See discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/270891359

The Boundaries of Urban : Towards a Political-Industrial

ARTICLE in PROGRESS IN HUMAN · DECEMBER 2014 Impact Factor: 4.39 · DOI: 10.1177/0309132514558442

CITATIONS READS 7 106

2 AUTHORS:

Joshua Peter Newell Joshua Cousins University of Michigan University of Michigan

22 PUBLICATIONS 250 CITATIONS 2 PUBLICATIONS 10 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Joshua Peter Newell Retrieved on: 12 October 2015 Article

Progress in 1–27 ª The Author(s) 2014 The boundaries of urban Reprints and permission: sagepub.co.uk/journalsPermissions.nav metabolism: Towards a DOI: 10.1177/0309132514558442 political– phg.sagepub.com

Joshua P. Newell University of Michigan, USA Joshua J. Cousins University of Michigan, USA

Abstract This paper considers the limits and potential of ‘’ to conceptualize processes. Three ‘’ of urban metabolism have emerged. Each privileges a particular dimension of urban space, shaped by epistemology, politics, and model-making. Marxist ecologies theorize urban metabolism as hybridized socio- that (re)produce uneven outcomes; industrial ecology, as stocks and flows of materials and ; and , as complex socio-ecological . We demarcate these scholarly islands through bibliometric analysis and literature review, and draw on cross-domain mapping theory to unveil how the metaphor has become stagnantin each. To reinvigorate this research, the paper proposes the development of political–industrial ecology, using urban metabolism as a boundary metaphor.

Keywords boundary object, industrial ecology, interdisciplinary research, metaphor, , urban ecology, urban metabolism

I Introduction relationships, borrowing heavily from Moles- chott’s (1852) work on cycling and exchange of Theory of metabolism in formalized written energy–plant–animal nutrients and from chemist word dates back to at least the time of Sanc- Justus von Liebig’s (1842, 1859) use of ‘metabolic torius (1712 [1614]), who spent some 30 years rift’ to characterize declining soil conditions due to eating, working, and sleeping on his famous the intensification of agriculture and the corre- balance seat, meticulously weighing his diet- sponding spatio-temporal disconnects between ary intake and bodily excretions. He concluded urban and rural (Swyngedouw, 2006a). that a substantial portion of his food was lost through his skin as perspiratio insensibilis or ‘insensible perspiration’ (Figure 1). Much later, in his treatise on cells, Schwann (1839, 1847) first used the term metabolische, from Corresponding author: Joshua P. Newell, School of Natural Resources and which metabolic derives. And of course Marx Environment, University of Michigan, 440 Church Street, famously deployed metabolism (stoffwech Ann Arbor, MI 48109, USA. sel) to characterize complex –society Email: [email protected] 2 Progress in Human Geography

city. Meanwhile, the Metabolist movement in Japan brought an avant-garde approach to as organic process, where buildings and other features of the city exhibited meta- bolic cycles. Although only a single building, the Nakagin Capsule Tower in Tokyo, was built (1972), the utopian visions of the organismic city – replete with prefabricated modules to replace outdated elements of the urban form – continues to influence architecture today (Lin, 2010). From these intellectual legacies, academia has witnessed a remarkable resurgence of ‘urban metabolism’ over the past two decades, sparked by increased attentiveness to processes of , to material and cultural con- sumption, and to notions of . How- ever, scholars deploying the metaphor ascribe very different meanings and models to it. Build- ing from Marx, human and sociol- ogists use it to denote the interwoven knots of social and natural processes, material flows, and spatial structures. Industrial ecologists in the Wolman tradition use mass–balance accounting methodologies such as (MFA) to quantify metabolism’s ‘stocks’ and ‘flows’. Then, there is the ‘’ research strand stemming from Eugene Odum’s and metabolism theory. Figure 1. Sanctorius (1561–1636) at his famous These diverse uses of urban metabolism beg balance seat, where he spent the better part of 30 the following questions: Given that scholarly years measuring his body’s metabolism. Source: University of Kansas Medical Center communities share the metaphor, do they have Clendening Library, 2000. similar intellectual lineages? Do these commu- nities interact? If not, why? How might geogra- phers apply theories and approaches in Geographers are well familiar with Burgess industrial ecology and urban ecology to advance (1925) and other human ecologists in the Chi- understanding of urban space? cago School of , who theorized a To begin to answer these questions, this essay transposition of ecological theory onto society reviews 47 years of academic literature on urban to explain (and in cases predetermine) urban metabolism (1965–2012) by using bibliometrics and spatial pattern. In 1965, to map the boundaries of scholarly communities sanitary engineer Abel Wolman wrote a seminal deploying the metaphor (Section II). We deline- article in Scientific American that quantified the ate and analyze the emergence of three ‘ecolo- metabolic inputs (e.g. water, food, and fuel) and gies’ of urban metabolism: industrial ecology, outputs (e.g. waste) of a hypothetical American Marxist ecologies, and urban ecology (Section Newell and Cousins 3

III). As the bibliometric analysis reveals, these But just as broader Marxist theory has thought traditions have effectively formed struggled to move beyond strictly a ‘social the- islands of urban metabolism, with little cross- ory’ (Swyngedouw, 2006a), there is an absence fertilization. The urban political ecology (UPE) of ‘ecology’ in much of the UPE metabolism metabolism community, relatively small in research. In addition, UPE urban metabolism terms of scholars and citations, is especially iso- typifies other perceived shortcomings in ‘first lated from the other clusters. wave’ UPE scholarship (Heynen, 2014), includ- These clusters are bounded by the ways in ing a preponderance of qualitative approaches which theories of metabolism are transposed and a ‘methodological cityism’ that ends up pri- on the city as metaphor. Stemming from the vileging a traditional, bounded conception of Latin metaphora, meaning ‘to carry over’ or the city (Angelo and Wachsmuth, 2014). ‘to transfer’, metaphor literally means to To reinvigorate the urban metabolism meta- understand and experience one thing in terms phor in geography and move UPE forward as a of another. Metaphors like urban metabolism part of a ‘second wave’ research agenda, we pro- have the power to structure entire research pose the development of political–industrial paradigms through metaphorical redescrip- ecology through the creative infusion of ideas tion (Hess, 2010). Geography is filled with and approaches from the other two ecologies such large metaphors (Barnes and Curry, (Section V). Industrial ecology offers a suite of 1992), from assemblages, networks, cyborgs, untapped methods, such as material flow analysis hybrids and circulation, to coupled natural– (MFA) and life cycle assessment (LCA), with human systems, the Anthropocene, adaptation, which to map out and quantify the networks of and resilience. The metaphorical generation flows that circulate within and beyond the city. process makes the unknown more familiar, yet Urban ecology, meanwhile, provides the theory results in an inevitable partiality (Richards, necessary to explore the ecological and biophysi- 1936; Lakoff and Johnson, 1980; Smith and cal of urban space and is nested in an Katz, 1993). Just as focusing light on part of ontological nature–society construction that a room highlights certain features, while hiding resonates with UPE. others, the partiality of metaphors gradually Cognizant of the challenges of interdisciplin- becomes canonized through habitual use. As ary scholarship, including the difficulty in Barnes (1996: 154) explains, the initial ‘murki- reconciling the divergent spatial politics that ness’ of the chosen metaphor is cleared up, mov- underlie the respective metaphorical deploy- ing from hermeneutic-to-epistemological. In the ments, we are not proposing any sort of frame- end, metaphors become literal, in essence ‘dead’. work that presupposes an integrative approach In this essay, we expose the partiality ende- to studying urban metabolism. Instead, urban mic to urban metabolism in the three ecologies metabolism can be collectively conceptualized and argue that the metaphor has reached the as a ‘boundary metaphor’ – drawing on the con- point of maturity, becoming stagnant –insome cept of boundary objects (Star and Griesemer, instances, ossified (Section IV). In the case of 1989), whereby scholarly communities might geography, Marxist urban political ecologists interact through empirical practice and as a have so successfully forged the conceptual and means to explore the friction between the varied methodological apparatus for urban metabo- epistemologies, methodologies, and framings of lism that it has become almost synonymous urban metabolism. Towards that end, we provide with urban socionatural transformations a hypothetical case of the political–industrial wrought by the uneven processes of capitalist ecology of urban water supply. Through these development. interdisciplinary engagements, a more robust 4 Progress in Human Geography metaphor of urbanization can emerge: the meta- following terms: city, , urban, rift, social, bolism of the urban . socio-economic, society, and industrial. These terms enabled the widest range of citation data possible to capture key papers, yielding 696 II Mapping disciplinary worlds entries. Then, to narrow the focus to relevant sub- Scholars use bibliometrics to map the dynamics ject areas in the social sciences, physical of research fields and to situate scholarly work sciences, and engineering, articles from fields in relation to their academic communities (see such as toxicology and pharmacology were Morris and Van der Veer Martens, 2008, and excluded. For example, we removed a technical Persson et al., 2009, for basic guidance on how study using urban samples to analyze the human to conduct a bibliometric analysis). The results metabolism of a substance. This process reduced provide a map of scientific knowledge that the total to 311 articles in the following ISI- Small (1999: 799) describes as ‘a spatial repre- defined subject areas: environmental sciences sentation of how disciplines, fields, specialties, (59%), engineering (24%), business economics and individual articles or authors are related to (12%), geography (11%) and (8%). one another as shown by their physical proxim- WoS does not include citations outside of the ity and relative locations’. As with , WoS-registered journals, which eliminated some bibliometric techniques map known structures influential articles, such as Fischer-Kowalski of a and reveal a limited view of ele- (1998), from the initial keyword search. WoS ments. Our analysis focused on mapping the also excludes many trade journals, books, and structure and interconnections between authors book chapters, making citation counts in WoS around the base concept of urban metabolism smaller than in Google Scholar. WoS is also pri- at a certain level of detail and inclusiveness so marily an English-language , giving the as to better understand the extent to which scho- results an Anglo-American bias. For this reason, larly communities are acting independently Chinese and Japanese urban metabolism scho- from each other and the degree to which larship, which is fairly active, is underrepre- cross-fertilization occurs around the concept. sented in the bibliometric mapping. To ensure Inherently a subjective process, choices to inclusion of as many publications as possible, remove or add links within the maps were based we used the ‘create citation report’ function and on a desire to balance the highest possible num- selected ‘full record’ with ‘cited references’ to ber of data points with visual clarity. The biblio- download the citation data of all articles that metric software BibExcel (Persson et al., 2009) cited the original 311 articles. These cited refer- was used to prepare the network data from the ences, including books and non-English jour- collected records; Pajek (2012) software was nals, provided as robust a picture of the used to map the field. structure and evolution of the fields employing To ensure germaneness to urban metabolism, the metaphor as possible, given the limitations we performed a keyword search using the Insti- of WoS. tute of Scientific Information’s (ISI) Web of To map scholarly networks, the bibliometric ScienceTM (WoS) database. WoS was chosen analysis incorporated three measures: direct over Google Scholar because the former allows citations, co-citations, and weighted-direct cita- downloading full citation records, including tions (Figure 2). Direct citation analysis is espe- cited references and cited reference counts, into cially helpful for identifying emerging research a .txt file compatible with BibExcel. In WoS we trends (Shibata et al., 2009) and clarifying infor- searched for articles with ‘metabolism’ or mation flows between two articles (Persson and ‘metabolic’ in the topic or title, plus one of the Ellega˚rd, 2012). To identify the specific direct Newell and Cousins 5

