The theoretical argument that lies at the heart of the claim that niche construction is an evolutionary process derives from the expectation that niche construction must be purposive and goal-directed, which follows from basics of thermodynamics. The great physicist, Erwin Schrödinger (1944) first pointed out that living organisms are far-from-equilibrium systems relative to their physical or abiotic surroundings. To preserve their out-of-equilibrium status, while creating order (structure, organization) in their bodies and their immediate surrounds, organisms must do work on their environment. Living organisms can only survive by constantly exchanging energy and matter with their …environments. Schrodinger explains why this work cannot be random; it must be directional. Random niche-constructing acts would not generate order in their bodies and immediate surrounds, and hence could not provide organisms with a basis for sustained life. It follows that organisms must change environments in systematic ways (Odling-Smee et al. 2003).

Over-and-above these thermodynamic considerations, niche construction generates environmental states that are coherent and integrated with the organism’s phenotype, and that are adaptive for the constructor or its descendants, largely because the organism’s capability to do so has evolved through earlier natural selection. Organisms in the same species possess broadly similar evolved capabilities for niche construction, so they are expected to modify environmental states in broadly similar ways, while the niche construction of offspring is expected to resemble that by their parents. For these reasons, niche construction theory expects niche construction to scale up across individuals in populations and over time to generate stable modification of environmental components, and as a result to generate consistent and sustained forms of natural selection. In this manner, niche construction imposes biases on selection.

This is probably the default assumption within evolutionary biology. There are two reasons why it is problematic. First, the niche construction of living organisms is only partly explained by earlier selection. Niche constructing capabilities can also arise in development.  

For illustration, the fact that humans possess an evolved capacity to learn does not provide a complete and satisfactory explanation for why some humans populations happen to take up dairy farming, and hence why those particular populations evolved adult lactose tolerance through self-imposed selection on LCT alleles.

The second reason is the causal intertwining of evolutionary processes. Niche construction can only be reduced to past natural selection in a model of evolution in which the causes of Lewontin’s (1970) three conditions of ‘phenotypic variation’, ‘differential fitness’ and ‘inheritance’ are treated as quasi-independent: conversely, where these processes are causally intertwined, this interpretation does not follow (Uller & Helanterä 2017; Laland et al. 2019).

Niche construction is defined as ‘organism-driven environmental modification’. It thus includes any and all changes that an organism’s activities have upon the world. Niche construction, so characterized, is a universal feature of all living creatures. Even after death niche-constructing effects may persist: for instance, as a bodies decompose, they change the chemical composition of the earth around them, and sometimes leave hard parts, such as shells (Erwin 2008).

The generality of this formulation raises the question, ‘What is not niche construction?’ If almost everything an organism does is niche construction, is it still a useful concept? Yes – many biological concepts (growth, development, natural selection) apply universally, but that does not mean they are not useful. The correct question here is ‘When is it useful to emphasize that organisms engage in niche construction, and when is it not?

Some aspects of niche construction, for instance, the local fluctuations in ambient temperature caused by the movement of homothermic organisms, are frequently so trivial that they are best ignored. Incorporating these processes adds complexity but little explanatory power. Where organisms change their environments in very minor ways, or in ways that are immediately dissipated, negated, swamped or counteracted by other more powerful processes, such that they do not accrue across individuals or over time, it is better to ignore those aspects of niche construction.

What is not niche construction, then, are changes that organisms bring about in their worlds that are of no evolutionary or ecological consequence. This is an empirical issue. Whether discarded shells or hoof prints need to be treated as an instance of niche construction depends on whether the shells or indented soils provide resources for other organisms, and accrue in space and time, to affect selection on descendant populations.

Odling-Smee et al. (2003) defined niche construction as

‘the process whereby organisms, through their metabolism, their activities, and their choices, modify their own and/or each other’s niche’.

They defined an evolutionary niche, as ‘the sum of all the natural selection pressures to which the population is exposed’.

