The philosophy of science is a subdiscipline of philosophy that utilizes the fields of epistemology (how we know what we know) and metaphysics (the fundamental nature of reality, often outside human observational experience) to study the principles and methods of science and the natural world. It seeks to understand the meaning, method, logical structure, and reliability of the scientific way of knowing about the world by developing and testing scientific theories. It also studies the relationships between different sciences and explores the issue of reductionism—the idea that theories of one science (such as biology) can be reduced to those of a more fundamental science (such as physics), or even a “Theory of Everything.”
Science uses a particular approach to knowing about the world and how it works by providing natural explanations for natural occurrences. Scientific explanations come from observations that can be repeated and confirmed by other researchers. Science utilizes the process of discovery; it is open to change as new information is discovered. According to scientific realists, all scientific thinking rests on just a few major principles: a real knowable universe exists, independent of human perception; the universe behaves according to certain predictable laws, without any outside influence; and these laws are knowable through observation and testing. This is the essential nature of science.
Or is it? Philosophers of science debate many of these points, and several schools of thought have emerged that represent or challenge the above position. These include logical positivism, the deductive-nomological model, the hypothetico-deductive method, social constructivism, and postmodernism.
Logical positivism developed from the ideas of positivism, epistemology, and mathematical logic. Positivism was derived from the work of French philosopher August Comte (1798-1857), who hypothesized that human culture evolved through three stages: theological, metaphysical, and positive. In the theological stage, humans understand nature through religious superstition and magic. During the metaphysical stage, magic and religion are replaced by impersonal but unobservable forces. The final positivist stage, according to Comte, is where understanding truly lies: humans rely on observation and experiment to understand nature. Positivists were skeptical of metaphysics in general and many theories involving phenomena that were not directly observable. Logical positivists, however, through the work of Gottlob Frege (1848-1925) and Bertrand Russell (1872-1970), added mathematical logic to the positivist philosophy. Mathematical logic is subject-neutral and, through the use of symbolic rather than linguistic terms, makes it possible to analyze objectively all kinds of concepts.
Logical positivists believe that all the events that take place in the natural world are controlled by the laws of nature. These laws can be determined through the observation of consistencies in the relationships between observable phenomena and make possible the explanation of natural phenomena. They developed the verifiability principle: a statement is meaningful if and only if it can be proven true or false through experience. These statements are proven true or false through mathematical logic, written as material conditionals (mathematical if-then statements). Logical positivists originally required that every element of every scientific theory must be expressed in precise mathematical ways, using definitions called correspondence-rules. These correspondence-rules explain how the nonobservable elements of the theory make a difference to observational experience. If there is no observable difference in result between a theory that includes an unobservable element X and one that doesn’t, then, say the logical positivists, there is no evidence that element X is present at all. The correspondence-rule requirement made it necessary to account for every element in the theory in a way that could be verifiable through observation. Therefore, one of the central issues of the positivist approach is the difference between observational terms and theoretical terms.
Logical positivism has developed certain difficulties. For example, the observational/theoretical distinction is often difficult to define. Some observables are too small or large to be observed by humans without instrumentation, which to some make them theoretical. Also, difficulties in writing the correspondence-rules to allow dispositional properties stimulated Carl Hempel (1905-1997) to argue against such a rule in favor of the idea that all the terms in a theory should be understood as an interconnected whole, rather than individually developed. Additionally, the act of observation or the presence of instrumentation can, in some cases, alter what is being observed, changing its behavior, its fundamental nature, or both. Finally, philosophers such as Thomas Kuhn (1922-1996) have argued that scientists do not, in fact, choose between competing theories in a rational manner using objective facts; rather, their observations are influenced by the theories they are testing.
The Deductive-Nomological Model
The deductive-nomological model of scientific explanation was developed by Hempel and Paul Oppenheim (1885-1977). In this model, the description of the phenomenon is deducible from a set of premises and the laws of nature. The deductive-nomological model sets out four main conditions that must be met for a scientific explanation: the explanation for the phenomenon must be deducible from the set of premises, using mathematical logic; the set of premises must contain at least one natural law necessary for the explanation; the statements composing the set of premises must be empirically testable; and all the laws of nature used must be independently confirmed. This model creates criteria for determining the explanatory power of the theory separate from any subjectivity of the practitioner. However, many philosophers of science argue that there is no true explanation without determination; mathematical logic alone is inadequate to represent the connections between the events in question. The model might allow the scientist to get the pieces of the theory right, but she will not be able to determine which piece is the determining one that explains the phenomenon.
