Encyclopedia of Educational Philosophy and Theory

2017 Edition
| Editors: Michael A. Peters

Science, Naturalism, and Education

Reference work entry
DOI: https://doi.org/10.1007/978-981-287-588-4_37

The relationship between science and naturalism has been a topic of an ongoing debate in both the philosophy of science and science education. That science and naturalism are somehow related is usually taken for granted. What is at issue, however, is the exact nature of this relationship as well as the particular variety of naturalism involved. More precisely, is naturalism a contingent outcome of science or a necessary presupposition for doing science, and is it an ontological or a methodological position? Although these questions belong in the philosophy of science, science education is concerned with them too, because it needs to know what to teach about the nature of science, and because it is faced with the challenge of supernaturalists who attempt to introduce creationism or intelligent design into the curriculum.


The evolution/creation controversy reveals the kind of naturalism involved here: naturalism as the opposite of supernaturalism. This naturalism is a set of assumptions about the furniture of the world and about the way the world works – and thereby about which entities and processes are admissible referents of scientific theories and explanations. Thus it is an ontological (or metaphysical) view. Roughly speaking, naturalism is the view that all that exists is our lawful spatiotemporal world. Its negation is supernaturalism: the view that our lawful spatiotemporal world is not all that exists because there is a nonspatiotemporal world transcending the natural one, whose inhabitants are intentional beings that are not subject to natural laws. Naturalism assumes that everything in the world works according to its own laws as opposed to laws imposed from above or no laws at all; that nothing comes out of nothing and nothing turns into nothing; and, consequently, that also our perceptions and conceptions, that is, the very processes of gaining knowledge, are not manipulated by external, in particular supernatural, agents but are themselves purely natural and lawful processes.

Note that a lawless world could be natural, too. But a lawless or chaotic world (in the traditional sense of “chaos”) would not be able to sustain the existence of complex beings with complex brains for a longer time, although they could randomly pop into (and out of) existence once in a while. For this reason, the concept of ontological naturalism usually includes the assumption of a lawful natural world, in which even randomness is not lawless as there are probabilistic and stochastic laws. Likewise, a supernatural world is not taken to be chaotic either.

The previous characterization is somewhat unsatisfactory because it seems to involve some circularity: natural is what is not supernatural, and supernatural is what is not natural. This problem can be avoided only by a full-fledged ontological theory describing the features of natural entities and elucidating the concepts of thing, property, law, event, process, and cause. Such theories, which must also address the status of mind and of abstract objects, do exist, but they are too complex to be summarized here (Bunge 1977; Mahner 2012).


In ordinary language, “supernatural” applies not just to religious entities, powers, or processes (gods, angels, demons, miracles) but also to profane ones (poltergeists, clairvoyance, psychokinesis). The latter might better be called “paranatural” because, possibly, they could just be yet unknown natural entities or abilities. For more than 150 years, parapsychology has unsuccessfully attempted to prove their existence so that the paranatural is most likely illusory (and parapsychology remains a pseudoscience). Yet if parapsychology were successful, the paranatural would probably be incorporated into normal science, thus becoming natural. Therefore, “paranatural” is mostly an epistemological attribute.

By contrast, the supernatural is something that supposedly exists and works beyond natural, spatiotemporal entities, laws, and processes. In fact, the supernatural is usually defined but negatively, that is, by negating certain natural properties. For example, “transcendence” is the negation of “immanence,” that is, not being located in our spatiotemporal world. Or, a first cause is nothing but an uncaused cause. Other properties of the supernatural are just natural properties raised to an absolute degree (omnipotence, omniscience). In this construal, the supernatural is only quantitatively different from the natural and as its attributes are still conceived of on the basis of familiar natural properties, it remains somehow intelligible. Perhaps it is this familiarity that blurs the line to the paranatural and that helps people to believe in the existence of such entities. As this conception is more or less anthropomorphic, many theologians consider it primitive and unacceptable. So they take the supernatural to be categorically different from the natural. God, then, is the Wholly Other, not someone or something to be understood even by the faintest analogy with anything known natural; his properties are genuinely transcendent and essentially mysterious, ineffable, incomprehensible. These two types of the supernatural may be called overnatural and transnatural, respectively (Spiegelberg 1951).

