Book Review: Robert Rosen's "Life Itself"

Review of Robert Rosen's book Life Itself

by Sarah Voss

About half a year ago a book called Life Itself (Columbia University Press, 1991) by Robert Rosen came to my attention, and I have been reading and rereading it ever since.  Rosen, who died in December, 1998, was a student of Nicholas Rashevsky and a product of the mathematical biology program at the University of Chicago and eventually professor emeritus of biophysics at Dalhousie University. He was the author of numerous articles and several earlier works, including "Principles of Mathematical Biology," and "Principles of Measurement. " In general, his work is very wide-ranging and for many years was largely undervalued.  In recent years however, the scientific community has begun to credit his work in new and potentially revolutionary ways, ways which impact our understanding of humankind, of artificial intelligence, of evolution, and even of consciousness itself.  By extension, I believe there will also be a corresponding impact on the religious community.  In this review I hope to introduce Meta readers to this man and to his provocative work.

Much of what Rosen said, at least in Life Itself, has to do with our view of animate/inanimate, and in some sense it is probably fair to say he sets forth a dichotomy between the two.  But I get the sense that his view, which completely reverses the traditional approach of animate as a narrow subset of inanimate, is much more in-keeping with a "process" orientation, perhaps even a Whiteheadian orientation, although this, alas is not really my field.  Others see other possibilities.  John Erskine, an experimental physicist recently retired from DOE who is now giving some public presentations discussing (in part) Rosen's work, sees a connection between Rosen's notion of  "life itself" and what Native Americans recognize as "animate."  William Dress, of Oak Ridge National Laboratory in Tennessee, believes Rosen's "seminal work in the biology of complex systems (organisms) [extends] to touch on the problems of thought and consciousness itself."

Donald Mikulecky, a physiologist at Medical College of Virginia Commonwealth University, notes (see that Rosen's answer to the question "What is Life?" carries with it a shift in contemporary science from a focus on method to one on content, and he says that the scientific world "is beginning to discover Rosen, not so much because it wants to or because it likes what he teaches, but because there may be no other way to proceed."

Brian Josephson (, the Nobel laureate who discovered coherent tunneling between superconductors (the Josephson effect; Josephson junctions), admires Rosen's attempt to treat the structure of biosystems on a rigorous basis, but also allows that some of his arguments are controversial.

For instance, Rosen was the subject of much controversy about a year ago in the NECSI discussion group ( has a link to the archives).  In case you might be as unfamiliar with NECSI as I was, it stands for New England Complex Systems Institute. Apparently, there was a split some time back in this group that stemmed from two different philosophical camps on complexity.  One, following the Santa Fe school, lost the original discussion group list, but kept the original address:  The other (the one cited above pertaining to the Rosen discussion) follows the ECHO school on complexity (after the series).  As I understand it, the difference in these two approaches  to complex systems is the difference between one that involves multi-entailment structures and relationships (which is what Rosen and, after him, Mikulecky champion) and one that is a single-entailment machine (of the sort that John Holland champions).

Controversy aside, Rosen wrote exquisitely of a subject matter so complex that I have on several occasions despaired of being able to comprehend his ideas, let alone review them.  I am reminded here of an occasion some years back when I gave a faculty review of Steven Hawking's  A Brief History of Time somewhat to the surprise of those who were better qualified than I to do so.  The thing was, though, those better qualified were unwilling to risk being wrong, and so they found reasons to avoid tackling the task themselves, preferring instead to hear what I had to say about the book.

It is in this spirit that I approach Life Itself.  Up front I admit to being grossly under-educated in the areas about which Rosen writes.  I have only the barest lay knowledge of biology and, while I do in fact have a mathematical background, it does not extend to the category theory Rosen specializes in and it barely touches on the chaos theory which peppers his thought.  Even those who follow in his footsteps -- and apparently there are a few such brave souls -- have not found this book easy to master.  Life Itself is frontline scientific research presented in a systematic, mathematically rigorous manner.  Quite frankly, it is never likely to duplicate the popularity of Hawking's Brief  History.  Nonetheless, heartened by Rosen's opening statement that he is unaware of any pragmatic purpose of the book and it is intended, thus, "for anyone who wants to claim it," I wish, if not to personally claim it, then hopefully to stimulate others to try their hand.

Perhaps I should mention here that a new Rosen book Essays on Life Itself was published posthumously in late 1999 and that a Web review of it (see Mikulecky, above) suggests it "fleshes out" the arguments Rosen set forth in Life Itself in a way that may lighten the reader's struggle with the material.  I haven't read this new collection yet, so I can't attest to how effectively it lightens or enlightens, but even if it makes Life Itself seem indecently laborious, I STILL recommend reading it.  Life Itself is a beautifully crafted book, provocative, insightful, challenging, and worthwhile even if you only comprehend portions of it.

