Alexandria French

Abstract

Once upon a time, all knowledge was written or spoken. Written knowledge was curated and stored away in massive universities and libraries. In the 21st century however, the digital archive plays a pivotal role in shaping collective memory and historical truth. The growing influence of algorithms and the corporations behind them raises important questions about how this power is used and who benefits from it. By drawing on Michel Foucault’s concept of the archive, I will explore how digital platforms curate and control information by preserving certain narratives and suppressing others. Once a tool for preserving history, the archive now often functions as a mechanism of surveillance where algorithms prioritize content that drive visibility and engagement as opposed to accuracy or inclusivity. By looking at the intersection of corporate interests, algorithmic decision-making, and the role of digital curation, I will argue that the digital archive reinforces existing inequalities by amplifying dominant perspectives while marginalizing dissenting voices. This analysis will focus on the ethical and philosophical consequences of these practices and question how power shapes what we collectively remember and understand.

​

Introduction

​

Understanding Foucault’s Archive

          Foucault’s concept of the archive gives us an incredibly compelling framework to understand how systems of power shape knowledge and discourse. The archive that Foucault (and ultimately this paper) is referring to is not the traditional notion of a physical location or repository of artifacts and records. Foucault’s archive is abstract in nature rather than a historical museum or library. It is a system of rules and relationships that govern what can be said, who gets to say it, and how these statements are preserved, transformed, and utilized. He argues that it is not just a collection of knowledge but rather a mechanism that defines the boundaries of discourse and determines what narratives are legitimized and which are marginalized and ultimately forgotten. In Foucault’s terms, the archive is not about what is stored. It is about the conditions that allow for certain forms of knowledge to come into being and be sustained. [1]

​

          Historically speaking, archives were tightly controlled institutions. Libraries, universities, and government repositories preserved records deemed important by those in power of those institutions. These historical archives were inherently biased and their contents reflected only the priorities and ideologies of their curators. The decisions about what to include or exclude were not neutral. These decisions were influenced by some of the most prevailing power structures such as the Catholic Church. By defining what was worth preserving these institutions shaped our historical narratives and societal memory. Unfortunately marginalizing voices and perspectives that did not align with the dominant ideologies in the process.

In the 21st century however, the notion or idea of the archive has shifted significantly. As the internet has grown into a force beyond imagine, so too has the archive. The now digital archive is governed by algorithms, extending Foucault’s insights into a domain he could have never imagined. These algorithms act as the contemporary rules of discourse and determine what information is visible, searchable or even hidden. Historical archives were built by identifiable curators, making bias an easier notion to track across the timeline. Algorithms however operate largely invisibly and closely guarded by those who own them. These digital systems built on corporate priorities and engagement metrics do more than preserve information. The algorithms actively shape how information is encountered, interpreted, and ultimately remembered. The transition to digital archives introduces an entire new layer of complexity. With the mechanisms of power and control becoming less transparent, their influence grows increasingly more pervasive.

Foucault’s concept of the archive is crucial in analyzing this shift from historical to digital archives as his emphasis on the archive as a site of power and knowledge shows us show digital archives do more than reflect reality. He argues that the archives actively construct it. By privileging certain narratives while suppressing others these systems shape our collective memory and the historical truth. Understanding the digital archive through Foucault’s lens will allow us to critically examine the forces that have determined what is preserved. Whose voice gets to be amplified? How is authority over knowledge maintained in the digital age? This perspective is essential for unpacking the ethical and philosophical implications of algorithmic curation the corporate control of information.

 

 

Algorithms as Curators

Because algorithms unarguably determine what content is visible and what remains obscured, it is no argument that in this digital age, they have become the primary curators of information. Functioning as modern day rules of discourse, the algorithms parallel Foucault’s concept of the archive. Just like Foucault argued that the archive is governed by a set of rules that define what can be said, the algorithms establish boundaries on the digital discourse by deciding what narratives are promoted to whom, what is suppressed entirely, or what is forgotten. This does not result in a neutral curation process; it is deeply embedded in systems of power that shape societal knowledge and our collective memory.

​

Algorithms as Rules of Discourse

            These algorithms are currently designed to process vast amounts of data and prioritize information based on certain criteria. These criteria, however, are not object. They are instead shaped by the underlying logic of the platforms that they serve. For example, social media algorithms prioritize content that generates engagement (likes, shares, and comments). This often amplifies sensational or polarizing material. [2] Similarly search engine algorithms rank content based on a combination of relevance, keywords, and advertising revenue which creates a hierarchy of visibility. Paralleling Foucault’s idea that the archive is not just a repository of knowledge, the digital archive created by algorithms governs what is knowable and by whom.

