Introducing the Pre-individual

I'm excited to get back to Simondon after the long detour his On The Mode of Existence of Technical Objects inspired.  I've only read the first 30 pages of Individuation in Light of Notions of Form and Information, but I can already see that this one is going to be epic.  Simondon aims to provide the philosophical framework that would support the vortical worldview at work in Macy's book.  In other words, he's going to develop the surprisingly deep implications of the idea that, "it's turtles all the way down".  Since this is the unofficial motto of Deleuze's philosophy of immanence, you are likely not surprised to hear that Simondon also promises to cast some light on the book that launched a thousand blog posts: Difference & Repetition.   But in fact as we'll see from examining the introduction, the connections between these two philosophers go far deeper than a shared general worldview.  I think Individuation was a massive influence on Deleuze in general, but on D&R in particular.  So pop some popcorn, this book is going to be a long one ... 

First off, skip the Foreword.  I don't know who Jacques Garelli is, but he produces the customary academic word salad.  It's not worthy of Simondon, who, while a very technical writer with an unusual fetish for semicolons, is about as clear and dry as French philosophy gets.  Unlike Deleuze, who often writes in an attempt to produce the effects his philosophy describes, Simondon remains firmly in the camp of expositing a theory.  He has a complicated idea which he'd like to transmit to us as best he can, and writes with barely more style than your average engineer.  The introduction breaks down into three parts: 1) A description of the problem of individuation and an orientation to the concept of the pre-individual necessary to resolve it.  2) A discussion of how individuation cannot be a single process accomplished at a stroke, but must consist in a cascade of individuations, together with an introduction to the four cascading levels Simondon will cover -- physical, living, psychic, and social or collective. 3) Some meta level reflections on the consequences of this viewpoint and a preliminary description of the type of thinking it represents.

1 ----------------

We can see this in the very first paragraph of his introduction, which states the problem he will tackle with admirable clarity.

The self-centered monism of substantialistic thought is opposed to the bipolarity of the hylomorphic schema. Yet both of these two ways of approaching the reality of the individual have something in common: both suppose that there is a principle of individuation prior to individuation itself that is capable of explaining, producing, and guiding it. We are prompted to revisit the conditions of the individual's existence starting from the constituted and given individual.  (I, 1)

Why is it that we appear to encounter a world filled with distinct, separated, individual things?  Our usual responses to this problem beg the question.  If we say that the things we find in the world, like clocks and spiders and the idea of Justice, are produced due to the interactions of smaller things like atoms, we have not answered the question.  Why are there individual atoms?  It doesn't turn out to matter whether we think the atoms are material or ideal, they are still pre-constituted individuals that we take for granted.  Likewise, if we adopt Aristotle's idea that an individual is the coming together of matter and form, we have again presumed that the reality of this particular individual is derived from another pre-existing individual, in this case the form.  But why does that individual exist?  Both these responses send us in circles, and satisfy us insofar as we quickly stop asking questions because we already recovered exactly what we were expecting -- the unity and identity of the individual we started with.

But the real question was why there are any individuals at all.  Maybe there are no individuals.  If we really want to understand how individuals come to be, we need to look for one of Nozick's "invisible hand" explanations that would tell us how individuals arise from somethings that are not individuals.  In short, we need to look for what Simondon calls an "operation" of individuation, a process by which individuals come to be.  Or do not come to be.  At least, do not come to be the fully distinct and finished entities we usually presume with this term.  If we take the reality of individuals for granted, if we presume we already know what this term means, and then search for a principle of individuation which produces exactly this, and only this, outcome, we've gone little beyond suggesting that, "Opium induces sleep because it has a soporific quality."   The principle of individuation -- the interaction of atoms or the marriage of matter and form -- produces individuals because it is an individual producing principle.  

The principle of individuation will be sought as a principle that is capable of accounting for the characteristics of the individual without a necessary relation with the other aspects of the being that could be correlative to the appearance of a real individuated entity. Such a perspective of research grants an ontological privilege to the constituted individual. (I,1)

Simondon instead proposes a theory of the dynamics of pre-individual being that would account for the production of individuals as in some sense a byproduct of its operation.  This operation does not presuppose individuals, does not produce completely finished individuals, and will not produce only individuals.  In this sense it represents a true onto-genesis, a process that spontaneously generates the being of an individual rather than slyly importing it as a given principle.

The individual would then be grasped as a relative reality, a certain phase of being which supposes a pre-individual reality prior to it and which, even after individuation, does not fully exist all by itself, for individuation does not exhaust in a single stroke the potentials of pre-individual reality, and, moreover, what individuation manifests is not merely the individual but the individual-milieu coupling (I, 3)

It's obvious that "pre-individual being" has got to be some strange stuff.  It can be neither form nor matter in the conventional sense.  It cannot contain individuals, nor itself be an individual.  Yet it must somehow be capable of giving rise to individuals.  So it is a type of being that already contains some inherent dimension of becoming.  At least at first, Simondon's preferred image for the pre-individual is a supersatured fluid out of which a crystal can precipitate.  There are a few important points to make about this metaphor. First, the pre-individual is not a homogenous entity.  It differs within itself or contains internal tensions or differences in potential that prevent it from completely coinciding with itself.  This internal heterogeneity is precisely why it can become individuals through an operation that Simondon likens to a phase transition.

... becoming is a dimension of the being and corresponds to the being's capacity to phase- shift with respect to itself, to resolve itself by phase-shifting; pre-individual being is being in which no phase exists; the being in which an individuation is completed is that in which a resolution appears through the division of the being into phases, i.e. becoming; becoming is not a framework in which the being exists; it is the being's dimension, the mode of resolution of an incompatibility that is rich in potentials (I, 4)

Second, as a result, notions of individual identity don't apply to it in the same way that they don't apply to Deleuze's conception of a multiplicity.  The pre-individual is irreducibly multiple not because it contains many things (units that could only be individuals) but because of its potential to overflow itself, so to speak, and in the process of produce individuals without exhausting itself.  Clearly we're talking about something that defies our usual zero-sum binary logic

In order to think individuation, we must consider being not as substance or matter or form, but as a tense, supersaturated system above the level of unity, as not merely consisting in itself, and as unable to be thought adequately by means of the principle of the excluded middle; the concrete being or complete being, i.e. pre-individual being, is a being that is more than a unity. (I, 4)

So the pre-individual is more like a field than an object, though Simondon seems to be reticent to use this term and instead usually describes it as a "system".  In any case, the point is again that 'the' pre-individual is not a single totality or whole that contains individuals as parts, nor is it the simple aggregate of all individuals.  As we'll see in more detail later, you can neither deduce individuals from it, nor induce it as a whole by combining individuals.  In short, the pre-individual is not the 'largest holon' which I think secretly lurks in the thought of those who employ this term.  It is not Being or Gaia or World or Whole.  And it's important to see that its distinction from these harmonious and unitary individuals is not just a technical one, but a whole matter of tone.  For Simondon, the pre-individual is under stress, riven with tension, animated by a dynamic that threatens to split it apart.  As his reference to Leibniz makes clear, it is anything but a pre-established harmony of beings.

