Individuation and Affectivity

Perception is an individuation in itself, but it also serves as a sub-individual in the ongoing individuation of a psyche.  In a way, Simondon already touched on this theme in the prior chapter when he discussed how important the orientation of the entire subject is for the act of perception.  In this section, he goes into more detail about how perception fits into its context as one half of a psychic synthesis that prolongs a vital individuation.  We already know that the other half of this synthesis is action, so the question is exactly how these two sides get coordinated in a feedback loop.  The answer is the central orienting force of what Simondon calls the "affective-emotive" regime.  This is the subconscious layer that acts as an interface between the actions of autonomic animal unconscious and the representational perceptions of a conscious psyche.  For Simondon, this sort of transitional zone is always the center of an individuation, in this case the individuation of the psyche.

Affectivity and emotivity would thus be the transductive form of the psyche par excellence, the intermediary between clear consciousness and subconsciousness, a continual link of the individual to itself and to the world, or rather the link between the relation of the individual to itself and the link of the individual to the world. At the level of affectivity and emotivity, the relation of causality and the relation of finality are not opposed: every affectivo-emotive movement is simultaneously judgment and preformed action; it is really bipolar in its unity (I, 273)

Since affectivity in general is just this connection between how it is on the inside and how it is on the outside, all living beings possess some form of it.  But since Simondon has now broached the psychic realm, it allows him his first chance to begin to talk about some of the more perennial philosophical puzzles of human experience.  For example, if it's really the subconscious affective-emotive level that articulates our subjective individuality, then communication between two subjects must run through this level, rather than the conscious symbolic level (pg. 275).  In addition, our own sense of personal identity, and in particular any sense of this identity's perserverance after death must also be a question of affective-emotive, and not substantial, continuity (pg. 276).  Our intimation of immortality is a completely real feeling (pg. 277).  In fact, the affective-emotive realm contains our entire spiritual life because it is precisely what connects what it within us to what surpasses us, what joins the here and now to the great beyond.  Often we imagine that spiritual life is some abstract life of the mind represented in tangible form by the books and music and films that outlive our biology.  But what is the point of knowing that 'your spirit will live on' if it is not connected to another spiritual being?  The joy we imagine from contemplating the Buddha's influence or our own 'legacy' arises only in the present moment.  Which is perhaps the only reason we give a shit about the 'future of the human race' -- if you all die tomorrow, who will read my blog!?

Culture gives too much weight to written, spoken, expressed, or recorded spirituality. This spirituality, which tends toward eternity through its own objective forces, is nevertheless not the only one; it is only one of the two dimensions of lived spirituality; the other, that of the spirituality of the instant, which does not seek eternity and shines like the light of a glance only to fade away afterwards, also really exists. Spirituality would have no signification if there were not this luminous adherence to the present, this manifestation that gives an absolute value to the instant and consummates within itself sensation, perception, and action. Spirituality is not another life, nor is it the same life; it is other and same, it is the signification of the coherence of the other and the same in a superior life. Spirituality is the signification of the being as separate and attached, as alone and as a member of the collective (I, 278)

With this mention of "the collective" we can start to see the beginnings of Simondon's theory of the transindividual.  Though it follows from his overall concept of individuation, this unique and somewhat counterintuitive theory casts a surprisingly light on what it means to be individuated human psyche, so it deserves to be explored in a bit of depth before we proceed with our discussion of affect and emotion. 

Right from the start, we've seen that one of the constants of Simondon's theory is the way the individual connects two distinct levels or scales or orders of magnitude.  The crystal connects a micro to a macromolecular scale by relating a singularity to an energetic system.  A biological individual (I think) connects a growing species (an aperiodic crystal) to the variations in its environment.  Both of these individuals resolve a problem of growth because they actualize a potential energy that allows the information of a singularity to propagate through a medium.  In the first case, this information propagates as the unchanged structure of the crystal, while in the second case, the information propagated is the local variations necessary for crystallizing a more complex milieu.  In both cases, the individual is an intermediate level that allows a larger and smaller level to communicate.

The individual is always between things, a structure that at once links and separates, an active polarized membrane.  The psychic individual is no different.  It links and separates the vital unconscious and the 'conscious' realm beyond the animal.  Since individual biological affect was already introduced as the way a species spreads by polarizing the environment along certain lines, it would seem that a fuller development of psychic life begins with the time variation of affect.  Psychic life begins when the preprogrammed affective reflexes of the species are not longer sufficient for its propagation.  [This isn't quite right, but we'll come back to the relations of sensation, affect, and emotion shortly]  Affections now have to be integrated, through a learning process, in order to propagate through a more rapidly varying environment.  But this environment is now filled with other psychic individuals undergoing the same process of individuation.  The now more dynamic environment is what Simondon means by the collective or the transindividual.  This shouldn't be confused with the inter-individual, which gives us the impression that the collective is a relation between already fully formed individuals, and might even make us think that we're only talking about a relationship between individuals within the same species (though it does seem to me that there is an outstanding question here about the relationship of the species to the transindividual).  The transindividual is closer to the pre-individual soup that of course never actually went away in this scheme.  Just like with the Tao, there's no individuation that exhausts the pre-individual or discharges all of its potential.  The psychic individual is the link between the biological individual and this yet-to-be-formed wider world, and it's this link that Simondon is calling "spiritual".  It seems that it's less about our relationship with other beings than a relationship between the process by which we are formed and the process by which these others are formed.

The spiritual or psychic levels seems to be precisely what moves us from pain and pleasure to suffering and release -- the first has only a narrow biological meaning, while the second involves our a whole orientation to our self and to the world.

Pleasure and pain are generally interpreted as signifying that a favorable or unfavorable life event emerges and affects the being: in fact, this signification does not exist at the level of the pure individuated being; there may be purely somatic pains and pleasures; but affective-emotive modes also have a signification in the accomplishment of the relation between what is individual and pre-individual: positive affective states indicate the synergy between the constituted individuality and the movement of the actual individuation of the pre-individual; negative affective states are states of conflict between these two domains of the subject. (I, 278)

In a sense, we might say that it's a question of whether we 'line up' with or resonate with our own becoming.  We might think of the transindividual as the collection of individuations-in-progress, which clearly gives our relationship to a reflexive character.  It seems that with the opening of the psychic realm, the individual itself, the individual as a process of individuation begins to intervene in its own ongoing individuation. This is why Simondon will also talk about the psychic as a synthesis (always via resonant feedback, and not through some a priori subject) of emotion and perception that leads to action.  [There's some confusion here since the beginning of the chapter (pg. 273) seems to make the affectivo-emotive the center of a synthesis that coordinates perception and action, while at the end of the chapter emotion appears to be integrated with perception through a collective action (pg. 290)]  A spiritual life cannot be merely contemplative because it is not the act of a finished subject confronting a finished world, but the act by which subject and world co-create one another (pg. 281).  In fact, when this feedback loop between emotion and action doesn't close because the subject comes to feel that it stands apart from the world, what we get is anxiety.  Simondon devotes a whole section to anxiety or existential dread or Heideggerean being-towards-death or whathaveyou, in which he beautifully articulates the problem -- our life, considered by itself, is utterly meaningless. 

... in anxiety, the subject feels itself to be a subject to the extent that it is negated; it bears its own existence in itself, it is weighed down by its existence as if it had to carry itself—a burden of the earth (ἄκθος αρούρης) [ákthos aroúres], as Homer says, but also a burden to itself, since the individuated being, instead of having the ability to find the solution to the problem of perceptions and the problem of affectivity, feels all problems flowing back into it; in anxiety, the subject feels as if it exists as a problem posed to itself, and it feels its division into pre-individual nature and individuated being; the individuated being is here and now, and this here and now prevent an infinity of other here and nows from coming into existence: the subject becomes conscious of itself as nature, as undetermined (ἄπειρον) [ápeiron], and as something that it will never be able to actualize into a here and now, that it will never be able to live; anxiety is diametrically opposed to the movement through which one takes refuge in one's individuality; in anxiety, the subject would like to resolve itself without going through the collective; it would like to come to the level of its unity by way of a resolution of its pre-individual being into an individual being, a direct resolution without mediation or delay; anxiety is an emotion without action, a feeling without perception; it is the pure reverberation of the being within itself. (I, 282)

What most interesting in all of this is not the complex details of Simondon's analysis of psychic individuation.  More important is simply the idea that, first, we are constructed as an individual subject, and second, that this individuality centers on an affective-emotional level and not the conscious rational level we usually take for granted.  These two ideas shift our whole orientation to the question of "what is the meaning of life?" in a much more useful direction.  It becomes more of a question of seeing, and feeling, how meaning is created though living.

