Instrumentalism is a view in philosophy of science that claims scientific theories are merely useful tools for predicting phenomena instead of true or approximately true descriptions of the physical world. Instrumentalism can best be understood as a position against scientific realism and its roots date back to the turn of the twentieth century with the work of Ernst Mach (1838-1916) and Pierre Duhem (1861-1916), who were both physicists and philosophers of science who held strong instrumentalist views. A contemporary form of instrumentalism is Bas van Fraassen’s (1980) constructive empiricism.


Scientific Realism

Scientific realism is the philosophical view that scientific theories are true or approximately true descriptions of the physical world. There are several types of scientific realists, but the most well known ones are realists about scientific laws and entities (Boyd 1983; Psillos 1999), and realists about scientific entities only (Hacking 1983; Cartwright 1983).

For instance, consider the Pauli exclusion principle, named after the Austrian physicist Wolfgang Pauli (1900-1958), which states that no two electrons in an atom have the same four quantum numbers. Some scientific realists believe that the Pauli exclusion principle is true and so believe in the entities referred to in the laws, in this case, electrons and atoms. However, other scientific realists just believe in the entities referred to in the Pauli exclusion principle because the principle itself is strictly-speaking false. At any rate, instrumentalists disagree with both sorts of scientific realists.

Early Instrumentalism

Ernst Mach and Pierre Duhem were the first philosophers of science to develop a sophisticated defense of scientific progress without accepting scientific realism. Although Mach’s instrumental views arose first, Duhem popularized instrumentalism among philosophers of science, largely because his thorough defense of it in his 1906 book The Aim and Structure of Physical Theory.

Duhem’s instrumentalism was centered on a firm separation between metaphysics and physics. Duhem claimed that physics could be and should be done independently of deep metaphysical assumptions. Instead, the aim of physical theory is to develop mathematical laws that predict phenomenological laws with as much precision, completeness, and simplicity as possible. In Duhem’s words:

A physical theory is not an explanation; it is a system of mathematical propositions whose aim is to represent as simply, as completely, and as exactly as possible a whole group of experimental laws (Duhem 1954, 19).

Thus physicists need not believe that the fundamental laws of physics (e.g. Einstein’s light principle) are true or that the fundamental physical entities (e.g. electrons) actually exist. Rather, these are just useful fictions that help physicists predict phenomena. Also, Duhem suggested that chemistry and biology can be viewed instrumentally since the methods and instruments of chemistry and biology depend on physical theories.

However, Duhem was well aware that it seems to be a spectacular coincidence that a merely instrumental physics can predict novel phenomena. Rather, some philosophers, including Duhem, were skeptical that physical theories were merely instruments due to this spectacular feature of physical theories. Philosophers of science call this view the No Miracles Argument against instrumentalism. However, Duhem formulated an amazing response to the No Miracles Argument, which has come to be known as the theory-dependence of observation.

Theory-Dependent Observation

Duhem’s claim is that just in order to interpret observational data from physical instruments and apparatuses, physicists need to use physical theory. For example, one event that made Isaac Newton (1643-1727) famous was the ability of his theory of gravity to predict new planets, such as Neptune in 1846. Literally, theoretical physicists in the 1800s used Newton’s theory to predict an eighth planet from perturbations in Uranus’s orbit. Then these theoreticians told astronomers where to look in the sky to find the eighth planet. Eventually, astronomers were able to find the planet in 1846, but only with the use of telescopes.

So, Duhem would point out that Newtonian gravitational theory was able to predict a new planet, but only with the aid of previous physical theory about telescopes as well as theory about the celestial landscape at that time. Thus when scientists predict and observe new phenomena with scientific theories, they are really observing theory-dependent phenomena that could be fictitious themselves.

Of course, realists had another challenge to offer instrumentalism. Namely, theories cannot just be instruments because physicists can often construct and execute “crucial experiments” that falsify a physical theory. However, Duhem had an even more spectacular argument against this claim. Among philosophers of science, Duhem’s argument against the possibility of falsification of scientific theories with crucial experiments has come to be known as The Quine-Duhem Thesis.

The Quine-Duhem Thesis

The Quine-Duhem Thesis, sometimes nicknamed “the underdetermination argument” is the view that any scientific theory is underdetermined by any empirical evidence we can offer in its favor. Duhem originally formulated the thesis for physical theories, but since then the Harvard philosopher Willard van Orman Quine (1908-2000) expanded it to all scientific theories.

The thesis was constructed from Duhem’s insight that background assumptions are needed to deduce observational predictions from physical theories. For example, Isaac Newton (1999) assumed that telescopes were reliable observational instruments and that planets can be idealized as point-masses in his prediction that the center of mass of the planetary system lied inside the sun. The former assumption Duhem would call an “observational assumption” and the latter he would call a “theoretical assumption.” Later, Quine (1951) noted that there are multiple metaphysical, mathematical, and logical assumptions that underlie theoretical predictions as well.

