A Tale of Two EPAs: Climate Change, Actor-Network Theory, and the Political Nature of Science

Zachary Lundgren, University of Northern Colorado

Published March 24, 2022

Following the 2016 presidential election, there came an immediate and widespread restructuring of United States policy, including within the Environmental Protection Agency (EPA), the primary government body tasked with safeguarding public and environmental health. While driven by science and empirical data, the EPA, being a government organization, is far from immune to political forces. The sweeping changes of the Trump administration were made manifest through various means, including the agency’s webpage (EPA.gov), where, practically overnight, links were deleted, topics altered, and information—especially that related to climate change—was nearly scrubbed entirely. However, as is unique to digital spaces, the previous version of EPA.gov still exists and is accessible via a banner at the top of the current EPA webpage. This results in two concurrent versions of the EPA’s digital presence, and  each represents a distinct understanding of the relationship among science, economics, and the role of the government in protecting both environmental and human health.

In this article, I analyze these two versions of EPA.gov as actor-networks, spaces “composed of many entities or actants that enter into an alliance to satisfy their diverse aims” (Spinuzzi 39). As actor-networks, the webpages become much more than neutral, digital conduits of science and environmental policy. Instead, they become political and rhetorical networks that articulate specific beliefs, values, and goals. As part of a tradition that examines the intersection between the environment and social discourse (Killingsworth and Palmer’s Ecospeak; Herndl and Brown’s Green Culture; Simmon’s Participation and Power; and Peterson’s Green Talk in the White House), this analysis highlights the role of political discourse in shaping our understanding of environmental topics. Focusing on the commonplace of climate change, I use Actor-Network Theory (ANT) to analyze the rhetoric of two versions of EPA.gov. This approach allows us to supersede knee-jerk claims of “good” or “bad” science to reach a more comprehensive and complex understanding of how power and scientific discourse operate in our society. 

INTRODUCTION

The primary purpose of the Environmental Protection Agency (EPA) is to “protect human health and the environment” (“Our Mission and What We Do” EPA.gov). This has been their standard since 1970, when Richard Nixon established the EPA in response to the growing environmental movement of the 1960s. This decade saw a series of important environmental landmarks, including: founding of the World Wildlife Foundation (WWF) (1961), publication of Rachel Carson’s Silent Spring (1962), and culminating with the first Earth Day (1970). Since 1970, the EPA has enacted their purpose through various means, from funding grants, educating the public, through, primarily, designing and enforcing environmental regulations. The EPA has achieved formative success in landmark legislations such as the Clean Air Act (1970), the Clean Water Act (1972), and the Toxic Substances Control Act (1976). In less than fifty years, the EPA has become a successful legislative force in protecting human and environmental health. 

In March 2018, when this research was conducted, a clickable banner at the top of EPA.gov stated the following: 

We’ve made some changes to EPA.gov. If the information you are looking for is not here, you may be able to find it on the EPA Web Archive [hyperlink] or the January 19^th, 2017, Web Snapshot [hyperlink]

January 19, 2017, is a significant date: it was the last day before Trump took office as the President of the United States, heralding a sea-change in US policy and political ideology. This sudden shift in policy became visible in the EPA almost overnight as Trump worked to overturn Obama-era clean-energy policies, and Scott Pruitt, a man who had sued the EPA nearly a dozen times as Oklahoma Attorney General, became the head of the agency. The shift also became visible on EPA.gov, the agency’s primary web presence. On their website, links were deleted, topics removed from or buried in the homepage, policies reframed, and information concerning climate change all but scrubbed. A quick visit to EPA.gov suggested that the EPA had become, seemingly overnight, a different organization. 

However, in a perplexing bit of digital limbo, these changes were not definite, as a version of EPA.gov that existed prior to the Trump administration remains accessible via the hyperlink at the top of the homepage. Clicking the link takes the user to a web snapshot of EPA.gov from January 19, 2017, the day before Trump took office. On this second version of EPA.gov, links still work, and content is still accessible. The only notable difference is that the web snapshot is static, not dynamic, and, therefore, the content does not update or change. Thus, in a unique state of affairs afforded to digital content, there exist, concurrently, two versions of EPA.gov—each representing distinct beliefs, values, and goals.  

Scientists, politicians, and the media were quick to respond to these changes. The Environmental Integrity Project claimed that “Trump’s war on environmental protections has been relentless,” as the administration had “targeted almost 80 environmental rules, including those meant to control greenhouse gases, coal ash waste, water pollution, mercury, and smog” (EIP). One example of such a rule was Obama’s 2015 “Clean Water Rule,” which aimed to protect rivers and waterways from pollution. In 2019, the rule was officially repealed by the EPA. Many were left wondering whether this was just one more example of Trump’s favoring the economy over the environment, as farmers and fossil fuel industries “complained the rule was too onerous and expensive to comply with” (Ebbs).

