Академический Документы
Профессиональный Документы
Культура Документы
1 Introduction
In [1] we characterized emergent phenomena as phenomena that may be described inde-
pendently of their implementations.1 We distinguished between static emergence (emer-
gence that is implemented by energy wells) and dynamic emergence (emergence that is
implemented by energy flows). We argued that emergence (of both forms) produces ob-
jectively real phenomena (because they are distinguishable by their entropy and mass
characteristics) but that interaction among emergent phenomena is epiphenomenal and
can always be reduced to the fundamental forces of physics. Our focus in that paper was
on the phenomenon of emergence itself. In this paper we explore the entities that arise as
a consequence of the two types of emergence, focusing especially on dynamic emer-
gence.
2 Material Entities
As human beings we seem naturally to think in terms of entities—things or objects. Yet
the question of how one might characterize what should and should not be considered an
entity remains philosophically unresolved. (See, for example, [Boyd], [Laylock],
[Miller], [Rosen], [Varzi Fall ‘04].) We propose to define a material entity as any instance
of emergence. What is fundamental to material entities is that one can identify the force
or forces of nature that bind them together and that cause them to persist in a form that al-
lows one to distinguish them from their environments—because of their distinguishable
entropy and mass properties.
Some material entities (such as an atom, a molecule, a pencil, a table, a solar system, a
galaxy) are all instances of static emergence. These entities persist because they exist in
energy wells. Biological entities (such as you and I) and social entities (such as a social
club, a corporation, or a country) are instances of dynamic emergence. These entities per-
sist as a result of energy flows.
1
In [1] we credited Anderson with being one of the first prominent physicists to argue that new laws of
nature, i.e., laws not derivable from physics, exist at various levels of complexity. While re-reading [Schrodinger]
we found the following. “[L]iving matter, while not eluding the 'laws of physics' … is likely to involve 'other laws
of physics,' hitherto unknown, which … will form just as integral a part of [the] science [of living matter] as the
former.” As we pointed out in the earlier paper, there are indeed new laws (such as the theory of computability),
which, while consistent with the laws of physics are not reducible to them.
On the other hand, what might be considered conceptual (or Platonic) entities—such as
numbers, mathematical sets (and other mathematical constructs), properties, relations,
propositions, categories named by common nouns (such as the category of cats, but not
individual cats), and ideas in general—are not (as far as we know) instances of emer-
gence.2 Nor are intellectual products such as poems and novels, scientific papers, or com-
puter programs (when considered as texts). Time instances (e.g., midnight December 31,
1999), durations (e.g., a minute), and segments (e.g., the 20th century) are also not in-
stances of emergence. Neither are the comparable constructs with respect to space and
distance. Since by definition every material entity is an instance of emergence, all materi-
al entities consist of matter and energy arranged to implement some independently de-
scribable abstraction. Since none of the preceding conceptual entities involve matter or
energy, none of them satisfy our definition of a material entity.
2
We simply do not understand how ideas as subjective experience come into being. When we learn how
subjective experience is connected to the brain, we may find that ideas (or at least their physical realizations) are in
fact instances of emergence and that one can identify the forces that hold ideas together. For now, though, we can’t
say that an idea as such is an instance of emergence since we don’t know how ideas are implemented physically.
Even if we did know how ideas are implemented in the mind, the physical instantiation of an idea would still not be
the same thing as the referent of the idea. (My thinking of the number 2 is not the number 2—assuming there is
such a thing as the number 2 as an abstraction.) So we maintain the position that concepts as such are not material
entities, and we do not discuss them further.
sum of the masses of the protons and neutrons that make it up when considered separ-
ately.3 The missing mass is released as energy.
The same entity-mass relationship holds for all static entities. An atom or molecule has
less mass (by a negligible but real amount) than the sum of the masses of its components
taken separately. The solar system has less mass (by a negligible but real amount) than
the mass of the sun and the planets taken separately. Thus the entropy of these entities is
lower than the entropy of the components as an unorganized collection. In other words, a
static entity is distinguishable by the fact that it has lower mass and lower entropy than its
components taken separately. Static entities exist in what is often called an energy well;
they require energy to pull their components apart. Static entities are also at an energy
equilibrium.
Manufactured or constructed artifacts also exhibit static emergence. The binding force
that holds manufactured static entities together is typically the electromagnetic force,
which we exploit when we use nails, glue, screws, etc. to bind static entities together into
new static entities. A house, for example, has the statically emergent property number-of-
bedrooms, which is a property of (a way of describing) the house from a perspective that
sees it as an entity. A house implements the property of having a certain number of bed-
rooms by the way in which it is constructed from its components.
A static entity consists of a fixed collection of components over which it supervenes. By
specifying the states and conditions of its components, one fixes the properties of the en-
tity.
But static entities that undergo repair and maintenance, such as houses, no longer consist
of a fixed collection of component elements thereby raising the question of whether such
entities really do supervene over their components. We resolve this issue when we discuss
Theseus’ ship.
3
It turns out that iron nuclei “lack” the most mass. Energy from fusion is possible for elements lighter than
iron; energy from fission is possible for elements heavier than iron.
4
Another common example of a dissipative structure is the Belousov-Zhabotinsky (BZ) reaction, which in
some ways is a chemical clock. We design digital clocks to tell time. We didn’t design BZ reactions to tell time. Yet
in some sense they both do. That one surprises us and the other doesn’t shouldn’t mislead us into putting them into
different categories of phenomena.
5
In all our examples, the form in which energy is delivered also matters. An electric current will produce
different effects from a thermal energy source when introduced into a digital clock and a Rayleigh-Bénard device.
astrously. When energy is flowing through it, a dissipative entity is by definition far from
equilibrium. So a dissipative entity is a static entity that is maintained in a far-from-equi-
librium state.
The sorts of dissipative structures we have been discussing are not fully qualified dynam-
ic entities, however, because they do not include mechanisms to repair their static struc-
tures. Their static structures are maintained by other forces than those produced by the
energy that flows through them. In particular, dissipative structures do not cycle material
through themselves as all fully qualified dynamic entities do. As we will see below, one
consequences of this fact is that dynamic entities do not easily supervene over their ma-
terial components. Static entities and dissipative structures do.