However, I would be remiss if I didn't jot down a few notes about Simondon's complex scheme in 2.2.2.6 (pgs. 285-291) in case they should come in handy later.  The basic equation here is sensation/perception = affection/emotion. Sensations and affections are like basic organic building blocks that result in a polarization of the external and internal worlds (respectively) according to a set of gradients.  They are related by the tropisms of simple organisms.  Move towards the light.  Follow that increasing concentration of food molecules. Perception and emotion are the higher level correlates of these same concepts, and they are related by action, properly speaking.  So perceptions integrate sensations that have become complex and self-contradictory, and emotions do the same for the wild variety of affections we feel.  Because of their integrative function, perception and emotion are more self-sustaining, more the cause of themselves, than sensation and affect.  This is how the constitution of the psyche returns to intervene in its own unfolding. 

Signification and the Individuation of Perceptive Units

Chapter 2 of Part 2 begins much more lucidly than the preceding chapter on biological individuation.  While I'm not completely clear why this should be so, I think one reason may be the unstable intermediate position of purely biological individuation in Simondon's overall theory.  The book begins with physical individuation, and the present chapter discusses psychic individuation as a part of the collective individuation that Simondon calls the transindividual.  It's not clear yet whether he conceives the transindividual as the end of the process of individuation, but it is certainly the end of the book.  So a purely biological individuation has to fit somewhere between these starting and ending points.  And since, as we'll see presently, Simondon considers perception to be an aspect of physic individuation, he draws the boundary of the psyche at a lower threshold than many might.  For example, it's clear that he would attribute an individualized psyche to each of my cats.  Once we let go of our anthropocentric bias in defining the psychic realm, it becomes pretty hard to limit its spread down the taxonomic tree of life.  The narrow space then left to the purely biological individual perhaps goes some way in explaining the difficulties of the last chapter, as well as its reliance on examples mainly derived from the life of colenterates; certainly no mammal lacks a psyche. 

In a way though, this same problem of instability effects the psychic realm as well.  In fact, since each type of individuation is inserted inside the preceding one, it seems that psychic individuation would be even less stable biological individuation.  This is part of the design of Simondon's system -- each new level is more metastable than the last.  That is, it represents a more precarious equilibrium, but one that, by the same token, can incorporate a much wider range of both internal and external information.  As the individuations cascade, each leads on to the next with increasing rapidity.  Each resolves a problem at the previous level ... by creating even more problems at the next level.  Simondon actually introduced us to this idea in the previous chapter, where he again used the concept of neoteny to relate psychic to vital individuation.

... the appeal to psychical life is like a slowing down of the living being, which conserves this slowing down in an extended and metastable state that is rich in potentials. The essential difference between simple life and the psyche consists in the fact that affectivity does not perform the same role in these two modes of existence; in life, affectivity has a regulative value; it dominates the other functions and guarantees this ongoing individuation that is life itself; in the psyche, affectivity is pressed on all sides; it poses problems instead of resolving them and leaves the problems of the perceptive-active functions unresolved. The entrance into psychical existence essentially manifests as the appearance of a new problematic which is higher and more difficult and which cannot receive any veritable solution from within the living being, properly speaking, conceived within its limits as an individuated being (I, 177)

This whole section (2.1.1.2 -- pgs. 177-180) was rather obscure at the time.  When we return to discuss affectivity in the next post we'll be able to make more sense of it.  But the basic idea we need from it now is just that perception is an integration of disparate sensations that solves some vital problem.

Simondon treats the perception of an object as a type of individuation, a specifically psychic individuation that occurs within an ongoing vital one.  This way of approaching things cuts across a lot of stale debates about truth and objectivity because it takes perception out of the domain of representation.  My percept of a cat is not a little picture of a cat in my brain, nor is it any coded version of that picture such as the firing of the 'cat neuron'.  Perception is not something that happens in the head of the subject, but between the subject and object, and its value does not lie in representing the world, but in organizing action in it.  In this way, perceptions are both structured by the world, and also return to structure it.  As with every other individuation, it's really the mutual resonance or feedback between two sides -- the structure of the subject and the structure of the object -- that allows for the individuation of a perception.  Or if we state this the other way around, taking the non-dual perspective, perception is a phase shift in the subject-object system that crystallizes a subject which sees and an object which is seen.

Before perception, before the genesis of the form that perception precisely is, the relation of incompatibility between the subject and the milieu only exists as a potential, similar to the forces that exist in the phase of metastability of the supersaturated solution or the supercooled solid, or even in the phase of metastability of the relation between a species and its milieu. Perception is not the grasping of a form but the resolution of a conflict, the discovery of a compatibility, the invention of a form. This form that perception is modifies not only the relation between the object and the subject, but also the structure of the object and the structure of the subject. (I, 259)

These days, after contributions from behavioral economics, psychology, and meditation, I think it's easier for us to appreciate how the structure of the subject contributes to perception in a way that has nothing to do with representation.  Put simply, we are good at perceiving things that are salient to us, and pretty bad at perceiving anything else.  We don't even see the gorilla if we're busy counting.  Stopping with this idea though can give us the impression that perception is nothing but selection.  As if all the possible objects were given in their full detail in advance and we merely chose to pick out some of them and not others.  This doesn't account for the generation of these possible objects as objects of perception.  What holds them together as objects?  Part of the answer lies again in the structure of the subject.  We select traits to characterize objects as relevant because we plan to do something with them.  Every perception is an incipient action, even if it remains unperformed.  But these perceptions are shaped by the success or failure of past actions, and so depend on the real structure of the object.  We are not just making this up.  If we perceive an object where the world does not afford one, we are in for a bad afternoon.  The feedback loop between subject and object cannot close to create a perception without contributions from both sides. 

There's a sense, then, in which the perception of an object is like an extension or continuation of the physical individuation that created it.  For perception to happen, this physical system must be coupled to the psychical system.  Objects of perception fall out of this coupled system as so many distinct frequencies of resonance, as it were.  We discover a bundle of correlated real properties of a physical individuation that make the object useful for us, in our ongoing vital individuation.  Simondon calls these resonances the quantum nature of perception.  The idea is that while physical reality is continuous (at least relative to our scale), we interact with it in a discreet manner.  But we can do this only because reality itself allows for it -- its continuity naturally comes  divided up by thresholds created by the phase shifts of the physical individuation. 

What humans perceive in objects when they grasp them as individual is therefore not an indefinite source of signals, an inexhaustible reality, like matter, which allows itself to be analyzed indefinitely; what they perceive is the reality of certain thresholds of intensity and of quality maintained by objects. ... the physical object is an organization of thresholds and of levels that is maintained and transposed throughout various situations; the physical object is a bundle of differential relations, and its perception as individual is the grasping of the coherence of this bundle of relations. A crystal is an individual not because it possesses a geometrical form or an ensemble of elementary particles, but because all of its (optical, thermal, elastic, electrical, piezo-electrical) properties undergo an abrupt variation when we pass from one facet to another (I, 264)

We don't see all of reality as it is.  But what we do see is reality, or at least, something that can really happen to an objective reality.  Amongst other things, many physical individuations are capable of being perceived because they have effects which propagate through the environment (and in particular, propagate through the part of the environment we call a nervous system) in a coherent way.

Perception then is like a resonance between what the object can do and what the subject can do.  Simondon talks about the details of this resonance in two distinct but related ways -- in terms of orientation, and in terms of information.  We've seen that since it is always characterized by a threshold or transition, every individuation involves a polarization or orientation of the milieu.  Consider the crystal in the quote above.  It's individuation as a structure distinguishes two sides where before there was one.  A previously amorphous milieu acquires a direction.  It's this orientation that actually allows the physical individuation of the crystal to continue or extend in the manner Simondon indicates.  The crystal is not so much a substantial unity of atoms with a certain structure as it is a propagation of the process by which is polarizes its medium.  That's why the crystal has identifiable 'properties' that make it distinct from its medium -- it has has different effects on other things than the uncrystalized solution, or perhaps better said, it's polarity means that it is taken up differently in other larger ongoing physical individuations.  But the living individual also has a polarization or orientation.   For the purely biological individual, Simondon seems to think that this orientation comes not from the individual itself but from the affectivity built into it in its role as a sub-individual which helps propagate the species as an individual (see pg. 177, and the distinction between drives and tendencies on pg. 183).  But for a more individuated species (ie. where each member seems more unique to us, ie. where a physic individuation has already begun to be layered onto the vital level) this affectivity expands into a whole emotional orientation to self and world.  What is it we personally fear and love?  What do we desire?  This individuation of a subject is also a sort of polarization or orientation.  And of course, it's the resonance or lining up of these two orientations that constitutes a perception.  To construct that object as desired is to couple the orientation we seek with the orientation the world offers to us.  Perception has an innately emotional component.