The selection of Pruitt as head of the EPA also raised red flags. As Oklahoma Attorney General, Pruitt had sued the EPA “14 times to block clean air and water safeguards” (EDF Action). As EPA Chief, Pruitt radically reinterpreted the commonplace of “protection” within the EPA—namely, who should be protected. Pruitt’s “Back-to-Basics” agenda focused on three Es: Environment (protecting the environment); Economy (sensible regulations that allow economic growth); and Engagement (with state and local partners) (EPA). While Pruitt claimed to seek a balance between the environment and the economy, declaring that “we can and we will achieve clean air and clean water and we will also have strong economic growth and job creation,” the fact that he launched his EEE agenda in front of an audience of coal miners is worth more than words (EPA).

Clearly, Trump is at odds with the scientific foundation that fuels not only the EPA, but other organizations, as well, including NASA and the National Science Foundation (NSF). Trump’s “attack on science,” as it is often presented, suggests a monolithic body—a material opponent that is capable of being attacked, almost like a rival politician (Union of Concerned Scientists). But this is a reductive understanding of Trump’s opponent. Science, like art or war, is a human endeavor, one method of many for creating knowledge. As Alan Gross proposes, “science may be progressively revealed not as a privileged route to certain knowledge but as another intellectual enterprise, an activity that takes its place besides, not above, philosophy, literary criticism, history, and rhetoric itself” (3). Thus, with a rhetorical understanding of science, it becomes much more complicated to “attack” this mode of thinking. In comparison, how would one go about “attacking” art or philosophy? Like these other human endeavors, science is complex, diverse, and far from a monolith. As such, the EPA and its website represent one platform for the intersection of science, human health, politics, and environmental policy.

In this article, I complicate Trump’s attack on science by analyzing the two versions of EPA.gov as distinct actor-networks, “assemblages of humans and nonhumans; any person, practice, or assemblage of these is considered a node in the network and indeed can be an actor-network in itself” (Spinuzzi 7). Using Bruno Latour’s Actor-Network Theory (ANT), I analyze climate change as a rhetorical commonplace to reveal the alliances and motivations of each actor-network. The Trump administration’s radical revisions to EPA.gov are prime examples of how communication and language can be shaped and manipulated in our post-truth age and why it is important to maintain vigilant study and research to be able to assess the rhetorical and political implications of these communication strategies. As Tarla Rai Peterson notes when discussing the work of James Cantrill and Christine Oravec, “because we are symbol-using creatures the natural environment we experience is largely a product of how we talk about the world” (4-5). Thus, examining the EPA’s bifurcated web presence in the wake of the Trump administration’s revision illuminates how environmental commonplaces, such as climate change, are considerably shaped by our discourse.

NETWORKS AND ACTOR-NETWORK THEORY

Networks have become an increasingly effective metaphor for understanding complex social phenomena, from healthcare systems and historical institutions to the production of scientific knowledge. One could say we now live in a networked society, as Jeff Rice proposes, one which “alters the concept of fixed identity by allowing individuals access to and production of multiple identities in multiple spaces”. This networked society is especially evident in online spaces, where Jeff Grabill and Stacey Pigg view “agency as constituted through interaction” with other actants (102). Identity, agency, and even power—in a network, these concepts are no longer fixed entities. Rather, they have become distributed and shaped by context and interaction. Networks have “shifted attention away from location as points of stability toward location as networks of people, buildings, events, data, and other human and non-human actors that interact” (Swarts).

Networks can also help us understand how science works. In applying network theory to the creation of scientific knowledge, Latour argues that “we are never confronted with science, technology and society, but with a gamut of weaker and stronger associations” (Science in Action 140). These associations form into networks, where “resources are concentrated in a few places—the knots and the nodes—which are connected with one another” and “transform the scattered resources into a net that may seem to extend everywhere” (180). For Latour, this is how a network gains strength and authority, through its extension and enrollment of various allies.

In Science in Action (1987), Latour presents the authority of “weathermen” as an example of a successful network. Forecasting the weather takes many shapes, such as professional forecasting on the news, farmer’s almanacs, empirical evidence, tribal histories, and even physical embodiment (the old woman feeling the weather coming in her bones). However, despite these diverse sources of information and authority, “the weather of the weathermen is strong enough to discount all the other weathers” (181). By enrolling the ethos of major news brands, standardized terminology and discourse, and the rhetorical velocity of technology like television and the internet, the network of weathermen is strong enough to override and deformalize all other networks.

To study the production of scientific knowledge and facts, Latour, along with John Law and Michel Callon, developed Actor-Network Theory (ANT), which questions what we mean by the “social” by rejecting the concept as “a homogenous thing” and analyzing it as “a trail of associations between heterogenous elements” (Reassembling the Social 5). ANT traces these connections and follows the actants themselves to “learn from them what the collective existence has become in their hands” (12). He claims that ANT “was devised as a reaction to the often too global concepts like those of institutions, organizations, states and nations, adding to them more realistic and smaller sets of associations,” such as the actions and discourse of individuals and objects (“On Recalling ANT” 2).