2.5 Petty reductionism fails for dynamic entities—for all practical pur-
poses
Petty reductionism is another way of saying that an entity supervenes over the matter of
which it is composed: fixing the properties of the matter of which an entity is composed
fixes the properties of the entity.7 Hurricanes illustrate a difficulty with supervenience and
petty reductionism for dynamic entities.
The problem is that from moment to moment new matter is incorporated into a hurricane
and matter then in a hurricane leaves it. Define as what we might call the hurricane’s su-
pervenience base the smallest collection of matter over which a hurricane supervenes.
Since matter cycles continually through a hurricane, a hurricane’s supervenience base
6
A characterization of a hurricane as a vertical heat engine may be found in Wikipedia. (URL as of
9/1/2005: http://en.wikipedia.org/wiki/Hurricane.) The preceding hurricane description was paraphrased from
NASA, “Hurricanes: The Greatest Storms on Earth,” (URL as of 3/2005
http://earthobservatory.nasa.gov/Library/Hurricanes/.)
7
Recall that a set of higher level predicates is said to supervene over a set of lower level predicates if a
configuration of truth values for the lower level predicates determines the truth values for the higher level predic-
ates. We are using the term supervene loosely to say that an entity supervenes over its components.
consists of the entire collection of matter that is part of a hurricane over its lifetime. It
would seem that a hurricane’s supervenience base must be significantly larger than the
amount of matter that constitutes a hurricane at any moment. Because a hurricane’s su-
pervenience base is so much larger than the matter that makes it up at any moment the
fact that a hurricane supervene over its supervenience base is not very useful. Other than
tracking all the matter in a hurricane’s supervenience base, there is no easy reducibility
equation that maps the properties of a hurricane’s supervenience base onto properties of
the hurricane itself.
Furthermore, the longer a hurricane persists, the larger its supervenience base—even if
the hurricane itself maintains approximately the same size during its lifetime. Much of
the matter in a hurricane’s supervenience base is likely also to be included in the super-
venience bases of other hurricanes. Like Weinberg’s example of quarks being composed
(at least momentarily) of protons, hurricanes are at least partially composed of each other.
Thus just as Weinberg gave up on the usefulness of petty reductionism in particle physics,
we must also give up on the usefulness of petty reductionism and supervenience for dy-
namic entities as well.
• They have reduced entropy (greater order) than their components would have on their
own.
• They depend on external sources of energy to stay in existence. Because of the energy
flowing through them, the have more mass than their components would on their
own.
• The material that makes them up changes with time. Their supervenience bases are
generally much larger than the material of which they are composed at any moment.
The longer a dynamic entity persists, the greater the difference. Petty reductionism
either fails entirely, or it becomes a historical narrative. One can tell the story of a
country, for example, as a history that depends in part on who its citizens are at vari-
ous times. One would have a difficult time constructing an equation that maps a coun-
try’s supervenience base (which includes its citizens over all time) to its state at any
moment unless that mapping were in effect a historical record.
• Most biological and social entities have other dynamic entities as components. These
component entities have “divided loyalties” in some sense—to themselves and to oth-
er dynamic entities of which they are also components.
Even though dynamic entities persist in time, and even though the properties of dynamic
entities are a function of the properties of their components at any moment, since the
components of which a dynamic entity is composed change from time to time, there is no
direct way to map the properties of the components a dynamic entity will have over its
lifetime to the moment-to-moment properties of the entity itself except as a narrative, i.e.,
a story which describes which elements happen to become incorporated into the dynamic
entity at various moments during its lifetime.
All entities are subject to the effect of interactions with elements they encounter in their
environments. Dynamic entities are doubly vulnerable. They are also subject to having
their components replaced by other components. To persist they must have defenses
against infiltration by elements which once incorporated into their internal mechanisms
may lead to their weakening or destruction. Social entities are more vulnerable still.
Some of their components (people) are simultaneously components of other social entit-
ies—often resulting it divided loyalties.
painted since by supervenience, its properties are fixed by the properties of its compon-
ents, and its components are different from what they were before.
This cycling of material through an entity wasn’t a problem when we were discussing
hurricanes or social or biological entities—we had already given up on the usefulness of
petty reductionism and supervenience for dynamic entities. In those cases we thought of
the entity as including not only its momentary physical components but as also including
the energy that was flowing through it along with means to slough off old material and to
incorporate new material into its structure.
To apply the same perspective to Theseus’ ship, think of the physical ship along with the
maintenance process as a social entity—call it the Theseus ship maintenance entity. That
social entity, like all social entities, is powered by an external energy source. (Since the
maintenance of Theseus’ ship is a governmental or societal function, the energy source is
either voluntary, conscripted, or taxation.) The Theseus ship maintenance entity uses en-
ergy from its energy source to do the maintenance work on the ship. Just as the material
that makes up a hurricane changes from time to time and the people who are employed by
a business change from time to time, the physical ship also changes from time to time.
But like a hurricane and a company, the ship maintenance entity persists over time.
8
Many agent-based and artificial life models acknowledge the importance of energy by imposing an artifi-
cial price for persistence, but we are not aware of any in which the cost of persistence is fully integrated into the
functioning of the entity.
Like interaction among static entities interactions among dynamic entities also depends
on the entities. Two obvious examples are interactions among people. Sex between biolo-
gical organisms (including people) is successful only when the participants are the appro-
priate organisms. This is the case at all levels—ranging from the physical (in which vari-
ous physical parts must fit together) to the cell level at which the gamete and the egg
must combine properly. Sex would not occur if only the components—e.g., organs or mo-
lecules, etc.—of the sexual partners were left along in a darkened room no matter how
sweetly the romantic music was playing.