authors who typically collaborate. To identify the primary works influencing the research field, we required a minimum of ten citations among the reference list. The co-citation net- work was mapped in Pajek using the Kamada-Kawai layout algorithm, a forced- directed layout utilized for graphing networks in two-dimensional space; it improves map readability by reducing the number of over- lapping and crossing links. After data impor- tation, complexity was further reduced by removing edges, or lines, with a value less than five (i.e. papers sharing fewer than five references). This improved legibility and highlighted the more prominent groups within the network, while retaining its overall structure. To identify additional research commu- nities, the third step used weighted-direct cita- tions, which integrates direct and co-citations with shared references (so-called bibliographic couplings) into a single metric of citation strength and further decomposes the network Figure 2. Bibliometric analysis measures: direct of papers (Persson, 2010). This is based on the citation; co-citation; and weighted-direct citation. rationale that the more shared references between two papers, the more similar they are citation links among the downloaded articles, in topic (Kessler, 1963). The strongest direct we omitted the second initial of the author’s first citation links between papers are those that name, thus minimizing the chance of missing a share many references and are frequently co- link due to spelling variations. We then linked cited. Weighted-direct citations calculate these articles based on author last name and first ini- relationships by assigning values to the cita- tial, year, journal, volume, and start page. To tions. A direct citation link between two articles illustrate the most influential scholars and pub- would be assigned a value of one, for example, lications in the field, the results were plotted but more shared references and co-citations can over time. strengthen the link. Thus, if a direct citation Co-citation links, introduced by Small link is accompanied by one shared reference (1973), posit that a stronger relationship is and one co-citation, the weighted-direct cita- established between articles if they appear tion has a value of three; for each additional in the references of a third document. This co-citation or shared reference, the value grows technique highlights instances of dense direct by a factor of one. Again using the Kamada- citation, enabling the network to be broken Kawai layout in Pajek to map the network, we up into meaningful groups or components removed the weaker components to improve (Small, 1999). The result is a more nuanced map clarity and to maintain consistency with view of the interaction structure between the co-citation map by requiring a minimum scholarly communities and the groups of value of four. 6 Progress in Human Geography

Figure 3. The three thought traditions of urban metabolism research: Marxist ecologies, industrial ecology, and urban ecology. The figure illustrates the direct citation network of the 35 most cited articles informing theory on urban metabolism. Arrows indicate a direct citation, and node size is proportional to the number of citations.

III Three distinct ‘ecologies’ and Newman (1999) proposed extending metabo- emerge lism to include livability and health. But in the end, planning was excluded because the meta- The bibliometric analysis, with subsequent map- phor is not infused with the same theoretical or ping, indicates three ecologies – industrial, Marx- methodological rigor and is not consistently visi- ist, and urban; each includes strands from multiple ble in the bibliometric analysis. disciplines but can be traced to a common line- Based on citations of the 35 most cited age.1 Labeling these three thought traditions as books and articles, the most influential of these such took careful consideration. A case could be ecologies in the urban metabolism field is made for rather than urban ecol- industrial ecology (49%), followed by the ogy. Girardet (1992) used linear vs. circular meta- Marxist ecologies (29%) and urban ecology bolism as a heuristic device to link ideas of (16%), which are briefly characterized in the to cities. Wackernagel next section. Fields such as architecture and and Rees (1996) used the metaphor symbolically urban planning (6%) form the bulk of the to characterize the of cities, Newell and Cousins 7 remaining citations. Figure 3 illustrates the be an energy or material source for another most influential thinkers, as indicated through (e.g. forging of bricks). citations. Wolman and Robert Ayres appeared The co-citation network map (Figure 4), early on and were seminal to industrial ecology which provides insight into the intellectual ter- formulations. By the late 1990s, urban metabo- rain of the metaphor and respective knowledge lism in Marxist ecologies began to emerge with structures, indicates two dense but distinct net- works by Swyngedouw and John Bellamy Fos- work clusters in industrial ecology. We label the ter, followed by those of Gandy, Kaika, Hey- first as ‘traditional’ urban metabolism and nen, and York. By the early 2000s, notable the second as the Vienna School of socio- urban ecologists such as Steward Pickett and economic metabolism, developed by sociologist Nancy Grimm were employing the metaphor. (and former president of the International Society of Industrial Ecology) Marina Fischer- Kowalski and her colleagues at Vienna’s Insti- 1 Industrial ecology tute of Social Ecology (Fischer-Kowalski, Pioneered by physicists and engineers (Ayres 1998; Fischer-Kowalski and Hu¨ttler, 1998). As and Kneese, 1969) in the late 1960s, industrial illustrated by the weighted-direct citation analy- ecology is largely a normative project whereby sis map (Figure 5), Fischer-Kowalski’s (1998) ‘nature’ serves as a model, both in terms of intellectual history of urban metabolism links structure and function, for the analysis of exist- the two clusters. Nonetheless, the Vienna ing industrial systems and for guiding new, School has evolved into being largely distinct. more efficient and resilient forms (Jelinski et al., 1992; Frosch, 1992). Building from Wol- ‘Traditional’ urban metabolism. The densest net- man, Ayres (1989, 1994) developed conceptual works of scholarly interaction are in this net- foundations for ‘’ that work cluster (Figure 5). MFA is traditionally specified more concretely how industrial sys- used to develop strategies to gradually demater- tems were analogous to organisms in terms of ialize the metabolism by optimizing resource processing waste and energy. Within industrial use. An influential early case study was the ecology, urban metabolism is defined as ‘the metabolism of Hong Kong (Newcombe et al., sum total of the technical and socio-economic 1978), which Warren-Rhodes and Koenig processes that occur in cities, resulting in (2001) bookended, leading them to conclude growth, production of energy, and elimination that Hong Kong’s ‘metabolic rates’ had of waste’ (Kennedy et al., 2007: 44). increased substantially. Similarly, in a meta- Intellectual underpinnings for industrial ecol- analysis, Kennedy et al. (2007) concluded that ogy stem from Lavoisier’s Law of the Conserva- eight metropolitan across five conti- tion of Mass (1789). Stating that in chemical nents exhibited ‘increasing per capita metabo- reactions mass is neither created nor destroyed, the lism’. But most case studies – Tokyo (Hanya law provides the basis for the mass-balance and Ambe, 1976), the Swiss Lowlands (Baccini, approaches and methodologies driving industrial 1997), Paris (Barles, 2007a, 2007b, 2009), ecology. In urban metabolism research, the most Vienna (Hendriks et al., 2000), Stockholm widely used is material flow analysis, or MFA (Burstro¨m et al., 1997), Toronto (Sahely (Baccini, 1996; Baccini and Brunner, 1991), et al., 2003), Singapore (Schulz, 2007), Beijing which is a fundamental tool for facilitating indus- (Zhang et al., 2011), and Lisbon (Niza et al., trial symbiosis (Chertow, 2000, 2008) – the 2009) – have been temporal snapshots, with strategic co-location of industrial activities that varying degrees of flow inclusiveness (usually enable waste from one factory (e.g. fly ash) to water, materials, and energy, sometimes food). 8 Progress in Human Geography

Figure 4. This co-citation map illustrates three network clusters that inform the urban metabolism knowledge base: ‘traditional’ industrial ecology, Vienna School of socio-economic metabolism, and Marxist ecologies. Note the important linkage played by Fisher-Kowalski (1998) between the two industrial ecology network clusters. The density of the interconnections indicates a greater degree of co-citation strength.

Within this cluster, an interesting research flows, but it is far more ambitious in that strand combines MFA with Eugene Odum’s sweeping historical materialism explains how systems ecology (1953, 1969) and H.T. Odum’s socio-economic transitions (such as from concept of ‘emergy’ (1983, 1996) – a measure- agrarian to industrial) have shaped stock– ment unit for a system’s energy (expressed as flow trajectories. The analyses make heavy equivalents). Early case studies use of metabolism as metaphorical expres- included Miami (Zucchetto, 1975) and 1850s sions (‘socio-metabolic regimes’, ‘metabolic Paris (Stanhill, 1976), and the most highly cited rates’ and ‘metabolic profiles’) and consider works are of Taipei, Taiwan, by Huang (1998) issues of labor, capital, , income dis- and Huang and Hsu (2003). tribution, and consumption. Some research links changes in land use to metabolism Vienna School of socio-economic metabolism. This dynamics, but those changes are rarely made second network cluster is smaller, but accord- spatially explicit, perhaps because the analy- ing to the bibliometric analysis scholars tical frame often extends to broader spatial within it are some of the most prolific scales: the socio-economic energy metabo- (Haberl and Krausmann with 17 records, lism of Austria from 1830 – 1995 (Kraus- Fischer-Kowalski with 10). Their metabolism mann and Haberl, 2002; Krausmann et al., work also uses MFA to quantify stocks and 2003); changes in societal metabolism and Newell and Cousins 9

