On Language in AI

This morning which I asked ChatGPT to edit a piece of my paragraph, it flattened its tone, complexity and density altogether. I was quick enough to point it out to the bot, and it was quite self critical. It said:

Many editing models—human and AI alike—are trained to assume that ambiguity is a problem to be solved. They tend to move toward clarity, linearity, and explicit causality. But in certain kinds of writing, ambiguity is not a flaw; it is the intellectual and affective work of the text.

This was reassuring as it echoed what my Yale advisor Eeva Liisa Pelkonen also brought to my notice after reading and interacting with me and my work during my masters; she had advised, "Channel your ambiguity in your writing." Not only did she realise that my doubtful voice was a form of thinking itself, but she also validated it as a powerful tool to produce discourse. I have often wondered how people - intelligent or otherwise become overtly assertive about their thoughts. I believe that the intelligent must know that there is more (to) knowledge beyond what they know, and the other has to know that they are asserting to hide their ignorance or in lack of knowing something. In such a scenario, how can assertion create any ground for productive thought-exchange? Arguments, debates, and reasoning happen precisely in this space of ambiguity, which allows us to challenge what we already know, or expand our thoughts into something we do not know. Does it then become a prerogative only of the curious minds to engage and sustain uncertainty?  ChatGPT further explains studying the past patterns in my writing:

The uncertainty is not incidental—it is the subject of the reflection. To resolve that ambiguity is to alter the thought itself.

The transitions are not fully explained. They are associative rather than argumentative. A conventional editor might ask for clearer links, but the gaps themselves create meaning. They allow the reader to inhabit the process of thinking rather than merely receive its conclusions. 

That 'ambiguity' alters thought is precisely the dialectical reasoning in action through a singular agency. It produces a stream of reflection in which all thoughts are constantly under scrutiny. This kind of writing does not burden itself with citations, rather indexes back to its own past experience, or its lack thereof, constantly invoking consciousness of thought. This is one way of producing knowledge. The Vedics arrived at their version of truth of external reality through disciplined practice of interrogation of the (inner) self. The Western tradition, on which the idea of knowledge largely rests today, tries to use the body as an instrument to arrive at an objective description of reality by minimizing the reliance of the self.  

Eventually ChatGPT told me what I consciously do:

Your writing often seems to think through language rather than merely report conclusions. The sentence is not just carrying an idea; it is exploring it. 

The simultaneous interrogation of thought and language is one within which I exercise meaning. In negotiating the various vectors of meaning within a singular word, I condition my text to hold more than an simple unilateral idea. This is the work of writing, and the task of labour. More importantly, it is conscious work at play - such that any piece of writing becomes accessible, or meaningful to a larger body of people. In other words, one must create multiple entry points within one's writing to expand its sphere of engagement, and influence. Thus, when I responded to the bot: "I like to make my sentences layered and dense with meaning often times rather than dumbing them down like scientific papers. I think that is also one of the dimensions of humanities work right?"; it went on to explain:

One of the distinctive features of much humanities writing is that language is not treated as a transparent vessel carrying information. Language is itself part of the inquiry. The sentence does not merely communicate a thought that already exists; it participates in the production of thought.

In many scientific disciplines, the ideal sentence aspires to be unambiguous, replicable, and resistant to multiple interpretations. Ambiguity is often considered noise because it interferes with the transmission of information.

Many strands of humanities scholarship, however, begin from the premise that the world itself is ambiguous, layered, and resistant to singular explanations. In that context, a sentence may deliberately sustain multiple possibilities because the object of study demands it. 

In the recent past, and recurrently, I have argued for institutional discourse to be more interpretive rather than instructional. The convening of the Graduate Symposium SEA 'Doubts' was to bring ambiguity to the centrestage of academic discourse. Incidentally, 'Noise Fields' was another Annual Conference thematic I had conceived for SEA, where noise was the metaphor to hold polyphonic perspectives together. In the subsequent year where SEA Annual Conference was on the subject of how must institutions grow together, I had presented along with Prasad the case for institutions to be situated amidst interpretive landscapes rather than between walls. The debate of all these ideas in their humanities spectrum and their odds with how traditional scientific discipline orders and utilises language is then quite revealing. It means that my endeavour has been to open up the overtly scientific underpinnings of what constitutes knowledge and bringing them under the purview of ambiguity. In analysing my ways of approaching writing further, ChatGPT says:

What I find interesting in your writing is that it seems to emerge from a tradition that sits somewhere between academic prose and essayistic reflection. It reminds me less of conventional social science writing and more of writers who use prose as a space for thinking—people who allow sentences to carry uncertainty, memory, atmosphere, and concept simultaneously.

The challenge for such writing is not to become clearer at all costs. The challenge is to ensure that every ambiguity is doing work.[...]

