Gayatri Raman
8 min readDec 1, 2021


  1. Functionalism:

In the initial years of what came to be called sociology, there came to be an influential school of thought, Positivism, led by August Comte (1798–1857). This paradigm (or in some ways supra-paradigm) epitomised the seeing-is-believing approach of the natural sciences, of which it considered itself a spiritual successor, seeking to “Enlighten” the new fields of study that were the social sciences. The spirit was to be objectivist, rationalist and interested only in the verifiable. Durkheim, a thinker from a few centuries later, would crystallise these ideas, within the school of thought that was Functionalism. There would come to be decades of dialectic conversations between these ideas, but especially in the latter half of the 19th and the first half of the 20th century, these ideas had gained a new wave of credibility. To the Functionalists of Sociology, of whom Durkheim is foremost, society was a living breathing organism. Like any organism, it had body parts (the family, the state, the Church or organised religion), and it worked well if these body parts were healthy individually and as a whole. And, like any organism again, any of these body parts could fall sick, just like social ‘order’ could ‘decay’. The role of the functionalist therefore was to identify sites of such malfunctioning, point them out and suggest ‘scientific’ measures for quick redressal — a doctor doing his/her best.

Nazism, born in this highly Functionalist age, believed firmly in the racial sciences or Eugenics,and had an openly stated mission of improving the “racial stock” of German society, and to do so it had to cleanse it of all “undesirables”.The poster on the right is from Nazi Germany, telling citizens how a disabled man costs the state 60,000 reichmarks over the course of his lifetime, driving home how the “useful” able-bodied citizens, were subsidising the “uselsess” (Khazan, 2014).

This is therefore the principal limitation of Functionalism; It may produce a class of extremely qualified, and dedicated professionals, with a thorough theoretical understanding of what being a PwD entails. And they may have the resources to “solve the problem” as it were, just as a Functionalist would be interested in “solving the problems” of divorce, of suicide, of crime, of mass poverty and so on, like a doctor treating a patient. But this lacks the one fundamental quality which is sacrosanct today in the social sciences — that of “lived experience”. Also just like with Nazism, this approach can easily veer to dangerous extremes. Why is “lived experience” so easy to ‘forget’? To devalue? To leave out of the equation? I posit that it is because of principally two intertwined reasons:

  1. The people with disabilities or those for whom such interventions are decided upon, are usually at a lower position on the socio-economic ladder, lacking the degrees, qualifications or entry requirements,and hence aren’t represented in these decision making bodies.
  2. The implicit or deeply rooted “web of common sense beliefs” which operates when designing such technologies may not account for the “who” and the “why”, but is concerned primarily with the “what”, being driven by a top-down approach of what Hirschmann calls “naive realism” — If we simply understand the problem we can solve it.

What would an intervention designed to “overcome stigma” from such a position look like? One example, would be of a bot, very similar to what we see on social media and the larger Internet today. Just like bots on Twitter, which are fed algorithms to detect problematic content, to alert account holders of such activities, and at a certain point, to suspend accounts altogether, our bot would also aim to uncover stigma in the conversations people have online. If it finds evidence of such stigma, say derogatory comments, under the post of a person with disabilities, it can behave in the way bots have been taught to behave. So its rationale is that it is seeking to create a stigma, prejudice and violence free online society for users who are PwDs. It has defined such users in a category similar to how racial, ethnic or religious minorities can be classified, and sees the problem as a “technological one, requiring technological solutions” (Gillespie, 2020, p-1).

The budgetary allocation would therefore be to increasingly sophisticate the algorithmisation of these platforms; more reliance on machine learning with the aim to eventually reduce human intervention or control altogether — the machine should learn from each successive case, how to better the tracing and removal of problematic content. The cornerstone of this approach is that if you simply remove visible traces of stigma, it no longer exists. The key actors include a class of psychological and technological experts, who would create both a repository of words and symbols that are problematic, and the code required to detect and deal with them. These teams would need to work parallely and conjunctly. Their actions will be verifiable — the number of accounts suspended, alerted, so on, is available — and as Gillespie outlines,to the technologists, this is an exercise that is “scalable”, with operation cost minimisation being the guiding principle (Gillespie, 2020, p-2). Why would such an intervention run into problems?

As outlined above, the “lived experience” of stigma is simply too multidimensional and complex, to be understood by a bot. Further, simply removing content that is stigmatic doesn’t erase stigma from society or even online society. This is borne from a simplistic requirements understanding; erase these words = stigma removed. Some of the problems we see bots running into — a lack of an understanding of contexts, sarcasm, and sometimes a disproportionate targeting of the very groups it is built to protect — can emerge here too (Gillespie, 2020, p-3). Somewhat more dystopically, how does a society reduce the mitigation of stigma, which emerges over long periods of inequalities and dysfunction, to simple, cost-efficient bot-driven content regulation — in this we can see echoes of the same reductionist thinking from the Nazi poster.