Obviously, if something is by definition transcendent and ineffable, it is unintelligible. Nobody, neither theologians nor scientists, should be able to know and say anything about it. So the transnatural is by definition beyond the reach of science. However, it is also useless for the ordinary believer – and ultimately also for theology – because every discipline needs a subject matter to investigate. To obtain a modicum of intelligibility, conceptions of the supernatural usually combine overnatural and transnatural features. This allows the believer to oscillate between these two conceptions, depending on his argumentative needs. Only by referring to the overnatural can he claim that the supernatural meddles with the natural world at least occasionally and that science is able to study the supernatural. After all, if some overnatural entity interacted with the natural world, it would have to work by partly naturalizing itself, that is, by being able to produce natural causes and, in turn, be affected by them (Pennock 2000). If the overnatural worked through sheer magic, it would work through principally incomprehensible processes.

Scientific Methods and Evidence

A popular view among scientists maintains that science need not bother with philosophy, let alone ontology, at all: scientists should just apply and follow the scientific method. If science is ultimately about finding the truth, all that counts is evidence. Whether it confirms the natural or points to the supernatural, we should follow the evidence wherever it leads. This empiricist view assumes that both scientific methods (such as observation, measurement, and experiment) and the evidence they produce do not depend on any ontological assumptions, such as naturalism.

Yet why would some scientists and philosophers maintain that science is based on naturalism? This becomes clear by analyzing a simple measurement, occurring in ordinary life or in science: measuring temperature with a thermometer. First, does this measurement occur in the real world or in our imagination only? Assuming that both the thermometer and the surrounding medium are real things, that is, being aware that the measurement in question is not merely a thought experiment, is an instance of ontological realism. Second, do scientists expect that the mercury in the thermometer moves capriciously or lawfully? Of course, scientists expect that there is a lawful relationship between some property of the surrounding medium (its kinetic energy or temperature) and some property of the mercury, namely, its density or expansion. As the mercury can expand only within the narrow glass tube, there is a lawful relationship between the height of the mercury column and the outside temperature. And this lawful relation always obtains under the same conditions and at all times so that the measurement can be repeated as often as needed. This is the lawfulness principle. Third, scientists also expect that they can cause the thermometer reading to go up or down, for example, by heating or cooling the surrounding medium. This requires some ontic causality principle (as opposed to an epistemological one).

Fourth, a tacit assumption is that the thermometer cannot be causally influenced in a direct way solely by our thoughts or wishes, that is, without the interposition of motoric actions of our bodies. Indeed, if the world were permeated by causally efficacious mental forces, scientists would have no reason to trust the reading of any measuring instrument or the results of any experiment. This is the no-psi principle. Fifth, what holds for natural entities applies a fortiori to supernatural entities. Scientists (tacitly) expect, then, that no supernatural entity manipulates either their scientific tests or their mental (neuronal) processes or both. This is the no-supernature principle. The reason for these last two assumptions is quite simple: the data obtained through observation, measurement, or experiment could not function as evidence if they were the telepathic product of wishful thinking or of supernatural manipulation.

Naturalism as an Ontological Presupposition of Science

The ontological expectations listed in the preceding are held mostly tacitly, and even if scientists are not aware of them, they are built in in the very design of any scientific technique: it would make no sense to conduct measurements or experiments in the first place if these ontological expectations were different. Of course, it is logically possible that the world does not conform to these ontological expectations, either entirely or partially. In the first case scientific methods should fail, in the second case they should work only wherever or whenever the world does possess the required ontic properties. However, if the very functioning of scientific methods depends on whether the world has certain ontic properties, the above “expectations” actually constitute ontological presuppositions: they are necessary ontological conditions for doing science. Again, this does not entail that the world actually exhibits such conditions, but that scientific methods can work only wherever they obtain (Mahner 2012, 2014).

