The causal criteria for being real

Ethan Siegel addresses a question on whether spacetime is real.

But there’s more to the Universe than the objects within it. There’s also the fabric of spacetime, which has its own set of rules that it plays by: General Relativity. The fabric of spacetime is curved by the presence of matter and energy, and curved spacetime itself tells matter and energy how to move through it. But what, exactly, is spacetime, and is it a “real” thing, or just a calculational tool?

After going through a quick grand tour of special and general relativity, as well as other physics, he comes to the conclusion that science can’t really provide an answer.

This question about whether something is real or merely a mathematical accounting convenience, is one that comes up all the time in science, and has throughout its history. When Copernicus published his theory of the Earth moving around the Sun instead of the other way around, many were willing to accept his mathematics since they made astronomical predictions easier, but insisted it was only a mathematical convenience, not reality. Max Planck, when he first introduced energy quanta into physics, only considered them a mathematical tool.

But for spacetime, I think Siegel actually answers the question in the quote above by paraphrasing John Wheeler (a physicist known for coming up with quick snappy terms and phrases): “Spacetime tells matter how to move; matter tells spacetime how to curve.” In other words, spacetime, whatever it is, has causal effects on things we can measure. It can affect and be affected by matter and energy. That, to me, is enough to consider it real in some sense.

That doesn’t mean it’s necessarily something fundamental. A number of physicists think it might be emergent from other things, such as time perhaps emerging from entropy, or space emerging from quantum entanglement. But just because something is emergent doesn’t mean it doesn’t exist. If it did mean that, then nothing would exist above quantum fields, and maybe not even them.

In this view, all that’s necessary for us to productively consider something real is that it participate in the causal chain that eventually effects what we measure.

This is why, in quantum physics, I generally consider the wavefunction to be modeling something real. Something causes the measured interference effects, and the various formalisms for modeling the wave dynamics accurately predict those effects. (Even if they only provide probabilities for particle positions.) That doesn’t mean the wavefunction is necessarily the complete story, or that it’s real in every respect, only that the overall phenomena is something that participates in the causal chain.

But this is also why I’m not a Platonic realist, someone who believes that abstract objects exist independently of the mind. In the Platonic view, abstract concepts are supposed to exist outside of time and space, be unchanging, and causally inert. Platonic objects, in and of themselves, do not participate in the causal chain. If we consider them to not exist, there’s nothing about them that forces us to reconsider that judgment. Any actual causal power they might have, only seems to happen through our mental models of them and the relations in the world that encourage us to form those models.

So, if something has causal effects, it is, at least in some manner, real. If it has no detectable effects, or at least theoretical ones, we can’t say conclusively that it isn’t real, but it may effectively not be real for us.

What do you think of the causal criteria? Does it miss anything real? Or does it include anything we commonly would say isn’t real?

78 thoughts on “The causal criteria for being real

  1. I think of reality as being a pairwise relationship. A exists to B if A can influence B. I guess this is the same as your causal chain. It does seem to imply that there could be cases when A exists to B but B does not exist to A, or A exists to B and B exists to C but A does not exist to C.

    Unless there is a god-like thing that influences everything, not everything is real to everything else, so there is no universal reality.

    However this does raise further puzzles because we are generally thinking A is something like an object or a person, and so complex and distributed across space and time; so what it means for A to exist to B is actually a bit more tricky than it might seem at first sight. Even whether A continues to exist as the same thing comes into question.

    Liked by 2 people

    1. I almost had a digression in the post about galaxies at the edge of the observable universe. The versions we’re seeing today can affect us by their light reaching us from over 13 billion years ago. However, those galaxies today are now beyond our cosmological horizon. The expansion of space is now moving them away from us faster than the speed of light. So for all intents and purposes, we’re now causally disconnected from them. Do they still exist for us? Do we exist for them? What about for galaxies far beyond the cosmological horizon where we’ll never have any interaction with them at all?

      What really makes my head hurt is that for those distant galaxies, relative to us, their time should now be going backward? What does that even mean? Relativity avoids paradoxes here because it’s impossible for us to ever interact with them.

      The same is true for someone falling into a black hole. For us, they slow down as they approach the event horizon, then freezes and become increasing redshifted. But for them, they cross the horizon without incident and continue moving toward the singularity. These are contradictory sequences, but again, physicists say it’s not a problem since us and the person who fell in can never compare notes. The urge to reconcile it into one reality is, apparently, misguided!


      1. “What really makes my head hurt is that for those distant galaxies, relative to us, their time should now be going backward?”

        Wait, what? Why?

        (FWIW, the redshift we see of objects falling into a BH is just due to how gravity affects the light waves coming from that object. The “freeze” is just the wavelength dropping to zero. It’s not in any way a “real” effect felt by the object, merely a matter of what distant observers see. There is no paradox there.)


        1. That’s my understanding of the relationship of something traveling faster than light relative to us. Is that wrong? If not, what is the time relationship between those galaxies and us? (My feeble attempts at the math just make the calculator spit out “ERROR”.)

          The black hole thing turns out to be something conjectural, black hole complementarity. I took Susskind’s description of it to be a general description of what general relativity said, but it turns out to be a speculative “solution” to the information paradox.

          My bad.