The unit is perceived when the reorientation of the perceptive field can be effectuated in line with the object's own polarity. To perceive an animal is to discover the cephalocaudal axis and its orientation. To perceive a tree is to see in it the axis that goes from its roots to the end of its branches. Every time the tension of the system cannot be resolved into a structure, into an organization of the subject's polarity and of the object's polarity, an uneasiness remains that habit is hard pressed to destroy, even if every threat has been removed. (I, 261)

Simondon also talks about this resonance of subject and object in information theoretic terms.  Here, his point is that perception is neither a process of finding the simplest and best forms in nature, such as the square and circle, nor in representing its complexity with perfect fidelity, but of discovering the most intense forms for the subject.  Gestalt theory tried to argue that we innately perceive simple geometric forms because these were the 'good' forms that couple be found everywhere and combined into everything.  In information theoretic terms though, these forms are highly compressible and therefore contain very little information in themselves.  So then how would we know how to combine them into a useful perception?  On the other hand, because of its technological roots, information theory imagines that more information -- more pixels, higher bit rates -- is always better, and that perception is just about gathering as much data as possible.  In this case though, how would we know what to do with all this data?  Representing the full complexity of reality is impossible for a finite nervous system and would moreover be completely useless to the biological organism.  Information and Gestalt theory are both hylomorphic theories of perception that obscure the way subject and object interact.  Instead, perception is all about finding the correct level of compressibility of the world's information for the particular needs of a given subject.  The forms perceived will contain an intermediate amount of information, one that resonates with the problem of choice faced by the subject as well as the coherent nature of the objects' effect on the world.  Simondon calls this the intensity of information, as opposed to its pure quantity or quality. 

We can trace how these two perspectives on resonance come together in the following quotes.  They also give you a sense of how Simondon's scheme could be compatible with the Bayesian inference model of perception (so long as we rid the foundations of the model of its atavistic reliance on the notion of representation -- note to self: is there some way to recharacterize the minimization problem of the free energy principle as a question of resonance?).

Above information as quantity and information as quality, there is what could be called information as intensity. The simplest and most geometrical image is not necessarily the most expressive; the image that has the most meaning for the perceiving subject is not necessarily the image that is most elaborated and meticulously analyzed in its details. The entire subject (with its tendencies, drives, and passions) must be considered in a concrete situation and not as a subject in the laboratory, i.e. a situation that generally has little emotive value. (I, 267)

The quantity of signals only produces an unpolarized ground; the structures of good forms only provide frameworks. It does not suffice to perceive details or ensembles organized in the unity of a good form: these details and ensembles must have meaning with respect to us and be grasped as intermediaries between the subject and the world, as signals that allow for the coupling of the subject and the world. The object is an exceptional reality; what is usually perceived is not the object but the world, which is polarized in such a way that the situation has a meaning. The object properly speaking only appears in an artificial situation that is somewhat exceptional. (I, 268)

It is not enough to simply say that perception consists in grasping organized wholes; in fact, perception is the act that organizes wholes; it introduces organization by analogically linking the forms contained in the subject to the signals received: to perceive is to retain the greatest possible quantity of signals inside the forms most deeply rooted in the subject; perception is not merely grasping forms or recording multiple juxtaposed or successive data; neither quality, quantity, the continuous, nor the discontinuous can explain this perceptive activity; perceptive activity is the mediation between quality and quantity; it is intensity, the grasping and organization of intensities in the relation of the world to the subject. (I, 269)

The overarching point is that perception is just like any other individuation.  It is a transduction of energy into (in this case psychic) structure, a crystallization of a milieu based on its internal resonance, a partial resonance of the world with itself that allows two sides to be simultaneously distinguished and coupled. 

Information and Ontogenesis

The first chapter of part 2 is so sprawling, difficult, and seemingly disorganized that I'm not sure how to handle it.  Basicailly, it investigates how the notion of individuation that Simondon developed with such care and consistency in his exploration of crystal formation might apply to the biological realm.  So, what is a biological individual and what makes it different from a physical individual?  Unfortunately, either life is so complicated that Simondon finds it difficult to consistently extend his schema into this area, or I am so dense that I haven't understood how the extension works.  Either way, while interesting in many respects, this chapter is confusing enough that I can't succinctly answer the question.  The best I'll be able to do is noodle for a while on the problem.

The root of the difficulty in defining the living individual is that life comes in so many levels.  There's what we usually call the living individual, namely myself or my cat.  But then there seem to be things that operate like individuals inside these -- to some extent, my cells have an individual 'life' of their own.  On the other hand, my individual life is pretty heavily dependent on the group in which I live, which forces us to ask whether this group is the true individual.  In fact, if we follow the analogy of crystal formation and attribute the reality of the individual to the capacity for an in principle unlimited propagation whose expending front phase shifts the matter it encounters, then it seems we should call the species the living individual.  After all, I cannot personally reproduce myself as an identity, nor expand indefinitely in time or space, but I do appear to be at least partly responsible for propagating (or in my case failing to propagate) the human species.  So perhaps what we usually call a living individual is better thought of as a sub-individual, a site at which the species individual crystalizes the environment, as it were. 

In short, the problem is that there are too many levels that could serve as candidates for the unit of vital individuation.  And what's worse, these levels are clearly all related.  Me, my cells, and the I writing to you somehow all operate together.  These interrelations seem to mire us in hopeless complexity, and that's definitely how I feel about much of this chapter.  But in a way, these complex interrelations seem to be part of the solution for Simondon -- vital individuation is precisely the sort of process that breaks down into these distinct but interrelated levels.  This is analogous to the way physical individuation linked together but also separated two different scales of organization, the micro (singular, structural) and the macro (energetic).  In this chapter, Simondon no longer mentions the micro and the macro, which seem to have been replaced with the distinction and interrelation between interior and exterior milieus and the paired concepts of integration and differentiation.  What before was a simple phase shift propagating itself by perfect repetition only at the edge of a crystal now becomes a reaction that can only advance by changing form.  But this propagation of difference, this constant mutation, nevertheless constitutes the same individuation.  It puts me in mind of Schrodinger's definition of life as an aperiodic crystal.  In other words, life seems to be a sort of individuation that can only continue if it breaks down into levels.  It's almost as if the unit of life can only propagate by splitting. 

The idea that individuation continues by fragmenting the individual into levels might seem like a sort of vitalism.  The sub-levels could appear to be mere tools of the higher levels, all the way up to the highest level where a mysterious Life runs the whole hierarchy.   But Simondon conceives of this process entirely in terms of matter-energy and information, so if it constitutes a vitalism it's of a peculiar materialist flavor.  We've already seen that Simondon's idea of 'inert' (ie. non-organic) matter is entirely different from the hylomorphic model.  Physical matter is already potentiated, already capable of forming itself under the correct conditions. So we don't need the vital to explain complex and interesting organizations of matter.  In fact, Simondon begins the chapter (pg. 170) by implying (somewhat contrary to what I've suggested above) that we don't necessarily even need the vital to explain a complex multi-level individual.  This fits with his overarching goal of showing us a continuity of individuation between the physical and the vital.  We need to always keep in mind the idea of neoteny that appeared at the end of the last chapter -- the vital individual is not constructed on top of the completed physical individual but instead inserted inside of an incomplete physical individuation.  A species is a very strange way to crystallize an environment into similar units, one that in turn requires the complex individuality of these similar units to proceed.  In this type of thinking there can be no sharp dividing line between inorganic and organic matter, since the latter merely extends the possible behavioral space of the former. 

However, as always, Simondon's in principle continuities are broken up by thresholds -- phase transitions where new stuff is created out of the same old stuff.  In this case the threshold seems to involve not the absolute level of organization, but what Simondon calls the "regime of information" (pg. 208) -- essentially a measure of how information propagates in a system.  While it's not completely clear to me how these two concepts interrelate, it seems that the regime of information refers to something like the topology of a causal network (as opposed to simply its number of levels -- though it seems like there ought to be some relation between these characteristics).  What parts of the system are densely interconnected and what parts sparsely?  Do distinct parts of the system function mostly independently, with only a weak coupling, or on the contrary, do we need to integrate information from a whole hierarchy of levels to understand what's going on at a single particular level?  Perhaps the idea is that with an increase in the absolute number of levels, we come to a point where the combinatorial possibilities of interlinking these levels reaches a phase transition?  I'm imagining something like phase transitions in network wiring diagrams.  At any rate, where the crystal only propagated information locally and at its edge in Euclidean space, in the living individual, information can propagate non-locally in both space and time (pg. 225).  The Euclidean interior of a crystal has nothing to with the operation of individuation happening at its edge.  You can hollow out the crystal and it will continue to grow.  However, the entire complex organization of the Euclidean interior of a living organism is necessary for its propagation.  In fact, we can say that this entire interior is in contact with the exterior in a living individuation (pg. 253).  The result is a non-Euclidean topology where the inside appears distinct from the edge of the individual.  In other words, the living individual has a sort of depth to it where the physical individual is totally superficial, it 'lives' only at its edge. 

