Archival Science 2: 87–109, 2002.

87
© 2002 Kluwer Academic Publishers. Printed in the Netherlands.

Colonial Archives and the Arts of Governance

ANN LAURA STOLER

Department of Anthropology, University of Michigan, Ann Arbor MI 48109-1382 USA
(E-mail: astoler@umich.edu)

Abstract. Anthropologists engaged in post-colonial studies are increasingly adopting an

historical perspective and using archives. Yet their archival activity tends to remain more an
extractive than an ethnographic one. Documents are thus still invoked piecemeal to confirm
the colonial invention of certain practices or to underscore cultural claims, silent. Yet such
mining of the content of government commissions, reports, and other archival sources rarely
pays attention to their peculiar placement and form. Scholars need to move from archive-as-
source to archive-as-subject. This article, using document production in the Dutch East Indies
as an illustration, argues that scholars should view archives not as sites of knowledge retrieval,
but of knowledge production, as monuments of states as well as sites of state ethnography.
This requires a sustained engagement with archives as cultural agents of “fact” production, of
taxonomies in the making, and of state authority. What constitutes the archive, what form it
takes, and what systems of classification and epistemology signal at specific times are (and
reflect) critical features of colonial politics and state power. The archive was the supreme
technology of the late nineteenth-century imperial state, a repository of codified beliefs that
clustered (and bore witness to) connections between secrecy, the law, and power.

Keywords: archives, archiving, bureaucracy, colonial archives, ethnography, knowledge

Genealogy is gray, meticulous and patiently documentary. It operates on a

field of entangled and confused parchments, on documents that have been
scratched over and recopied many times.1

This essay is about the colonial order of things as seen through its archival
productions. It asks what insights about the colonial might be gained from
attending not only to colonialism’s archival content, but to its particular and
sometimes peculiar form. Its focus is on archiving as a process rather than to
archives as things. It looks to archives as epistemological experiments rather
than as sources, to colonial archives as cross-sections of contested knowl-
edge. Most importantly, it looks to colonial archives as both transparencies
on which power relations were inscribed and intricate technologies of rule
in themselves. Its concerns are two: to situate new approaches to colonial
1 Michel Foucault, “Nietzsche, Genealogy, History”, in Daniel Bouchard (ed.), Language,
Counter-Memory, Practice: Selected Essays and Interviews by Michel Foucault (Ithaca:
Cornell University Press [1971] 1977), p. 139.
88 ANN LAURA STOLER

archives within the broader “historic turn” of the last two decades and to
suggest what critical histories of the colonial have to gain by turning further
toward a politics of knowledge that reckons with archival genres, cultures of
documentation, fictions of access, and archival conventions.2

Epistemological scepticism, archives, and the “historic turn”

Some four decades after British social anthropologist E.E. Evans-Pritchard’s

unheeded warning that anthropology would have to choose between being
history or being nothing, and Claude Levi-Strauss’ counter claim that
accorded history neither “special value” nor privileged analytic space,
students of culture have taken up a transformative venture, celebrating with
unprecedented relish what has come to be called “the historic turn.”3 Some
might argue that anthropology’s engagement with history over the last two
decades, unlike that recent “turn” in other disciplines, has not been a turn
at all, but rather a return to its founding principles: enquiry into cumu-
lative processes of cultural production. but without the typological aspirations
and evolutionary assumptions once embraced. Others might counter that the
feverish turn to history represents a significant departure from an earlier
venture, a more explicit rupture with anthropology’s long-standing complicity
in colonial politics.4 As such, one could argue that the historic turn signals
not a turn to history per se, but a different reflection on the politics of knowl-
edge – a further rejection of the categories and cultural distinctions on which
2 On the “historic turn,” see the introduction to Terrence J. McDonald (ed.), The Historic
Turn in the Human Sciences (Ann Arbor: University of Michigan Press, 1966). This essay
represents a condensed version of Chapter 1 from my book in progress, Along the Archival
Grain (Princeton: Princeton University Press). Parts of it are based on the 1996 Lewis
Henry Morgan Lectures delivered at the University of Rochester entitled “Ethnography in
the Archives: Movements on the Historic Turn.” A different version of this piece appears in
Carolyn Hamilton (ed.), Refiguring the Archive (forthcoming).
3 E.E. Evans-Pritchard, “Social Anthropology: Past and Present, The Marett Lecture,
1950”, Social Anthropology and Others Essays (New York: Free Press, 1951), p. 152. Claude
Levi-Strauss, The Savage Mind (Chicago: Chicago University Press, 1966), p. 256.
4 For some sense of the range of different agendas of the current “historic turn,” see Nich-
olas B. Dirks, Geoff Eley, and Sherry B. Ortner (eds.), Culture, Power, History: A Reader in
Contemporary Social Theory (Princeton: Princeton University Press, [1983] 1994), Terrence
J. McDonald (ed.), The Historic Turn in the Human Sciences (Ann Arbor: University of
Michigan Press, 1996); specifically on history in the anthropological imagination, see Gerald
Sider and Gavin Smith (eds.), Between History and Histories: The Making of Silences and
Commemorations (Toronto: Toronto University Press, 1997). Also see Richard Fox’s “For a
Nearly New Culture History”, in Richard G. Fox (ed.), Recapturing Anthropology: Working
in the Present (Santa Fe: School of American Research Press, 1991), pp. 93–114, and James
Faubion, “History in Anthropology”, Annual Review of Anthropology 22 (1993): 35–54.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 89

imperial rule was once invested and on which post-colonial state practices
have continued to be based.
Engagement with the uses and abuses of the past pervades many academic
disciplines, but nowhere more than in this burgeoning area of colonial
ethnography. Over the last decade, students of the colonial have challenged
the categories, conceptual frame, and practices of colonial authorities and
their taxonomic states.5 Questioning the making of colonial knowledge, and
the privileged social categories it produced, has revamped what students of
the colonial take to be sources of knowledge and what to expect of them.
Attention to the intimate domains in which colonial states intervened has
prompted reconsideration of what we hold to be the foundations of European
authority and its key technologies.6 In treating colonialism as a living history
that informs and shapes the present rather than as a finished past, a new
generation of scholars are taking up Michel De Certeau’s invitation to
“prowl” new terrain as they re-imagine what sorts of situated knowledge
have produced both colonial sources and their own respective locations in
the “historiographic operation.”7 Some students of colonialism are rereading
those archives and doing oral histories with people who lived those archived
events to comment on colonial narratives of them.8 Others are doing so
with photography, engravings, and documentary art.9 Some are attending to
how colonial documents have been requisitioned and recycled to confirm old
entitlements or to make new political demands. As part of a wider impulse,
we are no longer studying things, but the making of them. Students of coloni-
alisms in and outside of anthropology are spending as much time rethinking
what constitutes the colonial archive as they are reconsidering how written

