Incomplete Nature Part 12

What kind of work can be done with respect to this morphodynamic orthograde tendency? Notice that stirring with a paddle in parallel with this rotation will not disrupt it. But stirring in any other pattern, or merely impeding this pattern of flow, will tend to disrupt it. These disturbing patterns of interaction are in this way contragrade to the orthograde tendency of the system to regularize. The non-parallel patterns of interaction are doing morphodynamic work against this orthograde tendency, whereas the parallel pattern is not. Notice also that any pattern that results in morphodynamic work also involves thermodynamic work, whereas the parallel pattern does not. But this is relative to the orthograde tendency that is intrinsic to the system in question. Thus, if the flow tends to be chaotic because the geometric organization of the stream pattern with respect to the flow rates are not conducive to vortex formation, stirring in the appropriate direction can aid vortex formation, and decrease turbulence. This would also be morphodynamic work, since vortex formation was not the intrinsic orthograde tendency.

Consequently, the introduction of morphodynamic work also requires thermodynamic work. And notice that the amount of mechanical/thermodynamic work involved is strongly dependent on the form of the paddle-induced disturbance with respect to the form of the flow. This suggests a general rule: in order to perturb a dissipative self-organized dynamical form away from its spontaneous attractor tendency, a conflicting form must be introduced, and the combined amount of thermodynamic and morphodynamic work involved will be a function of the number of dimensions of asymmetry that are reversed in the process, adjusted according to their relative magnitudes in the two alternative dynamical processes. In this way, the parameters defining the higher-level morphodynamic work play a significant role in determining the correlated amount of thermodynamic work that is required.

Recognizing that there are both parallels and asymmetric dependency relations.h.i.+ps involved, we need to be clear about the a.n.a.logies and disa.n.a.logies between morphodynamic and thermodynamic work. For example, we might be tempted to describe the regular dynamics of simple self-organized processes as morphodynamic equilibria, on the a.n.a.logy of thermodynamic equilibria. Such an a.n.a.logy is complicated by the fact that many morphodynamic processes remain partially chaotic (like the stream example above)-describable only in terms of constraints, not geometric regularities-and even simple morphodynamic systems may have more than one quasi-stable dynamical attractor. This complicated attractor logic, which has become the hallmark of complexity studies, also complicates the a.n.a.lysis of morphodynamic work.

Because of the potential for explosive symmetry breaking in morphodynamic systems, describing the way that the interaction between morphodynamic processes can transform their orthograde dynamics into contragrade change in the morphodynamic domain (i.e., the description of a morphodynamic engine, if you will) is far more difficult than for thermodynamic systems. This is in part due to the hierarchic complexity of morphodynamic processes. It takes thermodynamic work to drive morphodynamic attractors, so morphodynamic interactions cannot undermine this thermodynamic base and still do morphodynamic work. Precisely organizing a mediating mechanism that is able to take advantage of the interactions between different morphodynamic orthograde attractors is thus limited by the need to align both thermodynamic and morphodynamic processes. Moreover, since morphodynamic attractors are not merely defined by quant.i.tative parameters (e.g., energy gradients) but also by formal symmetry properties, the possibilities for contragrade alignment of different morphodynamic processes are very much more restricted.

FIGURE 11.3: A diagrammatic depiction of the thermodynamic work performed by an organism to maintain its integrity with respect to thermodynamic degradation, and to support its higher-order orthograde (teleodynamic) capacity to replicate the constraints that support this process. Organisms must extract resources from their environment, e.g., by doing work (a) to constrain some energy gradient in order to access free energy to maintain their metabolisms (which maintain a persistently far-from-equilibrium state). Because the environment is often variable, they must also obtain information (i) about this variability in order to use it as a source of constraints (c) to regulate the work they perform. Constraints are depicted as right triangles deviating energy flows (arrows), and the constraints inherited genetically (g) are depicted as both within and outside the organism (since they are inherited from a parent organism).

Despite these conceptual difficulties, we have already described one special case of morphodynamic work: a simple molecular autogenic system. It is a special case, because of its precise recursive synergistic organization. But before reconsidering this special case in terms of the work involved, we need to examine some more generic examples in order to gain a general conception of what gets transformed during morphodynamic work.

Consider, for example, a resonating chamber such as a flute or pennywhistle that is continually supplied with energy by air pa.s.sed across an aperture. With this steady turbulent flow at one end, the air along the length of the resonating tube settles into a stable vibratory pattern, heard as a continuous tone. Changes in the effective length of the chamber, produced by opening or closing holes at various positions along the length, change which patterns of vibration (tones) are stable, even if the flow of blown air remains the same. Resonant vibrations are remarkably robust in linear chambers like the tube of a flute, but in irregular-shaped chambers they become less reliable and more sensitive to changes in the energy supplied. Even in a musical instrument like a flute, fluctuation of input energy can disrupt convergence to a stable vibration. A common experience for a novice flutist is to blow too hard, too soft, or at the wrong angle across the mouthpiece, with the result that the sound warbles between alternate tones, interspersed with the hissing sound of chaotic air flow. This demonstrates that the morphodynamic regularity exhibited by the tone being produced is sensitive to the rate and the form of the perturbing energy being introduced.

Once stable vibration is established, however, the sound of the flute can induce other objects (e.g., a winegla.s.s) to vibrate sympathetically, to the extent that they too are capable of regular vibration in the frequency range of the flute tone. Interestingly, if this sympathetic resonator has its own distinctive resonant frequency, an unstable interaction may result, with the consequence that it vibrates in a pattern that is different from but typically attracted to some regular multiple of the frequency of the flute tone: a harmonic. Not only constant energy but constantly dissonant vibrations must be provided to induce the gla.s.s to a.s.sume a vibration pattern that is different from its most robust spontaneous resonant frequency. Thus there are two levels of work that are necessary to maintain this non-spontaneous regularity: (1) thermodynamic work, which is responsible for the energy necessary to induce the sympathetic resonator to vibrate; and (2) morphodynamic work, which is necessary to cause the sympathetic resonator to vibrate at something other that its spontaneous frequency.

