It is time to develop stronger support for this unusual view I am advancing. I've offered some support from rather generic (if unpopular) epistemic and theoretical Galilean perspectives, but there is a deeper, more abstract level for us to explore. Both current evolutionary theory and value theory, I believe, can be reduced to a simpler conceptual level. We need a metatheoretical foundation for this reduction.

Though it must sound grandiose and overly presumptuous, the foundation I offer here applies nonetheless to all the sciences that examine the interactions among physical entities (such as subatomic particles, atoms, molecules, cells, organisms, stars, and galaxies) or among functional entities (such as personalities, species, psychological traits, economic interests, or political ideologies). Unless I'm mistaken, that's all the sciences, period. This foundation has been implicit in the work of those scientists and philosophers who have constructed today's prevailing Galilean framework -- but heretofore, as far as I have been able to determine, it has not been made explicit by anyone.⁸

This claim may seem less outrageous when placed in a familiar context. Numbers and arithmetic allow us to examine (in a way that is epistemically illuminating if not logically ampliative) certain relations among entities and groups of them, whether they be physical or abstract. Our metatheory operates at a comparable level of abstraction. It has an obviousness that I hope will engender a "Why didn't I think of that?" rather than a "So what?" reaction.

2.1 Valence

I want to adopt a term from chemistry, 'valence', to represent the key concept in this metatheory. Chemists treat valence (most commonly) as the number of hydrogen atoms with which a given kind of atom can combine in a chemical reaction. It is a dispositional property depending equally on the nature of the atom in question and the nature of hydrogen atoms with which it might bond. It might be tempting to say that valence simply is the nature of a given atom (or one aspect of it); but if hydrogen were structurally different in some important way -- such as having a nuclear charge of +2 instead of +1, two electrons filling its 1s orbital, and therefore no tendency to form covalent bonds (it would actually be helium, which, by any other name, would be just as inert) -- there would be no hydrogen atoms with which other atoms might bond, and the term 'valence' would be useless. Only the existence of an element having an unfilled 1s orbital allows us to talk about valence in this traditional way.

Even though chemists talk about valence as a property of the various elements, I think it is more useful and illuminating to treat it as a property of a situation (real or counterfactual). Put this carbon atom, say, near four hydrogen atoms, and barring external interference they will form a molecule that would require a significant amount of energy to break into its original parts. They will do so as though by common consent -- not by unilateral action as one entraps the others, but by mutual attractions dictated by their specific individual properties. We could say then that the situation -- of which the five atoms are components -- has this dispositional property: a valence for producing a methane molecule. If the molecule has actually been formed, we will say the valence is filled or realized or instantiated.

In fact, a valence is a defining property of a situation. Since each object in the universe supposedly interacts with every other object (e.g., every mass shares a non-zero gravitational interaction with every other mass, and every charge shares a non-zero electrical interaction with every other charge), we can't define 'situation' according to the presence or absence of interactions. Instead, we use the term as a conceptual convenience allowing us to talk about some collection of interactions, typically within a more or less well defined neighborhood, that can produce some physical behavior (or could produce some physical behavior in the presence of some additional component of interest). Absent some salient behavior, or the potential for it under modified conditions we can imagine, there is no apparent reason to conceive of any arbitrary collection of objects, along with the interactions they share, as a situation.⁹

Valence, generalized as I propose, is more than just a chemical property; it is a defining property of any situation whose component objects have a disposition (latent or realized) to behave in a way that is completely determined by the interactive¹⁰ properties of those components, even though external forces might introduce minor perturbations. 'Valence' is a term of functionality.

Natural laws describe valences of common physical situations -- or, for those who view natural laws as physical properties, they are kinds of valences. The law of gravity, for example, describes how matter accretes in space, how simple orbits are established, and how Lagrange points in a two-body gravitational system tend to trap smaller objects that might enter them. The law is general, but the relevant property of a specific situation -- say, in which the Earth, the Moon, and some tiny asteroid located at L4 maintain their gravitational dance -- is a valence.

It doesn't quite do to conceive of valence as a property of just the dominant component(s) of a situation -- say, of the carbon atom when four hydrogen atoms are nearby. To do so misses the mutuality of affinity these atoms have for bonding. The hydrogen atoms do attract the carbon atom, though to a lesser degree than it attracts them (if talking about such one-way attractions isn't completely nonsensical).

