Conservation biology and restoration ecology are scientific fields built on ethical foundations. Typically, specific interventions have a concrete intention, such as re‐establishing a particular species to a specific site, as well as a more abstract, value‐laden goal, such as increasing nativeness or integrity or biodiversity. Many projects are not explicit about the values underlying them, relying on broadly shared assumptions about values and goals. Others do justify their specific actions by reference to values like biodiversity or integrity, but the values themselves are rarely critically analyzed. Any analysis of whether conservation and restoration are “succeeding” or are having “intended consequences” should extend to an assessment of the moral justification for the field's activities.
The central value of both conservation and restoration is the value of biodiversity, from the diversity of genes in a population, up to and including the diversity of ecosystem types on Earth (Soulé, 1985). Another value held by many conservation biologists and restoration ecologists is the value of “ecological integrity.” (We use “ecological integrity” interchangeably with “ecosystem integrity,” following the general practice in the field.) Aldo Leopold famously axiomatizes the importance of ecosystem integrity in his moral principle, “A thing is right when it tends to preserve the integrity, stability, and beauty of the biotic community” (Leopold, 1989: 224–225).
The integrity of ecosystems is a property that contemporary conservation biologists and restoration ecologists believe that they can observe and measure, and integrity is often cited as a good reason for taking specific actions, including protecting high‐integrity areas and intervening to increase integrity by reintroducing lost species, removing non‐natives, and adjusting abiotic elements.
The value of integrity is thoroughly enshrined in policy. The U.S. National Parks Service and the State of Washington, for example, prioritize sites for conservation and management intervention using an Ecological Integrity Assessment developed by NatureServe (
In this article, we argue that integrity is a bad fit for ecosystems and that ecosystem integrity is a bad fit for conservation and restoration. We begin by surveying the concept as it is currently used, focusing our examples on Western North America.
Given their inherently changeable and variable nature, we question the idea that ecosystems are the kinds of things that can have integrity. Ultimately, we do not think they are. It then follows that “ecosystem integrity” cannot be valuable. Indeed, the features of ecosystems that make them unlikely to possess “integrity”—their dynamism, their gradual transitions in time and space, their adaptability—are among the things many find valuable about them.
Next, we acknowledge that not everyone will be persuaded that integrity simply does not apply to ecosystems. We thus do our best to sketch out a working definition. Using this definition, we argue that even if it is possible to say that ecosystems can have integrity, it is problematic to think that such integrity can explain losses of value when ecosystems change radically.
We propose that the value of ecological integrity is, in many cases, an imperfect placeholder for the values of biodiversity, complexity, and important cultural attachments to particular historical ecosystem states. Referring specifically to these values instead of using the vague and arguably incoherent concept of “ecosystem integrity” will improve the precision of our communications as we make arguments for or against particular environmental interventions.
Ecological integrity has been defined many different ways. Peter Bridgewater et al. note that the phrase “has a high degree of linguistic elasticity and should there ever be a legal challenge to its use, there are no precise and clear definitions for it” Bridgewater, Kim, & Bosselmann, 2014, p. 68). There are, however, certain common elements found in many definitions:
- Naturalness
- Wholeness
- Continuity through time
Often, the concept of ecological integrity includes the idea that the influence of humans, in particular, destroys integrity. Ecosystems with integrity are said to be “natural” (Noss, 2000) or “minimally influenced” by humans (Karr, 1996; Karr, 2000). As one integrity assessment tool puts it, “High integrity refers to a system with natural evolutionary and ecological processes, and minimal or no influence from human activities” (Theobald, 2013, p. 1859). This tool explicitly uses “indicators of the absence of human modification of habitat and alteration of ecological processes” (Theobald, 2013, pp. 1859–1860) to quantify ecosystem integrity. Another influential gloss on integrity comes from Parrish et al, who define it as “the ability of an ecological system to support and maintain a community of organisms that has species composition, diversity, and functional organization comparable to those of natural habitats within a region” Parrish, Braun, & Unnasch, 2003, p. 852).
Other definitions, however, reject naturalness as definitional and explicitly include some human influence as constitutive of ecological integrity, as we shall see.
