ETHICS IN SCIENCE AND ENVIRONMENTAL POLITICSESEP

2003:50–61 Published June 24

INTRODUCTION

The science of restoring damaged ecosystems hasbeen covered in many publications, but this articleprovides some illustrative examples of ethical issues inthe restoration process. Persuasive reasons exist forbelieving that precise replication of the predamagedcondition of an ecosystem is highly unlikely, thoughnot totally out of the question (Cairns 1989, Hobbs &Norton 1996). The most persuasive reason thatpredamaged conditions are not attainable is that thesequence of previous climatic and biological events isunlikely to be repeated. There is ambivalence between

saving nature for both present and future enlighteneduse by humankind (i.e. sustainable use of the planet)and saving natural systems because humankind has anethical responsibility for the fate of the 30+ million spe-cies with which it shares the planet. This ambivalenceis quite evident in the proposed principles for the newEarth Charter, which is being proposed as a guide forlocal, regional, national, and international efforts toprotect natural systems.1 Stone (1988) believes thatthis quest is for a single coherent and complete set of

© Inter-Research 2003 · www.int-res.com*Email: jcairns@vt.edu

PAPER

Ethical issues in ecological restoration

John Cairns, Jr.*

Department of Biology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061, USA

ABSTRACT: The acid test of humankind’s relationship to natural systems is the degree to which eco-logical damage caused by humans is repaired by humans. Technology and science are available, sothe remaining stimulus needed for implementation of ecological restoration is the ethical responsi-bility to do so. Ecological restoration can be regarded as enlightened self-interest for humankindsince it increases both natural capital and ecosystem services. However, well-designed ecologicalrestoration projects should have a major ethical component since the future of non-human life formson Earth requires more than self-interest. Although the field of science has provided various ratio-nales for ecological restoration, ethical issues associated with such activities must also be considered.If, as seems likely, human society and natural systems are co-evolving, restoration of damagedecosystems will improve both ecological and human health. The term ‘ecosocietal restoration’emphasizes this close relationship. However, if ecological restoration considers only human needsand does not emphasize ecological integrity, human-dominated ecosystems could become the norm.Such domination is already marked but the relationship could easily worsen. This article lists sevenmajor ethical issues in ecological restoration. This list is not encyclopedic but illustrative. Finally,there are five questions that human society must address that require robust scientific information tomake a sound ethical judgment.

KEY WORDS: Eco-ethics · Sustainability · Ecological restoration · Co-evolution · Species extinction

Resale or republication not permitted without written consent of the publisher

We must share in the action and passion of our time for fear of being judged not to have lived.Former US Supreme Court Chief Justice Oliver Wendell Holmes

The art of living consists not so much in complicating simple things as in simplifying those that are not.Frances Hertel

1Details are available in a special issue of Earth Ethics (Smith1996)

ESEP 2003:50–61

principles suitable for addressing all moral quandaries.If one views the planet’s ecological life support system(natural capital and ecosystem services) as essential tothe survival of humankind, there should be no ethicalambivalence.

Therefore, restoration ecologists have a number ofrestoration options: (1) assembling a naturalistic plantand animal community that closely resembles the struc-ture and function of the ecosystem in its predisturbedcondition, but without the identical species once present;it may be possible to use all of the predisturbance speciesthat are still available, together with other species, toproduce a naturalistic community more closely resem-bling the predisturbance community, (2) constructing anecosystem more able to withstand anthropogenic effects,since these are probably what damaged the ecosystemnow being repaired, and (3) allowing natural recoveryprocesses to determine the outcome, which most likelywill include exotic and pioneer species that may notreplace lost services and perhaps will have deleteriouseffects on adjacent ecosystems. As Berger (2003, pers.comm.) points out, ‘…achievement of that outcome maytake eons, or may produce a different stable state thanmanifested by, or tended toward by the ecosystemdestroyed’. All these options must be considered in thecontext of human society’s dependence on ecosystemservices (those ecological functions that are useful to hu-man society) for survival, as well as humankind’s ethicalresponsibility for the survival of non-human life forms.

The considerable attention given to sustainable useof the planet in recent years has increased the chrono-logical scope of restoration policy, and increasingglobal interdependence, both ecological and eco-nomic, has enormous policy implications for the geo-graphic dimensions of restoration policy. The NationalResearch Council (2003) report distinguishes betweenthe products of science (knowledge or data generatedby research) and the approach required to meet theneeds of decision makers. Although the geographicscale is not global, the processes used to supportrestoration, such as priority setting, identifying sciencegaps, and communicating research results, are quitesimilar. Since this is a large-scale undertaking, it de-serves serious attention. In addition, natural selectionjudges harshly the misallocation or unsustainable useof resources. Finally, humankind must remember itsethical responsibility for the damage done to naturalsystems and make reparations accordingly (e.g. Cairns2002).

GENESIS OF ETHICAL ISSUES

Humans cannot eradicate all species. Berger (2003,pers. comm.) cautions, ‘…if all domesticated species

were annihilated, the billions of people who dependon them for survival would likely starve…’. Suchlosses would fatally injure human society aspresently known (Cairns 1994, 1995a). If human soci-ety does not acknowledge its dependence on ecolog-ical life support systems, it will lose the species thatcannot resist or tolerate present practices and be leftwith those that can. Unwanted species (e.g. invasiveexotics) also pose a major threat that may be as seri-ous for natural systems as loss of species. Baskin(1996) provides an excellent discussion of the prob-lem of squelching the rising tide of unwanted spe-cies. Protection of indigenous species may well be asthreatened by invasive exotic species as by anthro-pogenic stress.

On the other hand, a compassionate relationshipwith natural systems would preserve and possiblyenhance ecosystem services, which surely are essen-tial for sustainable use of the planet. It is also a sine quanon that human society will not support large-scaleprotection of the environment or landscape-level eco-logical restoration unless the justification for doing sois both compelling and persuasive.