But our common-sense notions of causal agency lead us to make precisely this error. Natural languages tend to describe physical events using a subject-acts-on-object schema, and modern scientific language still reflects this thinking. Newton's choice of wording for his third law of motion thus invites a misunderstanding: what sounds like a sequence of two related events, an action and a reaction, is actually a single event -- an interaction. Armed largely with inappropriate transitive verbs to describe the physical events they study, it would seem that scientists come to understand the mutually interactive nature of those events despite the terms they use, not because of them.

One might wonder why 'disposition' shouldn't serve in lieu of 'valence' here. That term can apply to situations as well as to objects. But, for the psychological reasons just mentioned, to use 'disposition' for situations invites a chronic equivocation between disposition of object and disposition of situation. We are constantly placed in the position of needing to override the common-sense agent-object schemata with their Galilean analogs of interaction. Out of expediency, I strongly recommend abandoning 'disposition' to the realm of objects, and using 'valence' to deal with dispositions of situations.

To summarize, the actual dynamic interaction resulting from mutual affinity emerges from the situation in which the components are close enough for the forces among them to overcome their inertia or momentum, or to overbalance disruptive external forces. Unless our hypothetical carbon and hydrogen atoms are situated near one another, no bonding can occur. Furthermore, we naturally define situations according to the events they portend -- i.e., their valences for particular outcomes. Because of this, it makes sense to treat valence as a defining property of the situation.¹¹

My use of 'valence' instead of coining some new term might reasonably be criticized as a confusing expropriation, for I intend to introduce herewith another term that more closely matches the concept chemists express with 'valence': 'signature'. The signature of an object is the face it presents to the surrounding world, its complete interactive potential. This term is borrowed from the users of RADAR, who can identify objects by the RADAR signatures they present. The propensities of our carbon atom to bond with four hydrogen atoms, or with other carbon atoms in a diamond crystal, or to impart momentum to another object upon collision, are all parts of its signature.

Signature, then, though it deals with the properties of individual objects much as the traditional concept of valence does, is also a more general notion that goes beyond strictly chemical properties. Our practical concerns will not lead us to use 'signature' very often, for we will usually be interested in particular dispositions of an object to interact with certain things rather than its total interactive capacity. When we examine molecular bonds, we will typically forget about, say, the gravitational properties of atoms and their subatomic components, and focus on the electrochemical facet of each atom's signature. If we need a term for particular facets of a signature, we might specify, say, an atom's electrochemical initials (though I confess this might strain the metaphor, and 'disposition' works well enough here). But I think it is a valuable philosophical safeguard to dredge up signature from time to time as a reminder that we tend to ignore many minor aspects of interactive potential while discussing the salient ones -- even when some of those aspects are "engaged" and therefore either contribute to establishing or maintaining interaction between or among objects, or exert some countervailing pressure that, in concert with forces external to the system we are examining, may overcome the forces we had thought would dominate.

Valences can be strong or weak, and can be either "positive" (inclusive or attractive) or "negative" (exclusive or repulsive). Many of the valences we find particularly interesting are "positive" tendencies to form and maintain some form of organization. While it does make some sense to speak of "negative valences", such as the tendency of two protons to separate when placed close together, it is rarely useful -- for such situations self-destruct, and are therefore not the sort of situations whose persistence we might want to understand.

The objects of our experience, for example, exist as objects because there is a valence for their components to persist together as functional units (as opposed to some accidental collection of parts interacting independently of one another with their surroundings). Perhaps the best evidence in support of this claim is that we humans react to such material collectives as unitary objects. We recognize them, name them, use them, seek or avoid them -- not with respect to the distinct properties of their components, but with respect to their collective behaviors as units. If such collectives lacked the propensity to behave as a unit, and the parts thereof impinged upon us as distinct and separate from one another, we should expect our evolved perceptions to recognize the separateness rather than some non-functional unity.¹² Scattered objects and the like may be interesting philosophical conceptions, but they present no opportunities for organisms to evolve responses to them as objects.

Systems exist when there is a filled valence for objects to function persistently somehow in concert, even when the individual components are free to interact independently with their surroundings in other respects. This is just an explication of the standard definition of the term, so I will abstain from any detailed defense of it, and merely challenge any skeptical reader to cite a counterexample.