There is more unanimity in the second common definitional element—wholeness. This is unsurprising, since the word “integrity” is typically defined as “the state of being whole and undivided.” Although ecosystems are composed of millions of individuals, they are often seen as wholes. In many definitions of ecosystem integrity the ecosystem must be “entire” (Miller & Rees, 2000) “complete” (Noss, 2000), or “undiminished” (Miller & Rees, 2000). In the introduction of an edited volume on ecosystem integrity, Peter Miller and William E. Rees note that the concept of integrity is generally associated with a “valuable whole” or something that is “entire or undiminished… unimpaired or [in] perfect condition” (Miller & Rees, 2000, p. 10). Reed Noss claims that ecological integrity refers to “the healthy and complete condition of a natural landscape” 2000, p. 191).
Many definitions of ecosystem integrity include the idea that the ecosystem is self‐organizing or self‐repairing. It must be able to persist; that is, the system is able to “sustain” (Regier, 1993) and “organize” over time (Kay, 1993; Miller & Rees, 2000). When perturbed, it returns to a state of wholeness.
What this wholeness should look like or act like is usually defined by the past. Consider the Washington State Department of Natural Resources's official definition of ecological integrity as “the structure, composition, and function of an ecosystem operating within the bounds of natural or historical range of variation” (WA DNR, 2019). The NatureServe assessment tool, similarly, bills itself as “an assessment of the structure, composition, function, and connectivity of an ecosystem as compared to reference ecosystems operating within the bounds of natural or historical disturbance regimes” (Faber‐Langendoen, Nichols, Rocchio, Walz, & Lemly, 2016). The Forest Service's legally binding definition states that ecological integrity is “the quality or condition of an ecosystem when its dominant ecological characteristics (e.g.,, composition, structure, function, connectivity, and species composition and diversity) occur within the natural range of variation and can withstand and recover from most perturbations imposed by natural environmental dynamics or human influence” (36 CFR § 219.19).
Conservation biologists do not demand that ecosystems be completely unchanging to have integrity. No one denies that ecosystems are dynamic. Fluctuations in, for example, population sizes over time are widely regarded as “normal” and not a threat to the integrity of a system. Similarly, variations in space from one end of a given ecosystem to another are not considered to be problematic for integrity. But system‐wide changes in function, as when a grassland becomes a forest, are usually seen as losses of integrity. A widespread replacement of one set of species with another is also usually characterized as “degradation” or loss of integrity, even if the system's overall function and structure are not transformed. For example, grasslands that come to be dominated by introduced grasses are typically seen as having lost integrity, even if they are still grasslands, still support populations of herbivores, and so on. So what kinds of change, magnitudes of change, or rates of change are extreme enough to compromise integrity?
In the early 20th century, Western conservationists tended to use the past to judge what an ecosystem “should” look like. In these early days, the use of historical states as goals was justified by the assumption that the historical state represented a timeless, unchanging “natural” state that was more valuable because naturalness is inherently valuable. As Teddy Roosevelt famously said of the Grand Canyon, “Leave it as it is. You cannot improve on it. The ages have been at work on it, and man can only mar it” (Roosevelt, 1903).
Early ecological ideas held that ecosystems were by default static and that a “balance of nature” absorbed natural disturbance and returned the system to the steady state. These ideas were perhaps most famously expounded by Frederic Clements, who popularized the ideal of a recovery from disturbance through succession—a series of predictable compositions that lead inevitably to a “climax”—the inexorable community for each place, determined by climate, a whole that was nearly as internally coherent as a living organism (Marris, 2011).
In the 1990s, the role of “natural disturbance” including “lightning‐caused fires, floods, erosion, drought, disease, and insects” was increasingly recognized (Averill et al., 1994). Conservationists largely abandoned the single “historical baseline” in favor of a “range” of historical states. Many definitions of integrity thus refer to the “natural range of variation” (NRV) or, increasingly, “historical range of variability” (HRV). These range‐based concepts recognize that change occurs but describe a temporal envelope of conditions as the reference or target. In Western North America, this is, generally in the 100–300 years before European arrival (Douglas, 2002; Romme, Wiens, & Safford, 2012; Safford et al., 2017).