Unity, which is characteristic of the individuated being, and identity, which authorizes the usage of the principle of the excluded middle, do not apply to pre-individual being, which explains why the world cannot be recomposed after the fact with monads, even by adding other principles like sufficient reason in order to organize these monads into a universe  (I,4)

In fact, Simondon consistently notes how important it is that the pre-individual contain large enough differences within itself that it can support two irreconcilable orders of magnitude.  The individual is then precisely the structure that allows these disparate orders -- the molecular and the molar, just as D&G will call them in A Thousand Plateaus -- to communicate.  The individual (partially) resolves the tension that plies the pre-individual.  Its structuring represents the phase transition of a metastable system from one partially stable equilibrium to another.  The best metaphor is again the advancing limit of a crystal, the place where some of the potential energy of the supersatured fluid (a molar, systems levels variable) is dissipated into catalyzing a change in phase from liquid to solid (whose precise structure depends on the molecular components of the fluid).  The limit of an individual crystal comes into being as a mediation between these scales, as a macro-molecular phenomenon. The crystal is the structure that resolves the problem or tension posed by the fluid pre-individual system.

While a potential energy (the condition of a superior order of magnitude) is actualized, a matter is organized and divided (the condition of an inferior order of magnitude) into structured individuals on an intermediate order of magnitude that develops through a mediate process of amplification.  What leads to crystallization and underpins it is the energetic regime of the metastable system, but the crystals' form expresses certain molecular or atomic characteristics of the constituting chemical species.  (I, 6)

 

2 ----------------------

Of course, the metaphor of the crystal is just meant to start imagining what Simondon means by individuation.  The pre-individual system is not exhausted by a single process of individuation but produces an entire cascade of individuations.  So Simondon's ambitious program will proceed from the physical individuation of the crystal to the organic individuation of living beings, and from there to psychic and social individuations as distinct strata in their own right.  At each level, the basic idea is repeated; the pre-individual differentiates into an individual-milieu pair through a phase shift.  But each new individuation extends the previous one, rather than being built on top of the latter.  So it appears that Simondon's process of individuation never actually produces a complete and final individual! 

We get a preliminary sense of how this might work from his description of the individuation of living beings.  What makes them alive is precisely that they are not done becoming individuals.

The living being conserves within itself an ongoing activity of individuation; it is not merely a result of individuation, like the crystal or molecule, but a theater of individuation. (I,7)

Since we're already familiar with Simondon's ideas about technical objects, and his important distinction between life and even the most advanced, intelligent, and 'autonomous' technology, we can see that this ability to extend its individuation is what makes life so distinctive.  The machine cannot change goals or modify itself. Its individuation is complete the moment it is produced, and while, within limits, it can adapt to a changing environment, it cannot become.  It is only what it is, and never overflows itself into something novel.

At this point, some might be tempted to describe Simondon's cascade of individuations as the creation of a 'holarchy'.  It sounds like there are different levels, each higher level depends on the one beneath, and while there is no clear bottom or top, the pre-individual seems to be a Whole capable of embracing all of these perspectives.  However, we've already seen how Simondon's pre-individual is precisely not a whole.  And we can now clearly see that his cascade of individuations is precisely not one where finished and parts come together to form a whole, which in turn serves as a part of a higher whole.  In Simondon's scheme, what enables us to pass from one level to another is that the parts are not fully individuated.  They still contain within themselves some unresolved potential, and they still maintain an internal relationship with the pre-individual.  

This living being, which is both more and less than unity, conveys an interior problematic and can enter as an element into a problematic that is vaster than its own being. For the individual, participation is the fact of being an element in a vaster individuation through the intermediary of the charge of pre-individual reality that the individual contains, i.e. due to the potentials that it harbors. (I,8)

 

The fact that the living individual is not completely individuated is what allows for it to serve as the theater for a psychic individuation. But it's not so much that the living being serves a sort of pre-individual field in its own right as that it never really stopped being the larger pre-individual field.  The initial living individual is just a stage in individuation.  This psychic individuation arises from the living being's representation of its own problems and its own possibility for solving them, in other words, from its ability to model itself as subject.  But since this model would have to take into account other living individuals operating with this same model, the continued individuation of life into psyche happens in the context of a collective individuation which is "vaster" than a single individual.  Or at least this is how I read the very compressed sketch of what the table of contents indicates will occupy two-thirds of the book.

However, the psychical being cannot resolve its own problematic in itself; its charge of pre-individual reality, at the same time as it is individuating as a psychical being that surpasses the limits of the individuated being and incorporates the living being in a system of the world and the subject, makes participation possible as a condition of the individuation of the collective; insofar as it is collective, individuation turns the individual into a group individual that is associated with the group through the pre-individual reality that the individual bears, a pre-individual reality that, paired with the pre-individual reality of other individuals, individuates into a collective unit. The collective and psychical individuations are both reciprocal with respect to one another; they make it possible to define a category of the transindividual, which attempts to account for the systematic unity of interior (psychical) individuation and exterior (collective) individuation. (I,9)

 

I think this promises to be an interesting way of looking at things.  It emphasizes again that, while we might call new levels of individuation "emergent", they are not built on top of old levels the way the second floor is built on top of the first.  The second emerges here only because the first is incomplete and therefore unstable in itself, and only through the act of relating the instability or metastability in one person to that in another.  This seems to capture precisely the way my self is constructed as already so socialized that it seems a misnomer to call it "mine".  In short, there is a feedback loop between psychic and collective individuation.  Upon reflection It seems to me that this topological feature is the most important difference between the pre-individual and the holon (see ENDNOTE).  Relations between individual subjects are produced as a dimension of collective individuation not a reconnection of previously finished individuals.  Similarly, since the psychic subject is only individuated in a collective milieu, relations between a single individual subject and an objective world are also produced through the interaction of a psychic and collective individuations.  Simondon refers to this reciprocal relationship between psychic and collective individuation as  the "transindividual".

The notion of the transindividual corresponds to the collective taken as the axiomatic that resolves the psychical problematic. (I,11)

Finally, the introduction briefly sketches a new definition of information that seems to generalize the relationship between psychic and collective individuations.  The idea seems to be that all individuations proceed by passing information between two incompatible levels of a system so that they can coexist.  In-formation, internal to the pre-individual, is what allows form to appear.

... information is never relative to a single and homogeneous reality but to two orders in a state of disparation ... it is the tension between two disparate reals, it is the signification that will emerge when an operation of individuation will dis- cover the dimension according to which two disparate reals can become a system; information is therefore an initiation of individuation, a requirement for individuation, for the passage from the metastable to the stable, it is never a given thing ... information is that through which the non-resolved system's incompatibility becomes an organizational dimension in the resolution; information supposes a phase change of a system (I,11)

 

Clearly, this information refers back to the necessary heterogeneity of the pre-individual.  Information can only exist in a system that isn't stable (see footnote 12).  This may be another way of stating Bateson's definition of information as difference that makes a difference.  The walls in my house take in all kinds of "information" about the ambient air temperature, but without the thermostat, nothing much happens as a result of all this data collection.  Information must propagate.

3 -------------

The final section of the introduction is a meta level reflection on some of the peculiarities of Simondon's method.  We've already mentioned some of these oddities.  To think about the pre-individual, we have to leave behind the idea that relation is something that happens between two pre-existing substantial entities, and instead see it as a 'thing' in its own right, a mode or dimension of being.  This seems almost identical to Deleuze's notion of difference thought in itself, not as the difference between things or the remainder left by comparing identities.  We also have to leave behind classical logical concepts like the principle of the excluded middle that only apply to a realm where identity is already defined.  In particular, we have to leave behind the methods of both induction and deduction and create a type of thinking Simondon calls "transduction".

