Content uploaded by Hugh Desmond
Author content
All content in this area was uploaded by Hugh Desmond on Oct 08, 2021
Content may be subject to copyright.
Vol.:(0123456789)
Biology & Philosophy (2021) 36:47
https://doi.org/10.1007/s10539-021-09821-2
1 3
Niche construction andteleology: organisms asagents
andcontributors inecology, development, andevolution
BendikHellemAaby1 · HughDesmond2,3
Received: 28 January 2020 / Accepted: 1 September 2021
© The Author(s), under exclusive licence to Springer Nature B.V. 2021
Abstract
Niche construction is a concept that captures a wide array of biological phenomena,
from the environmental effects of metabolism to the creation of complex structures
such as termite mounds and beaver dams. A central point in niche construction the-
ory is that organisms do not just passively undergo developmental, ecological, or
evolutionary processes, but are also active participants in them (Lewontin RC, In:
DS Bendall (ed) Evolution: From molecules to men, Cambridge University Press,
Cambridge, 1983; Laland KN, Odling-Smee J, Feldman MW, In: KN Laland and T
Uller (eds) Evolutionary causation: Biological and philosophical reflections, MIT
Press, Cambridge, MA, 2019). In this paper, we distinguish between two fundamen-
tally different ways in which organisms are active participants: as agents and as con-
tributors. Roughly, organisms act as agents when niche constructing effects are a
result of a goal-directed behavior over which the organisms have some degree of
control. Organisms act as contributors when the niche constructing effects do not
arise from a goal to perform the constructive activity. As illustrative examples we
discuss plants altering leaf-morphology to optimize light exposure as reported by
Sultan (Organism and environment: Ecological development, niche construction,
and adaptation. Oxford University Press, Oxford, 2015) and bacteria creating novel
niches through excreting energy-rich metabolites (San Roman and Wager in PLoS
Comput Biol 14: e1006340, 2018). The difference between agential and contribu-
tional niche construction is important for understanding the different ways organ-
isms can actively participate in development, ecology, and evolution. Additionally,
this distinction can increase our understanding of how the capacity of agency is dis-
tributed across the tree of life and how agency influences developmental and evolu-
tionary processes.
Keywords Niche construction· Ecology· Evolution· Teleology· Agency· Goal-
directed behaviour
* Bendik Hellem Aaby
bendik.aaby@kuleuven.be
Extended author information available on the last page of the article
B.H.Aaby, H.Desmond
1 3
47 Page 2 of 20
Introduction
Niche construction, on its broadest construal, refers to processes where organisms,
through their activities—whether through metabolism or behavioral choices—alter
environmental conditions and consequently change the selection pressures acting
on natural populations (Odling-Smee 1988; Odling-Smee etal. 2003; Laland et al.
2000, 2016). This construal is intentionally broad and inevitably labels all organisms
as niche constructors (Laland etal. 2016). In fact, this is what niche construction
theory (NCT) refers to as obligate niche construction (Odling-Smee et al. 2003).
Obligate niche construction is categorized as a necessary feature of life, reflecting
how living organisms are open systems in a far from equilibrium state and need to
extract energy from their surroundings in order to maintain their internal order and
dynamic stability (Schrödinger 1944; Odling-Smee etal. 2003, 167–79). The idea
behind obligate niche construction was central to organism-centered approaches to
biology in the early twentieth century, and niche construction theory can thus be
seen, at least in this respect, as furthering these traditions (Baedke 2019; Nicholson
2018).
The concept of obligate niche construction casts the net very widely, and covers
disparate phenomena.1 Previous work has been done on structuring these phenom-
ena and making valuable distinctions between different types of niche construction
(Aaby and Ramsey 2019; Chiu 2019; Ikiri and Sakura 2008; Sterelny 2003, 2010;
Sultan 2015). In this paper, we would like to bring attention to a class of examples
pertaining to the distinction between “obligate” and “facultative” niche construction.
An example of the former is how organisms open up new niches for other organ-
isms through their byproducts (San Roman and Wagner 2018). The oxygenation of
earth occurred through waste products emitted from cyanobacteria around 3.6bil-
lion years ago. From the perspective of niche construction, this was significant as
it opened the possibility for heterotrophic lifestyles as well as aerobic bioenergetic
strategies in general (Laland etal. 2014; Stal 2000). In other instances, niche con-
struction activities can be characterized by “optionality” or “quasi-choice”. Thus,
beavers build dams with internal lodges (Jones etal. 1994, 1997), someprimates
play complex social roles and exhibit (pre)cultural traditions (Flack etal. 2006), and
lastly, humans engage in niche construction which is typically judged as unparal-
leled in terms of ecological consequences and behavioral complexity (Fuentes 2017;
Sterelny 2003). Without taking a stance on the metaphysics of determinism, it seems
fair to distinguish such behaviors from the type of obligate niche construction activi-
ties arising from metabolic waste.
1 In this paper we will leave the possibility of “ghost niche construction” out of consideration. An exam-
ple of ghost niche construction would be a decomposing carcass significantly altering the selection pres-
sures of nearby scavengers, detritivores, and a host of microbial organisms (Odling-Smee etal. 2003).
This is a difficult case, since if dead organisms can construct niches, then further questions can be raised
about the distinction between niche construction and causal influence in general.
1 3
Niche construction andteleology: organisms asagents and… Page 3 of 20 47
However, what grounds this distinction? We propose the concept agency as a
way to analyze purposive and non-purposive alterations to the environment.2 This
reflects two different roles organisms can play as active participants in ecology,
development, and evolution: either as agents or as contributors. Only when niche
constructing organisms are regarded as agents, a teleological explanation is required
to fully understand the niche constructing behavior. When organisms are regarded
as contributors, a teleological explanation is not needed. The distinction is impor-
tant because asking whether an instance of niche construction is agential or contri-
butional (or both) can potentially uncover novel explainable regularities that might
otherwise remain unidentified. Further, interpreting niche constructing behavior as
agential allows us to use niche construction theory to highlight important instances
in which organisms across the tree of life exhibit agency and how instances of
agency might have profound influences on ecological, developmental, and evolu-
tionary processes.
Dierent denitions andconceptions ofniche construction
The literature offers a variety of definitions and conceptions of niche construction,
the broadest one being obligate niche construction. As we saw above, on this con-
ception, niche construction is simply something that organisms must engage in to be
alive (Odling-Smee etal. 2003). This conception has been criticized for its breadth
(e.g., Brodie 2005; Okasha 2005), as it classifies all organismic activity as niche
construction. This criticism often amounts to worries about niche construction being
an empirical intractable, and a ubiquitous and hence trivial phenomenon (see e.g.,
Dawkins 2004).
