“Stand With Standing Rock Nov 11-15 2016” by Flickr user Leslie Peterson. Photo taken in November 2016 (CC-BY-NC-2.0).

A Media Cloud & Global Voices NewsFrames Collaboration

Written by Natalie Gyenes, Connie Moon Sehat, Sands Fish, Anushka Shah, Jonas Kaiser, Paola Villarreal, Simin Kargar, Cindy Bishop, Rahul Bhargava, Rob Faris & Ethan Zuckerman

As part of our collaboration with MediaCloud.org, this article can also be found on the MediaCloud website.

The deadline for Standing Rock campsite residents to depart their campsites along the Missouri River occurred last Wednesday. The evacuation deadline passed at 2pm local time, coincidently marking a two-year effort to prevent the construction of the Dakota Access Pipeline (DAPL), a conduit spanning 1,172 miles with the purpose of transporting crude oil from North Dakota to Illinois. Representatives from approximately 300 of the 566 recognized Native American tribes in the United States actively participated in the effort at the Standing Rock Reservation in North Dakota since April of 2016.

On December 5th, 2016, the self-declared water protectors and their allies celebrated the statement released by the US Army Corps of Engineers, blocking approval of an easement needed to continue working on the Dakota Access Pipeline. This celebration was short lived; within four days of his inauguration, President Trump signed two executive orders to advance the construction of the Dakota Access Pipeline, as well as the Keystone XL pipeline, which former President Barack Obama rejected in November 2015.

The facets of this movement that have been covered by mainstream and local media are broad – as complex and varying from the focus on Shailene Woodley’s or Scarlett Johansson’s role as activists, to the involvement of US veterans as allies, to water rights for indigenous peoples and the connection with climate change more broadly. Media stories focusing on issues like the Dakota Access Pipeline can be presented through a variety of narratives, and the framing used by media outlets helps to determine how we, as readers and information gatherers, understand a topic.

To illustrate how the media framing mattered in coverage of the Dakota Access Pipeline, we examined two different, and sometimes conflicting, ways that the participants were characterized: protectors versus protesters.  Overall, while the media tended to characterize participants as “protesters,” a close analysis of media coverage around the events shows a more complicated story, one that opens up questions about the representation of Native American sovereignty in the 21st century.

Protectors and protesters: which view of events?

“We are protectors, not protesters,” declared Iyuskin American Horse in August 2016 as he sought to explain his efforts to battle the pipeline.  Despite this and similar declarations, a macro-level view of media coverage shows us that this was not how the news tended to portray Standing Rock participants.  Using Media Cloud, an open source and open data media analysis platform developed at the Berkman Klein Center for Internet & Society and the MIT Media Lab, we analyzed total media attention, or the total number of stories mentioning the Dakota Access Pipeline from a digital media collection of 819 sources. It became clear that “protester” was clearly more favored by media sources than “protector” in the late Fall and early Winter timeframes:

Language and frequency of “protesters” vs. “protectors.”

And yet, a focus primarily on protesters provides a narrow view of events – specifically peaks in coverage around protest in October and November – and does not correspond with the overall depth and increase in media coverage about Standing Rock.

From its beginnings in April, representatives of the “spirit camp” at Standing Rock characterized their effort as one to “protect the water.”  Significant news coverage of the effort only began months afterward, as the attempt to halt the Dakota Access Pipeline began to result in increased conflict, arrests, and violence.

Overall coverage of “Standing Rock” in US mainstream, regional, top online, digital, and partisan news sources (04/16-05/16 coverage not shown was also near 0 sentences per day)

Several events in September led to the introduction of additional themes, outside of the ‘water protection’ framing. Despite the ongoing court cases related to the pipeline that expanded areas of protection to include sacred sites within disputed Sioux territory, the Dakota Access company began construction. With the destruction of burial sites, confrontations in early December “turned violent,” with guard dogs and pepper spray unleashed upon participants.

