FORGING THE NATION THROUGH RAILS: TRANSPORTATION
INFRASTRUCTURE AND THE EMERGENCE 
OF CHINESE NATIONALISM
by
DYLAN P. BRADY
A THESIS
Presented to the Department of Geography
and the Graduate School of the University of Oregon
in partial fulfillment of the requirements
for the degree of
Master of Arts 
December 2013
THESIS APPROVAL PAGE
Student: Dylan P. Brady
Title: Forging the Nation Through Rails: Transportation Infrastructure and the Emergence 
of Chinese Nationalism
This thesis has been accepted and approved in partial fulfillment of the requirements for 
the Master of Arts degree in the Department of Geography by:
Daniel Buck Chairperson
Alexander B. Murphy Member
and
Kimberly Andrews Espy Vice President for Research and Innovation;
Dean of the Graduate School 
Original approval signatures are on file with the University of Oregon Graduate School.
Degree awarded December 2013
ii
© 2013 Dylan P. Brady 
This work is licensed under a Creative Commons  
 Attribution-NonCommercial-NoDerivs (United States) License.
iii
THESIS ABSTRACT
Dylan P. Brady
Master of Arts
Department of Geography
December 2013
Title: Forging the Nation Through Rails: Transportation Infrastructure and the Emergence 
of Chinese Nationalism
While nationalism remains a vital element in the production of the political and 
economic landscape, it is often treated as a static container for other processes or 
neglected altogether. Rather, it must itself be treated as a process—a nationalizing project
—emerging from a constellation of often contradictory social forces. One such process of 
nationalization is the development of large-scale transportation infrastructure, such as 
railroads. These projects produce both new spheres of circulation and new understandings 
critical to navigating these novel environments, which together radically transform the 
relation between people, government, and territory. 
In early twentieth century China, the complicated contest over railroad rights 
produced and was produced by a fractured political economic geography. Understandings 
of both identity and space remained fragmented, cohering only partially into a singular 
entity, thus demonstrating the intimate interrelation between state power, political 
identity, and territories both real and imagined.
iv
CURRICULUM VITAE
NAME OF AUTHOR:  Dylan P. Brady
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon, Eugene
Whitman College, Walla Walla, Washington
DEGREES AWARDED:
Master of Arts, Geography, 2013, University of Oregon
Bachelor of Arts, Asian Studies, 2006, Whitman College
AREAS OF SPECIAL INTEREST:
Political Economy of Chinese Nationalism
Social Impacts of the Development of Modern Transportation Infrastructure
PROFESSIONAL EXPERIENCE:
Graduate Teaching Fellow, Department of Geography, University of Oregon, 
Eugene, 2010-13
GRANTS, AWARDS, AND HONORS:
Graduate Teaching Fellowship, Geography, 2010 - present
Foreign Language and Areas Studies Scholarship, University of Oregon, 2012-13
v
ACKNOWLEDGMENTS
I wish to express sincere appreciation to Professors Buck and Murphy for their 
patience and insightful critique at every step of this process. I would also like to 
acknowledge the support of the Department of Geography staff, whose timely assistance 
was indispensible, and the US Department of Education, whose FLAS Scholarship 
enabled me to take my studies to China and back. I would also like to acknowledge the 
support and sacrifice of my wife, Ruty Wherland, without which none of this would be 
possible.
vi
To Ruty, for coming with me.
vii
TABLE OF CONTENTS
Chapter Page
I. RAIL AS INVASION, RAIL AS NATION.............................................................. 1
II. NATION AS NATIONALIZING PROJECT.......................................................... 10
A New Materialist Approach to Nationalism......................................................... 13
Nationalism at the Edges....................................................................................... 17
Chinese Nationalism.............................................................................................. 30
What Is Needed Then............................................................................................. 38
III. FORGING THE NATION THROUGH RAILS.................................................... 41
Theorizing the Emergence of Nationalism............................................................ 43
Infrastructure and the National Imperative............................................................ 52
The Transformation of Skill................................................................................... 53
The Skilled Environment....................................................................................... 61
The State and Infrastructure................................................................................... 63
The Impact of Railroads on the Production of Space............................................ 66
State Impact at a Slant............................................................................................ 85
IV. THE FRACTURED GEOGRAPHIES AND IDENTITIES OF EARLY 
TWENTIETH CENTURY CHINA............................................................................. 87
Social Histories of Rail.......................................................................................... 90
Rail and the Transformation of Chinese Society................................................... 93
New Cities, New Hinterlands................................................................................ 97
The Provincial State and Provincial Identity......................................................... 100
viii
Chapter Page
Shanxi.................................................................................................................... 105
Sichuan................................................................................................................... 107
Manchuria.............................................................................................................. 108
An Infrastructural Lens.......................................................................................... 117
V. BUILDING BEYOND THE NATION.................................................................... 119
REFERENCES CITED................................................................................................ 126
ix
LIST OF FIGURES
Figure Page
1. Schematic of Chinese rail, 1911 & 1949. ............................................................. 93
x
CHAPTER I
RAIL AS INVASION, RAIL AS NATION
When the Boxer rebellion broke out in 1900, railroads were a major target of their 
anger. Seen as foreign pollutions, accused of being black magic driven by human 
sacrifice, rail lines were destroyed by Boxer forces as a way of purifying China and 
incidentally, slowing the incursion of foreign armies. In 1911 the eruption of the Railway 
Protection movement in Sichuan, sparked by an imperial attempt to seize control of 
privately- and locally-owned rails (in order to pay indemnities), forced imperial forces to 
leave their posts in Wuchang. Shortly thereafter, revolutionaries in Wuchang seized the 
opening and rose up, sparking a chain of events known today as the Xinhai Revolution, 
culminating in the establishment of the Republic of China. 
These two incidents, so closely juxtaposed, give a sense of not only the central 
position played by rail in the emergence of Chinese nationalism, but also its complex and 
shifting valence. Railroads are both an intolerable foreign incursion and also the people's 
inviolable right—much like nationalism itself. In the past decade Chinese nationalism has 
surged to the fore and has surprised many observers with its sudden and pressing political 
relevance, both within China and without. While the moment of its emergence is long 
past, it remains a vital force in China and the world, capable of dynamic changes and 
unexpected impacts. Yet understandings of nationalism, particularly Chinese nationalism, 
emphasize its structural and functional elements, leaving us with a theoretical apparatus 
poorly suited to understanding where it came from or what it is capable of. Examining the 
period when it first began to cohere into a movement is not therefore just a project of 
1
historical interest, but can provide insight into the processes which continue to sustain 
and shape it today. 
In this thesis I analyze the emergence of Chinese nationalism through the railroad
—not merely as a symbol, but as a material force in shaping the terrain of nationalism. 
This approach is atypical, particularly in the context of China. Most discussions of 
nationalism, particularly Chinese nationalism, focus on rhetoric, ethnicity, politics, 
government-led education campaigns, street protests and similar aspects. While these are 
important moments in the formation of nationalism and particularly in giving it direction, 
they are, in my mind, just the tip of a very large iceberg: the unseen bulk of which is not 
the explicitly nationalist but rather the perceptions and understandings that emerge from 
the everyday life of people living in modern society. People don't identify as part of a 
nation because they sang some songs in elementary school: they identify as part of a 
nation because it is simply obvious to them, in the daily course of their lives, that it is so. 
The multitude of ways it is not so, the ways the nation is a far more problematic and 
fragmented object than it appears, does not make its appearance of cohesion any less 
apparent.
A great deal of the work that focuses on Chinese nationalism concentrates its 
attention on the overt, rhetorical manifestations of nationalism. Much of the work that 
isn't explicitly about nationalism tends to either assume the nation's unproblematic 
existence or ignore it altogether. These tendencies produce narratives of the Chinese 
nation that too often reduce it to an uncontrollable popular, resentful chauvinism or the 
cynical manipulations of a plutocratic elite, missing not only the diversity and vitality of 
2
past and present incarnations of Chinese nationalism, but also isolate nationalism from 
the social context that gives it meaning.
My goal in this thesis is to sketch out a different approach, one that puts 
nationalism at the nexus of social, economic, and territorial change. To accomplish this 
task, I draw on the conceptual tools pioneered in the new materialisms scholarship. This 
field challenges many of the dichotomies—agency and structure; human and non-human; 
discursive and concrete; produced and natural—that have traditionally prevailed in the 
social sciences. Instead, the emphasis is in how these apparent opposites interpenetrate 
and mutually constitute material reality; this opens up a realm of analysis where 
physicality of rhetoric is as important as the social aspects of the built environment, and 
each is capable of acting on the other.
I focus on the built environment, and railroads in particular, because rail offers an 
unusually direct link between the tenuously-connected realms of state action and 
individual consciousness; it is intimately entangled both in vast movements of capital and 
state power and in the everyday experience of individuals. By producing both new kinds 
of territory and a popular awareness of those territories, rail transportation can be critical 
to creating the perception of a national territory, or geo-body: in the railways one can see 
the national unit being built. 
Ultimately, what I am striving for is a way of looking at Chinese nationalism that 
acknowledges its volatile relationship with its own nation and state. I'm looking for a way 
of studying the development of nationalism in China that takes the long view, and isn't 
obscured by fluctuating trends in political rhetoric. A way of exploring the interaction 
between large-scale political and economic shifts of state and capital and the experience 
3
of individuals, mediated through the construction and navigation of social space. A way 
that examines the shifting importance of territory and nation without granting their “self-
evident” validity on the one hand, nor, on the other hand, disregarding their existence 
altogether. With any luck, this thesis represents the first steps on that path.
Chapter II contextualizes my project within the current literature on Chinese 
nationalism—not just work which explicitly takes nationalism as its object of study, but 
even work that examines other objects but unknowingly relies on naïve conceptions of 
nation and territory when constructing its field of study. Indeed, the tendency to 
unproblematically use nations as the “natural” containers for societies, literatures, 
languages, economies, states, and so on is endemic even among post-colonial area studies 
specialists who ought to know better. 
Materialism is introduced here as a way of understanding the shortcomings of 
these approaches as well as directing attention towards alternative analytics. At closer 
examination, the production of the nation dissolves into a complex nest of whirring parts, 
a set of processes operating at different spatial and temporal scales. While they do come 
together, this union is far more happenstance and temporary than unitary models of the 
nation suggest. Rather than a nation, it is a nationalizing project, whose goal of 
nationhood is never reached but is always in the act of being produced.
The literature that focuses on the subject of Chinese nationalism also exhibits 
certain problematic trends. With some notable exceptions, the literature on Chinese 
nationalism focuses on the role of the state in founding and shaping nationalism in the 
past and today. The initial position of Chinese nationalism as a fierce critic of the 
contemporary Qing state, and the marshaling of nationalism by modern critics against the 
4
state receives far less attention. Even when nationalism is analyzed outside of the state 
context and populist, heterogenous nationalisms and sub-nationalisms are included, 
discourse remains the crux of the analytic. To understand nationalism's emergence and 
continued vitality as a social force, I maintain, its discourse must be analyzed side by side 
with its social and economic relations; nationalism is not just a discursive phenomenon 
but a tangible one as well. The state's influence on nationalism is considerable, but 
nationalism is never entirely within the state's grasp and often manifests in ways quite 
orthogonal to its interests.
In Chapter III, I trace out a theoretical apparatus capable, in theory, of executing 
this kind of analysis. This begins with a re-examination of two prominent theories of 
nationalism which emphasize the relation between nationalism and the great social 
transformations of industrialization and modernity. Benedict Anderson emphasizes the 
creation of the imagined community, conceived of by the mechanism of shared language 
and culture enabled by print capitalism; Gellner emphasizes the transformation of social 
structure and work set into motion by industrialization. Placing these different 
mechanisms of nationalization side by side suggests the fragmentary, heterogenous nature 
of nationalism's tendencies towards cohesion and homogenization, and also draws 
attention to the role played by skill in the emergence of nationalism.
Skill mediates the relation between the worker and the means of production. Long 
conceptualized by Marxists as being continuously degraded under capitalism, prompted 
by Gellner I suggest an alternative interpretation of skill's transformation: rather than 
destroying skill outright capital transforms it, homogenizes it. Homogenization 
accomplishes the function of eliminating capital's dependence on any particular subset of 
5
the working class, while maximizing that average level of skill maximizes the workers' 
efficiency as well. The line between skill and culture—as seen in the example of literacy
—is always quite blurry, and in creating a homogenous but complex skill set among 
workers, industrialization also creates a shared culture.
This culture is not unbounded, however. The sphere of circulation the broadly-
shared industrial culture creates only extends as far as that culture, and several factors 
conspire to limit its scope. One important factor is the influence of the built environment. 
As the infrastructure of the built environment is subsumed under capitalism, it is 
revolutionized just like any other means of production, also revolutionizing the skills 
which relate to its use. Particularly for large-scale infrastructure, the state plays a major 
role in enabling and supervising its construction and maintenance, and is often the 
instigator and owner of the projects. The population that lives within and depends upon 
this environment can only live and move about within the area they understand. Thus as 
the built environment becomes more similar over large spatial extents and demands more 
specialized knowledge to navigate, people are simultaneously freed to circulate beyond 
the local sphere, but are still constrained by the extent of the social environment their 
knowledge pertains to. As the state becomes increasingly involved in producing the 
environment, and the population becomes increasingly dependent on that environment, 
the triple imperative of nationalism—that demand for the coincidence of the national 
people, government, and territory—begins to emerge directly from the logic of capitalist 
development.
The means of transport play a particularly significant role in defining the sphere 
of circulation. Here I focus on the role of rail, the exemplar of industrial transport. The 
6
technical demands of rail create strong homogenizing, standardizing, and centralizing 
tendencies.1 Rail both produces new environments for people to navigate, in the form of 
rail carriages and train stations, and also drastically reworks the existing geography of the 
country. Any location accessible via the rails moves closer both temporally and 
perceptually, and other locations move further away. Rails create a sphere of circulation 
for people and commodities with far less friction than before, to the point that the 
distance between points begins to be forgot: commodities appear as products of the 
market through which they circulate, and places appear as products of the rails which 
carry one there. Paralleling Anderson's imagined community, rails contribute to the 
creation of an imagined territory.
Rail, then, is a site where the collision of state, capital and the individual is 
particularly abrupt: vast flows of capital and labor work to assemble a tremendous 
network of state-scale infrastructure, which then becomes an immediate and integral part 
of people's lives. Rail mediates the relation between the individual and the state in an 
uncommonly direct way. 
In Chapter IV, I take the theory worked out above and attempt to apply it to the 
history of Chinese rail development in the first half of the twentieth century. The theory 
indicates that governmental power, regional consciousness, and rail are entangled in a 
mutually-constitutive ensemble, and therefore will tend to manifest on a similar scale. 
Rail developed in China within a complex political environment where national 
governments, provincial governments and imperial powers all attempted to consolidate 
their power by controlling rail. This led to, and resulted from, the formation a set of 
1. Wolfgang Schivelbusch, The railway journey: the industrialization of time and space in the 19th 
century (Berkeley, Calif.: University of California Press, 1986).
7
partially overlapping spheres of circulation that reworked the previous spatial structure in 
dramatic ways, but did not create a single national sphere.
Early twentieth century China was characterized by weak central and strong 
provincial or regional government; nationalist movements battled with political 
movements at the provincial scale to become the primary form of political engagement. 
At times provincial identification superseded national identity; domestic rail projects 
were similarly conceived, administered, and supported at the provincial level. The strong 
influence of foreign powers on the development of Chinese rail in the pre-War period 
contributed to the fracturing of China's rail network: particularly in the Northeast, where 
Russia and Japan invested heavily in rail construction, rail created primarily regional 
rather than a national sphere of circulation. This demonstrates the potential for even as 
homogenizing a process as rail to exert heterogenizing influences under certain 
conditions.
Finally, in the concluding chapter I examine the limitations of this project and 
indicate some further directions this strand of research might be extended. Rail is, in the 
end, only one of a countless number of nationalizing mechanisms, each with their own 
specific history and ramifications. As I have argued previously, nationalizing projects are 
manifold, and their many parts are only in sync some of the time. Nor are railroads a 
singularly fitting exemplar: rather, their state-centric and capital-intensive nature 
emphasizes centralized and centralizing nation-building at the expense of relatively 
horizontal, multi-valent processes. Indeed, even within rail difference is produced anew: 
access to rail is uneven economically and politically. It is uneven geographically. The 
centralizing logic of nationalization is powerful, as the organization of the globe attests, 
8
but it is not monolithic. Other forms of transportation networks undoubtedly manifest 
different patterns, much less entirely different nationalizing mechanisms like mass 
education, modern militarization, or industrial work itself. 
After the triumph of the Communists in 1949, rail in China plays a very different 
role in Chinese history. United under the control of a strong central government, rail 
develops rapidly into a cohesive national network; after 1978 it serves as the conduit for 
unprecedented population flows. This thesis does not touch on this history, though it can 
suggest its importance. China's rail network continues to transform to this day; its 
influence on political consciousness has not come to a close. Rail, and this view of 
nationalism, offers us a way to talk about Chinese nationalism not just as something 
inherited from Sun Yatsen or as a disciplinary tool methodically deployed by today's 
Party, but as an ongoing project, shaped but not constrained by state action, capable of 
change and a vital force in its own right. 
This thesis makes the argument that nationalism must be studied, and studied as 
more than a rhetorical endeavor. Nationalism is a vital part of the larger discussion of 
political and economic geography, and is absolutely essential to any discussion of 
territoriality. 
9
CHAPTER II
THE NATION AS NATIONALIZING PROJECT
The nation continues to be a unit of analysis that is itself under-analyzed. This is a 
flaw extending into the very heart of area studies: too often, the nation is either the 
implicit area of analysis or is elided altogether from the analytic; rarely is the object of 
the nation itself, or the state with which it is mutually constitutive, the focus of analysis. 
Those approaches that do directly address the nation grant to it too easily the story of its 
own homogeneity—even those critical of the idea of an eternal national essence do not 
doubt its current cohesion. Yet to take seriously the idea that the nation was constituted at 
a particular point in history must also take seriously the ongoing and incomplete nature of 
that project: nations are no less being created and recreated today than they were at the 
moment of their ascendance. The focus is typically on points of origin and boundaries; 
yet reproduction and internal spaces are no less in need of explanation. 
A more complete approach to nationalism would avoid these pitfalls: rather than 
treating nationalism as a singular object, it would instead be treated as a set of distinct 
social and technological processes, each with its own relations with material and 
governmental practice. This approach would integrate sensitivity to nationalism's internal 
heterogeneity with an awareness of the pull of its homogenizing logic, without attributing 
to it a teleological force. The nation would be collectively constituted as a nationalizing 
project: an emergent phenomenon capable of generating novel forms and relations not 
reducible to original intentions or necessary laws of development. The nation would thus 
be treated as an open-ended, contingent process rather than an idealized, static object.
10
 This sort of approach breaks not only from traditional conceptions of the nation, 
but with traditional conceptions of objects. The challenge is therefore as much 
ontological as it is theoretical: how to articulate this sort of process-oriented 
understanding? In order to re-conceptualize the nation as a nationalizing project, this 
work draws from the renewed materialist tradition. This philosophic approach offers a 
number of ways of thinking about contingency, non-human agency, and emergent 
complexity. Within a new materialist framework, all objects are ongoing processes of 
materialization within which human agency co-participates: this frees objects from the 
trap of a static reductionism, unable to change or to impact the world. Instead, the 
relations that compose processes such as nationalism are capable of change—even of 
generating entirely novel and unexpected relations.
After laying out the materialist epistemology that lies at the core of my analysis, 
this chapter employs that analysis to critique previous approaches to Chinese nationalism. 
This section is composed of two parts. The first examines several recent works on China, 
but not explicitly on Chinese nationalism, and the ways that the national is treated 
therein. While nonetheless valuable works in their own right, their failure to address the 
question of the nation leaves a gap in their account of modern China. Some of the works 
put the cohesive state narrative squarely at the center but fail to adequately address 
dissenting or tangential discourses and realities, while others err too far in excluding the 
state and its homogenizing projects from analysis.  Understanding the nation requires 
addressing both its cohesion and its heterogeneities; without capturing both aspects, the 
relation between the two cannot be explained.
11
The second part of this essay focuses instead on the challenges faced by work that 
explicitly takes Chinese nationalism as its object of study. These accounts regularly take 
one of two approaches: either they focus on the Republican period, which emphasizes the 
evolution of Chinese nationalism towards a single whole, or they discussion the 
renaissance of Chinese nationalism in the early twentieth century, typically emphasizing 
the role of state power. Both of these approaches emphasize the discursive, state-centric, 
and unitary elements of nationalism at the expense of the socio-economic, non-state, and 
disparate elements. While the state—or in the Chinese case, states—play a crucial role in 
the constitution of nationalism, it is equally important to emphasize the role that 
nationalism (or competing nationalisms) play in constituting the state: any theory that 
casts nationalism entirely as a lackey of the state has misstepped. 
Some works avoid these pitfalls to some degree, but few are able to wholly 
integrate both the discursive and concrete aspects of nationalist development: it remains 
primarily a discursive object. While a few scholars grapple with popular nationalism, 
they do not invoke the society-wide cultural and economic shift that the theorists of 
nationalism such as Gellner or Anderson emphasize. Without explaining what grip 
nationalist rhetoric has on the masses—what it is in the lives of Chinese people that it 
resonates with—these accounts of nationalism can only be partial.
Nationalism remains a vital force on the global stage, and our understanding of it 
is far from complete. Further investigation is necessary not only for specialists in the 
field, but for anyone who wishes to understand today's world.
12
A new materialist approach to nationalism
This work turns to the materialist philosophic tradition as a means of 
incorporating non-discursive social and economic processes into our understanding of 
nationalism. Materialism takes the stubborn contingency, vitality, and unpredictability of 
matter as a central object of inquiry. Several strands of materialist thought have recently 
been put to useful work within human geography: Maria Kaika has used Haraway's 
notions of cyborg social/material hybridity to analyze the complex nature of municipal 
water supply; Meehan, Shaw and Marston have taken the object-oriented ontology of 
Graham Harmon to explicate the force-full potential of objects within the urban ecology.2 
While I pursue materialism from a different angle, these projects nonetheless point to the 
potential for materialist thinking to open new avenues of inquiry in modern critical 
geography. Coole and Frost, in their recent anthology, provide an overview of this 
approach's recent revitalization and its concerns.3 
Discussions of materialism are difficult. As Coole and Frost point out, “there is an 
apparent paradox in thinking about matter: as soon as we do so, we seem to distance 
ourselves from it, and within the space that opens up, a host of immaterial things seems to 
emerge.”4 To this one might add the inverse problem: metaphors derived from the 
physical (things emerging “within the space” opened up) are constitutive of the language 
we use to discuss “immaterial” things. Discourse and material are thus confused from the 
2. Maria Kaika, City of Flows: Modernity, Nature, and the City (New York: Routledge, 2005); Donna 
Jeanne Haraway, Simians, Cyborgs, and Women: The Reinvention of Nature (New York: Routledge, 
1991); Katharine Meehan, Ian Graham Ronald Shaw, and Sallie A. Marston, “Political Geographies of 
the Object,” Political Geography 33 (March 2013): 1–10, doi:10.1016/j.polgeo.2012.11.002; Graham 
Harman, Tool-Being: Heidegger and the Metaphyics of Objects (Chicago: Open Court, 2002).
3. Diana H Coole and Samantha Frost, New Materialisms: Ontology, Agency, and Politics (Durham 
[NC]; London: Duke University Press, 2010).
4. Ibid., 2.
13
very start. New materialist approaches do not reject this con-fusion but take it as an 
opening to explore the ways in which social discourse and corporeal existence are 
intertwined and mutually constitutive. This approach to materialism is not, therefore, an 
attempt to assert the dominance of some grossly-defined physical over the discursive. As 
Coole and Frost point out, “for critical materialists, society is simultaneously materially 
real and socially constructed.”5 Social constructions are also material, in the sense that 
they contain the same vital excess as any other object: they are not passive recipients of 
human agency, but capable of generating imperatives and complexities outside the 
intentions of the humans who compose and enact them. Nor are they free-floating, 
unmoored from concrete physical manifestations: social relations exist in a dialectical 
interpenetration with physical relations. 
My understanding of materialism is strongly influenced by Marx's. In Marx's 
work, physical objects, such as commodities, take on social relations that are as essential 
to their constitution as objects as their physical relations. Their social meaning is not 
separable from their physical reality: a commodity is not a commodity if it has no 
concrete use; nor is it a commodity if it does not exist within the social relations of 
exchange. The social and the physical are coupled, each capable of impacting and being 
impacted by the other. Nationalism is similarly a hybrid physical/social object: its 
constitution depends on both concrete physical processes and on social discourses.
Along with coupling the physical and social, new materialisms also refuse the 
traditional ontological distinction between passive matter and active humans. Rather, 
human and non-human actors are both capable of exhibiting agency: the “monotonous 
5. Ibid., 27.
14
repetition of inert matter from which humans are apart” is replaced with “active processes 
of materialization of which embodied humans are an integral part.”6 This opening up of 
the agency of non-human (or not entirely human) process has two major implications for 
the study of nationalism. First, it opens the question of the origin of nationalism to new 
influences: if processes are capable of generating novel effects beyond the intention or 
control of their instigators, then nationalism cannot be only understood as a result of the 
intentional work of nationalists (indeed—what inspired these originary thinkers?). 
Secondly, it opens the question of nationalism's own impact: if nationalism is itself a vital 
process, what novel relations is it generating? The purpose of this approach is not to 
displace human agency, merely to place it in the context of other forces. 
New materialisms thus opens up different ways of examining how individual 
experience and state power are mediated. “New materialist scholarship testifies to a 
critical and nondogmatic reengagement with political economy, where the nature of, and 
relationship between, the material details of everyday life and broader geopolitical and 
socioeconomic structures is being explored afresh.”7 Attempts to couple these disparate 
scales are not unique to new materialist approaches: Joe Painter, in “Prosaic geographies 
of stateness” emphasizes how mundane, everyday experiences are permeated with social 
relations of the state and create a perception of a singular state institution.8 However, with 
6. Ibid., 8.
7. Ibid., 7.
8. Joe Painter, “Prosaic Geographies of Stateness,” Political Geography 25, no. 7 (September 2006): 
752–774, doi:10.1016/j.polgeo.2006.07.004. Beyond the obvious substitution of nation for state, my 
project is differs in making a distinction between conceptions of a singular nation, and a singular 
conception of nation: simply because a nation (or state) is conceptualized as unitary does not mean 
that everyone conceives the same unitary object. Secondly, I give more weight to the reality of the 
nation-state than Painter: for all its heterogeneities of practice and conception, it does form a relatively 
dense nucleus around which these conceptions circulate. The unity of the state does have a real, 
though not an ideal, existence.
