Notley and the Beast: An Analysis of the Crisis Communication of Rachel Notley during the 2016 Wildfire

Chaseten Remillard and Sheridan McVean

On 1 May 2016 a small wildfire started southwest of Fort McMurray, Alberta. Wildfires are not unusual for the area; this particular fire was fueled by a combination of dry weather, high winds, and hot temperatures. This time, the combination proved catastrophic. The wildfire continued to grow in size, unpredictability, and power, and it’s magnitude and strength earned it the nickname “the beast.” By May 3, Fort McMurray faced imminent threat and at 5 p.m. that day a mandatory evacuation of the city was ordered.

As a growing cloud of smoke and flame engulfed the city, nearly 90,000 residents of the city began their exodus. Roads swelled with vehicles whose drivers navigated through walls of burning trees and buildings. Black smoke limited visibility and a hazy, nightmarish landscape prevailed. When, on July 1, the provincial state of emergency was lifted, the fire had raged for 66 days, destroyed 2,400 homes, consumed 590 acres of boreal forest, and caused over $3.5 billion in insurable damage. To date, it is the most expensive disaster on Canadian record.

Almost a year prior, a different type of news story dominated the Albertan public sphere. On 5 May 2015, the Alberta NDP scored an upset victory over the long-serving Alberta Progressive Conservative Party to become the government of Alberta. Leading the Alberta New Democrats was Rachel Notley, subsequently the first NDP premier of Alberta. The May 2016 Fort McMurray wildfire became the first and most significant test of the new government’s response to a crisis situation.

In general, crises often act as litmus tests for leadership legitimacy. How a leader responds to crisis can quite literally make or break their career, and in this case, their government. For Notley, the stakes were particularly high. In the early days of her government, important questions remained about its ability to shepherd the province out of an increasingly deep economic recession caused by depressed world energy prices, to address national and international stakeholders around important infrastructure projects such as pipelines, and the potential imposition of new taxes and government royalties on Alberta’s energy industry.

During the 2015 Alberta general election, the NDP had campaigned on a policy to review the royalties charged by the Alberta government on oil and natural gas produced in the province. Royalties are similar to taxation, but are premised on the fact that the vast majority of oil and natural gas in Alberta is owned by the provincial government. The government sets the price at which the energy industry is allowed to remove the oil and natural gas. Royalties are charged in addition to corporate or business taxes.

Fort McMurray and the surrounding area, called the Regional District of Wood Buffalo, is the centre of Alberta’s oil sands development and a lightning rod for critics of the tar sands, dirty oil, and climate change. Some of these critics saw “justice” in the fact that this area was suffering from the impacts of climate change since it is populated largely by those thought culpable for the effects of fossil fuel development.

Moreover, the vividness of the Fort McMurray fire was not just physical, but also virtual. Captured by smart phones and dash cams, the images of the wildfire streamed out to the world in high definition. The speed and ferocity of the fire and the rapidity of the evacuation had largely locked traditional news sources out of the city, but newsrooms swelled with visual documentation from thousands of embedded citizen journalists. These raw, uncut, and unfiltered first-person narratives of the disaster were visceral, shocking, devastating, and abundant. Countless images, videos, and personal accounts streamed out onto YouTube, Facebook, Twitter, and Instagram, and perhaps unlike any other Canadian natural disaster in history, the Fort McMurray evacuation went viral. Traditional media outlets, such as television network news, used the dark and dramatic video shot and posted on social media by citizens fleeing the wildfire to enhance their own news coverage of the fire.

The sheer magnitude of the fire, the expanded publicity facilitated by the viral nature of the images it produced, and the political climate of a newly established government makes the Fort McMurray fires a particularly unique case of crisis communication, and one that reveals important elements of Rachel Notley’s leadership and the impact of social media on public governance. In this chapter, we hope to determine the extent to which the premier’s initial personally communicated responses to the Fort McMurray wildfire addressed the visual and online framing of the event as set by social media images and, in so doing, reinforced the mandate and legitimacy of her premiership.

To do so, we consider three distinct sources of data: the images of the Fort McMurray evacuation, as recorded on several widely viewed YouTube videos posted during or immediately after the evacuation; the public comments posted to news stories that either incorporated or linked to footage from those privately produced videos; and Premier Notley’s first five press conferences and updates that occurred during the first three days after the mandatory evacuation was ordered.

We conclude that in her crisis communication, Notley used an effective strategy to emphasize “bolstering” and “corrective action” messages.1 Furthermore, although the narrative of the fire, as set by online images and commentary, framed the fire differently than Notley and her government did, her crisis communication efforts implicitly addressed many of those alternative frames, and did so in a manner that emphasized collaborative action and positive outcomes.

