Digilantism,discrimination,and punitive attitudes: A digital vigilantism model

Abstract

In this article, I propose and apply a digital vigilantism model to a specific incident that occurred in Mexico, where the death of two innocent people was filmed through Facebook Live. Using a mixed methods approach and content analysis, I analyzed digilante Facebook posts (N = 942) coding gender, digital vigilantism categories, discriminatory comments, and punitive attitudes aimed at the perpetrators and the inciter of the lynching. The categories include investigating, blaming, or rebuking, while the discriminatory comments include classism, racism, homophobia, and body-shaming. I coded the punitive attitudes distinguishing four categories: non-physical punishment (calling for God’s wrath and the guilty conscience of the targets), legal sanction, death, and other punishment. The findings reveal the key role gender played in digilantism: females tend to conduct more investigations and low level attacks (blaming) than males, but males tend to perpetrate more harsh attacks (rebuking) than females. The most popular punitive attitude is calling for the death of targets, revealing tensions between legal sanctions and digilantes’ desired punishment. This study suggests the presence of different expressions of discrimination and reasons to engage in digilantism, encompassing both legal and illegal behavior deployed in a mainstream social media platform such as Facebook.

Keywords

Digilantism digital vigilantism discrimination online harm punitive attitudes

Introduction: Accused, filmed, and burned due to a social media rumor

In August 2018, a rumor spread through WhatsApp and Facebook fueling the belief that kidnappers were terrorizing the community of Acatlán de Osorio, in the state of Puebla, Mexico. On August 27, F,¹ an amateur journalist, warned through Facebook of the presence of child kidnappers in the town, encouraging the villagers to take preventive measures and report suspicious behaviors on social media.

On August 29, R and A parked near a school in their black SUV to drink some beer. Unfortunately, this vehicle fitted the description of the rumor, provoking the suspicion of nearby villagers. A group of villagers accused R and A of being child abductors, attacking their car and beating them. The police intervened and took the suspects to a police station for “disturbing the peace” (McDonell and Sánchez, 2018). Alerted by Facebook and WhatsApp messages, a mob gathered outside the police station clamoring for justice. F filmed this situation on Facebook live, inciting social media users to join the mob and take justice in their own hands. Finally, the mob assaulted the police station, took the suspects from custody and burned them alive on the street while F continued to film how the suspects died in the fire.

The rumor spread about the two suspects was false. R (21 years old) was a law student and his uncle A (56 years old) a farmer; both were buying construction materials to build a fence for their relatives (Martínez, 2018). After the villagers realized that the accusation was a false rumor, a massive digital vigilante response occurred through Facebook against F and the mob lynching. Paradoxically, the video filmed by F was the evidence that digilantes used to attack him and investigate the identity of the perpetrators who were part of the lynching. This case is part of an increasing tendency of lynching in Mexico, where between 2015 and 2016 the percentage of lynching cases increased by 37%, and between 2017 and 2018 the cases increased by 190% (Comisión Nacional de los Derechos Humanos and Instituto de Investigaciones Sociales, 2019: 10).

This incident attracted attention from the local and international press (Ávila, 2018; McDonell and Sánchez, 2018; Martínez, 2018). However, this case has not been the subject of academic research. For criminologists and social scientists, this case is relevant for three reasons:

1) It exhibits a wide variety of digital vigilante practices 2) It reveals discriminatory comments and digilantes’ punitive attitudes aimed at F and the mob 3) It evidences a connection among digital vigilantism, social media, and lynching.

First, this case is relevant because it features a wide variety of digilante practices. Among the digital responses, there are posts published directly on F’s Facebook timeline criticizing the ignorance of the mob, memes mocking F’s sexual orientation, and death threats made against F and his family.

Second, this incident is important for measuring digilantes’ punitive attitudes, particularly those aimed at F and the mob lynching. Given the varying intensity of the digilantes’ posts, this case study allows exploring a wide range of attitudes, ranging from those calling for the guilty conscience of the perpetrators to those making death threats. Additionally, this case exhibits a wide range of discriminatory comments targeting F and/or the mob lynching, including from strong racist comments to body-shaming attacks.

Third, this case is important because it crystallizes the connection between digital and physical punishment mediated by social media. This phenomenon is part of a tendency where mob lynchings motivated by rumors spread through social media kill individuals. In India since 2015 there have been more than 100 hundred lynching cases where WhatsApp was used to disseminate information about the targets (Banaji and Bhat, 2019: 3). This connection between digital and physical punishment challenges governments and private companies to regulate the spread of false information” (Banaji and Bhat, 2019: 3).

Using content analysis, the objective of this article is to explore digital vigilantism categories and the punitive attitudes espoused by digilantes toward F and the mob behavior. To achieve that goal, I will first delineate basic concepts and categories of digital vigilantism, showing how this case study contributes to gaps in that body of literature. Secondly, I will describe the characteristics of the Facebook post sample and the methods used to conduct the analysis. Third, I will explore digital vigilante categories and the punitive attitudes held by digilantes toward F and the mob behavior, discussing the role of racism, classism, body-shaming, and homophobia in digital vigilantism. I conclude proposing a refined model of digital vigilantism.

Digital vigilantism: Concept, social context, risks, gender, and typology

Digital vigilantism: Concept and social context

Digilantism is a vague term used to analyze a wide variety of “do-it-yourself” aims to achieve justice in the online world (Jane, 2017: 3). The concepts netilantism (internet vigilantism), online vigilantism, and digilantism have been used interchangeably in academic research (Chang and Poon, 2017: 1915). Also, scholars have employed the terms cyber vigilantism (Chia, 2020), web sleuthing (Yardley et al., 2018), online shaming (Skoric et al., 2010), and human flesh engine (Cheong and Gong, 2010) in their research articles².

In this article I use the terms “digital vigilantism” and “digilantism” interchangeably. I chose this terminology because those terms conserve the idea of “vigilante” and “digital” as opposed to broader concepts such as online shaming.

Various digilantism approaches including those from cybersecurity, criminological, and social media studies, propose vigilantism frameworks to define digital vigilantism (e Silva, 2018; Loveluck, 2020; Smallridge et al., 2016; Tanner and Campana, 2020; Trottier, 2017). These studies tend to compare vigilante and digital vigilante activities, finding similarities and differences between them.

