1. Page references here are to the World's Classics Edition by Oxford University Press (Defoe 1972), based on the first issue. See also the Norton Critical Edition (Defoe 1975) for additional critical and contextual material. In best eighteenth-century fashion, the full title page of the work summarizes key elements: The Life and Strange Surprizing Adventures of Robinson Crusoe, of York, Mariner: Who lived Eight and Twenty Years, all alone on an uninhabited Island on the Coast of America, near the Mouth of the great River Oroonoque; Having been cast on Shore by Shipwreck, wherein all the Men perished but himself. With An Account how he was at last as strangely deliver'd by PYRATES. Written by Himself. The first of the sequels (Defoe 1903a) takes him around the world, and the second (Defoe 1903b) offers meditations on his condition.

2. Green (1990) gives a thorough account of the many reworkings of the Crusoe motif across centuries, including a long line of French variants, from Rousseau's Emile through Verne's Mysterious Island to Tournier's Friday.

3. For analysis of Verne, see A. Martin (1990: 22), and for Verne's influence on spaceflight pioneers, see McDougall (1985a: 22).

4. For a sampling of criticism, see Defoe (1975) as well as Bloom (1988), McKeon (1987), and Watt (1963), which remains the quintessential materialist rendering of Crusoe. Ray (1990) gives further background on the work's place relative to French and English traditions of the novel, Bender (1987) links the

5. Defoe (1972: 4).

6. Robinson sails the world, not a regional sea, and whereas Odysseus lends his name to a state of travel, Crusoe gives his to an island. In Dialectic of Enlightenment, Horkheimer and Adorno engage in an extended analysis of the character of Odysseus as he navigates between myth and enlightenment, making a comparison with Crusoe: “The wily solitary is already homo oeconomicus, for whom all reasonable things are alike: hence the Odyssey is already a Robinsonade. Both Odysseus and Crusoe, the two shipwrecked mariners, make their weakness (that of the individual who parts from the collectivity) their social strength. Delivered up to the mercy of the waves, helplessly isolated, their very isolation forces them to recklessly pursue an atomistic interest …. Odysseus and Crusoe are both concerned with totality: the former measures whereas the latter produces it. Both realize totality only in complete alienation from all other men, who meet the two protagonists only in an alienated form–as enemies or as points of support, but always as tools, as things” (Horkheimer and Adorno 1987: 61–62). In addition to distinguishing between measuring and producing reality, one might also note a distinction between the axis of gender and desire in each work: whereas Odysseus encounters numerous temptations in the form of women and always seeks escape to return to hearth and wife, Crusoe inhabits a world with women firmly at the margin, in which his real marriage is to his island. Tournier's (1969) rewriting of the Defoe myth plays on an elaboration of this subtheme, which resonates with some feminist accounts of knowledge, desire, and nature (for example, Merchant 1990).

7. Defoe (1972: 70).

8. Ibid.: 66, 137.

9. Indeed the spontaneous appearance of barley leads him to proclaim a miracle and find meaning in his suffering (ibid.: 78).

10. Witness the remarkably brief paragraph in which he weds, has three children, and sees his wife expire (ibid.: 305).

11. Ibid.: 277, 305–6.

12. Ibid.: 47.

13. Hulme (1986: 214).

14. Ibid.: 176.

15. Disciplinary doctrine in North America calls for a division into four “subdisciplines”: social or cultural anthropology, archaeology, physical or biological anthropology, and linguistic anthropology, in descending order of numerical representation and influence. This doctrine is frequently ignored in practice. Neither of the two institutions that have granted degrees to me in the subject adhered to the sacred even number, one counting three formal subfields (social anthropology, archaeology, and biological anthropology), and the other five (sociocultural anthropology, archaeology, physical anthropology, medical anthropology, and linguistic anthropology), with a quiet sixth (folklore). Nor is this confusion recent; even a cursory glance along the lines of descent show a

16. See Boas (1962), Malinowski (1944), and Kroeber (1963), to name a few forefathers who confronted contemporary existence. The comparative sensibility of foremothers Margaret Mead (1928) and Ruth Benedict (1959 [1934])—the best known of all to wider audiences, if institutionally marginized—also should not be forgotten.

17. On the importance of Europe to any postcolonial project, see D. Scott (1995). Here the focus on the edge of Europe seeks to recall the possibility of multiple modernities within the West, as well as the emergence of modern formations between colony and metropole (Cooper and Stoler 1997).

18. For the outlines of some of this literature, see Clifford and Marcus (1986), as well as Behar and Gordon (1995), Fox (1991), Geertz (1988), Kondo (1986), Marcus and Fischer (1986), Narayan (1993), Rabinow (1977), Stocking (1983), and Visweswaran (1994).

19. Asad (1973) and Said (1979) represent two sources of criticism; Stocking (1992) provides a more detailed and historically nuanced account. Appadurai (1996) and Gupta and Ferguson (1997a, 1997b) render the spatial logic transparent, while Fardon (1990) sketches a number of different area genealogies. Dening (1980) provides a lyrical study of islands and beaches in a Pacific branch of the discipline.

20. For anthropologically inclined commentary, see Fox (1991), Geertz (1983), Latour (1993), Marcus (1986), and Rosaldo (1989).

21. Here some differences in national traditions must be noted. While similar remarks might be made about the British tradition and (perhaps more weakly) about the French, other “native” anthropology traditions (such as the Mexican or Brazilian) concentrate less on a displacement of language or culture and more on an internal displacement of class, remaining within national boundaries. A more complete consideration of such issues lies outside the scope of this work, but historical legacies of the formation of academic traditions within differing contexts of national identity and imperial relations should be kept in mind.

22. See Geertz (1973: 5–6), Nader (1972), and Rabinow (1989, 1996a), as well as Gupta and Ferguson (1997a).

23. For an overview of such work, see Cooper and Stoler (1997) and Pels (1997). For a small sample, see Cohn (1987), Comaroff and Comaroff (1992), Dirks (1992), R. Price (1983, 1990), Rabinow (1989), Prakash (1995), Taussig (1987), and N. Thomas (1994). The more historical ends of postcolonial studies and varied interdisciplinary influences (for example, B. Anderson 1991, Guha and Spivak 1988, and Said 1993) have also weighed heavily in shaping the historical focus on modern empire.

24. For an overview of globalization in the context of anthropology, see Appadurai (1996), Featherstone (1990), Fardon (1995), Hannerz (1989), Kearney (1995), and Marcus (1995). A focus on consumption and commodities (for survey, see Miller 1995) constitutes another effort to integrate local material culture within wider systems of exchange. Wolf (1982) and Mintz (1985) remain the most influential introductions of world systems thought into anthropology, introductions that came, I would underline, through the greater Caribbean. The

25. Latour and Woolgar (1986) and Traweek (1988) are two particularly influential laboratory ethnographies. For more recent experiments within anthropology, see Gusterson (1996), E. Martin (1994), and Rabinow (1996, 1999). Downey and Dumit (1997) and Marcus (1995) collect a range of studies, Nader (1996) provides a tie with anthropology's heritage, and Hess (1992) and Franklin (1995) offer an overview. Many of the more provocative and controversial elements of science studies literature influencing this ethnographic detour can be found in Biagioli (1999). Bruno Latour's (1988, 1993, 1999) irreverent philosophical histories and Donna Haraway's (1989, 1997) transgressive explorations of cyborgs and related creatures between fields of science and fiction remain key interdisciplinary influences, as does critical attention to gender (such as Keller 1985).

26. For a provocative discussion of the possibilities of localization as a key concept in anthropological practice, see Gupta and Ferguson (1997).

27. I would also suggest that the internalist impulse displayed in many studies of science and technology only underscores this tendency, as the inner workings of a particular machine or logic come to represent a period and quickly unfold to hide its historical horizon. One of the aims of this work is to talk about technology's attachment to the world in a way that allows offshoots, secondary systems, and reverberations into view, while using a marginal vantage point to emphasize the plural character of historical space. Periodization has, of course, long been a preoccupation in branches of the humanities, but the conceit of cultural production to represent universal time (for example, the “Renaissance”) plays out rather differently than that of material culture, whose effects quietly litter and transform environments. Prehistoric archaeology, with its necessary methodological conflation of time and culture in the form of artifacts, illustrates an extreme of periodization, ruptured only by silence (for historical description of the “three age system” of prehistory and its significance in relative dating, see Trigger 1989).

28. Ortner (1995) draws a useful distinction between “ethnography” and the “ethnographic stance,” the latter allowing for a broader legacy of cultural sensibility beyond presentist monographs. Rabinow (1997: ix–x) proposes the term dia-ethnographic to describe studies traversing cultural fields. Here I am shifting focus from center to edge in “culture” to concentrate on key elements of social framing in modern spatial practices, but with a very literal eye on geography. For a classic example of geography in the ethnographic tradition, see Evans-Pritchard (1940).

29. See Bilby (1990).

30. The play on Clifford and Marcus (1986) would be “Writing Place before Writing Culture.” Lest this allusion be misunderstood, I underline that the

31. Lévi-Strauss (1966: 21–22). Rabinow (1996) follows an earlier definition of “incidental movement” for bricolage; here I adopt the more literal usage in colloquial French for improvised problem solving, endemic in French Guiana.

32. Between 1990 and 1994 I went on four research trips to French Guiana and two to France, totaling approximately fourteen months, split between each site.

33. The opposition of space and place has a lengthy lineage in and out of academe; for informative discussions, see Certeau (1984), Heidegger (1977), and Tuan (1977).

34. For etymology of the term, see The Oxford English Dictionary (see the 1989 edition, 3: 494–96). The French cognates are essentially the same; indeed the terms coloniser and colonisation may derive from their English equivalents (see Le Grand Robert de la Langue Française, 1985, 2: 711–12).

35. For discussion of such distinctions, see Salomon (1990).

36. See Chalifoux (1987) as well as Price and Price (1992).

37. This definition would not deny that people with strong attachment to locale can also live with reference to elsewhere. I would, however, draw a distinction between those who consider themselves (and are considered by others) to be “in” place and those who consider themselves (and are considered by others) to be “out” of place, in addition to noting varying rates, scales, and qualities of circulation.

1. Such disturbances have occurred intermittently through Guyanais history; for an account of a 1983 riot, see Bilby (1990).

2. See Ralegh (1928).

3. Whitehead (1995) offers some suggestive evidence about the historical roots of Manoa. For an earlier effort to support Ralegh's account in face of his skeptics, see Van Heuvel (1844).

4. V.T. Harlow in Ralegh (1928: cvi). For context and interpretation of Ralegh's career, as well as textual analysis related to the exploratory era of Europe's encounter with the New World, see essays in Greenblatt (1993).

5. The term Cayenne may have similar origins, although one etymology would give it a European lineage (see Huyghues-Belrose 1990). Grenand in Hurault (1989: xviii) provides demographics.

6. See A. Henry (1989) and Bilby (1990). In 1704 the tabulated population included 264 whites, 83 Amerindian slaves, and 1,132 or 1,137 black

7. Jolivet (1982).

8. A quick comparative census illustrates Guyane's marginal role in the plantation system. Whereas the slave population of Saint-Domingue stood at 80,000 in 1730, that of Martinique at 36,000 in 1720, and that of Dutch Guiana at 50,000 in 1738, French Guiana only counted 6,996 slaves as late as 1763 (Bilby 1990).