For Simondon, depth is a dimension that opens up when disparate things are integrated as different aspects of a single system.  Which means that individuation is almost defined as the creation of depth.  His favorite image for this is the way that the two distinct 2D optical images of our eyes can be integrated into a single 3D image that resolves their conflict.  While this integration of differentiation is at the core of all individuations, it seems to be particularly emphasized in the case of the multi-level living individuals because it seems it can be repeated in a fractal pattern.  This image begins to help us account for the difficulty we had in defining the proper unit of vital individuation at the outset.  In fact, it seems that the interior of the living individual can often function as the exterior milieu of another individuation.  Life seems prone to plunging into depth.

Form and Substance

I'm not going to cover chapter 3 in the detail to which loyal readers have become accustomed.  It would require too long of a detour into the history of quantum mechanics to do it justice.  In addition, most of this chapter was not published with the original edition of the book for good reason -- it's an extremely technical tangent to the main philosophical argument that also ends rather inconclusively.  So I just want to place this chapter in its context and extract its main contribution to Simondon's overall theory of individuation.

We ended chapter 2 by considering the most abstract version of this theory.  Individuation requires both a discontinuous or singular component and a continuous energetic one.  Crystal formation required both the energy available everywhere in the amorphous material as well as the elementary discontinuity of the periodic spacing of crystal atoms to serve as a seed capable of polarizing this material.  The crystal as an individual was actually the process of propagation of a discontinuity though a continuous milieu, the resonance between levels that allows a micro discontinuity to amplify into a macro phase shift in milieu.  In this sense, the crystal is the mobile limit at which more crystal can be made.  In principle, this limit has no largest upper bound; the crystal will grow indefinitely large as conditions allow.  However, it does have a smallest lower bound; the periodic spacing of crystallization sites means that the crystal can never grow between those sites, or at a scale smaller than them.  This is what Simondon means by the "elementary discontinuity" needed for the process of crystallization.  Every individual requires some singularity to get it started.  

In this context, it makes sense to ask if there is an absolute smallest elementary discontinuity that is required for any individual at all to form.  Notice though that this elementary discontinuity cannot possess the substantial identity of the classical atom.  It's role in Simondon's scheme is to serve as a singularity that is capable of propagating itself into a continuum.  Here we start to see something that reminds us of the wave-particle duality of quantum mechanics.  The electron isn't a simple particle with a given size, mass, and position but also a wave that can appear anywhere and even interfere with itself.  Somehow, when we get down to this very smallest level, we suddenly find it spontaneously spreading out everywhere.  And we also find that its structure is always associated with discontinuous changes in energy levels (energy quanta) -- Einstein's famous formula recapitulates Simondon's assertion that relation inter-converts with substance.  This may all sound intriguing, and many philosophers would surely have stopped here.  Simondon, of course, is never happy with this sort of vague analogy, so he spends the next 70 pages surveying the entire history of quantum mechanics (up to when he was writing in 1958) to try and flesh it out.  Unfortunately, most of the meat of it is quite simply above my pay grade.

In the end though, these long reflections on quantum mechanics bring us back to the question of continuity and discontinuity, energy and structure.  It's clear that Simondon's notion of individuation requires both, and that the general process is one of successive "transductions" of energy into structure.  A propos of some reflections on the quantum measurement problem that I won't go into, Simondon finally characterizes individuation as shifting between a deterministic mode in which a system evolves smoothly in time (wave mechanics), and an indeterministic one where the system undergoes an instantaneous structural transformation (measurement).   

Furthermore, if we contemplate microphysical reality directly, an interpretation of individuation starting from the phenomena of structural change would aim to consider becoming as essentially linked to the operations of individuation that are carried out in successive transformations; determinism would remain applicable as a borderline case when the system considered is not the theater of any individuation, i.e. when no exchange takes place between energy and structure (which would modify the system's structures), thus leaving it topologically identical to what it was in its previous states; on the contrary, indeterminism would seem like a borderline case when a complete structural change manifests in a system with the transition from one order of magnitude to another; this is the case, for example, of the modifications brought to a system by the fission of an atomic nucleus: intranuclear energies, which up to that point belong to the internal system of this nucleus, are unleashed by fission and can act as a gamma photon or a neutron on the bodies that belong to a system situated on a scale larger than that of the atomic nucleus. (I, 158)

These two modes of evolution characterize the system's "chronology" and "topology" respectively.  While we can conceive of each of them separately and independently, in general they are tangled together, and individuation is precisely the process by which these knots get tied in feedback loops.

The general case is that of a certain level of correlation between a system's chronology and topology, a level which is moreover variable due to the vicissitudes of its own becoming; a system reacts on itself not only in the sense of the principle of entropy through the general law of its internal energetic transformations, but also by modifying its own structure through time. The becoming of a system is the manner in which it individuates, i.e. essentially the manner in which it is conditioned according to the different structures and successive operations through which it reverberates within itself and phase-shifts relative to its initial state. Determinism and indeterminism are merely borderline cases, because there is a becoming of systems: this becoming is the becoming of their individuation; there is a reactivity of systems with respect to themselves. (I, 159)

Becoming, then is neither deterministic nor indeterministic, but a sort of punctuated equilibrium where each new topological exclamation point is produced by some internal resonance in the chronological dynamics of the previous stable era.  Fission serves as the perfect example of a phase shift because it is created through a runaway feedback loop which makes every atom resonate with the others in a way that completely alters the structure of the system, converting atoms back into energetic radiation.  The A bomb is a singularity.  Notice, though, that the A bomb is not really an individual.  Resonance here leads to a  complete conversion of structure back into energy.  Instead of a deterministic physical system operating at the atomic scale, we have a new deterministic system operating at a  planetary one.  But these two scales are connected by a single instant of discontinuity in the form of the critical threshold at which runaway fission begins to take place.  For Simondon, the true individual is the ongoing process by which different scales of reality can become related, the ongoing, semi-stable, self-modifying inter-conversion of energy and structure.  In this sense, the nuclear reactor is more individuated than the bomb because it provides for an ongoing communication between structure and energy and a regulation of their inter-conversion (I, 161).

At this point, we can see how the long detour through quantum mechanics leads us directly to the discussion of the living individual that will come next.  We wanted to know whether reality was fundamentally continuous or discontinuous at its lowest level.  And the clear answer is no.  These two aspects of reality always appear as complementary pairs not because they both characterize the real but because they are both the outcome of a real process of becoming that individuates a system.  Simondon calls them the "dimensions" of the system because they reflect the time and space, the chronology and topology, of the individuals it produces.  These are the existing (meta)stable structures that move around, or on the contrary the new structures that fill the system up as it evolves.

From this point of view, it seems possible to understand why the antagonistic representations of the continuous and the discontinuous, of matter and energy, of structure and operation, are not usable except as complementary pairs; this is because these notions define opposite and extreme aspects of the orders of reality between which individuation is established; but the operation of individuation is the active center of this relation; it is the latter that is the unity of this center that splits into aspects which are complementary for us, albeit in the real they are paired by the continuous and transductive unity of intermediary being, what we call here internal resonance; the complementary aspects of the real are extreme aspects that define the dimensionality of the real. Since we can only grasp reality through its manifestations, i.e. when it changes, we only perceive extreme complementary aspects; but, rather than the real, what we perceive are dimensions of the real; we grasp its chronology and topology of individuation without being able to grasp the pre-individual real that subtends this transformation. (I, 162)

The real process of individuation (which for Simondon is the real, full stop) lies in the ongoing back and forth between these dimensions of continuity and discontinuity.  And the ability to constantly shuttle back and forth between these becomes the defining characteristic of the living individual.  After all, it's distinctiveness doesn't lie in a special type of matter that makes it essentially different from the physical individual, but in a special organization of matter that leaves it more sensitive and responsive to changes in the material around it. In the living individual, structure and energy go back and forth constantly, not in one swift explosion of form. 