However, the criticism of ubiquity could be considered as weak. As Aaby and
Ramsey (2019) argue, other evolutionary processes like natural selection or ran-
dom genetic drift are also fairly ubiquitous. Yet these concepts do not refer to trivial
or intractable biological processes. It just means that there are cases in which the
effects of the processes are more or less significant. For example, the effects of drift
are inversely proportional to population size. As the population size increases, the
effects of drift decrease. Similarly, the strength of positive selection is relative to
how advantageous the phenotype under selection is relative to other phenotypes
in the population. In a large population, or one with few extreme phenotypes, the
effects of random genetic drift and natural selection may be negligible. But this does
not mean that they are trivial processes. It just means that there are instances in
which the effects are significant, but also instances in which they are negligible.
When the effects of a process are negligible, we usually background the pro-
cess when we study the dynamics of the system, while if they are significant, we
foreground them. Similarly, niche construction may be a ubiquitous process, such
2 In the rest of the paper, we will use the distinction between goal-directed vs. non-goal-directed ‘behav-
ior’ to flesh out the difference between agential and contributional niche construction. We come back
later to this definitional assumption.
B.H.Aaby, H.Desmond
1 3
47 Page 4 of 20
that we can background it when it has small effects. At other times, it is exactly the
effects of niche construction we want to study. This is reflected in Laland etal.’s
(2016) definition of niche construction as “the modification of selective environ-
ments by organisms” (191). While still broad, this definition picks out what kinds of
organism-induced changes are important—namely those that engender an alteration
of selective environments.
In response to the criticism of empirical intractability, Laland etal. (2016) fur-
ther operationalize the definition of niche construction by means of three criteria
suggested by Matthews etal. (2014, 247). These are: (i) the environmental modifi-
cation must be significant, (ii) the effects of the niche construction activities must
alter the selection pressures of some recipient organisms, and/or (iii) there must be
a detectable evolutionary response to the altered environmental states brought about
by the niche construction activities. The first two criteria are meant to capture all
cases where the organism-mediated environmental modifications are instances of
niche construction. The third criterion intends to capture the cases where evolution-
ary trajectories are influenced by organism-mediated changes to selection through
environmental modification.
While this conception is still broad and captures a vast array of organism-envi-
ronment interactions as niche construction, this is, as Laland etal. argue, intentional:
The breadth of our definition reflected the job we wanted the term to do. We
wanted to draw scientific attention to the diverse ways in which organisms
modify environmental states, and the myriad of important ecological and evo-
lutionary consequences that follow from these activities (Laland etal. 2019,
130).
A broad conception of niche construction thus serves to capture the many diverse
ways in which organisms can actively participate in their own ecology, development,
and evolution. ‘Niche construction’ could thus be considered a bucket term3 refer-
ring to an unstructured myriad of processes. In order to understand how, when, and
why niche construction is important, we should distinguish among specific kinds of
niche construction.
Specic forms ofniche construction
The literature offers quite a number of distinctions. For instance, Odling-Smee etal.
(2003) distinguish between several more specific forms of niche construction. First,
they distinguish between two ways in which organisms engage in niche construction:
through perturbation and relocation. Perturbational niche construction occurs when
organisms change the physical environment, while relocational niche construction
occurs when organisms move in space and time and consequently expose themselves
to different environmental variables.
3 We would like to thank an anonymous reviewer for suggesting this term.
1 3
Niche construction andteleology: organisms asagents and… Page 5 of 20 47
Odling-Smee etal. (2003) also make the orthogonal distinction between incep-
tive and counteractive niche construction. Inceptive niche construction occurs when
it is the organism itself that initiates the niche construction activity, while coun-
teractive niche construction occurs when an organism counteracts changes in the
environment. These can be combined to give us four categories of niche construc-
tion: inceptive perturbational, counteractive perturbational, inceptive relocational,
and counteractive relocational. These four categories can further be categorized as
positive, negative, or neutral, depending on their effects on the niche constructing
organism’s fitness. Finally, Odling-Smee et al. (2003) also identify cultural niche
construction. This refers to the effects cultural traits can have on other biological or
cultural traits, and on other species (Fogarty and Creanza 2017; Laland etal. 2012).
Sterelny (2003; 2010) uses the concept of niche construction to argue that a cru-
cial part of the story of human cognitive evolution is our ability to construct social
niches and modify our epistemic environment. Thus, social niche construction con-
sists in the modification of relationships between conspecifics, which consequently
alter selection pressures. An example of this is the different social roles of adult
meerkats (Suricata suricatta) in their troops. Epistemic niche construction consists
in altering the information flow from the environment, without necessarily altering
the environment itself. Predator-inspection strategies—when prey move closer to be
able to locate and track potential predators—is an example of this.
Another form of niche construction is experiential or mediational niche construc-
tion (Sultan 2015; Chiu 2019). According to Sultan (2015), there are two types of
niche construction: habitat construction and experiential construction. Habitat con-
struction occurs when the physical environment is altered (similar to perturbational
niche construction above). Experiential (or mediational) niche construction occurs
when an organism modifies its experienced environment without altering the physi-
cal environment. The experienced environment of an organism is a combination of
environmental cues and the response to such cues. The cue is a placeholder, and
what counts as a cue is relative to the organism’s specific sensory system and trans-
duction pathways, as well the environmental factors it engages with. Habitat choice
thus counts as an example of experiential niche construction.
Organisms can also change the way they experience their environment by chang-
ing their own constitution. Many plants change the morphology of their leaves
depending on light conditions. In low-light conditions, leaves become larger in sur-
face area in order to capture more photons. In high-light conditions it is beneficial
to have smaller but more numerous leaves, since too much exposure to ultra-violet
light can damage plant tissue (Sultan 2015). Thus, by altering their leaf morphol-
ogy, plants modify their experienced environment. This kind of niche construction,
through an alteration of the organism itself, has been called constitutive niche con-
struction by others (Aaby and Ramsey 2019; Walsh 2015).
The final kind of niche construction we will mention here is Ikiri and Sakura’s
(2008) intentional niche construction. This refers to an organism’s capacity to delib-
erately (in terms of higher-order intentional states, e.g., planning) manipulate their
environment. Ikiri and Sakura (2008) argue that tool-use and tool-manipulation
together with ecological and (proto)cultural inheritance and passive niche con-
struction (i.e., non-directed environmental modification by organisms) allowed the
B.H.Aaby, H.Desmond
1 3
47 Page 6 of 20
hominid linage to increasingly exploit the capacity for intentional niche construc-
tion. They argue that the rapid encephalization experienced by the hominid linages
over the last 2million year, which led to incredible diversity of cognitive capacities
we see in the hominids post Homo habilis was in part due to an increasing capac-
ity of intentional niche construction. The capacity of intentional niche construction
changed the selective environment these hominids experienced into one that was
predominantly determined by cognitive capacities. This in turn would have a ratch-
eting-effect on subsequent adaptation, leading to the rapid encephalization seen in
the later hominids.
While this is not an exhaustive discussion of different ways the concept of niche
construction has been utilized in the literature, it shows the extent to which the con-
cept has been adopted in different contexts. In a recent paper, Aaby and Ramsey
(2019) attempt to unify all these different instances under three fundamental cat-
egories of niche construction: constitutive, relational, and external. The reasoning
behind this tripartite categorization is that since a niche is defined as the functional
relationship between an organism and its environment, niche construction is any
alteration of this relationship. Thus, niche construction can involve a modification
to the organism itself, the environment, or the relations between the organism and
its environment. The thought is that since these categories track all the manipulable
parts of the niche itself, it will cover all specified instances of niche construction.