9. Marchand-Thébault (1986: 11–17, 56). See also Bruleaux, Calmont, and Mam-Lam-Fouck (1986) and Mam-Lam-Fouck (1987: 13–27)

10. Until the beginning of the twentieth century, French claims extended south to the Amazon, a largely theoretical reign ended by the arbitration of the tsar of Russia and the Swiss government. Had the parties involved ruled in France's favor rather than that of Brazil, French Guiana would be approximately three times larger than its current size (A. Henry 1989: 177–86).

11. The number of immigrants varies between eight thousand and sixteen thousand in different sources (Bruleaux 1992: 157). Jacques Michel (1989) provides the most documented figure of 11,082, while Chaïa (1958: 60), who gives us the most exhaustive effort at post facto diagnosis, believes that 10,446 arrived and 6,500 died, based on figures of three thousand who returned to France and nine hundred who stayed. Given the range of records involved, I will adopt a commonly repeated estimate of twelve thousand. The point is that the Kourou expedition represented a large and sudden influx of European immigrants—about equal to the total number of persons involved in the plantation colony (12,549, including 10,748 slaves, according to Chaïa 1958: xxi)–and that a shocking number of them died.

12. See Damas (1938: 24–25). Damas takes the account from Pitou (1807).

13. Lowenthal (1960a).

14. See Archives Départmentales (1989) and A. Henry (1989: 144). For an account of a revolutionary exile, see Pitou (1807).

15. Bilby (1990) and Mam-Lam-Fouck (1987: 29–48).

16. The British also blocked French efforts to recruit in India in 1877 (Mam-Lam-Fouck 1992: 31).

17. Estimates for the number of miners working the fields hover around ten thousand for the period between 1890 and 1930, with a peak of some twenty-five thousand in 1901 (Bilby 1990; Jolivet 1982: 121–22). Given the nature of the terrain and activities involved this accounting is understandably loose; however, it is probably safe to state that throughout the gold rush period, the population of miners roughly equaled that of convicts.

18. Mam-Lam-Fouck (1987: 85–151). Also see Strobel (1998) for a lyrical portrayal of the fading world of Creole miners.

19. Bilby (1990). The Inini was also the scene of one of the more poignant footnotes of the penal colony in 1931, when some five hundred political prisoners from Vietnam were transported to special camps along the edge of this interior territory (A. Henry 1989: 201; Mam-Lam-Fouck 1987: 146).

20. The period brought Guyane a small whiff of international significance: an American newsreel from 1940 expressed concern that Axis powers would

21. A. Henry (1989: 211); Mam-Lam-Fouck (1992: 44); and Alexandre (1988: 138–40). The airstrip at Rochambeau, which was built partly with Puerto Rican labor, subsequently developed into Cayenne's current airport.

22. For a general description of France's DOM-TOM system, see Aldrich and Connell (1992). The policy—an aberration in the history of decolonization—stems partly from the assimilationist thread within French colonization (Girardet 1972; Lewis 1962). It should be noted that, at the time, French Guiana possessed two political figures of clear national stature: Félix Eboué, who had rallied Free French forces while governor in Chad, and Gaston Monnerville, who served as president of what became the French Senate between 1947 and 1968 (Mam-Lam-Fouck 1992: 65, 415).

23. The report is quoted in Cooper (1970: 70). The context of the quotation is Senegal and general efforts of the French legislature to redirect colonialism through development.

24. Mam-Lam-Fouck (1992: 21–22, 36).

25. Ibid.: 86–109.

26. Electricity use rose 506 percent between 1949 and 1953 and 111 percent between 1953 and 1958. The number of cars in the colony tripled between 1952 and 1960, despite the lack of roads. At the same time the relative value of exports (largely gold, rum, and wood) to imports fell from 21.26 percent in 1949 to 10.4 percent in 1956. (Mam-Lam-Fouck 1992: 110–50.)

27. Bilby (1990) and Mam-Lam-Fouck (1992: 181–205).

28. Bilby (1990); Mam-Lam-Fouck (1992: 260–80); and Schwartzbeck (1986).

29. Sécretariat d'Etat aux DOM-TOM (1976).

30. For a discussion of the efforts of American magnate Daniel Ludwig to build an enormous paper plant in the Amazon jungle in the 1960s and 1970s, see Miles (1990).

31. Mam-Lam-Fouck (1992: 272–73).

32. The estimated surface area of French Guiana varies between 83,534 square kilometers and 91,000 square kilometers, depending on the source. For our purposes the common round figure of 90,000 square kilometers will serve.

33. See CNRS/ORSTOM (1979). Also see Groene (1990) and Croyere (1984).

34. CNRS/ORSTOM (1979); ONF (1992); and Saga (1988).

35. Estimated numbers are something on the order of twelve hundred vertebrates, seven thousand to ten thousand plants, and four hundred thousand insects (ONF 1992: 10–12).

36. Trade with France, Metropolitan and otherwise, accounted for 67.08 percent of imports in 1992 (total value 3,811,659,000 ff, or approximately 762,331,800 U.S. dollars) and 84.74 percent of exports (total value 541,971,000 ff, or approximately 108,394,200 U.S. dollars). Here I am following figures provided by the French Guiana Chamber of Commerce, CCIG (1994: 6); the national statistics institute, INSEE (1993: 15), gives slightly different figures, with exports only covering about an eighth of imports. The

37. INSEE (1993) and CCIG (1993, 1994).

38. The department is subdivided administratively into two districts (arrondissements), one governed from Cayenne and the other from St. Laurent. The legislative body of the department is composed by a general council (Conseil Général) with nineteen members, each elected to represent a canton for six years. Since 1975 French Guiana has also been a region (région), a larger political grouping normally uniting several departments in the Metropole. The decentralization policy of 1982 also established a regional council (Conseil Régional) with thirty-one members also elected every six years. In effect this placed another sovereign into the same realm, and the separation of powers between general council and regional council is not always simple. On a more local level there exist twenty-two communes (the most recent created in 1993), each with a municipal council and a mayor. On a national level French Guiana has two deputies, one senator, and an economic and social advisor (Bilby 1990; CCIG 1993: 7–8; and CCIG 1994: 9).

39. Bilby (1990).

40. Ibid., and Hurault (1989).

41. For more on the Aluku, see the comprehensive dissertation of Ken Bilby (1990). For the Ndjuka, see Vernon (1992). For the Saramaka, consult the numerous works of Richard and Sally Price (for example, R. Price 1983; S. Price 1984).

42. Chérubini (1988: 13); Price and Price (1992: 24); INSEE (1993: 14, 27); and CCIG (1994: 9).

43. The official birthrate was 31.7 per thousand in 1991, and the death rate was 4.7 per thousand. In addition to the number of people presently living in French Guiana, one can add those born in French Guiana who now live in Metropolitan France, a total of 12,198 in 1990 (INSEE 1993).

44. For more on tensions within the ethnic landscape of French Guiana, see Bilby (1990), Chalifoux (1987), Chérubini (1988), Jolivet (1982), Mam-Lam-Fouck (1987, 1992), and Price and Price (1992).

45. For an initial point of comparison, see Lowenthal (1960b).

46. Hulme and Whitehead (1992).

47. Trouillot (1992).

48. For example, Mintz (1985), Price (1990), and Wolf (1982).

49. Ralegh (1928: 73).

1. “In der Strafkolonie.” See Thiher (1990: 51). As Hayman (1982: 187) points out, Kafka was perhaps influenced by early reports of trench warfare and would surely have heard of Devil's Island from the Dreyfus Affair and German war propaganda.

2. The work was first published in French as Surveiller et Punir (Foucault 1975 and 1979). Foucault locates the central motif of mechanisms of discipline in Jeremy Bentham's plans for the Panopticon, a model penitentiary where inmates would live under the constant gaze of a central tower and each other. The

3. For a sampling of the reception of Foucault among French historians of the modern prison, see Perrot (1980), Petit (1984), and Petit et. al (1991). For more conventional narratives of French prison history in English, see O'Brien (1982) and Gordon Wright (1983). Ignatieff (1978) offers an account of English prison reform that strongly overlaps yet remains differently positioned than Foucault's. Garland (1990) surveys the place of punishment in social theory from a less threatening perspective. Semple (1993) provides an exhaustive study of Bentham's Panopticon project. Beyond debates over chronology and the role of class, some of the reaction to Foucault's metahistorical account, I would suggest, stems from theoretical discomfort with his ambivalent portrayal of the Enlightenment legacy of reform or simple misreading, such as that of those who assume he claims that torture has vanished from all of the contemporary world.

4. Foucault (1979: 272, 279). See also interviews in Foucault (1980a), especially pages 63–77, 146–65, and 224–25. T. Mitchell (1991: 35) and Kaplan (1995) contain discussions of Panopticon prototypes in colonial contexts, and Stoler (1995) gives a wider reinterpretation of Foucault's thesis on sexuality in light of imperial dynamics of race.

5. The most common etymology given for bagne (bagnio, in English) links it with a bathhouse, or more likely, simply a building in which prisoners were housed during the period of hostilities between Mediterranean Europe and the Ottoman Empire (Le Clère 1973: 16–17, 27; Petit et al. 1991: 169).

6. See Ignatieff (1978) and Gordon Wright (1983). Another line of descent would lie in Russia, where links between exile and the settlement of Siberia strengthened in the eighteenth and nineteenth centuries (Petit 1984). Because the Russian experience represents an alternative version of empire one step removed from the intercontinental ventures of England and France, and because references to Siberia appear far less frequently than Botany Bay in the French documents that concern us, I leave this thread to one side.

7. For a more detailed description and chronology of the Australian adventure, see Shaw (1966), Hirst (1983), Carter (1988), and especially R. Hughes (1986), the most readable and comprehensive account, from which these figures are largely drawn (R. Hughes 1986: 161–62).

8. Ignatieff (1978: 47). Lest the complexity of genealogies be lost, it should be noted that Howard's views owed much to Dutch prisons, and other British reformers were aware of earlier experiments in France and elsewhere.

9. See Bentham (1962), Ignatieff (1978: 75), and Jackson (1987: 2).

10. Jackson (1988: 45).

11. Cited in Jackson (1987: 12). This is a draft passage apparently intended for his Finance Committee report. For further detail on Bentham's Panopticon scheme and his opposition to Australian transportation, see Everett (1966), Hume (1973, 1974), Jackson (1987, 1988), and Semple (1993), in addition to Bentham (1962, 1977).

12. Hirst (1983: 21–27) and R. Hughes (1986: 162).

13. From Vidal de Lingendes, “De la colonisation pénale,” 1845, p. 11. CAOM H bagne 1.

14. Forster (1996) provides a focused survey of French interest in Botany Bay and Australia during the period. When asked in 1851 to prepare a report on prior French projects of penal transport, the head of the French archives traced the tradition back to the era of the initial settlement of Canada and a 1540 patent letter given to Jean Françoise de la Rouge that allowed him to recruit sailors from among those facing the death sentence (CAOM H bagne 3). Experiments in Louisiana between 1718 and 1722, however, foundered (Gordon Wright 1983: 31). See also Pierre (1982: 10–12).

15. Gordon Wright (1983: 31, 44).

16. Cited in Devèze (1965: 28, see also 25–26).

17. See Gordon Wright (1983: 45–46), Devèze (1965: 51), and Pierre (1982: 15).

18. Gordon Wright (1983: 48–53).

19. “Memoire sur le choix d'un lieu de déportation,” signed Forestier, dated 1816, in CAOM H bagne 1. Colin Forster reviews this same text, noting that Forestier was Councilor of State of the Committee of the Navy and Colonies (1996: 15–18).

20. T. Ginouvier, Le Botany-Bay Français, ou colonisation des condamnées aux peines afflictives[sic] et infamants et des forçats libérés, printed in Paris, 1856. CAOM H bagne 3. See also Forster (1996: 28).