5 See, for example, the introductions to and essays in Nicholas Dirks (ed.), Colonialism and
Culture (Ann Arbor: University of Michigan Press, 1992); and in Frederick Cooper and Ann
Laura Stoler (eds.), Tensions of Empire: Colonial Cultures in a Bourgeois World (Berkeley:
University of California Press, 1997).
6 See “Genealogies of the Intimate”, in Ann Laura Stoler and Frederick Cooper (eds.),
Carnal Knowledge and Imperial Power: Race and the Intimate in Colonial Rule (Berkeley:
University of California Press, 2002).
7 See Michel de Certeau, “The Historiographic Operation” (1974), in The Writing of
History (New York: Columbia University Press, 1988).
8 On archives in relationship to popular memory, see Richard Price, Convict and the
Colonel: A Story of Colonialism and Resistance in the Caribbean (Boston: Beacon Press,
1998); Luise White, Speaking with Vampires: Rumor and History in Colonial Africa
(Berkeley: University of California Press, 2000); Ann Laura Stoler and Karen Strassler, “Cast-
ings for the Colonial: Memory Work in ‘New Order’ Java”, Comparative Studies in Society
and History 42(1) (2000): 4–48, and the references therein.
9 On the power of images in the making of colonial rule, see Elizabeth Edwards, guest
editor, “Anthropology and Colonial Endeavour”, in The History of Photography 21(1) (Spring
1997).
90 ANN LAURA STOLER

documents collide and converge with colonial memories in the post-colonial

field.
If Evans-Pritchards’ warning some thirty-five years ago that “anthropolo-
gists have tended to be uncritical in their use of documentary sources” had
little resonance at the time, it certainly has more today. For however deep and
full the archival turn has been in post-colonial scholarship of the 1990s, what
is more surprising is how thin and tentative it can still remain.10 Anthropolo-
gists may no longer look at archives as the stuff of another discipline. Nor are
these archives treated as inert sites of storage and conservation.11 But archival
labour tends to remain more an extractive enterprise than an ethnographic
one. Documents are still invoked piecemeal and selectively to confirm the
colonial invention of traditional practices or to underscore cultural claims.
Anthropology has never committed itself to “exhaust” the sources, as
Bernard Cohn once chided the historical profession for doing with such moral
fervor. But the extractive metaphor remains relevant to both.12 Students of
the colonial experience “mine” the content of government commissions and
reports, but rarely attend to their peculiar form or context. We look at exem-
plary documents rather than at the sociology of copies, or what claims to truth
are lodged in the rote and redundant. We warily quote examples of colonial
excesses – if uneasy with the pathos and voyeurism that such citations entail.
We may readily mock fetishisms of the historian’s craft, but there remains the
shared conviction that access to what is “classified” and “confidential” are
the coveted findings of sound and shrewd intellectual labours.13 The ability
to procure them measures scholarly worth. Not least is the shared conviction
that such guarded treasures are the sites where the secrets of the colonial state
are really stored.
There are a number of ways to frame the sort of challenge I have in mind,
but at least one seems obvious: steeped as students of culture have been in
treating ethnographies as texts, we are just now critically reflecting on the
making of documents and how we choose to use them, on archives not as
sites of knowledge retrieval but of knowledge production, as monuments of
states as well as sites of state ethnography. This is not a rejection of colonial

10 E.E. Evans-Pritchard, Anthropology and History (Manchester: Manchester University

Press, 1961), p. 5.
11 See Carlo Ginzburg, Clues, Myths, and the Historical Method (Baltimore: Johns Hopkins
University, 1989).
12 Bernard Cohn, “History and Anthropology: The State of Play”, Comparative Studies in
Society and History 22(2) (1980): 198–221.
13 On the trips to archives as “feats of [male] prowess” in nineteenth-century middle-class
culture, see Bonnie G. Smith, “Gender and the Practices of Scientific History: The Seminar
and Archival Research in the Nineteenth-Century”, American Historical Review 100(4–5)
(1995): 1150–1176.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 91

archives as sources of the past. Rather, it signals a more sustained engagement

with those archives as cultural artifacts of fact production, of taxonomies in
the making, and of disparate notions of what made up colonial authority.
As both Ranajit Guha and Greg Dening long have warned, “sources” are
not “springs of real meaning,” “fonts” of colonial truths in themselves.14
Whether documents are trustworthy, authentic, and reliable remain pressing
questions, but a turn to the social and political conditions that produced those
documents, what Carlo Ginzburg has called their “evidentiary paradigms,”
has altered the sense of what trust and reliability might signal and politically
entail. The task is less to distinguish fiction from fact than to track the produc-
tion and consumption of those “facts” themselves. With this move, colonial
studies is steering in a different direction, toward enquiry into the grids of
intelligibility that produced those “evidential paradigms” at a particular time,
for a particular social contingent, and in a particular way.15
Students of the colonial have come to see appropriations of colonial
history as infused with political agendas, making some stories eligible for
historical rehearsal and others not.16 Troubling questions about how personal
memories are shaped and effaced by states too has placed analytic emphasis
on how past practices are winnowed for future uses and future projects.17
Such queries invite a turn back to documentation itself, to the “teaching” task
that the Latin root “docere” implies, to what and who were being educated in
the bureaucratic shuffle of rote formulas, generic plots, and prescriptive asides
that make up the bulk of a colonial archive. The issue of official “bias” gives
way to a different challenge: to identifying the conditions of possibility that
shaped what could be written, what warranted repetition, what competencies
were rewarded in archival writing, what stories could not be told, and what
could not be said. Andrew Ashforth may have overstated the case in his study
of South Africa’s Native Affairs Commission, when he noted that “the real
seat of power” in modern states is “the bureau, the locus of writing,” but he

14 Ranajit Guha, “The Proses of Counter-Insurgency”, in Nicholas B. Dirks, Geoff Eley, and
Sherry B. Ortner (eds.), Culture, Power, History: A Reader in Contemporary Social Theory
(Princeton: Princeton University Press, [1983] 1994), pp. 336–371. Greg Dening, The Death
of William Gooch: A History’s Anthropology (Honolulu: Hawaii University Press, 1995), p. 54.
15 Carlo Ginzburg, “Clues: Roots of an Evidential Paradigm”, in Clues, Myths and the
Historical Method Baltimore: Johns Hopkins University Press, 1989), pp. 96–125.
16 David William Cohen, Burying SM: The Politics of Knowledge and the Sociology of
Power in Africa (Portsmouth, NH: Heineman, 1992).
17 Joanne Rappaport, Cumbe Reborn: An Andean Ethnography of History (Chicago:
University of Chicago Press, 1994). Also see the contributions to Sarah Nuttall and Carli
Coetzee (eds.), Negotiating the Past: The Making of Memory in South Africa (Cape Town:
Oxford University Press, 1998).
92 ANN LAURA STOLER

may not have been far off the mark.18 That every document comes layered
with the received account of earlier events and the cultural semantics of a
political moment makes one point clear. What constitutes the archive, what
form it takes, and what systems of classification signal at specific times are
the very substance of colonial politics.

