Techniques of repair, the circulation of knowledge, and environmental transformation: Towards a new history of transportation

It is the aim of this article to put questions of maintenance and repair in the history of science and technology under scrutiny, with a special focus on technologies and methods of transportation. The history of transportation is a history of trying to avoid shipwrecks and plane crashes. It is also a history of broken masts, worm-eaten hulls, the flat tires of cars, and endless delays at airports. This introductory article assesses the technological, scientific, and cultural implications of repairing and maintaining transportation networks. We argue that infrastructures for maintenance and repair played just as important a role in the history of transportation as the wharves and factories where ships, cars, trains, and airplanes were originally built. We also suggest that maintenance and repair are important sites of knowledge production, and a historical account of these practices provides a new, decentered narrative for the development of modern science and technology.


Introduction
The philosophical paradox of the ship of Theseus asks what happens if a ship is exhibited in a museum, and its rotting pieces are replaced by replicas one by one, until no plank remains from the original construction. Is the exhibited object still the same as the original, or is it something entirely new?
It is not an accident that this abstract philosophical question was first posed about ships. Ancient and early modern ships, like twenty-first-century cars or airplanes, were constantly rebuilt, and it is anyone's guess what remained constant beyond the name. It is one of the paradoxes of transportation that although vehicles facilitate the movements of goods, people, and ideas across the globe, they themselves do not stay the same across many voyages. Consider the Avondster, for instance -an East Indiaman sunk in the Bay of Galle near Sri Lanka in the mid-seventeenth century. It was originally built in England, using local elm wood. After it was captured by the Dutch, its hull received double planking in Asian waters and its anchor was remade from non-European timber. 1 By the time it had gone down, many of the Avondster's parts had been replaced and repaired by skilled craftsmen across the world. Similar stories could be told of the twenty-first-century automobile and flight industry, which anticipate the need for constant repairs and part replacement. The size of the aircraft maintenance and repair business, for instance, is comparable to that of airplane production itself. 2 It is the aim of this special issue to put questions of maintenance and repair in the history of science and technology under scrutiny, with a special focus on technologies and methods of transportation. The history of transportation is a history of trying to avoid shipwrecks and plane crashes. It is also a history of broken masts, worm-eaten hulls, the flat tires of cars, and endless delays at airports. This special issue assesses the technological, scientific, and cultural implications of repairing and maintaining transportation networks. We argue that infrastructures for maintenance and repair played just as important a role in the history of transportation as the wharves and factories where ships, cars, trains, and airplanes were originally built. We also suggest that maintenance and repair are important sites of knowledge production, and a historical account of these practices provides a new, decentered narrative for the development of modern science and technology. As Cátia Antunes has recently emphasized, a focus on repair techniques in transportation history offers the opportunity to examine how non-European peoples and materials played an essential role in the development of large-scale technological enterprises. 3 Historians of science and technology have long preferred to study the emergence of new technologies, instead of examining the survival and maintenance of older ones. Yet, 4 5 In the field of human and especially urban geography, the work of Stephen Graham and Nigel Thrift has engineered a similar growth of interest, examining how issues of repair and infrastructure shed light on hidden inequalities in the neoliberal global order. 6 As these authors emphasize, technological repair is unpredictable and requires complex expertise: yet it often also relies on makeshift, bricolaged solutions to cope with breakdowns. The historiography of transportation has generally neglected to give attention to such breakdowns, instead traditionally focusing on inventions and constructions like the establishment of canal systems, the laying of Macadam roads, the development of steam ships, and the formation of railways. 7 The same approach pervades more recent histories of mobility and globalization that emphasize how inventions in mobility quickly spread across the world, presuming that the modern technologies that transform society run without needing much investment in their repair. 8 Of course, this is rarely, if ever, the case. Throughout history, a variety of different strategies have been developed for travelers to deal with issues of repair arising at distant locations, with important ecological and political consequences. We suggest that the history of science and technology would benefit greatly from attention to the ways in which different modes of transportation require different types of maintenance.