15
its singular understanding of social and physical, new materialisms offers a unique 
perspective on the problem. The capacity to clearly conceptualize the relation between 
extremes of political scale—the individual and the state—is of particular importance to 
the study of nationalism.
One aspect of this approach is the increased attention within new materialist 
ontology on the body. “For new materialists, no adequate political theory can ignore the 
importance of bodies in situating empirical actors within a material environment of 
nature, other bodies, and the socioeconomic structures that dictate where and how they 
find sustenance, satisfy their desires, or obtain the resources necessary for participating in 
political life.”9 The processes of acculturation, both pragmatic and ideological, by which 
people come to understand themselves as members of a nation (or not) are deeply 
entangled with corporeality. By collapsing distinctions between discursive and physical 
objects, a materialist approach is able to put the state and the individual in remarkable 
proximity.
A materialist approach thus offers a concrete set of approaches to rethink the 
nation as a nationalizing project. It is a method of analysis which can span the gap 
between the individual and the state, that can conceptualize the material as a 
simultaneously discursively social and concretely physical reality. This approach can 
revitalize the question of nationalism's emergence, examining the various processes of 
nationalization that collectively constitute the nationalizing project as both discursive and 
physical processes, capable of generating homogenizing effects at one level and 
heterogeneity at another. Nationalism itself can be understood as a vital object, constantly 
9. Coole and Frost, New Materialisms, 19.
16
in the process of re-creating itself, though never in quite the same way, and thus capable 
of unanticipated movement.
Yet the nation is rarely approached from this perspective. Instead, the nation is 
treated as a static container and otherwise ignored; it is treated as internally coherent and 
internal variation is waved away; or it is denied any relevance altogether. Emphasis is 
placed on the rhetoric of nationalism and the direction of politicians and thinkers to the 
exclusion of other social processes. In order to fully characterize the gaps left by previous 
approaches to nation, I turn several recent works that address Chinese nationalism 
directly and implicitly.
Nationalism at the edges
The question of nationalism continues to haunt area studies, even in works that do 
not directly address it. Many variations of area studies take the national form as their 
subject, yet when its boundaries determine the field of study, the nation-state's legitimacy 
is often taken for granted rather than entering the analysis as a question needing an 
answer. Alternately, some projects deliberately cut across national boundaries, making 
precisely that which falls outside state control the object of analysis. Even for studies 
which ostensibly avoid the nation-state however, the presence of those boundaries 
remains a constitutive condition that must be addressed. This section focuses on four 
recent works on Chinese subjects, two which take the nation as the field of study without 
ever asking why, and two whose avoidance of the question of national cohesion leaves 
startling lacunae in their otherwise insightful analyses. I argue that these two approaches 
must be synthesized, and that it is the relation between the nation-state's homogenizing 
17
project and the heterogenous reality (before, outside, and within the nation) that must be 
the center of analysis. How does the nation-state produce and reproduce itself against and 
within the non-state, the non-national? Research that fails to address this can only 
produce a partial view of its subject, even if that subject has only a distant relation with 
the nation.
One example of this is Haiyan Lee's Revolution of the Heart.10 Lee's book 
explores the production of new narratives concerning sentiment, or love, in the first half 
of the twentieth century. Lee's account is built around three “structures of feeling,” the 
term Lee uses for loosely-conceived discourses around the subject of love. The three 
structures—Confucian, enlightenment, and revolutionary—have a history, but are not a 
chronological progression, Lee stresses. Nor is each discourse conceptualized as a 
singular entity: each structure of feeling has its margins, full of heretics and critics. Nor 
are they entirely distinct: Lee pays close attention to the spaces where the structures 
mingle and interact, their attempted reconciliations and overt attacks. 
Of the three structures, only one can make a strong claim on the label “native.” 
The Confucian cult of sentiment, developing throughout the Qing dynasty, was itself an 
initially marginal critique of Confucian conceptions of emotion. Rebelling against a 
traditional emphasis on performative emotion, the cult of sentiment argued for the 
relevance of affective emotion, albeit still within the Confucian moral framework. The 
other two structures, however, Lee argues are heavily inflected by the influence of the 
European modern. The enlightenment view of love, as it was produced in Chinese 
literature, was a critique of and a rebellion against the oppression of Confucian practice; 
10. Haiyan Lee, Revolution of the Heart: A Genealogy of Love in China, 1900-1950 (Stanford, Calif.: 
Stanford University Press, 2007).
18
unlike the Qing cult of sentiment, the critique was an existential attack, not an attempt at 
reform. Conceptualizing the individual in terms of a psychosexual subjectivity, free love 
became the purpose and the means of Confucianism's overthrow. 
It is with the third structure, dominant in the thirties and forties, that the affective 
qualities of love and the affective qualities of nationalism become most closely entangled. 
The revolutionary structure of feeling paired romance, in variety of ways, with 
revolutionary fervor: sometimes as a synecdoche, sometimes as a supplement, and 
sometimes as an antagonist. These conceptions owed as much to foreign influences as 
earlier, individualistic “enlightenment” narratives, though the process of adaptation to the 
Chinese context makes characterizing them as imported ideals difficult. Here, the 
material from which the national culture was constructed was located not in China's past 
but in the foreign present.
As sensitive as Lee is to the heterogeneity of the portrayals of love in China in the 
early nineteen-hundreds, her perception does not extend to every aspect of her project. 
While the variations within every structure abound, presenting a seemingly endless set of 
permutations, as one steps back the discourse under examination seems more and more 
tightly bounded. Lee's conception of “the modern” serves as one entry point for critique: 
the term is used as a universal referent with a fixed and singular meaning. While the 
modern is applied in a multitude of different ways by Chinese writers, there is no sense 
that it might have multiple valences even before it reaches China's shores. Another telling 
gap in Lee's account of love in China is a lack of attention to its scope. Lee operates from 
a literary vantage, expressing the views of a variety of writers. The possibility that the 
views of these writers might not be representative of China as a whole remains 
19
unaddressed: despite the fact that these writers are overwhelmingly urban, 
overwhelmingly upper-class, and by definition extremely literate. What of the rural, the 
poor, the illiterate? There is a tremendous slippage in Lee's work between her actual area 
of study—a narrow group of elites located primarily in Shanghai and Beijing—and the 
area of study that she claims—all of China.
This, then is another example of the nation assumed, rather than interrogated, as 
the field of study. Lee's conception of China does not enter into her analysis of it: it is an 
object with pre-determined boundaries whose content is being investigated, rather than an 
object whose content is intimately involved in its self-production. This is particularly 
troubling in a study which is directly concerned with the period of time during which the 
boundaries of the field of study are being furiously debated and which is connected, in 
the form of the revolutionary structure of feeling, with efforts to constitute the field of 
study. Without an acknowledgement and consideration of this problematic, Lee's project 
cannot help but misrepresent the object of its study, to speak for a singular China while 
only addressing a fraction of it.
A work somewhat more attentive to the matter of nation, Paul Cohen's Speaking 
to History, follows the continual and varied deployment of an ancient Chinese story 
throughout the twentieth century. The story of King Goujian served as a symbol for the 
Kuomintang's struggle against Japanese and European imperialism, and then after the 
retreat to Taiwan, their struggle against the People's Republic. On the mainland, it was 
deployed by intellectuals first to valorize the CCP and then later as a form of 
underhanded critique; most recently it has again shifted valences and has become a model 
for individual, rather than national, perseverance.
20
Cohen's impetus for writing the book, he explains, was in discovering the story 
for the first time after decades of studying China—yet also discovering that the story was 
almost universally known among his Chinese friends and colleagues. Many of them 
considered learning the story one of the foundational moments of their growing national 
consciousness.11 Cohen argues that growing up conversant with this story—hearing it 
taught in school, transmitted through the media, shared within the family—is a 
surprisingly universal and important element of Chinese culture as it is understood by 
insiders. This does not mean that the story had a singular meaning within the Chinese 
cultural context; indeed Cohen stresses the variety of ways the story was used, often 
contradicting and critiquing each other. Rather, the story was a shared medium, creating a 
culturally-delimited arena of debate.12 
There is, however, in this account a certain tension. The story is an ancient one, 
going back to the Spring and Autumn period. Yet the story was deployed, in the cases the 
book discusses, in service of a nation-state that was in the late nineteenth century still in 
the process of being initially imagined. The mechanisms by which Goujian's story was 
spread, including state-run schools, newspaper advertisements, radio plays, and CCP-
sponsored theatre troops, were overwhelmingly institutions developed in China only in 
modern times. To what extent, then, is the emergence and constant reworking of the 
Goujian story something happening within a pre-existing Chinese cultural sphere, and to 
what extent was its spread—and the development of the mechanisms of its spread—the 
very process of the constitution of that cultural sphere?
11. Paul A Cohen, Speaking to History: The Story of King Goujian in Twentieth-Century China 
(Berkeley: University of California Press, 2009), 231.
12. Ibid., 169.
21
One of the most persistent problems faced by early Chinese nationalists was 
instilling, or awakening, a sense of Chinese-ness among the populace at large. This is not 
uncommon—a similar challenge faces most nascent nationalist movements. Cohen shows 
that the story of Goujian was self-consciously promoted as a way to do this; specifically, 
it was an example of awakening the Chinese people to their national humiliation.13 Yet if 
the Chinese populace failed to feel the shame of their ongoing national humiliation—a 
humiliation keenly felt by their nationalist intellectuals—it seems likely that it was the 
concept of nation that failed to resonate, not the concept of humiliation. The story of 
Goujian, then, isn't just a window into the nature and range of the Chinese national 
culture, but is an active, productive element of that national culture. They do not “bind 
[pre-existing] national communities together in the present,” but rather in binding 
produce those communities.14 
This may seem like a subtle distinction. Yet it makes a profound difference in how 
the nation is imagined: rather than a pre-determined, static object in which certain people 
exist and certain elements appear, the actions of the people within the nation and their 
choices about what elements to produce and reproduce becomes an active, ongoing set of 
processes. Even characteristics like what are the boundaries of this object are being 
actively produced. The apparently singular nation is opened and a tangle of moving parts 
is exposed for analysis. It becomes possible to imagine the nation as being produced not 
only with stories from before the nation, such as Goujian's, but also with narratives from 
outside the nation as well.
13. Ibid., 39.
14. Ibid., 240.
22
Both Lee and Cohen are exploring an aspect of Chinese culture, attempting to 
represent the reality of Chinese-ness. Yet, by treating the existence of the nation as a 
given, they both miss one of the most essential questions that must be asked about 
Chinese culture: how did it come to exist as any kind of coherent object at all? What are 
the processes by which that boundary was produced, and what are the ongoing processes 
that continue to produce Chinese culture as a coherent whole? Without asking these 
questions, it is impossible to explain, or in a profound way to even think about, their 
inverse: what are the ways in which Chinese culture is not a coherent whole? What are 
the ways in which Chinese culture is not reproduced? What, in other words, lies in the 
periphery of the Chinese nation, both internal and external? The failure to grapple with 
the nation as the natural and presumptive “container of societies” remains prevalent, 
hindering our ability to see the nation's messy margins and fractured internal geography.15
This failure has driven some scholars to adopt a radically opposite approach: to 
set aside entirely the concept of a unitary and homogenous object in order to foreground 
the fragmentary and heterogenous margins. This is a space characterized by a flat 
ontology: power structures emerge and dissipate rapidly, a constellation of different 
processes constantly in flux. Attention is paid to transformation and difference, rather 
than stability and sameness. By doing so, scholars find a very different reality than the 
ordered, often teleological structures typically used to explain the world—more agency is 
attributed to both individuals and material reality than in state-centric, structure-
emphasizing approaches.
15. John Agnew, “The Territorial Trap: The Geographical Assumptions of International Relations Theory,” 
Review of International Political Economy 1, no. 1 (April 1, 1994): 53–80, doi:10.2307/4177090; 
John Agnew, “Still Trapped in Territory?,” Geopolitics 15, no. 4 (2010): 779–784, 
doi:10.1080/14650041003717558. In the latter, Agnew emphasizes the role of the national in 
conceptions of state-as-container.
23
One, perhaps extreme example of a research project warped by its unwillingness 
to address the nation-state's role is Mei Zhan's Other-Worldly: Making Chinese Medicine 
Through Transnational Frames.16 In this book Zhan traces the set of interactions between 
a diverse set of practitioners connected in a network spanning Shanghai and the Bay Area 
and how their practice serves to “world” traditional Chinese medicine. By the process of 
“worlding,” Zhan does not mean the globalization of traditional Chinese medicine: that 
implies that there is an extent object that is being distributed on an increasing scale. Zhan, 
however, is interested in the way in which traditional Chinese medicine is made through 
this process of worlding—traditional Chinese medicine is in Zhan's conception 
constituted by the specific encounters through which it is enacted. Thus, Zhan's 
understanding of traditional Chinese medicine allows for (insists upon) a substantial 
amount of internal discontinuity and heterogeneity, as well as continual flux and 
evolution.
Zhan's approach emerges in close relation to her research topic. As Zhan shows, 
the worlding of Chinese medicine is taking place via a set of processes that do no easily 
fit into any neat analytical boxes. The process is stubbornly trans-local, emerging out of a 
flow of people and ideas between nexuses located in different countries. Nor does the 
practice itself admit to any easy generalizations: the relation between biomedicine and 
traditional Chinese medicine is hotly contested from both within and without, seen as 
both a corrupting influence and a source of validation. Traditional Chinese medicine thus 
provokes interesting questions about a number of deeply entrenched binaries: East versus 
16. Mei Zhan, Other-Worldly Making Chinese Medicine through Transnational Frames (Durham: Duke 
University Press, 2009), http://site.ebrary.com/id/10350247.
24
West, tradition versus modernity, science versus art. The present moment is without doubt 
a moment of great disruption and flux in the field of traditional Chinese medicine. 
Yet, in analyzing this moment, Zhan fails to examine with sufficient depth other 
moments in the history of Chinese medicine where the role of powerful structures, often 
in the form of states and nation-building projects, shaped Chinese medicine in a very 
different way—a legacy that continues to inform it today. This can be seen—as can 
Zhan's preference to minimize the institutional element of traditional Chinese medicine—
in her decision not to use the acronym “TCM” in her book. Her rationale is that TCM 
implies the particular form of Chinese medicine that was constituted and promulgated by 
the CCP in the post-WWII period. Yet this institutionalized lump in Zhan's effervescent 
worlding is not so easily ignored: the very sites in which she performed her ethnography 
owe their very existence to this legacy. 
To focus on the state-driven element of the institutionalization and standardization 
of Chinese medicine is perhaps to err again too far to the other side. More important, 
arguably, is the drive towards institutionalization and standardization that originates with 
the field of flux and dislocation where Zhan begins. One place where this might be seen 
is in the story of the founding of the first colleges of Chinese medicine in Shanghai, 
referred to as laosanxiao. Zhan mentions that these schools were founded as part of a 
nationalistic wave of protest against the growing presence of Western biomedicine: a self-
institutionalization, consciously appropriating the norms of its rival. The resonance 
between this and more explicitly state-building nationalizing projects is striking, despite 
the difference in scope and scale.
25
Nor is the American side of traditional Chinese medicine free of institutionalizing 
and standardizing processes: American schools, insurance companies, governments and 
even biomedical practitioners are all engaged in creating standards for Chinese medicine. 
Zhan's attention to the subtleties of the creation and recreation of traditional Chinese 
medicine is rewarding: she uncovers open-ended change where others would see only 
teleological, uni-directional development. Yet despite its lack of teleology, the 
development of Chinese medicine does evidence the consolidation of certain structures 
that do not stop being recreated, do tend to recreate something very like themselves. 
These self-reproducing objects also demands attention. While the disruptions and 
dislocations within the complex networks of traditional Chinese medicine are manifold, 
they are only meaningful in the context of coherence and locatedness. In focusing on 
these elements to the exclusion of the self-ordering and structuring processes, Zhan can 
only create a partial view of the terrain.
Jing Tsu's Sound and Script in Chinese Diaspora is an example of an approach 
emphasizing the margins of the national unit. Tsu's aim is to uncover the complicated and 
contentious process by which the modern Chinese language emerged in the form it has 
today: standardized around a single pronunciation, written in the scripts that it is written 
in, spoken by the population it is spoken by. To do this Tsu radically questions the 
concepts of “mother tongue” and “native speaker,” denaturalizing the assumption that 
language is something “each speaking subject ... come[s] already armed with.”17 Tsu's 
research shows that the production of modern Chinese and modern Chinese speakers was 
a process of mutual production throughout the modern Chinese history; a historical 
17. Jing Tsu, Sound and Script in Chinese Diaspora (Cambridge, Mass.: Harvard University Press, 2010), 
3. 
26
pattern of literacy and oral fluency had to undergo a massive, contentious re-constitution 
in order to become what we refer to as Mandarin today. 
The period Tsu begins in is the same late Qing, early Republican period that Lee 
and Cohen begin with, though her book continues into the modern era, tracing the 
persistently marginal participants in Chinese nativity. Tsu's focus is not the history of how 
the modern Chinese language came together—rather, it is the story of how it did not. 
What she finds in this early period is a wide range of mutually exclusive and much 
contested proposals for how to standardize Chinese, and for whom. The central question, 
echoed in Tsu's title, is the contradiction between orthography and pronunciation: while 
orthography (in the form of classical Chinese) was widely but shallowly dispersed, 
pronunciation varied widely even within dialect clusters, much less between mutually 
unintelligible linguistic families. Tsu pays close attention to the influences these 
seemingly insignificant but stubbornly material factors had on the development of 
modern Chinese. 
Tsu's field for this project is explicitly not the Chinese state: rather, it is the 
“Chinese-speaking world,” a field encompassing multiple state and trans-state 
populations. Tsu's term “literary governance,” she stresses, is not meant to imply top-
down control, but a horizontal power structure of multiple competing and cooperating 
alliances. One recurring theme of Tsu's work is the multiple scales on which linguistic 
identities are negotiated; often emphasizing similarity within to assert difference without. 
Here also Tsu pushes back against traditional state-based histories.
Yet Tsu errs too far in the other direction. In underplaying state power as a force
—a powerful force—in the formation of the Chinese national language, Tsu produces a 
27
distorted image of literary governance as it is and was practiced within the Chinese 
national sphere. One example is found in her treatment of Malaysian Chinese writing, her 
ultimate example of literature beyond the fold of the nation-state. Tsu analyzes at length 
the double marginalization such writers face when writing in Mandarin—yet little depth 
is given to the various projects that created the educational system in which these writers 
were “nativized” into that language. Tsu mentions only briefly the Guomindang-
sponsored schools that taught modern Chinese to Malaysians before the Anti-Japanese 
War, and doesn't speak at all about the powerful Chinese lobby that keeps Chinese 
language education alive in Malaysia.18 
Indeed, one question that Tsu's book does not even mention is that of why the 
drive to standardize a spoken and written Chinese language arose. Outside of the literate 
elite indoctrinated into classical Chinese, the diversity of Chinese dialects had 
presumably been only growing up until the late nineteenth century. Why the sudden 
reversal? As limited as the success of standardization has been in many ways, as many 
margins as it has created and sub-national dialects it has failed to assimilate, the sheer 
fact of the construction and subsequent dominance of a standardized Chinese remains. 
This is not a reality that Tsu's multi-polar, constantly negotiated literary governance can 
easily articulate.
In the cases analyzed above, we have seen how different approaches to the 
questions of nation and structure emphasize different elements of the field, foregrounding 
either processes of cohesion or of disruption. While Lee's analysis is acutely sensitive to 
the subtle variations and imbrications within early twentieth century Chinese literature, 
18. Alan Collins, “Chinese Educationalists in Malaysia: Defenders of Chinese Identity,” Asian Survey 46, 
no. 2 (April 1, 2006): 298–318, doi:10.1525/as.2006.46.2.298.
28
she fails to critically address the question of what binds that literature into a singular 
whole and what lies at the margins, politically and intellectually, of that whole. She thus 
misses the slippage that occurs when Chinese literatures of love are substituted for 
Chinese understandings of love. Cohen's focus on the recurring use of a single narrative 
thread can fill an internally complex but self-referential cultural field with touchstone 
concepts intuitive to insiders but inaccessible to outsiders. However, his assumption that 
the field—the Chinese nation—pre-dates and exists outside of the recurrent narratives 
and touchstones within it blinds him to the process in which the field is actively 
constituted by the narrative's recurring deployment, not merely operating within it. In 
both of these projects the question of structure is treated as an assumption, an organizing 
principle within which the project may be executed rather than as a problematic the 
project must address.
Zhan and Tsu, in contrast, emphasize the processes that are not manifest exercises 
of hegemony, that do not lend themselves to centralization and institutionalization. Zhan's 
analysis of traditional Chinese medicine captures the effervescent, disruptive 
transformation that dominates its present but addresses neither the powerful pull of 
institutionalization that characterized its past nor the processes of standardization that 
form an undercurrent within it today. Tsu's approach to the constitution of the Chinese 
national language is focused on the excesses that fall outside, prior, or are suppressed 
within the standardizing process. Yet in shining light on the periphery, the vast swathe of 
hegemonic space is cast into shadow; the structure-building processes within language, 
and the shifting balance between structure and chaos, remain outside the analytic. Both 
concentrate on the active process of their object's constitution, emphasizing the partial, 
29
contingent nature of its existence. Yet at the same time they fail to address the emergent 
processes by which these diverse, chaotic entanglements of disparate elements are 
constituted into fields; that there are processes which dominate and exert hegemonic, 
structural influence on the world.
It seems that scholars take two approaches to the question of the nation-state: 
either they work within the structure, describing the processes that occur within its 
confines but without addressing how it came to be so or why it continues to be so; or they 
emphasize the making-through, the active constitution of the national, but do not fully 
confront the efficacious structuring power these processes have. Each of these 
frameworks captures accurately one aspect of reality at the expense of the other. To 
understand nationalism, we need a framework that can capture both aspects 
simultaneously, understanding each in the context of the other. 
This is not, sadly, a common quality among the approaches to the study of 
Chinese nationalism. Indeed, Chinese nationalism is a subject which has only belatedly 
become a major area of research within and outside of China: the projects that balance 
recognition of the coherence and multiplicity of nationalism in China are few and far 
between.
Chinese nationalism
Nationalism presents itself as an ontological problem to scholarly projects that 
assume either its “fundamental” validity or its illusory superficiality. For most scholars of 
nationalism, however, the problem is less abstract: the emergence—and therefore 
dynamic movement—of nationalism is the central question. Instead, it is the nature of 
30
that dynamism that presents the challenge: when was nationalism constituted, how, and 
by whom? Consistently, accounts of Chinese nationalism provide rather simplistic 
accounts of when nationalism emerges and whose actions brought it forth. One factor in 
this is a poor integration with theories of nationalism outside of China. Two narratives, 
emphasizing different time periods but similar themes, occupy the bulk of the writings on 
Chinese nationalism. The first of these is the modern narrative, centered around the 
emergence of nationalism leading up to the Republican period under the leadership of 
Sun Yat-sen, Liang Qichao, and other luminaries. Here the anti-imperial impulse is 
foregrounded and the role of an intellectual vanguard is valorized: the history of this 
period emphasizes the role played by various intellectuals in hashing out the core values 
of Chinese nationalism and its gradual coalescence into a singular national consciousness. 
Some writers have challenged this narrative, calling attention to the variety of nationalist 
discourses erased post facto to create the appearance of a linear development towards a 
unified China. Even these accounts, however, do not accord the era's vast socio-economic 
changes a role in changing consciousnesses.
The second, contemporary narrative fast-forwards past the Maoist era straight to 
the opening of China. Here the CCP government plays the active role, propagating anew 
the narrative of humiliation and inculcating a new generation into a propagandistic belief 
in the nation's dependence on the state. Nationalism is here often modeled as an apparatus 
of state control—one, perhaps, on the verge of slipping the CCP's grip, but still 
considered primarily as a political discourse. 
What these narratives share is a discursive perspective: nationalism emerges as a 
result of the action and rhetoric of intellectuals and great leaders, literati and agitators. 
31
Almost entirely absent is a conception of nationalism as an emergent tendency of socio-
economic change; largely absent is nationalism as anti-state populist protest. Nationalism 
is understood as consisting primarily of intentional action on the part of powerful 
institutions or actors, and is analyzed primarily as a discursive object; nationalism's 
relation to material processes such as the development of mass media, popular education, 
mass transport, or industrial employment are only occasionally and unsystematically 
addressed.
These omissions become startling lacunae when these accounts of Chinese 
nationalism are compared with those by scholars of nationalism in the abstract. These 
accounts, heavily influenced by European nationalisms, emphasize the importance of the 
societal shifts resulting from industrialization in the emergence of nationalism. Rather 
than confining the analysis to the intellectual and popular media in which nationalism 
explicitly appears, this approach emphasizes the degree to which ideological and political 
currents are interconnected with every aspect of social organization, from educational 
institutions, modes of transport, and daily work. Nationalist discourse, in this view, is one 
aspect of a vast social transformation and cannot be understood apart from it; overt 
nationalist rhetoric is its most obvious manifestation, not its whole. 
Chinese nationalism, it is generally accepted, had its origin in the late Qing under 
the threat and influence of foreign imperialism. Though there has been some debate 
whether anti-Manchu sentiments—reflected in the actions of literati and secret societies
—evidences a nationalism of sorts, the general consensus has been negative.19 While 
19. Prasenjit Duara, Rescuing History from the Nation: Questioning Narratives of Modern China 
(Chicago: University of Chicago Press, 1995); Lucian W. Pye, “How China’s Nationalism Was 
Shanghaied,” The Australian Journal of Chinese Affairs no. 29 (January 1, 1993): 107–133, 
doi:10.2307/2949954.