Setting the Stage for Crisis: The Importance of Initial Communications

Our contention is that the initial organizational communication responses to a crisis situation can be very enlightening. Because public and media interest is so focused on the crisis, the initial communications messages from the organization deemed responsible can reveal both expected organizational characteristics and those that otherwise could have remained hidden. In other words, organizations dealing with the pressure of a crisis tend to make or break their public responses early on. Early organizational communication missteps or misstatements can reveal unintended negative organizational characteristics and cause long-term reputational issues. Conversely, successful crisis communication typically expands from timely, appropriate, and well-measured organizational responses early on in the crisis and reinforces existing key points of legitimacy related to that organization.

In those instances when a politician or leader speaks about the scope, severity, impact of a crisis and the mitigation strategy by which they hope to bring it under control, those statements function to define for the public what the crisis is and how it is best managed. By framing the crisis as such, successful crisis communication endeavours to set the agenda for the news coverage of a crisis.

Making the Crisis Meaningful: Agenda-Setting and Crisis Communication

Agenda-setting is a well-established and highly studied form of media effect.2 Substantial research over the last forty years has shown that the prominent agenda set by the media influences the expressed agenda of the public that consumes that media.3 Importantly, agenda-setting is not propagandistic, for “the press may not be successful much of the time in telling people what to think, but it is stunningly successful in telling its readers what to think about.”4 So, for example, during an election, the media may not tell the voters how to vote, but the media will set the agenda of what is important to “think about” when voting.5 Those policy issues given most attention by the media predictably become the expressed policy priorities for the public who have consumed that media. So, voters who watch a particular news channel or read a particular newspaper during an election will not have their voting decision directly determined by such coverage, but rather will rank the top issues of the election in alignment with the news coverage they consumed.

Although crisis communication literature uses a different vocabulary than agenda-setting, both share a central concern with message framing and communication effects. In general, two paradigms in crisis communications theory have been dominant in the public relations literature over the past two decades: Benoit’s image restoration theory and Coombs’s situation crisis communication theory.6 Benoit considered communication messages during crisis situations and defined five broad categories that organizations employ to repair their corporate images: denial, evasion of responsibility, reducing the offensiveness of an event, corrective action, and mortification (apology). Two key premises of Benoit’s theory are that an organization must be believed by a relevant audience to be responsible for an act (he claims that the perception is important, not the reality); and that the act be considered offensive.7

On the surface, natural disasters, such as wildfires caused by lightening, would not be seen by a company or organization or individual as responsible. The Fort McMurray wildfire was started by human activity rather than by a lightning strike,8 but police stated they were unable to identify a specific organization or individual who caused the wildfire. However, the evacuation of Fort McMurray and surrounding communities was very much a government action, with the Regional District of Wood Buffalo responsible for mandatory first evaluations and the Alberta government responsible for subsequent mandatory evacuations. As is discussed later in this chapter, social media chatter around the initial mandatory evacuation generated speculation on the causes of the wildfire as well as questions about the need for the evacuation. For this reason, and despite the fact that Benoit’s work was published more than two decades ago and, at that time, he envisioned his theory as one designed for corporations, we will discuss his types of crisis communication messages and assess their applicability to the Fort McMurray wildfire.

In creating situation crisis communication theory, Coombs utilized and added to Benoit’s strategies by distinguishing between strategies intended to change perceptions of the crisis and strategies intended to change perceptions of the organization experiencing the crisis. He also defined diminishment strategies as messaging intended to reduce the negative effects of the crisis or the organizational control over the crisis, and rebuilding strategies as messaging intended to improve the organization’s reputation.9 In addition, Coombs connects Bernard Weiner’s attribution theory with crisis communication. Attribution theory posits that people have a need to search for the causes of events, in this case, crisis events. In other words, people want to identify a cause for a crisis event and attempt to determine who is responsible for it.10

For these reasons, situation crisis communication theory directs a three-step process for communications managers in crisis situations to determine communication messages and responses that are appropriate to the individual crisis situation. First an assessment is made of organizational responsibility for creating the crisis as viewed by stakeholders and/or the public and if the organization is a victim, if an accident caused the crisis, or if the crisis was preventable by the organization. Next the crisis history of the organization is reviewed according to two measures: consistency—if the organization experienced similar problems/crises in the past—and distinctiveness—how well the organization has generally treated stakeholders/people in the past.11

Coombs notes other factors that are important when organizations select crisis response strategies: stakeholder and public assessment of organizational credibility—composed of the expertise and trustworthiness of the organization—and the prior reputation of the organization.12 He cites comments from other crisis experts that during a crisis, the organization must both establish control and show compassion.

In addition, believability of the organization is important, and the speed with which the organization can disseminate its communications messages helps increase believability, assuming stakeholders and the public will actually accept what the organization is stating in its communication messages. Coombs also points out that challenges to an organization and its messages can occur when a stakeholder or credible third party calls the organization’s actions or messages into question.13

Thus, what Coombs adds to Benoit’s categorization is the importance of responsiveness, context, and organizational legacy. Crisis communication messaging must be understood in the context of the specific crisis and in relation to the legacy of the organization communicating about that crisis. Messages may need to be adapted to accommodate or address alternative framings of the crisis, or existent public perceptions of the organization. Thus, timeliness, or the act of “stealing thunder” as Arpan and Roskos-Ewoldsen describe it, is an important consideration in successfully framing a crisis.14 Stealing thunder is the voluntary and proactive disclosure of potentially damaging information by an organization seen as responsible for a crisis situation. In Arpan and Roskos-Ewoldsen’s research findings, an organization was rated as more credible when it proactively steals thunder than when it does not.