From extraction to ethnography in the colonial archives

The transformation of archival activity is the point of departure and the

condition of a new history.19

If one could say that archives were once treated as a means to an end by
students of history, this is no longer the case today. The pleasures of “a
well-stocked manuscript room with its ease of access and aura of quiet
detachment” is a thing of the past.20 Over the last decade, epistemological
scepticism has taken cultural and historical studies by storm. A focus on
history as narrative, and on history-writing as a charged political act, has
made the thinking about archives no longer the pedestrian preoccupation of
“spade-work” historians or flat-footed archivists, nor the entry requirements
of fledgling initiates compelled to show mastery of the tools of their trade.
The “archive” has been elevated to new theoretical status, with enough cachet
to warrant distinct billing, worthy of scrutiny on its own. Jacques Derrida’s
Archive Fever compellingly captured that impulse by giving it a name and by
providing an explicit and evocative vocabulary for its legitimation in crit-
ical theory.21 But Natalie Zemon Davis’ Fiction in the Archives, Roberto
Ecchevaria’s Myth and Archive, Thomas Richards’ Imperial Archive, and
Sonia Coombe’s Archives Interdites, to name but a few, suggest that Derrida’s
splash came only after the archival turn had already been made.22
18 See Andrew Ashforth, The Politics of Official Discourse in Twentieth-Century South
Africa (Oxford: Clarendon Press, 1990), p. 5.
19 De Certeau (1988 [1974]), p. 75.
20 A phrase used by Jane Sherron De Hart to underscore the “problematics of evidence”
in contemporary historical reconstruction: see “Oral Sources and Contemporary History:
Dispelling Old Assumptions”, Journal of American History (September 1993), p. 582.
21 Jacques Derrida, Archive Fever: A Freudian Impression (Chicago: Chicago University
Press, 1995).
22 Natalie Zemon Davis, Fiction in the Archives: Pardon Tales and Their Tellers in
Sixteenth-Century France (Stanford: Stanford University Press, 1987); Thomas Richards, The
Imperial Archive: Knowledge and the Fantasy of Empire (London: Verso, 1993); Roberto
Gonzalez Echevarria, Myth and Archive: A Theory of Latin American Narrative (Cambridge:
Cambridge University Press, 1990); Sonia Coombe, Archives Interdites: Les peurs françaises
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 93

This move from archive-as-source to archive-as-subject gains its contem-

porary currency from a range of different analytic shifts, practical concerns,
and political projects. For some, as in the nuanced archival forays of Greg
Dening, it represents a turn back to the meticulous “poetics of detail.”23 To
others, like Michel-Rolph Trouillot, in his treatment of the archival silences
of the Haitian Revolution, and David William Cohen, in his “combings of
history,” it signals a new grappling with the production of history, what
accounts get authorized, what procedures were required, and what about the
past it is possible to know.24 For Bonnie Smith, research in archives, like the
university seminar, were the nineteenth-century sites where historical science
was marked with gendered credentials.25 Archivists obviously have also been
thinking about the nature and history of archives for sometime.26 What marks
this moment are the profusion of forums in which historians are joining
archivists in new conversations about documentary evidence, record keeping,
and archival theory.27 Both are worrying about the politics of storage, what

face à l’Histoire contemporaine (Paris: Albin Michel, 1994). See also Dominick LaCapra,
“History, Language, and Reading”, American Historical Review 100.3 (June 1995): 807,
where he also notes that the “problem of reading in the archives has increasingly become
a concern of those doing archival research.”
23 See, for example, Greg Dening, The Death of William Gooch: A History’s Anthropology
(Honolulu: University of Hawaii Press, 1995).
24 Michel-Rolph Trouillot, Silencing the Past: Power and the Production of History
(Boston: Beacon Press, 1995); The Combing of History (Chicago: Chicago University Press,
1994).
25 Bonnie G. Smith, “Gender and the Practices of Scientific History”, American Historical
Review 100(4–5) (1995): 1150–1176.
26 On the history of archives and how archivists have thought about it, see Ernst Posner’s
classic essay, “Some Aspects of Archival Development since the French Revolution”, in
Maygene Daniels and Timothy Walch (eds.), A Modern Archives Reader (Washington, D.C.:
National Archives and Record Service, [1940] 1984), pp. 3–21; Michel Duchein, “The History
of European Archives and the Development of the Archival Profession in Europe”, American
Archivist 55 (Winter 1992): 14–25; and Terry Cook, “What is Past is Prologue: A History
of Archival Ideas Since 1898, and the Future Paradigm Shift”, Archivaria 43 (Spring 1997):
17–63.
27 See, for example, Richard Berner, Archival Theory and Practice in the United States: An
Historical Analysis (Seattle: University of Washington Press, 1983); Kenneth E. Foote, “To
Remember and Forget: Archives, Memory, and Culture”, American Archivist 53(3) (1990):
378–393; Terry Cook, “Mind over Matter: Towards a New Theory of Archival Appraisal”, in
Barbara Craig (ed.), The Archival Imagination: Essays in Honour of Hugh A. Taylor (Ottawa:
Association of Canadian Archivists, 1992), pp. 38–69; James M. O’Toole, “On the Idea of
Uniqueness”, American Archivist 57(4) (1994): 632–659. For some sense of the changes in
how archivists themselves have framed their work over the last fifteen years, see many of the
articles in The American Archivist and Archivaria.
94 ANN LAURA STOLER

information matters, and what should be retained in the archive as paper

collections give way to digital forms.28
In cultural theory, “the archive” has a capital “A,” is figurative, and
leads elsewhere. It may represent neither material site nor a set of docu-
ments. Rather, it may serve as a strong metaphor for any corpus of selective
forgettings and collections – and, as importantly, for the seductions and long-
ings that such quests for, and accumulations of, the primary, originary, and
untouched entail.29 For those inspired more directly by Foucault’s Archae-
ology of Knowledge, the archive is not an institution, but “the law of what can
be said,” not a library of events, but “that system that establishes statements
as events and things,” that “system of their enunciabilities.”30
From whichever vantage point – and there are more than these – the
“archival turn” registers a rethinking of the materiality and imaginary of
collections and what kinds of truth-claims lie in documentation.31 Such a
“archival turn” converges with a profusion of new work in the history of
science, that is neither figuratively or literally about archives at all. I think
here of Ian Hacking’s studies of the political history of probability theory
and state investments in the “taming of chance;” Steven Shapin’s analysis of
the social history of scientific truths where he traces the power to predict as
one enjoyed by, and reserved for, cultured and reliable men; Mary Poovey’s
work on how the notion of the “modern fact” was historically produced;
Alain Desrosières study (among many others) on statistics as a science of the
state and Silvana Patriarca’s on statistics as a modern mode of representation;
Lorraine Daston’s analysis of the development of classical probability theory
as a means of measuring the incertitudes of a modernizing world.32 One could

