I’ve been thinking about “complicity” (the term comes from the writing of my friends Ian Stewart and Jack Cohen (see Figments of Reality and Collapse of Chaos); it refers to the idea that subsystems in complex organizations are thoroughly interactive), and how to put the concept on a somewhat more formal basis.

One way to view the unfolding of the universe is as a clockwork cosmos: Starting with the most fundamental LaPlacean elements—that is, from a “God’s-eye view”—the entire future of the universe, and every trajectory to arrive there, is laid out perfectly for observation. The trick, of course, is to be able to reduce things to those elements.

In this view, emergent phenomena are merely different magnifications of the abstractional telescope. Yet, to root out the needed fundamental elements, how far is “down”? Who knows? A first argument, then, against the reductionist approach is to point out its practical futility.

But I’m thinking there may be something else going on: Even given complete LaPlacean understanding at any given level (i.e., at whatever level of description we choose to define as “fundamental”), it is not possible, even in principle, to derive a complete understanding of a higher level. Indeed, if I am correct, it is, in any description that involves a temporal component, not possible to have a complete understanding of even a lower level of description. Further, this is not merely a function of the limits of parsing resolution, but is integral to how the universe goes about its business. That is, not only does it unfold from the bottom up, but also from the top down. In other words, feedback—both positive and negative—occurs not just among elements on a single level of abstraction (a difficult enough sorting-out challenge), but, crucially, across levels.

[It is important to note that feedback mechanisms involve randomness, or quasi-randomness, of two kinds. One is in the addition of noise in the signal that is fed back. The other is in timing jitter, so that the return delay is to some extent unpredictable. To schematicize this, think of noise generators both in series with the signal, and parallel to the delay generator within the feedback generator. Indeed, one can imagine many complications of this sort. These complications, in addition to the tuning of the loop’s resonance, can be viewed as a mechanism for amplifying randomness: If quasi-randomness is insufficient, then you can tune your system by amplifying real (quantum) randomness to any degree you wish.]

This is, of course, implicit in the gestalt idea that the sum is greater than the parts. But I propose to update this by striving for a formalization of the behavior of emergent systems. Consider, for instance, that the logic of an anthill cannot be derived easily (and maybe not at all) from a knowledge of the logic of all the component ants. The collective logic, from which is generated the next higher level of abstraction, is greater than that of the sum of the participants. (In a sense, each ant is a kind of stupid homunculus; as Dan Dennett points out, a hierarchy of decreasingly intelligent homunculi is no violation of infinite regress.) The logic of the colony operates on its own level, and “cares” nothing, or very little, about the fate of any one ant. Yet the logic of the collective profoundly influences the behavior of individual participants. A danger to the whole beats the war drums for the parts.

Put this way, we see complicitous feedback nearly everywhere we look. Cf, for instance, a line I often use: For all the yowling and screeching, there is no shortage of kittens. The logic of genetics and reproduction in on a different level from, and cares not more that a whit or two about, the logic of a pain-free existence for Tom and Thomasina, but deeply influences their conduct.

Genes, memes, and demes all are logic systems that are partly independent and partly interdependent upon each other. [That is, they have some finite impedance in their presentation.] On the level of human emotion, cheating partners can cause true pain, yet this logic is (often) subordinate to the logic of selfish genes. But memes (often) fight back, calling the shots at the organism level. What calls the shots at the social level?

At the scale of human consciousness (consciousness apparently being an emergent phenomenon), we cannot be overly concerned about the plight of any individual cell. When an influenza virus invades, we throw numbers at the problem—that and the Darwinian logic of the immune system. These cells do their patriotic duty, presumably knowing nothing about the collective logic (or, at least, no more than the average soldier about the reasons for being at war…). Still, they die hideous deaths in massive numbers for it. And we, the collective, are relieved, if not precisely “grateful”.

Many scientists, including Steve Pinker, deny the importance—even the existence—of “superorganisms.” (See The Blank Slate.) Thus, emergent systems are, to them, a non-productive way to think about (human) behavior. But I believe that understanding emergence is the key to understanding behavior, because it is a metaphor for what happens whenever there is an abstractional gradient in the phase changes just south of chaos—whenever, that is, there is exploitable fuel to power life.

