Can pollution problems be effectively solved by ... 5.pdf · Can pollution problems be effectively solved by environmental science and technology? An analysis of ... man ‘to establish

Ecological Economics 37 (2001) 271–287

ANALYSIS

Can pollution problems be effectively solved byenvironmental science and technology? An analysis of

critical limitations�

Michael H. Huesemann *Marine Sciences Laboratory, Pacific Northwest National Laboratory, 1529 West Sequim Bay Road, Sequim, WA 98382, USA

Received 7 December 1999; received in revised form 18 September 2000; accepted 27 November 2000

Abstract

It is currently believed that science and technology can provide effective solutions to most, if not all, environmentalproblems facing western industrial societies. The validity of this optimistic assumption is highly questionable for atleast three reasons: First, current mechanistic, reductionist science is inherently incapable of providing the completeand accurate information which is required to successfully address environmental problems. Second, both theconservation of mass principle and the second law of thermodynamics dictate that most remediation technologies —while successful in solving specific pollution problems — cause unavoidable negative environmental impactselsewhere or in the future. Third, it is intrinsically impossible to design industrial processes that have no negativeenvironmental impacts. This follows not only from the entropy law but also from the fact that any generation ofenergy is impossible without negative environmental consequences. It can therefore be concluded that science andtechnology have only very limited potential in solving current and future environmental problems. Consequently, itwill be necessary to address the root cause of environmental deterioration, namely, the prevailing materialistic valuesthat are the main driving force for both overpopulation and overconsumption. The long-term protection of theenvironment is, therefore, not primarily a technical problem but rather a social and moral problem that can only besolved by drastically reducing the strong influence of materialistic values. © 2001 Elsevier Science B.V. All rightsreserved.

Keywords: Environmental science and technology; Pollutant remediation; Entropy; Mechanistic reductionism; Industrial ecology;Paradigm shift; Post-normal science

www.elsevier.com/locate/ecolecon

1. Introduction

‘The deterioration of the environment producedby technology is a technological problem forwhich technology has found, is finding, and willcontinue to find solutions’ (Bereano, 1976, p. 10).

� Disclaimer: The views expressed in this article are solelythose of the author and do not necessarily reflect the officialposition of Battelle Memorial Institute, Pacific NorthwestNational Laboratory, or the U.S. Department of Energy.

* Tel.: +1-360-6813618.E-mail address: [email protected] (M.H. Huese-

mann).

0921-8009/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.

PII: S 0921 -8009 (00 )00283 -4

M.H. Huesemann / Ecological Economics 37 (2001) 271–287272

Although this statement was made more than 30years ago, little has changed since then with re-spect to our belief that technology is both theprimary cause of and solution for environmentalproblems. In fact, during the last three decades,tremendous efforts in science and technologyhave been made to better understand the natureof environmental deterioration and to developtechnologies to remediate the effects of past pol-lution. It is safe to say that hundreds of millionsof dollars have been devoted to improve thequality of the environment and that much morewill be spent in the future.

Despite these intense efforts expended in ‘sav-ing the environment’, it is questionable whethercurrent scientific and technological approachescan be sufficiently effective in solving our nu-merous environmental crises. This is because thecurrent practice of environmental science andtechnology is based on the following three as-sumptions, the validity of which are highlyproblematic:1. Science can provide us with enough detailed

knowledge about nature to solve and preventenvironmental problems.

2. Remediation technologies can successfullyremove pollution without causing otherunforeseen negative environmental impactselsewhere.

3. It is possible to prevent pollution in the fu-ture and develop ‘clean’ industrial processesthat have no environmental impacts.

It is surprising that most environmental scien-tists and engineers have never questioned thevalidity of these three assumptions, particularlyin view of the fact that the success of theirwork, namely, the protection of the environ-ment, critically depends on the accuracy of thesepremises. In addition, no rigorous examinationof the soundness of these hypotheses can befound in the literature. It is therefore the objec-tive of this paper to critically examine thesethree assumptions and to provide a better under-standing of the intrinsic limitations of technolog-ical approaches to solving environmentalproblems.

2. Mechanistic reductionist science is inherentlylimited in providing complete and accurateinformation about environmental problems

For more than 300 years, the western culturesof Europe and North America have placedheavy emphasis on the experimental sciences forgaining a deeper understanding of nature andthe universe. It is important to recognize thatthe philosophers of the Enlightenment did notpromote the sciences because it would expandknowledge for its own sake, but because it hadthe potential for improving a culturally definedquality of life. According to Eugene Schwartz,‘the principal aim of knowledge for Francis Ba-con, as for most of the other precursors of theschool of scientific progress, was the utilitariangoal of improving man’s condition by lesseninghis suffering and increasing his happiness. Cast-ing aside the viewpoint of the Greek philoso-phers that knowledge affords its own intellectualsatisfaction, Bacon defined the true goal of thesciences to be ‘the endowment of human lifewith new inventions and riches’. He called onman ‘to establish and extend the power and do-minion of the human race itself over the uni-verse’ (Schwartz, 1971, p. 23–24).

It is clear, then, that modern science was fromthe very beginning never value-free and objective(Huesemann, 2000a). By contrast, the pursuit ofscience was and still is strongly motivated bysocietal needs to improve living conditions and,according to the utilitarian dogma, to maximizethe happiness of the greatest number of people(DesJardins, 1993, p. 29). It is not surprising thatscientific knowledge was almost immediately con-verted, with the help of engineers, into usefultechnologies which promised to fulfill the wishesof society. The strong utilitarian emphasis ex-plains the heavy focus on mechanistic, reduction-ist research whose logic is based on two tacitassumptions, namely, that nature functions like amachine and that nature’s complexity can be un-derstood by investigating the properties of itsisolated parts. According to Longino, ‘mechanis-tic research is favored because it enables the ma-nipulation of nature’ (Longino, 1990, p. 94).Similarly, Sarewitz concludes that ‘reductionst sci-

M.H. Huesemann / Ecological Economics 37 (2001) 271–287 273

Fig. 1. Mechanistic, reductionist research focuses exclusively on cause and effect relationship A–B while ignoring all other effectsand interactions.

entific research helped to forge the path to thepost-industrial world by elucidating specific natu-ral phenomena that could be isolated, manipu-lated, and applied to practical tasks’ (Sarewitz,1996, p. 105).

Fig. 1 shows conceptually how mechanistic,reductionist research is used to focus on isolatedcause and effect relationships in order to exploitnature for some perceived benefit. In most cases,after the cause and effect relationships have beenelucidated by scientific methods, efforts are takento enhance desirable effects by manipulatingcausative factors. However, since the naturalworld more nearly resembles a large networkof intricately connected cause and effectrelationships1, enhancement of one isolated effectis certain to have unexpected consequences else-where in the system or at a later time. Theseundesirable effects often manifest themselves asenvironmental pollution or various eco-systemdisturbances. Thus, it can be seen that western

sciences have promoted the exploitation of natureand thereby have become a significant causativefactor in the current environmental crisis.