The role of morphodynamic work is demonstrated by the fact that even with the same energy, different driving frequencies will have very different capacities to push the resonant response away from its spontaneous frequency. In other words, the differential in regularity between the vibratory patterns of the two resonating objects results in a pattern in the one that would not occur if this specific pattern of excitation were different or random. Moreover, if these two differently resonant objects are rigidly connected, the effect can be bidirectional: the total system will likely a.s.sume a vibrational state that is a complex superposition of the two resonant patterns (in proportion to other relative properties, such as shape, relative ma.s.s, and vibrational rigidity) as each structure becomes a source of morphodynamic work affecting the other.

Let's be more specific about this subtle distinction between the two levels of work in this example. Separately, each of the resonating structures tends to converge to a different, relatively stable, global vibrational state when mechanical energy is introduced and allowed to dissipate. Resonance is a morphodynamic attractor: the resultant stable form of an orthograde tendency. It is produced because of the geometry of the resonating chamber, the vibration-propagating characteristics of the material, and the level and stability of the input energy. These are the boundary conditions responsible for the morphodynamic attractor tendency. For differently resonant bodies, the boundary conditions are different, and will determine different orthograde tendencies. The thermodynamic work-blowing-that induces vibration is potentially able to produce an unlimited number of vibratory patterns. What actually gets produced is dependent on the specific boundary conditions imposed by the flute. Although no vibration will occur without the introduction of a stable airflow to contribute the energy of vibration (constant thermodynamic work), the properties of the flute will constrain the domain of the possible spontaneous (orthograde), stable vibrational patterns. And this will be robust to modest changes in the flow of air, so that a range of input energies will converge to a single resonant frequency.

This many-to-one mapping of thermodynamic work to morphodynamic work is a characteristic feature of this dependency relations.h.i.+p. But it is not simply a many-to-one relations.h.i.+p; it is the mapping of a continuum to discrete states. We will return to this feature later, because it turns out to be a critical contributor to the discontinuity of emergent effects as we move up the hierarchy from thermodynamic to morphodynamic to teleodynamic processes.

The morphodynamic work produced by linking oscillators results from one set of boundary conditions affecting another. Specifically, their differences in geometry, ma.s.s, and the way they conduct vibratory energy all contribute to the total work of this transfer of form. The thermodynamic work component is roughly the same irrespective of whether the coupled oscillators reinforce each other's vibrations or rapidly damp all regular vibrations, transferring most of the energy into the irregular micro vibrations of heat. To the extent that their resonant features interact to produce a s.h.i.+ft in global regularities compared to the uncoupled condition, morphodynamic work is also involved, and can be judged more or less efficient on the basis of this transfer of regular global dynamics. But whereas thermodynamic coupling yields a combined system that dedifferentiates toward a state of global equilibrium-determined by the mean boundary condition of the total-morphodynamic coupling does not. There is nothing quite a.n.a.logous to a "mean" value, because of the relative discreteness of the morphodynamic attractors involved. The coupling of boundary conditions must be such that each reinforces the other in some respect in producing a third discrete orthograde tendency: one that is both amplified and amplifies each of the other two. For two oscillators, this can be a simple common multiple of the two resonant frequencies; but with additional couplings, the probability of simple and discrete dynamics quickly diminishes, and dynamical chaos results. And beyond the domain of simple oscillators this is far more likely to be the case.

This means that the ability to perform morphodynamic work can be quite easily disrupted. In coupled physical resonators, for example, if one structure is more regular in shape and form, and therefore more effective at form amplification, it will tend to drive the vibratory activity of the coupled system, though this will be resisted and constrained by the vibratory regularities or irregularities of the second structure to which it is linked. In this case, we can say that morphodynamic work is continually bringing the less resonant system to a non-spontaneous semi-regular vibratory state. In the case where both have different but nearly equally efficient resonant tendencies, the resulting vibratory state of the coupled system may converge to a pattern that combines the two, amplifying common harmonics and producing complex waveforms, or may never resolve a chaotic state, because of the incompatibility of their orthograde tendencies.

In such cases of competing resonant tendencies, we can discern another parallel with thermodynamic work: some systems are more difficult to perturb than others. In other words, the orthodynamic tendencies of different systems may be of different "strengths." Just as objects with greater momentum or inertia and thermodynamic systems with greater total specific energy require more mechanical work to produce equal changes of motion compared to less ma.s.sive or extensive systems, morphodynamic systems can differ greatly in the relative strength of their attractor dynamics. There are two potential contributors to this morphodynamic "inertia." First, one system may simply be more susceptible to thermodynamic work because it is physically smaller, less ma.s.sive, or better at conducting energy. This follows from the simple fact that morphodynamic work is entirely dependent on thermodynamic work. But second, one system may be more regular, such as the shaped body of a resonant musical instrument, or it may be more easily regularized, as is the minimally constrained flow of fluid in a Benard convection cell or vortex. This combination of factors will determine both the potential to do morphodynamic work and the tendency to resist morphodynamic change.

The potential to perform either thermodynamic or morphodynamic work is proportional to the divergence from an attractor maximum. But this can be a problem for the capacity to do morphodynamic work because systems with complex attractors tend not to exhibit consistent extended spontaneous change in any single direction. This means that only systems with highly reliable and relatively simple attractor dynamics are able to contribute any significant amount of morphodynamic work. This makes it difficult to find spontaneous examples of morphodynamic work, and makes it very rare for highly complex morphodynamic transformations to occur without highly sophisticated forms of human intervention. So, although simple examples such as the coupled resonators described above represent exceptions in nature, not the rule, simplicity is an advantage when it comes to making use of morphodynamic work. This is not, however, an absolute impediment, since elaborate webs of morphodynamic work are found in the metabolic networks of living organisms.

What about more complex morphodynamic processes that produce regularities with more dimensions of regularity? Consider these somewhat fanciful Rube Goldberg uses for Benard cell formation. The regular hexagonal tessellation of the surface of the water could, for example, be utilized to sort small floating objects into discrete collections of similar numbers, each collection sitting within the tiny hexagonal bowl of a Benard cell. Or the concave shape of these regular surface depressions could be used to focus incident light to dozens of individual points just above the surface. In these cases, there is very little thermodynamic work linking the two interacting substrates (especially in the case of reflected light), but the morphodynamic work occurs as the spontaneous regularization of fluid convection similarly regularizes something else that otherwise would never a.s.sume this configuration. Thus, via morphodynamic work two otherwise independent thermodynamic systems accomplis.h.i.+ng thermodynamic work can be coupled.