Similarly symbolic interaction works only between entities that are capable of operating
at the symbolic level. Science is not in a position to explain how people operate at a con-
ceptual/symbolic level. Yet it is clear that we do. It is also clear that as symbolic entities
we interact with each other symbolically through language. We each are capable (to some
reasonably good approximation) of communication to each other what ideas are occur-
ring in our consciousness.
that what we had postulated as a type of entity was not, perhaps because we found that
different instances were implemented completely differently.
Once this explanatory task is accomplished, the reductionist tradition has been to put
aside an entity’s functional/phenomenological description and replace it with (that is to
reduce it to) the explanation of how that functionality/phenomenology is brought about.
After all, once one can explain in terms of lower-level mechanisms how some behavior or
appearance comes about one can presumably reproduce that behavior or appearance by
means of those lower-level mechanisms. Of course one then has the task of explaining the
lower-level mechanisms in terms of still lower-level mechanisms, etc. But that’s what sci-
ence is about, peeling nature’s onion until her fundamental mechanisms are revealed.
In the first paper, we argued that contrary to the reductionist tradition, an understanding
of how functionality is brought about does not eliminate the significance of that function-
ality. Our example was the implementation of a Turing Machine on a Game of Life plat-
form. A reductive analysis of a Game-of-Life Turing Machine may help us understand
how the Turning Machine is implemented, but it doesn’t help us understand the function-
ality that the Turing Machine provides.
To illustrate this in terms of natural science, suppose that what will eventually be determ-
ined to be Game of Life Turing Machines (somehow) occurred in nature.11 We will as-
sume that when these creatures are first discovered in the wild, the scientific community
doesn’t know how they function. After we observe them for a while, we find out that they
are capable of transforming inputs into outputs in reliable ways. Specialties within the
fields of biology, zoology, and ecology develop to study these creatures. Scientists study
the functions they compute, how those functions relate to each other, and how those func-
tions relate to the environment and the ecological niches within which the creatures are
known to exist. In fairly short order, it is determined that these creatures compute exactly
the Turing computable functions. They are thereafter known as bioTMs.
The bioTM biologists and microbiologists study how bioTMs function internally. They
examine the bodies of dead bioTMs, and using the newest micro-surgical instruments,
they study the internal processes of live bioTMs. What they find are strange patterns that
pulse through the bodies of bioTMs.
Let’s also suppose that there is a separate science that studies Game of Life patterns. Sci-
entists in that field build catalogs of patterns and study how those patterns interact with
each other. Some interactions create new patterns. Others destroy patterns. Etc. These in-
teractions are called pattern reactions. This discipline is comparable to what we know in
our world as chemistry.
It was soon discovered that bioTMs are built upon Game of Life patterns. A new field of
bioTM biochemistry develops. Scientists in this field study those Game of Life patterns
that are found to be important for bioTMs. In a major breakthrough, scientists discover
that all bioTMs are universal and that each has an encoding of its operation as quintuples
11
As a conceptual construct a Turing Machine is not an energy-based entity. There are no forces—either
static or dynamic—that hold Turing Machines together. Furthermore, Turing Machines have no means of reproduc-
tion; so it is even more unrealistic to suppose that they simply “occur” in nature. But as an example, this nicely il-
lustrates our point about reductionism.
—which themselves are encoded as Game of Life patterns. This is comparable to our dis-
covery of DNA.
Once the biologists understand that Game of Life Turing Machines are all universal, they,
with the help of some of their mathematical colleagues are able to work out Computabil-
ity theory, and they are able to explain many of the features of Turing Machines that they
had previously only just been cataloged.
Let’s now suppose that there is yet another science which investigates the fundamental
nature of the Game of Life. These are the Game of Life physicists. Of course, since the
basic rules are not all that difficult, Game of Life physics soon reached a dead end. Once
the Game of Life physicists had discovered the Game of Life rules, the fundamental rules
underlying how nature works. There was no more to find out.
But once the Game of Life physicists work out the Game of Life theory of everything, the
Game of Life chemists were able to work out how Game of Life patterns come about as a
result of the Game of Life rules, and the Game Turing Machine biochemists were able to
figure out how Turing Machines functionality may be explained strictly in terms of Game
of Life rules.
Now that we know everything there is to know about Game of Life Turing Machines,
does it make sense to discard our understanding of Turing Machines as transducers that
transform input to output according to computability theory? The strict reductionists
claim that it does. After all, once one knows the fundamental facts about the Game of
Life, everything else is just a matter of historical accidents. Given an initial configuration
of what the chemists call patterns, it is only because of the Game of Life rules that the
Game of Life pattern reactions occur. Similarly, it is only because of an extremely un-
likely sequence of circumstances that Game of Life patterns happen to be configured in
such a way that Game of Life Turing Machines come into existence. But given that un-
likely historical event, the entire operation of the resulting Game of Life Turing Machines
is completely explained by the Game of Life rules. Because of the work of the Game of
Life physicists and chemists we now know that Game of Life Turing Machines are noth-
ing but Game of Life cells going on and off according to the Game of Life rules based on
the historical accident of an unusual initial condition.
What’s wrong with this point of view? This perspective throws away everything the
Game of Life biologists, mathematicians, and ecologists learned about Game of Life Tur-
ing machines. The functionality of Turing Machines as transducers is important on its
own. We still want to know which functions are computable, how those functions interact
with other functions, how they interact with the environment in which they were found,
etc. These are independent facts about functions computed by Game of Life Turning Ma-
chines that cannot be deduced from either a study of Game of Life patterns or Game of
Life rules.
We use the term the reductionist blind spot to refer to the doctrine that once one under-
stands how higher level functionality can be implemented by lower levels of functional-
ity, the higher level is nothing more than a derivable consequence of the lower level.
Significantly, the reductionist tradition does not dismiss all descriptions given in terms of
functionality. After all, what does reductionism do when it reaches “the bottom,” when
nature’s onion is completely peeled? One version of the current “bottom” is the standard
model of particle physics, which consists of various classes of particles and the four fun-
damental forces. This bottom level is necessarily described functionally. It can’t be de-
scribed in terms of implementing mechanisms—or it wouldn’t be the bottom level. The
reductionist perspective reduces all higher level functionality to primitive forces plus
mass and extension. This is not in dispute. As we said in [1], all higher level functionality
is indeed epiphenomenal with respect to the primitive forces.