Social media studies define digilantism as citizens coordinating digital retaliation through technological devices and social media sites (Gabdulhakov, 2018: 315; Trottier, 2017: 56). Jane (2017: 3) links digilantism to advocacy and activism, defining digilante activities as “politically motivated (or putative politically motivated) practices outside of the state that are designed to punish or bring others into account, in response to a perceived or actual dearth of institutional responses.” Social media scholarship places particular attention on online citizen empowerment (Jane, 2017: 5; Skoric et al., 2010: 182; Trottier, 2017: 56).

From a criminological perspective, Smallridge et al. (2016: 67) consider digital vigilantism as a variant of vigilantism, using Johnston’s (1996) six elements of vigilantism as the basis to develop their definition. According to Johnston (1996: 232), vigilantism is:

“a social movement giving rise to premeditated acts of force—or threatened force—by autonomous citizens. It arises as a reaction to the transgression of institutionalized norms by individuals or groups—or to their potential or imputed transgression. Such acts are focused upon crime control and/or social control and aim to offer assurances (or ‘guarantees’) of security both to participants and to other members of a given established order”.

The most significant modification of Johnston’s framework is the extension of the requirement of violence or threat of violence to “the causation of harm or threat thereof” (Smallridge et al., 2016: 66).

The limits between digilantism and other online harmful expressions are unclear. Digital vigilantism and cyberbullying are close concepts (Chang and Poon, 2017: 1929) since both are an expression of online persecution (Trottier, 2017: 62). The motivation of the perpetrators is a useful criterion to distinguish between digital vigilantism and cyberbullying (Smallridge et al., 2016: 66). Loveluck’s (2020) definition of digital vigilantism established that this motivation is related to justice, order, or safety.

In this article, I partially adopted Loveluck’s (2020) definition because it accurately reflects the aims and mechanisms that digilantes deploy:

“Direct online actions of targeted surveillance, dissuasion or punishment which tend to rely on public denunciation or an excess of unsolicited attention, and are carried out in the name of justice, order or safety” (Loveluck, 2020: 216).

Loveluck’s conceptualization highlights the critical elements of digital vigilantism: (1) Its denunciatory nature (Trottier, 2020: 197); (2) The idea of citizen policing (Chang and Poon, 2017: 1916) also called civilian policing (Huey et al., 2013: 83); and (3) The punishment element (Byrne 2013: 74; Chang and Poon, 2017: 1917; Gabdulhakov, 2018: 317; Jane, 2017: 3). In other words, digital vigilantism has both informative and punitive purposes (Trottier, 2017: 68).

However, Loveluck (2020: 214) states that a broad range of acts that can trigger vigilantism, including “perceived civil or moral transgressions, crimes, or injustices.” The term “injustice” is a broad and subjective notion. I argue that Loveluck’s definition of digilantism should be narrowed as per Johnston’s (1996) definition of vigilantism from a criminological perspective, focusing on the contravention of institutionalized norms upon crime and/or social control. Johnston (1996: 229) reduced vigilantism to “a reaction to real or perceived deviance,” distancing himself from mere political concepts that define vigilantism as “establishment violence.” Therefore, I define digilantism as a reaction to the infraction—or imputed infraction—of institutionalized norms via online mechanisms focused upon crime control and/or social control using surveillance, deterrence, and punishment to secure control, justice, or safety.

The conceptual comparison between digilantism and vigilantism also includes their explanations and risks. Regarding the explanations, both digital vigilantism and vigilantism are understood as responses to the inefficiencies of the police (Trottier, 2017: 63) and lack of security provided by the state (Byrne, 2013: 74). These reasons reveal the need to focus on the specific social context to conduct research on digilantism and vigilantism.

The relevance of the social and crime context is critical when analyzing online behavior and digilantism. In other words, the diffusion of messages on social media platforms does not occur in a social vacuum (Tanner and Campana, 2020: 6). In the United States, the majority of digilantes do not reside in “remarkably crime-ridden neighborhoods” (Lageson, 2020: 98). Conversely, Mexico has experienced unparalleled levels of violent crimes in 2019 (Justice in Mexico, 2020: 45–46). In particular, the state of Puebla is one of the states with the highest number of kidnappings reported in Mexico (Justice in Mexico, 2020: 16). These high trends of violence and crime are fundamental to understanding the violence and brutality deployed by the digilantes in this case study.

Digilantism: Risks and policing

Both digilantism and vigilantism share common risks. Among these risks, vigilantes and digilantes may eventually punish an erroneous target because of a wrong identification, inflict a disproportionate harm to the target, or refuse to give a chance to the target to explain a possible exculpatory reason (Jane, 2017: 578). Both vigilantism (Johnston, 1996: 233) and digital vigilantism can include legal and illegal activities (Jane, 2017: 3). The case studied in this article reflects these risks, including the killing of two innocent people and a massive digilante response against F and the mob lynching.

Furthermore, digilantism also involves particular challenges to the criminal justice system. Digital vigilante activities lack control (Chia, 2020: 657; Trottier, 2017: 60) due to the absence of norms ruling these actions (Chang and Poon, 2017: 1916). Digilantism is perceived as “lawlessness” since digilantes do not need legal authority to deed (Dunsby and Howes, 2019: 45).³ Digilantism shifts the legal presumption of innocence for the presumption of guilt (Gabdulhakov, 2018: 317). This lack of control and regulation on the online world is reflected in both the spread of the rumor that lead to the killing and the digilante response against F and the mob lynching.

Given that citizens do not have formal training in police procedures (Chang and Poon, 2017: 1916), digilante investigations could jeopardize legal procedures, such as exclusionary rules related to illegally obtained evidence⁴. Digilantes do not have policing strategies to prioritize investigations (Hadjimatheou, 2019: 13), giving them the freedom to target specific behaviors like vigilantes (Bateson, 2021: 924).