With some exceptions, the purely religious approachgenerally maintains that all species were created, ashumans were, and, therefore, deserve both compas-sion and esteem (Cairns 1995b). Obviously, humansociety has not acted as if this is a widespread, deeplyheld belief. If human society cannot survive withoutthe services provided by ecosystems, the need tochange the relationship with them is very compelling!

A beginning can be made by illustrating theservices currently recognized as provided by naturalsystems. A future possibility is that, as ecological lit-eracy improves, every function of natural systemswill be regarded as a service. Illustrative ecosystemservices include: (1) capture of solar energy andconversion into biomass that is used for food, build-ing materials, and fuel, (2) decomposition of wastessuch as sewage, (3) regeneration of nutrients informs essential to plant growth (e.g. nitrogenfixation), (4) storage, purification, and distribution ofwater (e.g. flood control, drinking water purification,transportation, etc.), (5) generation and maintenanceof soils, (6) control of pests by insectivorous birds,bats, insects, etc., (7) provision of a genetic library fordevelopment of new food and drugs through bothMendelian genetics and bioengineering, (8) mainte-nance of breathable air, (9) control of both microcli-mate and macroclimate, (10) provision of bufferingcapacity to adapt to changes and recover fromnatural stresses such as famine, fire, pestilence, (11)pollination of plants, including agricultural crops, byinsects, bats, etc., and (12) aesthetic enrichment fromvistas, recreation, and inspiration.

51

Cairns: Ethics in ecological restoration

GENESIS OF THE IDEA THAT HUMAN SOCIETYAND NATURAL SYSTEMS ARE CO-EVOLVING

Cairns (1996) asserts that human society has beendependent for most of its existence on an ecological lifesupport system that provides a variety of services.Cairns (1994, 1995a, c, 1996, 1997a) has furtherpostulated that human society and natural systems are co-evolving. Raven & Johnson (1986) have definedco-evolution as the simultaneous development ofadaptations in two or more populations, species, orother categories [emphasis mine] that interact soclosely that each is a strong selective force on theother.

The Agricultural and Industrial Revolutions haveprovided a variety of technological services that sup-plement the ecological life support system. In recentyears, the demand for natural resources, the encroach-ment of industrial and agricultural systems on naturalsystems, and the wastes of human society have endan-gered the ecological component of the life supportsystem. If sustainable use of the planet is the goal(meaning not depriving future generations of either ofthese life support systems), then the demand fortechnological services must be restrained so that eco-logical services are not threatened or damaged.

Schneider & Londer (1984) indicate that climate andlife are co-evolving. The World Commission on Envi-ronment and Development report (1987) and Cairns2

discuss sustainable use of the planet with the clearimplication that present usage will not result in sus-tainability. Vitousek et al. (1986) note that humansappropriate approximately half of the photosynthesison the planet; this percentage is surely a rather largeshare for one species out of many millions. The effectof this imbalance on the ecological life support systemis not clear, but losing so much photosynthetic energyto humans must be damaging. Until the effects areclear, this rate of use should not be exceeded. As Flan-nery (1994) notes in the dedication of his book to theAustralasians, ‘if they are to preserve their unique nat-ural heritage, their newly forged nations must cease tobe the realms of the future eaters.’ Until sustainabilityis achieved, all are ‘future eaters’ and, in order to ceasebeing so, must learn to cherish fellow species and nat-ural systems and to actively care for them.

If human society will acknowledge its dependenceon ecological services, then it is admitting that loss ofthese services can have an adverse effect uponhumankind (e.g. Cairns & Bidwell 1996a, b). Wilson

(1987) has noted that, if humans and other large verte-brates were to disappear, the ‘little things that run theworld’ (e.g. invertebrates) would not be seriouslyendangered and some, arguably, would thrive. How-ever, if the ‘little things that run the world’ disappearedentirely, humans and other large invertebrates wouldbe seriously threatened and probably driven to extinc-tion because of the absence of the services ‘the littlethings’ provide.

Unquestionably, humans can affect other speciessharing the planet by depriving them of habitat (e.g.water or other vital resources), exposing them to toxicsubstances, over-harvesting breeding stock, and thelike. Since each entity (human society and naturalsystems) can affect the other both beneficially andadversely, the relationship fits the definition of co-evolution. Therefore, it seems eminently reason-able to determine whether this relationship, as it nowexists, will lead to sustainability. If not, perhaps therelationship can be improved so that use of the planetover the next decade, millennium, or more would per-mit more humans and other species to live a quality lifethan would otherwise happen with the present rela-tionship. This concept leads to a number of issuesinvolving both science and ethics. One of the mostimportant issues concerns determining the conditionsto be met in order to achieve sustainability (Cairns1997b).

THE ROLE OF SCIENCE AND ETHICS IN SETTINGSOCIETAL GOALS

Both science and ethics are involved in making deci-sions on environmental assessment. Ideally, they arecalled into play sequentially (Suter 1993), i.e. a subjec-tive human perception of a problem leads to objectivescientific investigation of the causes and possible man-agement actions; then, alternate actions and theirprojected costs and benefits are evaluated for effec-tiveness and congruence with other societal goals. Asthis process of impact assessment and management isapplied to larger areas, longer time frames, and morecomplex and interconnected problems, distinguishingethical claims from scientific ones becomes increas-ingly difficult because the uncertainty inherent inscientific information increases with the scale of theenvironmental problem.

A decision that seems to be empirically based tosome scientists may seem to be based on ethics to oth-ers — because, while some professionals judge theuncertainty of the scientific data acceptable, othersjudge it excessive. Tolerance of scientific uncertaintyand tolerance of risk are both proper subjects fordebate before decisions are made. However, they are

52

2See Cairns J Jr (2002) Goals and conditions for a sustainableworld. ESEP Book 1. Inter-Research, Oldendorf/Luhe, Ger-many. Available at http://www.int-res.com/journals/esep/esepbooks/CairnsEsepBook.pdf

ESEP 2003:50–61

linked — acting with an intolerance of uncertaintyoften demands a high tolerance for risk. For example,there is uncertainty about the precise degree of tem-perature change due to anthropogenic greenhousegases, which is being used as a justification for inac-tion. However, this intolerance may result inhumankind being exposed to more severe conse-quences than it would be if precautionary measureshad been taken to reduce greenhouse gas production.