The difficulty arises because functionality must be described in terms of the interaction of
an entity with its environment. The fundamental forces, for example, are described in
terms of fields that extend beyond the entity. This is quite a different form of descriptions
from a structural and operational description, which is always given in terms of compon-
ent elements. When higher levels of functionality are described, we tend to ignore the fact
that those descriptions are also given in terms of a relationship to an environment. What
the reductionist blind spot fails to see is that when we replace a description of how an en-
tity interacts with its environment with a description of how an entity operates, we lose
track of how the entity interacts with its environment. The functionality of a Turing Ma-
chine is defined with respect to its tape, which is its environment.
This is particularly easy to see with (traditional) Turing Machines when formulated in
terms that distinguish the machine itself from its environment. The functionality of a Tur-
ing machine, the function which it computes, is defined as its transformation of an input,
which it finds in its environment, into an output, which it leaves in its environment.
What other formulation is possible? If there were no environment how would the input be
provided and the output retrieved? It is not relevant whether or not the computational tape
is considered part of the Turing Machine or part of the environment. All that matters is
that the input is initially found in the environment and the output is returned to the envir-
onment. A Turing Machine computes a function after all.
The same story holds for energy-based entities. Higher levels of functionality, the interac-
tion of the entity with its environment, are important on their own. An entity’s higher
level functionality is more than just the internal mechanism that brings it about. As higher
and more sophisticated levels of functionality are created—or found in nature—it is im-
portant to answer questions such as: how are these higher levels of functionality used and
how do they interact with each other and with their environment? Answering these ques-
tions fills in the reductionist blind spot.
The importance of describing entities from both perspectives was captured nicely by Eric
Jakobsson12 when he characterized biology as being “concerned equally with mechanism
and function.” Below we extend Jakobsson’s perspective beyond biology, but we agree
with his insight that mechanism and function are both significant.
The two most important questions to be answered about higher levels of functionality are
(a) how do they come about and (b) how do they persist? Entities that are a result of static
emergence come about as a result of clumps of matter finding their way into energy
wells. They persist because energy is required to pull them out of their energy wells. En-
tities that are a result of dynamic emergence present a much more complex story—the
12
At the Understanding Complex Systems Symposium, University of Illinois, Champagne-Urbana, Ill, May
2006.
short version of which is that dynamic entities persists as long as their functionality en-
able them to acquire from the environment the energy then need to maintain themselves.
The theory of evolution answers the questions about how new functionality comes into
existence and how it persists with respect to biological functionality. We explore these
questions with respect to dynamic entities more generally in the rest of this paper.
The whole plus its environment is more than the sum of the parts plus their environment.
is that refers to macro-level properties which arise from micro-level elements but are not
reducible to them. construct has a property that its component elements don’t have.
Similarly, the functionality of any entity is defined with respect to its environment. As we
will see later, the interaction of an entity with its environment is particularly important for
dynamic entities because dynamic entities depend on their environment for the energy
that enables them to persist.
More generally, consider the following from Weinberg.
Grand reductionism is … the view that all of nature is the way it is
(with certain qualifications about initial conditions and historical acci-
dents) because of simple universal laws, to which all other scientific
laws may in some sense be reduced.
And this.
[A]part from historical accidents that by definition cannot be ex-
plained, the [human] nervous system [has] evolved to what [it is] en-
tirely because of the principles of macroscopic physics and chemistry,
which in turn are what they are entirely because of the principles of
the standard model of elementary particles.
Even though Weinberg gives historical accidents, i.e., the environment, as important a
role in shaping the world as he does the principles of physics, he does so grudgingly,
seemingly attempting to dismiss them in a throw-away subordinate clause. This is mis-
leading, especially given Weinberg’s example—evolution. Contrary to his implication,
the human nervous system (and the designs of biological organisms in general) evolved
as they did not primarily because of the principles of physics and chemistry but primarily
because of the environment in which that evolution took place and in which those organ-
isms must function.
We would extend Jakobsson’s statement beyond biology to include any science that stud-
ies the functional relationship between entities and their environment—and most sciences
study those relationships. The study of solids, for example, is such a science—even
though solids are static entities. What does hard mean other than resistance to (external)
pressure. Without an environment with respect to which a solid is understood as relating,
the term hard—and other functional properties of solids—have no meaning. Without ref-
erence to an environment, a diamond’s carbon atoms would still fit together neatly, but
the functional consequences of that fact would be beyond our power to describe.
This really is not foreign even to elementary particle physics. The Pauli exclusion prin-
ciple, which prevents two fermions from occupying the same quantum state, formalizes a
constraint the environment imposes on elementary particles.13
Thus although neither Weinberg nor Fodor focuses on this issue explicitly—in fact, they
both tend to downplay it—they both apparently agree that the environment within which
something exists is important.
In summary, a functional description is really a description of how an entity interacts with
its environment. This is attractive because it ties both sorts of descriptions to the material
13
This was pointed out to me by Eshel Ben-Jacob [private communication].
ations are given in terms of an environment. Even input/output pairs are defined in terms
of the transformation of some input (in the environment) to some output. That’s how it
works on a Turing Machine. The environment is the tape; the input is found on the tape at
the start of the computation; the output is found on the tape at the end of the computation.
Thus for us emergence is defined in terms of the contrast between the effect of an entity
on its environment and the internal mechanism that allows the entity to have the effect.
3.8 Stigmergy
Once one has autonomous entities (or agents) that persist in their environment, the ways
in which complexity can develop grows explosively. Prior to agents, to get something
new, one had to build it as a layer on top of some existing substrate. As we have seen,
nature has found a number of amazing abstractions along with some often surprising
ways to implement them. Nonetheless, this construction mechanism is relatively ponder-
ous. Layered hierarchies of abstractions are powerful, but they are not what one might
characterize as lightweight or responsive to change. Agents change all that.