It would be difficult to overstate the ubiquity of valence. Each of the fundamental physical forces -- gravity, electromagnetism, and the strong and weak nuclear forces -- displays the mutual attraction or repulsion characteristic of valences, and one or more of them will underlie any physical situation with a valence for persistence: planetary orbits, chemical bonds, magnetic adhesion, atomic nuclei, etc. Let us now examine how valence governs the process of evolution and the nature of biological systems.

Why is valence an important new contribution to evolutionary theory? Because it changes the focus of attention from the organism to the situation -- from one element of the situation composed of organisms and their environment to the situation as a whole.

To illustrate, let's see how the notion of valence can illuminate the debate over whether the gradual introduction of genetic variation into a species over time (gradualism) or the periodic, sudden shifts in genetic composition of a species due to large-scale environmental changes (punctuated equilibrium) best accounts for the evolutionary record.

Punctuated equilibrium turns out to be a natural, unsurprising consequence of the nature of situations in which the environment remains largely stable, with brief dislocations resolving into new stability. We puzzle about the sudden changes in the fossil record that seem to deny the importance of steady variation in the genotype of a species, because it is the newly emerged traits of more recent fossils that we find interesting and important.

The only claim to "revolutionary" status for the theory of punctuated equilibrium rests on the supposition that significant variation in the character of the environment constitutes a force whose explanatory power concerning the origin of genotypic shifts in a species competes with (and defeats) the alternative explanation offered by the evolutionary force of genotypic variations produced gradually (by mutation, say) within the species.

This claim has a built-in appeal for anyone who wants to view the "creative" or "generative" nature of variation as an evolutionary force. But I think parsimony argues against the view. Variation, both within the organism and in the environment, has an apparently random quality that suggests it is less a force than a circumstance. Certainly both kinds of variation have their causes, and physical forces account for each instance of variation -- but should the variation itself be described as a force?

No. Punctuated equilibrium produces exactly the same kind of essential change that characterizes evolution on the gradualist view: a change in the relationship between organism and environment -- a change in valence. The apparent difference is due to the fact that the kind of variation in this case isn't the alteration of genetic material, but the adjustment of features of the environment. But environmental change does nothing creative. It doesn't cause Lamarckian genetic changes tailoring a species to its changed circumstances.

In fact, the only genotypic changes that can emerge as the result of environmental cataclysm are due to genetic variations that developed and remained, diluted in the population, over some indeterminate time preceding the event. Unless new traits compatible with the changed environment were already latent within the genotype, we would be talking about an extinction rather than an evolutionary change! (That is, of course, unless the species was able to tolerate -- if only barely -- its new circumstances; then any future genetic changes that were adaptive to the environment would spread through the population. But that's gradualism!)

A theory of valences resolves this dispute by pointing out that both camps are focused on just one component of an evolutionary situation. Gradualists see change in organisms as the driving force, while proponents of punctuated equilibrium see change in the environment as the driving force. Both miss the real point, which is that whenever the "fit" between organisms and their environment is damaged, whatever the source of change, there is a valence for an adjustment of the situation: either the organisms will adapt to their new circumstances (perhaps thanks to genes already in the population, or perhaps only when new genes emerge via normal, gradual, variation), or they will move to a new environment where they fit better, or they will die out.

That is how valence drives the evolutionary process.

2.2 Valence and Evolution

Adam Smith is widely regarded as one of the primary theoretical influences for Darwin. I believe this is because Smith studied valences of the marketplace while Darwin studied valences of biology. The analogy between market processes and evolutionary processes is built on the common focus on valences and how they operate. Malthus, too, influenced Darwin, with his study of how the arithmetic growth of food resources naturally limits the geometric growth of populations -- a study of competitive valences.

Evolution (as a biological process) is nothing more than the way certain valences play out in the world. There are physical valences leading directly to the emergence of biochemical changes (mutations, typically), which in turn affect the valences between an organism possessing a new mutation and its environment. (Does it tend to persist and reproduce better than it otherwise would have?)

That is the particular insight most useful to our Galilean project; but there is a more general one: nothing happens in the world that is not the result of some valence being realized. This claim is based on two presumptions: (a) that the causal order is a web of situations in which events are governed strictly by natural laws (that is, situations having valences for outcomes determined by natural laws); and (b) that nothing happens outside the causal order -- that there are no uncaused events in the universe.

This latter is a controversial claim. Probably most physicists believe that random events occur at the subatomic level, but I am not persuaded.¹³ If by chance I'm wrong, quantum randomness is not usually claimed to influence biological or behavioral events (except perhaps in unfathomably rare cases), and I will be satisfied with this modified claim: virtually nothing happens in the causal order that is not the result of some valence being realized, and biological and behavioral events are part of the causal order.