28 Terry Cook, “Electronic Records, Paper Minds: The Revolution in Information Manage-
ment and Archives in the Post-Custodial and Post-Modernist Era”, Archives and Manuscripts
22(2) (1994): 300–329.
29 This metaphoric move is most evident in contributions to the two special issues of History
of the Human Sciences devoted to “The Archive”, 11(4) (November 1998) and 12(2) (May
1999). Derrida’s valorization of “the archive” as imaginary and metaphor predominates both.
On the archive as metaphor, also see Allan Sekula. “The Body and the Archive”, October 39
(Winter 1986): 3–64.
30 Michel Foucault, “The Statement and the Archive”, The Archaeology of Knowledge and
the Discourse on Language, especially Part III (1972), pp. 79–134.
31 See, for example, Patrick Geary, Phantoms of Remembrance: Memory and Oblivion a
the End of the First Millennium (Princeton: Princeton University Press, 1994), especially
“Archival Memory and the Destruction of the Past”, pp. 81–114.
32 Ian Hacking, The Taming of Chance (New York: Cambridge University Press, 1990);
Steven Shapin, A Social History of Truth: Civility and Science in Seventeenth-Century
England (Chicago: Chicago University Press, 1994); Mary Poovey, A History of the Modern
Fact: Problems of Knowledge in the Sciences of Wealth and Society (Chicago: Chicago
University Press, 1998); Alain Desrosières, The Politics of Large Numbers: A History of Statis-
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 95

also add Anthony Grafton’s essays on footnotes as the lines that lead into
moral communities and their claims to authority and truth.33
What do these studies all have in common? All are concerned with the
legitimating social coordinates of epistemologies: how people imagine they
know what they know and what institutions validate that knowledge, and how
they do so. None treat the conventions and categories of analysis (statistics,
facts, truths, probability, footnotes, and so on) as innocuous or benign. All
converge on questions about rules of reliability and trust, criteria of credence,
and what moral projects and political predictabilities are served by these
conventions and categories. All ask a similar set of historical questions about
accredited knowledge and power – what political forces, social cues, and
moral virtues produce qualified knowledges that, in turn, disqualified other
ways of knowing, other knowledges. To my mind, no one set of concerns is
more relevant to the colonial politics of archives and their (parent) archiving
states.
But the archival turn can be traced through other venues as well,
suggesting that something resembling ethnography in an archival mode has
been around for sometime. Carlo Ginzburg’s micro-history of a sixteenth-
century miller, like Natalie Davis’ use of pardon tales in Fiction in the
Archives, drew on “hostile” documents of the elites to reveal “the gap between
the image underlying the interrogations of judges and the actual testimony of
the accused.”34 Neither were intended as ethnographies of the archive, but
both gesture in that direction. In Davis’ explicit attention to “how people
told stories, what they thought a good story was, how they accounted for
motive,” these sixteenth-century letters of remission are shown to recount
more than the bare facts of their peasant authors’ sober tales.35 Pardon tales
also registered the “constraints of the law,” the monopoly on public justice of
royal power, and the mercy that the monarchy increasingly claimed.36 Davis’
“fiction in the archives” demonstrated fashioned stories that spoke to moral

tical Reasoning (Cambridge: Harvard University Press, 1998); Silvana Patriarca, Numbers
and Nationhood: Writing Statistics in Nineteenth-Century Italy (Cambridge and New York:
Cambridge University Press, 1998). On the power of “suasive utterance” in the making of
scientific truth-claims, see Christopher Norris, “Truth, Science, and the Growth of Knowl-
edge”, New Left Review 210 (1995): 105–123; and Benedict Anderson, “Census, Map,
Museum”, in the revised second edition of Imagined Communities: Reflections on the Origin
and Spread of Nationalism (New York: Verso, 1991), pp. 163–186.
33 Anthony Grafton, The Footnote: A Curious History (Cambridge, MA: Harvard University
Press, 1997).
34 Carlo Ginzburg, The Cheese and the Worms: The Cosmos of a Sixteenth-Century Miller
(London: Penguin, 1982), pp. xvii, xviii.
35 Davis, 1987, p. 4.
36 Ibid.
96 ANN LAURA STOLER

truths, drew on shared metaphors and high literary culture, and depended on
the power of the state and the archived inscriptions of its authority.
While recent participants in the archival turn have been taken with
Derrida’s contention that “there is no political power without control of the
archive,” in fact, this insistence on the link between what counts as knowl-
edge and who has power has long been a founding principle of colonial
ethnography.37 Rolph Trouillot’s insistence in his study of the Haitian Revolu-
tion that “historical narratives are premised on previous understandings,
which are themselves premised on the distribution of archival power” allows
him to track the effacement of archival traces, and the imposed silences that
people have moved around and beyond.38 Nicholas Dirk’s observation that
early colonial historiographies in British India were dependent on native
informants, who were later written out of those histories, draws our attention
to the relationship between archiving, experts, and knowledge production.39
Christopher Bayly’s more recent attention to the ways in which the British
intelligence service in colonial India worked through native channels places
the state’s access to “information” as a nodal point in the art of governance
and as a highly contested terrain.40 My own work on “the hierarchies of
credibility” that contained colonial narratives in the Netherlands Indies, as
these constrained what were counted as having plausible plots, reads colonial
politics off the “storeyed” distributions of the state’s paper production and
through the rumors (spread by a beleaguered native population) that were
woven through it.41
As Foucault provocatively warned, the archive is neither the sum of all
texts that a culture preserves nor those institutions that allow for that record’s
preservation. The archive is rather that “system of statements,” those “rules of
practice,” that shape the specific regularities of what can and cannot be said.42
Students of colonialism have wrestled with this formulation to capture what
renders colonial archives as both documents of exclusions and as monuments
to particular configurations of power.

37 Derrida, 1995, p. 4.
38 Michel-Rolph Trouillot, 1995, p. 55.
39 Nicholas B. Dirks, “Colonial Histories and Native Informants: Biography of an Archive”,
in Carol A. Breckenridge and Peter van der Veer (eds.), Orientalism and the Postcolonial
Predicament: Perspectives on South Asia (Philadelphia: University of Pennsylvania Press,
1993), pp. 279–313.
40 Christopher Bayly, Empire and Information: Intelligence Gathering and Social
Communication in India, 1780–1870 (Cambridge: Cambridge University Press, 1996).
41 Ann Laura Stoler, “In Cold Blood: Hierarchies of Credibility and the Politics of Colonial
Narratives”, Representations 37 (1992): 151–189.
42 See Michel Foucault, “The Statement and the Archive”, The Archaeology of Knowledge,
pp. 79–134.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 97

Both Gonzalez Echevvaria and Thomas Richards follow Foucault in

treating the imperial archive as “the fantastic representation of an epistemo-
logical master pattern.”43 For Richards, that archive is material and figurative,
a metaphor of an unfulfilled but shared British imperial imagination. The
imperial archive was both the supreme technology of the late nineteenth-
century imperial state and the telling prototype of a postmodern one, predi-
cated on global domination of information and the circuits through which
“facts” move. Echevvaria locates the archive as both relic and ruin, a reposi-
tory of codified beliefs, genres for bearing witness, clustered connections
between secrecy, power, and the law.44 It was the legitimating discourses of
the Spanish colonial archives, he argues, that provided the Latin American
novel with its specific content and thematic form. For both Richards and
Echevvaria, the archive is a template that decodes something else. Both push
us to think differently about archival fictions, but reserve their fine-grained
analysis for literature, not the colonial archives themselves.45
Whether the “archive” should be treated as a set of discursive rules, an
utopian project, a depot of documents, a corpus of statements, or all of
the above, is not really the question. Colonial archives were both sites of
the imaginary and institutions that fashioned histories as they concealed,
revealed, and reproduced the power of the state.46 Power and control, as
many scholars have pointed out, is fundamental to the etymology of the
term.47 From the Latin archivum, “residence of the magistrate,” and from the
Greek arkhe, to command or govern, colonial archives ordered (in both the
imperative and taxonomic sense) the criteria of evidence, proof, testimony,
and witnessing to construct moral narrations. “Factual storytelling,” moral-
izing stories, and multiple versions – features that Hayden White ascribes
to what counts as history – make sense of which specific plots “worked” in
the colonial archives as well.48 It was in factual stories that the colonial state
affirmed its fictions to itself, in moralizing stories that it mapped the scope of
43 Richards, 1993, p. 11.
44 Echevvaria, 1990, p. 30.
45 Thus for Thomas Richards, Hilton’s Lost Horizon and Kipling’s Kim are entries in
a Victorian archive that was the “prototype for a global system of domination through
circulation, an apparatus for controlling territory by producing, distributing and consuming
information about it.”
46 This link between state power and what counts as history was long ago made by Hegel in
The Philosophy of History, as Hayden White points out: “It is only the state which first presents
subject-matter that is not only adapted to the prose of History, but involves the production of
such history in the very progress of its own being.” See Hayden White, The Content of the
Form: Narrative Discourse and Historical Representation (Baltimore: Johns Hopkins, 1987),
p. 12.
47 See Echevvaria (1990), p. 31, for a detailed etymology of the term.
48 See White, 1987, especially, pp. 26–57.
98 ANN LAURA STOLER