But contrary to deduction, transduction does not go elsewhere to seek a principle to resolve the problem of a domain: it extracts the resolving structure from the very tensions of this domain, just as the supersaturated solution crystallizes due to its own potentials and according to the chemical species that it holds, not with the contribution of some foreign form. It is also not comparable to induction, for induction truly conserves the characteristics of the terms of reality included in the studied domain, drawing the structures of the analysis from these terms themselves, but it only conserves what is positive in these terms, i.e. what is common to all terms, thereby eliminating what is singular from them; on the contrary, transduction is a discovery of dimensions whose system makes the dimensions of each of the terms communicate, such that the complete reality of each of the terms of the domain can become organized into newly discovered structures without loss or reduction (I, 15)

 

What makes this new logic especially interesting is that transduction is not simply a type of thinking, but an operation of being.  Transduction is the very process of individuation we've been talking about, the process by which a metastable pre-individual being (partially) resolves its tensions by splitting into two sides linked by, and separated by, an individual structure.  The individual here is less a thing, but a relation taken as a thing.  Which means that we should hear a technical overtone in this term --  eg. a photosensor transduces light into an electrical current, thus linking two distinct realms -- in addition to the logical one.  

Transduction corresponds to this existence of rapports that takes hold when pre-individual being individuates; it expresses individuation and allows for individuation to be thought; it is therefore a notion that is both metaphysical and logical; it applies to ontogenesis and is ontogenesis itself.  (I,14)

 

This idea gives very precise and technical form to a Deleuzian thought I've long been fond of -- thoughts are 'things' too.  Thinking is ontologically creative.  It doesn't represent being, it makes more of it.  Or as Macy put it, the knower and the known co-arise. 

Finally, Simondon briefly comments on the way that the concept of transduction means we have to modify Aristotle's influential hylomorphic model.  We'll have much more to say about this next time, but we can already see a few key points. In a sense, Simondon is more sympathetic to the hylomorphic model of individuation than to the substantialist or atomist one.  It's transparently obvious that atomism begs the question of how individuals are formed by presuming individual atoms exist, whereas with hylomorphism we have to think more carefully about the situation.  The problem is not so much the idea that both form and matter exist and are distinct.  The problem is with with our unstated presumption that individuals are created when a transcendent ethereal form is stamped on a compliant homogeneous matter.  What hylomorphism takes for granted is the already individuated form and the perfectly individuable matter.  Simondon's transduction is precisely not like stamping or molding a totally plastic material.  In this case, the form arises from within a 'matter' that is anything but homogenous.  As we've seen, this is an in-forming.

The notion of form must be replaced with that of information, which supposes the existence of a system in a state of metastable equilibrium that can individuate; unlike form, information is never a single term but the signification that emerges from a disparation.  (I,16)

At first this might come off as a relatively unimportant terminological distinction.  We just replace "form" with "information".  But thinking about this change highlights how subtly ingrained is the hylomorphic model.  Many of us who think we've left this scheme behind (eg. Macy) are still ensnared in it.  When someone describes the reality of an individual as a "pattern" that emerges in the flow of some "material" they may at first appear to be talking about the same thing as a Simondon.  And I think they are trying to, or talking loosely about the same idea.  But they often still have a tendency to substantialize this pattern, making it into the definition of a complete individual, distinct from all others.  We often equate an individual and a pattern in one-to-one correspondence -- as in, "my body is really just the pattern of continuity that keeps the ship of Theseus afloat".  What is this but the revenge of the hylomorphic model?  How is the reality of an individual pattern different from the reality of Aristotle's form?  And how could this abstract pattern that we can suddenly recognize when it "emerges" be helpful in explaining the creation, and not just the instantiation, of individuals?  Instead of form or pattern, Simondon suggests that we need to think about informing or patterning.  This is inherently an ongoing process in which the individual produced cannot be thought by themselves or in isolation.  For Simondon, there is no reality to "a" pattern, no form of an individual in itself.  There is only information that necessarily involves communication between disparate scales of pre-individual being.  In this case a single pattern is already multiple.

ENDNOTE ---------------

 I'm not sure why I'm so allergic to the terms holon and holarchy.  It may simply be an artifact my first encounter with it in Ken Wilbur shortly after college.  Everything about Wilbur's thinking fell into nice neat concentric circles.  But since nothing can simply fall this way, in fact everything needed to be rammed into this shape.  Immediately following Wilbur, the prison book club (thanks to my dear cellmate EA) read A Thousand Plateaus.  The contrast between Wilbur's neatly pruned tree and Deleuze's rhizome couldn't have been more plain.  

However, I've never actually read Koestler's The Ghost in the Machine,  so I don't know the whole history of the holon concept.  Still, recent encounters have made me cautious about the term.  Macy sometimes employs the term in the loose sense of asserting that it's turtles all the way down, a shorthand metaphor I obviously find unproblematic.  But she also sometimes employs the term in the more technical sense of 'autonomous sub-entity', which as I've pointed out leads us to imagine that each level is built on top of the stability of the previous one, with levels appearing in simple ascending order without any possibility of feedback between them.  Finally, contrary to her whole hypothesis, she also seems to think that calling nature a holarchy somehow magically turns it into a harmonious unity, an assumption I've already highlighted as particularly objectionable.  

But then I also recently came across a mention of holarchy in Niel Theise's Notes on Complexity, and he employs the term in a very different manner.  Near the end of Chapter 8 Theise, having talked about a descending scale of realities that passes from the body, through cells, molecules, atoms, to quantum 'wavicles', goes on to observe that we should consider the body a holarchy because it can be viewed equally accurately from any of these perspectives.

Similarly, our bodies are not solid objects at a "higher level" and cells at a "lower level" and molecules still "further down", though we have used that shorthand to construct our complexity view of the universe.  Instead, we should say that a body as a holarchy appears as a solid object from one perspective, as a community of cells from a different perspective, and as a cloud of molecules from a still different perspective. (NC, 47% on my Kindle)

This seems like a fine observation, and appears to be close to the quote from Koestler here, so perhaps this is closer to the original meaning.  But do we really need a whole new word to capture the idea that things can be seen from a variety of perspectives and scales?  What was the matter with perspectivism? And have we addressed the fact that the objects discovered at these scales need not neatly nest inside one another spatially as this examples suggest nor even coincide?  Are these even the same object?  If so, exactly what ties them together?  They seem related, but haven't we closed a lot of doors if we take their relation for granted.  Haven't we simply presumed that there is a single object when this is precisely the important question?  Indeed, Theise's next paragraph makes the gap in the logic here obvious, because he jumps from the idea of perspectivism straight to the idea of cosmic unity, in what is a clear logical non-sequitur.

If the universe is a unity, one vast holarchy of self-organizing complex systems, then we have to consider that what it true for any part is true for the whole.  (NC, 47%)

From here he will go straight into the ethical implications of the fact that 'we are one'.  This as I pointed out, is a reliable indicator of the tone of the concept of holarchy.  We should get along because we are all just parts of one great big universe!  But if there's one thing that Simondon is already uncovering, it's that the relationship of individual to world cannot be one of part to whole.  You can't get back to a pre-individual world just by adding up already constituted individuals. 

Without Feedback, How Would God Eat?

At the end of our philosophy of technology detour, and before returning to our investigation of individuation (inspired by Deleuze's book on Leibniz) I decided to insert a book Tucker recommended that I've been meaning to get to for a while now.  

It appears that Joanna Macy typically writes what you might call 'new age' books.  But her Mutual Causality in Buddhism and General Systems Theory is not your usual 'spiritual' fare.  I'm not sure what tone the other books strike, but here she adopts a high academic style.  Her goal is not so much a comparative study of the two areas in the title, however, as a productive mash-up that deepens our understanding of both feedback and dependent origination as processes of mutual causality, processes that, in a sense, make (two) something(s) out of nothing.  I think this central idea -- circular causality is another name for emptiness -- is so absolutely profound that I'm willing to overlook the fact that it actually wasn't a very good book in many respects.  While I'm quite fluent in academish, I loathe this style of writing.  I can only assume that Macy adopted it as a way to make her comparison of two marginal (from a Western academic point of view) sets of ideas sound more 'serious'.  Instead, it just makes the whole works sound stilted, repetitive, grammatical tortured, and just generally needlessly pompous.  Her framing almost seems to suggest that she would like to somehow 'prove' that Buddhist ontology is 'scientifically valid', though in the text she's actually quite explicit that this notion requires so many 'scare quotes' that it ceases to make any sense.  In any event, there are so many useful ideas here, and they fit so well with the vortex model of reality that I've been slowly articulating, that it feels foolish to get too distracted by stylistic criticism. 