21. CAOM H bagne 1. Another dated 1830 makes an argument based on what will be a familiar racial logic of labor: “French Guiana indeed is not in the least way proper to this end[;] … experience of several centuries has proved that all labor in the air of these climates is mortal to whites.” CAOM H bagne 2.

22. CAOM Guyane L1 (01), Lettre du Gouverneur de la Guyane française à direction, 11 September 1828.

23. Gordon Wright (1983: 70–81).

24. Documents dated July 1848, in Bodereau des pièces communiqués à la Commision présidée par M. Amiral de Mackau, Mars 1851. CAOM H bagne 1. For Marquesas reference, see Devèze (1965: 90) and Gordon Wright (1983: 92).

25. Cited in Clair (1990: 41) and Pierre (1982: 17). Louis-Napoleon was himself fascinated with Australia (Miles 1988: 21).

26. Gordon Wright (1983: 93, 295–96).

27. Devèze (1965: 119–29); Merle (1995: 40); Pierre (1982: 18–19); and Gordon Wright (1983: 93–94).

28. From L'Univers (Union Catholique), 16 January 1857. CAOM H bagne 1.

29. Witness a tract on penal deportation written in 1828 by a navy officer and later reprinted in 1840. The author criticizes British transportation for allowing prisoners excessive freedom and failing to encourage rehabilitation, stating that the real purpose of transportation must be “moral” health: “The goal of deportation is not only to distance from society those men that it has branded but also to return them to honest sentiments and to rehabilitate them in their own regard, by distancing them from places where the memory of the

30. CAOM H bagne 1.

31. Vidal de Lingendes, “De la colonisation pénale,” 1845, p. 14. CAOM H bagne 1. The passage concludes by considering the opposite extreme: “On the other hand, to exile the inhabitants of southern countries amid frozen deserts like those of Siberia would expose them to most excruciating suffering, and perhaps also to death.”

32. The same governor sought to encourage convicts to marry prostitutes from Martinique, but his plans met with little success, and he was quickly removed from office (Devèze 1965: 129; also Clair 1990: 19).

33. Cited in Devèze (1965: 130), who notes that the clearing of Ile Royale took only fifteen days. See also Clair (1990: 20).

34. Journal de Debats Politiques et Littéraires, 30 July 1852. CAOM H bagne 1.

35. Les Antilles, no. 20, 10 March 1852. CAOM H bagne 1.

36. Devèze (1965: 133) and Miles (1988: 25–27).

37. CAOM H bagne 4.

38. Letter of 21 December 1854. CAOM H bagne 14.

39. Devèze (1965: 135).

40. “Extrait du rapport de M. le Médecin en chef de Cayenne,” 1854. CAOM H bagne 14.

41. Letter of 15 February 1857. CAOM H bagne 14. On diet, see Notice sur la transportation à la Guyane Française et à la Nouvelle Calédonie, 1868–1870, p. 59, and Clair (1990: 49).

42. Pierre (1982: 24–27).

43. Another thousand or so women would be sent to New Caledonia; the majority in each case were sentenced under the recidivist laws of 1885 (Clair 1990: 39). Mortality among them was remarkably high: 23–54 percent in New Caledonia and 44–69 percent in French Guiana. For a statistical breakdown, see Krakovitch (1990b: 283–95), Clair (1990: 36–39), and Devèze (1965: 136), as well as the exhaustive early studies of Orgeas (1885, 1886).

44. Clair (1990: 23).

45. Devèze (1965: 142).

46. “La Verité sur les pénitenciers de la Guyane,” L'Economiste Française 21 (25 October 1882): 292–93, together with a lengthy rebuttal letter from the governor to the minister, 14 December 1862. CAOM H bagne 4.

47. 10 September 1856. CAOM H bagne 14.

48. From Bollot, “Un Pénitencier doit être une véritable maison de santé morale,” 1868. CAOM H bagne 4.

49. Gordon Wright (1983: 95). The shipment of about fifty Metropolitan women to French Guiana during the suspension of European arrivals (1867–1885) represents an interesting exception (Clair 1990: 36). Isabelle Merle (1995: 63) suggests that this redirection of convicts of European extraction to

50. Merle de Beaufont to minister, 21 December 1864, pp. 88, 102–3. CAOM H bagne 4.

51. Bollot, “Un Pénitencier doit être une véritable maison de santé morale,” 1868. CAOM H bagne 4.

52. “Etude sur la transportation à la Guyane.” CAOM Guyane L1 (08). Although undated, the draft copy (with many sections amended or crossed out, and footnotes added) makes references to the thirty-year history of the overseas bagne, which would place it in the early 1880s. Judging from content and style, it appears to be a draft of Orgeas's extended demographic study (1885 and 1886).

53. From Eribon (1991: 237), quoting a 1975 Le Monde interview with Foucault.

54. In this sense the penal colony serves as an example of an alternative modernity constructed around institutionalized failure, a place where governmental norms are suggested but not applied.

1. Marie-Antoinette Menier's article, “La Détention du Capitaine Dreyfus à l’île du Diable, d'après les archives de l'Administration pénitentiaire,” represents a significant exception (Menier 1977).

2. As the new venture in French Guiana gathered steam, the bagnes of Metropolitan France closed; Toulon, the last, shut down in 1873 (Le Clère 1973: 22; Pierre 1981: 77).

3. Pierre (1982: 35). Official death rates for the bagne in Guyane fluctuated between 4.5 percent and 26 percent (Pierre 1982: 311–12). As we shall see in chapter 7, 10 percent can be taken as a rough average for the time period involved.

4. The transportation of recidivists was a proposal of the First Republic, in part inspired by the British example of Botany Bay (Devèze 1965: 25–29). The terms of the 1885 debate were scarcely changed; one senator pointed to the success of transportation in Australia as justification for the new policy (Gordon Wright 1983: 143–45). For text of the law, see Pierre (1982: 309–10). For additional intellectual context, see Pick (1989: 182–83).

5. A number of efforts toward its abolition were made within as well as without the French government (Gordon Wright 1983: 150–51), but the eventual suppression occurred only with shifts in the political climate. The most famous campaign against the bagne was that of the journalist Albert Londres between 1923 and 1927 (Londres 1975). The efforts of prominent Guyanais, including M.G. Monnerville and L.-G. Damas, also played a significant role in abolishing transportation (see Donet-Vincent 1992, 1993). Though the final European shipment was in 1954, the last Indochinese prisoners of the interior

6. Alexakis (1979: 111–12) and Pierre (1981: 78–79).

7. Pierre (1981: 78). Also see Le Clère (1973: 68–71). For a thorough study of women in both the Guiana and New Caledonia Penal Colonies, see Krakovitch (1990b).

8. Mam-Lam-Fouck (1987: 141). It should be noted that the civil population contained a much smaller percentage of Europeans than the convict population (Mam-Lam-Fouck 1987: 163–227).

9. Michel Pierre provides a relative breakdown of the bagnard population (63 percent European, 25 percent Arab, 7 percent black, and 5 percent Asian) but without specifying total figures (Pierre 1982: 41). A sample from 1907 (a low year) shows 2,605 Europeans to 998 Arabs, 361 Africans, and 130 Indochinese (Le Clère 1973: 61). For further statistics, see Devèze (1965: 165–68), Krakovitch (1990b: 260–61), and Pierre (1982: 307–15). For a character profile of the petty French underworld, see the life of Arthur Rocques as reconstructed from documents by Claude Barousse (1989).

Born July 1852 (the same year as the bagne) of unknown father in Montpellier, Arthur Rocques was probably brought up as an orphan before leaving school at twelve. After spending his teens as a cabin boy, he left the navy in 1869 and was first arrested for theft in Tours. In 1870 he received a month in prison for rebellion and two more for vagabondage (his later memory was selective, recalling only the humiliation of France at the hands of the Prussians). Part of the army fighting against the Paris Commune, he deserted (which he also neglected to mention later), and received a ten-year sentence, probably spent in a disciplinary battalion in North Africa, although no record of his presence there remains. During this period he somehow learned grammar, arithmetic, geography, history, a good “dose d'anglais,” a little Spanish, and some literature, art, biology, and accounting. Books became his irreplaceable companions. Released, he wandered back through France and into petty crime. In 1886–1887 we find him in Sèle, engaged to work for a woman who had a small horse and buggy enterprise. Marie Vors, a widow, had two daughters—one drove coaches and the other was apprenticed as a milliner. After two years of wages, some drink, and gambling, Rocques married Marie, who was fourteen years his senior. When trams arrived in town, their business folded, so in 1891 they moved to Montpellier. He started to take “business trips,” gamble, and swindle, and then was arrested for posing as a police official in order to get into people's houses. Sentenced to five years, he escaped from police. By this point he had acquired a mistress—his stepdaughter Julie, the milliner. It is unclear how long the affair had been going on, but she was by then expecting his child. He moved to Vichy under a false name, then to Marseilles, and began passing counterfeit money. In 1900 and 1901 his two daughters were born; the couple left one with Marie near Bordeaux and gave the other to a wet nurse. Arthur and Julie were arrested in La Rochelle on November 16, 1901, for trying to use false two-franc pieces. He was sentenced to hard labor for life but gave an impassioned speech accepting blame and clearing Julie. This succeeded in saving

Aside from his striking autodidactic ability and tabloid personality, Arthur Rocques's life describes a fairly representative trajectory through the more marginal classes of France, bobbing above and below the line of criminality. Richard Price (1998) unearths a striking colonial corollary in the case of Médard Aribot, a Martiniquan carver exiled as a relégué. While even more remarkably idiosyncratic, Médard's life puts the bagne into regional context, and Price's reconstruction of the colonial order in Martinique provides an exceptionally nuanced panorama of the contemporary French Empire in its “old” colonies.

10. Clair et al. (1990: 27) describe white uniforms with red stripes without specifying relégué dress. Whatever the exact hues fading in the sun, accounts usually portray the clothing as poorly made, ill-fitted, and subject to wear and theft.

11. Clair (1990: 23–35) and Pierre (1981: 79–82). Reported successful escapes ranged from a low of three in 1935 to a high of 250 in 1906. An estimate for the years up until 1921 suggests that about one in six prisoners eventually escaped or disappeared in the attempt (Pierre 1982: 311–12).

12. Miles (1988: 155–66) and Pierre (1981: 82).

13. J. Munroe, “A Visit to One of the Prisons of Cayenne,” Good Works (October/November? 1878): 746–752. Copy contained in CAOM, H54. Hereafter, this work is cited parenthetically in the chapter text.

14. Tallach wrote: “At the Prison Congress in Stockholm I listened with much interest to your eloquent speeches in defense of transportation …. Pardon me, dear sir, for the liberty I now take in inviting your official consideration of the article in question.” The French administration did consider it, because the article is underlined and annotated. The letter, addressed to E. Michaux, Directeur des Colonies au Ministère de Marine et des Colonies, is dated November 12, 1878. This and all other documents related to the minor affair reside in CAOM, H54. The mention of the Sisters of Charity receives a “très bon,” and those related to the mentally ill contain many lines and crosses.

15. As an addendum, it is sometimes noted that the strong currents surrounding the remotest island may well have inspired thoughts that the Devil lurked nearby, grounding the choice of name in nature and locating the threat of damnation surrounding salvation. Support for the Kourou narrative can be found in early maps that refer to the entire group as “Diable.” See Adélaïde-Merlande (1986: 195), Bouyer (1990: 68), and Huyghues-Belrose and Bruleaux (1988). The convict forger Flag offers an interesting variant, suggesting that the name derives from flocks of black birds that would descend on the island—resolving the issue again to nature, though here in living form (Lagrange and Murray 1961: 9).