The notion of citizen policing is a double-edged sword when justifying and performing digilante practices. As Hadjimatheou (2019: 4) pointed out, “citizen-led policing initiatives have the potential both to undermine and to enhance democratic norms in policing” depending in the way in which digilantism is practiced. Organizations such as Reddit present themselves mimicking FBI logos, while Anonymous rejects any kind of association with police departments (Myles et al., 2020: 323)

Also, the idea of citizen policing is problematic since digilantes who engage in attempting “social justice” could also be simultaneously motivated by personal reasons such as pleasure (Chang and Poon, 2017: 1928–1929). The analysis of this case study will reveal the mixed nature of digilantism, including posts intended to assist police investigations but also comments mocking and attacking the targets.

Digilantism and gender

The majority of quantitative studies do not use gender as a variable when analyzing digilantism outcomes. The absence of gender may be due to the difficulties in gathering sociodemographic data on digilantes, since some online platforms allow people to participate using anonymous usernames (Yardley et al., 2018: 95).

Despite the lack of quantitative studies analyzing gender, qualitative findings reveal the relevance of gender for both digilantes and targets. Feminist studies have addressed digilantism as a reaction to cyber violence against women and girls (Jane, 2017: 3), creating new spaces for informal justice mechanisms, and narratives for victims-survivor of sexual violence (Powell, 2015: 583).

Gender also plays a key role when considering the levels of access to the technology required to perpetrate digilante acts. In countries such as India, access to smartphones is deeply gendered, where women become the principal target of digital violence, especially if they are part of a minority (Banaji and Bhat, 2019: 4). Based on these prior studies, I will test if there is a difference between males and females when engaging in digilantism activities using the model I will explain in the methods sections.

Digilantism typology

Loveluck (2020) provides the most detailed study of digilantism, analyzing about 50 different digital vigilantism cases from different parts of the globe and classifying digilante practices in four categories: flagging, investigating, hounding, and organized leaking. Flagging is a low intensity category, involving shaming a behavior without providing all the necessary information to identify the target. Flagging is usually committed via sharing images of the shamed behavior, and typical cases are bad parking, manspreading, and vandalism acts in neighborhoods (Loveluck, 2020: 217–219), while investigating includes naming and providing information to target the individual who committed the shamed behavior. The behavior shamed could include a broad variety of activities, from minor incivilities to serious crimes (Loveluck, 2020: 223). A step beyond investigating practices is hounding, which “not only [. . .] combine[s] an investigative dimension with a punitive intention, but it also involves a more sustained mobilisation against a specific target, triggered by intense outrage” (Loveluck, 2020: 227). Lastly, organized leaking is usually focused on systematic problems, targeting principally institutions and organizations, and involving well-structured organization to collect evidence and share incriminating information (Loveluck, 2020: 234).

Loveluck proposed a progressive harm scale from flagging, investigating, to hounding, combining punitive and investigative dimensions. In this regard, hounding aims to harm the target publishing incriminatory evidence against him/her (Loveluck, 2020: 227). However, I argue that this typology is mostly designed for cases where targets are not clearly identified. In this case study, the digital vigilante practices are directly deployed on F’s timeline, including investigative activities to obtain more information about him and to identify the members of the mob. In this article, I propose a conceptual distinction between attacks and investigations, since not all hounding attacks necessarily include crowdsourcing activities. I will briefly define my typology proposed in the methods section and examine those categories in the analysis section.

Methods section

Facebook is the most used social media platform in Mexico (Statista, 2019). I gathered data from Facebook because F filmed and shared the main incident via Facebook Live. The online transmission of this crime through Facebook explains why the digital responses took place on that platform.

I accessed F’s profile through my private Facebook account.⁵ F has a public profile, so every user with a Facebook account can access F’s timeline, read his posts,⁶ and even reply to his posts.⁷ Digilantes accessed F’s timeline and deployed their attacks replying to F’s posts. I did not code F’s posts because they are mostly unrelated to the case—they deal with social events, or publicize his work or private life—and he never replied to any of the digilantes’ posts. Therefore, I limited my sample to digilante posts that replied to F’s posts on his timeline.

To collect my data, I read F’s timeline from its latest post on August 29, 2018 back to three months before the lynching day. This search allowed me to identify the latest and the earliest digilante response on F’s timeline. I found the earliest digilante post on August 25, 2018, and the latest on August 29, 2018, and thus established the timeline of my research. Over this period, F published 26 posts with 1198 responses by online users. I copied all these posts to an excel sheet, coding the information from post responses in chronological order.⁸ This order allowed me to follow-up the dialog among the digilantes themselves.

From this original dataset, I excluded 231 post responses from the analysis following three criteria: (a) Posts that do not express any kind of support, balance, attack, or investigation regarding F or mob behavior and (b) Posts composed exclusively of links that are not currently available, and (c) Posts that are not possible to understand given the grammatical and syntax errors. Through these exclusions, I separated and excluded the posts unrelated to my research goals and information that I could not access. Examples of these exclusions are personal discussions among users where they insult each other, arguments discussing the celebration of religious events after the killing, and dead links.⁹

From this dataset comprising 967 post responses, I then excluded two further kinds of post responses: (1) Support for F or mob behavior (N = 4) and (2) Neutral comments about the incident (N = 21). This allowed me to narrow the data to my research interest: digital vigilante posts. The final study dataset comprises 942 post responses, including also 49 links and 139 images attached to the posts. I realized during the creation of the database that some users might change the names of their accounts over time. Therefore, I did not calculate the number of posts per person, but I did code for gender for each post when possible, based on the usernames.¹⁰

Through a mixed methods research, I integrated qualitative and quantitative data to acquire a complete comprehension of digital vigilantism (Creswell, 2014). Using an open and axial coding scheme with NVivo software I created digilante categories, discriminatory categories, and punitive attitudes using the literature available and emerging data contained in the posts. I used an inductive approach to create the categories, placing special attention on: (1) Activities employed by digilantes, (2) Contents of the post, (3) Intentions, aims, and desires expressed by digilantes.

As I will explain in the Analysis Section, I placed special attention on the particular context to interpret the meaning of the posts (Krippendorff, 2019: 29). Next, I used R statistical software to create excel sheets classifying the data and describing the connections among gender, digital vigilante categories, discriminatory comments, and punitive attitudes.