For Nathaniel Rivers, ANT is useful to “to trace the actors and to see the social as an emergent effect of the labors of many untold actors” (2014). ANT works to reveal those actants, those moving pieces, that are often concealed by broader social explanations and forces. For this analysis, the claim of “Trump attacks science” is exemplary of this concept. “Trump,” while functioning as an authoritative spokesperson as the President, is one actant within a broader network of the political distrust of the scientific community. When Trump claims that he wishes to “abolish the EPA or ‘leave a little bit,’” this must be understood within a long historical and rhetorical context (Neslen). Frederickson et al. note that Trump’s attack on the EPA “has precedents in the administrations of Ronald Reagan and George W. Bush,” as most Republicans have tried to severely cut funding to the agency (95). “Science,” in the example claim, is also a problematic term because, here, science is presented as a monolithic body—a singular entity that is capable of being attacked. It’s like saying someone is attacking the Catholic Church when they refuse to take communion. The statement may be true, in a sense, but the abstract reference clouds the involvement of individual actants and leads away from a more comprehensive understanding of why this social interaction occurred. Science, like the Catholic Church, is a vast and powerful network of individuals, institutions, and technologies. Like any institution, science is still vulnerable, but Trump versus Science is not a heavyweight boxing match. 

Network theories, and specifically ANT, help to reveal the complexity of social phenomena. “Machines,” Bruno Latour notes, which are networks in themselves, “are the concealed wishes of actants which have tamed forces so effectively that they no longer look like forces” (Pasteurization of France 204). Spinuzzi believes we can see “a network as woven and spliced, divergent and convergent, culturally-historically developed and anachronistically associated, material and performed” (60). For Spinuzzi, network enables and exposes activity. This is useful for an analysis of the two competing versions of EPA.gov, allowing for the study of information “woven and spliced,” interconnected and distinct, to be able to identify key discrepancies between the actor-networks (60). Alex Reid observes that this type of analysis is also critical for “our concerns in computers and writing,” which highlight “composition that recognizes its hybridity” (Reid). In the digital age, writing is collaborative, responsive, dynamic, and relational. Writing is everywhere and yet nowhere is it set in stone. This unique dynamic can be seen in the two versions of EPA.gov. In the digital era, understanding texts as networks helps to highlight their mercurial and capricious nature. 

ANALYSIS

For this study, I perform a rhetorical analysis on two versions of EPA.gov: one that existed prior to the Trump administration (EPA-1) and one that existed during (EPA-2). However, before beginning the analysis, there are certain limitations that should be addressed. Like any tool or theoretical framework, there are clear limitations to the use of ANT for this type of analysis. One of the most common criticisms of ANT and similar frameworks is that they are amoral. ANT, in particular, has been charged with having “little to say about the deep-seated political biases that can underlie the spectrum of choices that surface for relevant social actors” (Winner 370). Using ANT, you can trace a network of associations to conceptualize how a tool, scientific discovery, or belief system has come into being. But for most analyses using ANT, the project stops there. ANT does not provide a heuristic for action, for social engagement, or for determining the morality of a given actor-network. ANT traces, describes, and follows, almost like a disinterested, third-party observer.

In fact, Latour himself often critiques his own analytical tool. In Reassembling the Social, he notes that ANT “was perfectly fit for a blind, myopic, workaholic, trail-sniffing, and collective traveler” (9). We can think of ANT, then, almost like a bloodhound on the hunt. The dog will be able to expertly follow the trail (describing with its nose the shape of the network), but once our bloodhound has caught its prey, it doesn’t know what to do with it. Thus, ANT can help us construct a more honest, comprehensive, and inclusive (via nonhuman actors) understanding of the world. But the analysis stops there. Some may view this objective distance as an inherent tenet of ANT or as a mere cop-out, but the tool then requires others to determine the best course of action.

At the time of this study, logging onto EPA.gov reveals a banner at the top of the page that links to a version of EPA.gov from before the Trump administration. The banner reads as follows:

We’ve made some changes to EPA.gov. If the information you are looking for is not here, you may be able to find it on the EPA Web Archive [hyperlink] or the January 19^th, 2017, Web Snapshot [hyperlink]

Clicking through the second hyperlink, the web snapshot, presents a unique opportunity to visit two versions of the EPA at once. Roddy Scheer cites EPA spokesman J.P. Freire’s account of the changes to EPA.gov: “As EPA renews its commitment to human health and clean air, land, and water, our website needs to reflect the views of the leadership of the agency” (EPA). Freire claims that “we want to eliminate confusion by removing outdated language first and making room to discuss how we're protecting the environment and human health by partnering with states and working within the law” (EPA). To eliminate “confusion,” content has been removed, links and drop-down menus have changed, and almost any information concerning climate change has all but been erased. These two webpages serve as the primary data for this study. To examine how each webpage, or actor-network, articulates scientific information and knowledge (through an inherently political lens), I focus on climate change as a commonplace, and how its visibility and origins shape two divergent representations of this sociomaterial phenomenon.

Visibility

For digital spaces, accessibility of information is critical. In a sea of data and information, accessibility equates to efficacy, and in a sense, existence. While visibility and accessibility of information may seem like purely logistical issues, Douglas M. Walls notes that “access in computers and writing can be thought of as a dialogue between two major problems: technological materiality and ideology” (69). While logistics remain part of the issue, the ideology driving that digital space is also influential.