This special issue argues that repair in transportation matters for three different reasons, to be elaborated below. First, ships, cars, pack animals, and airplanes travel by definition. Their sites of repair are not at the sites of their construction. The materials and know-how involved in fixing and repairing these technologies, therefore, need to travel together with the thing itself or have to be created anew at a new site. Yet as sociologists of science have long argued, the transplantation of knowledge requires complex political and social coordination, and it raises concerns about the geographies of expertise and natural resources, as well. 9 The movement of knowledge about repair is not, then, simply a subsidiary function of the movement of information about ships and other transportation technologies. It travels in a way that is distinct from other forms of knowledge. Second, although recent scholarship on the circulation of scientific knowledge has emphasized the importance of travel and cross-cultural contact for the development of natural knowledge, it has not systematically considered how the material conditions of transportation affect the way knowledge develops on the road -or even how knowledge might fail to circulate at all, because of problems of transportation. 10 More work needs to be done to demonstrate how transportation conditions what kinds of knowledge are produced in cross-cultural exchanges. Third, a focus on repair helps us reorient and connect the literature on theories of modernity and environmental history. It reveals how increases in global connectivity may often result in the exploitation of local expertise and the depletion of natural resources across the globe. Travel significantly shapes what humans know about their natural environment, even as it destroys that environment. The detailed studies of transport technologies provided by the contributions of this issue make a powerful argument for the significance of movement and displacement, and attendant issues of maintenance and repair, to current narratives of the history of science and technology.

Repair strategies in the longue durée
Crashes and wrecks have always featured in the history of long-distance travel across the ages. Throughout the centuries, travelers developed a variety of strategies of maintenance and repair to ensure they reached their destination and then came back. Here, we survey the strategies of repair that have characterized technologies of transportation from the early modern period to the present day. Some travelers relied on multiple vessels or vehicles, hoping that at least one would make it through. A second, less costly alternative was to bring extra parts and skilled craftsmen on one vessel in order to rely on tried and trustworthy procedures of repair. Last, but not least, travelers could also decide to wing it, so to speak: to deal with repairs along the way, hoping to find materials and skilled workers at distant locations. These three strategies exploited natural and human resources in different ways, and had divergent impacts on the environment, local populations, and the development of natural knowledge.
As the accounts of Christopher Columbus' first voyage emphasize, exploratory travel in the age of sail often depended on taking several vessels on a voyage together at the same time. 11 If one ship broke down or sank, the others could carry the crew and the contents of the hold onwards. The strategy of sending out multiple vessels on such voyages was based upon the economics of waste, accepting that ships, and especially sailors, were disposable goods that could be replaced. By the nineteenth century, instructions to captains from the British Admiralty emphasized and repeated the message that no sailor was irreplaceable. 12 The strategy of bringing along multiple ships and sailors had obviously high material and personal costs, a defining feature of premodern travel. While most ships did survive oceanic travel, many humans died along the way. Some 15 percent of early modern travelers from Europe to the East Indies never made it to their destination, most of them dying of disease on board or in shipwrecks. 13 Arguably, it was only in the age of mass tourism that the idea of disposable ships and travelers went out of fashion. Once ships began to carry voyagers who paid for the experience, their risk of dying on the trip needed to be drastically reduced. In the twenty-first century, a time when the chance of airplane or railway catastrophes is approaching around one in ten million, the high value placed on passenger lives has turned vehicles into indispensable goods. 14 If tourism made it impossible to rely on strategies of sending out multiple vessels to ensure the safe return of at least one, ships could still continue to rely on the alternative strategy of carrying extra timber, hemp, and other materials on board to fix broken masts and rudders, as well as torn rigging, together with the carpenters who could make the repairs. 15 This strategy reduced the enormous costs of sending out multiple vessels at the same time, but it also necessitated the circulation of skilled workers and natural resources, 16 which in itself was a large undertaking. The expenses of provisioning ships with repair materials were enormous, as the surviving account books of the Navy Board attest for England. Such strategies also survived into the early age of car travel. Before the establishment of a nationwide system of mechanics, early American drivers had to carry extra repair materials and tools so that they themselves could deal with breakdowns. 