Discussion.

According to Bonilla and Rosa, “Social movements have long used media and technology

to disseminate, escalate, and enlarge the scope of their struggles” (2015, p.7). In the current age of widely accessible social media, the range and impact of these mediums have also increased. In a survey conducted by the Pew Research Center, a majority of Americans felt social media platforms were important for activism (69%), and a similar number felt these platforms created sustained movements for social change (67%).

However, this shift towards online communities, and resulting online discourse, has raised concerns about how effective these decentralized movements can be. Terms like “slacktivism” can be used to describe how online activism, particularly through social media discourse, lacks grounding in real action ― and sustained change. 

On the other hand, social media platforms such as Twitter, provide a uniquely situated space that is multivocal and dialogic (in which there are many views and channels collected within one platform, and there are active dialogues taking place around these diverse views). This situation allows users to feel as if they are directly participating in whatever discussion they are a part of ― allowing “users who are territorially displaced to feel like they are united across both space and time.”

Recognizing that hashtags can only offer a “limited, partial, and filtered view”, our group believes that the Tweets and actions of everyday people engaging with online activism still holds value in understanding:

  1. How public perceptions around the Black Lives Matter Movement have or haven’t shifted over this period and

  2. Whether the Tweets (online activism) we analyze reflect “slacktivist”, performative, and/or unactionable behaviors.

Summer 2020 Protest Lifecycle.

 

The protest life-cycle is the period in which the protests occupy the forefront of mainstream consciousness. We evaluate this using mainstream media coverage, social media trends, and high volumes of online activism (search indexes/Google Trends, Twitter trends). With this in mind, our team is aware that protests are a common phenomenon, and that many of the demonstrations we discuss in this study are ongoing.

Different channels are not comparable. Each channel can only compare to itself (regard the peak as 100%) but seeing how each channel goes up/down together or diverges might provide insights. Interact with the visualization here.

Short-term, viral social media "activities" mobilized the most people and brought this movement to a historic peak. In the long run, most people dropped from the online conversation, but some people have continued through the end of June.

According to our findings, the mass, anti-racist demonstrations of 2020 broke out following George Floyd’s murder on May 25, 2020. From then to June 1, 2020, the movement grew. The visibility of the protests hit its peak between June 2, 2020 to June 7, 2020. Some key events that occurred during this period were: 

The momentum of the demonstrations began to decline (aging stage) following June 8, 2020, but since then has remained somewhat stable ― quieter and smaller, but continuous. Since June 8, 2020, the (mainstream) conversation has shifted from racial justice broadly to more pointedly about police funding (#defundthepolice and #abolishthepolice grew in popularity).

Overall Sentiment.

 

During this period, there were changing sentiments around a myriad of issues across the United States (and abroad). The Black Lives Matter movement, which was established in 2012, has been steadily gaining support over the years. However, in this one month period in 2020, the BLM movement, statement, and mission gained historic, global attention ― and support.

The percent change in volume of search trends in the United States from 2012 to 2020 using Google trends’ “Rising Trends” and “Interest over time” features. Interact with the visualization here.

Part of our research was about the sentiments around these trending topics (as identified by trending hashtags). The discussion of our project revolves around analyzing the conversations that are related to the social movement of Black Lives Matter and the longevity of those conversations, thus our team was interested in using Natural Language Processing to understand what were the popular topics (trends) people were interested in discussing and what emotions (sentiments) were present during those emotions. 

Using Natural Language Processing, the qualitative content within the collected tweets were analyzed and categorized as either having more “positive” or “negative” sentiments. "Positive" refers to optimistic, hopeful, and/or positive language being used to discuss BLM. "Negative" refers to pessimistic, aggressive, and/or negative (or crude) language used about BLM. Neutral posts tend to be more informational, shorter, and or lack highly connotative language. 

As a note, positive is NOT defined as necessarily supportive​ attitudes towards BLM, nor does negative mean unsupportive attitudes towards BLM. "Positive" and "negative" are narrowly defined as the emotional state in which people are generally discussing issues around BLM. Tweets were separated based on hashtags attached to the tweet.

One topic that gathered massive attention during this period was law enforcement.

Looking at sentiments about specific topics, anti-law enforcement hashtags (#defundthepolice, #acab, #policebrutality) generally had more negative sentiments, while pro-law enforcement hashtags (#bluelivesmatter) had more neutral and positive sentiments. Looking at individual sentiments, pro-law enforcement hashtags had a higher percentage of positive sentiments associated with them (7.18% compared to 4.94%). (It should be noted that pro-law enforcement hashtags were “taken over” by anti-racist supporters and activists (including Kpop fans), who flooded the hashtag with non-related content). Pro-law enforcement hashtags also had a higher percentage of neutral sentiments associated with them (39.2% compared to 26%). Overall, anti-law enforcement hashtags had a greater number of non-neutral sentiments (95.1% non-positive sentiments). 