There is a stark contrast in the visibility and accessibility of information concerning climate change between the two versions of EPA.gov. While EPA-1 makes climate change information readily accessible through placement and sheer amount of content, EPA-2 effectively buries what is left of the climate-change content and removes it from the navigation menu.

One primary example is the top banner on the EPA homepage. The first link is titled, “Environmental Topics,” with a list of thirteen focal points for the EPA and an index for additional topic and content. In EPA-1, climate change is listed as one of these thirteen key points. However, in EPA-2, the topic of climate change is replaced by science. This discrepancy between topics is exemplary of the divergent actor-networks.

Listing climate change as one of the thirteen most prominent environmental topics is important in two critical ways. First, it highlights the importance of climate change as a topic of concern, situated between naturally occurring phenomena like air and water, with anthropogenic phenomena, including pesticides and lead. This situating of climate change amongst the natural and the anthropogenic, or social, phenomena recalls its hybrid nature as a phenomenon that cannot be solely attributed to the natural world or the social one, but rather, existing as an amalgamation of the two (Latour, We Have Never Been Modern 3). Second, in a matter that is only further highlighted by the emergence of EPA-2, listing climate change as an environmental topic confirms its validity as a scientific phenomenon. This positioning eliminates controversy and affirms the relevance and actuality of climate change. By being listed as an environmental topic, climate change becomes as uncontroversial as water or air.

EPA-2 replaces climate change as a topic with “science.” This shift suggests that science has become a piece of the puzzle, a component of the EPA and not the primary epistemological method. In EPA-1, science is the blood that pumps through the network; however, in EPA-2, listing science as a topic implies that science is subservient to the larger goals and aims of the EPA, which appear to be deregulation and economic gains.

EPA-2 does not espouse any radical notions such as the nonexistence of climate change or that it is a conspiracy concocted by China. Rather, the method by which EPA-2 articulates climate change is much more innocuous and perhaps more dangerous for anyone who seeks a real response to the environmental phenomenon. EPA-2 admits that the climate is changing but claims that they are part of “historically observed changes (e.g., temperature, precipitation)” (EPA). By claiming that climate change is a part of the earth’s natural oscillations, one immediately rescinds any responsibility by human populations to take action. Essentially, the position that climate change is a natural phenomenon takes the teeth out of any climate-change action or policy.

At the homepage of EPA-1, there is a list of “Popular Topics” just below the main rotating image carousel. Among those ten topics, with links leading to additional information, is climate change, listed among other important and other seemingly nonpartisan topics, such as drinking water, acid rain, and recycling. Through this positioning, climate change is an immediate focus for a user to EPA.gov. Additionally, its inclusion among other seemingly neutral, non-political topics, such as drinking water, adds an ethos of confidence and scientific neutrality to the topic. After all, who could argue the politics of safe drinking water? Clicking through takes the user to the climate-change homepage (epa.gov/climatechange), which hosts a plethora of content concerning climate change, including its origin, indicators, and impact assessments, and what the user can do to mitigate rising carbon-dioxide emissions (EPA).

In EPA-2, the climate change homepage does not exist. Searching for the URL, the user is redirected to the following message:

This page is being updated.

Thank you for your interest in this topic. We are currently updating our website to reflect EPA’s priorities under the leadership of President Trump and Administrator Pruitt. If you’re looking for an archived version of this page, you can find it on the January 19 snapshot [hyperlink].           

Not only has climate change been removed from the homepage, but the core source of related content has been erased and replaced with a “Climate Actions Benefit Report.” The content of this report will be analyzed in the next section; however, it’s notable that even this report is not easily accessible or visible. To access this report, the user must navigate through the A-Z index and search through the list, which includes a tedious amount of topics and links.

As Walls observes, the accessibility of online information is often guided by “ideology” (69). As Walls notes, rhetorical scholarship has “explored how underlying ideological assumptions drive… decisions about technological access” (69). The material and organizational move of burying certain links or content can reflect the ideological of the author or designer.

Here, the distinction is clear: in EPA-1, the user is presented with climate change; in EPA-2, the user must seek it out herself. The ideological difference is clear. This also correlates with Latour’s concept of the fact-builder. Exploring the ways in which allies are needed to transform a claim into a fact, an idea into a reality, he notes that “the construction of facts, like a game of rugby, is thus a collective process” (Science in Action 104). Use it or lose it, as the old saying goes. In the sciences, obscurity is worse than opposition, as a claim will not “become definitive if others do not take it up and use it as a matter of fact later on” (104). For a piece of information to gain validity, it must be taken up and used, supported, or even disputed. This is why the EPA-2 actor-network obscures climate-change information, making it practically inaccessible and invisible. EPA-2 works to bifurcate climate change with the rest of established science of the EPA, like water quality or asbestos.