For my purposes it will be sufficient to show that valence accounts for the evolution of biological and cultural structures (I hope the biological claim is now clear in principle, if not in detail; more on culture is immediately forthcoming), for the emergence of a natural property we can and should call 'value' (one I've explained considerably, but have more to say about), and for a metaethical functionalism that fits the Galilean scheme of theoretical reduction better than does Gibbard's norm-expressivism.

Having already discussed how valence drives the evolutionary process, I'll just offer a few examples of the countless valences to be found in biology. The African veldt has a valence for grazing animals (thanks to vast expanses of vegetation) and for predators (thanks to the presence of grazing animals). There is a valence for nectar-consuming creatures to play a role in pollinating the plant species they exploit, for doing so increases their food supply. There was a valence for early animals to develop an immune system, which lengthened their reproductive lives. The immune system itself works thanks to the chemical valences between antigens and antibodies.

Of most interest to us will be how valence governs the behavior of biological systems that are characteristically human -- economic, political, social, and cultural systems.

2.3 Valence and Human Systems

Consider a simple economic situation -- one that finds you in a department store. There will be valences of varying strengths for you to purchase certain items -- weak valences where the items are only moderately appealing or are a bit overpriced, and stronger valences when the items have a high perceived utility to price ratio. If you are in the market for a television set, you might fill the valence by purchasing the one unit that best matches features, quality, and price to your needs.

In the political realm, perhaps you voted in the last election (as people increasingly seem to do) for the "lesser of two evils" -- the candidate who least offended your sense of what constitutes proper governance. Though there was only a weak valence for casting your vote for one candidate, the valence for supporting the opponent was even weaker. Because your choice was limited to the two candidates, you filled the stronger of the two valences, however weak it may have been, by voting for that candidate.

As an example of social valence, let's look at the paradigm case of game theory, the Prisoner's Dilemma.¹⁴ Two prisoners must independently choose between two courses of action: cooperate with one another by refusing to confess to their joint crime, or defect and confess in hopes of obtaining the best personal outcome. They are presumed to be wholly self-interested and unconcerned with the other's welfare, and to be ideally rational -- capable of reasoning what their optimum choice will be. The payoffs for each course of action are arranged so as to make confessing the "dominant" choice for anyone motivated by rational self-interest and concerned only with the present situation. If only one prisoner confesses, he gets off scott-free, while his accomplice is sentenced to 12 years. If both confess, they each get 10 years. If neither confesses, both will serve two years. An ideally rational prisoner will recognize (it is claimed) that he will serve less time if he confesses, regardless of the choice his accomplice makes.

There is clearly a stong valence here for each prisoner to confess. This valence is predicated upon the immediate desire of each to minimize the damages he incurs. But, by expanding the scope of the situation beyond the immediate to include the possiblity that the prisoners may again face the same choice in the future (perhaps many times), the valence for cooperation becomes stronger as each realizes that early cooperation improves the likelihood their accomplice will cooperate, too, in subsequent interactions. So, if the prisoners are aware of the potential for repeated iterations of their problem, there is a stronger valence to cooperate.

There is another possibility, however. What if the prisoners are not ideally rational, but -- being human -- are often guided by nonrational (unreasoned) motivations? What if both have inherited a natural tendency to cooperate, regardless of the particular circumstances?

It turns out that such a shared tendency, if acted upon, minimizes the aggregate liability for all concerned. It is also helpful in other less contrived situations, especially when the other "players" are very likely close kin in whose reproductive success one has a biological interest. Most importantly, this is precisely the sort of tendency that can be inherited and, if valuable (as it clearly would be for the group), would spread due to the adaptive benefits it confers upon its owners. There is a strong valence, then, for populations to develop cooperative tendencies.

There is a final possibility that provides some insight into the sort of ethics I think is supported by this investigation. What if the participants are ideally rational, but have unusual interests -- say, they are less interested in narrow personal benefits than in long-term benefit for humankind? Maybe they have even adopted Life's telos as their own primary value, as I am recommending!

One might deny this is possible, claiming rationality dictates pursuing narrow self-interest over other concerns; but that begs the question as to whether narrow self-interest should dominate. My response is that actual self-interest may well be identical to other broader concerns, and that what typically passes for self-interest is instead myopic and potentially self-defeating. There is a strong valence for humans to cooperate as a matter of rational choice, despite apparent short-term losses.

