Seriously funny: The political work of humor on social media

Abstract

Research shows a clear intersection between humor and political communication online as “big data” analyses demonstrate humorous content achieving disproportionate attention across social media platforms. What remains unclear is the degree to which politics are fodder for “silly” content production vis-à-vis humor as a serious political tool. To answer this question, we scraped Twitter data from two cases in which humor and politics converged during the 2016 US presidential election: Hillary Clinton referring to Trump supporters as a “basket of deplorables” and Donald Trump calling Hillary Clinton a “nasty woman” during a televised debate. Taking a “small data” approach, we find funny content enacting meaningful political work including expressions of opposition, political identification, and displays of civic support. Furthermore, comparing humor style between partisan cases shows the partial-but incomplete-breakdown of humor’s notoriously firm boundaries. Partisan patterns reveal that the meeting of humor and social media leave neither unchanged.

Keywords

Discourse humor politics small data Twitter

A growing body of work shows a clear intersection between humor and politics online. (Driscoll et al., 2013; Hartley, 2012; Highfield, 2016a, 2016b; Wells et al., 2016). Using big data analytics and network analysis, studies show that parody accounts, sarcastic content, gifs, memes, and clever intertextual remixes garner significant attention during political media events. This trend intersects three converging forces—the rise of satirical political commentary (Baumgartner and Morris, 2006; Gray et al., 2009; Webber et al., 2013), a participatory media environment fostered by digital social technologies (Barnard, 2016; Prior, 2006; Vaccari et al., 2015; Vis, 2013; Wells et al., 2016), and internet culture in which humor is both expected and rewarded (Leavitt, 2014; Milner, 2013a; Miltner, 2014; Nooney and Portwood-Stacer, 2014; Phillips and Milner, 2017). Through this tripartite convergence, humor as a widespread political communication device emerges as not unique to the internet but distinctly of the internet. Questions remain, however, about how humor is deployed. Are people using humor to make substantive political claims or are they simply drawing on political content as source material for their jokes?

We take the well-documented pattern of humor in digitally mediated political communication as a jumping-off point and dig into how, specifically, political humor operates. We do so by exploring the use of humor and playfulness as part of political discourse on Twitter during key moments of the 2016 US presidential election. Specifically, we scrape and analyze tweets surrounding two political gaffes: Donald Trump calling Hillary Clinton a “nasty woman” and Hillary Clinton referring to Trump supporters as a “basket of deplorables.” With these data, we look for how humor works, and the work humor does, in digitally mediated political communication.

Using an interpretive abductive approach, we build and analyze two unique data sets—one from each political gaffe—that intersect politics and humor to discern if and how people utilize humor for meaningful political engagement in the social media context. Analyzing both form and content across data sets, we show that Twitter users rely on humor to express opposition, establish political subjectivity, and engage in direct and symbolic civic support. In turn, we show that humor style presents in patterned ways between data sets, a finding that both resonates with and diverges from cultural studies of humor as a persistent marker of distinction between groups. Together, these findings hold relevance for studies of digitally mediated democracy by weighing-in on debates about social media’s role in civic participation while contributing to humor studies by examining how humor takes shape between ideologically antagonistic groups on a prominent social media platform.

Political communication in a post-broadcast era

Political discourse has traditionally been dominated by a small number of elite actors with access to large-scale communication channels such as television, radio, and newspapers. These actors include journalists, politicians, and carefully selected political representatives. However, widespread access to information and communication tools afforded by emergent technological infrastructures have ushered in a “post-broadcast” democracy (Prior, 2006) represented by a shift in “communication power” (Castells, 2013). This new era is characterized by a move from bounded and defined news cycles to 24 hour information flows and from exclusive control of public discourse by powerful media gatekeepers to a distributed model of communication enabled by social media platforms and widespread use of personal digital devices (Gal et al., 2016; Shifman, 2007; Tufekci, 2017).

Within the political arena, media events have long played a central role, and these events are now marked by a strong participatory element. First described by Dayan and Katz (1994) as “high holidays of mass communication” (p. 1), media events are ritualized, pre-planned programming that monopolize media streams and reach mass audiences. In the United States, the presidential campaign season is characterized by many small media events, punctuated by periodic mass events, most notably, the scheduled televised debates. In 2016, the three debates between Hillary Clinton and Donald Trump broke new records for viewership, totaling 292 million viewers, including 84 million in the first debate alone (Nielsen, 2016). However, television viewership was not the only way in which people engaged. Rather, these media events were distinctly “hybrid” (Wells et al., 2016) in their inclusion of vast social media participation. According to Nielsen (2016) analytics, the first debate drew 17.1 million tweets from 2.7 million separate accounts, the second debate drew 5.5 million tweets, and the final debate drew 53 million interactions across social media platforms from over 16 million individual users.

Political media events are thus no longer performed by politicians, analyzed by journalists and political elites, and then discussed around the pub or the dinner table. Instead, media events are collaborative efforts in which performance, response, rebuke, and meaning-making include mass audiences engaging across multiple platforms asynchronously and in real time. In this way, everyday people join the “experts” to have their say. However, the inclusion of mass publics is not the only change brought forth by social media; social media have also shaped the tone of political interaction. Specifically, internet culture encourages a distinct playfulness which incorporates into the ways that publics communicate political messages online (Baumgartner, 2007; Jenkins, 2006; Milner, 2013b; Phillips and Milner, 2017).

Humor, politics, and social media

Silly citizens and the irreverent internet

Social media have fostered new forms of political communication and participation. These communicative forms take on the conventions of internet culture, which include wit, parody, sarcasm, co-optation, and playful memification as “economies of laughter have become inextricably entangled with … civic processes” (Henefeld, 2016). Hartley (2010, 2012) calls this style of interaction “silly citizenship,” while Highfield (2016a) positions “irreverence” as a core element of digitally mediated political engagement. Together, these theories of politics on the internet point to humor as a widely used and highly valued practice within political deliberations as they take shape through intersecting social platforms.

