First Monday

Archiving affect and activism: Hashtag feminism and structures of feeling in Women's March tweets by Kristi McDuffie and Melissa Ames

On 21 January 2017, over three million women participated in the Women’s March throughout the U.S., one day after President Donald Trump’s inauguration. This article investigates the digital component of this historic protest as a powerful moment of hashtag feminism, one that exemplifies the vital role of affect in contributing to social change. Through qualitative analysis of 2,600 #WhyIMarch tweets from the day of the March, we identify the rhetorical strategies that best leverage affect to further the social justice goals of the March — dedications, personal narratives, the use of first-person, and the use of humor — and describe the affective outcomes of these strategies, including motivational affect, vicarious affect, and collective affect. Using Raymond Williams’ concept of “structures of feeling,” we argue that these rhetorical strategies and their affective outcomes create a digital archive of affect that captures the cultural climate surrounding the Women’s March and mediates the way this cultural moment is affectively remembered. This study reveals that affect is vital for effective hashtag feminism.


Scholarly conversations
Findings: Prominent rhetorical strategies
Discussion: Affective outcomes
Implications for hashtag activism




On 21 January 2017, over three million women participated in the Women’s March throughout the U.S., one day after President Donald Trump’s inauguration. There were half a million marchers in Washington, D.C., more than doubling the projected attendance of 200,000 (Politi, 2017). The stated goals of the organizers — which included a robust 16 values and principles — ranged from reproductive rights to immigrant rights and voting rights (Women’s March, 2017). The March was both a collaborative movement for social justice and a targeted demonstration protesting the election of Donald J. Trump as the 45th President of the United States.

Along with this historic March emerged an impressive digital movement. There were 11.5 million tweets on 21 January alone, rivaling the 12 million tweets from the previous day’s inauguration (Cohen, 2017) [1]. Many of the tweets contained information and photos of the event, effectively creating an important and thorough archive for this monumental March. The majority of these tweets provided information and displayed photos of the crowds and protest signs from the March. Attached to these #WomensMarch tweets were a number of concurrent hashtags that drew attention to particular aspects of the March, including locations (i.e., #WomensMarchChicago) or particular issues (i.e., #NotMyPresident). Numerous users also articulated their reasons for marching under the #WomensMarch hashtag, and this theme emerged into the #WhyIMarch hashtag and, to a lesser extent, the #WhyWeMarch hashtag (Baker, 2017). #WhyIMarch tweets become an important part of the in-person and digital protest by detailing the issues and motivations for the protest. These tweets included a range of affective statements and images, and created a stand-alone digital movement based on personal and collective commitments to feminist issues. The #WhyIMarch movement is a powerful example of the ways that digital activism leverages affect to create social change. Papacharissi (2014) explains that the connections between “affect and ideology, feeling and belief, emotion and reason” all drive social movements and that affect — as the “energy that drives, neutralizes, or entraps networked publics” — especially drives social movements [2]. Affect is also inseparable from rhetoric. As Crowley explains, belief and affect are so deeply intertwined with rhetoric that these emotioned beliefs drive people’s values and ideologies [3]. These values and ideologies emerge in the rhetoric of social media movements like #WhyIMarch, which are driven by affect while simultaneously (re)producing it.

In this article, we study the intersections of affect, rhetoric, and social media movements by analyzing #WhyIMarch tweets for the rhetorical strategies that leverage affect to create social change, particularly within the context of hashtag feminism given that the organizing March principles were feminist goals (Women’s March, 2017). Our inquiry is driven by the following research questions:

  1. What were the most common rhetorical strategies used in #WhyIMarch tweets?
  2. What were the affective outcomes of these rhetorical strategies?
  3. What do these practices suggest about hashtag activism’s ability to archive affect and shape the experience and memory of social justice movements and the larger cultural moments from which they stem?

To investigate these questions, we collected over 200,000 #WhyIMarch tweets from the day of the March and coded a sample of 2,600 of those. We identify the rhetorical strategies that best leverage affect to further the social justice goals of the March — dedications, personal narratives, the use of first-person, and the use of humor — and describe the affective outcomes of these strategies, including motivational affect, vicarious affect, and collective affect. Using Raymond Williams’ concept of “structures of feeling,” we argue that these rhetorical strategies and their affective outcomes create a digital archive of affect that captures the cultural climate surrounding the Women’s March and mediates the way it is remembered and how it circulates in future hashtag feminism.



Scholarly conversations

Structures of feeling

This investigation into hashtag activism during the Women’s March is informed by Raymond Williams’ (1977) concept of “structures of feeling,” which captures the ways in which cultural experiences are felt and amalgamated into a collective affect state — the shared mood for a particular cultural moment, epoch, or generation. Williams (1977) points out that our “temporal present” is “active,” adaptable, “flexible,” and always in flux [4]. In Women’s March tweets, structures of feeling thus reflect the cultural climate, emotional atmosphere, and affective landscape of the moment. Furthermore, Flatley (2008) explains that structures of feeling operate subconsciously when he says that “we are often ignorant of the determinative effect our moods have on the world we see and how we relate to it” [5]. Flatley (2008) also points out that structures of feeling can be ephemeral or “just as durable and forceful as ideologies, perhaps even more so” [6]. Therefore, he argues, structures of feeling should be theorized as a “full-fledged parallel to ideology,” one whose function “is to narrate our relation to a social order so as to make our daily experience of that order meaningful and manageable” [7]. Our study considers how this narratorial role of structures of feeling plays out in and through hashtag activism, framing and shaping cultural narratives and collective memories.

Similar to Flatley, Best (2012) considers how Williams’ concept allows us to consider the interplay between emotion at both the micro and macro-levels. For her, a structure of feeling

addresses precisely the precarious balance between the forces of structure and agency, between the forces of the social process and the willing, intending, experiencing subject. “Structure of feeling” expresses the contradiction that our personal, intimate, individual experiences (feelings) are always, at the same time, informed by collective and historical prejudices, expectations, fears, desires, conventions, institutions, laws, and modalities of the social that transgress even the most extended view of the feeling subject. [...] The concept of a structure of feeling, therefore, for Williams, is an effort to capture the complex mediations between the particular and the general that animate any specific historical conjuncture. [8]

Best explains that a structure of feeling captures a variety of social, economic, political, and historical factors, personal and external, even when these factors are in conflict. Considering how these “complex mediations” work in digital spaces at particular cultural moments is one of the questions this study takes up.

