[Yesterday I presented a conference paper at the 112th Annual Meeting of the Pacific Ancient and Modern Languages Association (PAMLA). The conference theme for this year was “Familiar Spirits,” and I presented a paper titled “Philosophically Defining the Supernatural.” The topic relates to previous articles that I have written, both here in my blog series on metaphysical naturalism and in an earlier article here.
This article represents my most up-to-date view on how to metaphysically define “supernatural” phenomena in opposition to “natural” phenomena. I discuss five areas of metaphysical distinction between the two: 1) physicality, 2) uniformity, 3) open vs. closed causality, 4) mental objects & properties, and 5) teleology. Below is the transcript of my paper, with images added from the slides of my attending PowerPoint presentation.]
Halloween is a time of year when we celebrate the supernatural, being a holiday associated with the souls of the dead, witchcraft, and even (from certain quarters) the occult. The “supernatural” is something that is, by definition, different from ordinary “natural” phenomena. Often times we are able to distinguish between the “natural” and the “supernatural” prima facie, meaning that each can be identified at first glance. When we see an apple fall from a tree, we immediately recognize such an event to be “natural.” If, on the other hand, we were to see a ghost, magical spell, or vampire, we would not hesitate to call such phenomena “supernatural.” But what is the real difference that allows us to make such a distinction?
Whereas on the level of common sense the difference between the “natural” and the “supernatural” is often obvious, metaphysically distinguishing between the two on a philosophical level can be far more challenging. For example, if we were to see a witch cast a spell of fire, we would not hesitate to call such a superhuman ability “supernatural.” But how is this ability metaphysically different from a mutant (let’s say an X-man) who has evolved the ability to produce flames from his hands? Both would be superhuman abilities, but we would only consider the witch to be “supernatural.” Given that the two abilities would look the same, however, can we really make such a categorical distinction? Today I will discuss some of the different definitions of the “natural” and “supernatural” proposed by philosophers. My goal will be to show that there are certain attributes of “supernatural” phenomena that make them categorical different from “natural” phenomena, so that there can be a clear and meaningful metaphysical distinction between the two, justifying our use of these terms.
First, we need to understand the difference between the “supernatural” and the “paranormal.” The Parapsychological Association defines the paranormal as “Any phenomenon that in one or more respects exceeds the limits of what is deemed physically possible according to current scientific assumptions.” Notice how, under this definition, the paranormal is not characterized by what is “natural” or “supernatural,” rather than by phenomena that lies beyond the realm of scientific knowledge.
Such phenomena can include many things that we would consider to be “natural,” such as alien abductions, Sasquatch sightings, and the flame-casting mutant discussed earlier. We cannot, therefore, define the “natural” by what has been discovered and verified by the scientific method. Furthermore, there are many natural phenomena that were unknown in the past and have only been discovered in recent times, but we would not describe such phenomena as “supernatural” before their discovery.
Also there is the question of whether science is capable of discovering the supernatural. Sometimes it is claimed that science can only study the “natural” world and that “supernatural” phenomena cannot be known through the scientific method. Science, however, is a method that is above all empirical. Many phenomena that we would describe as “supernatural,” such as vampires, werewolves, and miracles like turning water into wine, would be empirically observable. So, while science has not discovered such supernatural phenomena yet (and perhaps never will), science could still conceivably discover the supernatural, meaning that the supernatural is not, by definition, paranormal. We have to therefore look to other criteria beyond scientific knowledge to define what is “natural” versus what is “supernatural.”
Part of the difficulty with defining these terms is that they are used to describe a wide range of phenomena. Rain is natural. Gravity is natural. Human beings and our evolution are natural. But what are the shared characteristics between these individual “natural” phenomena that allow us to form a universal category of the “natural”? Even if we recognize individual phenomena to be “natural,” that does mean that a real metaphysical distinction has been made, until the “natural” can be categorically defined.
The same problem holds true for the “supernatural.” The term “supernatural” is used to discuss highly diverse phenomena. A genie is supernatural. God is supernatural. Miracles are supernatural. Magical potions and witchcraft are supernatural. And even places like Heaven are supernatural. But what do all of these individual “supernatural” phenomena have in common? What allows us to metaphysically distinguish them from “natural” phenomena? To answer this question, professional philosophers have proposed a variety of criteria for defining the “natural” versus “supernatural.”
Perhaps the most common criterion used to define the “natural” is physicality. Physical materialism has long been associated with “nature” and the “natural” world, especially in the age of science. As philosopher Keith Augustine (“A Defense of Naturalism“) explains: “One obvious candidate for what is meant by the term ‘natural’ is physical. The earliest forms of naturalism, in fact, were versions of materialism or physicalism which maintained that everything that exists is physical.”