Given a Darwinian imperative and domesticatable fuel, logic systems—life—become not only possible, but mandatory.

Logic systems? Think of that which follows a dynamic “logically”, like drying mud or cooking oatmeal arranging itself into hexagons. “Life” is merely a complex logic system—complex in that it operates on multiple layers of abstraction, and so gets replication, or reproduction, into the bargain.

With sufficient fuel—exploitable, well-behaved “energy” gradients—logic systems will emerge out of, well, out of pretty much anything: a primordial soup, genetics, memetics, music, the playing fields of Eton, the killing fields of Cambodia, politics, the economy, even hagiography, for goodness’ sake. And, given the Darwinian imperative and sufficient fuel, logic systems can, and will emerge from collections of logic systems. (And it is logic systems all the way up, or at least a long way up; the higher one goes, the longer the supply lines, so, eventually, logic systems emergent from logic systems emergent from logic systems run out of exploitable fuel.)

“Given the Darwinian Imperative”? Perhaps natural selection is the fundamental logic system, from which all logic systems evolve (and will evolve, if numbers can be thrown at the situation, if there is a mechanism for “chance” variation, and if there is sufficient fuel and time enough). Literally, universal Darwinism?

Then, we come to the realization that fuel—exploitable resources, both as energy and as catalysts—are always a specific life’s limiting factor: Any logic system that has depleted its needed resources is a goner, and can take down a web of interdependent logic systems with it. Life needs fuel: A colony of bacteria in a petri dish will grow until it runs out of food. Then it dies, unless it can, somehow, migrate—to the refrigerator, say—or otherwise find a new (and larger) supply of food which the bacteria evolved to exploit. The alternative, a much slower one, is to evolve into something that can exploit a different, and (at least temporarily) more plentiful resource.

At this point, I’d like to take a large abstractional leap, to consider a kind of super-organism: the economy that allows 6 billion and more humans to inhabit the petri dish known as earth. As with all complex “econo-spheres”, resource density is not uniform. Indeed, whether one is a pauper, a prince, a body-politic, or (maybe) a black hole, the whole logical point is to have more exploitable resources than the next guy—a gradient that itself becomes a resource to fuel other super-organisms, like a social order. Thus, in the logic of personal continuance, a place at the trough is the crucial first requirement, and, when push comes to shove, niceties like human ethics and morality become a luxury—in the trenches, a quite expendable one.

Logic systems built from logic systems built from logic systems…may get “smarter” all the way up, but they have increasingly long “supply lines”, and are thus like a house of cards: strong enough with no stress, but what happens when the glue fails? (As Hannibal discovered, elephant feed is harder to come by in Switzerland than in Carthage.) Or, try another metaphor: the carnival rubber “castle” that maintains its structure by the constant influx of hot air. And…when the compressor is turned off? Still another metaphor: Life is, when resources are abundant, like a game of musical chairs, where more and more chairs are put out, allowing an increasing number of players. So…what happens when the chairs start getting taken away? And, not just one at a time, but in whole bunches…? And this is the only game in town? And it isn’t really a game and the losers really do lose, forever?

I am talking about oil, of course. Sure, I also mean uranium and fish in the sea and water in the aquifer and trees in the forest (especially those of the tropical rain forest; temperate forests can, if we let them, grow back in a few human lifetimes) and all the other resources non replete-able within individual lifetimes. But, for the present discussion, I mostly mean petroleum and other fossil fuels, which collectively are the major source of “hot air” to keep the economic rubber castle of six billion and more humans from collapsing. (“The economy”, far from monolithic, is really an emergence, a collective of little economies—and it is economies all the way down.) But the world’s supply of this lifeblood is replete-able only over the span of hundreds of millions of years, while its present rate of consumption will suck the dish pretty much dry within a couple hundred (from when the marvelous gushing teat was first sucked; many, including some with excellent credentials, feel that the great weaning has already begun…)

The world’s economies have evolved to be highly tuned, “professional” exploiters of this particular fuel. When we run out of oil (more correctly, when we run out of cheap oil), and we will, who will win? Who will lose? Who will have a chair when the music stops for good? And will the losers be good sports?

