Gregory R. Peterson is Associate Professor of Philosophy and Religion and Program
Coordinator of the Philosophy and Religion Department at South Dakota State University,
Box 504 Scobey 336, Brookings, SD 57007; e-mail greg.peterson@sdstate.edu.
SPECIES OF EMERGENCE
by Gregory R. Peterson
Abstract. The category of emergence has come to be of considerable
importance to the science-and-religion dialogue. It has become
clear that the term is used in different ways by different authors, with
important implications. In this article I examine the criteria used to
state that something is emergent and the different interpretations of
those criteria. In particular, I argue similarly to Philip Clayton that
there are three broad ranges of interpretation of emergence: reductive,
nonreductive, and radical. Although all three criteria have their
place, I suggest that the category of radical emergence is important
both for science and theology.
Keywords: emergence; nonreductive physicalism; reductionism;
supervenience; top-down causation.
Emergence has become a central and almost necessary category for many
theologians engaged with the sciences. One does not have to go very far to
understand why. To claim that there are emergent realities is to counter
reductionistic interpretations of science that are inherently hostile to theological
categories. Theories of emergence provide a framework that makes
scientific and theological claims compatible, even to the point of allowing
the two to be yoked together into a single synthesis. Small wonder, then,
that many of the most prominent theologians and philosophers within the
theology-science dialogue have invoked theories of emergence to varying
degrees.
At the same time, views as to what exactly constitutes genuine emergence
vary quite widely. Although the phrase “the whole is more than the
sum of its parts” is frequently invoked, it is taken to mean different things
by different scholars. Correspondingly, my goal in this essay is twofold:
first, to categorize the primary senses of emergence as they occur in relevant
[Zygon, vol. 41, no. 3 (September 2006).]
© 2006 by the Joint Publication Board of Zygon. ISSN 0591-2385
689
690 Zygon
fields of philosophy, theology, and science; and, second, to suggest how
these different senses may be found useful for the ongoing theology-science
dialogue. Although it is tempting to argue that only one sense of
emergence is the correct one, the more accurate observation is that different
senses of emergence have their place and value. The primary difficulty
is understanding which sense of emergence is appropriate to describe which
kind of phenomena.
WHY EMERGENCE?
All of philosophy, Alfred North Whitehead once observed with only mild
hyperbole, is a footnote to Plato. It is a claim that is at least partially true
of emergence theories. Plato and his contemporaries struggled with the
question of the one and the many, and Plato’s famous solution was to divide
the world in two, between the ever-changing material world and the
everlasting world of forms. The solution was a problematic one, and later
generations modified Plato’s doctrine of forms accordingly. Aristotle spoke
of formal causes; later neo-Platonists divided the two worlds into one world
with many layers, a great chain of being that provided a hierarchy of both
being and value. Although medieval scholars battled over the issue for
centuries, the more skeptical nominalist view eventually won out, influencing
the development and tenor of the scientific revolution.
Most analyses of emergence do not begin with Plato, but I emphasize
this connection at the outset for two reasons. First, at least one form of
emergence claims appeal to a kind of Platonic argument, and it is important
to recognize this in order to understand the complexity of the issues
that emergence theories present. Second, the historical rejection of
Platonism within the sciences (and those early philosophers and theologians
influenced by the sciences) is arguably at the root of the impetus for
the development of emergence claims to begin with.
The rise of emergence theories, beginning in particular with the works
of the British emergentists (Morgan 1923; Broad 1929), can be seen precisely
in light of a rejection of a reductionism that denied the reality and
relevance of higher-order principles and realities. The emphasis on mechanism
placed the primary reality at the level of the most basic material constituents.
For reductionists, the result was a worldview that ultimately
denies the reality of any higher-order pattern or entity, including the mind
and ethical values.
The British emergentists provided the initial impetus for a rejection of
reductionism that was consistent with the sciences. Later discourse on
emergence has proceeded along two different but linked lines of thought.
Emergence claims are most associated currently with the philosophy of
mind and the status of the mind in relation to the body. Simply put, if
both traditional dualism and materialism are false, there must be a sense in
which the mind is still real even though it is in some sense dependent on
Gregory R. Peterson 691
the body. Theories of emergence and its linked concept, supervenience,
provided one way to account for this reality. Similar issues arose for the
philosophy of biology in the middle part of the twentieth century, for, if
pure reductionism is true, what can one make of the complexity and novelty
of biological organisms and the laws that govern them? Emergence
thus became a way to speak of any kind of higher-order reality, so much so
that one could reconstruct a new kind of hierarchy of being composed of
the different levels of analysis that each scientific discipline represents.
In this vein, it is noteworthy that several of the early works that began
the modern phase of discourse on emergence and reduction were significantly
interdisciplinary in character, involving both scientists and philosophers
(Koestler and Smythies 1969; Ayala and Dobzhansky 1974; Globus,
Maxwell, and Savodink 1976). Indeed, emergence can be used as a general
category for understanding the physical world and not simply as a
category for explaining biological organisms. Physical phenomena such as
the Pauli exclusion principle, higher-order forms of physical complexity,
and even the properties of water as distinct from its molecular constituents
have been cited as examples of emergence (see Morowitz 2002).
For emergentists, the issues are twofold. In much of the literature, the
primary issue is an ontological one; there is a sense in which minds, complex
wholes, and the laws that govern them can be said to be real even
when composed of lower-level material objects. Thus, a scientifically informed
worldview and an ontologically reductionist worldview are not the
same thing and, on some accounts, are even opposed to each another. Therefore,
it is possible to believe that minds are real while simultaneously subscribing
to the truths found in neuroscience and to believe that cells and
the laws of evolution that govern them are real while simultaneously subscribing
to the laws of physics and chemistry.
Second, higher-order wholes not only are real; they also are causally
efficacious. This is particularly important for the philosophy of mind,
where issues of human freedom and mental causation are primary. A mind
that is real but is a passive observer (that is, epiphenomenal) is a possibility,
but a not very palatable one. The issue of causation is more important for
some accounts of emergence than for others, as we shall see.
Finally, it is worth mentioning here what is at stake (or perceived to be
so) for theology. Theologies that are strongly dualistic at the outset (both
with regard to human nature and with regard to the God-world relation)
have, arguably, little at stake in theories of emergence. Because mind and
God are completely outside the sphere of nature, emergence plays no role
for such thinkers. Rather, the primary problem is one of reconciling such
a dualism with science at all. Emergence is a primary issue for theologians
who reject traditional dualisms and, consequently, feel strongly compelled
to reject traditional reductionisms as well. It may be argued that this involves
almost all theologians involved in the theology-science dialogue.
692 Zygon
Doctrines of redemption make sense onlyifthereissomeonetobe redeemed.
Furthermore, if God is not completely separate and transcendent over the
world (as maintained by panentheists), the arguments concerning emergence
have possibly some bearing on the doctrine of God itself.
