The physical realm
Its extension
Items in this category are often called concrete or material. I’ll allow myself to use all those terms at times to avoid repetition. Yet the preferred adjective shall be physical.
Common-sense characterizations
of the physical realm seem natural, but unfortunately they are of little help to develop a more rigorous view. We’ll mention two:
Perceivability: "physical items are what can be seen, heard, touched, etc."
This view is likely fatal to any serious effort to differentiate between the mental and the physical. Perceptions are mental phenomena, so it seems peculiar to use those as characterizations of the physical realm. Kant, for one, was adamant that sense-data was "in our mind" and distinct from the thing in itself.
It also doesn’t sustain even a cursory examination very well. (a) I can’t see, touch or taste electrons, yet it would be peculiar to deny their physicality (if only because my own body couldn’t exist without electrons). (b) Electrons are still detectable via instrumentation. But we can only detect things within our lightcone. If there were structures outside of humanity’s collective lightcone (stars, galaxies...), it seems they should belong to the physical world all the same, even if they can’t ever be perceived. (c) Conversely, humans can see, hear, taste in dreams or hallucinations, although no one would argue those phenomenons are caused by physical objects.
Spatio-temporality: "physical items are entities located in spacetime."
This view has been serviceable enough during centuries; it amounts to equating the physical world with "spacetime + its content". Unfortunately, this doesn’t seem compatible with contemporary physics. (a) General relativity describes a curved spacetime with singularities such as black holes or the Big Bang; this does not fit well with the notion of spacetime as a medium "where" things exist. (b) The peculiarities of quantum ontology should also lead to phase out spatio-temporality as a foundational property. The standard model of quantum physics is set over the same 4D Minkovski space as special relativity, yet prospective successors use complex geometries under which our spacetime could be more of an emerging feature (therefore not suitable for a generic criterion). More fundamentally, the notion of location seems firmly anchored into a macroscopic view, which crucially can only emerge via wavefunction decoherence. Following some of these candidate theories, even spacetime could undergo quantization, putting macroscopic worldviews even further out of reach, and making this classic criterion totally unoperative.
Modern characterizations
rather focus on "physical properties"; in that representation a physical item would simply be an item with at least one physical property, and the physical world the collection of all such items.
The object-based conception: "A property is physical if it is the sort of property required by a complete account of the intrinsic nature of paradigmatic physical objects."
You may notice this is circular: to define what "physical" means, we need a pre-existing set of "physical objects", for instance a rock. The rock has the intrinsic property of being a rock, which (reassuringly) makes all rocks physical objects. Since we need to know the rock’s mass to accounts for its basic behavior, we can say that mass is a physical property too, thus all mass-having things are physical. To account for the atoms making up the rock holding together, we need electrostatic properties, etc. The whole conceit seems quite leaky in the first place, and in my view is invalidated by modern physics. The standard model predicts a number of particles, quantum invariants and modes of interactions that are thoroughly unrelated to the "intrinsic nature" of a rock. \(B^0\) bosons can only exist at energy levels very inhospitable to a rock (pre-electroweak symmetry breaking); they are chargeless except for "weak hypercharge" (for which all ordinary matter is neutral). Yet I need to make sure these particles belong to the physical world, lest my ontology yields some grave inconsistencies.
One may try to leverage the imprecise nature of this criterion and argue that all elementary particles are intrinsically essential to the rock, by virtue of their being bricks of a theory that explains why the rock exists. But that takes us a long way from the purported simplicity of this solution, and much closer to the next one.
The theory-based conception: "A property is physical if it is the sort of property that physical theory tells us about"
This is is circular again. Instead of positing that we already know some physical objects, we’re shamelessly positing that we already know physical theory. In the end, we need to accept this as the only solution we (or more likely, quantum physicists) cannot immediately poke holes in. It’s also quite an elegant one.
That being said, it comes with a jarring consequence: whatever we say about the extent of the physical world is false! Indeed, who would argue that contemporary physics is "finished" or "true" without qualification? As to future physics, we can’t assume anything about it. However, we can reasonably assume that our concept of the physical world is not just false in a random way, but rather incomplete. In general, a new physical theory would admit its predecessors as approximations of some sort, and every old physical property will either be re-used, or supervene on new ones. Therefore, "our" physical realm should be a proper part of our successors’.
Conversely, some "future physical item" would not qualify as such today given the state of physics, which may seem quite disagreeable. In principle, current mental phenomena or abstract entities could then migrate to the physical realm over time. This may seem to weaken the bedrock of objectivity that we’d hope to build our system of categories upon, but it’s important to remember it all hinges on human discovery, which is by essence incomplete and fallible.
A deeper probe into the consequences
Let’s now examine what comes out of accepting the theory-based conception. Placing ourselves in the contemporary setting, there are two fundamental theories: general relativity and the standard model of elementary particles.
General relativity (GR) represents the universe as (1) spacetime: a 4D curved Riemann space with negative metric signature and (2) matter content represented by a stress/energy tensor. Both elements are engaged in a dynamic relationship governed by Einstein’s field equation (\(G_{\mu\nu}=8\pi T_{\mu\nu}\), essentially: curvature = matter). This raises the possibility that what we interpret as matter is just a way of representing spacetime’s geometrical structures, thus making spacetime itself the fundamental physical reality.
The standard model of quantum field theory (QFT), on the contrary, uses a flat Minkovski space as a static background where quantum fields are defined. Elementary particles and fundamental forces are represented as local excitations of those fields. Quantum wavefunctions would thus be what characterizes physical content in that view.
Trying to anticipate, if not the "final theory", at least the next few steps, a family of prospective unified theories appears as the "heir" to each of these two.
In a GR-like perspective, this ultimate theory would successfully quantize Einstein’s theory and in the process on-board further geometrical dynamics that would, in "lower-energy" approximations replicate standard model quantum fields. Loop quantum gravity is the most developed attempt at the time of writing. The common denominator of all things, including quantum wavefunctions, would then be quantized spacetime geometry (which is distinct from "objects in spacetime").
In a QFT-influenced perspective, such as that of superstring models, spacetime itself could be an emergent property from the structure of quantum entanglement. Entities like strings or branes would then be characteristic of all physical content, independently of any spatio-temporal background.
Its structure: a universe full of things and stuff.
The first part of this discussion meant to define the extent of our first realm, the physical world. Our broad conclusion is that it encompasses the most fundamental objects from our best physical theories.
Yet conversations about the physical realm tend to not be about wavefunctions or spacetime geodesics. Instead they are about dogs, trees, cars, or people... what philosophers usually call "ordinary objects". On its face, it seems obvious that such things "should" exist in the ontology of our physical world, since not just our daily conversations but scientific discourse involve existence statements about non-fundamental objects such as stars, mountains, organisms, cells, rocks...
The common-sense response to anyone who would question that fact is that ordinary objects are made of the fundamental entities that we’ve already accepted in our ontology. My body is certainly made of elementary particles, as are my table, my pen, my clothes...so why worry? Well, one reason could be the status of more exotic entities, such as: "the sum of my nose and the Eiffel tower" or "the whole planet Earth, minus all monkey paws, wigs and Beatles records". Such baroque compositions may seem quite useless, except from a poetic perspective (if one feels generous). But is it at all legitimate to talk of them and to say they are?
As in our previous discussions of ontology, it’s important to keep in mind that deflationism is always an option: it may be that this whole line of questioning is of no philosophical import whatsoever, that it just describes a strictly verbal dispute. We shall consider this possibility in due course.