Half a century ago, Pierre-Paul Grasse invented [Grasse] the term stigmergy to help de-
scribe how social insect societies function. The basic insight is that when the behavior of
an entity depends to at least some extent on the state of its environment, it is possible to
modify that entity’s behavior by changing the state of the environment. Grasse used the
term “stigmergy” for this sort of indirect communication and control. This sort of inter-
play between agents and their environment often produces epiphenomenal effects that are
useful to the agents. Often those effects may be understood in terms of formal abstrac-
tions. Sometimes it is easier to understand them less formally.
Two of the most widely cited examples of stigmergic interaction are ant foraging and bird
flocking. In ant foraging, ants that have found a food source leave pheromone markers
that other ants use to make their way to that food source. In bird flocking, each bird de-
termines how it will move at least in part by noting the positions and velocities of its
neighboring birds.
The resulting epiphenomena are that food is gathered and flocks form. Presumably these
epiphenomena could be formalized in terms of abstract effects that obeyed a formal set of
rules—in the same way that the rules for gliders and Turing Machines can abstracted
away from their implementation by Game of Life rules. But often the effort required to
generate such abstract theories doesn’t seem worth the effort—as long as the results are
what one wants.
Here are some additional examples of stigmergy.
• When buyers and sellers interact in a market, one gets market epiphenomena. Eco-
nomics attempts to formalize how those interactions may be abstracted into theories.
• We often find that laws, rules, and regulations have both intended and unintended
consequences. In this case the laws, rules, and regulations serve as the environment
within which agents act. As the environment changes, so does the behavior of the
agents.
• Both sides of the evo-devo (evolution-development) synthesis [Carroll] exhibit stig-
mergic emergence. On the “evo” side, species create environmental effects for each
other as do sexes within species.
• The “devo” side is even more stigmergic. Genes, the switches that control gene ex-
pression, and the proteins that genes produce when expressed all have environmental
effects on each other.
• Interestingly enough, the existence of gene switches was discovered in the investiga-
tion of another stigmergic phenomenon. Certain bacteria generate an enzyme to digest
lactose, but they do it only when lactose is present. How do the bacteria “know” when
to generate the enzyme?
It turns out to be simple. The gene for the enzyme exists in the bacteria, but its expres-
sion is normally blocked by a protein that is attached to the DNA sequence just before
the enzyme gene. This is called a gene expression switch.
When lactose is in the environment, it infuses into the body of the bacteria and binds to
the protein that blocks the expression of the gene. This causes the protein to detach
from the DNA thereby “turning on” the gene and allowing it to be expressed.
The lactose enzyme switch is a lovely illustration of stigmergic design. As we described
the mechanism above, it seems that lactose itself turns on the switch that causes the
lactose-digesting enzyme to be produced. If one were thinking about the design of such
a system, one might imagine that the lactose had been designed so that it would bind to
that switch. But of course, lactose wasn’t “designed” to do that. It existed prior to the
switch. The bacteria evolved a switch that lactose would bind to. So the lactose must be
understood as being part of the environment to which the bacteria adapted by evolving
a switch to which lactose would bind. How clever; how simple; how stigmergic!
• Cellular automata operate stigmergically. Each cell serves as an environment for its
neighbors. As we have seen, epiphenomena may include gliders and Turing Ma-
chines.
• Even the operation of the Turing Machine as an abstraction may be understood stig-
mergically. The head of a Turing Machine (the equivalent of an autonomous agent)
consults the tape, which serves as its environment, to determine how to act. By writ-
ing on the tape, it leaves markers in its environment to which it may return—not un-
like the way foraging ants leave pheromone markers in their environment. When the
head returns to a marker, that marker helps the head determine how to act at that later
time.
• In fact, one may understand all computations as being stigmergic with respect to a
computer’s instruction execution cycle. Consider the following familiar code frag-
ment.
temp:= x;
x := y;
y := temp;
The epiphenomenal result is that x and y are exchanged. But this result is not a con-
sequence of any one statement. It is an epiphenomenon of the three statements being
executed in sequence by a computer’s instruction execution cycle.
Just as there in nothing in the rules of the Game of Life about gliders, there is nothing
in a computer’s instruction execution cycle about exchanging the values of x and y—or
about any other algorithm that software implements. Those effects are all epiphenomen-
al.
• The instruction execution cycle itself is epiphenomenal over the flow of electrons
through gates—which knows no more about the instruction execution cycle than the
instruction execution cycle knows about algorithms.
In all of the preceding examples it is relatively easy to identify the agent(s), the environ-
ment, and the resulting epiphenomena.
14
In passing, Brown claims that this is how most new technology develops.
[T]hat is the way we build almost all technology today, even though my lawyers don't want
to hear about it. We borrow things; we tinker with them; we modify them; we join them; we
build stuff.
In each of the examples mentioned above, one can identify what we have been calling an
autonomous entity. In most cases, these entities are self-perpetuating in that the amount
of money they extract from the environment (by selling either products, services, or ad-
vertising) is more than enough to pay for the resources needed to keep it in existence.
In other cases, some Internet entities run on time and effort contributed by volunteers.
But the effect is the same. As long as an entity is self-perpetuating, it becomes part of the
environment and can serve as the basis for the development of additional entities.
tems in that externally supplied energy continually flows through them. The overall
creative process can be summarized as consisting of finding increasingly innovative
ways of using the available energy. To facilitate this process, mechanisms must be
available to support the fungibility of energy—and its proxies such as money,
power, and attention.
• Standards. New products, services, and other items are almost always created
(composed) from existing products, services, and other items. Composition is
greatly facilitated when the elements to be composed adhere to widely accepted
standards.
• Communication and transportation infrastructures. Communication and trans-
portation infrastructures facilitate the exchange/transfer/flow of (a) information
throughout the environment and (b) energy (in one direction) and (c) products and
services (in the other) among trading partners.
• A reasonable level of confidence in the stability and continuity of the products
and services installed in the environment. Mechanisms must be available to allow
agreements to be made and for installed products and services to be relied upon.
