Next Article in Journal
Research on Spatial Planning of Petrochemical Industrial Parks from the Perspective of Symbiosis: Example of Yueyang Green Chemical Industry Park
Previous Article in Journal
Morphological and Biochemical Properties in Fruits of Naturally Grown Cornelian Cherry (Cornus mas L.) Genotypes in Northwest Bosnia and Herzegovina
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

How Strong Sustainability Became Safety

Sustainability Institute and Department of Agricultural, Environmental & Development Economics, The Ohio State University, Columbus, OH 43210, USA
Sustainability 2022, 14(8), 4578; https://doi.org/10.3390/su14084578
Submission received: 25 March 2022 / Revised: 5 April 2022 / Accepted: 11 April 2022 / Published: 12 April 2022
(This article belongs to the Section Resources and Sustainable Utilization)

Abstract

:
The core commitment of strong sustainability, SS, is that nature really is different: there are strict limits to the substitutability of natural and other kinds of capital. Initially, the threat to sustainability was perceived as human greed and impatience, and the goal of SS to address resource scarcity was to sustain resource stocks, the flow of environmental services, and/or the harvest for human benefit. For landscapes and ecosystems, the SS goal was preservation, often in a gestalt framing: preserved or not. Two developments beginning around the mid-20th century—increasing awareness of the variability of natural systems, and the revolutionary changes in thinking motivated by the study of complex dynamic systems, CDS—re-oriented SS toward Safety, i.e., minimizing exposure to risk defined as threat of harm. Around 2010, the sustainability agenda for CDS shifted from identifying early warning indicators enabling timely interventions to forestall adverse regime change to promoting resilience by expanding scale and encouraging patchwork patterns of systems in various stages of their adaptive cycles. Nevertheless, the need for natural resources to substitute for depleted exhaustibles suggests a continuing role for commercial agriculture, plantation forestry, and managed fisheries. I conclude with a paradox still to be resolved: the need for continued and increased production from renewable resources to replace depleted exhaustibles suggests SS-motivated management practices that seem obsolete from a CDS perspective.

1. Introduction

Worries about sustainability have been focused on renewable and exhaustible resources at various times in the last 500 years, but the Industrial Revolution substituted exhaustible for renewable resources (e.g., fossil fuels for firewood and charcoal) with great success. Consequently, by the early 19th century, depletion of exhaustibles became the dominant sustainability concern. With booming economies following World War II, fear of running out of resources seemed even more urgent and was focused mostly on exhaustibles. It is that discussion that laid the groundwork for current thinking about sustainability [1]. Two major streams of thinking emerged.
The weak sustainability, WS, literature arose directly from economic growth theory and conjectured that depletion of exhaustibles did not threaten sustainability so long as other kinds of capital could be substituted readily [2,3,4]. WS was motivated by the claim that reasonable expectations about the responsiveness of technology to research effort and the substitutability of different kinds of capital suggest a goal of sustaining human welfare rather than specific resources [2,3,4,5]. The WS narrative relies on discovery, but it is a different kind of discovery: not so much of new deposits of scarce resources, but of new technologies that use plentiful resources to satisfy human demands so that society could enjoy non-diminishing welfare, w(t), even as familiar resources become scarce.
WS posits that welfare can be sustained through the generations so long as inclusive wealth (IW) is non-diminishing. IW is an aggregate construct, and purposefully so: the whole point is that different kinds of capital are substitutable so that increases in one kind can compensate for decreases in another [6,7,8]. The fundamental equation of WS is
W 0 = 0 w ( t ) e g t d t
where W 0 is IW at t = 0, given an infinite time horizon; and w(t) is discounted at g, the growth rate of productivity such that time preference is zero. For a finite accounting period ending at T,
W 0 = 0 T w ( t ) e g t d t + W T ,   w h e r e   W T = T w ( t ) e g t d t  
In a typical accounting of genuine savings, the time period is one year, and accounts are updated annually. Thus, WS status is typically tracked relative to a two-part criterion: the sum of w(t) and W ( t ) should be non-diminishing. Ideally, for intergenerational equity, both components should be non-diminishing. That is, sustainability is possible if welfare can be maintained without diminishing wealth, generation after generation.
The strong sustainability, SS, stream insisted that nature really is different and the WS project of sustaining welfare by substitution among different kinds of capital was more a conjecture than a strategy [9,10,11]. SS thinking proceeded along two tracks: (i) identification and resolution, one by one, of specific threats to sustainability within the broader business-as-usual economy, and (ii) attempts to formulate a coherent comprehensive SS framework [5,12].
In what follows, SS holds center stage and mentions of WS serve mostly to provide contrast. My focus is on the increasing prominence accorded risk, i.e., threat of harm, as a perceived threat to sustainability, and the revolutionary changes in our understanding of natural systems that magnified the perceived risks. In the 1950s and 1960s, models of sigmoid growth were revised to consider stochastic regeneration and it was recommended that managers adopt a safe minimum standard, SMS, of conservation. By the 1970s, abstract complex systems models suggested the possibility of chaotic outcomes, unpredictable but explicable after the fact [13]. Modeling of ecosystems and coupled human and natural systems suggested similar systems properties [14], and confirming evidence began to accumulate in real-world systems for some of the model outcomes, suggesting risks that were greater and different in kind than previously recognized. Together, these influences motivated a major problem shift in SS from sustaining natural resource stocks, ecosystem services and/or the harvest upon which human populations depended, to Safety, i.e., minimizing the chance of system collapse, securing the minimum harvest required by dependent human populations, and assuring timely recovery of damaged systems.
That is not the end of the story. Safety for ecosystems is, of course, a human construct, and while early interpretations of CDS encouraged interventions to forestall regime shifts, the more recent panarchy interpretation suggests that we are likely to have got it wrong: changes in systems do not always justify intervention, and interventions aimed at encouraging virtuous cycles and patchwork spatial patterns are more likely to be successful [15]. Our understanding continues to evolve, and I note an emerging challenge: the apparent inconsistency between our developing understanding of CDS and our continued reliance on simplistic interventionist methods to sustain the enterprises—commercial farms, plantation forests, and managed fisheries—that we rely upon to replace depleted exhaustible resources with renewables.

2. Materials and Methods

This article seeks to develop an interpretive intellectual history of strong sustainability, showing how the basic framework of SS and the policy and management implications thereof changed repeatedly to accommodate the evolving understanding of complex systems and the increasing salience of risk. What follows is not a conventional literature review and makes no claim to be systematic and comprehensive. It is more an exercise in methodology, the study of how and why methods have evolved [16] and the implications of that evolution. Sustainability concepts, especially SS, are addressed to problem solving in the real world, which implies that they must pass two kinds of tests: consistency with emerging knowledge and unrelenting commitment to sustainability goals.
In its simplest form, strong sustainability is perhaps the most intuitive of the various sustainability concepts: identify an impending threat to sustainability involving a particular resource and address it directly with interventions to relieve the stress and permit the resource to recover. This formulation depends crucially on a simple model of the resource and its place in the broader scheme of nature, such that the problem can be solved readily by manipulating the system’s drivers. However, the complexity of the systems we seek to control magnifies the threat of harm from inadvertence and from getting policy and management wrong.
Early 20th-century notions of optimal management of renewable resources serve as a starting point. The focus is on shifts in the foundations of SS, developments in the understanding of human and natural systems, developments in our understanding of risk, and their implications for policy and management of natural resources. I make no claim that we have now figured it all out and, in this particular topic space, the future will be uneventful. There is a paradox still to be resolved: the need for continued and increased production from renewable resources to replace depleted exhaustibles suggests SS-motivated management practices that seem obsolete, and perhaps dangerous, from a CDS perspective.