Despite encompassing ~250 years of conditions, these ranges are essentially swollen historical baselines (Morgan et al., 1994; Wiens, Hayward, Hugh, & Giffen, 2012). The NRV and HRV concepts were designed to include more possible states and more possible historical moments as acceptable, but they ultimately just expand the target from a single instant to a very slightly longer moment. The HRV fits comfortably in a Clemensian world.
Although the Clemensian concept still has a remarkable hold on the popular imagination (Botkin, 1990), nearly all contemporary ecologists would admit that Clements' intellectual rival, Henry Gleason, painted a more accurate picture of how ecosystems function. Ecosystems are temporary assemblages of plants and animals, brought together by the vagaries of geology, history, climate, and a dash of pure happenstance.
Paleoecologists and environmental historians now paint a picture of an unspooling series of changes on the landscape, driven by changes in climate, evolution, and accident (Jackson, 2006; Marris, 2011). As the climate changes and evolution unfurls, species change and move—and they do not all move together. Pinyon‐juniper woodland is a well‐known ecosystem type that covers more than 40 million hectares in the high‐altitude American west (Romme et al., 2009), but 12,000 years ago, pinyon pine was found far to the south of the Great Basin, in today's Mojave Desert, while juniper was already found throughout the southern Great Basin and was well on its way north (National Forest Service). In North America, most ecosystems were unrecognizably different 14,000 years ago (Jackson, 2013). The same is true in Europe, where “vegetation development in the current interglacial has no analogue from the past 500,000 years” (Cheddadi et al., 2005, p. 13,939). Change is constant. Ecosystems are neither predictable nor inevitable. “Every species of plant,” Gleason wrote, “is a law unto itself” (Gleason, 1926, p. 26).
If ecosystems are always changing, there is no obvious scientific reason why one moment in time has “integrity” and another does not. So why choose the years 1,600–1850 for North American ecosystems? Clearly, the dates are pegged to colonization. Formerly, the presettlement state was considered “natural” or “pristine,” because the influence of Indigenous people on the landscape was considered minimal, in part due to lack of knowledge and in part due to racism. Conservation, thankfully, has begun to recognize the pervasive influence of Indigenous peoples on ecosystems.
Indigenous people in North America made substantial, transformative changes to the ecosystems they inhabited. Over the millenia, they likely helped bring about the extinctions of dozens of species of megafauna (Galetti et al., 2018; Gill, 2014; Johnson, 2009). They altered the ranges of species, controlled populations of game animals by hunting, influenced the evolution of species through domestication and semi‐domestication, directly propagated favored plants in key locations, and radically altered and controlled fire regimes in many places across the world, structuring entire ecosystems (Anderson, 2013; Johnson et al., 2016; Kimmerer, 2013). And these changes were profound enough to reshape the selection pressures acting on many species, altering their evolutionary trajectories.
But the realization that the pre‐settlement state was not “natural” or “pristine” has not led to its retirement as a reference point or goal. Instead, the pre‐settlement state is now described as having “integrity” in part because of Indigenous management.
The use of the word “historical” in the phrase “historical range of variation” intentionally includes effects of pre‐European humans on the landscape as part of the ecosystem (Romme et al., 2012). As a 2012 IUCN report on restoration explains, “traditional, ecologically sustainable human activities have shaped some ecosystems to the extent that cultural practice and ecological integrity are mutually reinforcing” (Keenelyside, Dudley, Cairns, Hall, & Stolton, 2012, p. 19). Or as Stephen Woodley writes, “ecological integrity can and should be understood outside the context of whether or not people are present in the system Woodley, 2010, p. 151).” This move helps address the problematic erasure of Indigenous land management on the landscape before European settlement implied by calling such states “pristine” or “natural,” but it also suggests that the high‐integrity state is not necessarily a “pre‐human” state or a “natural” state. It must be recognized or defined in some other way.
If “naturalness” is firmly set aside, how can we understand ecological integrity? The most common definitional elements that remain are wholeness and continuity through time.
Let us begin by looking at wholeness. For something to exhibit integrity, in the sense of wholeness, it has to be the kind of thing that can be a whole. Individual organisms are paradigm cases of wholes. For example, a single human seems like a clear whole, whose integrity can be compromised by, for example, lopping off an arm or leg.