Unfortunately, many environmental scientistsare not aware of the inherent limitations of theirresearch methodologies in solving environmentalproblems. This was clearly recognized by Sarewitzwho wrote: ‘Modern reductionist science wasborn explicitly of the moral obligation to knowthe polyphony of nature — and to conquer it.But even scientists who are passionately commit-ted to understanding and reversing the ecologiccrisis may strongly resist the notion that the char-acter of science itself is in any way connected tothe crisis’ (Sarewitz, 1996, p. 112). Commoner isalso convinced that ‘the reductionist methodol-ogy, which is so characteristic of much modernresearch, is not an effective means of analyzingthe vast natural systems that are threatened bydegradation’ (Commoner, 1971, p. 189). Similarly,in their research on post-normal science, Funtow-icz and Ravetz point out that ‘… the methodologyfor coping successfully with these novel problemscannot be the same as the one that helped createthem. Much of the success of traditional sciencelay in its power to abstract from uncertainty inknowledge and values… Now scientific expertise

1 Remember also Commoner’s First Law of Ecology: ‘Ev-erything Is Connected to Everything Else’ (Commoner, 1971,pp. 33–39).


has led us into policy dilemmas which it is inca-pable of resolving by itself. We have not merelylost control and even predictability; now we faceradical uncertainty and even ignorance, as well asethical uncertainties lying at the heart of scientificpolicy issues’ (Funtowicz and Ravetz, 1993, p.741–742).

It is therefore questionable whether current sci-entific methods that are strongly biased towardsthe exploitation of nature can be used to providesufficiently complete and accurate information toprotect against the effects of this exploitation. Inorder to remediate the effects of past pollution orto protect against future environmental damage, itis necessary to understand all possible cause andeffect relationships that might occur in nature as aresult of human manipulations. Even if only onecause and effect relationship is overlooked duringa scientific investigation, it may well prove to bean extremely critical one that could cause seriousharm if ignored. A good example is the massiveuse of fossil fuels for energy and transportationneeds. For many years, most researchers weresingle-mindedly focussed on reducing the impactsof various air-pollutants such as SO2 or NOx.Meanwhile, the potential effects of CO2 werecompletely ignored until recently when it wasfound that increasing atmospheric CO2 concentra-tions are a causative factor in global warming(Houghton, 1997). What if researchers were tohave gained this insight 50 years from now whenany remedial action would likely to be too late tobe effective (assuming that it is not too late at thispoint in time). For that matter, what other criticalcause and effect relationships are currently beingoverlooked?

Present scientific methodologies provide onlyvery limited knowledge about the workings ofnature as a whole, and environmental scienceelucidates only certain isolated aspects that werespecifically chosen for investigation. It shouldtherefore be clear that mechanistic, reductionistenvironmental research, by its very nature, cannever elucidate all the interconnected relationshipsthat exist in the natural world. For instance, itwill never be possible to determine all the adverseeffects synthetic chemicals may have in the envi-ronment. Scientific research has up to now fo-

cussed only on a few selected chemicals and theirdetectable effects on a very small subset of targetspecies or organs. In theory, if one would like toknow the total and complete impact of syntheticchemicals on the environment, it would be neces-sary to determine all possible adverse effects of allsynthetic chemicals, pure and in mixture, on allspecies. In practice, this is impossible because: (a)the detection of adverse effects is limited by thestate-of-the-art of current scientific knowledgeand methodologies (e.g. the endocrine disruptionpotential of chlorinated synthetic chemicals couldnot have been predicted 50 years ago due to lackof biological assays for the measurement of en-docrine disruption); (b) the testing of millions ofdifferent chemicals and chemical mixtures is cost-prohibitive; (c) there are millions of species, manyof which have not been identified; and (d) thepotential effects on future generations are impos-sible to measure or predict.

Considering the limited budgets for environ-mental research, it is obvious that only a fewselected cause and effect relationships can ever bestudied. But the elucidation of only a few selectedissues considered important to humans clearlydoes not provide sufficient knowledge to protectall of nature. Consequently, the focus of currentenvironmental science is too narrowly an-thropocentric to truly ensure the long-term pro-tection of human health and welfare. As a result,conservation biologist Ehrenfeld, who sees con-temporary science and technology as an expres-sion of an overly anthropocentric humanisticculture, is convinced that ‘there is no true protec-tion of Nature within the humanist system — thevery idea is a contradiction of terms’ (Ehrenfeld,1981, p. 202). This concern was also expressed bypolicy analyst Sarewitz who stated:

‘Because mainstream science is reductionistwhile the environment is an interconnected,complex system, much of modern science andtechnology may be intrinsically unsuited to suc-cessful confrontation of ecological crises … Ul-timately, therefore, that reductionist scienceand its technological consequences can save usfrom ecological crises should be viewed withskepticism’ (Sarewitz, 1996, pp. 107–113).


3. The conservation of mass principle inherentlylimits the effectiveness of physical remediationtechnologies

The conservation of mass principle is conciselystated in Barry Commoner’s Second Law of Ecol-ogy: ‘Everything Must Go Somewhere’ (Com-moner, 1971, p. 39). Thus, a pollutant, if nottransformed naturally by either chemical or biolog-ical reactions, will remain indefinitely in the envi-ronment and its total mass will not change withtime. Many physical treatment technologies at-tempt to reduce the risk posed by a pollutant usingvarious strategies such as: (a) limiting the dispersalof the contaminant (e.g. landfilling); (b) reducingtoxic effects by dilution (e.g. smokestacks); or (c)transferring contaminants from one medium toanother (e.g. air-stripping of contaminated water).

Since the total mass of pollutant is not reducedby physical treatment technologies, their respectiveeffectiveness in solving environmental problems ishighly questionable. It is clear that physical treat-ment technologies transfer risks either from oneplace to another or from the present to the future.For example, landfilling limits the dispersal ofcontaminants only at the present time. Since it isunlikely that landfill integrity can be maintainedindefinitely, contaminants can be expected toreenter the environment thereby causing negativeeffects for future generations.

The rationale of diluting pollutants or transfer-ring contaminants from one medium to anotherhas been justified by assuming that these treatmentstrategies make the pollutants less harmful. As waspointed out above, this assumption is very weakconsidering that mechanistic, reductionist scienceis inherently unable to elucidate all negative envi-ronmental consequences. For example, dilutionwill certainly reduce the risk of localized obviousacute toxicity, but is likely to increase the probabil-ity of more widespread subtle chronic effects thatare currently unknown or difficult to monitor.Similarly, it is uncertain whether transfer of apollutant from one medium to another will actu-ally reduce the overall risks. More likely, this typeof treatment strategy is chosen because the hazardsof a pollutant in one medium are well known (i.e.chlorinated solvents in groundwater used for

drinking) while very little data are available aboutthe same compound in the medium to which it istransferred (i.e. chlorinated solvents dispersed inair).

In summary, physical treatment technologies areunable to pro�ide effecti�e solutions to en�ironmen-tal pollution since they transfer risks either from oneplace to another or from the present to the future.Physical treatment technologies are often consid-ered effective only because spatially and tempo-rally transferred risks are not known and becausethere is little general concern about what happenselsewhere or at a later time.