CHARACTERISTIC FEATURES OF EMERGENCE
What does it mean for something to be emergent? Consider the following
three quotes:
Now it is true that the answer to “What else is there (other than atoms and molecules)
in, say, a living organism?” is “no-thing at all”, but this does not mean that
describing its molecular constituents and their properties is all there is to be said,
that there is nothing more to be said by way of description of the individuality of
a particular living organism, especially if it is a human one. (Peacocke 1993, 40–
41)
Systems at each hierarchal level have two properties. They act as wholes (as though
they were a homogeneous entity), and their characteristics cannot be deduced
(even in theory) from the most complete knowledge of the components, taken
separately or in other combinations. In other words, when such a system is assembled
from its components, new characteristics of the whole emerge that could
not have been predicted from a knowledge of the constituents. (Mayr 1988, 15)
Emergences thus occur both in model systems and in real world situations. If the
models are well chosen, the two kinds of emergences map onto each other. They
resonate with each other. In both cases, emergence leads to novelties: the whole is
somehow different from the sum of the parts. (Morowitz 2002, 20)
I select these three passages, not quite at random, to illustrate some of
the common threads of what emergence is taken to mean as well as the
widespread usage of the term. There are certain concepts (and slogans)
that are core to what emergence is about. At the same time, there is some
diversity as to what constitutes this core and what exactly these core concepts
and slogans imply. I suggest that, for emergentist positions to be
coherent, seven elements generally taken to be central claims of the emergentist
position need to be explored and enunciated carefully.
One requirement of an emergent entity is that it be capable of some
kind of higher-order description; there is some kind of whole to be analyzed.
Depending on the emergentist in question, the higher-order description
may be just about anything. On some accounts, even an individual rock
or properties such as the wetness of water may be taken to be emergent
(see, for example, Allen and Bekoff 1997, 8). More common, however, is
the claim that emergent entities be understood in some sense as functional
wholes. Thus, any living organism may be taken as an emergent whole.
So too may basic elements (because they reveal a structural organization),
complex ecosystems, or artifacts such as computers.
Once the wholes have been identified, it is typically claimed that these
wholes obey various sorts of higher-order laws. Stuart Kaufmann (1995),
Gregory R. Peterson 693
for instance, describes life as an emergent phenomenon that can then be
understood to obey certain laws of self-organization and complexity. Much
has been made of John Conway’s Life game, which stipulates a world composed
of discrete pixels that have two states, either on or off, whose states
are controlled by a simple set of rules. What is interesting about the game
is that the simple rules that govern the pixels individually give rise to recognizable
higher-order “organisms” that persist and even move across the
computer screen; these higher-order organisms, in turn, obey certain kinds
of regularities, or laws, so that their behavior can be described independently
of the lower-level “physics” of the life world. Both the entities and
the laws that govern them might be said to be emergent. (For a description
with relevance to these issues, see Dennett 2003; Sharpe and Walgate 2003.)
Similarly, higher-order regularities and principles such as the laws of evolution,
psychology, and economics can be said to also be emergent, governing
the behavior of the relevant higher-order entities already described.
Although the idea of higher-order laws and higher-order description are
conceptually distinct, they are typically seen as going hand in hand.
Claims of higher-order description and higher-order laws typically lead
to a third claim, that of unpredictable novelty. This point is made explicit
in the already cited quotes from Ernst Mayr and Harold Morowitz. The
novelty in question takes one of two forms. The last sentence of Mayr’s
quote (“In other words, when such a system is assembled from its components,
new characteristics of the whole emerge that could not have been
predicted from a knowledge of the constituents”) implies an organizational
novelty. That is, the higher-order whole could not have been predicted
from an analysis of the parts independently. Morowitz, by contrast, seems
to take an historical approach to novelty; what makes something emergent
is (in part) our inability to predict its occurrence beforehand. Presumably,
both kinds of novelty can exist side by side (perhaps necessarily so). In
either case, novelty is taken to be the opposite of predictability; what makes
an object or law emergent is, in part, the unpredictability of that object or
law in advance.
To have wholes, however, one must have parts, and it is in the characterization
of wholes and parts that emergentist theories achieve their primary
distinctiveness. First, emergence positions imply that parts are necessary for
the existence of the whole. There is a part-whole relationship that is essential
to understanding something as emergent, and without the parts the whole
cannot exist. This is particularly clear in the quote from Arthur Peacocke.
It is this claim that most centrally distinguishes emergentist positions from
dualist ones. In René Descartes’ understanding of the relationship of the
mind and body, the mind is not dependent on the body and can exist
independently of it. This contrasts significantly with an emergentist position,
which would imply that if there is no body, there can be no mind.
Second, and equally important, emergentist positions argue that the parts
694 Zygon
are not sufficient for the whole. Thus, after claiming that there is “no-thing”
above the parts, Peacocke also maintains that this “does not mean that
describing its molecular constituents and their properties is all there is to
be said, that there is nothing more to be said by way of description of the
individuality of a particular living organism, especially if it is a human
one.” While the claim of necessity opposes the emergentist position to
dualism, the claim of insufficiency contrasts emergentism with reductionism.
Thus, in a famous example, Donald Campbell argues that it is insufficient
to understand a soldier termite’s jaws solely in terms of its molecular
constituents. To understand the termite jaws fully, one must also have an
understanding of the emergent properties of the functional organization
of the termite, its role in the termite society, and the laws of evolution that
shaped that society as a whole (Campbell 1974).
These five elements may be enough to distinguish emergence as a position,
at least in an ontological sense. It is typical, however, to add one or
two other claims. For most emergentists, it usually is not enough to say
that emergent entities exist; it also is important to claim that they are causally
efficacious. Thus, emergent entities are said to employ a top-down
causality, in contrast to the bottom-up causality of its constituent parts. In
his example of the soldier termites, Campbell argues for a top-down causal
role of evolutionary laws. The notion of top-down causality is especially
central to emergentist theories of mind, for this allows the mind not only
to exist but also to have a causal influence. This causal influence is typically
understood as a constraint of the whole on the action of the parts.
Peacocke (1999) also speaks of top-down causation as downward flow of
information. Either way, the claim is that higher levels play a central role
in explaining why things happen.
Finally, but somewhat less frequently, emergent entities may be said to
be multiply realizable. This may be seen as a consequent of the joint claims
of necessity and insufficiency but also comes centrally out of the history of
thinking about the philosophy of mind. For example, two individuals
may have identical experiences of pain despite having differing configurations
of neurons. An extraterrestrial would (presumably) have a different
kind of brain but still be capable of thinking the same thoughts about
mathematics as human beings do. First elaborated by Hilary Putnam,
multiple realizability has played an important role in the philosophy of
mind as a way of conceiving of the mind as not simply identical to the
brain and therefore, in some sense, an independent reality (Putnam 1960;
for a summary, see Bickle 2002). Multiple realizability, among other qualities,
is what seems to give emergent entities a nonreductive character, because
there is no one set of physical things that they are said to be identified
with.
When we speak of emergence, then, it seems that we are speaking of
seven interrelated characteristics:
Gregory R. Peterson 695
1. Emergent entities are characterized by higher-order descriptions.
2. Emergent entities obey higher-order laws.
3. Emergent entities are characterized by unpredictable novelty.
4. Emergent entities are necessarily composed of lower-level entities.
5. Lower-level entities are insufficient for emergent entities.
6. Some emergent entities are capable of top-down causation.
7. Emergent entities are characterized by multiple realizability.
On any given account of emergence, many if not all of these claims are
included. Of the three authors cited above (Peacocke, Mayr, and Morowitz),
all seem likely to include the first six features in their understanding
of emergence, and perhaps the seventh as well. Philip Clayton (2004) has
provided a similar listing that captures many of these features. Beyond
this, however, that commonality in the characterization of emergence ends.