          1. Oh, I see what you mean. SR isn’t defined for velocity due to the expansion of space. SR forbids FTL (in several ways), which is why you’re getting an error. With a velocity faster than c, you end up trying to take the square root of a negative number, so things become imaginary.

            Liked by 2 people

    2. I think this is mostly right. But I think that if you say that A->B->C, where -> means ‘influences’, then you could think of A as existing for C also. Causal influence is presumably transitive.

      I think what you might be getting at though is that B might exist at different times, and maybe B’s influence on C comes from an earlier time than A’s influence on B.


  2. Causality itself is emergent, so if you want to use causality as a criterion of reality, you’d damn well better not deny that emergent things can be real. But it had better not be the only criterion of reality either. (The foregoing depends on reading “causality” as inherently asymmetric: if A causes B, then B doesn’t cause A. If you waive the asymmetry requirement, that allows a different description of the situation.)

    As for how spacetime affects matter: I think you’re right, but it’s tricky (warning: philosophy of science weeds! bonus: philosophy of science learning!)

    I have no problem with mathematical Platonism, as long as the Platonist doesn’t try to go all Tegmark about it and insist that every mathematical structure is “physically real.” And as long as they don’t go talking about a “realm” where abstract objects “dwell”. I’m rather fond of this argument: There are numbers between 3 and 7. Therefore, there are numbers.

    Liked by 2 people

    1. Interesting video. I especially like the point that something that has great leverage over the future is a cause while something that great leverage over the past is a record. I generally see information as causation, which fits with this view.

      What other criteria for being real would you add?

      That there are numbers between 3 and 7 doesn’t require mathematical Platonism to be true. It can be true in the sense that there are relations in the world involving between 3 and 7 entities, and this pattern occurs often enough that our brains model it. Numbers exist, but they don’t need to have an existence independent of physics.

      That said, when Platonists talk about realms where objects dwell, I don’t think they literally mean something like another universe with the objections floating around there. They’re trying to express a concept that is hard to put into words. Even Tegmark, I think, is trying to get across an idea that’s very difficult to put into words. The “physical” part I take to just mean “really exists”.

      Liked by 1 person

      1. My criteria for being real would center on being explanatory. Causal explanations are just one form of explanation.

        Stanford Encyclopedia of Philosophy says

        Platonism about mathematics (or mathematical platonism) is the metaphysical view that there are abstract mathematical objects whose existence is independent of us and our language, thought, and practices.

        –which seems like a pretty modest ask, and doesn’t mention independence from physics. But then the author later says “platonism entails that reality extends far beyond the physical world.” So go figure. I’m not sure what mathematical dependence on the physical would look like. If the universe had only one electron and no other particles, would that invalidate 2+2=4? It would make the equation a moot point, but that seems different.

        Liked by 1 person

        1. What would you see as an example of a non-causal explanation? Maybe mathematical relations? But what bring those relations into being? And what causes us to think about them?

          On the SEP article, you have to go through the entire thing. From section 1:

          Platonism is the view that there exist abstract (that is, non-spatial, non-temporal) objects (see the entry on abstract objects). Because abstract objects are wholly non-spatiotemporal, it follows that they are also entirely non-physical (they do not exist in the physical world and are not made of physical stuff) and non-mental (they are not minds or ideas in minds; they are not disembodied souls, or Gods, or anything else along these lines). In addition, they are unchanging and entirely causally inert — that is, they cannot be involved in cause-and-effect relationships with other objects.

          So if we view abstract objects as mental models constructed based on observed relations in the world, that’s not Platonic. The Platonic version are supposed to exist in addition to the mental models and physical relations.

          On the electron, if we think in terms of its wavefunction, maybe 2+2=4 would continue to have meaning, since the electron would likely spread out in an ever larger bubble. Assuming no dark energy, the electron would eventually be all over the universe. Of course, there’d be no one around to think about arithmetic.


          1. My first example of non-causal explanation would be explaining causality itself. That is, how it emerges from bidirectional physical laws plus the entropy gradient. Other emergence relationships, such as part-whole relationships, can also be explanatory.

            I don’t think anything brings mathematical relationships into being. Even if you deny abstract things exist, there has to be some physical feature not brought into being, even if it is only the whole sequence of events considered as a unit.

            I wouldn’t try to reduce mathematical objects to linguistic or conceptual ones, at least not in an asymmetric way. The reason why a word or concept refers to one thing and not another, depends on information theoretic properties of the word to world relations.


          2. So what should we call the symmetrical relationship between entities that exist before entropy makes it asymmetrical? Carroll just says “patterns” in the video, which I can understand but these are patterns which have a time sequenced relationship to each other that two other adjacent patterns often won’t have. Maybe that’s what I should be using as my criteria. Or maybe we can just say it’s causal asymmetry which is emergent?

            Liked by 1 person

          3. It seems like “patterns” underdetermines what we’re talking about, since there’s nothing that prevents a pattern existing consisting of elements that don’t have that relationship with each other. And the laws of nature are the rules that govern the time sequenced relationships, but not the relationships themselves. We could say “time sequenced relationships” but that’s a mouthful.

            In the absence of something better, I think “cause” still makes sense. We just have to understand that, without entropy, it’s fundamentally symmetrical.

