Skip to main content

Inventing needs: expertise and water supply in late eighteenth- and early nineteenth-century Paris


The aim of this paper is to investigate the notion of need, in this case an entire city's global need for water. This was a notion invented by Paris technicians between 1760 and 1804 in the context of several water supply projects, notably two river diversion schemes, those of the Yvette and the Ourcq, where the concept was much discussed. Different ways of considering the question of need – such as water resources, consumption and use, whether present or future – were strongly related to engineers' or scientists' conceptions of their own work. State engineers claimed they could make objective estimates of future needs with no reference to either value judgements or political intentions, a position which made it possible to keep strong control over the decision in the name of the state. In contrast, a practically trained engineer working outside the state corps claimed that in the case of expertise about the future, estimates would depend strongly on political intentions, norms and ideals, so the government should first give an outline of its intended actions. The paper studies the differences between these two approaches to the concept of need, especially how they articulated knowledge about what is and knowledge about what ought to be, present and future. The paper ends by linking these differences to conceptions of what was supposed to be technical or political in such projects and of what role engineers intended to play in the decision-making process.

Hide All

L. Beaumont-Maillet, L'Eau à Paris, Paris, 1991; Ph. Cebron de l'Isle, L'Eau à Paris au 19e siècle, Paris, 1991. These authors also insist on the growth of population during the eighteenth century while noting little infrastructural change before the last decades of that century.

Even historians who avoid the comparison of need with water resources are stunned by the small quantity of water at the disposal of the inhabitants of Paris during the modern period. The various estimates of this quantity are commonly between four and ten litres per person per day for the ordinary uses of daily life. Even Daniel Roche in his History of Everyday Things, who presents the modern period as an age of scarce water and rightly relates this scarcity to a whole system of social values and habits, describes water in Paris as a luxury, and sees a ‘probable deterioration’ of the situation during the eighteenth century. This is in fact questionable, because of a significant improvement in distribution in the last decades of the century. D. Roche, Histoire des choses banales – Naissance de la consommation dans les sociétés traditionnelles, XVIIe–XIXe siècles, Paris, 1997, 173–4.

K. Alder, Engineering the Revolution: Arms and Enlightenment in France, 1763–1815, Princeton, 1997, 12.

S. Schaffer, ‘Fish and ships: models in the age of reason’, in Models: The Third Dimension of Science (ed. S. de Charadevian and N. Hopwood), Stanford, 2004, 71–105, 71. Compare, more generally, The Moral Authority of Nature, a book concerned with ‘how humans use nature [things as they are] to think about standards of the good, the beautiful, the just, and the valuable [things as they ought to be].’ L. Daston and F. Vidal, ‘Doing what comes naturally’, in The Moral Authority of Nature (ed. L. Daston and F. Vidal), Chicago, 2004, 1–20, 1.

P.-S. Girard, Recherches sur les eaux publiques de Paris, les distributions qui en ont été faites, et les divers projets qui ont été proposés pour en augmenter le volume, Paris, 1812; Beaumont-Maillet, op. cit. (1), 85; Roche, op. cit. (2), 167. The quantity flowing from these fountains experienced major seasonal fluctuations; a minimum of two thousand cubic metres in 1700 and a maximum of eight thousand cubic metres at the end of the century seems a sensible, though rough, estimate.

Beaumont-Maillet, op. cit. (1), 15. They were probably over 20,000, but this is difficult to assess since the first inventory of the Paris wells was made in 1870, with an estimate of 30,000.

For a description of the various activities of these carriers see Mémoire des frères Vachette, frimaire year 9 (December 1800), Archives nationales, Paris (subsequently AN), box F14–685. For estimates of the number of carriers, compare with the situation in 1820 (1,338 carriers by foot registered, while the distribution by pipe had significantly increased); Recherches statistiques sur la ville de Paris et le département de la Seine, Paris, 1823. If one takes ten thousand carriers by foot in 1800 (which would be a high estimate) and an average of twenty trips per day and per carrier, they had a capacity of six thousand cubic metres a day.

Jussieu, ‘Examen des causes qui ont altéré l'eau de la Seine, pendant la sécheresse de l'année 1731’, in Mémoires de l'Académie royale des sciences, Année 1733, Paris, 1735, 351–60.