16 The same strategy of redundancy is still operational in the airline industry where, for instance, most commercial aircraft are equipped and operate with four engines to ensure that the plane can fly even if three of its engines break down. 17 If the two previous strategies of repair both relied on expertise and materials at the point of origin, a third strategy relied on resources and knowledge available at distant locations. In the age of sail, sailors frequently gathered timber to replace broken parts in the course of their journeys, chopping down trees wherever they went. Arguably, Columbus became assured of the possibility of establishing regular contact with the Americas when he discovered that the Caribbean contained enough timber to restock his ships for the return voyage. On November 24, 1492, a ship's boy discovered pine groves "like thick and slender pokers, from which, he recognized, ships could be built and an infinite number of planks and masts for the largest ships in Spain." 18 Repair en route necessitated engagement with local populations, local vessels of transport, and local expertise of repair. The issue was not simply to access natural resources, but also to engage in conversation with people who could provide alternative vessels, perform complex repair work, and advise on the best way to go forward, sometimes by kidnapping indigenous pilots. 19 These interactions literally transformed particular ships and even complete fleets. Such transformations are especially well-documented for the world of the early modern Indian Ocean. 20 Several historians have pointed out, for instance, how, once European navies were established as a major player in the Indian Ocean, they immediately began to adapt to local conditions. 21 In the case of the English Navy, the Bombay dockyards emerged as stations of ship repair in the early seventeenth century. There, expert Parsi shipwrights applied their knowledge toward repairing and rebuilding English ships, making them into hybrid vessels. 22 Throughout the early modern period, English and other European naval officers debated vigorously whether one could trust foreign timber and foreign workers at such sites of ship repair. During the time of the Napoleonic wars, the English Navy seriously considered relocating its shipbuilding industry to Bombay and a heated debate ensued in Parliament. London shipwrights expressed strong concerns about competition from Bombay, claiming that Indian wood was low quality and Indian workers low skilled and untrustworthy. 23 Local interactions about repair led to the development of travel infrastructure, but also to debates about the exact nature of skilled work and to racist diatribes about the trustworthiness of foreign people. 24 Concerns about the reliability and control of workers are omnipresent in the history of transportation. 25 Individual travelers and transportation companies still constantly oscillate between the strategy of relying on equipping vehicles and vessels with extra parts at the point of origin and that of relying on foreign expertise and materials. As Kevin Borg has documented, for instance, the history of American automobile repair is a history of tensions between the bricolage work of wayside mechanics, and car producers who have been making conscious efforts to standardize repair work and repairmen from the early 1930s onwards. 26 The rise of computer diagnostics in the car industry is the story of the rise of centralized car repair, standardized labor techniques, and the alienation of the independent mechanic from his work. 27 Controlling maintenance expertise and the location of repair became major issues for transportation companies on a global scale in the twentieth century. In Central and South America, the efforts of the United States to spearhead "Pan-American" projects of canal, railway, highway, and air travel foundered on project managers' perceptions of intractably inhospitable local environments, labor, and social conditions. 28  when they discover that aircraft maintenance is performed not in the United States, but in Central America or Asia. 29 Yet a history of repair is not only a history of labor; it is also a history of environmental transformation. Strategies of carrying extra supplies placed the strain on the natural environment at the point of origin. As Karl Appuhn has documented, the shipbuilding industry of Renaissance Venice quickly resulted in the deforestation and environmental degradation of the neighboring Dalmatian coast, and the depletion of England's forests was a major issue for the members of the Royal Society in Restoration England. 30 The strategy of repairing vessels using natural resources available while traveling, in contrast, shifted the environmental consequences of transport away from Europe to distant locations. An engagement with the expertise of local populations also meant the destruction of non-European nature. As Richard Grove has shown, the Age of Explorations meant deforestation and concomitant soil erosion for the islands of the Caribbean, together with St Helena, Mauritius, and Cape Town. 31 While historians and geographers often argue that repair and restoration can be panaceas against the large-scale production of waste that leads to climate change, it is important to realize that large-scale repair projects of transportation actually contributed to the transformation of the climate on a global scale. 32

Circulation, modernity, and producing knowledge of repair