            Liked by 1 person

          4. Sure, that’s a reasonable terminological choice. You just have to be careful which audiences you use it with, some will require extra explanation.

            In our universe, laws of nature govern time sequenced relationships (primarily?) In a logically possible universe – and perhaps in ours, if it turns out that time itself along with space is emergent – there may be other laws.

            Liked by 1 person

        2. I think the problem is that the word “real” is abused.

          Most people equate it with the physical. A tree in the park is real. It is very solid as I find if I inadvertently bang into it. A tree in my mind is not real in the same sense, although in the context of my mind it is real.

          The mathematical wouldn’t usually qualify for reality except to the extent that any mathematical truth probably has some representation in physical brain structure or process. However, approaching the mathematical truth at that level entails the loss of the meaning of the truth itself.

          On the other hand, if you take the view that the physical is really mathematical (it’s mathematical all the way down), then the mathematical would be the only thing that is real.

          But where does that leave Superman? We can talk about him as if he were real. We can read stories in comic books and see movies. In the context of the world of fiction, Superman could be real.


        3. Superman is real James, and so is the spaghetti monster; so are demons, gods, evil spirits, GR and spacetime because real-ness is a context. So at the end of the day, we do not have to endlessly debate what is real and what is not, all we have to do is identify the context in which some “thing” is real and then a consensus can be reached.

          Liked by 1 person

    2. Hi Paul,

      I’m a Tegmarkian, but I don’t always agree with how he describes the position. The way you describe it seems to suggest that at least part of your problem with it is how it is described. I think Tegmark’s ideas follow pretty much inevitably as long as you are willing to accept some sort of platonism, functionalism or computationalism about mind, and naturalism (by which I mean the idea that everything that happens does so according to natural laws describable mathematically).

      The way I would describe it, it is not that all mathematical structures are physically real. It is that the concept of objective, absolute physical reality is incoherent — physical reality only makes sense when construed as observer-relative. So, we shouldn’t think of Tegmark as claiming that somehow all mathematical structures are made physical, as if by magic. Rather the idea is that what we perceive as the physical reality of our universe is entirely explained by the fact that we are embedded in it. It is physical real only *to us*. In fact it’s just an abstract mathematical structure like any other. Mathematical structures without embedded observers are not physically real to anyone.

      I share your dislike of talk of abstract objects “dwelling” in a “realm”.


      1. People often state their own positions awkwardly, or they make mistakes which are separate from the main point and shouldn’t be used to impugn that point. So by all means, rephrase Tegmark.

        Well, I don’t accept functionalism about mind in general. I do accept it about “do this creature’s vocalizations *refer* symbolically to the world?” and “is this creature conscious?” But I don’t accept it about “what are the qualities of its consciousness?”

        On physical reality as observer-relative, it depends what you mean. Here’s what I do accept: when people say “physical”, that word gets its meaning from interactions between the language community and the rest of the world. So in *some* sense, physicality is observers-relative. Observers with an s.


        1. Hi Paul,

          I do disagree about the qualities of consciousness, but that’s another issue. It would indeed be reason to reject Tegmark if I conceded this controversial point.

          Just wanted to address the idea that it seems daft to think that mathematical structures somehow are made physically real as if by magic. Hopefully we’re more or less on the same page on this now.


  3. You’re speaking here, I take it, about physically real? As a contrast, in some sense, unicorns are real, because when I use that word, you know exactly what I mean. Likewise Sherlock Holmes or any known literary character.

    Physical reality beyond our horizon, those distant galaxies that can’t affect us, can affect matter in their vicinity, so they would satisfy your causal criteria.

    I can’t think of any exceptions at first blush, but in some sense the definition is circular. That which is physically real can have causal effects. I suspect a true definition of “real” will remain a philosophical and definition issue, but having causal effects is, at least, a property of what is (physically) real.

    Now what about someone reads Sherlock Holmes when they’re young and decides to become a detective as an adult because of that. Is that a case of something not physically real having a causal effect?

    Liked by 2 people

    1. I’m not sure if I’m following the circular point. It doesn’t necessarily seem circular to me. But maybe I’m missing something? I do agree that any definition of “real” is inevitably philosophical. (There are some scientists who take the attitude that only what they can directly measure is real, although that seems like a tough stance to hold consistently.)

      In the case of unicorns and Sherlock Holmes, it seems like they exist as mental models in our brains, the result of sensory impressions from numerous pictures, films, and books. Of course, a unicorn isn’t a real animal, and Sherlock Holmes was never a real person. But as concepts they are definitely real.

      Similar to Platonic concepts (actually identical to them), it’s often easier for us to just think of them as real in a non-physical sense, but I think that’s because the models don’t map to physical reality in the manner similar models typically do. So our model of a horse maps to reality in a certain way (it’s predictive of potential sensory impressions), but add a horn on its head and that mapping is no longer valid. But we still have the model of the horse with the horn.

      So it makes sense that a mental model can inspire someone to become a detective, particularly a model of a detective. Or we can just say Sherlock Holmes inspired them to become a detective, but we know that refers to an idealized model of a detective with phenomenal powers of logical deduction.