Instead, we find a great deal of diversity of interpretation, often using the
same or similar arguments and examples but coming to widely variant
conclusions.
CENTRAL ISSUES: EXPLANATION, CAUSATION, AND ONTOLOGY
Among emergentists there is widespread agreement on the explanatory significance
of emergent entities. One reason that purely reductionist approaches
are much criticized is because they seem to leave out so much,
that in emphasizing the individual trees one misses the significance of the
forest. Using Campbell’s example of the soldier termite’s jaw, to explain
the jaw in terms of its molecular constituents is to explain nothing. To do
so loses the understanding of the jaw’s function as well as the relation of
the soldier termite to the termite colony as a whole and the relation of the
termite colony to its evolutionary history. To understand the termite’s jaw,
it is not enough to understand its molecular constituents. One must understand
the jaw as an emergent whole, which in turn is part of larger,
emergent wholes, the colony and its evolutionary history guided by laws of
natural selection.
Arguments of this sort abound and are used to hammer at a naive reductionism
that understands every object solely in terms of its parts, employed
even by scientists in order to make sense of the phenomena that
they are describing. Stephen Jay Gould has famously argued for multilevel
or hierarchical account selection theory to more fully explain the evolutionary
process, as have Elliot Sober and David Sloan Wilson in a different
context (Gould 2002; Sober and Wilson 1998). Paul Glimcher (2003)
has argued that in order to understand why individual neurons behave the
way they do we need to employ a higher order of explanation, looking not
only at the individual neurons but at the kind of problems that the organism
as a whole is trying to solve. Examples such as these abound in science
696 Zygon
writing and indicate the important functional role that speaking of higherorder
entities plays in the practice of many scientific disciplines.
Speaking of emergent entities, therefore, is important not only for interpreting
science but also for the practice of science. To understand complex
entities, it is not enough to understand the parts of which an entity is
composed. The concept of emergence is therefore essential to the process
of understanding. What is not clear, however, is the ontology implied by
these various claims. Are emergent entities real, or are they simply useful
fictions?
One immediate problem we find is that of terminology. Note Peacocke’s
insistence that there is, properly speaking, no-thing at all above the material
constituents of an organism, implying that there is no extra substance,
no élan vital, that is at work and that is separate from the individual parts.
That would be the path of dualism, which emergentists disavow. But if
emergent entities are not independent substances, what are they? In the
philosophical literature, they are frequently described as properties or, in
specific cases, emergent laws or events. But what is an emergent property?
At this point we find the shadow of Platonism looming large, for it
seems that we have primarily two options. The first option is to understand
emergent entities as useful fictions. We can call this the nominalist
view. Nominalist William of Ockham argued that Platonic forms are mere
names with no further reality or, at best, concepts or conceptual relations
that exist only in the perceiving mind. Likewise, when one speaks of emergent
laws such as the law of supply and demand or of emergent entities like
a cell or a computer program, the nominalist would say that these are
nothing more than useful fictions—useful because we are middle-sized
creatures who perceive the world in a particular way. Because our eyes
have limited resolution, we see the table (or the rabbit) as a whole and give
it a name as a consequence; if our visual resolution was sufficiently higher,
we would not see the table at all but just a collection of atoms. On the
nominalist view, we speak of tables, cells, and economic laws because such
orderings of our experience are useful, not because they are real. To invoke
emergent entities because of their explanatory power, then, is to simply
acknowledge their explanatory usefulness. We speak of panda bears because
it is useful for us, as middle-sized beings with modest visual resolution,
to do so. But ontologically, any individual panda bear is a collection
of subatomic particles, and pandas as a species concept is, presumably, a
set of such particle arrangements.
Although this is a possible interpretation of emergence, most who invoke
the language of emergence seem to argue otherwise. For most emergentists,
there is an important sense in which pandas, people, and laws of
economics are real that is more than stipulating them as a mere aggregation
of their respective parts. Emergentists generally abhor statements that
include the phrase “nothing but” in them, such as “a panda bear is nothing
Gregory R. Peterson 697
but a collection of molecules.” So, the emergentist is saying that there is,
ontologically, something more to these emergent realities. What is this
something more?
This question of the ontology of emergent entities typically is linked to
the question of the causal efficacy. As already noted, a central claim of
emergentist positions is that emergent entities are capable of exerting a
kind of top-down causation. Neuroscientist Roger Sperry was one major
proponent of this view, arguing that any complete account of human action
must not only take into account the flow of causality from neurons to
brain to mind but must also conceive of the top-down causal effect of the
mind on the brain (Sperry 1987).
The prime question, however, has been exactly what is meant by topdown
causation. Normally, the idea of top-down causation is meant not
to replace or contradict the causal powers (usually understood in terms of
Aristotle’s efficient causation) of the lower levels of physics and chemistry
but rather to complement them. In explaining why a particular neuron
fires, for instance, we can give an explanation in terms of bottom-up causation,
enumerating the individual molecules of which the neuron is composed,
the laws that govern the individual particles, and the immediately
antecedent conditions that lead them to act as they do. But, observes the
emergentist, we have explained very little by making these observations,
because we have not, in a larger sense, understood why the neuron is firing,
which is because it is connected to a broader visual network in the
brain that is responsive to my higher-order decision to turn and look at the
clock. For the emergentist, bottom-up causation is always part of the story,
but when it comes to complex objects, it is only part of the story.
Frequently, notions of top-down causation hinge on claims about relevant
boundary conditions. Emergent entities provide the context in which
local, bottom-up causation takes place and is made possible. To refer back
to Campbell’s soldier termite, a bottom-up causal account of how the
termite’s jaw came to be would rely simply on the molecular constituents,
the relevant physical laws, and their antecedent conditions. But such an
account does not tell us why the molecules work together to form a jaw to
begin with. To understand that, we need to invoke the top-down causal
powers—the boundary conditions imposed by DNA encoding, cellular
organization, niche and colony specialization, on up to the laws of natural
selection. Similarly, in order to understand a computer network, it is not
sufficient to understand it as a linear sum of its parts; one must understand
a network as a higher-order whole that exerts a downward causal effect on
its individual components.
As in the case of ontology, there is a key link in many arguments regarding
top-down causation to claims about understanding and explanation.
Top-down causation is important because it is essential in order for us to
understand what is going on in many complex physical systems. It is, one
698 Zygon
might say, psychologically necessary. But is top-down causation real?
As with the ontological question, we can go two routes. We might
describe top-down causation as a useful fiction. That is, top-down causal
language is important because it helps us middle-sized beings of limited
cognitive capacity understand physical systems that would otherwise be
beyond our comprehension. We cannot see the molecular constituents of
the soldier termite’s jaw, let alone envision all the molecular constituents
that make up its environment and that environment’s past evolutionary
history. Concepts such as cells, ant colony, and natural selection and the
causal claims associated with them are useful approximations for the biological
sciences, but they are only approximations and, in the end, simply
fictions that help us in our tasks as scientists and philosophers.
Along this line of thought, it is worthwhile to point out how careful
many emergentists are in claiming that emergent entities and their topdown
causal effects do not imply a negation of lower-level efficient causation.
