Abstract
This research examines the linguistic choices that frame relationships between bicyclists and other parties involved in fatal crash events. Textual data were collected from media reports of all bicyclist traffic fatalities that occurred from January 2009 through June 2018 in Hillsborough County, Florida, which has a disproportionately high number of bicyclist deaths compared with other areas of the U.S. The media reports were coded with a qualitative data analysis software and analyzed using critical discourse analysis (CDA), a rigorous qualitative method used to analyze oral and written communication developed by Fairclough. Through CDA, the study examines how linguistic choices produce meaning and reinforce the “common sense” or “taken-for-granted” lexicon of transportation. Results show the majority of news reports were episodic rather than thematic, focusing on the traffic event and the parties involved in the crash, particularly the bicyclist. Vocabulary, grammatical structure, and narrative framing of news reports largely functioned to remove blame from the motorist and to highlight the bicyclist’s actions. These linguistic strategies reflect the assumption that responsibility for safety rests on the bicyclist and detracts attention from potential social policy reform that would lead to fewer bicyclist fatalities. A minority of articles written with thematic frames focused on broader issues such as social capital, safety education, and advocacy. This interdisciplinary study is a unique contribution to transportation literature, employing a methodology typically reserved for communication scholars and linguists.
According to the National Highway Traffic Safety Administration (NHTSA), the number of bicyclist deaths in the U.S. increased by 1.3 percent to 840 during 2016 from the previous year, the highest since 1991. Florida consistently leads the nation with the highest rate of bicyclist deaths per capita, reporting 16.6 percent of total U.S. bicyclist deaths during 2016 ( 1 ). Hillsborough County, Florida, located mid-state along the western boundary, is the geographic location for this research. Home to 1.4 million residents, the county has a larger population than ten other U.S. states. The county seat and largest city is Tampa ( 2 ).
As illustrated in Figure 1, from 2009 through 2016, Hillsborough County had a higher average rate of bicyclist fatalities than Florida’s rate, spiking at 10.92 fatalities per million during 2012 ( 3 ). When compared with the nation, and populous states such as California, Hillsborough County’s rate was also higher ( 4 ).
Bicyclist fatality rates per million residents for Hillsborough County, Florida, California, and the United States.
The high rate of fatalities makes Hillsborough County an appropriate site for transportation inquiry, which has the potential to provide insight into bicyclist fatalities and potential countermeasures statewide and nationwide. While pedestrian fatalities in the county follow a similar trend and represent a higher percentage of traffic-related deaths, this study focuses on bicyclist deaths as a primary exploration of transportation safety. In Hillsborough County, bicyclist fatalities were frequently reported by the media, yielding a comprehensive dataset through which to begin research about deadly traffic crashes.
Research shows that the perception of road traffic risks is influenced as much by media reporting as by personal perceptions of risk ( 5 ). To account for public understandings of bicyclist fatalities, it is important to examine the language in news reports about fatal crashes. Critical discourse analysis (CDA), a method used by discourse analysts and linguists to expose the subtleties of language and its function in relation to social structures, is an ideal method to analyze news reports ( 6 ). The purpose of this research is to use CDA to examine and critique common modes of discourse about bicyclist fatalities, particularly the ways news reports employ narrative framing, vocabulary, and grammatical structure, to observe and understand how language practices portray bicyclists.
In this qualitative study, textual data were collected via media reports for all bicyclist traffic fatalities that occurred from January 2009 through June 2018 in Hillsborough County, Florida. The reports were coded for repeated textual features and analyzed using the CDA procedure developed by Fairclough ( 7 , 8 ). The objectives of the research are: 1) to illustrate the nuances of language that shape institutional norms; 2) to evaluate the social effects of a “common sense” or “taken-for-granted” lexicon of transportation; and 3) to develop strategies for participating in transportation-related language practices with the aim of altering language use in the news reporting of bicyclist fatalities.
Because the study centers on language used in media reporting, the results section includes multiple excerpts from the articles in the dataset. These articles are cited in a report prepared by the Center of Urban Transportation Research at the University of South Florida, “Media Framing of Fatal Bicycle Crashes in Hillsborough County: A Critical Discourse Analysis” ( 9 ).
