Novelty, Adaptive Capacity, and Resilience

Author(s): Craig R. Allen and C. S. Holling

Source: Ecology and Society , Sep 2010, Vol. 15, No. 3 (Sep 2010)
Published by: Resilience Alliance Inc.

Stable URL: https://www.jstor.org/stable/26268188

REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/26268188?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
This content is licensed under a Creative Commons Attribution-NonCommercial 4.0
International License (CC BY-NC 4.0). To view a copy of this license, visit
https://creativecommons.org/licenses/by-nc/4.0/.

is collaborating with JSTOR to digitize, preserve and extend access to Ecology and Society

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
Copyright © 2010 by the author(s). Published here under license by the Resilience Alliance.
Allen, C. R. and C. Holling 2010. Novelty, adaptive capacity, and resilience. Ecology and Society 15(3):
24. [online] URL: http://www.ecologyandsociety.org/vol15/iss3/art24/

Insight, part of a Special Feature on Catastrophic Thresholds, Perspectives, Definitions, and

Applications
Novelty, Adaptive Capacity, and Resilience

Craig R. Allen 1 and C. S. Holling 2

ABSTRACT. We present a conceptual framework that explores some of the forces creating innovation
and novelty in complex systems. Understanding the sources of variability and novelty may help us better
understand complex systems. Understanding complex phenomena such as invasions, migration, and
nomadism may provide insight into the structure of ecosystems and other complex systems, and aid our
attempts to cope with and mitigate these phenomena, in the case of invasions, and better understand and
or predict them. Our model is broadly applicable to ecological theory, including community ecology,
resilience, restoration, and policy. Characterizing the link between landscape change and the composition
of species communities may help policymakers in their decision-making processes. Understanding how
variability is related to system structure, and how that generates novelty, may help us understand how
resilience is generated. We suggest that there are three primary opportunities for the generation of novelty
into complex systems. These sources of novelty are inherent in the cross-scale structure of complex systems,
and are critical for creating adaptive capacity. Novelty originates from the inherent variability present in
cross scale structures, within scale reorganization associated with adaptive cycles, and whole-scale
transformations resulting from regime shifts. Although speculative, our ideas are grounded in research and
observation, and they may provide insight into the evolution of complex systems.

Key Words: adaptation, cross-scale, extinction, innovation, invasion, speciation

INTRODUCTION Ecosystems are organized by interactions among

biotic and abiotic processes operating at discrete, or
Understanding complex phenomena such as nearly discrete, scales (O’Neill et al. 1989). In
invasion, extinction, migration, nomadism, and hierarchy theory, different scales are referred to as
speciation may provide us with a better grasp of the levels of the hierarchy in an abstract sense. We use
structures of ecosystems and other complex the term scale as nearly analogous with hierarchical
systems as well as aid our attempts to better levels and relate scales of hierarchical structure to
understand and/or predict these phenomena. All of space and time domains in ecosystems. A domain
these phenomena represent novelty. Understanding of a scale is defined as the spatial extent of a structure
how variability is related to the structures of or process of interest and its temporal frequency.
complex systems and how that structure helps For example, forest gaps resulting from winds are
generate novelty may help us understand how relatively small in space but occur quite frequently,
ecological resilience is generated. Understanding whereas the destruction of entire forests by
ecological resilience and the capability of an hurricanes is a rare event but occurs over large
ecological system to absorb disturbance without spatial extents. Within scales, biotic and abiotic
collapsing and reorganizing into a different interactions reinforce one another, creating
ecological state (Gunderson et al. 2010) is critical persistent structures and scale-specific patterns. For
to humanity given rapid landscape and climate example, fires in longleaf pine (Pinus palustris)
change. Here we present ideas that we hope provide ecosystems of the southeastern USA promote the
insight into the evolution of complex behaviors and regeneration of longleaf pine, which in turn further
the emergence of resilience and adaptive capacities. promotes fires by dropping highly flammable leaf
litter. Across scales, different patterns and processes

1
U.S. Geological Survey, Nebraska Cooperative Fish and Wildlife Research Unit, 2Department of Zoology, University of Florida