            By privileging certain types of content over others (content you are more likely to engage with, regardless of the morality of the subject) these algorithms are effectively dictating the scope of our digital discourse. They determine whose voices are amplified and whose are silenced. This mirrors Foucault’s observation that power operates through outright suppression as well as structuring the field of possible expressions. Algorithms define the boundaries of visibility and shape how knowledge is accessed and understood.

​

Corporate Influence on Algorithmic Priorities

            The design and operation of algorithms are deeply intertwined with corporate interests, pushing them well beyond simple technical matters. Most digital platforms are owned and operated by private corporations who’s primary goal is proft. Algorithms are optimized to serve these interest and prioritize content that maximizes user engagement and advertising revenue. This has created fundamental tension between the pursuit of profit and the ethical obligation held by private companies to provide accurate, inclusive, and diverse information.

            Social media platforms like Facebook and Twitter (X) have faced severe criticism for amplifying misinformation and hate speech because that content tends to drive high engagement. [3] Similarly Google’s search algorithms often prioritized paid advertisements and SEO-optimized content, regardless of its quality or credibility. [4] These practices demonstrate how corporate control over algorithms skew digital discourse and reinforces dominant narratives while marginalizing alternative perspectives. The profit-driven logic of these platforms exhausts existing inequalities as voices that lack the resources or influence to game the algorithm are left out of the conversation all together.

​

Case Studies

The biases inherent in algorithmic systems have profound implications for the fairness and inclusivity of digital discourse. One striking example comes from the field of facial recognition technology as highlighted in a 2018 article by Selena Silva and Martin Kenney in Phylon. They explore how these technologies often fail to perform equitably across diverse populations because of their limitations in training data.

​

         Silva and Kenney argue out that “facial recognition software often does not have a diverse set of training data”. They argue that this lack of training data “results in poor performance and low accuracy for the underrepresented sample, often compromising the minorities.” (Silva, Kenney 2018, p. 16) This observation is corroborated by a 2012 study by Klare, Burge, Klontz, Bruegge and Jain which identified systemic errors in recognizing individuals from minority groups due to inadequate representation in training datasets. [2]

These findings underscore a broader trend in algorithmic systems: the reproduction and reinforcement of existing inequalities. When datasets fail to reflect the diversity of the populations they serve, the resulting algorithms will disproportionately marginalized already vulnerable groups. This case illustrates how algorithms that are designed to be neutral tools can instead perpetuate systemic biases and shape digital and social realities in ways that exacerbate inequalities.

           Cases such as this show us how algorithms act as more than tools of efficiency. It shows us that they are powerful agents of discourse that determine whose experiences are validated and whose are overlooked. This example of facial recognition bias serves as a reminder that the rules governing algorithmic decision making are deeply entangled with societal structures of power and privilege.

​

Ethical and Philosophical Implications

​

            A digital archive that is shaped by algorithms and corporate interests, naturally raises significant ethical and philosophical concerns. As curators of collective memory, these systems can reinforce societal hierarchies, challenge the concept of truth and marginalize dissenting voices. When examining these issues through Foucault’s framework of power and knowledge we can better understand the implications of this curation in shaping what we remember, value, and deem credible.

​

Power and Inequality

            Digital archives have become powerful tools for reinforcing existing societal hierarchies. Because of their democratic nature, digital archives offer universal access to information. The reality is far more complex however. Algorithms that are driven by engagement metrics and corporate profit motives prioritize dominant narratives and mainstream sources which often leaves minority voices underrepresented.

​

            This systemic bias manifests in several ways. Content from well-funded organizations is often optimized for algorithmic visibility and smaller or marginalized groups lack the resources to compete. As a result, the digital archive amplifies the perspectives of the already powerful and reinforces societal inequities. This dynamic mirrors Foucault’s insight that power is embedded in systems of knowledge and shapes who has the authority to speak and be heard. By privileging dominant voices these digital archives entrench existing power structures and perpetuate inequalities in representation and thus access to information.

​

Erosion of Truth

            One of the most troubling consequences of algorithm driven archives is the erosion of truth. In a digital landscape that is dominated by engagement, content that elicits strong emotional responses (sensationalism, misinformation, polarizing rhetoric) is prioritized over content that is accurate or nuanced. This creates a tension between the philosophical concept of truth and the operational logic of digital platforms.