Ecosystems are quite a bit less obviously wholes than individual organisms. They are not organized by a single consciousness. They do not have a recognizable container or boundary like skin. Their elements do not always stay together physically—some birds migrate between multiple ecosystems each year, for example.
Ecosystems lack clear borders or boundaries. Although particular species assemblages may exist together over large swathes of the landscape, overlap will never be exact. As an example, pinyon‐juniper woodland is “associated with a range of different vegetation,” according to the National Parks Service, and this system can be seen “transitioning from grasslands or shrublands at lower elevations, and to ponderosa pine (Pinus ponderosa) or other montane forest associations at higher elevations,” implying that at least some land area is “transitional” between ecosystem types (National Parks Service, Chapter 1, 2015). Pinyon‐juniper woodland exhibits “a wide variety of stand structures and compositions, which are influenced by local climate, topography, growing conditions, and disturbance regimes,” and five different species of juniper and four different species of pinyon pine may be present in different areas (National Parks Service, Chapter 2, 2015). NatureServe has differentiated “over 10,000 vegetation communities and ecological systems” in the Western hemisphere (NatureServe, 2021). The hills behind our home in Klamath Falls appear to be “Western Juniper Open Woodland” in this classification scheme. Interestingly, this “group” is considered a sub‐type of the “Intermountain Pinyon ‐ Juniper Woodland” “macrogroup” despite the fact that it does not contain any Pinyons at all. Geographical borders between ecosystems are not hard or unambiguous; one shades gradually into another; categorization schemes are imposed from outside and every type contains considerable internal variability.
Hence, it is hard to conceptualize ecosystems as the kinds of things that are physical wholes like individuals. But, perhaps, there are other ways to conceptualize ecosystems as wholes.
One way to conceptualize “wholeness” in an ecological sense would be to suggest that an “intact” ecosystem has all its ecological niches filled. A “integral” ecosystem might better repel “invaders,” the theory goes, because there is no ecological niche available to them. This refers back to the notion that ecosystems with integrity can self‐organize and are resilient. However, that is not what we tend to see. Some systems, for example, the Hawaiian Islands, have seen their independently reproducing flora double in the human era (Mascaro, 2013). Most naturalizing species simply “slot in” to existing ecosystems, suggesting they were not “full” in the first place. Organisms can construct or reconstruct their own niches over time (Odling‐Smee, Laland, & Feldman, 2013). As Peter Bridgewater remarks, “there is much evidence in support of the idea that ecosystems may not ever be ‘whole,’ which brings further into question a simple notion of integrity” (Bridgewater et al., 2014, p. 71).
Conservation biologists have long argued that coevolution creates mutual dependencies that bind ecosystems together into a whole. Michael Soulé wrote that “the structure, function and stability of coevolved, natural communities differ significantly from those of unnatural or synthetic communities” (Soulé, 1985, p. 729). Soulé used the response of an insect to a host plant as an example of species that are “highly specialized” and thus vulnerable to extinction in the absence of their ecological partners. Here, ecosystems are whole in the sense that they require all or most of their parts to function. Remove enough components and the whole will collapse. The parts depend on the whole and the whole on the parts.
But more recent work has tended to underscore the flexibility of species to persist on their own or change ecological partners. For example, the North American butterfly Euphydryas editha switched hosts from a native wildflower, Collinsia parviflora, to an introduced weed Plantago lanceolata within a few decades (Singer & Parmesan, 2018).
Mutalisms can be formed with surprising alacrity. Eucalyptus is one of the most successful world‐travelers among the trees. A look at the ectomycorrhizal fungal partners of Eucalyptus globulus in Spain, far from its native range in Australia, found that out of the 26 fungal taxa it interacted with, only three were also from Australia. Interestingly, one Australian fungal species, Descolea maculata, was also making new connections in Spain, as the researchers found it interacting with native pines, oaks, and woolly rock rose (Santolamazza‐Carbone, Durán‐Otero, & Calviño‐Cancela, 2019).