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

dominate and are only loosely coupled with interact. Heightened variability at the species,
processes at higher or lower scales (Holling 1992, population and community levels has been observed
Peterson et al. 1998). Together, the suite of abiotic at those transitions between scales (Allen et al. 1999,
processes interacting with biotic elements produce Allen and Saunders 2002, 2006, Gunderson et al.
loosely structured hierarchical systems with 2007, Skillen and Maurer 2008, Wardwell and Allen
emergent qualities such as resilience. Reinforcement 2009), as indicated by phenomena such as invasion,
and inhibition among interacting processes drive extinction, nomadism, and migration. We believe
this organization. The partitioning of process, this reflects heightened variability at transitions
structure and function within and across scales between structures discernable at specific scales and
provides resilience to complex systems. that variability provides opportunities for and
generates novelty.
The changes in structures and patterns with changes
in scale in complex ecological systems provide Here, we offer a conceptual framework that
different templates at different scales with which addresses the forces that create novelty in complex
biota may interact. Within a scale, species strongly systems. We explicitly consider the generation of
interact, and the result is a diversity of species’ novelty in ecological systems, but believe the
lifestyles and thus functions, while actual and general concepts are appropriate to other complex
potential competition among species is reduced systems, such as societies, for which we might
(Peterson et al. 1998, Fischer et al. 2007, Wardwell consider innovation as well as novelty. Although
et al. 2008). This is simply because species with novelty and innovation are recognized as critical to
similar life spans and step lengths that live in the adaptation, the generation thereof is rarely
same ecological system are more likely to come discussed. It is, however, critical to consider
into contact with each other than with species that landcover and landuse change on Earth in this era
are considerably larger or smaller. Competition of rapid climate change. We believe that the
among species that exploit the same or similar generation of novelty, and hence adaptive capacity,
resources (i.e., members of the same functional is critical for maintaining resilience in complex
group) is reduced if resources are segregated by systems under stress. Therefore, we address how the
scale, thereby lessening the potential for generation of novelty enhances resilience. Our
competitive exclusion. In other words, species that model is broadly applicable to ecological theory,
exploit the same resource in the same or a similar resilience theory, community ecology, restoration,
manner are more likely to coexist if they are of and policy. Understanding the sources of variability
different body sizes. Species that exploit the same and novelty may help us better understand complex
resource in similar ways are members of the same systems.
functional group, and redundant function is
certainly present among species, but the co-
existence of ‘redundant’ species is facilitated when CROSS-SCALE ORGANIZATION IN
those seemingly redundant species live at different COMPLEX SYSTEMS
ecological scales. This produces a reinforcement of
functions across scales. The function of seed An ecosystem is the product of non-linear
dispersal, for example, is present at a wide range of interactions among its component parts. If all parts
scales in ecological systems, ranging from the of an ecosystem interacted equally, ecosystems
dispersal of a few meters by ants to the dispersal would be incredibly complicated and impossible to
across many kilometers by mammals such as tapirs. understand. However, ecosystems are complex
Such an arrangement of function within and across systems; the way in which they are organized makes
scales, regardless of origin, provides a robust them less complicated and more easily understood,
response to a variety of perturbations, especially to at least in an abstract sense. As complex adaptive
perturbations that tend to scale up, such as insect systems, ecosystems possess emergent properties
outbreaks (Peterson et al. 1998). Changes in such as resilience and discontinuous structures that
domains of scale are characterized by distinct scale vary across scales. This cross-scale structure has
breaks, reflecting abrupt changes in pattern and been recently described as a panarchy, a nested set
structure. Ants and tapirs live in the same of adaptive cycles with clearly differentiated
environments but do not interact with the same structures across scales. An adaptive cycle (Holling
structures, and there is not a continuous transition and Gunderson 2002) describes the process of
between the structures with which ants and tapirs development and decay in a system (Figure 1) and

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

is generally “self-organizing” because positive A given level may not be very resilient, but the larger
interactions reinforce its structures. The process system may be. Unlike the top-down control
that adaptive cycles describe is similar to the envisioned in traditional hierarchies, connectivity
process of succession, but the concept is important between adaptive cycles in a panarchy can be from
because a nested set of adaptive cycles each levels above or below.
occurring at a specific scale comprises a panarchy.
In an adaptive cycle, the initial stage of
development is of short duration and consists of a
rapid exploitation and garnering of resources by CROSS-SCALE STRUCTURE AND
system components. This stage has been termed the DISCONTINUITIES
r stage or function and is a period of rapid growth
in pioneering or early successional species. The r The structure of complex systems is strongly self-
stage is followed by the k stage or function, a stage organizing and may be quite conservative. This
of longer duration that is characterized by the structure provides the core ‘memory’ of a system in
accumulation of biomass or other system elements that its structure is unlikely to change. The
or energies as well as increasing connectivity and components of complex systems such as ecosystems
rigidity. Increasing connectivity and rigidity during interact to create conservative structures in time and
the k phase leads to decreased resilience and space – interactions that are reinforced persist,
eventual collapse. This stage of collapse, the whereas those that are not fade away. This is
omega, is rapid and unleashes the energy important for humans because complex systems
accumulated and stored during the k phase. The such as ecosystems often remain apparently more
omega phase is triggered, for example, by forest or less stable; thus, we can expect reasonably
fires or pest outbreaks. Collapse during the omega predictable dynamics and the relatively constant
phase is followed by reorganization during the provision of ecological goods and services. This
alpha phase, a relatively rapid period of the conservativeness and self-organization is due in part
assembly of components, analogous to the pioneer to the interaction of biotic and abiotic elements.
stage in ecosystems or a system under high gain Animals interact with the ecological structure that
(Allen et al. 2001). provides a distribution of necessary resources such
as food and space that they can successfully exploit
Adaptive cycles do not exist in isolation. Adaptive in space and time. In exploiting their environments,
cycles operate over limited ranges of scale, whereas animals often change ecological structures in ways
complex systems are characterized by a rich array that are favorable for themselves. For example,
of scales, each scale exhibiting characteristic large herbivores can alter the dynamics of
structures and dynamics. A panarchy is a nested set succession (and competition among grasses, bushes
of adaptive cycles (Figure 2). An ecosystem and and trees) such that the habitat is, in some sense of
other complex systems can be conceptualized as a the word, optimal for them. Self-organization
panarchy. For resilience theory, it is critical to involves other biotic system elements as well. For
understand the scales of interest and the scale of example, many grasses worldwide are pyrophilic
analysis because one level of a panarchy may and, therefore, highly flammable (Brooks et al.
collapse and cascade to lower levels, but the system 2004). In the absence of fire, succession would often
as a whole may be maintained. For example, at very eliminate these grasslands from the ecosystems they
small space and time scales, the leaves of trees occupy. However, the presence of these grasses
exhibit an adaptive cycle with an annual encourages fire, which favors their spread and
periodicity; forest stands also exhibit adaptive excludes competitors. In the absence of fire,
cycles, with a decadal or greater periodicity and ecosystems such as the longleaf pine savannah in
relatively large spatial extent. A forest fire that the Southeastern United States rapidly transform to
“resets” a forest stand starts that stand at the omega oak forest (Peterson 2002). Strongly interacting
phase of an adaptive cycle, but it does not species have been termed keystone (Mills et al.
necessarily affect larger scale structures, such as 1993) or driver species (Walker 1995). However,
the landscale in which the forest stand is embedded. we note the absence of scale in such definitions. Any
Each adaptive cycle operates over a discrete range keystone species (or process), because it affects a
of scales in both time and space and is connected system only over a limited range of scales, is a
to adjacent levels (adaptive cycles). Resilience is a keystone only at the scale in which it interacts
property that can exist at any scale in a panarchy. (unless it has truly cross-scale impacts, which is