Few studies of bicyclist fatalities focus on the language used in news reports and how it may influence social norms and the ability to reform transportation practices and enhance equity among all road users ( 10 , 11 ). CDA provides a rigorous method to analyze language and add to the current conversation about transportation safety. A primary objective of CDA is to draw attention to oppression that is inherent in cultural assumptions and unmask practices that perpetuate social problems. This research seeks to show how language can be studied to shift focus from bicyclist blame toward social reform.
Literature Review
How Discourse Shapes Social Practices
Discourse analysts posit that language shapes reality and the social practices that reproduce reality; discourses are language practices that construct a coherent system of meanings, reproduce power relations, and have ideological effects ( 12 ). Discourse analysis begins with the assumption that environments, objects, and subjects are things that both produce and are produced by conversation. Negotiating reality is a process of questioning language practices and problematizing the “taken-for-granted” nature of language ( 13 ).
In a 2015 article in The Washington Post, Emily Badger addresses the common use of the word “accident” to describe traffic crashes ( 14 ). “An ‘accident’ is, by definition, unintentional,” she writes. In the summer of 2015, an advocacy organization called Transportation Alternatives launched a campaign asking people to pledge to stop calling crashes accidents: “Language, they believe, shapes policy. The word ‘accident,’ they say, presupposes a conclusion that no one bears responsibility” ( 14 ).
Critics who have condemned the word “accident” are usually proponents of the word “crash.” A crash suggests a violent scenario in which damage has been done; crashes leave the public wanting answers, solutions, and, perhaps, someone to blame. However, Magusin ( 10 ) says the words “crash” and “collision” ignore the laws of physics. Crashes and collisions imply that two objects or persons bearing equal force have collided. But bicyclists are vulnerable road users who weigh exponentially less and are exposed on roadways, unlike motorists who are surrounded by vehicles weighing somewhere between 2,000 and 10,000 pounds.
This micro-level attention to language is at the heart of discourse analysis, as discourse analysts attempt to comprehend what language does by looking only at the data of conversation or texts.
This is often contentious; The Washington Post article that differentiated the word “accident” from the word “crash” sparked heated controversy on the website’s comments section. Some commenters argued that the use of the word “accident” was no different from the use of the word “crash” and that microscopic language choices such as these do not impact reality ( 14 ).
Fairclough considers language a form of social practice and contends that, via close analysis, linguistic devices are more than functions of grammar, style, and composition ( 7 ). In his 1989 seminal work, Language and Power, Fairclough outlines his procedure for critical discourse analysis with a series of questions:
What classification schemes are drawn upon?
Are there words which are ideologically contested?
Is there rewording or overwording?
What ideologically significant meaning relations (synonymy, antonymy) are there between words?
When examining texts for grammar, Fairclough’s method raises the following questions:
Is agency clear?
Are processes what they seem?
Are sentences active or passive?
Are nominalizations used?
In addition to the content of language, Fairclough’s method also draws attention to the general framing, or “schemata” of content. Fairclough breaks schemata into three categories: schema, frame, and script. A schema is a type of activity “in terms of predictable elements in a predictable sequence.” For instance, a newspaper report is predictably made up of contents such as: the cause of the incident, the consequences, and how it was dealt with. A frame also concerns the contents of a text in relation to its topic. According to Fairclough, a frame is a representation of a topic or subject matter within an activity. Frames can represent types of people (e.g., bicyclists), inanimate objects (e.g., vehicles), processes (e.g., dying), or abstract concepts (e.g., economy) ( 7 ).
“While frames represent the entities which can be evoked or referred in the activities represented by schemata, scripts represent the subjects who are involved in these activities, and their relationships,” Fairclough writes ( 7 ).
For instance, a police officer has a particular script to follow when responding to a traffic crash, and that script defines the relationship between the officer and civilians who are involved in the crash. A thorough discourse analysis extends beyond the micro-level content of language and attends to the ways that framing is employed to reinforce social practice.