4. According to the second law of thermodynamics,all remediation technologies have unavoidablenegative environmental consequences

The second law of thermodynamics states thatthe total entropy (S) within a closed system mustalways increase with time (Prigogine, 1961;Balzhiser et al., 1972; Rifkin, 1980; Atkins, 1984;Prigogine, 1989; Faber et al., 1995; Rebane, 1995;Ayres, 1998). Since entropy can be related to thechaos or disorder of a system via the Boltzmannequation (Atkins, 1984), this is equivalent to stat-ing that disorder must increase in closed systems.The second law of thermodynamics can be used toshow why many remediation activities generateunavoidable negative environmental consequences.

Consider, for example, the common problem ofcleaning up highly dispersed organic contaminantssuch as benzene or chlorinated solvents in ground-water (Bedient et al., 1994). As shown in Fig. 2A,the total entropy S0

t of the closed system prior tothe pollution event (t=0) is the sum of the entropyof the subsystem S0

s (consisting of the drum withthe contaminant, the soil layer and the cleanaquifer) and the entropy of the surroundingenvironment2 S0

e. As the chemical leaks from the

2 For simplicity, it is assumed that the surrounding environ-ment is a closed system. In reality, however, planet earth is anopen system that receives solar energy from the outside uni-verse. Nevertheless, the overall conclusions reached below (Eq.(3)) hold for both closed and open systems — see also Eq. (7).


drum and disperses through the soil layer and isdissolved in the groundwater (Fig. 2B), the en-tropy (disorder) of the subsystem increases to S1

s.Since the environment (outside the box) is notaffected, the entropy S1

e remains unchanged fromS0

e. However, as predicted by the second law ofthermodynamics, the total entropy (S1

t =S1s +S1

e)of the closed system increases as a result of thepollution event.

In a successful remediation effort such as

‘pump and treat’ (Fig. 2C), the dispersed pollu-tant is removed from the soil and aquifer, concen-trated and placed in the second drum for eithersubsequent disposal or treatment. From a thermo-dynamic perspective, the cleanup effort consists ofconverting a highly entropic, disordered environ-ment back to its original, low entropy condition(as in Fig. 2A). In short, during remediation, theentropy of the contaminated subsystem is reducedfrom S1

s to S2s. The question now arises, how can

entropy in the subsystem be reduced without vio-lating the second law of thermodymamics?

According to the second law, the entropy of thetotal system must increase from S1

t to S2t , i.e.

S2t �S1

t , (1)

Since St=Se+Ss, it follows:

S2e +S2

s �S1e +S1

s. (2)

Rearranging Eq. (2) and defining the change ofentropy in the environment and the subsystem as�S e=S2

e −S1e and �S s=S2

s −S1s, respectively, it

follows:

�S e� −�S s. (3)

Thus, remediation acti�ities can create order inthe subsystem (−�Ss) only at the expense ofgenerating more disorder in the rest of the en�iron-ment (�S e).

The main reason why disorder (entropy) is cre-ated in the environment is due to the fact thatenergy is needed to generate order inside thesubsystem. According to established thermody-namic principles, energy generation is only possi-ble by converting low entropy materials (coal,petroleum, natural gas, uranium) into high en-tropy wastes (carbon dioxide, sulfur dioxide, ni-trous oxide, radioactive materials, waste heat,etc.). Since energy production takes place outsidethe subsystem, the net entropy increase occurs inthe rest of the environment where it causes vari-ous undesirable effects.

The obvious next question is: How is entropyrelated to environmental pollution? Since entropyis a measure of chaos or disorder, it is not surpris-ing that a number of investigators have proposedentropy increase in the environment as an alterna-tive measure for pollution. For example, Kummel

Fig. 2. Changes in entropy of the subsystem (S s) and theenvironment (Se) from initial clean state (A) to polluted (B)and final remediated (C) state. As elaborated in more detail inthe text, the following relations hold: S2

t �S1t �S0

t ; S1s �S0

s;S2

s �S1s; S2

e �S1e =S0

e.


(1989) has used entropy as a pollution indicator inmacroeconomic modeling and Faber et al. (1995)have suggested the formulation of a pollutionfunction in which entropy increase due to disper-sal and reaction of specific contaminants is usedas an aggregated measure of pollution. The ratio-nale for using entropy as a substitute indicator forenvironmental disturbance has also been recom-mended by Ayres who notes: ‘As waste materialsapproach local equilibrium with the environment,the potential for future entropy generation is ameasure of their potential for driving uncon-trolled chemical or physical processes in environ-mental systems. Eco-toxicity is nothing more norless than environmental disturbance. Thus, poten-tial entropy can be regarded as a measure of the apriori probability of eco-toxicity’ (Ayres and Mar-tinas, 1995, p. 116).

A chemical that is released accidentally canincrease entropy in the environment by two pri-mary mechanisms, namely, physical dispersionand unwanted chemical reaction. It is interestingto note how these two entropy-increasing mecha-nisms have a parallel with the common riskparadigm (Klaassen et al., 1986) which states thatchemicals only pose a health or environmentalrisk if there is an exposure pathway (i.e. chemicalsneed to disperse to reach a sensitive receptor) andif the chemical is hazardous to the receptor (i.e.drives unwanted chemical reactions that are dele-terious to the organism). Based on this assess-ment, it appears appropriate to use the potentialfor entropy increase as a substitute measure forenvironmental pollution. In fact, the entropy-pol-lution paradigm can be used to understand reme-diation efforts from a new perspective.Remediation technologies generally attempt to re-verse or limit entropy increase in the affectedenvironment either by controlling or reversingpollutant dispersion or by having the contami-nants react to reach chemical equilibrium (thestate where entropy is maximized) before theyhave a chance to disperse in the environment andreach sensitive receptors. Examples of both reme-diation strategies will now be discussed.

One of the most common environmental reme-diation approaches for controlling and reversingdispersion of dissolved groundwater contaminants

is the ‘pump and treat’ technology (Bedient et al.,1994). Considering that there are more than250 000 leaking underground storage tanks in theU.S. (Bedient et al., 1994) and that — as a result— many aquifers are polluted with eitherpetroleum hydrocarbons or chlorinated solvents,many of which pose a health risk to water users,it is not surprising that environmental technolo-gies have been designed to limit or reverse thedispersion of these dissolved groundwater con-taminants. During ‘pump and treat’, contami-nated groundwater is pumped from the aquifer,the contaminant is removed from the water, andthe cleaned groundwater is reinjected. Since manygroundwater contaminants sorb to the aquifermaterials and are only slowly released into theaqueous phase, it often takes many years to re-move the groundwater pollutants.

For demonstration purposes, let’s assume that1 000 000 gallons of groundwater are pollutedwith benzene at an average concentration of 0.5ppm. Thus, a total of 1

2 gallon of benzene needs tobe removed from the aquifer. Assuming that aminimum of 5 years of ‘pump and treat’ is neces-sary to reach cleanup objectives and that thegroundwater pump is operated by a diesel enginerequiring five gallons of diesel per day, it followsthat a total of 9125 gallons of diesel fuel areneeded for the cleanup effort.