In order to consider these complex mediations in hashtag activism, Coleman’s (2018) concept of “infra-structures of feeling” is useful to consider. Coleman (2018) applies Williams’ theory to the affective elements of digital media and how they operate “across each other in a complex architecture of texts, textures, platforms and devices” [9]. Coleman’s notion of infra-structures of feeling captures the mass assemblage (or “architecture”) of online texts and the ways in which a particular structure of feeling might be produced and organized through them [10]. Infra-structures of feeling provide a productive lens through which to consider hashtag activism. Although hashtags are tools for metadata tagging and searching (Losh, 2014), they have been largely taken up for activist purposes of protest, community organizing, and creating social change. Most notably, the Black Lives Matter movement (#BlackLivesMatter, #BLM, and dozens of individual hashtags from #JusticeforTrayvon to #GeorgeFloyd and #BreonnaTaylor) reached astounding levels in summer 2020 with an average of 3.7 million tweets per day (Cohen, 2020) following the killing of George Floyd by police officers, in addition to other cases of police and civilian killings of Black Americans. On many scales, though, social media activist movements can demonstrate the affective results that emerge from the organizing and archiving features of hashtags. Activist tweets exist within and form infra-structures of feeling as well as circulate and produce affect. To consider this further, we turn to one subset of digital activism that capitalizes on affective strategies: hashtag feminism.

Affect in/and hashtag activism

While digital activism focused on feminism has been discussed for 20 years, hashtag feminism as named practice has erupted recently. Dixon (2014), for example, traced the different ways that hashtag feminism has been enacted, such as to share personal experiences and to challenge dominant discourses. Dixon (2014) found that personal narratives in particular were an important feature of hashtag feminism for the ways that users can re-author their lives using stories otherwise constructed by men [11]. Dixon (2014) also articulated how emotion facilitates the creation of community [12]. Aligning with her research, this study found both personal narratives and community were important in creating structures of feeling within Women’s March tweets.

Often hashtag feminist scholarship investigates particular movements. Clark (2016) studied the #WhyIStayed hashtag, which emerged in response to the Ray Rice domestic violence news in 2014, where activists challenged the victim-blaming rhetoric of domestic abuse narratives. Clark (2016) argued that the movement illustrated the power of hashtag feminism to intervene in normative, oppressive discourses: “Hashtag feminism, in its form, content, and production process, empowers its users to take control of the sociocultural narratives associated with their identities and subjective experiences” [13]. Drüeke and Zobl (2016) expanded on the focus on personal narratives in hashtag feminism when they researched the #aufschrei hashtag (German for #YesAllWomen). They explained that hashtag activism allows activists to gather personal experiences to negotiate and redefine values and norms [14].

Some scholars explicitly took up affect by wrestling with questions posed by Ahmed (2004), such as “How are emotions bound up with stories of justice and injustice? How do emotions work through texts not only to ‘show’ the effects of injustice, in the form of wounds and injury, but also to open up the possibility of restoration, healing, and recovery?” [15]. For example, Gong (2014) illustrated how emotion helps construct collective identify, motivates action, and achieves network solidarity [16]. Gong (2014) highlights particular strategies in which digital activists engage:

Digital activists take advantage of different forms of media online to engage in emotion work. They personalize statistics, offering users increased accountability and responsibility; they tell inspirational stories of survivors and activists whose individual efforts can have larger social impacts; they provide opportunities for mutually affirming interactions with users, and they maintain a steady flow of communications (updates and interactions) with their followers to keep the movement salient and build community. [17]

This narrative and community building work that Gong and others have identified are part of the emotion work that Ahmed (2004) articulated as orienting action toward others [18] and Coleman (2018) proposed as integral to infra-structures of feeling. This study takes up this emotion work in #WhyIMarch tweets during the Women’s March and demonstrates how rhetorical strategies used by hashtag feminists shape and are shaped by powerful structures of feeling that capture and archive a powerful collective affective moment.




The day of the Women’s March — 21 January 2017 — one researcher collected tweets using NodeXL, a Microsoft Excel plug-in that, under the premium pay version, allows users to collect up to 18,000 lines of tweet data every 15 minutes. (The other researcher proudly marched in Washington, D.C.) NodeXL pulls tweets from within seven days, but with the high Women’s March volume, our ultimate sample of 244,000 #WhyIMarch tweets is only a partial representation of all of the tweets sent that day. To sample a manageable number of tweets for coding, we narrowed the sample to the NodeXL scrape at 12:18pm EST with 25,648 tweets, the closest data pull to the official start time of the March. We further narrowed that sample by analyzing only the original tweets, rather than including the retweets, leaving a robust sample of 2,645 tweets for coding.

Our qualitative data analysis is influenced by grounded theory as outlined by Corbin and Strauss (2008). Grounded theory consists of building theory from data by identifying concepts through coding (assigning values to units of text) [19]. The basis of grounded theory is to build knowledge from the data itself through the coding process. That is, patterns and meaning discovered through the process lead researchers to further areas of study, further coding concepts, and findings from the data. We coded inductively (from the data) and were also informed by concepts drawn from feminist issues and affect scholarship. This process resulted in codes that ranged from in vivo (from the words, such as “sisterhood”) to higher levels of abstracted meaning (such as the theoretical concept of imagined community).

We used the qualitative data analysis software dedoose ( to code the tweets. The coding was a recursive process of creating and revising codes to best describe the data overall and analyze particular affective rhetorical strategies in detail. Our final dataset went through two rounds of coding. We concluded with 87 different codes organized under four categories: Feminist themes, Emotion, Rhetorical strategies, and Media. The details under Feminist themes and Media ended up being ancillary to the focus on emotion and rhetoric and are thus not discussed in this article. We focus this discussion on the most prominent Rhetorical strategies and Emotion themes, first by describing the most utilized rhetorical strategies that employed affect, and then by discussing the most powerful ways that affect emerged within the tweets.