A world without oil will, at the very least (that is, if a viable replacement is in place…quickly, and if we don’t trash our version of the world through other means), demand a terrific restructuring of economic logic, and, with it, individual fortunes. Likely, a world without oil will not support six and more billions with any recognizable level of affluence—and possibly not at all. The “natural selection” pressures for new economies will, without a doubt, be ferocious, especially as they must inevitably turn more local and isolated. As with all Darwinian processes, not all will win; most will lose.

All this talk about the end of oil is important, but, to return to the question of how complicity might work, I have a more academic—if, ultimately, no less eschatological—matter I now wish to muse upon. (I am a human, of course, and I have concerns and fears on a human scale, which is why the End of Oil is an important topic: We are, for all practical purposes, really discussing the End of Time—at least for many of us and our lineages. But…well, there is, I think, the unmissable metaphorical point about riding a few wild metaphors through many levels of punditry-space.)

The question is about catastrophe tracking, as opposed to Fisherian tracking, in evolution.

We begin by observing that a lineage is best served—i.e., it can continue to be a lineage—if its competencies track environmental change.

Now, in order to keep the discussion moving, let’s not get carried away; nevertheless, ruminating about what, exactly, “the environment” consists of is worth a few moments of our time. For genes, the most obvious—and, more to the point, immediate—environment is a genetic one. After all, it is the mutational variability among the genes in a common pool that gives natural selection the raw material to work with. But there is a lot of context “out there”: The flux of meteorological, seismic, atmospheric, astronomic…not to mention social, cultural, and memetic conditions—in short, events across all levels of abstraction—define the milieu in which lineages play their Darwinian roles. Furthermore, no gene is an island unto itself, but, I’m thinking, recruits others and joins coalitions to actively alter, or attempt to alter, at least the own local context, to skew it away from both idleness and chaos. All the while, the competition is doing pretty much the same thing, of course, so the dance is extremely complicated, and not entirely amenable to a reductionist exegesis, except, possibly, on a statistical basis. (But how reductionist it that?)

Back to my main argument. Fisher (as Richard Dawkins relates in The Blind Watchmaker) proposed a microscope focusing mechanism as an analog of environmental tracking. Following this train of thought, large jumps—maybe because the microscope operator has twitchy fingers—will not track the small changes assumed to occur to the plate relative to the system’s optics. Small shifts, however, usually do a fine job. The analogy works well, then, in an environment that is comparatively static (that is, it changes more slowly than the tracking system). Even though there will always be some tracking latency, it will be minimal. However, it is worth noting, for purposes of extending the image, that contexts flux on many temporal (and spatial) scales, and the speed and magnitude of any particular event are unpredictable—except, crucially, on a statistical basis.

Does evolution evolve? Natural selection may be blind, but in the competition among selection mechanisms, will one win where others lose? Given enough time evolution will hit upon a Good Idea (if it works!), and it is clear that a suite of skills to track, with minimal latency, contextual flux over wide magnitudes and slew rates (to borrow a term from electronics engineering) is a very fine idea indeed: The better an organism can thus track, the better off its lineage will be. To be sure, there is always that pesky issue of economic trade-off. Developing and maintaining a sophisticated mechanism for tracking types of changes that happen once in a blue moon may not be worth the cost. Also, it should be noted that hardwiring statistical predictions into a genome, though not impossible, nevertheless requires a lot of data. These data will not exist for very rare events, although what is considered “rare” depends on the life-span organism doing the “analysis”.