Now scientists have no reason to assume that the world is ontically patchy, but the starting point is that it is uniform: that naturalism applies everywhere. Thus they hold the ontological null hypothesis that there is no supernature. A null hypothesis usually negates that something is the case, in particular that something exists or that two variables are related. Examples: “Junk food is not the cause of obesity,” or “The Loch Ness monster does not exist.” The reason for this approach is simple: science cannot assert that something is the case without empirical evidence. So in order to prove some positive alternative hypothesis, its corresponding null hypothesis must be refuted empirically. However, there is an important difference between scientific and ontological null hypotheses: the latter are usually regarded as unfalsifiable by direct empirical evidence. But some ontological hypotheses may be disconfirmed indirectly. For example, science could fail as a cognitive enterprise, either in its entirety or in some particular area, so that naturalism as an ontological view applying to the whole world might have to be reconsidered.

Can Science Study the Supernatural?

Empiricists, who believe that scientific evidence can be had without ontological presuppositions, offer long lists of conceivable supernatural interventions that could be studied by science (Fales 2013; Fishman and Boudry 2013). Their argument: if science can confirm the existence of the supernatural, it does not presuppose naturalism. For example, if angels descended from the sky and raised the dead or if studies on the effects of intercessory prayer yielded significant positive results, we would have empirical evidence for the supernatural and hence a valid test. At first sight, this sounds convincing, but a closer look at the hidden assumptions of these two examples, abbreviated A and B, reveals some problems.

Both A and B still presuppose that our perceptions and conceptions of these occurrences are natural processes, that is, that they are not the result of supernatural manipulation. Otherwise they would not provide evidence for a real event, but could not be distinguished from hallucination. A requires that supernatural entities are able to naturalize themselves at least to such a degree that they are perceptible and able to interact with natural matter. B presupposes that both the praying as a neurophysiological process and the healing as a biological process are partly natural. Only the intermediate step would involve supernatural causation, and it would also have to be partly natural, as the praying would have to affect a supernatural entity, who in turn must stimulate the healing process. Obviously, what must be involved here are merely overnatural entities. By contrast, transnatural events would not be detectable. Think of transsubstantiation: whatever scientific analyses we undertake, we see nothing but bread and wine. Another example is continuous creation: there is no way we could prove scientifically that God recreates the world moment by moment out of nothing to sustain our existence.

Now what could science find out about overnatural entities? Not much. Science would be able to empirically study only the natural aspects of A and B, that is, the natural effects of overnatural interventions. It would not be able to study overnatural entities directly, unless we could catch, for instance, an angel and hold it for questioning or even dissection. But if that were possible, we would be faced just with some unusual natural entity, not an overnatural one. So science may confirm that some spooky events occurred, but the possible explanations would be entirely theoretical. As in many other areas, science would have to infer the unobservable from the observable by postulating so-called theoretical entities that are able to explain the observable. However, as is well known, an inference from the observable to the unobservable is never conclusive, because different unobservables may explain the same observable.

For example, if there were reproducible evidence that intercessory prayer works, there would be various alternative natural hypotheses compatible with the evidence, such as a superior alien civilization playing a prank on us. In the philosophy of science, this is known as the problem of underdetermination. Whereas in science underdetermination is an overrated problem because it can often be overcome in practice, it remains a problem for supernatural hypotheses. After all, in science it is ontologically innocuous that for some evidence e there are alternative hypotheses consistent with e, because all these alternatives refer to natural entities or processes. In the case of a hypothesis referring to an overnatural entity, however, there are also alternative natural hypotheses compatible with e, even if they are as outlandish as the alien prank hypothesis. So would e ever be good enough to make scientists opt for something that is even more outlandish for involving a supernatural entity?

Supernatural Entities and Scientific Explanation

At first sight, invoking a supernatural cause to account for some fact does seem to have explanatory power. For example, intelligent design creationists claim that the theory of evolution cannot explain how certain complex organs have originated. So they invoke a supernatural entity, an intelligent designer (who allegedly need not be, but is in fact God) who helped to create these organs. This answer appears to have explanatory power because, by analogy with human handicraft, we all understand what creating artifacts is about. Yet in fact it explains nothing because it explains too much. The problem is that an answer like “God made it the way it is” can be applied to all facts. Whatever exists and whatever happens can be explained by reference to the will and actions of some supernatural entity. But an explanation that explains everything explains nothing. Also, explaining the unknown by something magical and occult is an instance of the obscurum per obscurius fallacy. It may be argued that referring to supernatural entities is an appeal to ignorance, the respective explanations being therefore pseudoexplanatory. Indeed, as science would know nothing about the possible powers and intentions of such entities, explanations of some fact x would reduce to the form: some supernatural entity chose to do x for unknown reasons. Is this superior to “we do not know what caused x”?