... there is physical individuation when the system is capable of receiving information a single time, then develops and amplifies this initial singularity by individuating in a non-self-limited way. If the system is capable of successively receiving several inputs of information (of compatibilizing several singularities instead of iterating the single and initial singularity cumulatively and through transductive amplification), then individuation is vital, self-limited and organized. (I, 163)

Finally, this leads us to a really unique perspective on what makes the living individual different.  As we've seen, Simondon's understands the individual as an unfinished entity.  As far back as the introduction we observed that 'higher level' individuals like the living, psychic, or social, were not built on top of one another like layers of a pyramid where the completion of the previous layer is essential to support the following one.  Instead, each new level is a continuation or prolongation of what remains unfinished and unstable at the previous level.  It's exactly this instability that allows the individual to process any new information, to be affected by the world around it on an ongoing basis.  The more individual we become the more unstable we become as matter, and the more different things can affect us.  Higher levels are then characterized by a slowing down of the transition to a more stable state.  Just like the nuclear reactor, we have to catch the explosive phase transition somewhere in the middle and (partially) stabilize this inherently unstable moment.  What we're effectively describing here is a sort of physical neoteny.  The living individual prolongs the impressionable youth of a physical system's phase transition.

    According to this way of viewing things, vital individuation would come to be inserted in physical individuation by suspending its course, by slowing it down and by making it capable of propagating in the inchoate state. The living individual would be, in some sense and on its most initial levels, a crystal in the nascent state that is amplified without stabilizing.
    To relate this schema of interpretation to the most current notions, we can appeal to the idea of neoteny and generalize these types of rapports between classes of individuals by supposing a slew of possible neotenic developments in the category of living beings. In a certain sense, animal individuation can be considered more complex than vegetal individuation. However, the animal can also be considered an inchoate plant that develops and becomes organized while conserving the motive, receptive and reactional possibilities that appear in the reproduction of plants. (I, 164)

Form and Energy

In an effort to speed things up, I'm going to attempt to cover all of Chapter 2 in a single post.  This is a bit ambitious, since it contains both a number of subtle philosophical points and a bunch of detailed technical discussion.  However, I think the basic thrust of the chapter is pretty clear.  Last time we came to the conclusion that the individual, properly speaking, is whatever can continue a process of individuation.  We also know that this process requires both certain energetic conditions as a well as a singularity that allows for some of the system's potential energy to become actual.  But we're still not sure how this process might function in the absence of human intervention, because we only discovered the need to describe it as a result of discussing the hylomorphic schema's failure to adequately account for the individual physical brick produce in a technical operation.  So in this chapter, Simondon will move on from his critique of hylomorphism to provide a positive model of physical individuation that illustrates exactly how the process works in nature.  While he uses the term "allagmatic" to describe this operation, we could as easily call it the 'crystallization schema' of individuation, because a detailed consideration of crystal growth will serve as the fundamental analogy for all purely physical individuation.  Why does Simdondon choose to investigate crystallization?  Because it provides a very pure example of an individuation that immediately subverts the paradigm of substance.  Water vapor and ice are precisely not different substances, yet they are clearly different individuals.  In fact, it's not even clear if we should consider water vapor an individual at all, or if so, on what level.  So studying crystallization provides a perfect way to study how the individual is produced from the pre-individual, which is of course the overarching goal of the whole book.

Structures and Potential Energy

1.2.1.1 The Potential Energy and the Reality of the System; Equivalence of Potential Energies; Dissymmetry and Energetic Exchanges

With that orientation in mind, we can see that this chapter consists in Simondon separately taking up each part of the process of individuation -- the energetic conditions, the singularity, and their relation.  He begins by discussing what it means for a system to have potential energy.  While this is related to the standard definition of potential energy in physics, Simondon wants more precise than usual about how we use this term.  Potential energy only exists as such if it can be converted into some real change in the system.  What we're interested in is the system's potential to support individuation.  This effectively precludes a system resting at some stable equilibrium from having any potential energy -- by definition, these are systems that are no longer changing, and so by implication they cannot be involved in an ongoing process of individuation.  It's instructive to consider the physicist's immediate objection to this definition of potential energy.  For example, don't we usually think of a hot gas as having potential energy even though it may exist in a state of internal equilibrium where the average temperature is uniform throughout?  Of course we do, but notably only because we never imagine that the hot gas is the only thing in the system.  We imagine hooking it up to something, say another material that it heats up or a piston that it pushes.  But then it's not the gas itself that has potential energy, but this larger system that we imagined coupled to it, a system which we imagine changing precisely because it is not at equilibrium.  So if we try to imagine a system that has potential energy in itself, we are forced consider a non-equilibrium system that has some sort of asymmetry or heterogeneity built into it. 

The capacity for an energy to be potential is strictly linked to the presence of a heterogeneity, i.e. of dissymmetry relative to another energetic support (I, 55)

Simondon draws some profound consequences from this relatively simple observation.  Because it means that energy cannot be a simple scalar quantity that belongs to a single body in itself.  There is no absolute energy.  Energy requires duality, multiplicity.  A single energetic system must already have parts.  Energy then must be about the relation between parts, and not just characterize the individual substances that comprise those parts.  In fact, for Simondon as much as Deleuze, what is real is always relation, and never the in-itself.  However, what mainly interests Simondon in this chapter is the way that the reality of a relation like potential energy can actually be converted into 'substance' (though this is speaking loosely, since he's already described how the substance of substance, as it were, is relation).  Potential energy can be actualized.  As we'll see, this is what happens when a crystal grows in a suitable medium.  

But the reality of potential energy is not that of an object or a substance consisting in itself and "having no need of anything else in order to exist"; indeed, it requires a system, i.e. at least another term. No doubt, we must struggle against the habit that leads us to grant the highest degree of being to sub- stance conceived as absolute reality, i.e. reality without relation. Relation is not a pure epiphenomenon; it is convertible into substantial terms, and this conversion is reversible, like that of potential energy into actual energy. (I, 56)

But before we get to this point where potential energy is actualized in an irreversible way by literally crystallizing into a substance, SImondon seems to want to make sure we understand the subtleties of his definition by illustrating some cases where the actualization of a potential energy is reversible.  These are systems where potential energy can be converted into a structure and then converted back into potential energy with no loss.  Though, since Simondon illustrates the idea by considering different types of pendulums, his example ends up straining our normal usage of "structure".  

He considers three types of pendulums that transform energy in three different ways.  First, there's the familiar pendulum of freshman physics, which converts the gravitational potential energy of maximum deflection in one direction into the kinetic energy of the swinging bob and back into an equal gravitational potential at the point of maximum deflection in the opposite direction.  Second, there's the curious Holweck-Lejay pendulum, which inverts the situation and fixes the bob above the rotation point at the end of a flexible spring which oscillates back and forth as the pull of gravity plays tug of war with the bend in the spring.   This non-linear pendulum effectively converts a potential energy equal to the difference between the gravitational potential and the elastic potential into kinetic energy and back again.  So if the spring were perfectly balanced with gravity, the period of oscillation would be infinite, which is to say that it would not oscillate at all (observing the frequency of this pendulum apparently provides a much more accurate way of measuring gravity, based on choosing a spring with a potential very close that of gravity, so that the period produced is quite long and easy to measure accurately).  Third, Simondon discusses two coupled simple pendulums.  In this case, the underlying potential energy driver is again gravity, but if we start only one pendulum swinging we see that this potential is converted into a kinetic energy, which serves as a potential energy to start the other oscillator moving. So it appears as if the potential energy is being transferred from one part of the system to another.  All of these examples illustrate the way potential energy can be converted into a "structure" -- in this case the behavior of motion or rest of the pendulums -- and back into potential energy.  And in each case, this potential energy depends on some asymmetry in the system that prevents it from stopping at its point of greatest stability.  The simple pendulum is symmetrical with respect to displacements about the bottom position, but these positions together have an identical asymmetric potential with respect to that point.  The Holweck-Lejay pendulum relies on a carefully balanced asymmetry between gravitational and elastic forces.  And the coupled pendulums obviously represent a system with two distinct parts (as long as they are not started with equal amplitude and phase, in which case the problem reduces to the first type of pendulum). It's their asymmetry that allows each of these systems to change potential energy into kinetic and back again.  Which is basically the definition of oscillation.