[Jumping back to the matter of oil, the fact is that the “sudden” and catastrophic reduction of such a fundamental source of energy is extremely rare, and most desperately experienced only by those who did not pass on their reflections on the situation to their descendents, because there weren’t any. Still, there are a few models to consider and scare the pants off anyone who thinks too much about it all: The K-T catastrophe that was the bete noir of the large reptiles, for instance, along with whatever brought about the Permian and other earlier mass extinctions. If these are too distant, consider the more modern examples of the fall of Easter Island civilization , the end of the Mayans, the reindeer of St. Matthew Island (all wonderfully described by Jared Diamond in Collapse), and, of course, bacteria in a petri dish. But, while we are still clothed, back to the story:]

So slew rates and magnitudes can be plotted against time, as can economic conditions. The optimal trade-offs are, then, what competing lineages are also calculation and betting the farm on. But all gamblers really want to win, and extinctions do happen, so arms races are inevitable, and hedging of bets, poker faces, cheating, and all manner of complex advanced techniques will, eventually, develop.

I was musing about this a few years back when one of the kids brought home a video of Rambo. Interesting juxtaposition, that. Under most circumstances, it would seem, sociopaths and psychopaths are not “adaptive”, which is to say that a modern society would be best served by eliminating all sociopathology and psychopathology from the gene (or meme) pool…yes? The problem is that a “modern” society can turn into a Lord of the Flies scenario (or worse) very quickly. Noah had some notice of the impending flood, and migrational vanguards (e.g., the early nomadic settlers, or, in fact, the early European settlers, of the New World) can more or less predict the unpredictability of the new environment. (This is, of course, one super-genetic reason that both biological and cultural evolution take place faster at the frontier, in the small side channels of life rather than in the mainstream.) But Douglas Adams’ Arthur Dent (nearly the late Arthur Dent) had essentially no warning that his world was going to change drastically—actually, to go out of business—pretty much before his very eyes.

So, in an idealized world, we would no longer have any Rambos grunting and slashing among us…? Uh, what will Earth do when the Rigelians show up, hell-bent on finding a new source of protein, or a new planet to rule since they trashed the home world? As it turns out, of course, we have never figured out how to make a perfectly civilized society, and that may be a good thing. A metaphor stretched past its topological aptness is a dangerous thing, often tempting us to skin the bark off the wrong tree, but maybe we can, at least for now, get away with suggesting that societies can profitably hedge their bets by keeping a few misfits in their talent pools.

To recap, what is needed—what would be a Good Idea—is a mechanism to track contextual flux slowly most of the time, but to keep several layers of automatic back-up devices ready to kick in when normal tracking is too slow and too feeble to do the lineage much good. Does this happen?

Let’s say that it is likely that most genetic changes in a lineage are not adaptive, but, rather, they are neutral or otherwise non-participatory so far as natural selection is concerned. Now, perhaps we can make a small amendment to this proposition: “…they are neutral so far as natural selection is concerned most of the time.” Mutations come in many sizes, from micro to macro (I mean, of course, that individual mutations will, depending on the specific site altered, bring about all different sizes of phenotypical changes, which is what linked traits and regulatory genes are all about), and most must be non-adaptive—especially, like with Fisher’s microscope, the larger ones. Non-adaptive mutations, by definition, do not “direct” a lineage in any particular direction, so, under normal circumstances, are never noticed. In fact, these are vastly out-competed if not outnumbered by those mutations that are adaptive. But how about under extreme stress, when normally tracking micro-mutations are all goners? Then otherwise non-adaptive macro-mutations might suddenly become adaptive (since they would be optimized for conditions somewhat-to-quite distant from those of normal tracking) and now stand a chance, small, perhaps, but the best one for a tenable future, yes?

[To briefly chase down a rabbit hole, I’m thinking that the evolution of neural solutions (which, in turn, eventually lead to memetic mechanisms and consciousness) may have come about in this fashion, propelled, of course, by a fairly specific solution (like Bill Calvin’s rock throwing; see The Throwing Madonna) to a general need. This discussion could get quickly out of hand, to become a whole warren, as it would involve the concept of metaphor (particularly, but not only, memetically represented metaphor) as a way to warp-drive across abstractional possibility space, a favorite topic of mine. For now, then, I will limit my commentary to observing that consciousness is a fairly low-cost way to track many high-speed aspects of contextual flux.]

This weak form of catastrophe tracking becomes, then, a testable hypothesis. We need only the empirical observations that 1) most mutations are non-adaptive, and 2) mutations come in many “sizes”.