The first couple of sections of the book introduce the idea of mutual causality, articulate its difference from our more common one-way linear concept of causality, and introduce both the Buddha's dependent co-arising and Ludwig von Bertalanffy's  general systems theory as paradigms that make mutual causality central to their view of the world.  This, I think, is the weaker half of the book.  Her exploration of both these bodies of thought is probably a bit too compressed to be useful to folks who haven't encountered them, and at the same time, can come off as a bit superficial to those of us who have already spent some time with them.  For example, while we should respect the fact that she wrote this in 1991, well before meditation began entering the Western mainstream, I found that her explanation of dependent origination was not nearly as thorough, readable, or convincing as Leigh Brasington's version, despite the fact that their interpretations are largely similar.  That, admittedly, is a high bar for comparison, and perhaps writing today she would feel obliged to do no more than cite and summarize.  But I would make similar remarks about both her discussion of the philosophical history of the linear causal paradigm (both in the West and in ancient India) as well as her introduction to general systems theory.  The former doesn't really engage the subtleties of a very long and complex history, so ends up reading like little more than a wikipedia entry on causality (and irked me with the same misreading of Hume which, in fairness, I autocorrected only 5 years ago). And the latter leaves us with only a very high level sketch of how feedback works in general systems theory, mostly filtered through the rather cosmic perspective of Ervin László.  A lot of this doesn't go much beyond the few paragraphs I wrote at the end of my reflections on Winner.   

Perhaps this unsatisfying sense of cursory overview is inevitable in a short book that takes on two huge bodies of thought.  In any event, we still need to summarize the two parts of this section to lay the groundwork for understanding Macy's free mixing of their metaphors in the second half of the book.  

1) Dependent Origination

As I mentioned, her overall interpretation of dependent origination is very similar to Brasington's SODAPI -- there are no things, only streams of dependently arising processes interacting.  She too argues that the canonical 12 link list is somewhat arbitrary, not to be understood in linear order, and shouldn't be thought of as playing out over three lifetimes but as occurring at every moment.  And while she doesn't emphasize Mahayana teachings at any point in this book, she is clearly pointing to the same equation of dependent origination with emptiness that Brasington illustrates with his deep dive into Nagarjuna.  The main difference in these presentations would be one of emphasis.  Macy preserves the term "causality", instead of opting for Brasington's carefully distinguished and more accurate term "conditionality", because she wants to emphasize the reciprocal causality that we usually call feedback.  Her point is that things arise from no-thing when there is a condition of self-reinforcing causal closure, some sort of causal loop.  Since the whole thrust of this interpretation is that the Buddha articulated a theory of dependent co-arising, of mutual interaction in which A leads to B leads to A, it seems to me a mistake to preserve the old language of cause, freighted as it is with the image of a linear chain of colliding billiard balls. Calling something the "cause of itself" is really just a provocative paradox; it can be useful, but also lends itself to giving the phrase either a facile sense or dismissing it as nonsense.  

But while the terms may be confusing, Macy is clear that what is at stake is the way two things are coupled as necessary conditions of one another, such that if one disappears the other will not be able to stand up on its own, like a sheaf of reeds or teepee which collapses when any one of the supports is removed.  In a sense, this is the whole soteriological secret of dependent origination, which is emphasized by the way the suttas always discuss the links in both arising and ceasing order.  It would be near impossible to elucidate all the physical, biological, psychological, and social causes of our suffering.  Fortunately, we don't need to do this.  There's a long list of things that condition our suffering, that it depends on, which means that we can break into the 'causal' chain at any point and disrupt its arising by removing any of those conditions.  As Macy points out in her discussion of nibbana (pg. 60), this does not mean that we escape entirely from the causal world and enter some higher metaphysical plane.  We can only escape suffering, not the world.  But the way to do this is straightforward once you understand dependent origination.  Simply remove one of the legs of the stool, so to speak, and let the circle of mutual conditioning "blow out" for lack of fuel.  Things will still be happening, they just won't display the feedback loop or pattern of causal closure that conditions a suffering self.

2) General Systems Theory

The first thing to keep in mind about general systems theory is that it is not a theory of all systems in general, which would be a theory of everything, but a very general theory of how there can be something like a system.  Despite the genuinely wide application of these ideas and the cosmic pretensions of some of its proponents, it's important to keep in mind that the roots of the theory are specifically biological.  Basically, it hopes to be a theory of what all organisms have in common, which is their ability to maintain and adapt their organization in the face of the second law of thermodynamics.  The epigraph Macy chooses for this chapter conveys the point well, while also letting us glimpse how many different strands she hopes to tie together together under a single name; general systems theory is made synonymous with cybernetics, information theory, control theory, and, as we'll see later, the predictive processing theory of neural architecture.

The ideas were generated in many places: in Vienna by Bertalanffy, in Harvard by Wiener, in Princeton by von Neumann, in Bell Telephone labs by Shannon, in Cambridge by Craik, and so on. All these separate developments dealt with . . . the problem of what sort of thing is an organized system. ... I think that cybernetics is the biggest bite out of the fruit of the Tree of Knowledge that mankind has taken in the last 2000 years. (Bateson, Steps to an Ecology of Mind)

The solution to Bateson's problem is pretty easy to state -- an organized system is the sort of thing we call a pattern in flows of energy and information.  The fact that there's some relatively stable pattern is actually the definition of "organization" and hence also of "system".  The fact that this is a defintion is subtlety that often goes overlooked, one that relates directly to the ambiguity I mentioned about the scope of the word "general".  Soon, Macy will begin discussing how all systems show certain "invariances" that allow us to form a theory of them in general.  For example, all systems use negative feedback to maintain homeostasis.  And all systems use positive feedback for learning and self-modification.  Systems are also intensive (more than the sum of their parts) and open to an external environment.  In fact, those four points summarize the entire theory of general systems that Macy outlines.  When you examine them, however, you'll notice that most of these are really disguised versions of the definition of a system.  If there's some ongoing pattern we can recognize as a system or organism (in a general sense) then there's clearly some sort of stable homeostatic process through which this pattern maintains itself as recognizable in the face of the second law of thermodynamics.  That same law requires that any such system be an open one, since all closed systems are headed towards heat death.  And while it may not follow logically from the definition, since the system's internal equilibrium is maintained far from thermodynamic equilibrium, it's a safe bet that it will also be non-linear (are there any stable far from equilibrium systems that are linear?) which is quite literally what we mean by "more than the sum of its parts".  None of these three invariances then -- intensive, open, and homeostatic -- is an empirical statement based on scientific observation of the behavior of "systems" that have been selected according to some other criteria.  These are just descriptions of the notable properties we'd like to understand about the complex phenomena we usually call "organisms". 

The assertion that all systems learn however, does not seem to follow from the simple definition of what a system is.  Macy describes this in somewhat misleading terms as "self-organization".