This leads to some fuzziness in exactly how to construe what a
top-down cause is. Peacocke, for instance, speaks of top-down causation
as a “flow of information” (1999, 225–26). Philosopher of mind Fred
Dretske (1997) distinguishes between triggering causes and structuring
causes. In these and other cases, it is unclear to what extent we should
really regard top-down causation as a cause in the normal way we use the
term or if something else entirely is meant. If top-down causation is really
a flow of information, perhaps emergentists should abandon the language
of causality altogether in order to avoid confusion.
Many of those who have taken the idea of top-down causation more
literally have argued in favor of a supervenience relation between lowerlevel
and higher-level properties that would allow for a robust understanding
of top-down causation. Within the science-and-theology dialogue,
Nancey Murphy (1999) has been one of the most vocal supporters of this
account. Generally speaking, supervenience accounts of top-down causation
assume a (largely) deterministic and complete lower-level framework
out of which arise supervenient, emergent properties that in turn exert a
downward causal effect on the lower-level, subvenient elements. In the
specific case of the mind-body relationship, one may speak of a causal
relation not only from the physical to the mental but also from the mental
back to the physical. A typical diagram envisions the relationship this way:
Gregory R. Peterson 699
But does the mental really play a causal role here? There is good reason
to suppose that it does not (Kim 2000). As long as a physical system is
deterministic and nonchaotic (a big “if” to which we will return), it should
in principle be possible to determine the behavior of the system, however
large and however bounded, in terms of its most basic constituents alone
without invoking the top-down causal role of emergent entities. It may be
said that in doing so we do not sufficiently understand the system in question.
We have missed out, perhaps, on the interesting higher-order regularities
of natural selection, but the critic may with some safety dismiss
that as a merely psychological matter. If I can accurately model and predict
how a system operates, I have all the understanding that I need.
There may be room for rejoinder here, but it would seem that supervenience
is of limited use for the emergentist who endorses a robust notion
of top-down causation. If so, the emergentist has to look for an alternative
to strict supervenience or give up on the causal effects of emergent entities.
If the latter, emergent entities might be real but epiphenomenal, with the
result that emergent entities have no causal impact whatsoever. Most emergentists
would regard this as a disaster.
SPECIES OF EMERGENCE
At this point, we can observe that not all emergentists endorse supervenience
as a way to account for top-down causation. Indeed, once we get
beyond the primary seven characteristics that most emergentists have in
common, we find considerable diversity as to how to regard the more specific
ontological and causal claims. This diversity is tied fairly strongly to
other, prior, ontological commitments regarding the physical world. With
regard to such commitments, we may construe views of emergence as ranging
from the more conservative to the more liberal. Although each different
approach to emergence has its place, I suggest that if emergence is
going to have the kind of interesting implications that many of its advocates
claim, it is the more liberal accounts that will have to be endorsed.
Reductive Emergence. On the most conservative account, emergent
entities and their causal forces are nothing more than useful fictions. This
is the nominalist view alluded to above, and when it is endorsed it usually
is accompanied by a strong commitment to reductive physicalism. Biologist
Richard Dawkins and philosopher Daniel Dennett may both be seen
as exemplars of this kind of view. In his book The BlindWatchmaker (1996)
Dawkins briefly addresses the issue of reductionism, explaining that he
rejects a naive reductionism and endorses what he calls hierarchical reductionism,
which accepts the usefulness of higher-order descriptions and explanations.
Similarly, Dennett distinguishes between good reductionism
and greedy reductionism. Good reductionism acknowledges that objects
can be described in different ways and in terms of different hierarchical
700 Zygon
levels. Indeed, Dennett argues that for some objects (such as most animals
and sophisticated robots), we can take three stances—the physical stance,
the design stance, and the intentional stance—each with different kinds of
explanation associated with them (Dennett 1987; 1995).
Most emergentists will likely counter that neither Dawkins nor Dennett
is a genuine emergentist; they are reductionists, the very opposite of
what emergentism is about. In an important sense they are right. For
both Dawkins and Dennett, there is a strong sense that the whole is nothing
more than the sum of its parts and that emergent level entities and
explanations are simply useful fictions that aid in our understanding of
how things work. Dennett, for instance, sometimes makes this point elliptically,
as when he refers to the theory of natural selection as a universal
acid, but he also explicitly refers to higher-order entities as useful fictions
in other places (Dennett 1990). Yet, both Dennett and Dawkins are willing
to use such emergent language as a tool for explaining the complexity
that surrounds us, and this contrasts with some schools of thought that
would deny even this much—for example, the agenda of eliminative materialists
that wanted to completely replace what they dubbed “folk psychology”
with the language of neuroscience (see Churchland 1986). As
such, Dennett and Dawkins represent one logical if largely unpalatable
end of dealing with emergent entities short of a complete eliminativism.
This sort of view seems at least partially endorsed more recently by
Donald Wacome (2004).1
Wacome argues that the concept of pizza is not
simply reducible to material components. Pizzaness is multiply realizable
in the sense that there are many (perhaps infinitely many) kinds of pizza.
Nevertheless, Wacome observes, we generally do not feel compelled to endorse
pizza dualism. Pizza is, Wacome claims, nothing more than its physical
constituents. In other words, pizza as a term and concept is useful, but
that does not make pizzaness real in any relevant sense. Pizzaness is a
useful fiction.
Nonreductive Physicalist Emergence. To many, a more satisfactory approach
is to endorse emergence within the context of a nonreductive physicalism.
Generally speaking, nonreductive physicalists agree with other
physicalists that at the lowest level of analysis everything is physical and
everything is composed of physical particles (such as quarks and electrons)
and the laws that govern them. For the nonreductive physicalist, however,
emergent entities are both real and causally efficacious. Most important,
this provides room to speak of mind, spirit, and freedom in a way that is
not available to the simply nominalist approach to emergent entities. On
a nominalist approach, mind, spirit, and freedom are simply useful fictions.
The nonreductive physicalist claims something more: mind, spirit,
and freedom are real by virtue of being irreducible, and, being real, they
have causal impact.
Gregory R. Peterson 701
Centrally assumed in most accounts of nonreductive physicalism is the
notion that scientific accounts of nature are essentially accurate and complete
enough in such a way that new scientific discoveries are not going to
result in any major changes of how things work, particularly with regard to
biological organisms. Although there is much that is undiscovered, the
nonreductive physicalist generally is confident that new developments in
string theory (for instance) are not going to affect how we think about the
relationship of mind and brain, and that although new discoveries in neuroscience
will doubtlessly be important, they will not alter the basic logical
relations between the brain as a physical phenomenon and the mind as an
emergent entity or set of properties that emerge from the brain. Because
of this emphasis on the closure of the physical, I have elsewhere referred to
emergence within a nonreductive physicalist position as closed-system
emergence (Peterson 2002). As such, the nonreductive physicalist acknowledges
the necessity of a physical substratum for the existence of emergent
entities but emphasizes in particular the insufficiency of the physical to
properly account for the emergent entities and the relations between them.
There is an explanatory gap.