The contributions to this special issue also reveal that a focus on repair and maintenance in transport networks provides a new angle from which to look at the global circulation of knowledge, arguably the most important topic in recent historiographies of science. In the past two decades, historians of science have explored in detail how natural knowledge is produced in circulation, arguing that "science is a form of communicative action" born in transit. 33  so-called Age of Discoveries, with its emphasis on long-distance maritime travel, is often taken to spur forced, unequal, but nonetheless productive exchanges between different populations. From Beijing to Calcutta and Potosí, a new set of connected non-European locations emerged as crucial sites for the construction of knowledge. Much of this literature has provided a reinvigorating reorientation for traditional narratives of the emergence of modern science. It has highlighted how natural knowledge emerged in hierarchical exchanges where credit, authorship, and even freedom were not granted to all participants. Yet even this literature has not fully engaged with the materialities of transport that facilitate these exchanges. 35 This special issue claims that, if knowledge is produced in circulation, the material conditions of travel play a strong role in shaping it. Such an argument is inspired in part by the German tradition of media studies. As scholars such as Friedrich Kittler and Bernhard Siegert have argued, literature and culture are the aftereffects of developments and breakdowns in communication technologies. 36 In recent years, Siegert and others have expanded this approach by exploring how transportation networks condition cultural and artistic production. 37 Scholars in the maritime humanities such as Margaret Cohen and Steve Mentz have explored how oceanic travel often conditioned literature in the age of sail. As Mentz has argued, the shipwreck was the guiding literary metaphor of the early modern age, encompassing all the dangers and pitfalls of globalization while also exploiting the event's potential for creating narrative tension. 38 If Mentz is right, then early modern literature, from the Lusiads to Shakespeare's The Tempest and Robinson Crusoe, is a side-effect of repair techniques gone wrong, just as, in Siegert's account, modern literature is the side-effect of misfiring communications systems.
This special issue focuses, in contrast, on how natural knowledge is born when repair techniques do work. Transportation networks do not simply generate knowledge because they facilitate contact between different populations. Safe and secure transportation requires complex skills and knowledge of repair and maintenance, and much of the knowledge produced in long-distance networks focuses precisely on these practical concerns. Circulation produces the kinds of natural knowledge that facilitate further circulation, in other words. In recent years, scholars have already explored how much of the early modern mathematical, astronomical and cartographic sciences focused on solving the problems of navigation. Maps, astronomical instruments, and chronometers were all designed to ensure that ships reached their destination and found their way home. 39 Yet a parallel history that focuses on the major turning points of repair techniques, and their relationship to scientific knowledge, has not been written. The contributions to this issue are marking the first steps toward writing such a story.
For the age of sail, for instance, a focus on repair can complement the current, extensive, and excellent literature on European efforts at optimizing naval architecture with a more global collection of studies on the natural history of timber and the parasites that attack it. 40 As historians have begun to recognize, the shipworm teredo navalis was a major concern for early modern shipwrights, local experts, naval administrators, and natural philosophers because it bored into the hulls of ships and made them leak and sink. 41 In the 1660s, Henry Oldenburg and Robert Boyle discussed the possibility that Indian pear-wood, a foreign tree, could be used to build ships instead of English timber because it was resistant to shipworm. 42 The subsequent history of seventeenth-and eighteenth-century ship improvement presents a series of various failed or partially successful inventions to coat the hulls with protective paint or lead or copper sheathing, at enormous costs and expenses. As Frank James has documented, even nineteenth-century scientists were deeply involved in attempting to solve this problem. Humphry Davy turned his scientific prowess to practical purposes when he proposed adding zinc to the copper sheathing of ships, with catastrophic unexpected consequences that cost the British Navy more than a fortune before the project was abandoned. 43 From the perspective of repair, moreover, the coming of the railway was not a radical departure from the age of sail. In India and across much of South and Southeast Asia, white ants and other parasites ravaged any type of transport infrastructure that relied on wood, be it the sleepers of train tracks or the planks of a ship. 44 Technologies of transportation depended on each other to work, so that in China the first airplanes to make transnational journeys relied on gasoline that had been transported inland via the classic Silk Road transportation technology: camels. And a key phase in the early era of manned flight was the construction of seaplanes or "flying boats," which did not need runways to land or take off, adapting technologies of oceangoing navigation to the air.