As previously mentioned, Loveluck’s (2020) typology includes four categories: flagging, investigating, hounding, and organized leaking. I conserved only the investigative category and I created two new categories: blaming and rebuking. Blaming is a low intensity attack which usually describes F and the mob behavior motivated by digilantes’ disapproval, and rebuking constitutes an aggressive attack expressing anger. The creation of these codes allowed me to clearly distinguish between attacks and investigations, a distinction that is not clear in Loveluck’s category of hounding, that thus is not part of my typology. Organized leaking is a category that involves a completely different level of organizational structure that is irrelevant to this case study, so it was excluded as a category.

I coded the digilante activities under the three mutually exclusive categories mentioned: investigating, blaming, or rebuking. In the coding process, I took into consideration the phrases contained in the posts but also the use of capital letters, images, and links attached to every post. I also coded the total sample of posts identifying any social discriminatory comments regarding F and/or the mob lynching based on four categories: classism, racism, homophobia, and body-shaming. Each post might include any or several discriminatory comments since they are not mutually exclusive.

Figure 1.

Digital vigilantism model proposed.

I coded as classist those comments highlighting the ignorance of F and/or the mob lynching for acting based on rumors, focusing on their impoverished status, or underlining the misspelling errors of F’s Facebook posts. Regarding racism, I coded comments focused on the indigenous ethnicity—or imputed innate ethnicity—of the targets made with a pejorative intention. I coded as homophobia comments mocking the imputed homosexuality of F, and body-shaming encompasses any negative comments related to the physical appearance of the targets.

Finally, I coded the attacks—blaming and rebuking—of the digilantes who aimed, desired, or predicted a punishment for F and/or the mob lynching. I coded the punitive attitudes distinguishing four categories: non-physical punishment (calling for God’s wrath and the guilty conscience of the targets), legal sanction, death, and other punishment.¹¹ This coding scheme is not mutually exclusive, so each post can include any or several punitive attitudes. Using this typology proposed, I hypothesize that there is a difference between females and males when engaging in digitalism activities.

Data analysis

In this section, I will present a descriptive analysis of the sample composed of 942 digilante posts. I will describe the characteristics of my proposed typology and the role played by gender, discriminatory comments, and punitive attitudes in this case. Then, I will define and discuss my digilante categories and punitive attitudes using representative examples from the sample.

Sample characteristics

The results in Table 1 show that posts made by males comprised 50.63% of the sample, posts made by females constituted 42.25%, and for the remaining 7.11% of posts, gender is unknown. Among the digilantism categories, investigating comprised 24.94% of the posts, blaming 14.43%, and rebuking 60.61%.

Table 1.

Sample characteristics (N = 942).

Sample characteristics	% (n)
Gender
Male	50.63 (477)
Female	42.25 (398)
Unknown	7.11 (67)
Total	100 (942)
Digilantism categories
Investigating	24.94 (235)
Blaming	14.43 (136)
Rebuking	60.61 (571)
Total	100 (942)
Discriminatory comments	100 (145, contained in 129 posts)
Classism	53.10 (77)
Racism	28.27 (41)
Homophobia	15.86 (23)
Body-shaming	2.73 (4)
Total	100 (145)
Punitive attitudes	100 (286, contained in 268 posts)
Non-physical punishment	31.81 (91)
God’s wrath	16.08 (46)
Guilty conscience	15.73 (45)
Legal sanction	27.27 (78)
Death	35.66 (102)
Other punishment	5.24 (15)
Total	100 (286)

Table 1 shows the presence of 145 discriminatory comments contained in 129 posts from the total sample. Among the discriminatory comments, classism comprised more than one-half (53.10%), followed by racism (28.27%), homophobia (15.86%), and body-shaming (2.73%).

As shown in Table 1, 268 posts contained 286 punitive attitudes aiming a punishment at F and/or the mob lynching. Notably, death comprised 35.66%, followed by non-physical punishment (31.81%), legal sanction (27.27%), and other punishment (5.24%).

Gender differences emerge from analyzing the digilante categories, discriminatory comments, and punitive attitudes. As shown in Table 2, female posts outnumber males’ in two categories: investigating (50.21%) and blaming (48.52%). However, a larger proportion of male posts (54.46% compared with 37.47%) were present in the harshest digilante category: rebuking. A chi-square test of independence was performed to examine the relationship among gender and the digilantism categories, showing that females are significantly more likely than males to participate in investigating and blaming.

Table 2.

Digilante categories and gender.

Digilantism categories	Male % (n)	Female % (n)	Unknown % (n)
Investigating	45.95 (108)	50.21 (118)	3.82 (9)
Blaming	42.64 (58)	48.52 (66)	8.82 (12)
Rebuking	54.46 (311)	37.47 (214)	8.05 (46)
Total	50.63 (477)	42.25 (398)	6.47 (67)

Chi square results χ² (4, N = 942) = 17.0923, p = 0.001855.

Gender differences are also relevant for all types of discriminatory comments. Table 3 shows that males posted 54.54% of the total of classist comments, 65.85% of racist comments, 60.86% of homophobic comments, and 75% of the comments related to body-shaming.

Table 3.

Gender and discriminatory comments.

Discriminatory comments	Male % (n)	Female % (n)	Unknown % (n)
Classist	54.54 (42)	41.55 (32)	3.89 (3)
Racism	65.85 (27)	29.26 (12)	4.87 (2)
Homophobia	60.86 (14)	26.08 (6)	13.04 (3)
Body-shaming	75 (3)	25 (1)	0 (0)
Total	59.31 (86)	35.17(51)	5.51 (8)

Gender plays a crucial role in punitive attitudes. Table 4 details that females comprised 60.86% of the posts calling for God’s wrath (compared with males’ 30.43%) and 48.88% of the posts of guilty conscience (compared with males’ 35.55%). Conversely, the proportion of females and males change when we move to harsher punishments. Males comprised 62.82% of the punitive attitudes aiming a legal sanction (compared with females’ 29.33%), and 59.80% of the comments aiming the death of the targets (compared with females’ 33.3%).

Table 4.

Gender and punitive attitudes.