More practical examples of morphodynamic work include the use of specially shaped vessels or vibrating containers for sorting different shapes, weights, or sizes of particulate materials, such as pills or grains. Depending on the shape of the vessel, the way it is rotated or shaken to induce the contained objects to move with respect to one another, and the differences in object features (such as shape or weight), it is possible to automatically separate objects, transforming a well-mixed, uniformly distributed collection into a highly asymmetric distribution in which different types of objects occupy distinct positions relative to one another. Natural examples of this particulate sorting process occur with pebbles on ocean beaches and stones rising to the surface in soil as a result of periodic freezes and thaws; but other uses include ways of separating pills and minerals, as well as the cla.s.sic method of separating gold nuggets from sand and other pebbles, by "panning."

Morphodynamic work shares one very significant attribute with thermodynamic work: the law of diminis.h.i.+ng returns. As a consequence of the first and second laws of thermodynamics, and the constraints of doing work, perpetual motion machines are impossible. There are always some degrees of freedom of increasing entropy that cannot be fully constrained, and so the capacity to do work in one direction, and then reverse this organization and use that gradient to do work in the opposite direction, decreases with each step, making full reversal un.o.btainable. The potential to do iterated morphodynamic work also diminishes rapidly with increasing degrees of freedom and thus also with each interaction. There is something a.n.a.logous to nature's prohibition of perpetual motion machines when it comes to morphodynamic work as well. In fact, the efficiency problem is much worse, because of the discreteness issue. In most instances of coupled morphodynamic processes, the interactions between their distinct regularities result in complex dynamics that appear highly chaotic. Cla.s.sic examples of so-called deterministic chaos reflect the complexity that can result even as a result of coupling three otherwise quite simple morphodynamic processes into a larger system, as for example happens when different length pendulums are coupled with one another. Whereas the recursive dynamics in a simple self-organizing system amplify regular dynamic features, strongly coupled self-organizing processes can recursively amplify both concordant and non-concordant boundary conditions, producing complex and often extreme divergence and damping effects. This is especially true if thermodynamic energy is continually introduced, as in dissipative systems. This kind of coupling of organized dynamical processes is probably one of the factors contributing to the unpredictable and almost turbulent character of human social and economic systems; though, as we will see, this tendency to complexity becomes amplified to a far greater extend when we consider the superimposition of teleodynamic processes.

Given these limitations, and since morphodynamic regularities even with robust simple attractors only form under very limited boundary conditions, interactions between morphodynamic systems with different boundary conditions end up producing larger systems with complicated and irregular boundary conditions. So, for many reasons, morphodynamic work of any significant complexity and magnitude will tend to occur quite rarely under natural circ.u.mstances.

TELEODYNAMIC WORK.

There is, however, one cla.s.s of phenomena that presents glaring exceptions to this rarity: living processes. Indeed, self-organizing processes in living organisms and ecosystems defy the apparent problem of the chaos that should tend to result from coupling self-organizing processes to one another-and to an astounding degree-since even the simplest bacteria are composed of hundreds of strongly coupled cycles of chemical processes. Life appears to have cornered the market on morphodynamic work, and to have done so by taming the almost inevitable chaos that comes with morphodynamic interactions. Not only are living organisms themselves enormously complex webs of self-organizing processes, but they also tend to evolve to complement higher-order complex dynamical regularities made up of the large numbers of other organisms comprising their ecosystem, all embedded in semiregular patterns of climatic and resource change. So, it is within living processes that we must turn to find the greatest number and diversity of exemplars of morphodynamic work.

Besides energy and raw materials to maintain their far-from-equilibrium thermodynamics, living organisms also require incessant form production processes: production of specific molecular forms, specific patterns of chemical reactions, and specific structural elements. Morphodynamic work must be reliable and constant for life to be possible. This requires both thermodynamic and morphodynamic work cycles-engines of form production that are a.n.a.logous to human engines designed to perform thermodynamic work cycles. The process of biological evolution has not merely "discovered" and "remembered" how to set up a vast array of morphodynamic work processes; it has discovered complex synergies and reciprocities between them that enable repeatable cycling. We have encountered a simple example of this in the case of autogens, but to understand how the evolutionary process is able to mine the morphodynamic domain for these sorts of reciprocities and complementarities, we will first need to understand a yet-higher-order form of work: teleodynamic work.

Teleodynamic work can be defined a.n.a.logously to the prior levels of work we have described. It is the production of contragrade teleodynamic processes. Since this must be understood in terms of orthograde teleodynamic processes, the first step in describing this level of work is to define and identify examples of orthograde teleodynamics. In general terms, an orthograde teleodynamic process is an end-directed process, and more specifically, one that will tend to occur spontaneously. Although teleodynamic processes are incredibly complex, and an explanation of the structure of teleodynamic work is by far the most elaborate-since it is const.i.tuted by special relations.h.i.+ps between forms of morphodynamic work-it is also the most familiar. So it may be helpful to first consider the human side of teleodynamic work before delving into the underlying dynamical structure of this process.

Teleodynamic work is what we must engage in when trying to make sense of an unclear explanation, or trying to produce an explanation that is unambiguous. It is what must be produced to solve a puzzle, to persuade resistant listeners, or to conduct scientific investigations. It is also the sort of work that goes on in board meetings and in domestic arguments, and which leads to the design of machines and governments. And it characterizes what is difficult about creative thought processes. Although these examples could mostly be considered forms of mental work, they have a natural kins.h.i.+p with the simple process of communicating, and with biological adaptive processes as well. All share in common the work of generating new information and new functional relations.h.i.+ps, or of changing thought patterns or habits of communication and human intentional actions.