Science makes probabilistic statements about thenature of the world but does not offer a course of action.Science also helps to define problems and gather infor-mation about the extent and severity of environmentalchange and clarifies the links between environmentalchange and human self-interest. The basic ethical ques-tion here becomes: Does human self-interest differ fromthat of ecological integrity? An increased ability to mea-sure those ecological functions that, in the aggregate,constitute the ecological life support system upon whichhuman society depends would, arguably, provideincreasingly convincing evidence to support the processof restoring and conserving natural systems. While thisinformation is essential to effective management action,this scientific data cannot set goals.

Ethics, politics, and priorities are involved in settinggoals. Is there a consensus about what society shoulddo? Political action based on underlying ethical beliefsemerges as a result of consensus. In addition, there isnever enough money to do all the ecological preserva-tion and maintenance that would benefit human soci-ety over the long term. Ranking desirable goals andexpediting some, while delaying others, is thus a polit-ical process.

ETHICAL PROBLEMS IN ECOSOCIETALRESTORATION

Does restoration ecology represent a new trend inhuman society’s relationship with natural systems,enhancing a benign co-evolution? Or, are restorationecologists merely running a group of environmental‘body shops’ that repair damaged ecosystems withoutappreciable effect on either rates of ecological destruc-tion or human society’s guiding beliefs? Even if so, envi-ronmental consultants and their firms should carry outecological restoration at every opportunity because itadds to the body of knowledge on restoration method-ology and costs. At worst, ecological restoration could beused as another justification for continued damage tonatural systems. Furthermore, the global rate of ecolog-ical destruction is so enormous that the comparativelyfew attempts to repair ecological damage are dwarfed bycomparison. Indeed, there are similarly daunting ethicalproblems associated with ecological restoration.

1. Most ecological restoration is carried out to repairdamage caused by human mismanagement. If man-agement is the disease, how can it be the cure? Noss(1985) has said, ‘This is the irony of our age: “handson” management is needed to restore “hands off”wilderness character.’

2. Some mitigative restoration is carried out on rela-tively undamaged habitat of a different kind. Forexample, created wetlands may replace an upland for-est or an upland forest may be destroyed to attempt toreplicate the wetland that once occupied a particularlowland area. Logically, this secondarily damagedhabitat should be replaced by yet another mitigativeaction. Sacrificing a relatively undamaged habitat toprovide mitigative habitat of another kind may wellcause ecological harm. However, created wetlands, forexample, do have ecological value (Atkinson & Cairns2001)

3. The ecological life support system is viewed as acommodity. A homocentric viewpoint would justifyviewing the system as a commodity. An ecocentricview would emphasize the system’s intrinsic value andnatural rights. Sustainability ethics attempts to com-bine a homocentric and ecocentric viewpoint.

4. At the current state of knowledge, restoration pro-jects are likely to have unforeseen outcomes. Forexample, they may provide an opportunity for invasiveexotic species to become established in areas in whichthey had difficulty in doing so.

5. Well-meaning restoration efforts may displace thespecies best able to tolerate anthropogenic stress. Byattempting to return an ecosystem to its predistur-bance condition, the evolution of the species capableof co-existing with human society may be hampered.For example, some species might otherwise develop aresistance to anthropogenic stresses. Attempts tomanipulate the environment in such a way as to pro-mote the success of one or two species may impedeboth the natural successional process and alsoexclude other species that would otherwise be there.For example, restoring a stream to favor trout may notoptimize conditions for a wide variety of aquaticinvertebrates.

6. Similarly, if ecological restoration is carried out onan extremely large scale, human-dominated succes-sional processes could become the norm. For example,ecological reserves might be lost that preserve endan-gered and threatened species that may one day beextremely useful.

7. Finding sources of recolonizing species for dam-aged ecosystems is increasingly difficult. Should oneremove them from quality ecosystems and risk damag-ing that ecosystem or use pioneer species or, worse yet,exotics with the hope that the more desirable specieswill eventually colonize naturally?

53

Cairns: Ethics in ecological restoration

Ethical problem #1 — Human management: the cureor the disease?

Quammen (1996) discusses the biotic impoverish-ment caused by ecosystem fragmentation, and Dar-lington (1943) observes that restriction of area oftenlimits both the number and kind of animal species inisolated faunas. An ecological landscape, Quammen(1996) notes, is like a tapestry that, when fragmented,is not what it once was. Ecological fragmentation dueto barriers such as highways, parking lots, power lineright of ways, airports, and ubiquitous housing devel-opments is so commonplace in developed countriessuch as the US and much of Europe that it is acceptedas a norm.

At the State University of New York at Binghamton,a major campus road must be crossed by salamanderson the way to their spawning grounds. The roadsidecurbs are sufficiently high to represent an extremelydifficult exit barrier for the salamanders once they areon the roadway, while initially falling off a curb mightwell cause significant skin damage. Fortunately, envi-ronmentally sensitive persons installed ramps so thatthe salamanders have a more accommodating meansto cross the roadway. When I visited in 1994, the road-way itself had been closed during the annual spawningmigration, and presumably still is during the matingseason.

Forman et al. (2003) provide a superb analysis of theecological problems caused by roads. It illustrates theconflict inherent in simultaneously pursuing two goalsthat are incompatible: (1) optimizing the benefits of anexpanding road system, such as economic growth,more jobs, better access to schools, hospitals, and thelike, and the ability to live at some distance from one’splace of employment, and (2) the wish to preserve thenatural environment and reduce the threats to it aswell, such as improved air quality, reduced trafficcongestion, or preserved land from encroaching devel-opment. In short, too many human artifacts impair eco-logical integrity. Forman et al. (2003) provide a well-structured, useful synthesis for resolving one of thecurrent major problems, i.e. attempting to optimizetwo goals simultaneously.