Moreover, as Boin and his colleagues point out, this decision-making task for government is determined not only “by crucial leadership decisions but, to a considerable extent, also by the institutional context in which crisis decision making and implementation take place.”15 Wildfires in Alberta are frequent; between 2006 and 2015, Alberta experienced an average of 1,500 wildfires per year.16 Additionally, initial responses to wildfires are generally the responsibility of local municipal governments. Moreover, the Alberta government uses well-established and tested disaster management protocols and has experienced significant previous disasters involving wildfires and floods. In May 2011, for example, a wildfire burnt through the town of Slave Lake; like the Fort McMurray wildfire, the Slave Lake wildfire was also propelled by strong winds. In Alberta’s government emergency management circles, the experience of the Slave Lake wildfire became embedded in the “how to” manual for fighting urban wildfires in the province.

The institutional context in which the Fort McMurray wildfire began was that the local government in the Regional District of Wood Buffalo, and not the Alberta government, were engaged in the sense-making and decision-making tasks about the wildfire. Once the wildfire grew in size and became a crisis, the Alberta government also had to move through these critical tasks of sense-making, decision-making, meaning-making, before eventually declaring that the wildfire was under control. However, given that the fire had been identified as a crisis already, and that a well-established decision-making architecture was in place for fighting such fires, our analysis focuses on the meaning-making task that Premier Notley engaged in during her initial news conferences. Certainly, the Alberta government was responsible for the “on-the-ground” fighting of the Fort McMurray wildfire, and Premier Notley and her government made strategic choices to that effect, but what did they then communicate to Albertans and Canadians to make those choices meaningful? This question is at the heart of both agenda-setting and crisis communication literature, as it is the role of leadership in a crisis to “impute meaning to the unfolding crisis in a way that their efforts to manage it are enhanced.” If Notley failed to do this, her actions and “decisions will not be understood or respected.”17

In a crisis situation, leaders and spokespeople have the opportunity, through press conferences and news releases, to attempt to shape the agenda originally set by the media, and in doing so, to frame the crisis in a manner strategic to both the resolution of the crisis and the benefit of themselves and their organization.

In the three days that followed the announcement of the mandatory evacuation of Fort McMurray, Premier Notley held five separate news conferences. Typically each news conference update began with opening comments from the premier, followed by updates from non-elected government officials, and then a media question-and-answer period. To us, the opening comments for the initial wildfire updates are a particularly rich set of data for the following reasons: these comments were directly from Premier Notley and not filtered; the content of these comments were not set by the media (as questions from the media in the media question-and-answer component of the update could shape the discussion); and having the video record of the initial comments provided the ability to measure the length of the comments and sort the comments by subject category. Such content analysis enabled us to quantify what was prioritized in Premier Notley’s communicative management and agenda-setting of the crisis.

Seeing the Crisis Online: Social Media and a New Age of Crisis

Simultaneous to Notley’s news conferences, during the opening days of the crisis, an abundance of images and videos of the evacuation and wildfire became available through social media and other online sources. These images generated both a visual narrative of the evacuation and a growing online commentary on the wildfire, which in turn generated both supportive and skeptical discourses of the crisis.

Recently, scholars of agenda-setting have turned their attention to the impact of images of crises on public opinion, and the recalling of previous crises such as 9/11, Hurricane Katrina, and the Deepwater Horizon oil spill.18 In these studies, the type and frequency of news images were found to impact the recollection of the crisis event. In other words, the choice and repetition of news images functions to shape the public’s collective memory of an event. As Miller and LaPoe conclude, “society’s visual saturation is an important area of study because visuals can affect the way audiences respond to or prioritize responses to a crisis.”19

The power of images as a visual agenda-setting agent is amplified in moments of crisis for several reasons. First, audiences have a higher level of exposure to crisis coverage than regular news coverage. In general, audiences consume crisis information at higher rates than regular news.20 This higher rate of exposure is a result not only of extended coverage often related to crisis situations, but also new digital technology. The internet enables people to tailor their news diet and to search out information that is of interest to them.21 In times of crisis, audiences are therefore able to find and consume even more information across multiple news sources and through social network streams.