If you have read to this point, you have probably found some parts of the text quite difficult to follow. Perhaps you have even struggled without success to make sense of some claim or unclear description. But unless you are a very easily agitated reader, you will probably not have found yourself running out of breath or breaking a sweat because of the energy you have exerted to do this. While writing this chapter, I took a break to cut and split some wood for a fire. Doing so worked up a sweat. Though only a small fraction of writing time was devoted to this process, I no doubt expended far more energy chopping wood than in all my writing for the day. But which was the more total work? Obviously, the energy expended isn't the most useful means of a.s.sessment. Nevertheless, engaging in the effort of writing or reading does require metabolically generated energy, and the more difficult the task of creation or interpretation or the more stimulating or frustrating, the more thermodynamic work tends to be involved.

We have no trouble recognizing the capacity to do this kind of work. In an individual, we may describe it as intelligence. In a simpler organism, we may describe it in terms of its adaptability. This is the "power" that we recognize in great insights, influential ideologies, or highly developed a.n.a.lytical tools. It is the power to change minds and organize human groups. It can ultimately translate into "the power to move mountains," as the old adage implies, though its capacity to do this is necessarily quite indirectly implemented. It is commonly described as "the power of ideas."

FIGURE 11.4: Reading exemplifies the logic of teleodynamic work. A pa.s.sive source of cognitive constraints is potentially provided by the letterforms on a page. A literate person has structured his or her sensory and cognitive habits to use such letterforms to reorganize the neural activities const.i.tuting thinking. This enables us to do teleodynamic work to s.h.i.+ft mental tendencies away from those that are spontaneous (such as daydreaming) to those that are constrained by the text. Artist: Giovanni Battista Piazzetta (16821754).

So, engaging in teleodynamic work is one of the most familiar mental experiences of being a human agent. It is what characterizes what we describe as "intentionally willed action." It is naturally understood as work because of the resistances that it can overcome, and because it is required in order to modify otherwise spontaneous mental or communicative processes, such as an otherwise unquestioned belief or habitual process of reasoning. But as familiar as this experience is in everyday life, its relations.h.i.+p to physical work is often quite ambiguous, even though it is intuitively taken for granted. No one would confuse what must be done to raise a weight with what must be done to raise a question, but both involve effort, that experience of promoting contragrade dynamics. And while physicists and engineers can be incredibly precise at comparing the amount of work that is done to accelerate a car versus a baseball to 60 miles per hour, it seems almost a matter of idiosyncratic opinion how one should compare the work of solving a mathematical equation to that of solving a crossword puzzle clue. So we might be forgiven for thinking that the latter cases of mental work are merely metaphorical comparisons to physical work.

The production of teleodynamic work is, of course, totally dependent on the other forms of work we have been considering. Teleodynamic work depends on morphodynamic work depends on thermodynamic work. Mental work, for example, requires physiological processes doing thermodynamic work to maintain the constant activity of neurons producing and transducing signals, and also the production of spontaneous and non-spontaneously self-organized patterns of neural activity that recursively amplify and damp as they traverse complex changing neural networks. What is additionally involved, however, is the generation of new semiotic relations.h.i.+ps and new end-directed tendencies in the face of spontaneous habitual interpretive tendencies and countervailing end-directed processes. In the performance of difficult interpretive processes, for example, we directly experience both the energetic demands and the morphological demands of the supporting lower-level forms of work that are also involved. There is the fatigue that is generated by struggling to maintain attention on the interpretive problem until it is solved, and the challenge of trying to find an appropriate translation into other words or to conjure up an appropriate mental image among the many spontaneously arising but inadequate alternatives. But although this domain of work necessarily involves the others, it is intuitively not merely reducible to these simpler processes alone, as mere "mindless" physical labor demonstrates. Thus, the free a.s.sociation of daydreams or the nearly unconscious performance of a highly familiar task are not experienced as mentally effortful, even though they do require metabolic support and the generation of distinct appropriate patterns of neural activity. So we intuitively recognize the distinctive effortful character of teleodynamic work.

Although the exercise of mental effort is unquestionably its most familiar exemplar, teleodynamic work occurs in many forms that are neither cognitive nor even a.s.sociated with brains. Before we can hope to understand the processes responsible for the teleodynamic work that occurs in and between human brains, we will need to a.n.a.lyze its organizational details and its dependency on other forms of work in some far simpler systems. To do this, we can back down in complexity and familiarity, and at the same time incrementally unpack the complex contributions of other forms of work.

One of the more fundamental forms of teleodynamic work is that which occurs in the process of biological evolution. This is epitomized by novel functional adaptations and their heritable representations being constantly created where none previously existed. In other words, new teleodynamic systems and relations.h.i.+ps are generated from the interactions of prior teleodynamic systems with respect to their shared environmental contexts. Using natural selection as an exemplar, let's a.n.a.lyze what makes it teleodynamic work, using the generic logic that we have outlined above for thermodynamic and morphodynamic work.

To begin the a.n.a.lysis, we need to identify what const.i.tutes orthograde and contragrade processes in this teleodynamic domain. In the case of biological evolution, there are two very general cla.s.ses of orthograde teleodynamic processes. First, there are the actions of organisms that function to maintain them against degrading influences, such as thermodynamic breakdown of macromolecules and degradation of metabolic networks. Second, there are processes of growth, differentiation, and reproduction, which are involved in producing what amount to backup copies of the organism, in the form of daughter cells or offspring. These are teleodynamic processes because they are end-directed toward specific target states. They are orthograde because they are what organism dynamics produce naturally and spontaneously (given supportive underlying forms of morphodynamic and thermodynamic work). Organisms are of course highly complicated synergistic constellations of teleodynamic processes that each collectively contribute some fraction of the global teleodynamics of the organism. But for the purpose of this first level of a.n.a.lysis, we can lump all together as though the life history strategy of the organism is a single orthograde teleodynamic process. In this context, we can identify contragrade teleodynamic processes as those that are organized in such a way that they impede or contravene these orthograde processes. In other words, contragrade biological teleodynamic processes are those that are in some way bad for the organism, in that they are detrimental to survival and reproduction.