To be sure, humorous political engagement is not unique to the internet or social media. There is a long history of political satire (e.g. political cartoons printed in newspapers) and an entire broadcast genre that trades in humor while delivering social commentary and analysis (e.g. The Daily Show, Last Week Tonight, The Colbert Report, Adam Hill’s Last Leg) (Baumgartner and Morris, 2006; Gray et al., 2009; Webber et al., 2013). However, a robust infusion of “irreverent” and “silly” politics is distinctly of the internet, where humor and parody are woven into the fabric (Vis, 2013).

Existing analyses of major political campaigns and widely circulated news events show that playfulness looms large among digitally mediated publics, especially on Twitter. Using data from the 2012 presidential campaigns in France and the United States, Wells et al. (2016) find that while political and media elites maintained a disproportionate hold on public attention, parody and joke accounts—those depicting humorous versions of public figures and/or accounts that tweet primarily “silly” content—were highly retweeted and generated substantial interaction. In this vein, the 2012 US presidential debates showed a surge in Twitter activity during “meme worthy” moments, such as when Mitt Romney described his commitment to gender equity with a reference to “binders full of women” (Driscoll et al., 2013; Freelon and Karpf, 2015; Shah, 2015). These patterns persist in recent work showing the widespread following of, and engagement with, parody accounts in general and their central role as commentators on public events and political issues in particular (Highfield, 2016a).

This move toward “silliness” and “irreverence” in online political communication is significant, as Hartley (2012) suggests it may be an entry point among persons and groups who have been politically disempowered. Indeed, both humor (Hariman, 2008) and social media (Bode, 2016; Harlow, 2012; Mossberger et al., 2017) have been separately touted as means of lowering barriers to political participation. While humor “softens” discursive style, digital social platforms provide an alternative to elite-controlled broadcast conglomerates. Intersecting digitality and humor thus holds promise for increased access to political discourse and debate among diverse publics (Highfield, 2016b; Milner, 2013b). However, concerns remain that “silly” and “irreverent” political engagement online can foment cynicism, apathy, and general distrust in the political process, undermining digital media’s democratic potentials (Baumgartner, 2007; Baumgartner and Morris, 2006; Shifman et al., 2007). Moreover, digital inequalities of access, skill and media literacy threaten to exclude already marginalized populations (Hargittai and Walejko, 2008).

A tension thus exists between participatory potentials, on one hand, and the dissolution of silliness into cynicism and exclusion, on the other (Baumgartner, 2007; Phillips and Milner, 2017; Shifman et al., 2007). This tension leads us to our primary research question: in the context of Twitter, is humor a vehicle for serious political engagement or are politics simply fodder for funny content production? While our data are not suited for addressing related questions of inequality, understanding the work (or lack thereof) that political humor does will clarify the stakes for issues of access and inclusion.

Symbolic boundaries

Beyond this primary research question, examining politics and humor as they intersect on Twitter also illuminates key issues in cultural studies of humor. Specifically, the partisan nature of political communication lends itself to analyses of the role of humor in negotiating symbolic boundaries. Across social class, demographic markers, and ideology, cultural theorists have shown increasingly weak boundaries tied to taste preferences and practices, evoking an “omnivorous” metaphor to describe contemporary cultural consumption habits (DiMaggio and Mukhtar, 2004; Featherstone, 2007; Lamont and Fournier, 1992; Peterson and Kern, 1996). Yet, humor scholars find that comedy is an exception to the omnivorous trend, with humor style preferences maintaining clear status markers and moral meanings that draw hard lines between “us” and “them” (Friedman and Kuipers, 2013; Kuipers, 2009, 2015).

From internet studies, however, it seems that the relationship between humor and symbolic boundaries may shift as humor moves online. Indeed, social media scholars have demonstrated a shared vernacular in which users across and within platforms communicate through collective syntax, language, and grammar, with humor playing a significant communicative role (Burgess, 2006; Leavitt, 2014; Milner, 2013b). Thus emerges something of a contradiction in which humor itself has remained squarely within distinct cultural bounds, while digitally mediated communication practices—for which humor is an integral feature—rely on shared styles of speech. Resolving this contradiction helps answer questions about the mutual shaping effects of humor, on one hand, and digital vernacular on the other.

Research question 2 thus asks: how does humor style compare and contrast across ideological bounds in political communication on Twitter? Meyer’s (2000) origins and effects schematic is useful in this regard. Synthesizing empirical and theoretical literatures, Meyer thematizes humor’s origins (i.e. what makes something funny) and its effects. According to Meyer, humor is derived from three sources: relief, incongruity, and superiority. Relief cuts through an otherwise heavy moment, incongruity makes the familiar strange, and superiority is an expression of triumph through pointed deprecation of an “other.” These humor origins can serve two broad ends—division and unification (Meyer, 2000). Meyer argues that no single origin nor single effect can fully explain a humorous act. Rather, each act of humor includes multiple origins and effects with varying degrees of emphasis.

Our two cases, with their sharp partisan orientations, let us analyze not only how politics and humor intersect on social media but also how humor and social media converge into distinct cultural products and communicative forms. By paying attention to stylistic patterns employed across data sets, we can explore how humor practices reinforce and/or transcend ideological bounds. In doing so, our analyses answer key questions at the nexus of humor, politics, and digital culture.

Basket of deplorables and nasty woman: the events

We derive material for our analyses from two moments: the leak of Hillary Clinton referring to half of Donald Trump supporters as a “basket of deplorables” and Donald Trump’s reference to Hillary Clinton as a “nasty woman,” during the third presidential debate. Clinton’s political misstep came on 9 September 2016 while speaking at a fundraiser for lesbian, gay, bisexual, transgender, and queer (LGBTQ) supporters. Then candidate Clinton stated,

You know, to just be grossly generalistic, you could put half of Trump’s supporters into what I call the basket of deplorables. Right? The racist, sexist, homophobic, xenophobic, Islamaphobic—you name it. And unfortunately there are people like that. And he has lifted them up.