Further, Aaby and Ramsey (2019) make a distinction between intra- and intergener-
ational mechanisms. All three manipulable dimensions of the niche have important
mechanisms associated with them, both during development and across generations.
For example, an important intragenerational mechanism in constitutive niche con-
struction is phenotypic plasticity, while an important intergenerational mechanism is
epigenetic inheritance. For relational niche construction, behavioral plasticity is an
important intragenerational mechanism, while social learning is an important inter-
generational mechanism.
This tripartite categorization primarily aims at capturing the causal-mechanistic
structure of niche construction theory. We will argue that there is an additional tele-
ological dimension orthogonal to the tripartite categorization—namely whether or
not the niche constructing effects are a product of goal-directed behavior.4
Teleology andniche construction
In order to properly appreciate the difference between cases of niche construction
that are goal-directed and those that are not, we will outline two examples. The first
involves bacterial cross-feeding and the second developmental plasticity in leaf
morphology.
4 This distinction has been gestured at by Sterelny (2005) with his distinction between designed/adaptive
and accidental niche construction. Brodie (2005) and Dawkins (2004) made similar distinctions as well.
Such a distinction is problematic, however, as there are instances of accidental niche construction which
are also adaptive. See Kylafis and Loreau (2008).
1 3
Niche construction andteleology: organisms asagents and… Page 7 of 20 47
Niche construction incross‑feeding bacteria
In a recent paper on bacterial niche construction, San Roman and Wager (2018)
showed the surprising potential for niche construction in cross-feeding bacterial
strains. Cross-feeding refers to an interaction in which one organism depends
on the products of another organism for its subsistence. In the case of bacte-
ria, cross-feeding often happens when mutant strains are able to metabolize the
excrement of a different strain. For example, in evolutionary experiments on
genetically identical Escherichia coli populations grown in homogenous envi-
ronments where the only carbon source is glucose, it is common to see cross-
feeding E. coli strains emerge. One way this can occur is when some E. coli cells
that consume a primary carbon source (e.g., glucose) excrete a secondary carbon
source (e.g., acetate). After some time, a polymorphic E. coli cell turns out to be
able to metabolize acetate, and consequently a strain of cross-feeding E. coli is
established. Thus, in cross-feeding bacteria, new niches are constructed solely
out of the excretions of other organisms.
Experiential niche construction inplants throughalteration ofleaf morphology
An important instance of developmental plasticity in plants concerns the pro-
duction of different leaves by the same genetic individual under different light
conditions (Sultan 2010, 2015). Under low light conditions, the plant produce
thin, wide “shade-leaves.” Under high light conditions, the same individual will
produce thick, narrow “sun-leaves.” Depending on the cues from the environ-
ment, in this case the density of photons, the plant will alter its constitution such
that its “experienced environment” becomes different. Thus, the production of
shade-leaves will transform the experienced environment from one with a lower
to one with a higher photon density. Sultan (2015) interprets this as experiential
niche construction (or “mediational” niche construction sensu Chiu 2019).
An equally universal aspect of niche construction [obligate niche con-
struction being the other] is the way that an individual’s realized pheno-
type—including its morphology, physiology, and behavior in a given
environment—shapes and transforms how the individual experiences that
environment, apart from any measurable effects on external parameters
(Sultan 2015, 41).
Given that an environment is not restricted to physical features external to
the organism, but also includes how such features are experienced by the organ-
ism, it seems reasonable that an alteration of an organism’s experience of the
environment would be categorized as niche construction. In this case, the causal
basis of the niche construction activity is an alteration to the plant’s constitution,
which has the effect of modifying the experienced environment without alter-
ing the physical environment. In the conceptual landscape outlined above, this
would be experiential/mediational and constitutive niche construction.
B.H.Aaby, H.Desmond
1 3
47 Page 8 of 20
Explaining howandexplaining why
In providing an explanation of how both the cross-feeding bacteria and leaf-altering
plants are engaged in niche construction, standard causal-mechanistic explanations
are readily available. In the case of the bacteria, the explanation might cite mecha-
nisms responsible for the glucose and acetate metabolism of E. coli cells. In the case
of the production of shade-leaves, an explanation might cite biomass allocation and
other mechanisms responsible for developmentally plastic responses in the plant.
An explanation of why, on the other hand, would not share the same explanatory
structure. In the case of the bacteria, an explanation of why it emits energy rich
metabolites would be extremely general, most likely citing the necessity of metabo-
lisms to emit waste products. In the case of the plant, however, an explanation of
why the plant produces shade-leaves would provide the purpose that that particular
behavior serves in relation to attaining a specific goal.5 More specifically, the plant
produces shade-leaves in order to increase light capture. This again can be explained
in terms of biological function, i.e., the adaptive benefit higher light capture affords
the plant. Thus, the difference between the case of the cross-feeding bacteria and
the leaf-altering plant is that the latter needs a teleological explanation as well as a
standard causal-mechanistic explanation in order to properly explain the niche con-
structing behavior. In the former, an answer to the how-question is sufficient explain
why acetate excreting bacteria engage in niche construction.6 Why do E. coli bacte-
ria excrete energy-rich metabolites such as acetate? Because that is the waste prod-
uct of their bioenergetic strategy. The answer has no bearing on whether or not there
are, or can be, cross-feeding strains that consume it.
Agential niche construction
We are now in a position to make a distinction between cases with goal-directed
niche constructing effects, and those without such effects. For this, we will use a
concept of agency where agency is fleshed out in terms of purposiveness or goal-
directedness. (This is sufficient to capture the distinction between the examples
above; later we discuss in what ways goal-directedness is necessary but not suffi-
cient for attributions of agency.) For example, Walsh (2015, 210) defines agency in
the following way:
5 We consider the production of different kinds of leaves as a behavior: it is something that the plant
does and not something that merely happens to it. See Dretske (1988) for this approach to the concept of
behavior.
6 A clarificatory comment might be useful here. When we talk about cases of niche construction that are
a product of goal-directed behavior, we mean that the consequences of the niche constructing behavior is
the goal of the behavior that produces it. Thus, niche construction that is not the product of goal-directed
behavior can still be a consequence of behavior that is directed at something else. The cross-feeding bac-
teria example shows this, metabolism is clearly a goal-directed process, but the goal of the process is not
to excrete energy rich metabolites.
1 3
Niche construction andteleology: organisms asagents and… Page 9 of 20 47
[Agency is the] capacity of the system to pursue goals, to respond to the condi-
tions of its environment and its internal constitution in ways that promote the
attainment, and maintenance, of its goal state […] [Agency] consists in the
capacity of a system to cope with its setting, to attain its goal by responding to
its affordances as affordances.