With respect to evolution, where the critical end is reproductive success that is sufficient to guarantee continuation of one's lineage, reproductive compet.i.tion from other members of the species is the most directly relevant source of precisely contragrade influences. In this sense, each organism is doing teleodynamic work against its reproductive compet.i.tors. So far, this is intuitively familiar. The work required to compete with other organisms at many levels, over many kinds of resources, in order to achieve diverse ends, is in many ways the hallmark experience of being a living organism. But like other forms of work, teleodynamic work can also be used to transform one form of teleodynamic process into another, and to generate emergent phenomena at a higher order. This is what happens in the process of natural selection.

So now let's describe natural selection in terms of teleodynamic work. In the standard model of natural selection, variants of the same teleodynamic process (adaptive traits) that are represented in the different members of a species in one generation are brought into compet.i.tion over resources critical to reproduction. In this compet.i.tion of each against all (in the simplest case), work done to acquire resources, mates, and so on is also work that degrades the teleodynamic efficacy of compet.i.tors with respect to these same requirements. This work is both directly and indirectly a source of distributed contragrade effects on other organisms. Because the teleodynamics of organisms is supported by extensive morphodynamic and thermodynamic work, it is also the case that teleodynamic compet.i.tion ramifies to all these lower-level processes as well. And since all ultimately depend on thermodynamic work, this is the final arbiter of teleodynamic success. a.n.a.logous to the way the coupling and juxtaposition of non-concordant orthograde thermodynamic and morphodynamic processes can be utilized to generate specific contragrade patterns of entropy decrease and form production, respectively, the complex juxtaposition of non-concordant teleodynamic processes can generate teleodynamic systems that would not otherwise occur. Because the widespread integration of diversely contragrade teleodynamic interactions in one generation is mediated through an environment that is also the source for the resources supporting their underlying morphodynamic and thermodynamic work processes, the constraints and regularities intrinsic to that environment become the a.n.a.logue to the constraints of an engine for channeling the teleodynamic work into forms consistent with that environment.

In biological evolution, this ultimately results in an increasing asymmetry of the presence of these teleodynamic traits in succeeding generations. The differential reproduction and elimination of the less fitted variants from the population in each generation thus has a recursive influence on all levels of work involved, maintaining them in concordance with those constraints. More significantly, due to the inevitable spontaneous (thermodynamic) degradation of the capacity to do work at all these levels, new variant teleodynamic complexes continually arise and are entered into this evolutionary work cycle. The familiar result is the production of increasingly integrated, increasingly diverse, increasingly complex, increasingly well fitted teleodynamic systems.

What we can conclude from this is that evolution is a kind of teleodynamic engine, powered by taking advantage of spontaneous thermodynamic, morphodynamic, and teleodynamic processes, and which continually generates new teleodynamic processes and relations.h.i.+ps. Additionally, because teleodynamic processes are supported by the synergistic organization of morphodynamic and thermodynamic work cycles, these too evolve, as do the synergies that bind them together into the individual causal loci we know as organisms.

EMERGENT CAUSAL POWERS.

This brings us back to issues discussed in the introductory section of this chapter. We can now make a somewhat more detailed a.s.sessment of why there is a difference in the amount of work required to solve a difficult puzzle versus a simple one. Returning to the comparison between two jigsaw puzzles-one with fewer parts, distinctively different-shaped pieces, and forming a recognizable and heterogeneously organized picture when a.s.sembled, compared to one with more pieces, much more subtle shape differences, and a very h.o.m.ogeneous surface image-we can recognize that much more work at all three levels will be required to a.s.semble the second, harder puzzle. More pieces require more physical movements, less shape distinction between pieces will mean more errors of fit and will also translate into more movements, as will less pictorial distinctions. Hence, more thermodynamic work is required.

But now let's consider the cognitive challenge. The puzzler must generate mental images of each comparison, and will almost certainly generate many times more mental comparisons than actual trial fits. Each mental comparison requires the generation of at least two distinctive forms of neural activity to represent the compared pieces. Since there are inevitably other cognitive and mnemonic tendencies that would otherwise express themselves and could be potential distractions, this is a non-spontaneous process of form generation. This is the source of the feeling of resistance to focusing attention on the problem at hand and generating these highly constrained mental images rather than others. This then is a form of morphodynamic work: contragrade form generation against a background of competing orthograde form generation tendencies (we will talk about how this might be generated in brains in chapter 17). But of course this requires metabolic energy (thermodynamic work), and presumably more energy than is required to let one's thoughts wander in orthograde stream of consciousness patterns. In fact, this is made unambiguously evident in in vivo imaging studies of regional changes of brain metabolism when comparing active versus pa.s.sive interaction with stimuli (this too will be discussed in more detail in chapter 17).

But these forms are not merely forms; they are mental representations of objects being encountered in the context of a.n.a.lyzing and physically a.s.sembling the puzzle. The forms as representations are not then merely the result of morphodynamic work, but must be generated with respect to this context and tested against it. As these mental images are being generated, biases from past memory and from expectations will also compete with the generation of representations that are faithful to the task at hand, so there will be innumerable ways that the generation of adequately correspondent representations can go awry. Almost certainly, in the generation of these mental images, there will be multiple partial "drafts" that are produced and compared against the information provided by the senses. In this process, all but one will be rejected as not sufficiently constrained to correlate with incoming patterns. And this mental generation and testing of imagined comparisons will itself be iterative in advance of actually picking up a piece and trying its fit, looking for physical feedback to be the final arbiter of accurate representation. Indeed, this process also must be proceeding at many other levels in parallel, such as the generation of the teleodynamic predisposition to continue working on the problem, despite distractions and competing demands on one's time.

This generation and selection process is an expression of teleodynamic work. And given that it requires the generation, comparison, and context-sensitive elimination of vastly many partially developed early "drafts" of the mental representations involved, each of these was a product of morphodynamic work, and so on. So the level of work that must take place is significantly amplified with the puzzle difficulty. In short, it takes more work at all levels. But notice that this is far more efficient than randomly picking up pieces and testing their physical fit. This s.h.i.+ft of work up-level, so to speak, significantly decreases the total amount of thermodynamic work to achieve this end. And thermodynamic work is what supports the base of the whole hierarchy.