Origins of Climate Change

EPA-1 explicitly links climate change to anthropogenic causes. Under a section titled, “Why is the Climate Changing?” greenhouse gas emissions are listed as one of the four primary components. Additionally, under a section titled, “What can we do about this change?” reducing emissions is listed as the first of four components and provides a link to more information. In the EPA-1 actor-network, climate change and greenhouse gas emissions are clearly linked as cause-and-effect. Greenhouse gas emissions are understood as one of several anthropogenic factors leading to rising global temperatures, in addition to land use, reflectivity, and agriculture (EPA).

Under the “Basic Information” link on the climate-change homepage, one of the first sections is titled: “Humans are largely responsible for recent climate change” (EPA.gov/climatechange). This section reports that “human activities have released large amounts of carbon dioxide and other greenhouse gases into the atmosphere,” and that “the majority of greenhouse gases comes from burning fossil fuels to produce energy” (EPA). There is no controversy. The anthropogenic origin of climate change is presented as a scientific fact, and the focus of the information is on mitigation and response. Using ANT, one can also see how EPA-1 enrolls a number of powerful scientific allies, including Scripps Institute of Oceanography and the National Oceanic and Atmospheric Administration (NOAA). By including graphs and data stamped with the seals of these powerful organizations, this webpage is allowed to borrow rather expensive ethos.

Additionally, the reliance on visual data serves another purpose. Latour discusses the authority of scientific texts and observes how “the transformation of linear prose, into, so to speak, a folded array of successive lines [a graph] is the surest sign that a text has become scientific” (Science in Action 48). By including graphs that, for example, show the rise in carbon-dioxide emissions, one must now contend not only with the EPA but the actual science itself. It’s much easier to refute language than numbers, and as part of an actor-network that values the empirical evidence of modern science, there’s no stronger ally than scientific data.

EPA-2 presents a much more complex understanding of the origins of climate change. EPA-2 reframes these origins from anthropogenic to an inevitable, naturally occurring event. In EPA-2, climate change is not a scientific problem to be studied and resolved but, rather, a natural event to endure and overcome, much like a hurricane or tornado. While there is no main climate-change webpage (EPA.gov/climatechange) in EPA-2, there is a “Climate Change Adaptation Resource Center (ARC-X),” which is described as “an interactive resource to help local governments effectively deliver services to their communities even as the climate changes” (EPA). What’s most interesting about this description is the last phrase, “even as the climate changes,” which presents climate change as a seemingly innocuous event—innocuous, and inevitable (EPA). Of course, if an event is inevitable, then there is no need to take drastic measures to prevent it. All throughout the ARC-X, there is no mention of the human role in climate change. In fact, there is hardly any information concerning the emergence, origins, or general cause of climate change. Rather, the ARC-X positions itself as a responsive document, providing information and resources to help communities respond to the effects of climate change, such as extreme weather events (EPA).

In lieu of a climate change homepage, EPA-2 presents  “Climate Change Impacts and Risk Analysis (CIRA),” a report that “quantifies the physical effects and economic damages under multiple climate change scenarios” (EPA). The CIRA comes the closest to identifying a connection between climate change and human activity, citing greenhouse gases as a key influencer for the increase in extreme weather events. It is also interesting to note that throughout the entirety of the report, greenhouse gases are never explicitly named, but rather, referenced as “GHG.” The acronym is never broken down or explained, which, on a small scale, adds a layer of textual occlusion to the report.

Another intriguing aspect of the report is the framing of climate change and mitigation. Rather than listing the potentially hazardous implications of climate change, such as sea-level rise, biodiversity loss, or ocean acidification, the report focuses on the advantages and benefits of “GHG mitigation,” as if the issue has already been appropriately and thoroughly responded to (EPA). The focus throughout the mitigation report is on the success of projected mitigation. Overall, there is an unexpected, and compared to EPA-1, radically divergent tone from one of care and concern to one of positivity and success. For example, the report claims that “global GHG mitigation is projected to have a substantial effect on reducing the incidence of extreme temperature and precipitation events by the end of the century,” and that “2100 mitigation is projected to avoid 12,000 deaths annually associated with extreme temperatures” (EPA). This assessment is entirely positive and beneficial, reflective of the webpage’s title: “Climate Action Benefits.” 

Additionally, the report concludes on a crucial note of uncertainty. The final section, titled, “Impacts Vary across Time and Space,” concludes that regional and temporal differences may impact the scale of climate change-related effects. While this is a valid claim, it does open up a wedge for uncertainty and doubt. Leah Ceccarelli analyzes the use of this “wedge theory” when discussing the manufacturing of scientific controversy. Citing an intelligent-design advocacy group, Ceccarelli notes how the group knew that they couldn’t entirely cut down the “giant tree” of science but, rather, needed to only provide a wedge of doubt, which, “while relatively small, can split the trunk when applied at its weakest points” (209). Many of those advocating against mainstream science utilize this strategy, realizing that it is “unnecessary to outnumber the opposing establishment” but rather, apply just enough pressure to open up a wedge of concern (209).

This final section of the CIRA report does just that as the concluding sentence reads: “simply analyzing an impact in one time period (e.g., 2100) may mask important temporal dynamics that are relevant to decision-makers” (EPA). While not directly casting doubt on the existence of climate change, this does offer a rationale for decision-makers and policy that does not respond directly to climate change. As Ceccarelli demonstrates, sometimes you don’t need the whole ax, just the wedge.