The highly affective impact of crisis images is a second reason why visual agenda-setting is so powerful. In general, shocking imagery increases an audience’s attention to and consumption of news.22 Also, generally speaking, images function as important mnemonic aids and can stand as iconic representations for entire political events.23 Images can, as well, stimulate immediate and long-term emotional reactions to events, and “audiences respond to media messages using the same dimensions of emotions used in responding to real-life experiences.”24

In comparison with regular news images, images of crisis are more engaging and more threatening. However, despite the impetus of photojournalists to capture distinctive and unique images of crises, and a marked increase in the public’s appetite for images of crisis, some contend that even these images are conventional.25 Wright describes how images of disaster follow predictable narratives within the news, and that these characteristic images of disaster facilitate easier editorial decisions, aligning with audience expectations to “numb down” audience reaction: “the repetitious use of ‘TV codes’ and the reporting of disasters according to predetermined formulae has a numbing effect on the audience.”26 Therefore, although viewer ratings of news broadcasts increase, Wright questions whether or not news audiences have mentally “switched off,” even in the face of disaster.27

Here, once again, the impact of digital technology amplifies the power of images to function as agenda-setting agents. Advances in technology and the ubiquity of internet coverage now make the transmission of digital images easy. As a result, social media content and internet news is becoming more image rich. Moreover, the ease with which high-quality images can be distributed on contemporary digital networks means that audiences can easily monitor a crisis situation through continuous news updates and also watch it in real time as it unfolds. Importantly, these images of contemporary crises are not always vetted or curated through traditional news agencies. As such, these images function differently than traditional news images, as they triangulate the crisis from a multitude of citizen perspectives. The handheld smart phone in everyone’s pocket acts as a phalanx of embedded photojournalists. The imagery it captures is raw, immediate, personal, and palpable.

Moreover, the instantaneous, interactive, dialogic, and global characteristics of social media enable images to not only broadcast crises outside the parameters of traditional reportage, but also to stimulate and facilitate public debate and discussion around the meaning, direction, and consequences of the crisis. In other words, the capture and dissemination of crisis images through social media creates both a visual agenda-setting function, but also a dialogic and public agenda-setting one as well.

The internet, and by extension social media, has long been heralded, or feared, as a new public sphere. Since social media commentary is readily available and not controlled by editorial decisions to the extent that mainstream media is, some predict that social media comments can potentially enable minority opinions more voice than traditional news media.28 In terms of crisis situations, such alternative voices may set an agenda that quite starkly contrasts that of official proclamations. This poses an interesting question: To what extent should leadership monitor and respond to such frames and make sense of the crisis in relation to this visual narrative and the consequential agenda it initiates, as set by social media commentary?

As an aside, Marland has written about the use of branding and marketing techniques in government and politics, particularly in regards to the Canadian federal government.29 We believe that the sudden growth of the Fort McMurray wildfire severely restricted the Alberta government’s ability to brand or use the marketing techniques described by Marland in its early communication. The widespread availability of social media images, as described in the previous section, provided the context in which the crisis was subsequently understood in a way not planned or prescribed by Premier Notley or her government. And she had very little time to plan prior to her first news conference on the wildfire.

Method

Typically, agenda-setting effects are measured through a comparison of two distinct metrics, a quantitative content analysis of media content related to an event (e.g., an election, a policy change) and a quantitative or qualitative survey of audience opinions or recollections of that same event.30 The impact of the agenda-setting effect, therefore, is understood normatively, and these quantitative results are considered robust in terms of both generalizability and predictability. Our approach is different. Since we seek to analyze the extent to which the potential agenda-setting effects of images and social media commentary shaped the crisis communication of Premier Notley, we compare the content of images associated with the Fort McMurray wildfire, the content of comments associated with online news stories of the crisis posted during the opening days of the wildfire, and the content of Notley’s first five press conferences.

Content coding for the image set was facilitated through the use of Google Images. Using the search term “Fort McMurray wildfire,” a set of 150 images was collected. Once gathered, the images were coded inductively. This enabled us to approach the data without preconceived categories, and instead to let meaningful categories emerge from the data.

Inductive content coding was also used to develop meaningful content themes from the comments of two online news stories related to the wildfire, posted by the CBC on 4 and 5 May 2016.31 Although only a small sample of the vast amount of coverage the wildfire garnered, these news stories provide an insight into the emergent public response to the fire. The content of all comments posted to the two stories during the same time period as the initial five Notley news conferences, were collected as well. An initial open coding and word count was conducted; secondary coding then provided us a “means of describing the phenomenon, to increase understanding and to generate knowledge.”32

A similar process was used to code the content of Premier Notley’s addresses. We selected Notley’s statements from the official updates held at the Emergency Operations Centre.33 They were available on the “YourAlberta” YouTube channel, which is maintained by the government-run Alberta Public Affairs Bureau. The premier’s initial comments from the first five government updates were coded inductively. The length of the premier’s initial comments in these updates ranges from 3:00 minutes in the first update (the shortest) to 11:37 minutes in the fifth update (the longest of the five under study). During the third update, the Honourable Danielle Larivee, then the provincial minister of municipal affairs, made the initial comments instead of Premier Notley, as the premier was visiting the wildfire area and therefore unavailable.