Continents are, in a sense, large ecological islandsmade up of a mosaic of habitats with isolating barrierssuch as lakes, rivers, escarpments, and the like. How-ever, generally, ecological corridors permit speciesmovement or transport from one area to another. Giventhe degree to which ecological islands have been cre-ated on continents and the degree to which holes in theozone layer, acid rain, and transport of hazardouschemicals over considerable spatial scales haveoccurred, barriers exist to both invasion and successfulcolonization of damaged ecosystems. Ecological

restoration must necessarily be a crucial component ofenlightened management for sustainable use of theplanet.

The National Research Council (1992) notes thatrestored ecosystems are more likely to be self-regulat-ing at the landscape level. This statement is a reitera-tion of the well-established phenomena of islandbiogeography — namely, that large islands generallycontain more species than small islands. The seminalpublication authored by MacArthur & Wilson (1967)notes that all systems have a decolonization rate (i.e.species are lost from that area), and this phenomenonmust be offset by a countervailing colonization rate,which produces a dynamic equilibrium. Discussed atlength by MacArthur & Wilson (1967), this equilibriumcannot be maintained unless invading species haveaccess to all areas that are losing species. This processis more likely to occur at the landscape level thanbetween isolated patches that have significant anthro-pogenic barriers between them.

In short, given the effect of humans on ecosystems,enlightened management is the only viable solution toreach sustainability. For example, human societymight develop ecological management-derived solu-tions to avoid creating ecological islands by fragment-ing ecosystems through planning at a landscape orecoregional level.

Ethical problem #2 — Mitigative destruction ofecosystems

In Gunnison, Colorado, an airport taxiway affectedexisting wetlands, so mitigative (replacement) wet-lands were established west of the airport near thesewage treatment plant. These wetlands were adja-cent to already existing wetlands but, had they beenlocated on a grouse mating ground or some other cru-cial habitat, a reasonable person might question thejustification, even though the intent was the replace-ment of a particular lost habitat type. This issue willrequire robust professional judgment — prescriptivegovernment regulations should be avoided.

When mitigative restoration elsewhere replaces awetland that has been lost due to some developmentalactivity, the replacement may cause ecological dam-age to another habitat or ecosystem. If total ecologicaldestruction exceeds total ecological repair, then theecological damage has merely been shifted to anotherlocation. Arguably, the effort has not increased thequality or health of regional ecosystems collectively(bioregions), but merely replaced one habitat withanother (National Research Council 1992).

Decision makers need to get in a mental helicopterand rise above a highly localized situation to question

54

ESEP 2003:50–61

whether the ecological landscape has been improvedby the effort. Sometimes, mitigation is viewed aspurchasing or donating by setting aside a qualityecosystem in another location as a substitute fordamage that has occurred. However, nothing has beenadded to the total ecological capital of the region.Merely agreeing to protect an already existing ecosys-tem as compensation for proposed damage of anotherdoes not increase ecological capital. If the ecologicallydamaging development proceeds before mitigationoccurs, ecological capital will diminish during therestoration period, which may be many years (NationalResearch Council 1992).

Ethical problem #3 — The resource/commodity trap

In the US and much of the rest of the world, the term‘natural resources’ is commonly used when referring tonatural systems or ecosystems. Webster’s New Interna-tional Dictionary defines a resource as ‘a means of sup-plying a want; stock that can be drawn on; means ofsupport.’ In this view, the products of natural systemsare treated as commodities with a marketplace valuethat is relatively easily calculated. In some cases, theharvest from natural systems can be replaced withtechnological alternatives, i.e. natural wood can bereplaced with plastics (although petroleum does havebiological origins) and whale meat can be replaced bysoybean derivatives from agribusiness.

Another view is expressed by F. Henry Lickers,Director of the Department of the Environment for theMohawk Council of Akwesasne, Cornwall, Ontario(1990, pers. comm.): Earth is humankind’s mother andshould not be treated as a commodity. In this view,preventing ecological damage is a moral/ethical oblig-ation. If accidental damage occurs, healing it is acultural imperative (e.g. Leopold 1966) if one believesecosystems have intrinsic value beyond their useful-ness to humans. This is a subjective value as opposedto the ‘objective’ commodity context.

Clearly, some ecological restoration can be carriedout within a commodity context. For example, a clear-cut forest could be revegetated with the goal of yetanother harvest.

Although a clear-cut forest could be revegetated forcommercial ends, that would probably entail thereplanting of a single species, or at most a small num-ber of species, which is not ecological “restoration.” Inother words, trees could be recreated in this manner,but not a forest. Were an effort made to truly restore amulti-species, multi-age forest ecosystem, that wouldundoubtedly not be cost-effective for commercialforestry purposes. (Berger 2003, pers. comm.)

Although this type of restoration is perceived to havea fairly certain outcome, the time required is not easy

to predict accurately. In any case, human society’s rela-tionship with natural systems is dramatically differentfrom that espoused by Lickers. If natural systems areviewed as commodities, is restoration to predamageecological condition justified? Is restoration likely to besupported by a society viewing natural systems ascommodities?

In terms of sustainability, progress in thinking shouldmove from respect to esteem to acknowledgement ofdependence upon the planet’s ecological life supportsystem. Respect is optional. Esteem means highlyregarded and, thus, less likely to be viewed asoptional. Dependence means humankind cannot dowithout it. This progress in thinking would place soci-etal and individual behavioral norms in quite a differ-ent context than the resource/commodity point of view.However, policy must be based on how people behave(i.e. pragmatic and realistic). As pointed out by Berger(2003, pers. comm.), ‘Policy can also be normative andprescriptive as well; or it can be ameliorative and cor-rective in intent, i.e. neither may be “based on howpeople behave,” but on how we want them behaving’in particular contexts. This situation does not meanabandoning efforts to move toward an actively caringmodel for natural systems, but, rather, is an acknowl-edgement that, on the path to this model, human soci-ety must be able to recognize some short-termbenefits.

If the natural systems being restored are viewed ascommodities that can be harvested, exploited, oraltered in some major way, then restoration manage-ment practices must be altered accordingly (e.g. therepair of a clear-cut forest with one or a few species).This point of view will likely be incompatible withrestoration for the purpose of increasing habitat forrare and endangered species and the like. In addition,such systems are not likely to be self-regulating, thusincreasing management costs and efforts.