16. Despite the fact that in French its name was most frequently associated with the bagne, Cayenne, like Devil's Island, only played a minor role in the

17. Bredin (1986); Halasz (1955); and Gordon Wright (1987).

18. Following his rehabilitation, Dreyfus retired into a highly private and quiet life, suffering from occasional attacks of fever and a lingering aura of notoriety. In 1908, while attending the ceremonial induction of Zola's ashes into the Pantheon, he was slightly wounded by a gunman. With the outbreak of World War I he was mobilized in the reserve artillery command for the defense of Paris, and he later served in the battles of Chemin des Dames and Verdun, before finally rising to the rank of colonel in the reserves. The war claimed the lives of his nephew and son-in-law. While carefully collecting and classifying materials relating to his case (and apparently suffering from repeated nightmares), Dreyfus would rarely speak of his ordeal. In July 1935, at the age of seventy-five, he died quietly in bed (Burns 1991). Among the frequent complaints about his lack of public flair, we can note that even his champion Clemenceau is said to have described him as “looking like a pencil salesman” (Bredin 1986: 490).

19. Menier (1977).

20. Devèze (1965: 104–5) and Gordon Wright (1983: 149–50; 1987: 249–51).

21. Dreyfus (1901: 200). As Jean-Denis Bredin points out, religion—otherwise at the center of things—plays no role in Dreyfus's writing (Dreyfus 1899, 1901); his journal mentions God only twice (in reference to Schopenhauer) and Judaism not at all. Rather, his religion is that of the patriotic soldier, honor and country above all (Bredin 1986: 130).

22. Dreyfus (1899: 119, 137; 1901: 191, 226–67).

23. Dreyfus writes:

My suffering is at times so strong that I would tear my skin from my flesh, to forget in physical pain this too violent torture of the soul. I arise in the morning with the dread of the long hours of the day, alone, for so long, with the horrors of my brain; I lie down at night with the fear of sleepless hours …. My body is broken, my nerves are sick, my brain is crushed, say, simply, that I still hold myself erect in the absolute sense of the word only because I resolved to, so as to see with you and our children the day when honor shall be returned to us.

Ibid.: 47

24. Quotations from Dreyfus (1901: 173–74, 261, 226–27; 1899: 150). See also Bredin (1986: 132–33).

25. Quotations from (Dreyfus 1901: 154; 1899: 176); also Bredin (1986: 127).

26. E. Weber (1976) and Girardet (1972). For a discussion of race, class, and national identity in a later parallel to French Guiana, see B. Williams (1991).

27. See the essay “Truth and Power” in Foucault (1980b, esp. 126–33). Also see Rabinow (1989: 16, 251) for the distinction in a relevant context. “The Affair” continues to serve as a defining moment of the “engaged” intellectual in action. Ironically, Dreyfus would qualify for the specific category as an efficient military technician, concerned more with precision than passion. In this sense he was very much a figure linked to Kafka's world.

28. O'Brien (1982: 288–89).

29. In an extreme conflation of race and climate, 525 political prisoners from Indochina settled the interior Inini territory in 1930 under the worst of conditions (Clair et al. 1990: 45).

30. For example, Bouyer (1990).

31. There was also British, German, and South American fascination with and outrage over the bagne. Here I focus on the American strand, which would prove the most significant. Donet-Vincent (1992); Londres (1975); and Miles (1988).

32. See Batzler-Heim (1930), Belbenoit (1938, 1940), Allison-Booth (1931a, 1931b), Davis (1952), Krarup-Nielsen (1938), Lagrange and Murray (1961), Niles (1928), Rickards (1968), Seaton (1951), Sinclair (1935), and Willis (1959). For titles not directly referenced, as well as additional reading in English and French, see Miles (1988: 197–204). In addition to these written accounts, one could also add film: between 1925 and 1955 a number of Hollywood works found their setting in the French Guiana bagne, a tradition that would be recalled most successfully in the film version of Papillon in 1973. French cinema has also produced a range of works on the penal colony, one of the more recent of which, Seznec, resulted in the partial restoration of the prison ruins in St. Laurent (Miles 1988: 110–15).

33. For example, Roussenq (1957). At this point the most familiar reference to both English and French speakers is the account of the petty pimp Henri Charrière, known as “Papillon” for his butterfly tattoo. Charrière turned his own (and other people's) experiences into a runaway best-seller in 1969, later released as a Hollywood blockbuster (Charrière 1970). For more sober evaluations of the authenticity of his narrative (particularly his claims to innocence and prowess), see Ménager (1970) and Villiers (1970).

34. Allison-Booth (1931a: vi). This figure can be taken as a measure of the dubious hyperbole infusing the work.

35. On exposé literature, see Londres (1975: 13–197) and Donet-Vincent (1992, 1993). Another version of Devil's Island was published under the title Hell's Outpost, and in the preface to that edition the editor makes the connection to Blair Niles (Allison-Booth 1931b).

36. Allison-Booth (1931a: 59).

37. The convict is Roussenq, a figure legendary for his extraordinary incorrigibility and time spent in solitary punishment.

38. All convicts in Guyane received a number based on their “matricularion,” or sequential order of entry. The man born Henri Charrière, for example, may have been called Papillon, but upon arriving in the bagne he was officially known as No. 51367 (Clair et al. 1990: 56).

39. Niles (1928: 44). Here choice of metaphor echoes the manner in which this particular removal from civilized society involved natural order.

40. One of the more frequently noted items of convict material culture was the plan, a small screw container for precious items that was inserted into the body through the rectum (for illustration, see Miles 1988).

41. Papillon is far from the only convict to claim innocence. One favored motif on the part of foreign-born authors is to claim that they were tricked into joining the French Foreign Legion and then falsely judged by its rules (see Batzler-Heim 1930 and Krarup-Nielsen 1938).

42. Allison-Booth, whom I have chosen as a guide, is in this respect an interesting exception; rather than homosexuality he focuses on the presence of “fallen women” to entertain the guards, leaving only an ambiguous chapter title “Men Must Have Women” as a hint about the rest. This aspect of his account fits with its general omission of detail and suspiciously vague portrayal of everyday experience in the colony. For a period medical view of homosexuality among the convicts, see L. Rousseau (1930), and for further historical discussion, see Donet-Vincent (1992: 62).

43. Bourdet-Pléville (1960: 179).

44. None of the narrators admit to homosexual encounters themselves; rather, it is constantly spoken of in general or eyewitnessed terms. The sexual activity of choice for the narrator occurs either in the jungle with remarkably willing Indian women or in town with Creole women turned prostitute. A reassertion of heterosexuality and masculinity becomes necessary as a prisoner's autobiographical narrative makes the transition from “a jungle hell” to an overcoming, a survival. The hero must suffer, but in order to speak, to claim his identity as a man, he must remain unfallen. (For example, see Charrière 1970: 158–86, and Milani 1977: 207–9). In light of what follows, the tension of sexual conquest can also be read as an effort to reassert the imperial order of race.

45. “Oraput” was the name of a timber camp for incorrigibles. The song appears in several versions in different sources, not all of which include the slang reference to Arab command. Here I follow the rendition of a former prisoner, René Belbenoit (1938: 60) and adopt his free translation. The French: “De douleur, de dégoÛt nôtre coeur se soulève / Car la voix d'un arabe a crié: Roumi rö! / Ce supplice sans nom chaque jour se repète: Enfants des vieux gaulois, qu’êtes vous donc devenus? / le plus forts d'entre nous marchent en courbant la tête / Pleuvez, pleuvez foçats, vos coeurs ne battent plus! [your hearts beat no more].”

46. Herménégilde Tell, the father-in-law of Félix Eboué, became director of the penal administration in 1919 (Weinstein 1972: 72).

47. Belbenoit (1940: 18–19) and Allison-Booth (1931a: 71, 102). See also Batzler-Heim (1930: 95–96).

48. Seaton (1951: 60, 115) and Allison-Booth (1931a: 30–31).

49. See Batzler-Heim (1930), Krarup-Nielsen (1938), Londres (1975), and Belbenoit (1938, 1940).

50. Niles (1928: xiii).

51. Ibid.: 216.

52. Sinclair (1935: 20).

53. Ibid.: 22–24.

54. Ibid.: 23; Krarup-Nielsen (1938: 255).

55. Belbenoit (1940: 246–47). Mounted under glass and handsomely framed, butterflies remain a staple item in contemporary Cayenne tourist shops.

56. Seaton (1951: 25, 50).

57. Lagrange and Murray (1961: 174). See also Niles (1928: 366–67).

58. Belbenoit (1940: 37)

59. Niles (1928: 190). Another breathless example: “Before we know it we're surrounded by huge trees, ninety feet high, palisanders and mahogany

60. Miles (1988: 51).

61. Niles (1928: 206–7).

62. Cendrars (1958: 61). Jean Galmot, whose letter is cited here, was a Frenchman of many trades (gold seeker, rum merchant, writer, and politician), whose death (most likely by assassination) following an electoral victory sparked one of the worst periods of rioting in French Guiana's history (A. Henry 1989: 193–99).

63. Niles (1928: 251).

64. Ibid.: 199–200.

65. Ibid.: 133.

66. Pierre (1981: 76; 1982: 305).

67. Ann Stoler has questioned simple categories of “colonizer” and “colonized,” examining European moral narratives as conceived in gender and race. She concludes that “sex in the colonies was about sexual access and reproduction, class distinctions and racial demarcations, nationalism and European identity—in different measure and not all at the same time” (Stoler 1991: 87; see also Stoler 1989, 1995). In the case of Devil's Island a similar formulation would seem to hold, though here the moral questions were inverted: rather than methods to prevent degeneration and guard European prestige in the colonies, we find mechanisms to contain Metropolitan degeneration and transport it into a distant and rigorous colonial spectacle.

68. The comparison of nationalism to religion is an old reflex (see, for example, Hayes 1960). For a discussion of the development of purgatory within the context of medieval theology and social history, see Le Goff, who suggests that “the mobile frontier turned out to be the one between Purgatory and Hell” (1984: 226). Although terminology in play derives from the historical legacy of Christianity, conceptions of worldly and otherworldly place in relation to death can be found elsewhere, such as in Malinowski's description of baloma in the Trobriands, where the main spirit form of the dead person goes to a particular neighboring island (1954: 150). We should also note that the rise in transportation can be placed alongside a decline in the enforcement of the death penalty. “The new humanitarian system was still killing prisoners but in new ways and in remote places” (O'Brien 1982: 285). Thus another way of casting prison history would be to describe a vast expansion of the moment of execution.

69. See Burns (1991: 295, 328). When writing his memoirs, Dreyfus described the project to a friend as differing from that of Robinson Crusoe in that he (Dreyfus) had no goal to live on his island but rather must write to fulfill a particular duty: that of telling the story of “the five years I was cut off from the world of the living” (Dreyfus 1901: 294).

70. Mam-Lam-Fouck (1987: 139) and Pierre (1982: 285).

71. Londres (1975: 68). The “fifty years” would appear a rhetorical exaggeration, because construction only began in earnest in 1906–1907, less than

72. Mam-Lam-Fouck (1987: 142); Pierre (1982: 284); and Donet-Vincent (1992).

73. Quoted in Racine (1988: 49). Damas's experience was far from unique; Gaston Monnerville, the Guyanais politician instrumental in orchestrating the final closure of the penal colony and who eventually became president of the French Senate, worried equally about the land's reputation. See also Miles (1988: 194) and Weinstein (1972: 128). Forty years earlier, Clovis Savoie, an earnest historian of New Caledonia, concluded his work with the following appeal:

French Readers! Tell your friends that New Caledonia is no longer a penal colony. That since 1896 no more convicts have come to New Caledonia. That since 1896 the penitentiary has slowly passed away, by the constant reduction of its personnel of convicts and relégues. That the sale of the furniture and of the equipment of the penitentiary depot on the Ile Nou is today finished [1922]. That New Caledonia, this land blessed with beautiful sunshine and a good climate, has become free soil like any department of France. Make the legend that detracts from the good reputation of our magnificent island disappear. Tell all this to your friends. Thank you!