3. Strong Sustainability

The distinguishing feature of SS has consistently been the insistence that nature really is different, i.e., there are serious limits to the substitutability of other kinds of capital—financial, built, human, social, and governance—for natural resources [9,10,17]. This implies that SS restraints are not fungible in the way that components of IW are. In keeping with its concern about the uniqueness of natural resources, SS offers a limited array of remedies—often restraints on extraction, disturbance, and/or encroachment—focused on the resource itself. Insistence on in-kind remedies limits the options and tends to increase the cost of remediation. SS proponents would counter that in-kind remedies are the best—in the extreme, the only acceptable—remedies.
SS theory and practice have developed along two distinct tracks: (i) sustaining human welfare by ensuring reliable harvests that satisfy some human needs directly and others by substituting renewable resources for depleted exhaustibles, addressed in Section 3.1; and (ii) preserving iconic landscape and ecosystem features by protecting them from encroachment and disturbance (Section 3.2). Until the mid-20th century, there was agreement on the nature of the threat: human greed, which can be blamed for excessive consumption of raw materials and final goods, and for encroachment on landscapes that contain valued features and host valued ecosystems. The path-breaking economic analyses of intertemporal allocation of natural resources from Faustmann [18] through Hotelling [19] emphasized that resources were both potentially productive and a store of wealth, which generates incentives for harmonizing markets in natural and financial resources. Their message was implicitly a plea for taking the future seriously: greater rewards accrue to those who take the long view of their opportunities. However, economics has not always endorsed the activist sustainability agenda, being more inclined to rely on the incentives that encourage technological advances and substitution in production and consumption.

3.1. Strong Sustainability for Resource-Dependent Human Societies

Given the WS challenge, SS supporters motivated by resource scarcity concerns argued that WS proponents had seriously overstated the substitutability among assets of different kinds, especially substitutability between natural resources and other kinds of capital [17] and, in that context, SS is a reasonable restraint on human greed. Sustainable use of renewable resources is a coherent concept, but sustainable use of exhaustibles is an oxymoron. However, the idea of sustainability for a society that uses exhaustible resources could be salvaged if renewable resources could be substituted for depleted exhaustibles. This is not an easy lift: first, continuing technological progress would be needed to expand the domain of substitutability and, second, harvest of renewable resources for their established uses would need to be sustained and additional harvest secured year after year to compensate for increasing depletion of exhaustibles.

3.1.1. The Folk Theorem: Substituting Renewables for Depleted Exhaustibles to Achieve Sustainability

It has long been conjectured that sustainability can be achieved if renewable resources can be sustained and their production expanded enough to compensate for depletion of exhaustibles [20]. Despite continuing difficulties in formalizing this “folk theorem” [21], the message that sustainability can be achieved if enough renewable resources are substituted for depleted exhaustibles has continuing resonance. Additionally, it can be communicated simply, for example, “Extract exhaustibles, plant trees. Cut a tree, plant a tree”. The implication is that, while SS is directed at renewable resources, it addresses depletion of exhaustibles via compensating enhancements of renewable resources.

3.1.2. An Initial Exploration of the Relationships among Resource Stocks, Environmental Services, and Dependable Harvests

Define S as the stock of a renewable resource, E as a vector of environmental services not including harvest for raw materials and/or direct consumption, and H as the harvest. Then, suppress risk by ignoring it or taking refuge in the expected value theorem that recognizes risk but assumes effectively that it does not matter.
First, consider a simple case with substantial practical significance: a renewable resource harvested by humans indifferent to the level of E. At some particular time τ , 0 < τ   < ,   its stock, S τ , is equal to the initial stock, S 0 , minus cumulative net additions or subtractions, i.e., cumulative harvest, H, minus growth (net of spoilage and decay), G, all in per capita terms as suggested by Dasgupta [22]:
S τ = S 0 0 τ ( H ( t ) G ( t ) ) d t .
This is basically an accounting identity. With no additional restrictions, the resource will last forever so long as S0 > 0, S 0 0   ( H ( t ) G ( t ) ) d t   0 and H t 0 , t . That is, H t could drop to zero in some period(s) without extinguishing the resource. However, the notion of harvest suggests a human population that values the harvest and will seek to sustain H t > 0 , t . From a sustainability perspective, there is an optimal H (t) trajectory that, perhaps following a period of adjustment from a suboptimal baseline, stabilizes the harvest at the highest sustainable level thereafter. It is likely that G t = f ( S t 1 ); that is, growth depends on the quantity of the stock carried forward from the previous period, which is reduced by previous-period harvest. Given a positive first derivative G t / S t 1   and a negative second derivative, the optimal H (t), perhaps after a period of adjustment from a suboptimal base, sustains optimal S*(t) and G (t) thereafter. The goal of a SS policy would be to restrain H ( t )   H (t), t . It is important to note that SS does not require optimality in principle or in practice—for example, the common notion of SS as a defense against attrition of resource stocks suggests a goal of setting H ( t ) at the level that sustains S(t)   S 0 , t . Another harvest benchmark of interest for sustainability is Hmin(t), the minimum harvest of an essential resource to ensure the survival of a dependent human population. Essentiality is itself an assumption subject to challenge—are there really no substitutes and is there no way to obtain the resource other than producing it “in house”?—but in cases where the resource is truly essential, a minimal sustainability goal must be to maintain H ( t ) in the range: Hmin,0 = Hmin(t)   H ( t )     H (t), t .
It takes a credible threat to motivate active policy and management for sustainability. In the context of this section, the threat is greed: a generation of people, during their relatively brief time on earth, might seek to harvest more than the target Ht  H t , which would condemn at least one later generation to consuming less. It is also conceivable that a frugal generation might restrain H t   too much, effectively enriching later generations. That is, generations have some opportunity to depart from the SS path; but in so doing, they impose additional costs, or bestow additional benefits, on future generations [23]. While an accounting identity fixes the relationship between H ( t ) and S ( t ) through infinite time (Equation (3)), the opportunity to increase H ( t ) at the expense of S ( t ) and vice versa in finite time implies that real-world tracking of SS status requires monitoring both H ( t ) and S ( t ) .
Consistent with the fundamental premise of SS, the renewable resource may be expected to serve an additional purpose, compensating for increasing scarcity of exhaustibles [20]. So long as the pressure of this scarcity continues to increase, optimal H*(t) will be increasing in t, which requires that G*(t) and (given constant technology) S*(t) be increasing in t. If the resource is essential, Hmin(t) will also be increasing in t. As suggested parenthetically, improvements in technology may increase growth from a given quantity of stock. With this as the base case, now suppose that E(t) is valued.
(i)
Suppose E(t) is valued but H(t) is not, and E(t) = f ( S t ), t , with positive first derivative and negative second derivative. Then, there will be an optimal E (t) trajectory. In the simplest case, where technology of producing E(t) from S(t) is unchanging, the optimal time-path of E (t) implies an optimal time path of S*(t), from a beginning S0 that may be larger or smaller than optimal, that will converge to a constant for the remainder of the time path.
(ii)
Suppose E(t) and H(t) are both valued and they compete such that E t / H t < 0. When the trade-off is resolved, H t | E t   < H t | E t   = 0; that is, concern for E will reduce the optimal harvest. Hmin(t), if relevant, is a given and cannot be adjusted downward to make room for additional E(t).
(iii)
Suppose E(t) is valued but so is D(t), an activity that competes with E(t) and S(t) for a fixed supply of land so that E t / D t < 0. When the trade-off is resolved, E ( t ) | D ( t ) < E ( t ) |D(t) = 0; that is, demand for a competing land-using activity will reduce the optimal quantity of environmental services, an outcome that may be consistent with the value system of WS but is unacceptable to many SS proponents (Section 3.2).
The above analyses model highly stylized cases but suggest directional conclusions that provide insights more generally.