According to the second law of thermodynam-ics (see Eq. (3)), it follows that the increase ofentropy in the environment due to diesel combus-tion is greater than the entropy reduction in theaquifer due to the removal of 1

2 gallon of benzene.If entropy can be used as a measure for pollution,it follows that limited groundwater cleanup iscarried out at the expense of greater atmosphericpollution consisting of highly dispersed contami-nants such as SO2, NOx, particulates, and CO2,the latter of which contributes to global warming(Houghton, 1997).

The common rationale for carrying out thesetypes of remediation efforts is that the principleobjective is the cleanup of groundwater. Consider-ing the limited focus of environmental science, itis not surprising that any other side effects areseldom considered. Even if one would try to doso, mechanistic reductionist environmental science


would be hopelessly ineffective in providing usefulanswers. For example, how would it ever be possi-ble to determine all the risks associated with thewidespread dispersal of the numerous air-pollu-tants including the greenhouse gas CO2 whoseeffects on climate change, sea-level rise, and spe-cies extinction are extremely complicated? Andhow can these risks ever be objectively comparedto all the health and environmental risks thatresult from 1

2 gallon of carcinogenic benzene dis-solved in 1 000 000 gallons of groundwater?

What happens in practice is that only the obvi-ous problems are addressed, i.e. those situationswhere sufficient scientific evidence is available toindicate that environmental pollutants are presentand might pose a risk to humans or whateverhumans value. All other environmental problems,including those caused by remediation efforts, areignored either because mechanistic environmentalscience has not yet elucidated the cause and effectrelationships or because impacts occur elsewhere(preferably far away) or affect receptors that arenot currently considered important. In short,whatever is scientifically known and sufficientlyvalued is addressed, whatever is unknown or notvalued receives no attention. For example, ben-zene in groundwater is of concern because it isknown to be carcinogenic. By contrast, the factthat pump and treat technology causes distantair-pollution effects (acid rain, smog, ozone, etc.)and future global warming is ignored because theimpacts are not obvious and immediate. It cantherefore be concluded that remediation technolo-gies which attempt to reverse pollutant dispersionappear to be effective only when viewed from avery limited perspective. However, when one con-siders all side-effects, both spatially and tempo-rally, it is extremely questionable whether theseremediation technologies can truly improve thequality of the overall environment.

It should be noted that from a practical per-spective most highly dispersed persistent contami-nants are unlikely to ever be removed from theenvironment. Again, the second law of thermody-namics predicts that it requires tremendousamounts of energy to concentrate highly dispersedmaterials (Faber et al., 1995). This is why noserious attempts have been made to remediate

persistent chlorinated pesticides such as DDT,now found in the fatty tissue of almost everyorganism in the world, or lead contaminationalong U.S. roads and freeways that is the result ofearlier use of leaded gasoline, or the widespreadmetal and hydrocarbon contamination in manyharbor sediments. In short, as soon as contami-nants ha�e dispersed into the en�ironment, remedia-tion technologies are extremely ineffecti�e inaddressing the problem — in fact, o�erzealousattempts might well be counterproducti�e.

This leads to the second type of remediationtechnologies that are designed to limit entropyincrease in the environment by ‘neutralizing’ of-fending compounds before they disperse into theenvironment where they might drive harmfulchemical reactions. A simple example is the neu-tralization of acid waste with alkali to create amixture of neutral pH that is much less harmfulthan the original waste. There are many remedia-tion technologies that use either chemical or bio-logical reactions to render the waste stream muchless reactive prior to release into the environment.From an entropy perspective, the waste reacheschemical equilibrium (i.e. maximum entropy) inthe reaction chamber instead of the environment.Examples of chemical and biological treatmenttechnologies are incineration, neutralization, andbioremediation.

While chemical and biological treatment tech-nologies are certainly more effective than thoseremediation efforts that simply attempt to limit orreverse contaminant dispersal, it is important torealize that many of these technologies also haveundesirable side-effects (Barton et al., 1996; Pageet al., 1998). For example, incineration mightgenerate flue gases that contain minute quantitiesof dioxins that are highly toxic. In addition, ashescontaining toxic metals are generated and have tobe dealt with (see section on physical treatmenttechnologies). Even bioremediation, a low impacttechnology relying on native bacteria to metabo-lize pollutants, can cause the formation of inter-mediates that are more toxic than the originalcontaminant (e.g. carcinogenic vinyl chloride dur-ing chlorinated solvent bioremediation (Baker andHerson, 1994)). Also, the addition of certain pro-cess chemicals that are used to stimulate bioreme-


diation may have a negative impact on the localsoil or sediment ecology.

Consider finally the energy requirements andthe resulting waste heat that is emitted into theenvironment as a result of pollution-control pro-cesses. According to Kummel and Schussler(1991), it is possible to calculate the ‘heat equiva-lents of noxious substances’ which is defined asthe waste heat generated by pollution controltechnologies and which can be used as an aggre-gate pollution index in economic modeling. Thisapproach shows that substance emissions can inprinciple be converted via pollution control pro-cesses into heat emissions, which are generallyconsidered a more benign form of environmentalpollution. However, the entropy-exporting wasteheat emissions could become critical when theanthropogenic heat input into the biosphere,which is currently around 1013 W, reaches themaximum permissible ‘heat barrier’ of about 3×1014 W (Von Buttlar, 1975). For example, thewastewater treatment plant at the BASF chemicalplant complex in Germany uses as much energy asa city of 50 000 inhabitants (Faber et al., 1995).Not only are large amounts of waste heat emittedinto the environment as a result of this treatmentbut, as shown in more detail below, additionalenvironmental deterioration is always associatedwith any form of energy generation. In summary,e�en effecti�e chemical and biological remediationtechnologies are likely to ha�e negati�e en�iron-mental impacts, many of which may not be ad-dressed because of the narrow focus of currenten�ironmental science and technology.

5. It is impossible to design industrial processesthat have no negative environmental impacts

Because most remediation technologies are notvery effective in dealing with dispersed contami-nants and since — based on the second law ofthermodynamics — secondary negative conse-quences of cleanup operations are inherently un-avoidable, it has been proposed that industrialprocesses be redesigned in such a way as to makethem virtually emission-free so that environmentalpollution is prevented in the first place. Conse-

quently, during the last decade, environmentalresearch began focusing on pollution prevention,and even a new academic discipline, industrialecology, was born with the mission to designzero-emission industrial processes (Allenby andRichards, 1994; Ayres and Simonis, 1994; Graedeland Allenby, 1995; Ayres and Ayres, 1996; DeSi-mone and Popoff, 1997; Allenby, 1999).