Gibbard discusses how the bargaining situations studied in game theory have analogs in the evolutionary process (65-68), which he calls evolutionary bargaining situations. When inclusive fitness rather than personal preference is the governing goal, and a type of organism is subjected to recurring bargaining situations, it may evolve an evolutionarily stable strategy -- one in which each organism is genetically inclined to pursue a mutually fitness-enhancing combination of behaviors (a cooperative choice), holding in reserve a threatened combination of behaviors for response to those who are uncooperative in their interactions. An evolutionarily stable strategy derives its stability from remaining more fitness-enhancing for each individual than alternative strategies that might be introduced by mutation. To use the vocabulary I have introduced here, there is a stronger valence for an evolutionarily stable strategy than for any competing mutual fitness-enhancing strategy.

In the repeated prisoner's dilemma, full cooperation "is simple and symmetric, and it produces by far the largest joint gains," says Gibbard (66, note 12). But his real interest is in evolutionary bargaining situations in which "there is a genuine problem of how to divide the gains of cooperation." When there are multiple mutual fitness-enhancing combinations available, some options will provide greater advantages for specific individuals than for others. Gibbard expects such disparities to be ironed out among humans through normative discussion. But, because the spectrum of roles in these situations is limited, and individuals fit along that spectrum in groupings that can (in principle) be statistically determined, even rather complex problems of benefit sharing might find evolutionarily stable game-theoretical solutions.

Indeed, the sharing of food among apes may be an example. When a member of a chimpanzee tribe finds food or kills a small animal, the food is not divided according to rank within the tribe, as one might expect. Instead, the hunter doles out morsels to the rest, giving preference to those with whom he seeks to strengthen social bonds. More than half of the food exchanges among chimpanzees are giving rather than taking, and the vast majority (all but 2.6%) of takings are nonviolent -- typically the cautious and solicitous taking of small amounts from the owner's larger holdings.¹⁵

This strikes me as revealing a valence for property rights that is more apparent in humans, known to us (thanks to Adam Smith) as the invisible hand.¹⁶ Individuals who reap the benefits of their productivity not only have the opportunity to share with others in response to social or economic valences, but have renewed incentive to be productive in the future. Furthermore, the benefits they receive this way will often prepare them to be more productive, just as strength from one's food can aid in the next hunt. When such enhanced productivity surpasses the individual's capacity for personal consumption for many within a society, there is a valence for economic exchange which naturally leads toward everyone enjoying a greater quantity and variety of consumption.

This tendency among apes is probably not an inherited response to such situations, an innate respect for property rights. I think it is more likely a "forced move" in the individual psychological development of members of primate tribes. Given a valence for possessiveness (an uncontroversial aspect of narrow self-interest), infants and juveniles are faced with two solid facts: others like to keep whatever they have of value, and youngsters lack the power to take what they want from their seniors -- and face recriminations when they try. Before they grow large enough to break the rules, they have developed a deeply ingrained schematic norm: If it belongs to somebody else, don't try to take it. The best evidence that this norm persists through adulthood is that the aggressive taking of food occurs only 0.5% of the time.

Juveniles instead learn safe and harmonious ways to induce sharing by others: begging; reaching for small amounts (subject to the owner's veto); previous sharing of one's own bounty; and cultivating favor through grooming, sexual access, or political support. Generous individuals are rewarded in kind by their fellows, while the stingy often meet with rejection. Voluntary reciprocity, then, is the strategy the higher primates have adopted to bring about the important sharing of resources within the group.

The fact that learned tendencies such as voluntary reciprocity and respect for property rights can emerge predictably -- even automatically -- in primates demonstrates that the division between purely inherited dispositions and cultural dispositions can be subtle. Let me illustrate this with another example.

Trial-and-error is the method by which infants automatically refine their perceptual and motor skills. They select and repeat certain behaviors in preference to others in reaction to a generalized valence for apt responses to the environment. But this same valence governs our most sophisticated cultural development: science. We make theoretical predictions about how the environment will behave in well-defined situations, then test those predictions empirically against the actual responses of the environment. If the predictions are apt, we attempt to devise more demanding tests of our theories; if not, we revise our theories to conform to the results, and test the revised versions in their turn.

