Latest / Sleepy Facts About History / How Did the Census Shape British India? - Caste, Religion, and the Colonial Census | Sleepy Facts About History

Transcript

How Did the Census Shape British India? - Caste, Religion, and the Colonial Census | Sleepy Facts About History · Sleepy Facts About History

0:00
Welcome to Sleepy Facts About History. We tend
0:13
to think of a census as a simple mirror held
0:15
up to society, a neutral counting exercise that
0:19
discovers groups already waiting to be counted.
0:22
But the colonial census of British India reveals
0:26
a much more complicated reality. This episode
0:30
explores how the modern census functioned not
0:33
as a mirror, but as a mold. that helped create
0:36
and harden social identities. From the late 19th
0:40
century through the final colonial enumeration
0:43
in 1931, British administrators translated complex,
0:48
localized ways of life into standardized boxes.
0:52
In doing so, they transformed fluid systems of
0:56
caste, religion, and language into rigid, countable
0:59
categories with profound political consequences.
1:04
The story is not just about administrative imposition.
1:07
Indian communities quickly recognized the power
1:11
of these classifications and learned to play
1:13
the game. Cast associations organized campaigns
1:16
to secure higher rankings. Religious communities
1:20
mobilized around demographic numbers to demand
1:22
political representation. And individuals navigated
1:26
the enumerator standing at their doorway with
1:28
strategic answers. Deliberate ambiguity. or outright
1:33
resistance. The resulting data eventually provided
1:37
the vocabulary for drawing the borders of India
1:40
and Pakistan in 1947. I wanted to understand
1:45
why a simple printed form carries so much historical
1:48
weight. Some questions sound straightforward
1:51
until you try to answer them, and few objects
1:54
are as deceptive as the census schedule. Tonight
1:58
we look behind the statistical tables to examine
2:01
the human judgment. confusion, and negotiation
2:05
that produced the numbers. Exploring how the
2:08
act of counting people actively shaped the modern
2:11
world they lived in, the census form arrives
2:17
at the doorway like any other piece of paper.
2:20
But what follows is neither simple counting nor
2:23
neutral description. It is one of the most consequential
2:27
acts of translation in the modern state, and
2:31
the people of colonial India learn to treat it
2:33
as such. Imagine a printed form arriving in a
2:37
village somewhere in British India, perhaps in
2:40
the 1890s. A man stands at a doorway holding
2:44
this form. He has been trained, briefly, to ask
2:49
certain questions in a language that may not
2:51
be the one spoken inside. The form has boxes.
2:55
The boxes have labels. The labels whose words
3:00
that someone in an office far away decided would
3:03
cover everyone in the Empire. The person inside
3:07
the house answers as best they can. Or they refuse.
3:13
Or they answer differently than they did to the
3:15
last enumerator. Or they have already been told
3:19
by their caste association exactly what to say.
3:22
The man with the form writes something down.
3:26
That answer travels up through district offices
3:29
and provincial capitals. Eventually it reaches
3:32
Calcutta or Delhi, where clerks sort it, count
3:36
it, and publish it in thick volumes with tables
3:39
and footnotes. Those tables then shape who gets
3:43
represented on councils. They shape who gets
3:46
recruited into which army regiments. They shape
3:50
where schools teach, in which language. And eventually...