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Fig. 1. The adaptive cycle. A representation of the four ecosystem functions (r, k, omega, alpha) and the
flow of events among them. The arrows show the speed of the cycle where short, closely spaced arrows
indicate a slowly changing state and long arrows indicate a rapidly changing state. The cycle reflects
changes in two properties: (1) Y axis—the potential that is inherent in the accumulated resources of
biomass and nutrients; (2) X axis—the degree of connectedness among controlling variables. Low
connectedness is associated with diffuse elements loosely connected to each other whose behavior is
dominated by outward relations and affected by outside variability. High connectedness is associated
with aggregated elements whose behavior is dominated by inward relations among elements of the
aggregates, relations that control or mediate the influence of external variability. The exit from the cycle
indicated at the left of the Figure suggests where the potential can leak away or where a change of state
into a less productive and organized system is likely. From Panarchy: Understanding Transformations in
Human and Natural Systems L.H. Gunderson and C.S. Holling, eds. Copyright © 2001 by Island Press.

unlikely given current definitions of keystone discontinuity (Garmestani et al. 2009). Discontinuities
species). Because ecosystems are characterized by are rooted in the separation between levels of a
a rich array of scales, there may be a rich array of panarchy. Different adaptive cycles and different
keystone species present, and dominant interactions structuring processes are separated from one
at each scale are crucial. Furthermore, a keystone another by gaps in the domains of scale that they
or driver species often has a transient role; as system occupy, often an order of magnitude or more
or community dynamics change, the identity of key (Holling and Gunderson 2002). This separation has
species changes (Walker 1995, Walker et al. 1999). several important effects. First, as stated above, it
Self-organization is important not only for the means that variables within systems are distributed
provision of goods and services to humans but discontinuously. Second, it indicates that self-
because it means that understanding ecosystems organizing interactions and processes, such as
and other complex systems is at least somewhat community-level interactions for animals (including
tractable. Ecosystem structures and functions, interactions such as competition), are compartmentalized
though dynamic through time, are largely constant by scale. Therefore, similarly sized animals are
and educible at human time scales. The exception, more likely to strongly interact than animals of
of course, is when the resilience of systems is grossly different sizes, although exceptions occur
exceeded and rapid transformations occur. (for example, with predation). The compartmentalization
of systems along an axis of scale provides rich
Because adaptive cycles and self-organization opportunities for experimentation within levels, and
occur at discrete scales within a system, ecosystems this leads to the development of high levels of
and other systems are characterized by diversity within systems (O’Neill et al. 1986). This

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Fig. 2. Figure 2. A panarchy. Three selected levels of a panarchy are illustrated, to emphasize the two
connections that are critical in creating and sustaining adaptive capability. One is the "revolt"
connection, which can cause a critical change in one cycle to cascade up to a vulnerable stage in a larger
and slower one. The other is the "remember" connection, which facilitates renewal by drawing on the
potential that has been accumulated and stored in a larger, slower cycle. The number of levels in a
panarchy varies, is usually rather small, and corresponds to levels of scale present in a system.
Excerpted from Panarchy: Understanding Transformations in Human and Natural Systems L. H.
Gunderson and C. S. Holling, eds. Copyright © 2001 by Island Press.

results in a distributional pattern of function in separated by gaps (discontinuities; Allen and

which diversity is high within scales and is repeated Holling 2008). The body mass aggregations are
(i.e., the function contained within is redundant) at thought to correspond with the scales of structure
different scales, a pattern that adds to the resilience available in a system, and the gaps represent
of ecosystems (Peterson et al. 1998) and other transitions (scale breaks) between the available
complex systems (Garmestani et al. 2006). The ranges of scales. Discontinuities, or gaps, in
presence of discrete scales of pattern and process variables in complex systems have been described
in complex systems creates discontinuities, or scale as scale breaks where highly variable and
breaks, between ranges of scales. Theory and recent unpredictable behavior is expected (O’Neill et al.
empirical analysis suggest that these scale breaks 1989, Allen et al. 1999). Examples of high
generate novelty and innovation as a result of the variability at scale breaks in ecosystems include the
variable dynamics at these transitions. success of invasive non-indigenous species, the
failure or extinction of native species and an
association with migratory and nomadic species.
VARIABILITY AT SCALE BREAKS
The association of invasive species with scale
The body mass of vertebrates is strongly allometric breaks has been documented in a variety of
with many ecological attributes of species; ecosystems and for both birds and mammals (Allen
therefore, it is a useful proxy for the scale at which et al. 1999, Allen 2006). The association has been
an animal interacts with its environment. Body most thoroughly investigated for the Everglades
mass distributions of vertebrates from ecosystems ecosystem of Florida, where invasive species tend
are discontinuous and consist of body mass to occur at the edge of body mass aggregations. That
aggregations of species with similar body mass association is significant because it links an