Figure 5. This weighted-direct citation map expands urban metabolism research into four network clusters. The Marxist ecologies thought tradition is now divided into two network clusters: metabolic rift and urban political ecology. Primary nodes are labeled by author and date. Arrows represent direct citation links. land use in the United Kingdom from 1950 – and Sieferle (Weisz, 2011; Weisz and Clark, 2001 (Schandl and Schulz, 2002); social 2011). metabolism in agrarian Catalonia from 1860 – 1870 (Cusso´ et al., 2006); and even the socio-metabolic transition of the globe 2 Marxist ecologies (Haberl, 2006). Two isolated clusters, one on UPE and the Although the Vienna School draws strategi- other on Foster’s theory of metabolic rift cally from the social (e.g. sociology, history, (Figure 5), utilize metabolism to investigate economics, anthropology, geography) and natu- a range of related themes inspired by Marx’s ral (e.g. , ecology) sciences (Fischer- theorizations on metabolism2 –frombroad Kowalski, 1998; Fischer-Kowalski and Weisz, tensions between nature and society (Cronon, 1999), their theorization of the socio-economic 1991; Gandy, 2002) to the role of as a process of nature–society coe- power in shaping urban space and access to volution is rooted in sociology; Luhmann’s resources (Swyngedouw, 2004) and the rup- (1984) theory of social systems is particularly tures between human production and the nat- influential, as is the work of Boyden, Godelier, ural world wrought by urbanization and 10 Progress in Human Geography growth in long-distance trade (Broto et al., Metabolic rift. The second cluster also uses meta- 2012; Foster, 1999; Moore, 2000). bolism to characterize intertwined nature–soci- ety relationships but differs due to a sustained Urban political ecology. Urban political ecolo- focus on the metabolic rift between town and gists use metabolism to characterize nature– countryside. Drawing on Marx’s writings, Fos- society relationships as dynamic and net- ter (1999, 2000) frames this as an antagonistic worked circulations that reorganize social human–environment binary that emerges due and physical environments into socionatural to the processes of capitalism (such as how capi- assemblages (Swyngedouw, 2006a). Highly talist forms of urbanization create this rift). Col- influential in the ‘first wave’ of UPE (Hey- lectively, as indicated in Figure 5, Foster’s nen, 2014) and development of the metaphor elaborations on the rift have informed case stud- is Harvey’s (1996) notion of a ‘created eco- ies in a range of ‘spatio-temporalities’ (Shep- system’ (which interprets the ecology of cit- pard, 2010), from Clausen and Clark’s (2005) ies in a manner reflective of Marx’s examinations of depleted fish stocks and the metabolism as a nature–society dialectic), expansion of aquaculture in relation to capitalist Smith’s (2008) theorizations of the metabo- production to Wittman’s (2009) illustrations of lism of nature as embedded in the production how peasants’ movements can bridge the rift. of nature, and Cronon’s (1991) vivid account In this vein, McClintock’s (2010) work illus- of how the urbanization of Chicago trans- trates the potential for linking metabolic rift formed both city and countryside to produce with core UPE concerns as he reveals from eco- a unique political ecology. logical, social, and individual perspectives how The outcome has been a robust scholarship is both an outcome of the rift’s that shows urban space to be a socionatural alienating forces and an attempt to ameliorate it hybrid. Scholars have used the concept to by reconnecting urbanites with food production. explore both how nature is transformed by and enrolled into the political and socio-economic practices that shape urban form and its meta- 3 Urban ecology bolic relationship to other (Gandy, Evidenced by the lack of large network clusters 2002) and how the flows of water (Swynge- (Figures 4 and 5), urban ecology has used the douw, 2004), fat (Marvin and Medd, 2006), metaphor with less persistence, initially surpris- alcohol (Lawhon, 2013), and the technologies ing given the highly cited work on urban meta- and connecting these objects bolism by ecologists (Figure 3) and system and places (Gandy, 2005; Kaika and Swynge- ecologist EP Odum’s continued influence in douw, 2000; Monstadt, 2009) form a ‘socio- industrial ecology. Golubiewski (2012a) main- environmental metabolism’ entangled with the tains industrial ecologists misread Odum’s dynamics of social power and capital. Collec- work, asserting he referred to the metabolism tively, this community of scholars utilizes the of an individual organism, not the level of com- metaphor to destabilize binaries (e.g. nature– munity. But a close reading of Odum’s texts society, city–countryside) and unveil uneven reveals numerous references to a more exten- power relationships shaping urban space with sive metabolism (e.g. the ‘community metabo- the normative goal of fostering more sustain- lism of self-sustaining micro-’ able and democratic forms of urban environ- (1968: 16). mental governance and policy-making (Keil For many ecologists, Odum’s ideas are out- and Desfor, 2003; Swyngedouw, 2004; Swyn moded and misapplied with respect to the ecol- gedouw and Heynen, 2003). ogy of the city. A reading of the highly cited Newell and Cousins 11 work indeed reveals ambivalence towards the shared Marxist legacy? Scholarly islands extend metabolism concept, deploying it for its meta- beyond epistemology, ranging from the object phorical power yet being cognizant of its con- and method of study to the disciplinary-bound tested status vis-a`-vis cities within the ecology reward structures of the academy; all combine discipline. Thus Grimm et al. (2008) acknowl- to form a process of exclusion stemming from edge the debate about its appropriateness but knowledge production in disciplinary cultures highlight its utility as a method to quantify (Schoenberger, 2001). resource consumption trends. Decker and col- Space limits a detailed excavation of how and leagues (2000) use the term similarly in their why metabolism islands have emerged. The synthesis of the material and energy flows of focus here is on the use of metaphor because, 25 world cities, as do Ngo and Pataki (2008) unlike in more Kantian traditions where its role in their longitudinal mass-balance analysis of might be circumscribed to metaphorical expres- Los Angeles County’s fuel, water, and food sion (e.g. ‘it’s raining cats and dogs’), we view flows. the metaphorical process as fundamental to sci- Nevertheless, given the importance of ecolo- entific theorizing. This aligns with work on gical metabolism theory, the presence of highly metaphor by Richards (1936) and Lakoff and cited works, and the engaged debate by ecolo- Johnson (1980), by critical realist philosophers gists over the metaphor’s appropriateness, we who view it as the generative basis for making include urban ecology as the third ‘ecology’. sense of the world (Lewis, 1996), and by geo- Let’s now analyze how epistemology, politics, graphers who write of its conceptual power and and model-making are intertwined with the con- therefore the importance of exposing the poli- struction of the metaphor in each of the three tics associated with their construction and use ecologies. (Barnes and Curry, 1992; Smith and Katz, 1993; Robbins, 2013). Just as metaphors are inseparable from their IV Islands of urban metabolism ascribed meanings and politics, so too are the Purcell (2003) used a debate over scale, between models generated from them that make meaning Marston and Brenner, to make the case that of the world. Black (1962: 40) recognized this ‘islands of practice’ form not because of compet- early on: ‘to speak of ‘‘models’’ in connection ing epistemologies but instead because of research with a scientific theory already smacks of the practices (e.g. data collection, analysis, and dis- metaphorical’. As structured simplifications of course) that gradually limit outside engagement. complex phenomena, models need not only Purcell was referring specifically to schisms refer to statistical equations, graphs, flow within human geography, while our analysis charts, or maps (Pickett et al., 2004). spans disciplines and thought traditions that exhi- Richards (1936) identified metaphor as hav- bit a divide between more positivist approaches in ing a ‘tenor’ (its underlying subject) and a ‘vehi- urban ecology and industrial ecology and the cle’ (the terms for presenting that subject). range of critical realist, materialist, and construc- Thus, in the example, ‘the city is a metabolism’, tivist approaches found in the Marxist ecologies. the tenor is ‘the city’, and the vehicle is ‘meta- Epistemological differences clearly shape the bolism’. The generative power of metaphor is research questions posed, the processes by which through interaction or interanimation of these data are collected, and the parameters for what words and their associations. Richards (1936: counts as valid knowledge. But how does one 94) describes the process as ‘a borrowing explain the separate islands between ‘rift’ sociol- between and intercourse of thoughts, a transac- ogists and urban political ecologists despite a tion of context (rather than mere substitutions)’. 12 Progress in Human Geography

Thus, the emergent cognitive power of meta- (UPE), Kennedy (industrial ecology), and phor lies in its ability to conceptualize a murky Golubiewski (urban ecology). This section subject in a novel way. Tenor and vehicle, there- illustrates the ways in which the urban meta- fore, co-construct each other to give more var- bolism metaphor has become stagnant, due ied and powerful meaning to both (Richards, to repeated use through particular disciplin- 1936: 108). In this intercourse of thoughts, there ary constructs and situated knowledge. Table are not two subjects per se (urban and metabo- 1 summarizes the theoretical influences, lism), but rather just one (urban), and this sub- emphases, lexicon, methods/models, and cri- ject is described in lexicon appropriate to a tiques or perceived limitations of urban description of a metabolism. metabolism in the three ecologies. Lakoff and Johnson (1980) would later famously build off of Richards’ interaction theory of metaphor, replacing vehicle with 1 ‘Input–output model of the flow of things’ source domain and tenor with target domain and characterizing the metaphorical process Studies on urban metabolism have often uncriti- as ‘cross-domain mapping’, whereby a fixed cally pursued the standard industrial ecology per- set of ‘ontological correspondences’ between spective based on some input–output model of the entities in the source domain are mapped flow of ‘things.’ Such analysis merely poses the onto entities in the target domain (Lakoff, issue and fails to theorize the making of the urban 1993). Lakoff and Johnson (1980) would also as a socio-environmental metabolism. (Swynge- helpfully point out that by making a lesser douw, 2006b: 35) known domain better understood, metaphor has a utility, but the process invariably results In ‘traditional’ industrial ecology, the vehicle or in partiality. Cresswell (1997: 334) conveys source domain stems from an Odum-informed this partiality in spatialterms:‘Manymeta- ecology of metabolism as simplified biological phors are distinctly geographical acts that organism, with some internal complexity to be encourage spatial thoughts and actions while sure, but primarily organized around a series prohibiting others.’ Thus, constructions of the of inputs and outputs. This source domain selec- urban metabolism metaphor are useful but tion is influenced by a normative priority to inevitably partial, asymmetrical representa- quantify urban resource consumptions patterns, 3 tions of particular phenomena, processes, so as to reduce material ‘throughput’. As the sets of practices, and situated knowledge target domain, the city has organismic qualities (Haraway, 1988). in that it consumes resources to sustain itself, Using interaction theory of metaphor and transforms energy, and eliminates waste. The the cross-domain mapping concept, we now industrial ecologist generates and applies a analyze the source and target domains in the mass-balance model (e.g. MFA) to particular three ecologies of urban metabolism and the urban phenomena, namely by quantifying as varied epistemologies, politics, models, and many flows and stocks as available data allow. methods that undergird deployments of the The industrial ecology urban metabolism meta- metaphor. This is a discussion informed by phor is thus functional, linear, organismic and the few instances where the three thought based on an ontological construction of nature traditions have engaged each other in the lit- as distinct from society, in which nature is erature over appropriate use of the term, largely abstracted as a supply source of natural namely Syngedouw, Keil and Boudreau resources. Newell and Cousins 13