Ultimate responsibility for the events differ according to accounts, but this escalation of tension also resulted in the expansion of the story into the issue of Native American sovereignty and onto the international agenda. During his address to the United Nations Human Rights Council in Geneva, Chairman Dave Archambault II of the Standing Rock Sioux Tribal Council appealed to his audience for help in stopping the Pipeline, stating that “the oil companies and the government of the United States have failed to respect our sovereign rights.”  The result: a call to halt the construction of the pipeline and for the right of indigenous consultation by Special Rapporteur on Indigenous Issues Victoria Tauli-Corpuz on September 22.

In November, continued violence and continued international attention meant continued news coverage even as winter weather approached.  A second statement from the United Nations by Special Rapporteur on the Rights to Freedom of Peaceful Assembly and Association Maina Kiai on November 15 continued to call for the halt of the pipeline construction. It focused on the “inhuman and degrading conditions in detention” of some 400 people as well as the “militarized response” to protests.  Immediate events, at least, did not indicate that those on the ground were listening: on November 22, law enforcement officials justified the use of water cannons upon protectors in subfreezing temperature in addition to tear gas and rubber bullets in order to “control crowds” and fires in the area.

Coverage reached its peak in early December, with the halt to pipeline construction made possible through the denial of a permit by the Army Corps of Engineers on December 4.

National pipeline, but local and episodic coverage

Although the issues around Standing Rock eventually gained international attention, this attention came relatively late during the DAPL standoff.  News coverage began in April, but was infrequent and dominated by local papers or bloggers.  This trend continued throughout the year. By investigating how media sources link to one another online, we were able to investigate the contributors to the DAPL conversation in the digital media ecosystem.

The link network map below displays sources cited most frequently within this conversation, as well as list of most shared articles on social media. The map illustrates that the key sources driving the conversation around the Dakota Access Pipeline events were not mainstream news outlets. They were in fact local newspapers such as the Bismarck Tribune that provided live coverage of events around the pipeline. Fact-based outlets dedicated to the cause publicizing Dakota Access LLC's perspective such as daplpipelinefacts.com, and activist media like Counter Currents News continued to carry coverage of movement. These link network maps, as they are visualizations of web hyperlinks, generally reflect sources people perceive to be authoritative – the prominence of local news outlets in this map suggests that national media’s attention to the story has been episodic, rather than sustained throughout the year.

Link network map, visualizing the inlinking behavior between sources that covered DAPL.

Another interesting finding is the absence of Wikipedia as a frequently linked-to source. In a number of our other research projects using Media Cloud, Wikipedia shows up in virtually every network map, as the Wikipedia article is often seen as an authoritative source. Here, as mentioned above, the central information authorities are local news sources and fact-based outlets.

The late and episodic nature of Standing Rock coverage was exemplified also by the most popular story of the period on social media:  journalist and investigative reporter Amy Goodman facing jail time for her reporting on DAPL protests:

Buzzsumo, a social media analysis platform, shows the most shared stories on the topic of DAPL across Facebook (the query terms for this remain the same as for the Media Cloud search: DAPL or (dakota AND pipeline).

Of all stories published about DAPL in 2016, about 4% spoke about Amy Goodman. Yet the attention to Goodman’s reporting in the Fall contributed to increased attention to Standing Rock overall, since her celebrity and visibility likely enabled some readers to encounter DAPL issues for the first time.  Looking at the content most frequently shared on social media, and comparing differences with the stories most frequently linked to in this information network, coverage of Amy Goodman may also have contributed particularly to the visibility of the anti-DAPL actions.

No clear, authoritative narratives…

The absence of Wikipedia as a frequent linked-to source seems to indicate the difficulty in finding strong, single narratives in the coverage of Standing Rock, something we also found through clustering common language usage among different media sources. This can at times help identify the dominant narratives around DAPL in the digital media ecosystem, and which sources cover such narratives more frequently. To illustrate these clusters, we focus on three communities.

A community of DAPL coverage focused on the status, outcome and legality of the issue.

The multiplicity of perspectives and concerns can be seen above by looking at the community of media sources who are focused on the status, outcome, and legality of the issue. A mix of environmental, civil liberties, activist sites, local media, native media, and national primarily left-leaning media were active, giving a sense of complexity at work in this media landscape.

A community of DAPL coverage focused on the ‘environment’ narrative.