32
there is an idea of Chinese-ness at stake in this earlier period, it is reflective of a cultural 
and civilizational chauvinism rather than a nationalist imperative. Nationalism has 
therefore been largely understood in the Chinese context as at least initially a foreign 
import, though in its domestic development it took many unique turns. This foreign 
influence can be seen in the personal histories of many of China's early nationalists: 
Liang Qichao, known for introducing the concept of minzu, or nation, into Chinese, 
studied western ideologies and travelled in Canada, the United States, and Australia.20 
Sun Yat-sen studied in Hawai'i as a child, as well as traveling widely throughout his life; 
an entire generation of Chinese students were sent abroad to study in Japan, the US, and 
Europe during the early 19th century. 
As Prasenjit Duara documents, conceptions of nationalism in the first half of the 
twentieth century were unusually diverse.21 Several problematics animated early debates 
about Chinese nationalism: what was the proper scope of Chinese nationalism, i.e. did it 
include ethnic minorities (in particular the Manchu) or only the Han? What was the 
proper form of government for China: monarchy, republicanism, or federalism? Was 
nationalism already evident in among the Chinese, and if so, where? While it would be a 
dangerous mischaracterization to portray modern Chinese nationalism—on indeed any 
nationalism—as a homogenous, coherent whole, the debate over early Chinese 
nationalisms encompassed a menagerie of perspectives that were incompatible to their 
cores. It is difficult to find any point of universal agreement, other than a) China exists 
and b) it is under attack. This period was evidently a crucial one for the formation of 
20. Suisheng Zhao, “Nationalism’s Double Edge,” The Wilson Quarterly (1976-) 29, no. 4 (2005): 76–82.
21. Duara, Rescuing History from the Nation.
33
political identities.22 The newly formed identities did not always correspond, however: 
not even the primacy of Chinese identity over provincial identity was certain.23 However, 
even Duara confines his analysis to the rhetoric, rarely turning to the rapidly-shifting 
material conditions under which these debates were taking place.
The richness of these elite debates has led many writers to focus their attention on 
them, and to analyze the actions of the populace primarily in terms of whether or not and 
which rhetoric succeeded in mobilizing them.24 The apathy and general lack of national 
consciousness among the peasantry was posed as a central challenge of Chinese politics 
by elite nationalists.25 Counter-intuitively, some turned to tradition to awaken them: Sun 
Yat-sen for a time considered the anti-Manchu avowals of the secret societies, as 
attenuated as they had become by the late Qing, a prototype for ethno-nationalism. This 
avenue ultimately proved fruitless, and Sun abandoned his attempt to enlist the societies 
in his movement.26 
Anti-imperialism proved a more fertile ground for the instilling of national zeal. 
Zhao argues that it was precisely the ability of the Communists to mobilize a mass 
nationalism, in the form of anti-Japanese struggle, that allowed them to ultimately win 
out against the KMT.27 While Mao Zedong's leadership is often understood within the 
22. James Hugh Carter, Creating a Chinese Harbin: Nationalism in an International City, 1916-1932 
(Ithaca: Cornell University Press, 2002); Rana Mitter, The Manchurian Myth: Nationalism, Resistance 
and Collaboration in Modern China (Berkeley: University of California Press, 2000).
23. Wei Liu, “Wan Qing ‘sheng’ yishi de bianhua yu shehui bianqian [晚清‘省’意识的变化与社会变迁],” 
Shixue Yuekan 史学月刊 no. 05 (1999): 59–65.
24. Mary Backus Rankin, “Nationalistic Contestation and Mobilization Politics: Practice and Rhetoric of 
Railway-Rights Recovery at the End of the Qing,” Modern China 28, no. 3 (2002): 315–361.
25. Cohen, Speaking to History.
26. Duara, Rescuing History from the Nation.
27. Suisheng Zhao, A Nation-State by Construction: Dynamics of Modern Chinese Nationalism (Stanford, 
Calif.: Stanford University Press, 2004).
34
context of Marxist-Leninist ideology, for Mao communism was always firmly ensconced 
in nationalism. Particularly prior to Communist victory in 1949, raising national 
consciousness regularly took precedence over raising class consciousness.28 While 
Chiang Kai-Shek prioritized elimination of the CCP over fighting the Japanese, the CCP 
advocated a united front and executed a rural land reform that simultaneously bought 
loyalty from an enriched poor and middle-class peasantry and developed social 
institutions enabling those peasants to mobilize in support of the CCP. However, the 
agency of the peasantry in shaping that nationalism, and the material effects of the social 
changes wrought by mass mobilization are rarely central to the discussion: it remains 
Mao's story.
Many, if not all, accounts of Chinese nationalism minimize or leave out entirely 
the period of Mao's rule.29 This lack of interest has several possible explanations. Among 
foreign China specialists, the Chinese case was too particular, both in the nature of its 
nationalism and of its communism, to be straightforwardly compared to other countries: 
thus China escaped the modernization/nationalization analytic that dominated much of 
area studies.30 Within China, a similar lack of attention held, perhaps due to the internal 
emphasis on communist ideology. Only after 1980 did Chinese language work on 
Chinese nationalism begin to appear.31
28. L Ladany, The Communist Party of China and Marxism, 1921-1985: A Self Portrait (Stanford, Calif.: 
Hoover Institution Press, Stanford University, 1988).
29. Maria Hsia Chang, Return of the Dragon: China’s Wounded Nationalism (Boulder, Colo.: Westview 
Press, 2001); Peter Hays Gries, China’s New Nationalism: Pride, Politics, and Diplomacy (Berkeley: 
University of California Press, 2004); Zhao, A Nation-State by Construction.
30. Pye, “How China’s Nationalism Was Shanghaied.”
31. Houli Luo, “Cong sixiang shi guanjiao kan jindai Zhongguo minzu zhuyi [从思想史视角看近代中国民
族主义],” Strategy and Management 战略与管理 no. 01 (1998): 103–110.
35
Several trends have thrust nationalism back into the limelight post-1978: the 
government began the Patriotic Education Campaign in the early nineties, promulgating a 
new primary school curriculum designed to instill patriotism in Chinese youth.32 In the 
mid-nineties, the book China Can Say No became a massive bestseller in the PRC 
arguing a strongly nationalistic stance.33 Bridging popular and state-led nationalisms, 
popular protests against Japan went, over the course of the nineties and the oughts, from 
government-initiated to spreading beyond the government's control. There is no doubt 
that nationalism is playing a larger role in Chinese domestic and international politics; the 
only question is why.
The dominant narrative centers on an “instrumental” view of nationalism: 
“...nationalist consciousness is seen as a consequence of the historical context in which 
some interests or political forces successfully imagined a political community or national 
history and persuaded people of artificially shared origins that they were indeed one 
people: a nation.”34 Two elements deserve emphasis here: firstly, in this view nationalism 
is driven by the self-interest of an elite; secondly, it is driven primarily via suasion and 
rhetoric. This is the view mostly commonly found outside of academic writings, and 
quite often within it; the “resurgence” of Chinese nationalism is a self-conscious project 
of legitimation by the CCP to fill the gap left by the collapse of communist idealism. 35 In 
32. William A. Callahan, “History, Identity, and Security: Producing and Consuming Nationalism in 
China,” Critical Asian Studies 38, no. 2 (June 2006): 179–208, doi:10.1080/14672710600671087; 
Zheng Wang, “National Humiliation, History Education, and the Politics of Historical Memory: 
Patriotic Education Campaign in China,” International Studies Quarterly 52, no. 4 (2008): 783–806.
33. Gries, China’s New Nationalism.
34. Zhao, A Nation-State by Construction.
35. Kjeld Brødsgaard and David Strand, eds., Reconstructing Twentieth-Century China : State Control, 
Civil Society, and National Identity (Oxford: Oxford University Press, 1998); Chang, Return of the 
Dragon; Wang, “National Humiliation, History Education, and the Politics of Historical Memory.”
36
this discourse, the Chinese citizens enacting the nationalism are no more than “puppets in 
the hands of the Communist elite.”36 Nationalism is reduced to a function of state power.37
Less prominent, though influential, narratives argue a more “authentic” origin for 
popular Chinese nationalism; the stirrings of democracy in the eyes of some and the 
beginnings of a dangerous ethnocentrism for others. Gries argues that the balance of 
power, once tipped firmly in favor of the state, has gradually shifted towards the popular 
nationalists, requiring delicate negotiations between state and popular demands. 
Regardless of what position they take on mass nationalism, however, few of these writers 
look beyond nationalism's discourse about itself: it remains a rhetorical object. 
Throughout the period of its existence, Chinese nationalism has primarily been 
understood in terms of state-led development of nationalist consciousness, usually 
through the mechanism of overtly nationalist rhetoric. Even when the role of popular 
nationalism has been addressed, the analysis has still focused on rhetoric and self-
depictions of overt nationalism. This focus on the rhetoric of nationalism has left a large 
hole in the historiography of Chinese nationalism; i.e. the period between 1949 and 1978, 
when Maoist thought dominated the state rhetoric in China as well as the analysis of 
foreign observers. Lack of material, therefore, led to a dearth of analysis. 
However, it only makes sense to think of the Maoist period as a gap in the 
development of Chinese nationalism if one assumes that ideologically-driven state action 
is the only process necessary for or capable of producing nationalism. In contrast, 
theories of nationalization derived from the non-Chinese experience often include social 
36. Gries, China’s New Nationalism.
37. John Fitzgerald, “The Nationless State: The Search for a Nation in Modern Chinese Nationalism,” The 
Australian Journal of Chinese Affairs no. 33 (January 1, 1995): 75–104, doi:10.2307/2950089.
37
processes outside the realm of state propaganda, relating to changes in the fabric of 
everyday life, that nonetheless play a significant role in the emergence of nationalism. 
Rather than focusing on the deliberate inculcation of nationalism, they highlight the 
inadvertent and accidental ways that the exercise of state power and the subsumption of 
society under capital can create the conditions for the emergence of nationalism. 
The necessity for rethinking the state-nation relation becomes clear when 
contrasting the state's role in the early 20th century and its role in the late 20th. The state's 
impact in the formation of these early nationalisms is clearly significant, though more 
akin to the role of a grain of sand in the formation of a pearl than that of a puppet-master. 
Both foreign powers and the Qing were foils against which nationalists rose, not their 
keepers. This is not an unusual situation for nationalism; indeed it is far easier to find 
examples of nationalism emerging against the state than in support of it. It does pose a 
problem for those who view Chinese nationalism primarily as a tool of the state. 
The state's impacts on nationalism are not reducible to propaganda or 
manipulation—it works through the creation of an economic and cultural territory as 
much as through rhetoric. Yet in order to capture the manifold ways state power shapes 
nationalism, the analytic must be capable of articulating the impacts economic policy, 
infrastructural development, education, and other state projects can have on nationalism
—even when they are orthogonal to the state's explicit aims.
What is needed then
To date, the study of Chinese nationalism has centered analysis around the 
evolving discourse of Chinese nationalism, the rhetoric of nationalists, and the direct and 
38
explicit use of state power to instill nationalism. While these are important processes 
within the nationalizing project, they are only a partial view of nationalism. 
What is needed is an approach that recognizes the hegemonic aspirations and 
capabilities of the nation-state but is equally aware of the limits of those; that is attentive 
to the lacunae in state power and the heterogeneity of national identity, but does not elide 
the power that institutionalization can bring to bear. What is needed is an approach that 
conceptualizes structure not as a negation of individual agency, but as a form through 
which agency is expressed. What is needed is an approach that recognizes the complex 
composite that is “the nation” without ignoring that the nation is for all that no less an 
object—if for no other reason than that so many see it as such.
In order to accomplish this balancing act, the nation must be simultaneously 
understood as a range of disparate processes moving often at cross-purposes, distributed 
unevenly over space; yet also as operating under a not-coherent-but-cohering logic; a 
system shaped by but not solely composed of self-reinforcing cycles; a structure in which 
the state dominates but does not determine. 
A materialist ontology allows the study of nationalism to break out of a 
discursive, intentional model of nation-building, and incorporate a wider variety of 
processes and influences. How does our understanding of nationalist development, its 
advances and setbacks, change if rather than tracing rhetoric, we examine changes in the 
social organization of work under industrialization, or the character of mobility? By 
tracing the material social and technological processes by which nationalism is produced, 
a balance can be maintained between the particularities and heterogeneities of the 
different processes while still understanding their function within the larger, 
39
homogenizing process. It is the recognition of the existence of a nationalizing project, 
rather than the existence of a nation.
In the chapter that follows, the details of this analytical method will be worked 
out in greater depth. This begins with a close study of the theories about the social 
impacts behind nationalism's emergence and the transformation of skill that accompanies 
industrialization. The distribution of these skills and their connection to the built 
environment is crucial to producing the territorial element of the nationalist imperative. 
Ultimately these changes involve the state as well, uniting the state, the territory, and the 
people in the form of nationalism. Railroads are considered as an example of the 
simultaneously physical and discursive materiality that constitutes the nation both as an 
ideal, as a people, and as a specific territory. 
40
CHAPTER III
FORGING THE NATION THROUGH RAILS
The need for a theory-driven re-engagement with Chinese nationalism is clear. 
Recovering from a history of neglect, the literature on Chinese nationalism remains 
superficial in its focus on state action and nationalist rhetoric, the most obvious elements 
of nationalism rather than the most central. This misdirected attention distorts even 
projects that do not take nationalism as their core concern. Articulating a theoretical 
approach to Chinese nationalism that encompasses both state and popular action, both 
overt and subtle processes of nationalization, serves not only China specialists, but 
scholars of nationalism at large. China has been excluded from general theories of 
nationalism due to the particularity of the case it presents; yet all nationalism are 
ultimately particular, and the value of understanding the nationalizing project within 
China is self-evident.
Generalized models of nationalism tend to depict it as a suite of mutually-
reinforcing processes more or less centered around the changes of the industrial 
revolution. While some theorists have treated nationalism as a singular process, even a 
teleological evolution towards nationhood, these accounts have been thoroughly 
critiqued: more recent scholarly approaches emphasize the partiality, multiplicity, and 
incompleteness of the various mechanisms which produce the nationalist imperative. The 
first section of this chapter discusses nationalism theory, drawing primarily on Ernest 
Gellner and Benedict Anderson, still two of the best systematic approaches to the 
question of nationalism, in the context of the Chinese experience. The Chinese case is an 
excellent example of the multiplicity of the nationalizing project: some aspects of 
41
nationalism appear much earlier than theory assumes, and others much later. This 
provides a unique opportunity to analyze nationalizing processes not as a cohering whole, 
but as collection of disparate, sometimes conflicting processes operating on distinct sub-
populations at different chronological and spatial scales. By applying nationalism theory 
to China, and the Chinese case to theoretical predictions, our understandings of both are 
enriched and transformed.
A closer attention to heterogeneities within the nationalizing project centers 
analysis on the powerfully entwined concepts of the state, the people and most of all, 
territory. Tracing the origins of nationalism to the revolutions of industrialization and 
imperialism does not explain why these revolutions, which spread past border after 
border, did not obliterate those borders but instead caused them to grow stronger. The 
nationalist imperative demands the co-incidence of these three units: the national state, 
the national unit, and the national territory. The dominance of nation over class remains a 
major challenge for Marxist accounts of the world economy: if the dominant class 
relations are between the working class and the capitalist class, then why does the nation 
loom so large in the minds of workers and capitalists alike? To dismiss the nation as a 
fetish is to miss the point—fetishes still require a real basis for appearing as they do.
My contention is that this nexus can be explained by the immanent logic and 
internal contradictions of capitalist expansion. Following a Gellnerian explication of the 
demands made on the workforce by industrialization, I argue that the development of 
capitalist production simultaneously homogenizes and increases the skills, or knowledge, 
of the labor force working within it. This homogenization also, however, applies to the 
space of production: the human environment both becomes more similar over large 
42
spatial extents and has more abstract knowledge embedded within it. Labor is thus 
simultaneously freed to circulate within a larger social sphere, but still constrained by the 
extent of the social environment their knowledge pertains to. This environment does not 
build itself, and demands substantial maintenance: the state, even if it is not initially 
involved, plays a crucial role in bankrolling and maintaining this infrastructure. The state, 
the people, and their shared territory thus become increasingly interdependent, forming 
the nucleus of the nationalist imperative.
Focus then shifts to one particular infrastructural project: rail. Rail is uniquely 
relevant to the process of territorialization; not only is it a large-scale infrastructure 
project with which the state is inevitably involved, it is also heavily implicated in an 
unprecedented change in the nature of mobility and space.  The material qualities of rail
—the precise joining of rail and engine, the necessity of standardization and central 
coordination force the entire rail network to function as a single machine, producing 
centralization and homogenization across the entire space. Rail also gives rise to new 
perceptual spaces, transforming the relation of traveler and landscape, traveler and fellow 
traveler. Rail projects also help define a sphere of circulation for populations and 
commodities on a wider scale than ever before. Together, these changes give rise to a new 
perception of territory, one which is imbricated with conceptions of nation and identity.
Theorizing the emergence of nationalism
Despite the long history of the debate over nationalism, it has rarely been 
entangled with discussions of China, even of Chinese nationalism. Before attempting to 
apply this theory to China, we must summarize some of the major conclusions theorists 
43
have come to. This section focuses on the theories put forward by two of the best-known 
theorists of nationalism, analyzing their work from the Chinese context and applying 
their thinking to the historical development of nationalism in China. While both of their 
theories tend to underplay the heterogeneity and multiplicity of the nationalizing project, 
their work nonetheless is a crucial starting point for thinking through nationalism. We 
begin with Benedict Anderson.
In Imagined Communities, Anderson puts forward a theory of nationalism based 
around the idea of the “imagined community:” observing that any nation is far too large 
in population for every member to know even a small fraction of the members, he argues 
that national communities are essentially different from the tribal or familial (or in some 
cases, religious) communities that they superseded.38 One of Anderson's central questions 
therefore is: through what mechanisms is this community imagined? By what means is 
the idea of a nation spread? Anderson identifies the emergence of print capitalism as the 
primary dispersal mechanism, operating in three ways: first, it produced a sphere of 
linguistic circulation wider than the spoken dialect but still limited to a particular 
language-field; second, it gave a new fixity to language; and third, it pushed certain 
dialects (those most similar to the print-language) into a position of dominance.39 By 
creating a community of people who read the same books and newspapers, wrote the 
same script, who could participate in a shared culture, print capitalism produced a space 
within which nationalist imaginings could emerge, and a social institution around which 
38. Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism 
(London [etc.]: Verso, 2006).
39. Ibid., 44.
44
they could coalesce. The reach of nationalism is then defined by and confined to the 
reach of the shared print media.
What precisely was the reach of this media? It was delimited geographically by 
several factors, including dialect, distribution and government. But even within the 
territory so defined, print culture was not universally accessible. Printed media are 
constrained by both language and literacy: in order to access the ideas within, the 
audience must know the language and be literate within it. The first limits the reach of 
printed media to those conversant with the language and its associated culture; the second 
limits the media's reach to a particular class within that group. Literacy has always been 
unevenly distributed within society; historically it was confined to an 
administrative/intellectual elite, with its breadth gradually broadening to encompass the 
bourgeoisie and the working class. Apart from the role of formal education, pre-existing 
dialectical variation also created a set of winners and losers with the emergence of a print 
language. The community of literate readers was never representative of the population at 
large in either cultural or class terms: this disjuncture between the imagined and the 
imaginers lingers throughout the nationalizing project. The very process by which the 
nation emerges creates heterogeneity within the national group: it is from the outset 
privileging certain sub-groups above others. Even within the aegis of its homogenizing 
influence, the nationalizing project creates a new set of differences.
While Anderson's theory captures the interaction between capitalism, state-
formation, and cultural change within the development of nationalism, its analysis of 
territory is substantially lacking. This is symptomatic of its focus on language and 
literature, neither of which are bound tightly by state or natural boundaries. While 
45
Anderson pays an admirable attention to the material economic and technological 
structures which enable the seemingly unhindered flow of information, this attention is 
limited. While Anderson recognizes that linguistics are never a perfect match to nations 
(particularly, as Tsu shows, in the case of China and the Chinese language40), Anderson 
does not explain how territorial boundaries emerge independently of linguistic ones. 
Given the centrality of territory in the nationalist imperative, Anderson's focus on 
literature introduces a significant distortion into his model of nationalism: while the idea 
of the imagined community is a powerful one, and shared media is a powerful conduit 
through which that identity is forged and regulated, it is only one mechanism within the 
nationalizing project. Other theories take a broader view of the process: this bird's eye 
view of nationalism highlights a different set of processes, driven by different social 
forces.
One such theory is laid out by Ernest Gellner in his book Nations and 
Nationalism.41 Gellner's account of nationalism is centered around industrialization and 
the social changes that historic shift entailed. Gellner argues that the kind of work, and 
therefore the kind of laborer, demanded by industrial society is very different than that 
demanded by agricultural society. Where agricultural society emphasizes divisions 
between different kinds of work, cultivating specialists who spend substantial periods of 
their lives in dedicated training, industrial societies train their workers very differently: 
the vast bulk of the training people receive is generic, universal, common to every 
worker. Specialization only enters the training process at the very end, if at all.42 The 
40. Tsu, Sound and Script in Chinese Diaspora.
41. Ernest Gellner, Nations and Nationalism (Ithaca: Cornell University Press, 1983).
42. Ibid., 26–29.
46
nature of industrial society demands it—the mercurial, chimeric character of the 
industrial division of labor requires its workers to have a high degree of flexibility, to 
move between different jobs as they are destroyed and created. A second characteristic of 
industrial education is the nature of the skills being trained: no longer are physical skills 
(weaving, farming, smithing, building) the primary content of the education, but 
intellectual and social skills, the “manipulation of meanings” become central. 43 In order 
to perform their daily tasks, workers must be able to reliably communicate and cooperate 
with a vast and fluid range of people whom they have never before met and may not 
know tomorrow, and interact with workspaces embedded with symbolic meanings. Thus, 
a certain kind of cultural literacy (of which linguistic literacy is only one aspect) becomes 
the skill on which a vast amount of labor relies, and the educational system, formal and 
informal, that produces that literacy becomes a vital element of modern society. While 
specialized skills do not vanish, they are build around a shared core of skills of equal or 
perhaps greater complexity; a core that due to its very universality hardly appears to 
constitute “skill” at all.
The mobility of workers also increases dramatically: centers of production draw 
immigration, and as they are displaced by new centers, the population shifts again, 
resulting in the more rapid circulation of population over a wider sphere. Mechanisms 
that enable this heightened mobility—shared language and cultural norms, transportation 
systems and the ability to navigate them—become increasingly relevant to daily life. 
An interesting echo of this shift in the culture of work, albeit at the regional scale, 
can be seen in the work done by AnnaLee Saxenian on the differing trajectories of Silicon 
43. Ibid., 32–33.
47
Valley and Route 128 in the nineties.44 Saxenian locates Silicon Valley's edge over Route 
128 in the sharply divergent regional networks that emerged in each place. Silicon Valley 
culture begin with a core of developers with very similar backgrounds: overwhelmingly 
young, white men from the Midwest who had studied at MIT or Stanford and disdained 
the  Eastern establishment.45 This homogeneity served as the core around which a unique 
regional identity coalesced. Even as small firms multiplied and competed, a sense of 
shared culture and collegiality endured. Employees within Silicon Valley were highly 
mobile and highly networked: workers moved rapidly between firms and maintained their 
connections with past (and potentially future) colleagues. This unique regional structure 
was reinforced by a shared language of work, and also by a sense of identity tied to the 
region rather than the firm. Saxenian quotes one Silicon Valley CEO: “There are a lot of 
people who come to work in the morning believing that they work for Silicon Valley.”46 
Silicon Valley thus serves as an example of how homogeneity, mobility and a shared 
sense of identity interplay to create a highly productive work environment; Route 128, 
Silicon Valley's foil, serves as an example that the extent of this interplay remains 
variable. Silicon Valley, of course, due to its limited size functions as an actual 
community; at the level of the nation, pace Anderson, the shared identity is produced as 
much by imagination as by personal interaction.
Much like Anderson's print culture, however, Gellner's widened cultural sphere is 
still constrained: the culture of signs in which workers have been educated is only 
44. AnnaLee Saxenian, Regional Advantage: Culture and Competition in Silicon Valley and Route 128 
(Cambridge, Mass.: Harvard University Press, 1994).
45. Ibid., 30.
46. Ibid., 37.
48
functional within that culture's boundaries. Beyond those bounds, the use-value of their 
skills rapidly diminishes. In addition, the educational apparatus necessary to produce this 
generic cultural education is complex and expensive: Gellner argues only the state can 
manage it properly.47 Thus the workers are bound in turn to a particular cultural sphere 
and to a particular state. Culture and state become inextricably linked, not just in the 
minds and experiences of a hyper-literate elite, but in the minds and experiences of every 
worker dependent on that shared culture to perform their work. In schools, in the military, 
in the factories, the daily importance of this culture is hammered in again and again, not 
just in overt propaganda but in every facet of everyday life. 
Gellner's work suffers from an overemphasis on the homogenizing tendencies 
within nationalizing project, downplaying the ways in which nationalization fails to 
homogenize and in some ways even produces heterogeneity within the nation. Gellner's 
theory is structured around the demands of industrialization, and thus tends to treat the 
nationalization process as a coherent, goal-oriented process: nationalization happens 
because industry wants to create better workers. Furthermore, homogenization of work 
never goes as far as Gellner suggests: deep distinctions between the working and 
managerial classes persist, and even divisions within the working class (often entangled 
with gender and race) render industrial work an uneven field. Military service, picked out 
by several scholars as a potent process of nationalization, affects only one part of the 
male half of the population.48 Saxenian's work on regional advantage offers one example 
47. Gellner, Nations and Nationalism, 37.
48. Linda Colley, “Whose Nation? Class and National Consciousness in Britain 1750-1830,” Past & 
Present no. 113 (November 1, 1986): 101, doi:10.2307/650981; Daniele Conversi, “Homogenisation, 
Nationalism and War: Should We Still Read Ernest Gellner?,” Nations & Nationalism 13, no. 3 (July 
2007): 371–394, doi:10.1111/j.1469-8129.2007.00292.x. 