These previous examples from a long and unwritten history of repair, together with the stories told in this special issue, emphasize how transportation technologies rely on using local resources to solve breakdowns in local conditions. They hew closely to the vast historiography on how technologies are transformed when they are adapted to new locales. It is probably not an accident that Rudolf Mrazek's Engineers of a Happy Land, one of the groundbreaking works that explored how technology becomes transformed when moved across the globe, focused primarily on transportation when examining how Western technologies were adopted in Indonesia by different actors for widely differing purposes. 45 As things break down and are repaired, they slowly become different. A historiography of repair is also always a historiography of invention and innovation elsewhere.
Such an approach to transportation also problematizes historical claims about the key role of transportation techniques in the emergence of modernity. A variety of scholars such as Daniel Headrick, John Urry, John Law, and Bruno Latour have made powerful arguments about mobility and the establishment of long-distance networks as the key issue in the emergence of modernity. 46 Much of this scholarship tends to consider transportation as a story of innovations enacted in Europe and North America, and to make the assumption that the development of modernity is a story of how European ships, trains, airplanes, and cars could conquer the globe without undergoing adaptation to local circumstances. Even actor network theory, which is highly attentive to the problems of translation and adaptation, appears to offer a similar narrative when it comes to explaining the emergence of modern Western power or the circulation of vessels and information. The sixteenth-century Portuguese navy was one of the major case studies for John Law's theories of controlling networks, which he developed in tandem with Bruno Latour's concept of immutable mobiles. 47 Law argued that the success of Portuguese colonizers relied on their ability to keep their sailors and ships identical, faithful, and obedient across long distances, just as Latour's inscriptions maintained their identity throughout their travels. Yet recent historians have presented the story of Portuguese colonization in drastically different colors. 48 By the end of the sixteenth century, neither the ships nor the crew of the Portuguese were truly Portuguese. A focus on repair reveals that, in order to travel, survive, and succeed, things need to maintain their flexibility and mutability. To paraphrase Anna Tsing, whose work on globalization is key for the arguments of these essays, transportation is made possible by the productive work of friction. 49 Every vehicle is a ship of Theseus. If modernity is about global networks of circulation (and if there is such a thing as modernity), it is a shape-shifting modernity that shows a different face at every site of repair. At each locale, a new kind of knowledge relies on the expertise of different craftsmen and the availability of different natural resources to keep global circulation in perpetual motion.
The challenges that transportation technologies pose to actor network theory, and to theories of modernity, resonate with the arguments of the emerging field of critical infrastructure studies. As Cymene Howe and others have pointed out, infrastructures are by nature paradoxical and liable to escape any strict designations. They are both material, bearing the traits of large technological systems, and philosophical, laden with semiotic and aesthetic meaning: in the words of Brian Larkin, they are "objects that create the grounds on which other objects operate." 50 They deteriorate despite the fact that their purpose is typically to reproduce structures across time and space; they are fragile and require retrofitting, even as they are designed to withstand the test of time; and they mitigate risk but also generate it, often because of their anthropocentric design. 51  instance, not only a particular airplane, but also the airports, fields, runways, hangars, and factories in which airplanes are loaded and unloaded, repaired, and retrofitted, as well as the licensing regulations for pilots, air traffic control protocols, and regulatory laws applied by authorities. 52 Infrastructures of transportation, and the provisions they made for maintenance and repair, articulated and embodied the local forms of knowledge that globalization yielded. They also often provided the means by which transportation technologies changed local environments, labor markets, and cultures. 53

Conclusion
The contributions to this issue provide the first steps toward a long-term history of repair over the past millennium. They epitomize the complex political, social, and technological expertise that repair en route requires, and how these encounters have shaped the development of natural knowledge and modern science along the way. As Pepijn Brandon and Marten Dondorp show, shipbuilding and ship repair were large-scale industries that needed complex natural resources and large groups of skilled workers. 54 As a result, the establishment of complex repair facilities across the globe, and the transfer of shipbuilding technologies from one site to another, were highly difficult enterprises that tended to fail. Building and repairing ships in a cost-efficient manner, accessing the right resources at the right prices, and employing suitable skills of labor management required muchneeded expertise in management. Yet, despite the difficulties of transporting repair knowledge, ship repair and shipbuilding did happen across the globe. In her contribution, Bronwen Everill offers a case study of the establishment of a cluster of successful repair and building facilities in Freetown, Sierra Leone, in West Africa. She shows how Africans and Euro-Americans participated in the exchange of technologies related to shipbuilding and repair. Her story reveals how the presence of a local community of West African boat builders was essential for the success of establishing sites of repair for oceangoing ships built elsewhere; it also reveals the significance of canoe-building in coastal trade for Africans, Europeans, and Americans. 55 Her article therefore serves as a reminder that, as Euro-American sailors traveled across the globe, they did not only rely on the ships they brought with them; they also adopted and adapted local methods of transportation.