As shown in Fig. 3, current economic activitiesare causing serious pollution in the environmentfor at least two major reasons. First, more than90% (Pimentel et al., 1994) of the energy thatdrives the U.S. economy is derived from non-re-newable fossil fuels whose combustion byproductssuch as CO2, SO2, NOx, and particulates causewidespread air-pollution and global warming. Thesecond reason for the highly polluting nature ofthe current U.S. industrial system is related to thefact that materials flow in a linear fashionthrough the economy (Fig. 3A). Natural resourcesare extracted from the environment and refinedinto raw materials that are then manufacturedinto consumer products. After consumption, theresulting wastes are returned to the environmentwhere they often cause serious impacts toecosystems.

As was pointed out above, it is the ‘mission’ ofindustrial ecology to redesign current industrialprocesses in a way to make them more sustainableand compatible with the environment (Allenbyand Richards, 1994; Ayres and Simonis, 1994;Graedel and Allenby, 1995; Ayres and Ayres,1996; DeSimone and Popoff, 1997; Allenby,1999). In order to avoid the numerous negativeimpacts associated with the dispersal of wastesinto the environment, industrial ecologists havefocused most of their attention on ‘closing thematerials cycle’ (Graedel and Allenby, 1995;Ayres and Ayres, 1996; Allenby, 1999). As shownin Fig. 3B, it may be in theory possible to almostcompletely isolate the economy from the environ-ment by recycling all wastes into materials thatcan be manufactured into consumer products.While 100% recycling is thermodynamically notfeasible (Bianciardi et al., 1993), it may neverthe-less be possible to greatly reduce the dissipation ofwastes and the associated pollution effects. Inaddition, many industrial processes might also be


Fig. 3. Present unsustainable polluting linear flow economy (A) and future sustainable zero-emission circular flow economy (B).

redesigned to eliminate the presence of toxicbyproducts or wastes so that material releases tothe environment, if they were to occur by acci-dent, are relatively harmless. In summary, byclosing the materials cycle and eliminating therelease of toxics, it may well be possible to designfuture industrial processes that are virtually emis-sion-free and have a negligible impact on theenvironment.

However, a most important point that is com-monly overlooked by most industrial ecologistsand pollution prevention specialists is the fact thata significant amount of energy is required to runthe zero-emission industrial processes within thecircular flow economy, and that the generation ofany energy is associated with negative environ-

mental impacts. Considering the limited amountof remaining fossil fuels (Romm and Curtis, 1996;Campbell and Laherrere, 1998) and the extremelong-term hazards associated with nuclear energygeneration, it is clear that in the future all energymust come from renewable sources, i.e. it must bedirectly or indirectly derived from solar energy.Consequently, in an economy consisting of zero-emission industrial processes driven by renewableenergy, the only remaining environmental impactsare those related to the conversion of solar en-ergy. In other words, while it may be possible todesign industrial processes that are virtually emis-sion free, it is not possible to avoid the numerousnegative environmental consequences related tosolar energy production.


Fig. 4. Diversion of solar energy to human-related subsystems results in entropy increase in the environment.

It has been commonly assumed that renewableenergy generation is more environmentallyfriendly than non-renewable energy sources suchas fossil fuels or nuclear power (Brower, 1992;Boyle, 1996; Hayes, 1977; Lovins, 1977). Whilethis assumption may be correct, it must be real-ized that the capture and conversion of solarenergy will have significant negative environmen-tal impacts, especially if they are employed onsuch a large-scale as to supply nearly 100% of theU.S. energy demand.

Before discussing some of the potential negativeimpacts of different solar energy technologies, it isuseful to review again the implications of thesecond law of thermodynamics in order to showthat environmental impacts of renewable energygeneration are inherently not avoidable. This isbecause the flux of solar energy (or neg-entropy)onto earth is used to create highly ordered (i.e.low entropy) so-called ‘dissipative structures’ inthe environment (Nicolis and Prigogine, 1977;Atkins, 1984). Evidence of such structures can beseen in the complexity of organisms, ecosystems,biodiversity, and carbon and nitrogen cycles, all

of which are maintained by the constant in-flowof solar energy (Ayres and Martinas, 1995).

If the flow of solar energy were to stop — as itultimately will in a few billion years — all thesecomplex structures would decay and reach a finalequilibrium state where entropy is maximized.Similarly, if humans divert a fraction of solarenergy away from the environment to create or-dered structures for their own purposes (i.e.houses, appliances, transportation infrastructure,communication systems, etc.), less energy is avail-able to maintain these highly ordered dissipativestructures in nature. The disturbance of thesestructures translates into the various environmen-tal impacts that are associated with renewableenergy generation.

As shown in Fig. 4, the total amount of solarenergy (�Es) that is received on earth can beviewed as the sum of energy diverted for humanpurposes (�Eh) and energy that remains availableto maintain ‘order’ in the environment (�Ee):

�Es=�Ee+�Eh. (4)

According to the second law of thermodynam-ics, energy (�E) is used to decrease the entropy


(�S) (increase the order) of a system at tempera-ture T [°K] according to Faber et al. (1995)3:

�E= −T ·�S. (5)

Combining both Eq. (4) and Eq. (5) yields:

�Es= −T · �Se−T · �Sh, (6)

where �Se and �Sh are the change in entropy(order) in the environment and human-dominatedsub-system, respectively. Combining Eq. (4) andEq. (6), it follows that a change of entropy in theenvironment is related to a change of entropy inthe human-dominated subsystem according to:

�Se=−�Es

T−�Sh. (7)

Since the total flow of solar energy (�Es) isconstant, it follows that for each unit of ‘order’(neg-entropy) created by the diversion of solarenergy in the human-dominated subsystem, atleast one unit ‘disorder’ (entropy) is caused in theenvironment as evidenced by a wide range ofdifferent environmental disturbances.

Thus, the second law of thermodynamics dictatesthat it is impossible to a�oid en�ironmental impacts(disorder) when di�erting solar energy for humanpurposes.

This prediction, based on the second law ofthermodynamics, should be no surprise consider-ing the numerous roles solar-based energy flowsplay in the environment (Holdren et al., 1980;Clarke, 1994). For example, direct solar energyradiation is responsible for the heating of landmasses and oceans, the evaporation of water, andtherefore, the functioning of the entire climaticsystem. Wind transports heat, water, dust, pollen,and seeds. Rivers are responsible for oxygenation,

nutrient and organisms transport, erosion andsedimentation. The capture of solar energy viaphotosynthesis results in biomass that providesthe primary energy source for all living matterand therefore, plays a vital role in the mainte-nance of ecosystems (Clarke, 1994).

According to Holdren, the potential environ-mental problems with solar energy generation canbe summarized as follows: ‘Many of the poten-tially harnessable natural energy flows and stocksthemselves play crucial roles in shaping environ-mental conditions: sunlight, wind, ocean heat, andthe hydrologic cycle are the central ingredients ofclimate; and biomass is not merely a potential fuelfor civilization but the actual fuel of the entirebiosphere. Clearly, large enough inter�entions inthese natural energy flows and stocks can ha�eimmediate and ad�erse effects on en�ironmentalser�ices essential to human well-being’ (Holdren etal., 1980, p. 248).