On Walsh’s view, agency is found within an interdefinable triad of concepts:
goals, affordances, and repertoire. A goal is the end-state to which the behaviors
or activities of a system are directed. An affordance consists in the conditions that
are experienced by the goal-directed system as opportunities or impediments to the
attainment of a goal. Finally, a repertoire consists in the set of capacities available to
the system in pursuing its goal. An advantage of understanding agency as a capac-
ity located within this triad is that it allows agency to come in degrees. A repertoire
with a large set of behavioral capacities will increase the number of affordances,
which further increase the flexibility and freedom of the system in pursuing its goal.
We shall return to the relationship between teleology and agency in more detail
below. Motivated by approaches to agency such as Walsh’s, we will suggest term
agential niche construction to capture goal-directed forms of niche construction.
This does not mean that non-agential forms of niche construction—such as the
cross-feeding bacteria—do not involve organisms being actively engaged in niche
construction. The niche constructing effects are still a result of what organisms do.
The important aspects of these instances are not what the niche constructing behav-
ior is directed at, but rather who and what are affected by it. Such phenomena can
be termed contributional niche construction. Sometimes the contributions are global
and stable across evolutionary time, and even necessary for the viability of a large
chunk of the tree of life—as in the case of photosynthesis. Other times the contribu-
tions can be more local. We are thus left with two different ways in which organisms
can be active participants in development, ecology, and evolution through niche con-
struction: as agents and as contributors.
Why are teleological explanations important?
Now we need to show that the distinction between agential and contributor niche
construction plays a significant role for our understanding of niche construction and
the active role of the organism in ecology, development, and evolution. We will do
so by highlighting the special explanatory role that the category of agential niche
construction can play.
First, let us examine teleological explanations in general. Teleological explana-
tions point out regularities or dependencies between means and goals that cannot
be captured in standard sequential causal-mechanistic explanations (for a recent
account, see e.g., Walsh 2015). The classic Aristotelian example here is the differ-
ence between encountering a friend downtown by chance and a planned meeting.
In the case of encountering the friend by chance, seemingly innocuous changes to
the initial conditions might preclude the encounter from happening. For example,
stopping to browse in shop for a just a moment could be enough to ensure that the
encounter will not happen. Thus, in chance occurrences, it is necessary to explain
B.H.Aaby, H.Desmond
1 3
47 Page 10 of 20
the occurrence through a detailed account of the preceding mechanisms (and the
relevant initial conditions). By contrast, a planned meeting will occur across a broad
range of initial conditions and across a broad range of mechanisms which realize the
occurrence. You might go by car, bus, or by foot; you might travel from work, home,
or anywhere else. You may even be waylaid by an unforeseen delay. Either way, you
will most likely end up encountering your friend, not by chance but because that is
your goal. It is not explanatorily necessary to refer to initial conditions in detail, but
it is necessary to refer to the relevant goal.
Thus, merely citing the sequential mechanisms to explain your encounter leaves
out an explanatorily salient regularity—the fact that an encounter with your friend
is the goal and that a very different sequence of mechanisms could have equally
led to the encounter. This regularity stems from the dependence of means on goals.
The goal is available across a range of different means. In this case, a teleologi-
cal account of the occurrence explains why it was not a chance occurrence. Teleo-
logical explanations might thus uncover regularities which, on a purely mechanistic
account, would be considered the result of “blind” chance or variation. To neglect,
reduce, or otherwise explain away teleological regularities might lead to what Walsh
(2015, 194) calls a “selective blindness to a whole class of explainable regularities.”
Thus, by allowing teleological language to refer real, empirically tractable rela-
tionships, instead of treating it as terminological shorthand that ultimately can be
reduced and translated into the effects of prior mechanisms or causal processes, we
open up a whole class of potentially underappreciated but explainable regularities.
An example of such a regularity can be seen in the results of a meta-study of hun-
dreds of selection gradients measured in natural populations of different species by
Clark etal. (2020). The meta-study shows that when environmental variation is buff-
ered by organismic activity (e.g., nest building, pupal cases, burrowing), it generally
results in reduced variation in selection gradients and weaker directional selection
(compared to the selection arising from non-buffered environmental variation). This
is consistent with how niche construction is argued to alter the conditions of selec-
tion to favor traits that are beneficial to organisms controlling and regulating envi-
ronmental variation through their activities, instead of favoring traits that directly
address environmental heterogeneity (e.g., Laland etal. 2017).
The reason why this should be seen as an example of a regularity that is uncov-
ered by a teleological account is that it points to a general dependence of means on
goals, which moreover holds across a wide array of taxa. What are environment-
buffering activities for? The evidence suggests that organisms engage in such activi-
ties in order to reduce environmental variation. This regularity also helps explain
how goal-directed behaviors can influence selection. Environment-buffering activi-
ties influence selection gradients by reducing environmental variation. The reduc-
tion of environmental variation in turn alters the selective environment to favor traits
that are useful for regulation and control over environmental factors. This means
that not only does selection favor environment-buffering traits because they have
been beneficial in the past, but also because the selective environment has (partly)
been constructed by the environment-buffering activities themselves. Thus, if we try
to reduce teleological explanations to causal-historical explanation—as for example
the selected-effects of priori selection (Millikan 1984; Neander 1991)—we might
1 3
Niche construction andteleology: organisms asagents and… Page 11 of 20 47
gloss over the fact that the prior selective environment was itself partly the result of
the very purposiveness that the selected-effects approach is supposed to account for
by citing the prior action of selection. Instead of treating the purposiveness of organ-
isms exclusively as an explanandum (as one does in the reductive approach to teleol-
ogy), we should allow such teleological relationships to serve the role of explanan-
tia, especially in cases where we can observe goal-directed behavior.
What istheorigin of“goals” andwhat does it take tohave a“goal”?
Now that we have indicated why teleological explanations can be important, and
what the explanatory roles are of means and goals, we need to account for what
goals are and where they come from. In the context of this paper we can use Walsh’s
account: a goal is a stable end-state of system and goal-directed processes are pro-
cesses that aim to bring about and maintain those end-states (Walsh 2015). Organ-
isms are engaged in goal-directed behaviors when their behaviors are directed at and
conducive to a certain end-state. Another way to say this is to say that organisms
exhibit purposiveness.
Does the attribution of purposiveness to an organism also entail further capacities
such as foresight, deliberation, or other higher-order mental capacities in the goal-
seeking organism? Not necessarily. In attributing a goal to a behavior, we do not
have to simultaneously attribute any form of “knowledge” of the goal to the organ-
ism itself. For example, in case of the leaf-altering plant discussed above, the goal of
the behavior is “to increase light capture”. An individual plant, by common under-
standings of the concept “cognition” or “mentality”, is obviously not endowed with
the cognitive or mental capacities to “understand” why it performs the behavior, why
the behavior is conducive to a certain goal, or even that the behavior it performs
has specific goal. However, fortunately, one can assign goals to behaviors without
also assigning mentality or sophisticated cognition to the organisms in question. It
is sufficient that reaching the goal increases the relative reproductive output of the
behaving organism. A similar argument is provided by Laland etal. (2019) in their
discussion of purposiveness as a fundamental feature of life:
When we assert that organisms are “purposive” we mean nothing more than
that organisms exhibit goal-directed activities such as foraging, courtship, or
phototaxis, which are entirely natural tendencies with short-term local objec-
tives, and that have themselves evolved. The “goals” and “purposes” to which
we refer can be defined with respect to general aspects of biological function,
such as resource acquisition, stress avoidance, and reproduction (Laland etal.