      1. What I’m getting at is summed up in the phrase, “Ideas have the power to move mountains.” Even newly formed ideas, so it’s not the shared history of unicorns and Holmes, but that mental content can have causal power.

        Maybe “circular” isn’t the right word… it just feels like ‘causal power’ is a necessary property of anything real. Like it’s just another way of saying the same thing, although maybe ‘causal power’ is a larger category since it includes mental content?

        Or whatever. To be honest, I don’t seem to have much capacity for abstract thought these days.


        1. I’m definitely on board with the idea that mental content has causal power. It is caused by incoming sensory information and innate impulses and has both short term and long term motor effects on the environment. So it’s part of the causal chain. So definitely ideas have a lot of power.

          Maybe instead of “circular” you’re thinking that it’s just trivially true? Could be. Although the fact that people debate whether things like spacetime, wavefunctions, or Platonic concepts are real seems to put pressure on that idea.


          1. “So definitely ideas have a lot of power.’

            Are you talking about the joule, as in 1 Watt = 1 Joule per second (1W = 1 J/s)? And if so, would some ideas be quantified as having more joules than others?

            Liked by 1 person

          2. Hmmm. Well, all causal power is ultimately the ability to exert changes on things, directly or indirectly, physical changes if we’re operating under physicalism. So I suppose we can imagine in principle trying to measure it that way. Not sure how we’d go about measuring the amount of wattage produced by democracy, even in its effects on a signal individual’s life.


          3. Okay, yeah, “trivially true” might be a better phrase. The notion of ‘causal power’ doesn’t seem, at least to me, to have much utility as a razor to cut between real and unreal if it includes unicorns and Holmes and mental content in general as all real. What would it define as unreal?

            FWIW, spacetime, because we’re clearly embedded within it and move through it, has some sort of physical reality even if we don’t fully understand how it works. All we can say about wave-functions for sure is that they describe an aspect of reality and allow predictions. (Some physics classes start by comparing the Schrödinger equation to Newton’s F=ma, and I think that’s a good comparison.)

            As I’ve mentioned in the past, I’ve come to resolve the Platonic question as a consequence of existence in a lawful physical reality. The canonical example, a circle or sphere, the concepts of which seem to exist whether we discover them or not, ends up as just an observation of how physical 3D space works. Given space, there is a notion of location and distance, then of equal distance from some location, and thus circles and spheres. All we can really say about math is that it describes physical reality (because it reflects physical reality).

            Liked by 1 person

          4. In all seriousness Wyrd, I don’t see how conversations like this one can be productive if there is no consensus on a fundamental definition of power. Is power an objective state of the world, something that we discoverer; or is power a subjective state of mind, something that we make up like GR, spacetime and the joule?

            To me, it seems like power is the impetus and the driving force responsible for causation in the natural world, including the dynamics responsible for the motion and form of imaginative thought and yet, nobody wants to investigate it let alone discuss it. Instead, everybody seems absolute content to play in the sandbox ignoring the elephant in the room. Personally, you and others might consider the notion of power as too abstract, and that’s cool.

            In many ways, the notion of power reminds me of Michael Mark’s latest essay. I think most people appreciate the notion of power from the artist’s aesthetic perspective, a perspective that is clearly repulsed by the pure rationalist approach. I’m just thinking out loud here, so nobody should feel obligated to respond.


          5. Well, as I said last time, just substitute “ability” or “capability” for “power” and the confusion goes away.

            As to the actual physics notion, “power” is a derived quality, not a fundamental one. As you said above, 1 joule per second and has units of kilogram-meters-squared-per-second-cubed. (Electricians know it as the more familiar volts×amps.)

            I don’t see it as too abstract; I see it as too derived to be a fundamental social or physical property.


          6. Clearly Mike; when anyone asks how much “power” the POTUS has, no one is not asking how many joules that he possesses. For all practical purposes, “power” is a mystery, some “thing” that can only be appreciated for its aesthetic beauty and not understood from a rational reductionist perspective. Is that how you perceive it?


          7. Can’t resist channeling the wonderful Emily Litella:

            “What’s all this fuss I hear about how many jewels the POTUS has? Why do we care? This isn’t a monarchy, we don’t have Crown Jewels. Why I doubt the man has any jewels at all. Maybe his wife does, but jewelry just doesn’t…”

            “What’s that?”


            “Never mind.”


          8. Lee, I wouldn’t describe power that way. Certainly political power is something most people don’t have a good understanding of. But it’s been studied. Richard Neustadt’s “Presidential Power” is worth reading for anyone who wants to understand the POTUS version. In the end, all social power (including business or political power) involves influencing people to do what you want them to do.

            Sometimes it’s easy, with a legal order for people duty bound to obey. More often it involves persuading people, either directly or indirectly. In truth, it’s always about persuasion. It’s just that when you have line authority, you have extra tools. But fail to understand that people are not mere extensions of your will, that they each have their own values and agendas, and you will eventually flounder.

            In contrast, power or forces in physics are much simpler, even though social power is ultimately a special case of it.


          9. Wyrd,

            I keep forgetting that your own confirmation bias is rooted in some form of Spinozaism; so my use of the word power would not have the same meaning to you as it does to me. My own conformational bias is similar to that of Kant; with some essential a priori intuitions added which Kant’s ontology did not contain.