Punitive attitudes	Men % (n)	Female % (n)	Unknown % (n)
Non-physical punishment	32.96 (30)	54.94 (50)	12.08 (11)
God’s wrath	30.43 (14)	60.86 (28)	8.69 (4)
Guilty conscience	35.55 (16)	48.88 (22)	15.55 (7)
Legal sanction	62.82 (49)	29.33 (22)	8.97 (7)
Death	59.80 (61)	33.3 (34)	6.86 (7)
Other punishment	53.33 (8)	46.66 (7)	0 (0)
Total	51.74 (148)	39.51(113)	(25)

Table 5 shows the relation between discriminatory comments and digital vigilantism categories, showing clearly that discriminatory comments are concentrated in the rebuking category. Rebuking comprised 93.10% of the classist, racist, homophobic, and body-shaming comments posted against the targets. These kinds of discriminatory comments are an expression of the harmful dimension of rebuking compared to the lower level of attack represented by blaming.

Table 5.

Discriminatory comments and categories.

Discriminatory comments	Investigating	Blaming	Rebuking (n)
Classism	3.89 (3)	6.49 (5)	89.61 (69)
Racism	2.43 (1)	0 (0)	97.56 (40)
Homophobia	0 (0)	0 (0)	100 (23)
Body-shaming	25 (1)	0 (0)	75 (3)
Total	3.44 (5)	3.44 (5)	93.10 (135)

As shown in Table 6, rebuking contains 263 punitive attitudes representing 91.95% of the total of punitive attitudes. These data highlight the aggressive dimension of rebuking, expressing the outrage aiming a broad range of punishment at the targets.

Table 6.

Relation between categories and punitive attitudes.

Punitive attitudes	Blaming % (n)	Rebuking % (n)
Non-physical punishment	12.08 (11)	87.91 (80)
God’s wrath	10.86 (5)	89.13 (41)
Guilty conscience	13.33 (6)	86.66 (39)
Legal sanction	11.53 (9)	88.46 (69)
Death	0.98 (1)	99.01 (101)
Other punishment	13.33 (2)	86.66 (13)
Total	8.04(23)	91.95 (263)

Overall, this sample features the combination of different digital vigilante practices within a single case, including blaming, investigating, and rebuking. Gender has a significant statistical impact on the distribution of these digilantism practices.

Typology: Investigating, blaming, and rebuking

In this qualitative section, I analyze the digital vigilantism categories I proposed using representative examples from my sample, by including images and quoting comments posted by the digilantes. In the punitive attitudes section, I will examine specific examples of punitive attitudes aimed at F and/or the mob lynching, revealing different levels of intensity. My goal in these sections is to provide a brief overview about the content of the digilante posts, showing the complexities and risk involved in digital vigilante categories and punitive attitudes.

Investigating

Investigating includes activities aiming to identify and provide more information about the targets. Investigating is not restricted to sharing information but also includes discussing the relevance of the evidence, asking for more details, and giving feedback to the investigative findings of the rest of the digilantes.

Investigating ranges from cursory activities to more sophisticated and thorough research, including the elaboration of informational posters to collect evidence. Among the digilante cursory activities, one of the digilantes posted this picture without providing any information, or clue about how these people are related to the case:

We can guess that the two people portrayed in the picture participated in the lynching incident. However, we do not know if both participated, what kind of participation they supposedly had, or even their names. Digilantes reacted, harshly insulting the two people that appear in the picture after its publication on F’s timeline, showing how superficial finger-pointing in the online world can trigger outrage. Conversely, investigating can also be conducted thoroughly, including the diffusion of informational posters:

WE DEMAND JUSTICE

Two HUMBLE PEASANTS died at the hands of “judges” who convicted them to be burned to death, they only were drunk and they were ACCUSED OF BEING KIDNAPPERS without having any evidence

They could be your relatives

This informational poster in Image 2 includes more information than Image 1, promoting the diffusion of the information and suspects using emotional arguments. The inclusion of the burnt bodies in the poster is an additional effort to capture the attention of the readers and visibly demonstrate the seriousness of the incident.

Image 1.

Image captured from Facebook, 2018.

Image 2.

Translation provided to the right of the image.

Blaming

Blaming is a low intensity attack against an identified or suspected target, triggered by the digilantes’ disapproval of the target behavior. In contrast to Loveluck’s (2020) flagging category focused on behavior, blaming is directed to a person already identified or suspected of violating an institutionalized norm, ranging from minor incivilities to severe crimes as occurred in this case study. Blaming describes the target behavior or mocks his/her situation, including in some cases a punitive attitude aimed at the target.

For example, this digilante post combines the description of the target behavior and a punitive attitude aimed at F:

Se dice que tú fuiste el que hizo el “En vivo” [Facebook Live] pero la vida es justa y es un boomerang, espero que pagues lo justo por tu incitación mediante este medio. Y espero que tu conciencia día y noche te recuerde lo que hiciste.
[Translation]: It is said that you were the one who made the “In live” [Facebook Live] but life is fair and it is a boomerang, and I hope you pay fairly for your incitement through this means. And I hope your conscience reminds you day and night of what you did.

The post describes F behavior, adding a specific punitive attitude: F should feel guilty day and night. Blaming can also be expressed through ironic statements, including brief comments and elaborated memes as well. Contrary to a direct attack, irony uses a sarcastic tone to express an indirect thought that can also harm the target.

This digilante post accurately reflects the idea of irony, including prison as a punitive attitude aimed at F:

Próximamente, [apodo de F] en vivo desde la prisión. No se lo pierdan.
Coming soon, [F’s nickname] live from prison. Do not miss it.

This post implicitly aims a prison sanction at F, using an ironic tone recalling F’s behavior filming the killing of the innocents via Facebook Live. As I pointed out, blaming the target can also take the form of a more elaborated expression through memes, mocking, or reflecting on the situation:

THE NEW INQUISITION!

WITCHES!

KIDNAPPERS!

This meme in Image 3 reveals that blaming can include more profound reflections contextualizing the case into a broader historical framework. The meme compares the lack of rationality of the mob lynching to the hunting methods of the Inquisition. However, when the digilante attack is motivated by intense outrage or the punitive attitude turns harsher, we move to the next level of digital vigilante attacks: rebuking.

Image 3.

Translation provided to the right of the image.

Image 4.

Translation provided to the right of the image.