3.1.3. SS for Threatened Resources One by One

Early formulations of SS assumed, at least implicitly, separate accounts for each kind of natural resources [5,12,24]. However, a universal SS criterion—sustain baseline resource stocks and baseline flows of all ES—would amount to “absurdly strong sustainability” [25]. Obviously, some filtering is necessary: there are so many different kinds of natural resources in so many places, not all of them are threatened, not all of them are non-substitutable, and the set threatened by scarcity and/or greed is likely to change over time. Commonsense suggests that SS should be reserved for critical resources, i.e., those that are both essential and threatened. SS attention to critical resources has at least two considerable virtues: it recognizes and potentially mitigates the inherent hazards of reliance on WS assessments, and it is readily directed to monitoring and targeting interventions. Nevertheless, the critical resources approach is inherently unwieldy: it requires monitoring separate accounts not only for each critical resource but also for a set of resources that may become critical, and implementing SS interventions as necessary. Over time, some resources might be added to the critical list and managed sustainably, while others may recover enough to be de-listed.

3.1.4. SS for Constant Natural Capital

Various authors have suggested that the SS goal should be conservation of natural capital, NC, in aggregate. Suppose that (i) SS is motivated by concerns that substitutability among different kinds of resources is limited, (ii) substitutability prospects within the broad class of natural capital are better than between asset classes, and (iii) we are persuaded that SS for all critical resources one by one is too unwieldy and too restrictive given the prospects for substitution among natural resources. Then we have a rationale for conceptualizing SS as maintaining a constant stock of natural capital [12,17,26]. Given a commitment to maintaining constant NC, a couple of questions are immediately obvious.
First, since SS is predicated on the notion that nature really is different, and NC is a category of diverse assets that share the common trait of being natural, what exactly does natural mean? For example, is there a point at which human intervention in a landscape, an ecosystem, or a species makes it no longer natural? Ott [17] argues that nature is not limited to wilderness but extends a long way into the stocks and funds of cultivated natural capital. Furthermore, he writes, the constant NC rule goes beyond halting the on-going loss of NC: it is intended to revitalize and restore the natural world as a whole, and restoration ecology can contribute theoretically and practically.
Second, how is NC, a set of diverse natural assets, to be measured? Measurement is essential for setting a baseline for constant NC policy, which is difficult enough. However, the components of NC are adjusting continually, requiring methods of valuing continually changing sets of NC. Dietz and Neumayer [12] argue adamantly that the assumptions of NC, most notably the insistence on non-substitutability with other components of IW, preclude coherent valuation of NC. Ott [17] resolves the valuation question by delegating it to a process of deliberation among citizens committed to the public good: a proposed configuration of NC would be acceptable if the deliberative process ranked it higher than the current baseline set. To Ott, the constant NC set is continually adjusting but within rules of acceptability. Obviously, Ott’s solution is unassailable if we make the very strong assumption that the deliberative process always makes the right assessments.

3.2. SS Is a Big Tent in Terms of Motivations

Section 3.1 began by exploring the implications of SS motivated by concern for human well-being, which can be promoted by sustaining renewable resources in sufficient quantities to support direct demands for resources and environmental services, and substitute for increasingly scarce exhaustibles. Toward the end of that section, the concept of natural capital raises notions of a priori limits on the safe operating space for humanity [26,27] and a multiplicity of motivating values to be negotiated by deliberation among citizens committed to the common good [17]. In fact, SS has always been a big tent, in terms of motivating values: human welfare, the intrinsic value of some but not necessarily all components of nature, and the deep ecology worldview that nature has a good of its own that limits the scope of justifiable human activity [28]. Since many people hold some but not all of these values, and hold them in some but not all instances, the SS coalition can be fractious: a welfarist compromise reducing E(t) to accommodate a competing land use (Section 3.1.2) is likely to offend some groups of SS proponents, e.g., E-loving welfarists, people who attribute intrinsic value to E(t) and/or the resources that produce them, and deep ecologists. The practical work that leads to adoption of SS policies includes building and maintaining coalitions of people with different motivations; successful policies may combine elements addressing several kinds of threats, e.g., running out of resources, destroying nature, and encroaching on iconic historical and cultural sites.

SS for Landscapes and Ecosystems

Disturbance of, and encroachment upon, landscapes, ecosystems, and iconic sites and structures is typically motivated by demand for the land that supports them. Because WS is unsatisfactory to proponents of preservation—it attends to landscape and ecosystem values, but it does so from a welfarist perspective that offers no guarantees—the Safe Minimum Standard of conservation, originally developed to address stochastic regeneration (Section 4.1.2), quickly found application in these kinds of cases, where it was invoked to restrain disturbance and encroachment even in cases where stochastic risk was not the primary threat. Preservation is often framed as an all-or-none protection against unrestrained greed, but not always: in the case of preservation of species, ecosystem integrity, etc.—a significant motivator for some segments of the SS constituency—conservation goals may be focused on maintaining survivable stocks and adequate habitat. When baseline stocks are unsustainably low, restoration may play a role among the SS remedies.

4. The Emergence of a Safety Interpretation of SS

The early economic analyses of renewable resources identified greed—the motivation to consume more and sooner than a long-term view would recommend—as the primary threat to sustainability. It is hard to imagine that these authors were unaware of risk: Bernoulli had written compellingly on decision making in the presence of risk in 1738 [29]. Perhaps they took refuge, at least implicitly, in the expected value theorem and certainty-equivalence: the outcome of a long series of fair gambles will converge toward its statistically expected value provided the gambler has sufficient resources to survive a run of losses. That is, they most likely were aware of risk but thought it an unnecessary distraction from the general principles they were establishing.
Modern accounts of risk typically begin with Frank Knight’s influential treatise of 1921 [30]. While ordinary usage defines risk as “threat of harm” [31], Knight took a statistical perspective: suppose a stochastic system generates a normal distribution of outcomes. Then, the distribution may include gains and losses in decision maker welfare. The decision maker lacks clairvoyance but must decide, before the outcome is revealed, whether to “play the game” and what risk management strategy to adopt. This decision might seem daunting enough, but Knight also recognized the idea of uncertainty and applied that term to cases where the distribution of outcomes is unknown. Newtonian notions of orderly systems equilibrating following a disturbance have been challenged by complex dynamic systems, CDS, models suggesting that outcomes may be chaotic. As technology continues to enhance human capacity to intervene in systems we do not really understand, the more disturbing notions of deep uncertainty, gross ignorance, and unknown unknowns have gained currency and the domain of risk as threat of harm has grown dramatically. We have become more conscious of risk, and the risks we perceive have become vastly more threatening. It is unsurprising that risk now takes the lead among the perceived threats to sustainability. Greed, of course, has remained a threat to sustainability: while the spotlight shifts toward risk, greed motivates the pursuit of private profits while market failures socialize the risks. Thus, risk and greed are complements in magnifying the threat.

4.1. Stochastic Systems

Modern accounts of risk typically follow Knight’s lead, distinguishing between risk and uncertainty, and treating risk as stochastic, i.e., a matter of random draws from known and stable distributions. Obviously, stochastic regeneration changes the SS objective(s). Rather than sustaining a baseline or perhaps desired level of S(t), E(t), and/or H(t), the decision maker’s priority shifts toward safety, i.e., avoiding collapse of the resource and ensuring its recovery following a worst-case outcome. Sustainable, in this context, has come to mean survivable. In the case of a resource valued for its existence, the challenge is represented as a knife-edge: get it right and the stock is on a path to recovery, get it wrong and the resource is lost forever. In the case where a human undertaking—a farm or forest, a fishery, or an isolated society—is dependent on a renewable resource, the challenge is to sustain at least a minimally adequate regular harvest. In this formulation, SS is intended to protect against exhaustion or extinction of a critical resource, or failure of a critical harvest, as the case may be.