In the commodity context, the concept of ‘restorationof natural systems’ might not justify the name andmight be more analogous to agricultural systems orforestry systems with regular harvesting. Even thoughthe level of complexity might be greater followingrepair, the term ‘ecological restoration’ would proba-bly be inappropriate. Although this idea will beunpalatable to most ecologists, it may well be one thatis more and more accepted by society in general aspopulation pressures increase in the 21st century. Ifthis is to be avoided, the educational system must do abetter job in ensuring environmental literacy in alldisciplines!

Perhaps the commodity model could be replaced byone of compassion. If compassion and esteem are to beafforded to future generations and natural systems(including the other species with which humans share

55

Cairns: Ethics in ecological restoration

the planet), as well as compassion for presently livingindividuals, ecological restoration can be carried out ina sustainable use of the planet context. This wouldreduce or eliminate the resources available per capitaproblem hypothetically depicted in Fig. 1 if humanpopulation size and level of affluence are stabilized intime. The intersection point is now purely theoretical.Quality of human life, resilience of natural systems,and so on, need extensive discussion and evaluationbefore even a crude but reliable intersection point canbe determined. However, having compassion in onecategory modifies achievable compassion in the othercategories, making sustainability possible only whenall three are in balance. As Fig. 2 shows, focusing oncompassion in only one or two categories is notenough. Sustainability requires a balance of all threecategories. This also means making some difficult eth-ical choices and decisions that can be avoided if oneisolates each of these areas from the others. Surely, iso-lating interactive components is not an effective way ofeither carrying out ecological restoration or achievingsustainable use of the planet.

Ethical problem #4 — Uncertainty of outcome

A colleague once wore a tee shirt that was captioned‘Life is uncertain — eat dessert first!’ Clearly, uncer-tainty about the future of social security in the US, therate of global climate change, the number of humanssimultaneously living on the planet in the 21st century,the stock market, and most other areas of living isuncomfortably high. Human society’s lives are neitherrisk-free nor precisely predictable. Uncertainty accom-

panies almost every prediction (Lemons 1996), not onlyin ecological restoration activities but other areas ofscience as well. As Yogi Berra has reputedly stated,‘predictions are unreliable — especially about thefuture.’

Ecological restoration also will have unforeseen out-comes, even when carried out by the most skilled pro-fessionals. Important variables may be omitted;episodic events such as floods or droughts may occur atinconvenient times; exotic invaders may appearabruptly; and, almost certainly, the sequence of biolog-ical and meteorological events that resulted in thecharacteristics of the ecosystem before it was disturbedwill not be repeated. One may take comfort in the highprobability that the repaired ecosystem will almostalways be ecologically superior to the damagedecosystem. Furthermore, a healthy ecosystem is almostcertainly going to provide more services to humansociety than the damaged ecosystem.

Habitat restoration also should dampen the rate ofbiotic impoverishment (loss of species), which reducesuncertainty about the delivery of ecosystem services.However, present rates of species loss will almost cer-tainly have effects now difficult to predict. Diamond(1994) discusses species losses in an extremely broadpaleontological context. However, for this discussion,evidence on ‘recent losses’ is instructive.

Diamond (1994) estimates that 171 species and sub-species of birds are known to have become extinct

56

Fig. 1. Theoretical relationship between human populationnumbers and unimpacted natural systems. Figures given are

speculative

Fig. 2. Three pillars of sustainable use of the planet (S). Ecolog-ical restoration requires compassion for fellow species and willleave a more habitable planet for future generations. It will alsodemonstrate a compassion for living humans by improving thequality of the planet’s ecological life support system and bymaking damaged ecosystems more esthetically pleasing bymaking them more naturalistic. Both ecological restoration andsustainable use of the planet are acts of compassion if viewedfrom an ecocentric standpoint, or acts of enlightened self-

interest if viewed from a homocentric standpoint

ESEP 2003:50–61

since the year 1600. Of this total, 155 species and sub-species, or more than 90 %, lived and became extincton islands. Hawaii alone lost 24 species and sub-species. Roughly 20 % of the world’s species of birdsare confined to islands; therefore, nine-tenths of thehistorical bird extinctions have occurred in islandecosystems holding one-fifth of the total species. Dia-mond (1994) focused on terrestrial systems sur-rounded by water, although other kinds of ecologicalislands exist (e.g. a forest surrounded by grassland),and analyzed not only the effects of humans uponisland species but also the effects of ecosystem col-lapse upon human societies.

Mt. Kilimanjaro in Africa, a mountain ecosystemsurrounded by a totally different plains ecosystem, isalso an ecological island, as is Lake Malawi, which issurrounded by land. A patch of wetland surroundedby California freeways and freeway exits is just asmuch an ecological island as an island off the coast.Some species have access; others do not. Gettingthere can be hazardous and surviving there, if thearea is small, is less likely than it is on either a largeisland, such as Australia, or an even larger land masssuch as Eurasia.

Uncertainty about the effects of species loss will bereduced if recolonization occurs. Some articles in thepopular press suggest natural recovery of ecosystemsmay be noteworthy, but reaching the predisturbancecondition is unlikely. For example, McKibben (1995)discusses the reforestation of the eastern US. Someevidence indicates partial recovery of the foreststhemselves and, sometimes, much of the life theyonce supported. In such instances, however, Berger(2003, pers. comm.) points out that ‘…few trees arecomparable in height and girth to those of the aborig-inal forest, and certainly certain species, such as thechestnut, are now missing’. McKibben (1995) dis-cusses ‘quick devastation, quick recovery,’ but this isnot always true ecologically. For example, large oilspills may cause damage that cannot be quicklyovercome. If recolonization is achieved by species notpresent before ecological damage occurred, theuncertainty about the outcome may remain high.Berger (2003, pers. comm.) continues, ‘If the speciesare not the same, then assuredly we do not haveperfect structural (species) restoration, but we mayhave a restoration of certain if not many importantspecies and some ecosystem functions’.