There are a number of problems, however, with defining the “natural” solely by what is physical. To begin with, there are a number of phenomena that are difficult to describe and characterize in physical terms. Take for, example, human consciousness and internal subjective states. Even if we can describe the human brain solely with physical terms, how can we describe our internal awareness and mental experiences with only physical terms? There is also the question of whether universals and abstract objects can be explained physically. Can we say, for example, that mathematics is physical? Furthermore, would a lack of physicality always imply the “supernatural”? For example, if mathematical Platonism is true, and math really is immaterial, would we then call mathematics “supernatural” ? In the words of philosopher Stephen Law, this would be “odd,” since mathematics is not usually considered “spooky” enough to be “supernatural.” Then there are “supernatural” phenomena that involve physical states. Take, for example, a man transforming into a werewolf. Such a transformation would involve drastic physical change, but should we then describe werewolves as being “natural”?
One solution that philosophers have proposed for this problem is the idea of Supervenience. Supervenience occurs when lower order states determine higher order states, even if the higher order states cannot be described through the lower order states. Keith Augustine proposes Supervenience as one solution for making human consciousness compatible with physicalism. As Augustine (“A Defense of Naturalism“) explains, “In the contemporary philosophy of mind, an attractive alternative to reductive physicalism is some version of nonreductive physicalism or property dualism. According to nonreductive physicalism, mental states are not simply identical to certain physical states (such as brain states), as reductive physicalists hold; rather, mental states are supervenient upon those physical states … To say that mental states supervene upon physical states is to say that there can be no differences between mental states without a physical difference between the objects which instantiate those states.”
Accordingly, even if one cannot describe internal mental states through solely physical terms, we can still know that solely physical causes are determining mental states. Supervenience can also be proposed as a solution for how “supernatural” phenomena can still involve physical states. Even if a werewolf’s bodily transformation is physical, such physical change could supervene upon immaterial changes and causes. Assuming that there is no physical illness or genetic mutation that causes one to become a werewolf, we could still imagine that the lower order properties of a werewolf may be immaterial and non-physical. As such, the werewolf’s higher order physical change could still be caused by lower order non-physical properties.
Another criterion that is often associated with the “natural” is uniformity. As philosopher David Griffin (Religion and Scientific Naturalism, pg. 12) explains, “Naturalism in [a] minimal sense can be identified with what has historically been called ‘uniformitarianism,’ which is the assumption that the same general causal principles obtain for all events.” Natural events revolve around repeated patterns and consistent natural laws. The sun has risen today, and it will do so again tomorrow. Gravity holds us to the Earth this moment, and it will not suddenly cease in the next moment. One type of supernatural phenomena that involves a breach in this uniformity is miracles. Griffin (pg. 13) defines miracles “as supernatural interruptions of normal cause-effect relationships.”
For example, if you try to walk on water, you will always sink, because gravity is a uniform feature of our universe. If, however, you tried to walk on water and did not sink, we would probably describe such an occurrence as a miracle. As Christian philosopher William Craig (The Case for Faith, pg. 63) explains, “Natural laws have implicit ceteris paribus conditions … In other words, natural laws assume that no other natural or supernatural factors are interfering with the operation that the law describes.” Hence why you will always sink if you try to walk on water ceteris paribus, meaning all other things being equal. Nature is uniform. If such uniformity were broken, however, we would suspect that something more than nature, something “supernatural” was occurring.
The previous example also touches on the issue of open versus closed systems of causality. “Natural” phenomena typically occur within closed systems of causality. If you try to walk on water, the force of gravity and the lack of sufficient friction on the surface of the water will cause you to sink. This system of cause and effect is closed, meaning that there is nothing external to the system that is influencing or changing this pattern of cause and effect. With no external force changing the effect of gravity or the surface friction of the water, the forces within this closed system alone will influence the phenomena.
“Supernatural” phenomena, however, involve external influences on the natural world. Gravity and the lack of surface friction may cause you to sink within a closed system, but if an external force, such as God, intervenes in this system, then the phenomena may change drastically. As such, “supernatural” phenomena imply open systems of causality. When something more occurs than what is possible through the natural forces within a system alone, something external, something “supernatural” is influencing the system, meaning that it is not causally closed, but open.
Another feature of the supernatural is that it often involves irreducible mental states. Supernatural objects like souls, God, and ghosts are mental and have minds. Likewise, even supernatural phenomena without minds often involve mental properties, such as Karma, love potions, or magic that responds to the beckoning of a sorcerer. These “supernatural” mental objects, however, cannot be broken down into non-mental parts. The whole idea behind a soul’s immortality is that it is irreducible, meaning that it cannot be broken down into something less than a soul. “Natural” mental states, however, are reducible. Our mental states are dependent on our physical brains, and if those brain states are reduced, our mental states will be reduced. We can break down a human mind to the point that it is no longer a mind, but we cannot break down a soul to the point that it is no longer a soul. As such, the “supernatural” involves irreducible mental objects and properties.
