Abstract
As the impacts of climate change unfold, coastal cities are beginning to adapt to the emerging physical and financial risks. In our case study of climate adaptation in Boston, we advance the concept of hegemonic performativity, which shows how political pressures lead an assemblage – a network of human and nonhuman actors, including models, algorithms, instruments, market devices, and experts – to converge on a consensus in ways that privilege particular goals, actors, interests, and forms of knowledge. Our findings show how an assemblage is performative in building consensus around a particular climate response that tames uncertainty by excluding extreme risks and incorporating more palatable scenarios and parameters so that adaptation appears manageable and compatible with business-as-(almost)-usual. The mechanism of silencing facilitates consensus by downplaying community voices, equity concerns, and more extreme climate scenarios. Our study highlights how the operation of an assemblage is performative in shaping adaptation plans, physical interventions in urban infrastructure, and associated financial mechanisms with considerable effects on who and what is protected.
Introduction
Unprecedented disasters from climate change, including extreme heat, wildfires, and floods, are bringing destruction to lives and property (Allen et al., 2018). While emissions proliferate, intergovernmental organizations, nations, and cities are responding through adaptation processes that rely on experts, models, and calculations to construct scenarios of likely climatic impacts. In turn, these scenarios and analyses inform plans to protect assets and people in ways that will shape the built environment and lived experiences. The underlying data, models, and algorithms provide a sense of objectivity and legitimacy to these scenarios and plans (Callon & Muniesa, 2005; Millo & MacKenzie, 2009), but they also reflect organizational and political processes that select particular models, assumptions, and parameters, while silencing some voices and concerns. The adaptation process thus emerges from an assemblage of social and material elements – the networks of human and nonhuman actors including models, algorithms, instruments, market devices, and experts – that has performative effects by constructing consensus around projected climate risks and adaptation plans.
Recent discussions of assemblages in organization studies emphasize their performativity in bringing reality into being (Beunza & Ferraro, 2019; Cabantous & Gond, 2011; D’Adderio & Pollock, 2014, p. 640; Glaser, Pollock, & D’Adderio, 2021; Marti & Gond, 2018). That is, to be actionable and meaningful, any explanatory framework or theory requires infrastructure – an assemblage of rules, calculative devices, and other sociomaterial elements – that operates within or responds to a particular context (Callon, 2007; D’Adderio & Pollock, 2014). These studies show how ‘performative struggles’ among competing theories can influence assemblage realignments that produce new organizational routines (Callon, 2007; D’Adderio & Pollock, 2014).
This performative aspect also suggests how management theories, for example, do not simply describe the world, but actually create it in their image (Alvesson & Spicer, 2012; Carton, 2020; MacKenzie, 2007; MacKenzie & Bamford, 2018). One strand of this scholarship describes how theories and practice mutually adjust to achieve better societal outcomes (Carton, 2020; Glaser et al., 2021) while more critical interpretations have examined how power shapes contestation among competing theories. Bowden, Gond, Nyberg, and Wright (2021), for instance, discuss how local homeowners and real-estate interests mobilized concern about home prices to counter municipal efforts to translate climate science into adaptation plans. However, there is limited understanding of how assemblages and their performativity are influenced by organizational processes and political interests, especially when new phenomena challenge extant economic and political realities (Callon, 1998, 2007).
Our investigation of a climate adaptation assemblage takes a more critical approach to examine the pressures and processes by which an assemblage creates legitimacy and the appearance of consent around scenarios and plans in ways that align with dominant cultural and economic systems, thereby serving particular actors and interests while silencing others. The assemblage thus plays a hegemonic role in protecting these institutions and associated actors from threats (Jessop, 2010; Levy & Spicer, 2013). Consent, in the context of a hegemonic system, is a nuanced concept that implies ‘strategic acquiescence’ as much as full agreement (Levy & Scully, 2007, p. 978). For instance, climate adaptation processes rely on apparently objective and scientific analysis to construct projections and plans that appear to reconcile tensions between climate risks and economic growth, making adaptation appear manageable without disrupting the economy. To understand this process, we are guided by the following research question: How does an assemblage perform consensus within tensions between a new threat and the dominant system?
In answering this question, we investigated the case of climate adaptation in Boston, Massachusetts, USA. The City of Boston is an appropriate case because it recognized early on the high risks of sea-level rise and storms to people and property (Hallegatte, Green, Nicholls, & Corfee-Morlot, 2013). The complex and variable impacts of climate change present a significant organizational and political challenge that requires multiple public and private organizations to engage with the development of risk assessments, economic analyses, and adaptation plans (Wissman-Weber & Levy, 2018). Two authors collected data on adaptation processes led by the City, in conjunction with local consultants and business entities, which used model-based risk assessments and project evaluation. Our qualitative analysis of data drawn from interviews, observations, and archival sources shows how an assemblage mobilized theories and calibrated calculative practices to yield simplified and bounded constructions of climate risk.
Our paper makes three general contributions. First, our analysis develops the notion of hegemonic performativity, which conveys how the assemblage processes align interests and shared meaning to achieve consensual legitimacy, or ‘facticity’ (MacKenzie, 2007; Millo & MacKenzie, 2009). While existing performativity research focuses on how theory shapes practices and material reality (Cabantous, Gond, & Johnson-Cramer, 2010; D’Adderio & Pollock, 2014), our notion of hegemonic performativity instead emphasizes how organizational processes operate to construct consensus around a theory, embodied in models and scenarios, that aligns with pressures from the wider system. Second, we extend current discussions of assemblage theory in organization studies (Carton, 2020; Gehman, Sharma, & Beveridge, 2022) by showing how assemblage processes both create and partially mask contradictions and misalignments between the threat (i.e., climate change), planned responses, and the wider hegemonic system, creating dynamic feedback effects. Third, and practically, this study contributes to climate adaptation discussions by showing how the assemblage tames uncertainty through the choice of particular scenarios, parameters, and assumptions and is performative in delimiting climate adaptation processes with consequences for who and what gets protected (see Hayes, Introna, & Kelly, 2018). Climate risk is thereby rendered as a tractable, manageable problem, which facilitates the continuation of business-as-(almost)-usual but underestimates climate impacts, perhaps to a dangerous extent (Zscheischler et al., 2018).
Assemblages and Performativity
The concept of assemblage (or agencement) developed by Deleuze and Guattari (1987) aims to explain the complex relations between heterogeneous elements of a specific arrangement, and the continuous development of the arrangement as a rather fragmented, emerging configuration (Deleuze & Guattari, 1994). An assemblage represents a continuous process of matching or fitting together diverse components (Gehman et al., 2022) in contingent and dynamic ways that can link different assemblages together, creating unpredictable patterns and effects. In our case, models of sea-level rise were combined with financial models from insurance assemblages and cost-benefit analysis as part of the adaptation assemblage.
