Science as Culture Vol. 14, No.

2, 139 160, June 2005

Cloning Computers: From Rights of Possession to Rights of Creation

IVAN DA COSTA MARQUES
Rio de Janeiro, Brazil

Introduction Questions of originality and about the border between nature and artice are linked to the practices of international property rights and lie in the contact zone between the West and its others. The notion of intellectual property is strongly connected with standard epistemological assumptions, especially those pertaining to the existence of natural stable borders and instant discoveries. The notion of natural stable borders allows for the existence of pure, i.e. completely dened or unproblematic objects or things. The concept of discovery suggests the detachment of a relative instant of time as the denite moment of recognition or creation of a thing (a stable form). The epistemological assumptions around notions of stable borders and discovery are associated with the idea of the primacy of the origin, which invokes the precedence, priority, predominance, preference, prerogative, privilege, right-of-way, seniority, supremacy of the original over the copy, of the model over the imitated. Primacy of the origin is mobilized to legitimate granting intellectual property rights to those who rst recognize or invent a thing. Accordingly, by invoking institutionalized intellectual property rights, agents can side-step claims of liberty or rights of creation through naturalized or invisible processes of domination. However, during the second half of the twentieth century, strong philosophical movements, especially within French philosophy exemplied by the work of Foucault, Deleuze, Guattari, and Derrida, have assembled powerful tools of deconstruction which challenge the primacy of notions of origin. Also, in the last decades, a stream of STS (science, technology, and society studies) scholars working in the West, such as Madeleine Akrich, Geoffrey Bowker, Michel Callon, Donna Haraway, Sheila Jasanoff, Bruno Latour, John Law, Donald MacKenzie, Emily Martin, Annemarie Mol, Arie Rip, Susan Leigh Star, Marilyn Strathern, Lucy Suchman, Sharon Traweek, and Helen Verran, among others, have been translating, adapting and using philosophical tools linked to these philosophical movements to show just how problematic borders and origins can be in

Correspondence Address: Ivan da Costa Marques, Ave. Atlantica 822 apt 402, Leme, Rio de Janeiro, RJ 22010000, Brazil. Email: imarques@ufrj.br 0950-5431 Print=1470-1189 Online=05=02013922 # 2005 Taylor & Francis Group Ltd DOI: 10.1080=09505430500110887

140

I. da Costa Marques

the world of scientic and technological artefacts. Much of this STS work has been identied with what has been called controversially actor network theory, also referred to by the acronym ANT (see Law and Hassard, 1999). The tension between construction and deconstruction of origins leads to the following questions: in the context of the increasing investments in the framing of intellectual property, can postcolonial deconstructions of the Western concepts of discovery and authorship be relied upon to challenge relations of dominance? Can those deconstructions be relied upon to create new spaces of possibility for technological innovation under the conditions of global inequality? The Brazilian development of a clone of the Apple Macintosh computer speaks to these questions in a variety of ways. Indeed, a consideration of the Brazilian reverse engineering of the Apple Macintosh may transform the eld of meanings in discussions of technological innovation. The Unitron Case During the 1970s Brazil institutionalized a special policy of market protection for local minicomputer manufacturers. In the 1980s, the Brazilian dictatorship extended this policy to microcomputer market regulation.1 Through these policies, segments of the domestic computer market were to be reserved for manufacturers who had development laboratories in Brazil and who had locally designed their computer models. Prospective computer manufacturers were required to present their local development projects, schedules, and budgets to the Brazilian government, and have them approved before they started operations. SEI (Secretaria Especial de Informatica) was the name of the Brazilian government agency in charge of the computer industry policy in the 1980s. A special court, called CONIN (Conselho Nacional de Informatica), was also instituted to hear appeals against SEIs decisions. The Macintosh computer, the most important and successful alternative to the IBM PC, was developed by the Apple Computer Company in the USA. In contrast to IBM, Apple tried to enforce and maintain strict control over the property rights of its technology and over the functional characteristics of its products. Indeed, Apple has a history of taking aggressive legal measures to prevent the copying of its computers.

Figure 1. Mac da periferia. Credit: James Colburn

Cloning Computers

141

This story of Unitron, a Sao Paulo-based company, begins with its development of a Macintosh clone in November 1985. The rst version of the product, labelled MAC 512, was nicknamed Mac of the periphery, in an allusion to its origin in a peripheral country. As the nickname aptly suggests, this Brazilian production of an Apple clone directly addresses the spaces of possibility for technological creation or innovation in conditions of contemporary globalization. These are the spaces which are increasingly regulated by intellectual property rights. There are two ways to produce clones: simply copying or, alternatively, reverse engineering the original model. Through reverse engineering, it is possible to replicate the functional features of a computer system without exactly copying it.2 The SEI report on Unitrons bid to develop its clone of the Macintosh concluded that within technical bounds the project complies with current legislation and we recommend its approval.3 Moreover, according to a separate SEI report of November 1987 the software part of the project should be approved as an indigenous development (category A).4 As of November 1987, according to the same SEI report, Apple had not led any patent relative to the Macintosh in Brazil, and would not be able to do so in virtue of the time elapsed since the launching of the Macintosh in the market. After recommending approval of Unitrons request to manufacture a clone of the Macintosh, the same report concluded that Unitrons design would violate Apples rights in those countries in which Apple had led patents, although it did not violate Apples rights in Brazil. In the meantime, political and commercial pressure from the United States had built up. Unitron had made the external case of its clone an exact copy of Apples. Referring to this feature of the cloned computer, Apple publicized that Unitron was a pirate enterprise by placing two apparently identical machines side by side in a lobby on Capital Hill in Washington: one was the original made by Apple and the other was a Unitron clone, the latter adorned with a pirate ag.5 The US government became more directly involved in the dispute in retaliation for the Brazilian decisions and it threatened to impose commercial barriers on exports from Brazilian rms to the US. On 18 December 1987, under heavy political and commercial pressure from the United States, a new law regulating the software industry was passed in Brazil.6 On 22 January 1988, an addendum to the report on the software part of Unitrons project was drawn up referring to this new juridical statute.7 According to this addendum, approval of Unitrons project should be dependent on presentation of further information by Unitron, and possibly further developments should be required. From this point on, Unitron faced difculties: on 27 January 1988 Unitron formally requested that SEI approve its project, asserting that all formally requested information had been presented. On 15 March 1988 Unitron took legal action against SEI claiming the right to have its project immediately approved. On 21 March 1988 SEI denied Unitrons project approval based on Unitron having started commercialisation of the product prior to nal approval.8 But in spite of and possibly in contradiction with SEI expectations, Unitron did not abandon its initiative. On 29 March 1988 Unitron led for a new project to manufacture a clone of a new Macintosh model, designated Unitron 1024. As of August 1988, Unitron had changed both the external case and the internal characteristics in the new model of its computer. After further interactions with governmental institutions, universities, and National Semiconductors, Unitron claimed to have completed the design of a clone of the Macintosh using reverse-engineering techniques.