Given that these updates were consisted of speeches delivered by Notley, we used time rather than word count as our unit of measurement. This enabled us to account for emphasis expressed through the form of delivery, pacing, and non-verbal communication. By measuring the time the premier took to relay her messages we were able to remark on what the performative salience of each of her points were.

Findings

From the content coding of the visual data, commentary data, and the Notley press release data, we found several contrasting agendas. The visual data emphasized the evacuation, the scope and severity of the fire, and the urban context in which it took place. The online comments prioritized support and concern for those impacted, and the causes and magnitude of the fire. Importantly, the commentary also emphasized negative emotions, conspiracy theories related to the fire’s cause, and judgment of those impacted by the fire as responsible or deserving of the fire because of their association with so-called dirty oil. In Notley’s updates, she set an agenda that emphasized governmental and intergovernmental actions to address the evacuation and magnitude of the fire; she also offered support and sympathy for those impacted, and thanked first responders and industry for their efforts.

Seeing the Beast: The Visual Agenda

The visual data was categorized according to eight content codes: evacuation (28 per cent); scope and severity (23 per cent); urban fire (17 per cent); forest fire (10 per cent); rural fire (3 per cent); destroyed property (10 per cent); and first responders (9 per cent).

Evacuation (28 per cent) was the most prevalent image code. Evacuation images showed cars moving in long lines, often in front of or between large walls of flames. The images were variously composed; some were screen shots from dash cams and smart phones. Others were shot from a distance, revealing the number of vehicles, and by implication, people impacted by the fire. Overall, the evacuation code showed the size, speed, and embodied experience of those fleeing the fire.

Scope and severity images were comprised of several different visual representations of the fire, and these account for 23 per cent of the images. Some were aerial or satellite images of the fire, which documented its geographical enormity. Others consisted of maps, which again distilled the size and scope of the fire. Finally, still others depicted eerily beautiful vistas that present the landscape at a distance and the fire encompassing the horizon or plumes of smoke rising into the sky. These images all express the uncontrollable magnitude of the fire and do not specifically address the displacement of people.

Urban fire (17 per cent), forest fire (10 per cent), and rural fire (3 per cent) collectively account for 30 per cent of the images; as such they comprise the largest category of images. However, it was still important for us to differentiate the contexts in which the fire was pictured. Each of the “fire” categories depicts flames or smoke, or both, without a visual representation of evacuation, but they do so in different contexts. Urban fire images depict fire consuming or threating buildings, residential homes, and businesses. Forest fire images show trees and forests engulfed in flames. And rural fire images depict pasturelands or agricultural fields in flames, or under threat of flames.

The last two coded categories of visual data are destroyed property (10 per cent) and first responders (9 per cent). Images coded as “destroyed property” show the aftermath of the fire. These images show burned cars, furniture, or homes. Images coded “first responders” depict any first responder in the act of conducting their job during the fire.

Overall, when the top three categories of images are considered together, the visual agenda of the fire is evacuation, magnitude, and urban destruction.

The Beast Online: The Social Media Agenda

The textual data gathered from the comments sections of two news stories associated with the fire during the first few days of the crisis generated eleven different content categories: support (24 per cent); cause of fire (22 per cent); magnitude of fire (8 per cent); fire management (10 per cent); government distrust (9 per cent); judgment and mockery (9 per cent); evacuation (4 per cent); negative emotions (8 per cent); positive emotions (3 per cent); first responders (3 per cent); and media control and bias (2 per cent).

The support code was the most prevalent of all eleven content codes. Comments associated with that category included statements of empathy and sympathy for the people impacted by the fire. These included calls and suggestions for donations, as well as statements of solidarity and caring. The support code is differentiated from the positive emotion code in that the later captured statements of gratitude, thanks, and positive outcomes. In other words, “positive emotion” was a code we used to demark, typically, comments by people who had been impacted by the fire.

Interestingly, despite a dominant agenda set by the visual data, evacuation was a minor component of comment content. The code “evacuation” was used to categorize comments that referenced the logistics and undertaking of the evacuation. Instead, causes of fire was the second most discussed topic in the commentary of the news stories. The list of causes discussed in the commentary section ranged from arson to climate change. Of the causes listed or discussed, natural causes (30 per cent) was the most frequently cited. The fire’s natural causes were speculated to include warmer than usual weather, lighting storms, and high winds, for example. Almost equally present in the discussion of causes was a category we labelled “conspiracy” (24 per cent). Conspiracy causes ranged from the coming of the apocalypse to tailing ponds. The unifying element of this code was expressed in the assumption that the fires were caused by mismanagement, malfeasance, or malice. The fires were positioned as a result of wrongdoing.

Finally, climate change was categorized as its own separate cause category within the data because some comments framed climate change as a natural (albeit human-initiated) cause of the fire as it resulted in unseasonably high temperatures and low precipitation. Others, however, cited climate change as a direct result of petroleum extraction and use. These comments tended to align with the tone and intent of the conspiracy code, however they maintained a more implicit culpability.