The Hypothetico-Deductive Method
The hypothetico-deductive method, also known as falsificationism, was developed by Karl Popper (1902-1994). It is the model most commonly associated with science today. The method begins with the scientist developing a hypothesis that may explain some phenomenon. The scientist then develops a list of observable predictions that, if true, would support the hypothesis. The hypothesis is then assumed to be true, and a series of experiments are undertaken in an attempt to disprove, or falsify, it. Observations that are contrary to the hypothesis are considered evidence against it; observations in agreement with the hypothesis are considered evidence in support. It is possible to test competing hypotheses in this manner; a comparison of how well each is sustained by its predictions can lead to acceptance of one over another. Failure to falsify the hypothesis does not prove it, but rather gives it confidence. Repeated failed attempts to falsify a hypothesis lead it to become a more established theory.
However, critics point out that it is unrealistic to imagine that scientific experiments really operate this way. The complexity of the experiment and the world, they say, often make it impossible to isolate one variable to test. Willard Van Orman Quine (1908-2000) raised important concerns about this method. He argued that scientific terms and evidence have to be understood holistically, in that the meanings of scientific terms are introduced in interdependent networks, and only networks of hypotheses can be tested, not individual ones. Since observations, hypotheses, and theories are embedded in our overall understanding, it is not possible to make truly independent observations. Quine and Pierre Duhem (1861-1916) believed that it was often impossible to create experiments that would definitively support one hypothesis over another. The ideas of these two men were combined to form the Quine-Duhem thesis, which states that any observations seemingly contradicting one’s hypothesis can always be accommodated to any theory because humans adjust their interpretations of results by altering other, connected beliefs. In other words, it is possible to save the hypothesis by revising the assumptions to accommodate the observation. While this criticism sounds fatal to the method, Quine was pragmatic in his proposed solution: he contended that scientists adjust their hypothesis to maximize its productiveness in generating new predictions, consistency with former beliefs, overall simplicity, and the modesty of the claims. Adjusting the hypothesis in this way maximizes its likelihood of being “true.”
Despite these issues, the hypothetico-deductive method is the most popular scientific approach in use today. However, the controversial Quine-Duhem thesis persuaded some to reject the hypothetico-deductive method in favor of more relativistic approaches. Social movements such as environmentalism and postmodernism, as well as concerns for the social effects of technology, have heightened the interest in the social aspects of scientific inquiry. This has led many philosophers of science away from the realist view of how science is performed and toward relativist models such as social constructivism and postmodernism.
Social constructivists believe that reality and knowledge are not empirically discovered but rather constructed through organized human behavior. Most agree that the world is real and knowable, but they argue that scientific knowledge is socially constructed and people vary in how they understand that world. An extreme version of social constructivism is the idea that humans themselves, through scientific (or other) knowledge, invent the physical universe.
David Bloor, studying what he termed the sociology of knowledge, developed the “Strong Programme” to describe the development, distribution, and maintenance of knowledge. Bloor defined something as “knowledge” if it was taken for granted, institutionalized, or invested with authority by a group of people. The four tenets of the Strong Programme are (a) causality, or what causes knowledge or beliefs to become more or less widely held; (b) impartiality, concerned with all beliefs, true or false; (c) symmetry, in that the same kinds of concepts can explain true or false beliefs; and (d) reflexivity, in that the patterns of explanation can be applied to sociology of science. Bloor argued that truth alone is insufficient to explain why a true belief comes to be held by scientists, but rather social factors play a role. Critics argue that there must be some objective ways to compare true and false beliefs that, while filtered through a social lens, nonetheless reflect an independent material world.
Postmodernists reject the modernist positivist idea that there is an objective reality about which science holds the monopoly of truth. Instead, they posit that all humans subjectively interpret reality through their experiences. In this view, all science takes place through these filters, and both observations and theories are influenced by them. There is no one true way of understanding the universe, no universal narrative that would justify privileging the scientific explanation for a phenomenon over other explanations. Critics of this model argue that while it is useful to open science to pluralistic views, it is still critical to maintain an ability to discern legitimate truth from shared belief in untruth, and scientific explanation is not a democratic process.
Postmodernism is heavily influenced by the idea that language and narrative shapes human thought and experiences. David Orr, a postmodernist philosopher of science and proponent of ecological literacy, argues that scientific knowledge alone is insufficient to understand the universe; rather, social issues such as ethics and equity must be included.
While there are some scientists who believe that scientific explanations can be used to explain everything, most believe that science is useful for answering certain types of questions, leaving others to the realms of theology and moral philosophy. Scientific thinking, in whatever form, remains a powerful tool for humans to use to understand the world we live in.
- Curd, M., & Cover, J. A. (1998). Philosophy of science: The central issues. New York: Norton.
- Klee, R. (1997). The philosophy of science: Cutting nature at its seams. New York: Oxford University Press.