3:54
They shape how borders are drawn when empires
3:57
end. This is the modern census. Most of us have
4:02
filled one out. We tend to assume it is a mirror
4:06
held up to society. We assume it is a straightforward
4:10
counting exercise that discovers groups already
4:13
there, waiting to be counted. The categories
4:17
seem obvious, natural, pre -given. What else
4:21
would a census do but count what exists? Tonight
4:25
I want to show you that this assumption misses
4:27
almost everything that matters. The census is
4:31
not a mirror. It is a mold. It does not simply
4:36
discover stable groups. It helps create and harden
4:40
them. And the people being counted are not passive
4:44
objects. They negotiate, resist, strategically
4:48
adopt, and sometimes weaponize the very categories
4:51
designed to contain them. To make this concrete,
4:55
I will focus on one exceptionally well -documented
4:57
case. This is the census of British India from
5:01
the late 19th century through the final colonial
5:04
enumeration in 1931. India is not the only place
5:09
where this happened, but the records are rich
5:12
enough to let us see the machinery at work. They
5:15
also let us see the human responses to it. The
5:19
British colonial state faced a practical problem
5:22
of staggering scale. After the suppression of
5:25
the 1857 rebellions, the Crown took direct control
5:30
from the East India Company. Administrators suddenly
5:34
needed to govern a subcontinent of roughly 250
5:36
million people. These people spoke hundreds of
5:40
languages. They were organized into social forms
5:44
that British officials barely understood. The
5:47
administrators needed to tax, recruit soldiers,
5:51
maintain order, build railways, and eventually
5:54
managed political representation. They did not
5:57
know who lived where. They did not know what
6:01
people believed. They did not know what languages
6:04
people spoke. And they did not know how people
6:07
understood their own social identities. Local
6:11
knowledge certainly existed. Village headmen,
6:15
landlords, religious specialists, and merchants
6:18
knew their own worlds intimately. But this knowledge
6:21
was contextual and negotiated. It was often deliberately
6:26
opaque to outsiders. A person's identity in one
6:30
setting, such as their occupation -based status
6:33
in a particular village, might differ from how
6:36
they were known in a neighboring region. It might
6:39
differ from how they presented themselves to
6:41
a tax collector. It might differ from how they
6:44
understood their own religious practice. Caste,
6:48
in particular, operated as a complex, localized
6:52
system of relationships rather than a fixed,
6:54
ranked hierarchy that could be printed in a table.
6:58
Language shaded into dialect across continuous
7:01
territories. Religious practice blended and overlapped
7:05
in ways that resisted exclusive categorization.
7:09
The British needed to make this complexity legible
7:12
from a distance. The census was one tool among
7:16
several. It sat alongside cadastral surveys mapping
7:20
land ownership, vital registration tracking births
7:23
and deaths, and eventually identity documents
7:25
and passports. But the census was special because
7:29
it aimed at comprehensive enumeration. Everyone
7:33
counted. Everyone went in a box. And to put everyone
7:38
in a box, you first had to design the boxes.
7:42
The first all -India census was attempted in
7:45
1871 and 1872, though coverage was uneven. By
7:51
1881, the operation had become more systematic.
7:55
It continued every ten years thereafter. The
7:59
machinery was immense. Millions of forms were
8:03
printed in multiple languages. Thousands of enumerators
8:07
were recruited, often from local clerical classes.
8:11
They were given brief training and sent into
8:13
villages and city neighborhoods with schedules
8:16
to complete. Let us look more closely at that
8:19
schedule, because it is easy to miss how much
8:22
work is embedded in such an ordinary object.
8:25
The typical census form was a large sheet sometimes
8:28
folded with printed columns running down the
8:31
left margin. Name Age Sex Occupation Religion
8:39
Language Cast. Each column had to be filled in
8:45
ink, in a hand legible enough to survive the
8:49
journey upward. The forms were printed in English
8:53
and in vernacular languages, but the categories
8:56
themselves were designed in English and then
8:58
translated. A form in Hindi or Tamil or Bengali
9:03
carried the conceptual framework of the colonial
9:05
office into the village. The instructions printed
9:09
at the top or on the back told enumerators exactly
9:12
how to proceed. Ask the questions in order. Record
9:17
the first answer given. Do not argue. Do not
9:22
explain. But of course, enumerators did argue.
9:26
They did explain. They did make judgments when
9:30
answers did not fit the printed columns. The
9:34
journey of a single completed form reveals the
9:36
scale of the operation. An enumerator in a village
9:40
in the Punjab might complete 50 forms in a day,
9:44
walking from house to house with a sheaf under
9:47
his arm. Each evening, he submitted his forms
9:51
to a supervisor, who checked for completeness
9:54
and apparent errors. From there, the forms travelled
9:57
by bullet cart or train to the district headquarters.