The Effects of Media Framing
News stories are of narrative construction, arranging events in an order that is not always factually chronological but maintains “common ground” with readers via “common sense” or “taken-for-granted” assumptions ( 7 , 8 ). For instance, most news stories about bicyclist fatalities begin with the end of the event that is described: a bicyclist is dead. From there, assumptions are called upon (often unconsciously on part of both the author and the readership), rooted in ideals of journalistic objectivity.
Another narrative convention observed in news stories is the portrayal of social agents as characters in a story that originates from a particular point of view. In the case of most of the news stories collected as data for this project, the narrative of each bicyclist fatality is delivered in only a few brief paragraphs. This is what Iyengar calls “episodic framing,” which functions to isolate the story and its characters from larger contexts and complexities ( 15 ). The episodic framing of these stories may encompass a variety of viewpoints, but the viewpoint conveyed most is simple, clear, and requires no further thinking on the subject: these fatalities happen, and they happen often.
While episodic framing is used to report specific events and concrete circumstances, thematic framing presents news that accentuates political issues and events in some general context, reflecting collective evidence. Thematic frames point to larger structural and cultural issues instead of focusing solely on individual cases, lists of facts associated with those cases, and legal and/or personal responsibility. Thematic frames tend to overlap with episodic frames because the general public responds to concrete events more readily than it does to abstract ideas, concepts, or issues ( 15 ).
In examining the influence of television news framing of political issues, results indicated that episodic framing invoked individual rather than societal attributions of responsibility while thematic framing drew attention to society’s role or policy issues. Iyengar posits that, according to the way in which television news frames national issues, individuals’ interpretations of these issues can be altered ( 15 ). For example, in one study, participants viewed a videotape showing news stories centered around poverty, unemployment, and terrorism over the previous six months. Results found that individuals relied on contextual cues to reason about these national issues; individuals’ attributions of poverty depended on whether news took the form of particular victims (episodic) or focused on nationwide outcomes (thematic) ( 15 ).
Iyengar identifies “framing effects” as subtle changes in audience judgment ( 16 ). This is no new concept. In 1974, Goffman introduced the concept of framing as a tool for organizing information into a meaningful succession of events ( 17 ). Gitlin extended the concept specifically to journalists, who routinely create organized sets of information for audience interpretation ( 18 ).
Hidden Power and “The News”
According to Fairclough, newspapers possess “hidden power” through practices that embody assumptions that either directly or indirectly legitimize existing power structures. In other words, news articles hide power in plain sight because it is embedded in taken-for-granted language practices. Universal or “common sense” practices are practices which only appear to be common sense, but when discourse is examined closely, it becomes obvious that these assumptions are arbitrary and function to sustain unequal power relationships ( 7 ).
Fairclough argues that written language has a particularly nuanced quality, as it is one-sided communication, and communicators are generally separated by time and space when engaging with written texts ( 7 ). And, because media communicates to mass audiences, producers of media texts cannot know the audience or adapt to it, and therefore they write to an ideal subject. This alone is a form of control over discourse, as readers of texts must negotiate an ascribed identity of the ideal subject to understand the article or text (which is often taken at face value). This raises questions of whose perspective is adopted in news reports, though those reports are often considered “factual,” categorical truths ( 7 ).
“It is a form of the power to constrain content, to favor certain interpretations and ‘wordings’ of events, while excluding others,” wrote Fairclough ( 7 ).
Fairclough claims that discourse is a site of ideological social struggle. Therefore, when examining and interpreting texts, we must note that the production of a text and the interpretation of a text both have an interpretive character if we are to illuminate, understand, and transform these struggles and the positions of the parties involved. The producer of the text constructs the text as an interpretation of the world, or of the facets of the world which are then in focus; formal features of the text are traces of that interpretation (vocabulary, grammar, etc.) The traces constitute cues for the text interpreter, who draws upon her assumptions and expectations (incorporated in frames) to construct her interpretation of the text. Thus, text interpretation is the interpretation of an interpretation ( 7 ).