After looking at both the conspiracy theme and those comments related to climate change that implicated the oil sands as a contributor to global climate change, we observed that a significant amount of online commentary placed both implicit and explicit blame for the fire on the oil sands industry. These types of comments were reinforced by the code “judgment and mockery” (9 per cent), which included comments that framed the fire as retribution for working in the oil industry, or supporting the oil industry. These types of comments often made a karma connection that assumed that Fort McMurray was getting “what it deserved” because it had benefited from the extraction of “dirty oil.”

This sentiment also aligns with comments coded as “government distrust” and “media control and bias,” both of which voice skepticism about the truthfulness of the government and the media. Implicit in these comments is the suspicion that the whole story is not being shared with the public, and often, as with the more conspiratorial comments, an underlying sense that the fire was a result of an industry blunder or cover-up.

Finally, many comments expressed overt negative emotions associated with the fire. These “negative emotion” comments included fear, sadness, depression, shock, and horror. Although constituting only 8 per cent of the commentary content, these negative emotions reinforce the power of images to generate strong affective impacts on viewers.

Notley’s Response: The Premier’s Agenda

The content gathered from the first five government updates generated nine different categories: evacuation (23 per cent); scope and magnitude (16 per cent); support and sympathy (14 per cent); government actions, assessments, and plans (14 per cent); inter-government co-operation (13 per cent); motivation and gratitude (7 per cent); first responders (5 per cent); industry co-operation (4 per cent); and fire prevention (1 per cent).

The two most prevalent codes in Notley’s updates were predictably related to the evacuation of Fort McMurray and the scope and magnitude of the fire. The impact and logistics of the fire on Fort McMurray and its residents constituted nearly 40 per cent of the total time of Notley’s speech. These codes included topics such as instructions to evacuees, updates on the fire’s position, and statements about where evacuees were currently being housed.

Unlike the previously discussed content, government and inter-government actions featured prominently in Notley’s speeches. The codes “government actions, assessments, and plans” and “inter-government co-operation” collectively comprise nearly a third of Notley’s speaking time, accounting for 27 per cent of the total. These codes included statements about government decisions, explanations about implementing a state of emergency, details about how government agencies were assessing the safety of key infrastructure elements, and the premier’s personal plans to visit different locations or meet with different stakeholders.

The codes “support and sympathy” and “motivation and gratitude,” which both express positive emotional messages or material and emotional support for those directly impacted by the wildfire, or Albertans in general, combined to account for 21 per cent of the Notley updates. We continued to differentiate these codes, as the term “support and sympathy” denote statements or actions that have occurred and that are material or emotional in nature. By contrast, the code “motivation and gratitude” captures comments that are unifying in nature, such as “we are strong and will overcome this,” as well as statements of thanks to specific individuals, groups, or agencies.

Nearly equally represented in the content of Notley’s updates were industry and first responders. This content was categorized in the codes “industry” and “first responders,” respectively. The first responders code was used to categorize comments related to the efforts and progress of various first responders, police, fire fighters, and emergency medical services in their collective efforts to fight the fire and provide support for citizens. The industry code specifically refers to the oil industry and those companies that have operations in the region impacted by the fire. Those of Notley’s comments that were coded “industry” included statements of co-operative actions, updates on the support companies provided to evacuees, and the status of employees impacted by the evacuation order.

Taming the Beast: Trends in Notley’s Crisis Communication

Our primary finding is that Notley’s updates frame the wildfire crisis in a predictable manner that emphasizes “bolstering” and “corrective action” messages, as outlined by dominant theories of crisis communication. However, her updates did not fully address the alternative agendas set by the visual and online narratives of the wildfire. In avoiding certain points of the visual and online agendas, Notley was better able to emphasize a cohesive, collaborative, and positive response the fire.

When the content of Notley’s speeches are considered from the perspective of dominant crisis communications strategies, her communication followed a predictable pattern. The premier primarily used what Benoit has labelled “bolstering messages,” meaning she stressed positive aspects, for example by thanking the firefighters and emergency workers responding to the fire, as well as thanking the work of other governments and the energy companies with operations in the area. Overall, the tone of her speeches was positive, as the prevalence of the codes “support and sympathy” and “motivation and gratitude” reflect.

The premier also used what Benoit has termed “corrective action” messages, which state specific actions the government was or would be doing to help make the situation better and to support evacuees. Although the scope and magnitude of the fire was a prevalent content category of Notley’s statements, as it was in both the visual and the online commentary content, details of the wildfire’s spread were nearly always framed by Notley in relation to efforts to fight the fire and mitigate its negative impacts on citizens and property. This emphasis on action is reflected in the high prevalence of content that detailed the various government and intergovernmental actions taken.