Foucault’s exploration of truth as a construct shaped by power is particularly relevant here. Truth is no longer solely determined by traditional authorities like academics or journalist. It is increasingly influenced by algorithmic visibility. As algorithms reward content that drives clicks and shares they are contributing to the proliferation of echo chambers and the spread of misinformation. Someone who spends more time on article about right wing conservatism is more likely to continue to be shown information that follows that line of information. Conversely, if someone was an addict and is constantly shown rehab options, it can be argued that an agenda is being pushed for them to get clean. This undermines the collective ability to discern credible information and creates a fragmented and polarized understanding of reality as well as creating questions of true freedom of choice.

            Philosophically, these implications are profound. If truth is contingent on visibility and engagement, what happens to knowledge that is complex, subtle, or even challenging to dominant ideologies? The erosion of truth in the digital archive raises many questions about the role of algorithms as arbiters of knowledge and the ethical responsibilities of the corporations that design them.

​

Marginalization of Voices

            Algorithms suppress dissenting and minority perspectives by prioritizing dominant narratives. Content moderation practices (often automated now) disproportionately target voices that deviate from mainstream norms. This is particularly evident in cases where activism or advocacy for marginalized communities is flagged as “controversial” or “sensitive” resulting in reduced visibility or even removal. [5]

            The suppression of minority voices has far reaching consequences for the collective memory. Digital archives play a critical role in preserving the narratives that shape social identity as a whole. When certain voices are excluded that results in an archive that presents an incomplete and skewed account of history. This reinforces what Foucault described as the power and knowledge relationship where those in power shape knowledge that in turn legitimizes their own authority.

            The marginalization of voices in the digital archive contributes to a homogenized cultural memory that fails to reflect the diversity of human experiences. By excluding alternative perspectives the digital archives risk perpetuating cycles of exclusion and erasure that will silence the very voices that could challenge and enrich societal discourse.

Conclusion

Governed by algorithms and corporate interest, the digital archive is an incredibly powerful mechanism that shapes our collective memory and historical truth. By drawing on Michel Foucault’s concept of the archive as a locus of power and knowledge, this paper has explored how digital curation reinforces societal hierarchies, erodes the concept of truth, and marginalizes dissenting voices. These systems act as arbiters of discourse by determining what can be expressed, preserved, and later remembered in the digital age.

            These algorithms operate invisibly and structure the boundaries of visibility as well as shape societal understanding. By privileging engagement driven content they prioritize profitability over accuracy, inclusivity and diversity. This dynamic reinforces existing inequalities and amplifies dominant perspectives while silencing marginalized voices. The ethical and philosophical implication of these practices are far reaching and raise urgent questions about the responsibility of digital platforms in preserving the integrity of knowledge and fostering equitable access to information.

            Foucault’s insights into the interplay of power and knowledge remain strikingly relevant in this context. The digital archive serves as more than repository of information, but a site of contestation where power operates to shape what society remembers and values. As these algorithms increasingly dictate the flow of information, a potential for resistance and reform becomes critical. Scholars, technologists, and activists must work to challenge these systems by advocating for transparency, accountability, and ethical curation practices.

            When reflecting on the future of the digital archive we have to consider its potential as a medium for amplifying diverse voices and fostering pluralistic discourse in addition to its potential for a tool of control. By addressing the biases and inequalities embedded in algorithmic systems we can begin to envision a digital archive that serves as a true repository of collective knowledge that values inclusivity and truth over profit and power. Because the archive shapes not only what we know but who we become as a society, so arguably, the stakes are incredibly high.

 

 

 

 

Bibliography

[1] Foucault, M. (1969). The Archaeology of Knowledge. http://ci.nii.ac.jp/ncid/BA26532621

[2] Silva, S., & Kenney, M. (2018). Algorithms, Platforms, and Ethnic Bias: An Integrative Essay. Phylon (1960-), 55(1 & 2), 9–37. https://www.jstor.org/stable/26545017

[3] YERLİKAYA, T., & ASLAN, S. T. (2020). Social Media and Fake News in the Post-Truth Era: The Manipulation of Politics in the Election Process. Insight Turkey, 22(2), 177–196. https://www.jstor.org/stable/26918129

[4] Berman, R., & Katona, Z. (2013). The Role of Search Engine Optimization in Search Marketing. Marketing Science, 32(4), 644–651. http://www.jstor.org/stable/24545014

[5] Dorothy Kim, TreaAndrea M. Russworm, Corrigan Vaughan, Cassius Adair, Veronica Paredes, T. L. Cowan, Anna Everett, & Guisela Latorre. (2018). Race, Gender, and the Technological Turn: A Roundtable on Digitizing Revolution. Frontiers: A Journal of Women Studies, 39(1), 149–177. http://www.jstor.org/stable/10.5250/fronjwomestud.39.1.0149