142

I. da Costa Marques

Nevertheless, on 1 August 1988, SEI denied Unitron approval for the new model on grounds of technical deciencies.9 On 10 August 1988, Unitron appealed to CONIN to have the SEI decision reviewed. In its appeal, Unitron claimed that its model 1024 could be legitimately approved in Brazil or in any other country, since it is the result of an inestimable work of reverse engineering of the original American machine.10 Within a few months Unitron claimed to have made a new cloned machine avoiding violations of the new juridical statute. Unitron asserted its legal right before CONIN, against the SEI decision, to manufacture a clone of the Macintosh computer in Brazil. Nevertheless, on 19 December 1988, CONIN turned down Unitrons project in a vote of 8 to 7. It is noteworthy, however, that representation on CONIN was heavily weighted towards government interests. CONIN was composed of eight ministers of the federal government and eight independent representatives of civil society.11 All seven independent representatives who were present voted for approval of Unitrons project.12 All ministers voted against approval with the exception of the Minister of Aeronautics, who abstained.13 Geraldo Azevedo Antunes, the main Unitron shareholder, declared he would ght CONINs decision in the courts,14 but he never did. Unitron closed. Fiction or Fact? Who Chooses? Who Decides? The Unitron story related above suggests that, at least in some cases, once a technological frame of referencesuch as that of the Macintoshhas been circumscribed and xed, replicating the features of the relevant technology is feasible, even for products classied as high technology such as computers. Inside a laboratory, it is possible to clone a given computer at relatively low costs by replicating its functional characteristics within a welldened frame. But given the outcome of the Unitron case, can we still claim that cloning foreign technologies (in Sao Paulo, Brazil) is feasible? Is it relatively easy to copy or reverse engineer without violation of rights? If so, why are successful instances of such reverse engineering apparently so elusive? Legal Reverse Engineering is Relatively Easy I shall begin by making the argument that the reverse engineering of a high-tech product in a given legal frame is indeed relatively easy, even though Unitron lost its appeal by 8 to 7 votes. I shall go beyond this to suggest this is due not simply to technical issues, but also, and inseparably, for legal reasons. In his admirable study of what was involved in establishing and securing the patents of Schlumberger, Bowker contends that all a rm needs in such cases is a respectful enough account (Bowker, 1994, p. 124). I will make a threefold argument showing that Unitrons case provides evidence that a small rm with very limited resources in a peripheral country was able to produce a respectful enough account of its deeds to go to trial with. The three strands of my argument are outlined in the following sections. (A) A nuanced if bitter discussion proceeded from August to December 1988, when CONIN came to a verdict about Unitrons claim to clone the Macintosh computer. In 1988, the computer industry policy was already under domestic criticism for alleged obsolete, high priced products, and for not having developed Brazilian rms that would be able to export in the successful manner associated with Korean companies. Unitrons project, disentangled from its juridical appeals, might plausibly have turned into a golden

Cloning Computers

143

opportunity for export. However, important private rms in the Brazilian PC market feared eventual competition from Unitron, the lone company on the Macintosh market. Taking a practical approach, they did not want to witness a (possible) striking success for Unitron. So, in private, these rms did not hesitate to accuse Unitron of immoral behaviour. Inuenced by these larger local rms, and certainly eager to compromise over US demands, high ranking government ofcials of SEI argued that Unitrons problem was a moral and not [a] legal question, and that therefore the executive branch should act to avoid further international embarrassment. In other words, Unitrons account was respectful enough to bar larger Brazilian private capital investors from standing up in public against it. They had to work from inside the business/government network to exert inuence and make it easier for the pressure from the United States to become effective. (B) Mathias Machline was the owner of several rms, including SID, a computer manufacturer that was among the ve largest in the country. An SEI report in 1987 mentions that Richard Herson was a consultant working for Apple.15 In December 1988, Geraldo Azevedo Antunes, the founder of Unitron, told newspapers that Herson had selected Mathias Machline as the prospective Apple partner in Brazil, noting that Herson had been nominated to be vice-president of Mathias Machlines holding company.16 Hence, Machline would have had something to lose if a third group (Unitron, for instance) made a successful entry into the microcomputer market with a clone of the Macintosh. Machline was also well-known as a friend of the president of Brazil, Jose Sarney. On the eve of the CONIN meeting, the Minister of Science and Technology and CONINs head, said that the Unitron 1024 machine is substantially different from the Macintosh . . . all will depend on instructions to be received from Mr. Sarney.17 In other words, the Brazilian government was not much concerned with a balanced evaluation of the respectfulness of the stories of the contenders, but rather with the satisfaction of discretely manifested specic related interests. (C) The independent representatives in CONIN thought that Unitrons claims should be settled in the judiciary courts and not decided in CONIN, a court of appeal of the executive branch of government. This was a sign that Unitron had produced a respectful enough account of its deeds to go to trial with. An important gure among these representatives, Claudio Mammana, the president of ABICOMP (Brazilian Computer Industry Association), shared this view. He had had the opportunity to express ABICOMPs ofcial opinion in a meeting with the EIA (American Electronics Industry Association) in the United States. On this occasion, Apples representative admitted that his company had not led patents for the Macintosh in Brazil.18 On the day after CONINs decision, Claudio Mammana declared that: although the question of United States pressure had not been addressed directly by the government at the [CONIN] meeting, the correlation between Section 301 of [the US] Trade Act and the Unitron issue is too close to be overlooked.19 But ABICOMPs president was not a businessman. As a professor of the University of Sao Paulo, he was a qualied employee who was respected for his intellectual skills, but representatives of the local capitals associated in ABICOMP considered his views to be theoretical. Pressed by representatives of local capital, personal friends and the US government, and advised to do so by high-ranking SEI ofcials, in all likelihood Sarney did inuence the votes of the ministers, who, with one abstention (the Minister of Aeronautics), voted

144

I. da Costa Marques

consistently against Unitron. The unanimous vote of the independent representatives present in the court for Unitron suggested that the legal account was creditable. Hence, a respectful enough account was presented by Unitron. It appeared that reverse engineering had been achieved and that copyright laws had not been violated. Reverse Engineering was Barred from Outside of the Laboratory and the Courtroom But if it is easy to copy or reverse engineer without violation of rights as I claim, then why was success elusive for Unitron? The Unitron case suggests that the answer exceeds both the technical and the legal framing on which this claim rests. The independent representatives in the court thought that the dispute should be settled in the fully independent courts instead of in CONIN which was under the inuence of the executive. Furthermore, although it is difcult to prove, it is possible that Apple may not have been condent that the Brazilian (or even non-Brazilian) courts would nd in its favour, since it had not registered its patents in Brazil. In any case, the fact is that Apple and Unitron never faced each other in court. Apple never sued Unitron, instead it hired lobbyists. This shifted the dispute between Apple and Unitron to outside of the courtroom. In his history of Schlumberger, Bowker has suggested that in the world of corporate disputes over ownership of the origin, strategies for imposing patents outside of the courtroom involve all kinds of different uses for them (and sometimes workarounds for being able to do without them) (Bowker, 1994, p. 113). Inside of the courtroom only the veracity/respectability of history matters. The work of making a given story count as history is hidden. Outside of the courtroom, where the historical account is the central focus, questions concerning the legitimacy of a given historical account become irrelevant background noise to the real focus of company activity (Bowker, 1994, p. 113). My contention is that in a matter of a few months Unitron was able to make the cloning of Apples products legal. In the laboratory it was able to re-organize the relevant heterogeneous materials to separate unacceptable copying from acceptable reverse engineering, and through the extension of the laboratory to include the law ofce, it also made a respectful enough account of its actions. Technical framing intertwined with legal framing. But this was not enough. Techno-legal framing overowed into a political and an economic space in forms such as conict with the USA, competitors in the PC market, friends of the president, moral but not legal questions. It was this overow that nally allowed Apple to bar Unitronfrom outside of the laboratory and the courtroom. Attendant Translations The notion of translation is seminal to the stream of STS studies mentioned in the introduction to this article. Indeed, the eld may be referred to as the sociology of translation (Callon, 1980). Latour calls translation the interpretation given by the fact-builders of their interests and that of the people they enrol (Latour, 1987, p. 108).20 Opening the Black Boxes of Foreign Technology During the 1970s a proposal to develop a computer technology in Brazil deected the traditional process of industrialization by import substitution that had been prevalent in