1.2.1.2  Different Orders of Potential Energy; Notions of Phase Changes and of the Stable and Metastable Equilibrium of a State. Tammann's Theory

As we mentioned though, not all transformations of potential energy are reversible, and Simondon now takes up the example of crystal formation to illustrate exactly how energy can be more permanently converted into something we would typically call a structure.  In the context though, it's important to note that this doesn't entail any loss of energy.  Simondon still conceives of this as a conversion of one type of potential energy into another, only this time it's not an interconversion of the potentials at the tops of a pendulum's swing, nor a conversion of the difference between gravitational and elastic potential into motion, but a conversion of one order or scale of potential energy into another.  This changing of the order of potential energy turns out to be exactly what we mean my thermodynamic irreversibility.  Entropy always increases in a closed system because energy that began as macroscopic is converted into microscopic form.  A crystal of course is an example of the opposite sort of irreversibility because it constitutes a (local) entropy decrease. In fact, the arising of a structure and a local entropy decrease are almost synonymous.  A crystal structure results from the conversion of a microscopic potential energy into a macroscopic form.  It's a link between these two orders of energy.  In a sense, macroscopic form is nothing other than microscopic organization, but this organization is only possible with the emission or absorption of potential energy.  Which is precisely to say that energy can be converted into structure.

Simondon illustrates his philosophical idea that crystallization represents a substantialization of relation (potential energy) with a particularly obscure and subtle technical example.  He explains Tammann's theory of the crystallization of an amorphous solid in enough detail that I had to read it several times before I figured out what was going on.  I don't think it's worth going into the details except to say that he could have used a better phase diagram.  The overall point he's trying to convey is that there can be two forms of a material, an amorphous one like glass and a crystalline one, both of which we normally consider 'solid', that can convert into one another under certain energetic circumstances.  But these conversions are inevitably accompanied by the emission or absorption of energy in the form of latent heat or of volumetric change of the crystal.  So it's the energy of an overall system that defines the limits of where a structurally stable individual can form. 

The limits of a structural type of domain of stability are determined by energetic considerations. This is why, in order to broach the study of physical individuation properly speaking, we wanted to define the energetic aspect of the relation between two physical structures. An energetic characteristic is linked to every structure, but, inversely, a modification of the structural characteristic of this system can correspond to any modification of the energetic conditions of this system. (I, 67)

Conversely, the formation of an individual crystal corresponds to, and in fact can reveal, an energetic change in a system.  Some of the potential energy of an amorphous solid can be converted into a crystal.  And then this crystal can melt and release its structure as potential energy again.  But these types of conversions of potential energy are not generally reversible because they involve a discontinuity between two completely different states of the system. 

Conversely, the state changes undergone by the system forces us to consider a certain energy linked to the structure, an energy which is indeed a potential energy, but which is not capable of an ongoing transformation; for this reason, it cannot be considered suitable for the case of identity or of equality defined above. This energy can only be measured in a state change of the system; while the state remains, it is conflated with the very conditions of stability of this state. This is why we will choose to name those energies that express the limits of stability of a structural state as structural potential energies. These potential energies constitute the real source of the formal conditions of possible geneses. (I, 68)

The overall point is basically that structure = change in energy.

Individuation and System States

1.2.2.1 Individuation and Crystalline Allotropic Forms; Being and Relation

So some systems can transform their potential energy into a structure under some conditions.  This phase transition behavior is a necessary condition for individuation.  But are the energetic and material conditions alone sufficient to account for the individual produced?  Simondon thinks not, and he illustrates this conclusion by considering another situation where an amorphous matter can crystallize in either of two possible configurations (allotropy is probably best known as what distinguishes graphite from diamond despite the fact that they are both just carbon lattices).  Even at the same temperature and pressure, supercooled liquid sulfur can crystallize as either prismatic or octahedral crystals (or, it turns out, in a bewildering array of other forms).  Which one (if either) of these forms appears depends on whether a seed or germ of that type is present in the material as it cools.  Even super cold water won't freeze without a crystal nucleus to get it started.  Again, I'll skip the technical details because the important point is just that the process of individuation requires a singularity to get started in addition to some appropriate energetic conditions.  Which means that individuation isn't simple determined by the instantaneous conditions of the system, but depends on its entire history of development.

A stable individuality is thus formed when two conditions are met: a certain structure must correspond to a certain energetic state of the system. But this structure is not directly produced by the energetic state alone, for it is distinct from the latter; the initiation of structuration is critical; most often in crystallization germs are deposited from the exterior. Thus, there is a historical aspect to the manifestation of a structure in a substance, insofar as the structural germ must appear. Pure energetic determinism does not suffice for a substance to attain its state of stability. The beginning of structuring individuation is an event for the system in a metastable state. Thus, in general, even in the simplest process of individuation, a relation takes place between the body under consideration and the temporal existence of beings external to it that intervene as the evental conditions of its structuration. The constituted individual holds within it the synthesis of energetic and material conditions and of an informational condition, which is generally not immanent. (I, 70)

From this example we can also start to better understand how Simondon's notion of individuation doesn't really have anything to do with the individual as we normally use this term.  Individuation is always relative; the present individual is not the final individual, unless it happens to represent the lowest possible potential energy state of the system.  The same process of individuation can result in one crystal state or another, or first one then another (or vice versa).  Different individuals can result from the same individuation.  However, describing it this way presumes that we know what individuals we're talking about apart from the process of individuation.  Since this is precisely the thinking that Simondon wants to avoid, it might be more accurate to say that the individual is always the cutting edge of individuation.  The particular individual we find is always the most stable possible state of the system, given all the conditions and history.  It's what a system can do with the right energy and singularity.  When these conditions change, the individual will transform into another individual, making it a sort of 'unit' of becoming.  If, on the contrary, the individual persists, this is because that persistence represents the most stable state achievable by the system under the conditions -- conditions which necessarily include the presence and effect of the 'prior' individual, which acts as a singularity for the one following it.  We might say that the individual is constantly transforming, though sometimes into 'itself'. 

In terms of time, the individual is not in the past but in the present, for it only continues to conserve its individuality to the extent that this constitutive combination of conditions persists in and is extended by the individual itself. The individual exists such that the mixture of matter and energy that constitutes it is in the present. This is what could be called the active consistency of the individual. This is why every individual can be a condition of becoming: a stable crystal can be the germ for a metastable substance in a state of crystalline or liquid supercooling. (I, 74)

We have to use the infamous 'scare quotes' around that 'itself' because because this line of thinking opens up tricky but fascinating questions of identity.  Simondon only briefly addresses the topic in this chapter, but I suspect he will return to it in greater detail at some point (perhaps when he talks about quantum mechanics in 1.3.3.2).  The problem is that our crystal metaphor contains an ambiguity.  On the one hand, we have used the term individual almost synonymously with the precise state of the system constituted by given energetic and singular conditions.  We mean this particular crystal here, with a particular size and shape and set of defects, surrounded by a particular amount of amorphous material, and so on.  On the other hand, we sometimes seemed to mean the particular type of crystal (prismatic or octahedral) on the presumption that all crystals of that type are effectively the same individual, albeit perhaps of different sizes, which can be distinguished from other crystals and from the amorphous medium.  So are we talking about a point in phase space, or a whole region?  This ambiguity seems to be coupled to another ambiguity in the text or translation.  Sometimes the phrase "particular being" appears where we would expect "individual" to be.

... it is possible based on the discontinuities of conditions to define types that correspond to domains of stability or metastability; then, within these types, it is possible to define particular beings that differ from one another based on that which (within the limits of the type) is capable of a finer, sometimes continuous variation, like the speed of cooling. (I, 72)

Perhaps "mu" is the best response to this ambiguity.  The individual is neither a point nor a region in phase space, but something more like the possibility of movement through this space.  Unfortunately, I don't understand that answer well enough to avoid getting hit with a stick.  What seems clear to me at this point is that Simondon is trying to explain the origin of macroscopic individuals like a crystal we can see.  At this level of scale, the particular being we find acts like a point within a region marked off as, for example, "ETAT CRISTALLIN" on the phase space diagram on pg. 62.  That's because, as his example of volcanic rocks (pg. 72) better indicates, this region is never pure.  No macroscopic crystal consists in a perfect lattice of a single type (pg. 73).  However, it's important to note that this does not mean that the real individuals are the pure types of crystal that any mixed macroscopic crystal consists of.  From a macroscopic perspective, those pure type are merely abstractions.  All we have in front of us is a real crystal in all its impure complexity, whose detailed structure reflects the entire history of its genesis.  The trajectory of that history tells us about the type and the particularity of the crystal at the same time.  The fact that it passed into a certain region of phase space through a discontinuous structural transformation (ie. crystallization) determined its type, while the particular point in that region tells us about the finer details of the structure that can vary continuously from crystal to crystal.  So type and particularity aren't different level of ontology, they're just names for discontinuous and continuous properties of a real individual.