Aside from the more refined and systematic choice of behaviors to test, the selective process for science is the same kind as that for an infant learning to find and grasp objects. It is also the same kind of process by which variant organisms are tested against the environment by natural selection, or by which new products are tested in the marketplace. This is no coincidence, but merely an expression of this overreaching fact:

There is a universal algorithm for selection that is realized in countless ways in the world. Variants of some kind of thing are tested against the relevant environment, and tend to persist or spread whenever there is a stronger valence for them than for the rest of their kind.

With the last several examples above, we are clearly encroaching upon the realm of cultural valences -- those which are more likely to be filled by behavior in conformity with learned traditions than by innate tendencies triggered by the situation.

Let us consider one more -- the uncomfortable example of behaviors such as those exposed in Stanley Milgram's notorious psychological experiment in which subjects were ostensibly aiding researchers by administering negative reinforcement to test subjects (actually confederates) in a purported study of learning incentives.¹⁷ Subjects typically were willing to administer electrical shocks to the confederates, when instructed by the experimenter, far beyond the point at which severe pain was evident, and often beyond the point where a serious risk of death was apparent.

The driving motivation for such behavior is probably simple deference to authority -- something deeply ingrained in humans thanks to our evolutionary history of hierarchical primate social structures. What makes this case a useful example for us here is that the authority of the experimenter differs markedly from the authority of the alpha male in a primate tribe. It is based on certain defining elements of the particular situation, rather than upon habitual deference to a higher ranking individual of intimate acquaintance.

Once humans came to specialize in specific jobs rather than pursue the species-typical roles for their sex and age, there emerged a valence for specialized authority. Farmers would naturally be deferred to in matters of agriculture; doctors in matters of medicine; and psychologists concerning their experiments. This valence has been filled by a strong cultural motivation for individuals to defer to experts within their domains of expertise.

A general tendency to obey authority, then, coupled with a modern deference to specialized authority, explains why subjects were inclined to accept the commands of the experimenter over their own personal judgments.

This discussion of valence is intended to explain how organisms relate to their physical or social environments, and to make a sort of fundamental sense of how objects relate causally to one another in the first place. But there remains another large issue: how organisms differ from other things in the universe, and what that difference means for this project.

2.4 The Metaphysics of Physical Alternatives

The Existential Alternative

"There is only one fundamental alternative in the universe: existence or nonexistence -- and it pertains to a single class of entities: to living organisms."¹⁸ The first part of Ayn Rand's claim here is merely an application of the laws of noncontradiction and excluded middle: something cannot both exist and not exist, and there is no third alternative. If we accept that anything at all exists, we must accept that the only metaphysical alternative to its existence is that it not exist.

The second part of her claim is literally false, but Rand's meaning is not well expressed by it. A square formed by four rods of equal length can cease to exist; if separated, the rods remain in existence, but the square is gone. Similarly, a life can cease to exist, leaving behind all the matter and energy that had composed it. The square is a purely geometrical arrangement, whereas a life is a functional one. That's not the only important difference, though. Atoms, too, are functional arrangements, but where the persistence of atoms is concerned, nothing of importance is at stake; their components readily reform into other atoms at the first (frequently available) opportunity. Atoms are fungible commodities. But with a life, something of importance certainly is at stake, for a dismantled life can't be reconstituted by any likely physical process.

Our Galilean laws of conservation hold that matter and energy are never destroyed; their form changes. There is some challenge to laws of conservation from quantum theory,¹⁹ but (as with the concerns about randomness just discussed) purported exceptions are controversial, and apply almost exclusively to the behavior of subatomic particles. We can accept that, for material objects in the universe, their form alone faces what we may as well call the existential alternative.

Rand was defending the basis of her value theory, the claim that 'value' makes sense only with respect to the interests of a living organism -- a being with things to gain or to lose, with interests at stake. It is from this claim that I refined the view I am defending here.

Though Rand is not particularly clear on this point, I think it is both charitable and fair to say that, for her, life is at once the source of value and the ultimate value for this reason: when a life is lost, what remains is a different kind of thing -- and the kind of thing that life is differs from alternative kinds in that it is an end in itself. She even makes the point that this end is "a value gained and kept by a constant process of action," and that life is "a process of self-sustaining and self-generated action."²⁰ Though she doesn't speak of kinds, or of function, she might have agreed that life is a functional kind: it exists as a function, and it functions to exist. Put in perhaps the most economical terms, life has a reflexive functionality. Anything that has this property would be life, or would be of the same kind as life.

