Whether under the banner of innovation or efficiency, technology is evangelised as the disruptor that will solve development problems. International development practice barely acknowledges the idea of embedded bias in technologies it is pushing in formerly colonised countries.
Introduction: Disrupting the disruptor
Technology is rarely advertised as a conservative defender of the status quo; one that standardises a mostly white English-speaking Western cis male worldview as the invisible background (Birhane & Guest, 2020). But some have started to disrupt the disruptor. For example, computer scientist and the founder of the Algorithmic Justice League, Joy Buolamwini, has highlighted the failure of facial recognition technology to read the faces of black people (Hardesty, 2018). However, while researchers and activists – many of whom happen to be Black women – are unmasking the default whiteness and maleness of common technologies in the West, the idea of bias and oppression in technologies that the West is pushing in formerly colonised countries barely registers in the international development practice.
In my final assignment for the New Media, ICT and Development course, I will revisit some of the themes in my earlier blog posts bringing together the antiracist approach and digital for development practice. Borrowing from the antiracist approach in development practice can help make sense of and respond to digital colonialism, “a structural form of domination exercised through centralized ownership and control of […] the digital ecosystem” (Kwet, 2019). While digital colonialism can be framed within decolonial critique (e.g. Mohamed, Png, Isaac, 2020), the antiracist approach offers a much needed practical extension to the theory.
The road to tech hell is paved with all kinds of intentions
In ‘How to be an Antiracist’, Ibram Kendi (2019) argues that thinking that reduces racism to a pejorative, a characteristic of an unsavoury individual is preventing us from moving the conversation from interpersonal interactions to understanding how racism manifests in patterned behaviours, structures, and institutions. This thinking allows people to disengage under claims of colour blindness. However, one can only be a racist or an antiracist. This argument is similar to Hannah Arendt’s notion of political responsibility (Stonebridge, 2019), Iris Marion Young’s social connection model (Rothberg, 2019), and Rothberg’s writings about the implicated subject (ibid), which I explored during the previous course module Culture and Media Analysis. The above are less concerned with individual intentions and actions than the impact of our participation in exclusion and exploitation by the virtue of our position.
Similarly, technology is neither the utopia that it has been presented as nor the dystopia of our nightmares. It is both, depending on who you are. In ‘Black Communities Are Already Living in a Tech Dystopia, Janine Jackson writes “there was an earnest desire to see the power inequities reflected in corporate media – the racism, sexism, class and other biases – designed out of existence by some new delivery mechanism” (2019). However, technology has emerged to replicate the problems and harms that already exist in society (ibid). Earnest desires or intentions behind a technology are less relevant than the context, according to Ruha Benjamin, Associate Professor of African American Studies at Princeton University and the author of Race After Technology. It is not that technology is necessarily designed or implemented with explicit intention to cause harm – most are driven by profit or efficiency and simply lack any awareness of, or concern for, such collateral damage. It is that unchecked bias allows existing inequities to be replicated, automated and legitimised (Benjamin 2019; Mohamed, Png, Isaac, 2019).
Because technologies are proprietary, we cannot check the precise mechanisms that reproduce the bias (Benjamin, 2019). But we do know that development programmes don’t just apply but fetishise similar technology-centred problem solving logic in formerly colonised countries, often under celebrated public-private partnership models.
Digital for development or colonialism 2.0?
Tony Roberts distinguishes between digital in development where tech is used routinely in an operational way and digital for development, where technology is explicitly linked to the outcomes of development interventions (2019). Anyone working in the development sector is used to seeing indicators measuring the effectiveness of the latter – whether in relation to an e-Health app or machine learning to identify girls at risk of dropping out of school.
What could be wrong about deploying technology to improve health or education outcomes? As argued above, technology is neither good nor bad: intentions are less relevant than the context within which it operates. Yale Law School fellow Michael Kwet describes a number of examples where the US is reinventing colonialism in the Global South through the domination of digital technology (2019). For example, education projects in South Africa planted US tech products inside the classroom incorporating Big Data surveillance across the entire education system creating technological dependencies without any public debate (ibid).
One does not need to be an undercover journalist to find similar examples across the continent. In my own day job, I have been in meetings with some of the most unaccountable tech companies in the world discussing development collaboration to reform a failing national education system in sub-Saharan Africa and to make it more accountable. Imagine, what might happen if another powerful nation made the US or UK government dependent on its digital platform while collecting and owning data on their citizens! (Oh, wait! This would happen)
Of course, most large donors adhere to the Principles for Digital Development, but these seem more like a risk management tool to avoid a One-Laptop-Per-Child-type disaster, to mitigate against reputational harm, rather than to engage in the kind of self-reflexivity activists are calling for in the West. Development projects rarely, if ever, engage critically in questions such as how do unacknowledged systems of values and power inhibit our ability to assess harm (Mohamed, Png and Isaac, 2020); whose values are being represented, and how do structural inequities result in unequal spread of benefits and risk (ibid); and which technologies best promote privacy rights, transparency, collaboration, and local development (Kwet, 2019)
I am sceptical about leaving answering to these questions to Big Tech, donors and their contractors (who are driven by neoclassical theory and the gospel of economic growth where growth somewhere means growth everywhere). But ignoring these questions altogether does not just replicate, automate and legitimise coloniality (patterns of power that emerged as a result of colonialism that define culture, language etc) but reinvigorates colonialism (economic and labour exploitation etc). Yes, technology connects people, but so did the railroads the British built in their colonies to link up commercial and military outposts to the seaports to enable the exploitation of indigenous populations and their land (Kwet, 2019). A development project team with its digital literacy intervention is like the missionary a hundred years earlier arriving to teach English to the indigenous population: equally vital for taking part in society and to exist in a system that does not exist for you.