            Spinozaism is a good metaphysical model, one that I agree with to a large degree with one important exception. Spinozaism posits the notion of natural and physical laws as irreducible and fundamental, whereas my metaphysics unequivocally rejects the entire notion of law all together. So for now, we will have to agree to disagree; and I will try to keep your own metaphysical position in mind whenever I correspond with you.


          10. Wyrd,

            I suppose a good place to start would be to ask if you agree with Mike’s manifesto: “that we shouldn’t look to reality for meaning. We have to resolve to make our own meaning, and figure out how to bend reality to it.” I do not disagree that his manifesto is an explicit and succinct definition of subjective experience but personally, I unequivocally disagree with that position. What say you?


          11. Regrettably, Lee, this isn’t a discussion I have much strength for. My New Year’s resolution was to swear off “fantasy bullshit” (FBS). Not as innately wrong — I love me some FBS sometimes — but as having become so very problematic in our culture.

            I’ve been terrified ever since this culture started blithely talking about, and normalizing, “post-factual,” and as of last November my terror seriously ramped up and then blew up on January 6th (and again last Saturday). Our culture has sunk into too many forms of fantasy.

            I’ve been reading, with growing horror, Aldous Huxley’s essays in Brave New World Revisited (1959) which he penned about 30 years after he wrote Brave New World (a profoundly disturbing novel in the current social climate).

            Huxley saw it back in 1959 (if not 30 years before). I quote: “A society, most of whose members spend a great part of their time, not on the spot, not here and now and in the calculable future, but somewhere else, in the irrelevant other worlds of sport and soap opera, of mythology and metaphysical fantasy, will find it hard to resist the encroachments of those who would manipulate and control it.”

            We’ve seen that play out the last decades and culminate in the last months.

            I feel as someone who has been badly beaten, my mind has been harmed by all this. To try to heal that damage, I’m sticking firmly to the physically real.

            So it’s hard for me to answer your question. For one thing, what is “meaning”? Is it that New Age thing people are always looking for? “Meaning” can only come from within (or maybe from God if you swing that way). Secondly, I’ve never known what to make of the idea of “bending reality” — does that mean magic or just building a thing with wheels? I’m a hard-core realist, both emotionally and philosophically. I’m just a tiny, tiny piece of a very large physical reality. I define sanity as the degree to which my internal mental model matches the external world I perceive, and life is the process of building and refining that model in an attempt to remove dissonance between it and physical experience.


          12. Like you Wyrd, I am dreadfully disturbed by the prevailing trend taking place in our culture and I appreciate you being open and candid about your feelings as well. We do not have to engage in any serious discussions here. I read your own blog from time to time, and if there arises an opportunity for a productive discourse maybe we could collaborate on common ideas and goals such as the origin of meaning and where meaning actually resides; if that is acceptable to you. Your definition of sanity is a very good one as well.

            Take care my friend


  4. I’ve had to determine my approach to this topic for my project (understanding consciousness), so I’ll just put it here and see what you think. I’ve decided it is useful to be very clear what the terms “exist” and “real” mean. These definitions will certainly conflict with someone else’s. All I can do is explain how I use the terms.

    So, I say something exists if it interacts with other stuff. Interaction is a relation, and so stuff that cannot interact with you (those far flung galaxies) does not exist for you.

    Patterns are real. (See Dennett.). So, abstractions are real. Numbers are real. Some patterns are discernible in existing stuff. All existing stuff exhibit patterns: specifically, patterns of behavior. A physical thing (system) exists if and only if it exhibits a pattern of behavior, and this pattern determines what a thing “is”. I should point out here that any pattern of behavior is multiply realizable, so even if something “exists” you can’t know what it fundamentally “is”. But you can assign a name, like “electron”, to anything that exhibits that pattern.

    Re causation: An interaction is best described in the format input->[mech]->output. You can then say the mech “causes” output when presented with input. This pattern (input->output) could be described as a causal power. The mech exhibiting this pattern of interaction has the “causal power”. So a pattern does not have causal power, but a pattern may be the particular pattern associated with a mech and describe the causal power of that mech. Again, more than one mech can exhibit the same pattern, and so have the same causal power.

    So to rewrite your penultimate paragraph, I would say if something has causal effects, it, at least in some manner, exists. If it has no detectable effects, or at least theoretical ones, we can’t say conclusively that it doesn’t exist, but it may effectively not exist for us.

    [taking questions]

    Liked by 1 person

    1. That all sounds about right to me. But I’m wondering if I missed something, because it seems very similar to what I said. You did add stuff about multi-realizability, which I don’t have any problems with.
      Or maybe I should ask, what would you say distinguishes your view from mine? (Assuming something does.)


      1. I recognize that we pretty much have the same understanding, but I am, and want you to be, more precise with the term “real”. For example, you said “this is also why I’m not a Platonic realist, someone who believes that abstract objects exist independently of the mind.” You say you are not a Platonic realist, but I say Platonic forms are real things independent of the mind. They just don’t exist, except some of them are patterns detectable in things that do exist..

        I say unicorns are real, they just don’t exist. I say philosophical zombies are real, but they cannot exist (as their description requires contradiction).