• Minimum overhead. Cultural or other mechanisms must exist to discourage cor-
ruption along with enforcement mechanisms to make it harder to siphon off energy
flows for non-productive uses. More generally, the environment must incorporate
mechanisms that minimize the overhead of participating.
• Both (a) centralized but quasi-democratic and transparent governance of the
overall system, its infrastructure, and the standards making process and (b) decent-
ralized overall control (“power to the edge”) in which as much autonomy as pos-
sible is ceded to environment participants.
• Mechanisms that ensure that a certain amount of the available energy is de-
voted to the exploration of the space of new possibilities.
• Mechanisms that allow new products and services to be developed and in-
stalled in the environment and then made known to other participants in the en-
vironment.
• A (primarily, but perhaps not exclusively) bottom-up (i.e., market-like) means
for allocating energy (or its proxies) according to usefulness: the more (less) use-
ful a product or service is found to be (according to actual usage), the more (fewer)
resources it will have at its disposal. All of the participants in the environment must
be self-sustaining in terms of their overall energy transactions. This is possible be-
cause the environment is based on an available external source of “free” energy.
• An ability to form communities of interest (formal, informal, voluntary, and fee-
based) to facilitate the sharing of information, experience, and expertise. The value
of shared information is typically enhanced when it is shared in groups.
• Both (a) sufficient stability of the overall environment that participants can estab-
lish regularized modes of participation and (b) (generally collaborative) means to
allow the environment to evolve as conditions change.
artifact is expressed in terms that are not even a present in the ontological framework of
the lower level elements.
The question we pose in this subsection (and answer in the next) is whether such logic-
ally independent functionality occurs “in nature” at an intermediate level, at the level of
individual things. Or does this sort of phenomenon occur only in human (or chimpanzee)
artifacts?
Given the current debate (at least in the United States) about evolution, one might take
this as asking whether the existence of a design always implies the existence of a (pre-
sumably intelligent) designer.
ies as locomotion or material (i.e., skin) production for our own purposes. The exploita-
tion of existing capabilities for our own purposes is not a new idea.
An interesting example of this approach to engineering involves recent developments in
robotics. Collins reported [Collins] that a good way to make a robot walk is by exploiting
gravity through what he called passive-dynamic motion—raise the robot’s leg and let
gravity pull it back down—rather than by directing the robot’s limbs to follow a pre-
defined trajectory.
This illustrates in a very concrete way the use of an existing force in a design. Instead of
building a robot whose every motion was explicitly programmed, Collins built a robot
whose motions were controlled in part by gravity, a pre-existing force.
outsourcing functions that are not considered part of the core competence of the corpora-
tion illustrates this. Payroll processing is a typical example.
Because many organizations have employees who must be paid, these organizations must
provide a payroll service for themselves. It has now become feasible to factor out that
service and offer it as part of our economic infrastructure.
This outsourcing of internal processes leads to economic efficiencies in that many such
processes can be done more efficiently when performed by specialized organizations.
Such specialized organizations can take advantage of economies of scale. They can also
serve as focal points where expertise in their specialized service can be concentrated and
the means of providing those services improved.
As this process establishes itself ever more firmly, more and more organizations will fo-
cus more on offering services rather than functions, and organizations will become less
stovepiped.
We frequently speak of the “service industries.” For the most part this term has been used
to refer to low level services—although even the fast food industry can be seen as the
“outsourcing” of the personal food preparation function. With our more general notion of
service in mind, historians may look back to this period as the beginning of the age of
services.
Recall that a successful service is an autonomous entity. It persists as long as it is able to
extract from its environment enough resources, typically money, to perpetuate itself.
6 Observations
Our fundamental existence depends on taking energy and other resources from the envir-
onment. We must all do it to stay in existence. Raises fundamental ethical questions: how
can taking be condemned? Supports stewardship notions since we are all dependent on
environment.
Dynamic entities are composed of static and dynamic entities (bodies and societies).
That’s what makes them solid. But those static entity components are frequently replaced.
Competition for energy and other resources justifies picture of evolution as survival of
the meanest. Also justifies group selection since groups can ensure access to resources
better than individuals.
7 Concluding remarks
For most of its history, science has pursued the goal of explaining existing phenomena in
terms of simpler phenomena. That’s the reductionist agenda.
The approach we have taken is to ask how new phenomena may be constructed from and
implemented in terms of existing phenomena. That’s the creative impulse of artists, com-
puter scientists, engineers—and of nature. It is these new phenomena that are often
thought of as emergent.
When thinking in the constructive direction, a question arises that is often under-appreci-
ated: what allows one to put existing things together to get something new—and
something new that will persist in the world? What binding forces and binding strategies
do we (and nature) have at our disposal?
Our answer has been that there are two sorts of binding strategies: energy wells and en-
ergy-consuming processes. Energy wells are reasonably well understood—although it is
astonishing how many different epiphenomena nature and technology have produced
through the use of energy wells.
We have not even begun to catalog the ways in which energy-consuming processes may
be used to construct stable, self-perpetuating, autonomous entities.
Earlier we wrote that science does not consider it within its realm to ask constructivist
questions. That is not completely true. Science asks about how we got here from the big
bang, and science asks about biological evolution. These are both constructivist ques-
tions. Since science is an attempt to understand nature, and since constructive processes
occur in nature, it is quite consistent with the overall goals of science to ask how these
constructive processes work. As far as we can determine, there is no sub-discipline of sci-
ence that asks, in general, how the new arises from the existing.
Science has produced some specialized answers to this question. The biological evolu-
tionary explanation involves random mutation and crossover of design records. The cos-
mological explanation involves falling into energy wells of various sorts. Is there any
more to say about how nature finds and then explores new possibilities? If as Dennett ar-
gues in [Dennett ‘96] this process may be fully explicated as generalized Darwinian evo-
lution, questions still remain. Is there any useful way to characterize the search space that
nature is exploring? What search strategies does nature use to explore that space? Clearly
one strategy is human inventiveness.