Savoie 1922: endpage

74. Kesteloot (1974: 232) and Damas (1938: 202).

75. Belbenoit (1938: 60): “Chacun pour le tavail s'arme d'une bricole / Et dans le forêt sombre advance trébuchant / L'on direant des démons, la sarabande folle / Car l'enfer est au bagne, en non pas chez Satan!”

76. In Shamanism, Colonialism, and the Wild Man: A Study in Terror and Healing, Taussig seeks to rework colonial history along the lines of “mythic subversion of myth,” exploring the colonial reality of the New World in a more than realist fashion. “The formulation is sharp and important,” he writes, “to penetrate the veil while retaining its hallucinatory quality” (Taussig 1987: 10). Twin concepts that mark a point of entry to this veil are those of the “culture of terror” and the “space of death.” Understood as a social as well as physiological state, terror becomes the “mediator par excellence of colonial hegemony.” The space of death, populated by the social imagination with “images of evil and the underworld,” becomes the place where “the Indian, African and white gave birth to a New World” (ibid.: 5). Taussig then proceeds to read accounts of torture in Colombian rubber plantations for keys into their universe of terror and death. As he notes, “two interlacing motifs stand out in these stories: horror of the jungle and horror of savagery” (ibid.: 75).

77. Seaton (1951: 21).

78. Hammel (1979: 44). See also Petit et al. (1991: 258).

79. Allison-Booth (1931a: 227).

1. Verne was well read by all three of the major figures usually mentioned in connection to the development of the rocket: the Russian Tsiolkovsky, the American Goddard, and the German Oberth (Winter 1990: 1–27). Ironically, his main technical flaw was to choose a cannon over a rocket as a means of propulsion into space. See Verne (1958, 1966a, 1966b, 1995).

2. Sergey Korolev was the chief engineer of the Soviet space program after Sputnik's successful launch.

3. As W.W. Rostow wrote in 1960:

There is no clear analogy in American history to the crisis triggered by the launching of the Soviet satellite on October 4, 1957. This intrinsically harmless act of science and engineering was also, of course, both a demonstration of foreseeable Soviet capability to launch an ICBM and a powerful act of psychological warfare. It immediately set in motion forces in American political life which radically reversed the Nation's ruling conception of its military problem, of the appropriate level of the budget, and of the role of science in its affairs. The reaction reached even deeper, opening a fundamental reconsideration not only of the organization of the Department of Defense but also of the values and content of the American educational system and of the balance of values and objectives in contemporary American society as a whole.

Quoted in Bulkeley 1991: viii

4. Popular science accounts of the coming Space Age were also in evidence; see Bergaust and Beller (1956). Also see McDougall (1985a: 60–119).

5. See McCurdy (1997).

6. See D. Carter (1988), McDougall (1985a: 43), Neufeld (1995), and Winter (1990).

7. Historian Walter McDougall makes the technocratic argument most strongly: “The advent of spaceflight in our time, therefore, is not just a tale of the gumption of Russian, German or American rocketeers, but also of the progress of the idea of command technology as a tool and a symbol of the modern scientific state” (McDougall 1985a: 19). The fact that W.W. Rostow, a player in the Kennedy and Johnson administrations' space policy, was also the author of a central work of development theory should not be too quickly overlooked. For more on the history of rockets, see Winter (1990). McDougall's volume, The Heavens and the Earth (1985a), stands out as the single most comprehensive and literate history of the Space Age. Wolfe (1979) captures the essence of the Space Age within popular culture, while Arendt (1978) and Blumenberg (1987) offer skeptical counterpoints about its significance in human history. Leslie (1993) and Galison and Hevly (1992) provide additional context to situate the growth of the American space program amid general trends of Cold War research.

8. Collins (1990) and Naddeo-Souriau (1986).

9. See interview with Pierre Auger, “Le CNES a 30 ans,” CNESQUISEPASSE? (April 1992): 26. Auger, a specialist in cosmic rays who had been instrumental in the earlier creation of CERN (the European Center for Nuclear Research), went on to play key roles in the creation of CNES and cooperative European space programs. See also Pestre and Krige (1992), Collins (1990), and McDougall (1985b). Hecht's (1998) account of the French nuclear program nicely illustrates the marriage of technology and politics in postwar efforts to reinvent France as a technological power. See also Bess (1995).

10. Collins (1990: 9–10); McDougall (1985b); and Morton (1989). Other test sites considered were in Canada, South Africa, Australia, New Zealand, and the Bahamas (Morton 1989: 10).

11. A. Clarke (1968: 37).

12. Pierce (1968).

13. Ackroyd (1990: 1–2).

14. McDougall (1985a: 357–59). Because of its high latitude, the Soviet Union's early communications satellites used an odd elliptical orbit rather than a geostationary one. Given the curvature of the Earth, equatorial orbit serves areas near the poles less well than those in temperate or tropical zones, a factor that comes into play in developing systems of mobile satellite communication (Williamson 1990: 296). However, for the majority of the globe and the systems that seek to encircle it, geostationary orbit is the position of choice.

15. Blonstein (1987: 1–9); King-Hele (1992); and Pool (1990).

16. Pease (1991: 11); Lambright (1994); and Mack (1990). SPOT is an acronym for Système Pour l'Observation de la Terre (System for Earth Observation).

17. Pease (1991: 1–27). Orbits are tricky things, and their fine points require more calculation than here described. For our purposes the most significant point is that maintaining position aloft and performing designated functions requires energy; even when well placed in the initial launch, satellites have a limited life span. For more on communication satellites, see Ackroyd (1990), Blonstein (1987), L.J. Carter (1962), and Williamson (1990). For more on geostationary orbits, see Soop (1994). For observation satellites, see Mack (1990) and Pease (1991). For orbits, the shape of the Earth, and the early history of satellite theory, see King-Hele (1992).

18. Kern (1983) and Read (1992).

19. Shifting plans for the establishment of tropical launch centers can be followed in general and industry news stories (most consistently and specifically in Launchspace and International Space Industry Report, or more generally in Aviation Week and Space Technology). The journal of the Guiana Space Center itself, Latitude 5, regularly monitors active launch competition in the United States and Russia, as well as the potential threats in China, India, and Japan, and development of rival launch sites in places such as Brazil. For historical grounding, see the collection in CNES (1972); the most interesting alternative to a tropical land base remains a floating platform, once contemplated by the United States, built for test rockets by Italy and recently introduced for commercial launches by a Norwegian-Ukrainian-Boeing consortium known as Sea Launch (Reuters 1994). For a discussion of space technology relative to development, see Wise (1990). I would also like to thank John Leedom for providing me with regular clippings and local perspectives on the New Guinea situation.

20. The chapter text's subsequent description of the choice of a launch site appears in a 1994 CNES/CSG press release on the occasion of the space center's thirtieth anniversary (CNES 1994). The press release (also printed in Antilla 582 [29 April 1994]: 21–22) consists of excerpts from both an official history of CNES (l'Histoire du CNES, in preparation by Claude Carlier) and a 1964 CNES report by Raymond Debomy (“Recherche de sites de lancement”). Studies aside, it should be noted that both French and later European choices of Kourou have more than a whiff of political destiny about them.

21. Other criteria not explicitly mentioned in this document but often noted in later discussions of the location of CSG are stability of the physical (as

22. Indeed the space survey matches the penal colony survey in its relative disregard for the issue of whether the territories in question were under active control, and either could have been accomplished with a reasonable atlas.

23. CNES (1994).

24. Jolivet (1982: 443–49) and Mam-Lam-Fouck (1992: 293).

25. See interviews in Latitude 5 21 (July 1993): 11–13.

26. Desobeau (1990: 7).

27. Out of 1,016 workers employed in the construction in 1967, 206 were European, and 330 were from elsewhere in South America. In 1968 the operation employed 3,502 persons. The official ethnic breakdown was as follows: 26 percent European, 11 percent Guyanais Creole, 20 percent Maroons (primarily Aluku and Saramaka) and Amerindians, 3 percent French Antillean, 4 percent from English and other Caribbean islands, 6 percent Surinamese Creoles, and 30 percent Colombians and Brazilians. In 1971, after the end of the first phase of construction, the labor force had shrunk to 1,599 persons, 52 percent of whom were European, 16 percent Guyanais Creoles, and 8 percent French Antillean. (Figures taken from Mam-Lam-Fouck 1992: 301 and Jolivet 1982: 445.) Like all statistics in this context they may represent as much of a distortion as a record; oral evidence suggests that the number of Maroons was much higher than the figure given (R. Price, personal communication, 1995).

28. CNES (1987, 1990); Collins (1990); and ESA (1992a).

29. Mam-Lam-Fouck (1992: 296).

30. See the “Conférence d'Information, Jeudi Oct. 1, 1964,” published in Guyane's Radio Presse, 6–8 October 1964.

31. Figures vary between 30 and 40 percent. Mam-Lam-Fouck (1992: 297) and Jolivet (1982: 469).

32. All these comments (and a few more besides) are to be found in Jolivet (1982: 470–74). I have reordered their presentation slightly from the sequence she gives; however, all date from 1971.

33. According to a 1987 report commissioned by CNES, some 5,292 million French francs flowed through the space center between 1965 and 1974, 3,156 million in the form of investment (Rémondière and Colmenero-Cruz 1987). Figures are given in 1985 French francs equivalent. Given a rough currency exchange of five to one prevalent in the early 1990s, five billion francs would be about one billion dollars (somewhat less in 1985).

34. Mam-Lam-Fouck (1992: 302).

35. Ibid.: 299, 304, 306.

36. Vignon (1985: 365).

37. The following passage is to be found in Jolivet (1982: 446–47); sections of it are also translated in Bilby (1990) and Price and Price (1992: 42–44).

38. Certeau (1984: 34–39).

39. Jolivet (1982: 447).

40. From J. Charbonnel, quoted in CNESQUISEPASSE? 74 (July 1993): 36–37.

41. ESA (1992a).

42. Austria, Norway, and Finland only joined in 1987. It would be a mistake to consider all these separate entities equal in terms of weight within the organization, and it is no accident that ESA's headquarters are in Paris (ESA 1987). France has always been particularly interested and invested in Ariane, providing over half its budget, as well as its launch site.

43. Desobeau (1990) and ESA (1992a). A 90 percent success rate is considered an essential benchmark in designating a rocket “reliable.” After one hundred launches the Ariane family was at 93 percent.

44. That is to say, about seven thousand kilograms as opposed to about four thousand kilograms. Like the earlier series, however, it uses a transfer orbit to boost the satellite rather than reaching high orbit directly (ESA 1992a). The Ariane 5 story lies outside the main body of this work, because it only became (disastrously) operational in 1996. In its various configurations, the Ariane 4 has been the main workhorse of CSG during the 1990s. At the end of 1998 the mature design of the Ariane 4 clocked in at 96.4 percent (three failures in eighty-six attempts). The Ariane 5, with one failure in three tries, posted a less stellar 66.7 percent and has a long climb back to its promise of increased reliability. See International Space Industry Report 3, no. 1 (4 January 1999): 22.