Rebuking

Rebuking goes one step further than blaming, focusing on the causation of online harm to the target. Rebuking usually includes insulting the target, labeling his/her behavior as criminal, or showing a harsh punitive attitude. Digilantes express their outrage through opinions and judgments but also by using capital letters, exclamation marks, emoticons, images, and links. Rebuking can also include harsher punitive attitudes than those present in blaming, aiming physical retaliation against the target or his/her family and friends.

ya te cargo la verga cabrón a esta perra [novia de F] la van a violar en frente de ti y a ti te van a empalar por culero, asesino pocos huevos
you are fucked [.]this bitch [F’s girlfriend] is going to be raped in front of you and you are going be impaled for asshole, coward murderer

This post reveals that some digilantes are involved in the perpetration of criminal offenses, including death or rape threats¹². However, rebuking can also include more lenient punitive attitudes, such as legal sanctions and non-physical punishment.

In this post, the digilante attached a screenshot of a prior post of F where he spread the kidnapping rumor days before the killing:

Hasta adivino era el desgraciado! Ojalá y te pudras en el infierno.
This bastard was even a psychic! Hope you rot in hell.

IMPORTANT ANNOUNCEMENT IF YOU HAVE CHILDREN IN THE SCHOOL I HAVE BEEN ASKED TO INFORM YOU TO GET THEM, SEND FOR THEM, ASK FOR A FAVOR AND GET THEM, WE HAVE TO PREVENT [KIDNAPPERS] AND IF YOU SEE SOMETHING SUSPICIOUS UPLOAD IT TO SOCIAL MEDIA TO SUPPORT EACH OTHER, ESPECIALLY IN MORE VULNERABLE AREAS THIS ROBBERY, KIDNAPPING, AND EVEN FOR ORGAN SELLING COULD HAPPEN THANK YOU IN ADVANCE LET’S TAKE CARE OF OUR CHILDREN AND OUR SOCIETY WHERE WE LIVE THANKS IN ADVANCE [F‘S NICKNAME] IN VIVACHO TAKE CARE AND I SEND YOU KIND REGARDS

The post above expressed its outrage attaching the screenshot, using an ironic tone, and stating God’s wrath as the punishment called for F. The inclusion of F’s post provides more information about F’s prior behavior and details of the case. Also, through the attachment of the screenshot, we can see that this post was greatly shared (at least 107 times).

Punitive attitudes: Non-physical punishment, legal sanction, and death

Non-physical punishment

Non-physical punishment includes two subcategories: calling for God’s wrath and calling for the target’s guilty conscience. God’s wrath focuses on God and the idea that he will bring justice to the case, including thoughts related to God’s punishment, God’s unforgiveness, God as a judge, and divine justice. God’s wrath also includes ideas related to going to hell, which for some digilantes is a punitive attitude even harsher than prison:

La cárcel no, el infierno es mejor para ese gusano
Not prison, hell is better for that scum

Guilty conscience is the second subcategory of non-physical punishment, and it is focused on the target’s guilty conscience due to his/her behavior. This subcategory includes posts related to feelings of remorse and regret aimed at the target, also adding digilantes’ expressions regarding the mental wellbeing of the targets, such as that they will not ever have days of peace or nights sleeping easy:

Desgraciado asesino ojalá vallas preso ojalá nunca puedas dormir tranquilo tu y toda esa gente de mierda de tu pueblo q mataron a esas personas inocentes pudranse desgraciados crueles
Disgraceful murderer hopefully you go to prison hope you will never sleep peacefully you and all those shitty people from your town who killed those innocent people, may they rot miserable cruel

This post reveals that a single attack can include more than one punitive attitude, in this case, guilty conscience and a legal sanction: prison.

Legal sanction

Legal sanction includes attitudes related to criminal procedure: targets processed, convicted, or going to prison. I also included more general expressions aimed at the achievement of justice and hoping for the full force of the law in this case.

De todo corazón deseo que se haga justicia malditos perro mal nacido
With all my heart I want justice to be done you damn bastard dog

Death

Finally, death includes posts aiming at the lynching of the targets, predicting that they will be hunted, and wishing their suicide. In this category, the context plays a crucial role in interpreting the meaning of the post. I coded posts aiming based on the law of talion (eye for an eye), wishing that what happened to the victims happens to the perpetrator) as death:

ojo por ojo hijo de tu puta madre. ahora te toca a ti culero y será peor así que ponte verga porque te van a chingar
eye for eye son of a bitch. Now it is your turn and it will be worse so be ready because they are going to screw you

The different punitive attitudes described in this section reveal a broad range of varying degrees of punishment aimed at several levels of intensities, including motives based on religion, sanctions provided by the criminal justice, and using the law of talion as the appropriate punishment. The implications of these punitive attitudes will be discussed in the next section, including the quantitative data collected.

Discussion

The digital vigilantism model I propose and apply in this case study takes into consideration digital vigilantism a wide variety categories, discriminatory comments, and punitive attitudes aimed at the targets.

The variety and complexity of digilante categories suggest the presence of divergent motivations for digilantism. The qualitative findings evidence the existence of digilantes mocking and having fun with F’s future punishment. At the same time other digilantes are genuinely trying to solve the case, sharing incriminating evidence, and calling to the community to report the suspects. This finding is consistent with the results obtained through a survey finding that digilantes who engage in attempting “social justice” could also be simultaneously motivated by personal reasons such as pleasure (Chang and Poon, 2017: 1928–1929). The current study exhibits the simultaneous presence of pleasure and genuine attempts to solve the crime among different digilantes.

The digital vigilante categories proposed are also essential to analyzing this phenomenon’s legal dimensions since digital vigilantism can include legal and illegal practices (Jane, 2017). This study reveals that a single digilante case can consist of a wide variety of digital vigilante categories: blaming, investigating, and rebuking combined simultaneously. Within one case, digilantes committing crimes and digilantes expressing their opinions can interact with each other. These results are consistent with the policing perspective that digilantism can both support and weaken democratic rules in policing (Hadjimatheou, 2019: 4) within one single case.

The mixed nature of digilantism found within a single case challenges the police and prosecutors’ efforts to establish the boundary between legal and illegal practices in the online world, especially when investigating and determining the individual responsibilities in these types of events. The posts I found are in Spanish and Portuguese, showing that some digilantes likely participating in events from different locations around the world. The transnational feature of this digilante case study can lead to jurisdiction and enforcement challenges similar to what Yar (2006: 102) identified when analyzing other expressions of digital violence such as hate speech.