Given this gap, we can put the question this way: If nonreductive physicalism
is true, what kind of emergence is possible? Taking the ontology
claim first, we might ask what it is that makes a human mind real and not
simply reducible to the brain and its actions. One option is to define the
reality of emergent entities simply in terms of the seven criteria noted earlier.
That is, something is said to be real simply by virtue of higher-order
description, insufficient lower-level conditions, and so on. In particular,
the claim to the insufficiency of the physical substratum is key, because it
suggests that there is something more to the emergent entity than simply
its lower-level physical components. Multiple realizability is seen as buttressing
this point. In the case of the mind-body relation, although each
individual brain is different in minute and sometimes important ways,
nevertheless different brains can realize the same experiences (the sensation
of rough sand on the skin) and thoughts (knowledge of the Pythagorean
theorem). In principle, this could be true even among brains that are
radically different. If intelligent aliens landed on the planet tomorrow, by
virtue of their intelligence they too would be able to understand the
Pythagorean theorem, but they would presumably have brains very different
from ours. Similarly, computers with different physical configurations
and operating systems (PC, Macintosh, Unix) may differ in terms of the
physical states yet still produce the same emergent state, for example, a
Web page running a Java applet (a small program written in the Java programming
language). Indeed, we may say that the Java applet running on
each machine is identical, even though the physical procedures involved in
running the applet on each computer is different.
702 Zygon
This assertion of the reality of emergent entities may still seem troublingly
vague to some. If emergent entities are not simply identical to their physical
constituents, what are they? For nonreductive physicalists, the reality
of emergent entities cannot come from some other supernatural realm,
because then they would no longer be nonreductive physicalists; they would
be dualists of the standard Cartesian variety. Nor can nonreductive physicalists
claim that there is some unknown but relevant physical quantity
that stands unaccounted, because they have previously claimed that we
know pretty much all there is to know about the physical and that anything
new that is yet to be discovered promises not to affect the basic
emergence claim. Given that both of these avenues are closed, in what
sense are emergent entities real?
Nonreductive physicalists, as well as other emergentists, sometimes identify
emergent entities with information. We have already seen this connection
between information and top-down causation made by Peacocke.
Philosopher of mind David Chalmers (1997) claims that consciousness
can be understood in terms of information, invoking a property dualism
that falls within a nonreductive physicalist framework. A problem here is
that different authors mean different things by the term information, some
sticking close to a scientific definition as used in information theory and
physics, some understanding the term in a broader sense. In order to be
compatible with nonreductive physicalism, it would seem that something
closer to the scientific sense would be necessary. More generically, we might
think of emergent entities as patterns or as a patterned flow. From a nonreductive
physicalist point of view, my mind supervenes on my brain, and
what is relevant about my brain is its very specific pattern of neurons and
neuronal connections along with their associated support networks (glial
cells, capillaries) and connections to the body. On the nonreductive account,
it is the pattern that is important. In principle, if the pattern of my
brain could be reconstructed (for example, if I died but the pattern was
saved), I would be conscious once again, or a duplicate of me could be
created—with all the attendant identity problems this creates.
This emphasis on pattern is reminiscent of Plato’s use of mathematical
objects to argue for the existence of a separate world of forms. Just as Plato
argued that our knowledge of mathematics and geometric figures revealed
an existence prior to this one, connecting us ultimately to the world of
forms, so too the emergentist points to the multiple realizability of information-bearing
patterns as revealing their nonreducible character. Despite
this, the nonreductive physicalist framework has an Aristotelian edge,
for it is the shape (the pattern) and the flow of patterns through time
(information) that give emergent entities their reality. The pattern is not
completely separate from physical reality, as Plato would have it, but is
ultimately connected to the physical, as Aristotle seems to imply. These
connections suggest that the arguments both for and against Platonic and
Gregory R. Peterson 703
Aristotelian forms have some relevance here, although this connection has
been poorly explored.
If emergent entities are understood in terms of patterns or information,
in what sense are they causally efficacious? Here we have a much larger
problem. In invoking top-down causation, nonreductive physicalists have
hoped to provide an account of emergent properties that will preserve their
causal powers and, in particular, provide space for mental causation and
free will when the focus turns to philosophical and theological anthropology
(Murphy 1999). Higher-order, emergent entities, therefore, are not
only real; they are a cause among causes that need to be taken into account
when describing the physical world.
How can this work? Here, the nonreductive physicalist can only appeal
to supervenience arguments. As we have already seen, however, there is
good reason to believe that these are not going to provide the kind of
robust accounts of top-down causation as had been hoped. Nonreductive
physicalists are correct to point to the importance of boundary conditions,
the influence of the whole on its parts, and the explanatory significance of
higher-order entities and laws. But on a nonreductive physicalist account,
all of these considerations must be given a nominalist interpretation in
reference to causality.
Despite this, there are a few avenues available for the nonreductive physicalist.
One avenue suggested recently by Dennett (2003) is to define causality
in a way that is going to be relevant only at higher-order levels, not
necessarily lower-order ones. Dennett argues that for A to cause B, A must
be both necessary and sufficient for B. Necessary and sufficient conditions,
to put it as briefly as possible, cannot be given in the attempt to
relate prior, micro/subvenient events to later macro/emergent ones. It is
thus possible, according to Dennett, for an event to be determined yet,
technically speaking, uncaused, providing space for conceptions of mental
causation and free will. The form of Dennett’s argument, redefining causality,
seems to be the primary move available to nonreductive physicalists.
One can either define causality differently so as to include top-down causality
or one can add kinds of cause (formal cause, structuring cause) to
make it plausible. Either way, nonreductive physicalism would seem to
entail either epiphenomenalism or, at best, a form of compatibilism with
lower-level determinism.
Despite these concerns, emergence within the context of nonreductive
physicalism has its place, and there may be many areas where emergence of
this sort applies. Computers may, in fact, be the best examples of nonreductive
emergence at work. When analyzing how computers operate, we
have (in principle) complete knowledge of their lower-level workings and
can (again, in principle) connect the lower-level workings with the higherlevel
structures that emerge. Software programs and computer networks
704 Zygon
display most if not all of the features of emergent entities, including surprising
novelty at the emergent level. For a computer, top-down causation
is understandable in terms of supervenience relations, with all that implies.
Radical Emergence. For many advocates of emergence, nonreductive
physicalism does not go far enough. In particular, advocates of radical
emergence are inclined to reject the central premise of nonreductive physicalism
that our knowledge of the physical world is essentially complete
and that any new discoveries will not affect our understanding of the world
dramatically. Rather, radical emergentists argue, our knowledge of the
physical world is incomplete in important ways, and we should be wary of
hastily reducing higher-order, complex phenomena into their poorly understood
parts.
Radical emergentists share certain strategies with nonreductive physicalists,
because both oppose completely reductionist accounts of the world
and human beings. Radical emergentists point to the failures of the reductionist
program, to stubborn phenomena such as human consciousness
that resist attempts at reduction, and to the explanatory gap that results
when we try to explain complex systems simply in terms of their lower
components.
Radical emergentists differ from nonreductive physicalists in their more
expansive understandings of both epistemological and ontological emergence.