its philanthropic missions, and in multiple and contested versions that cultural
accounts were discredited or restored.
Viewed in this perspective, it is clear that the nineteenth- and early-
twentieth-century archives of the Dutch administration in the Indies were not
to be read randomly in any which way. Issues were rendered important by
how they were classed and discursively framed. Official exchanges between
the Governor General and his subordinates, between the Governor General
and the Minister of Colonies, and between the Minister and the King, were
reference guides to administrative thinking. Organized in folio forms, title
pages provided long lists of cross-referenced dossiers and decisions that
were abbreviated genealogies of what constituted relevance, precedent, and
“reasons of state.” With appended evidence that might include testimonies of
experts and commissioned reports, such folios contained and confirmed what
counted as proof and who cribbed from whom in the chain of command.
Attention to moments of distrust and dispersion, reversals of power, ruptures
in contract, have been the trade marks of critical political and social history
for some time. What has changed is an appreciation of how much the archival
practices of these “paper empires” signaled changes in their technologies of
rule.49
If it is obvious that colonial archives are products of state machines, it
is less obvious that they are, in their own right, technologies that bolstered
the production of those states themselves.50 Systems of written accountability
were the products of institutions, but paper trails (weekly reports to superiors,
summaries of reports of reports, recommendations based on reports) called
for an elaborate coding system by which they could be tracked. Colonial
statecraft was built on the foundations of statistics and surveys, but also out of
the administrative apparatus that produced that information. Multiple circuits
of communication – shipping lines, courrier services, and telegraphs – were
funded by state coffers and systems of taxation that kept them flush. Colonial
publishing houses made sure that documents were selectively duplicated,
disseminated, or destroyed. Colonial office buildings were constructed to
make sure they were properly catalogued and stored. And not unlike the
broader racialized regime in which archives were produced, the “mixed-
blood,” “Indo” youths, barred from rising in the civil service ranks, were the
scribes that made the system run. Employed as clerks and copyists in the
colonial bureaucracy, they were commonly referred to as “copy machines,”
and then disdained for their lack of initiative, their poor command of Dutch,

49 On this point, see Trouillot, 1995. On the relationship between state formation and
archival production, see Duchein (1992), cited above.
50 See my “Racial Histories and Their Regimes of Truth”, Political Power and Social
Theory 11 (1997): 183–255.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 99

and their easy adaptation to such imitative and degraded roles. Attention to
this sort of scaffolding of the colonial state renders an ethnographic reading
of the archives very different from what histories of the colonial looked like
several decades ago.

Along the archival grain

If one were to characterize what has informed a critical approach to the

colonial archives over the last fifteen years, it would be a commitment to the
notion of reading colonial archives “against their grain.” Students of coloni-
alism, inspired by political economy, were schooled to write popular histories
“from the bottom up,” histories of resistance that might locate human agency
in small gestures of refusal and silence among the colonized.51 As such,
engagement with the colonial archives was devoted to a reading of “upper
class sources upside down” in order to reveal the language of rule and the
biases inherent in statist perceptions.52
The political project was to write “un-State-d” histories that might demon-
strate the warped reality of official knowledge and the enduring consequences
of such political distortions. In Ranajit Guha’s formulation, colonial docu-
ments were rhetorical sleights of hand that erased the facts of subjuga-
tion, reclassified petty crime as political subversion, or simply effaced the
colonized. The political stakes were put on the analytic tactics of inversion
and recuperation: an effort to re-situate those who appeared as objects of
colonial discipline as subaltern subjects and agents of practice who made
– albeit constrained – choices of their own. Within this frame, archival
documents were counterweights to ethnography, not the site of it.53
But colonial authority, and the practices that sustained it, permeated more
diverse sites than those pursuing this “romance of resistance” once imagined.
If Marx’s insistence, that “people make their own history, but not exactly as
they please,” informed these early efforts to write histories of popular agency,
they also underscored that colonial rule rested on more than the calculated
inequities of specific relations of production and exchange. In looking more to
the carefully honed cultural representations of power, students of the colonial
have turned their attention to the practices that privileged certain social
51 For a more detailed account of these changes in research agenda, see the new preface
to my Capitalism and Confrontation in Sumatra’s Plantation, 1870–1979 (Ann Arbor:
University of Michigan Press, 1995).
52 I discuss some of these issues in “Perceptions of Protest: Defining the Dangerous in
Colonial Sumatra”, American Ethnologist 12(4) (1985): 642–658.
53 For a recent and sophisticated version of this culling project, see Shahid Amin, Event,
Metaphor, Memory: 1922–1992 (Berkeley: University of California Press, 1995).
100 ANN LAURA STOLER

categories and made them “easy to think.” Not least, we have become more
suspect of colonial vocabularies themselves that surreptitiously slip away
from their historical moorings and reappear as our explanatory concepts of
historical practice, rather than as folk categories that need to be explained.54
Focus in colonial studies on those tensions of empire that were at
once intimate and broad has placed sex and sentiment not as metaphors
of empire, but as its constitutive elements.55 Appreciating how much the
personal was political has revamped the scope of our archival frames: house-
keeping manuals, child-rearing handbooks, and medical guides share space
with classified state papers, court proceedings, and commission reports
as defining texts in colonialism’s cultures of documentation. Raymond
Williams’ pioneering treatment of culture as a site of contested, not shared,
meaning has prompted students of the colonial to do the same. In turning
from race as a thing to race as a porous and protean set of relations, colonial
histories increasingly dwell on the seams of archived and non-archived
ascriptions to redefine colonial subsumptions on a broader terrain.56 However
we frame it, the issues turns on readings of the archives based on what we
take to be evidence and what we expect to find. How can students of coloni-
alisms so quickly and confidently turn to readings “against the grain” without
moving along their grain first? How can we brush against them without a prior
sense of their texture and granularity? How can we compare colonialisms
without knowing the circuits of knowledge production in which they operated
and the racial commensurabilities on which they relied? If a notion of colonial
ethnography starts from the premise that archival production is itself both a
process and a powerful technology of rule, then we need not only to brush
against the archive’s received categories. We need to read for its regularities,
for its logic of recall, for its densities and distributions, for its consistencies
of misinformation, omission, and mistake – along the archival grain.
Assuming we know those scripts, I would argue, diminishes our analytic
possibilities. It rests too comfortably on predictable stories with familiar
plots. It diverts our attention from how much colonial history-writing has
been shaped by nationalist historiographies and nation-bound projects. It
leaves unquestioned the notion that colonial states were first and foremost
information-hungry machines in which power accrued from the massive
accumulation of ever-more knowledge rather than from the quality of it.
54 See the introduction, “Genealogies of the Intimate”, in my Carnal Knowledge and
Imperial Power: Race and the Intimate in Colonial Rule (Berkeley: University of California
Press, 2002).
55 See my “Sexual Affronts and Racial Frontiers”, Comparative Studies in Society and
History 34(3) (1992): 514–551.
56 See J. Chandler, A. Davidson, and H. Harootunian (eds.), Questions of Evidence: Proof,
Practice and Persuasion across the Disciplines (Chicago: University of Chicago Press, 1994).
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 101