Furthermore, there are two proliferation problems with overnatural explanations. First, if we admit one supernatural entity into the explanatory realm of science, we are on a slippery slope to admitting as many as we fancy. If we admit God as an explainer, we may as well admit demons, angels, fairies, and so on. Moreover, in most cases all these overnatural entities would do the same explanatory work, so it would be hard, if not impossible, to choose between competing supernatural explanations. The God hypothesis is not per se more plausible than others, it only appears to be, because in our culture we are more familiar with it. Yet this does not increase its scientific merit. Second, even if science could study the overnatural and incorporate it into its explanations, how would scientists know that those are final? Alternatively, what would they do if they also encountered explanatory gaps in the supernatural world? The analogous procedure would be to resort to super-supernatural entities to fill these gaps in the first-order supernatural world – and so on. This illustrates the problems of the god-of-the-gaps approach.

The Relation Between Ontology and Methodology

In a realist philosophy, being is prior to knowing. That is, the furniture and structure of the world must make cognition possible in the first place, and they must allow for the successful application of scientific methods. Hence, for a methodology to make any sense and to work successfully, there must be an ontology that helps to explain the functioning of this methodology. The ontology explaining the methodology of science is naturalism.

Empiricist philosophers who maintain that science can study the supernatural appear to believe that scientific methodology would remain intact if supernatural entities were admitted as explainers in scientific theories. But if there were solid evidence for the existence of supernatural entities, it would be hard to not also admit a nonnaturalist epistemology and methodology in which special forms of cognition, such as revelation, religious experience, or whatever nonnatural ways of communication with the supernatural may obtain, are accepted as legitimate sources of knowledge and means of justification. Indeed, some theologians make a case for a theistic science (e.g., Plantinga 2001). This illustrates that methodology is not free of ontology, and that the standards of evidence of science are bound to naturalism.

If its methodology (including its standards of evidence) is essential to science, science can work only in a naturalist world. If there is a supernature interacting with our world, then science can study at most the natural part of this world, including perhaps the quasi-natural features of overnatural entities, assuming of course that, in so doing, our cognitive processes are not manipulated by the overnatural. The alternative is to give up the naturalist definition of science and transform it into a supernaturalist, if not theistic, science. The question is, however, whether this “new science” would be a progress or rather a counter-revolution.


  1. Bunge, M. (1977). Treatise on basic philosophy, vol. 3: Ontology I. Dordrecht: D. Reidel.CrossRefGoogle Scholar
  2. Fales, E. (2013). Is a science of the supernatural possible? In M. Pigliucci & M. Boudry (Eds.), Philosophy of pseudoscience (pp. 247–262). Chicago, IL: University of Chicago Press.CrossRefGoogle Scholar
  3. Fishman, Y., & Boudry, M. (2013). Does science presuppose naturalism (or anything at all)? Science & Education, 22, 921–949.CrossRefGoogle Scholar
  4. Mahner, M. (2012). The role of metaphysical naturalism in science. Science & Education, 21, 1437–1459.CrossRefGoogle Scholar
  5. Mahner, M. (2014). Science, religion, and naturalism: Metaphysical and methodological incompatibilities. In M. R. Matthews (Ed.), International handbook of research in history, philosophy and science teaching (Vol. III, pp. 1793–1835). Dordrecht: Springer.Google Scholar
  6. Pennock, R. T. (2000). Tower of babel. The evidence against the new creationism. Cambridge, MA: MIT Press.Google Scholar
  7. Plantinga, A. (2001). Methodological naturalism? In R. T. Pennock (Ed.), Intelligent design creationism and its critics (pp. 339–361). Cambridge, MA: MIT Press.Google Scholar
  8. Spiegelberg, H. (1951). Supernaturalism or naturalism: A study in meaning and verifiability. Philosophy of Science, 18, 339–368.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Singapore 2017

Authors and Affiliations

  1. 1.GWUP e.V.RossdorfGermany