Cloning Computers

145

Brazil (Tavares, 1972). In traditional import substitution policies, local industrial activities are sought and successfully achieved using the factory, that is, the strict assemblage of products (and services). Innovation strategies revolving around import substitution policies did not encompass local development and design of products and processes. The origin of the technology was not an issue. Most import substitution policies, such as that implemented for the automobile industry in the 1950s in Brazil, took for granted that the technology would be imported and did not address the conditions under which import would occur. In contrast with traditional import substitution policies, the minicomputer industry policy adopted during the 1970s, which in part was conducive to Unitrons performance and other related initiatives, had its focus on the local design and development of minicomputer systems. The innovation strategy associated with the minicomputer industry policy of the 1970s was partially grounded on the idea of granting the high status of scientic research to the local activity of discovering the workings of foreign technologies and learning how to reproduce them. The relative originality of this activity was recognized by Brazilian computer professionals in the 1970s who claimed that discovering the works of foreign technology should be considered legitimate work of original research within a nation that did not know how to make the technological artefacts it uses.21 The effect of such attribution of status and meaning, a translation which turned commercially available products into legitimate objects of scientic research, exceeded the connes of the departments of linguistic and philosophy of the universities to become an important factor in the nancing of research in engineering departments in Brazil. It was in the 1970s that governmental nancial agencies started to provide funds for graduate schools of engineering to undertake the reverse engineering of products of foreign technology already available on the market. The decisions to allocate public money to open the technological black boxes that were commercially available came together with a rhetoric which, in the 1970s, granted the high status of academic scientic researchers to those who were researching such foreign technological black-boxes in universities. Traditionally such high status had been reserved for those who pursued the socalled disinterested pursuit of knowledge, or for research oriented towards knowledge for its own sake or concerned exclusively with the so-called advancement of human knowledge about nature.22 From Factory to Nation in the Division of Labor In the 1970s Brazilian computer professionals made use of a highlighted distinction that was embedded in the dominant discourse of the industrial engineering discipline, namely, the distinction between the activities of industrial assemblage of goods (and services) and what would more properly be called production (conception, design, fabrication, circulation and use) of goods (and services). They used this distinction to emphasize that, with the help of specic techniques in the organization of production, modern industrial engineering makes it possible for a specially guided group of people to perform the assemblage of industrially produced goods without expanding the context of their activities and without actually knowing (discovering) how these goods are made and created. In their journals and conferences of that period, computer professionals pointed out that IBM and Burroughs were assembling the latest generation of computers in Brazil with nearly no involvement of Brazilian engineering.

146

I. da Costa Marques

Although it was starting to lose its paradigmatic power, the Taylorist Fordist separation of tasks of conception from tasks of execution in industrial engineering was then (the early 1970s) still a dominant principle in labour and production organization. Brazilian computer professionals created new meanings and effects around this separation, translating it to the sphere and scale of the international division of labour. Brazilian computer professionals in the 1970s claimed that the separation of conception from execution reied an international division of labour which allocated the execution (i.e. the task of assembling computers) to Brazilians, while reserving conception (i.e. the tasks of discovering, inventing, constructing, creating, and organizing computers) to the developed countries. In the 1970s in Brazil computer professionals argued that development of local scientic and technological knowledgethe tasks of discovering, inventing, constructing, creating, and organizing computer systemswould produce better effects than imported knowledge. For a time, this vision of locating and localizing technological knowledge became most prominent among Brazilian computer professionals who rendered problematic the dominant meanings of efciency and not re-inventing the wheel which granted legitimacy to the adoption of ready imported technology.23 Thus computer professionals mobilized nationalist sentiments and partially deconstructed the universalism associated with sciences and technologies imported from the developed world. Lack of Science and TechnologyComparative Economic Disadvantage In the 1970s Brazilian computer professionals attached new specic meanings to the previously highlighted asymmetry of the international division of labour, arguing that: (1) the lack of scientic technological knowledge of how products were conceived and designed situated Brazil on the side of execution in the international division of labour; and (2) engagement on the side of execution resulted in comparative economic disadvantage. This double construction of meaning translated lack of scientic technological knowledge as the cause of the economic disadvantage in the context of the international division of labour, which was also translated as the cause of poverty. This double (indeed more complex) translation was illustrated by a short story repeatedly circulated and discussed among Brazilian computer professionals in the 1970s. The story of the Indian tells of a tribe who, dependent on hunting buffaloes to eat, had adopted the rie as a hunting tool, fascinated at rst by the prospect of gaining the free time that the rie seemed to offer (since free time was associated with the promise of the efciency of modern technology). But having adopted the rie for more than one generation, the Indians had forgotten the techniques of the bows and arrows. Moreover, they had taken no steps to discover how ries and the bullets were made (since re-inventing the wheel, they were told, would not be a good investment of their resources). Having convinced the Indians, the white man could then increase the price of the rie, demanding in exchange for it more and more products from the tribe. In fact, the Indians were assembling the ries and even exporting them to the white man to pay for the rights to the methods of fabrication (conception and design) of the rie. Hundreds of Indians had to work many months to barter and trade what they produced for something a few white men took much less time to produce. In short, this story enacted an economic explanation for the tribes trajectory from abundance to misery. Such an account, like every statement within scientic discourse,

Cloning Computers

147

would be true or false depending on what the others did with it (Latour, 1987). Yet, more important than its veracity or falseness, were its truth effects. One such effect was that it helped to elevate the status of the activity of researching into foreign computers that were already commercially available but which Brazilian engineers did not know how to make (conceive and design). The higher status of such activity helped to increase the funds for technological research in the computer engineering laboratories and graduate schools in Brazilian universities. A stock of prototypes required a channel to ow to the market. Computer professionals at universities, state owned laboratories and military sites expected that rms would transform prototypes produced in the laboratories into marketable industrial products. But there were no Brazilian rms or foreign computer manufacturers that would run the risk of adopting locally developed prototypes to complete their path to the market. That is, they were unwilling to take the risk of completing the construction of locally developed prototypes as technological products. An alliance with the state was forged by computer professionals to catalyze and attract entrepreneurs who would be willing to develop these technologies if they were protected in the market from other entrepreneurs in possession of foreign technologies. This was the basis for the industrial policy proposal made around 1975, which in 1977 resulted in a minicomputer market protection policy adopted by a federal government agency (CAPRE24), conceived and sketched out by a community of professionals that was not composed primarily of businessmen, but rather of university professors and managers of state-owned data processing bureaux (Dantas, 1988; Marques, 2003). Translating into New Spaces Unitrons case suggests new ways for the colonized to look at artefacts presented by the colonizers. Here, rather than using these terms in their usual historically specic meanings, I associate the words colonizer and colonized with an asymmetry in the processes of construction of the so-called modern global world. In general terms, this asymmetry allows for the (always provisional) identication of those who are seen as setting the pace in the construction of the modern world as opposed to those who are perceived as more willing or compelled to follow its rhythm, those who are taken to be active as opposed to those who are taken to be passive, those who are taken to be dominant as opposed to those who are taken to be dominated. In the way I perform it here the tension colonizer colonized may be identied with many contemporary divides such as the North South, the European non European, the division between white people and people of colour, or even the gender divide. In relation to modern science and technology, Portugal and Spainthe Iberian historical colonizers of Latin Americahave only initially set the tone of the construction of the modern world (see Law, 1986) and today can be as deeply entangled in the colonized space as their former colonies. The relativity of the tension colonizer colonized is easily visible in certain regions of Latin America, such as Sao Paulo. The Work of Division and the Division of Labour: Constructing and Deconstructing Boundaries Asymmetries and tensions in the construction of modern technoscientic knowledge emerge in the process of purication which has been well described in the ANT literature