The system is self-organizing. Where a mismatch between input and code persists, the system searches for and encodes a new pattern by which it can function. Incorporating positive feedback, differentiation and complexification of structure emerge. (MCBGST, 77)

What she really seems to me mean here is that the system is self-modifying.   Needless to say, not every system is capable of self-modification.  Plenty of intensive, open, homeostatic systems that we call individual organisms don't seem to learn.  They cannot modify themselves to adapt to an environment that falls outside of what their homeostatic mechanism is capable of coping with.  They don't self-modify through positive feedback in a way that increases their differentiation and "complexification".  When the departure from homeostatic equilibrium becomes too large, they just cease.  Consider a bacteria.  It certainly checks the first three boxes, but as far as I know totally fails the fourth.  I imagine there are probably various patterns of gene regulation that become active in high stress environments that allow bacteria to modify their internal chemistry and external behavior.  This isn't learning though, but just nested homeostasis.  So the only invariant property of systems which might have empirical content turns out to only apply to some systems, and not be at all general.

Of course, this fourth property definitely applies to the system we are most interested in, namely, us.  The cognitive system we call a self clearly fulfills all four of the criteria Macy discusses.  I don't think we should be surprised to find that the model example of a system is a self.  Nietzsche and Deleuze have shown us many times that our image of unity and wholeness is always a projection of our experience of self-unity.  And of course in Buddhism the self is the locus of all our suffering.  So the second half of the book (Part 3) will rightly focus almost exclusively on the arising and ceasing of the self system.  But I think it's important to pause here and think carefully about what I called a subtlety in the definition of a system.  The subtle part is that, at least in Macy's explanation of it, general systems theory often takes for granted the existence of an organism, of a unitary whole, and asks how this entity can stick around in an ever changing world.  It doesn't have as much to say about how there came to be an organism to begin with.  Nor does it often seem to concern itself with the tricky question of how we know that it's the same system once it has learned or evolved a new homeostasis.  After all, if the system is defined by the first homeostatic equilibrium, then doesn't the shift to a second one actually transform the system entirely?  Doesn't it become something new?  In short, as befits a theory that seeks to explain how unity, identity, and wholeness are maintained, general systems theory doesn't seem to concern itself with how these are created.

In a way, this omission is very sensible, especially for pedagogical purposes.  There's no point in wondering where 'the first' organism comes from -- its turtles all the way down.  If we say that a system has no substantial identity but just is a pattern in flows of energy and information, this doesn't reduce it to moving matter, but it does make it depend on the relatively stable units of proteins, molecules, atoms, etc ... These depend on other units all the way down the stack.  There's always already some smaller unit that can be woven into a pattern.  Likewise, once a relatively stable process has been embodied, it can in turn be woven into other higher level patterns.  Following Koestler, Macy calls these patterns holons, and their infinitely nested arrangement a holarchy.  

The holon concept is fine as far as it goes.  It means we needn't worry about what lies at the top and bottom of the chain.  Anywhere in the middle will work fine as a starting point, and everywhere is in the middle.  And the usual Russian doll or concentric circle image of a holarchy also keeps us aware not only that there are systems at many scales, but that there is also a separation of scales.  Nature does seem to organize itself into hierarchical levels of extension.  So even though I've grown allergic to the word "holon" I can understand why it was coined.  But in my mind this concept risks obscuring the most crucial idea in Macy's book -- the way things are created out of emptiness in a process of mutual causality.  The problem stems from taking for granted the separate coherence of each level, as if the tree of nature were preordained, as it were, to branch at the level of ... atoms, molecules, DNA, proteins, organisms, brains, societies, etc ...  Of course, all these levels exist, and no one is trying to argue that you can have actual organisms without atoms.  But this only looks like a simple branching tree or set of concentric circles when we examine them on the basis of extensive scale.  In other words, to see a tree and not a rhizome here is to fall for one of the classic blunders and smuggle a materialism into a realm where it does not belong.  With a fuller understanding of mutual causality, we see that every one of these levels is nothing but a pattern, and that these patterns can be and often are incarnated from 'materials' drawn from a whole array of different strata or levels, and not simply from the one directly beneath them.  And of course the same observations apply to these 'materials' in turn.  In fact, Simondon's idea of technical objects embodying a whole technical ensemble makes explicit that we can find 'materials' that could not exist were it not for the patterning effect of a 'higher' level.  Levels undoubtedly exist, but they are mixed together in a ramifying network of inter-relations that looks nothing like a nice neat tree, and can even include feedback loops between levels.  Which means that looking only at the levels we habitually recognize and asking how each maintains or even modifies itself is not going to be thorough enough.  We need to look across all the levels and follow the 'materials' drawn from any of them that cohere into a self-reinforcing structure -- that is, a structure that constitutes the self to be maintained.  We need to look at each causal loop as it closes into a self-sustaining circle.  We cannot presume to know what the world contains in advance, and just ask our selves what principles are needed for this stuff ot stick around.  [Simondon will take up this deep problem right off the bat in The Individual in Light of Notions of Form and Information]

In short, before there can be a negative feedback loop that maintains homeostasis, there has to first be a positive feedback loop -- a mutual or reciprocal causal loop -- that allows various materials to interact in a way that leads them to reproduce this same interaction at a later time.  A pattern has to emerge.  The organism has to be synthesized or contracted out of the not-organism.  This is not exactly positive feedback in the deviation-amplification sense that Macy uses, though I think these two are deeply related.  It's more like a qualitative or creative positive feedback that closes a causal loop and makes possible the very expectations by which deviation, whether positive or negative, can be measured.  Without a goal, there is no deviation, those are the fucking rules.  Ultimately, this suggests that the negative homeostatic feedback which inspired the whole theory of cybernetic systems is actually just a fleeting moment of negativity sandwiched between different types of positive feedback loops; the first  which creates and the second which transforms identity.

 ---------

That long discussion of the scope of general systems theory is really just meant to clarify the ideas about mutual causality or co-arising that Macy explores in Part 3 of the book (starting on pg. 107).  It's not that I think Macy doesn't understand or would object to my observations above.  They're really implicit in her whole approach.  But sometimes they recede far enough into the background that we risk missing what I think are the deepest messages of the book.  Since there are so many interesting ideas in this second half, I'll tackle it chapter by chapter.

6) Self as Process

As the title indicates, this chapter focuses on the impermanence of all things.  In fact, it would be better to call this the emptiness of all things, because the point is that there are no things, only processes.  There are various metaphors that convey this idea well -- the Buddha uses fire, general systems theory often cites fluid flow, and then there are gerunds, verbs in place of nouns.  The important thing about all of them is that they give us an image of a pattern that can only be sustained by a dynamic flux.  Curiously, emphasizing the irreducibility of time, the reality of process, also implies that there's no essential spatial boundary to a system.  Everything that gets involved in this self-sustaining flux is 'inside' the system.  One of the interesting corollaries to this is that systems at distinct levels need not share the same boundaries.  For example, the boundary of the cognitive system I usually take to be my self doesn't need to perfectly coincide with that of the physical system I call my body.  Here you can already see the concentric circles getting a little wobbly.  The important thing to see here is not that every whole is part of another whole, a metaphor which still encourages us to think of things as having an inside and an outside, albeit sides that are at least in communication with one another in this case.  The important thing to see is the constantly shifting loop of mutual causality that effectively draws the boundary of a system at any given time.  After mistakenly calling that boundary is arbitrary, Macy quotes Bateson's very clear explanation of just how precise, yet variable and potentially unlimited, is the process which draws this boundary.

