Geoffrey Hodgson's website
  Home     Career     Research     Books     Articles     Blogs     Citations     Debates     Webnotes     Interviews     Editorships     Lectures     Emailing List     Personal     Links     Site Map   


   Interpreting Veblen:

   Positivism, Metaphysics, Facts, and Values







Several prevailing interpretations of Veblen are incorrect. For example, he is sometimes depicted as theorist who focused on macro and technological issues, to the detriment of the individual, in part by excluding individual intentions and purposes from his analysis. These misinterpretations are patently false, as demonstrated in the following works and by multiple references therein to Veblen's own writing:

Camic, Charles and Hodgson, Geoffrey M. (eds) (2011) Essential Writings of Thorstein Veblen (London and New York: Routledge).

Hodgson, Geoffrey M. (2004) The Evolution of Institutional Economics: Agency, Structure and Darwinism in American Institutionalism (London and New York: Routledge).

Recently I cam across another case of misinterpretation, of both Veblen and myself. In their 2011 book, Political Ideas of Thorstein Veblen (Yale University Press), Sidney Plotkin and Rick Tilman (p. 220) attack the notion that "Veblen is to be taken at is word as a dispassionate, value-free scientist, a positivist, in effect." They then claim that this view of Veblen as "a dispassionate, value-free scientist, a positivist" is "adopted in the scholarship of Geoffrey Hodgson." But they give no reference to, or quote from, my own work to support this claim. In almost every feature it is inaccurate.

It is clearly evident that Veblen was not a "dispassionate" or "value-free" writer and I have never claimed that he was. Furthermore, he was not a positivist, by any reasonable and historically-grounded definition of that term.

Positivism and Metaphysics

"Positivism” is a widely-used word that has acquired several meanings. In its original, Comtean, sense it meant the rejection of metaphysics, and a view that science operates by the collection of evidence, with the supreme goal of prediction rather than causal explanation. Many accounts of positivism also stress its rejection of value judgments from science.

But in contrast to Auguste Comte, Veblen believed that metaphysical (ontological) assumptions were unavoidable. He explicitly countered the positivist rejection of metaphysics and asserted the importance of causal explanation. He rejected the view that science could be founded on experience or experiment alone, without additional presuppositions that themselves cannot be grounded upon observation. Veblen (1900a, p. 241 [Camic & Hodgson, p. 226]) argued that the “ultimate term or ground of knowledge is always of a metaphysical character.” For Veblen (1900a, p. 253 [Camic & Hodgson, p. 233]), “a point of view must be chosen” and consequently the “endeavor to avoid all metaphysical premises fails here as everywhere.” For Veblen, unlike the positivists, “metaphysical” was not a term of abuse. Veblen rightly held that some “metaphysical presuppositions” were necessary and unavoidable for science. This view is widely accepted among philosophers of science today.

Also against Comte, Veblen asserted the need for science to impute causal relations. David Hume had rightly pointed out that no causal connection can itself be observed. The imputation of causal connections must always involve preconceptions by the analyst, and such imputations cannot be derived from experience or data alone. Hence, for Comte, any investigation into causes is biased and futile. Taking a contrary view, Veblen (1906a, p. 597 [Camic & Hodgson, p. 363]) appropriately identified the “preconception of causation” as necessary for “the actual work of scientific inquiry.” As Veblen (1908d, p. 397 n.[Camic & Hodgson, p. 479]) elaborated:

Causal sequence … is of course a matter of metaphysical imputation. It is not a fact of observation, and cannot be asserted of the facts of observation except as a trait imputed to them. It is so imputed, by scientists and others, as a matter of logical necessity, as a basis of systematic knowledge of the facts of observation.

In The Instinct of Workmanship Veblen (1914, p. 260) repeated, in a similar vein, again redolent of Hume that: “The principle, or ‘law,’ of causation is a metaphysical postulate; in the sense that such a fact as causation is unproved and unprovable. No man has ever observed a case of causation.” Veblen (1904b, p. 67 [Camic and Hodgson, p. 202]) himself declared “a habit of apprehending and explaining facts in terms of material cause and effect.” This involved a materialist ontology: “Its metaphysics is materialism and its point of view is that of causal sequence.” These statements place Veblen at considerable distance from the original (Comtean) positivism.

Facts and Values

Whatever he claimed, Veblen’s writing is hardly "dispassionate." He used many ethically loaded words like “waste” and “sabotage.” In the Leisure Class he used the extreme word “invidious” several times. But he then claimed that “there is no intention to extol or depreciate, or to commend or deplore any of the phenomena which the word is used to characterise” (Leisure Class, 1899, p. 34). Veblen may have been ironic here, but there is no evidence that his irony masked a hidden belief that ideology and science are indistinguishable. Instead, Veblen was satirizing the many apologetic scientists of the Victorian era, who habitually invested their “scientific” work with claims that it endorsed the established order.


Veblen did not claim that judgments of value and judgments of fact are the same thing. He might have accepted that value judgments are unavoidable in science. But that does not mean that judgments of fact and value can be conflated. The final passages of Veblen’s (1899 [Camic and Hodgson, p. 279]) reply to John Cummings must be taken seriously. There Veblen calls for an approach to enquiry in economics that focuses on matters of cause and effect. Veblen had strong opinions and clearly upheld many normative propositions. But he repeatedly argued that for the social scientist, analysis and explanation should have priority over moral pronouncements. In his response to Cummings, Veblen made clear his desire to disentangle factual and moral issues as much as possible. He urged economists and other scientists “to keep the cultural value and the moral content … apart.” After provoking his readers of the Leisure Class with ethically loaded terms, Veblen declares that the foremost purpose of science is understanding and causal explanation of the existing and the possible.


This is no isolated example. He repeatedly argued that analysis and explanation should have priority over moral pronouncements in social science. There are several other cases where Veblen argues that judgments of fact and value must not be conflated. See Camic and Hodgson (2011, pp. 16-18) for several supporting quotations from his work.


We are warned not to to take Veblen's words at face value. Caution is indeed required, but we must be doubly careful. While Veblen was sometimes ironic, there is no basis to assume that he meant the opposite of what he actually and repeatedly wrote, including his criticisms of those who favor “homiletics and reformatory advice” over hard “analysis and exposition of the causal complex at work.” If we dismiss these remarks, on the grounds of possible irony or whatever, then we are in danger of abandoning his legacy. We might end up imputing to Veblen a set of propositions that stem from our own imagination, without any possibility of corroboration in his texts.


My own position is that value judgments are unavoidable in science. But we should not go so far as to regard judgments of fact and value as the same thing or indistinguishable. Although the conduct and content of science are affected by ideology, ideology is not science. As Irene van Staveren (Review of Political Economy, 19(1), January, 2007, pp. 21-22) rightly argues “there is a fundamental difference between distinguishing facts and values (which is necessary for conceptual reasons as well as for doing justice to the different meanings of these concepts) and placing them in a dichotomous relation.” Like van Staveren, I do not see them as necessarily dichotomous.