10:02
Clerks sorted them by village and taluka, checked
10:06
them again, and compiled summary tables. These
10:10
summaries went to the provincial capital, Lahore,
10:14
Bombay, Madras, Calcutta. Larger teams of clerks
10:20
extracted the data onto standardized summary
10:23
sheets. Finally, the provincial reports traveled
10:26
to the census commissioner in Calcutta or Delhi.
10:29
The commissioner oversaw the compilation of the
10:32
All India volumes. The 1901 census alone produced
10:37
28 provincial reports. and a general report of
10:41
several hundred pages. The 1931 volumes were
10:45
so extensive that the general report required
10:47
its own binding and shipped in multiple parts.
10:50
The published reports themselves were material
10:54
objects of considerable weight. Literally, a
10:58
complete set of the 1901 census reports would
11:01
fill several shelves. Inside, page after page
11:06
of dense tables showed the population broken
11:09
down by district by religion, by language, by
11:13
caste. The footnotes, often overlooked, are where
11:17
the real intellectual work shows through. Commissioners
11:21
used footnotes to explain why a category had
11:24
been changed. They explained why two groups previously
11:28
counted separately were now merged. They explained
11:31
why a particular term had proven impossible to
11:34
translate consistently. These footnotes are admissions.
11:39
In the bureaucratic voice that the census was
11:41
not simply counting what existed, it was struggling
11:45
to impose order on resistance. But the machinery
11:49
only worked if the categories made sense. Not
11:53
to the people being counted, but to the people
11:56
doing the counting. And here is where the real
11:59
work of the census happened. This is work that
12:03
most of us never see when we fill out our own
12:06
forms. Consider the category of caste. Census
12:11
Commissioner H .H. Risley, who oversaw the 1901
12:15
enumeration, made caste classification a central
12:18
project. He attempted to rank all castes by social
12:22
precedence. He tried to create a single all -India
12:26
hierarchy from thousands of local contextual
12:28
identities. This was not a neutral recording
12:32
of what existed. It was an enormous act of translation
12:36
and simplification. What did this actually look
12:39
like on the page? The 1901 cast tables arranged
12:44
groups in columns with attempted rankings indicated
12:47
by numerical order or descriptive labels. A page
12:52
might show Brahman at the top, followed by various
12:56
scribal and trading casts, then cultivating groups,
13:00
then laboring and menial categories. But the
13:04
footnotes reveal the confusion beneath the apparent
13:06
order. In one district, a group called themselves
13:10
Kermi and were recorded as such. In the next
13:14
district, a group with a similar occupation but
13:17
a different name were classified separately.
13:20
Then they were merged. Then perhaps separated
13:24
again in the next census. The 1901 report contains
13:28
pages of discussion about whether particular
13:31
groups should be classified as agricultural or
13:34
pastoral. It discusses whether they should be
13:38
classified as Hindu or tribal. It discusses whether
13:42
they should be classified as one caste or several.
13:46
Risley himself acknowledged that the ranking
13:48
project was imperfect. He acknowledged that local
13:51
variations made any All India scheme provisional
13:54
at best. The Enumerators Instruction Manual for
13:58
1901 told enumerators to ask, what is your caste,
14:03
and to record the answer as given. But it also
14:07
told them to use their judgment if the answer
14:10
seemed vague or evasive. It told them to consult
14:15
local informants if necessary. The manual warned
14:19
against recording subcasts that were too numerous
14:22
to tabulate. It instructed enumerators to fold
14:26
these into broader categories. An enumerator
14:30
standing in a doorway, form in hand, faced a
14:33
practical problem. The respondent said something
14:36
that did not match any category he recognized.
14:39
Did he write it down verbatim? Did he ask again?
14:44
Did he check with the village headman? Did he
14:48
simply force the answer into the nearest available
14:50
box? The published tables smooth away all this
14:54
variation. They present clean categories and
14:58
precise numbers. But the Commissioner's complaints
15:01
about vague answers and incomplete returns testify
15:04
to the messiness underneath. The translation
15:08
problems were constant and revealing. The English
15:11
word caste itself covers two quite different
15:15
concepts. One is vana, the fourth old brahmanical
15:19
scheme of priest, warrior, merchant, and laborer.