DISCUSSION

Why ANT? 

Since Thomas Kuhn’s The Structure of Scientific Revolutions (1962), contemporary scholars have studied the connection between science and rhetoric. Yet even Plato considered this relationship, asking Gorgias how “arithmetic is oratory” (450e). Science attempts to reveal the inner machinations of the natural world to demonstrate objective, predictable patterns. But the rhetoric of science has shown that this is not always the case. As a human endeavor, science is subject to perspective, persuasion, and even bias, leaving scholars and philosophers to continually debate: “What is the relationship between rhetoric and truth?” (Crick 1).

While science is a human endeavor, it is unlike other endeavors because it is founded on and grounded by empirical evidence. Gross argues that “science is constituted through interactions that are essentially rhetorical,” a claim that he modified ten years later, cautioning that “none of this suggests that science is only rhetoric; no sane person could reach so bizarre a conclusion” (78). At its core, there is some essential logic that, through persuasion, one cannot entirely alter or sculpt. As Gross explains, “no theory of physics can ignore flight; no theory of biology can turn its back on sex; no optics can dismiss refraction” (42). Thus, the rhetoric of science is caught between relativism and realism, between truth and persuasion. By using an analytical approach such as ANT, we can highlight this balancing act by recognizing the agency of nonhuman actants, like weather events and temperature extremes, and the implications for how humans represent them via discourse. ANT shows us that material objects and social discourse, together, make our world. 

For Latour, this last point means to dissolve the Modernist divide, which classifies the political, or social, and the natural into separate worlds. Using ANT, one sees the sociomaterial nature of phenomena, which can be understood as the “insistence on speaking of the social and the material in the same register… not reverting to a limited dualism that treats them as separate (even if interacting) phenomena” (Orlikowski 1437). This hybrid nature of phenomena is demonstrated in the articulation of climate change in EPA-1 and EPA-2.  While neither webpage contests the entire, empirical existence of climate change (EPA-2 does at least reference increased temperatures and extreme weather events), each actor-network strategically presents a form of climate change that better aligns with its political and rhetorical goals. But key differences arise when climate change is understood as a natural event, rather than an anthropogenic one, or when information concerning climate change is so effectively hidden that it becomes irrelevant. Even though the Trump administration has worked to divorce science from policy at the EPA, the organization cannot entirely escape the scientific method or scientific data. Just as Gross argues that “no theory of physics can ignore flight,” no version of the EPA can totally escape the house that the scientific method has built (42). But by using ANT, one can realize that even reality, from a sociomaterial perspective, is malleable, even multiple, with constantly shifting goal posts as actants shift, form alliances, betray those alliances, and build new networks.

Additionally, ANT is unique in that it describes phenomena, rather than offering up ready-made explanations. An actor-network, Latour notes, “is a concept, not a thing out there” and as such should be understood as a tool, not an entity (Reassembling the Social 131). A network is not a thing to be touched or analyzed but, rather, a tool to be used for understanding complex phenomena. To say that ANT describes “refers to an empirical tracing of the entire range of mediators, human and nonhuman, that support a particular and localized cultural activity” (Holmes 426). This is an important distinction in epistemological terms because “description means avoiding the reproduction of conventional topoi and discourse to explain a particular rhetorical situation” (427). Thus, ANT can move us beyond conventional claims of “good” or “bad” science to assess the fullness and intricacies of these actor-networks by refusing to use abstractions as shortcuts. Since ANT describes, it relies on the interactions of individual actants. An analysis using ANT can account for a more comprehensive understanding of the phenomena under investigation.

Critiques of Latour

While I find Latour and his theories to be valuable for this type of study, there are several valid critiques of his work, especially of his approach to sociology. Many find his radical approach to sociology (one that uses a flattened ontology to level the playing field between human and nonhuman actors) to be unusable, misguided, or as David Bloor claims, “a step backwards” (82). Many researchers claim that Latour simply doesn’t grasp many of the sociological theories he attacks, believing that his “criticisms are based on a systematic misrepresentation of the positions he rejects” (82). For example, Bloor decries Latour’s rejection of the subject-object schema (where a subject can observe an object in an independent reality), a rejection that is founded on Latour’s flattened ontology and inclusion of nonhuman actors within the realms of influence and agency. Latour works to dissolve the “anthro-centrism” of sociology, science, and many other fields, and yet, he may attempt to remove us too far from the human perspective.

Some scholars take issue with Latour, and especially ANT, as they claim that this flattened ontology approach to sociology also flattens and erases the embodiments of other, diverse voices. Zoe Todd, on attending a Latour lecture on climate change, noticed how his claim that the climate is a unifying force mirrors indigenous thinking: “Funny, I thought, this sounds an awful lot like the little bit of Inuit cosmological thought and legal orders that I have been taught” (5). Todd may be on to something.