When mutualisms are broken by the movement of one of the partners, extinction or even loss of abundance does not necessarily ensue. The Neotropical tree genus Cecropia is well known for its relationship with Azteca ants, which it “feeds” with special food bodies at the base of its petioles and houses in hollow stems. The ants defend the plant against animal herbivory and encroaching vines, which the ants meticulously trim back. However, Cecropia pachstachya has naturalized in Singapore despite the absence of Azteca ants there (Lok, Tan, Chong, Nghiem, & Tan, 2010). So far, no native ant has adapted to colonize the hollow stem habitat, but many take advantage of the food bodies. Other ant‐less Cecropia are found all over the tropics, including the Hawaiian Islands. As Dan Janzen wrote in 1973, “The ant‐free Cecropia demonstrate clearly that a complex mutualism can evolutionarily disintegrate without the loss of both partners if the proper habitats are available” (15).
These examples help illustrate the difficulty of thinking of ecosystems as wholes that necessarily shatter when the coevolved bonds between their components are broken. The most impressive coevolved mutualisms, such as that between the Star‐of‐Bethlehem orchid with its foot‐long nectar spur and the giant hawkmoth with its foot‐long proboscis, are frequently offered to students of ecology as examples of coevolution. Increasingly, it seems that such truly obligate mutualisms are actually outliers—coevolutionary adaptations so ornate that they function as evolutionary “traps,” endangering both parties should their partner disappear. Truly obligate mutualisms may be the exception, not the rule, just as island adaptations such as flightlessness in birds can be conceived of as evolutionary traps that make some species more vulnerable to extinction (Thomas, 2017).
Novel ecosystems, even those that exhibit high levels of diversity, typically have not yet developed deeply coevolved relationships. They may thus be seen as less valuable. But this lack of value can be characterized as a dearth of ecological complexity without requiring the concept of “integrity.”
Central to many definitions of integrity is the notion that changes to ecosystems do not threaten integrity if the ecosystem does not change radically or suddenly. Conservationists try to accommodate the tension between an ecosystem's integrity and the constant change of biotic systems by asserting that a system can have integrity but still be able to “adapt,” “develop,” or “evolve.” Definitions of ecosystem integrity often include the expectation that the system will change and evolve in the future.
So which changes or rates of change are acceptable and which are not? Often, anthropogenic change is treated differently than non‐anthropogenic change, but we have already argued that this is a mistake, and many contemporary definitions of integrity do include human influence. Simply differentiating “gradual” change from “sudden” change and “minor” from “major” change is not terribly specific. It also runs into a problem: sudden and significant ecological changes are neither uncommon nor limited to post‐colonization ecological dynamics.
One study of lake‐sediment pollen cores corresponding to the last 20,000 years of North American ecological history showed that ecosystems persisted for an average of just 230–460 years before transitioning to another ecosystem type (Wang, Shipley, Lauer, Pineau, & McGuire, 2020). Sixty‐four percent of these systems eventually reverted to their previous ecosystem type, but the rest did not. In particular, Pleistocene megafaunal extinctions “led to rapid changes where plant communities transitioned through multiple alternative states rapidly” (Wang et al., 2020, p. 5,923).
Humans are not the only species that can initiate extremely rapid ecosystem changes. Take as example the movement of beaver into low arctic tundra regions in Alaska and Canada (Jones et al., 2020; Tape, Jones, Arp, Nitze, & Grosse, 2018). Beaver, released from trapping suppression and perhaps encouraged by climate warming, are moving north at a rate of 8 km/year (Tape et al., 2018). As they create ponds, the deep water conducts heat and melts permafrost. Substantial vegetation changes and the colonization of “riverine plants, invertebrates, and fish” follow, along with changes in nutrient cycling and sediment dynamics (Tape et al., 2018). Within decades, the area can no longer be characterized as “tundra” at all. A majority of the components and interactions have changed.
Jones et al. (2020) also report on the presence of “fossil beaver dams and beaver‐gnawed wood dating to the early Holocene” (8) in their study area in northwestern Alaska, suggesting that this is not the first time the industrious mammals have flipped tundra into wetland. Presumably, no matter whether anthropogenic or non‐anthropogenic climate change nudges the beaver up north, all it takes is a single beaver generation to fundamentally alter the landscape.
If ecosystem integrity consists of the same components and interactions persisting through time, then these sorts of major changes presumably not only compromise integrity, but most likely extinguish it. The tundra is gone.