Although Clinton went on to empathize with Trump supporters who feel that “the government has let them down” and who are “desperate for change,” the Trump camp—both supporters and the official Trump team—pounced on the “deplorable” comment. Via senior communications adviser Jason Miller, the campaign put out the following statement:

Just when Hillary Clinton said she was going to start running a positive campaign, she ripped off her mask and revealed her true contempt for everyday Americans.

“Basket of deplorables” became a rallying cry for Trump supporters and continued to crop up formally and informally throughout the campaign. Clinton apologized for generalizing to “half” but never retracted the statement entirely.

Donald Trump’s “nasty woman” comment took place during the third and final presidential debate on 19 October 2016. Nearing the end of the debate, Clinton made a disparaging reference to Trump’s personal tax record. Trump leaned forward and, interrupting Clinton, muttered quietly into the microphone “such a nasty woman.” Clinton continued her line of thought but within moments, “nasty woman” emerged as a Twitter hashtag, a pro-Clinton website, merchandizing logo, and an activist feminist refrain. During post-debate interviews, Clinton dismissed the comment as unworthy of attention while her daughter, Chelsea, described the incident as a “sad” moment for women. Shortly thereafter, the comment was parodied in a widely viewed and shared sketch on the comedy series Saturday Night Live.

These incidents are ideal for studying political humor on social media for two reasons: first, they went viral and were quickly memeified while tapping into serious political contentions leveraged against each candidate—elitism for Clinton and misogyny for Trump—rendering them at once playful and politically relevant. Second, the sharp political divisions built into the “deplorable” and “nasty woman” incidents generated data from ideologically distinct groups. Thus, these moments were both funny and politically serious while illuminating questions about digitally mediated humor as it distributes along ideological lines.

Methods

Data for this study come from scraping a sample of Twitter content in the 24 hours following each key event—Hillary Clinton’s “basket of deplorables” comment and Donald Trump’s “nasty woman” reference in the third debate. Web scraping is the process of collecting data from the displayed information generated by a website. We employed Web Scraper (http://webscraper.io/), a free extension for Google’s Chrome web browser to collect the information contained in individual tweets that were listed as results by Twitter’s advanced search page (https://twitter.com/search-advanced). This information was saved into a spreadsheet. We searched the terms “deplorable” and “nasty woman” for the specified dates and scraped the resulting tweet text. Because Twitter’s advance search engine loads and displays tweets as a user scrolls down the list of results, we employed Web Scraper’s scroll feature to scroll down and load all results while scraping the information in an automated fashion. This process generated 6402 tweets for “deplorable” and 8854 tweets for “nasty woman.” We removed tweets that were not in English and those that used “deplorable” and/or “nasty woman” but were not referring to the candidates or election. This latter group of tweets was simply “caught up” in our scraping process but had no relevance to the research questions. In the end, our sample consisted of 14,324 total tweets (5708 from “deplorable” and 8616 from “nasty woman,” see Table 1).¹ These represent all publicly available tweets generated during the designated time period including retweets, those that use the terms as part of tweet text and those that use the terms in hashtags.

Table 1.

Tweet breakdown summary.

	Deplorable	Nasty woman	Total
Humorous tweets	1147	3793	4940
Humor with politics/percentage of humorous tweets that make serious political claims	870/76	2468/65	3338/68
Total tweets	5708	8616	14,324

Drawing on advances in computations methods, existing research has used “big data” to determine the prevalence of humor as a form of political engagement on social media, showing distinct patterns in which silly irreverence is both widespread and rewarded with attention and engagement (Driscoll et al., 2013; Hartley, 2012; Highfield, 2016a; Wilson, 2011). Building on existing work, our research questions address what people are doing with humorous political content on Twitter. This is a “small data” question, one that entails careful analyses of individual tweets and their specific contents. We thus take an interpretive approach to Twitter data, going beneath the top layer to answer questions about “what’s going on here?” (see Davis and Love, 2018; Love et al., 2018a, 2018b; Moloney and Love, 2017).

Our approach to the data was guided by abductive analysis which merges deductive theoretical insights with careful inductive techniques such as iterative analysis, de-familiarization, triangulation, and adjustment through review, critique, and debate (Tavory and Timmermans, 2014; Timmermans and Tavory, 2012). Concretely, we hand-coded each tweet identifying cases that were humorous, and of those cases, which ones projected a political message. Jokes, sarcasm, wordplay, irony, and mockery were all counted as humorous, as well as content the authors agreed had a humorous “feel.” Notably, we did not tie designations of humor to our personal comedic tastes. Thus, we coded tweets as “funny,” regardless of whether or not they made us laugh. We identified political tweets as those that expressed any sort of political messaging and/or revealed any kind of political agenda. The bar for inclusion as “political” was that the content had to meet at least one of the following criteria: conveys preference for a political candidate; articulates a position on one or more social issues; presents commentary on the political process; and/or indicates intention to, or evidence of, participation through voting, fundraising, and/or collective action. The authors first coded a segment of tweets independently and then conferred and conversed until clear criteria and examples were established. At that point, the third author (G.K.) completed the initial coding process.

Answering our research questions required data at the intersection of politics and humor. “Basket of deplorables” and “nasty woman” each represent political events. To ensure that their use on social media was, in fact, funny, we counted the proportion of tweets that used humor. In line with expectations, a substantial proportion of tweets incorporated some form of playfulness (N = 4940; ~35%), validating our use of these data to understand how humor and politics relate on social media.

