The Air-Propeller Tests of W. F. Durand and E. P. Lesley: A Case Study in Technological Methodology Author(s): Walter G. Vincenti Source: Technology and Culture, Vol. 20, No. 4 (Oct., 1979), pp. 712-751 Published by: Johns Hopkins University Press and the Society for the History of Technology Stable URL: http://www.jstor.org/stable/3103637 Accessed: 08-02-2016 19:59 UTC REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.org/stable/3103637?seq=1&cid=pdf-reference#references_tab_contents You may need to log in to JSTOR to access the linked references. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Johns Hopkins University Press and Society for the History of Technology are collaborating with JSTOR to digitize, preserve and extend access to Technology and Culture. http://www.jstor.org This content downloaded from 130.64.199.177 on Mon, 08 Feb 2016 19:59:59 UTC All use subject to JSTOR Terms and Conditions The Air-PropellerTests of W. F. Durand and E. P. Lesley:A Case Study in TechnologicalMethodology WALTER G. VINCENTI From 1916 to 1926, William F. Durand and Everett P. Lesley, pro- fessors of mechanical engineering at Stanford University, made a study of aircraft propellers that was described by a distinguished con- temporary as "the most perfect and complete . ever published."' This primarily experimental study supplied much of the data used by airplane designers in the 1920s to select the best propeller for their new designs. It also helped set the nature of American aeronautical research in the decades before World War II. It is therefore impor- tant for the history of aeronautics. More important, the Durand- Lesley work involved in a comprehensive way a methodology widely used by engineers but little studied by historians. Examination of the work in terms of this methodology provides a wealth of evidence on a number of fundamental problems in the history of technology gener- ally. This paper will first describe the background, events, and proce- dures of the Durand-Lesley study in the detail needed to exhibit the methodology. It will then analyze the methodology in light of the evidence and with an eye toward the broader problems. The foremost of these problems concern the nature of research in technology. They deal with such matters as the objectives of technological research, the peculiarities of research methodologies in MR. VINCENTI, an aeronautical engineer who has worked for many years in the Stanford University School of Engineering, currently teaches in the Stanford Program in Values, Technology, and Society (VTS). He writes: "The list of people who have helped in one way or another is too long to include. They will know who they are, and I thank them all. I am especially grateful, however, to Russell Robinson and the late Walter Bonney, onetime fellow employees of NACA, to Edwin Layton of the University of Minnesota, to Otto Mayr of the Smithsonian Institution, and to my present and past colleagues in the Stanford University Program in Values, Technology, and Society: Edwin Good, Stephen Kline, Robert McGinn, Nathan Rosenberg, and Howard Rosen." 1M. M. Munk, "Analysis ofW. F. Durand's and E. P. Lesley's Propeller Tests," Report No. 175, Annual Report of the National AdvisoryCommitteefor Aeronautics, 1923 (Washing- ton, D.C., 1924), p. 291. These volumes are hereafter cited as Annual ReportNACA. ? 1979 by the Society for the History of Technology. 0040- 165X/79/2004-0002$03.05 712 This content downloaded from 130.64.199.177 on Mon, 08 Feb 2016 19:59:59 UTC All use subject to JSTOR Terms and Conditions Air-Propeller Tests 713 technology, and the differences between research in technology and the basic sciences. We want to know if peculiarly technological methodologies exist, if so what their features are, and what objectives they have been created to satisfy. The limited attention given to these matters has been mostly theoretical and abstract;2 the Durand-Lesley study provides an opportunity for analysis in terms of specific events. We shall find that the technological methods in evidence differ in both form and object from those in the physical sciences. Our findings thus bear on the growing debate about the relation of technology and science and the inadequacies in the current models of technology- science interaction.3 A number of related secondary themes also appear. They include the determining influence of design in technology and technological research, the nature of technological knowledge, the convergence of knowledge and methodology in different fields of technology (spe- cifically, marine engineering and aeronautics), and the role of in- cremental and inconspicuous change in technological advance (as op- posed to discontinuous and dramatic invention).4 A virtue of a com- prehensive case study is that it can speak to a wide range of historical and philosophical concerns. Backgroundand Contentof the Methodology The technological methodology appears clearly-the first pro- minent example-in the work of John Smeaton early in the Industrial Revolution in Britain. His influential study of the performance of 2For examples by philosophers, see F. Rapp, "Technology and Natural Science-a Methodological Investigation," Contributionsto a Philosophyof Technology(Boston, 1974), pp. 93-114; and H. Skolimowski, "The Structure of Thinking in Technology," ibid., pp. 72-85, also in Technologyand Culture 7 (1966): 371-83. 