148

I. da Costa Marques

in general, and by Bruno Latour in particular.25 Traditionally, laboratories have been perceived as spaces of purication, restricted places where puried objects (theories and machines) are constructed. However, over the last few decades, ANT scholars have highlighted that hybridization proliferates intensively in such settings. Indeed, hybridization has never ceased to accompany purication. If one takes the standpoint that even the colonizers have never been modern, then it becomes apparent that there are no pure beings, that all actions are mediated and that there are only provisional juxtapositions of heterogeneous materials (Latour, 1993). What is the effect of such conceptualization of the presence and permanence of hybridization upon the perception of laboratories? One effect is the following: the previously invisible work necessary to create stable borders and origins of objects (purity as opposed to hybridization) becomes visible, and it is clear that it happens outside of the laboratory. Alternatively, it becomes apparent that places previously assumed to be outside of the laboratory are inside of it, or are, in fact, part of it. In effect, the laboratory space expands into the spaces of many other institutions, incorporating Nature Society in its entirety. Through these insights, laboratories and courtrooms of the West emerge as places that concentrate resources to carry out the work of division, classication, and purication in the modern world. Laboratories and courtrooms create boundaries or frontiers between puried objects and keep them stable. Western laboratories and courtrooms play a crucial role in constructing and updating purication divisions sketched by the modern colonizer, the boundaries between: Nature and Society, things in themselves and men among themselves (Latour, 1993). Through cloning the colonized replicates the functional features of a puried object by looking at it with their own eyes, capitalizing, as Strathern says, on the fact that you can not tell by looking at a motor vehicle [or a computer] whose it is, or the power it mobilizes (Strathern, 1999, p. 158). The boundaries between Apple and Unitron machines could not be conclusively established once and for all as if there were stable differences. As in all contacts and contracts, frontiers between designs of machines of different manufacturers cannot be specied to their denitive and ultimate limits because these are moving borders. At any given moment, these are the effect of the permanent movement of patent ofces, referees, lawyers, lobbyists, courts, engineers, chemicals, boxes, electrons, semiconductors, governments, subcontractors, competitors and users. Breaching the boundary between nature (the technical, machines in themselves) and society (the political, rms, governments, human collectives among themselves), Unitrons story suggests the making of new institutional spaces of science and technology and justice and law. It would appear that Apple had not protected its rights in Brazil and Unitron had not broken any Brazilian law with its clone of the Macintosh. The Brazilian government was compelled to approve Unitrons project. But this was unacceptable for Apple and for the US government. By the end of 1987, amid the disputes and controversies about the Brazilian computer industry policy, something had to give. The software law in Brazil seems to have been the weakest link in the network. However, the new legal order, that is, the new legal statute or framing provided by the new 1987 Software Law, did not push the cost of cloning the Macintosh computers high enough to make Unitron give up its initiative. Unitron had re-evaluated the situation and decided not to stop. Instead, it decided to abandon the 512 model and to add a law ofce to its laboratory. In this extended laboratory, Unitron had started to study/clone the Mac

Cloning Computers

149

1024 (Apples next model), when its 512 model was turned down. The costs of cloning undoubtedly had gone up for Unitron, who had to pay for new rounds of interaction with government, universities and other contractors. However, the extended laboratory allowed Unitron to appear condent before a court of appeal. As Bowker has argued in the context of Schlumberger ghting for its patents, Unitrons account was respectable enough to go to trial with, and that was all that was needed in the courtroom (Bowker, 1994, p. 124; emphasis in the original). In a few months Unitrons new extended laboratory had remade the accounts: watching attentively the new framing, carefully deciding what should be done based on the costs of development and reverse engineering. Laboratory-factory Law-ofce-courtroom Unitrons ability to perform amid this permanent movement suggests the creation of new spaces of fact and machine construction in the colonized world. These are new spaces created by the expansion of laboratories and law ofces to include each other in a new post-colonial world order: lab-factories attached to law-ofce-courtrooms. New translations for a new post-colonial world order suggested by Unitrons performance involve the proposal that the colonized might see artefacts constructed in the extended modern laboratories of the colonizers as if they were natural objects. This means that initially the colonized would research these natural objects as such, and in so doing integrate them into a previously unknown Nature Society whole. In this domain of the unknown, Nature and Society are not yet demarcated, but instead their boundaries are constantly being provisionally established according to changing and contested criteria of legitimacy. More specically, in this new space suggested for the colonized by Unitrons performance, hybridization is made explicit. The rst move towards the creation of this new space in the colonized world is the explicit deconstruction and crossing of the borders between factories, laboratories, law ofces and courtrooms. A laboratory law-ofce-courtroom is created. Developments of purication by the colonized in their laboratories become explicitly entangled with and inseparable from developments of purication by the colonized in their law ofces. Decisions about what and how to purify are taken with explicit concern for those forms ` that enhance the colonized vis-a-vis the colonizers: that is, decisions that better equip situated agents or interests who or which oppose the primacy of the origin. In this new space, the work of division, that is, the construction and stabilization of origins and boundaries that dene objects and subjects, would be made explicit taking into account the local conditions of the colonized. Every technoscientic elementa scientic fact, a technological artefact, a specialized profession, an academic discipline, a modern citizenis part of an open network and extends itself over an intricate mesh of present and absent patents, several colonizers technological border control policies, import export balances, job opportunities, labour practices, and other changes. The extended laboratory law-ofce space recognizes that all thisthis heterogeneous networkconstitutes the technoscientic element and should be part of its assessment when it is considered for incorporation into a machine or an agreement in the colonized world. The move to establish the extended laboratory law-ofce space is nothing more than the late recognition by the colonized of how technoscientic elements are really made or, in fact, constructed. While the always provisional form of a technoscientic element

150

I. da Costa Marques

may be incorporated into the colonized world, its origins and borders which have been constructed by the colonizers may be redened, even radically redened.26 It may even be, if an optimistic trend is conceded, that such late recognition can redress some imbalances in colonizer colonized relations and contribute to alleviate violence by transforming them.27 Fidelity of Unexpected Allies Bruno Latour uses the word actant to refer to technoscientic elements. I am focussing specially on technoscientic artefacts as the unexpected allies [of fact builders or colonizers]28 in the construction of the modern world. Latour (1987) describes how these actants come to the aid of the colonizers in the work of division, helping them constantly to reconstruct and keep stable the borders separating nature from society in their version of reality. However, the delity of these actants is not natural or guaranteed. The Unitron case shows that the delities of technoscientic artefacts are not spontaneously maintainedtechnoscientic artefacts are more or less easily diverted on the frontiers. The unexpected allies of the colonizers are prone to be unveiled, modied and cloned. The delity of these actants is maintained and imposed through the work of division, the construction of specic frontiers and links. Movements that alter links and cross frontiers may be regarded as unacceptable copying and rejected, or, alternatively, they may be seen as unavoidable (and socially benecial) reverse engineering. Callon (1998, p. 18) contends that the only absolute guarantee to prevent the possibility of reverse engineering is to abdicate commerce.29 Since the possibility of reverse engineering is inseparable from the exchange of goods and services, its possibility must be accepted in negotiations concerning the work of division between colonizers and colonized if economic transactions between them are to continue. One important dimension crucial to the work of division is the naturalization of the mechanisms of attribution of authorship and of the intellectual property rights that result from them. This analysis points to concerns with investments in intellectual property framing and the mobilization of studies of science technology society as resources with which to ght relations of colonizer colonized dominance. Ambivalence and Impasse in Facing Modernity The ambivalence around simultaneously copying and rejecting the models they imitate often brings those located in the contact zone between colonized and colonizers (such as Brazilian computer professionals) into a kind of impasse. They simultaneously imitate and are hostile to the models they imitate. They copy to the extent that they accept the standards diffused by modernity. The Mackintosh cloners are involved in at least two forms of rejection that affect non-Europeans when they approach modernity. Chatterjee describes these rejections and observes that they are: both of them ambivalent: rejection of the alien intruder and dominator who is nevertheless to be imitated and surpassed by his own standards; and rejection of the ancestral ways which are seen as obstacles to progress and yet also cherished as marks of identity (Chatterjee, 1993, p. 2).30