Next, we focused on those tweets that both employ humor and enact political engagement. There were 3338 tweets that fell into the funny and political category. We constructed broad and narrow themes to help identify patterns of “silly” political participation. Themes were refined and reworked until all data fit into at least one thematic. Notably, many pieces of data (i.e. tweets) spanned across themes, demonstrating multiple forms of political work. Therefore, the themes we present are not discrete but maintain porous and intersecting boundaries. In addition to combing through the tweets themselves, we also followed all links and read tweet-threads to gather context.

Coding humorous content means engaging with the various forms that communication takes through the Twitter platform and the varied ways that humor and political messaging can manifest in this space. Although Twitter places a strict character limit on text (140 characters at the time of data collection but 280 characters at the time of this writing), messages can be deeply layered, derivative, and polysemic. In turn, the content of a tweet may be straightforward, subtextual, intertextual, use only words, use only images, or rely on multimedia through gifs, videos, screenshots, and sound bites. These concise communications, then, are often tightly packed with meaning. Our coding process hinged on unpacking these texts and arranging them in meaningful ways. The exemplar tweets we display in the analysis are selected for their representativeness of content in a given category.

While our data present a neat case study at the intersection of politics and humor on social media, our “small data” approach and use of Twitter as a data collection site does entail limitations. Without network maps, we cannot identify key players driving content trends. We are also limited to our specific search terms and limited to those who use Twitter, thus excluding social media users who may contribute content on other platforms. We also have no way of verifying that accounts are connected to real people, and we therefore assume that at least parts of our data are bot generated. Results should be read with these limitations in mind.

Analysis

Generally, we are interested in what Twitter users do with their political humor. Our analytic strategy began by identifying data at the intersection of politics and humor.

As shown in Table 1, out of 4940 humorous tweets (1147 from “deplorable” and 3793 from “nasty woman”), we identified meaningful political messaging in 3338. This means that ~ 68% of humorous tweets maintained a political agenda. These data answer our initial research question by presenting evidence that humor on social media is indeed a vehicle for serious political engagement. This finding resonates with existing work that shows an infusion of serious politics into the seemingly mundane practices of digitally mediated life (Highfield, 2016b) and mitigates concern that political humor breeds only apathy and cynicism (Baumgartner, 2007).

As elucidated below, humorous political content on Twitter works in three key ways—expressing opposition, establishing political subjectivity, and bolstering civic support. Fleshing out these forms of political engagement brings deeper understanding to how humor operates in political communication online. In turn, by mapping these categories across the two data sets and paying attention to the political leanings of tweet content, we can start to address the place of symbolic boundaries in digitally mediated humor. Meyer’s schematic of humor origins (relief, incongruity, and superiority) and effects (unity and division) provides a useful framework for distinguishing between our three categories of political work and making sense of their distribution across the data sets. Specifically, we see incongruity and relief as humor mechanisms throughout, with a sharp emphasis on superiority in expressions of opposition. In turn, opposition emphasizes divisive effects, while identification and civic support emphasize unity (though maintain divisive elements). Because each data set is highly partisan, patterns and distinctions that emerge speak to humor style across ideological boundaries.

What does the humorous content do?

Although we find a substantial proportion of tweets that do some form of political work, there are also those that do not seem to have any discernible political agenda and, instead, use political source material to generate humorous content (N = 1602; 32% of all funny tweets). In these tweets, politics offer fodder for jokes, but jokes remain politically indifferent. For example,

Wish I had majored in deplorable basket weaving in college

“Basket of Deplorables” would be a good name for a rock group

I mean, it DOES sound like an 80’s superhero battle flick. Nasty Woman vs Bad Hombre: This time … it’s bigly serious

Alright who’s making the nasty woman and bad hombre Halloween costumes because I know somebody did …

Tweets such as these do not make a political argument, and one would have a difficult time discerning the political leanings of the tweeters. These exemplify instances in which political content serves playfulness, rather than instances of political messaging delivered humorously.

The remaining tweets in the data set (3338; 68% of all funny tweets) also use humor and, in some cases, appear quite similar to the previous examples. These remaining tweets, however, use humor to deliver political messages. That is, they do some form of political work and thus represent a form of political participation.

Discrediting the opposition

Twitter users employ humor to discredit the opposition through mocking, parody, and talking back. “Deplorable” and “nasty woman” threads are sites on which the merits of each candidate are hotly debated, and fitness for leadership aggressively questioned. While the data sets mostly follow partisan lines (i.e. “deplorable” tweets mock Clinton and support Trump; “nasty woman” tweets mock Trump and support Clinton), “opposition” is the one category in which discourses crisscross the political spectrum, at least to a small degree. Here, detractors issue harsh accusations while a contingent of supporters of each candidate push back and flip accusations in the alternate direction. Superiority features most prominently in this category as a humor mechanism, while expressions of opposition emphasize division vis-à-vis unity as humor’s main effect.

About two-thirds of all funny-political tweets contain elements of opposition, by far our most heavily populated category. Equally represented in “deplorable” and “nasty woman” data sets, both Trump and Clinton supporters utilize mockery and denigration to discredit the opposing candidate and that candidate’s agenda while often ridiculing the opposition’s voter base.

For many Trump supporters, “basket of deplorables” is evidence of Clinton’s elitism and disconnection from everyday Americans. With her well-documented personal wealth, the revelation of highly paid corporate speaking engagements, and her status as a “career politician,” Clinton was often cast as a member of the ruling class, thus calling into question her capacity to govern the general populace. Below, detractors reference expensive vacation spots and historical figures of royalty to highlight Clinton’s socialite status:

Someone help me out. Was Hillary in the Hamptons, on Martha’s Vineyard, or in Hollywood when she called working class Americans deplorable?