        I guess what bugs me is when people talk about “causal power”, and ask things like “does information have causal power?”. “Causal power” seems like it’s intuitive, but is ill-defined and causes misconceptions.


        Liked by 1 person

        1. Thanks for the clarification. I’m trying to see how we can make a distinction between being real and existing, but having a hard time. To me, those terms seem synonymous. (My working title for the post was actually “The causal criteria for existence”. I changed it to “being real” right before hitting Publish.)

          I do think we can make a distinction between ideas that most definitely exist which are about non-existent things like unicorns. Maybe that’s the sense in which you mean unicorns are real? If so, that seems strange, because it seems to imply concepts like the luminiferous aether or celestial crystalline spheres are real even though they don’t exist. I’m struggling to see how that use of language can be productive.

          Maybe if we say these concepts are abstractly real but not physically real? It’s all the same ontology, but different ways of talking about it.

          Liked by 1 person

          1. I am making a distinction between being real/existence by fiat. I’m saying, instead of using both words for the same thing, use one word for abstractly real (real) and the other for physically real (exists).

            This does bring up the question of what you mean when you say an idea of non-existent beings exists. But I translate that as saying an existent system in your brain recognizes the real pattern of a non-existent thing.

            Make sense?



          2. I follow what you’re saying. But I think it would be clearer to use those words with their common meanings and just use qualifiers to make what you’re saying explicit. So just preface with “abstractly” or “physically” for “real” or “exists”. If you use “real” in that fashion, it seems like you are obligated to constantly remind your audience of the special way in which you’re using it.


          3. >”I do think we can make a distinction between ideas that most definitely exist which are about non-existent things like unicorns.”

            I would argue differently. “Ideas” are the product of some specific species (humans) brain activities. They exist within that specific community, and, by extension, on any media they are recorded, if they could be deciphered (by other species?). Outside the mentioned group, “ideas” do not exist. If such species got into extinction, then their “ideas” got to extinction too.

            It makes sense to broaden this example and make a distinction between “reality” within and outside this specific group.

            Liked by 1 person

          4. It might depend on how we define an “idea”. A lot of mammalian and avian species, particularly social ones, can learn from each other. If a monkey figures out a new way to break open a nut, other monkeys will observe and copy. That troop will then have a cultural practice of how they break open the nuts that other troop lacks. In other words, culture, in the sense of shared concepts, isn’t unique to humans, at least unless we specifically define it to require symbolic communication.


  5. FYI In the nineteenth century and even the early twentieth, many scientists considered atoms a useful fiction, indicating that they weren’t real. This is why Einstein won his Nobel Prize for his work on Brownian motion, which was a physical manifestation of atoms/molecules that was definitive.

    Liked by 1 person

    1. Thanks. Didn’t know that. Pretty interesting. It’s amazing how often these useful fictions become real.

      I thought Einstein’s Nobel was for explaining the photoelectric effect, although I suppose it’s just as tied to atomism as well.


  6. Hi Mike, Wyrd,

    I definitely see the circularity, and it points to the fact that the concept of objective physical reality is empty and meaningless.

    If we make up an abstract toy universe with its own laws (as physicists will do with constructs such as Anti-de Sitter Space) then things will “play out” in that structure in something analogous to time and causality, albeit timelessly and causelessly from our perspective. But if you imagine a perspective within that universe, objects within that universe will appear to be real because they appear to engage in causality, whereas our universe will appear to be unreal and abstract because it does not. The circularity is that must presuppose that our perspective is objectively privileged, that what we observe is physically real to decide if it is in fact engaging in causality.

    If you make that assumption, then you foreclose the possibility of there being universes just as real as ours which are entirely causally disconnected. Whether or not such universes might exist, it seems unreasonable to rule them out a priori simply by defining them out of existence.

    Liked by 1 person

    1. Hi DM,
      You’re getting at why I hedged a bit toward the end, saying something might effectively not be real for us. We don’t even have to bring in other universes, just our own universe far beyond our cosmological horizon. (Which Tegmark actually considers another universe, so I guess I’m converging.)

      Is a galaxy a trillion light years away real for us? I suppose if cosmic inflation happened we could say we might still feel causal effects from the energy patterns that eventually became those galaxies. But what about galaxies 10^100 light years away? (Assuming such galaxies exist.)

      And somewhat tying in with the previous post, in a simulation, the simulated objects have simulated causal effects and are effectively real for any simulated entities within the simulation. But for those of us outside the simulation, they’re not.


    1. Good question. A naive answer might be something like having direct conscious interaction with a phenomena to determine things about it, particularly quantitative properties. But of course, in modern science that rarely happens. No one has ever seen an electron. Instead we have direct perception of a stand in, like a readout on a measuring device, which we use to infer things about the phenomena. We do this because we trust our theory about how the device works, but ultimately it’s an inference made using theory.

      However, before we allow ourselves to get too upset about this, it’s worth noting that direct conscious interaction is itself an inference based on preconscious sensory information coming into the brain. Those inference are themselves heavily dependent on our understanding, our model or theory of the world.

      In the end, we make predictions, note the errors and adjust, and make new predictions.


  7. Is there a sense in which our everyday concepts of time, space and causality are secondary, and behind the scenes it is quantum entanglement that is more fundamental, so that what is (potentially) real to us is everything with which we are entangled?