4.1.1. Stochasticity in Renewable Resource Systems—Safeguarding the Harvest

The first crack in the wall of certainty equivalence was attributable to Ciriacy-Wantrup in 1952 and 1968 [32], who worried that relationships such as outlined in Section 3.1.2 really are stochastic and that, while gains and losses often were modeled as statistically symmetrical, their human consequences were asymmetric: a worst-case loss could be devastating. One of the risks that are prominent in the SS discussion concerns regeneration of the resource. If all goes well, regeneration is sufficient to sustain the resource itself plus a desired harvest; if the regeneration outcome is bad enough, it may doom the harvest and the resource itself may not recover. Farms, forests, and fisheries, for example, with stochastic regeneration may experience worst-case regeneration outcomes. In open-access forests and fisheries, lack of incentives for restraint in harvest only exacerbates the problem—that is, greed in addition to stochastic risk is magnified by socialization of its consequences.
Ciriacy-Wantrup recommended SS protections in the form of a safe minimum standard of conservation, SMS, which should be adopted by farmers and foresters, and imposed by the fisheries regulator. A common exposition of the SMS rule assumes a perennial crop with stochastic regeneration normally distributed around a sigmoid growth curve such that harvest can be sustained in an average year without threatening regeneration. However, it is a feature of sigmoid regeneration that there exists a quantity of stock, Smin, below which the system collapses. Given a serious threat of collapse, the SMS would call for suspension of harvesting until the crop recovers enough to support safe resumption of harvest. If regeneration is deterministic a Safety constraint would maintain S(t)  Smin, but with stochastic regeneration a Safety constraint seeks to maintain a larger stock that would enable eventual recovery in the event of a worst-case regeneration outcome. If safety for a dependent human population requires H(t)  Hmin, the Safety constraint would maintain an even greater S(t) sufficient to secure Hmin in a worst-case season and eventual recovery of the stock.
Because the Safety constraint suspends harvest until the stock has recovered, a dependent human population with diversified income sources, savings and access to external markets may be able to wait several seasons until eventual recovery [23]. In the absence of these risk management opportunities, recovery in time for the next harvest may be essential. Thus, what counts as Safety is circumstantial, depending on whether the goal is to secure the ecosystem or the ecosystem and the H(t) path, and the risk management opportunities available to a dependent human population.

4.1.2. Adapting Ciriacy-Wantrup’s SMS to Preservation of Unique Natural Assets

SS is often explained and justified by appealing to special cases, e.g., an isolated community dependent on reliable harvests, an endangered species, or a unique ecosystem. Bishop, who was Ciriacy-Wantrup’s student, became the first to use respect for natural assets (in his case, an endangered charismatic species) per se, rather than the perceived need to secure the harvest by sustaining the resource stock, to justify an SMS application [33]. While the threat recognized by Bishop was stochastic risk of extinction, exacerbated by disturbance and encroachment, and his justification for an SMS intervention relied on risk aversion, there soon emerged cases, e.g., unique and iconic landscape features, and historic sites and structures, for which SS involves preservation as an all-or-none matter. Unique ecosystems may provide an intermediate case where there may be a minimum scale below which the ecosystem is impaired beyond recovery but, beyond that minimum, increasing scale is not necessarily an issue for survival.

4.2. Complex Dynamic Systems, Phase 1

Knight, in 1921, had left the door open to uncertainty, which does not obey the rules of stochasticity [30]. However, there seemed to be an implicit assumption that uncertainty was attributed to people: we are uncertain because we do not know enough. Two ways of learning more suggested themselves: we could collect more data to enrich our empirical knowledge (the frequentist approach to statistics), or we could learn by revising our prior beliefs in light of additional observations (as suggested by Thomas Bayes [34] in an essay read posthumously to the Royal Society in 1763 and eventually rediscovered by 20th-century statisticians). More recently, the concept of complex dynamic systems—introduced by Poincare in 1892 [35] and Birhkoff in 1927 [36] and pursued by Russian mathematicians during the Cold War years—raised the prospect of unexpected discontinuities in system configurations and emergent outcomes from natural systems. That is, uncertainty may be a property of natural systems instead of, or in addition to, a consequence of human ignorance.
Since approximately 1970, emerging concepts of CDS have induced a foundational problem shift in SS theory, whose proponents became more aware of the limitations of stochastic risk formulations and more concerned with threats posed by uncertainty, ambiguity, and gross ignorance. May’s abstract CDS models failed to support the diversity-stability hypothesis [13]—a widely held proposition among ecologists that a diverse ecosystem is likely to be more stable than a relatively specialized one—and subsequent ecological modeling has confirmed May’s conjecture. Holling [14,37] suggested that ecosystems, if stressed beyond their tipping points, may undergo regime shifts that are difficult to reverse. CDS have properties that introduce uncertainty well beyond stochasticity and cutting-edge scholarship shifted to address the implications of these emerging concepts of risk. Suggested remedies included interventions aimed at counteracting potential threats to ecosystem stability, and building and maintaining resilience to defend against system collapse and promote recovery.
CDS theory, phase 1, CDS1, suggested that complex systems may tend toward stability within a given range, called the regime, but beyond that range may explode or collapse. Even within the regime, systems may cycle rather than simply find an equilibrium and stay there until disturbed. Population from one period to next, traditionally modeled with simple growth functions, may take a variety of relationships. A ball and cup analogy is often used to illustrate the notions of tipping points, regime shifts, and the asymmetry of change and reversion [14,38,39,40]. Subjected to modest stresses the ball will stay in the cup, but given greater stresses it may fall, perhaps coming to rest in a lower cup. The accompanying narrative defines the fall from the prior cup as a regime shift, suggests that post-shift regimes tend to be worse than the prior regime, and observes that the energy needed to induce a regime change is often much less than the energy that would be required to reverse it. CDS1 and the ball and cup metaphor served to elevate the salience of SS as Safety aimed at preserving the regime.
The ball and cup metaphor encouraged a search for early warning indicators of potential regime shift that would signal the need for timely interventions to “keep the ball in the cup”, i.e., forestall adverse regime change: only then can we try to manage the resource for E(t) and/or H(t). The working hypothesis was that if a shallow lake, for example, is susceptible to harmful algal blooms when pollution loads overwhelm the ecosystem, a reliable early warning indicator may enable prompt and effective intervention to stabilize it [41,42]. However, the chaotic properties suggested by CDS theory are at odds with this agenda: CDS are inherently less stable and predictable than would be required to assure sustainable yields and reliable timely inventions. In retrospect, the high point of Safety—understood as targeted interventions to forestall regime change, stabilize systems, and manage them for sustainable yields—may have been reached around 2010. Since then, the concept of panarchy, noted by Holling decades earlier, has become more influential and researchers have become less optimistic about targeted stabilizing interventions and more inclined toward promoting resilience viewed as a systems property that perhaps can be promoted by management and policy.

4.3. Complex Dynamic Systems, Phase 2

Recent interpretations of CDS theory, CDS2, for natural systems place more weight on panarchy, which implies that cycles of flourishing and collapse may be the norm, and timely, purpose-driven stabilizing interventions are likely to be difficult and perhaps futile. These insights have encouraged “design with nature” approaches to policy and management that attend more to promoting resilience.