2019, 132).
So, having a goal does not require foresight or any other complex cognitive capaci-
ties on behalf of the organism.
However, we also need an account of how goals originate. A common strategy
in providing a naturalistic (in the sense of non-mentalistic) account of goals is to
highlight the autopoietic nature of life (Varela etal. 1974). This refers to the fact
that organisms are “self-building, self-regulating, highly integrated, functioning, and
B.H.Aaby, H.Desmond
1 3
47 Page 12 of 20
(crucially) ‘purposive’ wholes, which through wholly natural processes exert a dis-
tinctive influence and a degree of control over their own activities, outputs, and local
environments” (Laland etal. 2019, 132). Thus, goals arise in virtue of organisms
needing to actively maintain their internal order and dynamic stability in order to
stay alive. The simplest goal-directed processes thus have self-maintenance as the
end-state they are directed at. More complex goal-directed behavior follows from
organisms evolving more capacities (increased morphological, physiological, and
behavioral complexity) which consequently increase in what opportunities their
environments afford them (Walsh 2015). Thus, a perfectly naturalistic account of
goals and purposiveness is readily available.
Potential issues withtheagential andcontributional distinction
Now that we have established that there is a distinction between agential and con-
tributional niche construction, and that organisms across the tree of life can exhibit
goal-directed behavior, let us turn to some potential objections. First, there is the
question whether and how groups can engage in collective agential niche construc-
tion. Second, how does the agential and contributional distinction relate to the
difference between constructing one’s own niche and constructing that of others?
Finally, while agency requires goal-directedness, not all goal-directed processes
are instances of agency. What are we to make of non-agential, yet goal-directed,
processes?
Collective agential niche construction?7
Niche construction can happen collectively, through the activities of multiple organ-
isms. This is relatively unproblematic when the niche construction is contributional,
since the collective activity is simply the addition of individual contributions. The
oxygenation of the atmosphere would be an example of collective contributional
niche construction. By contrast, agential niche construction at the level of a group
seems more problematic. How are we to account for groups that engage in goal-
directed behaviors?
On this matter, the discussion of collective agency in action theory can be
helpful. In that context, collective agency is generally seen as action performed by
individuals in virtue of a shared goal and (often) a joint commitment to the reach
the goal (Gilbert 2010).8 Whether this shared goal must be held by the group
“over and above” individuals is a contentious issue. For some (e.g., Bratman
1993), it is sufficient that individuals share a goal and act as a group. In such a
case, a group-level action could simply be the aggregation of the individual-level
7 We would like to thank Jan Baedke for pointing this problem out to us when an earlier draft of this
paper was presented at ISHPSSB 2019.
8 Note that it is common to talk in terms of shared intentions and not shared goals in philosophy of
action, as it is primarily concerned with human social interaction.
1 3
Niche construction andteleology: organisms asagents and… Page 13 of 20 47
behaviors. Obligate shoaling in fish is a good example of this kind of collec-
tive action. The individuals share a goal (e.g., “staying close to conspecifics for
safety”) and act as a group. The group-level shoaling behavior is simply a func-
tion of individuals performing behaviors directed at a similar individual-level
goal. It is not the goal of the group to perform shoaling behavior. Shoaling behav-
ior is rather a consequence of the individuals performing behavior directed at
shared individual goals.
However, some philosophers of action argue that a joint commitment to reach
the goal is necessary for collective action (e.g., Gilbert 1989). What this means is
that the shared goal is attributed to the group, and not simply to the individuals.
A consequence of a goal shared at a group level is that the group-level behavior is
no longer simply an aggregate of individual-level behavior, but rather a function
of potentially varying individual-level behaviors each serving a role in the attain-
ment of the shared group-level goal. Complex hunting strategies in social ani-
mals or the different behaviors and morphologies of ants from different castes in
a single colony would be candidates for collective agential niche construction in
which the goal is shared at a group-level. What differentiates this kind of collec-
tive action from aggregate action is that it requires a minimal degree of orchestra-
tion amongst individual behaviors to reach the shared group-level goal. In social
hunting, an indication of this type of orchestration is the possibility that certain
individuals correct other group members when they fail to perform the appropri-
ate behavior necessary for achieving the shared group-level goal.
In some cases, groups of interspecific organisms such as holobionts seem to
act collective in purposive ways. Could such groups of organisms be said to have
the capacity for agential niche construction as a super-organism? A full explo-
ration goes beyond the scope of this paper, but the complexities of the interac-
tions between the different species that compose a holobiont (e.g., differences in
temporal and spatial organization) cast doubt on whether a shared goal can be
attributed, both at the individual and group level. For instance, it seems unlikely
that a eukaryotic host and its microbiome can share a similar goal, unless we con-
strue that goal in a very general manner, e.g., as “the goal of self-maintenance”
discussed above. By the same criteria, agential niche construction could be attrib-
uted to entire ecosystems, which are composed of interacting organisms, all of
which (except the very melancholic of us) share the goal of self-maintenance. Are
ecosystems a reasonable locus for collective action? It seems somewhat strange to
attempt to attribute agency to something simply by alluding to the fact that life is
interconnected and that organisms generally aim at self-maintenance.
In sum, collective agential niche construction can be grouped in two catego-
ries: aggregated and orchestrated collective action. Aggregated collective action
occurs when group members have similar individual goals and act as group.
Orchestrated collective action are cases in which the goal is shared at a group-
level. Whether or not different instances of group-level agential niche construc-
tion is aggregated or orchestrated, and what mechanisms are responsible for such
orchestration, is an empirical question which should be solved on a case-to-case
basis.
B.H.Aaby, H.Desmond
1 3
47 Page 14 of 20
Agential niche construction ofadifferent species’ niche
At first glance, it might seem that agential niche construction is limited to the con-
struction of the organism’s own niche. But there are many examples of goal-directed
behavior aiming at manipulating another species’ environment. A very obvious
example is the cultivation and domestication of plants and animals by humans
(Piperno 2017; Smith 2016; Zeder 2016). In such cases, the niche of the domesti-
cated organism is (partially) constructed by the domesticating organism. Such prac-
tices are not limited to humans, however. There are several instances of ant-fungus
mutualisms where ants cultivate fungi in their colonies as a food source.9 The most
studied species of fungus-cultivating ants are the leaf-cutter ants. These are com-
monly referred to as “higher” attines, because they cultivate highly derived fungi
that have no wild-type counterparts, and they form an obligate mutualism. These
ants actively propagate, nurture, and defend different species of fungi of the clade
Lepiotaceae. The fungi benefit from a steady supply of nutrition in the form of fresh
plant matter and protection against pests and mold, while the ants in turn feed on the
nutrient-rich hyphal swellings of the fungus called the “gongylidia” (Chapela etal.