While it’s a fruitless exercise to presuppose the intent behind Latour’s omission, any reader familiar with his work will know that his champions (e.g. Pasteur, Copernicus) tend to be white men from the European tradition. Without a doubt, this limits the scope and applicability of Latour’s thinking and ANT. Yet at the same time, I do not believe this is grounds for abandoning Latour and ANT. Rather, I think it highlights the fact that ANT is a useful framework for analyzing complex, social phenomena and scientific topics, but it is not complete. Incorporating other, diverse thinkers and modes of thought will only strengthen an analysis that begins with ANT.

Take the weather example from before. Latour describes how professional weather forecasters developed a stronger, more durable network than other modes of predicting the weather. As a result, professional weather forecasters have become the dominant network and wield most of the power, at least in most Western societies. However, instead of critiquing this hierarchy, one that may be built in part by colonialism, racism, and oppression, Latour stops with his description of how the network operates. He describes the evolution of this network but fails to explain many of its social implications or direct us in how we can make a certain realm of knowledge creation (weather forecasting) more equitable or effective.

By suggesting a flattened ontology where trees and hammers have agency on the same register as humans, some researchers argue that Latour fails to recognize the hierarchies and political imbalances that shape our social systems. It is easy to interpret a flattened ontology as one of equality that assumes we are all the same. Most likely, Latour understands the complexities and power structures of our social systems, but by expanding the classical agora of actors to include nonhumans, there is (perhaps inherently) less focus on the embodiment and agency of individual people. When we add trees, computers, carbon emissions, and snails into our discourse, it can be more of a challenge to emphasize the differences between human embodiments.  

Beyond Good and Evil (Science)   

In Beyond Good and Evil, Friedrich Nietzsche examines conceptualizations of good and evil, especially the oppressive role of Christian morality, claiming “There is no such thing as moral phenomena, but only a moral interpretation of phenomena” (64). We ourselves determine what is “good” and what is “evil” as our societies, and our times, require. For Nietzsche, there is no tangible connection between these abstract terms and the material events, people, and objects of our world. We determine morality, and like other abstractions, such as justice or peace, morality manifests through the people and objects of our material world.

In a similar vein, ANT seeks to understand “the exact content of what is ‘assembled’ under the umbrella of a society” (Latour, Reassembling the Social 2). Social explanations fall short for Latour as they fail to account for the specific actants and the relational nature of the actants involved in a given phenomenon. The Los Angeles Times claims that “Trump is killing the EPA by inflicting death by a thousand cuts” (2019). Vanity Fair addresses Trump’s “terrifying new plan to poison the air, water, humans” (Levin). It’s a common theme. Trump’s “attack on science” is repeated throughout countless media outlets, and while this discourse may be true, it also does little to describe our current state of affairs accurately and specifically or help us to understand the political nature of science. What does it actually look like to “attack science?” How does one attack an abstraction? Science is a method of human inquiry. As such, it is closer to a hammer than an objective truth. It’s a tool that can be wielded for a multitude of effects. Science (in the Western, post-Enlightenment tradition) has been wildly successful in developing vaccines, global communication, and renewable energy, but it has also brought us nuclear weapons and the Holocaust. In the scientific world, Crick argues, “Truths are not so much discovered as fought over, […] the caretakers of knowledge in any age are bound up with structures of power and authority, and […] the scientific community often accepts arguments in the short term (or even the long-term) for reasons other than pure rationality” (7).

A certain scientific method or viewpoint cannot truly be “bad” or “irrational” because those are abstractions that we invent. If the EPA fails to recognize the reality and complexities of climate change, calling the EPA (or those who drive it) “bad” or “evil” is a referential shortcut driven by a socially constructed morality. It leaves a lot out. However, if you claim that a study conducted by the EPA is based on inaccurate data or incomplete measurements, this statement can be traced through material actants and proven or disproven. There truly is no “good” or “evil,” but rather, the interactions of individual actants and how they influence additional actants. As Latour notes, “Belief is not a state of mind but a result of relationships among peoples” (On the Modern Cult 2).

EPA-1 is an actor-network of people, values, and goals. EPA-1 cannot be reduced to “good” science just as EPA-2 cannot be reduced to “bad” science because science is messy, complex, recursive, and most importantly, social (Latour and Woolgar 21). EPA-1 responds to and reflects certain values, such as those of scientific ethos, environmental protection, and a recognition of the anthropogenic origins of climate change. It is tempting to align these values with pure or objective science, but such a perspective falls short of understanding the complexity of science as a human way to engage with our natural world, one situated between both the cultural and the natural.

Thus, it also falls short to simply claim that the EPA-2 webpage is irrational or exemplary of censorship and “bad” science. Stopping there would not allow a comprehensive understanding of the network of associations and the actual work that they do. Latour discusses the trials of rationality, wherein beliefs and knowledge are separated through the successful expansion of some networks over others (Latour, Science in Action 53). The successful networks become knowledge and the unsuccessful are left to mere beliefs. Latour uses the example of weather forecasters, who become “the only official mouthpiece of the earth’s weather” as a myriad of alliances (major broadcasting rights, distribution, funding) works to expand their networks and ethos (182). Consider whose forecast you would most likely trust, the weatherman’s or your neighbor’s, and why. This trial of rationality can be applied to the EPA webpages as competing actor-networks. EPA-1 and EPA-2 represent radically different networks of actants, values, and goals. However, deeming one actor-network “rational” and the other “irrational” fails to reveal anything meaningful about each actor-network. What becomes more meaningful is assessing how one network might become more widespread, more accepted than the other and how.   

