- Purchase/rental options available:
History of Political Economy Annual Supplement to Volume 33 (2001) 4-22
[Access article in PDF]
Economics and the History of Measurement
Theodore M. Porter
In a classic historical essay on quantification in physics, Thomas Kuhn remarked on the use by social scientists of Kelvin's famous dictum on the decisive importance of measurement. Had it not carried the authority of physics, he argued, social scientists at the University of Chicago would never have chiseled it into the stones of their new building. Kelvin's phrase was displayed in abbreviated form as follows: “When you cannot measure, your knowledge is meager and unsatisfactory.” Not inscribed was Frank Knight's gently satirical translation for social scientists: “If you cannot measure, measure anyhow” (Kuhn  1977, 178, 183; see also Merton, Sills, and Stigler 1984). Knight's version captures something fundamental about the motto that is a matter of less urgency for physicists, but reflects the particular ambitions and anxieties of the social sciences. The common presumption, implied also by the Chicago social scientists, was and remains that measurement and quantification were developed and proven first in physics and that the measurement activities of economists and psychologists have followed in this great scientific tradition. But this view fails to credit the practical purposes of measurement, which have been central to its history.
Themes from the Historiography
The understanding so often presumed of the arrangement of the sciences dates back to a formulation from about 1830 by Auguste Comte, whom [End Page 4] we have also to thank for the neologism sociology. He claimed as his discovery the doctrine that the sciences pass in turn through the three stages that lead to positivity. He listed the sciences in a hierarchical order, from abstract simplicity to concreteness and complexity: mathematics, astronomy, physics, chemistry, physiology, and sociology. Yet he firmly refused to privilege mathematics or measurement as motors of the progress of science. Sociology was to be a broadly historical discipline. Political economy and psychology had no place on his list—in part because he opposed their individualism, but more generally because he insisted on the need for a unified, integrative science of society rather than abstract, specialized ones.
Comte railed against a universal ethic of measurement, insisting that every science required its own special methods, which could never be reduced to those further up the hierarchy (Heilbron 1995). Far, then, from anticipating the campaign of the Chicago social scientists, Comte had defined an alternative ordering. Social science was more complex than the sciences of nature. It could, in a limited way, follow the model of physiology, whose decisive advance into the positive stage around 1800 followed from its rejection of mathematics and mechanics for a distinctively vitalistic perspective. Comte was far from alone in his philosophical tastes. From the late eighteenth century to the early twentieth, methodologically reflective pioneers of social science generally saw their charge not in terms of following the one true path of science toward mathematical exactitude but in terms of a consequential choice between mechanical and organic models, requiring more or less modification to be applied to society. Somewhat paradoxically, as we will see, measurement in social science was not decisively linked to physics. In political economy, especially, historicist and organic conceptions of the economy provided more encouragement to quantification than did abstract mechanical ones.
To pose the issue in terms of such broad conceptions, however, is to set out on the basis of grand and dubious assumptions about the character of scientific, including economic, measurement. It supposes that successful measurement is quintessentially the achievement of pure, basic, or disinterested science. In effect, it identifies theoretical understanding as the proper goal of science and values measurement for its contribution to the making of mathematical theories. The simplest version of this doctrine—that measurement is the first stage of mathematization, which [End Page 5] proceeds by abstracting to general principles from quantitative data—has long seemed naive to historically informed interpreters of science.
Among the most acute of such interpreters was Alexandre Koyré, the French philosophical historian whose works inspired the first...