How does Rosen's book merit such praise?   A few examples will help illuminate his challenges to us:

1) "First and foremost," he declares, "it should always be borne in mind that this book is about biology.  More specifically, it is a report of where the question 'What is life?' has taken me in the quest for an answer.  I am well aware that most of the ideas developed herein seem, in isolation, to have little to do with conventional biology.  But ideas fold too; even those that seem most remote in terms of their initial origin, and even their content, may turn out to lie very close together indeed in some appropriate topology."  (xv)

2) "In order to be in a position to say what life is, we must spend a great deal of time in understanding what life is not. Thus, I will be spending a great deal of time with mechanisms and machines, ultimately to reject them, and replace them with something else. This is in fact the most radical step I shall take, because for the past three centuries, ideas of mechanism and machine have constituted the very essence of the adjective 'scientific'; a rejection of them thus seems like a rejection of science itself." (xv)

3. "Physics as we know it today is, almost entirely, the science of mechanism, and mechanisms, as I argue, are very special as material systems.  Biology involves a class of systems more general than mechanism.  In fact the relative positions of physics and biology become interchanged; rather than physics being general and biology special, it becomes more the other way around." (xvii)

4. "My subject matter herein is... [a] duality, that between 'hard' science and 'soft' science, between quantitative and qualitative, between 'exact' and 'inexact.'  This duality is not to be removed by any kind of tactical accommodation, by any superficial effort of conciliation or ecumenicism....  It is thus not a matter of logical argumentation or persuasion that is involved here; it is a matter more akin to religious conversion." (2)

5. "[I]n the sciences this dichotomy [between qualitative and quantitative] rests on (generally unrecognized) presuppositions about the nature of material reality and on how we obtain knowledge about it. ...[T]hese presuppositions themselves have formal, mathematical counterparts, which allow us to reflect this scientific dualism into an exactly parallel one that exists within mathematics itself." (2)

These five quotations, taken (as the page credits denote) from the earliest portions of the book, set the stage for the rest of the work.  Over and over in the remaining 279 pages, Rosen paraphrases, re-phrases, and expands upon these ideas, much as though he had learned every minister's fundamental rule of sermonizing: Tell 'em what you're going to say, then tell 'em, then tell 'em what you said. In the process, he reveals the depth of his own vast knowledge and the mind of a truly independent thinker, though he does both in such a modest way that one hardly realizes he's doing it.  What is life? Even to frame this question, he writes, requires an "almost infinite audacity."  To attempt an answer "compels an equal humility."  Rosen, clearly, has both.

His exploration of this question is delivered in eleven chapters as follows: 1)   Prolegomenon (and this after a preface, a note to the reader, and a "praeludium") 2)   Strategic Considerations: The Special and the General 3)   Some Necessary Epistemological Considerations 4)   The Concept of State 5)   Entailment Without States: Relational Biology 6)   Analytic and Synthetic Models 7)   On Simulation 8)   Machines and Mechanisms 9)   Relational Theory of Machines 10) Life Itself: The Preliminary Steps 11) Relational Biology and Biology

In chapter 1, Rosen lays out the history of the "What is life?" question, showing that physics has been largely silent on the issue while biologists, following physicists, have answered it with the machine metaphor: Life is a machine.  Such reductionism, however, is not something Rosen can leave unchallenged:  "...the machine metaphor is not just a little bit wrong; it is entirely wrong and must be discarded." (23)  In chapter 2 he begins the process of metaphorical and analytic comparison with which he justifies and elaborates upon this stance.

As the fifth of the five quotes I noted above suggests, it is to mathematics that Rosen turns for both his metaphors and his analysis. Thus, even as we have a "predilection for rational numbers" far out of proportion to their actual abundance among numbers, so, too, do we inaccurately perceive physics to be more universal than biology.  And just as Gödel showed that the syntactic aspects of mathematics can not effectively replace the semantic aspects, the syntactic aspects of physics cannot fully embrace the semantic aspects of life.  Says Rosen:

  ... just as formalization in mathematics [initially] believed that everything could be formalized without loss, so that all truth could be captured in terms of syntax alone, so particle mechanisms came to believe that every material behavior could be, and should be, and indeed must be, reduced to purely syntactical sequences of configurations in an underlying system of particles.  Hence the power of belief in reductionism, the scientific equivalent of the formalist faith in syntax. (68)

In other words, and again as in the relationship between mathematics and its Hilbert-like formalizations, the organism possesses a kind of semantic meaning totally absent in the machine.   It has something more, something additional, something which the machine lacks precisely because it bears a certain, limited type of causality, or what he calls an "impoverishment of entailment." (e.g., 245-246)

