Book Review: Essays on Life Itself

Essays on Life Itself by Robert Rosen is a collection of essays that were mainly written after Robert Rosen’s other book Life Itself was published. Robert Rosen (1934—1998) was an American theoretical biologist. His research was concerned with the most fundamental aspects of biology, specifically questions about life itself.

Basic questions
Robert Rosen’s main curiosity and research revolved around the question, “What makes living things alive?” (p.iii). Robert Rosen’s view was that this kind of question “becomes unanswerable if we do not permit ourselves a universe large enough to deal with the question” (p.2). Whatever “our world may be, it is not simple” (p.60). The basic questions of life “are not empirical … at all, but rather, conceptual ones” (p.274).

Primitive physics
Both Erwin Schrödinger and Albert Einstein pointed to “a conundrum about … physics itself, and … its relation to life” (p.7). Albert Einstein wrote, “One can best feel in dealing with living things how primitive physics still is” (p.7). Organisms “are more general than the nonorganisms comprehended in the old physics” (p.27). “Contemporary physics … rests on assumptions that limit it profoundly” (pp.255–56). “No amount of sophistication within these limitations can compensate for the limitations themselves” (p.256).

Limits within mathematics itself
Interestingly, Robert Rosen himself pointed to limits “arising within mathematics itself” in answering questions about living things (p.2). He claimed that “an organism … is more generic than an inorganic system rather than less” (p.4), and that “organisms possess noncomputable, unformalizable models” (p.4). The “world of algorithms, consisting of finite-hardware executing algorithms embodied in programs or software” is “a very small world, a world of machines” (p.99).

Mathematics’ wrong turn
Robert Rosen’s “contention [is] that mathematics took a disastrous wrong turn … in the sixth century B.C” (p.63), and that the “impact of that wrong turn … has spread far beyond mathematics” (p.63). It was at that time “mistakes began to be made … which have haunted us ever since” (p.71). Mathematics is “much more than … symbol manipulation” and “transcends algorithms” (p.92). “All the foundation crises that have plagued mathematics … arose from the discovery that set theory was impredicative, that it was itself not formalizable” (p.136).

Limits with algorithms
Problems such as “the mind-body problem appear hard because they involve properties (e.g., life and mind) that depend on impredicativities within the systems … that manifest them” (p.136). “Just as an attempt to break open an impredicative loop in a mathematical system, and to replace it by a finite syntactic … algorithm, destroys all properties of the loop itself, so any attempt to fractionate a material system containing closed causal loops destroys all of its properties that depend on such an internal loop” (p.136).

Organisms have maximal entailment
Robert Rosen argued that “the systems we call organisms are … maximal in their entailment structures” (p.161). They are thus “inherently semantic and contain causal counterparts of impredicative loops” (p.162). Rosen contended that “organisms take us … into a realm where particulars do not necessarily entail generalities and where knowledge of how a system works does not entail how it is created (p.258).

The essence of reductionism
“The essence of reductionism is … to keep the matter of which an organism is made and throw away the organization, believing that the latter can be … recaptured from the former” (p.260). The “fabrication of something (e.g., an organism) is a vastly different thing than the simulation of its behaviors” (p.269). “It takes a lot more than we presently have” to build an organism (p.269). “That is why the problem is so hard” (p.269).

Social vs. biological modes of organization
Robert Rosen believed that “biology provides us with a vast encyclopedia about how to solve complex problems, and also about how not to solve them” (p.271). He believed that “there are many deep homologies between social modes of organization and biological ones” (p.271). One of the “deepest lessons of biology” is that “complex organizational problems can be solved via cooperation and not by power and competition” (p.272).

Machines vs. organisms
The “machine metaphor is false” (p.295) and “leads us in exactly the wrong direction” (pp.254–55). “Machine and organism are essentially different in kind” (p.284). “We generally construct things sequentially” (p.291). “That is how we build a machine” and it is “the only way we know how to construct anything” (p.291). However, we cannot build an organism. The world of “machine and mechanisms” is “a very nice, tidy, orderly world,” but it is also a “closed” and “impoverished world” (p.292). “Machine is simple, organism is complex” (p.295).

Formalism and reductionism
There is “a deep parallel between the formalist view of mathematics … and the reductionistic view of the material world” (p.305). The “reductionistic view seeks to capture all causality in laws that tacitly amount to … computability” (p.305). However, computability in mathematics is “an excessively strong stipulation that characterizes a vanishingly small class of … systems” (p.305).

Simple vs. complex systems
If “we try to replace a complex system by a simple one”, then “we necessarily miss most of the interactions of which it is capable” (p.338). It is possible to “approximate, but only locally and temporarily” (p.338). And it will be necessary to “keep shifting from model to model, as the causal structure in the complex system outstrips what is coded into any particular dynamics” (p.338). “The situation is analogous to trying to use pieces of planar maps to navigate the surface of a sphere” (p.338). A “sphere is in some sense the limits of its approximating planar maps” (p.338).

Conclusions
If Robert Rosen is right — and I think he is — then it does change the way we look at our world. The mathematical machinery is limited in what it allows us to capture about the world and necessarily misses most of what is really going on. Rosen doesn’t say that we can learn nothing from computable models. His point is that inherently computable dynamical systems miss most of reality.

Computability imposes very strong limitations on entailment. It is precisely these limitations that allows them to be expressed as ‘software.’ However, meaningful distinctions cannot be made between ‘software’ and ‘hardware’ in organisms. Almost everything about organisms is entailed from something else. Rosen asserts, furthermore, that organisms and human systems are very much alike.

The ramifications of Robert Rosen’s ideas are very deep indeed. It’s a brilliant book!