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

independent biological attribute, invasiveness, with represent unpredictable or annual turnover in

a particular location on a body mass axis, at the edge community composition and are also novel and
of discontinuities. It is strengthened by the finding poorly understood behaviors, in an evolutionary
that declining species are also associated with scale sense. Both allow the exploitation of resources at a
breaks. The significance of the association is further level that could not otherwise be achieved, at least
strengthened because the analysis of successfully not without some other novel approach to secure
versus unsuccessfully introduced avian species in resources that vary so strongly in both time and
the Everglades demonstrates that introduction space. In Mediterranean-climate Australia, near the
success is best predicted by distance from the edge city of Adelaide, the climate is highly variable, and
of a discontinuity and not by intrinsic traits of the avifauna has among the highest, perhaps the
species or communities. Introduced species whose highest, incidence of nomadism in the world. There,
body mass places them close to the edge of a body nomadism is best predicted by a combination of
mass aggregation are more likely to become body mass, nectivory and proximity to
established than an introduced species whose body discontinuity. Birds that are bigger, feed on nectar
mass places it toward the center of an aggregation and have body masses that place them closest to
(Allen 2006). Other potential predictors of discontinuities are most likely to be nomadic (Allen
introduction success based on intrinsic or and Saunders 2002, 2006, but see Woinarski 2006).
community-level hypotheses of success fail in that Similarly, migrant birds have body masses that tend
continental data set. It should be noted that despite to place them close to discontinuities in body mass
a flurry of books and articles focusing on invasive distributions. Weeks and Allen (unpublished data)
species over the last 15 years, invasion biology is investigated the relationship between migrant
a discipline that is in its infancy, largely because species and body mass distributions for the migrant
the ability to experiment is lacking and because bird species that breed in South Carolina, USA, but
inference is largely based on positive cases that annually depart for the winter. They did the
(successful introductions or invasions). There is same for species that migrate to Costa Rica (La
limited information on what species were Selva) during the winter but that breed elsewhere.
unsuccessful in becoming established, although They found that in both South Carolina and Costa
there are exceptions (Allen and Starr 1988). One of Rica, migrants tend to have body masses proximate
those exceptions is the south Florida avifauna, to discontinuities. Interestingly, the South Carolina
where we, with some surety, know the pool of migrants that winter in Costa Rica, which have body
unsuccessfully introduced species. The Everglades masses close to scale breaks in the body mass
provides an additional clue to the link between distribution of South Carolina resident birds, are not
ecological structure and novelty. Declining species close to scale breaks in the body mass distribution
comprise approximately 25% of the fauna in three of Costa Rica birds. (South Carolina migrants are a
vertebrate taxa (mammals, birds and herpetofauna; small subset of the migrant species present in Costa
Forys and Allen 1999), and non-indigenous species Rica.) Analysis of migrant South African birds also
comprise a further 25%. Forys and Allen (2002) suggests an association with discontinuities (Alai
examined historic (no non-indigenous species), 2010). This suggests that their existence near
current (non-indigenous and declining species discontinuities and their variable resources
included) and hypothetical future (non-indigenous represent an unexploited, but risky, opportunity. It
species included, declining species eliminated) may be overly risky to exploit ‘at the edge’ on both
body mass and functional patterns for mammals, the wintering and summering grounds.
birds and herpetofauna of the Everglades. They
found that neither the distribution of function nor
overall body mass pattern was substantially NOVELTY AND INNOVATION
changed by the large species turnover in their time
series. Body mass pattern was conserved, and the Novelty and innovation are required for systems to
large species turnover was primarily limited to remain dynamic and functioning. Without
areas of discontinuity. innovation and novelty, systems may become over-
connected and dynamically locked, and the capital
Species invasions and extinctions represent therein may be unavailable (Gunderson and Holling
turnover in animal community composition, but 2002). Novelty and innovation are required to keep
there is other documented variability at scale breaks existing complex systems resilient and to create new
in animal communities. Migration and nomadism structures and dynamics following system crashes.

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

This is true in all complex systems, and the either locally or globally novel. Locally novel
importance of novelty is recognized as much (if not additions are novel to that particular system. For
more) in the management and business world as it example, the addition of an invasive species to an
is in scientific fields. ecosystem adds novelty to that system, but the
invasive species was in existence prior to its
Speaking of management hierarchies, Pierce and invasion of a new ecosystem. On the other hand,
Delbecq (1977) described the organizational globally novel additions did not previously exist and
elements required for innovation. One of those are new not only to the particular system within
elements is differentiation, which is necessary for which they are generated or added but are also new
the initiation of innovation. We find differentiation to the globe. Speciation within a system represents
present in ecological systems in many forms, the addition of global novelty.
genotypically, phenotypically and functionally,
and this is paralleled in social and social-ecological
systems. Decentralization is another important Background
element described by Pierce and Delbecq. It is a
key component of complex systems and is related Novelty is generated as a result of the normal
to Pierce and Delbecq’s concept of stratification, dynamics of complex systems. In terms of a
which we interpret as levels within a hierarchy. panarchy with a discontinuous structure, novelty is
Pierce and Delbecq (1977) also discuss contextual generated at the edge of scale breaks (at the
attributes of innovation, specifically environmental transitions between domains of scale) as a result of
uncertainty, larger size and age. Larger hierarchies the highly variable distribution and occurrence of
have more opportunities for innovation, and older resources in space and time, which in turn is
hierarchies are less open to innovation. Clearly, reflected in the high variability in biotic components
novelty is important for the maintenance and health of the system (e.g., Allen et al. 1999, Skillen and
of a wide variety of systems—ecological, social, Maurer 2008). This generation of novelty creates
cultural, and combinations thereof. Novelty and options for systems, is critical in maintaining
innovation are needed to maintain not only adaptive capacity, and serves as a reservoir of
maintenance functions, but also the adaptive potential functions that may be required following
capacity of systems and to allow complex systems transformations or as normal system dynamics
the latitude to ‘explore’ alternative structures and evolve. In the thermodynamics and gain literature,
dynamics, that is, to evolve. Below, we describe the this has been termed internal complexification
generation of novelty as related to the structure of because it builds upon extant structures (Tainter
discontinuous, panarchically organized complex 1990, Tainter et al. 2003). Such novelty is at the
adaptive systems. heart of resilience.