There is a noted affinity to actor-network theory (ANT) (see, e.g., Callon, 1998, 2007; Latour, 2007; Law, 2004) in describing how orders emerge and cohere, with both assemblages and networks explaining how heterogeneous entities interact and function as collective, emergent entities that are performative in producing new realities. This idea has been used to understand complex forms of agency (Carton, 2020; Gehman et al., 2022), where an action, strategy or initiative cannot be attributed to a particular person or group of actors; rather, it is the assemblage, or assembling, of networks (Callon, 2007; MacKenzie, Muniesa, & Siu, 2007) that embeds theories (Carton, 2020; Ligonie, 2018; Marti & Gond, 2018) with performative effects.
These scholars explain how complex assemblages of actors, including algorithms, instruments, and experts, direct possibilities and limit alternative actions (Callon, 1984, 1998; MacKenzie, 2017). As such, assemblages provide infrastructure for explanatory frameworks or theories to become actionable and meaningful (Callon, 2007; D’Adderio & Pollock, 2014). The legitimacy of the process and ‘facticity’ of the results derive, in part, from the status of the experts involved (Carton, 2020; Ligonie, 2018; MacKenzie, 2007). For instance, management gurus develop theories that influence organizational processes (Carton, 2020; Ligonie, 2018). Despite the uncertainty and inaccuracy associated with predictions, Millo and MacKenzie (2019) observe that ‘hybrid human–machine networks . . . produce results . . . perceived as valid and accurate descriptions of a reality’ (p. 641).
Delimiting alternatives is accomplished through a process of commensuration and translation that employs calculative practices to simplify complex phenomena and various types of risk into quantitative data and a common financial metric (Hayes et al., 2018; Levin & Espeland, 2002). For instance, MacKenzie (2009) describes how calculative practices commensurate diverse greenhouse gas emissions into the standard metric of ‘carbon dioxide equivalent’, which is then used without further examination, or ‘blackboxed’, for market trading. The common metric facilitates communication and convergence among the heterogeneous actors, but also oversimplifies complex, entangled phenomena and marginalizes considerations that are difficult to quantify or monetize (Callon, 2007). In this instance, blackboxing emissions metrics underestimates the shorter-term potency of gases such as methane. As such, blackboxing delivers organizational and political goals but mystifies the complexity of phenomena.
For assemblages to be performative, their practices need to be ‘in connection with’ (Gherardi, 2016, p. 694) the broader operating context and ‘rules at large’ (D’Adderio & Pollock, 2014). Callon (2007) describes co-performation where an assemblage is mutually constituted by ‘economics-at-large’ – the broader rules of governing (p. 26). For the assemblage to gain traction, its configuration is realized only in relation to other systems (D’Adderio & Pollock, 2014). Thus, the existing economic and political infrastructures are constraining and enabling forces that shape the construction, maintenance, and performativity of an assemblage (Callon, 1998, 2007; Hardie & MacKenzie, 2007).
The Politics of Performativity
Assemblages often have unintended, or even counter-performative effects, and their structure, processes, and performativity can be influenced by political and economic processes. The complexity and indeterminacy of assemblages creates the potential for misalignments between the theories the assemblages embed and the outcomes they perform. Conceptually, this misalignment has been called ‘errors’ when the theory does not align with reality (D’Adderio & Pollock, 2014) and ‘overflows’ when referring to aspects of the world that the theory neglects (Callon, 2007; Callon & Muniesa, 2005). Most treatments of these misalignments suggest that they prompt assemblage reconfiguration to increase the assemblage’s productive capacity (Carton, 2020; Glaser et al., 2021).
More critical perspectives point to the potential for counter-performativity, when the ‘model does not simply fail to produce a reality that is consistent with the model, but actively undermines the postulates of the model’ (MacKenzie & Bamford, 2018, p. 99). The model (or theory) and associated assemblage might be undermined unintentionally, as occurred with the financial crisis of 2008 (Carter, 2013), or more actively by actors ‘gaming’ the system to their advantage (MacKenzie & Bamford, 2018). For example, Nyberg and Wright (2016) use the term ‘misfire’ to describe how the fossil fuel industry has tried to misrepresent the likely magnitude of climate impacts. Bowden et al. (2021) likewise demonstrate how actors mobilized local concerns with house prices to challenge the translation of scientific understandings of climate impacts. These ‘misfires’ and counter-performativity highlight the deep entanglement of assemblages and performativity with politics, as the models are tuned to serve particular actors and interests (Bowden et al., 2021; Nyberg & Wright, 2016). Similarly, Gond and Nyberg (2017) advance the concept of political performativity to explain how the constitution of new assemblage entities such as corporate social responsibility rankings are inherently political in their attempt to mask tensions between profitability and societal goals.
These political perspectives illustrate how an assemblage can generate theories that internalize elements of the dominant economic system, and so construct or ‘perform’ realities that align with and reinforce this system (Bowden et al., 2021; MacKenzie, 2007). This process echoes the concept of hegemony, which refers to the contingent stabilization of a sociomaterial system, but with an overt political characterization. Hegemony refers to the alignment of economic, discursive, and political forces that mobilizes an array of actors and generates consent, cooperation, and legitimacy around a set of ideas and projects that serve their interests (Haugaard, 2006; Mumby, 1997). Hegemonic formations always contain tensions and contradictions, but these are managed through minor accommodations and the projection of intellectual leadership. MacKenzie (2017) demonstrated how the mobilization of experts and complex models could secure legitimacy even when financial models ‘perform’ inaccurately.
These tensions generate dynamics as dominant groups try to protect the system while challengers employ strategies to exploit contradictions and misalignments to create change (Levy, 2015). As Garud and Gehman (2019) observe in their discussion of the relationship between performativity and practice, ‘the world is always in flux and transformation; identities, functionalities, theories, and practices are materialized and transformed along the way’ (p. 681).
One way that assemblages attempt to maintain hegemonic stability is to marginalize or silence certain voices, concerns, and forms of knowledge, such as social equity or extreme outcomes in climate adaptation (Anguelovski et al., 2016). Such silences have been noted in contested phenomena such as workplace safety (Hardy & Maguire, 2016) and exposure to nuclear waste (Cable, Shriver, & Mix, 2008). Social equity concerns can be silenced by relegating them to ‘mundane calculative practices’ that reify unequal relations (Hayes et al., 2018). From a performative perspective, silences represent an ‘organizing agent of change’ that reshapes distributions of power (Dupret, 2019, p. 699). These silences can render risks as ‘manageable’ in complex human–technological systems, but then end in disastrous consequences when stressed systems fail (Perrow, 2011).