Consider a man felling a tree with an axe. Each stroke of the axe is modified or corrected, according to the shape of the cut face of the tree left by the previous stroke. This self-corrective (i.e. mental) process is brought about by a total system, tree- eyes-brain-muscles-axe-stroke-tree; and it is this total system that has the characteristics of immanent mind. More correctly, we should spell the matter out as: (differences in tree)-(differences in retina)-(differences in brain)-(differences in muscles)- (differences in movement of axe)-(differences in tree), etc. What is transmitted around the circuit is transforms of differences. A n d a difference which makes a difference is a n idea or unit of information. (MCBGST, 112)

Cybernetics would go somewhat further and recognize that the "self" as ordinarily understood is only a small part of a
much larger trial-and-error system which does the thinking, acting, and deciding. This system includes all the informational pathways which are relevant at any given moment to any given decision. The "self" is a false reification of an improperly delimited part of this much larger field of interlocking processes. (MCBGST, 113)

As an aside, this chapter also contains what I think is a revealing philosophical error.  Just saying panta rhei isn't enough. Perhaps there is a completely self-consistent way to maintain that there are no real entities at all.  This is a tricky question whose subtlety we only grasp when we confront its self-reflexivity; isn't my statement or even my thought that there is nothing still something?  But this is explicitly not the view of Buddhism, general systems theory, or Macy herself.  She thinks there are entirely real and abiding entities outside the flux; they're just called patterns. 

The universe is seen as made up not of things but of flows and relationships. It is these relational patterns, not any "stuff," that abides. Therefore, Laszlo, in drawing out the meta- physical implications of general systems theory, is disinclined to posit any immutable essence aloof from time and change. "Platonic ideas, or Whiteheadian eternal objects, are rejected as uncalled for." He argues that Whitehead's eternal objects enjoy "a one-way causal connection to actuality. ... They are externally related to actuality, i.e. are themselves immutable and unqualified by their exemplifications in the actual world." But in the causal interdetermination of the systems view, the "ordering is from within." Where all is mutually conditioned, "there are only internal relations and no basis for positing an unaffected agent, external to the causal play of process. (MCBGST, 108)

Despite several attempts, I can't claim to completely understand Whitehead's eternal objects.  But I do think I understand the problem to which they apply, which is the mode of existence we should attribute to patterns.  According to Macy, these patterns, "abide".  That sure makes them sound like they're eternal, so long as we understand this term as Spinoza defined it -- eternity is outside of time, not just the amount of time it takes to speak to a human at the cable company.  This isn't a small semantic debate either.  The reality of patterns provide for our notion of repetition, and hence for that of continuity and even of time itself passing.  I don't think the question is irrelevant from a soteriological perspective either.  Is the spiritual life not, at bottom, a way to get in touch with a reality that is always 'more than meets the eye'?  Whether we think there is one cosmic pattern called God or Dhamma, or many individual patterns that constitute the suchness of things, we are trying to get in touch with some dimension of reality that is not 'given', even if it is paradoxically right in front of us all the time.  We can't make this problem go away just by claiming that all the ordering is "from within".  Within what?  To get anywhere with this line of thought, we need patterns to be real in the broadest sense, even if they're reality is empty.

7). The Co-Arising of Knower and Known

One of the things that repeatedly struck me about Macy's book was how many of her examples illustrate the principles behind what we would now identify as the predictive processing or Bayesian inference or free energy minimization theory of brain architecture.  This is probably not a result of Macy somehow anticipating this theory.  Instead it appears that the free energy principle theorists simply don't often credit all the related theories that preceded their own.  So when we read Macy relating the Buddhist idea of saṅkhāras to the cybernetic concept of gestalts upon which we fabricate a world, we find we're suddenly in precisely the modern world of the Bayesian brain.

The gestalts and constructs by which the world is interpreted are understood, in the systems view, to be coded by past experience. As such, they are functionally equivalent to the Buddhist notion of sankhāra, the volitional formations constellated through earlier activity, which co-condition the content of perception and cognition. As "accumulated sense experience fermented by ignorance", these constitute, in Nanananda's words, "the ruts and grooves of our mental terrain," and "influence every moment of (our) living experience. The perceptions we take for granted as given turn out, under examination, to be synthetic and composite, sankhata. While the Buddhist view highlights, in a way that systems does not, the role of attachment in the perpetuation of these formations, both bodies of thought recognize their significance for perception, and both see them as subject to alteration. Formed by experience, these codes are modified by experience. (MCBGST, 124)

Since Macy's goal is less to expound this theory of cognition than to draw out its consequences for our worldview, she's gratifyingly explicit in describing the circular causality at work here.  The world has certain statistical regularities which the brain extracts and encodes.  These become the priors by which that brain processes perception and so fabricates a world relevant to it in the form of a generative model.  However, this model is not built for the purposes of representing the world, but of acting in it.  It tries to create a world that will allow it to sustain its prior models.  There's a loop here, some of whose feedback circuits exists only inside the skull of hairless chimps, but many of which extend out into the 'external' world.  As we saw in the previous chapter, when we consider a scheme like this we cannot draw a fixed boundary around where exactly the cognition happens.  The object of knowledge and the subject doing the knowing can no longer be clearly separated.  They appear simultaneously as mutually adpated, mutually conditioning.  This opens the door not only to a much more embodied notion of thinking in general, but also to considering what happens when we externalize some of our generative models of how reality works as concrete physical objects.  With a cybernetic (or predictive processing or general systems or whatever you'd like to call it) view of self-reinforcing causal loops in mind, we will naturally be forced to have a discussion about the 'thought' embodied in technical objects. 

In this chapter, Macy extracts three corollaries of the predictive processing theory that the Bayesian brain folks who are interested in meditation might finally be catching up to.  The first is that since knowing in this model is actually ontologically creative, there can be no final knowing.  If knowing is the feedback interaction that produces subject and object of knowledge, then every act of knowing creates even more world to be known.  Of course, it's theoretically possible for this to settle into a very static loop where everything in the environment we encounter simply reconfirms our priors because we've shaped this environment to do exactly that.  But this stable state rarely lasts long -- the world is big enough and we are small enough that something inevitably comes along to dislodge us from our narrow equilibrium.  Perhaps this is another way of stating the first noble truth.  You will depart from equilibrium.  There will be error you cannot minimize.  Even the mostly deeply grooved sankhara is only metastable.  Its self-reinforcing loop exists in the context of a whole ecology of other loops , all of which are transforming the world as they too know it.  There's no way to get to the end of this infinitely ramified flux and know 'it all' since that very act would change the totality we propose to know.

The second corollary is perhaps an extrapolation of the first.  Macy presents the idea of positive feedback as a deviation amplifying mechanism.  When the system is forced far enough away from its usual homeostatic equilibrium for long enough, it can stop applying the negative feedback mechanism that leads to its usual stable orbit, and begin allowing the initial departure to drive an even larger departure.  If it works, this process can drive the system to a new equilibrium, or perhaps a meta-equilibrium that contains several possible equilibrium points which include but extend the original one.  We call this learning.  It may be the same process as evolving.  While the difference between these seems to revolve around who exactly is the agent before and after the transition (and is it the same agent?) I think they are both mathematically equivalent to the Bayesian idea of updating the generative model itself.  When the model doesn't fit the data, one option is to call the data noisy and gather more data until it does fit the model.  Failing this, we start to decide that the data must be right after all and we should come up with a better model to predict it.  Of course, this doesn't always work.  If we think in biological terms, most of the time an organism pushed too far from equilibrium just ceases.  Given the similarity of things like nirodha to death, and the likening of Nirvana to a "blowing out", it's intriguing to wonder whether a similar logic is not also true of the cognitive system.  Ego death certainly seems like a transition away from our usual equilibrium.  In short, learning, or evolving, is risky for a system.  