What about a slightly stronger form, employing a “telenomic” mechanism (to use Mayr’s term)? That is, perhaps it would be interesting to speculate about an “exaptational drive”, wherein organisms in a lineage have “learned” to hedge bets by generating not single purpose solutions, but somewhat generalized ones that, in extremis, can be converted to other functions? [Remember linked traits? Perhaps some mutations can unlink them…?]  In other words, where the convertibility is built in, as it were? This, too, seems possible, and rather likely. (Indeed, I hope very much that this is the case, at least where humans are concerned: These frail bodies and lame brains were “invented” over the long haul of evolution, most of which happened a long time ago. But, over the next few years [!] they will have to deal with some very sophisticated and thorny problems. If those bodies are as hard to alter as they seem to be, I’m hoping our minds can work with the flexibility necessary to see our lineages through. [To be sure, however, mere hope may not cut it: Unlike the Judeo-Christian Divinity, Nature guarantees us no special privileges.])

How could this come about? I believe this idea is implicit in Richard Dawkins’ “evolution of evolvability” concept. In the long run, probability analysis based on historical statistics must tend to get hardwired into the genome (‘corrections’, needed to account for “quickly breaking news” can be dealt with socially, culturally, or memetically). And one of the statistical predictions is that “change” is unpredictable but inevitable (particularly so in small, isolated gene pools that are candidates for becoming new species), so, rather than fight it, go with it. This is a limited form of predicting the future based on past history. Moreover, it is, I believe, why the blind watchmaker analogy eventually bites the dust: While the watchmaker, or Watchmaker, may indeed be truly blind, the protegenic apprentices will, some fine eon, hit upon both figurative and literal vision as a Good Idea. Humans, of course, carry the vision thing to an extreme, as I’ve discussed elsewhere, even purposely attempting to design their own futures….

We know that many genes are expressed differently in different contexts (which is, of course, why the “recipe” metaphor works well while the “blueprint” one does not), and not only different genetic contexts, but those including the Outside World as well. In a known, slowly changing milieu, different genetic expression is largely predictable, but, under stress, all bets are off. Moreover, different expressions often lead to phenotypical and behavioral differences, which in turn lead to greater variation in sexual selection, which feeds back to influence the variability natural selection has to work at the genetic level. Thus, via positive feedback, environmental stress could engender significant (macro-) genotypical changes with extreme rapidity—in just a few generations in some cases. All this would occur with lowered resolution and redundancy within a lineage, to be sure, but with its continuance at stake, who cares?

In any case, my hunch is that the robustness of a species in coming to terms with contextual flux (i.e., its resistance to extinction) has a great deal to do with its exaptational skills—its flexibility, or its ability to generalize in some fashion. At the molecular level, I surmise that some parts of the genome, or even, perhaps, some “junk DNA” could be cannibalized for raw parts in an emergency. (In principle, this is testable, I’m pretty sure.) At the neural level, this would seem to indicate that some part of the slate must be blank, or at least re-writable (sort of like an E-PROM). And, at a cultural level, we can muse about all sorts of Donald Brown’s “universals”—such as music—that don’t seem to have an obvious reason to exist so far as standard evolution theory is concerned. That is, while our primary memetic logic is built around language (in cats, visual images, maybe; in dogs, olfactory ones), a logic or meta-logic built around music, say, is entirely possible, and surely happens in many cases, especially in dire straights. This is true regardless of the “origin” of the “universal” in question. Calvin cites sequencing as the most important precursor to music. I’m inclined to go along with him, although I believe that there are other precursors (it is the collective of these precursors that is important); one can’t, for example, build polyphony and counterpoint out of sequencing talents alone, but there it is. Withal, the point is that life, on any level of abstraction, has to be opportunistic, to ride the dynamic wave that washes it, to build a furnace to burn the fuel it is presented with. And this fuel can be a musical logic system. It can also be any philosophic logic system, including science….