A feature of this characterization of niche construction is its breadth. The definition encompasses the construction of environmental modifications that have evolved as adaptations (such as birds’ nests) as well as environment-changing by-products (such as plant litter). It includes physical perturbations of the environment (e.g. digging a burrow) and relocations in space (e.g. migrations). It includes environment changing activities that have evolved through standard biological evolution as well as those that are learned and culturally transmitted (e.g. human agricultural practices).

Odling-Smee et al. coined this definition because they wanted to draw scientific attention to the diverse ways in which organisms modify environmental states, and the important ecological and evolutionary consequences that follow from these activities. What the above activities have in common is that they modify selection, as well as flows of energy and matter. A narrower definition risks neglecting important processes and feedbacks that fall outside it.

There is now extensive data showing that even organisms with modest individual effects on their environments can nonetheless strongly affect ecological and evolutionary events (Jones et al 1994, 1997; Odling-Smee et al. 2003; Sultan 2015). That is because a variety of processes scale up the impact of niche construction.

Organisms frequently repeat their niche-constructing activities across their lifespan. Organisms of the same population often modify their environments in similar ways, which means that even organisms with small per capita effects can have huge ecological effects where population densities are high. The effects of individual organisms can also accumulate in environments often over long periods of time, spanning multiple generations. In addition, some constructs or artifacts (such as termite mounds or beavers’ dams) are very durable or long-lasting, and can continue to have ecological consequences long after their constructors have died.

For all of these reasons, niche-constructing activities can act like a persistent unidirectional “pump” to eventually cause considerable change in environments.

No. The impacts that organisms have on environments are taken into account in countless evolutionary and ecological analyses. Ecologists have built large bodies of theory that explore the ecological consequences of many aspects of niche construction, for instance, models of resource depletion, succession, or competition for resources. Likewise, evolutionary biologists have developed models of frequency and density dependent selection, habitat selection, coevolution, sexual selection, maternal effects, and many other phenomena, in which the activities of organisms modify the action of selection. This literature is reviewed in considerable depth in Odling-Smee et al. (2003) and Sultan (2015), and is frequently discussed in articles and talks on niche construction.

Why then does Brodie (2005) write: ‘Natural selection is depicted [by niche construction theory] as resulting only from inanimate and abiotic features of the environment’. From where might this misunderstanding derive?

Possibly one source of confusion stems from a misreading of the above/adjacent figure, versions of which appear extensively in articles and Odling-Smee et al.’s (2003) book on niche construction. Here the horizontal arrows represent causal processes not effects in the world. Hence the absence of an arrow from organism to environment in figure 1a should be read as implying that niche construction is not recognized as an additional cause of evolutionary change, not that organisms are not recognized to alter environments.

This distinction between ‘process’ and ‘effect’ is important here. Evolutionary biology textbooks and articles frequently list a number of processes (selection, drift, mutation) that are causes of biological evolution, and invariably ‘niche construction’ is not found in such lists. The characterization of ‘niche construction’ as ‘the neglected process in evolution’ (Odling-Smee et al. 2003), implies only that it is not typically treated as process in evolution, and not that evolutionary biologists fail to recognize that niche construction happens.

Another potential source of misunderstanding may stem from the use of a pair of differential equations to illustrate the difference in perspective between niche construction theory and standard evolutionary theory. This heuristic dates back to Lewontin (1983), but has been reproduced widely (e.g. Odling-Smee et al. 2003).

Standard evolutionary theory is presented as dO/dt = f(O,E), dE/dt = g(E), but niche construction advocates argue that the following is more accurate: dO/dt = f(O,E), dE/dt = g(O,E).

It is possible to see how this characterization might be misleading, since on the surface it seems to deny any recognition by the standard account that environmental change can be caused by organisms. This was not Lewontin’s intention (personal communication to Kevin Laland).