Marie Antoinette “Let them eat cake.” Hillary “Deplorable.” She just lost. Keep the fire burning! Trump 2016. (https://twitter.com/seanhannity/status/774742748985454592) (link to a statement by conservative TV personality Sean Hannity, accusing Clinton of waging a “Holy War” against Donald Trump and listing key issues on which the candidates differ: immigration, supreme court nomination, and energy)

For Clinton supporters, Trump calling Clinton a “nasty woman” further entrenched concerns over Trump’s misogyny. Questions about Trump’s treatment of women have long plagued the candidate due to a public history of attacking the physical appearance of women who leverage critiques against him. Trump’s troublesome gender relations came to a head during the campaign with the release of old audio and video in which Trump brags about grabbing women’s genitals. Tweeters, like the one below, connect key lines from that released tape (i.e. “grabbing women by the pussy”) with Trump’s nasty woman comment during the third debate. They also replace “nasty woman” with “bitch,” unveiling the supposed real meaning behind Trump’s words:

“Such a Nasty Woman,GRAB THEM BY P*SSY, Nobody has more respect for women than me”—donald trump

Clinton “bitch slapped” him last night; I mean “nasty woman slapped” him … Pretty sure that’s the euphemism Trump implied.

While elitism and misogyny feature prominently in “deplorable” and “nasty woman” tweets, respectively, additional issues come to the fore as well. For instance, Trump supporters reference a scandal with Hillary Clinton’s emails, her supposed failures as Secretary of State (and the resulting deaths of American operatives in Benghazi), her physical appearance, supposed character flaws, and alleged bodily weakness. In turn, Clinton supporters make claims that Donald Trump is racist, lacking intelligence, immature, and unprepared for the duties of president. We sample some of this content below:

Mrs Deplorable will have to take a few days off from parties in Hollywood, she’s in the bed, deplorabley tired. #LockHerUp #TrumpTrain

It’s a Great Honor to be called “Deplorable” by this Lying Degenerate Criminal Racist Bigot! #BasketOfDeplorablespic.twitter.com/Vey3SyrqyY (unflattering image of Clinton overlaid with text: “she belongs in prison, not the Whitehouse” [referencing investigations into her email server and possible mishandling of national security issues])

I guess if I say he’s a misogynistic bigot, that makes me a Nasty Woman? Huh … @RepBrianBabinhttps://twitter.com/tpm/status/789573733656715264 … (link to story about Republicans defending Trump’s comment)

@HillaryClinton “Nasty Woman” takes care of business including a whining old baby Trump.

Finally, opposition tweets are unique among the data for their cross-partisanship. While the candidates’ gaffes were primarily used against them, about 10% of tweets in each data set show Clinton and Trump supporters doubling down on their candidates’ claims. The following are examples of Clinton supporters defending the characterization of Trump voters as a “basket of deplorables,” and Trump supporters claiming that calling Clinton a “nasty woman” is an accurate (or even kind) description:

What half of #LoserDonald Trump supporters are REALLY pissed off about is that they had to look up what “deplorable” means

White racists, birthers & Trump supporters must feel betrayed by Hillary. “Since when is white supremacy deplorable??” they gasp in unison.

This CRAP is what kills me. MSM [mainstream media] luvs 2 lick this end of the turd. #TRUMP says ‘sucha nasty woman.’OMG world! THE BITCH IS A VERY NASTY WOMAN!

She IS a nasty woman! pic.twitter.com/lEiBw7Jj7i (image of Clinton giving a speech behind a podium overlaid with text: “Listen, we’ve got to ban guns to save the toddlers, and we’ve got to have late term abortions to kill the babies”)

In short, Clinton and Trump supporters use humor—subtle, explicit, text-based, and multimedia—to discredit the opposition. This includes claims about elitism and misogyny, representing ongoing threads of critique leveraged against Clinton and Trump, respectively. These claims intermix with general and specific contentions about each candidate’s fitness for the office of president. In turn, supporters of each candidate also populate the data sets, doubling down, talking back, and reasserting their preferred candidate’s legitimacy. Stylistically, superiority is a chief driver of the tweets in the opposition category. While they do employ incongruity and offer comic relief, the opposition tweets play a primarily divisive function of denigrating political figures and political factions.

Identification

A second way that Twitter users wield humor for political ends is through identification. Identification includes constructing the self as a political subject by “reclaiming” derogatory labels, connecting one’s political preferences with other valued statuses, and establishing oneself as part of a political bloc. We read identification as the flip side of opposition. While oppositional humor relies heavily on superiority with a strong divisive effect, identification pushes back on denigration and maintains an in-group unifying function.

About one-third of all funny-political tweets contain an element of identification, with relatively even proportions of this category across data sets. Within each data set, the use of identification distributes cleanly along partisan lines, with all identification tweets in the “deplorable” data set leaning toward Trump and/or the conservative platform and all identification tweets in the “nasty woman” data set leaning toward Clinton and/or the progressive platform.

Reclaiming represents a key form of identification in both data sets. Just as “queer,” “bitch,” and “crip” have been reclaimed by sex–gender minorities, feminists, and people with disabilities, respectively, “deplorable” and “nasty woman” emerge as monikers of pride. Contradicting the devaluation entailed in each candidates’ gaffe, “deplorable” is reconfigured into patriotic and hardworking American, while “nasty woman” comes to signal strength and intelligence with a feminist bent:

So now if you love ur country, ur a #deplorablesinabasket.

#BasketOfDeplorables we are a beautiful bunch of deplorables!!