49
on the contingent elements of this process as well as its variability.49 Homogenization is 
thus properly understood as a tendency rather than a certain outcome; nationalizing 
processes are partial and contingent, and generate discontinuities as well as cohesion.
In outlining the mutually-reinforcing feedbacks between the transformation of 
work, on the one hand, and education and mobility on another, Gellner leaves little room 
for the independent development of any of these processes: the emergence of widespread 
literacy outside the context of industrial education, for instance, is not considered. The 
transformations come all at once, or not at all. Arguably this is symptomatic of Gellner's 
default Euro-centrism: as is clear from his models of the organization of pre-industrial 
society, his theory is centered on the European context. This Euro-centrism does not 
render Gellner's account inapplicable in the Chinese context; it merely requires judicious 
application. 
Examining the historical development of Chinese nationalism through the lens of 
Anderson and Gellner's theories yields an interesting set of questions and opens a number 
of avenues of investigation. Anderson's theory establishes a new vantage point from 
which to examine the well-trampled literature of Chinese nationalism. Rather than 
focusing on the content of the dialogue within which Chinese nationalism coalesced as a 
political force, Anderson's theory suggests that the infrastructural skeleton supporting that 
dialogue is equally important: how was the shared community of literacy produced? 
While the discourse of nationalism takes place within this matrix, it needn't be explicitly 
nationalist rhetoric that spurs nationalism: romance novels, advertisements or railway 
signage may be equally important in creating the sense of a shared cultural space. If 
49. Saxenian, Regional Advantage.
50
media and language play such a significant role in nationalism, then it becomes 
imperative to establish the scope, both spatial and socioeconomic, of its influence. How 
widespread was literacy in print Chinese? What role did it play in everyday life? How 
important was literacy to navigating different social spaces? The impact of literacy 
needn't be confined purely to the realm of literature: linguistic signs become embedded in 
the human environment in a multitude of ways. Anderson's theory would predict very 
different levels of nationalist engagement depending on these variables, and at first 
glance the relation holds: the coastal regions, where print media flourished and 
abstraction-rich environments like factories and train stations first appeared, were also 
hotbeds of nationalism.
Gellner's theory offers a different angle of entry. While Anderson opens literature 
and signs to analysis from a new angle, Gellner establishes a link between nationalism 
and a myriad of social changes. What has the impact of industrialization as a whole had 
on China over the past century? How has mobility altered, thereby altering people's 
perception of the environment around them? How has industrialization changed the 
breadth and intensity of the circulation of goods and people? Where has industrialization 
been most concentrated, and has there been a correlation with the development of 
nationalism? These two approaches, emphasizing literacy and print culture, and the other 
the everyday experience of industrial work and society, each concentrated in a different 
class, suggests that perhaps different classes are nationalized differently, via distinct and 
even competing processes: how do these heterogenous experiences of the nation affect 
debates over the nature and responsibilities of nationalism? 
51
Combining the study of Chinese nationalism with theories of nationalism 
generated outside China can produce insights for scholars of both fields, suggesting 
avenues of empirical inquiry for those studying the history of Chinese nationalism and 
providing opportunities for fine-tuning the theory of nationalism. However, nationalism 
is not just a transformation of societal organization: it is also intimately connected with 
concepts of the state and the territory. The relation of these to national identity remains 
murky. The relation of territory to nationalism has long been a bugaboo of political 
geography: constantly assumed, but rarely adequately theorized.50 Both Gellner and 
Anderson provide clues, but territory remains the weakest link in our understanding of 
the nationalist imperative; the boundedness of the national space is assumed more than 
explained. In the following section, I argue that the key element relating state, nation, and 
territory is large-scale infrastructure. By developing further the transformation of skill 
that occurs under industrialization, the mediating role infrastructure plays in embedding 
skill into space becomes apparent, clarifying the relation between nationalism's cultural, 
political, and territorial claims.
Infrastructure and the national imperative
In this section I argue for placing infrastructure at the heart of the nationalizing 
project. Infrastructure relates questions of territory, governance, and culture: as its impact 
and necessity grows with industrialization, it exerts a gravitational pull, drawing the three 
closer together. Questions of national infrastructure precede and inform debates about 
nationalism, offering a glimpse into how concretely physical and discursive processes 
entwine. 
50. Agnew, “The Territorial Trap.”
52
The central problematic here is that of skill, and how it is transformed as the labor 
process is subsumed within capitalism. I argue that this transformation has been subtly 
misrepresented in Marxist theory (though not by Marx), which emphasizes the 
diminishment of skill under capitalism and the transformation of abstracted labor power 
into a universal attribute of the working class. Rather than shifting to the working class as 
a whole or disappearing entirely, I follow Gellner in arguing that it is only the 
differentiated guild and craft skills that vanish and they are quickly replaced with equally 
sophisticated, albeit generalized skills that are held in common by the national unit. The 
territorial focus of nationalist movements is accounted for by exploring the nature of the 
built environment as a means of production from the vantage of skill. Finally, the role of 
the state in developing and maintaining an increasingly elaborate built environment and 
thus disseminating new forms of skill entangles the state with the coalescing 
nationalizing project. All of this, however, begins with a re-examination of the nature of 
skill.
The transformation of skill
“Labour,” for Marx, “is, first of all, a process between humanity and nature, a 
process by which humanity, through its own actions, mediates, regulates and controls the 
metabolism between itself and nature.”51 This labour is distinguished from the production 
of animals by its intentionality: “At the end of every labour process, a result emerges 
which had already been conceived by the worker at the beginning, hence already existed 
ideally.”52 With intentionality, we catch a glimpse of the role of skill: how does the 
51. Karl Marx, Capital: A Critique of Political Economy (London; New York, N.Y., USA: Penguin Books 
in association with New Left Review, 1990), 283. Edited to avoid an implicit gendering of labour.
52. Ibid., 284.
53
worker conceive of production, other than through a past knowledge of its details? Marx, 
however, rarely foregrounds skill in his analysis. Instead it is discussed primarily in terms 
of the “average” skill of labour, a decision that still shapes Marxist understandings of 
skill.
The concept of skill as used here has a some degree of commonality with the 
concept of performativity, or Bordieu's habitus, Foucault's discipline or other related 
theoretical apparatuses. Skill emphasizes several important elements particularly 
important when working from a materialist perspective: relationality, agency, and 
embodied-ness. Skill does not exist in a vacuum; it describes a relation between a person 
and an object, or a series of objects. Skill is an inherently active, constitutive force; it 
endows its possessor with new capabilities and widens their choices. Finally, skill is 
materially embodied not just in its possessor of the skill, but also in the objects to which 
it relates, having thus both an abstract and a concrete existence. Skill centers analysis 
around an active, volitional constitution of material reality by agents in a realm they do 
not wholly control but neither are they powerless within.
How then, does skill change under the subsumption of capitalism? One 
interpretation, laid out by Braverman and followed by Harvey, centers around the concept 
of “deskilling.” The term deskilling is misleading; the word refers to the shift from 
traditional craft or artisanal skills to the generalized education characteristic of the 
modern economic mode. This shift, crucially from the perspective of capital, also entails 
a shift from highly specialized and therefore monopolizable skills to generalized and 
therefore substitutable skills.53 This shift has many ramifications: firstly, it removes an 
53. David Harvey, The Limits to Capital (Chicago: University of Chicago Press, 1982), 109.
54
important negotiating advantage from labor in the struggle over wages by making the 
skills necessary to the production process harder for sub-sections of the working class to 
monopolize: workers are always under the threat of being replaced by their fellows in the 
industrial reserve army.54 Secondly, it affords capital much more freedom in the 
revolution and expansion of the labor process, moving workers as market shifts and 
technological advances demand. 
This understanding of the process does not rest comfortably with the term 
“deskilling,” which implies more an absolute loss of skill than a transformation in the 
nature of skill. It is not clear that Braverman or Harvey appreciates fully the 
consequences of this distinction: certainly their language suggests that despite their 
acknowledgement that fully-subsumed, homogenized skill is still a skill, they still view it 
essentially as a diminishment. Braverman says the worker “would sink to the level of 
general and undifferentiated labour power [emphasis mine]”;55 Harvey calls the transition 
to “mere machine-minding skills”56 a “reduction of skilled to simple abstract labour”.57 
The language of reduction conceals the potential for the transformation of necessary 
production skills to increase the absolute level of skill as easily as to decrease it. Indeed, 
it is hard to argue that the intensity and length of the educational process has decreased 
since the industrial revolution.
54. Marx describes the manner in which guilds were structured as designed specifically to avoid 
developing along capitalistic lines and ultimately devaluing their skills; see the example of guild skills 
as “the mysteries” in Marx, Capital, 616.
55. Quoted in Harvey, The Limits to Capital, 110.
56. Ibid., 108.
57. Ibid., 110.
55
While he foregrounds the violence of this transformation, Harvey is not blind to 
its emancipatory potential: “By 'liberating' workers from their traditional skills, capital at 
the same time generates a new and peculiar kind of freedom for the worker”.58 Yet a fuller 
understanding of deskilling as a transformation of rather than a negation of skill
—“reskilling,” perhaps—remains stubbornly in the shadows. Crucially, if skill does not 
disappear but are merely transformed, the role it plays in wage negotiation and other 
aspects of the labour process do not vanish but simply appears at a different scale.59 
Harvey mentions two levels at which skill continues to be relevant: its peculiar liberation 
of the working class, and the lingering artisanal quality of scientific and managerial craft. 
Yet a closer consideration of the levels at which the new skills operate—“literacy, 
numeracy, the ability of follow instructions and to routinize tasks quickly”60—shows that 
they inhere to neither of these scales, but to another altogether: that of the nation. 
Literacy in a print language, the ability to follow instructions given in a common 
language and according to a particular set of norms; all of these are aspects of a national 
culture.61 In slipping the bonds of guild specialization, labour nonetheless remains bound 
up in a more spacious, but no less limited sphere. What is it about skill, which exists 
within individuals, ties it to the particular community of the nation, and to the nation's 
particular territory? To understand why skill takes on the nation's shape, it is necessary to 
examine more closely the nature of the skill-tool relation. 
58. Ibid., 109.
59. Such scale-shifting is, from a certain perspective, implicit in the exhortation of the global working 
class to organize. However, this positions labour skills as shared in common by the entire working 
class—I argue it inheres to a more limited scale.
60. Harvey, The Limits to Capital, 109.
61. Numeracy is notably left out here; while I argue that a great deal of skill and work-discipline adheres 
to the national culture, this is of course not true of all skills.
56
Skill is not abstract—it is concrete, and requires specific tools and particular 
working conditions.62 A spinner requires a spindle; a weaver a loom. Far from 
annihilating this material relation, the transformation to general skills merely alters the 
dimensions along which this interdependency manifests. This need not, and has not, 
produced a general deskilling any more than the drive to reduce labor content in 
commodities has resulted in a Keynesian paradise of fifteen-hour work-weeks: resources 
freed by simplification in one area are simply absorbed elsewhere. One element from the 
above quotation from Harvey provides an excellent example: literacy. Literacy in one 
sense operates to universalize skills; a literate worker with access to manuals can quickly 
learn a range of quite sophisticated skills, operating in a variety of mechanized work 
environments. Literacy itself is no simple skill, however, and indeed the literate worker is 
no less dependent on her tool—the book—than the specialized workers whom she 
replaces. A manual written by someone who does not share her language is useless to her 
in a way that a hammer manufactured by someone who does not share her language is 
not. Language thus serves as a barrier through which certain kinds of labor may pass and 
other kinds may not, strengthening certain relations even as it dissolves others. A number 
of the other skills central to the subsumed labor process are similarly constrained: 
language serves here as an exemplar, but less obvious elements such as a sense of time 
discipline or styles of social organization are also crucial. At this point it becomes clear 
that the line between skill, with its association with tools and work; and culture, with its 
association with custom and society, is more apparent, and less real, than is commonly 
thought.
62. Marx, Capital, 285–86.
57
Thus when Braverman writes that labor “comes ever closer to corresponding, in 
life, to the abstraction employed by Marx in an analysis of the capitalist mode of 
production,” he jumps too quickly from the concrete to the abstract, missing how even 
the most abstracted of concrete labor remains embedded in a complex web of material 
social relations.63 Even in Capital this distinction is less unidirectional, and more 
ambiguous than is commonly understood. Marx ends the chapter on machinery not with a 
prediction that hyperspecialization will continue, but that, like the reckless extension of 
the working day, must reverse itself in order for capitalism to continue.
...large-scale industry, through its very catastrophes, makes the recognition 
of variation of labour and hence of the fitness of the worker for the maximum 
number of different kinds of labour into a question of life and death. This 
possibility of varying labour must become a general law of social production, and 
the existing relations must be adapted to permit its realization in practice. That 
monstrosity, the disposable working population held in reserve, in misery, for the 
changing requirements of capitalist exploitation, must be replaced by the 
individual human who is absolutely available for the different kinds of labour 
required of them.64 
This is not, crucially, Marx making a normative claim about what ought to occur, 
but an observation about what the immanent drives of capitalism demand: an individual 
capable of any kind of work is more useful for capital than one suited, ultimately, to 
nothing.
While capital wishes to avoid complex labor, complex labour is defined by Marx 
as that which has some additional inputs beyond the average, therefore requiring more 
training time and higher wages. The word “average” is crucial: complexity in this sense is 
a purely relative term. If the “average” level of social labor is seven year's of schooling, 
then work requiring eight years is complex; if the average is nine years, then what was 
63. Quoted in Harvey, The Limits to Capital, 110.
64. Marx, Capital, 618. Edited to avoid an implicit gendering.
58
complex has become overly simple. Insofar as more highly trained labor is more efficient, 
there is an incentive for capital to increase the average level of skill. Thus while there is a 
constant drive to force all labor into the form of simple labor, to homogenize it, there is 
equally a drive to improve the efficiency of “simple” labor by raising the level average 
level of skill. 
This immanent drive manifests clearly in the training and disciplinary regimes 
discussed by Foucault: the elaboration of military training in the seventeenth century, in 
which every movement was broken down and carefully instilled in each soldier, and the 
systematization and regulation of education, in which each student was observed and 
controlled, both reflect a considerable increase in level of formal training necessary to 
function in society.65
The possibility of reskilling is reflected in the literature on post-Fordist 
production. Piore and Sabel pioneered the arguments for “flexible specialization” in The 
Second Industrial Divide, sparking a long running debate over the nature of skilled labour 
in the post-Fordist world.66 In their model, the production line model (and its associated 
deskilling tendency) characteristic of Fordist production is being superseded by an 
alternate production method emphasizing dynamic production methods relying on a 
workforce with flexible skillsets. While there has been considerable debate about the 
precise nature and meaning of this shift, the “embeddedness” of economic production in 
social and cultural context has become increasing apparent—a context largely, though not 
65. Michel Foucault, Discipline and Punish: The Birth of the Prison (New York: Vintage Books, 1979).
66. Michael J Piore and Charles F Sabel, The Second Industrial Divide: Possibilities for Prosperity (New 
York: Basic Books, 1984).
59
exclusively, thought spatially.67 In its emphasis on skill, culture, and territory, this body of 
scholarship presents an interesting parallel to the work being done here.
Yet in its treatment of regionality, and of skill, embeddedness scholarship focuses 
on a narrower sense of region, and a more specialized sort of skill, than is our object of 
analysis. Any skill which appears as the regional advantage of one region in a country 
versus another region in that country is, by that measure too specialized to constitute the 
sense of common understandings and acculturations that on the one hand form a sense of 
national identity, and on the other delineate the national territory. It is the skills that do 
not appear at all because they are so omnipresent, that we must excavate. It is the skills of 
basic literacy, ability to work under discipline, to show up on time, and interact within 
workplace culture, that without which an individual appears deficient and cannot function 
in any workplace at all, no matter how “deskilled” the work appears, that constitute our 
object of study here. 
It is from this perspective we address the “freedom” of the worker. The peculiar 
liberation the subsumed labor process offers workers is still constrained by language, 
custom, education—it is by no coincidence that these social boundaries, defined from 
within or without, become increasingly relevant in the political sphere. Yet the 
characteristic claims of nationalism are not yet fully articulated; in particular, the 
territoriality that is so central to the nationalist drive remains obscured. In order to 
understand the importance of territory the skill and tool relation must be applied on a 
67. Martin Hess, “‘Spatial’ Relationships? Towards a Reconceptualization of Embeddedness,” Progress in 
Human Geography 28, no. 2 (April 1, 2004): 165–186, doi:10.1191/0309132504ph479oa; Andrew 
Jones, “Beyond Embeddedness: Economic Practices and the Invisible Dimensions of Transnational 
Business Activity,” Progress in Human Geography 32, no. 1 (February 1, 2008): 71–88, 
doi:10.1177/0309132507084817.
60
macro-scale, bringing the built environment, its function as a means of production and its 
function as a tool in the hands of the populace into the analysis.
The skilled environment
A rich literature addresses the built environment, noting among other 
characteristics its mingling of productive and consumptive uses, its progressive 
deepening under the demands of capital, and its use as a spatial fix. Harvey opens up the 
built environment to analysis as an arena of capitalist investment: the vast slow 
realization of the value embedded in infrastructure serves as a sink for excess capital, 
delaying the onset of crisis.68 For our purposes, the crucial point is that it is not only the 
machinery of the factory floor or cubicle farm that is subsumed by capital and undergoes 
transformation, but the entire infrastructure that society inhabits; by the same token it is 
not just within the narrowly-defined production process that increasingly abstract (but 
still bounded) knowledge becomes necessary to properly function.69 Storper and Walker 
drive this point home: the geographies produced by the built environment are just that: 
produced. They stress the role of capitalism in producing these built geographies and 
driving its expansion, alteration, and even abandonment.70
Thus capital's impact on skill stretches beyond the workroom into the skills that 
mediate the entire built environment: navigation, modes of public sociality, and countless 
68. Harvey, The Limits to Capital.
69. The question may be asked whether the built environment is more complex to navigate or use than the 
natural environment. I do not argue that it is: rather, the character of the skills necessary undergo a 
transformation in two important ways. First, the complexity is no longer produced by non-human 
processes, but by human ones; secondly, the complexity is encoded in abstract symbols—language, 
etc. This entangles an abstract cultural element into the environment-skill relation.
70. Michael Storper and Richard Walker, The Capitalist Imperative: Territory, Technology, and Industrial 
Growth (Oxford, UK; New York, NY, USA: B. Blackwell, 1989).
61
other aspects of everyday life. For all its size, the built environment is simply another set 
of tools, requiring a certain set of skills. Even as from the perspective of capital it is 
necessary that the built environment yield value, and therefore necessary that they be 
properly employed, from the perspective of those inhabiting that environment acquiring 
those skills is just as vital. 
In the same moment the sphere of capital's subsumption of the labor process 
becomes society as a whole, it is inextricably joined to a particular space. The spatiality 
of the built environment necessarily territorializes its corresponding skill set: the skills 
which allow individuals to take advantage of that particular environment will not transfer 
perfectly, and in some cases not at all, to other settings. Insofar as it needs to realize the 
value of the built environment, capital is tied not only to a particular territory but to a 
particular population, and the population is tied to that territory as well. Their interests 
are entangled in a particular space, and with a particular set of skills, or culture. In order 
to maintain people's livelihoods, the built environment must be maintained: if it is too 
radically transformed or let fall into disrepair their hard-won skills will become 
worthless. The outlines of nationalism begin to emerge out of the logic of capital.
One might argue: such society-shaping infrastructure might not be maintained on 
a national scale; cities in particular are characterized by complex lumps of interdependent 
infrastructural systems that do not extend into the nation as a whole. Sometimes the 
cultural skills embedded in the built environment have a tighter scope than the nation. 
Indeed, much of the literature on the embeddedness of economic production has focused 
on a regional or trans-regional scale.71 In her classic study, Saxenian contrasts two regions 
71. M Storper, “The Resurgence of Regional Economies, Ten Years Later: The Region as a Nexus of 
Untraded Interdependencies,” European Urban and Regional Studies European Urban and Regional 
62
within the United States with very different regional environments that produce very 
different systems of production; her collaboration with Jinn-Yuh Hsu focuses on linkages 
between Silicon Valley and Hsinchu, Taiwan.72 While testifying to the power of “untraded 
interdependencies,” in Storper's words, the regional focus of these studies highlights the 
absence, at the regional scale, of the nationalizing skills I am interested in excavating. 
These skills cannot appear at the regional scale because due to their very universality, 
they do not appear as “skill” at all. By the same token, they cannot create regional 
advantage, because they are held in common across the nation.
The state and infrastructure
The state's role in the production and maintenance of this nation-producing 
infrastructure is without question significant: case after case of nationalism coalescing 
within state boundaries—even when the ethnic characteristics upon which nationalism is 
purportedly founded do not correspond—suggest that state influence over the formation 
of the nation is profound. There are several mechanisms through which this influence is 
exercised. The first is the role played by the state in incubating capitalism in its earliest 
stages—the extent to which capitalism began in Europe as a tool of the state in intra-state 
conflict is rarely adequately addressed in Marxist theory.73 The state system, while 
Studies 2, no. 3 (1995): 191–221; Trevor J Barnes and Meric S Gertler, The New Industrial 
Geography: Regions, Regulations and Institutions (London; New York: Routledge, 1999); Costis 
Hadjimichalis, “Non-Economic Factors in Economic Geography and in ‘New Regionalism’: A 
Sympathetic Critique,” International Journal of Urban and Regional Research 30, no. 3 (September 
2006): 690–704.
72. Saxenian, Regional Advantage; Jinn-Yuh Hsu and AnnaLee Saxenian, “The Limits of Guanxi 
Capitalism: Transnational Collaboration between Taiwan and the USA,” Environment and Planning A 
32, no. 11 (2000): 1991 – 2005, doi:10.1068/a3376.
73. Charles Tilly, Coercion, Capital, and European States, AD 990-1990 (Cambridge, Mass., USA: B. 
Blackwell, 1990).
63
eventually modified substantially by capitalism, precedes it and profoundly shaped its 
structure. More tangibly, the state plays an essential role in shaping the skills of the 
national working class via the construction formal, standardized education. While the 
informal habituation and indoctrination of the workplace is also critical to shaping the 
new working class, as production grows increasingly sophisticated the educational 
limitations of working children from the age of five became increasingly clear, as Marx 
records.74 It is no more beneficial for capital than the fourteen hour working day; greater 
profit is to be had from an educated workforce. As with many of the collective action 
problems that confront capital (where the benefit is too diffuse for a single capitalist to 
profitably capture), the state takes the lead in building the infrastructure from which all 
capitalists benefit. In doing so, the state plays a significant role in delineating the space 
within which that skillset holds sway: coincident with the state boundaries. 
Most relevant to our focus, however, are the many state-led or -associated 
infrastructure projects: investments in rail networks, roadways, communication systems 
and so on often involve a substantial state regulatory role, if not direct state investment or 
management. Consequently, they are also typically fitted to state boundaries. The 
management and maintenance of infrastructure may be seen as one arena for the 
everyday production of “state effects.”75 Yet, operating within a materialist ontology, the 
origin of these projects in the state does not mean that their result—the gradual 
coalescing of a nationalist consciousness—was in any way intended. Linda Colley's 
examination of early nationalism in England draws this distinction quite clearly: while 
state projects, such as the military mobilization of the male populace played a key role in 
74. Marx, Capital, 370.
75. Painter, “Prosaic Geographies of Stateness.”
64
the emergence of a popular nationalism, that nationalism was an unintended and 
unwelcome consequence.76 The state's influence on nationalism thus occurs not directly, 
but at a slant.
This infrastructure often relies on the standardized education discussed above: 
communication is transmitted in the state language, or signs are written in the state script. 
While these signs are the most obvious way that cultural skills are materially embedded 
in the built environment, they are in some sense the least significant: it is that which goes 
entirely unmarked and implicit that demands the greater skill on the part of the 
individual. Taken as a whole, this process of investment and education creates a 
homogenized space and a homogenized population within which there is a great deal of 
flexibility, inter-communicability and circulation of population—all of which end sharply 
at the state's border.77
The relation between the cataclysmic restructuring of the production process 
during its subsumption under capital and the emergence of the nationalist impulse has 
been sketched in rough outline. The demands of capital to reshape the labor process to its 
own needs and away from the monopoly of artisan craft entail a drastic homogenization 
and generalization of skill. Yet the abstraction of skill continues to depend on a concrete 
set of linguistic and social relations, displacing but not eliminating the specificity 
inherent to the production process. This shift relates not only to the tools within the 
immediate production process but also applies to the human environment within which 
76. Colley, “Whose Nation?”.
77. The implicit contrast here is with pre-capitalist models of territorial control, where both state influence 
and cultural homogeneity are concentrated in state centers and blur out or vanish entirely towards the 
edges of allegedly state-controlled spaces. cf. James C Scott, The Art of Not Being Governed: An 
Anarchist History of Upland Southeast Asia (New Haven: Yale University Press, 2009).
65
both productive and consumptive activity take place; the state's role in developing that 
large-scale infrastructure that comes to characterize that environment creates a substantial 
co-incidence between it and the state's borders. This built environment, studded with 
culturally-specific abstractions and assumptions, plays an ever greater role in the 
everyday life for its population: work, leisure, communication and circulation all take 
place within its confines and are bounded by its boundaries.
All kinds of state-scale infrastructure projects contribute their own nationalizing 
effect, creating somewhat real and somewhat imagined shared spaces, shared senses of 
culture, shared spheres of communication. While they are never as complete as they 
represent themselves to be, and they are collectively less coherent than they seem to be, 
they nonetheless create a tangible push towards homogenization and standardization, 
coalescing into what I call the nationalizing project. One type of infrastructure, however, 
plays a double role, both creating a homogenization of environment and experience over 
a broader space and also increasing the circulation of commodities and populations 
throughout that newly unified territory. Transportation infrastructure, particularly rail, 
thus serves as a particularly apropos starting point for examining how these abstract 
nationalizing processes materially produce new kinds of spaces and experiences.