It takes as a given that colonial statecraft was motivated and fueled by a

reductive equation of knowledge to power, and that colonial states sought
more of both. Not least, it makes irrelevant failed proposals, utopian visions,
and improbable projects because they were “non-events.” Reading only
against the grain of the colonial archive bypasses the power in the production
of the archive itself.

Civilities and credibilities in archival production

If colonial documents reflected the supremacy of reason, they also recorded

an emotional economy manifest in disparate understandings of what was
imagined, what was feared, what was witnessed, and what was overheard.
Such a reading turns us to the structures of sentiment to which colonial
bureaucrats subscribed, to the formulaic by which they abided, to the mix of
dispassionate reason, impassioned plea, cultural script, and personal exper-
ience that made up what they chose to write to their superiors and thus
place in the folds of official view. Dutch colonial documents register this
emotional economy in several ways: in the measured affect of official texts,
in the biting critique reserved for marginalia, in footnotes to official reports
where assessments of cultural practice were often relegated and local knowl-
edge was stored. Steven Shapin’s set of compelling questions in his social
history of truth could be that of colonial historians as well. What, he asks,
counted as a credible piece of information; what was granted epistemological
virtue and by what social criteria? What sentiments and civilities made for
“expert” colonial knowledge that endowed some persons with the credentials
to generate trustworthy truth-claims that were not conferred on others?
Colonial archives were, as Echevvaria notes, legal repositories of knowl-
edge and official repositories of policy. But they were also repositories of
good taste and bad faith. Scribes were charged with making fine-penned
copies. But reports on the colonial order of things to the Governor General in
Batavia, and to the Minister of Colonies in The Hague, often were composed
by men of letters, whose status in the colonial hierarchy was founded as much
on their display of European learning as on their studied ignorance of local
knowledge, on their skill at configuring events into familiar plots, and on
their cultivation of the fine arts of deference, dissemblance, and persuasion.
All rested on the subtle use of their cultural know-how and cultural wares.
As Fanny Colonna once noted for French Algeria, the colonial politics of
knowledge penalized those with too much local knowledge and those with
not enough.57 In the Indies, civil servants with too much knowledge of things
57 See Fanny Colonna, “Educating Conformity in French Colonial Algeria”, in Frederick
Cooper and Ann Laura Stoler (eds.), Tensions of Empire (1997), pp. 346–370.
102 ANN LAURA STOLER

Javanese were condemned for not appreciating the virtues of limited and
selective familiarity.
Christopher Bayly, in a thoughtful study of the development of an intelli-
gence system by the British in India, argues that the mastery of “affective
knowledge” was an early concern of the British colonial state, that diminished
throughout the nineteenth century as that state became more hierarchical and
governing became a matter of routine.58 But I would argue the opposite: that
affective knowledge was at the core of political rationality in its late colonial
form. Colonial modernity hinged on a disciplining of one’s agents, on a polic-
ing of the family, on Orwellian visions of intervention in the cultivation of
compassion, contempt, and disdain.
The accumulation of affective knowledge was not then a stage out of
which colonial states were eventually to pass. Key terms of the debates on
poor whites and child-rearing practices from as late as the 1930s, just before
the overthrow of Dutch rule, make that point again and again. When classified
colonial documents argued against the support of abandoned mixed-blood
children – that “mother care” (moederzorg) should not be replaced by “care of
the state” (staatszorg) – they were putting affective responsibility at the heart
of their political projects. When these same high officials wrote back and
forth about how best to secure “strong attachments” to the Netherlands among
a disaffected, estranged, and growing local European population, “feeling”
is the word that pervades their correspondence. Dutch authorities may never
have agreed on how to cultivate European sensibilities in their young, and just
how early in a child’s development they imagined they needed to do so. But at
stake in these deliberations over “upbringing” and “rearing” were disquieted
reflections on what it took to make someone “moved” by one set of sensory
regimes and estranged from others. Colonial states and their authorities, not
unlike metropolitan ones, had strong motivation for their abiding interest in
the distribution of affect and a strong sense of why it mattered to colonial
politics.

Cultural logics and archival conventions

The archive does not have the weight of tradition; and it does not consti-
tute the library of libraries, outside time and place – it reveals the rules of
practice . . . its threshold of existence is established by the discontinuity
that separate[s] us from what we can no longer say.59
58 Christopher Bayly, Empire and Information 1996.
59 Foucault, 1972, p. 130.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 103

One way to re-configure our uses of the colonial archive is to pause at, rather
than bypass, its conventions, those practices that make up its unspoken order,
its rubrics of organization, its rules of placement and reference. Archival
conventions might designate who were reliable “sources,” what constituted
“enough” evidence and what – in the absence of information – could be filled
in to make a credible plot. Conventions suggest consensus, but it is not clear
what colonial practitioners actually shared. Archival conventions were built
upon a changing collection of colonial truths about what should be classified
as secrets and matters of state security, and what sorts of actions could be
dismissed as prompted by personal revenge and ad hoc passion or accredited
as a political subversion against the state.60 Such conventions exposed the
taxonomies of race and rule, but also how skilfully, awkwardly, and unevenly
both seasoned bureaucrats and fledgling practitioners knew the rules of the
game.
Attention to these conventions may lead in two directions: to the consen-
sual logics they inscribed, but also much more directly to their arbitrary
rules and multiple points of dissension. Political conflicts show up in the
changing viability of categories and disagreements about their use. But as
Paul Starr suggests, “information out of place” – the failure of some kinds of
practices, perceptions, and populations to fit into a state’s ready-made system
of classification – may tell as much or more.61 Commentaries on European
nurseries in the colonies might be expected to turn up in reports on education,
but the very fact that they consistently showed up elsewhere – in reports on
European pauperism and white poor relief, or in recommendations to quell
creole discontent, suggest that what was “out of place” was often sensitive,
and that is was children cued to the wrong cultural sensibilities that were
dangerously out of place.