Interestingly, the ideological burden is often on readers, as they must call upon contentious assumptions if they are to make sense of the text. This is not to suggest that producers of texts embed these assumptions deviously—it is quite the opposite; usually, neither producers nor interpreters of texts are aware of the assumptions that are evoked through discourse. This study of discourse does not call upon those in power or demonize particular members of society. Instead it points to the power itself as it hides in language practices. Of course, power is not a “thing,” but rather an abstract concept that is operationalized by language; the consequences of power can only be transferred to the concrete world through language that structures our assumptions about the nature of that world. This approach removes “blame rhetoric” from discourse analysis and allows for a creative but detached conversation to emerge, perhaps leading to critical awareness and positive change.
Methodology
Data Collection
The dataset for this study consists of 190 news reports of 94 bicyclist fatalities that occurred from January 2009 through June 2018 in Hillsborough County, Florida, to create a complete dataset of all fatalities within this timeframe. First, Florida’s Integrated Report Exchange System (FIRES) was used to document the number of reported bicyclist fatalities ( 3 ). Next, the names of the bicyclists, and the date and locations where the deaths occurred were compiled using internet searches and a public list compiled by a local safety advocate. Finally, data was verified using the Florida Department of Transportation District Seven’s Crash Data Management System (CDMS). The information collected was used to filter the results on a database inquiry of local news sources that reported on the incidents, including the Tampa Bay Times, the Tampa Tribune, the St. Petersburg Times, WSTP 10 News, Bay News 9, Fox 13 Tampa Bay, WUSF Public Media, ABC Action News, and other news outlets appearing in print and online.
The search terms used were: (bike OR bicycle OR bicyclist OR crash OR collision OR struck OR killed OR dead OR death OR dies OR died OR fatal OR fatality OR accident) AND Tampa or Hillsborough County AND (“Name of Bicyclist”). The results were filtered by the month and year of each death. All published stories were collected, excluding data from blogs or non-official news sources (e.g., lawyers’ websites reproducing news as advertisement for their services). Search results were thoroughly examined, and data were cleaned, checked, and cross-checked to ensure accuracy. Stories were converted to a .pdf file format and imported into a qualitative data analysis software. All news reports referenced in this research are in the public domain.
Data Coding and Analysis
Fairclough’s method of discourse analysis (outlined in the literature) was used for coding the data. Coding categories, or “nodes,” were created based on Fairclough’s general categories of “experiential” and “expressive” codes. Experiential codes focus on vocabulary, words/wording, classifications, grammar, and sentence structure. Expressive codes center on the producer of the text’s content, particularly related to the evaluation of subjects, identities, and outcomes. Based on this organization, general parent and child nodes were identified to begin the coding process.
First, a process of open coding was employed using queries based on the general nodes. For example, queries were run for particular terms like “crash” and “accident.” The query results were saved and the articles they returned were assigned to the corresponding nodes.
Then, selective coding was employed to mark textual features and language patterns according to Fairclough’s procedure. Each article was coded manually and data within each article were assigned to general nodes. During this process, the researchers made note of emerging patterns in the text and identified new categories and subcategories of nodes, which were added to the scheme of coding categories. The process was repeated in a second round of coding.
When the coding was finished, the nodes were refined by combining repetitive nodes and revising the organization of parent and child notes. To ensure coding accuracy, coding was performed by two different coders, and the Kappa coefficient was calculated to measure the degree of agreement between the coders, returning a value of 0.8. This means 80% of the coding matched, which reflects a high intercoder reliability. Table 1 represents the final organization of nodes.
NVivo Experiential and Expressive Nodes
Finally, a process of axial coding was employed to identify recurring relationships between nodes. These were saved and provided the foundation for analysis of the data. Data analysis centered on the frequency of experiential nodes, which include the most commonly used terms and grammatical structures. The researchers also focused on the most common relationships between nodes and the framing of the articles.
Episodic frames were differentiated from thematic frames; episodic frames constituted the majority of the data, providing a formulaic body of news reports on bicyclist fatalities, while thematic frames provided contrast and context. The results of this study focus primarily on episodic frames, since they make use of language structures that reveal the cultural assumptions surrounding bicyclist fatalities. Thematic data were used during analysis to provide insight into language about transportation safety, and to make conclusions and provide suggestions for reform and future research.