First, radical emergentists emphasize that our knowledge of the
physical world is incomplete, so to claim that human consciousness can be
reduced to the activity of neurons or (at the presumably lowest level) a
particular arrangement or set of arrangements of subatomic particles is to
assert that we already have a complete account of these physical and biological
levels. Clearly, we do not yet have complete knowledge of these
lower levels, and so, in particular cases such as that of human consciousness,
the radical emergentist will argue that it is premature to claim that
neurons as we now understand them can serve as the proper subvenient
base to provide an account of consciousness, as the nonreductive physicalist
will have it. Note what is being argued here. The radical emergentist is
not denying that neurons (for example) are necessary for the emergence of
human consciousness but is asserting in this case that our current understanding
of neurons is deficient in an important way—so important that it
prevents us from being able to understand neurons as a subvenient base.
Radical emergence, therefore, has an important epistemological component,
which opens up possibilities for more radical ontological claims.
Recall that a premise of nonreductive physicalism is that the universe is
closed and that the scientific description of the physical world as we now
have it is essentially correct. Once this premise is accepted, it follows that
any emergent entity must be understood as arising out of the physical
world as we now understand it. So, in the case of consciousness, the non-
Gregory R. Peterson 705
reductive physicalist is pretty much committed to saying that consciousness
is a form of information processing arising out of the activities of
neurons, because this seems the avenue of interpretation most consistent
with contemporary cognitive science. Because radical emergentists do not
share the premises of closure and completeness, they are not compelled to
make the same move. Indeed, in a difficult case such as consciousness, the
radical emergentist may argue not only that we are ignorant of the particular
physical basis of consciousness (epistemological emergence) but that
there remain important, unknown physical properties that are yet to be
discovered (ontological emergence). Radical emergence thus leaves the
door open for radical ontological claims in a way that nonreductive physicalism
does not.
Implied in this is an understanding of science and reality very different
from that held by the nonreductive physicalist. Because nonreductive physicalists
are committed to closure and completeness, there is a strong commitment
to a unified worldview that is comparatively seamless, emphasizing
the smooth flow from the lowest levels of subatomic physics to chemistry,
biochemistry, cell structure, multicellular organisms, on up to laws, minds,
and the organization of complex societies. The radical emergentist, by
contrast, will be more inclined to observe the disunity of science. On this
view, there is not a seamless flow from one discipline of scientific study to
another. Rather, each level of scientific inquiry should be interpreted as a
slice of reality, partially continuous with adjacent slices but partially discontinuous
as well. Because of the emphasis on both epistemological and
ontological openness, I have referred to radical emergence as open-system
emergence (Peterson 2002).
In the theology-and-science dialogue, Philip Clayton has perhaps been
the most prominent advocate of a form of radical emergence (1997; 2004).
Clayton also distinguishes between reductive, nonreductive, and radical
forms of emergence (referred to in Clayton 2004 as façon de parler emergence,
weak emergence, and strong emergence respectively). Clayton rejects
both the reductive and nonreductive forms of emergence in favor of
radical (strong) emergence, arguing that radical emergence is required to
sufficiently explain consciousness and perhaps other natural phenomena
as well. For Clayton, the argument for radical emergence is in no small
part an empirical one, arising out of a need to explain the stubborn data
generated by the sciences. The reason we come to emergence as a philosophical
position, Clayton argues, is that we observe emergence in the world
and find it to be scientifically useful. For Clayton, human consciousness is
one such radically emergent phenomenon, arising out of the natural but
not reducible to the physical as understood within the context of contemporary
science. Clayton concludes, “Emergent monism makes mental properties
strongly emergent out of a substrate that is neither ‘physical’ nor
‘mental’ (2004, 158). On Clayton’s analysis, current neuroscience cannot
706 Zygon
explain consciousness precisely because neuroscience is not a complete science
and as such gives us only a slice of the reality that makes up the
human person. Likewise, physics may not be able to explain the complex
networks created by brains not because physics is false but because our
knowledge of physics is incomplete and, perhaps, limited by the very presuppositions
that physics makes.
Another example of the use of radical emergence within the context of
the science-theology dialogue is the work of John Haught. Although
Haught does not embrace the language of emergence, his approach to science
in general and evolutionary biology in particular employs the very
kind of arguments typical of radical emergence. In Deeper than Darwin
Haught provides an extended argument that nature, like a book, can be
read on several different levels but also emphasizes the incompleteness of
individual levels. In this vein, he argues that the subatomic world that
physics describes should be understood not as the concrete, subvenient
base out of which everything else arises but as the most abstract of the
different views of the world that science provides (Haught 2003, 44).
Physics provides a slice of reality, but only a slice, implying that higherorder
physical realities cannot be simply understood in terms of arrangements
of subatomic particles.
A somewhat different approach is taken by Paul Humphreys (1997).
Like Clayton and Haught, Humphreys focuses on the incompleteness of
the scientific worldview as we now have it, noting that we can never provide
an adequate subvenient base because we cannot be sure what the base
actually is. While the subatomic world currently described by modern
physics may be the ultimate, lowest-level base, tomorrow’s new discoveries
may render our current understandings obsolete. Humphreys furthermore
argues for the possibility of emergent laws and properties that appear only
when lower-level entities come together in a particular way. He cites the
Pauli exclusion principle as a prime example of this sort of emergence.
This principle forbids any two electrons from occupying the same orbital
around a nucleus. There is nothing, however, in the basic laws of physics
governing single particles to predict or explain why two electrons cannot
occupy the same orbital in an atom. The Pauli exclusion principle would
appear to be, following Humphreys’s analysis, an emergent law that comes
into play only when some subatomic particles interact in a particular way.
On his analysis, the individual particles do not provide a subvenient base;
rather, they are part of a set of conditions necessary for the Pauli exclusion
principle to appear.
Other examples may be adduced, either as exemplars of radical emergence
or as bearing important family resemblances. William Hasker’s
emergent dualism (1999), for instance, shares some of the attributes of
radical emergence. So do aspects of the metaphysical worldview of process
philosophy as enunciated by Whitehead ([1929] 1985). Although the
Gregory R. Peterson 707
particular routes these thinkers use to arrive at their conclusions are somewhat
different, the result is often very similar, if not the same.
EMERGENCE, RELIGION, AND SCIENCE
At this point, the concern over theories of emergence should be clear. An
important problem is that when emergence is invoked as an explanatory
category, a number of quite distinct claims are actually being made. Generally
speaking, most arguments for the existence and nature of emergent
entities share seven primary characteristics. Beyond this, however, there is
significant disagreement as to how to construe the meaning and significance
of emergent entities, ranging from the more reductive approaches of
Dennett and Dawkins to the more radical approaches of Clayton and
Humphreys, with nonreductive physicalists’ accounts occupying a middle
position. This diversity is muddied by the fact that some authors seem to
be making arguments for more than one kind of emergence while failing
to distinguish the kinds of emergence being argued for. Morowitz (2002),
for instance, argues for the existence of twenty-eight kinds, or levels, of
emergence in the natural world, ranging from the laws governing the origins
of the universe to the emergence of social systems and spirituality
among human beings, including cells, neurons, and technology. It is not
clear, however, that Morowitz’s twenty-eight examples are all emergent in
the same way, and a good case can be made that they are not. A similar
tension (although not explicit contradiction) appears in the recent approach
of Ursula Goodenough and Terrence Deacon (2003), who argue that there
are three levels of emergence that correspond, roughly, to simple interactions,
complex systems, and cognitive systems with memory and feedback.