Findings: Prominent rhetorical strategies

In this section, we speak to the first research questions in our project: What were the most common rhetorical strategies used in #WhyIMarch tweets? How did these rhetorical strategies leverage affect to address social justice? Although our analysis process identified over eighty codes, ranging from expressed emotions (love, hate) to feminist issues (domestic abuse, wage gap), we focus on the rhetorical strategies that best utilized affect in order to address social change. These strategies include the use of dedications, personal narratives, first person plural pronouns, and humor.


One of the most frequent strategies used to supply an “answer” to the #WhyIMarch prompt was dedication statements that listed a particular person or group as the reason for the user’s participation. Among the variety of dedication tweets, generational posts focusing on children (those known to the poster or youth/girls more collectively) and mothers/elder women were the most common.

The posts dedicated to children especially invoke emotion, especially when children are included in the photos. For example, the first post in Figure 1 shows a baby wearing a sign reading, “leave a little planet left for me.” This sign is written from the child’s point of view, adding empathy and guilt to the message. The text of the tweet says “for him,” which demonstrates the user’s emotion, but the visual rhetoric of the included image adds another dimension to the affective positioning of the tweet. The author is marching explicitly for the future of this child (and perhaps by extension youth more generally) and the planet as a whole


Examples of tweets utilizing children for affective purposesExamples of tweets utilizing children for affective purposesExamples of tweets utilizing children for affective purposes
Figure 1: Examples of tweets utilizing children for affective purposes [20].


The next tweet in Figure 1 emotes hope and empowerment. The text of the tweet says, “these girls right here, that’s #WhyIMarch.” The image reveals a crowd at a March with young girls on top of a platform. A girl holds a ”Girl Power” sign, and the distance between the photographer and the girls indicates that the dedication is referring to children who are not personally known to the user. The dedication seems to refer to the girls at the March in specific and even the future of feminism in general.

The final tweet in Figure 1 reads, “Godbaby’s first demonstration #maevethebrave #whyimarch #formygoddaughter” and has an accompanying image of a group shot of two women holding an infant, all wearing hot pink “pussy hats.” The hats are a form of protest against President Trump’s well-quoted sexist and abusive comments that include “Grab ’em by the pussy” (Trump, 2016). The tweet ties together the hope of the dedication and the anger of the pussy protest to convey that the user is marching to reduce sexual abuse in the child’s — and all children’s — futures.

There were also a number of posts dedicated to older women, including users’ relatives and older generations of feminists. One example is a post dedicated to a recently passed-away grandmother. The text reads, “#IIMarchFor Jo Rita Murray. #whyIMarch #WomensMarch” and there are two images in the tweet: one of a typed dedication to the grandmother and one of a backpack with a “WhyIMarch” pin and a memorial flyer for the woman. The dedication includes information about the woman’s grandmother and explains, “I walk for all the women (dead or alive) like my grandmother who have fought for our rights.” This post, and others like it, reveals grief and inspiration and can inspire similar emotions in readers who have also experienced loss or who also feel sadness and frustration about earlier generations of feminists who expected the U.S. to have achieved more gender equality than it has.

By sharing their reasons for marching as being embedded in emotional relationships with children and elders, activists both personalize the feminist issues and create a shared affect of a commitment to change. These emotions circulate and resonate with other Twitter users and contribute to the structures of feeling during the March. These posts reveal motivational forms of nostalgia and optimism which, as we discuss below, serve as calls to action for other participants.

Personal narratives

Earlier hashtag feminist scholarship established personal narratives as important to digital protest. Personal narratives as a feminist rhetorical strategy brings visibility to a multiplicity of experiences, facilitates resistance to dominant narratives, and demonstrates strength and confidence [21]. For example, #WhyIStayed allowed users to share their own stories of domestic abuse, #EveryDaySexism provided a space for women to share their personal encounters with everyday misogyny, and #MeToo brought visibility to the sexual harassment and assault that individual women experienced. The #MeToo movement is one of the largest hashtag feminist movements in recent history and produced over 1.7 million tweets from users in 85 countries, with a third of these posting within the first 24 hours of the hashtag’s 2017 re-activation (Park, 2017). In this study, the personal narratives in the #WhyIMarch tweets were indeed powerful for their affective impact. The narratives in our dataset ranged from stories of the user’s experiences the day at the March to past experiences explaining what she was protesting against.

Numerous stories from the day of the March relayed anger about misogynistic encounters with men opposed to the event. The encounters included stereotypical accusations of women being overly emotional (e.g., “because in a cab on my way to #WomensMarch TO the cab driver told me I was overreacting”), comments that reinforced the patriarchal status quo (e.g., “#WhyIMarch: a man just yelled ‘Men rule the world. You can’t change anything.’ to a crowd of women on their way to the March. #WatchMe”), examples of “mansplaining” (e.g., “On our way to the Women’s March and our Uber driver feels he should lecture us #womensmarchonwashington #menknowall #whyimarch”), and even statements that encouraged violence against women (e.g., “On way to #WomensMarchboston when guy tells train conductor to run over any protesters on the tracks. #whyimarch”).There were also posts revealing positive interactions during the March, such as inspirational connections between strangers (e.g., “a little girl gave me this sticker to thank me for marching with her,” “A woman at the March said my Nana is an ‘inspiration’ I’m sobbing,” “I met a 77 year old woman who said it was her first time marching, ever. She said it was too important & she couldn’t stay home”). These latter examples demonstrate emotions of hope, nostalgia, and encouragement.

Finally, there were personal narratives in the tweets that provided background motivation for why the user was protesting. For example, one user explained she was marching because she was sexually assaulted and now experiences the stigma of being a single mom. In contrast to the multiplicity of tweets that detailed feminist issues as reasons for marching, such as domestic violence more generally, this personal example conveys significant emotion and invites audience members to share in those emotions of anger, frustration, and shame. As Gong (2014) explained, personal stories add detail to impersonal statistics, ask readers to take more accountability and responsibility, and inspire others to take action [22]. The narratives sharing these deeply personal experiences draw on emotion to implicitly ask other Twitter users to be accountable and responsible for improving women’s lives in the future and to inspire them to take action. Together, these narratives coalesce into a positive structure of feeling, largely due to the collective empathy they evoke. As discussed below, this Twitter dialogue documents a range of women’s experiences in a form of “herstory,” while also archiving the larger cultural climate created as these women and their stories came together to join in social protest at this particular historical moment.