Discussion of Results
Taken-for-Granted Vocabulary
Overall, the vocabulary used to describe fatal traffic events draws upon taken-for-granted, common-sense assumptions that motorists and bicyclists (the subjects) bear equal responsibility for traffic safety ( 7 , 8 ). When examined closely, these assumptions make little sense in light of physical realities. While equal responsibility may be valid in relation to personal choices, in reality, the motorist drives a much more powerful object than a bicyclist. This imbalance in power is removed through the use of terms such as “accident” and “incident.” These words are what linguists call “nominalizations,” or verbs that have been transformed into nouns to abstract the action of the subjects. Nominalizations are often used in passive constructions of sentences in which there is no subject or the action is detached from the subject.
Accident/Incident
The word “accident” was used to represent traffic fatalities 62 times in 48 articles. Through the literature review, the researchers presented the argument that the word “accident” removes blame from the motorist and suggests that the event was unavoidable. The word “incident” was used 13 times in 12 articles, which Merriam-Webster describes in more abstract terms as a noun that is synonymous with an event, occurrence, episode, or circumstance ( 19 ). This may leave out the assumption that the incident is unintentional or unavoidable, but does little to ground the event in ways that suggest the existence of context, or indicate the event’s connection to a concrete and embodied reality.
Crash/Collision
The word “crash” was used 168 times in 91 articles, and the word “collision” was used 26 times in 22 articles. This means a vast majority of the news stories in the dataset use the word “crash” to describe motor vehicle crashes that involve bicyclist fatalities. There is no significant change over time in the use of the word “crash” compared with the word “accident.” They are often used interchangeably; the decision to use the word “crash” does not seem a conscious choice to replace the word “accident,” as they often appear together in the same article. As presented in the literature review, the term “crash” is preferred among language activists, though it still removes action from subjects and suggests equal force between objects ( 10 ).
Grammatical Strategies to Remove Agency
Agency is conscious action by a clear actor. It is communicated in language through grammatical structure. In almost all the news articles analyzed for this project, motorist agency was eliminated from the story through a variety of grammatical choices.
Passive Sentences
A sentence is written in a passive construction when the subject is absent or included after the verb, as in “A bicyclist was killed by a car.” In addition to nominalizations, passive constructions of sentences serve to remove agency by emphasizing the action rather than the actors. More than half of the articles in the sample used passive sentences when describing the traffic event. Often, the motorists were not included in the sentence as the agent, and when they were, they were often referred to as the vehicle instead of by name or any human identifier. In contrast, most of the active sentences in the articles were used in reference to the bicyclist, as in “The bicyclist was riding at the intersection of…”
Motorist Agency
The main way that motorist agency was eliminated was to omit the motorist’s name and/or obscure the presence of a driver and emphasize the activity of the motor vehicle the driver was operating. In 55 of 189 sources, the motorist was replaced at least once with the word “truck” (usually multiple times within a source), totaling 111 references to the vehicle as an agent in bicyclists’ deaths. The following are examples of this: When a bicyclist died Nov. 1 on Fowler Avenue, investigators said a pickup truck entered a triangular safety zone and collided with the bike (
9
). Boger was standing with his bicycle on the corner of Waters Avenue and Florida Mining Boulevard at 2:14 p.m. when a Mack truck made a sharp right turn and hit Boger, deputies said (
9
). When Campbell started to cross the road, her bicycle went into the path of the truck (
9
).
In addition to the 55 trucks that are given agency in the dataset, cars are also agents in 50 additional articles and the word “car” is referenced a total of 138 times: Teen on bicycle killed after being hit by car in Apollo Beach: A 13-year-old riding a tricycle died Saturday (Aug. 12) night when hit by a car, the Hillsborough County Sheriff’s Office said (
9
). The sheriff’s office reports that the first boy, 12-year-old Mitchell Bowers, died Tuesday evening. He had been riding in the bicycle lane when he reportedly turned left in front of a car and was hit (
9
). Bryan Lebron Jr., 11, of South Tampa, died Wednesday afternoon after he was struck by a car at South Dale Mabry Highway and Bay Avenue (
9
).
The word “vehicle” is used in place of the motorist 103 times over 70 of the articles. More common than replacing motorists with these generic terms is replacing them with the specific make and/or model of the vehicle, making the naming of the agent more specific. The make and model of the vehicle was the descriptor of the agent in 97 articles and was referenced a total of 106 times:
A 2002 Dodge truck approached from behind near Jana Drive, traveling about 45 mph.