In our study, we investigate the development of an assemblage to address the threat of climate change. Climate change adaptation is a complex and political issue. There is a patchwork of misalignments, both within the assemblage itself and with the external context it helps to construct. Indeed, the lines between the assemblage and its external context might be blurred. An important aspect of performativity is then how the assemblage stabilizes and reproduces itself. This leads us to our research question: How does an assemblage perform consensus within tensions between a new threat and the dominant system?
Methods
Case context: calculating climate risks in Boston
This paper is part of a larger research project initiated by two of the authors of climate change adaptation in Boston, Massachusetts, USA (2015–2019). During this time, a series of multi-year climate adaptation projects was actively funded that culminated in reports detailing likely climate impacts such as hotter days and storm surges, populations at risk, and potential economic losses. Two authors gained access to interviewees and events because their affiliated university was involved in these projects, which allowed us to follow the climate adaptation process ‘in real time’ using interviews, observations, and archival data. A preliminary study using the initial 14 interviews (out of 48) focused on the first phase of the climate adaptation project in Boston (Wissman-Weber & Levy, 2018). That study described the emergence of a ‘risk regime’ that structured the political economy of regional climate change. We build upon that study, using a much larger dataset to examine assemblage performativity in the case of climate change adaptation in Boston.
A case study is an appropriate methodology for novel and complex phenomena such as climate adaptation (Yin, 2003). The case is revelatory in that Boston has been at the forefront of climate work, is at severe risk from climate change impacts, and there is regular cross-sector collaboration around climate adaptation. As such, it presents an early-stage opportunity to explore the heterogeneous elements and their relations that comprise the assemblage. The city was part of the Rockefeller Foundation 100 Resilient City Initiative (beginning in 2014), has philanthropic support for adaptation planning, and houses firms that model climate risks, such as AIR Worldwide, and consulting groups that engage in developing climate change impact models, reports, and tools (100 Resilient Cities, 2019; City of Boston, 2017).
Boston is ranked fourth in terms of flood risk in the United States (Hallegatte et al., 2013), with large low-lying areas reclaimed from the harbour. This risk is amplified by a construction boom in recent decades, particularly in areas susceptible to flooding. Awareness of climate risk grew dramatically in the wake of Hurricane Sandy in 2012, which generated losses of at least $50 billion in the US and would have caused substantial damage to Boston had it hit at high tide. Considering potential losses, financial markets have been signaling that municipalities should pay closer attention to climate impacts or risk being downgraded by credit rating agencies (Moody’s Investor Services, 2017).
The variety of actors engaged in ongoing processes presented an opportunity to explore political and organizational processes. For instance, Boston’s Green Ribbon Commission (GRC), a group of high-ranking public and private sector actors, convened to support the City’s 2007 Climate Action Plan. Nearly a decade later, GRC launched the Climate Ready Boston (CRB) initiative, with funding from local philanthropic organizations and in collaboration with the City, including the Mayor’s Office and the Environment Department. CRB comprises a series of projects, conducted in collaboration with local universities and private consultants, to assess climate risks and make recommendations. A broader network of business actors was engaged in the process, including insurance, finance, and developers. At the heart of CRB’s process was a series of linked models to forecast climate impacts and consequent losses, and to evaluate the value of adaptation projects. There were also activities connected to CRB and the Mayor’s Office focused on social issues and highly diverse and poor neighborhoods in climate adaptation.
Data collection
Interviewees were identified by developing a target list of organizations involved in climate adaptation. Between 2015 and 2019, the first author conducted 48 interviews that involved actors connected to climate adaptation, including consultants, nonprofit organizations, philanthropic organizations, business associations, city government officials, activists, and social movement groups that focused on areas such as risk assessment, climate modeling, consulting projects, and affordable housing and displacement (see Table 1 for a summary of data sources). Since adaptation involved private, nonprofit, and government organizations, we aimed for adequate representation from each sector. We conducted interviews in each sector until saturation was reached, that is, informational redundancy or lack of further relevant interviewees (Corbin & Strauss, 2015).
Data sources.
Interviews consisted of open-ended questions and lasted 1 to 2 hours. Interviewees were asked about approaches to adaptation such as climate risk assessments (e.g., sea-level rise, emissions scenarios), how and why specific factors were chosen, and approaches to social impacts (e.g., gentrification). When available, interviews were audio-recorded and professionally transcribed, and when not, detailed notes were taken. Interviews took place after interviewees received and agreed to the study parameters outlined in the informed consent documents, which included details about the research project, procedure and risks, confidentiality, data use, and authors’ contact information. Careful anonymization of all interviewee and organizational data was critical because Boston has a small network of individuals working on climate adaptation. For instance, one individual was typically responsible for climate adaptation in a particular agency or organization.
Two authors collected observational data from public and private meetings organized by the City of Boston, local Boston-area universities, private organizations, and the public sector (see Table 2 for list of observations). Public meetings covered adaptation topics such as social equity, community perspectives, property and insurance information, and the release of studies such as the ‘Feasibility of Harbor-wide Barrier Systems’ (Sustainable Solutions Lab, 2018). Private meetings involved interested city representatives, developers, and insurers, and were typically hosted by a consulting group or city agency and tended to focus on issues of financing adaptations, insurance, and infrastructure. Meeting convenors were contacted prior to attendance and given details about the research project that included procedure, confidentiality, and contact information. Participants were made aware that a researcher was present and collecting data. Observations were an essential source of detail about cross-sector engagement because they provide a more natural state of observation than a researcher–informant interview (Ritchie, Lewis, McNaughton Nicholls, & Ormston, 2013), and in this case, provided a snapshot of misalignments among assemblage components, particularly regarding social equity concerns.
List of observations.
Archival data added a layer of verification to interview and observational data. The archival data provided information about climate impact modeling and allowed for corroboration of data to confirm, for instance, methods, calculations, and events provided by interviewees. All archival data was publicly available and included information from meetings, news articles, and reports developed by research institutions, consulting groups, government agencies, and community groups, and internal reports produced by insurance and consulting firms (see Table 1 for the summary of data sources and Table 3, archival record examples).
Archival record examples.
Data analysis
The data were analysed abductively using the qualitative analysis software Nvivo 12.2. An abductive approach recognizes the interplay between theory, method, data, and inference (Van Maanen, Sørensen, & Mitchell, 2007). Two authors read all data, including interviews, observation notes, and archival data. The interviews and observations were central to our analysis, and a selection of archival data relevant to our research question was analysed and used to corroborate data (Saldaña, 2015). In line with our abductive approach, we simultaneously iterated between data, theory, and our knowledge of the field to interpret the emerging themes (Timmermans & Tavory, 2012).