The final corollary might almost be construed as a combination of the first two.  Macy suggests that the creative process of knowing can be turned on itself, as a process, in a way reminiscent of video feedback.  The camera watches the movie.  She proposes this as a means to account for meditative experiences of pure consciousness without objects.  Perhaps we should see these states not as the move to some new equilibrium where sensory data is turned off, but as a kind of meta-level model of the creative act of knowing itself.  This doesn't give us access to some new ontological realm, some final foundation of reality.  But perhaps it does make us fractally isomorphic to the ongoing process by which it is constructed? 

8) The Co-Arising of Body and Mind

Macy has two objectives in this chapter.  First, to address critics who see Buddhism as an ascetic world-denying religion.  Second, to aim the theory of mutual causality at the problem of mind-body duality, and argue for a non-dual conception of these 'opposites'.  The overarching theme is that we needn't denigrate the particularly of the body in favor of the universality of some abstract mind. 

The first question is one of textual interpretation that's above my pay grade.  People have been wrestling over the inner core of Buddhism for thousands of years.  Did the Buddha describe a negative or affirmative vision of the world?  Was he trying to live better in it, or escape it?  The answer is obviously yes.  It seems to me that we could read the Pali Cannon through either lens.  And over the years many different Buddhists have read it in different ways.  I prefer to see what the Buddha taught as an affirmative philosophy, albeit of an unusual sort akin to Nietzsche's affirmative idea that nihilism overcomes itself.  But I see it that way because I think it's the most useful way for me to see it, not because I'm convinced its objectively correct.  Your mileage may vary.  I'm only concerened with 'what the Buddha really taught' insofar as its obvious that this guy thought and practiced this stuff way way more than I have.  That hardly makes his advice infallible, nor his philosophical disposition beyond critique.  However, it does call for a little humility, and a belief that its worth listening to all of what he said very carefully, since there very well may be things in there that I would like to ignore, but should not. 

The second question, to understate the situation, has also been the subject of some controversy.  Macy argues that systems theory understands mind and matter as "two side of the same coin" and appeals in passing to panpsychist thinkers like de Chardin, Spinoza, and Whitehead.  The basic idea seems to be that anything that comes together as a system is a sort of incipient mind.  It reminds me of the way Deleuze called the passive syntheses of habit that create forms "contemplations".  Nature, in closing the causal loops that create systems, contemplates itself.  This implies a clear continuity between life and mind, where what we call consciousness is simply the life of a system that contemplates other systems, and ultimately contemplates its own arising as a (conscious) system.  The idea fits neatly with the creativity of knowing we discussed in the previous chapter.

I find this an appealing, even exhilarating, idea, but one whose feeling of resolution tends to dissipate under closer scrutiny.  For example, the philosophers mentioned are certainly related, but they are far from sharing the same theory.  In the end, perhaps the only thing they would completely agree upon is the inadequacy of naive materialism.  Then there's the knotty question of why a mutual causal feedback loop should constitute a quality of thinking and feeling.  If we consider this thought an emergent property of matter, then it seems we are not non-dualists but end up as materialists or at best epiphenomenalists.  If, following our objections to the idea of a 'first' holon, we flip the question over and ask why thoughts offer themselves as concrete and particular phenomena that can be taken up in other thoughts, then we end up as idealists instead.  Finally, the solution raises very difficult issues around the question of the One and the Many.  Are we talking about a single and same consciousness that manifests itself in a variety of systems, or are there an infinity of distinct consciousnesses, one for every closed loop, that only share a similar process -- essentially the process of coming together?  Here, Macy seems to opt for the latter, and views the concept of suchness in this light.

Buddhist thought also recognizes in the natural world an extension of mentality beyond the human realm. Ven. Sangharakshita, an English-born monkand scholar, emphasizes that in the Buddhist world view, in contrast to the Semitic religions, humans are not seen as unique in their possession of soul or mind, nor does this possession constitute a gulf between them and other forms of life.  The human is but "one manifestation of a current of psycho-physical energy, manifesting now as god, now as animal, etc."  This belief in psychic continuity underlies, he points out, the compassion for other creatures, the "boundless heart" which the Buddha manifested and enjoined.In the early Buddhist view, the consciousness endemic in life forms is not unitary and undifferentiated, because it does not derive from a supraphenomenal source.

By the logic of dependent co-arising, however, the consciousness manifest throughout the worlds and planes of existence is in every instance particular, characterized not by sameness but by "thatness" or "suchness" (tathatä). (MCBGST, 152)

But there are other places in the book where she seems to take for granted a more wholistic new age view of the inherent harmony of all things.  Does her beautiful soul consider all these distinct tathata as parts of a single whole?  And doesn't this leave us with a form of transcendence we've been trying to avoid?  These are complicated question that I don't think Macy ever completely resolves.

9) Co-Arising of Doer and Deed

I found this the most challenging chapters  The basic idea is straightforward -- karma describes a feedback loop between past actions and present possibilities.  The choices we make create both the world in which future choices will be made and the self who will make them.  While this sounds simple and incontrovertible, like everything in Buddhism and general systems theory, it involves a subtle question of agency.  We'd like to ask whether the self that decides is the same self that receives the fruits of the decision.  But by this point we can see that this question is nonsensical.  These two selves are involved in a mutual causal feedback loop across time. 

Just exactly how is this idea supposed to work though?  Because if we say that our current actions determine the world our future self will encounter, we are espousing the simplest type of determinism.  Even if we do not interpret karma as this law-like one-to-one correspondence, but simply observe that our actions at least have some influence on the future, we have only advanced to a slightly more sophisticated stochastic determinism, without undermining our usual feed-forward understanding of this concept.  Macy's initial explanation of the way our actions create the sankharas through which our future self encounters the world seem to fall into this category.

... our acts co-determine what we become.
    They do so through the formation and operation of the sankharās. These subconscious drives and tendencies condition the ways in which we interpret and react to phenomena. Like kaya, they too are composite, the term meaning "put together," compounded, organized. They are accrued and constellated by previous volitional acts of interpretation and response. I. B. Horner, in qualifying them, employs the term "potential energy" Their potency shapes not only consciousness, but perceptions and feelings. Their reciprocal relation to psycho-physical behavior leads Oldenberg to wish to identify them directly with action itself. "We might translate sankhara directly by actions," he writes, "if we understand this word in the wide sense in which it includes also at the same time the internal 'actions', the will and the wish."  More precisely, sankharà, both as a khanda and as a factor in the nidâna series, represents the reflexive or recoil effects of actions: the tendencies they create, the habits they form and perpetuate, the latent energies they bear. (MCBGST, 165)

We met the sankharas back in chapter 7 as the priors through which the predictive brain understands the world.  These are not 'innate', but the learned habits that enable us to cope with the world successfully enough to maintain a homeostatic equilibrium.  If we back up a step along the life-mind continuum, we can describe them as the self-stabilizing action of being an organism, the habits of nature itself that allow an action pattern to be repeated again and again, and in the process constitute a stable organism.  These are nature's own habits or tendencies.  We're looking at the same mutual causal structure in chapter 7 from an active rather than a representational lens.  

But this change of perspective raises an interesting question.  Previously, we examined the way this reciprocally causal feedback loop simultaneously gave rise to the gestalt of the knower and the ground of the known.  But knower and known co-arose in space.  By analogy, we should now be forced to think about the co-arising of present and future selves in time.   Is this really what Macy's equation of sankhara with action does though?  It seems that we're actually just discussing a scheme where actions, including 'internal' mental actions, lead to more actions.  Over time this snowballing accumulation of actions, or more precisely this trajectory of successive actions describes who we are.  Thinking of the self as a trajectory rather than a substance seems like a good idea.  But I don't think this view takes us away from determinism, since it does not seem to imply a reciprocal causality that would in this case need to operate backwards in time.  In other words, we see how actions shape who we are, but how does 'who we are' (or who we might become) return to shape our actions?  If sankharas just accrue and their consequences unfold in time, it seems we are sent back to the linear causal paradigm we were trying to escape.  Where, then, is the room for choice or freedom or liberation?