TYPES OF NOVELTY Incremental

The generation of novelty and innovation is a Self-organizing processes among the biotic and
characteristic of dynamic complex systems. It is abiotic elements of complex systems, such as
generated at all levels. For example, in biological ecosystems, add complexity over time. Some of that
systems, it is generated at the genetic level through complexity is added during the r and k stages of an
random processes of mutation, at the species level adaptive cycle in the form of new connections, new
through the selective processes of evolution, at the functions and new arrangements of elements.
community level as a result of assortment and However, new levels (new adaptive cycles) can also
changes in the species pool, and at the ecosystem be added during r stages. The addition of new layers
level as a result of changes in key driving processes to a hierarchy adds new scales of opportunity for
and self-organizing (reinforcing) interactions. elements of complex systems such as species or
firms. In a thermodynamic sense, the emergence of
There are fundamental similarities around which a new layer of structure in a hierarchy results from
the concept of novelty can be organized. We self-organization with a steep thermodynamic
organize novelty into three types: background, gradient, termed external complexification because
incremental and punctuated. These are discussed it results partially from an external gradient (Tainter
below. All three types of novelty generation can be et al. 2003). The addition of new levels of adaptive

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

cycles may make a complex system more resilient innovation and novelty may be a double-edged
and thus less prone to cross-scale collapse (i.e., the sword in some circumstances. In ecosystems, for
collapse of all levels of a panarchy at once, example, in addition to being a cause of major
including the top level, termed a regime shift or a extinctions, ’innovation’ is also the prime source of
flip to an alternative state)(Allen et al. 2005). recovery (Jain and Krishna 2002).

Punctuated THE GENERATION OF NOVELTY AT

SCALE BREAKS
When the resilience of a complex system is
exceeded, a system may collapse, and all levels of The novelty generated at scale breaks, those regions
a panarchy may experience the omega phase between adaptive cycles in a panarchy, is critical
simultaneously (Gunderson and Holling 2002). for complex systems. Furthermore, these regions
Novelty may be added to or introduced to a system may represent ‘novelty pumps,’ regions of the actual
during reorganization. Here the novelty added (for and potential production of innovation and novelty.
example, new animals to a community or As discussed above, highly variable phenomena are
ecosystem) may be local or global. While this type associated with discontinuities in animal body mass
of transformation and novelty generation is not distributions. Body mass distributions reflect the
likely to spawn globally novel elements, it does scales of available structures in an ecosystem, which
provide opportunities for globally novel arrangements in turn reflect the panarchical structure of the system
of elements. The addition of novelty in a punctuated (Figure 3). Thus, discontinuities in body mass
manner may also occur within adaptive cycles, patterns reflect the location of scale breaks in
when individual cycles enter the omega – alpha ecological structure. A similar structure is also
phases, and this may also build overall resilience present in other complex systems, such as regional
over time and in response to changing conditions. urban systems (Garmestani et al. 2005) and regional
economic systems (Garmestani et al. 2006).

NOVELTY AND RESILIENCE AND Phenomena such as invasion, extinction, nomadism,

ADAPTIVE CAPACITY and migration in animal communities reflect high
variability, but also represent the creation or
Adding levels to a panarchy (or other incremental insertion of novelty. Invasive species have subtly or
novelty), the generation of novelty at scale breaks grossly different ways of interacting with their new
(background), and punctuated novelty may all build environments, as compared to native species, and
resilience in systems. The novelty generated at scale their addition may reflect a system in transition
breaks, and as importantly, the potential for the (Allen et al. 1999). Their addition may not alter, but
generation of novelty at scale breaks, builds rather reinforce existing ecological organization
adaptive capacity in complex systems. Jain and (Forys and Allen 2002) and thus build resilience, or
Krishna (2002) documented how dormant they may be destructive and transformative forces.
innovations in complex graph networks can
takeover system dynamics at times when other We assert that the generation of novelty at scale
dominant elements become weak (i.e., when the breaks results from a high variability in resources
resilience of the system is diminished). Having a (Figure 4; O’Neill et al. 1989, Allen et al. 1999).
constant source of innovation and novelty is clearly While high variation in resource abundance and
important for systems, both following transformations location in space and time is a hardship for some
and during their normal dynamics, if a system is not species (see, for example, the propensity of
to become over-connected and over-capitalized. declining species to have body masses proximate to
Without a continual source of novelty, complex discontinuities (Allen et al. 1999, Skillen and
systems such as ecosystems cannot be adaptive or Maurer 2008)), it is an opportunity for other species
dynamic. Scale breaks are a key source for such that successfully invade and exploit these locations/
innovation. It should be noted, however, that resource or that develop novel and innovative
innovation and novelty may be destructive forces behaviors (Figure 5). However, a strategy that
as well. Invasive species, for example, can alter focuses on resources that are highly predictable,
basic process and structure in ecosystems and be a especially when the structure of a system is dynamic
source of collapse and transformation. Thus, and when the location of scale breaks may shift over