Table 1. Characteristics of the three ecologies of urban metabolism. Marxist ecologies Industrial ecology Urban ecology Metaphorical Metabolism as Metabolism as Metabolism as conception nature–society relations biological organism ecosystem Network clusters 1) Urban political ecology 1) ‘Traditional’ urban — 2) Metabolic ‘rift’ metabolism 2) ‘Vienna School’ of socio-economic metabolism Theoretical Marx; Harvey; Bellamy-Foster; Lavoisier; EP Odum; HT Complex systems influences/key Smith; Swyngedouw; Gandy; Odum; Wolman; Ayers; theory; EP Odum; figures Kaika; Keil; Heynen Kennedy; Fisher-Kowalski; Pickett; Grimm; Weisz; Schandl Alberti Emphases How dynamic nature– Quantifying ‘flows’ and Internal complexity society relationships shape ‘stocks’ of urban ecosystem outcomes, including Optimizing and reducing processes (re)production of material ‘throughout’ Sub-system inequality and rift interactions Social power Ecosystem function to inform sustainability Language Movement as networked Movement as input– Movement as circulations output feedback loops Urban as created Flows as throughput Flows as structure ecosystems Beyond the city as –function linkages Nature–society dialectic ‘hinterland’ Internal Production of nature transformations Nature–society hybridity Methods/models Historical materialism Mass-balance approaches Ecologically Qualitative approaches Material flow analysis informed complex (MFA) systems models Emergy analysis Life cycle assessment (LCA) Critiques/ Isolated scholarly network ‘Black boxing’ of urban Bounded sense of perceived ‘Methodological cityism’ processes ‘urban’ limitations Social at expense of Aspatial rendering of Apolitical ecological ‘hinterland’ Complexity at Marxist dominated Nature–society dualism expense of distal Flows instead of stocks flows Apolitical 14 Progress in Human Geography

Although the resource flows are hypothe- these flows’, and addresses ‘social factors’, such tically connected to distal production spaces as habits of consumption and modes of regula- (e.g. ‘the hinterland’), the work has been tion. Although not because of the underlying spa- largely aspatial with respect to grounding tial politics, urban ecologists also take issue with the origins of water, food, materials, and the ‘black-boxing’ of urban nature–society pro- energy flows in specific geographies. In cesses, which we turn to next. addition, by privileging the flows, the inter- nal processes within the city are ‘black- boxed’ in that the ‘stocks’ that gradually 2 City as ecosystem vs. organismic accrue in the urban space receive short metabolism shrift, as do the socio-economic and ecologi- cal processes that principally shape stock– Rather than an organismic metabolism, urban flow dynamics. Reduced to a series of input ecologists such as Golubiewski (2012a, 2012b) and outputs, the city, as Swyngedouw points propose ‘ecosystem’ – an assemblage of organ- out, is divorced from the historical material- isms interacting with the physical environment ism and power geometries perpetually within a specified area (Likens, 1992; Tansley, [re]shaping urban material and energy flows 1935) – as the appropriate metaphor to charac- over time–space. terize and model urban space. Although indus- In this conceptualization, space is treated as trial ecology and urban ecology are both rooted ‘absolute’, relegating the city to a ‘field, a con- in systems thinking, the source domain differs tainer, a coordinate system of discrete and in that the ecosystem metaphor Golubiewski mutually exclusive locations’ (Smith and Katz, proposes is informed by a complex systems 1993: 73). This apparent apolitical construction theorization in which humans are integral com- is actually highly political. As Smith and Katz ponents of an urban socio-ecological matrix (1993: 73) explain, ‘absolute space is politically (Alberti, 1999; Grimm et al., 2000). Biotic and charged’, whereby ‘a specific tyranny of power abiotic entities interact within a fairly bounded is expressed through the capitalist production of system to generate the emergent patterns and it and in which the spatial metaphor is implicit’. processes that define an urban ecosystem In UPE discourse, therefore, the industrial ecol- (Golubiewski, 2012a). ogy enterprise is framed as a naı¨ve promoter of The city is conceptualized as a relational, capitalism, and more specifically a weak form hybridized set of nature–society processes. Cit- of (Desfor and Keil, ies, writes Alberti (2008: 252), are ‘hybrid 2004; Gibbs and Deutz, 2005) – that is, phenomena’ that emerge from ‘interactions techno-managerial approaches that endorse between human and ecological processes’. These market as a means to create urban ecosystems are contingent, constantly win–win scenarios between economic growth [re]shaped by drivers, patterns and processes. and environmental sustainability, but preserve The likeness of language (‘assemblages’, ‘emer- the status quo by failing to identify or challenge gent properties’, etc.) is striking and, in stark the structural changes necessary to address the contrast with industrial ecology, it is an ontologi- underlying problem of uneven capitalist devel- cal construction of nature–society that resonates opment (Harvey, 1996; Swyngedouw, 2013). not only with UPE (Gandy, 2005; McFarlane, Thus, following Swyngedouw (2006b), Keil 2011, 2013; Swyngedouw, 2006a; Whatmore, and Boudreau (2006: 43) prescribe a political 2002), but also with co-constructivist actor–net- ecology of flows that pays attention to ‘political work theory and, to a lesser degree, with the context’, analyzes the ‘capitalist systems driving Vienna School. Newell and Cousins 15