A significant environmental impact is likely tobe related to the fact that large areas of terrestrialecosystems have to be converted to either biomassmonocultures or flooded to create lakes for hy-droelectric energy generation. For example, Pi-mentel et al. (1994) have estimated that ca. 20% ofthe U.S. land area would have to be dedicated forsolar energy generation to produce 37 quads,which is only ca. 40% of current total U.S. energydemand. From this it can be seen that theavailability of land might become a limiting factorin solar energy generation if close to 100% of thefuture U.S. energy demand has to be supplied byrenewables.

It should also be noted that a large amount ofrenewable and non-renewable resources will berequired to manufacture the solar energy capturetechnologies. For example, how much steel andconcrete will be required to build tens of millionspassive solar energy collectors and photovoltaicsolar panels or several million windmills? (Bezdeket al., 1982; Bezdek, 1993). Even using the bestprecautions, some pollution will occur during themanufacture and use of solar energy technologies.A particular concern may be the handling ofhundreds of million lead batteries used in energystorage and the leaching of fertilizers and pesti-cides that are applied in biomass plantations.

3 This relationship reflects the ideal case of equilibriumthermodynamics when heat transgresses across systemboundaries in a reversible manner. Clearly, the flow of solarenergy onto earth is a non-equilibrium process in which ‘addi-tional’ entropy is generated as result of irreversible phenomena(Prigogine, 1961; Kummel, 1994). Since the mathematicaltreatment of irreversible processes would greatly complicatethe current analysis, it is omitted here for the sake of simplic-ity. The overall conclusions reached from the interpretation ofEq. (7) are the same assuming either equilibrium or non-equi-librium conditions.


Finally, one of the most difficult problems toassess are potential eco-system impacts. Thesemay range from simple interventions such as theremoval of shade trees to bird and insect kills inwindmills. Since photosynthetically fixed energy(i.e. biomass) supports the great diversity of spe-cies inhabitating ecosystems (Vitousek et al., 1986;Wright, 1990), it follows that removal of thisenergy source will result in the extinction of spe-cies. For example, it has been determined that areduction of energy flow through an ecosystemwill result in a concomitant loss of species asshown in the species-energy curve in Fig. 5. Thus,the greater the diversion of solar energy for hu-man purposes (�E), the greater the loss of speciesdiversity (�SP) (Wright, 1990).

In summary, a wide range of negative environ-mental impacts are associated with the capture,generation, and storage of solar energy. Theseimpacts are likely to be very significant if close to100% of the U.S. energy demand has to be sup-plied by renewables, a condition that is becomingmore probable as non-renewable fossil energysources become much more expensive, especiallyafter 2010, and as the search for sustainable en-ergy solutions becomes more intense (Holdren,1990; Romm and Curtis, 1996; Campbell andLaherrere, 1998). Consequently, while it may inprinciple be possible to design future industrialprocesses that are virtually emission-free, it isinherently impossible to produce solar energywithout causing significant environmental im-

pacts. Thus, all industrial processes — howe�er,well designed to be emission-free — will ha�euna�oidable negati�e en�ironmental impacts relatedto energy generation.4

The critical reader might at this point expressreservations about the above analyses since com-plex environmental phenomena are reduced to asingle abstract entity such as entropy. Clearly,much more research is needed to experimentallydefine precisely the correlation between entropygeneration and environmental pollution. How-ever, a thermodynamic analysis of environmentaltechnologies has a distinct advantage in that itdeals with the entire system and not just oneisolated aspect of it as traditional reductionistscience. The second law of thermodynamics —like the law of gravity — applies everywhere andcannot be ignored when examining the effective-ness of environmental technologies.

6. Conclusions

Based on the above assessment, it is clear thatmodern science and technology have very limitedpotential to alleviate the numerous environmentalproblems facing western industrialized countries.This is due to the fact that: (a) reductionist,mechanistic science is inherently unable to providecomplete and accurate information about currentenvironmental problems; (b) the conservation ofmass principle and the second law of thermody-namics both dictate that negative environmentalconsequences of remediation activities are in mostcases intrinsically unavoidable; and (c) all indus-trial processes, even if designed to be emission-free, will have significant negative environmentalimpacts related to energy generation. In manycases, en�ironmental science and technology appearto be successful only because attention is focussednarrowly (in space and time) on specific objecti�eswhile wider, long-term impacts are ignored.

Given these serious limitations, are there anyroles that science and technology can play in

Fig. 5. Reduction in the energy flow through the ecosystemwill result in a reduction of species in the respective ecosystem.

4 For a more complete discussion of the limits of technolog-ical solutions to sustainable industrial development, see Huese-mann (2000b).


guiding the development of environmental poli-cies? Since the greatest strength of contemporaryscience is the detection of environmental distur-bances and the elucidation of isolated cause andeffect relationships, the main role of environ-mental science and technology should be to in-form policymakers of the many negative impactsof past and current industrial activities so thatactions can be taken to greatly modify or com-pletely discontinue them as soon as possible. Inaddition, the development of ever more precisemonitoring technologies will reduce the manyuncertainties associated with the estimation ofthe extent and magnitude of environmental pol-lution.

As was pointed out earlier, most environmen-tal problems are extremely complex and tradi-tional reductionist scientific methods have beenvery limited in revealing the intricate web offar-reaching environmental interactions. In re-sponse to this limitation, environmental sciencehas evolved in recent years to improve the un-derstanding of the ‘whole’ instead of its isolatedparts. This has been attempted by encouragingmore interdisciplinary research, carrying out sys-tems analyses, and by developing complex math-ematical models that are run on supercomputers(Funtowicz and Ravetz, 1994a). Despite thesenew trends, it is still highly questionable whetherthese methods will ever be able to provide infor-mation that is sufficiently accurate to guide pol-icy decisions. Consider, for example, theemergence of a new discipline called ‘earth sys-tems engineering’ (Allenby, 1999, p. 3) whosemission is to find technological solutions toglobal problems such as the greenhouse effect.Despite eager attempts by scientists and engi-neers to develop various technological ap-proaches to lower atmospheric CO2

concentrations (DOE, 1999), it is almost guaran-teed — given the high degree of scientific uncer-tainties — that these technological measures, ifimplemented, will cause serious, unexpected con-sequences elsewhere.

In their pioneering work on post-normal sci-ence, Funtowicz and Ravetz have proposed thatthe uncertainty of policy decisions related to themanagement of complex systems be decreased by

extending the peer community to include notonly scientists but also a variety of stakeholderssuch as representatives from industry, govern-ment, citizen groups and environmental organi-zations (Funtowicz and Ravetz, 1993, 1994a,b,c;Ravetz and Funtowicz, 1999). While this ap-proach clearly has the advantage of reducingbias (Huesemann, 2000a) and enhancing demo-cratic decision-making, it is still not clear howcomplex environmental problems can be solvedgiven the high degree of scientific uncertaintiesand the unavoidable negative impacts of most, ifnot all, environmental technologies. However,the inclusion of many different stakeholders inpolicy decisions might finally lead to the recog-nition of the enormous importance of non-tech-nical issues and to the discovery that changes inpersonal and social behavior may provide theonly possible solutions to many complex envi-ronmental problems.