Lewontin rejected the notion that the environment was ‘out there’ as an external context for evolutionary events that logically pre-exists the organism. The traditional accounts’ ‘externalism’ is well-recognized as a widespread metaphysical framework within biology by philosophers of science (e.g. Godfrey-Smith 1996).

Lewontin’s differential equations were never intended to apply to those aspects of the selective environment that constitute other organisms; for instance, sexual selection or coevolutionary theory. Nor did Lewontin’s equations deny ecological change caused by organisms. Rather, his equations were intended to challenge the standard separation of ecological and evolutionary processes and time-frames; the notion that one could study evolution without considering ecology, and vice-versa; that organismal effects on physical environments are too trivial or transient to exert any lasting impact; or even that a niche can be characterized without an organism.

Oxford evolutionary biologist Richard Dawkins (1982) introduced the term ‘extended phenotype’Biological adaptations expressed outside of the body of the organism. to represent the effects that genes can have outside of their bodies.  Some, but not all, aspects of niche construction are also extended phenotypes. The latter is a narrower term because it is restricted to forms of environmental modification that are biological adaptations, such as birds’ nests or termites’ mounds. 

In contrast, niche construction theory emphasizes that niche construction also incorporates evolutionary by-products. This broad characterization is vital, since in ecosystems the ecological and evolutionary consequences that flow from niche-constructing by-products are likely to be as substantial as those flowing from niche-constructing adaptations.

For instance, Post & Palkovacs (2009) describe how various patterns of guppy evolution are thought to have resulted from the different patterns of excretion observed in populations with different size distributions, while variant forms of root growth are observed in Populus trees and surrounding flora that have been differentially exposed to leaf litter with different tannin levels, since tannins impede nutrient release. They write: ‘eco-evolutionary feedbacks at the community and ecosystem level can emerge from both direct engineering and the by-products of living organisms (contrary to the arguments of Dawkins and Brodie).’

In addition, since extended phenotypes are restricted to biological adaptations, they automatically exclude aspects of niche construction that depend on acquired knowledge, for instance, human cultural processes. This is an important limitation for scholars of anthropogenic change, since very little human niche construction can accurately be characterized as a biological adaptation.

Most instances of human-induced habitat degradation – harvesting, deforestation, urban development, agricultural practices, livestock grazing, and so forth, – are products of cultural knowledge acquired and transmitted by humans, and expressed in their use of tools, technology and, engineering. In contrast to the extended phenotype, niche construction theory emphasizes how the acquired knowledge of organisms can trigger ecological cascades or drive evolutionary episodes.

Thirdly, extended phenotypes are properties of individuals, not of collections of individuals, yet some forms of niche construction are the results of actions by many populations, even many species. Both terrestrial and marine bioturbation influence the geochemistry and productivity of sediment, in turn influencing the evolutionary environment of a host of other species. The extended phenotype model is not a good description of the role of bioturbators, both because many effects are probably not adaptations, and because the sediment is a collective rather than individual product, comprised of the by-products of several species’ niche-constructing activity.

Finally, the environmental modifications produced by niche-constructing organisms, such as beaver’s dams or termite mounds, may persist through repetition and reconstruction for longer than the individual constructors, and may continue to modulate the impact of these effects on subsequent generations of the same or other populations, a legacy described as an ‘ecological inheritance’The inheritance, via an external environment, of one or more natural selection pressures previously modified by niche-constructing organisms.. 