15:24
The other is jati. the thousands of localized,
15:28
occupation -based, often regionally specific
15:30
groups. Indian languages did not necessarily
15:34
make this distinction the same way. Respondents
15:37
might answer in terms that made sense locally,
15:40
but confused enumerators trained in a different
15:43
framework. Census reports are full of official
15:47
complaints about respondents who did not know
15:51
their caste or who gave different answers at
15:53
different times. These complaints reveal not
15:57
the ignorance of the counted. They reveal the
16:00
mismatch between administrative categories and
16:03
lived identity. Religion presented similar difficulties.
16:08
The census required exclusive categories. Hindu,
16:13
Muslim, Christian, Sikh, Buddhist, Jain, Tribal,
16:17
and eventually others. But religious practice
16:20
in many parts of India did not fit these boundaries.
16:24
People might worship at shrines associated with
16:27
multiple traditions. They might follow practices
16:30
that scholars later labeled syncretic. They might
16:34
simply not experience their devotional life in
16:37
terms of exclusive membership. The census forced
16:40
a choice. And because the categories had political
16:44
consequences, the choice was never merely descriptive.
16:47
The categories affected separate electorates,
16:50
representation, and access to certain legal jurisdictions.
16:55
Language was equally problematic. The census
16:59
asked for mother tongue, but what counted as
17:02
a language versus a dialect. Hindi and Urdu,
17:06
for instance, were linguistically very close
17:09
but politically charged. Their treatment in the
17:13
census became a matter of intense contestation.
17:16
The 1901 census counted them separately in some
17:20
provinces and together in others. This depended
17:24
on local political pressure and the Commissioner's
17:26
judgement. Regional languages with continuous
17:30
dialect variation had to be mapped onto territorial
17:33
units. This created boundaries where speakers
17:37
experienced none. A villager in the Hindi -Urdu
17:41
belt might speak a local dialect that shared
17:44
features with both, yet the census required a
17:47
single answer, a single box, a single classification.
17:52
That classification would appear in the tables
17:54
and eventually influence educational policy and
17:57
state boundaries. The census commissioners knew
18:01
these problems. Their reports are remarkably
18:04
frank about the arbitrariness of many decisions.
18:07
They are frank about the impossibility of consistent
18:11
classification. They are frank about the gap
18:14
between official categories and local realities.
18:18
They revised categories between censuses. Sometimes
18:22
they merged groups that had previously been separated.
18:26
Sometimes they created new categories in response
18:29
to lobbying or apparent confusion. Each revision
18:33
was an admission that the previous census had
18:36
not simply counted what existed. It had imposed
18:39
a framework that now needed adjustment. But each
18:43
revision also created new realities, new incentives,
18:47
new ways for people to understand and present
18:50
themselves. By the early 20th century, the census
18:55
had become something more than an administrative
18:57
exercise. It had become a site of political contestation,
19:02
and Indian communities had learned to play the
19:05
game. The Nadars of southern India provide one
19:09
instructive example. In the late 19th and early
19:13
20th centuries, Nadar caste associations organized
19:16
actively to secure higher census classification.
19:20
Nadar elites had historically engaged in Palmyra
19:24
cultivation and toddy tapping. Now they sought
19:28
recognition as closer to kshatriya or other high
19:32
status vana categories. They printed pamphlets
19:36
instructing members how to answer census questions.
19:39
These pamphlets survive in the historical record.