Through ontological correspondence,‘eco- 3 Marxist urban metabolism system’ becomes the preferred metaphor, To build the initial architecture for the UPE focusing attention on a target domain of pat- urban metabolism, Swyngedouw (1996) fused tern and process of biotic and abiotic interac- traditional Marxist discourses with lexicon from tions in a city that is complex, adaptive, and actor–network theory (e.g. networks, assem- emergent. Thus, ecosystem-based urban mod- blages, cyborgs). Nevertheless, Marxist logic els focus on simulating dynamic interactions has so successfully permeated the metaphor that between these socio-ecological entities. it is hard to imagine it without this epistemolo- When operationalized empirically as a gical apparatus. The generative source domain model, the ecosystem metaphor results in a is dominated by a framing of metabolism as partiality of a different sort, one that privi- an ongoing dialectical process of transforming leges urban ecosystem complexity (and spa- the raw materials of nature into ‘things’ tially bounded processes) at the expense of through human labor and their circulation the more unbounded flows coursing through within and through urban socio-natures. Hybri- the city. dized nature–society relations of these perpe- As with industrial ecology, the urban ecology tually emerging things – whether they be water enterprise is largely devoid of political concerns mains, concrete, urban green space, or some (e.g. uneven capitalist development, equality, other object of focus – are excavated to pursue and justice), but it resonates in that its epistemo- a particular political project: namely to unveil the logical foundations in systems thinking leads to unevenness and power relations behind their pro- similar metaphorical optics politically. Despite duction and consumption under capitalism, all in the conceptual hybridity, the urban ecosystem the pursuit of a more egalitarian politics. For is nonetheless composed of identifiable and dis- some, the predominance of Marxist framing hin- crete subsystems that are quantifiable within a ders future development of the UPE field. Holi- larger systemic whole. As Robbins (2013: field (2009), for example, rejects a synthesis of 313) explains, ‘People, trees, monkeys, and car- and actor–network theory (ANT), in bon all enter a system mix’, and those nature– favor of an elaborated ANT approach when tack- society relationships that resist quantification ling questions of urban and are typically excluded (Robbins, 2012). The politics. In a related vein, Grove (2009) argues that political implications of systems metaphors are UPE extend theorizations beyond a Marxist epis- problematically reductionist in the sense that temology by including conceptual approaches the outcomes from the models can be used to offered by post-structural and feminist political reproduce a narrow technocratic agenda in sup- ecology and critical geopolitics. port of state and expert power, thereby silencing In addition to the permeation of Marxist other forms of knowledge and action (Robbins, framings, three emphases have gradually 2012). This can have disastrous consequences, imbued the metaphor and typify first-wave as Robbins (2013: 314) insightfully recognizes, UPE: 1) a privileging of the ‘social’ at the in situations where the ‘behavior of real-world expense of the ‘ecological’; 2) a methodological socio-environments’ is incongruent with poli- cityism (Angelo and Wachsmuth, 2014); and 3) cies and approaches rooted in the ‘meta- the preponderance of qualitative approaches. technics of system logic.’ First, although theorized as a hybridized Let’s now turn to urban political ecology, the socio-natural process, the focus is clearly on the third ecology of urban metabolism, and con- social and political dynamics shaping urban sider what cross-domain mapping reveals about . ‘Environment’ and ‘nature’ are its use of the metaphor. 16 Progress in Human Geography often used interchangeably with ‘ecology’ to across a large swath of UPE, the ontological foreground or provide context for an urban envi- correspondence is indeed a particular thesis ronmental politics. For example, drawing on of urban space that is contrasted with generally Latour (1993) and Haraway (1991), Loftus presumed nonurban spaces (e.g. farms, forests, (2012) describes the cyborg relationships that mines, or dams) (Brenner, 2013). This has link social and ecological processes around resulted in case studies that focus on the city water in Durbin, South Africa, but throughout proper (e.g. Cooke and Lewis, 2010; Gandy, the text the actual biophysical properties of 2002; Hagerman, 2007; Heynen et al., 2006; water are largely absent. Concerns about access Desfor and Keil, 2004), rather than how the to and struggles over water provide the context processes of urbanization occur at other sites to propose an situated in and scales, or in delineating and interrogating the everyday experience. Similarly, the urban spatiotemporal linkages between cities and the ecology of Los Angeles and Toronto explored more distal sites of resource extraction and by Desfor and Keil (2004) relies on a discursive production that help sustain them. analysis of the politics of urban environmental Third, qualitative approaches dominate most change, but neglects how the city’s biophysical UPE frames of urban metabolism, relying on ecology might co-construct the various social either some combination or individual aspects and political struggles over the environment. of discourse and document analysis, archival Other examples that explore the politics of methods, interview data, and participatory urban environments without a strong emphasis observation (e.g. Grove, 2009; Ioris, 2012; on biophysical ecology include representations Lawhon, 2013; Kooy and Bakker, 2008). Some of post-industrial ruin (Millington, 2013), the UPE does engage with quantitative approaches role of social power in shaping access to water (e.g. Heynen et al., 2006; Buzzelli, 2008; Cou- (Swyngedouw, 2004; Kaika, 2005), the politici- sins and Newell, 2015), but these studies are a zation and maintenance of urban infrastructures minority in the tradition. (Broto and Bulkeley, 2013; Domı´nguez and Fogue´, 2013) and (Quastel, 2009), among others. V Advancing urban metabolism in Second, a methodological cityism that theo- geography and beyond rizes the spatio-temporal dimensions of plane- So how might, but the urban metabolism meta- tary urbanism, but paradoxically maintains an phor be reinvigorated in geography (and the empirical focus that privileges bounded concep- broader academy)? As noted earlier, post- tualizations of the city, prevails in UPE (Angelo structural, ANT, and feminist political ecology and Wachsmuth, 2014). The city becomes rei- offer vehicles (i.e. source domains) in which to fied as a specific socio-natural artifact produced do so. But if one observes the three emphases through the metabolization of nature (Heynen, (social at the expense of the ecological, methodo- 2006; Swyngedouw and Heynen, 2003) – ironic logical cityism, and preponderance of qualitative given the spatial dimensions the metabolism approaches) as shortcomings to be addressed in a metaphor implies (Heynen et al., 2006) and the ‘second-wave’ UPE research agenda, then urban widespread understanding in UPE that globa- ecology and industrial ecology are especially lized socio-natural relationships are integral well-suited for such an effort. Indeed, growing to contemporary urbanism. Methodological numbers of scholars are calling for interdisciplin- cityism does not permeate all UPE scholarship, ary collaboration under the urban metabolism as Cronon (1991) and Gandy (2004) provide banner so as to leverage respective disciplinary two excellent examples to the contrary. But strengths associated with diverse theorizations Newell and Cousins 17 of the city (Broto et al., 2012; Pincetl, 2012; Pin- assert that ‘one can be a Marxist, postmodernist cetl et al., 2012; Kennedy et al., 2011, 2012; or post-structural critic and make use of mathe- Ramaswami et al., 2012). matics and formal theory’. Building from a shared theorization of For all the well-trumpeted benefits of inter- urbanization as hybridized socio-ecological disciplinary research (National Academies, processes, urban ecology provides an impor- 2005), however, the challenges are often under- tant grounding to explore the ‘ecology of estimated (Lau and Pasquini, 2008; Massey, cities’ and how the aggregated human and 1999; Schoenberger, 2001). Shared spaces of non-human systems sum in the urban envi- protected intellectual experiment need to be ronment (Grimm et al., 2008). Given the created to foster real engagement (Brenner dualistic formulations of nature–society rela- et al., 2011; Lele and Norgaard, 2005; Schoen- tions, the initial utility of industrial ecology berger, 2001; Wolman, 1997). Therefore, we for geographers is likely methodological. propose urban metabolism be conceptualized Specifically, quantitative approaches to mea- as a boundary metaphor, much like the concept sure stocks and flows could be used to of a boundary object. advance understandings of the networked impacts of urbanization processes on ecolo- gies, people, and places. In addition to MFA, 1 Urban metabolism as a boundary LCA offers possibilities for spatializing metaphor metabolisms, not necessary for inputs–out- Boundary objects are concepts or items (such as puts as a whole, but instead for specific policy documents or maps) that have a malle- water, food, wood, waste, and energy flows ability or ‘plasticity’ so that they ‘inhabit sev- that link (and potentially destabilize) notions eral intersecting social worlds ... and satisfy of urban and rural (Newell and Vos, 2011, the informational requirements of each of them’ 2012). The Vienna School metabolists have (Star and Griesemer, 1989: 393). By enabling long coupled the quantification of flows heterogeneity and multiple attitudes to co- using MFA with core themes of critical exist, boundary objects enable ‘collaboration . Although the underlying tele- without consensus’ on difficult epistemological ology and sweeping historical narratives of issues between disciplines (Clarke and Star, socio-economic metabolic transitions that 2008: 222). Similar to watersheds (Cohen, permeate much of this work is tendentious, 2012), resilience (Brand and Jax, 2007; Brown, the Vienna School metabolists nonetheless 2014) or conservation corridors (Goldman, demonstrate the potential of IE tools for pro- 2009), urban metabolism as a boundary meta- gressive praxis, such as the development of a phor offers the potential to bring together multi- ‘geographical politics of consumption’ ple attitudes and beliefs around a shared and (Hartwick, 2000). recognizable concept, while allowing the meta- Recent calls in geography for epistemologi- phor to maintain its particular meaning in each cal and methodological pluralism and bridging social world or island of disciplinary practice. qualitative–quantitative divides are helpful Although boundary metaphors (and objects) when considering such collaboration (Barnes are meant to enable translation and communica- and Sheppard, 2010; Sui and DeLyser, 2011). tion across thought traditions, their interpretive Geographers are combining diverse intellectual flexibility requires a focus on a particular object traditions, such as mathematical theory and crit- of study to retain some degree of coherence of the ical human geography (Bergmann et al., 2009; metaphor across scholarly communities. Socio- Bergmann, 2013), leading Barnes (2010: 3) to technical infrastructures such as water and energy 18 Progress in Human Geography systems and highways and streets are a material set by water source and life cycle phase (Newell of objects with which research paradigms might and Vos, 2011, 2012). Recommendations would fruitfully interact. Explored as a series of shared then follow, such as a shift away from high- empirical experiments, these objects of study carbon water sources and technological fixes could provide the conditions for a useful ‘friction’ to make water supply more efficient. As dis- between often contested and unequal interactions cussed earlier, this traditional industrial ecology across epistemological differences (Tsing, 2005). approach ‘black boxes’ the city, both in terms of the lack of areal differentiation in the LCA mod- eling used to quantify the carbon footprint and 2 An urban political–industrial ecology: The in rendering opaque the urban processes that case of water drive these flows. Water provides an excellent initial object of As an antidote, we crafted an urban political– study for deploying urban metabolism as a industrial ecology approach to analyze the water boundary metaphor. In contrast to globalized supply metabolism of Los Angeles (Cousins product commodities, urban water flows are and Newell, 2015). To better delineate spatially bounded by proximate ecosystems, the supply system, including ‘downscaling’ providing a definite set of hydrological systems to ascertain the specific grid mixes that drive for urban ecologists to study and a relatively the pumping and treatment plants along constricted system boundary for industrial ecol- the water network, we coupled ogist to locate (and quantify) flow origins. LCA with GIS. This intervention led to two Water is also the most well-studied flow in UPE major outcomes. First, it unveiled how aspatial (Gandy, 2002; Kaika, 2005; Desfor and Keil, assumptions about the system boundary, activity 2004; Swyngedouw, 2004), thereby leveraging data, and emissions factors embedded in con- a rich body of work, such as emergent research ventional urban water metabolism modeling on the ‘hydro-social cycle’, which recognizes can fundamentally shape the end result. The the circulations of these flows as dependent outcome provides a means of opening up the upon the hydrological cycle as well as institu- black box of the carbon modeling, measure- tions, infrastructures, and social practices ment, and calculation processes that drive (Bakker, 2003, 2005, 2013; Budds, 2009, urban climate governance. Second, it makes 2013; March, 2013; Swyngedouw, 2009). To LCA more political by revealing the uneven give the reader a sense of how collaboration spatiality of water supply burdens (and car- might proceed, we expand on our own efforts bon emissions) across demographics, along to develop a political–industrial ecology using supply chains, and among resource users. urban water supply as the object of focus. These results were then linked to a broader In a typical study of the urban water metabo- qualitative analysis of the socio-economic and lism, the industrial ecologist would first deline- political factors that shape how geographic ate the system boundary by identifying the life complexity is scoped in the production and cycle phases (sourcing, conveying, treating, application of industrial ecology approaches, etc.) of water supply sources (e.g. imported, thereby providing important insights into its groundwater, recycled) to be quantified using political role in (re)configuring urban water MFA or life cycle assessment (LCA). This metabolisms and the production of science would include selection of ‘activity data’ and (Forsyth, 2003). The metrics used to measure ‘emissions factors’, usually based on regional urban water metabolisms support urban infra- or national averages, to quantify the carbon structural transitions that are embedded within footprint of the supply system, broken down a broader political economy and reflect the Newell and Cousins 19 social histories of place, discourses of develop- islands or ‘ecologies’ of urban metabolism have ment, and the networks of power that solidify emerged, each with their own framings and urban water supply networks (Cousins and understandings, this does not pose an insurmoun- Newell, 2015). Linking these complex socio- table barrier to cooperative interdisciplinary political issues to a quantitative measure work. That the metaphor is employed by scholars required a careful analysis of the discursive with different epistemological and disciplinary strategies, both historical and contemporary, commitments highlights its utility in achieving employed to build social consensus so as to multiple and, likely, contradictory objectives. facilitate the transformations of urban water Similar to traditional formulations of boundary metabolisms (Kaika, 2003) and to deflect objects, the extent of future engagement among attention away from potential issues of social scholarly communities will depend on the infor- justice, land distribution, and the environment mational needs found outside of their traditional to benefit elite actors. For example, the legacy discipline and the specific work requirements of of the Los Angeles Aqueduct continues to those collaborating in an interdisciplinary envi- shape water politics in both Los Angeles ronment (Star and Griesemer, 1989). Above all, and California as a whole through regulatory these engagements require moving beyond disci- drivers that reallocate water for environmental plinary boundaries to embrace multiplicity and a mitigation to control dust on the dried up portions spirit of openness, patience, and even periodic of Owens Lake. Yet other strategic new para- compromise to enable geographers to better cap- digms have emerged in the Los Angeles , ture what Thrift (2002: 297) recognizes as the from stormwater capture to water , ‘myriad geographies perpetually emerging’. which impact socio-ecological systems while reshaping urban carbon emissions in potentially contradictory ways. VI Conclusion This coupling of urban metabolism approaches The metabolism metaphor reflects particular provides a means for the field of UPE to move lenses through which urban space is compre- beyond a ‘methodological cityism’ by addressing hended. Through bibliometric analysis and liter- the networked impacts of urban metabolisms on ature review, we have mapped out the evolution ecologies, peoples, and places both inside and out- and crystallization of three islands of urban side of the ‘city’. The approach infuses UPE with a metabolism: industrial ecology, Marxist ecolo- new set of quantitative methods (e.g. MFA, LCA) gies, and urban ecology. Industrial ecology to calculate and map out the uneven spatiality of focuses on quantifying material and energy environmental and social impacts associated with flowing into, within, and out of cities, using production and consumption. From the standpoint accounting methodologies such as MFA. Urban of industrial ecology, a geographically informed ecology models the complex patterns and pro- analysis that takes seriously the spatiality and cesses of urban human–ecological systems. political ecology of GHG emissions destabilizes Marxist ecologies such as UPE emphasize the apolitical acceptance of technological adop- power relations that constitute uneven socio- tion and innovation to fix the carbon emissions ecological production. problem. Interaction and cross-domain mapping theories This is just a brief example of how utilizing of metaphor have unveiled how epistemological, urban metabolism as a boundary metaphor offers methodological, and political commitments and the potential to advance the concept within geo- normative priorities shape the urban metabolism graphy and to extend the discipline’s influence to of the three ecologies accordingly. Each privileges other scholarly communities. Although three a particular dimension of urban space, while 20 Progress in Human Geography excluding others. This is endemic to the generative metaphor, a metabolism of the urban ecosystem, process of metaphors, as is their canonization one that blurs boundaries by blending elements through habitual use, at which point they often of the three ecologies and could be defined as fol- stagnate, their generative power greatly reduced. lows: a global circulatory process of socio-natural First-wave UPE urban metabolism has reached relations that transforms and (re)creates urban this stage, simultaneously empowered and limited ecosystems through the exchange of resources, by a particular Marxist apparatus that privileges capital, humans, and non-humans into and out socio-political dynamics and by methodological of the spaces of global urbanization. How exactly approaches that are largely qualitative and this metaphor would take shape hinges on those ‘cityist’. who would collectively build it. To remain alive and vibrant, metaphors need to be continually infused with creativity, namely Acknowledgements through the circulation of new thoughts, ideas, and applications. Interdisciplinary engagement The authors would like to thank Christopher Ken- can reinvigorate the urban metabolism metaphor nedy and Stephanie Pincetl, three anonymous reviewers, and editor Noel Castree for comments as part of a ‘second wave’ of UPE that creatively and insights that greatly improved this essay. The articulates the emergent socio-natures that (re)de- research was financially supported by the Univer- fine urban space and that develops the political sity of Michigan’s M-Cubed program, which fos- and methodological means necessary to produce ters interdisciplinary scholarship, and the John more sustainable and just urban worlds. As the Randolph and Dora Haynes Foundation. water supply case illustrates, collaboration with industrial ecology offers a means to spatialize and interrogate the carbon calculus that drives urban Notes climate governance, while simultaneouslyadvan- 1. Unlike the ‘three ecologies’ of urban metabolism iden- cing efforts to quantify the uneven material bur- tified by Wachsmuth (2012), our analysis excludes the dens associated with urbanization processes. ‘’ of Burgess and the Chicago School of Crafting this urban political–industrial ecology Sociologists because the bibliometric analysis indicates that it is no longer influential among contemporary entails an importing of method and an exporting metabolism formulations. Instead, we identify the third of spatial sensitivity and critical political econ- ecology as urban ecology, whose influence is evident– omy. Although beyond the scope of our water based on the citation analysis. case, complex-systems urban ecology would 2. Marx (1976: 283) uses metabolism to define the labor enrich this effort by offering a relational view of process as a relationship ‘between man and nature, a socio-ecological systems grounded by knowl- process by which man, through his own actions, med- edge of ecological processes. Scrutiny of urban iates, regulates and controls the metabolism between metabolism as a spatial metaphor vis-a`-vis ecolo- himself and nature’. gical theory accompanies this understanding. 3. Reducing the material and energy burdens associated Geography has long been rightly suspicious of with resource consumption is a research focus that per- naturalistic and organismic metaphors, especially vades industrial ecology, beyond work on urban metabo- when applied to cities, as their use brings to mind lism. See, for example, work by Lugschitz et al. (2011), Konar et al. (2011), and Wiedmann et al. (2013). the tainted past of their misapplication (e.g. Bur- gess, 1925; Park, 1936). Through this engagement, a richer, deeper, References more inclusive, and yet still politically engaged Alberti M (1999) Modeling the urban ecosystem: A con- metaphorical conceptualization of urban space ceptual framework. Environment and Planning B: can emerge. This may even take form as a new Planning and Design 26(4): 605–630. Newell and Cousins 21