The concept of the existential alternative provides a foundation for a naturalized, functionalist metaethics. When existence is at stake for a kind of thing whose distinguishing property is that it works to maintain its existence, what is good or bad for that kind of thing seems reasonably measured by how it either promotes or impedes the function of existing as that kind of thing.

I have been examining this issue in more detail than Rand attempted, imputing a second, more fundamental function of life -- being part of Capital-L Life -- which is also an end in itself, yet has a better claim to being the ultimate value for a living organism. For Life also faces the existential alternative, and I think all other concerns, existing only because it does, fade to insignificance by comparison.

The Temporal Alternative

In addition to the existential alternative, there is another physical alternative that I will call the temporal alternative. As time passes, a state of affairs will either remain the same or will change. Change, from a long-term perspective, is the normal course of events, and it is resisted only by some strongly established structure. As far as we have been able to determine, nothing is absolutely static. Even protons are thought to decay in the fullness of time. The meaningful alternative to change, I think, is stability, continuity, or persistence -- a temporary condition of negligible (or no) change over some defined time period.

There are two modes of change quite familiar to philosophers: qualitative and quantitative change. Change is measured either by kind or by degree. For a living organism, change entails either being born with (or developing) new properties (mutations or alterations) or being born with (or developing) an enhancement or diminution of a property its kind normally has.

Both qualitative and quantitative change exhibit what we might call positive or negative 'direction'. We are naturally concerned with the positive changes that might emerge to redefine an organism's success, so we naturally focus on excellence as the positive quantitative pole, and diversity as the positive qualitative pole, for the changes a kind of organism might undergo. To improve, an organism must do something better than, or different from, its forebears. More abstractly, it must either fill a new valence (perhaps only weakly) or fill an existing valence with a stronger "fit" than others of its kind.

There are also two mixed modes of improvement. If an organism is good at exploiting a variety of situations, we say it is versatile. Versatility is diversity of excellence. If an organism can change to meet a variety of environmental circumstances, we say it is adaptable. Adaptability is excellence at diversity.

Let's consider now some relations among valence, continuity, and the modes of change for living things. The fundamental valences for biological continuity are the persistence of individual lives and the ongoing process, for all Life, of reproduction -- generation after generation. There is also a valence for kinds of organisms to resist change -- once they fill a valence in the environment (the sort we often call a 'niche') so strongly that any change they might undergo would either render them less fit or fail to spread through the population for lack of any advantage it could confer. Sharks and cockroaches are standard examples of change-resistant creatures.

A valence for quantitative change is embedded in situations in which a kind of organism weakly fills a niche in the environment, for the appropriate change will strengthen the way the niche is filled. A valence for qualitative change exists when some biologically possible change (or series of changes, provided no intermediate stage seriously diminishes the organisms' fitness) permits a kind of organism to fill a new niche more strongly than it fills its current one (or to fill an additional niche that is beneficial to it). For example, a change in a moth's coloration may prove to be protective in a new food-rich environment where its former coloration would attract predators.

One might think there would be a valence for versatility to emerge in any kind of organism, but that isn't quite true. Organisms that depend upon individual skill and strength for success will usually benefit from newly developed versatility, provided the change doesn't require passing through a stage of seriously reduced fitness. But some creatures -- most notably the social insects -- rely on group strength for their success. For such creatures there is a valence for the individual organisms to be specialized rather than versatile.

One variety of adaptability applies to populations rather than to individuals: numerous and rapidly reproducing organisms such as bacteria, which collectively undergo frequent mutations, have the capacity to adapt in a few generations to a radically changed environment. But this is really nothing more than the process of evolution itself in action. Adaptability as a trait of individuals requires a plasticity of functionality that will exist only in rather complex creatures. Simple creatures, for whom the only possible new behaviors must be just as hard-wired as their current repertoire, are not candidates to develop adaptability. Once a species develops flexible behavior responsive to changes in the environment, however, adaptability will tend to spread once it emerges -- for it allows organisms to exploit new valences to which they haven't been specifically adapted by evolution.

The reason that humans appear to us to be the apex of the evolutionary process is that intelligence is the most potent source of both versitility and adaptability we know of; it wouldn't be surprising to discover that it's the most potent source of these goods that is possible for organisms to acquire. Clearly, the benefits it has conferred upon humankind are dramatic.