As this one simple example ill.u.s.trates, the domain opened by teleodynamic work is enormous. The diversity of forms of teleodynamic work is as extensive as the fields of biology, psychology, human social behavior, and all the arts combined. Not only is it far beyond the scope of this book, much less this chapter, to do more than cherry-pick a few ill.u.s.trative examples of the process; a thorough a.n.a.lysis of teleodynamic work in such cases would likely be redundant and superficial compared to a.n.a.lyses pursued in these many specialized fields. Indeed, one might argue that throughout this chapter we have merely redescribed selected phenomena drawn from the domains of mechanics, acoustics, evolutionary theory, and now cognitive and behavioral science in terms of the concept of work. But if this were merely redescription and the renaming of otherwise well-understood phenomena, there would be little gain over knowledge gained in these already well explored territories. What I hope this a.n.a.lysis accomplishes is not the introduction of conceptual tools to replace those already employed in the various sciences we have touched upon, but rather a road map for following the common thread that links these hitherto disconnected and independent realms. What has been missing is a full understanding of how the power to effect change at all levels is interconnected and interdependent. I believe that something like this expansion and generalization of the concept of work is the critical first step.

One important contribution of this perspective is that it untangles the notion of causal power from undifferentiated notions of causality. Whereas our commonsense conceptions of cause tend to be applied w.i.l.l.y-nilly to all manner of physical changes, attributions of causal power are typically invoked in a more restricted sense. I believe that this distinction parallels the distinction explored in this chapter between changes due to work and changes in general. The addition of the term power is the clue that the issue has to do with work and not just causality in general-in other words, it invokes a sense of overcoming resistance, of forcing things to change in ways that they wouldn't otherwise change. And this is what is important to us. It is the emergent capacity to reorganize natural processes in ways that would never spontaneously occur, which is what we have been struggling to understand with respect to life, and with respect to mind.

The term causal power has become particularly a.s.sociated with debates about emergence and mental agency. For example, the contentious debates about the status of human agency and the efficacy of mental representations are not merely concerned with whether physical changes of state occur in conditions a.s.sociated with these phenomena; they concern the question of their status as sources of work, that is, the production of non-spontaneous change in some conditions. So, while there is no serious debate over whether mental events are a.s.sociated with such physical events as the release of neurotransmitter molecules, or the propagation of action potentials down neuronal axons, there is concern about whether the effects of mental experiences are anything more than this, and whether the experience of mental effort really plays any role in initiating physical changes that would not otherwise be attributable to the spontaneous consequences of physiological chemistry and locomotor physics. Putting these issues in the context of an expanded theory of work, we can appreciate that although all the physiological changes do in fact involve (for example) molecular interactions, the real challenge is determining in what way a person's spontaneous molecular, formative, and end-directed processes have combined to produce non-spontaneous mechanical, organizational, and semiotic changes. The issue is not whether the changes introduced into the world due to a person's considered actions are caused or not-of course they are-but whether there is something physically non-spontaneous about the effect. Debates concerning the status of human conscious agency are about the proper locus of this causal power.

Causal power is also a code word for what is presumed to be added to the causal architecture of the universe as a result of an emergent transition. But as we've seen, when this idea is conflated with more generic notions of causality, it yields a troubling implication: that such phenomena as life and cognition might be changing or adding to the fundamental physical laws and constants, or at least be capable of modifying them. The presumed restriction against this is the postulate of causal closure, discussed in chapter 5. We can now re-address this issue more precisely. Although the fundamental constants and laws of physics do not change, and there is no gain or loss of ma.s.s-energy during any physical transformation process, there can be quite significant alterations in the organizational nature of causal processes. Specifically, work can restructure the constraints acting as boundary conditions that determine what patterns of change will be orthograde in some other linked system. This is the generation of new formal causal conditions, and because the resulting orthograde dynamics will determine the possible forms of work that can result, it sets the stage for the emergence of unprecedented organizations of efficient causality, and so forth, with the generation of yet further new constraints, and new forms of work. As we have seen, this can also occur in ascending levels of dynamics, with a correlated increase in the possibilities of organizational complexity. So to restate the closure or conservation laws a bit more carefully: the universe is closed to gain or loss of ma.s.s-energy and the most basic level of formal causality is unchanging, but it is open to organizational constraints on formal cause and the introduction of novel forms of efficient cause. Thus we have causal openness even in a universe that is the equivalent of a completely isolated system. New forms of work can and are constantly emerging.

The concept of causal power is also of particular interest to social scientists, since it has become a critical issue for discussing the capacity for semiotic processes to control behavior and shape the worldview of whole cultures. Indeed, it is often argued that the very nature of the interpretive process is subject to the whims of the power of hegemonic semiotic influence. The problem has been that although this sort of power is an intuitively recognized and commonplace feature of human experience, it is at least as difficult to define as the concepts, like interpretation, that are supposed to be grounded in it. This difficulty may be mitigated by recasting this notion of power in terms of teleodynamic work.

Specifically, as we have seen, teleodynamic work is defined contra to orthograde teleodynamic processes. In cognitive terms, orthograde teleodynamic processes may be expressed as goal-directed innate adaptive behaviors, spontaneous emotional tendencies, learned unconscious patterns of behavior, stream-of-consciousness word a.s.sociations, and so forth. In social terms, orthograde teleodynamic processes may be expressed as common cultural narratives for explaining events, habits of communication developed between different groups or cla.s.ses of individuals, conventionalized patterns of exchange, and so on. As a result, we can easily recognize how it is that these orthograde predispositions to change might come into conflict, why they might encounter resistance, and how they might become specifically linked to afford transformation of work from one domain to another or become engaged in a form of semiotic and social evolutionary dynamic.

As we will see in subsequent chapters, these predispositions and the work that can be generated thereby are the basis for the generation of new forms of teleodynamic relations.h.i.+ps: that is, new information and new representations. In fact, the very concept of interpretation can be cashed out in terms of teleodynamic work. This will be the subject of the next two chapters.