Cloning Computers

151

However, although such rejections were certainly in play, Schwarz argues that since the nineteenth century those who he calls the educated people of Brazil31 feel that they are living among ideas and institutions that are copied from abroad and which do not reect local reality (Schwarz, 1987, pp. 38 39). He indicates that, for this reason, they feel inadequate. As members of the social category educated people of Brazil, living between colonizers and colonized, Brazilian computer professionals in the 1970s experienced this inadequacy in various ways. Sometimes they submitted to it. However, they also created prototypes, the design of which reected the inadequacy of foreign products for local conditions.32 In one way or another, then, Brazilian computer professionals translated inadequacy into technological dependence, and they tried to experiment with it. In this respect, the position of Brazilian computer professionals of the 1970s resonated with the tradition of a well known cultural movement in Brazil, the modernist Oswald de Andrades anthropophagous Pau-Brazil programme. According to Schwartz the: anthropophagous Pau-Brazil programme sought to give a triumphalist interpretation of Brazils distance from modernity, with the disharmony between bourgeois models and the realities of rural patriarchy at its very heart. Its novelty lay in the fact that the lack of accord was seen as a source not of distress but of optimism, evidence of the countrys innocence and the possibility of an alternative, non-bourgeois historical development. This sui generis cult of progress was rounded out with a technological wager: Brazils innocence (the result of Christianization and bourgeoisication barely scraping the surface) plus technology equals utopia; modern material progress would make possible a direct leap from pre-bourgeois society to paradise (Schwarz, 1987, p. 37).33 Schwartz argues that Oswald de Andrades modernist programme, put forward in the 1920s, introduced a change of values and advocated cultural irreverence in place of subaltern obfuscation, using the metaphor of swallowing up the alien: a copy, to be sure, but with regenerating effect. A MAC of the periphery. Why should the appropriation of an original be respected? Why should an original be attributed more value than a copy? What would be a legitimate mode of challenging a boundary that establishes legal differences between original and copy? In Oswald de Andrades programme, Brazilian local primitivism would give a modern sense back to tired European culture, liberating it from Christian mortication and capitalist utilitarianism. Brazils experience would be a differentiated and utopian cornerstone in contemporary history. Brazilian modernism therefore brought about a profound change of values: for the rst time processes in Brazil were said to have something to offer the modern world. Schwartz claims that: [h]istorical distance allows one to see that the programmatic innocence of the Antropofagos, which allows them to ignore the malaise, does not prevent it from emerging anew. Tupi or not Tupi, that is the question!Oswalds famous saying, with its contradictory use of the English language to pursue the search for national identity, a classical line and a play on words, itself says a great deal about the nature of the impasse (Schwarz, 1987, p. 37).34

152

I. da Costa Marques

Although the Antropofagos may have been programmatically innocent, a non-European approach to the attributes of modernity (including the study of options in the construction of sciences and technologies) faced consistent opposition from colonizers. At least until the last decades of the twentieth century few things tended to provoke more opposition than attempts to denaturalize or de-neutralize scientic and technological knowledge and progress. This suspicion regarding any attempt to retard technological progress did not go unnoticed by Albert Hirschman in 1971 when he wrote his inuential A Bias for HopeEssays on Development and Latin America. He remarked that: otherwise rather enterprising United Nations experts who wrote the report on commodity trade and economic development [after the Second World War said]: We are strongly opposed to retarding technological progress for the sake of avoiding the pains of adjustment which inevitably attend progress. And they went on to advocate a father-knows-better attitude in case the industrial countries were to encourage the production of substitutes through subsidies: Industrial countries are not in the habit of following such a course unless there are weighty reasons (Hirschman, 1971, p. 67). Among colonizers a prospect even more distasteful than interfering in market pricing is interfering with supposed neutral technical progress! On the other hand, the Unitron case suggests that colonizers act differently if the prospects of winning contests with established standards and within established spaces are not favourable. In such cases, they do not hesitate to abandon their own standards and established spaces.35

Ontological Politics Unitrons case suggests that the colonized explore the ontological politics36 involved in the debate about property rights. The opportunity to perform ontological politics ows from the fact that the delity of technoscientic artefacts to the colonizers partially depends on the possibility of being able to attribute authorship to products of the intellect, and thus turn debate about property rights from rights of possession to rights of creation (Strathern, 1999, p. 161). According to Latour, unexpected and indispensable allies do not look like men or women, and, with the benet of hindsight, the colonized can verify that this is true for an iron axe, a hunting rie, a motor, a microbe, a medicament, a vaccine, a paved road, a telephone, or an economic theory of comparative advantage (Latour, 1987, p. 121). So now the question for a new post-colonial order is: how can the colonized better negotiate their encounter with a computer, a virus, a drug, a cell phone, a piece of diagnostic equipment, a scientic or an economic theory? How can these unexpected and indispensable allies of the colonizers be seen? I have claimed that Unitrons performance suggests that the colonized may see these contemporary artefacts constructed in the extended modern laboratories-factories law-ofce-courtrooms of the colonizers as natural objects. Colonized may research new modern objects by initially integrating them into a not-yet-demarcated Nature Society whole. In this process boundaries may be redened, even radically redened, bringing greater equality into the balance of resources between the colonizer and the colonized.

Cloning Computers In/visibility

153

When a friend of mine who is a professor in England rst went to Rio de Janeiro initially he saw a child begging by a car as a child who belonged in that car. Though he had no difculty in seeing the adult beggars straight away (they immediately t into the categories of beggars in the big cities of rich countries) he did not see that attractive and androgynous gure as a child wandering alone in the night, begging in a big city. It may be that for someone coming from the afuent North, the category street child is not strongly or immediately stabilized (Law and da Costa Marques, 2000). In the absence of stabilized categories things become invisible, they are not recognized and may be more easily repressed or denied. That child who belonged to no adult might be an entity that remained invisible, outside of the universe, an entity that did not exist.

