Individuals Killed in Mass Protests and Demonstrations, Chile 1983-1989 and Contextualizing Data

After years of economic turmoil and political instability, socialist Salvador Allende was elected president of Chile in 1970. Allende began land reform and nationalized $1 billion of US investment within Chile (Craig and Logevall, America’s Cold War, 281). On September 11, 1973, the Chilean military removed President Salvador Allende from power, replacing him with a junta. That same day, Allende took his own life as those in his administration were arrested, killed, or exiled. Congress was immediately closed and military rule was installed. General Augusto Pinochet would lead Chile through an era of repression until 1990. During the almost two decades of his power, tens of thousands of Chileans were killed and disappeared. Though mass violence and political suppression plagued the country, Chileans opposed Pinochet’s policies through protests and demonstrations during his presidency.

Following Pinochet’s tenure, the Rettig Commission was tasked with documenting human rights abuses during his presidency. The commission, formally known as the National Commission for Truth and Reconciliation, found 3,428 disappearances, murders, kidnappings, and cases of torture from 1973 to 1990, which is nowhere near the up to 40,000 unofficial cases as historian Joshua Nadel asserts. This official account details the stories of almost every victim.

Between the years of 1983 and 1989, the commission found that 141 individuals were killed during mass protests and demonstrations. These individuals were counted based on their relation to major political demonstrations. Although the fatalities have relations to political incidents, many of the individuals were not involved in the protests, but were still victims. The commission included cases of individuals that they classified as human rights violations or victims of political violence.

Figure one shows the fatalities that the commission included in their definition of human rights violations and victims of political violence (aside from two individuals who were not provided a location). Each dot represents a life lost during these demonstrations and each reference the age, profession, and narrative provided about each individual by the Commission. According to figure two, the majority of fatalities among demonstrators were in the Santiago region. But, more interestingly, it shows that there were violent protests across the very extremes of the country. Chileans mobilized and died in the name of resistance from the northern border to Chile’s southern island.

According to Thomas Miller Klubock, Chile entered an economic crisis in 1982 when the regime’s economic model failed. (Klubock, 122). The liberalization of trade after Allende’s fall combined with a worldwide recession caused the crash in Chile’s economy (Klubock, 122). By 1983, 30% of Chileans were unemployed and 10% of the labor force were employed in minimal government programs (Klubock, 123). Klubock associates the economic turmoil with the rise of protests in the 1980s. Figure one could include a layer to show the economic trends during the era. It could also highlight which profession was most common at these violent protests. Average age could be examined as shown in figure three. Gender discrepancies could also be analyzed– but what is the value of these possible layers on top of figure one’s model?

Catherine D’Ignazio and Lauren Klein question data and its communication in Data Feminism. The concept of data feminism challenges individual to think of data and its communication as a result of “direct experience, by a commitment to action, and by the ideas associated with intersectional feminist thought” (D’Ignazio and Klein, chapter 1). The two show that while visualizations based on data may appear objective, they are the product of a design process. “Data feminism can also show us how the categories of data collection matter deeply, especially when dividing people into groups” (D’Ignazio and Klein, chapter 1).

D’Ignazio and Klein remind us that “feminism is about power… And in our contemporary world, data is power” (D’Ignazio and Klein, chapter 1). In the case of the Rettig Report, power must also be questioned. Who determined the definitions of those lost to human rights abuses and political violence? What individuals are yet to be reported? Whose voices are lost because of the power dynamic?

The Open Data Movement, as described in Data Feminism, is exampled by the Rettig Report. Reports and datasets made readily available to the public democratizes information, and also poses concerns. If anyone can access this information, such as those who perished in protests, what narratives can they produce with this information? Economist Daniel Kauffman presents the issue of “Zombie data”: “data that has been published without any purpose or clear use case in mind” (D’Ignazio and Kein, chapter 5). Does all data need to be accessible to the public, including personal narratives?

Continuing their exploration of big data, D’Ignazio and Klein count the infinite datasets that one can attain online. “There are test data sets – for network analysis, machine learning, social media, and image recognition. There are fun data sets, curious data sets, proliferating newsletters of numbers and data sets and so on. This would all appear to be a “good thing” in the sense that we, in our contemporary moment, tend to think of unfettered access to information as a kind of inherent good. And, in general, it is kind of amazing that one can just go download pigeon racing statistics, social networks of jazz musicians, the lengths of guinea pig teeth and truckloads of tweets for various hashtags… there’s a larger problem at work here that has to do with context” (D’Ignazio and Klein, chapter 5).