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Fig. 3. Discontinuous body mass distributions (black circles on x-axis representing animal body masses
along a body mass axis) reflect the levels of scale (grey circlular regions, representing space-time
domains of structures) available in a given system, which in turn reflect the underlying panarchy. The x
axis represents both spatial scale of processes and a body mass axis for species inhabiting the system.

time, is likely not an optimal long-term strategy. basic similarities. In each case, innovation was
However, it is the best strategy when the stable constructed through an evolutionary triad of
resources far from a discontinuity are effectively challenge, potential and opportunity. Each of these
sequestered by others and niches are saturated. conditions is met within a panarchy, specifically at
the discontinuities that define regions between
Mutations and other novelty that occur as levels (scale breaks).
background have little chance of success in the
center of an aggregation, far from discontinuities Evolutionary innovation is the acquisition of novel
and scale breaks. At the center of an aggregation, morphologies and/or behaviors that open new
resources are stable and thoroughly exploited and niches, providing new ways to successfully exploit
competition is intense. However, at discontinuities, the environment. In ecosystems, an extreme case of
the high fluctuation in resources is more likely to the generation of novelty is represented by
lead to the success of random mutations. Over time, speciation. Scale breaks may offer the opportunity
species or strategies successful at the edge may for parapatric or sympatric speciation. However, the
migrate into the center of an aggregation, where isolation is not due to linear distance per se, as in
resources are more stable and secure (Figure 6). parapatric speciation (Knapp and Mallet 2003), any
This is analogous to the Taxon Cycle in more than it is geographic barriers. Rather, it is
biogeography (Wilson 1961). Many of the major isolation driven by the interaction with the
innovations in the history of life, such as the spread environment at different ranges of scale. This may
of novel metabolic activities in the first billion years strike some as unlikely. However, consider an
of Earth’s history, the spread of photosynthesis, the extreme mammal example: least shrews (Cryptotis
development of multi-cellular organisms, and the parva) and moose (Alces alces) in a boreal
spread of life to terrestrial ecosystems, all share ecosystem. Clearly, these animals are sympatric, yet

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Fig. 4. The relationship between discontinuities, variability and stability. Body size distributions of
animals are characterized by aggregations and discontinuities (upper graphic, circles representing
species). Aggregations reflect domains of scale and discontinuities reflect scale breaks. Within domains
of scale, resource variation is low (lower graphic, dotted line), and at scale breaks resource variation is
high. Stability (solid line) is inversely related to variability and is high within scales and low at scale
breaks.

they have little (if any) interaction. The life span of population and community level at scale breaks
a least shrew is less than two years, and that of a hints of potential individual phenotypic and
moose is more than twenty. The resources and genotypic variability.
structure that a shrew interacts with during the
process of its life are, we believe, viewed as noise Despite a substantial volume of papers on speciation
to the moose. Similarly, the structures and from field, laboratory and theoretical perspectives,
processes that a moose interacts with are slow the process leading to the creation of new species
relative to the life span and step size of a shrew and, remains poorly understood (Sepkoski 1998). While
therefore, they are simply background to the shrew. sympatric speciation is (may be) controversial for
Sympatric speciation would not lead to such some (Via 2001), habitat choice has been shown to
extreme size divergence quickly. However, scale- produce assortative mating and sympatric
dependent isolation would likely lead to increasing speciation under disruptive selection (Rice 1984,
size divergence between closely related species. Kondrashov and Mina 1986, Rice and Salt 1990),
Life at discontinuities means constant adaptation to including selection on traits associated with
fluctuating resources. The high variability inherent competition or predation (Schluter 1996, Via 2001).
at scale breaks, and more stable opportunities for Sympatric speciation can arise from reproductive
resource acquisition at higher or lower scales, isolation associated with adaptation to alternative
provides ample opportunity for differentiation of resources or habitats (Turelli et al. 2001). Scale
lineages within a species (by habitat, by scale, by segregation may produce ecological speciation by
resource, or by a combination thereof). Because isolating two populations based on their scale of
species at discontinuities are at the edge of an habitat use; this may be sympatric or parapatric. The
available range of scale, it is perhaps possible for rebound of species diversity (increased speciation
these species to shift between scales (Allen and rates) following mass extinctions (Seposki 1998)
Saunders 2002). This provides an opportunity for suggests that incipient species are always present
sympatric speciation based on scale. Furthermore, but the ‘success’ of species is greater following such
the biological variability witnessed at the events. Thus, a novelty pump, as it were, may be

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Fig. 5. Novelty and innovation mirror discontinuities and resource variability in complex systems, and
are high at the edge of discontinuities.

extremely important in providing a source APPLICATIONS

following such events. Niche construction
(lineages modifying their environments to At the molecular, community, ecosystem, and other
construct their own ecological role and levels, novelty is constantly created and
consequently construct niches for other species) extinguished. Scale breaks provide a robust pump
seems to be especially important following mass for the generation of novelty, and thus a key source
extinctions. Why? Panarchy theory provides a of adaptive potential following transformation, and
possible explanation. We cannot consider species adaptation to changing conditions such as rapid
as passive elements within an ecosystem or climate change.
landscape, but rather as critical engineers of their
environment. Species interact in an often Global climate change results in rapid transformations
reinforcing manner with abiotic structure and in the organization of the complex systems that we
process, traveling upon a shared trajectory, where inhabit, we create, and we rely upon. The novelty
a change by either the biotic or by the abiotic pump inherent in the structure of complex
elements is equally important. Thus, mass panarchies is necessary for adaptation to changing
extinctions either cause or reflect systems that have environments (for example, species undergoing
been transformed. The large loss of species means long-range migration or nomadic movements may
a disruption in self-organizing processes that are be better able to cope with global climate change).
responsible for structure in complex systems. The
loss of elements responsible for self-organizing Species in the center of aggregations may have
processes with abiotic elements leads to the ‘built’ niches that are stable. Given the conservative
generation of new inter-relationships, new self- nature of the structure of body mass distributions
organizing processes, and thus keystone (or niche- (Havelcik and Carpenter 2001, Forys and Allen
building) species. 2002) and the high degree of resilience in many
systems, this is adaptive for most circumstances. It
is analogous to the K phase of the adaptive cycle.
However, when the resilience of a system is
exceeded and it transforms during other
transformative events at one or more scales, this
strategy is apt to be ‘brittle’ and vulnerable to