When interpreting the meaning of emergence, they write, “Metabolism
and mentality are nothing but their constitutional parts. But they are also
something more, something new and emergent” (2003, 803). So strongly
embracing nothing but seems to imply the reductionist emergence accounts
put forward by Dennett, Dawkins, and others. Yet, the insistence on something
more seems to imply at minimum a nonreductionist account. Which
is being advocated is not clear.
Such conflations seem to be rare, but they illustrate the problems that
can arise that in turn create problems for critiques of emergence theories.
Critiques that emphasize supervenience relations apply primarily to nonreductionist
emergence but not necessarily to radical emergence. Charges
that emergence theories are merely a reintroduction of an élan vital or a
new kind of dualism apply more to the radical emergentist accounts than
to nonreductionist ones.
It is important to note that the three broad kinds of emergence delineated
do not necessarily conflict until they are made into universal claims.
For example, a radical emergentist may argue that certain kinds of systems
708 Zygon
are best construed in terms of reductionist emergence (emergent properties
are useful fictions), other kinds of systems are best construed in terms
of nonreductive physicalist emergence (for example, computer hardware
and software), and some few (such as the Pauli exclusion principle or human
consciousness) are instances of radical emergence. They come into
conflict only when, for instance, the claim is made that all emergent entities
are of the nonreductive variety or when there is disagreement over
what kind of emergence applies to a particular phenomenon such as human
consciousness. In discussing emergence, clarity is precious.
Given these complexities, what is the utility of theories of emergence
for theology? If reductionist emergence is taken as a universal account of
emergence, it is clearly antithetical to any theological project that makes
appeal to the reality of the person and to God, and it is no accident that
there is a strong tie between reductionist accounts of emergence and avowed
atheism, as the work of both Dawkins and Dennett exemplifies.
Nonreductive emergence presents a more complicated case. It does allow
room for the reality of the person and higher-level objects in general
but at the price of genuine freedom and causal efficacy. Certainly, nonreductive
physicalism has its theological advocates, as exemplified by the
work of Murphy, Joel Green, and others (Brown, Murphy, and Maloney
1998; Green 2004). As seen above, however, the idea of mental causation
seems problematic at best within a nonreductive physicalist framework, as
do claims to human freedom. Murphy claims that nonreductive physicalism
is consistent with human freedom, but it would appear that the freedom
we are speaking of is the compatibilist rather than the incompatibilist
variety. That is, freedom in the nonreductive physicalist context must be
compatible with microdeterminism. In addition, because nonreductive
physicalism emphasizes causal closure and completeness, God’s relationship
to the world must be at first blush either completely immanent or
completely transcendent. If completely immanent, God is an emergent
property of the world as a whole, giving rise to a form of panentheism or,
more properly, a panenworldism. An emergent God within the context of
nonreductive physicalism would be a God that is determined by the world
rather than the reverse (Bielfeldt 2000). The alternative would be to understand
God as completely transcendent, with divine action rendered as
impossible (deism) or as completely miraculous.
Quantum indeterminacy may solve some of these problems. Robert
Kane (1996), for instance, appeals to quantum indeterminacy as a means
for supporting a notion of free will that is not bound by microdeterminism.
Murphy and others have appealed to quantum indeterminacy as a means
of providing an account of divine action compatible with nonreductive
physicalism (Murphy 1997). In addition, it is worth pointing out that a
significant segment of the Christian tradition, from Augustine through
Luther, has endorsed accounts of freedom and predestination that may be
Gregory R. Peterson 709
seen as compatible with nonreductive physicalism in a way that many current
positions are not.
I do not fully evaluate these options here but simply point out that they
do have their problems. Quantum indeterminacy does not provide a solution
to the problems with top-down causation in a nonreductive physicalist
context. Although freedom may be defined in such a way as to be
compatible with microdeterminism, I would suggest that stronger accounts
of freedom are to be preferred. Beyond this, I also suggest that not only is
nonreductive physicalism not good for theology, it is also of limited use for
science. With its emphasis on closure and completeness in terms of our
contemporary understanding of science, nonreductive physicalism puts a
limit on the scientific imagination, a straitjacket that confines interpretation
of research into that which is currently conceivable.
At this point radical emergence reveals its strengths. Radical emergence
is, arguably, good for both theology and science. It cracks open the world,
allowing for the possibility of that which is truly undiscovered. On one
hand, radical emergence does not force artificial solutions to existing problems,
so there does not need to be, for example, a rush to explain the
phenomenon of consciousness in terms of existing scientific categories and
models. On the other hand, it does not artificially foreclose scientific investigation
the way that, for instance, substance dualism does by proclaiming
certain domains (mind, ethics, beauty) as being automatically immune
to scientific analysis. Theologically, radical emergence allows for concepts
of God that are both transcendent and immanent and provides a better
framework for panentheism than nonreductive physicalism.
Nevertheless, radical emergence also has its dangers, possibly leading to
what might be called an emergence of the gaps. This is seen most prominently
in arguments for the emergence of human consciousness. For many
theologians it is sufficient to claim that human consciousness is an emergent
phenomenon and leave it at that. This seems to be the route, more or
less, that Peacocke takes in his Theology for a Scientific Age (1993). While
this statement may be satisfactory at the theological level, it is not very
informative for the practice of neuroscience and related disciplines. Claims
of radical emergence can lead to a disincentive for scientific research, as
any insufficiently explainable phenomena can be labeled as emergent and
left at that. For radical emergentists, an important question is when to
label something as emergent. What makes the Pauli exclusion principle,
for example, an emergent law is precisely its inexplicability in terms of
lower-level laws and particle interactions. In other words, the evidence for
emergence in this sort of case is lack of evidence; a lower-level explanation
does not currently exist. Could one exist in the future, so that the Pauli
exclusion principle could be reduced to lower-level, more universal regularities?
This is a possibility, although the principle has been so thoroughly
investigated that such a reductionist solution seems unlikely. There is a
710 Zygon
disturbing parallel here between radical-emergence arguments and arguments
made by advocates of intelligent-design theory that also depend on
the lack of evidence for reductionist explanations to defend concepts of
irreducible complexity (Behe 1996, for example).
A second issue for radical emergence is whether or not the sciences provide
much room for radical-emergence claims. Laws of energy conservation
and entropy seem to hold universal sway and must govern any emergent
phenomena that are claimed to appear. Although our knowledge of the
human brain is far from complete, it is sufficiently detailed to make claims
for the emergence of consciousness and human freedom complicated.
There is perhaps no easy way around this. In the absence of any neat
algorithm for determining the completeness and strength of scientific theories
and their relation to one another, individual judgment with all of its
subjective vagaries must be employed. This does not mean that determining
emergence relations is an irrational activity, only that it requires a careful
and detailed analysis, an openness to revision, and an awareness of when
reductionist approaches are promising and when they are not. Indeed, the
strength of radical emergence claims comes from the very precise weaknesses
of reductionism in specific instances. In this light, it may be better
to think of radical emergence not so much as a sweeping ontological claim
as a practical heuristic. Radical emergence is a reminder to us all that all is
not yet known and that the complexities of experience need to be explained,
but not explained away.