Police say Edward Nelson Lewis, 55, was struck by a Honda CRV on North Nebraska Avenue and East Emily Street just before 5 a.m. (
9
). Ivan Miller, 47 of Tampa was standing in the inside lane of U.S. Highway 41 south of 10th Street near the white painted line divider when he was struck by a 2012 Honda Civic at 7 p.m. (
9
).
It is important to note that, in some of these articles (none of which are in the examples above), the death was caused by a hit-and-run driver and, in these cases, the agent was otherwise unidentifiable, so the vehicle was presented in place of the individual. However, these articles are of the minority, as only 41 use the phrase “hit-and-run,” and in many of those articles the driver had been subsequently caught and identified. When a hit-and-run death is reported, the agency of the driver is made clear when the driver is caught. However, in preliminary articles about hit-and-runs, the police are often looking for the vehicle, and still, the motorist or the fact that a driver was involved are omitted. This is probably because there are no witnesses or ways to describe the driver, and in standard journalistic and police reporting practice, attribution of a crime is avoided until a person has been charged.
Relationship between Bicyclist Class and Bicyclist Safety
Besides using vocabulary that contains assumptions about responsibility and grammatical strategies that remove agency, the episodic articles revealed a strong relationship between bicyclist class and bicyclist safety.
There are various ways to indicate the approximate socioeconomic status of a bicyclist. A relationship exists between unsafe bicycling behaviors and lower-class identity markers, while an inverse relationship exists between safe bicycling behaviors and higher-class identity markers. The first of the two examples to follow are indicative of the former relationship: Weeks, riding against traffic, was on his way home from a landscaping job (
9
).
In this example, the bicyclist, Anthony Weeks, was riding against traffic, which is illegal and unsafe. In the same sentence, there is an association with his vocation, a landscaping job. This line of work is not considered a career, but a trade, and Anthony is identified as a laborer with little economic power: Police said Gutierrez was riding north against a red light at Woodland Corporate Blvd., in front of the Waterford at Cypress Lakes apartment complex, where he lives (
9
).
Another way to mark the socioeconomic identity of a person is to describe their living conditions and/or the location of their homes, as the above sentence does in the case of Jose Baez Gutierrez. Conversely, when bicyclists are of a higher status, they are more often portrayed as having practiced safe behaviors at the time of their deaths, as in the case of Kayoko Ishizuka, a cancer researcher at the University of South Florida: At the time of the crash, Ishizuka had lights on her bike, wore a helmet and rode in the bike lane, Coleman said (
9
).
Ishizuka was known as one of the “best and the brightest” in her field: She bicycled for the love of it, not out of necessity (
20
): Rivera lost control and struck cyclist Robert Niedbalec, 52, a Temple Terrace veterinarian and black-belt instructor, who was riding in a designated bicycle lane (
9
).
In this example, Robert Niedbalec is identified as a cyclist, a veterinarian, and a black-belt instructor who was riding his bike in the designated bike lane. This demonstrates his contribution to society and his choice to follow the law. However, it is interesting that it is not mentioned in any of the myriad articles about his case whether he wore a helmet when he was killed in the designated bike lane. In cases where bicyclists are of lower socioeconomic status, it is frequently mentioned that they were not wearing helmets when they died, even if they were not breaking the law (it is not a violation of Florida law to ride without a helmet unless the bicyclist is underage).
Opportunities in Thematic Framing
A significant portion of the dataset are written with thematic frames (56 of 189), suggesting that bicyclist fatalities happen for reasons that can be prevented, and that the prevalence of bicyclist fatalities in Hillsborough County is undesirable for some of the more prominent social actors in the surrounding communities, particularly bicyclist advocacy groups, political figures and organizations, engineers, law enforcement, and urban planners.
Thematic frames in the dataset highlighted issues associated with social capital, safety education, and bicycle advocacy.