    1. It’s a definite possibility, particularly if you subscribe to the idea of a universal wave function, that is, a quantum universe with no Heisenberg cut. Of course, that view implies many worlds, so most people reject it out of hand.


  8. Your previous post was on a simulated universe and this one is on what is real.

    Is a simulation real? I mean real in the sense that it is more than real as a simulation. Would simulated consciousness be real also in the sense of more than real as a simulation?

    Liked by 1 person

    1. I think the contents of the simulation would be real for any simulated entities within the simulation. Simulated wetness for a simulated being would be real wetness. Simulated pain would be real pain.

      For us on the outside, they would be real in the sense of being a real simulation. Of course, you could arrange for the simulated beings to have access to physical robot bodies in the outer world, which would graduate them from just simulation status to something much more real.

      Liked by 1 person

        1. I don’t know. Would it? I suppose if you reject that a simulation of consciousness can be conscious, it might be. Essentially it would be a philosophical zombie. I personally don’t think p-zombies exist, so for me giving it a physical body makes it as physically real as we are. You could, of course, then argue that it was always physically real, since it was always implemented by some kind of physics.

          Liked by 1 person

          1. Would the simulated consciousness using a simulated body execute the exact steps and processes that the simulated consciousness executes with a real body?

            If the steps/processes are identical, why would one be more real than the other? They would be indistinguishable.

            Liked by 1 person

          2. Hi James,

            > If the steps/processes are identical, why would one be more real than the other?

            It wouldn’t be realer from an objective point of view. But I’ve been arguing that what is physically real is a matter of perspective. Putting a simulated consciousness in a robot body may make it physically real to us for this reason.

            If it’s just running in a simulated world and does not interact with the real world in any way (e.g. the program takes no inputs), then from our perspective it isn’t physically real. I actually think we have no moral responsibility for beings in such a simulation (because I’m a Tegmarkian platonist and I think the worlds we are simulating, no matter how horrific, must all exist out there in the multiverse independently of whether or not we want to explore them with simulations — our simulating them creates no additional suffering).

            But as soon as you start interacting with a simulated consciousness, then you are a part of the mathematical world it inhabits. You are physically real to it and it is to you. You are in the same relationship to it as you are to any other physical being, and so I think you do have moral responsibility for it.


          3. So uploaded minds to a computer wouldn’t be real?

            But if the uploaded mind somehow instantiates itself in a physical body then suddenly it becomes real. Because perspective. But the only perspective would be our own perspective or the perspective of non-simulated mind. So in the end it is only our mind that makes it real.

            I don’t know.


  9. Hi Mike,

    I meant to give an overall comment rather than just responding to comments of others but didn’t have time.

    I think you raise some very interesting issues and so I enjoyed and appreciated the article very much.

    But I think it is a mistake to get too hung up on what is and isn’t real — a mistake many philosophers have been making for too long in my view. What is real and what is not depends on what you mean by “real”, and different definitions are appropriate in different contexts. As long as you are clear about what you mean there is no problem. I don’t think there is a fact of the matter on which definition is correct and so what is “really real”.

    The question of whether some concept refers to something real or is just a calculational tool strikes me as entirely meaningless. I genuinely cannot make sense of it.

    The closest I can come is to take the examples of calculational tools from the past that are since discarded, such as caloric theory or geocentrism/epicycles. From my point of view, to the extent that these theories disagree with experiment (as caloric theory does when it claims that caloric is a gas), they are not physiclaly real and that’s all there is to it.

    If on the other hand they can be patched and amended (e.g. epicycles within epicycles) and so made to agree with experiment to the point where their predictions cannot be falsified, then they are as real as any other model but fail to be as useful or elegant as simpler theories. So in my view, it is possible for both geocentrism and heliocentrism to be true (i.e. there is no fact of the matter on what is actually at the centre), with the latter only being far more elegant and more useful.

    A more current debate is which of the Newtonian framework or the principle of least action is the more fundamental description of physical law:

    In the Newtonian framework, we have objects a certain point in time and space, and laws that describe how they evolve over time. In the framework of the principle of least action, the rule is that some quantity (the “action”) is minimised or maximised, and what happens will be whatever achieves this. The latter ends up being more apparently teleological and less intuitive to humans (but more intuitive to the Aliens in Arrival or Ted Chiang’s original short story “Story of your Life”), despite being very useful mathematically in some circumstances. But the two frameworks are mathematically equivalent, in that you can derive one from the other. So the question arises, which is prior? Which is the “real” one and which is the derived one? In my mind, there is no answer to this question. Each framework is equally real.

    Causal criteria for what is physically real to us makes sense to me, with some caveats.

    I think it’s better to require the relationship to be bidirectional. Our far future descendents can exhibit no causal influence on us, but we should consider them to be real in some sense or else we have no more responsibility for them than we would for fictional characters, and that doesn’t seem right. The fact that we can causally influence them makes them real, I think.

    Adopting this rule has some other benefits. We can trace the chain of causality backwards to the Big Bang and forwards again to parts of the universe that are no longer causally connected to us. So those parts of the universe are also physically real. That also seems right.