First person plural pronouns

One of the most direct rhetorical strategies that leveraged affect in #WhyIMarch tweets was the use of first person plural pronouns. The pronoun “we” implies a multiplicity of subjects that, together, impact the rhetorical uptake of the tweet and build affective community among users.

There were five main ways that authors utilized first person plural within this dataset. As the majority of the tweets reported participation and described the Marches, one of the primary uses of first person was to create an audience for the protests (e.g., “You will not disrespect and treat women like second class citizens, nah! Not on our watch,” emphasis ours). The next common use of first person plural was simply to refer to women as the subject of the March (“bc the statement ‘but you’re a woman’ should never make us feel inferior or ashamed like we’re at some sort of disadvantage,” emphasis ours). First person plural was also often used to denote American citizenship. “We the people” appeared as a phrase or hashtag within multiple tweets, as did other statements wherein the focus on U.S. citizens is relatively obvious, for example with references to “our president” (e.g., “Because she deserves to grow up in a world where our Pres doesn’t #grabherbythepussy,” emphasis ours). We/our was also used to suggest that the author’s anticipated audience shared the same political ideology, implying that the reader would be a Democrat or liberal. Consider, for example, the following tweet: “This is what the #snowflakearmy looks like. We stand for women’s rights! We stand for human rights! We will not sit down!” (emphasis ours). This example also reclaims a pejorative label (“snowflake,” referring to the way that President Trump has referred to liberal Democrats) and turns it into a symbol of power (#snowflakearmy). The use of “we” suggests that the reader, too, is a self-identified liberal. Finally, some tweets used first person plural in a broad sense to represent people and humanity in general (e.g., “Girls are our future,” emphasis ours).These posts using “we” emote excitement, anger, joy, and more, but more than just portraying these emotions, the use of “we” implies that audience members share these emotions.

Combined, these uses of first person plural create an imagined affective community, one with shared values, motivations, and experiences. In some situations, these pronouns infer a shared benefit of collective action (e.g., “60% of minimum wage workers are women. Let’s #RaisetheWage and give our economy a massive boost,” emphasis ours). This rhetorical strategy may resonate with readers by implying that they are invested in an issue or a recipient of an injustice, thereby encouraging readers to act on this injustice. But these pronouns carry a risk of being exclusionary. Using “we” to refer to women excludes the many men and gender-non-binary people who participated in the Marches or care about the feminist issues. Similarly, the use of “we” to refer to U.S. residents fails to acknowledge the allies around the globe, including those who attended the many sister Marches around the world. Despite these limitations, the rhetorical strategy of first person plural is both subtle and pervasive, and is perhaps one of the most effective ways that the various tweets function together to collectively create structures of feeling. These archived posts reveal (assumed) shared sets of beliefs and world outlooks, as well as highlight how those world views go hand-in-hand with shared feelings about issues and societal circumstances.


Humor was another dominant theme in the #WhyIMarch tweets. Activists used humor to draw attention to problematic issues while using a jovial tone in their tweets and promoting a positive tone for the protest overall. Sowards and Renegar (2006) explain that invoking humor in the face of oppression and discrimination is a show of strength and conveys hope and creativity [23]. Humor is also a tactical rhetorical device given that Twitter privileges humorous and inflammatory posts in that they are more likely to be retweeted [24].

Much of the humor in the #WhyIMarch tweets appeared in the photographs of signs. Users often posted photos of signs that they were carrying or that they saw, and some humorous signs were tweeted without added commentary or with hashtags as commentary. Some examples include: “So bad, even introverts are here” and “CTRL ALT-RIGHT DELETE.” Some of the most effective uses of humor are tweets that utilized interplay between the text and image, even using the photographs as punch lines to the textual set-up. Figure 2 shows an example of interplay between the linguistic and the visual modes, where the first tweet reads “My to do list” and the photograph reveals signs stating, “Dismantle White Supremacy” and “Smash the Patriarchy.” These humorous tweets reveal anger, but the humorous tone refocuses the emotion to inspiration and hopeful action.


Examples of humorous signsExamples of humorous signs
Figure 2: Examples of humorous signs.


In addition to revealing political commitments and invoking change as activism strategies, the humorous signs facilitated an affective connection between the in-person Marches and the digital protests. Tweeting photos of the signs, especially with added commentary, continued the conversations started in-person by furthering the reach of the message and multiplying the audiences participating in the collective community. In addition, the sense of hope conveyed through this (critical) humor circulates emotion and contributes to the shifting structures of feeling created by the hashtag feminism of the March tweets overall. Humor-infused posts arguably played a role in “lightening” the pervasive mood of the moment. However, similar to the strategies discussed above, they also point toward a toggling point wherein protesters shifted from negative to positive affect, from reaction to the recent election to action for future change.



Discussion: Affective outcomes

Now that we have identified the most prominent rhetorical strategies utilizing affect in #WhyIMarch tweets, we turn to addressing the second research question: What were the affective outcomes of these rhetorical strategies? In other words, what emotional effects are evident as a result of these strategies? While many tweets in the sample elicit emotion, we focus on three particular outcomes that emerge from the rhetorical strategies discussed above: calls to action, expressions of vicarious affect, and evocations of imagined communities. These outcomes build on previous research that consider social media’s role in triggering, intensifying, and circulating affect [25]. Pybus (2015) explains the circulation of affect as “a relational, affective model ... for understanding social networks,” one that attends to the ways “each message, note, and photo that gets uploaded carries with it the ability to affect not only a friend in the network but equally the individual user, based on the way this object is received by the members of his or her respective community” [26]. This analysis considers the affective impact that tweets have on the user, those engaging directly with the post, and those consuming the post (as participants of the digital and/or in-person protest). When users tweet, retweet, and respond to tweets, they forward emotion as much (or perhaps more than) content and collectively construct structures of feeling related to the moment (in this case the March). Arguably, these circulating positive emotions and the structures of feeling they create can have tangible outcomes. The Women’s March was the first in a series of protests that unfolded during Donald Trump’s presidency. Below we discuss some affective outcomes from this particular instance of hashtag activism that contributed to the years of political activism that followed.