We first proceeded with open coding and memo-ing by reading through the empirical material to understand how climate impacts were described and calculated (Saldaña, 2015). For instance, we noticed particular points of struggle, such as property development, social issues, and assumptions in models. As clear patterns in the data emerged, we consulted assemblage and performativity theory to make connections between climate adaptation, the context of markets and regulatory structures, and the effects.
We then developed the second-order categories by iterating between the data, our theoretical framework, and the literature. We mapped connections between first-order codes by consulting with our theoretical framework and allowing for unexpected patterns to emerge (Van Maanen et al., 2007). To finalize and operationalize higher-order categories, we engaged in multiple rounds of coding that involved conversation between the authors and consultation with theory. In developing our categories, it became clear that calculative practices including hybrid networks of human and nonhuman actors (e.g., experts, algorithms, instruments, and market devices) were central in developing the assemblage. We noted various tensions, for example, how the City attempted to reconcile flood scenarios with plans to continue property development along the shoreline (City of Boston, 2015).
In the final stages of coding, we referred to our theoretical framework to analyse how an assemblage develops consensus in responding to the climate threats. In doing this, we axially analysed the processes of assembling human and non-human actors that, in response to misalignments, silenced voices, legitimized actors, and tamed uncertainties. This produced performative effects that were calibrated with and reinforced the wider hegemonic system (see Table 4 for further details of the categories and the effects). Finally, our research team engaged our theoretical framework to map how the categories fit together to develop a model of hegemonic performativity (see Figure 1).
Categories, description, and representative quotes.
Although the multiple data sources produced a robust picture of adaptation in the cross-sector context, there are limitations to our dataset and analysis. First, we relied on expert interviews, observations, and archival records that detailed the methods and instruments that structured calculative practices. Thus, we did not directly observe the material elements. The data are representations, recollections, and forecasts of climate impacts on systems. Another limitation in our analysis is that the marginalization of community groups in adaptation deliberations is reflected in our empirical material. However, we do not directly analyse marginalization processes. In the next section, we outline the case context and present our findings.
Findings
Our findings demonstrate how the adaptation assemblage – networks of human and nonhuman actors, including models, algorithms, instruments, market devices, and experts – embedded complex calculative practices that projected future sea-level rise and storm surge to estimate flooding and consequent social and economic harms. These projections were based on multiple models linked together, which made the projected outcomes highly sensitive to the selection of scenarios and parameter calibration. In what follows, we employ the five categories from our analysis to show how the assemblage was performative in constructing a consensus view of climate risks around a particular imagined future, one that rendered climate threats manageable without major disruptions to the local economy or municipal finances, and hence aligned with the wider hegemonic system.
Assembling data-driven risk productions
Assembling data-driven risk productions involved enrolling experts from multiple sectors in a process driven by models and specialized expertise. After Hurricane Sandy narrowly missed Boston in 2012, the City developed several reports such as ‘Preparing for the Rising Tide’ and ‘Greenovate Boston: Climate Action Plan’ to identify climate risks and discuss adaptation strategies. Other initiatives included an international design competition, ‘Boston Living with Water’, in collaboration with other organizations, to plan for a ‘resilient, more sustainable, and more beautiful Boston adapted for end-of-the-century climate conditions and rising sea levels’ (City of Boston, 2015).
The first Climate Ready Boston (CRB) project was the Boston Research Advisory Group (BRAG), tasked to review scientific literature and develop consensus estimates of climate impacts such as sea-level rise, extreme heat, and other risks, under various scenarios (Climate Ready Boston, 2016). The estimates of potential flood depths and probabilities were derived from global circulation models (GCMs) that forecast future average temperatures and precipitation, hydrodynamic models of storm surge, models that link sea-level rise to temperature, and simulations of storms with various intensities and tracks.
Aside from the inherent limitations of modeling, GCM temperature forecasts are dependent on assumptions regarding future emission trajectories, population and economic growth, technological innovation, and climate policies (Shackley, Young, Parkinson, & Wynne, 1998). These temperature forecasts are the main input for sea-level rise models, which are even more uncertain because of the complex dynamics of ice melt. Estimates of flooding also need to account for storm surge, which can be up to ten feet for major storms (Climate Ready Boston, 2016). Climate change is likely to increase the intensity and frequency of major storms, but predicting the severity or paths of storms is difficult (Allen et al., 2018). The BRAG report predicted flood depths by relying on sophisticated hydraulic modeling that accounts for not just sea-level rise but also storm surge and tidal flows, though the report noted that limited computational resources compelled the use of a simple median estimate of sea-level rise rather than a probability distribution, which could underestimate flooding significantly (Climate Ready Boston, 2016)
A consulting firm translated the BRAG flood projections into financial loss estimates by combining flood maps with detailed maps of property values and applying insurance models that estimate losses to various types of buildings at different flood depths (City of Boston, 2016b). The CRB report gave a range of 2.4–7.4 feet of sea-level rise by 2100 under moderate emissions reductions scenarios, with a low (but unspecified) probability of 10.5 feet related to business-as-usual emissions. The CRB report also noted that the ‘business-as-usual’ scenario estimate of seven feet of sea-level rise is ‘within the likely range by the end of the century’ (City of Boston, 2016b, p. 19). The report, however, focused on the scenario of 36 inches of sea-level rise by 2100, and estimated average annual losses for Boston for three periods (in US dollars): about $140 million for 9 inches of sea-level rise, representing 2030s to 2050s; about $500 million for 21 inches, representing 2050s to 2070s; and $1.5 billion annually with 36 inches, projected for after 2070. These costs mostly comprised damage to building structures and contents.
Following the 2016 CRB report, which received substantial media attention, Boston embarked on neighborhood-level adaptation plans. Consultants were contracted to plan for a single target, 40 inches of sea-level rise, and to evaluate options using cost-benefit analysis. Separately, CRB commissioned an analysis of a 4 mile-long harbor barrier to protect the region, which was rejected due to high costs and feasibility questions (Kirshen, 2018). Cost-benefit analysis attempts to monetize and commensurate diverse types of costs and benefits across temporal scales, and is also highly dependent on assumptions regarding what is included, how valued, and the choice of discount rate (Wegner & Pascual, 2011). Consequently, cost-benefit analysis generally favors protecting wealthier neighborhoods with higher property values and neglects issues of displacement, community disruption, or adaptation-driven gentrification, because these are hard to quantify and usually considered out-of-scope (see Anguelovski et al., 2016).
Overall, the adaptation planning process reflected a commitment to data-driven analytical processes to optimize adaptation investments. As one high-level city official stated during a multi-stakeholder meeting intended to generate consensus around planning: Mayor Walsh’s overall vision is to make the city a data-driven process. We need to understand where we are most vulnerable. This needs to be translated into dollars and annualized based on what we are willing to spend yearly for protection.