1994).10 Theants perform behavior such as cutting fresh plant material and placing
it in close proximity to the fungus in order to provide nutrition to the fungus, namely
so that the fungus can excrete digestive enzyme and absorb the nutrients from the
plant material. In this case, we would say that the ants are engaged in agential niche
construction of another species’ niche as it results from goal-directed activity.
Thus, in cases of cultivation and domestication, there are organisms engaged in
agential niche construction of a different species’ niche. However, cultivation can
also happen indirectly. What are we to make of such cases? In such cases it might
not be clear whether or not an organism is engaged in agential or contributional
niche construction. An example can be found in the male spotted bowerbird, Chla-
mydera macultata. These birds are famous for their elaborate bowers used primarily
as a display to attract mates. A recent study found that male spotted bowerbirds use
the fruit of a species of nightshade (Solanum ellipticum) as an important part of
their sexual display (Madden etal. 2012). Males that have a large number of fruits
on display (and consequently in and around their bowers) experience high mating
success.11 The S. ellipticum plant, being a perennial pioneering species, benefits
from being collected by the bower birds in virtue of being deposited in areas with
little surrounding vegetation, which in turn is favorable for seed germination and
9 Cultivation of fungi have been found in termites as well, so agricultural practices in eusocial insects
is more prevalent than just amongst different species of the tribe Attini, which is by far the most studied
(Mueller etal. 2005).
10 There is much more complexity to ant-fungus mutualism. For example, there is mounting evidence
that a host of microbial organisms, e.g., gut-bacteria in the ants, are intertwined in these mutualistic rela-
tionships (Aylward etal. 2012).
11 Madden etal. (2012) also speculate that the male spotted bowerbird might benefit from having a local
supply of S. ellipticum fruit is by having to spend less time foraging away from their bower, thus reduc-
ing the risk of marauding neighboring rivals.
1 3
Niche construction andteleology: organisms asagents and… Page 15 of 20 47
pioneer establishment. Are bowerbirds agentially or contributionally constructing
the niche of S. ellipticum?
When the male birds arrive at new sites with unconstructed bowers, they do not
choose sites with a higher number of S. ellipticum plants than any other random site
suitable for bower construction. Most male bowerbirds are also relatively sedentary,
occupying the same bower for periods up to 10years and remain close to their previ-
ous year’s bower site (sometimes less than 10m). Finally, there is no evidence that
the fruit is consumed or that it serves any role beyond ornamentation during sexual
displays. In this way, Madden etal. (2012) found no evidence that the bowerbirds
engage in any sort of soil manipulation or other activities that would increase the
fruit yield or germination success of S. ellipticum. They instead argue that it is most
likely a by-product of males removing fruit that has turned brown from their dis-
plays and dropping them in the immediate vicinity of their bowers. In other words,
there is no evidence that S. ellipticum cultivation is a result of bowerbird behavior
directed at cultivation. Thus, until more evidence is gathered and we see a good
reason to revisit the question of whether or not the cultivation is a result of goal-
directed behavior, no teleological explanation of that particular niche constructing
activity is necessary.
In general, many instances of mutualisms are thought to start out as one species
exploiting another (Conner 1995; Odling-Smee et al. 2003). Such cases begin as
contributional niche construction, but as the exploited species begins to exploit its
exploiter, a mutualistic relationship evolves. Such cases are likely to end up as agen-
tial niche construction, at least in the case of the original exploiting species (e.g., the
leaf-cutter ants discussed above, and it seems to be true for most cases of domesti-
cation).12 A distinction between contributional and agential niche construction can
be helpful in describing the way in which some mutualisms end up as instances of
domestication, and how goal-directed behaviors of agential niche construction might
alter the selective environment in a different manner than what would we expect if
niche construction behavior was purely contributional.
Teleology does notequal agency
Until now we focused primarily on goal-directedness in order to distinguish between
agential and contributional niche construction. By doing so, we adopted a broad def-
inition of organismic agency, where goal-directedness is sufficient to make an attri-
bution of agency (e.g., Walsh 2015, where the environment is also conceptualized as
a landscape of affordances). The resulting distinction between agential and contribu-
tional niche construction, we argued, helps better account for explainable regulari-
ties that might otherwise remain unidentified. However, one could ask whether the
definition of agency could still be further fine-grained.
How precisely the metaphysics of organismic agency should be understood is still
an open question, both how it relates to causal-mechanistic processes (physiological
12 We would like to thank the reviewer for pointing this out.
B.H.Aaby, H.Desmond
1 3
47 Page 16 of 20
or neural processes) and how precisely it relates to natural selection. After all, inso-
far a process of natural selection produces an adaptation, the subsequent goal-direct-
edness of the adaptation could be accounted for solely in terms of an etiology of
fitness differences. Agency as a concept does not seem to need to figure in such
explanations. For instance, in behavioral ecology, the “phenotypic gambit” refers to
the modeling assumption that organismic behaviors can be modeled as fitness maxi-
mizing (Grafen 1984, 2014). Building on this work, in Okasha’s account of agency,
attributing agency to an organism is simply an expression of adaptationism, in
which the various traits of the organism all contribute to a single process of fitness
maximization of the whole (Okasha 2018, p. 5). Okasha defends some attributions
of agency as explanatorily justified, but is much more qualified in attributions of
agency than for instance Walsh (2015).
We cannot explore this debate in this paper, but as a closing discussion we would
like to examine how the concept of agential niche construction changes as a more
narrow definition of agency is adopted. In action theory, agency is thought to require
something more than mere goal-directedness, namely an influence of the agent over
the behavior it performs. Frankfurt (1998), for example, defends a position in which
agency consists in goal-directed behavior which is under the guidance or influ-
ence of the agent. In a similar vein, Laland etal. suggest that agential activities are
goal-directed activities over which an organism exerts a degree of control (Laland
et al. 2019). Organismic agency can thus be seen as the capacity of an organism
for goal-directed behaviors over which (to a lesser or greater extent) the organism
itself exerts a degree of control. On this narrower definition of agency, agential niche
construction refers to cases in which the niche constructing behavior is goal-directed
and where the organism performing it exerts a degree of control over that behavior.
One candidate strategy of translating the concept of “control” into a biological
context would be in terms of developmental canalization, where the prior action
of selection on genetic variation produces behavioral programs over which the
organism has little control. Burrowing and nest building behaviors have often been
described as canalized in this way. However, it does not follow that an organism can-
not exert any influence over such behavior. For example, if the canalized behavior in
question is directed at the alteration of the environment, it is likely that the organism
will be sensitive to a whole range of conditions that requires an alternate behavio-
ral output for the alteration to obtain.13 As a simple example, think of an organism
attempting to construct a burrow in a surface that is too hard. It will, most likely,
after a short while attempt to burrow in a different location where the surface has
different properties. In fact, recent studies have revealed considerable plasticity in
burrowing and nest building behaviors (e.g., Hansell 2007).