I was going to be a nasty woman for Halloween, but I am already sexy, smart and generous

In addition, tweeters connect the (reclaimed) label to additional positively valued identities, infusing those identities with political relevance. For example,

Folks I’m not a Major.’Major Lee D Plorable’ read fast is Majorly Deplorable. I was only corporal in USMC #BasketOfDeplorables lol

Mother of 2 strong girls, business owner, feminist, caring, does for others, stays informed, leader, voter, proud “NASTY WOMAN” #imwithher

Finally, Twitter users identify themselves as part of a political bloc by drawing connections between themselves and other supporters through text, images, videos, and tagging. For example,

Hi #Hillary from a #Deplorable friend. Just one of the 10K+ at the FL event

smash that like if ur a nasty woman who supports other nasty women #NastyWomanForPresidentpic.twitter.com/dWZkNBo0ZK. (image of pop singer Rihanna wearing a Hillary Clinton shirt with images of Clinton’s face in the background superimposed over clouds)

In addition, Trump supporters on Twitter often reference the need for a “bigger basket” alongside images of packed rallies, while Clinton supporters sarcastically “thank” Trump for bringing them together, often with links to rallies and fundraising efforts.²

By reclaiming “deplorable” and “nasty woman” as markers of pride, connecting these labels to other valued identities, and establishing the self as part of a political bloc through images, links, and social affordances of the Twitter platform,³ the data show humor as a mechanism of identification. Unlike opposition, identification deemphasizes superiority and division and instead, places primary focus on unity and in-group formation.

Civic support

Discourse itself is an important form of political participation but in a democratic system, what matters most is who gets elected and which policies go into effect. “Basket of deplorables” and “nasty woman” worked to not only discredit oppositional candidates and identify the self as a political subject but also became forces of individual participation and collective action, fostering the broad category we call civic support. Civic support takes the form of voting (intention to and evidence of), fundraising, rallying, and mobilization.

Civic support makes up the smallest category of political action found in the data, featuring in about 20% of all funny-political tweets. In addition, civic support is unique among our three categories for its uneven split between data sets. While about a quarter of “nasty woman” tweets contain an element of civic support, less than 10% of “deplorable” tweets are grouped in this category. Like identification, civic support distributes along clear partisan lines between the data sets, with “deplorable” tweets showing support for Trump and/or the conservative platform and “nasty woman” tweets showing support for Clinton and/or the progressive platform. Civic support emphasizes superiority to a moderate degree (less than it is emphasized in “opposition” tweets but more than it is emphasized in “identification” tweets). Civic support maintains both unifying and divisive effects as tweets ubiquitously proclaim who stands together, but also, take shape through articulations of who supporters stand against.

Voting is the most straightforward way to participate in the democratic political process. Both data sets feature ballot box declarations, ballot box selfies, and calls on others to vote for a particular candidate. For example,

How’s Go “F” yourself, from a deplorable Independent who just changed her vote from Her to Him

#BasketOfDeplorables #Deplorable lives matter canvassing in Florida today! pic.twitter.com/HPYrbMWwvU (image of two women wearing Trump gear at a campaign center)

This nasty woman is taking my pussy to a voting booth to vote for @HillaryClinton Too bad we both can’t vote. #ImWithHer #NastyWomen

This nasty woman voted today. From a swing state. #ImWithHerpic.twitter.com/pO8uNwpl4.z (image of filled-in absentee voting ballot)

We also see advertisements and displays of merchandise that benefit politically relevant groups. Notably, while merchandise ads are present in both data sets, connections between the products and monetary donations to politically relevant organizations are unique to the “nasty woman” data set. These derive primarily from a viral campaign promoting shirts, hats, mugs, and even a fragrance with proceeds benefiting Planned Parenthood—a women’s health organization that conservatives (including Trump) continuously threaten to defund—and Emily’s list—a political action committee supporting pro-choice candidates:

There’s now a Nasty Woman perfume that the proceeds benefit Planned Parenthood. I love the internet #nastywoman

Just ordered a Nasty Woman shirt because of course I did. (https://googleghost.com/collections/all/nasty-woman … [half of the proceeds goes to Planned Parenthood! (link to product)])

Civic support from both camps, though tied to the presidential candidates, also spread down ballots and over to related conservative and progressive issues and affiliations. For example, Trump supporters called for a boycott of Hollywood films due to a perceived progressive orientation in the entertainment industry, and a boycott of the National Football League (NFL) due to players’ conscientious objections to standing during the National Anthem. In this vein, Clinton supporters rallied for down ballot Democrats and distributed information about organizations that specifically benefit girls and women.

Civic support is the smallest and most partisan category of the three. Its relatively minor role vis-à-vis opposition and identification indicates that perhaps Twitter is most widely used for expressions of self rather than forms of concrete mobilization. Yet, the presence of mobilizing behaviors (voting, rallying, fundraising, etc.), however small, show the potential use of Twitter as a medium not just for political speech but also political action (Tufekci 2017). At the same time, it is curious that while civic support features about two-and-a-half times more often in Clinton/“nasty woman” tweets than Trump/“deplorable” tweets, the final election results fell in favor of Trump. Of course, Twitter is not representative of the voting public, civic support makes up only a small proportion of our content overall, and the category includes participatory acts other than voting (e.g. fundraising and rallying). However, the discrepancy between shows of civic support in our data and ultimate election outcomes presents a puzzle that should be teased out in future research.

The clear partisan lines along which civic support distributes also speaks to questions about humor and symbolic boundaries. The gap between Trump and Clinton supporters in this category exemplify those hard cultural boundaries long-documented by humor scholars in which ideological differences foster distinctions in humor sensibilities. This boundary distinction in civic support is notable, as humor style in the other two categories show no partisan patterns. Thus, while opposition and identification indicate that the internet’s shared vernacular may have boundary breaching effects upon humor, the partisan split of civic support tweets shows that symbolic boundaries remain at least somewhat intact.

Discussion and conclusion

Through analyses at the intersection of politics and humor on Twitter, we posed two research questions. First, we examined the degree to which humor is a vehicle for serious political engagement vis-à-vis politics as fodder for humorous content. Next, we explored how humor styles distribute along ideological lines.