In conclusion, being able to trace the thread of causality that links these domains avails us of the ability to discern whether methods and concepts developed in different scientific contexts are transferable in more than merely a.n.a.logous forms. It also makes it possible to begin the task of formalizing the relations.h.i.+ps that link energetic processes, form generation processes, and social-cognitive processes. Most important, it shows us that what emerges in new levels of dynamics is not any new fundamental law of physics or any singularity in the causal connectedness of physical phenomena, but rather the possibility of new forms of work, and thus new ways to achieve what would not otherwise occur spontaneously. In other words, with the emergence of new forms of work, the causal organization of the world changes fundamentally, even though the basic laws of nature remain the same. Causal linkages that were previously cosmically improbable-such as the special juxtapositions of highly purified metals and semiconductors const.i.tuting the computer that is recording this text-become highly predictable.

This causal generativity is a consequence of the fact that higher-order forms of work can organize the generation of non-spontaneous patterns of physical change into vast constellations of linked forms of work, connecting large numbers of otherwise unrelated physical systems, spanning many levels of interdependent dynamics. Although I have only described three major cla.s.ses of work, corresponding to thermodynamics, morphodynamics, and teleodynamics, it should be obvious from previous discussions of levels of emergent dynamics that there is no limit to higher-order forms of teleodynamic processes. Thus, the possibilities of generating increasingly diverse forms of non-spontaneous dynamics can produce causal relations.h.i.+ps that radically diverge from simple physical and chemical expectations, and yet still have these processes as their ground. This is the essence of emergence, and the creative explosion it unleashes.

12.

INFORMATION.

The great tragedy of formal information theory is that its very expressive power is gained through abstraction away from the very thing that it has been designed to describe.

-JOHN COLLIER, 2003.

MISSING THE DIFFERENCE.

The current era is often described as "the information age," but although we use the concept of information almost daily without confusion, and we build machinery (computers) and networks to exchange, a.n.a.lyze, and store it, I believe that we still don't really know what it is. In our everyday lives, information is a necessity and a commodity. It has become ubiquitous largely because of the invention, perfection, and widespread use of computers and related devices that record, a.n.a.lyze, replicate, transmit, and correlate data entered by humans or collected by sensor mechanisms. This stored information is used to produce correspondences, invoices, sounds, images, and even precise patterns of robotic behavior on factory floors. We routinely measure the exact information capacity of data storage devices made of silicon, magnetic disks, or laser-sensitive plastics. Scientists have even recently mapped the molecular information contained in a human genome that corresponds to proteins, and household users of electronic communication are sensitive to the information bandwidth of the cable and wireless networks that they depend on for connection to the outside world.

Despite this seeming mastery, it is my contention that we currently are working with a set of a.s.sumptions about information that are only sufficient to handle the tracking of its most minimal physical and logical attributes, but are insufficient to understand either its defining representational character or its functional value. These are serious shortcomings that impede progress in many endeavors, from automated translation to the design of intelligent Internet search engines.

The concept of information is a central unifying concept in the sciences. It plays critical roles in physics, computation and control theory, biology, cognitive neuroscience, and of course the psychological and social sciences. It is, however, defined somewhat differently in each, to the extent that the aspects of the concept that are most relevant to each may be almost entirely non-overlapping. The most precise technical definition of information has come from the work of Claude Shannon, who in the 1940s made precise quant.i.tative a.n.a.lysis of information capacity and transmission possible. As we will see, however, this progress came at the cost of entirely ignoring the representational aspect of the concept that is its ultimate base. As the epigraph to this chapter hints, this reduced definition of information is almost devoid of any trace of its original and colloquial meaning. By stripping the concept of its links to reference, meaning, and significance, it became applicable to the a.n.a.lysis of a vast range of physical phenomena, engineering problems, and even quantum effects. But this reduction of the concept specifically obscures distinctions which are critical to identifying the fundamental features that characterize living and mental processes from other physical processes. To begin to make sense of these world-changing transitions, then, we need a concept of information that is both precise and yet also encompa.s.ses the referential and functional features that are its most distinctive attributes.

In many ways, we are in a position a.n.a.logous to the early nineteenth-century physicists in the heyday of the industrial age (with its explosive development of self-powered machines for transportation, industry, timekeeping, etc.), whose understanding of energy was still laboring under the inadequate and ultimately fallacious conception of ethereal substances, such as caloric, that were presumably transferred from place to place to animate machines and organisms. Even though energy was a defining concept of the early nineteenth century, the development of a relational rather than substantive concept of energy took many decades of scientific inquiry to clarify. The contemporary notion of information is likewise colloquially conceived of in substancelike terms, as for example when we describe the "purchase" and "storage" of information, or talk about it being "lost" or "wasted" in some process.

The concept of energy was ultimately tamed by recognizing that it was a quite general sort of physical difference, which could be embodied in any of many possible substrates (heat, momentum, magnetic attraction, chemical bonds, etc.), and which could give rise to non-spontaneous change if able to interact with another physical substrate. Eventually, scientists came to recognize that the presumed ethereal substance that conveyed heat and motive force from one context to another was an abstraction from a process-the process of performing work-not anything material. Importantly, this abandonment of a substance-based explanation did not result in the concept of energy becoming epiphenomenal or mysterious. The fallacious conceptions of an ineffable special substance were simply abandoned for a dynamical account that enabled precise a.s.sessment.

A complete account of the nature of information that is adequate to distinguish it from merely material or energetic relations.h.i.+ps also requires a s.h.i.+ft of focus, but the figure/background s.h.i.+ft required is even more fundamental and more counterintuitive than that for energy. This is because what matters is not an account of only its physical properties, or even its formal properties. What matters in the case of information, and produces its distinctive physical consequences, is a relations.h.i.+p to something not there. Information is the archetypical absential concept.

Consider a few typical examples. Information about an impending storm might cause one to close windows and shutters, information about a stock market crash might cause millions of people to simultaneously withdraw money from their bank accounts, information about some potential danger to the nation might induce idealistic men and women to face certain death in battle, and (if we are lucky) information demonstrating how the continued use of fossil fuels will impact the lives of future generations might affect our patterns of energy use worldwide. These not-quite-actualized, non-intrinsic relations.h.i.+ps can thus play the central role in determining the initiation and form of physical work. In contrast, the material sign medium that mediates these effects (a darkening sky, a printed announcement, a stirring speech, or a scientific argument, respectively) cannot. The question is: How could these non-intrinsic relations.h.i.+ps become so entangled with the particular physical characteristics of the sign medium that the presence of these signs could initiate such non-spontaneous physical changes?