The lesson here is that “control” does not necessarily require a large degree of
plasticity or behavioral flexibility (i.e., a large repertoire of capacities). An organism
could have a relatively limited set of capacities in its repertoire—and consequently
13 More likely, a behavioral disposition for engaging in nest building would be what is developmentally
canalized, while the actual nest building behavior is under the influence of the organism as it interacts
with its environment.
1 3
Niche construction andteleology: organisms asagents and… Page 17 of 20 47
fewer affordance to respond to—while still exerting an influence or control over its
behavior (cf., persistence and plasticity in Lee and McShea 2020). Thus, a large
degree of behavioral flexibility is not necessary for organismic agency, but could
rather be understood as an indication of the organism’s “freedom of action” in pur-
suit of its goal. Here “freedom” should be understood not in terms of volition, but
rather as possessing different means to deal with impediments and opportunities for
achieving a goal.
Another option would be to understand “control” in contrast to instinctual and
reflexive behaviors. Triggered by stimuli and producing behaviors in a mechanis-
tic way, such behaviors may be goal-directed (adaptive) without being under the
control of the organism performing them. In fact, the rationale in referring to those
responses as reflexive or instinctual is to indicate that they are outside the control of
the organism. Such behaviors could potentially count as non-agential on the narrow
definition of agency.
Nonetheless, also here it is not possible to simply classify all instinctual behaviors
as non-agential. For example, inappropriate or unnecessary reflexive and instinc-
tual behavioral responses can at least sometimes be overcome through habituation,
desensitization, or other forms of learning and experience (Ginsburg and Jablonka
2019). Habituation has even been observed in non-neural organisms, such as the
flowering plant Mimosa pudica (Gagliano etal. 2014) and slime molds (Boisseau
etal. 2016). Thus, some instinctual and reflexive behaviors can be rectified when
unnecessary or inappropriate. In such cases, the control or guidance of the behavior
is not concomitant with the performance of the behavior, but rather exerted at a later
stage as a consequence of learning or experience. In this way, the judgment whether
a specific niche construction behavior is agential or not becomes complex, since the
specific niche construction behavior might be latently regulated by other capacities
or behaviors.
Similarly, goal-directed processes of sub-organismal parts may seem automatic
(and non-agential) but yet can be indirectly influenced by the activities of an agent
as whole. For instance, digestive processes are automatic and are in general not
under the direct control of the organism. Yet, animals that feed infrequently and in
large portions may remain stationary after feeding in order to allow more blood to
go to the digestive system rather than to their extremities. Another example is how
thermoregulation of endotherms is an automatic and adaptive process, even if some
mammals without fur may use mud to influence thermoregulation. If the organ-
ism has the appropriate behavioral capacities and the environment affords the right
opportunities, the organism as a whole can influence such processes indirectly. Just
as with reflexive and instinctual behaviors, the organism may produce goal-directed
yet automatic behaviors, but that still may be indirectly agential through being inter-
twined with other capacities or behaviors under the direct control of the organism.
In sum, it is possible to adopt a narrower definition of agency, such as one that
requires control in addition to goal-directedness. The indicators of agency then
become properties such as persistence and flexibility. However, under what condi-
tions precisely a niche construction behavior can be definitely judged to be non-
agential is a complicated question beyond the scope of this paper. One difficulty
we pointed to lies in the reciprocal influences that different organismic behaviors
B.H.Aaby, H.Desmond
1 3
47 Page 18 of 20
and capacities can have on each other. This means that many behaviors which seem
beyond the control of the organism may turn out to be only partially so, or only indi-
rectly so. In any case, the main cases considered in this paper—cross-feeding bacte-
ria and alteration of leaf morphology in plants—do not require these added consid-
erations and so also on a narrower definition of agency the previous analysis holds.
Conclusion
We have seen that the concept of niche construction has been conceived and defined
in a number of different ways reflecting the theoretical utility of the concept across
a broad range of phenomena and disciplines. We argued that there is an important
distinction to be made between agential and contributional niche construction. The
distinction is fleshed out in terms of goal-directedness, where agential niche con-
struction arises from a goal-directed behavior, while contributional niche construc-
tion does not. In instances of agential niche construction, an account of why the
niche construction occurs requires a teleological explanation in order to fully appre-
ciate how organisms are active participants in development, ecology and evolution
as agents. In instances of contributional niche construction, a teleological explana-
tion of why the niche construction occurs does not have any bearing on the organ-
ism’s role as contributors. The distinction is important because asking whether an
instance of niche construction is agential or contributional (or both) can potentially
uncover novel explainable regularities and phenomena that might otherwise remain
unidentified. Further, interpreting niche constructing behavior as agential allow us
to use niche construction theory to highlight important instances in which organ-
isms across the tree of life (in bacteria, fungi, plants and animals) exhibit agency and
get a better grasp on how organismic agency influences evolutionary, developmental
and ecological processes, such as natural selection.
References
Aaby BH, Ramsey G (2019) Three kinds of niche construction. Br J Philos Sci. First online 05 December
2019, doi:https:// doi. org/ 10. 1093/ bjps/ axz054
Aylward FO, Currie CR, Suen G (2012) The evolutionary innovation of nutritional systems in leaf-cutter
ants. Insects 3:41–61
Baedke J (2019) O organism, where art thou? Old and new challenges for organism-centred biology. J
Hist Biol 52(2):293–324
Boisseau RP, Vogel D, Dussutour A (2016) Habituation in non-neural organisms: evidence from slime
moulds. Proc R Soc B 283:20160446
Bratman M (1993) Shared intentions. Ethics 104(1):97–113
Brodie ED (2005) Caution: niche construction ahead. Evolution 59:249–251
Chapela IH, Rehner SA, Schultz TR, Mueller UG (1994) Evolutionary history of the symbiosis between
fungus-growing ants and their fungi. Science 266(5191):1691–1694
Chiu L (2019) Decoupling, commingling, and the evolutionary significance of experiential niche con-
struction. In: Laland KN, Uller T (eds) Evolutionary causation: Biological and philosophical reflec-
tions. MIT Press, Cambridge, MA, pp 299–323
Clark AD, Deffner D, Laland K, Odling-Smee J, Endler J (2020) Niche construction affects the variabil-
ity and strength of natural selection. Am Nat 195(1):16–30
1 3
Niche construction andteleology: organisms asagents and… Page 19 of 20 47
Conner RC (1995) The benefits of mutualism: a conceptual framework. Biol Rev 70:427–457
Dawkins R (2004) Extended phenotype–But not too extended. A reply to Laland. Turner Jablonka Biol
Philos 19:377–396
Dretske F (1988) Explaining behavior: reasons in a world of causes. MIT Press, Cambridge MA
Flack JC, Girvan M, de Waal FBM, Krakauer DC (2006) Policing stabilizes construction of social niches
in primates. Nature 439:426–429
Fogarty L, Creanza N (2017) The niche construction of cultural complexity: interactions between innova-
tions, population size and the environment. Phil Trans R Soc B 372:20160428
Frankfurt HG (1998) The importance of what we care about. Cambridge University Press, Cambridge
Fuentes A (2017) Human niche, human behaviour, human nature. Interface Focus 7:20160136
Grafen A (1984) Natural selection, kin selection and group selection. In: Krebs JR, Davies NB (eds)
Behavioural ecology, 2nd edn. Blackwell Scientific Publications, Oxford, UK, pp 62–84
Grafen A (2014) The formal darwinism project in outline. Biol Philos 29:155–174
Gagliano M, Renton M, Depczynski M, Mancuso S (2014) Experience teaches plants to learn faster and
forget slower in environments where it matters. Oecol 175:63–72
Gilbert M (1989) On social facts. Princeton University Press, Princeton NJ
Gilbert M (2010) Collective action. In: T O’Conner and C Sandis (eds) A companion to the philosophy of
action. Wiley-Blackwell, Malden, MA, pp. 67–73.