To the first question, our data show that nearly 70% of funny-political tweets demonstrate some form of political work, advancing clear evidence that humor acts as a vehicle for meaningful political participation. This is relevant to debates about the effects of humor and emergent technologies on the democratic process. While both humor and digital media may lower barriers to political participation (Bode, 2016; Hariman, 2008; Harlow, 2012; Hartley, 2010), they can also foster cynicism and apathy (Baumgartner, 2007). Although not all political engagement is humorous and not all humorous content does political work, we show a substantial proportion of funny tweets that engage politically through various forms of opposition, identification, and civic support. Not only does this finding answer our research question but also clarifies the stakes for issues of digital inequality. If social media are increasingly relevant to political action and if humor is central to social media communication practice, then exclusion from either due to access, skill, or media literacy will be increasingly detrimental to already marginalized voices. While we do not interrogate this latter issue in the present work, it is an important avenue for future research.

Our data illustrate three ways in which humor is used to engage the political sphere. We show tweeters discrediting the opposition, identifying as political subjects, and exercising civic support through voting, fundraising, and collective action. Because the data sets are so tightly partisan, analyzing the distribution of categories across cases gives us insight into the place of symbolic boundaries as humor is enacted online. While scholars have shown humor withstanding the general move toward cultural omnivorousness (Friedman and Kuipers, 2013; Kuipers, 2009, 2015), social media has been touted as the greater leveler, with a shared vernacular converged upon by diverse networks (Jenkins, 2006; Milner, 2013b). We therefore examined how antagonistic groups utilize humor in similar and dissimilar ways.

Using Meyer’s (2000) schematic, we show that humor style looks similar across partisan lines for both opposition and identification. However, civic support distributes unevenly between the two data sets, suggesting the kind of boundary distinction that resonates with traditional humor studies. Thus, in answering our second research question, this article shows humor on social media as a distinct cultural form. Humor remains tied to cultural bounds, but those bounds are perhaps loosened through articulation on the Twitter platform. At the same time, social media’s shared vernacular and convergence culture may emerge more bounded when articulated through humor, politics, and/or their combination.

Establishing the serious political work of humor on Twitter is an important first step toward understanding a participatory political environment. In this way, our study contributes to studies of both politics and media. At the same time, our use of partisan cases illuminates the place of cultural boundaries as humor is enacted online. Together, these findings set the stage for additional work at the intersection of politics, digital media, and humor.

For example, in comparing data sets, we noticed that “nasty woman” contained vastly more content than “deplorable.” There are several possible reasons for this—the public nature of the debate, its parody on Saturday Night Live, proximity to election day—but another possible reason is of theoretical interest. It may be that Clinton supporters/progressives are more invested in social media participation, more vigorous in their use of humor, or some combination of these factors. The disproportionate use of humor among the progressive faction in our data resonates with the progressive bent of US comedians in both stand-up comedy and political commentary shows. While the data do not offer clear answers in this regard, future work would do well to examine not only how people engage in digitally mediated political humor, but also how political affiliation affects one’s relationship to humor. In other words, for whom is humor considered a viable and likely form of political participation?

Future work would also benefit from tracing the effects of political humor as enacted on social media. Our data indicate that Twitter is an important forum for political participation, but it remains unclear how Twitter engagement translates into votes. Our findings highlight the complexity of this question, as we saw a disproportionate show of civic support in favor of Clinton, yet Clinton ultimately lost the election. So, while the presence of civic support in our data along with meaningful political claims-making in the other two categories counter concerns about humor causing apathy, the disconnect between expressions of civic support and the outcome of the election underscores the need to understand the relationship between humor on social media and concrete political dynamics.

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author biographies

Jenny L Davis is a lecturer/assistant professor in the School of Sociology at the Australian National University and co-editor of the Cyborgology blog. She studies the interplay of emergent technologies and social dynamics.

Tony P Love is an assistant professor of Sociology at the University of Kentucky. A social psychologist and a criminologist, he has expertise in digital data methods and is interested in juvenile delinquency, intimate partner victimization, and the relationships between power, gender, and the ability to accurately take the role of the other.

Gemma Killen is a PhD Candidate in the Research School of Social Sciences at The Australian National University. Her research focuses on intersections of gender and sexuality in the contemporary West and her current work explores the ways in which internet technologies are taken up as tools of community building, particularly for young queer people.

References

Barnard

(2016) “Tweet or be sacked”: Twitter and the new elements of journalistic practice. Journalism 17: 190–207.

Baumgartner

(2007) Humor on the next frontier: youth, online political humor, and the JibJab effect. Social Science Computer Review 25: 319–338.

Baumgartner

Morris

(2006) The Daily Show effect: candidate evaluations, efficacy, and American youth. American Politics Research 34: 341–367.

Bode

(2016) Political news in the news feed: learning politics from social media. Mass Communication & Society 19: 24–48.

Burgess

(2006) Vernacular creativity, cultural participation and new media literacy: photography and the Flickr network. Available at: http://creativitymachine.net/downloads/publications/JeanBurgessAoIR2006.pdf

Castells

(2013) Communication Power. Oxford: Oxford University Press.

Davis

Chouinard

(2016) Theorizing affordances: from request to refuse. Bulletin of Science, Technology & Society 36: 241–248.

Davis

Love

(2018) Women who stay: a morality work perspective. Social Problems 65(2): 251–265.

Dayan

Katz

(1994) Media Events. Cambridge, MA: Harvard University Press.

10.

DiMaggio

Mukhtar

(2004) Arts participation as cultural capital in the United States, 1982–2002: signs of decline? Poetics 32: 169–194.

11.

Driscoll

Ananny

Bar

et al . (2013) Big Bird, Binders, and Bayonets: humor and live-tweeting during the 2012 U.S. Presidential debates. Available at: https://spir.aoir.org/index.php/spir/article/download/736/326

12.

Featherstone

(2007) Consumer Culture and Postmodernism. London: SAGE.

13.

Friedman

Kuipers

(2013) The divisive power of humour: comedy, taste and symbolic boundaries. Cultural Sociology 7: 179–195.

14.