Naturalist philosopher Richard Carrier (“On Defining Naturalism as a Worldview“) has also argued that the “natural” is best defined as being reductively non-mental. As Carrier explains: “Naturalism is true iff everything that exists is either ontologically reducible to the nonmental, or causally reducible to the nonmental, or both … A mental object is any object that is distinctive of the contents or activity of a mind, in contrast to what we do not consider as such. The most obvious examples of mental objects in this sense are thoughts, perceptions, and emotions.”
“Natural” mental states, such as our thoughts and emotions, are contingent upon lower order physical states. If those physical states change in our brains, then so do our mental states. As such, our mental states do not exist at the lower order level, but are only emergent properties at the higher order level. “Supernatural” mental states, however, are irreducible. They are not emergent properties, but exist at the lower order level and can thus be described as ontologically basic. In naturalism, the physical matter in our brains is the basic state that determines our higher order mental states. In supernaturalism, the mental states themselves are basic and are not contingent upon any lower order states.
Supernatural phenomena are not usually described as occurring without any purpose or intention. Miracles are not just violations of physical laws, but are violations that achieve specific results, such as raising the dead or healing the sick. Teleology is the study in philosophy of goal-oriented, intentional phenomena . Mental states are very often characterized as being teleological. Our thoughts and emotions direct us towards certain ends. One consequence of defining the “supernatural” as having irreducible, lower order mental properties is that it implies that teleology is a lower order property as well. Take, for example, the supernatural view that God created the universe. Such a creation is not blind, mechanical, or accidental, but is done willfully, with discretion, and is directed towards a certain end. The concept of God is thus very teleological.
In fact, teleology is feature of almost all world religions today. As anthropologist David Eller (The End of Christianity, pgs. 274-275) defines religion across cultures: “Religion is that worldview, that paradigm, which sees nonhuman/superhuman minds/wills/intentions at work and which ‘explains’ events and legitimizes relations and institutions in terms of these beings and their wills.” Religions almost always involve underlying intentionalities at work, such as polytheistic gods who must be appeased to prevent natural disasters, or even more abstract intentions, such as Karma, which works towards specific ends of justice. The teleology common in world religions is a reason that we characterize them as being “supernatural.”
In contrast, “nature” is usually described as being unintentional and mechanical. Nature is blind. When something happens without any intention and only as the result of accidental causes, we describe it as a “natural” phenomenon. Take, for example, the difference between “natural” abiogenesis and evolution, verses “supernatural” creation and design. Abiogenesis and evolution are the result of accidental biochemical events. They are blind forces that drive biological change. This is very different from the notion that God created life and designed it in a specific way. Such a creation would be very teleological, whereas natural abiogenesis is non-teleological.
Defining the supernatural as “teleological,” however, raises the question of whether human agency and teleology is “supernatural” as well. I argue that this is not the case. The reason why is that human agency, as a higher order contingent property, is predicated on lower order non-teleological causes. We are the result of accidental abiogenesis and mutations. Even if our behavior is willful and teleological, we are still reducible to non-teleological causes.
In contrast, “supernatural” teleologies, such as God, cannot be reduced to non-teleological causes. God did not gradually evolve through accidental processes into a complex mental state with intentions. God is instead irreducibly teleological. God’s teleology is a basic, lower order property. Instead of using the term “supernatural,” then, to describe human agency, I propose the term “unnatural” in its place. If someone is struck by lightning, it is a “natural” death. If someone is killed by a human, we can call it an “unnatural” death. But since human agency is reducible to non-teleological causes, a human killing another human is not “supernatural.” If, in contrast, God or a demon killed a human, we could call this a “supernatural” death. The reason why is that God or demons are irreducibly teleological at the lower order level.
A final point of this discussion today is the question of where “superhuman” abilities fit in to the “supernatural.” Supernatural phenomena often involve superhuman capabilities. Miracles, magic, and witchcraft all involve powers that go beyond what humans are ordinarily capable of doing. But simply being “superhuman” does not mean that something is “supernatural.” This distinction can be justified based on whether the superhuman ability in question is contingent upon gradual, accidental processes, or whether it exists at the basic level.