Steven Holmes posits that “networks have to be composed of writers through a rather laborious and difficult process of determining unique and local configurations of agency, connectivity, affect, influence, and many other factors” (426). ANT helps to reveal this complexity by following the actants and describing their interactions and working to avoid shortcuts or explanations. Networks, as understood with the resources of ANT, are never static, but rather, constantly dynamic, “woven and spliced, divergent and convergent” (Spinuzzi 60). Actants can betray one another, and alliances fall apart. As Latour notes, one can even “force the opponent’s allies to change camp” as new resources or actants are engaged (183). This is evident as EPA-2 reinvents the focus of the EPA, shifting from environmental health to economic health.

The dynamic nature of actor-networks should also offer hope. Even if the Trump administration works to bleed the EPA by inflicting death by a thousand cuts” the EPA will not ultimately die. From a sociomaterial perspective, the EPA is much more than a government organization, it is the work of countless scientists, Superfund projects, clean rivers, and ecologically diverse habitats throughout the country. The EPA can be attacked, yes, but it cannot so easily be defeated. And for those who understand how history repeats itself, conservative efforts to diminish the EPA should not come as much of a surprise. During the Reagan administration, “the EPA’s operating budget fell by 27%. The science budget tumbled 58%” (Frederickson et al. 96-97). Despite these massive cuts, the EPA has persisted.

Despite the evolving nature of the EPA, many still see Trump’s actions (and his entire presidency) as almost inconceivable. We can look at Trump as a Black Swan, a concept developed by Nasim Nicholas Taleb to account for unpredictable phenomena, and how we rationalize them. For Taleb, a Black Swan is defined by three characteristics: “It lies outside the realm of regular expectations … carries an extreme impact … [and] human nature makes us concoct explanations for its occurrence after the fact” (xxii). Suffice it to say, many people would agree that Trump would fall under this category and then work to explain his election. In a sense, this project enacts a similar mode of thinking in citing Trump as evidence of our “post-truth age.” But ANT helps us work against this instinct to explain things after the fact and, instead, carefully describe the network that gave rise to this phenomenon. Sure, this is a much more complex, slow, and messy project, but proponents of Latour and ANT would argue that it also creates a much fuller and more honest understanding of our sociology.

Discussing the concept of redundancy in complex systems, Taleb introduces a concept called tinkering, which describes “when an organ [or object or network] can be employed to perform a certain function is that is not its current central one” (317). He uses aspirin as an example. Originally developed to reduce fevers, aspirin is now widely used for its analgesic and anti-inflammatory properties. New situations call for new responses. The same may be said of the EPA. It is not a static network, and thus, may adapt to new uses and purposes amidst its evolution. While one can see this as justification for fearing that the EPA could become a toothless agency that merely works to stay out of the way of economic progress, one could also see it as an opportunity for the EPA to adapt to new issues, such as climate change, climate justice, and environmental racism.

The changes to EPA.gov, the transition from EPA-1 to EPA-2, should not be seen as linear, progressive, or irreversible. Each is an actor-network, capable of flux, deterioration, or new alliances. There is no straight, progressive path but, rather, opportunities for splicing and revisions. This is why the project of rationality should be abandoned when trying to understand science. EPA-2 is not a conclusive representation of science, the environment, and governmental policy. Neither was EPA-1. Rather, by tracing the associations, the alliances, the networks, one can see that each version of EPA.gov articulates specific goals and values. The politics cannot be separated from the science, and the science cannot be separated from the politics. It is up to the actants within the network (politicians, scientists, carbon emissions, lab equipment, American citizens, tax dollars) to “act.” Understanding sites of discourse, such as the two versions of EPA.gov, as actor-networks reveals the potential for agency that can build the type of EPA that most effectively “protects human health and the environment” (EPA).

In Post-Truth Rhetoric and Composition, Bruce McComiskey references the Oxford Dictionary’s definition of post-truth: “Relating to or denoting circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief.” He then goes on to call this type of discourse an “unfair rhetoric” (6). But one must ask: unfair for whom? With a Latourian approach to post-truth rhetoric, one will reject this division between objective facts and personal belief in recognition that those two aspects work together to form our modern agora, our sociomaterial discourse. There can be no full assessment without the realization that the subjective and objective are two sides of the same coin. Even if there is a purely objective reality, we cannot access it as humans given our reliance on filtering reality through our senses. This is why we cannot disregard the emotional, the personal—the human—when analyzing science. The post-truth age shines a light on the shifting balance between objective information and subjective beliefs. Which side will win? A Latourian approach to this form of rhetoric demonstrates that, in fact, the answer must always be both.