And I suspect that what you are often pursuing is not obscurity but what one might call semantic abundance—the feeling that a sentence contains more than can be exhausted in a single reading. That is very much a recognizable aspiration within many traditions of humanities writing, and it is one of the reasons certain essays, catalog texts, and theoretical works reward rereading decades after they were written.

 And the above, has been my pursuit, like mentioned before. I really like the conceptualisation of 'semantic abundance' - for writing to be able to hold multiple meanings at once. I have always found the 'allegorical' form in literary texts the most fascinating, after metaphor and pun. The allegorical in writing is the most artful way of being political, or rather, most political way of being artful...Here, language works like a chameleon, changing colours in cascading contexts - camouflaging, yet out there, hidden yet visible. In employing this, language becomes mirror - "Do we see the same person in the mirror every day?", Eeva had very early on asked me. Despite that mirror is simply doing its work like it has always done, it is at the same time presenting a different picture of us, only if we are willing to see. And it is here, that writing becomes a work of art, which the mathematical models of AI may never be able to calibrate accurately...for it is only the human mind that knows what ghosts to project for the dreams of the future. 

of Dipesh Chakraborty and Gayatri Spivak

a comparison between the historiographic ideas of Dipesh Chakraborty and Gayatri Spivak

For a very long time, I have had difficulty understanding premise of Dipesh Chakraborty's 'Provincialising Europe.' I have reached out to my academician friends to explain, and they have really tried well - however, conversations evaporate and learnings do not settle in the mind. Some time ago therefore, I reached out to ChatGPT for those questions, and it really laid out the stakes well. I wanted to understand quite straightforwardly: what is the key argument of Dipesh Chakraborty in Provincialising Europe. 

In quite direct and simple language, it answered that the key argument of Provincialising Europe by DC is the pursuit of how we can understand modernity, history and political life without treating Europe as the automatic center or universal model for all societies. Chakraborty challenges that idea that concepts such as "capitalism", "democracy", "modernity", "citizenship", "secularism" developed in Europe from specific European histories and are often presented as universal. These (should not) automatically explain societies in Asia, Africa or Latin America. Post colonial societies live in both at once. They are modern but not in purely European ways. Further, his concern is that Europe dominates how we think about time, history, and modern life.

Following this explanation, ChatGPT prompted me if I was interested in understanding how his proposition differs from Said, Spivak or Fanon. 

and I jumped into that provocation. I really like how it laid out the difference:

The difference between Dipesh Chakrabarty and Gayatri Chakravorty Spivak lies mainly in what they think the core problem of Western theory is—and how it can be addressed.

Put simply:

Chakrabarty asks: How can we rethink history and modernity beyond Europe?

Spivak asks: Who gets to speak, and who is silenced, even in critical theory?

It says that Spivak's concern comes from the realisation that all theory (whether western, or not) is embedded in power. is that "even progressive thinkers often silence the most marginalised people" She asks us to be suspicious of all theory, and skeptically suggests that intellectuals often reproduce heirarchy unintentionally. Thus, if Chakrabarty suggests "localisation" of theory indirectly, Spivak does not trust easy "localisation" of theory, because she asserts that even postcolonial scholars may dominate subaltern voices. For example, how do the lives of workers and peasants challenge our theories when they do not fit western categories? Further she says that the "subaltern" is often spoken about, but rarely allowed to speak for themselves. When they do speak, elites translate and distort them, leading to structural silencing. 

Thus, Chakraborty is saying - Don't reject western theory, rather "provincialise" it. Meaning, as Gautam Bhan said in one of his talks, "particularise" it in its region and dont assume it as a universal standard for everywhere. 

On the other hand, Spivak says that critical scholars too, often silence the oppressed, so we must question our right to speak for others.

Marxist vs. Socio-economic Approaches

 a few days ago, I asked ChatGPT if a "Marxist" method of narrative of life is same as that we call as "socio-economic' way of thinking about life? 

It explained me that partly, it may be right, but full not so. 

The Marxist way of understanding life comes mainly from the ideas of Marx and Engels which views life through:

• Material Conditions (work, income, housing, resources)
• Class relations (who owns, who labours)
• Power and exploitation
• Historical change through conflict 

pretty standard. 

So a Marxist narrative asks:

How do economic systems shape people’s lives, identities, dreams, and suffering? Who benefits, and who is disadvantaged?

All in all, it concluded that while a socio-economic reading asks "Who lives where and at what cost?"; a Marxist reading asks: "who profits from this spatial arrangement and why?" Thus Marxist reading will necessarily weave it into the politics of capitalism as a conspiracy. 

When I further asked if 'socio-economic' thinking as a derivate of Marxist thought, it elaborated that Socio-economic analysis did not grow from Marx alone. It also comes from other traditions. For example, Max Weber argued that society is shaped by ideas as much as money, and hence culture, religion and values matter. Therefore power is not purely economic. This is different from Marx. 

Emelie Durkheim on the other hand was focusing on social cohesion through institutions and moral norms; and was looking at how societies hold together - not mainly in class conflict but shared collective life. 