While I’m thinking about it: Some commentators have theorized that “consciousness”, at least as it is conceived from introspection, is impossible without language. Now, I am not qualified to discuss this from a neurological standpoint, but I can say, with a fair amount of confidence, that a representational “mind” is such an astoundingly Good Idea, in spite of the obvious propagation delay, that evolution must have hit upon it very early—long before humans, with their vaunted sequencing skills upon which language is largely built. Again, when I talk about a “representational mind”, I mean a metaphor machine. True, linguistic metaphors are largely temporal (sequential) orderings (often) temporal “micro-phors”. But this linguistic usage of a deeper metaphor system may have essentially hitched an evolutionary ride: Non-linguistic animals—the rest of the lot—surely engage reality at a representational level as well, just without the fancy sequencing hardware. And we humans, too, engage the first order modalities as the front end for representation, which must be, at the bottom of it, comparison with what is stored in memory. In this way of looking at metaphor, language (a higher-order modality) is important, to be sure, but mostly for, as I say, the primary engine to “warp-drive” across possibility space. Pre-language “engines” run at “impulse speed” only. The other thing to say about this is that humans often engage other higher-order modalities, like spatial mapping, to generate metaphor. And this, too, may have hitched a ride: Higher order representations require a lot signal processing horsepower to generate enough speed to be truly useful, and so have surely been less used in the days before big brains.

[To fully explicate these ideas at the human level requires an aside into the philosophical concept of object vs. subject, or the I/me dichotomy. I’m sorry to leave the connection cryptic here, for the topic deserves better. But in brief we can trace the evolution of what I am calling the “metaphor system” from “object proxies” (Antonio Damasio’s term; see The Feeling of What Happens: Body and Emotion in the Making of Consciousness) to an open-ended, multi-dimensional abstractor/generator. At this level of discussion, we are thereby provided one type of positive feedback explanation for the evolution of a large, cognitively generalized brain. But where the idea really takes off is the next level of discussion, for there we consider “metaphor” as not only a way of mapping, via proxies, the unfolding of the universe, but as an active (!) participant in the unfolding.]

There is another form of catastrophe tracking that is so obvious that it is easy to miss: sex. Dither, to use another of my favorite metaphors, makes any digital system more robust. Genetics is digital; shuffling genes from one generation to the next is a serious hedge against specialist, “professional” genes that are extremely successful in a relatively fixed environment, but stand no chance if the game suddenly changes. This, in fact, is closely related to the Rambo thing: Professionals—specialists—will out-compete amateurs in a slowly changing milieu; the fortunes are reversed when things speed up and approach chaos.

The strongest form of catastrophe tracking—a putative set of genes that somehow become “aware” that something has gone terribly wrong with the world out there and, in essence, pushes the panic button to allow macro-mutations that would have otherwise been suppressed—is, admittedly, the most speculative as well. Still, there are, I feel, good reason to explore the possibilities, at least to generate blue-sky hypotheses. The main problem has been the lack of a theory how higher order events can influence lower order ones—how top-down explanations can work in an otherwise bottom-up universe. In fact, much progress toward such an understanding has been made in the last few decades. I won’t bother here to develop the full thesis, though I will outline some of the topics needed for its exposition:

Emergence theory

    Hierarchies

        Nested

        Non-nested

    Feedback

        Static

        Dynamic

            Negative

            Positive

    Constraints

 

Levels of Abstraction

    Metaphor system: micro- to meta-proxies

        Features and object proxies

        Pattern proxies

        Structure proxies

    Fungibility

        The “collapse of fungibility”

        Reification

        Resonance

 

In short, the idea is that the universe unfolds not only in a bottom-up sense, but a very real top-down one as well (one does not need to posit a participatory Creator to make this work), and that there are lines of communication between these them. That life is an emergent phenomenon is no longer especially controversial, of course; what is less agreed is that higher levels of emergence influence (direct? design?) lower levels. I’m pretty sure I believe this latter thing to be true, though I will not here try to convince anyone else to concur. I’ll only mention a few trivial examples. 1) The health of the anthill affects the roles taken on by individual ants; 2) thoughts alter the states of individual neurons; 3) the economy influences the plights of individual citizens.