There are types because these conditions vary discontinuously by delimiting domains of stability; but because within these domains of stability certain parameters, which are part of the conditions, vary more finely, each particular being is different from a certain number of others. The original particularity of a being is not different in nature from its typological reality. The particular being does not possess its most singular characteristics any more so than its typological characteristics. Both the former and the latter are individual because they result from the encounter of energetic conditions and singularities, the latter of which are historical and local. (I, 73)

Those final two lines articulate a very subtle point that might seem somewhat academic in the case of a macroscopic crystal.  I would actually have been tempted to state it almost the opposite way Simondon did.  It's not that a type possesses a certain range of possible variations in characteristics which then possess individuals that instantiate those particular variables.  This is how we imagine the model of genus and species working -- like a top down tree.  Instead, its that an individual possesses a type and a particularity because of how it was formed. The real given is the individual, or rather the process of individuation which at any moment results in a more or less stable individual.  The problem is not that type and particularity don't mean anything real or are merely human imposed abstractions.  The trajectory of individuation has real continuities and discontinuities.  But an individual isn't built from these types and their variation or from a synthesis of types.  The individual is simply the unique trajectory.

As I say, this distinction is subtle, and doesn't become really interesting until Simondon considers the possibility that certain systems at a microscopic level might actually possess regions of phase space that have only a finite number of points.  If we imagine that only a single value is occasionally possible, then the region would reduce to a point, and an individual passing into it what be a real and direct incarnation of a type.  In that case, every process of individuation that exhibits that discontinuous structural change would end up producing an identical individual. 

If, within the interior of the same domain of stability, conditions that are still variable are not capable of an infinity of values but merely a finite number of values, it will have to be acknowledged that the number of effectively different beings able to appear is finite. In a certain quantity of substance, there could then be several identical beings that seem indiscernible.(I, 73)

This sort of situation would represent a quantized or atomized reality.  All the hydrogen atoms are equivalent.  And Simondon considers this a perfectly real variety of individuation that often happens in nature at very small scales.  His point is just that the reality here is still the process of individuation that combines energetic and singular historical conditions.  The reality is still the individual hydrogen atom produced in just this way at just this time and place by quantum vacuum fluctuations or whatever the physicist are into these days.  But this individuality is degenerate in the mathematical sense because it reduces to a type in this particular system.  So the apparently simple individuals that atomism takes for granted are actually complex special cases of a general process of individuation.

To seek the principle of individuation in matter, form, or force is to be condemned to only explaining individuation in these seemingly simple particulars, like, for example, that of the molecule or the atom. Instead of constituting the individual's genesis, this would be to suppose this genesis as already formed in the formal, material, or energetic elements and, due to these elements already harboring individuation, to generate through composition an individuation that is in fact simpler. (I, 74)

Now we can appreciate the reason Simondon began his discussion of physical individuation by talking about allotropic crystals and not atoms.  It doesn't make sense to start a discussion of spheres with a study of the point.  Of course he believes that atoms exist and that all physical matter is a combination of atoms.  But this type of explanation doesn't get at how the individuals are made, regardless of whether its the individual matter or the individual atom that interests us.  By considering an allotropic crystal, we bypass a host of possible objections.  It's the same stuff, made of the same atoms in the same conditions that nevertheless can sometimes crystallize in two possible forms.  We can't explain the individual crystal in a reductive bottom up manner as just determined by a combination of atoms and their forces specified in some equations. Instead we have to talk about how the energetic and material conditions interact with a seed singularity in a historical process. 

Finally, Simondon rounds out this section with some light epistemology.  Though several pages were taken out of the later edition, they contain such a fascinating swipe at Kant's theory of the a priori that I can resist reflecting on them.  The basic idea is straightforward, but has some pretty startling consequences.  We've seen that for Simondon the being of the individual is fundamentally relation.  It's not a term but a process that establishes a relation between 'terms' of different orders of magnitude.  So things are relations.  But our knowledge of things is also a relation, in this case a relation between ourselves and the thing.  Which is to say that knowledge is a relation of relations.  If we assume that this second order relation is still a true relation -- an individual that connects two disparate orders of reality -- then we've discovered that our knowledge itself is actually the same sort of reality as what we usually call a thing.  In short, thoughts are things because things aren't like we thought. 

Now that we've removed the categorical gulf between thoughts and things, Kant's prohibition on our knowledge of the thing-in-itself falls apart.  The noumenon isn't a substance separated from all other substances, and in particular from a substantial self.  It's a relation.  And we have direct experience of what relation is like, both through our experience of our self and of the world around us.  In fact, it's doubtful we are or can experience anything other than relation.  So it turns out that we can know the thing-in-itself, so long as we understand that the thing is precisely not in-itself, and neither is our knowledge of it.  The a priori forms of our sensibility are not distinct from the a priori conditions of possibility of anything else, and this condition is simply relation -- the linking of asymmetrical sides of an energetic system across a singularity.  And as we'll discuss a bit more in the next section, the simplest analogy for relation is Time, whose form is pure asymmetric succession.

If noumena are indeed not pure substance but also consist of relations (like exchanges of energy or passages of structures from one domain of reality to another domain of reality), and if relation has the same status of reality as the terms themselves, as we have tried to show in the preceding examples—insofar as relation is not an accident relative to a substance but a constitutive, energetic and structural condition that is extended in the existence of constituted beings—then the a priori forms of sensibility that allow us to grasp relations because they are a power of organizing according to succession or according to simultaneity do not create an irremediable relativity of knowledge. If relation effectively has the value of truth, then both the relation within the subject and the relation between the subject and the object can have the value of reality. (I, 75)

I think this recasting of the problem of knowledge is profound because it immediately steers us away from the problem of how to judge the truth of a pure reflective knowledge.  There is no such thing.  All knowledge is as active as any other thing in the world.  In fact, the world contains only actions, only verbs, that create passages from one state to another.  And as a part of the world, our knowledge is no different.  There's nothing, including our thought, that sits above or outside the world and has no impact on it.  The difference between true knowledge and error then lies not in its representational accuracy, but in whether the relation it creates or expresses is stable or merely metastable and subject to change as circumstances change.  As Kuhn demonstrated long ago, our knowledge can undergo paradigm shifts that look just like phase transitions.  But this doesn't 'disprove' or 'invalidate' the old knowledge any more than the transition to a new allotrope 'disproves' the prior crystal structure.  What's happened is simply that a previously stable equilibrium has become unstable due to some change in social, psychological, or physical conditions, ie. some change in the energetic and singular conditions that individuate knowledge. 

1.2.2.2 Individuation as the Genesis of Crystalline Forms Starting from an Amorphous Mass

Let me recapitulate the trajectory of the argument to set up Simondon's final step in this chapter.  The question is how to make an individual.  Chapter 1 showed us that you cannot make an individual from abstract matter and form, but that these two have to be brought together as such by some system that encompasses both.  Chapter 2 begins by examining what Simondon has started to call the "hylomorphic situation" in which an asymmetric non-equilibrium system contains potential energy.  This is the first positive requirement for individuation -- the existence of a potential energy of structural transformation.  These energetic conditions alone are not sufficient however.  The particular type of sulfur crystal which precipitates depends on the presence of a particular historical singularity in the form of a seed of one type or another.  This is the second positive requirement for individual -- the presence of a singularity that crystallizes the system by transforming potential energy into actual structure.  However, it turns out that these two conditions are not the complete story of individuation.  The third positive requirement for individuation is that the singularity present be capable of catalyzing this particular system's transformation.  In other words, there has to be some sort of match or fit or resonance between the singularity and the energetic conditions.  They have to belong to one another.

This third condition might seem kinda obvious given the examples Simondon has used so far.  Of course supercooled sulfur will form a prismatic sulfur crystal in the presence of a prismatic sulfur seed.  But where did this initial seed come from?  What triggers the first step where we just begin to pass from amorphous mass to crystalline solid?  There has to be some sort of singularity that is expressly not a tiny crystal.  In fact, this singularity may have nothing to do with sulfur at all.  Consider again the instructions for supercooling water.  The flask has to be cleaned with acid and free of scratches because all kinds of impurity can serve as seeds for ice formation.  All kinds, but of course, not any kind of impurity or asymmetry (for example, we can use a cylinder, not a sphere for the experiment, despite the former's lower degree of symmetry).  The particular singularity has to correspond to the particular energetic system in question.  This is why Simondon again takes up the examination of crystals, and specifically considers how we move from the pre-individual amorphous state to a state where there is an individual crystal seed.