Taming uncertainty
Taming uncertainty involved bounding the complexity of climate change in ways that constrained apparent uncertainty, rendered adaptation manageable and affordable, and kept stakeholders onboard. This process involved a series of decisions about scenarios, parameters, calibrations, and outputs, which helped make the process tractable but generated potentially problematic consequences.
The CRB process relied on an assemblage of diverse models that were originally developed for a variety of purposes, from setting insurance rates to modeling the global climate. Each model has significant sources of uncertainty, error, and scope for subjective selection of parameters and assumptions. Global climate models, for example, do not provide the high spatial resolution needed for city-level planning. These uncertainties are multiplied because of the way that the output of one model provides the input to the next, leading to a very wide range of possible outcomes. Indeed, an analysis of the pricing of catastrophe bonds, which rely on a similar set of models, indicated that ‘the modeling. . .is not demonstrably better than guesswork at predicting the financial consequences of extreme events’ (Etzion, Kypraios, & Forgues, 2019, p. 530).
One approach to reduce this uncertainty and stabilize the assemblage was to select just a few ‘most likely’ scenarios or even a single number. This occurred at every stage of the modeling, from the global climate modeling to sea-level rise to property damage estimates. The choice of 36 inches to represent sea-level rise after 2070 is less than half of the high-end estimate of 7.4 feet, to which the report assigned a probability of 15% (City of Boston, 2016b). The 36 inches was then ‘black-boxed’ (Callon & Latour, 1981) as the input to the loss-estimation stage (though another four inches was added rather arbitrarily for waterfront planning). The expected annual losses from flood damage were likewise presented as a single estimate rather than a range of potential values. Although the average was high enough to justify action, it still appeared relatively ‘manageable’.
The assemblage also tamed uncertainty by ignoring factors that were hard to model and quantify, such as potentially catastrophic feedback effects that could dramatically accelerate warming and sea-level rise (see Zscheischler et al., 2018). Estimates of property loss assumed that there would be no new development in Boston and that property values would not increase over the coming decades. The insurance-based models were not designed for and did not account for systemic city-scale losses following widespread and prolonged disruption from a flood disaster. As a result, forecast losses are likely to be significantly underestimated.
The City itself is also an important economic stakeholder, as development grows the tax base and revenues, with high-end shoreline development being particularly lucrative. One modeler noted how the desire to control adaptation costs affected modeling parameters and assumptions, stating: ‘In terms of informing what the real risk would be for inland flooding and how high levees should be, no-one wants to build levees that are too high because of costs.’ A consultant similarly reflected that: You can set your protection levels and define what is included. . .tidal, wave dynamics, worst case scenarios. This way you get the most value out of the investment, and you did something good. When you overprotect, it costs too much money, and that investment is never going to happen.
Cohering and legitimating the assemblage
Simplifying the models and choosing mid-range rather than more severe estimates helped to bring coherence and legitimacy to the assemblage. Similarly, the drive to commensurate and translate climate risks into a common financial metric facilitated communication and consensus building among the actors. However, it also led to the exclusion of particular impacts that are difficult to quantify and model. For example, the City’s planning process narrowed risk considerations, neglecting the impact of extreme heat, excess precipitation, and the spread of infectious diseases, even though the earlier CRB report had mentioned these risks. One executive for a global water management consulting firm explained the drive for simplicity and consensus: Sea-level rise is the easiest to plan for and with less contested data. Extreme rain, storm-surge, and rain patterns that require complex models and wave dynamics is more complex to translate to the city in a way that they can use the data and plan for adaptation. This whole step is skipped because it is too complex to grasp. They [the City] move forward not knowing what they are planning for.
Our interviews highlighted the organizational and political reasons for choosing somewhat arbitrary values from the range of model outputs. One city official suggested that the convergence around planning for 40 inches of sea-level rise was based on the need for a ‘reasonable’ outcome: . . . you’ve got to draw a line somewhere. You use a number that is reasonable. I mean, 40 inches for the most part matches the historic shoreline. It’s easier to rationalize than just 9 inches or the thousand-year storm event which is catastrophic but very unlikely.
Another city official was explicit that the choice of sea-level rise value reflected a concern to keep stakeholders at the table, and that more pessimistic forecasts might have caused alarm regarding potential future costs and regulations, such as extending designated flood zones: ‘[36 inches] is not a worst-case scenario, but there is greater than 50 percent likelihood. We came out not knowing if we’re going to be getting blowback from the development community here.’
The assemblage thus relied on sophisticated models and credentialed experts to create an aura of objectivity and scientific legitimacy, but in practice translated the calculations into simpler estimates that could be presented as reasonable and manageable. A notable element of the performativity of the assemblage was, therefore, to keep a diverse array of actors engaged with the process and secure buy-in across the business and policy communities.
Financial actors engaged with the assemblage and proposed innovative financial mechanisms to fund adaptation projects, such as insurance-linked securities, risk-based insurance rates tied to building resilience metrics, specialized ‘green bonds’ for infrastructure, and catastrophe bonds. These mechanisms draw from the wider legitimacy of economics to position financial actors and expertise as key to climate risk governance. An insurance executive that was active in the adaptation process commented: Think of insurance as an institution in society rather than an industry. Insurance is the gear to look at the way the capital investors and insurers look at risk. It is the gear to speak the language of risk. . .From systems to society – we help articulate risk to markets and lenders in ways they can understand – it is about data and analytics.
The financial actors strongly emphasized metrics and quantitative analysis, to the exclusion of more qualitative measures. One director for a multinational risk management company stated, ‘Let’s quantify the economic impact that we are talking about. . .only when we do that can we take action that makes a difference.’ A global insurance CEO further reflected on the foundational role of quantitative models in structuring the adaptation process to create hegemonic consensus: It is essential to be a master and not a servant of the modeled world. As a city you can create the rules of the game – insurers can work for cities and be the Boston algorithm to bring these communities together. . .We need to take everything we know and write the rules of the game for a shared set of challenges that we all face.
Insurers were keen to establish partnerships with cities to develop shared understandings of climate impacts and legitimize a common financial and data-driven language for adaptation.
Misalignments
Misalignments were evident across elements of the assemblage itself, for example, between the wide range of outcomes predicted by the climate models and the single number used for neighborhood planning, and between the assemblage and the external context, such as future flood damage. Several misalignments emerged in a dynamic manner as the adaptation project progressed and they could also represent potential drivers for change.