The impact of railroads on the production of space
The emergence of railroads radically altered the relation between time and 
distance, between humans and nature, collapsing vast chronological and psychological 
gulfs to a fraction of their former distance. It was essential in producing a new kind of 
travel, over a new kind of geography. While many heralded the arrival of the rail as the 
66
“annihilation of time and space”, it produced instead a new time and a new space. For rail 
did not open all the world all at once: rail collapsed distance only selectively, only where 
massive infusions of capital laid the groundwork. The demanding precision of rail 
infrastructure—the precise joining of railway and railcar, the increasing fine tolerances 
within the machine, the demand for ever clearer lines of communication, and the 
horrendous damage incurred when any of these failed—all combined to make the rail 
network function less like a disparate collection of roads and more like a single machine 
operating under centralized control. The development of rail instantiates in reality the 
abstract logic of industrialization and capitalism outlined earlier: the drive towards 
homogenization of space and standardization of skill under centralized control. Where the 
rail ran, circulation of people and goods accelerated as the friction of distance decreased, 
creating an ever wider, less differentiated marketplace; consumptive and productive uses 
entwined in a uniquely close embrace. Rail, as a mechanism within the nationalizing 
project, calls to the fore its centralizing, homogenizing, and standardizing tendencies. 78 
Not only was the relation between existing spaces radically reworked, entirely 
novel spaces were also produced in the form of the rail carriage and the rail station. 
Inhabiting these new spaces required new modes of perception, new norms of 
interpersonal interaction: in sum, a novel set of skills. Skill, here, should not be 
understood merely or even primarily as that of the rail operators, but of the ridership: 
those skills which, due to their very ubiquity, tend to go unrecognized. The public's 
perceptions of the rail journey, as well as all those skills necessary to successfully 
78. This is not to be understood as a claim that all mechanisms of nationalization have these qualities, or 
even that rail contains solely these qualities. An attention to the heterogeneity produced even by 
homogenizing processes is a central goal of this thesis.
67
navigate the rail system, also constitute an important body of skill with society-wide 
implications. Other state interventions collaborated to shrink this perceptual distance 
further: standardized language and orthography ensured that travelers could speak and 
read where ever they traveled—within the state, of course. Without traveling to every 
stop within the state, one can nonetheless get the sense that travel anywhere would be 
essentially similar, allowing the traveler to imagine far-off locations in a new way: as in 
some sense, part of the same spatial entity, the same geo-body. This sense may not be 
accurate, either in the ease of travel or in the similarity upon arrival; the perception is 
nonetheless potent. Rail thus creates an imagined territory, in the same vein as Anderson's 
imagined community: while there is no way to visit every part of the national geo-body, 
just as there is no way to know every member of the nation, it becomes possible to 
vividly imagine a relation to, and identity with that unknown place.
In the section on rail my engagement with materialism is clearest. As discussed in 
Chapter II, materialism takes seriously the active capacities of matter to shape the world: 
a vital excess inheres in everything. A core assumption of this work is that rail, and the 
nation, are material processes and thus partake of the same vitality. Rail, in its 
reorganization of territory and rerouting of social life, has the potential to work changes 
far beyond the imaginings of those who build it; and the nation, composed of the 
countless imaginings of countless individuals, is capable of transforming in ways 
unanticipated by those whose rhetoric gave it shape. Technological limitations have 
ideological implications, and economics can shape psychology; gauge changes shape 
social boundaries and psychological needs can impact carriage design. 
68
It is around this core that I attempt to integrate economic, political, technical and 
social aspects of rail's development into a single analytic in order to understand the 
impact of rail on nationalism. While the large-scale social impacts of rail networks are 
widely acknowledged in economic geography, rail's role in nation-building is rarely 
analyzed in more than passing. In Capturing the Horizon, James Vance details the 
development of rail networks within the container of the nation—appropriately, as the 
state played a direct role in many lines—but dwells on their economic, not social, impact. 
This is despite such headings as “The Railroad as Liberator and Unifier: Italy.”79 Meinig, 
in the magisterial The Shaping of America, notes unreflectingly that the foremost goal of 
a transcontinental railroad was to “'bind society together in all its parts' until it becomes 
'coextensive with the boundaries which embrace the American family,'” but offers little 
analysis on how rail accomplishes this other than through brute economic union.80 
Schivelbusch's The Railroad Journey is rare in the equal weight it gives to rail as both a 
technological and a social subject; this section relies heavily his account.81 
More recently, some of the work associated with the “new mobilities” paradigm 
has begun to unpack the implicit skills of travel and the role of the traveler in producing 
the rail experience.82 The mobilities paradigm, however, often begins from the 
79. James Elmon Vance, Capturing the Horizon: The Historical Geography of Transportation since the 
Transportation Revolution of the Sixteenth Century (New York: Harper & Row, 1986), 246.
80. D. W Meinig, The Shaping of America: A Geographical Perspective on 500 Years of History, vol. 3 
(New Haven: Yale University Press, 1986), 6.
81. Schivelbusch, The railway journey.
82. Laura Watts, “The Art and Craft of Train Travel,” Social & Cultural Geography 9, no. 6 (2008): 711–
726; Laura Watts and John Urry, “Moving Methods, Travelling Times,” Environment and Planning D: 
Society and Space 26, no. 5 (2008): 860 – 874, doi:10.1068/d6707; Mimi Sheller and John Urry, “The 
New Mobilities Paradigm,” Environment and Planning A 38, no. 2 (2006): 207 – 226, 
doi:10.1068/a37268; David Bissell, “Vibrating Materialities: Mobility–body–technology Relations,” 
Area 42, no. 4 (2010): 479–486, doi:10.1111/j.1475-4762.2010.00942.x.
69
assumption that everything is always already mobile, ignoring the importance of relative 
immobility in producing other kinds of mobility: for example, the role of travel 
infrastructure, itself constantly evolving, in producing the mobilities of travelers.83 To 
differentiate between mobility in these two registers, this section introduces the concept 
of spheres of circulation: spaces produced by the construction of travel infrastructure 
which enables circulation of populations and commodities within. 
Rail operates in an essentially different way than most forms of travel. Vance 
draws a distinction between the facilities of travel, the road or sea over which travel takes 
place, and the technologies of travel, the means of mobility themselves.84 In most 
transportation networks, these interlocking systems exhibit some degree of autonomy, 
and the management and operation of the facility and the management and operation of 
the technology can be, and typically are, separate. Early in the history of the the railroad, 
it became clear it could not function this way: a multitude of factors conspired to force 
the ever-closer joining of conveyance and roadway both technically and administratively. 
The technical demands of the railroad, the required close conjoining of the railroad and 
the engine on it, allowed scarce leeway even at its earliest stages. As rail technology 
developed, the coupling between rail and the engine became only more precise. The 
precision needed went beyond the contact-point between wheel and rail: as trains moved 
faster, with more momentum, carrying more people, a revolutionary level of  
communication was also necessary to avoid the most hideous of accidents. Trains could 
not give way, nor easily slow; preventing collisions required the construction of a 
83. Peter Adey, “If Mobility Is Everything Then It Is Nothing: Towards a Relational Politics of 
(Im)mobilities,” Mobilities 1, no. 1 (2006): 75–94, doi:10.1080/17450100500489080.
84. Vance, Capturing the Horizon, 4.
70
secondary network of telegraphs paralleling the rails.85 Safety depended not on the 
vigilance or skill of the engineer, but on the proper functioning of a vast network of 
signals. The refrain of “one machine” was heard again and again, re-iterating not only the 
strength of this imperative, but also its novelty. The implications of this technology were 
only gradually worked out, and different answers emerged in different locales.
Rail's unity of facility and technology, to borrow Vance's terms, emerged as a 
technical necessity before it became a legal reality. It was first given legal sanction in 
Britain in 1840 with a law granting railways a legal monopoly on transport on their own 
line, an unusually direct intervention by the laissez-faire British state.86 (Previously, any 
operator was allowed to use the track—a vision that now strikes terror into the heart of 
any railway passenger.) It is in Britain that the consolidating logic of rail is most evident: 
emerging piece-meal, without precedent, in an atmosphere strongly inclined towards 
minimal government intervention and individual property rights, Britain's rail network 
nonetheless developed extensive coordinating mechanisms between companies, and 
ultimately under state administration, standardizing such issues as inter-line ticketing, 
time, and gauge.
The first rails to be built in Britain were all private, local concerns, with no 
pretensions towards national scope or interconnection. As the system developed, this led 
to a chaos of competing standards and incompatible networks. Transfers between lines 
were nigh impossible for passengers or goods, with no impetus for any line to cooperate 
with any other. In 1842, the Britain's rail companies came together to found the Railway 
Clearing House (RHC), initially to coordinate ticket sharing but the institution became a 
85. Schivelbusch, The railway journey, 30.
86. Ibid., 28.
71
forum for determining technical standards.87 The independent origin of the RHC 
evidences the importance of standardization in rail quite apart from any state influence; it 
is often the desire for enforceable standards that drives companies to seek state 
intervention. 
Rail also brought about the centralized administration of time.88 Prior to the 
development of rail in England, each city or town had its own time, tens of minutes or 
seconds apart. The difficulties first appeared even among the disparate private lines of 
early England: each station keeping its own time presented an nigh-insurmountable 
difficulty to schedulers, forcing each line to establish a standard time along its length. 
Eventually, as inter-line traffic grew, country-wide coordination became unavoidable and 
the RHC established a single standard time zone for the whole country, selecting the as-
yet quite esoteric Greenwich standard.89
Inter-line operability was hindered by more than incompatible timing, however; 
arbitrary yet profoundly consequential decisions regarding leeway (the necessary 
clearances beside and above the rail) and, more importantly, rail gauge cast up obstacles 
to moving engines and carriages between lines. Despite standards agreed on by the RHC, 
a “gauge war” broke out between supporters of “standard” gauge and broad gauge, used 
solely by the Great Western Railway. Supporters of each were unwilling to squander the 
fixed capital of already-built lines and attempted to expand their own network. Ultimately 
87. Philip Sidney Bagwell, The Railway Clearing House in the British Economy 1842-1922, (London: 
Allen & Unwin, 1968).
88. I mean this is in the practical sense, but Latour draws attention to how rail travel contributes to a 
perceptual disconnect between time and distance, time and experience, thereby causing time to appear 
as an abstraction. Bruno Latour, “Trains of Thought: Piaget, Formalism, and the Fifth Dimension,” 
COMMON KNOWLEDGE 6, no. 3 (1997): 170.
89. Schivelbusch, The railway journey, 43.
72
standard gauge won out; whatever benefits broad gauge offered could not compete with 
the benefits of interoperability.90
Schivelbusch emphasizes the degree to which this centralization and 
standardization of rail administration happened in England quite against the dominant 
ideology, which centered around competition and laissez-faire economic policies. 
Thus the railroad as a machine ensemble finally became institutionalized 
even in this respect, despite the resistance offered by the economic thought of the 
period, which was based on the principle of competition. The machine ensemble, 
consisting of wheel and rail, railroad and carriage, expanded into a unified railway 
system, which appeared as one great machine covering the land.91
England's example thus evidences one of the more liberal interpretations of rail 
development, yielding to centralization and standardization only with the utmost 
reluctance. In France and Belgium, for comparison, the state played a far more direct 
role.92 Regardless of the inclinations of those in charge, a multitude of factors inherent to 
the operation of rail yield advantages exponential in proportion to the level of investment, 
standardization, and administration. Whether present from the beginning in the form of 
direct state investment and control, or emerging from voluntary cooperation between rail 
companies, such standardization and centralization were logical complements of the 
development of rail. Rail was in many cases a key moment in the production of the 
modern state: it posed novel problems that demanded new forms of state intervention and 
new state structures.93 However arbitrary the standards embedded in rail were—gauge 
90. Vance, Capturing the Horizon, 218–220.
91. Schivelbusch, The railway journey, 29.
92. Vance, Capturing the Horizon, 231; 227.
93. Cristina Purcar, “Designing the Space of Transportation: Railway Planning Theory in Nineteenth and 
Early Twentieth Century Treatises,” Planning Perspectives 22, no. 3 (2007): 325–352, 
doi:10.1080/02665430701379142.
73
size, time zone, ticket format, language(s) of communication—once established, they 
exerted a tangible homogenizing force on the institutions they now partially constituted.
The idea of the railroad as “one great machine” deserves a moment's 
consideration. If the preposition is granted, then rail is most certainly the world's first 
country-, even continental-scale machine. While other (typically state) projects—for 
example the Great Wall or the Three Gorges Dam in China—mobilized the huge amounts 
of labor and capital towards the construction of a single structure, the railroad is unique in 
its geographical scope. Its function is dependent not only on its coherence, being 
regulated and maintained on a massive scale, but on its ubiquity, existing simultaneously 
at a local and regional scale. Vance notes that in the case of British rail, “it is striking how 
strong the drive for ubiquity became in the face of the original notions of regionalization 
or even specialization of lines.”94 Lines intended to improve connections between 
adjacent cities, or to connect provincial cities with the metropole, became the core of an 
unprecedented network interconnecting vast swathes of the country. Vance records that 
early rail builders, extrapolating from the technology's origins in coal-hauling, saw them 
primarily as commodity transport; they were caught off-guard by the demand for 
passenger services.95 Rail not only creates new centrally-administered institutions, but 
recreates the very environment in which people live.
This environment is not merely a new environment for its inhabitants to navigate, 
but a new type of environment: one where space functions in a novel manner, where the 
pace of circulation accelerates so radically that it constitutes a qualitatively new thing, 
necessitating a new set of skills of perception, understanding and navigation.
94. Vance, Capturing the Horizon, 220.
95. Ibid., 198; 234.
74
Two absolutely novel spaces emerged from the railroad: the carriage and the 
station. While each took pre-existing structures as their model, it took a long period of 
mutual adjustment and learning before they and their occupants settled into mutual 
accommodation. In addition to these absolutely novel spaces, the rail transformed the 
relation between existing places in profound and profoundly disorienting ways: speed and 
enclosure conspired to isolate the train passenger from the spaces she traveled through, 
presenting the space in-between origin and destination as a dreamscape, a panorama, or 
as nothing at all. All of these changes necessitated learning new ways of seeing, and new 
skills of navigation.
The physical design of the rail carriage began, as the name would suggest, with 
the stage coach that formed the backbone of the contemporary English overland 
transportation system. This design immediately encountered problems both technical and 
psychological. The small enclosure of the stage coach proved unsuitable when transposed 
to the rail: the inability to move or communicate easily between cars, access hygienic 
facilities, or see or control any aspect of the train's movement inspired in many 
passengers a feeling of helplessness and anxiety. A good part of the general anxiety and 
boredom of train travel can be traced to the forced passivity in the face of vast and 
terrible forces: despite the pleasant aspects of the journey, there remained the “close 
possibility of an accident, and the inability to exercise any influence on the running of the 
cars.”96
96. Schivelbusch, The railway journey, 130. See also Nicholas Daly, “Blood on the Tracks: Sensation 
Drama, the Railway, and the Dark Face of Modernity,” Victorian Studies 42, no. 1 (October 1, 1998): 
47–76, doi:10.2307/3829126.
75
The technical and psychological aspects here are not separate, but closely joined. 
The mode of sociality that had emerged in the environment of the stage couch, which 
consisted of intense sociability, vanished with the arrival of the train.97 Some combination 
of shorter travel times, greater variety of people, more rapid speed, or other, subtler 
factors rendered this sociability a thing of the past.  Instead, the necessity of sharing a 
compartment with a stranger inspired, among first- and second-class passengers, intense 
social discomfort to be escaped as quickly as possible, most often through reading. This 
highlights a key point: an essential element of these new landscapes—of any human 
landscape—was the people within it, requiring as much skill to navigate as a complicated 
station layout. The forced intimacy of the carriage provoked a crisis of how to be, in this 
new space: 
Before the development of buses, trains, and streetcars in the nineteenth 
century, people were quite unable to look at each other for minutes or hours at a 
time, or to be forced to do so, without talking to each other.98 
Schivelbusch documents repeated complaints over the boredom and awkward 
atmosphere of the carriage as well as the concurrent rise of train station book stalls. 
Reading quickly emerged as the foremost method of avoiding both social discomfort and 
boredom. The uncertain social landscape of the train can be seen most clearly in the 
hysteria generated by a pair of murders on board several years apart; these incidents 
sparked a hunt for a carriage design that allowed for movement without sacrificing 
privacy that only after several decades yielded the passage/compartment design used in 
Europe today. Watts suggests that the passenger's actions, while seemingly meaningless 
97. Itself decried as the ruination of the virtue of an earlier mode of transportation, and later accepted as 
normal and benign.
98. Georg Simmel, Soziologie (Duncker & Humblot, 1908); quoted in Schivelbusch, The railway journey, 
75.
76
and passive, nonetheless are an active “crafting” of the rail experience, paralleling my 
own focus on skill as a productive element of travel.99
If carriages posed a problem of being, stations posed a problem of navigation. 
Stations served as a point of transition between two very different spaces: the established 
space of city street traffic, and the industrialized space of the railroad, wherein rules of 
conduct were largely yet to be established. Like airports today, rail stations were spaces 
of transition between travel on different scales, catapulting people and things into global 
relationships.100 Schivelbusch shows that this transitional character was reflected in the 
architecture of the buildings themselves: “half-factory, half-palace;” one end consisted of 
train platforms covered with glass and steel, and the other end built of traditional stone or 
brick, facing the city101. One end was located in the city or town whose name the station 
bore; but the other existed as a standardized node in a network whose purpose was the 
homogenization of space. Upon arrival, one was greeted with a structure eerily like the 
one just departed; to travelers, stations became synecdoche for their respective places yet 
at the same time, were perforce far more self-similar than the places they represented.102 
The local character, thus, was symbolized foremost in the name written on the station 
wall: geography became abstract, homogenized, and legible only to the literate.  
There is a great deal of implicit knowledge in the navigation of the rail station: 
station layouts, the meaning of signage, cultural norms of flow and conduct are here 
another example of the homogenization of structure and skill railroads produced. In the 
99. Watts, “The Art and Craft of Train Travel.”
100. Mark Gottdiener, Life in the Air: Surviving the New Culture of Air Travel (Lanham, Md.: Rowman & 
Littlefield, 2001), 10–11.
101. Schivelbusch, The railway journey, 70.
102. Ibid., 40, 55.
77
initial period of rail expansion, the  transitional space of the station was tightly regulated 
due to a concern that passengers would be unable to safely or successfully navigate an 
industrial space: passengers had to gather in waiting rooms, and were only allowed to 
board immediately before departure. As trains became integrated into everyday life, 
competent railway navigation became part of the average, assumed social skill set and 
regulations eased—though these skills were unevenly distributed; the fumbling country 
bumpkin remained a staple trope of railway scenes. Rail stations were and have remained 
a space where a remarkable cross-section of society interacts.103 
Stations, with an emphasis on the plural, also became sites where rival lines 
competed and, in many cases, eventually merged. Each company entering a city had its 
own terminus, offering confusion and inconvenience to through traffic; Chicago in the 
1890s had seven separate stations.104 In some places these were replaced by union 
stations; in others, complex local networks managed the interchange of passengers and 
freight.
Rail thus produced a new set of spaces with their own dangers and challenges, 
requiring a new set of skills to successfully inhabit and navigate. These new skills were 
not universally endowed, however: the spaces of the rail were always shared between 
experienced travelers and inexperienced, foreign and/or rural first-timers, between 
cosmopolitan elites and the poor. Even as standardization produces a homogeneity of 
space, new heterogeneities of skill and affect are produced as well. 
103. Orvar Löfgren, “Motion and Emotion: Learning to Be a Railway Traveller,” Mobilities 3, no. 3 
(2008): 340, doi:10.1080/17450100802376696.
104. Meinig, The Shaping of America, 3: 247.
78
Discussions of the impact of technology on culture, particularly one with as strong  
standardizing and centralizing tendencies as rail, can often veer into or be mistaken for 
technological determinism: rail required such and so forth changes in culture, and 
humanity could not help but oblige. This is not my argument: rather, technological 
change creates conditions to which humans must adapt, but the manner of the adaptation 
is undetermined, and the change itself is the product of human action. Two examples 
highlight the contingency inherent to the process of techno-cultural change and the 
heterogeneity which inevitably results: the different modes of sociality that emerged 
among different classes in European rail cars, and the differences between the American 
rail culture—and its subtly different technological complement—and that which emerged 
in Europe.
The challenge that the rail car environment posed to travelers was discussed 
above, along with the strategies—reading, careful avoidance of social contact—with 
which riders responded. However, this was incomplete: these norms developed only 
among the first- and second-class passengers. In third-class, an entirely different mode of 
sociality, characterized by constant conversation and mutual entertainment, prevailed. 
“How often...I have..., while traveling alone or with people with whom it was impossible 
to start a conversation, envied the travelers of the third or fourth class, from whose 
heavily populated carriages merry conversation and laughter rang all the way into the 
boredom of my isolation cell.”105 Schivelbusch gives several possible reasons for this 
difference: the lack of reading habits, no prior expectations of travel, and the denser 
population of the lower class compartments. All of these seem reasonable enough—yet in 
105. Schivelbusch, The railway journey, 67.
79
a different section, it is the density of population that drives the development of social 
avoidance strategies such as reading. In any case, the reasons why the development of rail 
car sociality took such a different direction among the lower classes is less important than 
the bare fact that it did: whatever problems rail posed to previous forms of sociality were 
amenable to multiple solutions. Even as the infrastructural environment around society 
was homogenized, heterogenous responses emerged.
Across the sea, American carriage design took a very different path than that in 
Europe. Rather than multiple small compartments based on stage coaches, Schivelbusch 
argues American carriages were modeled after steamboat saloons and canal packet boat 
cabins, preferring open, shared spaces to confined, private ones. This was undergirded, 
literally, by a technical innovation called the bogie, which allowed the front and rear 
axles of the carriage to rotate independently without derailing. This in turn permitted the 
construction of longer compartments despite the curves of American rail lines. American 
carriages, even the luxurious Pullman cars, were therefore more like third-class European 
rail cars in spatial layout than first. A different technical precedent led to a different 
design, avoiding the crisis of sociality and hysteria that has such a profound impact on the 
European experience: no doubt the American rails had their own issues, but Schivelbusch 
does not say. 
Rail functions to create homogenous territories and standardizes behavior, 
machinery and culture over unprecedentedly broad spaces; yet even as it does so, it 
produces new forms of heterogeneity within and between. Acknowledging its centralizing 
tendencies does not mean ignoring its heterogenous impacts; indeed, each is necessary 
context for understanding the other.
80
Rail's impact was not limited to the new spaces it had produced. It also radically 
reworked the pre-existing geographies of travel and trade, producing something which 
appeared to contemporary commentators as “the annihilation of space and time.” While 
doubtlessly hyperbolic, this sentiment does demonstrate the shocking impact the railroad 
had on perceptions of space and time. The sheer speed of rail travel, as well as cutting 
transit time, also cut travelers off from their surroundings. Numerous writers complained 
that the scenery simply flew by too fast to be taken in: the same phenomenon that 
presented as a safety risk to the train's crew was here a less dire but nonetheless 
unsettling disconnection from the traveler's environment. Schivelbusch documents many 
bitter complaints of the difficulty early travelers had seeing out of the train: their habits of 
perception were simply unsuited to rail. One text explicitly recommends that passengers 
focus their eyes on more distant objects: that such advice needed giving reinforces the 
sheer novelty of rail's speed, and necessity of cultivating new visual skills.106 The fatigue 
engendered by the effort to track one's surroundings no doubt played, for some, a role in 
the rise of the habit of travel reading. Other eyes, perhaps younger, found the panorama 
of rail thrilling, presenting “in a few hours...all of France...a vast succession of charming 
tableaux.”107 The minute study of landscape allowed (and enforced) by stage coach or 
foot travel, which joined the traveler to the landscape, gave way to a constantly changing, 
necessarily impressionistic, evanescent panorama. The space between origin and 
destination did not vanish, but rail changed it from an unavoidable element of the travel 
experience to one that could be escaped, and often was: nineteenth-century Parisians 
heading to the Mediterranean coast paid “no attention to the invisible landscapes of the 
106. Ibid., 56.
107. Ibid., 61.
81
journey. To leave Paris and to get to where the sky is clear, that is their desire.”108 What 
came in between was irrelevant; the journey took place not in the landscape but on the 
train. These factors combine to shrink not just the time of the journey, but also the 
perceptual distance, the imaginary distance.
In the eyes of contemporary observers, rail if not annihilated then at least 
condensed geography, as if 
...the whole population of the country [would]...at once advance en masse, 
and place their chairs nearer to the fireside of their metropolis by two thirds....As 
distances were thus annihilated, the surface of our country would, as it were, 
shrivel in size until it became not much bigger than one immense city.109
Yet even within the confines of this metaphor geography reasserts itself in the face 
of annihilation: the rail's influence manifests as pulling the hinterlands in towards the 
nucleus of the metropole; geography condenses unevenly.110 In reality the process was far 
less even: rail shrank distance only where lines ran, incorporating some hinterlands 
earlier and more thoroughly than others. Transport between hinterlands was more 
difficult than between hinterland and metropole: “So secondary were [services between 
cities other than London], and so difficult access to them, that a whole literature of the 
horrors incident to boarding them...became a special genre for nineteenth-century British 
writers.”111 The rail produced a new hierarchy of distance, more chronologically compact 
but no less differentiated. 
108. Ibid., 38.
109. Quoted in ibid., 34.
110. It is worth noting that this condensation did not, ultimately, empower the metropole: the Age of Rail 
was dominated by nation-states, not the great cities that defined the political geography of earlier 
centuries. Tilly, Coercion, Capital, and European States, AD 990-1990.