2. Radical Structuralism:

This paradigm seeks ‘total change’ — a definitive, and final, shift away from the sources and sites of inequality or power imbalances. Arising from Marxist thought, and the idea of “class struggles”, technology as a highly probable tool and site of oppression, and the need for a complete structural change or ‘revolution’, are its key underpinnings. To Radicals, stigma, is ever visible and a constant reminder of the tremendous injustice that it arises from. Thus, unlike the functionalists, they seek more visibilisation, rather than invisibilisation, to raise consciousness about these injustices, and their primary aims would be to achieve social justice collectively, rather than maximising efficiency or cost effectiveness. In this paradigm, the developer would therefore seek “emancipation” and would need a heightened and deepened sense of the obstacles to human communication and development as outlined by Hirscheim and Klein — “authority and illegitimate power, peer opinion pressure, time, space and resource limitations, social differentiation, bias and limitation and language use” (Hirscheim and Klein, 1989, p-1208).

Thus, like Functionalism, the Radical response to the process is also Positivist — it also seeks actual, deep-rooted, meaningful change. However the ways in which it qualifies, measures, and seeks to achieve that change can differ greatly. For example, to the Radical, there will be a more fundamental set of questions that need to be answered — who are we designing this platform for? Who are we NOT designing it for? Why? Why is profit maximisation or cost-efficiency the ideal? If they are to be replaced as ideals, what is it that should take their place? If it is inclusivity and diversity, what sorts of inclusivity and diversity do we mean? Particularly in the case of People with Disabilities this means being very aware of the range of disabilities, the ways in which they can manifest as socio-economic and partcipatory barriers, and also of how technology cannot always be the solution, i.e, a sense of caution around techno-solutionism. Thus the key actors include not just developers but also the stakeholders; here they are not always degree-holders, but it is their ‘lived experience’ and the firsthand knowledge they have about their unique disadvantages which is more valued. There is emphasis on increasing the number of participants in the process, and making it as inclusive as possible — from the visually impaired to those in whose languages the social media platforms are not available, the aim should be to maximise representation from all possible groups, pointing to a growing, keener understanding of the different kinds of disabilities. In fact, the truest radical structuralism would contrast heavily with the functionalist emphasis on an expert driven process — here the process is driven by a very diverse set of stakeholders, each of whose unique voices is taken equally into account, and the technologists must build, with their priorities in mind. The success of the process depends on:

  1. How inclusive and representative the stakeholder selection process is
  2. The effort to ensure every concern is valued, raised and eventually addressed
  3. The keenness and awareness of the developers and technologists about the need for and strategic ways to achieve emancipation

Continuing on the example of social media platforms, Radicals would make, as the name suggests, radical changes to overturn stigma. This could be from something as symbolic, as changing the name of the platform, or publishing a manifesto on the home page which outlines the aim to be inclusive of PwD alongside other similarly disadvantaged groups, to something structural — raising the character limit, or delimiting the number of strikes for account holders of a particular group, especially when they post sensitive information in an effort to raise consciousness, that may have otherwise been deemed problematic — for eg: an activist using slurs or expletives that are commonly associated with PwD to point out how they are problematic, should not get banned. These measures would be constantly revised and updated with directions from an ever increasing group of stakeholders; with time the effort should be to elevate democratisation of this decision-making process. This could mean facilitating mechanisms such as enabling users to become part of the moderation and decision-making process, such as moderator accounts, etc. At each step of the process, the developers must constantly critique the platform they are helping create, and see if it is measuring up to the kind of emancipation they had sought to achieve. For example, Twitter, a popular social media platform is primarily geared around short text pieces called tweets. These may not be accessible for the visually impaired, and hence the developers, having realised this through their interactions with and feedbacks from the stakeholders should seek to include an easily accessible feature enabling automatic voice reading of tweets for the same. Similarly for the hearing impaired, there should be availability of subtitles in the language of choice for all video clippings or posts. While these are iterative processes, they should be driven and constantly interrogated by a firm desire for positive changes. The truest forms of radical structuralism would however, seek to create new platforms altogether, where such interventions and mechanisms are not add-ons, but the main feature, and each useror account holder is expected to be aware of and committed to the cause of the restructuring, in this case, overcoming stigma.

While this approach may seem the most inclusive an appropriate, it too can run into problems. The selecting of stakeholders, ensuring that their interactions with developers are meaningful, and that they result in a coherent set of changes and actions is a process that has scope for dissatisfaction or clashes among the groups if not managed carefully. Also the kinds of funding and support required for such endeavours may not always be available in the amounts required, and to have priorities independent of the funds and funders available is difficult. Thus we often find that while radical structuralism is admirable to many, it may not always entail pragmatic solutions.


Gillespie, T. (n.d.). Content moderation, AI, and the question of scale. Big Data & Society, July- December, 1–5. doi:DOI: 10.1177/2053951720943234

Hirschheim, R., & Klein, H. K. (1989). Four paradigms of information systems development. Communications of the ACM, 32(10), 1199–1216. doi:10.1145/67933.67937

Khazan, O. (2014, September 3). Remembering the Nazis’ Disabled Victims. The Atlantic.



Gayatri Raman

I am a research inclined student at IIIT -B currently seeking opportunities in Accessibility and HCI research within the larger ICT4D discourse