For example, in the aforementioned theoretical prediction by Newton, Euclidean geometry was used to represent actual physical space, classical logic (including the law of excluded middle) was used in the deduction, and absolute motion was presupposed as a metaphysical assumption.

Thus as a matter of logic, when an observational prediction from a scientific theory is not observed or a conflicting observation is observed, this fact does not imply that the theory is false. It implies that the theory or one of its many background assumptions conflicts with observation. Notice that a theory plus its background assumptions roughly cover the whole of the science up to that time. This fact prompted Duhem to develop the philosophical view known as confirmation holism, which states that no scientific theory can be tested in isolation and, rather, when we test one scientific theory we are testing the whole of science with that test.

Given the Quine-Duhem Thesis, confirmation holism, and the theory-dependence of observation, Duhem’s instrumentalism teaches us that so-called novel predictions about the physical world are illusions constructed from entrenched scientific theories and a so-called falsification of a scientific theory is just incoherence in our web of beliefs.

Contemporary Instrumentalism

Despite the appeal of Mach and Duhem’s instrumentalism, science became so successful in the twentieth century, that it became more and more difficult to defend instrumentalism. For instance, realists developed convincing arguments identifying certain theoretical virtues (such as explanatory power) as constitutive of good scientific theories and so underdetermination could be overcome after all (Psillos 1999). Nevertheless, a new wave of instrumentalism arose under the Princeton philosopher Bas van Fraassen. His view, known as constructive empiricism claims that while scientific realism is a reasonable philosophical view about scientific theories, instrumentalism is equally reasonable (van Fraassen 1980).

Notice the difference in Duhem’s instrumentalism and van Fraassen’s instrumentalism. Duhem claims that the aim of physical theory should be instrumental. Van Fraassen claims that the aim of physical theory could be instrumental. In other words, constructive empiricism claims that it is just as rational for a philosopher of science to be an instrumentalist as it is for her to be a realist. It makes no claims about how scientists should view scientific theories. Perhaps it is best for science (i.e. scientific progress) that scientists honestly believe in scientific realism. Constructive empiricism can admit this much. However, what it does not admit is that instrumentalism is not a perfectly reasonable way to make sense of science from a philosophical viewpoint.

There are a number of questions that contemporary instrumentalists must answer, such as: How we cannot be ontologically committed to the entities postulated in our scientific theories if the theories describe causal processes? Nevertheless, there are a number of questions that scientific realists must answer, such as: What does it mean to say that a scientific theory is approximately true?

Also, some of the biggest figures in twentieth century science were instrumentalists. For example, Nobel laureates Niels Bohr (1885-1962) and Werner Heisenberg (1901-1976) developed the very instrumentalist Copenhagen interpretation of quantum mechanics in the 1920s, which remains the most popular interpretation of quantum mechanics in contemporary physics. Therefore, instrumentalism is still a vibrant philosophical view and will be around for a long time to come.

See also


  • Boyd, Richard. 1983. On the Current Status of Scientific Realism. Erkenntnis 19: 45-90.
  • Cartwright, Nancy. 1983. How the Laws of Physics Lie. Oxford: Oxford University Press.
  • Duhem, Pierre. 1954. The Aim and Structure of Physical Theory. Philip Wiener, ed. Princeton: Princeton University Press.
  • Hacking, Ian. 1983. Representing and Intervening: Introductory Topics in the Philosophy of Natural Science. Cambridge: Cambridge University Press.
  • Newton, I. 1999. The Principia, Mathematical Principles of Natural Philosophy: A New Translation, trans. I.B. Cohen & A. Whitman. Berkeley: University of California Press.
  • Psillos, Stathis. 1999. Scientific Realism: How Science Tracks Truth. London: Routledge.
  • Quine, W.V.O. 1951. Two Dogmas of Empiricism. Philosophical Review 60: 20-43.
  • van Fraassen, Bas. 1980. The Scientific Image. Oxford: Clarendon Press.

External links

All links retrieved April 16, 2014.

General Philosophy Sources


New World Encyclopedia writers and editors rewrote and completed the Wikipedia article in accordance with New World Encyclopedia standards. This article abides by terms of the Creative Commons CC-by-sa 3.0 License (CC-by-sa), which may be used and disseminated with proper attribution. Credit is due under the terms of this license that can reference both the New World Encyclopedia contributors and the selfless volunteer contributors of the Wikimedia Foundation. To cite this article click here for a list of acceptable citing formats.The history of earlier contributions by wikipedians is accessible to researchers here:

Note: Some restrictions may apply to use of individual images which are separately licensed.