What he uncovers is a fairly subtle point that is easy to overlook on the first reading.  Because the real question is not so much how the seed itself forms as how some symmetry breaking cascades into further symmetry breaking.  We said that all kinds of things can serve as seeds; there are always asymmetries in a system if we go looking for them.  But we never know whether these asymmetries will serve as singularities by looking at them in-themselves.  We can only discover in retrospect, as it were, that this asymmetry constituted a singularity for this system.  We only find out that this was the initial crystal seed by watching how it amplifies itself, by how it propagates itself or spreads through the milieu.  The singularity 'in-itself' is truly a sort of nothing.  But under the right conditions, that singularity can constitute a crystal seed whose advancing front is defined by a series of points that each serve as the same type of singularity with respect to the uncrystallized medium.  The crystal propagates at its edge, or better yet, the crystal as an individual simply is this mobile edge that serves as interface between two different states of the system.  So the individual isn't really a spatial phenomenon contained in the initial seed which just happens to grow larger.  It would be more appropriate to say that is the the act of growth itself, or the capacity of the system to crystallize under certain energetic and singular conditions.  In other words, the individual is resonance, relation, not substance. 

Thus, a third condition is manifested that we have not been able to note in the preceding case because it was necessarily fulfilled, since the structural germ and the metastable substance were of the same chemical nature. Here it is no longer a question of the scalar quantity of potential energy nor of the pure vectoral properties of the structure carried by the germ, but a ques- tion of a third type of rapport (which can be called analogical) between the latent structures of the still amorphous substance and the germ's actual structure. This condition is required for there to be a veritable amplifying relation between this structure of the germ and this potential energy carried by an amorphous substance. (I, 81)

This relation is information.  It's what allows a tiny crystal seed to grow into a macroscopic crystal.  The energy for this transformation doesn't come from the internal power of the seed, but from the way that its tiny energy resonates with the larger energy initially external to it, thus amplifying the source.  It's the same type of energy my words have when they cause you to fetch me a beer.  I'm not driving the car, handling the money, and getting the beer all with my own energy.  I'm using the tiny energy of my speaking to exploit yours.  And just like the definition of information as surprisal, the individual relation begins with an act of symmetry breaking, or what Simondon calls polarization.  The individual properly speaking is the ability of this symmetry breaking polarization to propagate, that is, for an initial difference to make a difference, at Bateson put it.

... the properties of a crystalline individual express and actualize the polarity or bundle of polarities that have presided over its genesis by pro- longing this polarity. A crystal, which is a structured matter, can become a structuring being; it is both the consequence and the cause of this polarization of matter, without which it would not exist. (I, 84)

I think this idea of resonance really helps us understand Simondon's conception of the individual as process or relation in a deeper way.  The individual crystal is not in the seed nor the amorphous milieu. It doesn't have defined physical 'properties' of 'characteristics'.  It is nothing but its ability to transform one set of substantial properties into another set, a transformation that always happens at the limit of these two phases. And this ability is the same thing as a potential amplification of an initial singularity.

The genetic properties of a crystal are prominently manifested on its surface; these are the limit's properties. Thus, if we want to be rigorous we cannot say the "properties of the crystal"; they are instead modalities of the relation between the crystal and the amorphous body. It is because the crystal is perpetually unfinished, in a maintained state of suspended genesis, that it possesses what can be uniquely called "properties"; these properties are in fact the ongoing disequilibrium manifested by the relations with the polarized fields or by the creation (at the limit of the crystal and around it) of a field that has a polarity determined by the crystal's structure. (I, 84)

At this point, as I alluded to earlier, Simondon makes his way back to the issue of time.  Time is the perfect model of symmetry breaking.  For this reason, we might say that the real limit that defines the crystal is the present.  The present is where the structured past has a chance to crystallize the amorphous future. In this sense the individual is where being undergoes a phase change, where the actual crystallizes from the potential in a becoming, "... becoming is not opposed to being; it is the constitutive relation of being qua individual" (I, 85).  But this means that the individual is neither potential nor actual, neither possible form nor actual matter, but exists only at the moment of transition, and in the way this moment, because of the resonance it creates, gives rise to different moments that are nevertheless the same, in a process of amplification.  Thus the real individual doesn't lie in the hunk of structured sulfur that results from a process of crystallization, but is instead the whole family of moments where a point on the edge of the crystal served as seed for the amorphous milieu surrounding it.  The macroscopic crystal is not so much the statistical average of these real individual moments as a depiction of their family tree.

It can certainly be said in a derivative sense that a certain amount of sulfur is individualized by the fact that it is presented in a determined allotropic form. But this determined state of the overall ensemble does nothing but express on the macroscopic level the underlying and most fundamental reality of existence in the mass of real individuals that have a community of origin. The individualized characteristic of the ensemble is merely the statistical expression of the existence of a certain number of real individuals. If an ensemble envelops many physical individuals from various origins and different structures, it is a mixture and remains poorly individualized. The veritable support of physical individuality is effectively the operation of elementary individuation, even if it only appears indirectly at the level of observation. (I, 85)

1.2.2.3 Epistemological Consequences: Reality of Relation and the Notion of Substance

Simdondon concludes this chapter with some reflections on the difference between the crystalline and the hylomorphic schemas of individualization.  The most obvious one is that their definitions of the individual don't share the same understanding of interiority and exteriority.  The substantial individual we associate with the hylomorphic model has a clear spatial inside and outside. It is a finite being whose limit separates self from not-self.  By contrast, for the crystalline schema, the individual exists only in its constantly displaced limit, which is no longer what separates it from all other things, but what produces it as a connection or transition between system states. 

The finite being is the exact contrary of the limited being, for the finite being is self-limiting, since it does not possess a sufficient quantity of being to grow endlessly; on the contrary, in this indefinite being that the individual is, the dynamism of growing does not stop, since the successive stages of growing are like a number of relays due to which increasingly large quantities of potential energy are captured in order to organize and incorporate increasingly considerable amounts of amorphous matter. (I, 88)

Though it sounds odd to put it this way, because the crystalline individual is a being of the limit, it is by rights unlimited, infinite.  The amplifying process of individuation gets as big as it can get, and the individual it produces thus has no inherent largest limit.  In a sense, since the essence of the crystal is the feedback process by which the seed resonates with and converts the amorphous milieu into more crystal, the individual is eternal and self-producing.  These words are easy to misinterpret though, so Simondon simply points out that the spatial growth of the crystal is always indefinite.  And he gives a couple of interesting examples of how this same logic applies to the crystal shrinking through something like corrosion.  Both the arising and the ceasing of the individual occur at the limit, in the eternal present.

So the crystal has no largest scale.  But does it have a smallest scale?  Indeed it does.  Since it is produced by a transformation of an amorphous milieu into a regular lattice, the crystallization can only happen according to the spacing of the lattice.  Or as Simondon puts it, the crystal milieu is periodic.  The macroscopic edge of the crystal can be indefinitely large, but if we look carefully, this edge is composed of many microscopically ordered sites of crystallization with predictable gaps between them.  Even though it can multiply itself through the crystal's growth, the smallest limit of the crystal can't be just anywhere.  In fact though, this regular microscopic limit of the crystal is part of what allows it to grow into an indefinitely large macroscopic crystal that remains self-similar.  The momentary sites of growth that are properly speaking the 'limited being' of the crystal are spaced periodically, which is what allows the macroscopic crystal to grow as large as it can without changing structure.

In fact, the shared source of the limit and the structuration is the milieu's periodicity. Here, with a more rational content, we rediscover the already indicated notion of the indefinite possibility of growth; the crystal can grow while conserving all its characteristics because it possesses a periodic structure; the growth is therefore always identical to itself; a crystal has no center that allows us to measure the distance of one point of its exterior contour with respect to its center; relative to the crystal's structure, its limit is no more distant from the center than the other points; the crystal's limit is in virtually every point, and it can really appear in each point through a cleavage. The words interiority and exteriority cannot be applied with their usual meaning to this reality that the crystal is. (I, 91)

The crystal as a structure isn't possible without the discontinuity of the periodic lattice at the microscopic level.  If the molecular level were purely isotropic, with no polarization or symmetry breaking, and every point were equivalent to every other, the molar crystal could not form.  In this sense, the individual always requires some small scale discontinuity or singularity.  However, it also requires the energetic milieu that surrounds this discontinuity, and which it bathes in as a continuous potential energy that this singularity can draw from in crystallizing a structure.  So we return to a point Simondon made in his introduction.  The individual is an intermediate level of reality that links two disparate scales, two orders of magnitude. It's a relation that can create a substantial system with multiple parts.  

This supposes that individuation exists on an intermediate level between the order of magnitude of the particulate elements and that of the molar ensemble of the complete system; on this intermediate level, individuation is an operation of amplifying structuration that makes the active properties of initially microphysical discontinuity pass to the macrophysical level; individuation is initiated on the level at which the discontinuous of the singular molecule is capable (in a milieu in a "hylomorphic situation" of metastability) of modulating an energy whose support is already a part of the continuum in the population of randomly arranged molecules, i.e. in a superior order of magnitude relative to the molar system. (I, 94)