4.3.1. Panarchy

By 1986, Holling had introduced adaptive cycles as models of natural change in ecosystems [43]. There may be several cycles at different levels of scale. Each cycle, often depicted as the number 8 on its side, has four distinct phases: (i) growth which may support exploitation, (ii) conservation of established patterns, (iii) breakdown and release, and (iv) reorganization. Within-scale structures and processes interact across scales at key phases of the adaptive cycle. These cross-scale interactions include some that are bottom-up, so that lower levels drive the action, and some that are top-down as we might expect. By 2002, Holling and colleagues [44] adopted the label panarchy for this pattern of adaptive cycles driven by cross-level interactions in both directions. For the standard depiction of adaptive cycles and panarchy in ecology, see Figure 1 in [15]. The exemplar is a coniferous ecosystem and the levels are organized hierarchically, ranging from a needle on a branch, to the crown of a conifer tree, a small patch of forest, and a broad stand of forest. Each level has its own adaptive cycle, and the cycles are linked by cross-scale interactions suggesting that power and influence travel in both directions. The system works better when these linkages are strong.
Panarchy is a conceptual framework that accounts for the dual and seemingly contradictory characteristics of all complex systems: stability and change. Allen et al. [15], noting that panarchy has often been used as a metaphor rather than a rigorous scientific concept, identified three core testable propositions derived therefrom and summarized the available empirical evidence as of 2014:
  • Complex systems are discontinuously structured. The hypothesis that key variables are distributed discontinuously has been confirmed in a wide variety of contexts including animal body mass, city size, firm size, and aquatic communities. Several tests in shallow lakes have confirmed that variability increases when regime change is approached.
  • Complex systems undergo cycles of renewal and destruction. Cycling is observed in a wide class of complex systems, and there is some evidence that cycles of renewal and collapse occur at distinct spatial and temporal scales.
  • Cross-scale linkages are critical to system structure. Panarchy predicts the existence of cases where small-scale variables control system dynamics. Confirming evidence has been observed for contagious processes, e.g., pest outbreaks and wildfire. There is also some empirical support for the conjecture that cross-scale distribution of system functions is critical to maintaining system
Allen et al. [15] also note that some implications for management have emerged. Consider two types of regime changes. First, regime changes that occur at fixed scale, e.g., eutrophication of shallow lakes, may be foreshadowed by increased variability in system properties, a indicator that may facilitate management interventions. Second, cross-scale changes are more likely to occur in systems at the edges of their scale, where variability increases. Examples include invasions and extinctions: increasing variability may challenge struggling incumbent species and provide opportunity for potential invaders. Interestingly, given the impulse common among ecologists to repel the invaders, Allen et al. are open to possibilities that invaders offer a kind of renewal opportunity for the system.
Since 2014, the literature has become richer, especially in number and diversity of case studies, but its tone is little changed. Recent case studies have examined grass-trees transition in the great plains [45], floodplain ecosystems [46], marshlands [47], and limed lakes [48]. The basic message of CDS2—complex systems are discontinuously structured and experience cycles of destruction and renewal, and cross-scale linkages really matter—is confirmed in case after case. Much like Allen et al. in 2014, Chaffin [49] encourages empirical testing of relationships among fast and slow variables in panarchic systems. He hypothesizes that governance variables are slow and suggests that even tentative conclusions regarding policy and management will be relatively few until we know more about how these systems work.

4.3.2. The Concept of Resilience Evolved along with the Understanding of CDS

Holling, in 1973, introduced the notion of resilience [14] as the ability of a system to absorb disturbances and maintain or recover form, structure, and relationships. This looks a lot like the popular notion of bouncing back from adversity, but it provides more clarity about what bouncing back means. Parenthetically, this concept of resilience remains current among engineers and planners. By 1978, Holling had amended his earlier definition to include “… persist through, and even benefit from, change” [50]. More recently, resilience has incorporated notions of systems exploiting instability for adapting and evolving, and humans intervening to build or maintain resilience in systems [51]. As Canizares et al. note, this amounted to a slow-motion embrace by ecologists of resilience as antifragility [51]: a property of systems that increase in capability to thrive as a result of stressors, shocks, volatility, noise, mistakes, faults, attacks, or failures [52]. To the extent that panarchy has implications for policy and management, they are mostly about building and maintaining resilience and recognizing the capacity of systems to exploit instability in positive ways.

4.4. Implications for Policy and Management

We have seen how awareness and understanding of risk and its implications have evolved and responded to revolutionary changes in the way humans view the workings of nature. Stochastic risk can be threatening, even though the outcome distributions are symmetric, when losses are more devastating than gains of similar dimension are beneficent. Potential harm is perceived as greater yet, and more disturbing, when we view ecosystems through a complexity lens that denies simple cause-and-effect Newtonian logic, thereby challenging reliance on effective interventions to restore and/or stabilize systems that seem damaged or threatened. The motivations for reconfiguring SS as Safety are clear but the evolving understanding of risky and complex systems provokes questions about the meaning of Safety as well as the means and challenges of attaining it.

4.4.1. Stochastic Risk in Newtonian Systems

The SMS is designed to avoid worst-case outcomes in stochastic Newtonian systems. Thus, it assumes a knowable and stable distribution of outcomes in a system that tends toward equilibration and stability, and is likely to recover following modest stresses but vulnerable to shocks from excessive exploitation, disturbance and encroachment. SMS interventions are intended to avoid system collapse and, if relevant, secure adequate H(t) and/or E(t). A typical intervention would restrict H(t) to truncate the bad tail of a stochastic regeneration distribution, resuming harvest only when the system has recovered sufficiently to tolerate it without risk of collapse. For iconic landscape and ecosystem features, the goal of SMS intervention is preservation by eliminating intrusive shocks. SMS is potentially applicable at geographical scope from the pond to the globe: from save my little fishery to save the species from local extirpation or global extinction.
SMS made a significant shift in the orientation of SS, tilting it away from maintaining baseline or desired levels S(t), E(t), and/or H(t), and toward risk avoidance, i.e., Safety. It is a is a primitive kind of SS for risky systems. Nevertheless, SMS—motivated by extreme risk aversion to avoid outcomes deemed unacceptable—is a precursor to the precautionary principle, PP: avoid actions with potential consequences that include unacceptably bad, even if unlikely, outcomes. The whole SMS project assumes implicitly that stabilizing the system is good and managers know how to do it. In contrast, as early as 1970s, Holling thought that such interventions amounted to “living dangerously” [14]—that is, intervening confidently in systems we fail to understand comprehensively.

4.4.2. Complex Dynamic Systems, Phase 1

Modeling demonstrated that CDS can be much less predictable, and consequently much less manageable, than Newtonian models suggest. Discontinuities are commonplace, system behavior may be chaotic and unpredictable even if it can be explained after the event, and cross-scale linkages play important roles. The ball and cup analogy played an outsized role in CDS1 thinking: systems have a tendency to vary while remaining within a range of values for key variables (called the regime) but, if the system is stressed beyond that range, a regime shift may occur. In the standard telling, the post-shift regime is objectively less desirable than the baseline regime, and the energy needed to precipitate regime shift is much less than would be required to reverse the change and restore the prior regime.
Risk management strategies based on stochastic games were obviously inadequate to address the implications of complex dynamic systems. Nevertheless, the quest for effective intervention strategies continued. The working hypothesis that systems experience increased variability as they approach regime shift motivated attempts to validate early warning indicators that would enable stabilizing interventions, to “keep the ball in the cup”. Attention also turned to resilience and its place in policy and management. Initially, resilience was conceived in Newtonian terms—a resilient system absorbs shocks, resists change, and eventually recovers prior form and function, i.e., “remains in the cup”—although Holling and others raised the possibility that resilience might permit some adaptation and continued evolution. Safety involved successfully resisting regime shift, i.e., avoiding the presumably less desirable regime that would follow a shift. In that respect, Safety adapted to the initial understanding of CDS while retaining its proactive interventionist flavor. However, Safety-based management also included promoting the resilience of systems considered desirable, which in the case of ecosystems was and perhaps still is more a goal than a program.