At the extreme, niche-constructing activities, such as sediment bioturbation or the accumulation of shell beds, can accumulate over geological time, modulating macroevolutionary patterns and diversity (Erwin 2008). This temporal dimension to niche construction, in shaping the pattern and intensity of selection experienced by distant descendants, is entirely missed by the theory of the extended phenotype. However, numerous theoretical models have shown that it strongly affects evolutionary dynamics

Richard Dawkins (2004) criticized niche construction theory for ‘confusing’ byproducts and adaptations:

‘The problem I have with niche construction is that it confuses two very different impacts that organisms might have on their environments. As Sterelny (2001, p. 333) put it, ‘Some of these impacts are mere effects; they are byproducts of the organisms’s way of life. But sometimes we should see the impact of organism on environment as the organism engineering its own environment: the environment is altered in ways that are adaptive for the engineering organism. Niche construction is a suitable name only for the second of these two (and it is a special case of the extended phenotype). There is a temptation, which I regard as little short of pernicious, to invoke it for the first (byproducts) as well.’

In fact, niche construction theory does not ‘confuse’ adaptations and byproducts. Scientific articles regularly distinguish between niche-constructing adaptations and niche-constructing byproducts.

The difference in perspective stems from the fact that Dawkins’ focus is on the evolution of niche-constructing traits, where his distinction between adaptations and byproducts is most relevant. However, niche construction theory has a broader focus that also encompasses the ecological and evolutionary consequences of niche construction for other species, where this distinction is only sometimes relevant.

Dawkins’ extended-phenotype stance, in restricting evolutionary feedback to the alleles underlying a niche-constructing trait only, may arguably be adequate as a conceptual simplification with which to understand the evolution of a specific niche-constructing trait (although even here the extended phenotype position may overlook important processes). However, Dawkins’ extended phenotype stance ignores additional forms of feedback that flow to the constructor, and to other populations in its local environment, from its niche-constructing activity, indirectly via the environment.

In practice, it is frequently (almost always) not known whether an instance of niche construction is an adaptation or a byproduct. Demonstrating that something is an adaptation is a very challenging thing to do. For this reason, many researchers find a neutral umbrella term like ‘niche construction’ useful in practice.

For the evolution of niche construction the appropriate focus is on adaptations, and on the direct and indirect fitness benefits accrued to niche constructors as a result of modifying their environment. However, for the ecological or evolutionary consequences of niche construction (for instance, the evolution of other traits, or of other populations, or on ecological flows and resource distributions, or eco-evolutionary feedbacks, or on the structure and characteristics of ecosystems), it is vital that analysis is not restricted to adaptations, but includes the many important consequences that flow from time-lagged effects, by-products, acquired characters, and collective activity.

‘Niche construction is not just an important addition to current evolutionary theory; it requires a reformulation of evolutionary theory.’ (Odling-Smee et al. 2003)

‘The changes to the evolutionary process brought about by niche construction… are sufficiently important and occur sufficiently frequently to warrant an overhaul in evolutionary thinking.’ (Day et al. 2003)

Claims, such as the above, made about niche construction by its advocates, have been viewed as extravagant overstatements, hyperbole or ‘hard sell’ by its critics. Conversely, advocates view such statements as justified, on the grounds that the refinement in evolutionary thinking that they advocate is genuinely radical.

For instance, the inclusion of a general second inheritance system (ecological inheritance), the explicit recognition of the role of acquired characters in directing evolution, and the emphasis on reciprocal causation, are three ways in which niche construction theory differs from standard accounts quite distinctly.

Niche construction theory describes the evolutionary process in a fundamentally different way. If an advocate draws the conclusion that their science is better served by recognizing niche construction as a distinct evolutionary process then to portray it as a radical departure is not exaggeration.

The claims that niche construction is an evolutionary process, a source of adaptation, and requires an extended evolutionary synthesis, have all excited controversy. These issues are discussed by Laland & Sterelny (2006), and by Scott-Phillips et al. (2014). The latter article is a collaboration between some critics of the niche-construction perspective and one of its advocates that attempts to pinpoint their differences. The authors conclude that their disagreements reflect a wider dispute within evolutionary theory over whether the neo-Darwinian synthesis is in need of reformulation, as well as different usages of some key terms (e.g., evolutionary process). Finally, Uller & Helanterä (2017) discuss the different conceptions of biological causation that separate niche construction advocates and critics.