If technological solutions to environmentalpollution are only of limited effectiveness, thequestion arises whether other alternative ap-proaches can be found to deal with this seriousand persistent problem. Again, the second lawof thermodynamics can be used to provide an-swers by shedding light on the most fundamen-tal causes of environmental disruption.According to the second law (see also Eqs. (3)and (7)), for each unit of ‘order’ that is createdby human activities, more than one unit of dis-order must be created in the surrounding envi-ronment.5 As was mentioned earlier, thehuman-generated ‘order’ is generally related tothe many physical artifacts and activities thatare considered signs of civilization such as theendless array of consumer goods. Conversely,the concomitant disorder created in the environ-ment manifests itself in the wide range of healthand environmental impacts. Consequently, envi-ronmental deterioration is directly related to thetotal quantity of human artifacts so that an in-crease in the quantity of artifacts will automati-

5 This concept is also well captured by Rifkin (1980, p. 123)who wrote: ‘Each technology always creates a temporaryisland of order at the expense of greater disorder in thesurroundings.’


cally increase environmental disruption. The latterpoint is succinctly summarized by Georgescu-Roegen, an economist who was the first to includeentropy considerations in economic theory(Georgescu-Roegen, 1971, 1977), who stated: ‘noone has realized that we cannot produce ‘betterand bigger’ refrigerators, automobiles, or jetplanes without producing also ‘better and bigger’waste’ (Georgescu-Roegen, 1980, p. 55).

For simplicity sake, if we equate the total quan-tity of artifact production and related serviceswithin a nation to the gross domestic product(GDP), it is clear that environmental deteriora-tion is directly related to the magnitude of theGDP according to the following equation(Graedel and Allenby, 1995):

Environmental Impact

=GDP×Environmental Impact/Unit GDP(8)

= Population×GDP/Person

×Environmental Impact/Unit GDP.

The term ‘Environmental Impact/Unit GDP’ isoften referred to as the ‘technology factor’, reflect-ing the idea that technological improvements ineco-efficiency can be counted on as the mainstrategy in reducing the environmental impact ofcurrent economic activities (Allenby andRichards, 1994; Ayres and Simonis, 1994; Graedeland Allenby, 1995; Ayres and Ayres, 1996; DeSi-mone and Popoff, 1997; Allenby, 1999). However,as has been pointed out repeatedly in this paper,the extent to which technology can improve theenvironmental performance of industrialeconomies is bounded by the second law of ther-modynamics, i.e. the technology factor can ne�erbecome zero.

Because of these thermodynamic limitations, itfollows from Eq. (8) that the total environmentalimpact can only be effectively decreased by reduc-ing either the size of the population (Ehrlich andEhrlich, 1991) or the per capita consumption ofgoods and services (Georgescu-Roegen, 1980;Daly, 1996) (or both). If economic (GDP) growthis allowed to continue without limits, any poten-tial reductions in environmental impact due to

technological improvements in eco-efficiency willnecessarily be short-lived. For example, let usbegin with the rather optimistic assumption that itwould take only 15 years6 of intensive R&D toimprove the overall eco-efficiency of the U.S.economy by 50%. This technological effort wouldcut the total environmental impact in half — butonly if there is no growth in GDP during that15-year period. Given the current addiction toeconomic growth, even a modest annual growthrate of 4% would double the GDP within 18years. At that point, the total environmental im-pact would be the same as today and additionalimprovements in eco-efficiency would be againneeded. At some point, eco-efficiency improve-ments are no longer possible because they becomecost-prohibitive due to the law of diminishingreturns characteristic of technological innovationsor because of thermodynamic constraints (i.e. sec-ond law). In summary, while science and technol-ogy can in a limited way reduce the environmentalimpact of current economic activities, long-termprotection of the en�ironment is only possible byreducing and limiting both the human populationsize and per-capita consumption (GDP/person).

At this point it becomes clear that the rootcause of the environmental crisis is the prevalenceof materialistic values that are the driving forcefor the present — as Daly aptly called it — ‘orgyof procreation and consumption’ (Daly, 1980,p.123). Thus, long-term protection of the environ-ment is not primarily a technological problem buta social and moral one. As was pointed outearlier, short-term techno-fixes are basically use-less unless the limited time they buy is used toradically change our values and behavior. Unfor-tunately, the opposite has been the case: Thepublic is led to believe that all problems are goingto be solved by science and technology and thatno change in values or behavior is necessary.Nothing could be farther from the truth.

6 For comparison, it took ca. 25 years (1959–1984) toreduce the energy intensity (energy use per unit GDP) of theU.S. economy by 50% — see Figure 2.3 in Graedel andAllenby (1995). During the same time period, the GNP grewmore than 200% (Figure 9–2 in Samuelson and Nordhaus,1989), indicating that total U.S. energy use did not decreasedespite improvements in energy efficiency.


References

Allenby, B.R., 1999. Industrial Ecology: Policy Frameworkand Implementation. Prentice-Hall, Englewood Cliffs, NJ.

Allenby, B.R., Richards, D.J. (Eds.), 1994. The Greening ofIndustrial Ecosystems. National Academy of Engineering,National Academy Press, Washington, DC.

Atkins, P.W., 1984. The Second Law. Scientific AmericanBooks, W.H. Freeman and Company, NY.

Ayres, R.U., 1998. Eco-thermodynamics: economics and thesecond law. Ecol. Econom. 26, 189–209.

Ayres, R.U., Ayres, L.W., 1996. Industrial Ecology: TowardsClosing the Materials Cycle. Edward Elgar PublishingCompany, Brookfield, VT.

Ayres, R., Martinas, K., 1995. Waste potential entropy: theultimate ecotoxic? Econom. Appl. 48, 95–120.

Ayres, R.U., Simonis, U.E. (Eds.), 1994. IndustrialMetabolism: Restructuring for Sustainable Development.United Nations University Press, New York, NY.

Baker, K.H., Herson, D.S., 1994. Bioremediation. McGraw-Hill, New York, NY.

Balzhiser, R.E., Samuels, M.R., Eliassen, J.D., 1972. ChemicalEngineering Thermodynamics — The Study of Energy,Entropy, and Equilibrium. Prentice-Hall, EnglewoodCliffs, NJ.

Barton, J.R., Dalley, D., Patel, V.S., 1996. Life-cycle assess-ment for waste management. Waste Manag. 16, 35–50.

Bedient, P.B., Rifai, H.S., Newell, C.J., 1994. Ground WaterContamination: Transport and Remediation. Prentice-Hall, Englewood Cliffs, NJ.

Bereano, P.L., 1976. Technology as a Social and PoliticalPhenomenon. John Wiley & Sons, NY.

Bezdek, R.H., 1993. The environmental, health, and safetyimplications of solar energy in central station power pro-duction. Energy 8 (6), 681–685.

Bezdek, R.H., Wendling, R., Bennington, G.E., Chew, H.R.,1982. National goals for solar energy: economic and socialimplications. Natural Res. J. 22, 337–360.