45. The factor for indirect employment increases for the period between 1975 and 1985, from 1.9 to 2.3, and francs spent on the project increasingly stayed within the department (Rémondière and Colmenero-Cruz, 1987).

46. Mam-Lam-Fouck 1992; Rémondière and Colmenero-Cruz 1987; CNES 1988; and INSEE 1995. With the dramatic demographic expansion of the department, the new projects related to the space center were estimated in 1987 to represent a smaller percentage of the total investment, falling by half to around 27 percent. More recent estimates of the economic weight of the space program made in 1995 by the French bureau of statistics, however, suggest that it accounts for a startling amount of all formal economic activity in French Guiana: 49.8–57.2 percent of production, 28.3–33.7 percent of added value, 28.3–35.4 percent of revenue, 26.7–33.7 percent of employment, 19.9–25.7 percent of the population supported, 59.1–72.9 percent of imports, 20.3–27.7 percent of customs revenues paid, and 41–49.8 percent of local taxes (INSEE 1995: 107–9).

47. The Ariane 5 program has given birth to an even larger assemblage known as ELA 3.

48. The third stage of Ariane 4 has also proved the trickiest, being implicated in a number of launch failures. Ariane 5 has a cryogenic main stage powered by liquid oxygen and hydrogen and two large solid fuel boosters strapped to each side (Arianespace 1993; ESA 1992a).

49. Isakowitz (1991). Different arrangements of boosters, either liquid or solid fuel, are added for extra thrust depending on the weight of the particular load.

50. ESA (1988; 1992a); Isakowitz (1991); and Soop (1994).

51. The gold standard comes from A. Miles in the summer of 1993, as based on a price of gold at roughly 360 dollars per ounce versus a two-thousand-kilogram Hughes 601 satellite at about two hundred million dollars (or over nine thousand dollars per ounce), a factor of around twenty-five. For general estimates, see Blonstein (1987).

52. CNES, Palais de la Découvert, and SEP (1991).

53. From M.J. Jamet in CNES (1972: 333). Translation as given in abstract.

54. CNES (1988); CSG (1994); and ESA (1991).

55. CNES (1990). Compared to a similar breakdown for CNES as a whole, this distribution displays a lower proportion of engineers to technicians and laborers, as well as a surprisingly high proportion of administrators and secretaries.

56. The majority of influential administrators of the space center come out of the elite grande école system of postsecondary education. This fits with a wider French educational and employment pattern. As the French embassy explains in an official publication: “In France, engineers occupy the greatest number of top management posts, competing with other specialists, such as economists and business school graduates, for the highest positions in government and industry. Prestige and salary put the French engineer near the top of the social ladder. The vast majority of French engineers are trained in prestigious institutions called grandes écoles, often separate and independent from the university system, sometimes with administrative ties, but always jealous of their autonomy” (Embassy of France 1989). Alfred Dreyfus, we should not forget, was himself a graduate of the most prestigious of the technical grandes écoles, the Ecole Polytechnique. For a profile of an exceptional local space technician, see “Paul Henri: Un Amerindien dans le siecle,” Latitude 5 18 (October 1992): 29–31.

57. On the Challenger accident, see Jensen (1996) and Vaughan (1996).

1. The charisma of rockets is perhaps best encapsulated in the story of a “conversion by launch” of a black activist at Cape Canaveral, who arrived to protest Apollo 11 and left a believer (McDougall 1985a: 412).

2. CNES and ESA (1994a).

3. See D. Carter (1988: 177).

4. McDougall (1985a: 413).

5. See interview with J.-D. Levi in CNESQUISEPASSE? 67 (October 1991): 4–7. Mailer (1970), Pynchon (1973), and Wolfe (1979) offer literary–and often insightful—versions of the Space Age. The collected presentations in ANAE (1992) and Schneider and Léger-Orine (1987) provide a relevant sampling of lofty views on European Space.

6. ESA (1989b).

7. Like most ESA documentation, this brochure is available in both English and French, the two official languages of the organization; I am quoting directly from the official English version, as I have done whenever possible.

8. Wolfe (1979).

9. For example, we have the following passage on the technical importance of “the vast natural laboratory” of space (ESA 1989a): “Scientists have two main reasons for wanting to get their instruments into space. First, they very naturally want to explore virgin territory accessible to their space vehicles in the same way that Christopher Columbus wanted to explore the new world. … The second main reason for getting into space is to escape the obscuring effects of the Earth's atmosphere and to look at the stars in all their glory.” See

10. This account of the naming appears in an ESA publication (1992a) and can thus be taken as an official memory of sorts, however inflected.

11. ESA (1992a). As a quip of questionable—but representative—taste had it, the feminine name was appropriate because both rockets and women “require lengthy preparation and deliver brief pleasure.” For several variants and embellishments of the christening, see CNESQUISEPASSE? 51 (May 1988): 3–4. Although there is no mention of why Ganymede lost favor, one can imagine the complications a rocket recalling male homosexual desire would pose for an organization like ESA.

12. ESA (1988).

13. Ibid.

14. For Arianespace figures, see Arianespace (1994). All told, six Apollo flights (11, 12, 14, 15 and 17) made the voyage between 1969 and 1979, and a dozen astronauts (white male explorers all) bounded about the lunar surface. In discussing a possible return to the moon, the French weekly Le Point compares this brief romp to that of an undisciplined baboon (Ponchelet 1992: 79). A 1991 map distributed by the Hughes Corporation reveals the degree of crowding along some segments of the geosynchronous orbit, namely those that serve North America, Europe, and Asia.

15. Ariadne remains on Naxos, either dead, pregnant, or simply abandoned. In some versions she marries Dionysus, and in others she dies (Graves 1960: 339–48).

16. Private collection of author.

17. Embassy of France (1989).

18. Gordon Wright (1987: 442).

19. Hubert Curien in Naddeo-Souriau (1986: 7).

20. See M. Hauzeur, “L'Europeanisation du CSG,” Latitude 5 16 (April 1992): 4–5. Francophone anxiety runs deep within CNES as a whole, because English is by far the dominant technical language of space. See the call for the use of French in scientific settings in “La Langue Française et les réunions scientifiques internationales,” CNESQUISEPASSE? 74 (July 1993): 34–35. For the numbers of non-French European employees, see R. Borel, “CSG: L'Europe avant l'heure,” Latitude 5 16 (April 1992): 6–8. The count specifies fifty-four Italians, twenty-four Belgians, eighteen Germans, eight Spaniards, five Britons, one Irish citizen, and one Swiss, adding up to 111 rather than the 106 total given.

21. “Nous ne sommes pas prets pour l'Europe,” Latitude 5 17 (July 1992): 8.

22. In an ethnography of ESTEC, the ESA science center in the Netherlands, Stacia Zabusky (1995) describes the intricate choreography of space scientists and engineers negotiating nationality, expertise, and common goals. Although many of their dilemmas echo at CSG, the context of Kourou is far more French.

23. An estimate of the ethnic demographics of Kourou in 1991 ran as follows: 3,500 Metros, 6,500 Creoles (both Guyanais and from the French

24. For additional flavor of Kourou and French Guiana as experienced by Maroons, see Bilby (1990) and Price and Price (1992).

25. In 1993 some twenty thousand members of the general public visited the base, as well as sixteen hundred schoolchildren in 160 classes. See CNESQUISEPASSE? 78 (July 1994): 31. Given that CNES also owns the Iles du Salut, and that many visitors to French Guiana are drawn by friends and relatives working at CSG, the space program constitutes a significant actor in the local tourist industry (CCIG 1993).

26. ”Quand le ciel s'enracine dans la terre Guyane … La voie lactée et la cité des étoiles / La voie lactée, un projet pour l'espace d'une route.” Printed in France for the Direction Départmentale de l'Equipement. No date given, but its subject matter would place it between the late 1980s and the beginning of the 1990s.

27. In 1990 an ambitious seven-year project, Le Plan Phèdre, called for over one hundred million dollars of investment in the region, of which nearly half would come from CNES. An article in CNESQUISEPASSE? describes Kourou with the help of two photos, one of the “lively and pleasing” new Quartier Monnerville, with its resplendent fountains, and the other of the Village Saramaca—”that little African village of certain charm”—soon to be renovated and sanitized (CNESQUISEPASSE? 65 [April 1991]: 13). In the mid-1990s, new housing for the village Saramaca was finally under construction.

28. These comments stem from both fieldnotes and materials produced internally by the space center, such as in Latitude 5 17 (July 1992): 8. Because of the repetition involved I present them as an anonymous, self-reinforcing cultural voice.

29. “Un paradis pour les Enfants,” CNESQUISEPASSE? 65 (April 1991): 14.

30. M. Hauzeur in Latitude 5 21 (July 1993): 3–5.

31. Interview with M. Mignot, Latitude 5 15 (January 1992): 5–8.

32. M. Hauzeur in Latitude 5 21 (July 1993): 3–5.

33. The organization “Freelance” both shelters clients from inconvenience and seeks to make their stay pleasurable. See also chapter 7, as well as CNESQUISEPASSE? 60 (March 1990): 38–40.

34. Interview with A. Rémondière, director of CSG between 1986 and 1991, “Emotions Guyanaises,” CNESQUISEPASSE? 69 (April 1992): 31–32.

35. A cartoon published in the CSG employee magazine, Latitude 5, nicely parodies the director's encounter: A European man lost in the jungle asks three Amerindians: “Do you know the place where the White Man shoots the big stick that spits fire?” One of them replies: “Ah! Yes! No doubt you mean the Guiana Space Center. As well as our launcher Ariane IV. After our hunt, allow us to take you there. We have to go back to work there in a few hours to participate in the last controls before the launch.” Meanwhile, one of the other Amerindians thinks: “Another one fresh off the boat!” Latitude 5 5 (July 1989): 29.

36. The primary local newspaper of recent years, France Guyane, the product of a conservative chain, contained a special commemorative insert.

37. CSG and CLAMFUK (Club de Mini Fusées de Kourou) 1994. Lancement de la Fusée RG, 21 July 1994, Site Fusées Sondes.

38. See Latitude 5 25 (July 1994): 21–22 and France Guyane (12 August 1994): 6. For additional discussion of the islands relative to CSG, see Latitude 5 10 (October 1990): 10–12, and 14 (October 1991): 19–21. The anniversary grew even more surreal when Charles Dreyfus, Alfred's grandson, arrived on Devil's Island in the spring of 1995—his way paid by the space center (Miles 1995; also CBC interview on “As It Happens,” 2 May 1995).

39. “In fact there are more endangered and threatened species at the Kennedy Space Center than at any other refuge in the continental United States” (NASA 1992: 46). See also NASA (n.d.). The Kennedy Space Center hosts a large museum and tour operation and displays striking parallels (if on a different scale) with the Guiana Space Center in terms of practices of commemoration as well as landscape.

40. See Blumenberg (1987), Cosgrove (1994), and Ingold (1993).

41. See CNESQUISEPASSE? 65 (April 1991): 15–18 and Latitude 5 15 (January 1992): 21–23.

42. Verne left his readers in suspense, before returning with a sequel, Autour de la lune (Around the moon), several years later, in 1870 (like the first volume it was first published in the periodical Journal des Débats). See Verne (1958, 1966a, 1995).

43. Only the English version of Desobeau (1990) contains the line used here.

44. For a description of a launch viewed from ESA's main research and technology center in the Netherlands (it is broadcast at all centers of the European and French space worlds), see Zabusky (1995: 4). In French Guiana one has several options for observing Ariane live, including a general public observation site, any location near the coast with a clear horizon, and the television broadcast.