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Fig. 6. The novelty cycle in discontinuous panarchical systems is analogous to the taxon cycle described
by Wilson (1961). Variability in resources (line above the x axis that represents the discontinuous body
mass distribution of species) is highest at scale breaks, as in Figure 5. In time step one, novel behaviors
or species (black) are added at scale breaks. As time progresses (time step two) the novel behaviors or
species have ‘migrated’ towards the center of body mass aggregations where resources are less variable.
By step three, this migration has completed. In time steps two and three new novelty is being introduced
at the edges (grey). This diagram does not show the increased extinction rates also expected to be
associated with scale breaks.

failure. On the other hand, the novel strategies Responses to this article can be read online at:
generated at scale breaks are likely to become http://www.ecologyandsociety.org/vol15/iss3/art24/
responses/
dominant during such systemic crises.

Continuously pumping out novel solutions to

current or potential challenges ensures a Acknowledgments:
maximization of energy/resource use within a
system without reorganizing the system. It also The Nebraska Cooperative Fish and Wildlife
allows the system to be dynamic both in its internal Research Unit is jointly supported by a cooperative
structure and connectivity and in its relationship agreement between the United States Geological
with other systems. Most of the novel solutions Survey, the Nebraska Game and Parks Commission,
created by mutation and other sources do not have the University of Nebraska - Lincoln, the United
a challenge to respond to, and thus are conceived States Fish and Wildlife Service and the Wildlife
of as failures. However, when challenges arise, it Management Institute. Support was provided to
is critical to have a solution available. Thus, CRA by the James S. McDonnell Foundation 21st
generation of novelty at scale breaks helps ensure Century Research Award/Studying Complex
that the evolutionary potential of both species and Systems. An earlier version of this manuscript was
systems is maintained. improved by comments from T.F.H. Allen and two
anonymous reviewers.

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

LITERATURE CITED bird communities in regions with different land use

intensities. Ecosystems 10:964-974.
Alai, A. 2010. The textural discontinuity hypothesis
and its relation to nomadism, migration, decline, Forys, E. A., and C. R. Allen. 1999. Biological
and competition. Thesis, University of Nebraska, invasions and deletions: community change in south
Lincoln, Nebraska, USA. In press. Florida. Biological Conservation 87:341-347.

Allen, C. R. 2006. Predictors of introduction Forys, E. A., and C. R. Allen. 2002. Functional
success in the South Florida avifauna. Biological group change within and across scales following
Invasions 8:491-500. invasions and extinctions in the Everglades
ecosystem. Ecosystems 5:339-347.
Allen, C. R., E. A. Forys, and C. S. Holling. 1999.
Body mass patterns predict invasions and Garmestani, A. S., C. R. Allen, and K. M. Bessey.
extinctions in transforming landscapes. Ecosystems 2005. Time series analysis of clusters in city size
2:114-121. distributions. Urban Studies 42:1507-1515.

Allen, C. R., L. Gunderson, and A. R. Johnson. Garmestani, A. S., C. R. Allen, and L.

2005. The use of discontinuities and functional Gunderson. 2009. Panarchy: discontinuities reveal
groups to assess relative resilience in complex similarities in the dynamic system structure of
systems. Ecosystems 8:958-966. ecological and social systems. Ecology and Society
14(1): 15. [online] URL: http://www.ecologyandso
Allen, C. R. and C. S. Holling. 2008. ciety.org/vol14/iss1/art15/.
Discontinuities in ecosystems and other complex
systems. University of Columbia Press, New York, Garmestani, A. S., C. R. Allen, J. D. Mittelstaedt,
New York, USA. C. A. Stow, and W. A. Ward. 2006. Firm size
diversity, functional richness and resilience.
Allen, C. R., and D. A. Saunders. 2002. Environment and Development Economics 11:
Variability between scales: predictors of nomadism 533-551.
in birds of an Australian Mediterranean-climate
ecosystem. Ecosystems 5:348-359. Gunderson, L., C. R. Allen, and C. S. Holling.
2010. Foundations of ecological resilience. Island
Allen, C. R., and D. A. Saunders. 2006. Multi- Press, New York, New York, USA.
model inference and the understanding of
complexity, discontinuity and nomadism. Ecosystems Gunderson, L., C. R. Allen, and D. Wardwell.
9:694-699. 2007. Temporal scaling in complex systems:
resonant frequencies and biotic variability. Pages
Allen, T. F. H., and T. B. Starr. 1988. Hierarchy: 78-89 in J. A. Bissonette, and I. Storch, editors.
perspectives for ecological complexity. University Temporal dimensions in landscape ecology: wildlife
of Chicago Press, Chicago, Illinois, USA. responses to variable resources. Springer, New
York, New York, USA.
Allen, T. F. H., J. A. Tainter, J. C. Pires, and T.
W. Hoekstra. 2001. Dragnet ecology, “just the Gunderson, L. H., and C. S. Holling. 2002.
facts Ma’am: the privilege of science in a post- Panarchy: understanding transformations in
modern world. BioScience 51:475-485. human and natural systems. Island Press,
Washington, D.C., USA.
Brooks, M. L., C. M. D'Antonio, D. M.
Richardson, J. B. Grace, J. E. Keeley, J. M. Havlicek, T. D., and S. R. Carpenter. 2001.
DiTomaso, R. J. Hobbs, M. Pellant, and D. Pyke. Pelagic species size distributions in lakes: are they
2004. Effects of invasive alien plants on fire discontinuous? Limnology and Oceanography 46
regimes. Bioscience 54:677-688. :1021-1033.