8 Acknowledgement
We are grateful for numerous enjoyable and insightful discussions with Debora Shuger
during which many of the ideas in this paper were developed and refined.
References
Abbott, R., “Emergence, Entities, Entropy, and Binding Forces,” The Agent 2004 Confer-
ence on: Social Dynamics: Interaction, Reflexivity, and Emergence, Argonne National
Labs and University of Chicago, October 2004. URL as of 4/2005: http://abbott.cal-
statela.edu/PapersAndTalks/abbott_agent_2004.pdf.
American Heritage, The American Heritage® Dictionary of the English Language, 2000.
URL as of 9/7/2005: http://www.bartleby.com/61/67/S0146700.html.
Anderson, P.W., “More is Different,” Science, 177 393-396, 1972.
BBC News, “Gamer buys $26,500 virtual land,” BBC News, Dec. 17, 2004. URL as of
2/2005: http://news.bbc.co.uk/1/hi/technology/4104731.stm.
Bedau, M.A., “Downward causation and the autonomy of weak emergence”. Principia 6
(2002): 5-50. URL as of 11/2004: http://www.reed.edu/~mab/papers/principia.pdf.
Boyd, Richard, "Scientific Realism", The Stanford Encyclopedia of Philosophy (Summer
2002 Edition), Edward N. Zalta (ed.), URL as of 9/01/2005:
http://plato.stanford.edu/archives/sum2002/entries/scientific-realism/.
Brown, J.S., Talk at San Diego State University, January 18, 2005. URL as of 6/2005:
http://ctl.sdsu.edu/pict/jsb_lecture18jan05.pdf
Carroll, S.B., Endless Forms Most Beautiful: The New Science of Evo Devo and the Mak-
ing of the Animal Kingdom, W. W. Norton, 2005.
Chaitin, G. Algorithmic Information Theory, reprinted 2003. URL as of Sept. 6, 2005:
http://www.cs.auckland.ac.nz/CDMTCS/chaitin/cup.pdf.
CFCS, Committee on the Fundamentals of Computer Science: Challenges and Opportun-
ities, National Research Council, Computer Science: Reflections on the Field, Reflections
from the Field, 2004. URL as of 9/9/2005:
http://www.nap.edu/books/0309093015/html/65.html.
Church, G. M., “From systems biology to synthetic biology,” Molecular Systems
Biology, March, 29, 2005. URL as of 6/2005:
http://www.nature.com/msb/journal/v1/n1/full/msb4100007.html.
Ciborra, C. "From Thinking to Tinkering: The Grassroots of Strategic Information
Systems", The Information Society 8, 297-309, 1992.
Clarke, A. C., "Hazards of Prophecy: The Failure of Imagination,” Profiles of The Future,
Bantam Books, 1961.
Cockshott, P. and G. Michaelson, “Are There New Models of Computation: Reply to
Wegner and Eberbach.” URL as of Oct. 10, 2005:
http://www.dcs.gla.ac.uk/~wpc/reports/wegner25aug.pdf.
Colbaugh, R. and Kristen Glass, “Low Cognition Agents: A Complex Networks Perspect-
ive,” 3rd Lake Arrowhead Conference on Human Complex Systems, 2005.
Collins, Steven, Martijn Wisse, and Andy Ruina, “A Three-Dimensional Passive-Dynam-
ic Walking Robot with Two Legs and Knees,” The International Journal of Robotics Re-
search Vol. 20, No. 7, July 2001, pp. 607-615, URL as of 2/2005: http://ruina.tam.cor-
nell.edu/research/topics/locomotion_and_robotics/papers/3d_passive_dynamic/3d_pass-
ive_dynamic.pdf
Comm Tech Lab and the Center for Microbial Ecology, The Microbe Zoo, URL as of Oct
10, 2005: http://commtechlab.msu.edu/sites/dlc-me/zoo/microbes/riftiasym.html
Comte, A. “Positive Philosophy,” translated by Harriet Martineau, NY: Calvin Blanchard,
1855. URL: as of 7/2005:
http://www.d.umn.edu/cla/faculty/jhamlin/2111/ComteSimon/Comtefpintro.html.
Cowan, R., “A spacecraft breaks open a comet's secrets,” Science News Online, Vol. 168,
No. 11 , p. 168, Sept. 10, 2005. URL as of 9/9/2005:
http://www.sciencenews.org/articles/20050910/bob9.asp.
Dennett, D. C., The Intentional Stance, MIT Press/Bradford Books, 1987.
Dennett, D. C. “Real Patterns,” The Journal of Philosophy, (88, 1), 1991.
Dennett, D. C., Darwin's Dangerous Idea: Evolution and the Meanings of Life, V, 1996.
Dick, D., et. al., “C2 Policy Evolution at the U.S. Department of Defense,” 10th Interna-
tional Command and Control Research and Technology Symposium, Office of the Assist-
ant Secretary of Defense, Networks and Information Integration (OASD-NII), June 2005.
URL as of 6/2005: http://www.dodccrp.org/events/2005/10th/CD/papers/177.pdf.
Einstein, A., Sidelights on Relativity, An address delivered at the University of Leyden,
May 5th, 1920. URL as of 6/2005:
http://www.gutenberg.org/catalog/world/readfile?fk_files=27030.
Emmeche, C, S. Køppe and F. Stjernfelt, “Levels, Emergence, and Three Versions of
Downward Causation,” in Andersen, P.B., Emmeche, C., N. O. Finnemann and P. V.
Christiansen, eds. (2000): Downward Causation. Minds, Bodies and Matter. Århus: Aar-
hus University Press. URL as of 11/2004:
http://www.nbi.dk/~emmeche/coPubl/2000d.le3DC.v4b.html.
Fodor, J. A., “Special Sciences (or the disunity of science as a working hypothesis),” Syn-
these 28: 97-115. 1974.
Fodor, J.A., “Special Sciences; Still Autonomous after All These Years,” Philosophical
Perspectives, 11, Mind, Causation, and World, pp 149-163, 1998.