Alberti M (2008) Advances in Urban Ecology: Integrating Barnes TJ (2010) Taking the pulse of the dead: History and Humans and Ecological Processes in Urban Ecosys- , 2008–2009. Progress in tems. New York: Springer. Human Geography 34(5): 668–677. Angelo H and Wachsmuth D (2014) Urbanizing urban Barnes TJ and Curry MR (1992) in eco- political ecology: A critique of methodological cityism. nomic geography: Metaphor and the construction of International Journal of Urban and Regional alterity. Environment and Planning D: Society and Research. DOI: 10.1111/1468-2427.12105. Space 10(1): 57–68. Ayres RU (1989) Industrial metabolism. In: Technology Barnes TJ and Sheppard E (2010) ‘Nothing includes every- and Environment. Washington, DC: National Academy thing’: Towards engaged pluralism in Anglophone eco- Press, 23–49. nomic geography. Progress in Human Geography Ayres RU (1994) Industrial Metabolism: Restructuring for 34(2): 193–214. Sustainable Development. Tokyo: United Nations Uni- Bergmann L (2013) Bound by chains of carbon: Ecologi- versity Press. cal–economic geographies of globalization. Annals of Ayres RU and Kneese AV (1969) Production, consump- the Association of American Geographers 103(6): tion, and externalities. The American Economic Review 1348–1370. 59(3): 282–297. Bergmann L, Sheppard E and Plummer PS (2009) Capital- Baccini P (1996) Understanding regional metabolism for a ism beyond harmonious equilibrium: Mathematics as if sustainable development of urban systems. Journal of human agency mattered. Environment and Planning A Urban Technology 4(2): 27–39. 41(2): 265–283. Baccini P (1997) A city’s metabolism: Towards the sus- Black M (1962) Models and Metaphors: Studies in Lan- tainable development of urban systems. Journal of guage and Philosophy. Ithaca NY: Cornell University Urban Technology 4(2): 27–39. Press. Baccini P and Brunner PH (1991) Metabolism of the Brand FS and Jax K (2007) Focusing the meaning(s) Anthroposphere. Berlin: Springer. of resilience: Resilience as a descriptive concept Bakker K (2003) Archipelagos and networks: Urbaniza- and a boundary object. Ecology and Society tion and water privatization in the South. The Geogra- 12(1): 23. phical Journal 169(4): 328–341. Brenner N (2013) Theses on urbanization. Public Culture Bakker K (2005) Neoliberalizing nature? Market environ- 25(1/69): 85–114. mentalism in water supply in England and Wales. Brenner N, Madden DJ and Wachsmuth D (2011) Assem- Annals of the Association of American Geographers blage urbanism and the challenges of critical urban the- 95(3): 542–565. ory. City 15(2): 225–240. Bakker K (2013) Constructing ‘public’ water: The World Broto VC, Allen A and Rapoport E (2012) Interdisciplin- Bank, urban water supply, and the biopolitics of devel- ary perspectives on urban metabolism. Journal of opment. Environment and Planning D: Society and Industrial Ecology 16(6): 851–861. Space 31(2): 280–300. Broto VC and Bulkeley H (2013) Maintaining climate Barles S (2007a) Feeding the city: Food consumption and change experiments: Urban political ecology and the flow of nitrogen, Paris, 1801–1914. The Science of the everyday reconfiguration of urban infrastructure. Inter- Total Environment 375(1–3): 48–58. national Journal of Urban and Regional Research Barles S (2007b) Urban metabolism and river systems: An 37(6): 1934–1948. historical perspective – Paris and the Seine, 1790– Brown K (2014) Global environmental change I: A social 1970. and Earth System Sciences Discus- turn for resilience? Progress in Human Geography sions 4(3): 1845–1878. 38(1): 107–117. Barles S (2009) Urban metabolism of Paris and its Budds J (2009) Contested H2O: Science, policy and poli- region. Journal of Industrial Ecology 13(6): tics in water resources management in Chile. Geoforum 898–913. 40(3): 418–430. Barnes TJ (1996) Logics of Dislocation: Models, Meta- Budds J (2013) Water, power, and the production of phors, and Meanings of Economic Space. New York: neoliberalism in Chile, 1973–2005. Environment Guilford Press. and Planning D: Society and Space 31(2): 301–318. 22 Progress in Human Geography

Burgess EW (1925) The growth of the city: An introduc- ecosystem. Annual Review of Energy and Environment tion to a research project. Publications of the American 25: 685–740. Sociological Society XVIII: 85–97. Desfor G and Keil R (2004) Nature and the City: Making Burstro¨m F, Brandt N, Frostell B and Mohlander U (1997) in Toronto and Los Angeles. and information for environ- Tucson: University of Arizona Press. mental policies in the city of Stockholm. In: ConAccount Domı´nguez RF and Fogue´ U (2013) Technifying public Conference, Parallel Session C: MFA for Regional and space and publicizing infrastructures: Exploring new Local Materials, 11–12 September, Wuppertal, Germany. urban political ecologies through the Square of General Buzzelli M (2008) A political ecology of scale in urban air Vara del Rey. International Journal of Urban and monitoring. Transactions of the Institute of Regional Research 37(3): 1035–1052. British Geographers 33(4): 502–517. Fischer-Kowalski M (1998) Society’s metabolism: The Chertow MR (2000) : Literature and intellectual history of materials flow analysis, Part I, taxonomy. Annual Review of Energy and the Environ- 1860–1970. Journal of Industrial Ecology 2(1): 61–78. ment 25(1): 313–337. Fischer-Kowalski M and Hu¨ttler W (1998) Society’s meta- Chertow MR (2008) ‘Uncovering’ industrial symbiosis. bolism. Journal of Industrial Ecology 2(4): 107–136. Journal of Industrial Ecology 11(1): 11–30. Fischer-Kowalski M and Weisz H (1999) Society as hybrid Clarke AE and Star SL (2008) The social worlds frame- between material and symbolic realms: Toward a theore- work: A theory/methods package. In: Hackett E, tical framework of society-nature interaction. Advances Amsterdamska O, Lynch M and Wajcman J (eds) in Human Ecology 8: 215–251. The Handbook of Science and Technology Studies. Forsyth T (2003) Critical Political Ecology: The Politics Cambridge, MA: MIT Press, 113–37. of . New York: Routledge. Clausen R and Clark B (2005) The metabolic rift and Foster JB (1999) Marx’s theory of metabolic rift: Classical marine ecology: An analysis of the ocean crisis within foundations for . American capitalist production. Organization and Environment Journal of Sociology 105(2): 366–405. 18(4): 422–444. Foster JB (2000) Marx’s Ecology: Materialism and Cohen A (2012) Rescaling : Nature. New York: Monthly Review Press. Watersheds as boundary objects at the intersection of Frosch RA (1992) Industrial ecology: A philosophical science, neoliberalism, and participation. Environment introduction. Proceedings of the National Academy of and Planning A 44(9): 2207–2224. Sciences 89(3): 800–803. Cooke J and Lewis R (2010) The nature of circulation: Gandy M (2002) Concrete and Clay: Reworking Nature in The urban political ecology of Chicago’s Michigan New York City. Cambridge, MA: MIT Press. Avenue Bridge, 1909–1930. Gandy M (2004) Rethinking urban metabolism: Water, 31(3): 348–368. space and the modern city. City 8(3): 363–379. Cousins JC and Newell JP (2015) A political–industrial Gandy M (2005) Cyborg urbanization: Complexity and ecology of water supply infrastructure for Los Angeles. monstrosity in the contemporary city. International Geoforum. DOI:10.1016/j.geoforum.2014.10.011. Journal of Urban and Regional Research 29(1): 26–49. Cresswell T (1997) Weeds, plagues, and bodily secretions: Gibbs D and Deutz P (2005) Implementing industrial ecol- A geographical interpretation of metaphors of displace- ogy? Planning for eco-industrial parks in the USA. ment. Annals of the Association of American Geogra- Geoforum 36(4): 452–464. phers 87(2): 330–345. Girardet H (1992) The Gaia Atlas of Cities: New Directions Cronon W (1991) Nature’s : Chicago and the for Sustainable Urban Living. London: Gaia Books. Great West. New York: W.W. Norton. Goldman M (2009) Constructing connectivity: Conserva- Cusso´ X, Garrabou R and Tello E (2006) Social metabo- tion corridors and conservation politics in East African lism in an agrarian region of Catalonia (Spain) in rangelands. Annals of the Association of American 1860–1870: Flows, energy balance and land use. Eco- Geographers 99(2): 335–359. logical Economics 58(1): 49–65. Golubiewski N (2012a) Is there a metabolism of an Decker EH, Elliott S, Smith FA, Blake DR and Rowland urban ecosystem? An ecological critique. Ambio FS (2000) Energy and material flow through the urban 41(7): 751–764. Newell and Cousins 23