Perhaps the most glaring misalignment was between the plans to protect Boston neighborhoods, based on three feet of expected sea-level rise, and the likelihood of future flooding overwhelming these plans. There are two temporal aspects to this misalignment: first, evolving climate science increasingly points to faster and more severe sea-level rise; second, there is a multi-decadal time lag between planning and future climate impacts. The assemblage’s silence regarding extreme precipitation, which is causing devastating floods around the world, exacerbates this misalignment (Tradowsky et al., 2023). Indeed, the Boston Seaport waterfront district flooded twice in early 2018 during winter storms, leaving cars frozen in water, lending some urgency to adaptation planning.
The City envisaged that the CRB process would drive new building codes and push developers to be proactive in designing more resilient buildings. One misalignment that surfaced, however, was that most developers do not plan to own buildings long term, and they take advantage of imperfect markets to pass on the risks by selling to asset managers. While there were exceptions, one consultant for a climate modeling firm reflected upon the thinking of developers: We have talked to a lot of developers and sometimes they don’t care. They know what’s happening, but they’re, like, ‘Well, I’m building this to rent it out, to make money and I’ll move or flip it before things turn south.’ That can be a bit of a challenge. Regulations and policies need to change so that development projects going forward integrate this thinking into their work.
This was contrary to the aim of CRB’s forward-looking risk projections, which they assumed would encourage insurance markets to reflect climate risks. This would provide incentives to developers and building owners to make buildings more resilient and discourage development in high-risk areas. However, insurance markets were not tightly coupled to the adaptation assemblage; traditionally, insurance rates are determined using actuarial mechanisms based on historical data as well as competitive conditions. Moreover, insurance is typically a very small portion of building operating costs and unlikely to change location decisions. At one climate forum, a global insurance executive was asked: ‘Can you name any example where insurance has dissuaded a commercial developer from developing in risky areas?’, to which he responded ‘No’. Indeed, misaligned market-based risk signals produced contradictions such as continuing development along highly desirable but flood-prone waterfront areas.
Misalignment was also evident between the CRB estimates and the risks indicated by the Federal Emergency Management Administration (FEMA). FEMA is the primary federal agency that manages disaster response and mapping for zoning and insurance requirements. Mortgage companies and federal agencies rely on FEMA flood maps to designate insurance requirements, and these maps also use historical data to indicate the 100-year floodplain (areas subject to 1% or greater flood probability), neglecting future climate risks (Kalaith, 2016). Efforts to update and improve the maps have met political resistance and opposition from property owners concerned about insurance costs, creating a disjuncture between FEMA maps and climate-based models. One government resilience officer described the ensuing difficulty in working with developers: There are a lot of conflicts between what the City is projecting versus the FEMA maps and what Global Insurer Inc. uses to determine flood risk. . .We are relocating a public building and according to Global Insurer Inc., this area is not in danger of flood risk. So, from the developer’s standpoint there’s no problem. . .But according to the flood maps I’ve looked at, the new location will be more vulnerable and could flood even from a 100-year storm – that becomes our agency’s problem.
These misalignments could also be generative, as developers, insurers, and federal and municipal employees discussed the various maps and risk estimates and came to a better understanding, if not a consensus. We observed this dialogue during the CRB’s many meetings and documents circulated. One professional, who had worked in various capacities on adaptation, including in government and for environmentally focused nonprofits, discussed how a major winter storm highlighted misalignments between recognizing climate impacts and the finance that sparked conversations: In the context of having had an almost six-year conversation about climate change, winter storm Grayson was a motivating thing. The City said, ‘this is climate change’. That’s a really big deal. The other big deal is that it’s really hard to do something about. It’s expensive. Who is going to pay for it?
While these conversations helped to build consensus around municipal plans, they did not resolve the misalignments discussed above regarding risk forecasts and pricing. Some expressed worry that reality would eventually catch up with these contradictions, causing a dramatic impact on markets. A senior consultant for a nonprofit organization stated that ‘I am concerned that we are in a bubble – a real estate, insurance, market bubble. We create bubbles and then it all comes crashing down.’
A similar sentiment was reflected by a nonprofit director for East Boston, a diverse, disadvantaged yet gentrifying area that is highly vulnerable to flooding. He stated, We don’t need to just ensure that things and buildings are resilient, we want to make sure people are too. That they have the networks and the social cohesion to be able to help each other and bounce back in situations like big climate events.
The official reflected on the impact of these silences: ‘If you look at New Orleans and Houston, the people that suffered the most were people of color or people who didn’t have a lot of money. How do we prevent that from happening here?’ Ultimately, the planned infrastructure adaptations for East Boston ‘beautified’ the waterfront but priced out residents from their community (see McDonald, 2022).
Silencing
Silencing was a mechansim observed particularly around community interests of social equity or difficult-to-model phenomena. For instance, in building consensus and legitimacy for the adaptation process, the City had expressed its intent to engage the community and consider social and economic equity in the process. Some tension was evident, however, between the desire for inclusiveness to gain broad legitimacy and the coherence of core elements of the assemblage. While CRB held a couple of open events for the community, their concerns regarding housing affordability, employment, and mobility were largely absent from the process. One former activist turned Boston City official commented: ‘The social equity work did not get incorporated into the City plans the same way Climate Ready Boston did. They never became actionable in the same way.’
Community representatives from nonprofits and social movement groups did not participate in the key meetings, particularly those regarding finance and the built environment, and were rarely consulted as reports were compiled by the consultants. In one instance, the planning group for East Boston did engage a particularly vocal community organization to obtain their input, but the final report neglected this and instead presented the standard cost-benefit analysis (City of Boston, 2016a). Although community engagement was viewed as important for legitimacy, the assemblage marginalized these actors and their concerns.
The City considered developers and insurers to be key stakeholders in structuring the built environment and the market for risk, and they were active participants in CRB meetings and in commenting on adaptation discussions. The dominance of technical experts and financial approaches to adaptation tended to exclude the voices of low-income and minority urban communities from decision-making spaces. At an exclusive meeting of city officials, consultants, and business representatives, a global insurance representative acknowledged that ‘Certain populations are more vulnerable than others. It is a major issue and a major point’, but then continued: ‘At the city level you can allow the private sector to decide what they can contribute. . .that will help with broader protection.’ This assertion that the private sector can represent general community interests is central to the notion of hegemonic consensus.
The financial approach to climate risk and reliance on cost-benefit analysis led to attempts to identify ‘optimum’ levels and targets of risk protection. Boston contracted with design and engineering consultants to work on neighborhood plans and identify projects with calculated benefits that exceed the costs. These calculations raised a host of issues relating to what is included and how it is valued (see Anguelovski, Connolly, Garcia-Lamarca, Cole, & Pearsall, 2019; Sustainable Solutions Lab, 2018). Some of the neighborhood reports did mention community impacts, including the potential for gentrification, flood-induced pollution, and health impacts from heat stress. However, these were considered difficult to quantify and not included in the cost-benefit analysis. As one nonprofit program officer stated: We are defining a particular face of the problem [climate change] that’s limited because of who’s talking about it. . .it’s tricky to translate social issues into concrete solutions. I think that they are going to need a broader, more diverse advocacy community to do some of that problem solving and work.