Some might argue that no loss of integrity occurred as the frozen tundra was replaced by wetlands because there was some kind of continuity between the two systems. There was no “sudden break” or single moment when all organisms were killed and the system wiped clean. Thus, one might suggest that ecosystems are like the famous ship of Theseus, a wooden boat that has all its planks replaced, one by one, as they rot away. Even after the ship has had all the planks replaced, some philosophers say, it remains the same ship, because the turnover in individual planks was gradual and never represented a clean break. It is true that all ecosystems have their physical components gradually replaced as generations of organisms give rise to new generations of the same kinds of organisms. In this way we can conceive of an ecosystem persisting for centuries or longer, even though none of the organisms are identical, since its components are the same type of components and performing the same functions. But in the Arctic beaver scenario, both components and functions are changing. It is more akin to seeing planks replaced with wheels, eventually transforming the wooden ship into a chariot. It seems strange to say that the boat has not disappeared or at the very least lost considerable integrity.
To claim that “integrity” can be preserved even across complete transitions in ecosystem form and function over just a few decades suggests a very quixotic definition of “integrity.” Such “integrity” seems to simply be code for “unmanaged and biodiverse.” And, as we will argue in our conclusion, if you are using the term in such a unique way, it would be simpler to just come out and say “unmanaged and biodiverse.”
We have argued that ecosystems are not whole in the sense that they have all their niches saturated, since many “natural” ecosystems seem to admit entrance of a surprising number of new components. We have argued that they are not wholes in the sense that they are so tightly coevolved that they would collapse if one or a few components are removed. And we have argued that they are not integral in the sense of being continuous through time, because rapid and significant ecological changes are common. Ecosystems simply change too often and too thoroughly to be easily described as wholes with obvious integrity. We think the concept is simply a poor fit for the thing it seeks to describe and should not be guiding our conservation actions.
Indeed, prioritizing integrity can compromise other values, including biodiversity. For example, removing non‐native species from a site is a common intervention undertaken to preserve ecosystem integrity. But where the incoming species are “climate refugees” or species that are threatened in their home ranges for other reasons, controlling them could increase the chances of their global extinction.
We think that the best path forward is to simply discard “ecosystem integrity” as a goal or metric. But we understand that the idea is deeply entrenched in the field and woven into policy. In the next section, we will sketch out what we think the most plausible working definition could be for “ecosystem integrity,” using the work of philosopher Jay Odenbaugh as a guide.
Jay Odenbaugh (2007), a philosopher of science, argues that ecosystems are real things whose components (the individual organisms) are bound together via their causal interactions into an entity that can persist through time.
We can use Odenbaugh's definition of what an ecosystem is to imagine what its integrity might consist of. Perhaps the wholeness of an ecosystem is neither a web of obligatory relationships nor a fully‐stocked cabinet of niches. Perhaps it is simply the typical causal interactions between an ecosystem's components as they exist right now.
This may be the most workable definition, but such integrity might be more ephemeral than commonly understood, since, as we have seen, ecosystems are so variable in both space and time. No matter the system, the causal connections between components will shift as the components evolve and change in response to each other, climate changes, geological changes, and so‐called “disturbances,” including storms, fires, volcanic eruptions, floods, and other forces. Links between components can be strengthened, weakened, and even disappear. New links can form with components that were not previously present. Some components will disappear altogether. Odenbaugh (2007) acknowledges this, noting that if ecosystems are constituted and maintained by causal connections, then perhaps ecosystems are simply “much smaller and more ephemeral than ecologists have typically considered” (p. 12 of preprint). If Odenbaugh is correct that ecosystems are relatively small and short‐lived, then the recent geological history of any location may include dozens or hundreds of distinct ecosystems, each with their own distinct integrity, none of which last much more than a few hundred years. In this case, an emerging system or a new system may be misidentified as the degradation of the old ecosystem or a loss of integrity of the old system.
We think this definition—the specific components and their specific relationships at any specific moment—is the most plausible way to understand how ecosystems could have “integrity.” It is close to what ecologists often term “community structure.” In the next section, we will use this definition to examine whether such a concept actually helps describe the loss of value that conservationists see in many contemporary, rapid changes in ecosystems.