Social Capital
Networks of people come forth to grieve together, support one another, and, in the case of some bicyclists’ deaths, plea for safer roads for bicyclists when a loved one dies. These people have social capital and are considered of a higher socioeconomic status than other bicyclists. In other words, some bicyclists were reported to be valuable assets in the networks to which they belonged and, as a result, their deaths were framed as tragic and senseless in opposition to episodic “hard news” stories. The following excerpts are expressions of emotional hardship that develop the meaning of the abstract concept of “public health” and help audiences to understand the implications of our current epidemic: Bryan’s mother said the family was close. The boys shared a bunk bed. Bryan had the top where a big stuffed green frog remains, along with an open book: 20,000 Leagues Under the Sea. Donté listed his brother’s favorite things: basketball and the colors red, blue and black. What did he like most? “Me,” Donté said (
9
). Each Friday morning, at the exact minute her bicyclist sister Diane was hit and killed on a Tampa sidewalk, JoAnn Vega stands at the street corner and rings three cowbells. Her sister loved the Rays…Tuesdays, instead of meeting her daughter Diane for their once-weekly lunches, Amelia Vega visits the corner to tidy the memorial. She rearranges the candles and brings fresh flowers (
9
).
Both the above examples illustrate social capital via family networks and the deep-seated grief that is suffered by real people when bicyclists die and leave beloved kin behind. The following excerpts include family, professional, communal, and friendship networks: Calonge’s father opened Joe Haskins Bicycle Shop in Tampa in 1953. Diane Vega was a regular at the shop, chatting with the employees and asking that Haskins do the repairs. “She was more than just a customer,” Calonge said (
9
). “Dr. Niedbalec, by all accounts, was a wonderful man,” Battles said. “To characterize this as anything less than absolutely senseless would be a mistake…” They were judo students, veterinary partners, friends, and his widow, Kathryn, and his daughter, Kate. They spoke of a man who loved people as much as he loved life (
9
). A fifth-generation Floridian, Mr. Collins was the son of LeRoy Collins, who, as governor from 1955 to 1961, appealed to heads and hearts of average Floridians to reject racial segregation (
9
).
Social capital garners respect by adding value to the bicyclist through association with family, professional, communal, and friendship networks.
Safety Education
Most of the thematic articles focused in some way on safety education or bicycle advocacy. The following is an example of these articles’ focus on pragmatic information about the rules and laws for safe cycling and sharing the road: But also the need for riders and drivers to understand the rules of Florida roads. Under state law, bicycles are considered vehicles and have the same rights as motorists, said Tim Bustos, executive director of the Florida Bicycle Association (
9
).
The following example elaborates on the importance of visibility, and the responsibility and agency of the bicyclist is emphasized: Bicycling safety advocate Alan Snel said motorists and bicyclists must learn to properly share the road. And bicyclists need reflective clothing and lights on both wheels when riding in the dark. “Light yourself up like a Christmas tree,” Snel said (
9
).
Bicyclist advocates and local transportation leadership work to inform the public of cyclists’ legal rights and garner political support for the protection of cyclists on Hillsborough County roads. For example, Gena Torres, a senior transportation planner in Hillsborough County, stresses the legal status of the bicycle as a vehicle deserving of its own infrastructure: “The solution is not just to try and rid the roads of bicyclists and put them on a trail,” she said. “A bicycle is a vehicle” (
9
).
Bicycle Advocacy
Advocacy also takes the form of joining with international movements, like the Ride of Silence, written about in the excerpts below: The bikes are personal memorials honoring the dead, but they’re also symbols, part of an international movement aimed at reminding motorists to watch out for bicyclists (
9
). Bicyclists are a common sight along the stretch of road. Bicycle safety advocates already were planning a 15-mile bike ride Friday to honor five bicyclists killed in about two months. They said people need to speak up to politicians about the need to improve bicycle safety. About 60 people attended the Honor the Fallen Bike Ride, which traveled into downtown Tampa and Ybor City before returning to Seminole Heights (
9
).