19:43
Members were told to emphasize connections to
19:46
ancient ruling lineages. They were told to describe
19:49
their traditional occupation as agricultural
19:52
rather than toddy tapping. They were told to
19:56
present a unified community identity that masked
19:59
the internal diversity between wealthy Nedar
20:01
merchants and poor Palmyra climbers. The Medar
20:06
Mahajana Sangam, founded in 1910, coordinated
20:10
this campaign. It collected subscriptions to
20:14
fund representations to census authorities. It
20:18
distributed printed instructions in Tamil. The
20:21
census category became a target of collective
20:23
action, not a neutral description. The Kayasthas,
20:28
a scribal community spread across northern and
20:31
eastern India, similarly organized through the
20:34
Kayastha Conference. They sought to secure favorable
20:37
census classification and to resist attempts
20:40
to subdivide or downgrade them. Their petitions
20:44
to census commissioners argued for recognition
20:46
as a single. high -status community with ancient
20:49
origins. They contested official proposals to
20:53
split them into regional subgroups with different
20:56
rankings. The Jats, the Marathas and numerous
21:01
other groups followed similar paths. Caste associations
21:05
increasingly understood that census numbers and
21:08
classifications translated into social prestige,
21:11
political representation and eventually government
21:14
employment and educational access. Muslim political
21:18
leaders, too, engaged strategically with the
21:22
census. The growing concern with Muslim demographic
21:26
numbers shaped political mobilization. Leaders
21:29
tracked Muslim numbers relative to Hindus, within
21:32
particular provinces and in relation to government
21:35
employment. This fed directly into demands for
21:39
separate electorates and eventually the partition
21:41
of India. The 1909 introduction of separate electorates
21:46
for Muslims made census numbers directly consequential
21:49
for political power. Communities that could demonstrate
21:53
sufficient numbers could claim separate representation.
21:56
The census was no longer merely descriptive.
21:59
It was constitutive of political possibility.
22:03
This was the feedback loop that transformed colonial
22:07
governance. The state created categories for
22:10
administrative convenience. People recognized
22:14
the power of those categories and organized to
22:17
influence them. The state adjusted categories
22:20
in response to pressure or doubled down on them
22:23
or created new ones. Each iteration hardened
22:27
certain identities. Each iteration made certain
22:31
claims possible and foreclosed others. The census
22:36
did not invent caste, religion or language as
22:39
forms of social organization. but it transformed
22:43
them into countable, all -India categories with
22:46
standardized meanings and political weight. The
22:50
enumerators themselves were crucial intermediaries
22:53
in this process. They are often the most invisible
22:57
actors in the story. They were recruited locally,
23:00
sometimes from the same communities they counted,
23:03
sometimes from outsider clerical classes. They
23:07
carried the forms, asked the questions, and made
23:10
countless small decisions about how to record
23:13
ambiguous answers. The 1901 instruction manual
23:17
told enumerators to record cast as stated, but
23:22
it also told them to use their judgment when
23:24
answers were unclear. An enumerator who knew
23:28
the local situation might record what he thought
23:31
was really meant. One who did not might force
23:34
an answer into the nearest available box. A respondent
23:38
might answer differently to an enumerator from
23:41
their own community than to an outsider. They
23:44
might answer differently depending on whether
23:47
they perceived the census as connected to taxation,
23:50
conscription, or some other state purpose. The
23:54
physical act of enumeration was the moment when
23:57
abstract categories became concrete entries.
24:00
The enumerator stood in the doorway, pencil in
24:04
hand, scanning the columns, deciding where to
24:07
place this particular household. This was the
24:10
moment when lived identity was translated into
24:13
official data. Some people resisted outright.
24:18
There are reports of communities refusing to
24:21
answer. There are reports of villages that could
24:24
not be reached. There are reports of whole regions
24:28
where enumeration remained incomplete. Resistance
24:31
took quieter forms too. Deliberate misdirection.
24:37
Strategic ambiguity, the simple refusal to fit
24:41
one's complex social reality into someone else's
24:44
boxes. These resistances are harder to document
24:49
than organized lobbying, precisely because they
24:51
often left no paper trail. But the Census Commissioner's
24:55
complaints about evasive answers and incomplete
24:58
returns testify to their reality. By the 1931
25:04
Census, the last full enumeration before independence
25:07
the political landscape had been profoundly shaped
25:11
by decades of census operations communal categories
25:15
had become entrenched in political representation
25:18
caste associations had become normalized features
25:22
of public life language claims had become central
25:26
to regional politics The census had helped teach
25:30
Indians to see themselves in categories that
25:33
would have been unfamiliar or unimportant to
25:35
their grandparents. It had helped teach them
25:39
to organize politically through those categories.