Larger socio economic thinking comes from welfare and policy studies which are studied in development economics, public policy, statistics and demography. These look at socio economics through poverty index, literacy rate, HDI (human development index) or uemployment data. Thus, 

one can imagine this at three levels, says Chat GPT:

Level 1: Descriptive socio-economic
“Low-income groups have less access to healthcare.”
(No Marxism here.)

Level 2: Analytical socio-economic
“Economic structures reproduce inequality.”
(Some Marxist influence.)

Level 3: Marxist
“Capitalism structurally produces inequality through exploitation.”
(Fully Marxist.)

In fields like architecture, urbanism, and cultural theory:
Using a socio-economic lens = mapping class, income, access
Using a Marxist lens = questioning land ownership, speculation, capital flows

Many scholars use the first without committing to the second.

I asked further: Do people fall into using Marxist frames because it gives them a sense of purpose of "equalising" society?

It said:

Yes—for many people, Marxist frameworks are attractive partly because they offer a strong sense of moral and political purpose, especially around equalising society. But that is not the only reason people use them. People are drawn to Marxist frames because they combine explanation, critique, and purpose in one system.

Some people use Marxist language without deep commitment to equality. Sometimes when people adopt Marxism mainly for purpose, some risks appear: For example, there is a danger of oversimplification in framing everything as  “class struggle”. There is a projection of Moral certainty wherein the People feel automatically “right” when they are posing a certain situation as Marxist. And lastly, sometimes it can feel quite dogmatic when Other perspectives are dismissed.

This discussion seemed insightful, and worth keeping in notes here.

on Fetish

 The word "fetish" is used in different ways in psychology, culture and social theory. At its core, it means giving special emotional or symbolic power to something that is actually just an object or detail.

1. Freudan

object or detail becomes more important than the whole person

eg: someone feels especially attracted to shoes, or is very focused on the hair/fabric, or someone feels excitement mainly because of a uniform...

2. Cultural / Religious

an object believed to have special powers. The object itself is ordinary, but people believe it has hidden power.

eg: A statue thought to contain spiritual energy, a necklace believed to protect someone, etc.

3. Social/Economic Meaning (Marxist)

The commodified object looks magical..

eg: a phone might feel desirable, but one forgets the workers who made it, the factory, the conditions of working in valourising it. 

4.  Everyday Use

people having an intense liking for something.

eg: coffee, fitness

to summarise:

Fetish happens when we give too much meaning, power, or emotion to something small, symbolic or ordinary

German School and the French School of Critical Theory

I was introduced to the two schools of Critical Theory during my masters at Yale School of Architecture. While Peggy Deamer, our teacher mentioned that her course would be more to do with Marxist analysis of architecture (and that those who did not believe in it should strongly reconsider being in the course); and that there were two schools of Critical Theory - the Frankfurt School (German Tradition) and the French School - I had not quite understood what was the difference between the two.

Over the last many years after doing the course, the question took a back seat, and did not come to resolution even if I must have tried to relook specifically what separates the two in their ideological orientation. Finally last week, as a matter of time pass, I asked this question to ChatGPT and it gave me some of the most convincingly understandable response, which I want to note down here for my own understanding.

Very briefly, 

The German tradition of the Frankfurt School of critical theory has Marxist roots with strong engagement with capitalism and class. It is focused on the critique of mass culture that emerges out of it - how media and consumer culture shape consciousness. It is also concerned with the question of reason and emancipation - in other words "rationality". This is necessarily the Western notion of scientific rationality that developed through the renaissance. 

In summary, they analyze how modern society produces domination especially through capitalism, bureaucracy, technology and culture industries. Their work is generally systematic, moral-political, and reform oriented.

The French tradition mainly develops after the 1950s reacting against structuralism and classical Marxism. 

Their core features can be located in their distrust of universal explanations (like Marxism or Enlightenment theory) - in other words, the suspicion of "grand theories". They focus on how meaning, truth, and knowledge are constructed with a focus on discourse and language. They conceptualize power as "diffused", meaning, power is not only in the state of economy, rather everywhere. Furthermore they are anti-essentialists, with no belief in fixed "human nature" or stable subject. 

They examine how power operates through knowledge, language, institutions and norms. Their work is often fragmentary, experimental, and skeptical of political blueprints 

ChatGPT articulates for me that in simple terms.

German Critical theory asks:

How does capitalism and modern rationality dominate us, and how can we overcome it.

whereas, French Critical theory asks:

How do language, knowledge and institutions produce what we think is normal or true.

I came to articulate then that, the essential different between the two is that one emerges from the critique of mass, whereas the other from the critique of the body. The AI helped me lay out that the key entry points for the German and French theory  are mass/system versus body/micropower. In a more precise academic formulation, the German critical theory begins with the problem of mass domination, while French theory begins with the problem of bodily and subjective regulation. One critiques how power works from above and through systems, the other how power works from within and through bodies.