Although the number of discriminatory comments found is lower than the number of posts expressing punitive attitudes, these results differ from prior descriptive studies (Nhan et al., 2017) that have found a practical absence of racist comments since moderators and users were allowed to delete comments. This case study revealed that discrimination is present in digilantes’ behavior on a mainstream platform like Facebook when there is not a person moderating the posts. This finding is consistent with prior studies focused exclusively on digilante platforms (Byrne, 2013), advancing to a digilante model that differentiates among different types of discriminatory posts.

The literature on digilantism reveals that despite the development of detailed categories (Loveluck, 2020) or the quantification of posts (Nhan et al., 2017), there are no studies exploring the impact of demographic variables on digilantism categories. This study contributes to filling this gap, revealing that females tend to conduct more investigations and lower level attacks (blaming) than males. Overall, males tend to perpetrate more harsh attacks (rebuking) than females.

This study suggests that females and males tend to behave differently when approaching digilantism. This finding is consistent with the differences regarding punitive attitudes: Females comprised the majority of the posts calling for God’s wrath and the target’s guilty conscience. However, most of the harsher punitive attitudes aiming for a legal sanction and the death of the targets were espoused by males.

These findings suggest that gender plays a key role in digilantism. This conclusion is also reflected in prior qualitative studies that address women as either targets or perpetrators of digilantism (Banaji and Bhat, 2019; Jane, 2017; Powell, 2015). More research is needed to address what social and cultural factors lead women to conduct more investigations, lower attacks, and less punitive punishments aimed at the targets than males. This research perspective could potentially justify the future implementation of educational programs and policies specifically directed to educate male’s violent online behavior and digilante practices.

The analysis of punitive attitudes in this study shows how social context and the criminal justice system impacts the use of digital vigilantism in Mexico. Notably, the most popular punitive attitude is calling for the death of the targets. However, the Mexican Constitution, in article 22, explicitly forbids extra capital punishment. This contradiction between citizens’ desires and the catalog of sanctions provided by the Mexican criminal justice system reveals a tension between legal punishment and justice mobilized by the digilantes.

As I analyzed before, the majority of the literature has focused on the relationship between citizens and police when explaining digital vigilantism. The findings of this study indicate that death and non-physical punishment represent 61.47% of the total of the punishments aimed at targets, far beyond legal sanction which comprised of 35.66% of the posts. Since both death and non-physical punishment are sanctions that cannot be provided by the criminal justice system, these punitive attitudes reveal the need to study the legal catalog of sanctions and cultural contexts for engaging in digilantism. Future research should explore the idea that digilantism provides the opportunity to demand solutions that the criminal justice system cannot offer event in the most harmful crimes.

Moreover, the punitive findings show how digilantes navigate the “interactional space” in a virtual/networked place (Hayward, 2012: 456). The violence expressed by the digilantes should be understood within the broader violent social context of Puebla, where the number of kidnappings reported is among the highest in Mexico.

The social context in Mexico may promote this cycle of digilante-citizens where both can be targets of violent crimes and perpetrators of digital violence. The larger context of wound culture is crucial to understanding this case, where the ideas of spectacle and repetitive violence (Seltzer, 1998; 2017) are present. In addition to including harsh punishment aimed at the targets, the digilantes also use memes and ironic messages in their discourse. These strategies reinforce the idea that violence and humor are part of the same digilante phenomenon.

Finally, this study measured digilantes’ punitive attitudes, categorizing and quantifying online posts without the need to design and implement a survey to collect data. This type of measurement reveals the possibility of exploring new ways to measure punitive attitudes using online data available to researchers through digital vigilante cases. More research using this technique is needed to examine punitive attitudes in the online world and compare the findings with the results obtained through traditional research techniques such as surveys.

Conclusion

The digital vigilantism model proposed takes into consideration digital vigilantism categories, discriminatory comments, and punitive attitudes aimed at the targets. The findings suggest that gender impacts how males and females engage in digilantism and punitive attitudes. Females tend to conduct more investigations and low attacks (blaming) than males, but males tend to perpetrate more harsh attacks (rebuking) than females. Females represent the majority of the posts calling for God’s wrath and the target’s guilty conscience. However, most of the punitive attitudes aiming for a legal sanction and the death of the targets were espoused by males. More research addressing the social and cultural factors explaining these differences is crucial and could potentially lead to policies and educational programs focused exclusively on males’ online behavior.

The analysis of the digilantes’ posts revealed the mixed nature of digilantism: wherein some digilantes attempt to solve a crime; while, others are motivated by pleasure. Additionally, digilantes who commit crimes and digilantes who express legitimate opinions as citizens can interact with each other within a single case, challenging the police, and prosecutor’s efforts to investigate these cases.

This case study showed that discrimination is present in digilante practices on mainstream social media platforms such as Facebook when there is not a moderator deleting or censoring digilantes’ posts. These findings demand a more comprehensive understanding of digilantism, including the broader cultural and social contexts where it is deployed. In this case, the social context of Puebla—among the Mexican states with the highest number of kidnapping—suggest that the digilantes’ behavior should be interpreted within this context of violence.

This study found that the most popular punitive attitude measured is calling for the death of the targets, a harsh punishment explicitly forbidden by the Mexican constitution. This contradiction reveals the need to include in future research not only the policing context, but also the legal catalog of sanctions of the criminal justice system when analyzing digilantism.

Finally, this study created and developed a digital vigilante model without the need to design and implement a survey to collect data. More research is needed using online data to explore, categorize, and measure directly punitive attitudes and digital vigilantism practices instead of traditional research techniques such as surveys.

Footnotes

Acknowledgements

This study would not be possible without the outstanding mentorship of Dr. Rosemary Barberet. I am grateful to Sebastián Rodríguez, Sofia Larrazabal, Katy Pugliese, Dr Verónica Michel, and Dr Deborah Koetzle for their valuable advice and critique. I also thank to two anonymous reviewers for their thoughtful comments.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Graduate Research Scholarship, John Jay College of Criminal Justice (CUNY).

ORCID iD

Sebastián Galleguillos

Notes

Author biography

Sebastián Galleguillos is a doctoral student at John Jay College (CUNY), where he also completed an MA in International Crime & Justice. He is a research associate at the Centro de Estudios de Derecho Penal (Universidad de Talca, Chile). His research interests include social media & crime, comparative criminology, and alternatives to incarceration.