Golubiewski N (2012b) The power of language in feed- and the Politics of Urban Metabolism. New York: back metaphors: A response to Kennedy. Ambio Routledge, 1–20. 41(7): 767–768. Heynen N, Perkins HA and Roy P (2006) The political Grimm NB, Faeth SH, Golubiewski NE, Redman CL, Wu ecology of uneven urban green space: The impact of J, Bai X and Briggs JM (2008) Global change and the political economy on race and ethnicity in producing ecology of cities. Science 319(5864): 756–760. environmental inequality in Milwaukee. Urban Affairs Grimm NB, Grove JM, Pickett STA and Redman C (2000) Review 42(1): 3–25. Integrated approaches to long-term studies of urban Holifield R (2009) Actor-network theory as a critical ecological systems. BioScience 50(7): 571–584. approach to environmental justice: A case against Grove K (2009) Rethinking the nature of urban environ- synthesis with urban political ecology. Antipode mental politics: Security, subjectivity, and the non- 41(4): 637–658. human. Geoforum 40(2): 207–216. Huang SL (1998) Urban ecosystems, energetic hierarchies, Haberl H (2006) The global socioeconomic energetic meta- and of Taipei metropolis. Jour- bolism as a sustainability problem. Energy 31(1): 87–99. nal of Environmental Management 52(1): 39–51. Hagerman C (2007) Shaping neighborhoods and nature: Huang SL and Hsu WL (2003) Materials flow analysis and Urban political ecologies of urban waterfront transfor- emergy evaluation of Taipei’s urban construction. mations in Portland, Oregon. Cities 24(4): 285–297. Landscape and Urban Planning 63(2): 61–74. Hanya T and Ambe Y (1976) A study on the metabolism of Ioris AAR (2012) Applying the strategic-relational cities. In: Science for a Better Environment. Tokyo: approach to urban political ecology: The water man- Science Council of Japan. agement problems of the Baixada Fluminense, Rio de Haraway DJ (1988) Situated knowledges: The science Janeiro, Brazil. Antipode 44(1): 122–150. question in and the privilege of partial per- Jelinski LW, Graedel TE, Laudise RA, McCall DW and spective. Feminist Studies 14(3): 575–599. Patel CK (1992) Industrial ecology: Concepts and Haraway DJ (1991) Simians, Cyborgs, and Women: The approaches. Proceedings of the National Academy of Reinvention of Nature. New York: Routledge. Sciences 89(3): 793–797. Harvey D (1996) Justice, Nature, and the Geography of Kaika M (2003) Constructing scarcity and sensationalising Difference. Malden, MA: Blackwell. water politics: 170 days that shook Athens. Antipode Hartwick ER (2000) Towards a geographical politics of 35(5): 919–954. consumption. Environment and Planning A 32(7): Kaika M (2005) City of Flows: Modernity, Nature, and the 1177–1192. City. New York: Routledge. Hendriks C, Obernosterer R, Mu¨ller D, Kytzia S, Baccini P Kaika M and Swyngedouw E (2000) Fetishizing the mod- and Brunner PH (2000) Material flow analysis: A tool ern city: The phantasmagoria of urban technological to support environmental policy decision making: networks. International Journal of Urban and Regional Case-studies on the city of Vienna and the Swiss low- Research 24(1): 120–138. lands. Local Environment 5(3): 311–328. Keil R and Boudreau JA (2006) Metropolitics and meta- Hess G (2010) The ecosystem: Model or metaphor? Jour- bolics: Rolling out environmentalism in Toronto. In: nal of Industrial Ecology 14(2): 270–285. Heynen N, Kaika M and Swyngedouw E (eds) In the Heynen N (2006) Green urban political ecologies: Nature of Cities: Urban Political Ecology and the Poli- Toward a better understanding of inner-city environ- tics of Urban Metabolism. New York: Routledge, 41–62. mental change. Environment and Planning A 38(3): Keil R and Desfor G (2003) Ecological modernisation in 499–516. Los Angeles and Toronto. Local Environment 8(1): Heynen N (2014) Urban political ecology I: The urban 27–44. century. Progress in Human Geography 38(4): Kennedy C, Baker L, Dhakal S and Ramaswami A (2012) 598–604. Sustainable urban systems. Journal of Industrial Ecol- Heynen N, Kaika M and Swyngedouw E (2006) Urban ogy 16(6): 775–779. political ecology: Politicizing the production of urban Kennedy C, Cuddihy J and Engel-Yan J (2007) The chang- natures. In: Heynen N, Kaika M and Swyngedouw E ing metabolism of cities. Journal of Industrial Ecology (eds) In the Nature of Cities: Urban Political Ecology 11(2): 43–59. 24 Progress in Human Geography

Kennedy C, Pincetl S and Bunje P (2011) The study of urban Lugschitz B, Bruckner M and Giljum S (2011) Europe’s metabolism and its applications to urban planning and global land demand. A study on the actual land embo- design. Environmental Pollution 159(8–9): 1965–1973. die in European imports and exports of agricultural Kessler MM (1963) Bibliographic coupling between sci- and products. Vienna: Sustainable Europe entific papers. American Documentation 14(1): 10–25. Research Institute. Konar M, Dalin C, Suweis S, Hanasaki N, Rinaldo A and Luhmann N (1984) Soziale systeme. Frankfurt am Main: Rodriguez-Iturbe I (2011) Water for food: The global Suhrkamp. virtual water trade network. Water Resources Research March H (2013) Taming, controlling and metabolizing 47 W05520. DOI:10.1029/2010WR010307. flows: Water and the urbanization process of Barce- Kooy M and Bakker K (2008) Splintered networks: The lona and Madrid (1850–2012). European Urban and colonial and contemporary waters of Jakarta. Geoforum Regional Studies. DOI: 10.1177/0969776412474665. 39(6): 1843–1858. Marvin S and Medd W (2006) Metabolisms of obecity: Krausmann F and Haberl H (2002) The process of industria- Flows of fat through bodies, cities, and sewers. Envi- lization from the perspective of energetic metabolism: ronment and Planning A 38(2): 313–324. Socioeconomic energy flows in Austria 1830–1995. Marx K (1976) Capital, Vol. 1. New York: Vintage. Ecological Economics 41: 177–201. Marx K (1981) Capital, Vol. 3. New York: Vintage. Krausmann F, Haberl H, Schulz NB, Erb KH, Darge E and Massey D (1999) Negotiating disciplinary boundaries. Gaube V (2003) Land-use change and socio-economic Current Sociology 47(4): 5–12. metabolism in Austria: Part I: Driving forces of land- McClintock N (2010) Why farm the city? Theorizing urban use change, 1950–1995. Land Use Policy 20: 1–20. agriculture through a lens of metabolic rift. Cambridge Lakoff G (1993) The contemporary theory of metaphor. In: Journal of Regions, Economy and Society 3(2): 191–207. Ortony A (ed.) Metaphor and Thought, 2nd edn. New McFarlane C (2011) The city as assemblage: Dwelling and York: Cambridge University Press. urban space. Environment and Planning D: Society and Lakoff G and Johnson M (1980) Metaphors We Live By. Space 29(4): 649–671. Chicago: University of Chicago Press. McFarlane C (2013) Metabolic inequalities in Mumbai. Latour B (1993) We Have Never Been Modern. Cam- City 17(4): 498–503. bridge, MA: Harvard University Press. Millington N (2013) Post-industrial imaginaries: Nature, Lau L and Pasquini M (2008) ‘Jack of all trades’? The representation and ruin in Detroit, Michigan. Interna- negotiation of interdisciplinarity within geography. tional Journal of Urban and Regional Research Geoforum 39(2): 552–560. 37(1): 279–296. Lavoisier AL (1789) Traite elemntaire de chimie. In: Pre- Moleschott J (1852) Der kreislauf des lebens. Mainz: Von sente dans un Ordre Nouveau et d’apres les Decou- Zabern. vertes Moderns. Paris: Chez Chuchet. Monstadt J (2009) Conceptualizing the political ecology of Lawhon M (2013) Flows, friction and the sociomaterial urban infrastructures: Insights from technology and metabolization of alcohol. Antipode 45(3): 681–701. urban studies. Environment and Planning A 41(8): Lele S and Norgaard RB (2005) Practicing interdiscipli- 1924–1942. narity. BioScience 55(11): 967. Moore JW (2000) Environmental crises and the metabolic Lewis PA (1996) Metaphor and critical realism. Review of rift in world-historical perspective. Organization and Social Economy 54(4): 487–506. Environment 13(2): 123–157. Likens GE (1992) The ecosystem approach: Its use and Morris SA and Van der Veer Martens B (2008) Mapping abuse. In: Kinne O (ed.) Excellence in Ecology. Olden- research specialties. Annual Review of Information Sci- dorf: Ecology Institute, 166. ence and Technology 42(1): 213–295. Lin Z (2010) Kenzo Tange and the Metabolist Move- National Academies (2005) Facilitating Interdisciplinary ment: Urban Utopias of Modern Japan.NewYork: Research. Washington, DC: National Academies Press. Routledge. Newcombe K, Kalma JD and Aston AR (1978) The metabo- Loftus A (2012) Everyday Environmentalism: Creating an lism of a city: The case of Hong Kong. Ambio 7(1): 3–15. Urban Political Ecology. Minneapolis, MN: University Newell JP and Vos RO (2011) ‘Papering’ over space and of Minnesota Press. place: Product carbon footprint modeling in the global Newell and Cousins 25