Ginsburg S, Jablonka E (2019) The evolution of the sensitive soul. MIT Press, Cambridge MA
Hansell M (2007) Built by animals: the natural history of animal architecture. Oxford University Press,
Oxford
Iriki A, Sakura O (2008) The neuroscience of primate intellectual evolution: natural selection and passive
and intentional niche construction. Phil Trans R Soc B 363:2229–2241
Jones CG, Lawton JH, Shachak M (1994) Organisms as ecosystem engineers. Oikos 68:373–386
Jones CG, Lawton JH, Shachak M (1997) Positive and negative effects of organisms as physical ecosys-
tem engineers. Ecology 78:1946–1957
Kylafis G, Loreau M (2008) Ecological and evolutionary consequences of niche construction for its agent.
Ecol Lett 11:1072–1081
Laland KN, Odling-Smee J, Feldman MW (2000) Niche construction, biological evolution and cultural
change. Behav Brain Sci 23:131–175
Laland KN, O’Brien MJ (2012) Cultural niche construction: an introduction. Biol Theory 6:191–202
Laland KN, Boogert N, Evans C (2014) Niche construction, innovation and complexity. Environ Innova-
tion Soc Transitions 11:71–86
Laland KN, Matthews B, Feldman MW (2016) An introduction to niche construction theory. Evol Ecol
30:191–202
Laland, K., Odling-Smee, J., & Endler, J. (2017). Niche construction, sources of selection and trait coev-
olution. Interface Focus, 7, 20160147.
Laland KN, Odling-Smee J, Feldman MW (2019) Understanding niche construction as an evolutionary
process. In: Laland KN, Uller T (eds) Evolutionary causation: Biological and philosophical reflec-
tions. MIT Press, Cambridge, MA, pp 127–156
Lee JG, McShea DW (2020) Operationalizing goal directedness: an empirical route to advancing a philo-
sophical discussion. Philos Theor Pract Biol 12:5
Lewontin RC (1983) Gene, organism, and environment. In: Bendall DS (ed) Evolution: From molecules
to men. Cambridge University Press, Cambridge, pp 273–285
Madden JR, Dingle C, Isden J, Sparfeld J, Goldizen AW, Endler JA (2012) Male spotted bowerbirds
propagate fruit for use in their sexual display. Curr Biol 22(8):264–265
Matthews B, De Meester L, Jones CG, Ibelings BW, Bouma TJ, Nuutinen V, van de Koppel J, Odling-
Smee J (2014) Under niche construction: an operational bridge between ecology, evolution, and eco-
system science. Ecol Monogr 84(2):245–263
Millikan R (1984) Language, thought, and other biological categories: new foundations for realism. MIT
Press, Cambridge MA
Mueller UG, Gerardo NM, Aanen DK, Six DL, Schultz TR (2005) The evolution of agriculture in insects.
Annu Rev Ecol Evol Syst 36:563–595
Neander K (1991) Functions as selected effects: the conceptual analyst’s defense. Phil Sci 58(2):168–184
Nicholson DJ (2018) Reconceptualizing the organism: From complex machine to flowing stream. In:
Nicholson DJ, Dupré J (eds) Everything flows: towards a processual philosophy of biology. Oxford
University Press, Oxford
B.H.Aaby, H.Desmond
1 3
47 Page 20 of 20
Odling-Smee J (1988) Niche Constructing Phenotypes. In: Plotkin HC (ed) The role of behavior in evolu-
tion. MIT Press, Cambridge, pp 73–132
Odling-Smee J, Laland KN, Feldman MW (2003) Niche construction: the neglected process in evolution.
Princeton University Press, Princeton
Okasha S (2005) On niche construction and extended evolutionary theory. Biol Philos 20:1–10
Okasha S (2018) Agents and goals in evolution. Oxford University Press, Oxford
Piperno DR (2017) Assessing elements of an extended evolutionary synthesis for plant domestication and
agricultural origins research. Proc Natl Acad Sci USA 114(25):6429–6437
San Roman M, Wagner A (2018) An enormous potential for niche construction trough bacterial cross-
feeding in a homogenous environment. PLoS Comput Biol 14(7):e1006340
Schrödinger E (1944) What is life? The physical aspect of the living cell. Macmillan, New York
Smith BD (2016) Neo-Darwinism, niche construction theory, and the initial domestication of plants and
animals. Evol Ecol 30:307–324
Stal LJ (2000) Cyanobacterial mats and stromatolites. In: Whitton BA, Potts M (eds) The Ecology of
cyanobacteria: Their diversity in time and space. Kluwer Academic, London, pp 61–120
Sterelny K (2003) Thought in a hostile world: the evolution of human cognition. Blackwell publishing,
Malden
Sterelny K (2005) Made by each other: organisms and their environment. Biol Philos 20:21–36
Sterelny K (2010) Minds: extended or scaffolded? Phenomenol Cogn Sci 9:465–481
Sultan SE (2010) Plant developmental responses to the environment: eco-devo insights. Curr Opin Plant
Biol 13:96–101
Sultan SE (2015) Organism and environment: ecological development, niche construction, and adapta-
tion. Oxford University Press, Oxford
Varela FG, Maturana HR, Uribe R (1974) Autopoiesis: the organization of living systems, its characteri-
zation and a model. BioSystems 5(4):187–196
Walsh DM (2015) Organism, agency, and evolution. Cambridge University Press, Cambridge
Zeder MA (2016) Domestication as a model system for niche construction theory. Evol Ecol 30:325–348
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
Authors and Aliations
BendikHellemAaby1 · HughDesmond2,3
Hugh Desmond
hugh.desmond@gmail.com
1 Institute ofPhilosophy, KU Leuven, Leuven, Belgium
2 Institute fortheHistory andPhilosphy ofScience andTechnology (IHPST), CNRS/Paris I,
Paris, France
3 Department ofPhilosophy, University ofAntwerp, Antwerp, Belgium
A preview of this full-text is provided by Springer Nature.
Content available from Biology & Philosophy
This content is subject to copyright. Terms and conditions apply.