Macy wants to address this problem by appealing to the way systems theory predicts the gradual "complexification" of organisms.  In these terms the feedback loop is between structure (sankhara) and function (action), and the dynamics of their interaction over time is essentially a cybernetic reworking of the idea of evolution.

Structure, then, represents not something fixed, but a slice in time, the system's spatial order or organization at a given moment. As von Bertalanffy puts it, "What is described in morphology as organic forms and structures, is in reality a momentary cross-section through a spatio-temporal pattern." (MCBGST, 166)

Laszlo describes the relation between structure and function in a way that permits us to see with particular clarity its relevance to the Buddhist idea of karma.

     Structure is the record of past functions and the source of new ones. Function in turn is the behavior of the structure and  the pathway leading to the formulation of new structures. .. Not what a thing is, what it is made of, or for what purpose it exists, defines it, but how it is organized. (MCBGST, 167)

Over time, new structures can become so complex, and can accumulate a memory so deep, that their functions are determined more by factors 'internal' to the structure than by the environment it is placed in.  Macy would like us to see this internal determination as synonymous with choice, with freedom.  The idea seems to be that once a system becomes complex enough, its actions in the world no longer directly determine its future actions (since, by hypothesis, the external world as a whole no longer simply determines its actions), but instead tilt its internal landscape of determination, so to speak, in one direction or another.  In this in-determination lies both our free choice as well as the means by which this free choice is shaped by prior 'internal' acts (karma). 

In the early Buddhist view, then, a person's identity resides not in any enduring substance or self (atta) but in his acts (kamma) that is in the choices which shape these acts, which in turn through the conditioning effect of the sankharas, shape him. Because this causal process is reciprocal - the dispositions formed by previous choices modified in turn by the present acts they influence -- this identity as choice-maker is fluid, its experience alterable. While it is affected by the past, it can also break free of the past. (MCBGST, 173)

I like this idea in many ways, but I think it may be incomplete.  Consider our earlier discussion of positive and negative feedback loops.  Negative feedback loops are what allow an organism to settle into a stable homeostatic equilibrium.  The actions these loops entail are ones that enable this same action to repeat itself.  If we think of a simple organism as itself an action of nature, we can see that this action is the same as the organism repeating itself in time.  More complicated organisms have a wider variety of actions compatible with their equilibrium, but a similar idea applies -- we habitually act in a way that reproduces the possibility of our being able to continue acting in this way.  In other words, these homeostatic sankhara are re-actions.  They are the endless return of actions that have already happened.  This of course is an idea familiar to any meditator; most of our 'actions' are actually quite passive, mere reactions.  And one of the goals of sitting still for an hour is simply to short-circuit these reactions, to introduce a gap into their functioning that allows us to see the process of a reaction unfold before our eyes without actually engaging or identifying with it.  But where does choice enter into this scheme?  Even the most complex and contingent set of reactions doesn't seem as if it could ever add up to a true action.  What we're calling "choosing" here seems to correspond only to a positive feedback loop where in-action creates more inaction.  In this way we are somehow popped out of the rut of our sankharas and new possibilities open up for us.  

Macy characterizes this observation of our own skittering across the landscape of our habits as the process of choosing itself, and she emphasizes that what's important here is not this or that particular choice but the intention behind these choices.  But it seems to me that the only intentional or volitional or selective activity here is actually to fully experience our passivity and the mystery of how productive doing nothing can be.  In other words, the only true intention is to experience a release from the process by which we appropriate reactions as what we intended and begin building a self around what particular thing we 'chose'.  What happens when we ignore the negative feedback of the homeostatic loops and allow the error to grow instead of reacting to minimize it is anyone's guess.  But I'm not sure the right term for this is "choice".   Positive feedback takes us somewhere new, but it seems that almost by definition this cannot be where we wanted to go.  Of course, if we want to get somewhere genuinely new, letting go may be the only way to do it.  Fundamentally though, we can only choose to open up or not, to say yes or no, to be the kind of systems that overcome themselves or the kind that cling to their limits of life.  Progress along the path is in making this opening a habit, at the limit, perhaps the only habit.

--------

The last few chapters of Mutual Causality in Buddhism and General Systems Theory are devoted to arguing that the ontology of mutual causality necessarily entails a morality of open mindedness, reverence for life, compassion, and economic sharing.  While these are all noble goals that fit well with the deeply interrelated vision of reality she has developed, I don't believe that Macy comes even close to demonstrating that a particular moral stance follows necessarily from her ontology.  The argument would be that all these values are just alternative names for the habit of openness I just described, and that systems in general tend to evolve in this direction.  I've already articulated how I think the generality of this premise is faulty, or at least in need of significant explanation (is there a cosmic system of the World or Life that tends to produce consciousness?).  And it seems obvious that Macy is making the category error of attempting to get morality from ontology, is from ought.  One can argue that being has a tendency, but if this tendency is a law of nature, its hard to see how it could carry and moral force.  Since I find line of thought unrewarding, I'm going to skip these chapters and end on a speculative note related to the last chapter that I think may again emphasize how narrow our choice would be in this sort of moral deduction.

We noticed earlier that when we apply Macy's basic metaphor of a reciprocal causal feedback loop in a temporal, rather than a spatial, sense, a funny thing happens.  The knower here and the known there, the inside and the outside, are the differentiated poles of a circle of mutual conditioning.  But we still think of this circle operating in a single direction in time.  This becomes more obvious when we consider her explanation of the co-arising of doer and deed.  Karma means that our past intentional choices structure our possible present actions, and that this structuring deepens or accumulates every moment that time's circle passes again through the present.  So our present choices structure future possible actions for which these choices will in turn be past.  This makes sense, but does it constitute the same type of feedback loop that we discussing in the context of knower and known, mind and body?  It seems that this model describes a linear type of causality, and that the whole feedback involved is contributed by time itself.  In particular, it doesn't seem we're articulating the co-arising of doer and deed, a co-arising that could only happen in the present moment, so much as simply the arising of future doers from past deeds the way a bigger snowball arises from a smaller snowball with every revolution.  If we want to contemplate a directly temporal form of feedback strictly analogous to the spatial version, one that actually simultaneously produces the two distinct 'opposites' of present doer and past deed as its outcome, then we would need to consider some way the future could influence the past.  This would constitute a true temporal feedback, not just the normal feed-forward accumulation we already associate with time.  I know, it sounds weird.  But since we usually already see time as a feedback loop in which the output of the present is used as input to the future, we have to come up with some other way to think about true mutual causality in time. 

Now, where are we gonna find somebody wild enough, somebody crazy enough to ... helllo!  Nietzsche's Eternal Return is meant to generate precisely this kind of loop. What if we could choose the past, will it to be exactly as it was?  In this case, the past and future would not be essentially distinct but would appear as differentiated dimension of the present, just as the knower and know, the subject and object differentiate themselves from some non-dual ground.  The Eternal Return is temporal non-duality, truly temporal feedback.  While this idea gets pretty weird pretty quick, I think it may help us to understand what I was saying earlier about the limitations of our free choice.  The choice that allows us to keep going, to keep growing, is the same choice that dissolves us, that affirms everything all at once, including our powerlessness.  Whe have to choose for nothing to be different.  We have to desire the endless return of the rut we got stuck in.  To affirm or resist this moment of decision is our only freedom.  Which is why Nietzsche always portrays the Eternal Return as a the greatest weight, as capable of crushing us as of liberating us.  In a sense, we need to create our suffering.