Un ruisseau large et noir comme le Styx, épais et limoneux.’ L.-S. Mercier, Tableau de Paris, Paris, 1994, 848. Mercier also criticized the laundresses working on the river inside Paris, the trash and the sewers running into the Seine. This and all other translations are my own, unless otherwise noted.

10  A. A. Parmentier, Dissertation sur la nature des eaux de la Seine, avec quelques observations relatives aux propriétés physiques et économiques de l'eau en général, Paris, 1787. Parmentier did not ignore the pollution, which was obvious to all, but developed a theory where joint stream and air motion could produce a regeneration both of water and air; pollution thus became a necessary part of the complex production of good water and the Seine water, though muddy, should be preferred to transparent water, whose apparent purity could hide the worst qualities. But Parmentier recommended taking water at some distance from the bank and let it rest in a ‘fontaine’, a traditional vessel used in Paris homes to ‘purify’ the water, to eliminate floating and sinking materials. More generally, on water purity in the late eighteenth century, see F. Graber, ‘La Délibération technique – Disputes d'ingénieurs des Ponts et Chaussées sous le Consulat – L'Affaire du canal de l'Ourcq’, thèse, EHESS, Paris, 2004, 259–310. On the quarrel between the ‘Sequanistes’ and the ‘Anti-sequanistes’, the pros and cons of Seine water, see Beaumont-Maillet, op. cit. (1), 23.

11  For an account of these projects see Girard, op. cit. (5), 60–108.

12  On the respective roles of monarchy and municipality in the late eighteenth-century water supply projects for Paris see D. Massounie, ‘Monuments hydrauliques urbains: aqueducs, châteaux d'eau et fontaines dans la ville moderne, 1661–1791’, thèse, University Paris 1, 2000, 18–87. Massounie (ibid., 27), notes that the multiplication of private projects in the second half of the eighteenth century rather benefited the king's authority, for all profit-oriented water supply projects had to obtain a privilège d'exploitation by the royal authorities.

13  On both the pumps and the Yvette project see Beaumont-Maillet, op. cit. (1), 96–115; J. Bouchary, L'Eau à Paris à la fin du 18e siècle – La Compagnie des eaux de Paris et l'entreprise de l'Yvette, Paris, 1946; Girard, op. cit. (5), 66–100. After the failure of the Périer brothers, mainly due to over-speculation on their shares, a former artillery officer, Fer de la Nouerre, proposed building the Yvette canal without surfacing, thus significantly reducing costs. Work began in 1788 but quickly came to a halt due to protests by residents over questions of property and subsistence since the project also intended to divert the Bièvre, a river very important for the manufacturing activities of the Faubourg Saint-Marceau.

14  Antoine Deparcieux was a mechanic and mathematician, creator of hydraulic machines for several noblemen.

15  Girard, op. cit. (5), 63–70.

16  The traditional unit for flow measurements, le pouce d'eau, literally ‘inch of water’, is about nineteen cubic metres per day. Cebron de l'Isle, op. cit. (1), 75.

17  A. Deparcieux, ‘Mémoire sur la possibilité d'amener à Paris, à la même hauteur à laquelle y arrivent les eaux d'Arcueil, mille à douze cents pouces d'eau’, in Histoire de l'Académie Royale des Sciences, année 1762, Paris, 1764, 342.

18  ‘Citerne’, in Encyclopédie, ou Dictionnaire raisonné des sciences, des arts et des métiers, 17 vols., Paris, 1753, iii, 487. Jaucourt borrowed the entire example from a 1703 mémoire by the astronomer La Hire (‘Sur l'eau de la pluie et sur l'origine des Fontaines; avec quelques particularités sur la construction des citernes’, in Mémoires de l'Académie Royale des Sciences, année 1703, Paris, 1720, 68).

19  Water, however, was not a public utility in the eighteenth century. A first step towards such a conception of water was made by Bonaparte's government, which argued that Paris water should be the government's property since people could not be submitted to the power of private companies in such an important subject. Graber, op. cit. (10), 116.

20  The project of the Périer brothers, for example, was even more uncompromising towards the water carriers because it intended to bring water directly to houses.

21  More precisely, these two entrepreneurs had taken the idea from another, Brullée, who first proposed an Ourcq project in 1785. Girard, op. cit. (5), 107–10.