Advocacy and awareness are important elements of public health education that make the safety needs of cyclists known. However, the road systems available to cyclists also contribute to local safety concerns. Some of the news stories highlighted problems with local infrastructure that may have contributed to the fatal episodes reported in the articles. For instance, inadequate bicycle lanes that contain hazards are problematic, along with the absence of bicycle lanes in areas that need them: Torres said part of the solution is to add more bike lanes and take steps to slow traffic in certain areas (
9
). Bayshore Boulevard now has a dedicated bicycle lane, but many do not use it because of debris and fast cars that sometimes drift to the right (
9
). Sheriff’s officials noted that the accident happened in an area where there is no bike lane or paved shoulder. That particular stretch of Gunn Highway is narrow, as the road curves around Lake Calm (
9
).
Conclusion
A Formula for Reframing Bicyclists’ Deaths in News Reports
To start a creative conversation that might begin to prevent bicyclist deaths, the researchers have identified linguistic strategies that may change the ways people view these deaths. Hillsborough County has taken great strides to prevent bicyclist deaths in the past ten years, yet fatalities continue to occur at the same rate. According to Bike/Walk Tampa Bay, a coalition of community partners formed in 2015, what is desired is cultural change ( 21 ). The results of this study indicate that it is possible to reform the news reporting of incidents involving bicyclists, which may impact the ways our institutions are willing or able to catalyze change. Recently, transportation researchers found in a study where participants self-identified as both drivers and as bicyclists, that aggressive driver behavior toward bicyclists focused on blaming bicyclists for behaving improperly and punishing them accordingly ( 22 ).
It is time to start talking publicly about how bicyclist deaths are framed and to collectively attend to how language shapes beliefs about power and control. That conversation starts with one basic premise: In each of these articles, a human being has been killed. A person has died of unnatural causes. In episodic reports newspaper readership often learns the name, age, and gender of this person. The reader sometimes learns where they lived, about crimes they once committed, or if they were under the influence of drugs or alcohol. What is not learnt in these articles is what is read about in all the thematically framed news stories; the reader does not learn about their families, their favorite sports teams, or of the rough hand life may have dealt them.
The researchers recommend that news reports employ thematic framing when reporting traffic events, moving beyond the details of the events and the parties involved toward a larger discussion of social responsibility for transportation infrastructure and policy reform. In addition to a discussion of reform, there should be respect for the dead and their families and acknowledgement of their personhood and the emotional burden the drivers of the vehicles shoulder after being involved in these tragic and often gruesome deaths. Though drivers are often not charged with crimes for bicyclist fatalities, there are other consequences that may be much worse than citations or jail time.
In addition, when the event or the parties involved are mentioned, the researchers recommend employing active sentences in which the subjects and their actions are clear, in an effort to avoid overemphasizing the bicyclist’s responsibility for traffic safety. For example, instead of “A bicyclist was hit by a car,” an active sentence would center on the subject doing the action: “A driver hit a bicyclist.” These articles may be repetitive because of the frequency of these kinds of traffic deaths, but they are newsworthy. Each bicyclist and each motorist is a person, a person with a unique story that ended or was forever changed.
This project provides a method and dataset with diverse potential for conducting further research with similar objectives. Future transportation studies should compare textual features of news reports in Hillsborough County with news reports in other areas. Regional differences in vernacular may affect local transportation culture and policy, and henceforth correlate with the frequency of bicyclist fatalities. Other factors viable for comparison include differences in urban environments, infrastructure, and law enforcement policies. For instance, police reports are institutional texts that function to inform media and influence representations of bicyclists, providing yet another source of linguistic data. Finally, both this project and potential future projects are viable for replication with a focus on pedestrian fatalities.
Footnotes
Acknowledgements
This research was supported by the National Center for Transit Research at the University of South Florida. We thank Dr. Mariaelena Bartesaghi, Dr. Robert L. Bertini, and our colleagues from the Center for Urban Transportation Research who provided insight and expertise that greatly assisted the research. The interpretations and conclusions of this paper are those of the authors alone.
Author Contributions
The authors confirm contribution to the paper as follows—study conception and design: JB, ES; data collection: JB, ES; analysis and interpretation of results: ES, JB, LEM; draft manuscript preparation: ES, JB, LEM. All authors reviewed the results and approved the final version of the manuscript.
The Standing Committee on Bicycle Transportation (ANF20) peer-reviewed this paper (19-01484).