Since 1961, I have worked intermittently at RockyMountain Biological Laboratory, situated at the miningtown of Gothic, Colorado, which was abandoned by allbut one inhabitant and his dog in the late 1800s whenthe early promise of mining proved illusory. Convertedinto a biological station in the 1930s, this area has sincebeen occupied during summer by varying numbers of

biologists, rarely exceeding 250. In recent years,between two and four people have been winteringthere. Thus, the Laboratory had a period of intense usewith accompanying dramatic changes in the ecosys-tem, followed by disuse, and, subsequently, over half acentury of occupation by persons likely to inhabit thearea with as little disruption to the native plants andanimals as possible. In spite of this nearly ideal oppor-tunity for recovery, the ecosystem does not resemblethe predisturbance ecosystem, nor is it likely to in theforeseeable future. Recovery is not automatic, but israther an accident of climate, soil, and economics(McKibben 1995). Assisted recovery (restoration) ‘…isalso dependent on the success of restoration planning,implementation/management, and the cooperation ofnatural forces’ (Berger 2003, pers. comm.).

In addition, many species of wildflowers in thesouthern Appalachians have not returned (McKibben1995), almost 100 years after the forests were last cut.While some species may return, they must come fromelsewhere or be brought there by human manage-ment. The fewer barriers to the recolonization ofspecies, the more likely natural recovery to somesemblance of predisturbance conditions will occur.Landscape fragmentation makes recolonization diffi-cult, even if the original species remain in the region.

Ethical problem #5 — Displacement of species bestable to tolerate anthropogenic stress

Well-meaning ecological restoration efforts mighteliminate those species that had initially colonized dis-turbed areas and were able to tolerate anthropogenicstress. The restoration efforts might replace tolerantspecies with species intolerant of the present practicesof human society; these intolerant species will subse-quently be eliminated unless human society alters itspresent behavior. Some species that are needed mostfor long-term sustainable use of the planet are almostcertainly included in the group of species with poorresistance to present practices (for example, speciesthat control ‘pests’ or pollinate plants). These speciesare in the most need of protection and are, collectively,the ones most needed to provide a vast array of ser-vices useful to human society. These species may notyet even be recognized as valuable to the interdepen-dent web of life because, in many cases, they have notreceived scientific names. If they are not even named,it is unlikely that substantive information is availableabout their ecological functions, some or possibly all ofwhich are of unperceived value.

Unquestionably, many of these intolerant speciescontrol population densities of those species that resisthuman control (i.e. pests). They may do so by preying

57

Cairns: Ethics in ecological restoration

upon them, successfully competing with them forspace or nutrients, or by favoring other competitivespecies. On the other hand, if human society’s prac-tices and behavioral norms are not changed, ecologicalrestoration carried out with species sensitive to anthro-pogenic effects will result only in a repetition of theecological disasters that necessitated the restoration inthe first place. For example, restoration with speciesintolerant of oil spills is useless if oil spills continue.

It is difficult to accept that damaged ecosystems maynot be restored to some naturalistic assemblage ofplants and animals resembling the predisturbancecondition. This realization is especially unpleasant if itmeans the loss of restored habitat essential to themaintenance of a bioregion or landscape. Probablymost important is the strong possibility that ecosystemstolerant of anthropogenic stress may not furnish eco-logical services that bear a close resemblance to thosefurnished by the indigenous ecosystem before it wasdamaged. Most ecologists would undoubtedly choosethe restoration model that closely resembles the pre-disturbance condition. However, if there is no assur-ance that the conditions that caused the damage areunlikely to be repeated, society will cease to supportecological restoration efforts; the present level of sup-port is hardly overwhelming.

It is distressing to think that species that are removedas landscapes are restored might actually turn out tobe quite desirable some day. However, the species thatmight be removed in favor of establishing predistur-bance species are not those that would be eliminatedfrom the larger landscape. For example, r-selectedspecies are highly tolerant of disturbance and, thus,are doing very well in frequently disturbed land-scapes. In the future, society might have uses for somespecies that are currently defined as weeds (e.g.tobacco has found practical value in biotechnology).Adaptive management must incorporate new scientificevidence, but it must be done in a systematic andorderly fashion.

An additional important ethical problem arises rela-tive to restoration and climate change. Climatechange, for example, will flood restored wetlands assea level rises and obliterate them. The ethical prob-lem becomes whether to restore an ecosystem, even ifit will only function in the short term, or whether to putthe money to other valid ecological purposes. Addi-tional related problems concern how much moneyshould be put into restoration, how to prioritizeresources for restoration, how much of a social/politicalfuror should be created if society fails to get itself on asustainability trajectory (one in which rates of resourcerestoration are greater than or equal to rates ofdestruction), and how to sanction those who wantonlydestroy nature and fail to restore it.

Ethical problem #6 — Should human-dominatedsuccessional processes become the norm?

Without question, if ecological restoration with spe-cies able to tolerate anthropogenic stress is chosen,whatever successional processes result will be human-dominated (i.e. exogenously managed). Even if thepractices that caused the ecological damage arereduced, the damaged ecosystem still may not berepaired to the predisturbance condition in both struc-ture and function.

Substantial portions of most countries’ landscapesare human-dominated (i.e. urbanized, developed, non-wild). As a consequence, these ecosystems will almostcertainly be inhabited by species that are resistant to,or tolerant of, human activities and which will invadethe ecologically damaged areas even if there are on-going restoration attempts. Additionally, exotic speciesoften thrive in stressed or altered ecosystems and arelikely to invade areas that are undergoing ecologicalrestoration.

Continual management is needed to keep exoticsunder even partial control during some ecologicalrestoration attempts. Nonsuch is a 14.5-acre island inBermuda. Even with its small size, intensive continualmanagement is necessary to keep the exotic speciesinhabiting most of the other portions of the Bermudansystem from overwhelming the indigenous species (e.g.Bermuda cedar, yellow-crowned night heron) beingreestablished on Nonsuch. The restoration effort is valu-able because it demonstrates the difficulties of restoringa portion of a landscape with severe anthropogenicstress (e.g. non-native plants established on the main-land, whose seeds are transported by birds that roost onNonsuch at night) that is dominated by exotics, althoughthis stress is not directly exerted on the island.