25:41
The consequences extended into independence and
25:45
partition. When British India was divided in
25:48
1947, the boundaries drew on decades of census
25:52
data about religious demography. The Radcliffe
25:56
Commission charged with drawing the boundary
25:58
between India and Pakistan in the Punjab and
26:01
Bengal, faced a mapmaker's problem. Districts
26:05
had mixed populations. No clean line was possible.
26:10
Census data provided the numbers that made certain
26:13
boundary choices more defensible than others.
26:16
In the Punjab, the district of Gurdaspur presented
26:19
a particular difficulty. Muslim majorities lived
26:23
in some tessels. Hindu and Sikh majorities lived
26:28
in others. Populations were so intermingled that
26:31
any line would leave minorities on both sides.
26:35
The census tables showed percentages, ratios,
26:38
and demographic profiles. These became arguments
26:42
for drawing the line here rather than there.
26:45
The numbers did not determine the boundary. Political
26:49
negotiation, military pressure, and administrative
26:52
convenience all played their parts. but the numbers
26:56
provided the vocabulary in which the boundary
26:59
could be justified. They provided the apparent
27:02
objectivity that made a particular division seem
27:05
rational. The census table became a border. The
27:09
linguistic reorganization of Indian states in
27:12
the 1950s used census language data as a key
27:16
input. Postcolonial India continued and modified
27:20
colonial census categories. It added new ones.
27:24
notably scheduled castes and scheduled tribes
27:28
for affirmative action purposes. These carried
27:31
forward the logic of classification, even as
27:35
they redirected it toward different political
27:37
ends. The census remained a mold, not a mirror.
27:42
What should we make of this history? Several
27:45
important distinctions help prevent misunderstanding.
27:49
First, to say that the census helped create and
27:52
harden categories is not to say that these categories
27:55
were pure fictions. It is not to say that people
27:59
had no collective identities before enumeration.
28:02
Cast, language and religious community mattered
28:06
enormously in pre -colonial India. But they mattered
28:10
differently. They mattered more locally, more
28:13
flexibly, more relationally. The census did not
28:18
invent them. It transformed them into standardized,
28:22
all -India, politically consequential units.
28:26
Second, to emphasize Indian agency in engaging
28:29
with the census is not to imply equality of power
28:32
between colonizer and colonized. The British
28:36
colonial state had vastly greater resources,
28:39
coercive capacity, and control over the final
28:42
categories. Indians operated within constraints
28:45
they did not choose. But they were not merely
28:49
passive objects. They recognized the system's
28:52
logic and worked within it. Sometimes they subverted
28:56
it. Sometimes they turned it to their own purposes.
29:01
Third, the census was never the only force shaping
29:04
identity. Law courts decided cases about caste
29:08
status and religious boundaries. Army recruitment
29:12
practices channeled particular communities into
29:15
particular regiments. Educational institutions
29:18
taught in certain languages and not others. The
29:22
census interacted with all of these. Sometimes
29:26
it reinforced them. Sometimes it complicated
29:30
them. It never operated alone. Fourth, we should
29:35
be careful not to treat the outcomes as inevitable.
29:38
The census did not make partition inevitable.
29:41
It did not make communal politics inevitable.
29:46
It created resources and constraints within which
29:48
people made choices, often terrible choices,
29:52
under conditions they did not fully control.
29:55
Historical contingency matters. Things could
29:59
have developed differently, even if the census
30:01
made certain paths easier and others harder.