In a similar way, a nonprofit director who was deeply involved in the city’s adaptation processes highlighted how the city’s emphasis on the built environment and flooding neglected community. She stated: Are they taking into consideration the neighborhood or are they just going to push their water on to the rest of us? Are we going to get priced out of our neighborhood? Where are we going to go? It all ties back into one conversation. If we are doing all these really wonderful projects to protect the neighborhood, is that just going to accelerate gentrification? Adaptation planning is happening, resilience planning is happening, and it should not happen without the community.
Community advocates expressed that adaptation plans were being developed by ‘experts’ who employed technical language that was difficult to understand or engage with. Moreover, this expertise was expensive, constraining who was able to commission detailed analysis and establish the parameters. One executive of a global nonprofit mentioned how an insurance company had told her: ‘We could help you, but we are not in the business of providing information to cities for free.’
Discussion: A model of hegemonic performativity
Our study explains how an assemblage of experts and models engaged in a process of managing emerging climate impacts by forecasting climate risks, particularly flooding and consequent economic impacts. We find that the climate impact assessment process served several political functions for the City and other actors; the risks needed to appear sufficiently credible and serious to mobilize action, but not so severe as to be unmanageable and threaten major disruptions to urban economic life. The projection of relatively moderate risks was important in keeping stakeholders, especially private developers and financial institutions, at the table. Further, moderate risk outputs facilitated the development of solutions at reasonable costs that allowed for continued development, without a retreat from the shoreline or substantial increases in insurance rates. In returning to our research question, we observed that the centrality of experts, models, and financial metrics in the assemblage served to generate a hegemonic consensus regarding climate impacts and adaptation plans. This consensus emerged from an alignment of material interests and discursive frames among the core actors in the assemblage, and broad legitimacy for the process, but in a manner that masked tensions and silenced more extreme climate outcomes as well as community groups and non-financial concerns.
Our analysis led us to develop a model of hegemonic performativity that extends a dynamic and dialectical conception of assemblage performativity. Whereas scholarship on assemblage theory shows a more evolutionary progression of performativity (Carton, 2020; Gehman et al., 2022), our model builds on critical performativity that emphasizes the political process (Bowden et al., 2021; Nyberg & Wright, 2016) where alternative and dominant theories concomitantly exist (Cabantous, Gond, Harding, & Learmonth, 2016; Spicer, Alvesson, & Kärreman, 2009). Our model leaves open possibilities for contestation (Levy & Scully, 2007) from alternative or silenced views. Hegemonic performativity (see Figure 1) shows how an assemblage maintains contingent stability through a process of accommodation and consensus that calibrates new climate threats to the wider hegemonic system to produce and maintain an imagined climate future.
A model of hegemonic performativity.
The model shows how the warnings of climate science, potential economic disruption, and direct experiences and awareness of climate impacts set in motion efforts to establish forecasts of the physical effects, likely costs, and adaptation resilience plans (indicated in Figure 1 by the box labeled new threats). These threats galvanize a network of actors, models, and calculations (depicted as the center circle). This stage of building a data-driven process is driven primarily by city authorities who decide which actors to include (finance and insurers), what methods to use (moderate climate scenarios and cost-benefit analysis), and to establish broad parameters and goals for the process.
The wider hegemonic system is a set of actors, ideas, and political and economic structures that include continued economic growth, the supporting financial system, and dominant actors such as financial actors and developers, city officials, and researchers. This wider system serves as a stabilizing force by both constraining and enabling the assemblage, through its influence on who and what is included in the assemblage, as well as the development of accepted methods and processes of organizing for climate change (Wissman-Weber & Levy, 2018).
The assemblage is performative in generating consensus across disparate actors and ideas, and in informing adaptation plans on where and how to protect specific assets and neighborhoods. Organizing a data-driven process establishes baseline adaptation parameters that can speak across dominant actors and calculative mechanisms (i.e., insurers, developers, and financial actors). Hegemonic performativity reflects the process by which a data-driven assemblage aligns with the wider hegemony by taming uncertainty to bound complexity in ways that increase the tractability of the assemblage. The simplification and commensuration of the risks coheres and legitimates an assemblage that can speak across contexts such as insurance, finance, and development. However, misalignments are created between different parts of the assemblage as well as in relation to the wider hegemonic system, which generate dynamics in the system.
Establishing a data-driven process helps reinforce the wider hegemony but creates tensions and misalignments. Silencing is a key mechanism in managing contestation and conflicting elements such as considerations of social equity, shoreline retreat, or more extreme outcomes. As an example, we observed that while the use of science, complex models, and professional experts brought credibility and legitimacy to the process, a series of choices needed to be made to simplify the results and constrain uncertainties if the process was to serve as a basis for planning and action. For climate adaptation, models project an unknown future ‘reality’, so acceptability leans not only on the legitimacy of the underlying science but also on the desirability of the imagined future. As MacKenzie (2007) observed, a model ‘must be an acceptable representation of the reality of which it speaks’.
The model shows how consensus is performed by developing forecasts and adaptation plans that are broadly acceptable to the dominant actors and aligned with the wider hegemonic system. Since this consensus relies on the marginalization and silencing of certain voices and concerns, there is always the possibility for new tensions to develop with new threats to the wider hegemonic system. This explains the process through which a new threat becomes malleable to the interest of dominant actors by using established ways of approaching a problem that aligns institutionalized models and calculative practices to manage it. As such, the hegemonic performativity can be employed to understand how cities, and even nations, deal with adaptation to climate change and other unpredictable events. Further, our findings help explain the stubborn tendency to rely on a limited set of tools that support a consensus in responding to threats in line with maintaining the current economic system.
Contributions
First, our analysis develops a model of hegemonic performativity, which conveys how the assemblage calibrates processes to align with a wider system to achieve consensual legitimacy, or ‘facticity’ (MacKenzie, 2007; Millo & MacKenzie, 2009). Where Bowden et al. (2021) took the modeling of climate impacts on coastal property as science-based theory, here we open up the process by which rather malleable models of climate risk are combined and then massaged in particular organizational and political contexts to project a favored future scenario. The purpose is not to create this future reality; the physics of climate dynamics are not perturbed by the models that attempt to describe them. Rather, the assemblage is performative in building consensus around a particular climate response, one that tames uncertainty by excluding extreme risks and choosing more palatable scenarios and parameters so that adaptation appears manageable and compatible with business-as-(almost)-usual.