Calls to action

Within the #WhyIMarch data set, a number of users explicitly named an emotion, such as hope or pride, and these named emotions were overwhelmingly positive. These positively named emotions resulted in a great deal of motivational tweets, or what we name calls to action. These calls to action take the form of invitations to join the March (e.g., “Women’s March NYC today. Get out ladies”) or invitations to join feminist causes (e.g., “Who ready for a revolution? Join us and let Trump know we’re not backing down,” “I’m committed to using my many privileges to fight back this year. I hope those of you who can will join me”). These motivational posts, some heavily punctuated with exclamation marks, served multiple roles: some were expressing and encouraging enthusiasm (e.g., “Gonna be a great day for a protest!”), some were celebratory and encouraging (“We’ve got work to do. Good thing we’re so bad ass”), and others were encouraging about the future specifically (e.g., “Building #BridgesNotWalls at the #Womensmarch in DC! Because another world is possible!”). The numerous generational references — particularly photographs of children, as discussed earlier — sometimes overlap with calls to action, adding layers of rhetorical strategies that leverage and circulate affect even more.


Example of motivational tweet with allusion
Figure 3: Example of motivational tweet with allusion.


Some calls to action even more explicitly invoked girl power and resistance. Readers were encouraged to “roll up [their] sleeves, to be courageous to be radical.” One of the most popular (and recurrent) battle cries was an allusion to Beyoncé, the pop singer who appeared at the 2014 MTV Music Video Awards in front of glowing lights that spelled out the word feminist. This lyric from her politically charged 2016 album Lemonade appeared in multiple posts under the #WomensMarch hashtag: “okay ladies, now let’s get in formation.” Taken at face value, these tweets called for growth and development of the women’s movement or directed women to (metaphorically) get together and fall in line, much like fighter pilots would “get into formation” during flights. However, additional meaning emerges when readers know the context of this song lyric. In this Beyoncé song, she tackles a range of issues including police brutality. Even those not familiar with her work may have seen her controversial 2016 Super Bowl performance of it wherein “she appeared flanked by black women in berets, a direct homage to the Black Panthers” (Howard, 2016). This lyric is not only a call to action; it is also an allusion to the social injustices that women — and particularly women of color — often face. Utilizing allusions magnifies the affective possibilities of a tweet within its short character limit. Ultimately, we find that some of the most rhetorically effective tweets are those that are layered with affective meaning.

Vicarious affect

Another effect of the emotion relayed through the #WhyIMarch tweets is the conveyance of vicarious affect, or the ways that users feel or experience the event through others. Tweets from marchers and non-attendees alike referenced the positive feelings and energy that the March and accompanying tweets conveyed. Participants referenced “feeling the energy” of the crowd in person while those following the March via Twitter made similar claims (e.g., “There is so much strength, power, solidarity and hope in this picture. It must be so much more energetic on the ground!”). Comments such as these showcase a visceral reaction, felt affect, sparked by engagement with the March tweets. In addition, many users explained that they were participating in the March virtually with phrases like “I’m there in spirit” and “Oh I wish I could have been there. I am with you all in love, support and solidarity.” Oftentimes users explained their absence by referencing physical issues (pregnancy, illness) or other obligations (work, school) and expressed their solidarity from afar. These vicarious affect posts — and more nuanced tweets that contributed to the vicarious affect — again contributed to the structures of feeling created by the March. That is, despite the negative political circumstances that drove participants to in-person and online protests, these comments reveal an overarching positive sentiment and, arguably, a shift from despair and anger to hope and inspiration.


Examples of vicarious affect tweetsExamples of vicarious affect tweetsExamples of vicarious affect tweets
Figure 4: Examples of vicarious affect tweets.


One main feature of the vicarious tweets is that they highlighted the increased accessibility of digital participation. Users who could not travel to or participate in a March protested virtually. The first example in Figure 4 shows a young woman breastfeeding an infant with a speech bubble saying “Marching w/ You Sisters.” The speech bubble conveys support and elicits solidarity and sisterhood. Furthermore, the image itself is a feminist act of breastfeeding in public. Finally, the presence of the baby invokes the generational affective rhetorical strategy. This tweet is layered with rich affective meaning and demonstrates the vicarious experience an activist can gain from engaging with live tweets from a protest, as well as the solidarity that marchers can receive from tweets posted by digital participants.

The ability to be impacted emotionally by an event miles away was regularly highlighted in tweets by non-marchers, and their tweets often mirrored the emotions of those participating on the ground. However, this is not a unidirectional affective exchange, as marchers can be encouraged by off-site support. Gong (2014), for example, reports that social justice activists in the field felt emotionally supported by reading messages from digital participants. Overall, then, we find that there is a productive relationship between the vicarious tweets and the marcher tweets.

Collective affect

The next affective theme that emerged in the #WhyIMarch tweets was collective affect, or tweets that projected the author’s feelings onto a crowd or community, erasing a sense of separation between self and other. While these posts occasionally referenced negative shared affective states, such as the “soul crushing” election of Trump, more often these were expressions of shared positive affect. One example of these collective affect tweets were proclamations of shared love (e.g., “We’re here and we are full of love,” “Love brings us together today and love is what will continue to move us forward tomorrow”). Some of these tweets convey the mood of the crowd (e.g., “Heavy rain, high spirits and much laughter!”), while others assign an emotion to a group. For example, the tweet “We raise feminists. Proud, loud, strong, angry, and indignant” associates feminism generally with these particular emotions. As described earlier in the first person plural section, presumptuous statements like this could be problematic and overreaching. There is a fine line between interpreting emotions and assigning them, but these variations ultimately have the same affective outcome in portraying a group — whether March participants or feminists overall — as a collective affective community.

Our determination about the way that users created a collective community speaks to previous research on Twitter demonstrating how users create an imagined community (Chen, 2011; Gruzd, et al. 2011; Xu and Yan, 2011; Harrington, et al., 2012). But not all researchers agree on the idea of online community. Dean (2015) argues, “Affective networks produce feelings of community, or what we might call ‘community without community.’ They enable mediated relationships that take a variety of changing, uncertain, and interconnected forms” [27]. Such affective attachments, she notes, “are not in themselves sufficient to produce actual communities” [28]. Whether the feelings that Twitter users are expressing represent real or imagined community, affect plays a vital role. Paasonen (2015) explains that “affective intensities both drive online exchanges and attach people to particular platforms, threads, and groups” [29]. She goes as far as to suggest that “social media users are largely driven by a search for intensity — a desire for some kind of affective jolt, for something to capture one’s attention” [30]. The numerous affective rhetorical strategies within the #WhyIMarch tweets (e.g., the felt emotions reported by virtual participants discussed in the vicarious affect section) seem to confirm this desire for the “affective jolt” available through digital activism.