22  This was not an uncommon attitude among entrepreneurs who seem to have had greater difficulties in formulating numerical estimates of needs. When they knew the classical estimate of Deparcieux, they usually repeated it, but mostly they simply proposed a dramatic description of the difficulties in getting water. See for example Mémoire des frères Vachette, frimaire year 9 (December 1800), AN box F14–685.

23  On Bonaparte's water policy see Graber, op. cit. (10), 113–18.

24  The state did not ask the Ponts-et-Chaussées in this case, since this corps had no special expertise on water supply and waterways at the time. On Bralle's interesting career see his administrative file, AN box F14–21802, especially Bralle to the Conseiller d'Etat chargé des Ponts et Chaussées, 20 October 1807. In 1807, when the various Paris water services were unified, he was at last incorporated in the Ponts-et-Chaussées corps.

25  On the Machine de Marly see T. Brandstetter, ‘“The most wonderful piece of machinery the world can boast of”: the water-works at Marly, 1680–1830’, History and Technology (2005), 21, 205–20.

26  Bralle to Laplace (minister of the interior at the time), 19 frimaire year 8 (10 December 1799), AN box F14–685.

27 Pour bien remplir les vœux du ministre j'aurais eu besoin de connaître quelles peuvent être les intentions du gouvernement à cet égard, c'est-à-dire, s'il veut se borner au strict nécessaire, ou si convaincu des avantages … d'une abondante distribution d'eau … il se déterminait à profiter de la construction du canal dont il s'agit pour en amener la plus grande quantité possible; manquant de renseignements positifs qui m'eussent été nécessaires, je prendrai pour base de mes évaluations un terme également éloigné de la prodigalité et de la parcimonie’. Bralle, op. cit. (26).

28  On the question of expertise in late eighteenth-century Europe see E. J. Engstrom, V. Hess and U. Thoms (eds.), Figurationen des Experten – Ambivalenzen der wissenschaftlichen Expertise im ausgehenden 18. und frühen 19. Jahrhundert, Frankfurt am Main, 2005, especially 7–17; H. Vérin, ‘Expertise et Etat fondé en raison au XVIIIe siècle’, in L'Expertise (ed. R. Damien), Besançon, 2001. A recent contribution to the issue of science and the state is S. Jasanoff (ed.), States of Knowledge: The Co-production of Science and Social Order, London, 2004.

29  Aristote, Rhétorique, Book 1, Paris, 1998, 30–3.

30 La consommation d'eau est sans doute beaucoup moindre aujourd'hui parce que la disette rend économe; mais lorsque l'abondance écartera les difficultés qu'on éprouve à se procurer de l'eau dans les quartiers éloignés de la rivière, les besoins se multiplieront et il est sage de prévoir qu'alors la consommation pourra atteindre l’évaluation ci-dessus.’ Bralle, op. cit. (26).

31  Horses were very important water consumers: 20,000 to 30,000 in number consumed twenty to thirty litres a day each. Roche, op. cit. (2), 159.

32  There was a general danger in presenting different possibilities, since this could reveal the arbitrary character of a project or lead to difficulties in comparing different projects. See Graber, op. cit. (10), 567–9.

33  On the Ponts-et-Chaussées corps see A. Picon, L'Invention de l'ingénieur moderne – L'Ecole des Ponts et Chaussées, 1747–1851, Paris, 1992; J. Petot, Histoire de l'administration des Ponts et Chaussées, 1599–1815, Paris, 1958; Graber, op. cit. (10).

34  Picon, op. cit. (33), 209–15; Graber, op. cit. (10), 36.

35  Graber, op. cit. (10), Chapters 5 to 7.

36 Il est temps de ne plus s'abuser sur cette évaluation.’ L. Bruyère, Rapport du 9 floréal an 10 sur les moyens de fournir l'eau nécessaire à la ville de Paris, Paris, 1804, 13.

37  Here ‘objective’ is not an actor's category. I use the word in the sense of a pretence to deal with facts and truth.

38 On a été dans l'habitude de supposer’. Bruyère, op. cit. (36), 13.

39 Il est de fait, et chacun peut s'en convaincre, qu'une famille composée de 10 individus, ne consomme moyennement que … environ 69 litres.’ Bruyère, op. cit. (36), 13.