Bianciardi, C., Tiezzi, E., Ulgiati, S., 1993. Complete recyclingof matter in the frameworks of physics, biology, andecological economics. Ecol. Econom. 8, 1–5.

Boyle, G. (Ed.), 1996. Renewable Energy: Power for a Sustain-able Future. Oxford University Press.

Brower, M., 1992. Cool Energy: Renewable Solutions to Envi-ronmental Problems. MIT Press, Cambridge, MA.

Campbell C.J., Laherrere, J.H., 1998. The end of cheap oil.Scientific American March, 78–83.

Clarke, A., 1994. Comparing the impacts of renewables. Int. J.Ambient Energy 5 (2), 59–72.

Commoner, B., 1971. In: Alfred A. Knopf (Ed.), The ClosingCycle: Nature, Man, and Technology. New York, NY.

Daly, H.E. (Ed.), 1980. Economics, Ecology, Ethics: EssaysToward a Steady-State Economy. W.H. Freeman andCompany.

Daly, H.E., 1996. Beyond Growth: The Economics of Sustain-able Development. Beacon Press, Boston, MA.

DeSimone, L.D., Popoff, F., 1997. Eco-Efficiency: The Busi-

ness Link to Sustainable Development. MIT Press, Cam-bridge, MA.

DesJardins, J.R., 1993. Environmental Ethics: An Introduc-tion to Environmental Philosophy. Wadsworth PublishingCompany, Belmont, CA.

DOE, 1999. Carbon sequestration: research and development.U.S. Department of Energy Report, Office of Science andOffice of Fossil Energy, DOE/SC/FE-1.

Ehrenfeld, D., 1981. The Arrogance of Humanism. OxfordUniversity Press.

Ehrlich, P.R., Ehrlich, A.H., 1991. The Population Explosion.Touchstone Books.

Faber, M., Niemes, H., Stephan, G., 1995. Entropy, Environ-ment, and Resources. Springer Verlag, Berlin.

Funtowicz, S.O., Ravetz, J.R., 1993. Science for the post-nor-mal age. Futures 25 (7), 739–755.

Funtowicz, S.O., Ravetz, J.R., 1994a. Emergent complex sys-tems. Futures 26 (6), 568–582.

Funtowicz, S.O., Ravetz, J.R., 1994b. The worth of a song-bird: ecological economics as a post-normal science. Ecol.Econom. 10, 197–207.

Funtowicz, S.O., Ravetz, J.R., 1994c. Uncertainty, complexityand post-normal science. Environ. Toxicol. Chem. 13 (12),1881–1885.

Georgescu-Roegen, N., 1971. The Entropy Law and the Eco-nomic Process. Harvard University Press, Cambridge, MA.

Georgescu-Roegen, N., 1977. The steady-state and ecologicalsalvation: a thermodynamic analysis. BioScience 27 (4),266–270.

Georgescu-Roegen, N., 1980. The entropy law and the eco-nomic problem. In: Daly, H.E. (Ed.), Economics, Ecology,Ethics: Essays Toward a Steady-State Economy. W.H.Freeman and Company.

Graedel, T.E., Allenby, B.R., 1995. Industrial Ecology. Pren-tice Hall, Englewood Cliffs, NJ.

Hayes, D., 1977. Rays of Hope: The Transition to a Post-Petroleum World. W.W. Norton, New York, NY.

Holdren, J.P., 1990. Energy in transition. Scientific AmericanSeptember, 157–163.

Holdren, J.P., Morris, G., Mintzer, I., 1980. Environmentalaspects of renewable energy sources. Annu. Rev. Energy 5,241–291.

Houghton, J.T., 1997. Global Warming: The CompleteBriefing, 2nd. Cambridge University Press.

Huesemann, M.H., 2000a. The inherent biases in environmen-tal research and their effects on public policy. Futures(accepted for publication).

Huesemann, M.H., 2000b. The limits of technological solu-tions to sustainable development. J. Ind. Ecol. (Submitted).

Klaassen, C.D., Amdur, M.O., Doull, J. (Eds.), 1986. Toxicol-ogy: The Basic Science of Poisons, 3rd. Macmillan Publish-ing Company, New York, NY.

Kummel, R., 1989. Energy as a factor of production andentropy as a pollution indicator in macroeconomic model-ing. Ecol. Econom. 1, 161–180.

Kummel, R., 1994. Energy, entropy — economy, ecology.Ecol. Econom. 9, 194–196.


Kummel, R., Schussler, U., 1991. Heat equivalents of noxioussubstances: a pollution indicator for environmental ac-counting. Ecol. Econom. 3, 139–156.

Longino, H.E., 1990. Science as Social Knowledge: Values andObjectivity in Scientific Inquiry. Princeton UniversityPress, Princeton, NJ.

Lovins, A.B., 1977. Soft Energy Paths: Toward a DurablePeace. Ballinger Publishing Company, Cambridge, MA.

Nicolis, G., Prigogine, I., 1977. Self-Organization in Non-Equilibrium Systems: From Dissipative Structures to Or-der Through Fluctuations. John Wiley & Sons, New York,NY.

Page, C.A., Diamond, M.L., Campbell, M., McKenna, S.,1998. Take a life-cycle view of site remediation. ChemicalEngineering Progress August, 63–70.

Pimentel, D., Rodrigues, G., Wang, T., Abrams, R., Gold-berg, K., Staecker, H., Ma, E., Brueckner, L., Trovato, L.,Chow, C., Govindarajulu, U., Boerke, S., 1994. Renewableenergy: economic and environmental issues. Bioscience 44,536–547.

Prigogine, I., 1961. Thermodynamics of Irreversible Processes.John Wiley & Sons, New York.

Prigogine, I., 1989. What is entropy? Naturwissenschaften 76,1–8.

Ravetz, J.R., Funtowicz, S.O., 1999. Post-normal science —an insight now maturing. Futures 31, 641–646.

Rebane, K.K., 1995. Energy, entropy, environment: why isprotection of the environment objectively difficult? Ecol.Econom. 13, 89–92.

Rifkin, J., 1980. Entropy — A New World View. The VikingPress, New York, NY.

Romm, J.J., Curtis, C.B., 1996. Mideast oil forever? AtlanticMon. 277 (4), 57.

Samuelson, P.A., Nordhaus, W.D., 1989. Economics, 13th.McGraw-Hill, New York, NY.

Sarewitz, D., 1996. Frontiers of Illusion — Science, Technol-ogy, and the Politics of Progress. Temple University Press.

Schwartz, E., 1971. Overskill — The Decline of Technology inModern Civilization. Quadrangle Books, Chicago, IL.

Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H., Matson, P.A.,1986. Human appropriations of the products of photosyn-thesis. BioScience 36 (6), 368–373.

Von Buttlar, H., 1975. Umweltprobleme. Phys. Bl. 31, 145–155.

Wright, D.H., 1990. Human impacts on energy flow throughnatural ecosystems, and implications for species endanger-ment. Ambio 19 (4), 189–194.

.

Documents

Can pollution problems be effectively solved by ... 5.pdf · Can pollution problems be effectively solved by environmental science and technology? An analysis of ... man ‘to establish