Colonial commissions as stories that states tell themselves

As Ian Hacking says of social categories, archives produced as much as

they recorded the realities they ostensibly only described. They told moral
stories, they created precedent in the pursuit of evidence, and not least they
create carefully tended histories. Nowhere is this history-making work more
60 On the administrative distinctions between the “political” and the “private,” and the
“criminal” versus the “subversive,” see my “Perceptions of Protest: Defining the Dangerous
in Colonial Sumatra”, American Ethnologist 12(4) (1985): 642–658; and “Labor in the
Revolution”, Journal of Asian Studies 47(2): 227–247.
61 Paul Starr, “Social Categories and Claims in the Liberal State”, in Mary Douglas and
David Hull (eds.), How Classification Works: Nelson Goodman among the Social Sciences
(Edinburgh: Edinburgh University Press, 1992), pp. 154–179.
104 ANN LAURA STOLER

evident than in the form of the commission of inquiry or state commission. By

definition, commissions organized knowledge, rearranged its categories, and
prescribed what state officials were charged to know. As the anthropologist,
Frans Husken, notes of Dutch commissions in colonial Java, “‘when nothing
else works and no decision can be reached, appoint a commission’ was a
favorite response of colonial authorities.”62 But commissions were not just
pauses in policy and tactics of delay. Like statistics, they helped “determine
. . . the character of social facts” and produced new truths as they produced
new social realities.63 They were responses to crisis that generated increased
anxiety, substantiating the reality of that crisis itself.64 By the time most
commissions had run their course (or spawned their follow-up generation),
they could be credited with having defined “turning points,” justifications for
intervention, and, not least, expert knowledge.
The various commissions produced on the problem of poor whites in
the Indies between the 1870s and early 1900s, and those carried out in
South Africa between the early 1900s and the late 1920s, are exemplary of
what I have in mind. There are certain general features which they share.65
Both produced published and publicized volumes: Pauperism among the
Europeans (published 1901–1902), and The Problem of Poor Whites in South
Africa (published 1929–1932).66 Both commissions were about indigent

62 Frans Husken, “Declining Welfare in Java: Government and Private Inquiries, 1903–
1914”, in Robert Cribb (ed.), The Late Colonial State in Indonesia (Leiden: KITLV, 1994),
p. 213.
63 Ian Hacking, “How Should We Do the History of Statistics?”, in Graham Burchell, Colin
Gordon, and Peter Miller (eds.), The Foucault Effect: Studies in Governmentality (Chicago:
Chicago University Press, 1991), p. 181.
64 A good example of what Ian Hacking calls “dynamic nominalism” or “the looping effect”
in categorization.
65 I discuss the politics of colonial comparisons elsewhere and therefore will not do so here.
I have used the 1902 Indies Pauperism Commission, commentaries around it, and enquiries
that preceded it, in much of my writing over the last fifteen years on the construction of
colonial racial categories. The South African Carnegie Commission and the enquiries that
preceded it are compared in a chapter in my forthcoming book, Along the Archival Grain. A
more general discussion of the politics of comparison can be found in my “Tense and Tender
Ties: American History meets Postcolonial Studies,” paper delivered to the Organization of
American Historians in April 2000; and in my “Beyond Comparison: Colonial Statecraft
and the Racial Politics of Commensurability,” paper delivered as a keynote address to the
Australian Historical Association in Adelaide, July 2000.
66 Students of colonialism could come up with a host of others. For an unusual example of
someone who deals with the commission as a particular form of official knowledge, in this
case with the South African Native Affairs Commission, see Adam Ashforth, The Politics of
Official Discourse in Twentieth-Century South Africa (Oxford: Clarendon Press, 1990). Also
see Frans Husken’s discussion of the Declining Welfare Commission in Java, cited in footnote
62.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 105

Europeans and their inappropriate dispositions toward work, racial distance,

sexual propriety, and morality. Each requisitioned administrative energy and
expertise, entailed several years of labour, produced thousands of pages of
text, and involved scores of interviewers and hundreds of interviewees. In the
case of the Indies, its probing questionnaires on sexual unions, illegitimate
children, and domestic arrangements sparked the wrath of hundreds of irate
colonial Europeans, who condemned the Indies government as an “inquisi-
tionary state.” Both commissions were repositories of colonial anxieties –
unsettling testimonies to the insecurity of white privilege, to the ambigu-
ities of membership in the privileged category of “European,” and to the
making of a public welfare policy solidly based on race. Both worried over
increasing numbers of impoverished whites because they were worried about
something else. As stated in the Carnegie Commission, their “propinquity
of . . . dwellings” to “non-Europeans” tended to bring native and white into
contact, “counteract miscegenation,” weaken the color line, and promote
“social equality.”67
These commissions could and should be read for their extraordinary
ethnographic content, but also for the content evident in their form. Like other
colonial commissions, they marked off clusters of people who warranted state
interest and state expense. Secondly, they were redemptive texts, structured
to offer predictions based on causal accounts of exoneration and blame.
And thirdly, both commissions were documents to the making of state
historiography and monuments to why history mattered to consolidating and
imperial states. In writing the past, they produced dramatic narrative histories
based on select chronologies, crystallizing moments, and significant events.
In defining poverty in the present, they also dictated who in future would
count as white –and therefore who would be eligible for state aid.
In doing all of the above, they wrote, revised, and over-wrote genealo-
gies of race. Neither of these commissions were the first of their kind. On
the contrary, they were made credible by how they mapped the past onto
prescriptions for the present and predictions of the future. They also showed
something more. How social practices were historically congealed into events
and made into things: how an increase of unemployment and impoverishment
among European colonials became a “problem” called “poor whiteism,” with
attributes of its own. “Poor whiteism” defined, physiologically and psycho-
logically, distinct sorts of persons, with aggregated ways of “being in the
world,” with specific dispositions and states of mind. Like other colonial
commissions, these commission were consummate producers of social kinds
and social categories.
67 The Poor White Problem in South Africa, Report of the Carnegie Commission (Stellen-
bosch: Pro Ecclesia Drukkerij, 1932), p. xx.
106 ANN LAURA STOLER

Commissions and statistics were features of statecraft in similar ways.

Both are eighteenth-century inventions consolidated by the nineteenth-
century liberal state.68 Both were products and instantiations of the state’s
investment in public accountability. But commissions commanded more
moral authority as they purported to scrutinize state practice, reveal bureau-
cratic mistakes, and produce new truths about the workings of the state itself.
Moreover, these “poor white” commissions were quintessential products of
“biopolitical” technologies. Not only did they link the relationship between
parent and child, nursemaid and infant, to the security of the state; they
sought ethnographic substantiation, eye-witness testimonies from participant-
observers that what individuals did – whether they wore shoes, lounged on
their porches, spoke Dutch or Malay, or made their children say morning
prayers – were practices linked directly to the state’s audit of its own viability.
Both commissions and statistics were part of the “moral science” of
the nineteenth century that coded and counted society’s pathologies. While
statistics used deviations from the mean to identify deviations form the
norm, commissions joined those numbers with stories culled from indi-
vidual “cases” to measures gradations of morality.69 Commissions in turn
affirmed the state’s authority to make judgements about what was in society’s
collective and moral good. Both were prescriptive and probabilistic tools
whose power was partially in their capacities to predict and divert politically
dangerous possibilities.
Like statistics, the commission demonstrated the state’s right to power
through its will to truth. In the Indies, the Pauperism Commission conferred
on the state moral authority by demonstrating its moral conscience and
disinterested restraint, its willingness and commitment to critically reflect on
its own mishaps, to seek the truth “at whatever cost.” But its power rested in
more than its calculation of the moral pulse of the present and its implications
for the future. The commission justified its licence to expend funds, time,
and personnel in part by rehearsing the past and remembering and reminding
its readership of its enduring weight. Historical narratives shape these texts
with stories that deflected the causes of deprivations and inequities away from