Timeline: A look at the trends.

We tracked trending hashtags and search keywords from May 25, 2020 to June 15, 2020, the first month of protests following George Floyd’s murder by Minneapolis police officers. Hashtags carry important messages about Black Lives Matter. 

 

Categorizing the tweets more specifically, we separated hashtags as either actionable and unactionable. Actionable hashtags are defined as boosting (tweeting about) in-person direct action (ex. Joining in a protest) and online direct action (ex. Donating money, hosting a fundraiser, boosting a related activist organization) Example hashtags would include: #protest, #BuyBlack, #StreamForBLM, #petitionstosign. Unactionable hashtags: defined as not explicitly boosting in-person direct action and/or online direct action. Examples hashtags would include: #BlackLivesMatter, #dearwhitepeople, #NoJusticeNoPeace.

Note: While these hashtags are defined by our project as “unactionable”, this does not mean that these hashtags are not spaces for productive, transformative, and antiracist conversation. The definitions of “actionable” and “unactionable” align narrowly with how we have decided to organize and categorize the data collected for this project.

We extracted 20k tweets with hashtags #BLM (10k) and #BlackLivesMatter (10k) for each day from May 25, 2020 to June 25, 2020 and analyzed what hashtags were talked about around this movement. Interact with the visualization here.

Considering the data collected at the end of June, our team found that unactionable hashtags generally faded from their trending status after a short period of time. Many of the unactionable hashtags contained time-sensitive information, addressing the issues of the hour. “Activities” and celebrations (such as #blackouttuesday and #juneteenth) also had short term trending periods. However, these short-lived, unactionable hashtags were highly visible, pushing the momentum and energy of the anti-racist demonstrations (and the Black Lives Matter movement) further during this period in early June. 

An exceptional hashtag was #defundthepolice. It stayed on our visualization (of top Twitter trends) for about two weeks after it’s rise on June 7, 2020 (compared to the typical trending hashtag’s lifespan of several days). On June 23, 2020, the hashtag surpassed #georgefloyd.

Below is a timeline of social media trends that stood out from our data (Google Trends and Twitter trends). Brown represents trending topics.

  • 5/25: Murder of George Floyd

  • 5/25 to 5/30: George Floyd (#georgefloyd), ACAB [all cops are bastards] (#acab)

  • 5/30: Protests maintain its trending status, but drops out of Google’s top 3 trends

  • 6/2: Blackout Tuesday (#blackouttuesday). Surpasses George Floyd/#georgefloyd in trends

  • 6/4: Breonna Taylor (#breonnataylor), Police Brutality (#policebrutality)

  • 6/7: Defund the Police (#defundthepolice). Remains on the chart continuously through our tracking period (June 25, 2020)

  • 6/19: Juneteenth (#juneteeth)

  • 6/22-6/25: George Floyd (#georgefloyd) drops from the top 1 position during this period

  • Black Lives Matter protests (#blmprotest) trends for one day (6/22)

  • Defund the Police (#defundthepolice), while not in the top position, remains on the trending chart

Future Considerations.

From our sentiment analysis around Black Lives Matter between May 25, 2020 to June 25, 2020, our team has found notable results that support our hypothesis that George Floyd’s murder and the current influx of protests and activism around racial injustice has garnered greater mainstream attention compared to similar tragic events. 

Viewing Google Trends data, we were able to visualize the spike of attention given to the Black Lives Matter movement during this period. Through analyzing trending Twitter hashtags, we were able to document the sustained discussions around George Floyd, police brutality, and racial injustice during this period. And finally, through our analysis of Tweets using Natural Language Processing, we were able to categorize discussions around the Black Lives Movement as either positive, negative, or neutral ― and thus gather greater insight about how different trending topics (related to BLM) are viewed.

However, our greatest insight is that through tracking Twitter trends, we found #georgefloyd and #justiceforgeorgefloyd were the top hashtags for much of early June, but as the month progressed, more and more trends related to what actions could be taken (such as #defundthepolice and #blmprotest) rose to the top position. ​

Seeing that the discussions during this one-month period has not only garnered historically [publicized], mainstream attention about a single highly-visible police brutality case, but also discussion about what can be done to address racism in the United States,  this may be a sign that the current momentum of Black Lives Matter do not reflect temporary, performative activism (as is stereotyped of online activism), but is resilient (has potential for change and has already influenced change). 

As the voices around these protests and, more broadly, Black Lives Matter have not diminished to zero, are those who are standing the true advocates? How can we account for burnout among activists, and how can we cultivate more sustainable models of demonstrating? And most importantly, while the mass protests of early June have faded from the media and from mainstream consciousness, people are still working towards justice ― how much does media coverage matter, if it matters at all?