If Sierra Leone shows how to make repair infrastructures work, Sara Caputo's paper on voyages of discovery in the late eighteenth century British Empire reveals the limits of these infrastructures. 56 As Caputo shows in great detail, port cities across the world did not only provide the opportunity to fix ships, but they could also be sites of danger where inexperienced pilots could wreck a vessel and seamen could be poached. Even then, repair was a necessity. Caputo argues that the discovery ships of the British Navy, arguably the most powerful naval power in the eighteenth century, needed constant care and attention to travel at excruciatingly slow speeds.
Similarly, in Mary Brazelton's study of Republican China, the limitations of technical infrastructures were constant concerns for authors and administrators who sought to establish and promote aviation ventures. 57 Airplanes could not fly across China without the construction of airfields and airports, weather observation and reporting systems, or radio communications networks. Brazelton argues that in the 1920s and 1930s, the education of skilled mechanics, engineers, and other staffers therefore became a priority for a number of transnational aviation ventures, because technical infrastructures required capable labor corps -thereby suggesting that histories of transportation maintenance and repair cannot be easily disentangled from histories of the labor associated with these activities.
Examining transport technologies across a variety of local environments also reveals the frequent coexistence and competition of different technologies of transportation over long periods of time. A global history of travel does not offer a linear history of unitary modernity. In his contribution, Stefan Tetzlaff shows how the coexistence and competition of motor cars and bullock carts in twentieth-century India shaped networks of transportation there. 58 He suggests that those who advocated for and those who protested against the rise of motor transport in India collectively over-emphasized the transformative power of automobiles upon agrarian society. Tetzlaff presents a scenario in which the juxtaposition of old and new technologies, their designers, and their users reflected the competition and collaboration of various underlying ideas of modernity and technoscientific knowledge.
Taken together, the essays of this special issue reveal the immense amount of work that goes into transportation repair across the globe. Production and repair sites have long been unequally distributed across the globe, contributing to the production and reproduction of social and political inequalities. While the story of repair presents a history of science and technology that moves beyond the traditional confines of Europe and North America, it does not ignore the fact that practices of repair did not eliminate social and political hierarchies. Ships and places of repair were always sites of complex and mostly hierarchical labor organization, leading to the unequal distribution of profit and the unequal attribution of credit across the centuries. When repair was performed at a distance, it not only contributed to the production of hybrid types of knowledge, but it could also fuel the nationalist and racist tendencies of colonial powers. While repair and maintenance are often presented as ecological alternatives to an economy of waste, this essay has argued instead that fixing ships and other vehicles often had disastrous consequences for island and port environments in the early modern period. 59 And by the dawn of the twenty-first century, global transportation and its sites of maintenance have become not only the foundational building blocks of a connected and networked society based on exchanges of knowledge, but also one of the main causes of human-made climate change. Repair provides a means to consider the checkered and connected histories of globalization across the past several centuries, and to reconsider how the material culture of transportation matters for the rewriting of that history.