Rosen posits that life is a material system closed to efficient causation.  More precisely (and there is a subtle difference here), he claims that "a material system is an organism if, and only if, it is closed to efficient causation." (244)  In arriving at this conclusion, he draws  parallels between Aristotle's three traditional categories of causation -- material, efficient, and formal -- and three mathematical counterparts axioms, production rules, and specific algorithms/ programs. (48)  These three formal mathematical notions likewise have their own analogues in  a) the figures, b) the gears and springs, and c) the particular rules for arranging these gears and springs which characterize the Newtonian mechanistic or "clockwork" worldview. (185)

By examining the inferential entailments typically found in formal mathematics  where "we can see inside... to an extent that would be Godlike in the external world" (153) Rosen distinguishes two significant kinds of modeling relationships relevant to his definition of an organism: analytic and synthetic.  The analytic is "intimately tied to" the notion of efficient causation and, not incidentally, to the direct (Cartesian) products found in category theory and the semantic understandings of natural language systems. The synthetic modeling relationship, on the other hand, is metaphorically equivalent to material causation, direct sums, and syntactic considerations.

In the proverbial nutshell, Rosen claims that the whole notion of mechanism follows from the assumption of equivalence between these two kinds of modeling relationships.   Eliminating this assumption leads to a new, more holistic realm, one he labels  "complex."  The assumption of equivalence is common to reductionistic Newtonian insights; the opposing belief, that analytic and synthetic models are not equivalent, is typical of what he calls relational biology.  In reading these chapters, you may not understand all the intricacies of this distinction, but you will be left with no doubt as to which, in Rosen's mind, is preferable.  "In a sense, it is the thrust of this entire work that this hypothesis of analysis equals synthesis must be dropped." (154)

When we lift the limitations of equivalence (analysis equals synthesis) from a natural system, the results are most obvious in terms of the kinds of causal entailments that result.  In particular, the efficient cause of some event in a mechanistic system can not be found within the system itself, whereas in an organism it can. (235) In different words, entailment in a machine is too "impoverished" to allow for what Aristotle called "final" causation (his fourth category).  "In short, efficient causation of something *inside* the [mechanistic] system is tied to final cause of something *outside* the system.  As Voltaire once succinctly put it, 'a clock argues a clockmaker,' from which he then went on to conclude, by extrapolation, '*therefore* a universe argues a God.'  This kind of dialectic ... is inherent in the concept of a m chine... one of the main intentions of the machine metaphor was to dispense with final causation entirely.  Instead, it inevitably reappears in a worse form than before." (246)

One significant area in which trouble arises from misunderstood entailment is in evolution.  To Rosen, today's biological weltanschauung totally (but wrongly) denies any entailment of evolutionary processes.  To do so would seem to imply that biology is either a form of mechanism or a form of vitalism.  Most contemporary biologists, stuck with one or the other of these two alternatives, choose (incorrectly, if you accept Rosen's argument) the former over the latter.  What Rosen offers instead is an alternative, "right" direction (108), a "complex," relational approach which ultimately offers both the possibility of a new biorelational technology (245) and a wiser, perhaps more "mystical" sense of connectedness. (279)

Rosen not only talks about complexity, he practices it, at least in the sense that his writing overflows with a plethora of ideas, many of which seem at first to be unrelated and all of which tempt the reader to wander off into any one of countless tangential arenas. Indeed, any "review" of his work is necessarily going to be highly selective of just which of his many ideas are most important to stress.  Since my own research interests emphasize the metaphorical use of mathematics in religion, it is no surprise that I present his ideas through this filter.  Another reader will have other options, possibly too many options.  But then, it is said that one's greatest strength is also one's greatest weakness.  In Rosen's case, this translated into a career in which his work was often undervalued by his colleagues and little known otherwise.  As one reviewer said (see, "Dr. Rosen was one of those unfortunate scientists who worked on problems that to the rest of his community were non-existent."  Yet Rosen himself, in an obscure 1997 interview for Belgian television ( uleck/rsntpe.html) indicated a keen awareness of the potential impact of his ideas.  Allowing that he believed there was "a massive lack" of wisdom inherent in our species and society, he acknowledged that he might never publish all of his ideas because he considered some of them potentially dangerous.  "I may leave them somewhere," he added. "You know, like Leonardo da Vinci, write it in code backward in the mirror... in some kind of cipher..."

To some, Life Itself may already seem to be written in cipher.  Yet, in this reviewer's opinion, it is well worth the effort to try to decode it.

Sarah Voss Unitarian Universlist minister and author of several books, including What Number is God? Metaphors, Metaphysics, Metamathematics and the Nature of Things and Zero: Reflections about Nothing.  Also former Templeton course winner and recipient of "2nd Rounder" grant development for her course "Math: A New Language of Theology."


Join Metanexus Today

Metanexus fosters a growing international network of individuals and groups exploring the dynamic interface between cosmos, nature and culture. Membership is open to all. Join Now!