111. Vance, Capturing the Horizon, 222.
82
The hinterland was not only brought into closer contact with the metropole in a 
perceptual sense, but in a material sense as well. Rail carried individuals and 
commodities more quickly and cheaply over a wider area than ever before: while this 
change began as qualitative, it became a quantitative one when a commodity lost “the 
context of the original locality of its place of production” and instead was “seen as 
products of the market, just as Normandy seemed to be a product of the rail road that 
takes you there.”112 Schivelbusch follows Marx in emphasizing the role of transportation 
in the production of commodities. “This locational movement—the bringing of product to 
market...could more precisely be regarded as the transformation of the product into a 
commodity.”113 It is intriguing that Marx isolates this moment as the point at which 
commodities are produced as such: marketplaces play a necessary role in the creation of 
commodities. Yet it is not entirely clear whether the market Marx refers to is a concrete or 
an abstract marketplace: it is precisely this ambiguity which it is the market's function to 
create. Commodities enter the market not by appearing at a particular location, but by 
entering the sphere of circulation: a place characterized by its placelessness. Tremendous 
work goes into producing this sphere: the construction of roads, rails, shipping lines, 
warehouses, distribution centers: the marketplace which product must enter in order to 
become commodities thus only functions as an abstract, placeless place within the 
context of the transportation network within which it is bound and upon which it is 
112. Schivelbusch, The railway journey, 41.
113. Karl Marx, Grundrisse: Foundations of the Critique of Political Economy (rough Draft) 
(Harmondsworth London: Penguin “New Left Review,” 1973), 534; quoted in Schivelbusch, The 
railway journey, 40.
83
dependent. The scale, speed, and ubiquity of rail render it indispensable in the creation 
and maintenance of a homogenous, country-wide market.114
Yet it is not only goods that circulate within the rail network, but people as well, 
and Schivelbusch argues that on the train that individuals—particularly bourgeois 
individuals—had their closest contact with the industrial production process. As an 
industrial process, Schivelbusch and Marx argue, transportation is unique in that the act 
of production and the act of consumption are simultaneous, thus forcing the productive 
and consumptive uses into unusually close proximity. The consumer, accustomed to 
enjoying the products of industry at a remove from their manufacture is in traveling via 
rail forced to participate directly in the industrial process. This reality is reflected in the 
perceptions of the rail traveler: “It transmutes a man from a traveller into a living parcel,” 
complained one critic, echoing a common refrain.115 In contrast to the individualistic 
“artisan” travel on foot, horse-back, or carriage, rail brought industrial scale to travel. 
People, not just goods, circulate through this network; just as goods appear to be a 
product of the market in which they are sold, and locations a product of the transport on 
which it is reached, so too do populations appear as an aspect of the industrial 
transportation network they circulate within.
Yet, markets are themselves still bound by the limits of the transportation 
infrastructure which produces them, and reproduce both its external boundaries and its 
internal differences. When rail is a state project, which to some extent it always is, the 
boundaries and differences being reproduced take on a state character—or perhaps it 
would be equally accurate to say that the state takes on those boundaries. State space is 
114. Meinig, The Shaping of America, 3: 265.
115. Schivelbusch, The railway journey, 54.
84
thus produced anew by rail as the substance through which social life is navigated. This is 
not entirely novel: roads and waterways also shaped territory and its navigation. Yet rail, 
because of the precise joining of engine and track and the necessity of centralized 
coordination, brings centralized administration into the core of the structure; because of 
its ease, cheapness, and reliability, it expands and quickens circulation to an 
unprecedented degree. Paradoxically, the extent to which the territory within the rail 
network is rendered homogenous and standardized does not appear within the network: 
there, homogeneity and standardization appear normal. Rather it appears at the 
boundaries of that network, as a stark contrast between within and without. As station and 
carriage are standardized, and the behavioral norms of rail travel are established, one 
journey within the network becomes increasingly like another: without traveling to any 
given location, it becomes possible to imagine the journey there, or the journey to 
anywhere the rail connects, producing an imaginary territory—imaginary, but based on a 
very material reality. The gaps in that reality, where circulation is not as easy as 
imagined, where political, cultural, or infrastructural barriers hinder free circulation of 
goods and populations appear as problems to be solved, creating a political impetus for 
the perfection of the national territorial network. The production of a national territory 
and the production of a national consciousness thus constitute mutually-reinforcing 
processes.
State impact at a slant
By reworking the relation between state, population, and territory through the lens 
of infrastructure, and rail in particular, a theoretical framework for analyzing the 
85
emergence and constitution of Chinese nationalism that is not confined to the rhetorical 
realm has been assembled. This theory enables a way of talking about nationalism that 
encompasses not just what people say but how they live. Critically, it provides an entry 
into the relation between state and nationalism that doesn't reduce either to a direct 
function of the other: they can interact in complex and non-determinative ways. State 
action can now be understood as producing nationalism even when producing 
nationalism is not the intent; nationalism can be seen as making demands on the state 
which are not in harmony with the state's interests. The impact of rail on the development 
of nationalism is therefore undeniably state influenced, but that does not mean that the 
nationalism thereby instilled was either the goal of the state nor made nationalists the 
state's creatures. The state's impact, though considerable, is at a slant.
Rail encompasses aspects of language, material culture, social interaction, and 
perception; thus it can mediate between material and discursive definitions of national 
consciousness. What remains, now, is to take this theoretical apparatus and turn it on the 
actual history of Chinese nationalism and see what insights it yields. In examining the 
early history of Chinese rail, however, what is seen is not primarily a unitary Chinese 
state creating the conditions for the development of a singular national identity, but the 
failure of such a centralized project. Any theory attempting to explain why a phenomenon 
arose should be just as capable of explaining why, in another context, it did not. In the 
case of early Chinese rail, the parallel development of identity, state institutions and rail 
took place in a manner that failed to create a singular sphere of circulation of people and 
goods—but nonetheless illustrates their interdependence.
86
CHAPTER IV
THE FRACTURED GEOGRAPHIES AND IDENTITIES OF EARLY TWENTIETH 
CENTURY CHINA
Taking rail seriously as a meaningful site for the creation of state effects and 
territorial identities entails understanding not just when it leads towards the creation of a 
centralized nation-state, but also addressing the conditions where the entanglement of 
these processes leads to an altogether different outcome. Such is the situation with 
Chinese rail development in the first half of the twentieth century. Successive waves of 
devolution, first as a deliberate policy of the Qing government, then as a consequence of 
revolution and the emergence of warlordism, led to a distinctive history of political and 
economic development where even as the notion of a singular Chinese nation emerged, 
the central state's ability and responsibility to manage the country's infrastructure shrank 
drastically.
The theoretical structure assembled in the previous chapters argues that 
governmental power, territorial identity, and capital-intensive, state-affiliated projects 
have strong, mutually constitutive effects, and predicts that they will therefore tend to 
manifest at the same scale. This pattern is indeed evident—yet counter-intuitively, the 
scale for these phenomenon was not China as a whole. Rather from the late Qing until the 
Japanese invasion provincial or regional movements, companies, and governments 
increasingly dominated political and economic life. In China prior to the Japanese 
invasion, the legitimacy of provincial identities and governments versus their country-
wide equivalents was hotly debated: the very existence of such debate demonstrates the 
diversity and complexity of identity and governance during the period. It is the 
87
prevalence of regional, largely provincial identities, even in the face of a powerful 
nationalist movement that demands an explanation, and that an examination of rail can 
help illuminate.
Early development of rail in China was scattered and fragmentary: foreign 
interference, war, and technical and social challenges to rail construction prevented the 
development of a coherent country-wide rail network. Rail's impact on China during the 
pre-war period was nonetheless dramatic, causing new cities to rise, old cities to decline 
or be transformed, creating new centers and new hinterlands, spurring the 
commercialization of agricultural production and the rise of new industries. Rail 
produced fine-scale heterogeneity between rail-adjacent and suddenly isolated and 
peripheral areas: rail brought changes in clothing and hairstyles, senses of time, and 
social mores. The experience of journeying by rail caused re-evaluation of certain social 
norms: carriage design led to the increased interaction between gender and social classes. 
Rail superseded previous transportation networks in speed, price and reliability, 
which led to a radical reorganization of China's economic geography. Harbin did not exist 
prior to the arrival of the Eastern China Railway by Russia; Shijiazhuang also arose as a 
metropolis only after it became the junction of two major lines. As part of the same 
movement, traditional economic centers were suddenly pushed to the periphery; 
prominent regions along the Grand Canal found themselves transformed into hinterlands 
in a matter of decades. While a common sphere of circulation gradually accreted during 
the pre-war period, it remained partial, with sharp boundaries between localities and vast 
difference between regions.
88
These large but diffuse social shifts were not the only impact rail had in the 
development of national consciousness. Rail, and the control of rail, became central 
issues in two of the largest protest movements in the late Qing, the Rights Recovery 
Movement and the Railway Protection Movement.116 Rail became a proxy for intense 
conflicts over the proper scale for territorial identity and state legitimacy: some argued 
that the province was only a momentary proxy for a weak and corrupt Qing China, while 
others argued for a genuine province-first mentality. 117
Due to the devolution of state power over the course of the early nineteen 
hundreds, centralized administration and funding of infrastructure projects fell into the 
hands of the provincial governments; accordingly different provinces exhibited very 
different trajectories of rail development. Generally, the more commercially developed 
provinces, such as Guangzhou or Zhejiang, had merchant-led rail lines, and had far more 
success in actually building rail and turning a profit than those in predominantly 
agricultural provinces. This exacerbated existing differences between provinces; 
agricultural provinces fell further behind the economic powerhouses, rendering China a 
more heterogeneous, not a more cohesive, space.
In Shanxi, where rail wasn't built until the warlord era, both the physical structure 
of the railway and its administration evidence an autarkic regional identity. Sichuan's 
history of rail development is perhaps the strangest: despite being the center of the 
railway protests that led to the downfall of the Qing, no rail was built until after World 
War II. In Manchuria, Russian and Japanese rail lines tied the region into spheres of 
116. 收回利权运动 ( shōuhuí lìquán yùndòng) and 保路运动 (bǎo lù yùndòng), respectively.
117. Duara, Rescuing History from the Nation; Liu, “Wan Qing ‘sheng’ yishi de bianhua yu shehui 
bianqian.”
89
circulation outside China, developing a regional economy and identity distinct from the 
rest of China and hindering economic integration. 
The course of China's rail development exhibited a strong regional or provincial 
impact in the early nineteen hundreds, influenced by and influencing political 
consciousness and economic change. During this period, regional identities challenged 
the legitimacy of Chinese nationalism as the primary political identity, often on the basis 
of the (mis)management of large-scale infrastructure such as railroads. This historical 
period provides an interesting case for studying the role played by infrastructure in 
mediating the relationship between identity, territory, and governance.
Social histories of rail
Chinese scholarship on the social history of modern Chinese rail has been limited. 
Practical limitations, such as the destruction of materials during the war and lack of 
applicable expertise, has hindered research, and a “revolutionary view of history”118 
confined discussions of rail's impact to subjects such as foreign loans, international 
relations, and politics. However, in the last thirty years, along with the broader rise of 
interest in social history, studies of rail's social impacts have drawn more attention from 
the Chinese scholarly community.119 In this chapter I draw on both Western scholarship 
and the small but growing group of Chinese scholars researching the economic and social 
changes engendered by rail. 
118. 革命史观，or “gémìng shǐ guān”
119. Pei Jiang, “Zhongguo jindai tielushi yanjiu zongshu ji zhanwang: 1979 ~ 2009 [中国近代铁路史研究
综述及展望:1979～2009],” 过去的经验与未来的可能走向——中国近代史研究三十年（1979-2009) 
(2009): 22; Quanyou Su, “Jin shinian lai woguo jindai tielu shi yanjiu zongshu [近十年来我国近代铁路史研究综述],” Journal of University of Science and Technology of Suzhou 苏州科技学院学报(社会科
学版) no. 02 (2005): 110–114+127.
90
Prior to the appearance of railroads, Chinese inland transportation consisted of a 
wide-spread but minimally maintained road network designed for use with the distinctive 
Chinese wheelbarrows,120 and a complex water transport system utilizing both natural 
rivers and human-built canals. While these transport systems did not integrate space to 
the extent that rail later did, they nonetheless supported a certain degree of economic 
integration at a macro scale.121 
Rail was initially championed in China by foreign powers interested in 
penetrating the Chinese economy, and as such drew ire both from court elites and 
populist rebels. The first major rail projects in the Northeast were spearheaded and 
controlled by foreign governments, and were oriented towards incorporating those areas 
into Russian or Japanese economic and political spheres. In the Yangtze river basin and in 
Guangdong, early railroads were primarily built by Western commercial interests to 
facilitate import and export, and thus were designed to integrate the hinterlands into the 
trade concessions of Shanghai and Hong Kong.122 
Even domestic efforts to build rail were focused on building local connections 
rather than forming a single national network. Chinese domestic railroads encountered 
frequent difficulties raising capital, avoiding corruption and meeting the technical 
requirements of rail construction: the success of engineer Zhan Tianyou stands in stark 
contrast to the general state of affairs. During the period of warlordism following the 
establishment of the Republic of China, some provincial warlords pursued rail projects 
120. Joseph Needham et al., Science and Civilisation in China (Taipei, Taiwan: Caves Books, 1986).
121. G. William Skinner, The City in Late Imperial China (Stanford, Calif.: Stanford University Press, 
1977), 275–288.
122. James Zheng Gao, Meeting Technology’s Advance: Social Change in China and Zimbabwe in the 
Railway Age (Westport, CT: Greenwood Publishing Group, 1997).
91
limited to their own sphere of influence, in some cases deliberately building rail of a 
gauge incompatible with the rest of China.123
In the period leading up to 1911, less than ten thousand kilometers of track had 
been built, primarily in the Lower Yangtze and around Beijing. In the next decades, the 
transport in these areas was quickly transformed, though construction proceeded 
unevenly throughout the pre-war period and came to a standstill during the war. Wang 
Jiao'e et al. divide the development of Chinese rail into four periods: preliminary 
construction, before 1911; network skeleton, from 1911 to 1949; corridor building, from 
1950 to early 1990s; and deep intensification, from mid-1990s to present (Figure 1). At 
the end of the “network skeleton” period, there were no lines into Sichuan or Fujian, and 
Yunnan was connected to French Indochina but not to the Chinese network.124 Yunnan, 
Tibet, and Xinjiang were more accessible from outside of China than from within.125 
As a consequence of these myriad competing agendas, prior to 1949 rail 
infrastructure had only begun the process of forming a cohesive Chinese sphere of 
circulation. The peripheries were poorly integrated into the rail network, and in some 
areas were more tightly integrated into foreign spheres than Chinese ones. The 
differentials in access and incorporation created by these disconnected railroads 
exacerbated existing social disparities, widening the gap between integrated and isolated 
areas. While the social and economic consequences of rail development over the first half 
123. Zhankui Jing, “Yan Xishan yu Tongpu tielu [阎锡山与同蒲铁路],” Vicissitudes 沧桑 no. 03 (1993): 
26–29.
124. Jiaoe Wang et al., “Spatiotemporal Evolution of China’s Railway Network in the 20th Century: An 
Accessibility Approach,” Transportation Research Part A: Policy and Practice 43, no. 8 (October 
2009): 765–778.
125. Prasenjit Duara, Sovereignty and Authenticity: Manchukuo and the East Asian Modern (Lanham: 
Rowman & Littlefield Publishers, 2003), 188.
92
of the twentieth century were considerable, but profoundly different in character and 
degree in different localities, cities, and regions.
Rail and the transformation of Chinese society
Rail enabled the transmission of goods and populations on an unprecedented 
scale, transporting novel commodities and ideas deep into the countryside. This spread 
also had a specific geography, producing a fine-scale heterogeneity between rail-adjacent 
and more isolated areas. Counties adjacent to rail lines went through changes in various 
aspects of everyday life and social customs earlier than neighboring counties not along 
the track. Rail served as the mechanism for the dispersal of western-style clothing and 
hairstyles: for example, the county gazette of Xushui County, which was adjacent to the 
Ping-Han line, records that by the nineteen-thirties machine-woven cloth had replaced 
hand-made; in neighboring but not railway-adjacent Mancheng County, machine-woven 
93
Figure 1: Schematic of Chinese rail, 1911 & 1949. Note the 
northern bias and disconnectedness. Source; Wang et al. 2009.
cloth was too expensive for common use.126 Hairstyles and housing followed similar 
patterns: men stopped wearing queues and women cut their hair short sooner where the 
rail ran; earthen houses endured longer the further they stood from the rails. This was also 
the moment the bicycle became a common mode of transport.
Along the rail lines, social shifts occurred sooner and penetrated more deeply. 
This was no simple process of Westernization or modernization: the circulation of 
commercial commodities adapted to local markets even when the commodities in 
question had foreign origins. Wearing Western clothes, riding bicycles, living in Western-
style houses all entailed learning new skills related to but distinct from their European 
analogues.
The changes that accompanied rail were not limited to the superficial or the 
material: profound social shifts are also incurred when changing material culture. The 
train, as in Europe, also altered the perceptions of time; while watches had been known in 
China long before as a rich dandy's plaything, with the train's increasing importance and 
its inflexible, exacting schedule, understanding and using clock time became an everyday 
skill for those Chinese who interacted with the railways.
The number of people who had such contact increased steadily. In 1931, 
government-owned rail lines carried 180,000 passenger trips; by 1934 that number had 
increased to over two million.127 Much like in Europe, the experience of riding the train 
posed certain problems for existing social mores. One such more was the principle of 
126. Zhancai Li, “Tielu yu Zhongguo jindai de minsu shanbian [铁路与中国近代的民俗嬗变],” Shixue 
Yuekan 史学月刊 no. 01 (1996): 57.
127. Li, “Tielu yu Zhongguo jindai de minsu shanbian.”
94
men and women avoiding contact.128 While it would not have been unimaginable to 
designate men-only and women-only cars, Chinese passenger rail by and large followed 
the spatial layout of European carriages and offered only mixed-gender seating.129 Those 
who rode the rails perforce learned a new norm of social interaction, and the importance 
of gender segregation declined.
With easy, frequent travel the cultural bias against traveling away from one's 
parents—“while your parents are alive, don't go far”130—also weakened. With people 
traveling for business, to seek work, to go to school, to visit relatives, or simply for the 
sake of traveling, the traditional adage lost its force. When a journey from Nanjing to 
Shanghai that via traditional water transport took thirteen or fourteen hours could be 
completed in two, the costs and risks of travel greatly diminished. It was possible even to 
leave in the morning for Shanghai and return the same night: long-distance commuting, 
with its associated habits and skills, became possible.131 
Rail not only affected the lives of rail travelers and inhabitants of the transformed 
landscape, but produced a new workplace as well. By 1937, the railways collectively 
employed over three hundred thousand workers.132 This necessitated the development of 
entirely new methods of workforce management, employee training and worker retention. 
128. “男女授受不亲”，or “nánnǚ shòushòu bù qīn”
129. Jie Zhang, “Huocheshang de shehui dengji [火车上的社会等级],” Art Museum 美术馆 no. 01 (2010): 
166–177.
130. ”父母在，不远游“，or “fùmǔ zài, bù yuǎn yóu”
131. Zhancai Li, “Tielu yu jindai Zhongguo chengzhen bianqian [铁路与近代中国城镇变迁],” Journal of 
Suzhou Railway Teachers College 铁道师院学报 no. 05 (1996): 34–38. 
132. Stephen L. Morgan, “Personnel Discipline and Industrial Relations on the Railways of Republican 
China,” Australian Journal of Politics & History 47, no. 1 (March 2001): 24. Morgan does not 
distinguish clearly between the rails administered by the Republican government, imperial powers, 
and regional warlords, blurring the complexity of this period. 
95
While labor was plentiful, managers were challenged to find workers with the skills and 
attitudes necessary to function within a complex technical bureaucracy, and the early 
railroad departments were an important site for the pioneering of modern management 
techniques for the modern Chinese state. In order to produce the sort of skilled, complaint 
workers necessary to run a railroad, the ministries developed a comprehensive system of 
Foucauldian “micropenalties” to regulate the punishment and reward of workers, while a 
complex system for awarding raises and pensions provided incentive for workers to stay 
with the railroad long-term.133 The regulatory apparatus and the work force emerged in 
tandem.
Within the rail network, heterogeneities were produced as well. As mentioned 
earlier, the Chinese rail carriages were modeled after European carriages: including their 
system of three separate classes. Initial ridership of trains skewed heavily towards the 
upper classes: a first-class roundtrip ticket on the Wusong Line cost the equivalent of 100 
liters of rice; a third class ticket cost one sixth that.134 While ticket prices declined as rail 
became more common, the internal distinctions inherent to the three tier system 
remained: a new set of social distinctions for Chinese people to learn to navigate.
The social shifts that accompanied the spread of rail were multifarious and 
complex, and produced new forms of heterogeneity even as they produced homogeneity 
in others. In addition to changes on an individual level, the coming of the railroads also 
altered China's economic geography at the level of city and region.
133. Ibid., 31.
134. Zhang, “Huocheshang de shehui dengji.”
96
New cities, new hinterlands
The new transportation economy engendered by rail had tremendous impact on 
China's urban geography. Some of the China's largest metropolises trace their genesis 
back to this period. Early railways were torn between the sometimes competing and 
sometimes complementary imperatives of foreign powers and the Qing court (later, the 
warlords and the Republican government.) None were interested in preserving the 
traditional hierarchy of Chinese cities. 
The greatest transfers of population took place where the railways were most 
developed: in Manchuria and in Jiangbei. The city of Harbin didn't exist prior to the 
arrival of the Russian-owned China Eastern Railway in 1897; after Harbin's future 
location was designated as the railway's administrative headquarters and the juncture of 
the Dalian-bound southern spur, the area rapidly urbanized, developing a substantial 
industrial sector and reaching a population of one hundred thousand by 1905.135 
In the Jiangbei region, the rail-driven urbanization was quite significant: in much 
the same way that the juncture of the Eastern Railway and its spur line gave rise to 
Harbin, the junction of the Beijing-Hankou line and the Taiyuan-Shijiazhuang line136 
spurred massive industrialization and urbanization in Shijiazhuang.137 Railway towns also 
became the site of new manufacturing hubs. Zhengzhou, at the intersection of the Ping-
Han and Longhai rail lines,138 became the linchpin of a regional cotton industry, 
135. Li, “Tielu yu jindai Zhongguo chengzhen bianqian.”
136. 京汉铁路 (Jīng hàn tiělù) and 正太线 (zhèngtài xiàn), respectively.
137. Pei Jiang and Yaping Xiong, “Tielu yu Shijiazhuang chengshi de jueqi [铁路与石家庄城市的崛起:1905
—1937年],” Modern Chinese History Studies 近代史研究 no. 03 (2005): 170–197.
138. 平汉铁路 (Píng hàn tiělù) and 陇海铁路 (Lǒng hǎi tiělù), respectively. 
97
accelerating the integration of the surrounding agricultural area into commercial 
production.139
On other cities, however, rail's impact was detrimental or simply negligible. 
Zhenjiang, once a major hub along the Yangtze river trade, was gradually eclipsed by 
Nanjing after the Shanghai-Nanjing Railway's construction at the end of the 19 th century, 
particularly after the completion of the Tianjin-Pukou Railway in 1912.140 Pomeranz's 
The Making of a Hinterland also demonstrates the deeply uneven impacts of rail: the 
region of his study, centered on the once-vital lifeline of the Grand Canal, was bypassed 
by rail and subsequently became a backwater.141 Pomeranz uses this example to illustrate 
the contingency of political economic geography: it was the state's decision to relinquish 
canal maintenance to local authorities unequal to the task that pushed the region into 
economic decline, and varying social structures in different counties produced different 
outcomes. One city, Jining, was able to lobby successfully for a spur line: it stands out, in 
the region, for its relatively security and prosperity.142 It is a lonely exception in a region 
delineated by its lack of railroads. 
Further from the coast, rails were few and far between. Despite the existence of a 
Sichuan Railway Company beginning in 1905, there was no line connecting Sichuan to 
the rest of China until after World War Two. French-owned lines connected Kunming in 
Yunnan province to Hanoi in 1910, but the province did not connect to the Chinese 
139. Hui Liu, “Tielu yu jindai Zhengzhou mianye de fazhan [铁路与近代郑州棉业的发展],” Shixue Yuekan 史学月刊 no. 07 (2008): 102–109.
140. Li, “Tielu yu jindai Zhongguo chengzhen bianqian.”
141. Kenneth Pomeranz, The Making of a Hinterland: State, Society, and Economy in Inland North China, 
1853-1937 (Berkeley: University of California Press, 1993).
142. Ibid., 147–149.
98
network until after the war. The French line was also built on a narrow-gauge track—
different national standards and nomenclatures, brought by the various imperial powers, 
hindered the consolidation of the rail network even after nationalization.143
Thus, the circulation of goods and populations through developing rail network in 
China was highly uneven: mutually constitutive of increased economic development and 
social mobility in the northeast, the North China Plain, the Yangtze river basin, and the 
southern coast, but leaving the geography of the interior relatively unchanged. This 
uneven assimilation of provinces and regions into the emerging rail network meant that 
the sharp transitions between interior homogeneity and external heterogeneity didn't fall 
on the country's borders but instead divided China internally. This had important 
ramifications for the development of Chinese national identity: residents of the 
modernized and westernized coastal cities whose lives were most conducive to 
nationalism were at the same time seen as the least authentically Chinese.144
Altogether, rail produced a radically different economic and social geography in 
China at the regional, county, and individual level. The preconditions for expanded 
circulation of populations and commodities were created, though the sharp boundaries of 
that circulation are as important to remember as the relatively homogenous territory 
within. The circulation is deeply entwined with the new forms of regional consciousness 
that emerged during the same period: as larger portions of the population began to travel 
more frequently over a wider area, and interacting commercially and socially with a 
wider sphere, traditional local identities no longer fit people's everyday experience. It is 
143. Bruce A Elleman, Manchurian Railways and the Opening of China: An International History 
(Armonk, N.Y.: M.E. Sharpe, 2010), 107.
144. Pye, “How China’s Nationalism Was Shanghaied.”
99
no surprise that gaining control over the rails that mediated that circulation became a 
focus of popular consciousness. The weakness of the central government and the 
consequent fractured political landscape made a conception of China as a whole difficult 
to imagine. Instead, it was at the provincial level, where governmental authority, railroad 
development, and political identity were strongest, that the political ramifications of rail 
are most apparent. Beginning in the late Qing, a historical shift in the balance of power 
between the central and provincial governments gave provincial governments an 
unprecedented amount of autonomy.