Moreover, Notley also addressed the actions of both first responders and the energy industry in their efforts to combat the fire. In considering the role of government during crisis situations as what Boin and his co-authors describe as “meaning-making,” the premier actively generated a narrative of collaboration between government, industry, and first-response agencies. She also heralded firefighters, police, and emergency response personnel as superheroes fighting against “the beast.” She consistently praised the energy industry’s contributions during the initial and subsequent evacuations. She discussed the work of municipal and local leaders, and she outlined her government’s actions and plans. Taken together, nearly 40 per cent of the entire content of her updates was dedicated to chronicling the different actions, agents, and collaborative efforts undertaken to stop the fire and keep people and property safe. Furthermore, over the course of the five news conferences, the premier changed from expressing only sympathy for the evacuees in the first two conferences to providing more information and reassurance to evacuees during the three subsequent addresses. This underscored the government’s determination to take action to further and deepen its support of the evacuees.

Notley emphasized both “bolstering” and “corrective action” messages, and set an agenda that was both positive in tone and anchored in action. Her key messages were rooted in positive emotions, concrete and corrective actions, co-operation, and progress. Thus, although Notley did not specifically address the shocking visual nature of the fire, which in the visual content was expressed through images of fiery escape, expansive horizons of smoke, and urban destruction, she was able to combat that visual agenda through her own agenda of sustained and collective action and positive, motivational, and gracious sentiment.

A major category of the online commentary agenda was the causes of fire. The premier made no comments about the cause of the wildfire, although she did slightly discuss fire prevention (1 per cent). Similarly, despite the significant distrust in the government voiced in the online content, Notley did not attempt to justify or defend the mandatory evacuations and did not mention the possibility that the government may have done a better job preventing wildfires. In this way she avoided altogether any discussion of culpability in her updates. Again, through an emphasis on bolstering and corrective action, Notley set an agenda that did not prioritize looking backward at causes, but rather focused on current actions aimed at improving and mitigating the situation. So, while the government took no responsibility for anything related to the start of the wildfire or the initial mandatory evacuation, the frequency of the news conferences and the premier’s comments at the news conferences demonstrated that the government was taking action to deal with the wildfire situation and the plight of the evacuees.

Finally, Premier Notley seized an opportunity to emphasize collaboration, gratitude, and co-operation with the oil industry. Her promises to re-evaluate royalties and carbon taxes convinced most Albertans that her government would strike a more adversarial role with oil industry than previous governments. To some, this was a benefit, as it showed that the Notley government would strike a seemingly more responsible path in terms of carbon emissions and environmental oversight. To others, such policy changes could only deepen the growing recession in Alberta and stymie economic growth. The online commentary raises this debate in the content coded in the category “judgment and mockery” and the conspiratorial elements of the “causes of fire,” which include such things as industry pollution or the connection between industry, global warming, and increased wildfires. In other words, albeit in a more extreme, insensitive, and vitriolic manner, some of the sentiments raised in the online commentary speak to the very issues that propelled Notley to power and made her a controversial figure in relation to the oil industry. In not addressing the causes of the fire, and through emphasizing the responsible actions of the oil industry, Notley once again emphasized messages that were both “bolstering” and “corrective.” In so doing, she unified Albertans as a collective and avoided potentially divisive topics of culpability.

Putting the Beast to Rest

“The beast” raged for over sixty days, destroyed homes, and displaced thousands. Images of the fire generated powerful emotions and brought the shock and horror of the event to countless smart phones, tablets, and television screens. Online commentary and social media enabled citizens to comment and question the events of the fire. Premier Notley, new to power, faced an unparalleled test of her leadership. But in those opening days of the crisis, through consistent and purposeful crisis messaging, Notley set the agenda of the fire in terms of government action and co-operation and positive support and sympathy. Although she did not explicitly address some of the most prevalent concerns raised by the narrative set by online images and commentary, her emphasis on “bolstering” and “corrective action” messages aligned with crisis communication best practices and enabled her to set an agenda that reinforced her leadership style and capacity.