The lesson of this restoration effort is that human-dominated successional processes would clearly be thenorm if it were not for continual intervention byresearchers, land managers, government officials, andenvironmentalists on behalf of the reestablishedindigenous species. There are several justifications formaking this restoration effort: (1) Bermudians shouldalways be able to see what their country once lookedlike before intensive occupancy by humans; (2) indige-nous species characteristic of the islands should bemaintained so that, if recolonization efforts were to beundertaken elsewhere in the Bermudan system, colo-nizing species would be available; (3) it is always help-ful when examining ethical and scientific questions,such as the ones posed in this discussion, to have somehard evidence of the difficulties involved and thedegree of management necessary (e.g. during my visitdecades ago, two employees worked full time remov-ing exotics; Wingate 1988, pers. comm.).

58

ESEP 2003:50–61

Ethical problem #7 — Obtaining recolonizing speciesfor damaged ecosystems without damaging

quality ecosystems

In many parts of the world, ecological landscapes areheavily dominated by human society’s artifacts —highways, power lines, shopping malls, altered landcontours — and the number of relatively pristineecosystems is small and diminishing. Even these rela-tively untouched ecosystems are threatened by air-borne contaminants (such as acid rain), waterbornecontaminants, or exotics living in greatly alteredecosystems. Habitat fragmentation has reducedgenetic diversity for many species, and other problemsare associated with diminished habitat area. As a con-sequence, one is reluctant to remove indigenous spe-cies from quality habitats for the purpose of recoloniz-ing areas that have been damaged, even when theprobability of success is fairly high. Removing a num-ber of individuals of an already diminished populationwould have a variety of adverse effects, and movingtoo few might result in unsuccessful recolonization.

Obviously, considerable ethical judgment will benecessary in making these decisions about recoloniz-ing species, and, inevitably, the decisions will not bethe same from one site to another. These issues mustbe discussed and evaluated before the restoration isever started so that the risk to the species sources areexplicitly stated and the probability of success isrelated to the risks of doing further damage.

On the other hand, if recolonization is successful,another potential source of recolonizing species forother damaged ecosystems will have been established.However, microhabitat differences are often difficult todetect and may be responsible for the success or failurein a recolonization effort.

The National Research Council (2003) has provideda useful review of the Critical Ecosystem Studies Initia-tive (CESI), a project launched by the US Departmentof the Interior to provide scientific information toadvise restoration decision making and to guide itsown land management responsibilities for SouthFlorida ecosystem restoration. However, CESI shouldbe a useful information source about the complexitiesof large-scale restoration projects. CESI’s most impor-tant contribution to the ecological restoration process isthe graphic depiction of organization, cost sharing, andsums of money involved in major restoration projects.

FAKING NATURE

A small but insistent group has suggested that eco-logical restoration (as opposed to natural ecologicalrecovery) is ‘faking nature’ (e.g. Elliot 1997). Most spe-

cies modify their environment to some extent, but noneon the scale of humans, particularly over the last fewcenturies. In addition, ecosystems are dynamic, sochange is the norm rather than the exception. Speciesthat adapt to the changes survive; those that do notbecome extinct. Finally, ecosystem restoration is basedon natural processes subsidized in various ways byhumankind (for example, use of hatcheries to reestab-lish a fishery).

The ultimate goal of ecological restoration is to pro-duce a naturalistic, self-regulating community ofplants and animals. Ideally, human assistance shouldonly be required for the initial stages; that is, until theecosystem becomes self-regulating. If the ultimategoal is to have natural processes take over, how canthe assistance be considered ‘faking nature’? Animportant criterion for restoration success is the degreeto which the restored system accumulates natural cap-ital and provides ecosystem services of benefit to amajority of species, not just humans.

Many ecosystems being restored will require subsi-dies for a considerable time period, but ultimately, inevolutionary time, all will become self-regulating,although they may not always be perceived as benefi-cial to humankind. Stated another way, if humansbecame extinct, all ecosystems would become self-regulating, just as they have become self-regulatingfollowing a number of mass extinctions that occurredbefore humans appeared on the planet. However,future self-regulation measures may not favor humansas much as present ecosystems have in the lastcenturies. The fact that humans are responsible for thepresent biotic impoverishment (species extinctions)does not diminish, in the long term, the prospects forultimate recovery when the stressor (humankind)disappears.

CONCLUSIONS

This discussion has offered a few illustrative exam-ples of the interface between science and ethics withregard to ecological restoration. Some of the questionsthat human society must address follow.

1. Is a world consisting entirely of human-dominatedecosystems desirable?

2. If not, under what circumstances should restora-tion to predisturbance be chosen or rejected?

3. If ecological restoration is carried out with speciestolerant of anthropogenic activities, ecosystems willnot resemble the predamaged conditions, and theirservices may not correspond with those of the ecosys-tem in the predisturbance condition. What informationis needed to make an informed judgment in this situa-tion?

59

Cairns: Ethics in ecological restoration

4. Is it possible to have sustainable use of the planetfor 1000 or more years and under conditions not appre-ciably worse than those at present if ecological repairdoes not equal the rate of ecological destruction?

5. Does sustainable use of the planet mean ecosys-tems that fill minimal expectations of services andother amenities or, in addition to quantity, is quality anexpected attribute of sustainability?

Other questions raised by Berger (2003, pers. comm.)that are beyond the scope of this article include: Howcan fraudulent restoration efforts, which are completedmerely to give cover to developers, be detected andprevented? Do environmental consulting firms thatoffer and promise mitigative restoration often servemerely to facilitate development? Who should pay forrestoration? Does society need to adopt a variant of ‘thepolluter pays’ principle to this problem? Other impor-tant ethical problems regarding restoration includedetermining when to discontinue a restoration effort —when is restoration done and regarded as a success —and when monitoring should cease (Holl & Cairns2002).