45. I do not mean to suggest that a launch site is the only example of modern technology involving place, or incorporating an open horizon. Nuclear test sites present interesting inverse parallels. See Gusterson (1996).

46. The quip works more richly in French, because one of the terms used for a rocket launch, tir, also describes the shooting of an arrow. While the artifacts in question are of the souvenir variety, primarily intended to authenticate a jungle experience, hunting with guns and the consumption of exotic game animals remain popular and controversial activities in French Guiana.

1. Mam-Lam-Fouck (1992: 306–7).

2. Orgeas (1885: 5)

3. As Clarence Glacken (1967) meticulously documents, concepts of the environment have had a long history in Western thought, maneuvering between philosophy and science. A concern for climate runs through the lineage of medical thought from ancient Greece to the emergence of disease theory.

4. See Kupperman (1984: 215), Kennedy (1981: 51), Curtin (1989), and W. Anderson (1992, 1996a, 1996b).

5. An example: On the fifteenth of March 1855, the governor writes to the minister reporting that out of a total of 492 convicts at Montagne d'Argent during the month of February, fifty-three were sick and ten had died, an improvement of 19.92 percent over the number ill during the same month the year before and 0.02 percent over the previous death rate. CAOM H bagne 14. For different views on the emergence and use of statistical reason, see Hacking (1990), Poovey (1998), and Porter (1995).

6. Montesquieu (1900: book 15, section 7: 240).

7. See W. Anderson (1996a) and Rabinow (1989: 129–34). It should be noted that French science came late and grudgingly to Darwinian versions of evolution, and that the story described here is presented in crude and summary form. For a richer account of nineteenth-century evolutionary debates, see Bowler (1989). For a more nuanced presentation of acclimation and opposition to it in French science, see Osborne (1994). For a theoretical frame through which to interpret the emergence of milieu, see Rabinow (1989) and Canguilhem (1989).

8. Osborne (1994: 83–97). Also see Gould (1981).

9. Cited in Osborne (1994: 94).

10. Examples given in the OED for cosmopolitan date to 1860 (Gosse) and 1875 (Lyell). Orgeas's use of the term, while archaic to the ear of contemporary cultural studies, is quite in keeping with the tradition of natural history. See also the Tresor de la langue française, 6: 254 (Paris, 1978), which provides the following gem from Teilhard de Chardin's 1955 Le Phénomène humain:“Zoologiquement considérée, l'Humanité nous présente le spectacle unique d'une ‘espèce’ capable de réaliser ce à quoi avait échoué toute autre espèce avant elle: non pas simplement être cosmopolite,—mais couvrir, sans se rompre, la Terre d'une seule membrane organisée [From a zoological perspective, humanity presents us with the spectacle of a ‘species’ capable of realizing that which has eluded all others before it: not simply being cosmopolitan—but covering the Earth with a single, unbroken organized membrane].”

11. Orgeas (1885: 20–27).

12. Ibid.: 63.

13. Ibid.: 122.

14. Orgeas (1886: 416).

15. Ibid.: 416–18.

16. From Bollot, “Un Pénitencier doit être une veritable maison de santé morale,” in CAOM H bagne 4. His statistics are drawn from the Notice Officiel and the paper Toulonnais. Because the other death rates for French Guiana do not factor out accidental deaths, I prefer to use these; however, it should be stressed that although the rates given in different sources vary slightly, depending on whether the official numbers are modified for accidents or compared to naval physicians' reports, they vary only slightly (see also Devèze 1965: 142; Orgeas 1885: 26). For uniformity between nineteenth- and twentieth-century sources, as well as the comfort of the average reader, I have translated later demographic statistics based on a unit of one thousand into simple percentages.

17. Based on statistics found in Mam-Lam-Fouck (1987: 31–32) as adopted from the Almanach de la Guyane française. Figures in the original are

18. The figures available for the 1834–1847 period only differentiate the population by civil status, slave versus free, rather than by racial category. During the earlier period of 1807–1828, freed slaves outnumbered white colonists, eventually by as much as two to one; thus it is likely that free blacks represent the larger statistical mass in the figures for nonslaves between 1834 and 1847. The number of white colonists recorded for 1842—the one year available—is 1,215, virtually unchanged from the 1,280 present in 1828, and only about a fifth of the total free population (Mam-Lam-Fouck 1987: 32).

19. Curtin (1989: 7–8).

20. Figures are drawn from Curtin (1989: 8), where original calculations are per thousand.

21. R. Price (1990: 300), original rates given per thousand.

22. Orgeas (1886: 311).

23. Crosby (1986).

24. Curtin (1989: 9–10)

25. Figures from Curtin (1989: 10).

26. “With the yellow fever danger gone, Jamaica began to encourage tourism to exploit the comfort of its winter climate and its new reputation for healthfulness” (Curtin 1989: 132). For a cogent summary of disease, race, and empire, see W. Anderson (1996a) and M. Harrison (1996). Also see Headrick (1981) for technical shifts associated with late-nineteenth-century European imperialism, and Rabinow (1989) on colonial planning and administration. For more on malaria, see Desowitz (1991) and Humphreys (1996), and for more on yellow fever, see Delaporte (1991). Needless to say, the story presented here is highly abbreviated.

27. L. Rousseau (1930: 355) gives the following death rates for hard labor convicts: 1924–7.77 percent; 1925–6.14 percent; 1926–7.54 percent; 1927–8.15 percent; 1928–10.30 percent; and reports that the rate for those sentenced under the recidivist laws was over 10 percent every year, rising to 11.45 percent in 1927.

28. (INSEE 1993: 30–31). The range in the official INSEE 1993 book of statistics is between 0.56 percent in 1987 and 0.41 percent in 1992. All original figures are given per thousand.

29. For an overview of interest in wilderness in the United States and historical patterns relative to urbanization, see R. Nash (1982).

30. As the late critic Raymond Williams writes, nature “is perhaps the most complex word in the language” (R. Williams 1983: 219). See also Arnold (1996), Evernden (1992: 18–35), Glacken (1967), Oelshlaeger (1991), K. Thomas (1983), R. Nash (1982), and Merchant (1990). Bess (1995) provides a survey of recent French attitudes about nature precisely in relation to high technology.

31. Guyane Excursions, “Le Maroni–Le Tapanahony (Descriptif),” collected 1994. Touristic fascination with the “original” peoples of French Guiana, the Amerindians and Maroons, can be read as reflecting a general Western obsession with romanticized origins (see, among others, Fabian 1983 on anthropology's role in this project).

32. The specific examples and phrases used in this discussion are taken from an article in the CNES magazine by Anne Paradis, “SOS Missionaires,” CNESQUISEPASSE? 60 (March 1990): 38–40.

33. Recall the visit of the director among the Wayana described in chapter 6. For more on tourism, see MacCannell (1989) and Graburn and Jafari (1991).

34. A tour group I accompanied in the summer of 1994 (fewer than a dozen people) managed to consume some seven liters of rum and twenty liters of wine over four relatively exertion-free days. Amid new conditions of controlled adventure, echoes of colonial degeneration remain.

35. For a classic example, see Cognat (1977); for a more recent sketch, see Bilby (1990). Lezy (1989) provides a history of exploration in the area as well as an account of his own attempts to replicate such efforts in the mid-1980s. For another survey of the typical tourist itinerary in French Guiana, see Doucet (1981). For more on art collecting and river trips in French Guiana, see Price and Price (1992).

36. See Le Pou d'Agouti 1 (June 1990): 8–9; also see ONF (1992).

37. Sanite (1988).

38. The periodical Le Pou d'Agouti, published by the organization of the same name, offers the most lively overview of environmental issues in French Guiana. Named for a local biting insect, the Pou does its best to serve as an ecological gadfly. Together with the WWF, it comprises the most vocal element of environmental politics in French Guiana; organizations such as SEPANGUY (Société d'Etude de Protection et d'Aménagement de la Nature en Guyane) and the LPO (Ligue pour la Protection des Oiseaux) tend to be a tad more sedate. In 1994 another publication appeared, Le Tamouchi, with views more in keeping with official accounts linking architecture, urbanism, and the environment.

39. Wood (1995).

40. R. Harrison (1992). In Tristes Tropiques,Lévi-Strauss (1973: 91) finds the alterity of the New World amid trees: “This forest is different from Western forests because of the contrast between foliage and trunk. The foliage is darker, its shades of green suggesting the mineral rather than the vegetable kingdom.”

41. Romier (1962: 11–12).

42. Laporte (1983: 175).

1. Vidal de Lingendes, “De la colonisation pénale,” 1845. CAOM H bagne 1.

2. Mam-Lam-Fouck (1987: 215).

3. This ethnic hegemony is such to have been coded into speech: in Guyane le chinois also means a corner market. The number of these establishments is quite remarkable, particularly in Cayenne.

4. For background on the position of climatic debates within the history of professional geography, see Livingstone (1992: 216–59). S. Hussein Alatas (1977) discusses the “lazy native” in the context of Southeast Asia; see also Certeau (1984), Ong (1987), and J. Scott (1990) on patterns of resistance to enforced labor. Sherry Ortner (1995: 173–93) provides cogent criticism of easy

5. Ellsworth Huntington (1876–1947) was a Yale geographer who both advocated restrictive immigration policies and served as president of the American Eugenics Society between 1934 and 1938. He was also involved in early work in human ecology (see Kingsland 1993). Civilization and Climate went through many printings; the 1924 version cited here is the third edition.

6. Huntington (1924: 30).

7. Ibid.: 220.

8. Ibid.: 411.

9. Huntington composed his map of civilization on the basis of evaluations provided by a number of “well-informed persons”—distinguished scholars and experienced travelers. He sent a total of 213 inquiries to twenty-seven countries and received 138 responses, of which fifty-four satisfied his requirements sufficiently to represent a contribution. In the end his contribution list included twenty-five Americans, eight Britons, eight “Teutonic Europeans,” six “Latin Europeans,” six “Asiatics,” one “Non-Teutonic and Non-Latin European” (a Russian), and no Latin Americans. Given the distribution, the resulting maps are rather predictable: England and the North Atlantic states score a perfect one hundred, closely followed by Germany and northern France at ninety-nine, whereas central Italy (outside of Florence) receives an eighty-five, Greece a seventy-two, central China a sixty-six, the United Provinces of India a fifty-three, New Guinea a fifteen, and the poor Kalahari Desert a mere twelve. The Guianas, French, British, and Dutch, together rate a collective thirty-four (ibid.: 240–74, 415–32). Among the Americans who participated we find the names of two anthropologists of note: Ale? s Hrdli? cka and Edward Sapir. Two others declined the invitation to contribute to his larger project. Alfred Kroeber wrote that he had to “frankly confess that I believe you will obtain misleading results.” Franz Boas put it even more strongly:

I feel … quite unable to comply with your request, for several reasons …. It has been my endeavor, in my anthropological studies, to follow … the same principles that are laid down for the natural sciences; and the first condition of progress is therefore to eliminate the element of subjective value; not that I wish to deny that there are values, but it seems to me necessary to eliminate the peculiar combination of the development of cultural forms and the intrusion of the idea of our estimate of their value, which has nothing to do with these forms. It seems to my mind that in doing so these obtain subjective values, which in themselves may be subject of interesting studies, but which do not give any answer to the question you are trying to solve.

Livingstone 1992: 226;Huntington 1924: 249, 415

10. Markham (1944: 20).

11. For a more recent and detailed history of air-conditioning in the United States, see Cooper (1998), who notes that early systems developed in industrial settings were intended to dehumidify as much as lower temperature (a significant point when considering the operation of delicate bits of technology as well as more massive forms of industrial production). Arsenault (1990) considers the cultural implications of the technology in the American South, as windows closed and porches emptied.