Categories Construct the In/visible This little story illustrates that categories perform the visible and also the invisible. They shape, conform and conrm objects (and subjects as well) and categories also hide.37 In addition to making in/visibility possible,38 categories are like Pierre Bourdieus schemes of domination: when their form is sufciently stabilized, they acquire the opacity of things,39 rendering that which is behind them invisible. The struggle for power is the struggle for the imposition of categories of perception of the world. Categories of technoscience40 are constructed in the (extended) laboratories of the colonizers, and the latter try to ensure that the tests of strength (in controversies) are squeezed through laboratories and courtrooms because in these locations they are better equipped to stabilize their own categories. First, it is in laboratories and courtrooms41 that the asymmetry of scales of the colonized and the colonizers relative capacities to create meaning seems to be most overwhelming. The resources assembled in the colonizers (extended) laboratories are so great that they claim to dene reality. The colonized are left without power, that is, without resources to contest, to reopen the black boxes, to generate new objects, to dispute the [colonizers] authority (Latour, 1987, p. 93). Moreover, modern colonizers have learned to construct a rhetoric so powerful that those who insist on denying their categories are delegitimized in one way or anotherhistorically, they have been ridiculed, isolated and even considered mad (Latour, 1987, p. 59). If different ontologies create different entities that populate Nature Society, then one battle frontthe tests of strength of their categoriesin the contact zone between colonized and colonizers concerns in/visibility. Science and technology, or more precisely, technosciences are ontologies (systems of belief) that originate from the material work of purication that takes place in the colonizers extended laboratories courtrooms. To understand how they increase scales and accumulate knowledge seems vital for non-Europeans who approach modernity. One needs to understand how the attribution of meaning to categories such as Nature and Society work. This political urge pertains to categories that create, for instance, the opaque wall which is the truth effect of the claim that: modern Western sciences and technologies are the legitimate spokesmen of Nature. By using apparatuses of objectication from metropolitan centres of calculation and purication42 (laboratories and courtrooms), extended-science-and-technology professionals align heterogeneous interests and appropriate other agents labour, services, and

154

I. da Costa Marques

deference. In relation to science and technology, this is how subjection of the countries approaching modernity is realized and reproduced. The colonized need to ght an ontological battle in order to deconstruct this opaque wall and to construct a different scenario with new visibilities. Unitrons Case and Ontological Politics The Unitron case provides evidence to show that the meanings attributed to copy and authorship shape apparatuses of modern objectication and domination that acquire the opacity of things. Accordingly, a copy is regarded as secondary, dependent on an original, is considered worthless, etc. In this version of reality, the author would be the creator of the original, of the independent, of what is worth more, etc. This places a minus sign before the scientic and technological efforts of the colonized. But authorship pertains more to authority than to originality. There was a time when artefacts circulated with no mention of their creators, freely used and modied [in part] by other creators. There was a time when the authora modern category that would identify the original creatordid not exist, and therefore there was no need to respect him or her.43 In order to acquire legitimacy, the right to be paid for or otherwise dispose of their work, the creators of Unitrons Macintosh clone would have to assert their authority (and their authorship) over the artefact through controversies and tests of strength. They would have to achieve what Latour calls the secondary mechanism of attribution and its spaces of in/visibility.44 As indicated previously, some recent French philosophers have challenged notions of the primacy of the origin and the naturalized notion of copy. Moreover, a stream of STS scholars have shown that hierarchies of the sort original copy are provisional stabilizations. Furthermore, they have shown that the alignment of human and non-human interests that juxtapose materials to realize those provisional stabilizations may be researched and put to tests of strength according to the materiality of each case. Values can be researched, analysed, studied and explained in the same way or terms as facts. Colonized/national is opposed to colonizer/foreign and original to imitated, and these oppositions hide a great deal: parts of the alien in the autogenous and parts of the imitated in the original. Of course, understanding these oppositions is not the same as undoing them. There is a large gap between this promise of relief and its fullment: breaking the intellectual bewilderment of the colonized does not by itself undo the situation. It is clear that the colonizers unexpected allies (technoscientic innovations) do not become immediately dispensable once the colonizers prestige deriving from originality has been undermined. Situated Encounters The preceding analysis suggests that the Brazilian computer business entrepreneur and manufacturer of the 1970s was an imagined semiotic character. In terms of ontological politics, this Brazilian computer business entrepreneur and manufacturer was an interlocutor for a material discourse (texts and prototypes) created in a community of professionals. This community of professionals sought their interlocutor, searching for and actually constructing her/him together with a government agency, namely, CAPRE, which was in charge of implementing a computer industry policy. A new identity for a class of entrepreneurs started to be performed by businessmen who invested in the local

Cloning Computers

155

development of technology under locally organized market competition among local equals. By means of non-tariff import controls the Brazilian government through CAPRE was able to protect this new class of entrepreneurs, Brazilian minicomputer manufacturers, and demand from them, in return, investments in local conception and design of their products. This worked well, for within ve years the Brazilian minicomputer market was supplied by local rms with locally developed products (see Tigre, 1983; Schmitz and Cassiolato, 1992). Unitrons case can then be understood as a result of the situated encounter between two colonizer colonized discourses. One is the discourse of economic development (globalization, Enlightenment); the other concerns the generation of science and technology in Brazil (autonomy, Romanticism). The different translations of the story of the Indian led to concerns about the technological autonomy of the nation. Concern for the technological autonomy of the nation suggested that artefacts produced by foreign technologies should be studied as natural phenomenathat this study was legitimate scientic work. The push for technological autonomy of the nation materialized the idea implicit in the story of the Indian, that what is made by (is an artefact for) the white man ( foreign) may belong to Nature for the Indian ( Brazilian). That is, it may be non-Indian, or non-human for the Indian. Indeed, the slogan technology is imported magic is a Brazilian saying. The foreign computer, something that can be felt and sensed but not understood, integrated itself into Nature and, thus, initially became non-human. Foreign computers were thereby othered, becoming part of nature. Hence, their commercial entanglements could be initially unmade and they could be studied in the realm of things in themselves, separated from the realm of men among themselves. This move would also constitute some Brazilian scientists engineers lawyers as fact constructors whose work was directed initially towards foreign technological black-boxes as if they were natural phenomena. Moreover, Brazilian scientists engineers lawyers were constituting themselves as professional subjects in the same process that was constituting their objects. These scientists engineers lawyers were complex entities, Brazilian citizens, men and women, fathers and mothers, sons and daughters, lovers and beloved, with many other links which might or might not resist the tests of strength that would follow. The more these professional subjects, persons, complex entities or agencies, would be able to resist and endure the consequences of their move to treat foreign technological black-boxes initially as if they were natural phenomena, the sturdier would result the fact they were constructing. Conclusion: Inevitable Hybrid Embraces and Dangerous Dialogues Strathern shows how the hybrid embrace promoted by intellectual property rights entails new practices of purication, while the right to intellectual property seeks its own new separations of Nature from Society (Strathern, 1999, p. 161). In the 1970s, Brazilian computer professionals, in defending what I have depicted as the incorporation of the unknown in the form of foreign technology (that is, artefacts which are the intellectual property of an author) into the unknown in Nature (that is, non-appropriated natural phenomena), committed what seemed to be an unbearable transgression in the imperial eyes of the colonizers. But they were, in fact, retracing the division between Nature (which expanded to encompass new entities or objects) and Society (which contracted to lose, for example,