For example, we discussed at the beginning a witch being able to cast a spell of fire in comparison to an X-man who has evolved the ability to produce flames from his hands. Both abilities are superhuman, but only one is supernatural. The reason why is the X-man’s superhuman ability can be explained as the result of gradual biological change and mutation, which eventually resulted in abilities that exceed those of ordinary humans. Likewise, even we humans today have capabilities (or at least technology) that supersede the abilities of ancient humans. If I were to go back in time and fly a fighter jet over ancient Rome, to the ancient Romans that might appear to be a supernatural event, but we would know that it was only the result of gradual changes in human capabilities. The process behind it is still reducible. In contrast, a witch did not “evolve” the ability to use witchcraft through gradual, mechanical changes, but instead witchcraft exists at an irreducible, basic level. The witch’s abilities cannot be described as being the contingent result of blind processes. We can therefore describe the witch’s superhuman abilities as “supernatural.”
To conclude this discussion I would like to summarize and compare the various attributes of the “natural” versus the “supernatural” that we have covered today. “Natural” phenomena tend to be physical and materialistic (or at least to supervene upon lower order physical states). “Natural” phenomena also tend to be uniform, and to exist within closed systems. Things that are “natural” are reducibly non-mental and non-teleological. Hence the “blindness” of nature.
In contrast, “supernatural” phenomena tend to be non-physical or immaterial. Or, if a supernatural phenomenon has physical components, it still tends to be influenced by non-physical lower order properties. The “supernatural” is irregular and involves phenomena that would not be possible ceteris paribus, or all other things being equal. Instead, the “supernatural” often involves external intervention that interferes with normal natural processes. As such, the “supernatural” implies an open system of causality. The “supernatural” also tends to be irreducibly mental, and to involve mental objects that cannot be broken down into non-mental parts. Finally, the “supernatural” tends to be teleological and directed towards certain ends, contrasting it with the blindness of nature.
These are just some ways to think about the differences between the “natural” and the “supernatural,” and I hope that my discussion here today has shed light on how we can define these terms, especially since it is a central issue for this year’s conference. Perhaps by knowing metaphysically and categorically what the “supernatural” really is, we may also understand our “Familiar Spirits” better as well.
 Incidentally, I am a mathematical nominalist, and thus do not think that mathematics consists of abstract, non-physical objects, as mathematical Platonists maintain. Instead I maintain that all mathematical knowledge is tautological and constructive.
 In using teleology as a criterion to contrast the “natural” from “supernatural,” I am referring to the Platonic sense of teleology and not the Aristotelian sense. The Platonic sense of teleology is more akin to Paley’s teleological argument for God, which maintains that the universe reflects a willful process of design that points to an intelligent creator. The Aristotelian sense of teleology pertains more to substance being directed toward certain ends, which can be formulated as “having the potentiality to change in a predictable way or to move towards a predictable goal.”
As professional philosopher André Ariew in “” (pg. 8) explains:
“Insofar as Aristotle’s teleology pertains to explanations of natural items, it is misleading to cast off Aristotle’s teleology as reading purposive behavior into natural events. This perception of Aristotle’s teleology is the result of conflating Aristotle’s naturalistic teleology with Plato’s … Plato’s natural teleology is and Aristotle’s is not creationist, anthropomorphic, and externally evaluative … Aristotle’s natural teleology is and Plato’s is not naturalistic, immanent, and functional.”
Ariew (pgs. 8-9) goes on to identify two distinct types of teleology in Aristotle’s writings:
“Teleological explanation in Aristotle pertains broadly to goal-directed actions or behavior. Aristotle invokes teleology when an event or action pertains to goals … [W]e can distinguish two distinct conceptions of teleology in Aristotle’s writings and at least two sub-categories:
I. Agency-centered teleology:
i. Behavioral. Activities undertaken for the sake of something, which may be either a state or further action.
ii. Artifactual. Activities undertaken for the sake of producing an object of a certain sort (artificial).
II. Teleology pertaining to natural organisms.
i. Formal. Biological developmental processes that occur for the sake of self-preservation or preservation of the species (form).
ii. Functional. Parts of organisms that are present for the sake of the organism possessing them.
I and II are distinct notions of teleology … Agent-specific teleology (I) is purposive, rational, and intentional, and represents external evaluation. The goal is the object of an agent’s desire or choice … Teleology pertaining to natural organisms is distinct: non-purposive (though seemingly so), non-rational, non-intentional, and immanent — that is, an inner principle of change. The goal is not an object of any agent’s desire.”
This paper makes the argument that agency-specific teleology (I) does not exist at the basic, irreducible level in naturalism, but does not argue that naturalism necessarily excludes Aristotle’s teleology pertaining to natural organisms (II). The agency-specific notion of teleology (I) also relates to my 4th criterion of “irreducible mental” properties, since I am specifically referring to the sort of teleologies exhibited by minds and agents.