Of course, one can argue that this is merely looking through the other end of the abstractional telescope. But, to recapitulate, I counter-argue that it is not possible, even in principle, to understand the workings of a higher level from even a complete understanding of a lower one (the whole is truly greater than the sum of its parts). Thus, the workings of that lower level system cannot, out of an open-ended context, be completely known either!

Keeping all this in mind, we can conclude, I believe, that it is the human condition, not science fiction, to want to know the future, to plan for it, and to attempt to change it, even when that change involves tinkering with genes and their associated molecules. (Humanity is the Designing Ape, the God-playing Ape.) Furthermore: “human condition”, surely, but on a far more limited basis, that of every living thing. As alluded to earlier, Dennett has somewhat rehabilitated the “homunculus” concept by pointing out that a sort of virtual, emergent homunculus, made up of increasingly more stupid agents “all the way down” violates no important principle. This is, of course, the flip side of Darwin’s “appearance of design” idea. Natural selection and mutation are about as “stupid” as you can get. Yet, out of those Darwinian raw parts, highly complex, even intelligent results can be had. Out of blindness comes vision, and from vision, an “improved” genotype, perhaps, may ensue. This is how “nature does it”, we know, for the simple reason that we, too, come from those same natural, stupid, Darwinian raw parts!

At the same time…

We think we are smarter that the bacteria…? I’m not totally convinced….

Another hypothesized example is less trivial, and fairly close to home. It can be inferred from a reading of standard evolutionary theory, and is testable. It also fits into Stu Kauffman’s conceptual framework regarding extinctions (see Investigations): Perhaps a killer comet did do in the dinosaurs, but, if it hadn’t, endogenous processes would have eventually accomplished the same thing. (As the Kingston Trio pointed out, years ago, “What nature doesn’t do to us will be done by our fellow man.”) Here is what I am thinking: If an organism senses impending doom, the tendency is to throw numbers at the problem. In other words, K-type reproductive strategies work best in relatively stable environments, which large mammals specifically evolved to exploit—to attempt to bring about (and in the case of humans, to attempt to assure!). For most other critters—insects, bacteria, and so on, existence is always a dicey issue, so R-type strategy is always appropriate. However, “K” and “R” strategies are not discreet, but on a continuum. “Stability” varies, so, if there is a mechanism to estimate relative stability, it most likely evolved to help reproductive strategies track it.  (The main way to increase the tracking speed of natural selection is to increase the selection pressure. This happens naturally when higher numbers compete of more difficult to obtain resources.)

What could such a stability sensor look like? Many things. There are limitless numbers of potential calamities ready to befall an organism. Thus, a sensor, like Granddad’s bunion that could predict a thunderstorm could be useful. So could a nose that knew a noxious product when it smelled one. Indeed, any sense organ—primary or, like language, a virtual (higher order) one—can, and will be, pressed into service. Is there any evidence that such a feedback loop exists, and operates among humans? I mean, past the obvious sensory ones? Possibly. It is apparently documented that families are smaller in urban (from a health-care standpoint, stable) milieu than in a rural (dicey) one. Also, low-income (from a social-success standpoint, unstable) families seem to have larger families than the (relatively stable) working middle class. Teen pregnancy is far more common among the poor (life’s potential losers) than among the rich (the obvious winners).

None of this is proves my conjecture, of course, but we may now guess where to look for a smoking gun. And, if this is at all close to how things actually work, it bodes ill for trying to take the pressure of a population that will suddenly be far higher than the environment can support simply by appealing to “reason”.

“Life on Earth” is not endangered, to be sure. The presence of some 150 varieties of bacteria in the Hanford radioactive waste tanks attests to that. And, just perhaps, even human life has a future, although that may be small comfort to those “losers” who face personal extinction. I mean their lineages, of course; we all face personal extinction. Indeed, one way to view death is as an act of creation, as Heinz Pagels pointed out with wrenching eloquence (see The Dreams of Reason). Neotony may be one way life (in general) tunes itself: Youthful plasticity keeps the “bits” from stagnating, while death keeps the living from slipping over the edge of chaos.

[Dancing on the scarp with Death? What, a pavane? Maybe a tango? But that is another discussion, or maybe the same one….]