    But I think you should be aware that adopting your criteria might (like me) commit you to the physical reality of epicycles and geocentrism, assuming that epicycles can be used to make correct predictions. If so, then they can be said to have as much of a causal influence as spacetime or the wavefunction or whatever. If you want to exclude them, you may need to introduce an additional criterion that no simpler or less ad hoc model can produce the same predictions. But then we’re back where we started. Perhaps spacetime does not exist because there is a simpler model that can produce the same predictions, and it becomes an open question whether spacetime is real or just a calculational tool.

    Liked by 1 person

    1. Hi DM,
      I know where you’re coming from. It’s basically why, until recently, I was comfortable calling myself an instrumentalist, although I recently covered why I’ve become uneasy with that label. (Too much baggage, with people projecting positions on me I don’t hold.) But I think it’s important to be able to put on the instrumentalist hat at times and assess theories in that light.

      I would note that I don’t consider causality to be the only criteria for reality. Aside from making more accurate predictions, parsimony also comes into it. That’s how we can dismiss geocentrism today. It’s just a much more complex theory given all the data now available. It’s always possible to add variables to any theory to make it consistent with observations, but at the cost of increasing complexity. If there is a simpler theory, one with fewer assumptions, then it has a better chance of remaining reliable.

      I do think asking whether something is real or just a mathematical convenience is productive though. We shouldn’t expect a mathematical tool to necessarily reconcile with other theories, or worry if it outright contradicts them. It’s just a tool, so no one should care. For example, the weak (epistemic) Copenhagen interpretation is basically an anti-real theory, so any concerns about its contradictions with cosmology would be misplaced. (Stronger more ontological versions of Copenhagen are a different matter.)

      But if it does reconcile with other well established theories, then that seems to increase the chance of it being real. Of course, that’s always with the possibility that a cluster of theories that reconcile with each other may constitute a paradigm that eventually ends up being overturned.

      In the end, we have theories that make more or less accurate predictions. If it’s the simplest theory, it may be reliable. And if its components are compatible with other reliable theories, they may be “real”. At least until a better theory comes along. It seems to be the best we can do.


  10. The problem for me with concepts of “reality” is that the great majority of what scientists talk about can not be “experienced”. At least not by mere mortals such as myself without any skill in mathematics beyond the ten times table.

    I can readily experience (most) human qualia and emotions. I can understand the colour red in two ways: firstly in the scientific explanation of light waves of a certain frequency but secondly (and very importantly, to me at least) in my qualitative experience of colour.

    I can not experience the warping of spacetime. Nor the minute particles of matter of which (it is said) everything is made. As a very pedestrian mortal I have five very limited senses and can only really understand and accept what I can personally “feel” and “experience”.

    Perhaps these shortcomings are limited to just myself. Perhaps once Elon Musk has implanted the necessary extra processing chips in my brain I will be able to feel such matters as easily as I can feel and understand light or heat.

    Bring it on, Elon.

    Liked by 1 person

    1. I’m tempted to do an appeal to the stone and point out anytime you’ve fallen on to the ground you’ve experience the warping of spacetime, but of course the idea that that is the warping of spacetime is what is so far outside of our experience. And we only experience photons and electrons en mass, such that the idea they were composed of units was controversial for a long time.

      The thing about Musk implanting extra processing chips is, if it becomes common, we might someday wonder what it was ever like to only experience reality with natural senses. Of course, that could also be when we start experiencing simulated reality. (Assuming we’re not already experiencing it.)


  11. Space, time, space-time, particle, laws of physics – All that are physical terms. Our understanding and our discussions are based on variations of that physical view of the Universe.

    I think, there is a a new sheriff in town, so to speak. Please look at this article “ (New machine learning theory raises questions about nature of science).

    Here is a long quote from this article. -“The algorithm, devised by a scientist at the U.S. Department of Energy’s (DOE) Princeton Plasma Physics Laboratory (PPPL), applies machine learning, the form of artificial intelligence (AI) that learns from experience, to develop the predictions. “Usually in physics, you make observations, create a theory based on those observations, and then use that theory to predict new observations,” said PPPL physicist Hong Qin, author of a paper detailing the concept in Scientific Reports. “What I’m doing is replacing this process with a type of black box that can produce accurate predictions without using a traditional theory or law.””

    What we see here? There is a new way to describe the Universe and its contents without a use of physical terms and laws. That has implications on how we define “real”, “existence”, the underlying laws of the world, and so forth.

    Liked by 2 people

    1. I’ll have to check out that article, but making predictions with a black box seems like it will have limited utility. One of the benefits of actually having a theory, a model, is that we can then apply it for various purposes, like technology. Having an oracle just make predictions won’t do that. Although it might provide a useful test for possible theories.

      Now, if the black box can produce actual theories, then we might be on to something. But then we basically have an AI scientist.

      Liked by 1 person

    2. I think I linked to the paper you reference on another thread.

      I also had a blog post on this. Quote from paper:

      “We discuss a possibility that the entire universe on its most fundamental level is a neural network…This shows that the learning dynamics of a neural network can indeed exhibit approximate behaviors described by both quantum mechanics and general relativity. We also discuss a possibility that the two descriptions are holographic duals of each other”.


Your thoughts?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.