References

Arun

(2019) On WhatsApp, rumours, and lynchings. Economic & Political Weekly 54(6): 30–35.

Ávila

(2018) Linchados en Puebla eran campesinos; no eran robachicos: Fiscalía. El Universal, 29 August.

Banaji

Bhat

(2019) WhatsApp Vigilantes: An Exploration of Citizen Reception and Circulation of WhatsApp Misinformation Linked to Mob Violence in India. London, UK: Department of Media and Communications, The London School of Economics and Political Science, pp.1–60.

Bateson

(2021) The politics of vigilantism. Comparative Political Studies 54(6): 923–955.

Byrne

(2013) 419 digilantes and the frontier of radical justice online. Radical History Review 2013(117): 70–82.

Chang

Poon

(2017) Internet vigilantism: Attitudes and experiences of university students toward cyber crowdsourcing in Hong Kong. International Journal of Offender Therapy and Comparative Criminology 61(16): 1912–1932.

Cheong

Gong

(2010) Cyber vigilantism, transmedia collective intelligence, and civic participation. Chinese Journal of Communication 3(4): 471–487.

Chia

(2020) Seeking justice on the web: How news media and social norms drive the practice of cyber vigilantism. Social Science Computer Review 38(6): 655–672.

Comisión Nacional de los Derechos Humanos and Instituto de Investigaciones Sociales (2019) Informe especial sobre los linchamientos en el territorio nacional.

10.

Creswell

(2014) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, 4th edn. Thousand Oaks, CA: Sage Publications.

11.

Dunsby

Howes

(2019) The NEW adventures of the digital vigilante! Facebook users’ views on online naming and shaming. Australian & New Zealand Journal of Criminology 52(1): 41–59.

12.

e Silva

(2018) Vigilantism and cooperative criminal justice: Is there a place for cybersecurity vigilantes in cybercrime fighting? International Review of Law, Computers & Technology 32(1): 21–36.

13.

Facebook (2021) https://www.facebook.com/help/203805466323736 (accessed 20 May 2021).

14.

Gabdulhakov

(2018) Citizen-led justice in post-communist Russia: From comrades’ courts to dotcomrade vigilantism. Surveillance & Society 16(3): 314–331.

15.

Hadjimatheou

(2019) Citizen-led digital policing and democratic norms: The case of self styled paedophile hunters. Criminology & Criminal Justice. Epub ahead of print 16 October 2019. DOI: 10.1177/1748895819880956

16.

Hayward

(2012) Five spaces of cultural criminology. The British Journal of Criminology 52(3): 441–462.

17.

Huey

Nhan

Broll

(2013) ‘Uppity civilians’ and ‘cyber-vigilantes’: The role of the general public in policing cyber-crime. Criminology & Criminal Justice 13(1): 81–97.

18.

Jane

(2017) Feminist digilante responses to a slut-shaming on Facebook. Social Media+ Society, 3(2): 1–10.

19.

Johnston

(1996) What is vigilantism? The British Journal of Criminology 36(2): 220–236.

20.

Justice in Mexico (2020) Organized crime and violence in Mexico: 2020 special report. University of San Diego, Department of Political Science & International Relations, pp.1–68.

21.

Krippendorff

(2019) Content Analysis: An Introduction to Its Methodology, 4th edn. Thousand Oaks, CA: Sage Publications.

22.

Lageson

(2020) Digital Punishment: Privacy, Stigma, and the Harms of Data-Driven Criminal Justice. Oxford: Oxford University Press.

23.

Loveluck

(2020) The many shades of digital vigilantism. A typology of online self-justice. Global Crime 21(3–4), 213–241.

24.

McDonell

Sánchez

(2018) When fake news kills: Lynching in Mexico are linked to viral child- kidnap rumors. Los Angeles Times, 21 September. Available at: https://www.latimes.com/world/la-fg-mexico-vigilantes-20180921-story.html (accessed 19 May 2021).

25.

Martínez

(2018) Burned to death because of a rumour on WhatsApp. BBC, 12 November. Available at: https://www.bbc.com/news/world-latin-america-46145986

26.

Myles

Benoit-Barné

Millerand

(2020) ‘Not your personal army!’Investigating the organizing property of retributive vigilantism in a Reddit collective of websleuths. Information, Communication & Society 23(3): 317–336.

27.

Nhan

Huey

Broll

(2017) Digilantism: An analysis of crowdsourcing and the Boston marathon bombings. The British journal of Criminology 57(2): 341–361.

28.

Powell

(2015) Seeking rape justice: Formal and informal responses to sexual violence through technosocial counter-publics. Theoretical Criminology 19(4): 571–588.

29.

Seltzer

(1998) Serial Killers: Death and Life in America’s Wound Culture. Psychology Press.

30.

Seltzer

(2017) Wound culture. In: Oxford Research Encyclopedia of Criminology and Criminal Justice. Oxford University Press.

31.

Skoric

Chua

JPE

Liew

, et al. (2010) Online shaming in the Asian context: Community empowerment or civic vigilantism? Surveillance & Society 8(2): 181–199.

32.

Smallridge

Wagner

Crowl

(2016) Understanding cyber-vigilantism: A conceptual framework. Journal of Theoretical & Philosophical Criminology 8(1): 57–70.

33.

Statista (2019) https://www.statista.com/statistics/449869/mexico-social-network-penetration/ (accessed 6 January 2020).

34.

Tanner

Campana

(2020) “Watchful citizens” and digital vigilantism: a case study of the far right in Quebec. Global Crime 21(3–4): 262–282.

35.

Trottier

(2017) Digital vigilantism as weaponisation of visibility. Philosophy & Technology 30(1): 55–72.

36.

Trottier

(2020) Denunciation and doxing: Towards a conceptual model of digital vigilantism. Global Crime 21 (3–4): 196–212.

37.

Yar

(2006) Cybercrime and Society. Sage Publications.

38.

Yardley

Lynes

AGT

Wilson

, et al. (2018) What’s the deal with ‘websleuthing’? News media representations of amateur detectives in networked spaces. Crime, Media, Culture 14(1): 81–109.