From ‘do no harm’ to undoing the harm in digital for development
If digital for development took the antiracist approach, rather than settled with the administrative “do no harm”, development programmes would acknowledge the harm that already exists. They would invest resources into investigating how a programme might not just replicate this but actively reduce it. As Benjamin puts it, this would mean discerning the difference between a tech fix and fundamental questioning of the infrastructure and the society in which the technology is being deployed (2019).
At the outset, the ugly underside of the tech industry in formerly colonised countries may seem different from that of the West. Of course, specific issues to do with mining for raw materials, labour exploitation, and dumping digital waste in Africa also need addressing. But there are also close parallels with the design and use of technology: the coding of existing biases, the promotion of technical fixes to social problems, and where few profit at the expense of the most vulnerable.
Yet the dots are not well connected. This reflects the broader disconnect between critical race theory and decolonial studies. In addition, both seem far removed from every-day development policy and practice. I would like to see more discussion of what the different issues mean in different contexts.
The antiracist approach requires self-evaluation and awareness regardless of intention. A few Black female academics and computer scientists have been pushing antiracist thinking into the field of technology because “coded inequity is perpetuated because those who design and adopt such tools are not thinking carefully about systemic racism” (Benjamin, 2019). A similar radical approach is needed in digital for development. The intention of a development programme not to do harm is not the same as being “anti-harm”. Digital for development needs to become a more reflexive practice. I don’t think this achieved by adherence to the Principles for Digital Development. A more interesting list to start with could be the Design Justice Principles, where step one is to seek liberation from exploitation and oppression. Instead of teaching digital literacy as “assimilation into US products” (Kwet, 2019), development implementers and funders should demand – both of themselves and the designers of technologies – literacy in the effects of power and privilege.
***
Personal reflections
This is my final assignment on the New Media, ICT and Development module. In some ways, I wish it would have been the first, as it is one where I have most engaged with the topic. The module was mostly delivered through teamwork. Consensus isn’t always the most effective way to explore a topic, and depth was sacrificed for the ‘universal’ to accommodate each team member’s preferences and limitations. There were also various requirements related to content creation and mechanical elements of running the blog, which were valuable in their own right, but which nonetheless diverted my attention from the subject matter. I have already written about these practical aspects and challenges in the Group Report and course feedback, so in my final reflections, I will use the opportunity to revisit the topic itself.
Both the positive and negative aspects of tech and datafication are often framed at an individual or household level. This module has been a fascinating intellectual journey into considering the impact of the above at the societal level. The reading allowed me to connect the theme with some of my previous Communication for Development course assignments – particularly on the politics of memory, implication and social justice. It also helped me join the dots between my academic background in sociology of race and ethnicity with my professional experience in international development. Race studies have always been strongly US-centric, whereas, in development, sociology has played the second fiddle to economics, a field which seems to be deeply uninterested in themes around power and privilege. The disconnection between my academic background and every day work has often left me questioning whether I am in the right place.
I saw this essay as an opportunity to argue for looking at these issues through the two lenses. I did this by discussing the application of the antiracist approach in development practice. Although, I didn’t have the space in this post, I do think that the reverse – applying a decolonial lens in antiracist practice – would offer an equally interesting discussion. As Steve Biko said: “the most powerful weapon in the hands of the oppressor is the mind of the oppressed”. Biko’s home country South Africa may in fact offer an example where antiracist and decolonial schools of thought have by necessity worked together: all the more reason to add Michael Kwet’s work to my UK/US -centric reading list!
The exercise has also been somewhat cathartic. For most of the time, the reality of international development programming couldn’t be further removed from the visions of my younger Freire-reading self – which most certainly didn’t include meetings with tech giants about profit margins. But you can’t blow a whistle at a system that is doing what it is designed to do. One person can’t change the system. But one person can at least blog about it and hope that it might spark something else.
References
Birhane, A., Guest, O. (2020) ‘Towards decolonising computational sciences’, submitted to Computers and Society on 29/09/2020 available at https://arxiv.org/abs/2009.14258 accessed on 20/11/2020
Hardesty, L. (2018)’ Study finds gender and skin-type bias in commercial artificial-intelligence systems’ in MIT News 11/02/2018 available at https://news.mit.edu/2018/study-finds-gender-skin-type-bias-artificial-intelligence-systems-0212 accessed on 21/11/2020
Jackson, J. (2019) ‘Black Communities Are Already Living in a Tech Dystopia’ CounterSpin 15/08/2019 available at https://fair.org/home/black-communities-are-already-living-in-a-tech-dystopia/ accessed on 14/11/2020
Kendi, I. (2019) How to be an Antiracist. Penguin Random House LLC.
Kwet, M. (2019) ‘Digital Colonialism: US Empire and the New Imperialism in the Global South’ Race & Class Volume 60, No. 4. Available at file:///C:/Users/DulcePedroso/Downloads/SSRN-id3232297%20(1).pdf accessed on 14/11/2020
Mohamed, S., Png, M-T., Isaac, W. (2020) ‘Decolonial AI: Decolonial Theory as Sociotechnical Foresight in Artificial Intelligence’ in Philosophy & Technology published online on 12/07/2020, available at https://arxiv.org/abs/2007.04068 accessed on 20/11/202
Roberts, T. (2019) ‘Digital Development: what’s in a name?’ in Appropriating Technology blog 09/08/2019 available at http://appropriatingtechnology.org/?q=node/302 accessed on 14/11/2020
Rothberg, M. (2019) The Implicated Subject. Beyond Victims and Perpetrators. Stanford University Press
Stonebridge, L. (2019) ‘Hannah Arendt is the philosopher for now’ New Statesman 22/03 available at https://www.newstatesman.com/culture/books/2019/03/hannah-arendt-resurgence-philosophy-relevance accessed on 09/08/2020