40  One can link this attitude with a conception of the world centred on stability and ruled by statics, characteristic for the French Enlightenment around figures such as Condillac, Condorcet, Lagrange or Laplace and around concepts of balance or equilibrium, where the time factor was explicitly eliminated from all analysis. But at the turn of the century some individuals like Bralle and Girard were beginning to introduce a more dynamic conception of the world. On statics as typical of French Enlightenment see M. N. Wise (with the collaboration of Smith, C.), ‘Work and waste: political economy and natural philosophy in nineteenth century Britain (I)’, History of Science (1989), 27, 263301.

41 L'agrément général’, Bruyère, op. cit. (36), 13.

42 On n'entreprend pas de déterminer une telle quantité.’ ‘Il serait sans doute fort beau de voir des eaux jaillir et couler en abondance dans toutes les places publiques; mais avant de se livrer à un grand luxe, il faut d'abord pourvoir à l'objet le plus nécessaire.’ Bruyère, op. cit. (36), 13.

43  On P.-S. Girard and on the Ourcq canal affair more generally see Graber, op. cit. (10).

44  See, for example, and among many others, the responses of the engineers of the Ponts-et-Chaussées assembly to the consulting request of the Conseiller général, in January–April 1804, Bibliothèque historique de la ville de Paris (subsequently BhvP), manuscripts M 1177.

45  P.-S. Girard, Mémoire sur une modification proposée au projet général du canal de l'Ourcq, 27 germinal year 12 (17 April 1804), BhvP M 1177.

46  Bertrand, Réflexions sur le canal de l'Ourcq, 3 pluviôse year 12 (24 January 1804), BhvP M 1177.

47  Independently of the accuracy of the various apparatus, the main problem appears to have been the choice of the right moment, depending on the regularity of the weather, the season and the artificial activity of the river to be measured. On flow measurement techniques and debates among these engineers see Graber, op. cit. (10), 425–65.

48  Liard, Notes relatives à la dérivation de l'Ourcq, 19 germinal year 12 (9 April 1804), BhvP M 1177.

49  It was, for instance, important to argue that the need was as high as the whole river flow, if one wanted to divert the Ourcq entirely, which was an important argument in favour of the navigability of the new canal.

50  The engineer Lecreulx could even say, ‘In vain has the law ordered the construction [of the canal], its impossibility due to the nature of the ground made this law necessarily powerless.’ ‘En vain la loi avait ordonné l'exécution [du canal], son impossibilité qui tenait à la nature du terrain rendait nécessairement cette loi sans pouvoir et sans application exécutable.’ Lecreulx, Canal de l'Ourcq, 2 pluviôse year 12 (23 January 1804), BhvP M 1177. During the Ourcq debates the Ponts-et-Chaussées would develop an aggressive tone towards the government, mainly explicable by the relative absence of administrators' authority over a project launched by Bonaparte, therefore strongly linked to the emperor's personal discretion but left unattended during his increasing absence in wartime.

51  Besides the work of Picon, see for instance T. M. Porter, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life, Princeton, 1996, 114–47. Porter's main thesis, applied to the later nineteenth century but generalizable, is that the Ponts-et-Chaussées did not need to ground their decisions on quantified knowledge because they were powerful enough to impose them. This thesis can be questioned, because these engineers had to fight for their position and competences and because it is a negative argument: one can find many reasons why they did not need to ground their decisions on quantification alone. See Graber, op. cit. (10), 574–80.

52 L'universalité des circonstances.’ Register of the settings of the assembly of the Ponts-et-Chaussées, 15 March 1801, AN box F14* – 10910.

53  The extremely polemical cases of the Consulate, like the Saint Quentin canal or the Ourcq canal, can be considered both as a result of this autonomy and as a reason for its revocation. For a more general discussion of the complex relationship between technical administration and state see Graber, op. cit. (10), 119–20 and 136–54.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

The British Journal for the History of Science
  • ISSN: 0007-0874
  • EISSN: 1474-001X
  • URL: /core/journals/british-journal-for-the-history-of-science
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 1
Total number of PDF views: 20 *
Loading metrics...

Abstract views

Total abstract views: 125 *
Loading metrics...

* Views captured on Cambridge Core between September 2016 - 22nd March 2018. This data will be updated every 24 hours.