We still do not think ecosystems are the kinds of things that have integrity. But in this section we show that even if you insist on using the concept, integrity has a hard time explaining common judgments about loss of ecological value. In fact, in this section we argue that other goals, such as biodiversity and complexity seem to do a better job of accounting for common judgments about the loss of value.
The change from ecosystem A to ecosystem B is usually understood to be a bad thing, a change that amounts to a loss of value, since the integrity of system A has been lost. But, if we accept that ecosystems are fleeting as Odenbaugh proposes, then at any given location there will be many different ecosystems, each with their own integrity, following each other through time. Ecosystem A was preceded by many other previous ecosystems and will be followed by many more. Hence the loss of integrity of ecosystem A cannot explain a loss of value when going from A to B, since when B comes into existence as a whole it will have its own integrity. And there is no reason to think that the integrity of one ecosystem is more valuable than the integrity of another.
A relevant example of this is found in “urchin barrens.” Urchin barrens result when urchins uncontrolled by predation overwhelm a kelp forest. The urchins graze the kelp out of an area and when the kelp goes many of the other organisms in the system go as well. Marzloff and Johnson (2015) characterize this change from a kelp forest to urchin barren as a loss of integrity—“Widespread formation of sea urchin barrens in eastern Tasmanian waters poses a significant threat to the integrity, productivity and biodiversity of shallow (<40 m) rocky reef systems” (ICES CM 2015/B12). However, the urchin barrens are also self‐sustaining systems with causally interacting components that can persist through time. Hence, urchin barrens conform to the most plausible notion of ecosystem integrity. Some have lasted for 90 years (Watanuki et al., 2010). The urchin barrens are thus not highly degraded versions of a kelp forest. They are new systems with their own components, functions, and causal interactions that exhibit persistence through time and resilience against change—and as such they have their own integrity. To be clear: We agree with the judgment of conservationists that in transitioning from the kelp forest to the urchin barrens there is a loss of value; however, we contend that this loss of value cannot be explained by reference to integrity.
Claiming that the integrity of ecosystem A is more valuable than the integrity of ecosystem B seems odd, given that they are both wholes that can persist through some length of time. One might respond and say that ecosystem A existed for much longer and that is why its integrity is more valuable. We contend that this is an error, the result of a confusion between valuing history itself and valuing what history creates. Often, older systems have more diversity and complexity. But it is the diversity and complexity—not the time that it took to develop it—that is valuable. An ecosystem with 20 species and 400 interactions that came together 1,000 years ago does not have higher integrity than an ecosystem with 20 species and 400 interactions that came together 100 years ago.
Likewise, arguing that the urchin barrens has less integrity simply because it has less diversity seems to be a category error. If diversity made for integrity, then all tropical forests would have higher integrity than all boreal, Arctic, or Antarctic ecosystems. And that seems bizarre.
We can easily make sense of two sequential ecosystems having different values if they have different levels of biodiversity and complexity and if biodiversity and complexity are valuable. Indeed, we think this is the real reason we value the kelp forest more. The new system lacks the rich diversity and more complex causal interactions that its predecessor possessed. But if we are talking about diversity, we should just say diversity.
Even if one denies that the urchin barren is a different system and one wants to claim that it is the same system, just in a highly “degraded” state, it seems that what explains the loss of value here is that the system has become less diverse in terms of its components and that the complexity of interactions has decreased. Again, the loss of value seems better explained by focusing on the value of diversity and complexity. Referencing a loss of integrity does not give us a more complete or more useful way to account for the loss of value than referencing the values of diversity and complexity.
We believe that as it is currently used in conservation and restoration, the value of integrity is often an imperfect placeholder for the values of biodiversity and complexity—and in some cases, cultural attachment to specific historical states. Indeed, occasionally one will see this equivalence made explicit in the literature, as when Woodley writes that, “there is no real conflict between many of the terms used to define an ecosystem condition. The terms of “ecological integrity,” “ecosystem health,” “biodiversity,” and “resilience” are really just subsets or derivatives of each other” (Woodley, 2010, p. 153).