Another way that users in our dataset invoke an affective imagined community was by referencing Hillary Clinton’s campaign slogan, “Stronger Together.” The saying itself invokes solidarity and community, and using it in March tweets adds layers of shared ideologies, politics, and continued support of Clinton (and opposition to Trump). An additional strategy for constructing imagined community was references to sisterhood. Figure 5, for example, reads, “We’re here with 1000s of our friends.” This tweet, too, has multiple layers of meaning in that it uses first-person plural to invoke collective affect while hyperbolically addressing the 1,000 fellow marchers as friends. This constructed community conveyed a powerful sense of solidarity and support that we argue was one of the best outcomes of the March and its virtual protest. These collective affect tweets thus generate structures of feeling that are considerably positive, motivational, and supportive.


Example of imagined community tweet
Figure 5: Example of imagined community tweet.


Despite all of the affective community work, there was minimal interaction between Twitter users in our sample. While much online communication tends to be interactive, these posts were largely parallel, rather than responsive. This lack of dialogue may result from the multiplicity of marches around the world. However, community building can take place in person and by reading, retweeting, and mirroring posts, rather than replying directly to a post. This may suggest that the “collectives” formed by online affective practices such as these “may never quite come together in stable ways” [31]. Or, as Lauren Berlant explains, the affective infrastructures that bind people together are symbolic as well as practical, straddling “the conceptual and material organization of life” (McCabe, 2011). Gleaning more information about the interactions between digital protestors would be a fruitful project for further research.

Archiving affect

The final — and most compelling — effect of the #WhyIMarch tweets pertains to the ways that these tweets created an archive of affect of the event. A majority of users in our sample digitally recorded the March by capturing their attendance at protests; they shared photos or video footage of crowds or themselves or simply stated where they were. Additional users employed the hashtag to share information, such as by posting a link of the March Web site. These tweets circulated not only this information, but also the affective themes like hope and solidarity to create an immense archive of affect. Here we employ Ames’ (2020) phrase “archiving affect” [32], which builds upon Ann Cvetkovich’s (2003) term “archive of feelings,” to describe how cultural texts act “as repositories of feelings and emotions, which are encoded not only in the content of the texts themselves, but in the practices that surround their production and reception” [33]. Pybus (2015) similarly built on Cvetkovich’s work when she explained that archives of feeling capture “discrete moments when we are affected by intimacy, sexuality, love, and activism” [34]. We argue the Women’s March tweets captured the “discrete moment” of the protest through the significant archival posts and the affect that those posts conveyed.

The archiving of affect is perhaps the most powerful outcome of the digital Women’s March protest. Most hashtag movements are confined to a particular time period — sometimes resurging with a new tragedy or revealing news — and a digital protest accompanying a March has an even shorter life span. But the lasting impact of the #WhyIMarch tweets is visible in the thousands of digitized attendance records and the collective affect that persisted after the March. The participation and rhetorical strategies are archived, but the structures of feeling built that day are not only captured through digital archives; they also live on beyond the confines of 140 characters in the collective memories and feelings of the millions of people who participated that day (even vicariously). The participants formed an imagined community and built solidarity that can survive the ephemeral Twittersphere — and even sitting presidents.



Implications for hashtag activism

The findings from this study of #WhyIMarch tweets have several implications for research into hashtag activism, particularly for those interested in studying digital activism through the lens of affect. To begin, this study highlights women’s voices and feminist issues and shows how hashtag feminism can serve as a form of herstory — a contestation and challenge to traditional, patriarchal historical documentation and narratives. One of the functions of structures of feeling “is to narrate our relation to a social order so as to make our daily experience of that order meaningful and manageable” [35]. Discussing archives of feelings, Pybus (2015) similarly argues that they are “an important space of interpretation and contestation that has the power to make meaning through its ability to privilege certain discourses over others. Who and what gets remembered and who gets to make these existential decisions, are issues with important social, political, and economic ramifications” [36]. The collective #WhyIMarch tweets provided a space for women to claim agency and rearticulate problematic norms. There were also references to silence and voice that attest to digital activism being an important space of witnessing where social inequalities can be acknowledged and challenged. Future scholarship on feminist hashtag activism might attend to the ways that activists can leverage the archival function of social media to create more visible and enduring herstories.

The feminist narratives within the Women’s March tweets are not without fault, however. Despite the Women’s March proclaimed goals of recognizing, supporting, and furthering intersectionality and diversity, there was not significant evidence of diversity in the March tweets. The March was criticized from its inception for its initial lack of inclusivity and dominance even thereafter of white cisgender voices (Mosthof, 2017). Beyond the representation at the Marches and in accompanying digital dialogue, the rhetoric within some tweets reveals problems with inclusivity. For example, tweets within our sample document occasional slips into essentialism and presumptuousness, such as the hashtags #caringmajority that reinforces a gendered stereotype associating women with caring and nurturing. Other tweets used first person plural pronouns that sometimes universalized experiences and values among all marchers, despite the variety of genders, beliefs, and goals of the marchers (i.e., “we are marching so that we can decide when we parent and control our birthing options. #whyimarch #womensmarch #march4r #birthjustic” (@colorlatina)). Future scholarship should further attend to intersectionality and the affective consequences of amplifying and silencing particular voices within online activism.