30:05
What remains uncertain in this story? Quite a
30:09
lot. The precise mechanisms by which census categories
30:13
influenced everyday self -understanding are difficult
30:17
to document we have petitions newspaper commentary
30:21
and caste association records but the inner experience
30:25
of ordinary people encountering the census especially
30:28
illiterate villages far from political centers
30:30
is largely unrecoverable we know that strategic
30:35
engagement happened at elite levels how far it
30:39
penetrated How much the census actually changed
30:42
how a poor farmer in Bihar or a weaver in Gujarat
30:46
understood himself, these questions resist confident
30:49
answer. The motivations of individual enumerators
30:53
and respondents are similarly opaque. The archives
30:57
preserve official complaints and occasional glimpses
31:00
of local practice. They do not preserve the thousands
31:04
of small negotiations at village doorways that
31:08
made the system function. or fail to function.
31:11
We are left with structural patterns and occasional
31:15
illuminating fragments. The relationship between
31:19
census categories and later political outcomes,
31:22
particularly partition, is complex and contested
31:26
by scholars. The census was one factor among
31:29
many. Colonial policy, economic change, political
31:34
mobilization, communal violence, world war and
31:38
its aftermath. all played their parts. Isolating
31:42
the census's specific contribution requires careful
31:45
qualification. It enabled certain claims and
31:49
made certain calculations possible. It did not
31:53
determine what would happen. Some scholars emphasize
31:57
continuity between pre -colonial and colonial
32:00
forms of enumeration. They argue that the British
32:03
census was less novel than often claimed. Indian
32:07
states had long -conducted surveys. tax registers
32:11
and population counts. There is truth in this.
32:15
But the scale, standardization and political
32:19
integration of the modern census were genuinely
32:21
new. Others emphasize Indian agency more strongly.
32:26
They see the census as a site of negotiation
32:29
rather than imposition. This too has merit. But
32:34
it must not obscure the fundamental asymmetry
32:37
of colonial power. Still others question how
32:41
much the senses actually mattered compared to
32:43
other institutions. They suggest we not overstate
32:46
its influence. These debates are healthy. They
32:51
keep us from oversimplification. Let me return
32:56
to where we began. The printed form, the boxes,
33:00
the labels, the enumerator standing at the doorway,
33:04
pencil in hand, translating a life into a category.
33:09
We still live with this technology. Every census
33:13
cycle, debates erupt about which categories to
33:16
include. Debates erupt about how to ask about
33:20
ethnicity, or religion, or gender. Debates erupt
33:25
about what to do with people who do not fit the
33:28
available boxes. The specific categories have
33:31
changed. The underlying dynamic has not. The
33:36
census remains an act of translation. simplification
33:39
and standardization it remains a moment when
33:43
the state asks us to fit our complex selves into
33:47
its framework it remains a moment when we decide
33:50
whether to comply resist or strategically adapt
33:55
the indian colonial census teaches us to look
33:59
behind the tables to ask who designed the boxes
34:03
who trained the questioners who translated the
34:06
terms who checked the answers who published the
34:09
results, and who used them for what purposes.
34:12
To recognize that the numbers that seemed so
34:15
objective, so solid, so merely descriptive, emerged
34:19
from a process full of human judgment, human
34:22
confusion, human negotiation, and human power.
34:26
To understand that when we are counted, we are
34:30
also being shaped. And that we, in turn, can
34:33
shape the counting. The census was not a mirror
34:37
held up to Indian society. It was a mold that
34:41
helped to make it. It made populations legible
34:45
to a state that needed to rule them. And in doing
34:49
so, it taught people to see themselves through
34:52
categories that would outlast the empire that
34:55
created them. The statistics became an active
34:58
force in nation building. They influenced who
35:02
could claim representation. They influenced which
35:05
languages would be taught. They influenced where
35:09
borders would run. They influenced how people
35:12
would understand their place in the social order.
35:16
The modern census does not simply count what
35:19
exists. It helps constitute what exists. And
35:24
the people being counted have never been merely
35:26
passive in the face of that power. They have
35:29
negotiated, resisted, adapted. and sometimes
35:33
turn the state's own categories against it. This
35:37
is the real story of enumeration, not a simple
35:41
counting but a complex making in which the counted
35:44
and the counters are bound together in the production
35:47
of social reality.