Fischer, J., D. B. Lindenmayer, S. P. Blomberg, Holling, C. S. 1992. Cross-scale morphology,

R. Montague-Drake, A. Felton, and J. Stein. geometry, and dynamics of ecosystems. Ecological
2007. Functional richness and relative resilience of Monographs 62:447-502.

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Rice, W. R., and G. Salt. 1990. The evolution of

Holling, C. S., and L. H. Gunderson. 2002. reproductive isolation as a correlated character
Resilience and adaptive cycles. Pages 25-62 in L. under sympatric conditions: experimental evidence.
H. Gunderson, and C. S. Holling, editors. Evolution 44:1140-1152.
Panarchy: understanding transformations in
human and natural systems. Island Press, Schluter, D. 1996. Ecological causes of adaptive
Washington, D.C., USA. radiation. American Naturalist 148:S40-S64.

Jain, S., and S. Krishna. 2002. Large extinctions Sepkiski, J. J., Jr. 1998. Rates of speciation in the
in an evolutionary model: the role of innovation and fossil record. Philosophical Transactions of the
keystone species. Proceedings of the National Royal Society of London B 353:315-326.
Academy of Science 99:2055-2060
Skillen, J. J., and B. A. Maurer. 2008. The
Knapp, S., and J. Mallet. 2003. Refuting refugia? ecological significance of discontinuities in body
Science 300:71-72. mass distributions. Pages 193 – 218 in Allen, C. R.
and C. S. Holling, editors. Discontinuities in
Kondrashov, A. S., and M. V. Mia. 1986. ecosystems and other complex systems. Columbia
Sympatric speciation: when is it possible? University Press, New York, New York, USA.
Biological Journal of the Linnean Society
27:201-223. Tainter, J. A. 1990. The collapse of complex
societies. Cambridge University Press, Cambridge,
Mills, L. S., M. E. Soulé, and D. F. Doak. 1993. UK.
The keystone-species concept in ecology and
conservation. Bioscience 43:219-224. Tainter, J. A., T. F. H. Allen, A. Little, and T. W.
Hoekstra. 2003. Resource transitions and energy
O’Neill, R. V., D. L. DeAngelis, J. B. Waide, and gain: contexts of organization. Conservation
T. F. H. Allen. 1986. A hierarchical concept of Ecology 7(3): 4. [online] URL: http://www.consecol.
ecosystems. Monographs in Population Biology 23, org/vol7/iss3/art4/.
Princeton University Press, New Jersey, USA.
Turelli, M., N. H. Barton, and J. A. Coyne. 2001.
O’Neill, R. V., A. R. Johnson, and A. W. King. Theory and speciation. TRENDS in ecology and
1989. A hierarchical framework for the analysis of evolution 16:330-343.
scale. Landscape Ecology 3:193-205.
Via, S. 2001. Sympatric speciation in animals: the
Peterson, G. D. 2002. Estimating resilience across ugly duckling grows up. TRENDS in Ecology and
landscapes. Conservation Ecology 6(1): 17. Evolution 16:381-390.
[online] URL: http://www.consecol.org/vol6/iss1/art17/
. Walker, B. 1995. Conserving biological diversity
through ecosystem resilience. Conservation
Peterson, G., C. R. Allen, and C. S. Holling. 1998. Biology 9:747-752.
Ecological resilience, biodiversity and scale.
Ecosystems 1:6-18. Walker, B, A. Kinzig, and J. Langridge. 1999.
Plant attribute diversity, resilience, and ecosystem
Pierce, J. L., and A. L. Delbecq. 1977. function: the nature and significance of dominant
Organizational structure, individual attitudes and and minor species. Ecosystems 2:95-113.
innovation. The Academy of Management Review
2:27-37. Wardwell, D., and C. R. Allen. 2009. Variability
in population abundance is associated with
Rice, W. R. 1984. Disruptive selection on habitat thresholds between scaling regimes. Ecology and
preference and the evolution of reproductive Society 14(2): 42. [online] URL: http://www.ecolog
isolation: a simulation study. Evolution 38:1251-1260. yandsociety.org/vol14/iss2/art42/.

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms
 Ecology and Society 15(3): 24
http://www.ecologyandsociety.org/vol15/iss3/art24/

Wardwell, D., C. R. Allen, G. D. Peterson, and

A. D. Tyre. 2008. A test of the cross-scale resilience
model: functional richness in Mediterranean-
climate ecosystems. Ecological Complexity
5:165-182.

Wilson, E. O. 1961. The nature of the taxon cycle

in the Melanesian ant fauna. American
Naturalist 95:169-193.

Woinarski, J. C. Z. 2006. Predictors of nomadism

in Australian birds: a re-analysis of Allen and
Saunders (2002). Ecosystems 9:689-693.

This content downloaded from

110.234.19.164 on Sun, 24 Mar 2024 06:31:02 +00:00
All use subject to https://about.jstor.org/terms