Peter Bridgewater notes that Parks Canada seems to think along similar lines: “The goal of conserving ecological integrity is best addressed by maintaining or restoring the diversity of genes, species and communities native to the region” (Parks Canada, supra note 56, quoted in Bridgewater et al., 2014). Here integrity is glossed as essentially equivalent to biodiversity, while the specification of “native” indicates the cultural attachment to a particular historical complement of species.
The real reason that conservationists and restorationists in Western North America have chosen the years 1,600–1850 to use as a reference is because such systems were biodiverse and complex and many Americans have a cultural attachment to these states. All of these seem to us to be good reasons to use this historical moment as a reference and, in some cases, as an explicit goal. We believe that these personal and cultural preferences are important; we also believe they should be presented as they are, not cloaked with the imprimatur of objective ecological fact (Rohwer & Marris, 2016).
For example, in the urchin barrens case, many conservationists, divers, fishers, Indigenous groups, and other stakeholders have deeply felt cultural and personal connections to the kelp forest and its many inhabitants. This attachment, as well as the high diversity and complexity of such systems, are good reasons to conserve them. We do not actually need to appeal to “integrity.”
There is no objective, empirically measurable property of integrity in North American ecosystems that can be shown to be high in 1850 and low in 2020 that is not actually a list of speicies present at a culturally important historical moment or a measurement of biodiversity or complexity of interactions. In short, the concept of “integrity” is a red herring. Ecosystems do not need to be considered as wholes with integrity to be valuable.
Conservation biologists and restorationists should determine what they are really trying to preserve or promote in any given project. Is it really “integrity” or is integrity just a placeholder for complexity, diversity, or a preferred historical state? Humans form strong emotional connections to their environments and have strong preferences about those places, often wanting beloved ecosystems to stay the same. But ecosystems are not valuable because they have integrity. To assert as much sells ecosystems short. They are valuable because they are diverse, because they are complex, and because they are dynamic. Their unending and unpredictable unfolding through space and time in response to changing conditions is not troublesome noise that makes it harder to successfully categorize their essential wholeness; it is central to what makes them fascinating and valuable. Dynamism and complexity are also what makes the endeavors of conservation and restoration so difficult—and so interesting. The entities we seek to protect and care for are weirder and wilder than we can often comprehend.
Before we intervene in the nonhuman world, we should analyze why we seek to act and what we hope to achieve. Allowing the idea of “ecosystem integrity” to guide us is risky. Where it is used as a rough proxy for complexity, diversity, and preferred historic states, it may not lead us astray in the short term. But where actions to protect or promote “integrity” prompt us stop ecosystems from adapting in the face of change, we may find ourselves unwittingly compromising other values we hold dear, including biodiversity.
We would like to thank Stephen Palumbi, Philip Seddon, and Christopher Preston for helpful feedback on earlier versions of this article. We also like to thank the participants at the June 2020 Intended Consequences Workshop, sponsored by The Nelson Institute for Environmental Studies at the University of Wisconsin‐Madison, Gerry Ohrstrom, The Nature Conservancy of California, and Revive & Restore. Special thanks to Michele Weber for facilitating the process. Thanks to the Office of the Provost at Oregon Tech for funding intial research with a Faculty Creativity Grant.
The authors have no known conflict of interest.
Yasha Rohwer and Emma Marris contributed equally in all aspects of developing and writing this article.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
© 2021. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Abstract
We argue that “ecological integrity” is a bad fit as a value for conservation biology and restoration ecology. Both fields are organized around shared values, but it is important to be clear about the specific values and reasons motivating protection of or interventions in specific ecosystems. In practice, appeals to ecological integrity often fail to account for losses in value when ecosystems change. Ultimately, we do not believe ecosystems are the kinds of things that have integrity. Ecosystems are simply too dynamic in space and time, their complex interconnections, including coevolved relationships, ultimately fleeting at the geological scale. Any impression of “wholeness” is an artifact of the brevity of human lives and the shallowness of our historical records. We believe “ecological integrity” as it is currently used is typically a proxy for the values of diversity, complexity, and cultural connections with beloved ecosystem states. We should simply say what we mean and retire the concept of “ecological integrity.”
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
; Marris, Emma 2 1 Department of Humanities and Social Sciences, Oregon Institute of Technology, Klamath Falls, Oregon, USA
2 Klamath Falls, Oregon, USA