These serious ethical questions are but a few of themany that cannot be answered by science but willrequire robust scientific information in order to make asatisfactory judgment. Ecological and environmentalliteracy for the general population and its represen-tatives must be greatly improved to deal with thesecomplex multivariate issues. Furthermore, both thetemporal and spatial scales are much greater thanthose with which most political systems are accus-tomed to coping. The complexity level of the problemsrequires multidimensional approaches. While thechallenge is great, the opportunity for systems levelscience and thinking has never been greater, nor hashuman society’s stake on the outcome.

Acknowledgements. I am indebted to K. Cairns for typing thefirst draft of this manuscript and to D. Donald for her usualskillful editorial assistance in preparing this manuscript forpublication. H. Cairns Chambers prepared the two figures. J.Berger provided useful comments on the manuscript.

LITERATURE CITED

Atkinson RB, Cairns J Jr (2001) Plant decomposition and litteraccumulation in depressional wetlands: functional perfor-mance of two wetland age classes that were created viaexcavation. Wetlands 21(3):354–362

Baskin Y (1996) Curbing undesirable invaders. BioScience46(10):732–736

Cairns J Jr (1989) Restoring damaged ecosystems: Is predistur-bance condition a viable option? Environ Prof 11:152–159

Cairns J Jr (1994) Ecological restoration: re-examining humansociety’s relationship with natural systems. The AbelWolman Distinguished Lecture. US National Academy ofSciences, Washington, DC

Cairns J Jr (1995a) Eco-societal restoration: re-examininghuman society’s relationship with natural systems. Annalsof Earth 13(1):18–21

Cairns J Jr (1995b) Respect for the inherent worth of eachindividual versus the web of life: an unresolved conflict.Spec Sci Tech 18:294–301

Cairns J Jr (1995c) Re-examining human society’s relation-ship with natural systems: maintaining ecosystem servicesand sustainable use. Corp Environ Strat 3(1):69–74

Cairns J Jr (1996) Determining the balance between techno-logical and ecosystem services. In: Schulze PC (ed) Engi-neering within ecological constraints. National AcademyPress, Washington, DC, p 13–30

Cairns J Jr (1997a) Global co-evolution of natural systems andhuman society. Rev Soc Mex Hist Nat 47:217–228

Cairns J Jr (1997b) Defining goals and conditions for asustainable world. Environ Health Perspect 105(11):1164–1170

Cairns J Jr (2002) Rationale for restoration. In: Davy AJ,Perrow MR (ed) Handbook of ecological restoration,Vol. 1: principles and restoration. Cambridge UniversityPress, Cambridge, p 10–23

Cairns J Jr, Bidwell JR (1996a) Discontinuities in technologi-cal and natural systems caused by exotic species. Biodi-versity Conserv 5:1085–1094

Cairns J Jr, Bidwell JR (1996b) The modification of humansociety by natural systems: discontinuities caused by theexploitation of endemic species and the introduction ofexotics. Environ Health Perspect 104(11):2–5

Darlington PJ Jr (1943) Carabidae of mountains and islands:data on the evolution of isolated faunas, and on atrophy ofwings. Ecol Monogr 13:1

Diamond JM (1994) Ecological collapses of ancient civiliza-tions: the golden age that never was. Bull Am Acad ArtsSci XVLII(5):37–59

Elliot R (1997) Faking nature: the ethics of environmentalrestoration. Routledge, London

Flannery TF (1994) The future eaters: an ecological history ofthe Australasian lands and people. George Braziller, NewYork

Forman RTT, Sperling D, Bissonette JA, Clevenger AP, Cut-shall CD, Dale VH, Fahrig H, France R, Goldman CR,Heanue K, Jones JA, Swanson FV, Turrentine T, WinterTC (2003) Road ecology: science and solutions. IslandPress, Washington DC

Hobbs RA, Norton DA (1996) Commentary: towards a con-ceptual framework for restoration ecology. RestorationEcol 4(2):93–110

Holl KD, Cairns J Jr (2002) Monitoring and appraisal. In: DavyAJ, Perrow, MR (eds) Handbook of ecological restoration,Vol. 1: principles and restoration. Cambridge UniversityPress, Cambridge, p 411–432

Lemons J (ed) (1996) Scientific uncertainty. Blackwell Sci-ence, Cambridge, MA

Leopold A (1966) A Sand County almanac, with essays fromRound Rivers. Ballantine Books, New York

MacArthur RH, Wilson EO (1967) The theory of island bio-geography. Princeton University Press, Princeton, NJ

McKibben B (1995) An explosion of green. Atl Mon 275(4):61–75

National Research Council (1992) Restoration of aquaticecosystems: science, technology, and public policy.National Academy Press, Washington, DC

National Research Council (2003) Science and the greatereverglades ecosystem restoration. National AcademyPress, Washington, DC

Noss RF (1985) Wilderness recovery and ecological restora-

60

ESEP 2003:50–61

tion: an example for Florida. Earth First 5(8):18–19Quammen D (1996) The song of the dodo: island biography in

an age of extinction. Scribner, New YorkRaven PH, Johnson GB (1986) Biology. Times Mirror/Mosby

College Publishing, St. Louis, MOSchneider SH, Londer R (1984) The co-evolution of climate

and life. Sierra Club Books, San Francisco, CASmith JA III (1996) Earth charter. Earth Ethics 7:3–4Stone C (1988) Earth and other ethics. Harper and Row Pub-

lishers, New York

Suter GW II (ed) (1993) Ecological risk assessment. LewisPublishers, Boca Raton, FL

Vitousek PM, Ehrlich PR, Ehrlich AH, Matson PA (1986)Human appropriation of the products of photosynthesis.BioScience 36:368–373

Wilson EO (1987) The little things that run the world (theimportance and conservation of invertebrates). ConservBiol 1(4):344–346

World Commission on Environment and Development (1987)Our common future. Oxford University Press, Oxford

61

Editorial responsibility: Mary Batson (Managing Editor),Oldendorf/Luhe, Germany

Submitted: February 21, 2003; Accepted: June 18, 2003Proofs received from author(s): June 23, 2003Published on the web: June 24, 2003