Gal

Shifman

Kampf

(2016) “It gets better”: Internet memes and the construction of collective identity. New Media & Society 18: 1698–1714.

15.

Gray

Jones

Thompson

(2009) Satire TV: Politics and Comedy in the Post-Network Era. New York: NYU Press.

16.

Hargittai

Walejko

(2008) The participation divide: content creation and sharing in the digital age. Information, Communication & Society 11: 239–256.

17.

Hariman

(2008) Political parody and public culture. Quarterly Journal of Speech 94: 247–272.

18.

Harlow

(2012) Social media and social movements: Facebook and an online Guatemalan justice movement that moved offline. New Media & Society 14: 225–243.

19.

Hartley

(2010) Silly citizenship. Critical Discourse Studies 7: 233–248.

20.

Hartley

(2012) Digital Futures for Cultural and Media Studies. Chichester: John Wiley & Sons.

21.

Henefeld

(2016) Laughter in the age of Trump. Flow. Available at: http://www.flowjournal.org/2016/12/laughter-in-the-age-of-trump/

22.

Highfield

(2016a) News via Voldemort: parody accounts in topical discussions on Twitter. New Media & Society 18: 2028–2045.

23.

Highfield

(2016b) Social Media and Everyday Politics. Cambridge: Polity Press.

24.

Jenkins

(2006) Convergence Culture: Where Old and New Media Collide. New York: NYU Press.

25.

Kuipers

(2009) Humor styles and symbolic boundaries. Journal of Literary Theory 3: 219–239.

26.

Kuipers

(2015) Good Humor, Bad Taste: A Sociology of the Joke. Berlin: Walter de Gruyter.

27.

Lamont

Fournier

(1992) Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago, IL: The University of Chicago Press.

28.

Leavitt

(2014) From# FollowFriday to YOLO: exploring the cultural salience of Twitter memes. In: Weller

Bruns

Burgess

(eds) Twitter and Society. New York: Peter Lang, pp. 137–154.

29.

Love

Davis

Calvert

(2018a) A Mixed-Methods Examination of Morality Work through Sentiment Analysis and Qualitative Coding of Twitter Data. London. SAGE. Available at: https://dx-doi-org.web.bisu.edu.cn/10.4135/9781526435323

30.

Love

Moloney

Bunting

(2018b) Analyzing Virtual Manhood: Qualitative Analysis of Fappening-Related Twitter Data. London. SAGE. Available at: https://dx-doi-org.web.bisu.edu.cn/10.4135/9781526429797

31.

Meyer

(2000) Humor as a double-edged sword: four functions of humor in communication. Communication Theory 10: 310–331.

32.

Milner

(2013a) FCJ-156 hacking the social: Internet memes, identity antagonism, and the logic of Lulz. The Fibreculture Journal. Available at: http://fibreculturejournal.org/wp-content/pdfs/FCJ-156Ryan%20Milner.pdf

33.

Milner

(2013b) Pop polyvocality: Internet memes, public participation, and the Occupy Wall Street movement. International Journal of Communication 7: 2357–2390.

34.

Miltner

(2014) “There’s no place for lulz on LOLCats”: the role of genre, gender, and group identity in the interpretation and enjoyment of an Internet meme. First Monday 19.

35.

Moloney

Love

(2017) #TheFappening: virtual manhood acts in (homo) social media. Men and Masculinities. Epub ahead of print 9 March. DOI: 10.1177/1097184X17696170.

36.

Mossberger

Tolbert

Anderson

(2017) The mobile Internet and digital citizenship in African-American and Latino communities. Information, Communication & Society 20: 1587–1606.

37.

Nielsen (2016) Election central: Nielsen’s hub for the data behind the 2016 presidential election. Available at: http://www.nielsen.com/electioncentral

38.

Nissenbaum

(2011) A contextual approach to privacy online. Daedalus 140: 32–48.

39.

Nooney

Portwood-Stacer

(2014) One does not simply: an introduction to the special issue on Internet memes. Journal of Visual Culture 13: 248–252.

40.

Peterson

Kern

(1996) Changing highbrow taste: from snob to omnivore. American Sociological Review 61: 900–907.

41.

Phillips

Milner

(2017) The Ambivalent Internet: Mischief, Oddity, and Antagonism Online. Hoboken, NJ: Wiley.

42.

Prior

(2006) Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections. Cambridge: Cambridge University Press.

43.

Shifman

(2007) Humor in the age of digital reproduction: continuity and change in Internet-based comic texts. International Journal of Communication 1: 23.

44.

Shifman

Coleman

Ward

(2007) Only joking? Online humour in the 2005 UK general election. Information, Communication & Society 10: 465–487.

45.

Tavory

Timmermans

(2014) Abductive Analysis: Theorizing Qualitative Research. Chicago, IL: The University of Chicago Press.

46.

Timmermans

Tavory

(2012) Theory construction in qualitative research: from grounded theory to abductive analysis. Sociological Theory 30: 167–186.

47.

Tufekci

(2017) Twitter and Tear Gas: The Power and Fragility of Networked Protest. New Haven, CT: Yale University Press.

48.

Vaccari

Chadwick

O’Loughlin

(2015) Dual screening the political: media events, social media, and citizen engagement. Journal of Communication 65: 1041–1061.

49.

Vis

(2013) Twitter as a reporting tool for breaking news: journalists tweeting the 2011 UK riots. Digital Journalism 1: 27–47.

50.

Webber

Sarlos

Eckhardt

et al . (2013) The Cultural Set up of Comedy: Affective Politics in the United States Post 9/11. Bristol: Intellect Ltd.

51.

Wells

Van Thomme

Maurer

et al . (2016) Coproduction or cooptation? Real-time spin and social media response during the 2012 French and US presidential debates. French Politics 14: 206–233.

52.

Wilson

(2011) Playing with politics: Political fans and Twitter faking in post-broadcast democracy. Convergence 17: 445–461.