioning of the whole. This notion was recently reintroduced into literature as causal spreading.

Moreover, in their work the temporal aspect of these processes is left intentionally unspecified to maintain generality. This points towards a potential weakness in the authors’ framework: their emphasis on abstract, time-independent representations may overlook the dynamical properties that are essential for the self-maintenance and evolution of living systems. In contrast, we can find similar approaches such as the closure of constraints proposed by Montévil and Mossio. In the latter, closure refers to a network of constraints with an associated temporal scale that mutually maintain one another, establishing a form of systemic cohesion that enables the persistence of biological systems over time.

Some core concepts

The authors present a detailed conceptualization of several fundamental ideas that characterize living systems. These notions, which are meant to capture the essential dynamical properties of biological organization, include:

1. Conservation: In the context of the authors’ framework, conservation refers to the preservation of specific entities throughout processes. Catalysts are viewed as entities that remain unchanged across transformations, allowing processes to unfold without their structural integrity being altered. This concept is related to Rosen’s earlier work on relational biology, where conserved entities contribute to maintaining the causal structure of an organism. The authors argue that conservation is not necessarily absolute; in open systems, weaker notions of conservation are relevant. They emphasize that some parts of a system may remain unchanged even as the system as a whole undergoes dynamic transformations.
2. Cyclicity: Processes that exhibit cyclicity are essential for maintaining the organization of a living system. Cyclicity in their text is connected to the idea of ‘organizers’, which are structures that simultaneously exhibit conservation and cyclical behavior. This concept is reminiscent of Rosen’s idea of metabolic-repair systems, where processes are arranged in cycles that ensure the system’s persistence. The authors argue that these cycles are not mere repetitions but rather essential dynamical motifs that sustain the organization of the living system.
3. Autonomy: Autonomy arises from the interaction between conservation and cyclicity. The authors define autonomy as the capacity of a system to maintain itself through the continuous interplay of conserved entities and cyclic processes. This perspective aligns with previous work on autopoiesis but expands it by acknowledging the necessity of both production and consumption processes. Autonomy is not a static property; rather, it emerges from the continuous dynamical activity of the system.
4. Autosynthesis: Autosynthesis refers to the network of entities and transformations that collectively ensure the self-production of the system’s components. This idea parallels autopoietic models but distinguishes itself by explicitly including a broader range of transformations that contribute to the generation of new components. Autosynthesis is fundamental for understanding how living systems sustain themselves despite the constant turnover of their material constituents.
5. Autoanalysis: This concept, which the authors introduce as a complement to autosynthesis, refers to controlled degradation or consumption processes necessary to maintain self-organization. In living systems, processes of degradation, dissolution, or breakdown are not random but are actively regulated to contribute to the overall organizational structure. In this context, autoanalysis is particularly important for maintaining dynamical stability and resilience, preventing the system from growing unchecked or becoming overly rigid.

The combination of these concepts allows the authors to define self-organization as a process wherein autosynthetic and autoanalytic mechanisms coalesce, producing a resilient and dynamically stable system. This conception is in line with their broader aim of developing a general framework for understanding life that transcends specific material substrates. However, how true is that the substrate does not matter?

Conclusion

The authors’ attempt to formalize the concept of biological organization is both ambitious and thought-provoking. By emphasizing the importance of autoanalysis alongside autosynthesis, they provide a novel perspective that diverges from previous theories, mainly tracing back to the work by Rashesky-Rosen and Maturana-Varela. However, several challenges remain, particularly in clarifying how their framework integrates with existing theories of organizational closure and constraint-based approaches. Relational biology has not remained static since Rosen’s death, so it is important not to reinvent the wheel but focus on synthesizing what has already been done so far in this century.

Moreover, while their formalism is intended to be broadly applicable, the lack of explicit temporal modeling may limit its utility in describing real biological processes. In principle, it seems like another relational model invariant in time and incapable of evolving. If not, where are the evolutionary and anticipatory aspects of life in this new approach? The authors’ decision to avoid detailed physical constraints may also be a double-edged sword, as it offers generality at the expense of specificity. Overall, The Nature of Organization in Living Systems is a significant contribution to the ongoing discussion of what it means for a system to be alive. It offers valuable insights that merit further exploration and integration with other formal approaches in theoretical biology.