4.4.3. Complex Dynamic Systems, Phase 2

At this point in the evolution of CDS2, the key lesson for management seems to be: if we really understand what CDS is teaching us, we are now a lot less sure about how to manage ecosystems and coupled human and natural systems than we may have thought we were. CDS2 questions whether post-shift regimes are generally worse than their predecessors; whether biological invasions are always bad and interventions to resist them are desirable; whether managers should intervene to fortify a declining system, or welcome its decline; and whether, if cycles of destruction and renewal are the norm, managers should intervene even if only to hasten the renewal? Should policy and management focus on interventions or on establishing the preconditions for virtuous cycles, e.g., expanding the spatial and temporal dimensions, which would accommodate patchwork patterns that stabilize in the large what cannot be stabilized by interrupting the cycles? People play an outsized role in so many coupled human and natural systems, and many of them approach natural systems from an interventionist perspective. It follows that stakeholder engagement is important in gaining acceptance for radically different approaches, e.g., using fire as a beneficial force when spatial patterns are patchy. In all of this, Holling has been prescient, noting much earlier that efficiency-based management dominates, but it encourages “living dangerously” [14]. He suggested design with nature at regional scale, to allow and encourage patchwork landscape patterns, all within a policy and management environment that is decentralized and participatory.
Given the current tentative and incomplete understanding of the implications of CDS2, there is an understandable reticence regarding interventionist policy and management. There is more confidence in recommendations for enhancing resilience by, for example, planning and managing at expanded spatial and temporal scales and encouraging mosaic patterns of systems in various stages in the adaptive cycle. This may achieve a kind of stabilization, but only if viewed at a more aggregate scale. Stabilizing my plot may be hopeless: SS is now about design with nature.
Perhaps policy and management interventions can be categorized as fast and slow, much like panarchic processes. Fast interventions would include timely actions aimed at avoiding system collapse and forestalling regime shifts; slow interventions would include those that aim to allow the system more space and time in which to complete its adaptive cycles. Both of those approaches are interventions, and I make no assumptions about which might be more costly in aggregate. We might imagine a proliferation of fast (i.e., targeted and timely) interventions and fewer slow interventions, but the latter are likely to incur substantial opportunity costs, may be harder to calibrate, and the results may be less predictable.

4.5. Can SS Be a Comprehensive Sustainability Program?

SS, as introduced in Section 3, is a set of specific and perhaps uncoordinated interventions to address specific threats. The idea is to troubleshoot business as usual, BAU, in order to avoid or forestall specific threats to sustainability. Not much attention is paid to BAU itself. Much of the discussion simply assumed that BAU means a fairly well-articulated economy responding to the usual economic incentives. It is easy to imagine circumstances in such an economy that call for some additional attention to sustainability, but it is less clear that the BAU default is close enough to sustainability that a manageably few SS patches can make it sustainable. As early as 1991, I proposed explicitly that the BAU economy should be governed so as to maximize social well-being subject to SMS constraints as necessary [53], and more recently Irwin et al. [8] proposed what they call WS-plus, a set of SS restraints that would be superimposed on a WS economy to protect specific stressed resources.
Section 3 also acknowledges proposals that SS be framed as sustaining natural capital in aggregate. NC approaches were introduced as positing that nature really is different from other forms of capital and substitutability prospects within the broad class of natural capital are better than between asset classes. NC approaches obviously are more comprehensive than piecemeal SS, but they risk raising costs and reducing well-being by proscribing cross-category substitutions that would have worked out well. Furthermore, substitution within NC proceeds with relatively few restrictions, so sustaining NC calls for regular judgments about the relative worth of different NC packages. That is, it is the value of NC that must be sustained while the fundamental assumptions of SS are honored, raising an apparently insoluble problem [12].
The planetary boundaries, PBs, framework [54,55] is a more plausible attempt at a comprehensive SS framework. There is no guarantee that the safe operating space, SOS, for human activity will sustain human ambitions, although Rocktrom et al. [27] are optimistic in this respect. Rather, the prevailing message is that the human population and way of living need to adjust to the SOS since there are limits to the reverse adjustment. At least two challenges to the PBs framework are obvious. (1) The comprehensiveness of the PBs at the planetary scale is strained. For example, assimilation capacity for atmospheric carbon and greenhouse gases, C-GHGs, is unambiguously a global resource; but freshwater is used and managed at local and regional scales; and land resources are meaningful at multiple scales from global for food supply to local for urban greenspace. (2) Global-scale constraints provide useful accounting at the global level, but much of the appeal of SS lies in its applicability to local and regional issues. Furthermore, a comprehensive SS program based on the PBs would also need to accommodate specific SS provisions for unique and treasured landscapes and ecosystems [26].

4.6. SS as Safety

CDS1 accelerated the problem shift away from SS defined as sustaining baseline stocks and/or harvest, and toward SS as pursuit of Safety defined around system collapse and recovery. Sustaining a non-diminishing stock of any resource or flow of resource services may not be the primary goal of Safety—in many applications, Safety does not do that at all. What, exactly, constitutes Safety depends on the circumstances. For ecosystems valued for their existence and uniqueness, Safety might involve avoiding collapse or, if the chance of collapse cannot be reduced to zero, maximizing the chance of recovery. Where an isolated human society depends on harvest, risk management considerations might be tilted toward survival and welfare of the human population. The need for Safety to sustain a dependent human society is mitigated to various extents by increasing scale, diversification, and interconnections that facilitate saving, borrowing, and trade [23].
The implications of CDS2 for policy and management remain a work in progress, raising misgivings regarding Safety in the form of timely interventions to stabilize the regime. There is more enthusiasm for enhancing resilience by expanding spatial and temporal scales and encouraging mosaic patterns of systems in various stages in the adaptive cycle. This may achieve a kind of Safety, but only if viewed at a more aggregate scale.

5. Conclusions

Barbier [1] has recently revisited the distinction between Ricardian (relative) and Malthusian (absolute) perspectives on resource scarcity. WS is more about adjustments to changes in relative scarcity, while SS is more about sustaining critical resources. This article has traced the broad outlines of SS from its beginnings as a program to sustain the levels of environmental services and harvestable output produced by renewable resources, through a series of revisions motivated by revolutionary developments in CDS modeling of ecosystems and heightened concern about the risks facing people in coupled human and natural systems. SS has developed along two distinct tracks—(i) sustaining human welfare directly and by substituting renewable resources for depleted exhaustibles, and (ii) sustaining landscapes and ecosystems for a variety of motivations—which attract different coalitions of supporters depending on the framing of the threat and the solutions under consideration. Initially, the goal of SS for resource scarcity was to sustain resource stocks, S, the flow of environmental services, E, and/or the harvest, H, for human benefit. For landscapes and ecosystems, the SS goal was preservation, often in a gestalt framing: preserved or not. The threat was understood as greed—people are tempted to exploit resources and disturb natural systems without adequate concern for future generations—often exacerbated by market failures.
Ciriacy-Wantrup in the 1950s changed the conversation by focusing on stochastic risk to regeneration as the threat to sustainable harvest [32]. He suggested a safe minimum standard of conservation, SMS, as an intervention to avoid worst-case outcomes, ensuring regeneration and eventual recovery of sustainable harvest. Bishop, in the 1970s, applied the SMS to threatened landscapes and ecosystems where the objective is likely to be preservation. Around 1970, Holling applied the emerging study of complex dynamic systems, CDS, to ecosystems, contrasting the prevailing simple models of S, E, and H with the chaotic properties suggested by CDS theory [14,50]. Stressing a natural system beyond its tolerance may induce a regime shift that is difficult to reverse. Until approximately 2010, the sustainability agenda for CDS emphasized identifying early warning indicators that would prompt timely interventions to forestall adverse regime shifts. SS had become Safety, aimed at saving the regime: only then can we try to manage the resource for E or H. Recent renditions of CDS place more weight on panarchy, which implies that cycles of flourishing and collapse may be the norm, resilience makes sense as a goal of policy and management, and human interventions may best take the form of expanding scale and encouraging patchwork patterns of systems in various stages in the cycle. Stabilizing my small plot may be hopeless in the long run: SS is now about design with nature, but the panarchy framing has generated relatively little detailed policy and management advice.
The increasing concern about risk, the evolving understanding of complex dynamic systems, the reframing of SS as Safety, and newly recognized limits on policy and management interventions combine to elevate the policy salience of resilience, particularly the kinds of resilience that would allow natural systems more space and time in which to complete their adaptive cycles. However, one of the key motivations for SS is to replace depleted exhaustible resources with sustainable renewables. It seems likely that commercial agriculture, plantation forestry, and managed fisheries—the sorts of economic undertakings for which SMS was first proposed—will continue to play substantial and important roles in this quest.
Engineers, fully aware that Einstein’s theory has replaced Newton’s among physicists, have noted that it takes Einstein to go to the moon but Newton is good enough for building a bridge. Is there an analogy in SS policy? Does something similar apply to ecosystems? Is there a role, perhaps a need, for brute-force interventionist management of farms, plantation forests, and commercial fisheries, despite our emerging understanding of CDS2 and its implications? Casual observation and inference suggest that there are niches where earlier versions of Safety still are serviceable, perhaps even the simple model where full recovery is attainable given time [56] and the SMS narrative in which recovery is assured if the SMS is observed but impossible if it is breached. The ball in cup model applies to a considerable set of agricultural situations, e.g., perennial pastures (e.g., atriplex spp. in the semi-arid zones of Australia [57] and the great plains grasslands in the US) that withstand intermittent and low-intensity grazing but collapse under over-grazing and are very difficult to restore; and salinization in arid-zone irrigation areas [58] and semi-arid drylands where annual cropping has replaced deep-rooted woody species [59]. Perhaps this broader menu of management options will persist, and the managers’ task will include fitting the appropriate interventions to the various challenges encountered.
My conclusions are not of the “end of history” kind; that is, there is no claim that we have now figured it all out and, in this particular topic space, the future will be uneventful. The “Einstein and Newton” argument tends to support the kinds of interventions that Holling called “living dangerously” [14], which suggests that a full research agenda lies ahead. Rather, I conclude with a paradox still to be resolved: the need for continued and increased production from renewable resources to replace depleted exhaustibles suggests SS-motivated management practices that seem obsolete from a CDS perspective.