156

I. da Costa Marques

copyrights and patents), which is something that moderns are constantly doing (Latour, 1993). Thus, we may think about two Natures, one delineated by the colonized (computer professionals of Brazil) and the other by the colonizers (western international capital and technoscience): initially the former incorporates as natural objects the artefacts of foreign technology, while the latter denies that these are hybrid objects with complex boundaries. This is the case, even though the colonizers have succeeded in partially disentangling them to assign the corresponding rights to an author (creator). Neither of the two natures (nor their respective societies) refers to a denite entity given in a pre-discursive reality. In other words, the struggle for the delity of the allies of the colonizers may be displaced to new (more even and less violent?) spaces of negotiation by increasing the visibility of the ontological policing (surveillance and punishment) of the colonizer. Colonizers appeal to a transcendent Nature and repress other ontological possibilities in order to legitimate and impose specic social rules and hierarchies as universally legitimate behaviour. It is at this point that we can appreciate the ontological political character of the proposal that the colonized might initially see artefacts constructed in the extended modern laboratories (factory-laboratories attached to law-ofce-courtrooms) of the colonizers as natural objects and research into them by integrating them into an unknown Nature Society whole. In this unknown, Nature and Society are not yet demarcated. Moreover, some STS scholars may be learning how to move around on this ground, treating these heterogeneous phenomena even-handedly. But for this treatment to become effective it is clear that the colonized need to build their own extended counter-laboratories which will construct new ontologies, veritable new worlds that are not just nagging and inauthentic replications, colonies of Europe (Latour, 1987). This requires, too, that the colonized become equipped with semiotic apparatuses that can account for the forked tongues of the colonizers (moderns). Dialogue between colonizers and colonized is inevitable but it is also dangerous. This is because the material capacity of the apparatuses of production of meaning of the colonizers ([extended] centres of calculation in Bruno Latours terms) is incomparably greater than that of the colonized. An asymmetrical situation is easily established, where the colonizers categories stabilize and acquire Bourdieus opacity of things hiding what lies behind them. Initial integration of the unknown in foreign technology into an undifferentiated Nature Society seeks instead to maximize the possibilities of different hybrid embraces. This is the best hope at the start of a dangerous dialogue between colonizers and colonized, that is, the best initial step to overcome the illusion and the costs of a monologue in the imperative mode: the one-dimensional or rather one-natural vision of a civilization that ows from Europe to the rest of the world. Notes
1

I have discussed the details of the Brazilian computer industry policy and stressed the differences between developments in the 1970s and those emerging in the 1980s elsewhere. For texts in English see da Costa Marques (2002 or 2000). 2 According to the IEEEUSAs Intellectual Property Committee, [t]he term reverse engineering means the discovery by engineering techniques of the underlying ideas and principles that govern how a machine, computer program or other technological device works. Engineers use this information for many purposes, including making other products interoperate with the target product that is the subject of the reverse

Cloning Computers

157

engineering. Engineers also use this information for the purpose of designing competing products that are not substantially similar in expression, as well as to discover patentable subject matter and ideas not otherwise disclosed in the literature provided with the product by the originator. We further believe that lawful reading, analysis, or disassembly of machine language is a reverse engineering technique by which an engineer can reconstruct the ideas of a computer program. See http://www.ieeeusa.org/ forum/POSITIONS/reverse.html, accessed on 11 April 2004. 3 SEI report of process F-026398/85 approved on 20 November 1987, p. 5. 4 SEI report referent to process 07824/87-4 (Registro do Sistema Operacional do Microcomputador Mac512), dated 11 November 1987. 5 VEJA, Micro vetadoCONIN proibe a venda de computador Unitron, Sao Paulo: 28 December 1988, p. 42 (VEJA is the local equivalent to TIME magazine in Brazil. It sells over one million copies of its weekly edition all over the country with a high concentration of subscribers in the richer southern states). 6 Law no. 7646, known as software law. 7 Addendum to technical report of 11 November 1987, dated 22 January 1988, relative to process 07824/ 87-4. 8 In an interview with the author (Sao Paulo, 3 September 2001), Unitrons main shareholder and manager, Geraldo Azevedo Antunes, denied SEIs accusation that Unitron had sold 100 machines before approval of the project. According to him Unitron had only provided demonstration machines for special clients and had not made regular commercial deliveries. 9 Source: Appeal to CONIN on part of Unitron for re-evaluation of SEIs decision to revoke the project of manufacturing a clone of the Macintosh in Brazil, dated 10 August 1988, p. 11. 10 Ibid., p. 1. 11 In the case of a tied vote, the Minister of Science and Technology, president of CONIN, would have the nal word. 12 The representative of the Union of Data Processing Industry Employees, APPD (Associacao dos Prossionais de Processamento de Dados), missed the meeting. 13 Jornal do Commercio, 20 December 1988. 14 Jornal do Brasil, 20 December 1988: CONIN decide que Unitron nao pode fabricar micro. 15 SEIs report on process F-026398/85, dated 20 November 1987, p. 6. 16 Jornal do Commercio, 20 December 1988 and O Estado de Sao Paulo, 20 December 1988. 17 O GLOBO, 19 December 1988, p. 15: Conin decide hoje se libera Unitron. 18 Interview with Claudio Mammana, President of ABICOMP during 1987 88. Braslia: 3 September 1996. 19 O GLOBO, 20 December 1988, p. 24: CONIN veta fabricacao do Unitron. 20 An incomplete list of seminal papers about translation would include Callon (1986), Law (1986), and Star and Griesemer (1989). 21 Proceedings of conferences and seminars (SECOMU, SECOP, SUCESU) and journals (DADOS & IDEIAS) of the 1970s. See also Adler (1987), da Costa Marques (2000, 2002), and Evans (1995). 22 That is, human knowledge imagined as something unied and metaphorically made equivalent to a stock of knowledge of the human species, Man. 23 The vision of local frontiers of technological knowledge of the Brazilian computer professionals of the 1970s resonates with the later feminist STS notion of situated knowledges of the late 1980s. See Haraway (1991, pp. 183201). 24 Coordenacao das Atividades de Processamento Eletronico de Dados, a federal agency under the Minister of Planning. 25 Latour (1987, 1993). For a very relevant treatment of hybridization in economics, see Callon (1998). 26 Note that radical redenition of origins and borders is what the colonizers usually do with certain elements of knowledge of the colonized. One instance of such redenition is the handling of the healing properties of plants, possessed by certain members of tribal societies, such as the shaman. A redenition of origin (which requires huge scales) translates the shamans mere belief into the Western laboratorys scientic discovery entitled to intellectual property rights. 27 For an exploration of related hope and optimism see Latour (2003). 28 Latour (1987, pp. 124 128): Tying up with new unexpected allies. 29 About this important point see especially Introduction and An essay on framing and overowing in Callon (1998). 30 Anti-Americanism amid the preference for elements of the modern life requiring the individual home, car, telephone, computer, etc. and jeitinho brasileiro would be two examples of such ambivalent rejections