Gardner, M., Mathematical Games: “The fantastic combinations of John Conway's new
solitaire game ‘life’," Scientific American, October, November, December, 1970, Febru-
ary 1971. URL as of 11/2004: http://www.ibiblio.org/lifepatterns/october1970.html.
Grasse, P.P., “La reconstruction du nid et les coordinations inter-individuelles chez Belli-
cosi-termes natalensis et Cubitermes sp. La theorie de la stigmergie: Essai d'interpretation
des termites constructeurs.” Ins. Soc., 6, 41-83, 1959.
Hardy, L., “Why is nature described by quantum theory?” in Barrow, J.D., P.C.W. Davies,
and C.L. Harper, Jr. Science and Ultimate Reality, Cambridge University Press, 2004.
Holland, J. Emergence: From Chaos to Order, Addison-Wesley, 1997.
Hume, D. An Enquiry Concerning Human Understanding, Vol. XXXVII, Part 3. The
Harvard Classics. New York: P.F. Collier & Son, 1909–14; Bartleby.com, 2001. URL a of
6/2005:: www.bartleby.com/37/3/.
Kauffman, S. “Autonomous Agents,” in Barrow, J.D., P.C.W. Davies, and C.L. Harper, Jr.
Science and Ultimate Reality, Cambridge University Press, 2004.
Kim, J. “Multiple realization and the metaphysics of reduction,” Philosophy and Phe-
nomenological Research, v 52, 1992.
Kim, J., Supervenience and Mind. Cambridge University Press, Cambridge, 1993.
Langton, C., "Computation at the Edge of Chaos: Phase Transitions and Emergent Com-
putation." In Emergent Computation, edited by Stephanie Forest. The MIT Press, 1991.
Laughlin, R.B., A Different Universe, Basic Books, 2005.
Laycock, Henry, "Object", The Stanford Encyclopedia of Philosophy (Winter 2002 Edi-
tion), Edward N. Zalta (ed.), URL as of 9/1/05:
http://plato.stanford.edu/archives/win2002/entries/object/.
Leibniz, G.W., Monadology, for example, Leibniz's Monadology, ed. James Fieser (Inter-
net Release, 1996). URL as of 9/16/2005: http://stripe.colorado.edu/~morristo/monado-
logy.html
Lowe, E. J., “Things,” The Oxford Companion to Philosophy, (ed T. Honderich), Oxford
University Press, 1995.
Maturana, H. & F. Varela, Autopoiesis and Cognition: the Realization of the Living.,
Boston Studies in the Philosophy of Science, #42, (Robert S. Cohen and Marx W.
Wartofsky Eds.), D. Reidel Publishing Co., 1980.
Miller, Barry, "Existence", The Stanford Encyclopedia of Philosophy (Summer 2002 Edi-
tion), Edward N. Zalta (ed.), URL as of 9/1/05:
http://plato.stanford.edu/archives/sum2002/entries/existence/.
NASA (National Aeronautics and Space Administration), “Hurricanes: The Greatest
Storms on Earth,” Earth Observatory. URL as of 3/2005
http://earthobservatory.nasa.gov/Library/Hurricanes/.
Nave, C. R., “Nuclear Binding Energy”, Hyperphysics, Department of Physics and As-
tronomy, Georgia State University. URL as of 6/2005: http://hyperphysics.phy-as-
tr.gsu.edu/hbase/nucene/nucbin.html.
http://oaspub.epa.gov/trs/trs_proc_qry.navigate_term?p_term_id=29376&p_term_cd=TE
RMDIS.
Varzi, Achille, "Boundary", The Stanford Encyclopedia of Philosophy (Spring 2004 Edi-
tion), Edward N. Zalta (ed.), URL as of 9/1/2005:
http://plato.stanford.edu/archives/spr2004/entries/boundary/.
Varzi, A., "Mereology", The Stanford Encyclopedia of Philosophy (Fall 2004 Edition),
Edward N. Zalta (ed.), URL as of 9/1/2005:
http://plato.stanford.edu/archives/fall2004/entries/mereology/ .
Wallace, M., “The Game is Virtual. The Profit is Real.” The New York Times, May 29,
2005. URL of abstract as of 6/2005:
http://query.nytimes.com/gst/abstract.html?res=F20813FD3A5D0C7A8EDDAC0894D-
D404482.
Wegner, P. and E. Eberbach, “New Models of Computation,” Computer Journal, Vol 47,
No. 1, 2004.
Wegner, P. and D.. Goldin, “Computation beyond Turing Machines”, Communications of
the ACM, April 2003. URL as of 2/22/2005:
http://www.cse.uconn.edu/~dqg/papers/cacm02.rtf.
Weinberg, S., “Reductionism Redux,” The New York Review of Books, October 5, 1995.
Reprinted in Weinberg, S., Facing Up, Harvard University Press, 2001. URL as of 5/2005
as part of a discussion of reductionism: http://pespmc1.vub.ac.be/AFOS/Debate.html
Wiener, P.P., Dictionary of the History of Ideas, Charles Scribner's Sons, 1973-74. URL
as of 6/2005: http://etext.lib.virginia.edu/cgi-local/DHI/dhi.cgi?id=dv4-42.
WordNet 2.0, URL as of 6/2005: www.cogsci.princeton.edu/cgi-bin/webwn.
Wolfram, S., A New Kind of Science, Wolfram Media, 2002. URL as of 2/2005:
http://www.wolframscience.com/nksonline/toc.html.
Woodward, James, "Scientific Explanation", The Stanford Encyclopedia of Philosophy
(Summer 2003 Edition), Edward N. Zalta (ed.). URL as of 9/13/2005: http://plato.stan-
ford.edu/archives/sum2003/entries/scientific-explanation/.
Zuse, K., “Rechnender Raum” (Vieweg, Braunschweig, 1969); translated as Calculating
Space, MIT Technical Translation AZT-70-164-GEMIT, MIT (Project MAC), Cambridge,
Mass. 02139, Feb. 1970. URL as of 6/2005: ftp://ftp.idsia.ch/pub/juergen/zuserechn-
enderraum.pdf.