Answering this question poses a double problem. Not only do we need to give up thinking about information as some artifact or commodity, but we must also move beyond an otherwise highly successful technical approach that explicitly ignores those aspects of information that are not present (i.e., referential content and significance).

The currently dominant technical conception of information was important for the development of communication and computing technologies in response to the need to precisely measure information for purposes of transmission and storage. Unfortunately, it has also contributed to a tendency to focus almost exclusively on only the tangible physical attributes of information processes, even where this is not the only relevant attribute. The result is that the technical use of the term information is now roughly synonymous with difference, order, pattern, or the opposite of physical entropy, and thus is effectively reduced to a simple physical parameter entirely divorced from its original meaning. Indeed, something like this was implicit in Boltzmann's account of the concept of entropy, because at maximum entropy, we have minimal information about any given microstate of the system. This redefinition of the concept of information as a measure of order has, in effect, cemented the Cartesian cut into the formal foundations of physics. More perniciously, the a.s.sumption that information is synonymous with order has been used to implicitly support the claims of both eliminativism and panpsychism. Thus, if it is a.s.sumed that we can simply replace the term information with order, thought becomes synonymous with computation, and any physical difference can be treated as though it has mentalisitc properties.

To step beyond this impa.s.se and make sense of the representational function that distinguishes information from other merely physical relations.h.i.+ps, we will need to find a precise way to characterize its defining non-intrinsic feature-its referential capacity-and show how the content thus communicated can be causally efficacious, despite its physical absence. And yet this a.n.a.lysis must also demonstrate its compatibility with the precise technical conception of information, which although divorced from this key defining attribute, has been the basis for unparalleled technical advances in fields as different as quantum physics and economics.

Tacit metaphysical commitments have further hindered this effort. This more encompa.s.sing concept of information has been a victim of a philosophical impa.s.se that has a long and contentious history because of its entanglement with the problem of specifying the ontological status of content of thought. We have progressively chipped away at this barrier in previous chapters, so that at last it can be addressed directly with fresh conceptual tools. The enigmatic status of this relations.h.i.+p was eloquently, if enigmatically, framed by the German philosopher Franz Brentano's use in 1874 of the curious word "inexistence" when describing mental phenomena: Every mental phenomenon is characterized by what the Scholastics of the Middle Ages called the intentional (or mental) inexistence of an object, and what we might call, though not wholly unambiguously, reference to a content, direction toward an object (which is not to be understood here as meaning a thing), or immanent objectivity.

This intentional inexistence is characteristic exclusively of mental phenomena. No physical phenomenon exhibits anything like it. We can, therefore, define mental phenomena by saying that they are those phenomena which contain an object intentionally within themselves.2 As I will argue below, neither the engineer's technical definition of information as physical difference nor the view that information is irreducible to physical relations.h.i.+ps confront the problematic issues raised by its not-quite-existent but efficacious character. The first conception takes the possibility of reference for granted, but then proceeds to bracket it from consideration to deal only with physically measurable features of the information-bearing medium. The second conception disregards its physicality altogether and thus makes the efficacy of information mysterious. Neither characterization explicitly accounts for the relations.h.i.+p that the "inexistent" content has to existing physical phenomena.

OMISSIONS, EXPECTATIONS, AND ABSENCES.

The importance of physical regularity to the a.n.a.lysis of information is that it is ultimately what enables absence to matter. Where something tends to regularly occur, failing to occur or in other ways diverging from this regularity can stand out. A predictable behavior (and more generally any tendency to change in a redundant way or to exhibit regularity or symmetry) is presupposed in many circ.u.mstances where something absent makes a difference. The source of this regularity need not be exemplified by human action. It could, for example, be exemplified by the operation of a machine, as when stepping on the brakes fails to slow one's vehicle and thus provides information about damage. It could be provided by some naturally occurring regularity, like the cycling of the wet and dry seasons, which in a bad year may be interpreted as information that certain deities are angry. Even systems that appear mostly chaotic and only partly constrained in their behavior can offer some degree of predictability against which deviation can be discerned. So, although financial markets are highly volatile and largely defy prediction on a day-to-day basis, there are general trends with respect to which certain deviations are seen as indicative of major changes in the economy that cannot be discerned otherwise.

In the realm of social interactions, we are quite familiar with circ.u.mstances where omissions can have major consequences. For example, after April 15, if a U.S citizen has not prepared and submitted a U.S. tax return, or an extension request, serious consequences can follow. In a legal context that requires producing a tax return by this date, its non-existence will set in motion events involving IRS employees coercing the delinquent taxpayer to comply. They may write and send threatening letters and possibly contact banks and credit agencies to interfere with the taxpayer's access to a.s.sets. Similarly, so-called sins of omission can also have significant social consequences. Consider the effect of the thank-you note not written or the RSVP request that gets ignored. Omissions in social contexts often prompt deliberations about whether the absence reflects the presence of malice or merely a lack of social graces. Failure to initiate an intervention can also provide information about the actor. A failure to prepare or to attend to important warnings can be the indirect cause of a disaster, and this failure can be grounds for punishment, if intervention is expected. Even ignorance or the absence of foresight due to lack of appropriate information or a.n.a.lytic effort can be blamed for allowing "accidents" to occur that might otherwise have been avoided.

Intuitively, then, we are comfortable attributing real-world consequences to not thinking, not noticing, not doing, and so on. In these human contexts, we often treat presence and absence as though they can have equal potential efficacy. But these omissions are only meaningful and efficacious in a context of specific expectations or processes that will ensue if certain conditions are not met, or in the context of undesirable tendencies that will ensue in the absence of opposition. Where there is a habit of expectation, a tendency that is intrinsic, a process that needs to be actively opposed or avoided, or a convention governing or requiring certain actions, the failure of something to occur can have definite physical consequences.

These familiar examples are, of course, special cases that invert the general case of representation-something present that is taken to be about so