158

I. da Costa Marques

noted by Partha Chatterjee. In his prized study of nationalism Chatterjee takes issue with John Plamenatzs distinction between two types of nationalism (western and eastern) to clarify the premises the liberalrationalist dilemma in talking about nationalist thought (my emphasis). 31 Schwartz stresses that the concept does not mean a homage but refers to a social category. 32 In the data processing centre of the Brazilian internal revenue service, SERPRO, they developed a keyboard concentrator that was inexpensive and simple. However, designed for SERPROs needs, this technology advantageously substituted the imported data entry systems equipped with expensive video terminals that were used abroad. In the data processing centre of the Federal University of Rio de Janeiro (NCE/UFRJ) they developed a oating point electronic processing unit, a hardware piece that executed oating point arithmetic operations typical of scientic calculations much faster than software subroutines. This kind of hardware processor was not an option offered by IBM for the IBM 1130 computer. Since most of the workload of university IBM 1130 computers was scientic calculus, this development more than doubled the throughput of the machines, extending their use period and postponing the importation of new machines. In the Brazilian Navy, military security was made problematic by military computer professionals. They claimed Brazilian engineers did not have the technological capacity to maintain Ferranti computers on board the frigates recently acquired from England. Brazilian computer professionals translated cultural malaise into technological dependence and interacted with it by constructing local prototypes. 33 Schwartz (1987, p. 37) notes that Marx himself, in his famous letter of 1881 to Vera Zasulich, came up with a similar hypothesis that the Russian peasant commune would achieve socialism without a capitalist interregnum, thanks to the means made available by progress in the West. Similarly, albeit in a register combining jokes, provocation, philosophy of history and prophesy, the Pau-Brazil programme set itself the aim of leaping a whole stage. 34 Tupi (pronounced too-pee in Portuguese) was the name of the Indian nation that at the time of the arrival of the rst Portuguese lived in the territory that is today the southeast coast of Brazil. 35 In one of his most compelling paragraphs Latour (1993, p. 38) remarks that [b]y separating the relations of political power from the relations of [extended] scientic reasoning while continuing to shore up power with reason and reason with power, the moderns have always had two irons in the re. They have become invincible. . . . Native Americans were not mistaken when they accused the White [colonizers] of having forked tongues. 36 A politics about what there is in the world: see Law (1996, 2002), and Mol (1999). 37 To illustrate a correlative point in a simple way, Thomas Kuhn refers to experiments in which people had difculties recognizing playing cards the colour of which did not correspond with their suits (a black king of hearts, for example) (Brumer and Postman, 1949, pp. 206 223; Kuhn, 1962, pp. 6264). 38 [W]e seek to understand the role of invisibility . . . how . . . categories are made and kept invisible, and in some cases, we want to challenge the silences surrounding them (Bowker and Star, 2000, p. 5). 39 . . . domination no longer needs to be exerted in a direct, personal way when it is entailed in . . . social formations in which, mediated by objective, institutionalised mechanisms, such as those producing and guaranteeing the distribution of titles (titles of nobility, deeds of possession, academic degrees, etc.), relations of domination have the opacity and permanence of things and escape the grasp of individual consciousness and power (Bourdieu, 1977, pp. 183184). 40 Categories of technoscience are (n1)th . . . n . . . (n 1) order forms constructed in heterogeneous and dominating centers of calculation (Latour, 1987, p. 245). See also Latour (1999). 41 By the capacity to provide scientic or technical evidences or proofs. 42 Purication opposes hybridization (see Latour, 1993). 43 I thank Marcia J. Bossy for her comments on this point. 44 To follow these trials [of responsibility, where the winner takes all,] a distinction had to be made between the primary mechanism that enlists people, and the secondary mechanism that designates a few elements among the enlisted allies as the cause of the general movement (Latour, 1987, p. 174; also see p. 134).

References
Adler, E. (1987) The Power of IdeologyThe Quest for Technological Autonomy in Argentina and Brazil (Berkeley: University of California Press).

Cloning Computers

159

Bourdieu, P. (1972, 1977) Outline of a Theory of Practice (Cambridge: Cambridge University Press). Bowker, G. C. (1994) Science on the RunInformation Management and Industrial Geophysics at Schlumberger (Cambridge, MA: The MIT Press). Bowker, G. and Star, S. L. (2000) Sorting Things OutClassication and its Consequences (Cambridge, MA: The MIT Press). Brumer, J. S. and Postman, L. (1949) On the perception of incongruity: a paradigm, Journal of Personality, XVIII, pp. 206 223. Callon, M. (1980) Struggles and negotiations to dene what is problematic and what is not: the sociology of translation, in: K. D. Knorr, R. Krohn and R. D. Whitley (Eds) The Social Process of Scientic Investigation: Sociology of the Sciences Yearbook, vol. 4, pp. 197 219 (Dordrecht and Boston, MA: Reidel). Callon, M. (1986) Some elements of a sociology of translation: domestication of the scallops and the shermen of Saint Brieuc Bay, in: J. Law (Ed.) Power, Action and Belief: A New Sociology of Knowledge? Sociological Review Monograph, 32, pp. 196 233 (London: Routledge and Kegan Paul). Callon, M. (Ed.) (1998) The Laws of the Markets (London: Blackwell). Chatterjee, P. (1986, 1993) Nationalist Thought and the Colonial WorldA Derivative Discourse (Minneapolis: University of Minnesota Press). da Costa Marques, I. (2000) A new look at an old devil: the computer market reserve in Brazil, in: Proceedings of The Third Triple Helix International Conference, Rio de Janeiro. da Costa Marques, I. (2002) A new look at an old devil: the computer market reserve in Brazil, in: G. Szell and G. P. Cella (Eds) Labor, Work & Social SciencesResearch and Trade Unions (Frankfurt, Germany: Peter Lang International). da Costa Marques, I. (2003) Minicomputadores brasileiros nos anos 1970: uma reserva de mercado democratica em meio ao autoritarismo, Historia, Ciencias, SaudeManguinhos, 10(2), (MaioAgosto), pp. 657 681. Dantas, V. (1988) Guerrilha tecnologica: a verdadeira historia da Poltica Nacional de Informatica (Rio de Janeiro, Brazil: Livros Tecnicos e Cientcos Editora Ltda). Evans, P. (1995) Embedded AutonomyStates & Industrial Transformation (Princeton: Princeton University Press). Haraway, D. J. (1991) Simians, Cyborgs, and WomenThe Reinvention of Nature (New York: Routledge). Hirschman, A. O. (1971) A Bias for HopeEssays on Development and Latin America (New Haven and London: Yale University Press). Kuhn, T. (1962, 1970) The Structure of Scientic Revolutions (Chicago: The University of Chicago Press). Latour, B. (1987) Science in Action (Cambridge: Harvard University Press). Latour, B. (1993) We Have Never Been Modern (Cambridge: Harvard University Press). Latour, B. (1999) Pandoras HopeEssays on the Reality of Science Studies (Cambridge: Harvard University Press). Latour, B. (2003) The promises of constructivism, in: D. Ihde and E. Selinger (Eds) Chasing Technoscience (Bloomington: Indiana University Press). Law, J. (1986) On the methods of long distance control: vessels, navigation and the Portuguese route to India, in: J. Law (Ed.) Power, Action and Belief: A New Sociology of Knowledge? Sociological Review Monograph, 32, pp. 196233 (London: Routledge and Kegan Paul). Law, J. (1996) Organizing accountabilities: ontology and the mode of accounting, in: J. Mouritsen and R. Munro (Eds) Accountability: Power, Ethos, and the Technologies of Managing, pp. 283306 (London: International Thompson Business Press). Law, J. (2002) Aircraft StoriesDecentering the Object in Technoscience (Durham, NC: Duke University Press). Law, J. and da Costa Marques, I. (2000) Invisibilities, in: S. Pile and N. Thrift (Eds) City A Z, pp. 119121 (London: Routledge). Law, J. and Hassard, J. (1999) Actor Network Theory and After (Oxford: Blackwell Publishers/The Sociological Review). Mol, A. (1999) Ontological politics: a word and some questions, in: J. Law and J. Hassard (Eds) Actor Network Theory and After, pp. 74 89 (Oxford: Blackwell and the Sociological Review). Schmitz, H. and Cassiolato, J. (Eds) (1992) Hi-Tech for Industrial DevelopmentLessons from the Brazilian Experience in Electronics and Automation (London: Routledge). Schwarz, R. (1987) Nacional por subtracao, in: Que horas sao?Ensaios, pp. 2948 (Sao Paulo: Companhia das Letras). Also published as Schwarz, R. (1988) Brazilian culture: nationalism by elimination, New Left Review, 167 (JanuaryFebruary), pp. 7790 (translated by Linda Briggs).

160

I. da Costa Marques

Star, S. L. and Griesemer, J. R. (1989) Institutional ecology, translations and boundary objects: amateurs and professionals in Berkeleys Museum of Vertebrate Zoology. 1907 39, Social Studies of Science, 19(3), pp. 387 420. Strathern, M. (1999) Property, Substance and EffectAnthropological Essays on Persons and Things (London: the Athlone Press). Tavares, M. (1972) Da substituicao de importacoes ao capitalismo nanceiroensaios sobre Economia Brasileira (Rio de Janeiro: Zahar Editores). Tigre, P. (1983) Technology and Competition in the Brazilian Computer Industry (London: Frances Pinter).