The provincial state and provincial identity
Prior to the devolution of the late Qing, reaching its apex during the warlord 
period, China typically had a highly centralized government with minimal autonomy 
given to the provinces. This high degree of centralization can be seen in the management 
of major infrastructure, which was traditionally funded directly by the imperial court. 
The premier example of the state's role in infrastructure construction and 
maintenance is undoubtedly the Grand Canal: began in earnest in the Sui dynasty, was 
gradually expanded, until in the Yuan dynasty it reached from Hangzhou to Beijing. The 
maintenance requirements for this structure—constant dredging, the rebuilding of levies, 
the construction of dams and locks—was enormous, and its decline and rejuvenation 
tracked the fall and rise of dynasties. The Grand Canal was certainly the greatest piece of 
hydraulic infrastructure of its age, but it was first among many: imperial projects also 
maintained coastline and levies along the Yellow River. Government involvement in 
these projects varied, occasionally playing little more than a coordinating role; when it 
100
became more involved, it was often as a result of petitioning by local communities unable 
to deal with increasing maintenance costs.145 In the late Qing, Kenneth Pomeranz argues, 
the state began to refocus its priorities away from maintaining the hydrological 
infrastructure, long considered a key state obligation, leaving provincial or local 
government in charge.146 This was part of a larger devolution of state prerogative and 
obligation towards local levels, understood by some as a weakening of the imperial 
government, and by others as a triage of less vital projects in favor of those crucial to 
resisting imperialism.147
In the wake of the first Sino-Japanese war, the Qing government began devolving 
unprecedented powers to the provincial level, including the power to levy taxes and raise 
armies.148 Previously the balance of power between the central government and local 
government had been tilted firmly and resolutely in favor of the central government. 
After their decisive defeat by the Japanese, the court became aware of the urgent need to 
spur the development of China's domestic economy stave off imperialist invasion, and 
empowering the provinces to pursue such development soon became central to that goal.
Accordingly, along with increased power came increased responsibilities: the 
provincial governments were required to develop industry, open schools, print money, 
and open mines.149 As many of these projects were quite profitable, the provincial 
145. Mark Elvin, The Retreat of the Elephants: An Environmental History of China (New Haven: Yale 
University Press, 2004).
146. Pomeranz, The Making of a Hinterland.
147. Ibid., 21–22.
148. Liu, “Wan Qing ‘sheng’ yishi de bianhua yu shehui bianqian.”
149. While I do not touch on the role of educational institutions in the formation of regional consciousness 
it is clearly an important site as well. See Zheng Wang, Never Forget National Humiliation: Historical 
Memory in Chinese Politics and Foreign Relations (New York: Columbia University Press, 2012).
101
officials were happy to oblige, though non-compliance with imperial orders became more 
and more of a problem.150 Provincial officials were often personally involved with the 
establishment of manufacturing and other commercial concerns and used their power to 
forward their interests; the expansion of commercial and industrial circulation happened 
largely under the aegis of the province. Thus, a great deal of the infrastructure mediating 
the relation between state, territory and people was being constructed and maintained at 
the provincial level—not at the imperial level. 
Railroads were among the most prominent of these provincial economic 
development projects. Along with mines, they became an important symbol of provincial 
identity as much as a factor in their economic development. The Rights Recovery 
Movement refers to a series of protest movements across China beginning in the first 
decade of the twentieth century aimed at recovering mining and rail-building rights from 
the foreign companies that had purchased them from the Qing government. The history of 
the Rights Recovery Movement demonstrates how rail reflects and embodies the 
interconnections between territory, identity and governance: as provincial governments 
gained power relative to the central government, not only did provincial identities 
become more prominent (often in direct opposition to country-wide identities), provinces 
also become the primary scale for state projects such as rail lines. This relationship was 
both material and symbolic: capital was raised and rail companies were organized at the 
provincial scale, and the construction of rail lines was seen as a matter of provincial pride 
and inter-provincial rivalry.151
150. Liu, “Wan Qing ‘sheng’ yishi de bianhua yu shehui bianqian.”
151. Rankin, “Nationalistic Contestation and Mobilization Politics.”
102
They were most prominent in the lower Yangtze basin and Guangdong, though 
mining rights recovery movements were significant in Shanxi and Shandong.152 These 
movements were driven at the provincial scale, even when the rights being recovered—to 
build a Guangdong-Hankou rail line, for example—were not so neatly delineated. They 
also ran directly counter to the central government: they sought to recover rights the 
central government had (in their view) cavalierly sold off to foreign concerns.
As part of their bid to recover railway rights, the provinces set up rail companies 
to raise capital in order to buy back rights and to construct rails. While the ultimate fate 
of all these rail companies was the same—nationalization—their trajectories beforehand 
were quite diverse: some raised sufficient capital to build and operate lines that seemed 
on their way to profitability and self-sustainability; others foundered under the weight of 
corruption and incompetence. Two interlinked factors seemed to play the strongest role in 
determining their success or failure: the companies where merchants provided most of the 
capital and exerted the most influence on the running of the company did well, while 
those that relied on capital collected via taxes and other coercive means and where 
officials controlled the management tended to fare poorly. The relative balance of power 
between officials and merchants in turn depended on the economic strength of the 
province in question. Provinces like Zhejiang or Jiangsu with strong commercial and 
industrial sectors invested a great deal of merchant capital in their respective companies, 
which then fared relatively well; provinces like Sichuan with largely agricultural 
152. Roger R. Thompson, “‘If Shanxi’s Coal Is Lost, Then Shanxi Is Lost!’: Shanxi’s Coal and an 
Emerging National Movement in Provincial China, 1898–1908,” Modern Asian Studies 45, no. 05 
(2011): 1261–1288, doi:10.1017/S0026749X10000119.
103
economies depended on official taxes to fund its company and tended to fare poorly.153 In 
the end, the impact of the provincial rail companies was to cause the economic conditions 
of their respective provinces to diverge more widely over time in a self-reinforcing cycle.
The discourse of provincial consciousness rang true to many ears in the early 
twentieth century in China. Its relationship with nationalist consciousness was complex: 
some held that it was only nationalism on a temporarily smaller scale, necessary when the 
national state was too weak or corrupt to stand up for its people. “To preserve Yunnan is 
to also preserve China. If Yunnan is lost then a number of Southeastern provinces will be 
lost; if a number of Southeastern provinces are lost, then China is lost.”154 Here the 
provincial level is a necessary element in a larger struggle. Yet others believed that the 
province truly was a appropriate scale for political identity and autonomous rule. 
“Guangdong is the Guangdong of Guangdong people!” went one rallying cry, 
demonstrating with remarkable brevity the triple coincidence of territory, people, and 
governance characteristic of the nationalist imperative.155 In the late Qing and early 
Republican period, implicit contradictions between these positions could remain dormant: 
in the face of foreign aggression and central government impotence, nationalists and 
provincial loyalists shared enough goals to make cooperation possible.
It is undoubtedly true that provincial consciousness developed to a large extent 
because of the lack of leadership from the central government. It would be a mistake to 
paint this as a purely rhetorical lack, however: the lack of leadership also had tangible, 
153. Shixuan Jin, “Ershi shijiechu gesheng shangban tielu jiqi jieju [二十世纪初各省商办铁路及其结局],” 
Journal of Beijing Jiaotong University 北方交通大学学报 no. 02 (1977): 80–99.
154. Liu, “Wan Qing ‘sheng’ yishi de bianhua yu shehui bianqian.” Translation mine.
155. Ibid. Translation mine.
104
physical impacts on the development of China's internal coherence. While national 
identity is often analyzed purely as a discourse, it is also intimately involved in the 
production of a set of structures, structures that regulate the circulation of populations, 
commodities and ideas within. Prasenjit Duara's Rescuing History from the Nation lays 
out the diversity of opinions within the debate on Chinese nationhood which were 
retroactively sidelined by teleological narratives of nation-building.156 However, his 
analysis remains confined to the realm of discourse; here I am interested in exploring the 
material aspect of these conceptions of nation alongside the rhetoric and examining their 
complex interplay.
Despite their loose alliance against the Qing government, these provincial 
movements exhibited an idiosyncratic mix of supporters, goals, and resources, and 
engaged in a certain amount of inter-provincial competition: during the warlord period, 
this rivalry extended to all-out war. In different provinces, merchants, intellectuals, and 
government officials exerted a different amount of influence and pursued different goals. 
In the following, three different provinces with very different histories of rail 
development will be examined.
Shanxi
The development of Shanxi provides an especially interesting case. Early attempts 
by British companies to develop the province's mineral resources were met with a strong 
and ultimately successful Rights Recovery Movement. While connections with country-
wide protests were undeniably present, the organization of the protests was done at the 
provincial level by provincial elites and new-school students, supported in Beijing by 
156. Duara, Rescuing History from the Nation.
105
fellow provincials. The rhetoric of the movement reflected this: criticisms of the officials 
responsible for selling the concessions and rationales for protecting Shanxi's coal both 
emphasized provincial identities and provincial needs.157
Shanxi's provincial trajectory continued to follow its own path after the Xinhai 
Revolution. Under the leadership of Yan Xishan, Shanxi maintained political 
independence from the nascent Republic and pursued its own course of economic 
development. Along with the opening of mines and factories, a rail project also enjoyed a 
prominent role in the province's development.  Directed by Yan himself, the Datong-
Puzhou railroad158 had several unique characteristics: first, it avoided foreign ownership 
entirely. Second, Yan worked to manufacture construction supplies, from concrete to 
hammers, within Shanxi as much as possible. Most intriguingly, Yan deliberately chose a 
different gauge than that which predominated in the rest of China. This was partially an 
economic decision, as the narrower gauge was cheaper, and also partially military: the 
gauge difference prevented any invaders (at this point almost certainly rival warlords or 
Chiang Kaishek's Nationalists) from using rail in Shanxi. Yan Xishan's engines, however, 
could be adapted to the wider rail quite easily, leaving his aggressive capabilities 
unchecked.159
The example of Shanxi shows how rail development is entangled with broader 
issues of economic development. It also exemplifies the role the state plays in shaping the 
scale that rail and economic change manifests on and the boundaries of the circulation 
157. Thompson, “‘If Shanxi’s Coal Is Lost, Then Shanxi Is Lost!’.”
158. 同蒲铁路 (tóng pú tiělù)
159. Jing, “Yan Xishan yu Tongpu tielu.”
106
they enable; in Shanxi the scale that dominated all these aspects was the provincial, not 
the national.
Sichuan
The 1911 Railway Protection Movement erupted in response to the Qing 
government's unilateral decision to nationalize the provincial rail companies: the edict 
transformed associations assembled to raise consciousness and capital into nascent 
revolutionary cells.160 Often seen as part of the nationalist push that culminated in the 
Xinhai Uprising and the establishment of the Republic of China, the theoretical approach 
laid out in the previous chapter suggests a re-evaluation of the Railway Protection 
movement in Sichuan.
In Sichuan, the protests against the Qing dynasty's move to nationalize the 
railways was fierce, forcing the Qing government to divert an army to put down the 
uprising, inadvertently leaving Xinhai vulnerable. Yet unlike other provinces such as 
Zhejiang or Guangdong where the rail company had received substantial popular, 
voluntary support, Sichuan's rail company had been funded largely via coercive, feudal 
means, including taxes on basic foodstuffs that caused widespread hardship.161 
Despite having raised a considerable amount of capital this way, the company had 
succeeded in building only seventeen kilometers of rail line, as officials were busy 
investing the money elsewhere for personal profit. The best most Sichuanese hoped for 
was that the rail wouldn't lead to disaster.162 Far from being the place where rail had the 
160. Rankin, “Nationalistic Contestation and Mobilization Politics.”
161. Jin, “Ershi shijiechu gesheng shangban tielu jiqi jieju,” 94.
162. Jin, “Ershi shijiechu gesheng shangban tielu jiqi jieju.”
107
most nationalist support, it was the place with the least. There is another interpretation of 
the uprising: less a fervent nationalism than a more traditional anger at the exploitative 
excesses of feudal overlords: it was bad enough that their harvests were taxed to build a 
railway, but then they took even the railway! 
The problem of the Sichuan Railway Protection Movement is symptomatic of 
many discussions of political movements: are they motivated by nationalist or provincial 
consciousness, or by simple self-interest? In other instances, nationalist rhetoric was used 
to attract investment from scattered, minor shareholders, or used as a palliative to quiet 
unrest when rail companies found it difficult to pay interest. It can easily be argued that 
officials and business people in charge were simply taking advantage of nationalist 
ideology or provincial consciousness to forward their own interests. That is ultimately a 
trivial observation. The more interesting question is: why did that particular rhetorical 
strategy appeal, and why did it succeed in persuading its audience? What part of it rang 
true to its listener's experience?
Manchuria
Manchuria in the first half of the twentieth century was a region of complex 
loyalties. Chinese nationalists were a vocal, but not particularly powerful contingent; 
powerful elites in the region preferred a status quo of relative autonomy, and were as 
interested in strengthening ties with Japan or devolving power to the individual provinces 
as they were with integrating with greater China; and a great number of peasants, soldiers 
and bandits were loyal to little beyond their immediate community. Nor did external 
powers, including the KMT, necessarily regard Manchuria as of a piece with China; 
108
Japan and Russia, rather strategically, viewed Manchuria's connection with Beijing or 
Nanjing as somewhat ambiguous. 
Politically and economically, this connection was in fact tenuous: governed by an 
independent warlord and foreign railroad companies, selling its exports to Japan and the 
world, only populated by Han Chinese in the late Qing, Manchuria was a distinct space. 
Manchuria was also fractured internally: in the first half of the twentieth century, 
Manchuria was home to China's most cosmopolitan and its most isolated communities. In 
The Manchurian Myth, Rana Mitter argues that it was only in the wake of the Japanese 
invasion that Manchuria came to be largely seen by Chinese within the region and 
without as an unalienable part of China. While the narrative of nationalist resistance is 
what has been remembered, the reality consisted as much of collaboration and co-
existence as resistance, and much of the resistance was less than strictly nationalist in 
ideology.163 
The form and function of Manchuria's railroads worked against political and 
economic integration, widening divisions between Manchuria and the rest of China and 
tying the region into  separate spheres of circulation. The fastest growth of Chinese rail 
infrastructure happened in the Northeast, which by 1949 had over 40% of China's entire 
track length.164 The Japanese-run South Manchuria Rail was designed to tie its associated 
territories into the Japanese imperial economy in the south, and the Russian-operated 
China Eastern Rail made Harbin into a unique melting pot of Russian, Chinese and other 
cultures. Rail construction had a dramatic impact on economic and social structure, 
163. Mitter, The Manchurian Myth.
164. Wang et al., “Spatiotemporal Evolution of China’s Railway Network in the 20th Century.”
109
giving rise to rapid urban and economic growth in what had previously been a pastoral, 
sparsely populated region.165
Manchuria was not, under the Qing, a typical region of the Chinese empire. 
Rather, as the homeland of the ruling Manchus, it was closed off to settlement by Han 
farmers and kept as a preserve of Manchu culture.166 In the late nineteenth century, this 
policy was relaxed and eventually abandoned in an attempt to counter Russian and 
Japanese designs on the region; over the next five decades millions of migrants from 
Shandong and Hebei poured into the region. Many of these migrants were temporary 
sojourners, returning home after several years of work, but enough settled to increase the 
population of Manchuria by 8 million by 1942.167 While the influx of Han population 
rendered the society of the Northeast increasingly like that of Northern China, in some 
key respects the pattern was quite different: in contrast to the clear delineation of 
economy and politics of the North China plain, in Manchuria the economic and political 
were far more integrated, giving county seats an unusual level of power.168 
The Manchurian labor market was therefore relatively well-integrated into China, 
but the commodities produced had a very different sphere of circulation: the railroads 
necessitated both a large industrial sector, to service their own needs, and an export-
driven economy, to justify their existence to their imperial masters. Not only did this give 
rise to the urban centers of Harbin and Dalian, but a number of depot towns sprung up as 
165. Elleman, Manchurian Railways and the Opening of China.
166. Duara, Sovereignty and Authenticity, 41.
167. Thomas R. Gottschang, “Economic Change, Disasters, and Migration: The Historical Case of 
Manchuria,” Economic Development and Cultural Change 35, no. 3 (April 1, 1987): 461–490, 
doi:10.2307/1153926.
168. Duara, Sovereignty and Authenticity, 44.
110
well. The South Manchuria Railway (SMR)'s own need for iron and steel drove the 
development of the largest ironworks in East Asia in Anshan, which then served as a 
nucleus for further industrialization. The demands of the SMR also served to catalyze a 
timber industry that became one of the world's top exporters.169 The SMR also expanded 
into glass, ceramics, and other industries, driving out Chinese competition and 
monopolizing the industrial sector as a whole.170 The SMR played nearly as important a 
role in the Japanese economy as in Manchuria: in the 1920s, the SMR provided a quarter 
of the Japanese government's tax revenue.171
The single largest sector of the Manchurian economy, however, was the growth 
and export of soya—an endeavor in which the SMR was deeply involved. Despite a 
continuing use of traditional agricultural methods, the presence of rail transport 
nonetheless radically transformed Manchurian agriculture, stimulating increased 
specialization in and dependence upon a single crop. This had impacts both global—tying 
the fates of farmers into changes in global commodity prices—and local—farmers were 
dependent on county soya merchants.172 Soya-based commodities represented 80 percent 
of Manchuria's exports, and provided 59 percent of the global supply in the late 
twenties.173 
Manchuria was thus already an important part of the economy of the Japanese 
empire long before it was politically assimilated. Manchuria's economics were controlled 
169. Ibid., 172.
170. Mitter, The Manchurian Myth, 67.
171. Louise Young, Japan’s Total Empire: Manchuria and the Culture of Wartime Imperialism (Berkeley: 
University of California Press, 1998), 32.
172. Duara, Sovereignty and Authenticity, 46. 
173. Ibid., 48.
111
by Japanese interests, backed by Japanese capital, and sold to Japanese markets. Other 
than the remittances sent home by workers, very little of Manchuria's economy had any 
relation to the rest of China. In the late 1920s, Zhang Xueliang attempted to re-orient 
Manchuria's economy towards China and away from Japan by building an alternate deep-
sea port, refused to sell land to Japanese citizens, and developing alternate rail lines.174 
This substantive economic re-orientation, as contrasted with the largely nominal political 
re-orientation towards Nanjing, posed a significant threat to Japanese interests, and set 
the stage for the occupation—an attempt to reconcile the economic and political spheres.
Manchuria's separation from the rest of China was not only economic, however. 
Politically, early twentieth century Manchuria also had a distinctly regional sphere. As 
will be discussed later in more detail, the government of the late Qing, unable to manage 
needed military and economic reforms, devolved substantial power to provincial officials. 
The assortment of provincial and local movements that precipitated the collapse of the 
Qing dynasty retained a great deal of autonomy under the Republic, and after Yuan 
Shikai's death, the central government dissolved into a shifting series of regional 
warlords. Manchuria was no different: from 1916 to 1928, it was ruled by the military 
and civil governor Zhang Zuolin, and then by his son, Zhang Xueliang. Chiang Kai-shek 
and the KMT fought through the twenties to bring the warlords under his control, but 
Chiang was forced to rely on co-option as much as conquest. 
Manchuria was by no means a uniform or internally homogenous space. Zhang's 
control of the region was not only compromised by the autonomy of the SMR and CER 
zones, but challenged from within by provincial elites, banditry, and self-interested 
174. Ibid., 51; Mitter, The Manchurian Myth, 67.
112
localities. Provincial elites in Heilongjiang and Jilin were dissatisfied with and regularly 
challenged Liaoning-based rule—fracture lines eagerly exploited by the Japanese in 
1931.175 Zhang Zuolin's multiple failed attempts to seize Beijing, and the tax burden 
imposed for this purpose, inspired a strategic disinterest in Chinese affairs among 
Northeastern elites.176 Even within the military, the Zhangs' base of support, the lines of 
loyalty were blurry: Mitter documents the long history of banditry in Manchuria and the 
circulation of soldiers through banditry and back again.177 Soldiers were often more loyal 
to their individual commander—or former bandit leader—than to the chain of command. 
Similarly, Shenyang's control over local-level polities was limited, often no more than 
formalizing existing authority, and tapered off sharply with distance.178 In Harbin, where 
the currents of social transformation ran strong, the foreign presence served as a catalyst 
for the emergence of active student- and merchant-led nationalist movements—not 
always with quite the same strategies and goals.179 
Local, provincial, and regional identities thus provided nationalism with stiff 
competition. Competing loyalties did not only come from within: the influential role 
played by the Japanese inclined some elites to consider Japan as a rational alternative to 
the KMT. The most prominent such politician was Yu Chonghan, who had long been an 
advocate of separatism and had long-standing Japanese connections.180 Mitter argues that 
his allegiance to Japan was both the product of personal financial benefit and of genuine 
175. Mitter, The Manchurian Myth, 37–38; 79–81.
176. Duara, Sovereignty and Authenticity, 50; Mitter, The Manchurian Myth, 36.
177. Mitter, The Manchurian Myth, 38–41.
178. Ibid., 63.
179. Carter, Creating a Chinese Harbin.
180. Mitter, The Manchurian Myth, 36.
113
belief in the preferability of Japanese rule: disillusioned with corruption, taxes, and 
military-centric policies under warlordism, he advocated alliance with Japan for the 
purpose of governmental reform and infrastructural development.181 The potency of 
Japanese reputation for governmental efficiency should not be underestimated. Mitter 
repeats an anecdote showing even those who cheerfully cursed the Japanese occupation 
were forced to admit, once across the border into China proper, that it had some 
advantages. Upon hitting a savage bump: “Aiya, if this was still in Manchoukuo [sic], it 
would be a proper road. They are not allowed to steal the road-making money there!” The 
Japanese government of Manchuria enjoyed very little legitimacy. What little it did 
accumulate, however, was largely the result of its good governance and infrastructure
—“good roads as propaganda.”182
This is the situation under which Zhang allied himself with the KMT. Rana Mitter 
argues that when Zhang agreed to join forces in 1928, the relationship was more of an 
alliance between powers than the re-union of a nation.183 Zhang continued to enjoy near-
complete autonomy in governing the Northeast as well as in pursuing his own foreign 
policy. The CER Incident illustrates the degree of both his freedom and his isolation. In 
1929, the Chinese governor of Harbin made a play to bring the Chinese Eastern Railroad, 
the Russian-operated line in northern Manchuria, under his (and therefore Zhang's) 
control. This plan backfired when the Soviets retaliated with surprising force, deploying 
aircraft to back their claim to the CER. Mitter documents that the KMT government in 
181. Ibid., 95, 109.
182. Ibid., 123–24.
183. Ibid., 52–53.
114
Nanjing paid little interest to this minor debacle: “even an international conflict such as 
the CER incident registered as only a regional problem.”184 
The complexities of identity in pre-occupation Manchuria can be seen clearly in 
the work that went into producing resistance as a nationalist endeavor. Mitter shows that 
a sense of Manchuria as an inextricable part of China, and therefore the invasion as a 
violation of Chinese territorial integrity, had to be actively produced. We should therefore 
understand “the move to nationalism among the resistance as a dynamic process and not a 
preexisting situation.”185 One center for this process was, unsurprisingly, the armies of 
national resistance in Manchuria. The other location of anti-Japanese agitation was 
Shanghai. In Shanghai, public opposition to Japanese aggression solidified as their own 
struggles with Japan deepened and a corps of dedicated propagandists worked to 
construct a discourse of pan-Chinese anti-Japanese resistance. Resistance fighters against 
the Japanese such as Ma Zhanshan initially pursued the prevalent bandit-soldier strategy 
of patronage-seeking to maximize their own power, but as the struggle wore on, 
increasingly understood themselves in a nationalist sense. 
Initial responses to Japanese aggression were subdued both in the KMT 
government and the Chinese public. Chiang Kai-shek's reasons for adopting a strategy of 
non-resistance, and Zhang Xueliang's reasons for going along, are impossible to know for 
certain.186 It was enabled, however, by the general disinterest in the Manchurian Incident 
on the part of the Chinese public at large. This changed, however, both because of 
184. Ibid., 50.
185. Ibid., 201.
186. Chiang's reasoning may be illuminated by his interpretation of the Goujian story discussed in Chapter 
II. See Cohen, Speaking to History, 71. 
115
dedicated efforts by Manchurian partisans to link the Manchurian situation into the larger 
nationalist narrative and due to the expansion of Japanese aggression into the larger 
Chinese sphere. The Shanghai Incident, where the Japanese Navy attacked Shanghai, 
gives an example of how relatively small groups of nationalized elites spread their ideas 
into general circulation. A boycott of Japanese goods, organized by nationalists, provoked 
the Japanese navy into an attack on Japan. This assault, quickly linked with the 
Manchurian Incident by the pro-Manchurian propagandists, spurred an increase of public 
support for the Manchurian resistance, with whom Shanghainese had had little 
sympathy.187 
Mitter picks out the example of Ma Zhanshan to illustrate shifting ideology 
among the resistance fighters. Ma's initial resistance to the Japanese was self-interested 
rather than principled: he resisted in order to extract concessions from the Japanese and 
preserve his own power. He was nevertheless represented as a nationalist, and 
consequently attracted considerable support from anti-Japanese nationalist 
organizations.188 Ma is here quite representative: despite propaganda to the contrary, only 
five percent of the resistance armies were committed nationalists. The bulk was made up 
of landless peasants and soldier-bandits; many “resistance armies” were more interested 
in plunder than in fighting the Japanese.189 Nonetheless, his perception outside Manchuria 
made Ma's 1932 decision to join the Japanese a powerful symbol of betrayal.190 Mitter 
187. Mitter, The Manchurian Myth, 140.
188. Ma, Mitter explains, was illiterate and therefore his communications were written by more educated 
advisors—typically quite nationalist themselves. There is thus an ambiguity to what extent Ma 
intended to represent himself in such nationalistic terms, much less how he viewed his own actions. 
See Ibid., 204. 
