Duhem made a number of enduring contributions to thermodynamics andphysical chemistry. Among these were the Duhem–Margules andGibbs–Duhem equations, which deal with reversible processes inthermodynamics as quasi-static limiting processes and give a generalproof of the Gibbs phase rule. These results were obtained in thecontext of a program of generalized thermodynamics called“energetics.” Indeed, Duhem's entire scientificprogram was driven by the conviction that a generalized thermodynamicsshould be foundational for physical theory, thinking that all ofchemistry and physics, including mechanics, electricity and magnetism,should be derivable from thermodynamic first principles. Duhem startedfrom the concept of the thermodynamic potential (the topic of hisfailed thesis), deploying it in a manner similar to that of potentialsin mechanics, so as to represent all physical and chemical changes. Theprogram finds its mature statement in his Traitéd'énergétique of 1911; it was well received bylate-nineteenth-century energeticists, such as Wilhelm Ostwald andGeorg Helm. So important was energetics for Duhem, that his work in thehistory and philosophy of science has been viewed as an attempt todefend its aims and methods (see Lowinger 1941). More recently, NiallMartin and others have argued for the importance of religious motivesin Duhem's work (see Martin 1991, Jaki 1991) and it has becomeclear in the course of Duhem's writings that he expected theendpoint of science to harmonize with the teachings of the CatholicChurch.
Whatever was Duhem's initial motivation, his historical andphilosophical work took on a life of its own. One cannot readDuhem's numerous historical and philosophical tomes and thinkthat his labor was only in the service of energetics and that the solegoal of his works was but a defense of its methods and its historicalposition. No doubt energetics might be a thread running throughDuhem's various works, and no doubt these works harmonize withthe method of energetics as he conceives it, but energetics cannot bethe whole story.
Duhem banishes model building from physical theory (as he previouslybanished Maxwell's rashness) because model building breaks withhistorical continuity; in fact, model building is not only historicallynon-continuous, but present models are even often“non-continuous” among themselves. Some model builders evenfind pleasure in building two or more models of the same law. The factthat the English physicist can accept disparate models, breaking up thehistorical continuity of science and even its present unification, iswhat shocks Duhem; it is what reconfirms for him that English physicsis not the work of reason, but the work of imagination.
For Duhem there is a crucial difference between representing andexplaining. He divides theories into two large categories, explanatoryand purely representative theories, and argues that physical theoriesshould not be considered as explanatory, but as purely representativeor classificatory. The argument, as we have seen, is that, in order forphysical theory to be explanatory, it would have to be subordinate tometaphysics and not autonomous. The reference to the two words havingthe same meaning for Maxwell and English scientists is thus a referenceto what Duhem would consider a confusion about the aim of physicaltheory, one that arises in the identification of the model with thetheory, in thinking that what is represented by the theory and/or modelis real.
A principle of historical continuity is invoked in the conclusion ofDuhem's primary work on Maxwell. Duhem evaluates there aninterpretation of Maxwell's work he attributes to Heaviside,Hertz, and Cohn, among others. He quotes Hertz as stating that:“what is essential in Maxwell's theories is Maxwell'sequations.” He takes this to be Hertz's way of salvagingwhat is valuable in Maxwell from the midst of logical errors andincoherence, which are not only difficult to correct, but which havefrustrated many illustrious mathematicians. But Duhem cannot acceptHertz's implied criterion of identity for physical theories. Heasserts that he might accept such a criterion for algebra but “aphysicist is not an algebraist”:
Global realism is a very weak philosophical thesis. It does not convey any information concerning external reality. According to parsimonious scientific realism, we do have epistemic access to some observable and unobservable portions of the world, but such knowledge is always partial. It is obvious that even scientific fails to constitute a complete knowledge of reality. The universe may well be the “biggest” or “most encompassing” object of scientific study, but cosmological models are constructed by taking into account only a few properties (radius of the visible universe, rate of expansion, mean density of matter etc.) and standard cosmology relies heavily on elementary particle physics. Cosmology remains appropriately silent on the behavior of most systems, which are studied by biology, psychology, physiology, demography, and economics, among other disciplines.
The principled reason why some objects, characterized by some properties, are retained despite theory change is that we have managed to forge strong causal connections with them (Chakravartty 2007). The electron can be detected by the impressively precise measurement of some specific properties, using different methods. Moreover, some mathematical formulae, which state relations between these properties, have been repeatedly verified. Even though the mathematical objects in a theory (unlike the terms used in ordinary language) are defined with exactitude and provide a complete characterization of existing systems, this situation does not preclude novelty. Some new empirical evidence will be linked to theoretical features, provided that we modify the structure of the old theory so that what we call “electron” becomes another theoretical object, albeit one possessing many properties in common with the old electron. For example, according to the old theory, the electron has a mass and a charge; in the new theory it retains these same properties but gains a spin. Given the possibility of completing theories in such a way, the truth of a theory remains only partial. It is also approximate, in the sense that the value of certain quantities is known only within a certain margin of experimental error.
3. The Underdetermination of Theories by Data (UT). Perhaps the most serious challenge to the scientific realist lies in the alleged possibility of constructing interesting – that is, uncontrived and non-artificial – alternative, mutually incompatible theories that save all the data as satisfactorily as any currently accepted theory. This is the celebrated Duhem-Quine thesis of the underdetermination of theories by the observations. What does it say?
In the metaphysics of science, things are referred to as “particulars.” A particular is anything that has specific properties and can enter into relation with other particulars. Entities (such as an electron), systems (a crystal), events (a gas having a specific pressure) or causal processes (the fall of a body), are all particulars. With regard to science, our concern is with the properties of particulars that are objective, which differ from holistic properties possessed by singular things or singulars. In the metaphysics of science, we are therefore concerned with particulars endowed with objective properties and relations. In what follows, we will concentrate our discussion on entities, which form a special class of particulars.
A distinguished scholar and writer in the field of philosophical studies, he has earned pre-eminence in the study of philosophy of logic throughout forty years of dedication to teaching, research, and writing.
In recognition of his scholarship in the field of logic and for his contributions to the literature of philosophy, I now present Willard Van Orman Quine for the honorary degree, Doctor of Laws.