Our study also demonstrates how structures (or infra-structures) of feeling are mediated through screens. Doane (1990) studied the ways in which catastrophes and crises have historically drawn people to television screens to witness these moments of impact that disrupt “ordinary routine” [37]. Studying television and cultural anxiety, Ames (2020) details “the emotions television has captured in regard to 21st century tragedies: live footage of people at ground zero on September 11th, family members mourning outside the Sandyhook Elementary School, displaced persons returning to their destroyed homes in the wake of Hurricane Katrina, the last moments of victims of police brutality (captured on smartphones and then broadcasted on television screens)” [38]. Such mediated moments not only disrupt our routine, drawing us to the screen to consume them in endless (even painful) repetition, but they also end up acting as time markers for those who witness them, epochal reference points, and arguably play a large role in a generation’s structure of feeling. Doane (1990) claims that “catastrophic time stands still” [39]. Hence the event, the frozen moment in time, becomes one potentially shared with an entire generation. More recent digital activism suggests a similar process is at play because many movements are sparked by tragic events, such as #ICantBreathe and #COVID-19. While not all hashtag activism is sparked by catastrophic events, they are often tied to cultural trauma in some form and can viewed as a disruptive moment, such as #MeToo. The hashtag feminism from the Women’s March was the result of such a moment and reveals how users leveraged their screens to mediate the complex emotional terrain following the 2016 election. Future studies can take up these themes and investigate how users mediate trauma through hashtag activism and other forms of digital protest. Scholars might also pay attention to not just what is digitized and remembered about these watershed moments, but also to the processes and consequences of this storytelling mode, or how this compulsive pull to our screens during such times, and our repetitive consumption of digital affect and messaging, creates and reinforces particular structures of feeling. Given these possibilities and its affective potential, hashtag feminism is sure to continue capturing a variety of powerful cultural moments in the future. End of article


About the authors

Kristi McDuffie is the Director of Rhetoric at the University of Illinois at Urbana-Champaign, where she administers one of the largest general education programs at the University. In this role as writing program administrator, she directs curriculum, supervises, and trains over 70 writing teachers a semester, oversees professional development, leads staffing and scheduling, liaises with campus partners to support students, and more. One of her biggest accomplishments is editing, producing, and composing a large portion of an in-house textbook, I write: A writing guide for the Undergraduate Rhetoric Program at the University of Illinois, now in its third edition. She primarily teaches composition theory and pedagogy and first-year writing. Her research interests center on digital writing and rhetorical theory, and her recent projects and publications investigate hashtag activism and digital rhetorics. One example is her forthcoming edited collection, Hashtag activism interrogated and embodied: Case studies on social justice movements (University Press of Colorado, 2022). Finally, she is a section editor of PraxisWiki at Kairos: A journal of rhetoric, technology, and pedagogy, where she has worked with authors to develop, edit, and publish dozens of pieces of digital scholarship over the past six years.
E-mail: kristimcduffie [at] gmail [dot] com

Melissa Ames is a Professor at Eastern Illinois University specializing in media studies, television scholarship, popular culture, Internet studies, feminist theory, and pedagogy. Her most recent and forthcoming publications include her books, Women and language: Gendered communication across media (McFarland, 2011), Time in television narrative: Exploring temporality in 21st century programming (University of Mississippi Press, 2012), How pop culture shapes the stages of a woman’s life: From toddlers-in-tiaras to cougars-on-the-prowl (Palgrave, 2016), Television & cultural anxieties: Emotion’s role in 21st century programming & viewing practices (University Press of Kentucky, 2020), and Hashtag activism interrogated and embodied: Case studies on social justice movements (University Press of Colorado, 2022); chapters in Grace under pressure: Grey’s Anatomy uncovered (2008), Writing the digital generation (2010), Bitten by Twilight: Youth culture, media, and the Twilight saga (2010), Manufacturing phobias (2016), Adventures in Shondaland: Identity politics and the power of representation (2018); Young adult literature in the composition classroom (2018), Computers & writing proceedings (2018), and articles in the Journal of Dracula Studies (2011), Women and popular culture encyclopedia (2012), High School Journal (2013), Journal of Popular Culture (2014), Feminist Media Studies (2017), and Pedagogy (2017).
E-mail: mames [at] eiu [dot] edu



1. Researcher Nick Ruest (2017) puts the tweet volume at 14 million.

2. Papacharissi, 2014, pp. 3, 7.

3. Crowley, 2006, p. 80.

4. Williams, 1977, p. 128.

5. Flatley, 2008, pp. 21–22.

6. Flatley, 2008, p. 26.

7. Ibid.

8. Best, 2012, p. 194.

9. Coleman, 2018, p. 601.

10. Coleman, 2018, p. 609.

11. Dixon, 2014, p. 36.

12. Ibid.

13. Clark, 2016, p. 11.

14. Drüeke and Zobl, 2016, p. 50.

15. Ahmed, 2004, p. 191.

16. Gong, 2014, p. 88.

17. Gong, 2014, p. 100.

18. Ahmed, 2004, p. 4.

19. Corbin and Strauss, 2008, p. 21.

20. To honor ethical practices in Internet research studies, we only collected tweets that were publicly posted. When images are used in this article, we blurred user names to protect the identity of authors. We also blurred faces of minors. Finally, we reached out to as many users as possible to notify them of their use in our project.

21. Sowards and Renegar, 2006, p. 67.

22. Gong, 2014, p. 100.

23. Sowards and Renegar, 2006, p. 63.

24. Holton and Lewis, 2011, p. 6.

25. Ahmed, 2004, p. 11; Paasonen, 2015, p. 31.

26. Pybus, 2015, p. 240.

27. Dean, 2015, p. 91.

28. Ibid.

29. Paasonen, 2015, p. 28.

30. Paasonen, 2015, p. 30.

31. Coleman, 2018, p. 610.

32. Ames, 2020, pp. 244–245.

33. Cvetkovich, 2003, p. 7.

34. Pybus, 2015, p. 239.

35. Flatley, 2008, p. 26.

36. Pybus, 2015, p. 239.

37. Doane, 1990, p. 228.

38. Ames, 2020, p. 244.

39. Doane, 1990, p. 231.



Sara Ahmed, 2004. The cultural politics of emotion. New York: Routledge.

Melissa Ames, 2020. Small screen, big feels: Television and cultural anxiety in the 21st century. Lexington: University Press of Kentucky.

Mike Baker, 2017. “The Woman’s March: Why we marched,” Crimson Hexagon (26 January), at, accessed 28 September 2017.

Beverly Best, 2012. “Raymond Williams and the structure of feeling of reality TV,” International Journal of Humanities and Social Science, volume 2, number 7, pp. 192–201, and at, accessed 3 January 2021.

Gina Masullo Chen, 2011. “Tweet this: A uses and gratifications perspective on how active Twitter use gratifies a need to connect with others,” Computers in Human Behavior, volume 27, number 2, pp. 755–762.
doi:, accessed 3 January 2021.