12. Markham (1944: 200–6). Cooper (1998) distinguishes between systems incorporated into building design (important to institutions and manufacture) and commercial mass production, also noting a strong countercurrent of open-air enthusiasm. Because the focus here is on the possibility of control rather than the history of the technology per se, I am simply following Markham's broader progressive account.

13. Markham (1994: 206).

14. Roussenq (1957: 55–57).

15. For an example from a very different climatic zone, see Solzhenitsyn (1963).

16. Certeau (1984: 34–39).

17. Ibid.: 30.

18. Revert (1949: 269–71).

19. See Rabinow (1989: 12–13; also 1994; 1996: 59–79). In “techno-cosmopolitanism” planners recognize history and nature; consequently, their designs—while modern and technical—retain a flavor of local specificity. In “middling modernism” planners abandon ties to a historical and natural milieu in favor of a social and technical one; consequently, their designs—also modern and technical—are pure and universal, designed for an abstract human subject. The essential difference between the two planning moments thus emerges in clear anthropological terms: a concern for particular groups as opposed to a concern for humankind. Rabinow's schema provides us with an additional vocabulary for development, particularly the norms of modernization. While the structures of “developing” society are less clearly drawn than those of the built environment, the generalized condition of “underdevelopment” and general efforts to act on the third world certainly replace history and nature with society and science. In this sense institutionalized development represents the final colonization of local conditions by universal principles, producing a new empire of experts. “Middling modernism's project was more audacious, seeking to create New Men freed, purified and liberated to pursue new forms of sociality which would inevitably arise from healthy spaces and forms” (Rabinow 1994: 403). For the emergence of such “New Men” worldwide, the space requiring redesign becomes the globe.

20. After the hundredth launch in 1997 the Ariane program (excluding the next-generation Ariane 5 rocket) had seven failures, giving it an overall success rate of 93 percent. Failed launches were 02, 05, 15, 18, 36, 63, and 70 (Reuters, Cayenne, personal communication).

21. Paradis (1993).

22. See the collection of anecdotes gathered by Joëlle Brami, in Brami (1989).

23. Ibid.

24. Ròt Kozé 45 (July/August 1994): 4.

25. Quoted in Le Pou d'Agouti 12 (March-April-May 1994): 29.

26. See A. Calmont (1987), EDF and CNEH (1993), and Jules (1994). In 1988, EDF reported a consumer base of some thirty thousand clients, drawing a total of forty-five megawatts. By 1992 that total had grown to seventy megawatts, and estimates for 1995 predicted a demand for one hundred megawatts. Operations surrounding Ariane 4 required twelve megawatts, a figure expected to increase to seventeen megawatts with the advent of Ariane 5. (EDF and CNEH 1993: 7). In relative terms, the growing energy use in French Guiana remains well below North American standards, if well above those of Africa (CCIG 1994: 53).

27. EDF and CNEH (1993) and Jules (1994: 29).

28. Some 2,700 million French francs. The dam design called for the use of rolled, compact concrete, a total length of 750 meters, and a reservoir of 315 square kilometers at 35 meters. The resulting structure can produce 116 megawatts, a relatively high 368 kilowatts per square kilometer of submerged surface (EDF 1993; Jules 1994: 35).

29. Jules (1994: 34–35). For comparison, see Hennigsgaard (1981) and Wali (1989).

30. Ayangma (1990: 4).

31. EDF (1992).

32. For example, the cover of one issue of the activist periodical Le Pou d'Agouti (no. 12, March/April 1994), transforms the initials EDF into “Enterprise de Destruction de la Forêt [Forest Destruction, Inc.].”

33. From Ròt Kozé 45 (July 1994): 2.

34. Quoted in B. Villeneuve, “La Dot du spatial,” France Guyane (16 July 1994).

35. Ròt Kozé 45 (July 1994): 2. Material for this and the following synopsis stems from direct observation as well as reports on local TV (RFO and ACG) and in France Guyane during the months of July and August 1994. Special acknowledgment must be given to the Reuter Bureau of Cayenne, which followed the road stories assiduously and shared public information most generously.

36. WWF Communiqué de Presse, Operation “Acajou,” dateline Versailles, 19 July 1994.

37. Intersyndicale des Socio-Professionels de Guyane. Communiqué de Presse, 21 July 1994.

38. See Parti Socialiste Guyanais Communiqué, 21 July 1994, and letter by L. Bertrand published in France Guyane (2 August 1994), as well as Ròt Kozé 45 (July 1994).

39. “SEDTP et intersyndicale répondent au WWF,” France Guyane (2 August 1994).

40. Le Pou d'Agouti 13 (July, August, September 1994): 19.

41. Despite all the fuss, state permission had only been granted for the first twenty-three kilometers of deforestation, and the eagerness of the French government to facilitate the flow of persons over the Brazilian border was far from clear. As the paper in Martinique reported, even some local officials had private reservations about the benefits of the road, and those with interests in the local

42. The “Debate Route Regina–St. Georges” was broadcast on ACG the evening of August 1, 1994. I both watched it live and later repeatedly reviewed a video recording. However, since a number of comments were inaudible (at some points several people were talking at once), and others repetitious, I have condensed sections and spliced once or twice, trying to trim excess while retaining adequate detail and an accurate portrayal of each position, as reflected in written statements made by each side.

1. See Latour (1993: 100–3). Here I am suggesting a slightly different symmetry of analysis than that Latour calls for, because the focus rests on the divide between “home” and “tropics” rather than on a repatriated discipline per se.

2. The strategy of incorporating the human and nonhuman symmetrically into studies of science and technology (as in Callon 1999, Latour 1993, and Haraway 1997) is an instructive balance to the Marxist legacy of focusing on production (see Harvey 1989 and Lefebvre 1991 for relevant discussions with spatial sensibility). With respect to concerns about mystification, I would recall the equal error of veiling the active presence of machines, animals, plants, dust, and wind behind terms such as “social relations.” Neither nature nor artifice is simply a passive prop in a human drama, even less so when the categorical distinctions between them grow increasingly blurred.

3. This is not to imply that yielding to design equals success and persisting in improvisation equals failure. Bricolage is ultimately the most functional of approaches in that it seeks practical solutions to immediate problems. But as Lévi-Strauss (1966: 17) indicated, it operates within a closed universe, whereas planning, however disastrous, projects itself outward.

4. The crucial line of Hegel: “Through this rediscovery of himself by himself, the bondsman realizes that it is precisely in his work wherein he seemed to have only an alienated existence that he acquires a mind of his own” (1977: 118–19). Alexandre Kojève summarizes the importance of craft to the Hegelian slave by describing work as a kind of bildung: “Therefore—once more—thanks to his Work, the Slave can change and become other than he is, that is, he can–finally—cease to be a Slave …. Thanks to his work, he can become other; and, thanks to his work, the World can become other” (Kojève 1969: 52, 41–55). See also Winner (1977: 188). Here we have the kernel of the Marxist tradition of liberatory labor, pitting the materially engaged worker against abstract capital. Unfortunately, the dynamics of high technology do not appear to

5. On colonial domination, see Fanon (1967, 1968), Memmi (1965), Mannoni (1990), and Nandy (1983). Fanon's (1968: 39–41) description of a divided Algeria, settler and native, provides a stark and powerful vision of Settler and Native against Marxist tradition.

6. Fanon (1967: 220–21).

7. See Arendt (1959). In Arendt's terms, Crusoe's realm is a private matter as much as a public one and can only become truly political once he is no longer alone. Thus his career on the island moves across each of her categories in turn, as Crusoe survives, builds, and finally governs.

8. Frederick Cooper and Ann Stoler (1997: 33–34) sound an appropriate cautionary note about not taking colonial categories and archives simply on their own word, or ignoring the dynamics of empire in attempting to distinguish the fluid, hybrid nature of the present. Here the image I would evoke is less that of a break than a layering, a further folding of a system over itself and subsequent rearrangements.

9. Even without an isolating shipwreck, the traveling expert inherits aspects of Crusoe's condition: restricted mastery and a habit of improvement. Not unlike the virtuous gentlemanly witness of science in seventeenth-century England (Shapin 1994), the twentieth-century international expert defines a kind of limitless neutrality within limited social terms. While the mobile space of travel opens to anyone with sufficient funds, authorizations, and inoculations, the status of natural modern (a transcendent success in biological and social mobility) remains restricted, marked primarily by neutral expertise and sensibilities of design and unmarked in formal terms by gender, race, or even a shred of culture. The very neutrality involved opens a door of universal possibility while obscuring the steps necessary for actors located within different fields of probability to cross the threshold. Members of historically marked groups, struggling against greater social gravity in the form of disparities in wealth and education, acceptance of appearance, psychology, and culture, must work harder to become neutral. They must also detach themselves from local expectations. The essence of a “career” remains motion and the ability to imagine a life alone, at home on standardized islands amid shifting landscapes. Thus while knowledge may always be locally constituted and applied at its end points in the world, its incorporation into professional expertise depends on abstraction and mobility.

10. Since the adoption of a microchip system in French credit card technology, cards from elsewhere in the world can pose problems and arouse suspicions. Early versions of the chips reportedly suffered in high humidity, and–as everywhere in the world—the equipment to read them sometimes fails to function. And yet, many a ride in a taxi collectif involves negotiation between a driver demanding a standard fare and a passenger seeking alternatives. Conflict only increases in encounters between official and unofficial realms: I once witnessed a long and painful interaction between a Maroon with a fantastically crumpled American hundred dollar bill and two skeptical Creole postal clerks reluctant to exchange it for French francs.

11. Tournier (1969, 1972).

12. Tournier (1969: 120).

13. For relevant criticism of development discourse, see Ferguson (1990) and Escobar (1995). The discussion of ages here can be extended further into the academic vocabulary of geography, periodization, and formal history and anthropology. In addition to Fabian (1983) and Said (1979), see Bentley (1996), Chakrabarty (1992), Cooper (1994), Coronil (1995), and Prakash (1995).

14. For sometimes hopeful renderings of the possibilities of a mobile present, see Appadurai (1996) and Haraway (1997). Gupta (1998) provides additional testimony that repeated failure itself can constitute an alternative modern tradition.

15. From Soeurs de Saint-Paul de Chartres, 250 ans en Guyane, STI Roma (no date but from contents circa 1978). The sisters of Saint Paul de Chatres ministered in the penal colony. The work opens on the third page with these lines.

16. Verne (1966b: 248). Verne would later expand this idea in a 1889 work Sans dessus dessous (see A. Martin 1990: 179–91; Verne 1995: 109).

17. The combined polar difference amounts to about forty-five meters. See King-Hele (1992: 178–81) for illustration and explanation, and Fischer (1995) for the human context of early work in geodesy. Thurston Clarke (1988: 14) provides a romantic summary of the essential facts:

But since the earth is an imperfect sphere, rotating around the poles and bulging in the middle, the equator, like a river, desert or mountain range, can only be exactly where it is: equidistant from the poles and perpendicular to the earth's axis, at 24,901.55 miles the longest circle that can be thrown around the earth. It divides the world into climatic and vegetative mirror images. On the equator at sea level, gravity is weakest, barometric pressure is lowest, and the earth spins fastest. To its north, winds circulate clockwise around zones of high pressure; to its south, counterclockwise. Where it crosses oceans, placid seas spin unpredictable hurricanes into the hemispheres; where it crosses land, predictable temperature and rainfall nurture life in sensational abundance and variety.