NOTE
1. Wacome’s article addresses nonreductive physicalism specifically, but his thesis is that
nonreductive physicalist accounts offer no advantage over reductionist ones.
REFERENCES
Allen, Colin, and Marc Bekoff. 1997. Species of Mind: The Philosophy and Biology of Cognitive
Ethology. Cambridge: MIT Press.
Ayala, Francisco J., and Theodosius Dobzhansky, eds. 1974. Studies in the Philosophy of Biology:
Reductionism and Related Problems. Berkeley: Univ. of California Press.
Behe, Michael J. 1996. Darwin’s Black Box: The Biochemical Challenge to Evolution. New
York: Touchstone.
Bickle, John. 2002. “Multiple Realizability.” In The Stanford Encyclopedia of Philosophy (Fall
ed.), ed. Edward N. Zalta (http://plato.stanford.edu/archives/fall2002/entries/multiple-
realizability/).
Bielfeldt, Dennis. 2000. “The Peril and Promise of Supervenience for the Science-Theology
Discussion.” In The Human Person in Science and Theology, ed. Niels Gregersen, Willem
Drees, and Ulf Görman, 117–52. Grand Rapids, Mich.: Eerdmans.
Broad, C. D. 1929. The Mind and Its Place in Nature. New York: Harcourt, Brace.
Brown, Warren S., Nancey C. Murphy, and H. Newton Malony, eds. 1998. Whatever Happened
to the Soul? Scientific and Theological Portraits of Human Nature. Minneapolis:
Fortress.
Campbell, Donald T. 1974. “‘Downward Causation’ in Hierarchically Organised Biological
Systems.” In Studies in the Philosophy of Biology: Reductionism and Related Problems, ed.
Francisco Jose Ayala and Theodosius Dobzhansky, 179–86. Berkeley: Univ. of California
Press.
Gregory R. Peterson 711
Chalmers, David. 1997. The Conscious Mind: In Search of a Fundamental Theory. New York:
Oxford Univ. Press.
Churchland, Patricia Smith. 1986. Neurophilosophy: Toward a Unified Science of the Mind/
Brain. Cambridge: MIT Press.
Clayton, Philip. 1997. God and Contemporary Science. Grand Rapids, Mich.: Eerdmans.
———. 2004. Mind and Emergence: From Quantum to Consciousness. New York: Oxford
Univ. Press.
Dawkins, Richard. 1996. The Blind Watchmaker: Why the Evidence of Evolution Reveals a
Universe without Design. New York: W. W. Norton.
Dennett, Daniel. 1987. The Intentional Stance. Cambridge: MIT Press.
———. 1990. “The Interpretation of Texts, People and Other Artifacts.” Philosophy and
Phenomenological Research 50:177–94.
———. 1995. Darwin’s Dangerous Idea: Evolution and the Meanings of Life. New York:
Touchstone.
———. 2003. Freedom Evolves. New York: Viking.
Dretske, Fred. 1997. Naturalizing the Mind. Cambridge: MIT Press.
Glimcher, Paul. 2003. Decisions, Uncertainty, and the Brain: The Science of Neureconomics.
Cambridge: MIT Press.
Globus, Gordon G., Grover Maxwell, and Irwin Savodink. 1976. Consciousness and the Brain:
A Scientific and Philosophical Inquiry. New York: Plenum.
Goodenough, Ursula, and Terrence W. Deacon. 2003. “From Biology to Consciousness to
Morality.” Zygon: Journal of Religion and Science 38 (December): 801–19.
Gould, Stephen Jay. 2002. The Structure of Evolutionary Theory. Cambridge, Mass.: Belknap.
Green, Joel B., ed. 2004. What About the Soul? Neuroscience and Christian Anthropology.
Nashville: Abingdon.
Hasker, William. 1999. The Emergent Self. Ithaca, N.Y.: Cornell Univ. Press.
Haught, John. 2003. Deeper than Darwin: The Prospect for Religion in the Age of Evolution.
Boulder, Colo.: Westview.
Humphreys, Paul. 1997. “How Properties Emerge.” Philosophy of Science 64:1–17.
Kane, Robert. 1996. The Significance of Free Will. Oxford: Oxford Univ. Press.
Kaufmann, Stuart. 1995. At Home in the Universe: The Search for Laws of Self-Organization
and Complexity. New York: Oxford Univ. Press.
Kim, Jaegwon. 2000. Mind in a Physical World: An Essay on the Mind-Body Problem and
Mental Causation. Cambridge: MIT Press.
Koestler, Arthur, and J. R. Smythies, eds. 1969. Beyond Reductionism: New Perspectives in the
Life Sciences. London: Hutchinson.
Mayr, Ernst. 1988. Toward a New Philosophy of Biology: Observations of an Evolutionist.
Cambridge: Harvard Univ. Press.
Morgan, C. Lloyd. 1923. Emergent Evolution. London: Williams and Norgate.
Morowitz, Harold. 2002. The Emergence of Everything: How theWorld Became Complex. New
York: Oxford Univ. Press.
Murphy, Nancey. 1997. “Divine Action in the Natural Order: Buridan’s Ass and Schrödinger’s
Cat.” In Chaos and Complexity: Scientific Perspectives on Divine Action, ed. Robert John
Russell, Nancey Murphy, and Arthur R. Peacocke, 325–57. Berkeley, Calif.: Center for
Theology and the Natural Sciences (dist. Univ. of Notre Dame Press).
———. 1999. “Supervenience and Downward Efficacy of the Mental: A Nonreductive Physicalist
Account of Human Action.” In Neuroscience and the Person: Scientific Perspectives
on Divine Action, ed. Robert J. Russell, Nancey Murphy, Theo C. Meyering, and Michael
Arbib, 147–64. Berkeley, Calif.: Center for Theology and the Natural Sciences (dist.
Univ. of Notre Dame Press).
Peacocke, Arthur. 1993. Theology for a Scientific Age: Being and Becoming—Natural, Divine,
and Human. Enlarged ed. Minneapolis: Fortress.
———. 1999. “The Sound of Sheer Silence.” In Neuroscience and the Person: Scientific Perspectives
on Divine Action, ed. Robert J. Russell, Nancey Murphy, Theo C. Meyering,
and Michael Arbib, 215–48. Notre Dame, Ind.: Univ. of Notre Dame Press.
Peterson, Gregory R. 2002. Minding God: Theology and the Cognitive Sciences. Minneapolis:
Fortress.
Putnam, Hilary. 1960. “Minds and Machines.” In Dimensions of Mind, ed. S. Hook, 148–
79. New York: New York Univ. Press.
712 Zygon
Sharpe, Kevin, and Jonathan Walgate. 2003. “The Emergent Order.” Zygon: Journal of Religion
and Science 38 (June): 411–33.
Sober, Elliott, and David Sloan Wilson. 1998. Unto Others: The Evolution and Psychology of
Unselfish Behavior. Cambridge: Harvard Univ. Press.
Sperry, Roger. 1987. “Structure and Significance of the Consciousness Revolution.” The
Journal of Mind and Behavior 8:37–66.
Wacome, Donald H. 2004. “Reductionism’s Demise: Cold Comfort.” Zygon: Journal of
Religion and Science 39 (June): 321–37.
Whitehead, Alfred North. [1929] 1985. Process and Reality. New York: Free Press.