3Much of this debate has also been in the abstract. For a number of recent papers dealing with specific instances, however, see N. Reingold and A. Molella, eds., "The Interaction of Science and Technology in the Industrial Age-Proceedings of the Burndy Library Conference, March 23-24, 1973," Technologyand Culture 17 (1976): 621-742. 4For the influence of design and the nature of technological knowledge, see E. T. Layton, Jr., "American Ideologies of Science and Engineering," Technologyand Culture 17 (1976): 688-701, and "Technology as Knowledge," ibid., 15 (1974): 31-41. The term "technological convergence" was introduced by Nathan Rosenberg to describe the ap- plication of common productive processes in different industrial technologies, but it seems useful also for knowledge and methodology (see N. Rosenberg, "Technological Change in the Machine Tool Industry, 1840-1910," Journal of Economic History 23 [1963]: 414-46). For the role incremental change, see A. P. Usher, "Technical Change and Capital Formation," reprinted in The Economics of Technological Change, ed. N. Rosenberg (Harmondsworth, Middlesex, 1971), p. 63. See also G. H. Daniels, "The Big Questions in the History of American Technology," Technologyand Culture 11 (1970): 1-21. This content downloaded from 130.64.199.177 on Mon, 08 Feb 2016 19:59:59 UTC All use subject to JSTOR Terms and Conditions 714 Walter G. Vincenti waterwheels and windmills, presented before the Royal Society in 1759, contained the two main methodological components: (1) a sys- tematic method of experiment and (2) the use of working scale mod- els.5 A full history of these matters has not been written and is hardly possible here; a brief survey, however, will help establish ideas needed later. Since Smeaton's time the two components, usually (but not always) together, have formed the basis of an autonomous tradition of engineering research. Smeaton's method of experiment, exemplified in his tests of model waterwheels, was to alter separately the conditions of operation of the wheel (speed and quantity of flow of water and speed of rotation of wheel) and measure the output of power. He thus followed what today may be called the method of "parameter variation,"6 which can be defined as the procedure of repeatedly determining the performance of some material, process, or device while systematically varying the parameters that define the object of interest or its conditions of opera- tion. Apart from earlier use in science, the method goes back in en- gineering at least to the ancient Greek catapult designers who established the best proportions for their devices "by systematically altering the sizes of the various parts of the catapult and testing the results."7 In addition to his tests of waterwheels and windmills, Smeaton employed parameter variation to arrive at the best composi- 5J. Smeaton, "An Experimental Enquiry concerning the Natural Powers of Water and Wind to Turn Mills, and Other Machines, Depending on a Circular Motion," PhilosophicalTransactions of the Royal Society 5 1, pt. 1 ( 1759): 100-174. For discussion and assessment of Smeaton's work, see D. S. L. Cardwell, Turning Points in WesternTechnol- ogy (New York, 1972), pp. 79-84; A. Pacey, The Maze of Ingenuity (London, 1974), pp. 205-15; and N. Smith, Man and Water (London, 1975), pp. 153-58. For an illuminating discussion of Smeaton's waterwheel experiments (including matters referred to later in notes 7, 102, and 104), as well as of the more specialized and less well known tests of his contemporary, Antoine de Parcieux, see the recent article by Terry S. Reynolds, "Scientific Influences on Technology: The Case of the Overshot Waterwheel, 1752- 1754, Technologyand Culture 20 (1979): 270-95. 6This terminology is my own. The method is so much taken for granted by engineers that they rarely call it by name. In the theory of the statistical design of experiments that has matured since World War II, the procedure in the elementary form to be described here is called "factorial experiment." For general purposes, however, parameter varia- tion seems more descriptive and closer to other engineering terminology.