189. Ibid., 197, 199 .
190. Ibid., chap. 7.
116
argues that this blowback forced Ma to reconsider his strategy: it was no longer tenable to 
switch sides in pursuit of the most advantageous patron; instead he had to choose 
between nationalism and treason.191 While he ultimately chose the former, many chose 
the latter: only about one percent of the local population engaged in anti-Japanese 
resistance.192 
The work that went into creating a perception of Manchuria as a part of China, 
and the spaces where it failed, ought to be as notable as its successes. While nationalism 
was far from powerless, its pull was tempered by political and economic realities that 
pulled Manchuria into the Japanese sphere, Russian sphere, or divided it into yet smaller 
circulations. The Manchurian rail system, the most developed in China, played no small 
part in shaping this reality: Manchuria was populated by Chinese but incorporated into 
the Japanese and Russian economies, and its power brokers had much to lose by adhering 
to a strict nationalism. Rail created spheres of circulation that cut against the cultural and 
linguistic characteristics that form such important parts of nationalism, putting into 
conflict different nationalizing processes. The resolution of this conflict did not take place 
until after 1949.
An infrastructural lens
As can be seen from the above, an understanding of the role played by large-scale 
infrastructure projects in mediating the relation between governance, territory, and 
popular consciousness is essential to understanding the political terrain of early twentieth 
century China. In this period, the lack of central government power led to the devolution 
191. Ibid., 218–19.
192. Mitter, The Manchurian Myth.
117
of considerable state power to provincial governments; these governments then played an 
influential role in pushing local economic change. This economic change had profound 
social consequences, radically altering the cultural and political geography of China and 
creating spheres of circulation on a much wider scale than previously possible. The 
power exerted by provincial governments, on the one hand, and foreign powers on the 
other, produced an incredibly uneven, fractured geography: the impacts of increased 
circulation were as negligible in some places as they were profound in others. Rail 
projects in particular took on symbolic and practical importance, producing a conception 
of the province as the legitimate and effective guarantor of rights even as it became a 
mechanism for the centralization of capital and political power. 
Understanding the relation between state infrastructure and political 
consciousness can clarify the ways in which the fractured transportation network and the 
diverse scales of political consciousness in early twentieth century China were mutually 
reinforcing, each serving as the preconditions to the reproduction of the other. The 
relation between changing social and economic realities and the evolving political 
rhetoric is also clarified by this approach: nationalist and provincial rhetoric was both 
responding to and shaping the nature of social and political participation. This theoretical 
approach is not limited to explaining heterogenous political environments but can also be 
applied to analyze the impacts of consolidation under the CCP and the boom in mobility 
in the wake of the 1978 reforms. This approach, by highlighting the impact of different 
geographies of circulation have on political consciousness, provides a valuable lens 
through which to analyze the relationship between economic and social change and the 
emergence of nationalism.
118
CHAPTER V
BUILDING BEYOND THE NATION
This thesis re-centers the discussion of Chinese nationalism around the question 
of material social transformation and moves it away from a focus on discourse and the 
impact of state power. Where other scholarship dwells on intellectual work done by 
prominent nationalist thinkers and explicitly nationalist social movements, this work 
shifts the emphasis in the direction of the transformation in the everyday experiences of 
the Chinese populace. Where other scholarship emphasizes the role of state power in 
manufacturing or controlling nationalism, this work highlights the character of 
nationalism as an unintended, capricious by-product of state power, with its own logics 
and drives separate from, and often at odds with, the state.
Rather than being a unitary whole, this thesis treats nationalism as a nationalizing 
project, composed of many different mechanisms with different scopes and impacts. 
These mechanisms emerge out of capitalism and industrialization: as work and society 
are transformed, so too are cultures and identities. Homogenization of work and 
environment deepens the role of culture in mediating between people and the human 
environment, thereby increasing the scale at which populations can circulate from the 
local to the regional, or national—but differences between national cultures are therefore 
drawn all the more plainly. The impacts of increasing scale and deepening cultural 
dependency can be seen clearly in the development of transportation infrastructure, 
particularly rail. A sophisticated understanding of rail travel and how to navigate it frees 
individuals to travel quickly and easily over a wide swath of the earth—but only that 
119
swath which has been transformed by massive infusions of capital and labor. The 
imagined territories thus defined play a critical role in shaping the national geo-body.
Rail's tendencies towards the creation of wider spheres of circulation produced a 
fractured geography of political consciousness in China. The devolution of state power 
towards provincial governments and foreign colonies during the end of the Qing, and the 
influence of regional warlords and imperialist powers during the Republican period led to 
rail primarily being conceived of and executed at regional scales, producing only regional 
spheres of circulation, reinforcing regional governance and political consciousness. Only 
after 1949 was a China-wide rail network built.
The overview of Chinese rail development in Chapter IV is necessarily 
superficial, given the gap between the limited scope of this thesis and the depth of the 
subject. The impact of rail on developing nationalist consciousness in the northeast would 
be a particularly fruitful avenue of exploration: the role played by Japanese rail in the 
development of Chinese identity would benefit greatly from an approach that emphasizes 
disjuncture between intent and outcome in the nationalization process. Even Japanese rail 
can be understood as contributing to the formation of Chinese identity, both within 
Manchuria and without. 
The development of the rail network in the post-1949 era, under Mao, is another 
direction this research agenda could be developed—nationalization did not cease or reach 
completion at the end of the Chinese Civil War. The development of national identity 
under Mao is poorly understood: the period's rhetoric was dominated by Party messages 
on international capitalism and class struggle, and a discourse-focused analysis is 
necessarily stymied. Rail construction, on the other hand, proceeded steadily, more than 
120
doubling between 1949 and 1974.193 Can the intensification of the rail network be used as 
a proxy for an ongoing process of nationalization? Or did the tight controls on travel limit 
its effect? One advantage of this analytic is its ability to continuously trace some aspects 
of nationalization through the early Republican period, Mao, and into the post-reform era 
despite the rhetorical confusion.
Changes in transportation have also played a major role in China's post-reform 
period, most noticeably in the form of China's floating population. This is not only a huge 
population traveling long distances regularly, but they are largely drawn from rural 
populations where other nationalizing projects like standardized language and education 
had the weakest influence. If the experience of rail travel plays as significant a role in the 
formation of a sense of national identity as I have argued, it ought to be evident here. 
Taking seriously the argument that nationalization is a constantly ongoing project, the 
constitutive moment of Chinese nationalism is as much now as 1911: as more and more 
Chinese travel regularly, their understandings of themselves as Chinese and China as a 
territory are undergoing vital changes. The recent spike in nationalism needn't, be due to 
government manipulation, but to social shifts beyond the government's direct control.
Rail can be a powerful entry into discussions of Chinese nationalism's evolution 
over the past century. It can also serve as a productive entry into analyses of state power. 
Given the significant role of the state in instigating and regulating large-scale 
infrastructure projects like rail and the centrality of those projects in the popular 
perception of modern society, I am tempted to propose a new definition of the state: the 
state is that which regulates and even creates the built environment upon which society is 
193. Wang et al., “Spatiotemporal Evolution of China’s Railway Network in the 20th Century.”
121
dependent for its existence. This definition cuts across other definitions of the state, such 
as its territorial monopoly on violence, in intriguing ways: a monopoly on violence is one 
of the preconditions for the function of society, and its territorial scope comes from the 
the inherently territorial nature of the built environment. Institutions are state-like insofar 
as they produce and maintain social infrastructure: accountability to popular opinion, 
distinctions between state and civil society, their unitary or fragmentary nature, are 
merely characteristics they may or may not exhibit. This approach seems highly 
compatible with calls to think through the state as a social relation, with an added benefit 
of focusing attention on its often poorly-conceptualized territorial aspect.194
This thesis also suggests ways in which the theory of nationalism can be moved 
forward. While transportation infrastructure plays a particularly clear role in producing 
the concept of national territory, a multitude of other large-scale infrastructure projects, 
including mass communication, mass education, and urban planning can have similar 
effects. Each of these projects will have different scales at which they operate, different 
sections of society they impact and will create different senses of national consciousness. 
One crucial ramification of disassembling nationalization is the realization that 
even when the bounds of the nation are agreed upon, not all parts of the nation will have 
been nationalized by the same mechanisms. To reiterate one example, the working class 
and the managerial class have very different experiences of industrial work—their 
conceptions of nation are unlikely to be entirely identical. Foucault discusses the 
importance of the military in creating modern disciplined subjects; some recent 
scholarship also suggests that this is an important mechanism for the creation of a 
194. Painter, “Prosaic Geographies of Stateness.”
122
national identity—yet one that does not affect every member of the nation equally. 195 
Examining the different means by which national identity is instilled in different 
geographic, ethnic, or socio-economic portions of the population may shed light on 
certain struggles over the meaning of the nation and how nationalism becomes the 
unquestioned terrain on which domestic political disputes are fought.
Take, for example, the wage struggle. The group of workers who share the same 
working language, the same work habits, the same sense of time discipline, who navigate 
and dwell within the same built environment can be seen as a vast guild: their shared 
skills are, collectively, a monopoly that may be wielded against capital in the struggle 
over wages and working conditions—assuming that they can be organized effectively. 
Quite aside from the bonds of sympathy between members of the same society and 
culture (though the importance of those bonds cannot be discounted), developing the 
national group as a self-aware entity is a powerful negotiating strategy in the struggle 
over wages.
Yet capital cannot comfortably oppose the development of a national 
consciousness. For even as the skills upon which the national identity is built are the 
property of the workers, their character and distribution are bespoke by capital. The 
infrastructure those skills interface with are investments that capital must realize—in a 
sense, so are the skills themselves. While an inadvertent and contrary consequence, the 
development of nationalism is inseparable from the drive to realize capital. Thus capital 
finds itself entangled with particular assemblages of infrastructure, language and people; 
the same processes that develop its freedom of action within that arena draw the 
195. Conversi, “Homogenisation, Nationalism and War.”
123
boundaries of that arena ever starker. This suggests one explanation as to why class 
struggle has consistently taken place within the confines of nations, rather than adopting 
the international character Marxists always insisted it ought.
This angle on nationalism can also be used to analyze differences between as well 
as within different countries. Just as different groups within the nation are impacted 
unevenly by nationalizing processes, so too are entire countries nationalized by different 
mixes of nationalizing processes. One relatively well-examined distinction is between 
“traditional” nationalisms and post-colonial nationalisms. An aspect of nationalism that 
has been neglected by this thesis is the crucial role played by the international (more 
properly but awkwardly termed inter-state) state system in producing and structuring 
nationalism. As the first nationalisms were coalescing, the lack of precedent meant that 
internal conditions were the driving force. As the concept of the nation-state was more 
fully articulated, inter-state rivalry and outright imperialism played a more central role in 
disseminating ideas of nationhood: particularly in colonial regions, external conceptions 
of national boundaries and identities shaped domestic understandings. To take China as 
an example, external conceptions of the Chinese nation as delineated by the boundaries of 
the Qing empire clashed with domestic conceptions emerging out of pre-existing Han 
ethnic chauvinism. National identity as mobilizing to resist foreign imperialism is a very 
different thing than national identity as mobilizing to survive the impacts of 
industrialization. 
Once nationalism's self-representation as a singular object, a historical subject 
developing along teleological lines is understood as just that—a self-representation—the 
diverse, uneven, at times self-contradictory field that constitutes the nation is opened as a 
124
vast field of study. Its implications for the understanding of territory, the state, the built 
environment, and capitalism are immense, and this thesis represents only the first steps 
towards understanding. Nationalism is thus a vital component of the modern world, and 
anyone who uses the idea of nation, even to define an unrelated research subject, must 
take a closer look that this complex, crucial field.
125
REFERENCES CITED
Adey, Peter. “If Mobility Is Everything Then It Is Nothing: Towards a Relational Politics 
of (Im)mobilities.” Mobilities 1, no. 1 (2006): 75–94. 
doi:10.1080/17450100500489080.
Agnew, John. “Still Trapped in Territory?” Geopolitics 15, no. 4 (2010): 779–784. 
doi:10.1080/14650041003717558.
———. “The Territorial Trap: The Geographical Assumptions of International Relations 
Theory.” Review of International Political Economy 1, no. 1 (April 1, 1994): 53–
80. doi:10.2307/4177090.
Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of 
Nationalism. London [etc.]: Verso, 2006.
Bagwell, Philip Sidney. The Railway Clearing House in the British Economy 1842-1922,. 
London: Allen & Unwin, 1968.
Barnes, Trevor J, and Meric S Gertler. The New Industrial Geography: Regions, 
Regulations and Institutions. London; New York: Routledge, 1999.
Bissell, David. “Vibrating Materialities: Mobility–body–technology Relations.” Area 42, 
no. 4 (2010): 479–486. doi:10.1111/j.1475-4762.2010.00942.x.
Brødsgaard, Kjeld, and David Strand, eds. Reconstructing Twentieth-Century China : 
State Control, Civil Society, and National Identity. Oxford: Oxford University 
Press, 1998.
Callahan, William A. “History, Identity, and Security: Producing and Consuming 
Nationalism in China.” Critical Asian Studies 38, no. 2 (June 2006): 179–208. 
doi:10.1080/14672710600671087.
Carter, James Hugh. Creating a Chinese Harbin: Nationalism in an International City, 
1916-1932. Ithaca: Cornell University Press, 2002.
Chang, Maria Hsia. Return of the Dragon: China’s Wounded Nationalism. Boulder, 
Colo.: Westview Press, 2001.
Cohen, Paul A. Speaking to History: The Story of King Goujian in Twentieth-Century 
China. Berkeley: University of California Press, 2009.
Colley, Linda. “Whose Nation? Class and National Consciousness in Britain 1750-1830.” 
Past & Present no. 113 (November 1, 1986): 97–117. doi:10.2307/650981.
Collins, Alan. “Chinese Educationalists in Malaysia: Defenders of Chinese Identity.” 
Asian Survey 46, no. 2 (April 1, 2006): 298–318. doi:10.1525/as.2006.46.2.298.
126
Conversi, Daniele. “Homogenisation, Nationalism and War: Should We Still Read Ernest 
Gellner?” Nations & Nationalism 13, no. 3 (July 2007): 371–394. 
doi:10.1111/j.1469-8129.2007.00292.x.
Coole, Diana H, and Samantha Frost. New Materialisms: Ontology, Agency, and Politics. 
Durham [NC]; London: Duke University Press, 2010.
Daly, Nicholas. “Blood on the Tracks: Sensation Drama, the Railway, and the Dark Face 
of Modernity.” Victorian Studies 42, no. 1 (October 1, 1998): 47–76. 
doi:10.2307/3829126.
Duara, Prasenjit. Rescuing History from the Nation: Questioning Narratives of Modern 
China. Chicago: University of Chicago Press, 1995.
———. Sovereignty and Authenticity: Manchukuo and the East Asian Modern. Lanham: 
Rowman & Littlefield Publishers, 2003.
Elleman, Bruce A. Manchurian Railways and the Opening of China: An International 
History. Armonk, N.Y.: M.E. Sharpe, 2010.
Elvin, Mark. The Retreat of the Elephants: An Environmental History of China. New 
Haven: Yale University Press, 2004.
Fitzgerald, John. “The Nationless State: The Search for a Nation in Modern Chinese 
Nationalism.” The Australian Journal of Chinese Affairs no. 33 (January 1, 1995): 
75–104. doi:10.2307/2950089.
Foucault, Michel. Discipline and Punish: The Birth of the Prison. New York: Vintage 
Books, 1979.
Gao, James Zheng. Meeting Technology’s Advance: Social Change in China and 
Zimbabwe in the Railway Age. Westport, CT: Greenwood Publishing Group, 
1997.
Gellner, Ernest. Nations and Nationalism. Ithaca: Cornell University Press, 1983.
Gottdiener, Mark. Life in the Air: Surviving the New Culture of Air Travel. Lanham, Md.: 
Rowman & Littlefield, 2001.
Gottschang, Thomas R. “Economic Change, Disasters, and Migration: The Historical 
Case of Manchuria.” Economic Development and Cultural Change 35, no. 3 
(April 1, 1987): 461–490. doi:10.2307/1153926.
Gries, Peter Hays. China’s New Nationalism: Pride, Politics, and Diplomacy. Berkeley: 
University of California Press, 2004.
Hadjimichalis, Costis. “Non-Economic Factors in Economic Geography and in ‘New 
Regionalism’: A Sympathetic Critique.” International Journal of Urban and 
Regional Research 30, no. 3 (September 2006): 690–704.
127
Haraway, Donna Jeanne. Simians, Cyborgs, and Women: The Reinvention of Nature. New 
York: Routledge, 1991.
Harman, Graham. Tool-Being: Heidegger and the Metaphyics of Objects. Chicago: Open 
Court, 2002.
Harvey, David. The Limits to Capital. Chicago: University of Chicago Press, 1982.
Hess, Martin. “‘Spatial’ Relationships? Towards a Reconceptualization of 
Embeddedness.” Progress in Human Geography 28, no. 2 (April 1, 2004): 165–
186. doi:10.1191/0309132504ph479oa.
Hsu, Jinn-Yuh, and AnnaLee Saxenian. “The Limits of Guanxi Capitalism: Transnational 
Collaboration between Taiwan and the USA.” Environment and Planning A 32, 
no. 11 (2000): 1991 – 2005. doi:10.1068/a3376.
Jiang, Pei. “Zhongguo jindai tielushi yanjiu zongshu ji zhanwang: 1979 ~ 2009 [中国近代
铁路史研究综述及展望:1979～2009].” 过去的经验与未来的可能走向——中国近代史研究三十年（1979-2009) (2009): 22.
Jiang, Pei, and Yaping Xiong. “Tielu yu Shijiazhuang chengshi de jueqi [铁路与石家庄城市的崛起:1905—1937年].” Modern Chinese History Studies 近代史研究 no. 03 
(2005): 170–197.
Jin, Shixuan. “Ershi shijiechu gesheng shangban tielu jiqi jieju [二十世纪初各省商办铁路及其结局].” Journal of Beijing Jiaotong University 北方交通大学学报 no. 02 
(1977): 80–99.
Jing, Zhankui. “Yan Xishan yu Tongpu tielu [阎锡山与同蒲铁路].” Vicissitudes 沧桑 no. 
03 (1993): 26–29.
Jones, Andrew. “Beyond Embeddedness: Economic Practices and the Invisible 
Dimensions of Transnational Business Activity.” Progress in Human Geography 
32, no. 1 (February 1, 2008): 71–88. doi:10.1177/0309132507084817.
Kaika, Maria. City of Flows: Modernity, Nature, and the City. New York: Routledge, 
2005.
Ladany, L. The Communist Party of China and Marxism, 1921-1985: A Self Portrait. 
Stanford, Calif.: Hoover Institution Press, Stanford University, 1988.
Latour, Bruno. “Trains of Thought: Piaget, Formalism, and the Fifth Dimension.” 
COMMON KNOWLEDGE 6, no. 3 (1997): 170.
Lee, Haiyan. Revolution of the Heart: A Genealogy of Love in China, 1900-1950. 
Stanford, Calif.: Stanford University Press, 2007.
128
Li, Zhancai. “Tielu yu jindai Zhongguo chengzhen bianqian [铁路与近代中国城镇变迁].” 
Journal of Suzhou Railway Teachers College 铁道师院学报 no. 05 (1996): 34–38.
———. “Tielu yu Zhongguo jindai de minsu shanbian [铁路与中国近代的民俗嬗变].” 
Shixue Yuekan 史学月刊 no. 01 (1996): 56–61.
Liu, Hui. “Tielu yu jindai Zhengzhou mianye de fazhan [铁路与近代郑州棉业的发展].” 
Shixue Yuekan 史学月刊 no. 07 (2008): 102–109.
Liu, Wei. “Wan Qing ‘sheng’ yishi de bianhua yu shehui bianqian [晚清‘省’意识的变化与社会变迁].” Shixue Yuekan 史学月刊 no. 05 (1999): 59–65.
Löfgren, Orvar. “Motion and Emotion: Learning to Be a Railway Traveller.” Mobilities 3, 
no. 3 (2008): 331–351. doi:10.1080/17450100802376696.
Luo, Houli. “Cong sixiang shi guanjiao kan jindai Zhongguo minzu zhuyi [从思想史视角看近代中国民族主义].” Strategy and Management 战略与管理 no. 01 (1998): 103–
110.
Marx, Karl. Capital: A Critique of Political Economy. London; New York, N.Y., USA: 
Penguin Books in association with New Left Review, 1990.
———. Grundrisse: Foundations of the Critique of Political Economy (rough Draft). 
Harmondsworth London: Penguin “New Left Review,” 1973.
Meehan, Katharine, Ian Graham Ronald Shaw, and Sallie A. Marston. “Political 
Geographies of the Object.” Political Geography 33 (March 2013): 1–10. 
doi:10.1016/j.polgeo.2012.11.002.
Meinig, D. W. The Shaping of America: A Geographical Perspective on 500 Years of 
History. Vol. 3. New Haven: Yale University Press, 1986.
Mitter, Rana. The Manchurian Myth: Nationalism, Resistance and Collaboration in 
Modern China. Berkeley: University of California Press, 2000.
Morgan, Stephen L. “Personnel Discipline and Industrial Relations on the Railways of 
Republican China.” Australian Journal of Politics & History 47, no. 1 (March 
2001): 24.
Needham, Joseph, Peng Yoke Ho, Gwei-Djen Lu, and Ling Wang. Science and 
Civilisation in China. Taipei, Taiwan: Caves Books, 1986.
Painter, Joe. “Prosaic Geographies of Stateness.” Political Geography 25, no. 7 
(September 2006): 752–774. doi:10.1016/j.polgeo.2006.07.004.
Piore, Michael J, and Charles F Sabel. The Second Industrial Divide: Possibilities for 
Prosperity. New York: Basic Books, 1984.
Pomeranz, Kenneth. The Making of a Hinterland: State, Society, and Economy in Inland 
North China, 1853-1937. Berkeley: University of California Press, 1993.
129
Purcar, Cristina. “Designing the Space of Transportation: Railway Planning Theory in 
Nineteenth and Early Twentieth Century Treatises.” Planning Perspectives 22, no. 
3 (2007): 325–352. doi:10.1080/02665430701379142.
Pye, Lucian W. “How China’s Nationalism Was Shanghaied.” The Australian Journal of 
Chinese Affairs no. 29 (January 1, 1993): 107–133. doi:10.2307/2949954.
Rankin, Mary Backus. “Nationalistic Contestation and Mobilization Politics: Practice and 
Rhetoric of Railway-Rights Recovery at the End of the Qing.” Modern China 28, 
no. 3 (2002): 315–361.
Saxenian, AnnaLee. Regional Advantage: Culture and Competition in Silicon Valley and 
Route 128. Cambridge, Mass.: Harvard University Press, 1994.
Schivelbusch, Wolfgang. The railway journey: the industrialization of time and space in 
the 19th century. Berkeley, Calif.: University of California Press, 1986.
Scott, James C. The Art of Not Being Governed: An Anarchist History of Upland 
Southeast Asia. New Haven: Yale University Press, 2009.
Sheller, Mimi, and John Urry. “The New Mobilities Paradigm.” Environment and 
Planning A 38, no. 2 (2006): 207 – 226. doi:10.1068/a37268.
Simmel, Georg. Soziologie. Duncker & Humblot, 1908.
Skinner, G. William. The City in Late Imperial China. Stanford, Calif.: Stanford 
University Press, 1977.
Storper, M. “The Resurgence of Regional Economies, Ten Years Later: The Region as a 
Nexus of Untraded Interdependencies.” European Urban and Regional Studies 
European Urban and Regional Studies 2, no. 3 (1995): 191–221.
Storper, Michael, and Richard Walker. The Capitalist Imperative: Territory, Technology, 
and Industrial Growth. Oxford, UK; New York, NY, USA: B. Blackwell, 1989.
Su, Quanyou. “Jin shinian lai woguo jindai tielu shi yanjiu zongshu [近十年来我国近代铁路史研究综述].” Journal of University of Science and Technology of Suzhou 苏州科技学院学报(社会科学版) no. 02 (2005): 110–114+127.
Thompson, Roger R. “‘If Shanxi’s Coal Is Lost, Then Shanxi Is Lost!’: Shanxi’s Coal and 
an Emerging National Movement in Provincial China, 1898–1908.” Modern 
Asian Studies 45, no. 05 (2011): 1261–1288. doi:10.1017/S0026749X10000119.
Tilly, Charles. Coercion, Capital, and European States, AD 990-1990. Cambridge, Mass., 
USA: B. Blackwell, 1990.
Tsu, Jing. Sound and Script in Chinese Diaspora. Cambridge, Mass.: Harvard University 
Press, 2010.
130
Vance, James Elmon. Capturing the Horizon: The Historical Geography of 
Transportation since the Transportation Revolution of the Sixteenth Century. New 
York: Harper & Row, 1986.
Wang, Jiaoe, Fengjun Jin, Huihui Mo, and Fahui Wang. “Spatiotemporal Evolution of 
China’s Railway Network in the 20th Century: An Accessibility Approach.” 
Transportation Research Part A: Policy and Practice 43, no. 8 (October 2009): 
765–778.
Wang, Zheng. “National Humiliation, History Education, and the Politics of Historical 
Memory: Patriotic Education Campaign in China.” International Studies 
Quarterly 52, no. 4 (2008): 783–806.
———. Never Forget National Humiliation: Historical Memory in Chinese Politics and 
Foreign Relations. New York: Columbia University Press, 2012.
Watts, Laura. “The Art and Craft of Train Travel.” Social & Cultural Geography 9, no. 6 
(2008): 711–726.
Watts, Laura, and John Urry. “Moving Methods, Travelling Times.” Environment and 
Planning D: Society and Space 26, no. 5 (2008): 860 – 874. doi:10.1068/d6707.
Young, Louise. Japan’s Total Empire: Manchuria and the Culture of Wartime 
Imperialism. Berkeley: University of California Press, 1998.
Zhan, Mei. Other-Worldly Making Chinese Medicine through Transnational Frames. 
Durham: Duke University Press, 2009. http://site.ebrary.com/id/10350247.
Zhang, Jie. “Huocheshang de shehui dengji [火车上的社会等级].” Art Museum 美术馆 
no. 01 (2010): 166–177.
Zhao, Suisheng. A Nation-State by Construction: Dynamics of Modern Chinese 
Nationalism. Stanford, Calif.: Stanford University Press, 2004.
———. “Nationalism’s Double Edge.” The Wilson Quarterly (1976-) 29, no. 4 (2005): 
76–82.
131