Funding

Research support was provided by National Science Foundation Innovations at the Nexus of Food Energy Water Systems grant INFEWS #1739909 and the National Institute for Food and Agriculture award #2018-68002-27932.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

I appreciate continuing discussions on these topics with Elena Irwin and colleagues on the above-mentioned projects, and helpful suggestions from this journal’s reviewers.

Conflicts of Interest

The author declares no conflict of interest.

References

  1. Barbier, E.B. The Evolution of Economic Views on Natural Resource Scarcity. Rev. Environ. Econ. Policy 2021, 15, 24–44. [Google Scholar] [CrossRef]
  2. Solow, R. Intergenerational Equity and Exhaustible Resources. Rev. Econ. Stud. Symp. Econ. Exhaustible Resour. 1974, 41, 29–45. [Google Scholar] [CrossRef] [Green Version]
  3. Hartwick, J. Intergenerational Equity and the Investing of Rents from Exhaustible Resources. Am. Econ. Rev. 1977, 67, 972–974. [Google Scholar]
  4. Turner, R.K. Speculations on Weak and Strong Sustainability; CSERGE Working; Centre for Social and Economic Research on the Global Environment: London, UK, 1992. [Google Scholar]
  5. Neumayer, E. Weak Versus Strong Sustainability: Exploring the Limits of Two Opposing Paradigms, 4th ed.; Edward Elgar: Londin, UK, 2013. [Google Scholar]
  6. Pearce, D.; Atkinson, G. Capital Theory and the Measurement of Sustainable Development: An Indicator of Weak Sustainability. Ecol. Econ. 1993, 8, 103–108. [Google Scholar] [CrossRef]
  7. Arrow, K.J.; Dasgupta, P.; Goulder, L.H.; Mumford, K.J.; Oleson, K. Sustainability and the Measurement of Wealth. Environ. Dev. Econ. 2012, 17, 317–353. [Google Scholar] [CrossRef] [Green Version]
  8. Irwin, E.G.; Gopalakrishnan, S.; Randall, A. Wealth, Welfare, and Sustainability. Annu. Rev. Resour. Econ. 2016, 8, 77–98. [Google Scholar] [CrossRef] [Green Version]
  9. Ekins, P. Strong Sustainability and Critical Natural Capital. In Handbook of Sustainable Development, 2nd ed.; Atkinson, G., Dietz, S., Neumayer, E., Agarwala, M., Eds.; Edward Elgar: Cheltenham, UK, 2014; pp. 55–71. [Google Scholar]
  10. Ott, K. The Case for Strong Sustainability. In Greifswald’s Environmental Ethics; Ott, K., Thapa, P.P., Eds.; Steinbecker Verlag Ulrich Rose: Greifswald, Germany, 2003; pp. 59–64. [Google Scholar]
  11. Cohen, F.; Hepburn, C.J.; Teytelboym, A. Is Natural Capital Really Substitutable? Annu. Rev. Environ. Resour. 2019, 44, 425–448. [Google Scholar] [CrossRef]
  12. Dietz, S.; Neumayer, E. Weak and Strong Sustainability in the SEEA: Concepts and Measurement. Ecol. Econ. 2007, 60, 617–626. [Google Scholar] [CrossRef] [Green Version]
  13. May, R.M. Will a Large Complex System Be Stable? Nature 1972, 238, 413–414. [Google Scholar] [CrossRef]
  14. Holling, C.S. Resilience and Stability of Ecological Systems. Annu. Rev. Ecol. Syst. 1973, 4, 23. [Google Scholar] [CrossRef] [Green Version]
  15. Allen, C.R.; Angeler, D.G.; Garmestani, A.S.; Gunderson, L.H.; Holling, C.S. Panarchy: Theory and Application. Ecosystems 2014, 17, 578–589. [Google Scholar] [CrossRef] [Green Version]
  16. Hepburn, B.; Andersen, H. Scientific Method; Zalta, E.N., Ed.; The Stanford Encyclopedia of Philosophy: Stanford, CA, USA, 2021. [Google Scholar]
  17. Ott, K. On Sustaining the Conception of Strong Sustainability; KIT Scientific Publishing: Karlsruhe, Germany, 2011. [Google Scholar]
  18. Faustmann, M. On the Determination of the Value Which Forest Land and Immature Stand Pose for Forestry. Allg. Forstund Jagdztg. 1849, 15, 441–455. [Google Scholar]
  19. Hotelling, H. The Economics of Exhaustible Resources. J. Political Econ. 1931, 39, 137–175. [Google Scholar] [CrossRef]
  20. El Serafy, S. The Proper Calculation of Income from Depletable Natural Resources. In Environmental Accounting for Sustainable Development; Ahmad, Y.J., Serafy, S., Lutz, E., Eds.; A UJiNEP-WorldBank Symposium: Washington, DC, USA, 1989. [Google Scholar]
  21. Asheim, G. Hartwick’s rule. Encycl. Energy Nat. Resour. Environ. Econ. 2013, 2, 314–320. [Google Scholar]
  22. Dasgupta, P. Time and Generations: Population Ethics for a Diminishing Planet; Columbia University Press: New York, NY, USA, 2019. [Google Scholar]
  23. Randall, A. Intergenerational Commitment, Weak Sustainability, and Safety. Sustainability 2020, 12, 5381. [Google Scholar] [CrossRef]
  24. Helm, D. Natural Capital: Assets, Systems, and Policies. Oxf. Rev. Econ. Policy 2019, 35, 1–13. [Google Scholar] [CrossRef]
  25. Daly, H.E. On Wilfred Beckerman’s Critique of Sustainable Development. Environ. Values 1995, 4, 49–55. [Google Scholar] [CrossRef]
  26. Barbier, E.B.; Burgess, J.C. Natural Resource Economics, Planetary Boundaries, and Strong Sustainability. Sustainability 2017, 9, 1858. [Google Scholar] [CrossRef] [Green Version]
  27. Rockström, J.J.D.S.; Sachs, J.D.; Marcus, C.Ö.; Guido, S.-T. Sustainable Development and Planetary Boundaries; JSTOR: New York, NY, USA, 2013. [Google Scholar]
  28. Randall, A. Environmental ethics for environmental economists. In Encyclopedia of Energy, Natural Resource, and Environmental Economics; Shogren, J., Shortle, J., Eds.; Elsevier: Amsterdam, The Netherlands, 2013; Volume 3, pp. 25–32. [Google Scholar]
  29. Bernoulli, D. Hydrodynamica. Dulsecker. Consultable En Ligne. 1738. Available online: https://pure.tudelft.nl/ws/files/4414689/Westerweel2016b.pdf (accessed on 25 March 2022).
  30. Knight, F.H. Risk, Uncertainty and Profit; Houghton Mifflin: Boston, MA, USA, 1921; Volume 31. [Google Scholar]
  31. Randall, A. Risk and Precaution; Cambridge University Press: Cambridge, UK, 2011. [Google Scholar]
  32. Ciriacy-Wantrup, S. Resource Conservation: Economics and Policies; Division of Agricultural Science University of California: Berkeley, CA, USA, 1968. [Google Scholar]