With these information networks, we were also able to identify a community of coverage around discussions of the environment.  The above language map identifies a content focus by some media sources that included “global”, “climate”, “renewable”, “pollution”, and “environmental.” The word “protectors” is featured here as well, pointing to a choice by these media sources to feature the label chosen by the movement.

A military, government, and DAPL fact-based community based on shared language use.

Finally, another cluster of word usage emerged around a group of military-focused sources, including the official website of the US Army as well as justice.gov (shown below).  These sources all use similar language when they cover the Dakota Access Pipeline. This is unsurprising, though notable. More noteworthy is the way that the protector movement site sacredstonecamp.org also used some of the same language as the military and governmental sites, as seen below. It was also linked to most frequently by daplpipelinefacts.com, one of the prominent fact-based key information authorities in the digital media network, but also sites sympathetic to the protectors such as sacredstonecamp.org. In the end, this mix of sources represents a ‘monitorial’ frame, in sources are monitoring the language of official pronouncements and responding to them using those same words.

Rather, a contest of narratives:

Reinforcing the multiplicity of communities and shared languages were the multiple topics of narratives themselves.  Using topic model visualizations of text-mining algorithms to highlight key thematic structures embedded within media stories in the Media Cloud database, we found several narratives emphasized at different time points over the course of 2016.   

Topic model visualization, showcasing the prominence of different narratives overtime.

By examining the above visualization, we were able to identify that the narrative focusing on energy, gas, and the oil companies themselves, represented by the red line, was much more prevalent in early 2016. This tapered off by April of 2016, and was replaced with numerous narratives of protesters at Standing Rock (blue line). It is interesting that, at this time, the narrative bridging gaps between related pipeline proposals and overall climate change was significantly smaller (purple line), as was the narrative focusing on water rights (blue-green line).

And the protector narrative?  As shown by this network view of dominant terms around Standing Rock in November 2016, the notion of protectors did finally rise to become a significant feature of news coverage.  In addition to the diversification of tribal identities (alongside the use over native american or indigenous) by the late fall, we can also see how media struggled over other ways to define members of the Standing Rock camps:  

November 2016 network map analysis of most prolific media producers and the contest over terms: activists, protectors, protestors [sic].

Why does the “protector” compared to “protester” frame matter?

On first blush, a high-level view of media coverage surrounding Standing Rock in the latter half of 2016 shows the tendency to characterize the participants as protesters. But looking deeper, we find other factors at work. Analyzing news coverage further, the overall coverage of Standing Rock events reveals inconsistencies and struggles over narrative: late and episodic focus on the standoff, more local than national attention, and the existence of multiple shared language-use communities without resulting in a set of authoritative narratives.

Although the characterization of protester certainly did allow for increased media coverage in the latter part of the year, especially as events became more violent, it privileged a view of events around defiance.  Attempts to characterize the activity around Standing Rock as undertaken by “anti-energy protesters,” for example, allowed an interpretation of events that narrowly focused on environmental concerns in comparison to broader resistance. On the other hand, the language of protectors goes beyond a semantic argument to emphasize what these individuals and communities were fighting for, not fighting against.

In the end, perhaps the news coverage – which includes the earlier inattention and struggles over narrative – reflects the state of our national inexperience with discussions over Native American sovereignty.  The issue of sovereignty includes the right to choose the terms and ultimate meanings of the pipeline dispute. Over the course of 2016, it emerged that choosing the word ‘protector’ privileged the Sioux’s, and any native american tribe’s, right to define the very meaning of their activity as “protection” instead of “protest,” and ultimately their legal right to be consulted with regards to their lands.  

Sovereign consultation may at times include decisions to accept environmental risk and also needs to consider the diversity of opinion among tribal representatives with regards even to DAPL (such as Archambault’s call to activists and protectors to go home after the early December pipeline halt).  The increased significance of the “protector” language by the year’s end allows us to hope that no matter the outcome of DAPL construction, members of the Standing Rock camps, and news articles about them, have helped United States citizens to become more practiced at thinking about continued relevance of land issues for Native Americans.  Future media coverage could continue to help us think about issues related to tribal sovereignty with more depth and complexity.