Context is key. Without context, where would that leave these individuals? Without context, what would be interpreted about these individuals?

What are the social, cultural, historical and material conditions in which knowledge is produced? What are the identities of the humans making the knowledge? Rather than seeing knowledge artifacts – like datasets – as neutral and objective fodder to use for more knowledge making, a feminist perspective advocates for connecting them back to their context, to better understand their limitations and ethical obligations. And, ultimately, to better understand the ways in which power and privilege may be obscuring the truth (D’Ignazio and Klein, chapter 5).

All knowledge is “situated,” according to Donna Haraway (D’Ignazio and Klein, chapter 5). These violent protests must be situated in a Cold War context. The capitalist world had been fighting communism for over two decades by the time that Allende was deposed. By this time, the fight against communism had become a fight against left-leaning governments and movements. Action that challenged capitalist norms within a nation were squashed because of their possible communist-leanings or their susceptibility to communist influence. This had become a theme in Latin America, as well. In 1951, the United States supported a coup against President Arbenz in Guatemala, who had a history of social and agrarian reform. Cuba had become communist in 1959, and in 1960, Nikita Khrushchev declared that the Monroe Doctrine was dead, further escalating the threat of communism in the Western Hemisphere (LaFeber, The American Age, 568). This not only affected US foreign policy, but domestic policy in Latin America, as well.

Johanna Drucker also urges audiences to be critical of data in “Humanities Approaches to Graphical Display.” Drucker suggests the usage of a “humanities approach” to visualization to understand that data is not quite objective. Data is in fact “capta” according to Drucker. While data is “given,” recorded and observed, capta is “taken” actively (Drucker, 2). As made clear by D’Ignazio and Klein, Drucker asserts that knowledge is “constructed, taken, not simply given as a natural representation of pre-existing fact” (Drucker, 2). The constructedness of data, making it capta, is based on interpretations of the world. For example, in a data set, if there is a breakdown of gender into male and female, the data contains an interpretation, making it capta. Assuming a binary distinction of gender renders it capta (Drucker, 3). The data presented by the Rettig Report contains many interpretations. Defining what constitutes a “protest,” “demonstration,” “human rights abuse,” “victim of political violence” are interpretations laid on top of the data. Drucker concludes that “phenomena of human experience are constituted as interpreted acts” and thus must be regarded as capta (Drucker, 17). Of course each individual life taken by the Pinochet regime can not be reduced to units, and thus cannot be construed as simple “data.”

If we did not contextualize the data regarding protest deaths, how could it be construed? Would these deaths simply appear to be a result of a random conflict? Placing these deaths within the framework of Pinochet’s regime and Cold War politics, this capta displays the narrative of political violence characteristic of the era. Without recognizing the interpretations placed on this data, these fatalities would lose their significance. Power shapes understanding, and by contextualizing this data power is shifted to the hands of those who were slain.

There is no correct way to display data, especially data with that represents human life. But humanity can be lost if it is reduced to simple metrics. By contextualizing data, by both recognizing it as capta and by utilizing a feminist lens, visualizations become better representations of reality. D’Ignazio and Klein argue for investment of providing and maintaining context around data so that numbers do not have to speak for themselves (though this could also cause limitations). While resources for this kind of investment may not be readily available, it is important to remember the human experience behind the data.

Posted by USIP Library on: October 4 2002 Source: Report of the Chilean National Commission on Truth and Reconciliation (Notre Dame, Indiana: University of Notre Dame Press, 1993), vol. I/II, Foreward, xxi-xxii. Note: Digitized and posted by permission of the University of Notre Dame Press, February 22, 2000.

Drucker, Johanna. “Humanities Approaches to Graphical Display.” Digital Humanities Quarterly 5, No. 1 (2011).

D’Ignazio C, Klein L. Chapter Five: The Numbers Don’t Speak for Themselves. MIT Press Open [Internet]. 2018 Nov 28; Available from:

Further Reading:
Klein, Lauren F. “The Image of Absence: Archival Silence, Data Visualization, and James Hemings.” American Literature 85, No. 4 (1 December 2013):661–688.

Padilla, T. (2017). On a Collections as Data Imperative. UC Santa Barbara.

Padilla, Thomas G. “Collections as data: Implications for enclosure.” College & Research Libraries News79, No. 6 (2018).