The process serves a hegemonic function by using models and experts and a nominally open process to generate legitimacy and project ‘intellectual and moral leadership’ (Gramsci, 1971, p. 57), but in ways that privilege particular goals, actors, interests, and forms of knowledge over others. Our critical approach illustrates how a data-driven assemblage aligned the material interests and discursive frames of diverse actors to create consensus and legitimacy around ‘manageable’ climate risks and adaptation projects. Such consensus does not imply full agreement; indeed, several actors indicated how they ‘went along’ with the process despite some reservations, resonating with Levy and Scully’s (2007) description of hegemonic consensus as representing strategic acquiescence rather than full concurrence. Assemblage performativity then extends into the material realm with physical interventions in urban infrastructure and associated financial mechanisms.
This notion of hegemonic performativity contributes to these debates by challenging the perceived linearity of self-fulfilling theories (Marti & Gond, 2018). What is seen as taken for granted shapes the possible models and calculations employed to enact the future, and in a dialectical movement, the calculative practices (e.g., cost-benefit analysis) produce the imagined future. Hegemonic performativity does not just construct consensus in the assemblage; the assemblage serves to maintain the wider hegemony of reified assumptions of financial models, economic growth, and climate impacts. In other words, the assemblage is shaped by the wider hegemony, and in turn, the performativity of the assemblage fortifies and reproduces it. This shows the limits to experimentation (Marti & Gond, 2018) and alternative imagined worlds (Garud & Gehman, 2019), since the experimentation and imagination occurs in relation to a broader hegemonic system.
Second, we contribute to a more complex perspective on assemblage performativity. Recent studies have described how assemblages may generate ‘misfires’, ‘errors’, ‘overflows’ or ‘counter-performativity’ when the theories embedded in the assemblage do not align with the expected outcome or even undermine the premise of the assemblage, prompting revision of the theories (Carton, 2020; D’Adderio & Pollock, 2014). Inherent in these terms is a rather dualistic perspective where the assemblage either performs or fails to enact the expected reality. By contrast, our study paints a more nuanced canvas in which the assemblage constructs a particular version of reality that aligns well with some purposes and interests but also creates misalignments and contradictions.
In this regard, we observed that particular voices and elements were silenced or obscured in reconciling tensions or misalignments within the assemblage. The assemblage then produced delimited risk assessments that became concretized or ‘blackboxed’ and incorporated into plans, policies, and tools for managing climate impacts. The adaptation also excluded community voices and concerns, and focused on calculative practices that could build consensus around financialized metrics, pathways for continued economic development, and the building codes and neighborhood plans to enable these goals. This contributes to assemblage and performativity theory by emphasizing how silences and exclusions are crucial to understanding how models and theories are constructed so that they produce consensus around a particular version of a climate changed future, with associated ideas, practices, and planned interventions.
The actors and the calculative practices they employed did not maliciously or forcefully distort the portrayal of risks, but the overall operation of the assemblage had the effect of making adaptation appear tractable and manageable, obscuring but not eliminating contradictions between untrammeled economic growth and environmental concerns. The potential impacts become limited in scope and internalized in mundane calculative practices (Hayes et al., 2018). Climate adaptation as a process is influenced by powerful institutions and vested interests that strive to prepare for climate impacts while avoiding major economic disruption, such as retreating from at-risk and high-value shoreline development. This is important in order to understand the limits of climate adaptation assemblages. Future climate shocks could well undermine the legitimacy of the assemblage, with potentially catastrophic feedback effects from interactions among complex social and spatial climate drivers and hazards that could dramatically accelerate sea-level rise (see Zscheischler et al., 2018).
Third, the assemblage is performative in constructing climate adaptation processes in particular ways, which in turn prompt adaptation plans that do not protect communities equitably or consider a full range of risks. The emerging assemblage privileges adaptation as a financial and data-driven process supported by technology and innovation, while marginalizing community groups and non-financial considerations. These power effects tend to be masked by the trappings of scientific objectivity and a process that was at least nominally open to community input. This was most apparent around the built environment. For example, the CRB engaged the East Boston community – one of Boston’s most diverse and at-risk areas – to determine their climate priorities (e.g., safety and livelihoods were significant concerns). However, the value of disruption to communities is hard to quantify in monetary terms and is generally neglected (see Hayes et al., 2018). Consequently, the East Boston analysis focused on business interruption, regional economic impacts, and property values from flooding (City of Boston, 2016a), which belies the breadth of impacts, particularly those significant to marginalized communities. In fact, there is evidence that green adaptations have intensified shoreline development, contributed to displacement and gentrification (McDonald, 2022), and situated new residents on a shoreline at risk from significant storms.
Conclusions and Implications
Our findings have important practical implications in that as climate risks deviate from historical precedent, it becomes more difficult to determine future climate impacts and how to manage them. Storms and droughts are becoming more frequent, but the complex dynamics of the climate system do not allow for the estimation of these shifting probabilities. Climate adaptation illuminates a fracture between the current institutional infrastructure for managing risks, with its limited capacity and mechanisms rooted in a more stable climatic era (see Wissman-Weber & Levy, 2018), and the need to assess and plan for future climate impacts generated from human activity.
In developing our case study of Boston’s climate adaptation process, we advance an understanding of how the centrality of technical models and financial metrics tended to privilege technical consultants, financial actors, and economic concerns. Experts had control over technology, regulation, and knowledge and who was included in decision-making processes. Private market actors such as insurance companies, developers, bond rating agencies, and risk modeling companies play a substantial role in shaping how climate risks are perceived, translated, and addressed. Importantly, this shows how an assemblage creates infrastructure for acting on climate adaptation but is constrained through political and organizational interests. Even in a city like Boston at the forefront of developing inclusive climate initiatives, climate adaptation is emerging in ways that are constrained and reliant on approaches that support business-as-(almost) usual. Thus, hegemonic performativity illuminates how the operation of an assemblage subtly obfuscates uncertainties, higher-risk scenarios, and qualitative social dimensions, constraining the profound and urgent changes needed to prepare for climate impacts.
Future research may consider the disconnection of marginalized populations from adaptation processes and the silencing of their concerns. This study illuminates that, although social equity was included in some adaptation processes, it was not readily integrated into practices or was inadequate in remediating inequities. The needs of vulnerable or marginalized communities, and the trauma they may face in a climate-changed future (Anguelovski et al., 2016), are vital areas of inquiry. Importantly, climate adaptation organizing will continue to play an essential role in a climate changed future, with significant implications for social (in)equity, making this a critical area of investigation for organization scholars.
Footnotes
Acknowledgements
We would like to thank the editors, managing editor Timon Beyes, and the three anonymous reviewers whose thoughtful comments and valuable insights significantly enhanced the quality of this manuscript. We thank them for their time, expertise, and commitment to improving this manuscript.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