David Cohen, 2017. “Twitter: 12 million inauguration tweets Friday, 11.5 million women,” (23 January), at, accessed 28 September 2017.

Jason Cohen, 2020. “#BlackLivesMatter hashtag averages 3.7 million tweets per day during unrest,” PC Mag (20 July), at, accessed 19 November 2020.

Rebecca Coleman, 2018. “Theorizing the present: Digital media, pre-emergence and infra-structures of feeling,” Cultural Studies, volume 32, number 4, pp. 600–622.
doi:, accessed 3 January 2021.

Juliet Corbin and Anselm Strauss, 2008. Basics of qualitative research: Techniques and procedures for developing grounded theory. Third edition. Los Angeles, Calif.: Sage.

Rosemary Clark, 2016. ”’Hope in a hashtag‘: The discursive activism of #WhyIStayed,” Feminist Media Studies, volume 16, number 5, pp. 788–804.
doi:, accessed 3 January 2021.

Sharon Crowley, 2006. Toward a civil discourse: Rhetoric and fundamentalism. Pittsburgh: University of Pittsburgh Press.

Ann Cvetkovich, 2003. An archive of feelings: Trauma, sexuality, and lesbian public cultures. Durham, N.C.: Duke University Press.

Jodi Dean, 2015. “Affect and drive,” In: Ken Hillis, Susanna Paasonen, and Michael Petit (editors), Networked affect. Cambridge, Mass.: MIT Press, pp. 89–102.

Kitsy Dixon, 2014. “Feminist online identity: Analyzing the presence of hashtag feminism,” Journal of Arts & Humanities, volume 3, number 7, pp. 34–40, and at, accessed 3 January 2021.

Mary Ann Doane, 1990. “Information, crisis, catastrophe,” In: Patricia Mellencamp (editor). Logics of television: Essays in cultural criticism. Bloomington: Indiana University Press, pp. 222–239.

Ricarda Drüeke and Elke Zobl, 2016. “Online feminist protest against sexism: The German-language hashtag #aufschrei,” Feminist Media Studies, volume 16, number 1, pp. 35–54.
doi:, accessed 3 January 2021.

Jonathan Flatley, 2008. Affective mapping: Melancholia and the politics of modernism. Cambridge, Mass.: Harvard University Press.

Rachel Gong, 2014. “Indignation, inspiration, and interaction on the Internet: Emotion work online in the anti-trafficking movement,” Journal of Technology in Human Services, volume 33, number 1, pp. 87–103.
doi:, accessed 3 January 2021.

Anatoliy Gruzd, Barry Wellman, and Yuri Takhteyev, 2011. “Imagining Twitter as an imagined community,” American Behavioral Scientist, volume 55, number 10, pp. 1,294–1,318.
doi:, accessed 3 January 2021.

Steven Harrington, Tim Highfield, and Axel Bruns, 2012. “More than a backchannel: Twitter and television,” In: José Manuel Noguera (editor). Audience interactivity and participation. Brussels: COST Action Transforming Audiences, pp. 13–17.

Avery Holton and Seth C. Lewis, 2011. “Journalists, social media, and the use of humor on Twitter,” Electronic Journal of Communication, volume 21, numbers 1–2, at, accessed 3 January 2021.

Adam Howard, 2016. “Beyoncé doesn’t shy away from controversy with ‘Lemonade’,” MSNBC (25 April), at, accessed 25 November 2018.

Elizabeth Losh, 2014. “Hashtag feminism and Twitter activism in India,” Social Epistemology Review and Reply Collective, volume 3, number 12, pp. 10–22, and at, accessed 3 January 2021.

Earl McCabe, 2011. “Depressive realism: An interview with Lauren Berlant,” Hypocrite Reader, number 5, at, accessed 3 January 2021.

Mariella Mosthof, 2017. “We need to talk about the Women’s March criticism,” Bustle (30 January), at, accessed 3 January 2021.

Susanna Paasonen, 2015. “A Midsummer’s bonfire: Affective intensities of online debate,” In: Ken Hillis, Susanna Paasonen, and Michael Petit (editors), Networked affect. Cambridge, Mass.: MIT Press, pp. 27–42.

Zizi Papacharissi, 2014. Affective publics: Sentiment, technology, and politics. New York: Oxford University Press.
doi:, accessed 3 January 2021.

Andrea Park, 2017. “#MeToo reaches 85 countries with 1.7M tweets,” (24 October), at, accessed 16 February 2018.

Daniel Politi, 2017. “Women’s March on Washington beats expectations: Half a million descend on Mall,” (21 January), at, accessed 28 September 2017.

Jennifer Pybus, 2015. “Accumulating affect: Social networks and their archives of feelings,” In: Ken Hillis, Susanna Paasonen, and Michael Petit (editors), Networked affect. Cambridge, Mass.: MIT Press, pp. 235–250.

Nick Ruest, 2017. “Exploring #WomensMarch,” (10 February), at, accessed 28 September 2017.

Stacey K. Sowards and Valerie R. Renegar, 2006. “Reconceptualizing rhetorical activism in contemporary feminist contexts,” Howard Journal of Communications, volume 17, number 1, pp. 57–74.
doi:, accessed 3 January 2021.

Donald Trump, 2016. “Transcript: Donald Trump’s taped comments about women,” New York Times (8 October), at, accessed 3 January 2021.

Raymond Williams, 1977. Marxism and literature. Oxford: Oxford University Press.

Women’s March, 2017. “Our mission,” at, accessed 7 December 2018.

Huimin Xu and Ruoh-Nan Yan, 2011. “Feeling connected via television viewing: Exploring the scale and its correlates,” Communication Studies, volume 62, number 2, pp. 186–206.
doi:, accessed 4 January 2021.


Editorial history

Received 25 November 2019; revised 8 December 2020; revised 28 December 2020; accepted 1 January 2021.

Creative Commons License
“Archiving affect and activism: Hashtag feminism and structures of feeling in Women’s March tweets” by Kristi McDuffie and Melissa Ames is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Archiving affect and activism: Hashtag feminism and structures of feeling in Women’s March tweets
by Kristi McDuffie and Melissa Ames.
First Monday, Volume 26, Number 2 - 1 February 2021