Notes

1. William L. Benoit, “Image repair discourse and crisis communication,” Public Relations Review 23, no. 2 (1997): 177–86.
2. Maxwell McCombs and Donald Shaw, “The agenda-setting function of the mass media,” Public Opinion Quarterly 36, no. 2 (1972): 176–87.
3. Ibid.; see also Shanto Iyengar and Donald Kinder, News that matters: Television and American opinion (Chicago: University of Chicago Press, 1987), and Kimberly Goss and Sean Aday, “The scary world in your living room and neighborhood: Using local broadcast news, neighborhood crime rates, and personal experience to test agenda-setting and cultivation,” Journal of Communication 53 (2003): 411–26.
4. Bernard C. Cohen, The Press and Foreign Policy (Princeton, NJ: Princeton University Press, 1963), 13.
5. McCombs and Shaw, “The agenda-setting function of the mass media.”
6. See Benoit, “Image repair discourse,” and Timothy Coombs, Ongoing crisis communication: Planning, managing, and responding (Thousand Oaks, CA: Sage, 2015).
7. Benoit, “Image repair discourse,” 178.
8. Nia Williams, “Humans probably caused the Fort McMurray Wildfire: Police,” Reuters, 14 June 2016, http://www.reuters.com/article/us-canada-wildfire-cause-idUSKCN0Z02OO (accessed 22 November 2016).
9. Coombs, Ongoing crisis communication, 147.
10. Timothy Coombs, “Attribution theory as a guide for post-crisis communication research,” Public Relations Review 33, no. 2 (2007): 135–9.
11. Ibid., 137.
12. Coombs, Ongoing crisis communication.
13. Ibid.
14. Laura Arpan and David Roskos-Ewoldsen, “Stealing thunder: Analysis of the effects of proactive disclosure of crisis information,” Public Relations Review 31, no. 3 (2005): 425–33.
15. Arjen Boin, Paul Hart, Eric Stern, and Bengt Sundelius, The politics of crisis management: Public leadership under pressure (Cambridge: Cambridge University Press 2005), 12.
16. Alberta Agriculture and Forestry, “Wildfires 10-Year Average,” 15 April 2015, https://wildfire.alberta.ca/resources/historical-data/documents/Wildfires10YearAverage-Mar08-2017.pdf (accessed 14 August 2018).
17. Boin et al., The politics of crisis management, 13.
18. Shahira Fahmy, Sooyoung Cho, Wayne Wanta, and Yanghoi Song, “Visual agenda-setting after 9/11: Individual emotion, recall and concern about terrorism,” Visual Communication Quarterly 13, no. 1 (2006): 4–15; Andrea Miller and Victoria LaPoe, “Visual Agenda-Setting, Emotion, and the BP Oil Disaster,” Visual Communication Quarterly 23, no. 1 (2016): 53–63; and Andrea Miller and Shearon Roberts, “Visual agenda-setting and proximity after Hurricane Katrina: A study of those closest to the event,” Visual Communication Quarterly 17, no. 1 (2010): 31–46.
19. Miller and LaPoe, “Visual Agenda-Setting,” 53.
20. Ibid.
21. David Taras, Digital Mosaic: Media, Power, and Identity in Canada. (Toronto: Toronto University Press, 2015).
22. Silvia Knobloch, Matthais Hastall, Dorf Zillmann, and Coy Callison, “Imagery effects on the selective reading of Internet newsmagazines,” Communication Research 30, no. 1 (2003): 3–29.
23. David Perlmutter, Photojournalism and foreign policy: Framing icons of outrage in international crisis (Westport, CT: Greenwood, 1998).
24. Fahmy et al., “Visual agenda-setting after 9/11,” 8.
25. See Andrew Mendelson, “Effects of novelty in news photographs on attention and memory,” Media Psychology 3 (2001): 119–57, and Terence Wright, “Collateral coverage: Media images of Afghan refugees, 2001,” Visual Studies 19, no. 1 (2004): 97–112.
26. Wright, “Collateral coverage,” 100.
27. Ibid., 101.
28. Roger Silverstone, “Finding a Voice: Minorities, Media and the Global Commons,” Emergences: Journal for the Study of Media & Composite Cultures 11, no. 1 (2001): 13–27.
29. Alexander J. Marland, Brand Command: Canadian Politics and Democracy in the Age of Message Control (Vancouver: UBC Press, 2016).
30. See most notably McCombs and Shaw, “The agenda-setting function of the mass media.”
31. “Fort McMurray special May 5th, 2016,” YouTube video, posted by Canadian Broadcasting Corporation, https://www.youtube.com/watch?v=jfsXF4v3Zeo, and “Fort McMurray special, May 4, 2016,” YouTube Video, posted by Canadian Broadcasting Corporation, https://www.youtube.com/watch?v=Xn2ydiQctus&t=501s.
32. Satu Elo and Helvi Kyngas, “The qualitative content analysis process,” Journal of Advanced Nursing 62, no. 1 (2007): 111.
33. These are online as follows: “Fort McMurray Wildfire Update #1 – May 3, 2016 at 5 pm,” YouTube video, posted by YourAlberta, https://youtu.be/Wp9rhagUcUM; “Fort McMurray Wildfire Update #2 – May 4, 10:15 am,” YouTube Video, posted by YourAlberta, https://youtu.be/kX8um8SYKiY; “Fort McMurray Wildfire Update #3 – May 4, 2016 at 3:15 pm,” YouTube Video, posted by YourAlberta, https://youtu.be/9Xp7o3-dQ_4; “Fort McMurray Wildfire Update #4 – May 5, 2016 at 11:20 am,” YouTube Video, posted by YourAlberta, https://youtu.be/B8UJ-OLBsZA; and “Fort McMurray Wildfire Update #5 – May 5, 2016 at 6:05 pm,” YouTube Video, posted by YourAlberta, https://youtu.be/Z_-9oOW9Uuo (all urls accessed 22 November 2016).