  33. Bishop, R.C. Endangered Species and Uncertainty: The Economics of a Safe Minimum Standard. Am. J. Econ. 1978, 60, 10–18. [Google Scholar] [CrossRef]
  34. Bayes, T.L., II. An Essay towards Solving a Problem in the Doctrine of Chances. By the Late Rev. Mr. Bayes, FRS Communicated by Mr. Price, in a Letter to John Canton, A. M. F. R. S. Philos. Trans. R. Soc. Lond. 1763, 53, 370–418. [Google Scholar]
  35. Poincaré, H. The Value of Science: Essential Writings of Henri Poincaré; Modern Library: New York, NY, USA, 2012. [Google Scholar]
  36. Birkoff, G. On the Periodic Motions of Dynamics. Acta Math. 1927, 50, 359–379. [Google Scholar] [CrossRef]
  37. Holling, C.S. Surprise for Science, Resilience for Ecosystems, and Incentives for People. Ecol. Appl. 1996, 6, 733. [Google Scholar] [CrossRef]
  38. Willems, J.L. Stability Theory of Dynamical Systems; Wiley and Sons: New York, NY, USA, 1970. [Google Scholar]
  39. Scheffer, M.; Carpenter, S.R. Catastrophic Regime Shifts in Ecosystems: Linking Theory to Observation. Trends Ecol. Evol. 2003, 18, 648–656. [Google Scholar] [CrossRef]
  40. Walker, B.; Holling, C.S.; Carpenter, S.; Kinzig, A. Resilience, Adaptability and Transformability in Social-Ecological Systems. Ecol. Soc. 2004, 9, 5. [Google Scholar] [CrossRef]
  41. Folke, C.; Carpenter, S.; Walker, B.; Scheffer, M.; Elmqvist, T.; Gunderson, L.; Holling, C.S. Regime shifts, resilience, and biodiversity in ecosystem management. Annu. Rev. Ecol. Evol. Syst. 2004, 35, 557–581. [Google Scholar] [CrossRef] [Green Version]
  42. Scheffer, M.; Bascompte, J.; Brock, W.A.; Brovkin, V.; Carpenter, S.R.; Dakos, V.; Held, H.; van Nes, E.H.; Rietkerk, M.; Sugihara, G. Early-Warning Signals for Critical Transitions. Nature 2009, 461, 53–59. [Google Scholar] [CrossRef] [PubMed]
  43. Holling, C.S. The Resilience of Terrestrial Ecosystems: Local Surprise and Global Change. In Sustainable Development of the Biosphere; Clark, W.C., Munn, R.E., Eds.; Cambridge University Press: Cambridge, UK, 1986; pp. 292–317. [Google Scholar]
  44. Gunderson, L.; Holling, C.S. Panarchy: Understanding Transformations in Human and Natural Systems; Island Press: Washington, DC, USA, 2002. [Google Scholar]
  45. Garmestani, A.; Twidwell, D.; Angeler, D.G.; Sundstrom, S.; Barichievy, C.; Chaffin, B.C.; Eason, T.; Graham, N.; Granholm, D.; Gunderson, L.; et al. Panarchy: Opportunities and challenges for ecosystem management. Front. Ecol. Environ. 2020, 18, 576–583. [Google Scholar] [CrossRef]
  46. Shilpakar, R.L.; Thoms, M.C.; Reid, M.A. The resilience of a floodplain vegetation landscape. Landsc. Ecol. 2020, 36, 139–157. [Google Scholar] [CrossRef]
  47. Nacher, M.E.; Ferreira, C.S.S.; Jones, M.; Kalantari, Z. Application of the Adaptive Cycle and Panarchy in La Marjaleria Social-Ecological System: Reflections for Operability. Land 2021, 10, 980. [Google Scholar] [CrossRef]
  48. Angeler, D.G.; Allen, C.R.; Garmestani, A.; Gunderson, L.; Johnson, R.K. Panarchy and management of lake ecosystems. Ecol. Soc. 2021, 26, 7. [Google Scholar] [CrossRef]
  49. Chaffin, B.C. Panarchy and the Governance of Social-Ecological Systems. In Applied Panarchy: Applications and Diffusion Across Disciplines; Island Press: Washington, DC, USA, 2022; p. 253. [Google Scholar]
  50. Holling, C.S. Adaptive Environmental Assessment and Management; John Wiley & Sons: Hoboken, NJ, USA, 1978. [Google Scholar]
  51. Cañizares, J.; Copeland, S.; Doorn, N. Making Sense of Resilience. Sustainability 2021, 13, 8538. [Google Scholar] [CrossRef]
  52. Taleb, N.N. ‘Antifragility’ as a mathematical idea. Nature 2013, 494, 430. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  53. Randall, A. The Value of Biodiversity. Ambio 1991, 20, 64–68. [Google Scholar]
  54. Rockström, J.; Steffen, W.; Noone, K.; Persson, Å.; Chapin, F.S., III; Lambin, E.F.; Lenton, T.M.; Scheffer, M.; Folke, C.; Schellnhuber, H.J.; et al. A safe operating space for humanity. Nature 2009, 461, 472–475. [Google Scholar] [CrossRef]
  55. Steffen, W.; Richardson, K.; Rockström, J.; Cornell, S.E.; Fetzer, I.; Bennett, E.M.; Biggs, R.; Carpenter, S.R.; De Vries, W.; De Wit, C.A.; et al. Planetary boundaries: Guiding human development on a changing planet. Science 2015, 347, 1259855. [Google Scholar] [CrossRef] [Green Version]
  56. Barfuss, W.; Donges, J.; Lade, S.; Kurths, J. When optimization for governing human-environment tipping elements is neither sustainable nor safe. Nat. Commun. 2018, 9, 2354. [Google Scholar] [CrossRef]
  57. Clift, D.; Dalton, K.; Prior, J. Bladder Saltbush (Atriplex vesicaria Heward Ex Benth) Regeneration on the Riverine Plain of South-Eastern Australia since 1983. Rangel. J. 1989, 11, 31–39. [Google Scholar] [CrossRef]
  58. Smedema, L.K.; Shiati, K. Irrigation and Salinity: A Perspective Review of the Salinity Hazards of Irrigation Development in the Arid Zone. Irrig. Drain. Syst. 2002, 16, 161–174. [Google Scholar] [CrossRef]
  59. Feng, X.; Smith, M.S.; Zhang, L.; Lyu, N.; Lyu, Y.; Fu, B.; Xu, Z. Global Dryland Ecosystem Programme (Global-DEP): Australasian consultation report. J. Soils Sediments 2020, 20, 1807–1810. [Google Scholar] [CrossRef] [Green Version]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Randall, A. How Strong Sustainability Became Safety. Sustainability 2022, 14, 4578. https://doi.org/10.3390/su14084578

AMA Style

Randall A. How Strong Sustainability Became Safety. Sustainability. 2022; 14(8):4578. https://doi.org/10.3390/su14084578

Chicago/Turabian Style

Randall, Alan. 2022. "How Strong Sustainability Became Safety" Sustainability 14, no. 8: 4578. https://doi.org/10.3390/su14084578

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop