Query on Twitter:

https://x.com/FibraDomingos/status/2015231522089910656

I got a good response from ChatGPT so am sharing here in the hope of provoking some more:

There is a small but real ‘bridge literature’ that takes Rosen’s notion of complexity (impredicativity, closure to efficient causation, and the modelling relation) and places it into live debates in philosophy of science (models vs simulations, mechanism vs organisation, computability, self-reference, and the status of explanation in biology).

If you want direct, explicit connections, these are the most on-the-nose starting points.

Christensen and Hooker, ‘Anticipatory systems and time: a new look at Rosennean complexity’ (2005). This is explicitly pitched as “consequences … on science as a whole” and works through how anticipation forces a rethink of time, causality, and what counts as an adequate scientific perspective for biology. (Wiley Online Library)

Cárdenas, Letelier, Gutiérrez, Cornish-Bowden, Soto-Andrade, ‘Closure to efficient causation, computability and artificial life’ (Journal of Theoretical Biology, 2009). This is one of the clearest places where Rosen’s “organisms aren’t mechanisms / no simulable model” claim is treated as a philosophy-of-science issue about computation, modelling vs simulation, and what “closure” amounts to logically. (DCC Universidad de Chile)

Siekmann, ‘An applied mathematician’s perspective on Rosennean complexity’ (2017). Not philosophy-of-science in the narrow sense, but it directly reframes Rosen’s anti-mechanism stance as a claim about a particular kind of mechanistic modelling, and so lands right in the mechanistic explanation debate (what mechanisms can and can’t capture). (ScienceDirect)

On the “Rosen ↔ autopoiesis / second-order cybernetics / closure” axis (which is probably the most travelled bridge into philosophy of science of biology):

Recent autopoiesis work that explicitly uses Rosen’s modelling-relation framing to sort out what is “natural system” vs “formal system”, and relates operational closure in autopoiesis to Rosen’s closure to efficient causation. This is squarely about explanatory status (definition, theory, model) and how to interpret formalisations. (ScienceDirect)

For the “Rosen ↔ self-reference / impredicativity / logic” axis:

The Springer chapter ‘Impredicativity, Dynamics, and the Perception–Action Divide’ (2003) treats Rosen’s ‘epistemology of complexity’ as a central source and ties it to broader work on impredicativity and self-reference. That’s one of the more direct routes from Rosen into mainstream philosophy-adjacent discussions of self-referential structure. (Springer)

For the “Rosen ↔ modelling / representation / semantics” axis:

Pattee’s line on the ‘epistemic cut’ and ‘semantic (semiotic) closure’ overlaps strongly with Rosen’s concerns about modelling, description, and closure, and Pattee explicitly frames it as about the conditions for objective scientific description when symbols and dynamics intertwine. (casci.binghamton.edu)

A slightly wider map of “where Rosen sits” in philosophy-of-science terms

1. Models vs simulations, and what it means to model at all
   Rosen’s modelling relation pushes you toward the semantic/pragmatic view of models (models as mediating structures with interpretation), and away from “simulation = model”. The computability/closure debate papers are where this gets cashed out technically and methodologically. (DCC Universidad de Chile)
2. Mechanism vs organisation
   Rosen’s slogan ‘throw away the physics, keep the organisation’ ends up adjacent to contemporary arguments about whether mechanistic explanation can capture organisation-level closure. Siekmann is useful precisely because it tries to translate Rosen into the language of current modelling practice and mechanistic explanation. (ScienceDirect)
3. Closure concepts across traditions
   Rosen’s closure to efficient causation is often compared (sometimes too quickly) with autopoietic operational closure. The autopoiesis “cartography” paper is useful because it is explicitly about avoiding category mistakes: when are we defining a natural organisation vs building a formal surrogate. (ScienceDirect)
4. Complexity as impredicativity (not “many parts”)
   This is the thread that links Rosen to philosophical discussions of self-reference, circularity, and limits of formalisation. The impredicativity chapter above is a decent doorway into that conversation. (Springer)