paper industry. Annals of the Association of American Pincetl S (2012) Nature, urban development and sus- Geographers 101(4): 730–741. tainability: What new elements are needed for a Newell JP and Vos RO (2012) Accounting for forest car- more comprehensive understanding? Cities 29(2): bon pool dynamics in product carbon footprints: Chal- S32–S37. lenges and opportunities. Environmental Impact Pincetl S, Bunje P and Holmes T (2012) An expanded Assessment Review 37: 23–36. urban metabolism method: Toward a systems approach Newman PW (1999) Sustainability and cities: Extending for assessing urban energy processes and causes. Land- the metabolism model. Landscape and Urban Planning scape and Urban Planning 107(3): 193–202. 44(4): 219–226. Purcell M (2003) Islands of practice and the Marston/Bren- Ngo NS and Pataki DE (2008) The energy and mass bal- ner debate: Toward a more synthetic critical human ance of Los Angeles County. Urban Ecosystems geography. Progress in Human Geography 27(3): 11(2): 121–139. 317–332. Niza S, Rosado L and Ferra˜o P (2009) Urban metabolism. Quastel N (2009) Political ecologies of gentrification. Journal of Industrial Ecology 13(3): 384–405. Urban Geography 30(7): 694–725. Odum EP (1953) Fundamentals of Ecology. Philadelphia: Ramaswami A, Weible C, Main D, Heikkila T, Siddiki S, Saunders. Duvall A, Pattison A and Bernard M (2012) A social- Odum EP (1968) Energy flow in ecosystems: A historical ecological-infrastructural systems framework for inter- review. American Zoologist 8(1): 11–18. disciplinary study of systems. Journal Odum EP (1969) The strategy of ecosystem development. of Industrial Ecology 16(6): 801–813. Science 164(3877): 262–270. Richards IA (1936) The Philosophy of Rhetoric. Oxford: Odum HT (1983) System Ecology: An Introduction. New Oxford University Press. York: Wiley-Interscience. Robbins P (2012) Political Ecology: A Critical Introduc- Odum HT (1996) : Emergy and tion. 2nd edn. Malden, MA: Wiley-Blackwell. Environmental Decision Making. New York: John Robbins P (2013) Choosing metaphors for the anthropo- Wiley & Sons. cene: Cultural and political ecologies. In: The Wiley- Pajek (2012) Pajek: Program for large network analysis. Blackwell Companion to . Chiche- Available at: http://pajek.imfm.si/doku.php?id¼pajek. ster: John Wiley & Sons. Park RE (1936) Human ecology. American Journal of Sahely HR, Dudding S and Kennedy C (2003) Estimating Sociology 42(1): 1–15. the urban metabolism of Canadian cities: Greater Tor- Persson O (2010) Identifying research themes with onto area case study. Canadian Journal of Civil Engi- weighted direct citation links. Journal of Informetrics neering 30(2): 468–483. 4(3): 415–422. Sanctorius S (1712 [1614]) Medicina Statica: Being the Persson O, Danell R and Wilbor Schneider J (2009) How Aphorismi of Sanctorius, trans. Quincy J. London: to use Bibexcel for various types of bibliometric anal- William Newton. ysis. In: Astrom F, Danell R, Larsen B and Schneider Schandl H and Schulz N (2002) Changes in the United JW (eds) Celebrating Scholarly Communication Stud- Kingdom’s natural relations in terms of society’s meta- ies: A Festschrift for Olle Persson at his 60th Birthday. bolism and land-use from 1850 to the present day. Eco- Leuven: International Society for Scientometrics and logical Economics 41: 203–221. Informetrics, 9–24 Available at: http://www8.umu.se/ Schoenberger E (2001) Interdisciplinarity and social inforsk/Bibexcel/. power. Progress in Human Geography 25(3): Persson O and Ellega˚rd K (2012) Torsten Ha¨gerstrand in 365–382. the citation time web. The Professional Schulz NB (2007) The direct material inputs into Singa- 64(2): 250–261. pore’s development. Journal of Industrial Ecology Pickett STA, Cadenasso ML and Grove JM (2004) 11(2): 117–131. Resilient cities: Meaning, models, and metaphor for Schwann T (1839) Mikroskopische untersuchungen u¨ber integrating the ecological, socio-economic, and die uebereinstimmung in der struktur und dem planning realms. Landscape and Urban Planning wachsthum der thiere und pflanzen. Berlin: San- 69(4): 369–384. der’sche buchhandlung. 26 Progress in Human Geography

Schwann T (1847) Microscopic Investigations on the Swyngedouw E (eds) In the Nature of Cities: Urban Accordance in the Structure and Growth of Plants and Political Ecology and the Politics of Urban Metabo- Animals. London: Syndenham Society. lism. New York: Routledge, 21–40. Sheppard E (2010) Geographical political economy. Jour- Swyngedouw E (2009) The political economy and nal of 11(2): 319–331. political ecology of the hydro-social cycle. Journal Shibata N, Kajikawa Y, Takeda Y and Matsushima K of Contemporary Water Research and Education (2009) Comparative study on methods of detecting 142(1): 56–60. research fronts using different types of citation. Journal Swyngedouw E (2013) Into the sea: Desalination as hydro- of the American Society for Information Science and social fix in Spain. Annals of the Association of Technology 60(3): 571–580. American Geographers 103(2): 261–270. Small H (1973) Co-citation in the scientific literature: A Swyngedouw E and Heynen NC (2003) Urban political new measure of the relationship between two docu- ecology, justice and the politics of scale. Antipode ments. Journal of the American Society for Information 35(5): 898–918. Science 24(4): 265–269. Tansley AG (1935) The use and abuse of vegetational con- Small H (1999) Visualizing science by citation mapping. cepts and terms. Ecology 16(3): 284–307. Journal of the American Society for Information Sci- Thrift N (2002) The future of geography. Geoforum 33(3): ence 50(1973): 799–813. 291–298. Smith N (2008) Uneven Development: Nature, Capital, Tsing AL (2005) Friction: An Ethnography of Global and the Production of Space, 3rd edn. Athens, GA: Connection. Princeton, NJ: Princeton University University of Georgia Press. Press. Smith N and Katz C (1993) Grounding metaphor: von Liebig J (1842) Animal or Organic Chem- Towards a spatialized politics. In: Pile S and Keith istry in its Application to Physiology and Pathology. M(eds)Place and Politics of Identity.NewYork: New York: Johnson Reprint. Routledge, 66–81. von Liebig J (1859) Letters on Modern Agriculture. Stanhill G (1976) An urban agro-ecosystem: The example London: Walton and Maberly. of nineteenth-century Paris. Agro-Ecosystems 3: Wachsmuth D (2012) Three ecologies: Urban metabolism 269–284. and the society-nature opposition. The Sociological Star SL (2010) This is not a boundary object: Reflections Quarterly 53(4): 506–523. on the origin of a concept. Science, Technology & Wackernagel M and Rees WE (1996) Our Ecological Human Values 35(5): 601–617. Footprint: Reducing Human Impact on the Earth. Gab- Star SL and Griesemer JR (1989) Institutional ecology, riola Island, Canada: New Society Publishers. ‘translations’ and boundary objects: Amateurs and pro- Warren-Rhodes K and Koenig A (2001) Escalating trends fessionals in Berkeley’s Museum of Vertebrate Zoology, in the urban metabolism of Hong Kong: 1971–1997. 1907–39. Social Studies of Science 19(3): 387–420. Ambio 30(7): 429–438. Sui D and DeLyser D (2011) Crossing the qualitative- Weisz H (2011) The probability of the improbable: quantitative chasm I: Hybrid geographies, the spatial Society-nature coevolution. Geografiska Annaler: turn, and volunteered geographic information (VGI). Series B, Human Geography 93(4): 325–336. Progress in Human Geography 36(1): 111–124. Weisz H and Clark E (2011) Society-nature coevolution: Swyngedouw E (1996) The city as a hybrid: On nature, Interdisciplinary concept for sustainability. Geogra- society and cyborg urbanization. Capitalism Nature fiska Annaler: Series B, Human Geography 93(4): Socialism 7(2): 65–80. 281–287. Swyngedouw E (2004) Social Power and the Urbanization of Whatmore S (2002) Hybrid Geographies: Natures, Cul- Water: Flows of Power. Oxford: Oxford University Press. tures, Spaces. London: SAGE. Swyngedouw E (2006a) Circulations and metabolisms: Wiedmann TO, Schandl H, Lenzen M, Moran D, Suh S, (Hybrid) natures and (cyborg) cities. Science as Cul- West J and Kanemoto K (2013) The material footprint ture 15(2): 105–121. of nations. Proceedings of the National Academy of Swyngedouw E (2006b) Metabolic urbanization: The Sciences of the United States of America. DOI: 10. making of cyborg cities. In: Heynen N, Kaika M and 1073/pnas.1220362110. Newell and Cousins 27

Wittman H (2009) Reworking the metabolic rift: La Vı´a Zhang Y, Yang Z, Liu G and Yu X (2011) Emergy analysis Campesina, agrarian citizenship, and food sovereignty. of the urban metabolism of Beijing. Ecological Model- Journal of Peasant Studies 36(4): 805–826. ing 222(14): 2377–2384. Wolman A (1965) The metabolism of cities. Scientific Zucchetto J (1975) Energy-economic theory and mathe- American 213(3): 179–190. matical models for combining the systems of man and Wolman MG (1977) Interdisciplinary education: A con- nature, case study: The urban region of Miami, Florida. tinuing experiment. Science 198(4319): 800–804. Ecological Modeling 1(4): 241–268.