68 Royal commissions have a longer history still. See, for example, David Loades, “The
Royal Commissions”, in Power in Tudor England (New York: St. Martin’s Press, 1997),
pp. 70–82. On statistics and state-building, see Alain Desrosieres, “Statistics and the State”,
The Politics of Large Numbers (1998): 178–209. For the twentieth century, see William J.
Breen, “Foundations, Statistics, and State-Building”, Business History Review 68 (1994):
451–482.
69 See Arjun Appardurai’s discussion of numerical representation in colonial India as a “key
to normalizing the pathology of difference.” “Number in the Colonial Imagination”, Modernity
at Large: Cultural Dimensions of Globalization (Minneapolis: University of Minnesota,
1996), pp. 114–138.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 107

the present, even as they rehearsed the enduring burden of earlier policies of
administrations past.
Finally, these commissions were quintessential “quasi-state” technolo-
gies, both part of the state and not, at once a product of state agents but
constituted invariably by members outside it. If modern states gain force in
part by creating and maintaining an elusive boundary between themselves
and civil society, as Tim Mitchell has argued, such commissions exempli-
fied that process.70 Their specific subjects were state generated, but often
researched and written by those not in its permanent employ. Both the
Indies and Carnegie Commissions delegated bodies of experts equipped to
assess morality (religious experts), deviance (lawyers, educators), and disease
(doctors), to whom the state conferred short-term and subject-specific voice
and public authority. They instantiated the ways in which the state exercised
its will to power by calling on “outside” expert authorities to verify the state’s
ability to stand in for public interest and its commitment to the public good.

Archival seductions and state secrets

As archivists are the first to note, to understand an archive one needs to

understand the institutions that it served. What subjects are cross-referenced,
what parts are re-written, what quotations are cited, not only tell us about
how decisions are rendered, but how colonial histories are written and re-
made. Information out of place underscores what categories matter, which
ones become common sense and then fall out of favor. Not least they provide
road maps to anxieties that evade more articulate form.
The commission is one sort of archival convention; “state secrets” are
another. States traffic in the production of secrets and the selective dissemi-
nation of them. In this regard, the Dutch colonial state was gifted at its
task.71 As Weber once noted, the “official secret” was a specific inven-
tion of bureaucracy that was “fanatically defended” by it. The designations
“secret,” “very secret,” and “confidential” registered more than fictions of
denied entry and public access. Nor did they mostly signal the pressing polit-
ical concerns of the colonial state. Rather, and more importantly, such codes
70 See Gramsci’s discussion of “state and civil society,” in Quintin Hoare and Geoffrey
Smith (eds.), Selections from the Prison Notebooks of Antonio Gramsci (London: Lawrence
and Wishart, 1972), esp. 257–264; and Timothy Mitchell, “The Limits of the State”, American
Political Science Review 85 (1991): 77–96.
71 George Simmel once wrote that “the historical development of society is in many respects
characterized by the fact that what at an earlier time was manifest enters the protection of
secrecy; and that, conversely, what once was a secret, no longer needs such protection but
reveals itself”, in Kurt Wolff (ed.) The Sociology of George Simmel (London: Free Press,
1950), p. 331.
108 ANN LAURA STOLER

of concealment were the fetishized features of the state itself. State secrets
named and produced privileged knowledge, and designated privileged readers
while reminding the latter what knowledge should be coveted, and what
was important to know. The secreted report, like the commission, created
categories it purported to do no more than describe. In the Indies, the classi-
fied document commanded a political weight that called for secret police,
paid informants, and experts.
Secrets imply limited access, but what is more striking in the Dutch
colonial archives is how rarely those items classified as “confidential”
(vertrouwelijk, zeer vertrouwelijk, geheim, and zeer geheim) were secrets
at all. Some surely dealt with clandestine police and military tactics (such
as preparations for troop movements to protect planters against an attack),
but far more of these documents were about prosaic, public parts of Indies
life.72 If one could argue that documents recording the disquieting presence
of European beggars and homeless Dutchmen in the streets of Batavia were
“secrets” to those reading them back in the Netherlands, these presences
certainly were not secrets to the majority of Europeans who lived in the
colony’s urban centers.
What was “classified” about these reports was not their subject-matter
– in this case, indigent “full-blooded” Europeans and their mixed-blood
descendants – but rather the conflict among officials about how to act on the
problem, their disparate assessments of what was the cause, and how many
such indigents there were. Some reports were “classified” because officials
could not agree on whether there were twenty-nine mixed-bloods in straitened
circumstances or tens of thousands.73 In short, documents were classified as
“sensitive” and “secret” sometimes because of the magnitude of a problem
– other times because officials could not agree on what the problems were.
But perhaps what is more surprising are the range of confidential informa-
tion that students of colonialism expect them to divulge. State secrets are
not necessarily secreted truths about the state, but promises of confidences
shared. If state secrets are more attention-getting annotations than conven-
tions of concealment, then how state secrets were produced, what was a
secret at one time and later not, may index the changing terms of what was
considered “common sense,” as well as changes in political rationality. As
Marc Ventresca argues in a study of why and when states count, statistical
information in the eighteenth century was considered a source of state power
and therefore not published. Public access to state statistics was a nineteenth-

72 Algemeen Rijksarchief (The Hague) Ministerie van Koloniën. Geheim No. 1144/2284.
From the Department of Justice to the Governor General, Batavia, 29 April 1873.
73 Algemeen Rijksarchief. Verbaal 28 March 1874, no. 47. From the Department of Justice
to the Governor General.
 COLONIAL ARCHIVES AND THE ARTS OF GOVERNANCE 109

century phenomenon.74 State secrets made up a basic feature of the colonial

archive, a telling element in the production of fictions of access displayed by
their content as well as their form.

Colonial archives as “systems of expectation”

To take up Jean and John Comaroff’s invitation to “create new colonial

archives of our own” may not only entail, as they rightly urge, attention to
new kinds of sources, but also to different ways of approaching those we
already have, different ways of reading than we have yet done.75 In turning
from an extractive to a more ethnographic project, our readings need to move
in new ways through archives both along their fault lines as much as against
their grain. De Certeau once defined the science of history as a redistribution
in space, the act of changing something into something else. He warns us that
our historical labours in the archives must do more than “simply adopt former
classifications,” must break away from the constraints of “series H in the
National Archives” replaced with new “codes of recognition” and “systems of
expectation” of our own. But such a strategy really depends on what we think
we already know.76 For students of colonialisms, such codes of recognition
and systems of expectation are at the very heart of what we still need to learn
about colonial polities. The breadth of global reference and span of lateral
vision that colonial regimes unevenly embraced suggest that an ethnographic
sensibility, rather than an extractive gesture, may be more appropriate for
identifying how nations, empires, and racialized regimes were fashioned –
not in ways that display confident knowledge and know-how, but in disquieted
and expectant modes.

74 Marc Ventresca, When States Count: Institutional and Political Dynamics in Modern
Census Establishment, 1800–1993. Ph.D. thesis: Stanford University (1995), p. 50.
75 Jean and John Comaroff, Ethnography and the Historical Imagination (Boulder: West-
view Press, 1992).
76 De Certeau, 1988 [1974], p. 107.