There is a famous question by Einstein which was reported by his biographer, the physicist Abraham Pais, and which expresses his concern with quantum physics:

We often discussed his notions on objective reality. I recall that during one walk Einstein suddenly stopped, turned to me and asked whether I really believed that the moon exists only when I look at it.

However, it is interesting that Einstein's intuition is problematic in Special Relativity as well.

When we look at the moon we see it one second into the past, so to speak.

But what do we mean when we ask if the moon exists or assert that it does?

We may mean that it exists as an idea in our minds, or as an object in a so called block universe in which past and future and present all exist in some unintelligible way.

But we naturally and intuitively mean that it exists Now!

But alas, special relativity does away with the concept of absolute simultaneity, and as a result it renders physically meaningless the idea of a metaphysical or "real" Now-somewhere-else.

1) In what sense can we ask if the moon exists, if it is 1 light second away?

2) Is there a notable discussion of this problem in philosophy?

Note 1 - I am aware of the Andromeda paradox and the block universe view, but I believe that they fail to resolve the problem — if past, present, and future, all have the same status of existence, then that kind of existence is doubly unintelligible and cannot correspond to our intuition of existence as in a thing that exists-Now.

Note 2 - this question is a "response" to @ChrisDegnen's take on this problem. I believe he got SR wrong, but nevertheless his take (which I do not subscribe to) seems prima facie valid — that there exists an absolute plane of simultaneity even if this plane has no place in science and physics.

  • I think this can be put into: Do the objects of our perception exist or only what physics tells us to be in the same space-time? Or another take: What are the intakes on the notion of "existance" by naive and scientific realism? Well, it is clearly nonsense to declare everything nonexistant only because presentism is scientifically questionable. Our talk does not depend on what physicists tell us to be 100% accurate. Or you would necessarily endorse scientific realism, which would lead to reductionalism (or dualism).
    – Philip Klöcking
    Jun 9, 2016 at 8:56
  • We can ask "if the moon exists" - independently/regardless of how many seconds away it is. If we know that the moon has existed in the recent past and most likely will exist in the near future, then it exist Now!
    – Guill
    Jun 13, 2016 at 20:01
  • @Guill, 1) so are you saying that we can say it exists Now, even though we do not know what Now somewhere else means? 2) what if it is not the moon, but an astronaut doing a highly risky maneuver on Mars, 12 light minutes away (see my comment to Hurkyl's answer).
    – nir
    Jun 14, 2016 at 6:47
  • My use of now, is my local reference frame. If I am interested in now-somewhere-else, all I need is the distance (time) to the new frame of reference. For Mars, if I send a command now, it will be delayed 12 minutes, and when I receive a reply, I know that it was sent 12 minutes ago. So I can know "my now" and "their now."
    – Guill
    Apr 15, 2017 at 7:26
  • bad question title which refers to the body of the question
    – user25714
    Apr 24, 2017 at 16:23

6 Answers 6


But we naturally and intuitively mean that it exists Now!

I don't know about you, but my intuitive notion of "now" is heavily tied to a system of clocks. Each clock I reference basically defines its own local time domain basically independent from all others, connected only by the readings given by the clocks.

What constitutes "local" can occasionally get stretched; e.g. when speaking on the phone or chatting over the internet, if the communications are sufficiently low latency, I could consider the person I'm talking to be covered by my local time domain.

Furthermore, my notion of "now" becomes distinctly murky when lacking clocks of sufficient precision and accuracy, or when clocks disagree. e.g. if there was a one second latency between chat messages, the extension of my local notion of "now" to cover the recipient is not meaningful at a resolution of less than a few seconds.

The idea that there is an objective, global notion of "now" that the various local time domains aspire to measure is a pleasant and convenient one for organizing, simplifying, and understanding our experiences of time, but it is not a necessary one.

Special relativity (and general relativity even moreso) strongly suggest that such a notion of "now" doesn't hold up when stressed.

My general opinion on such conflicts is that we should seek to refine our intuitive notions, rather strictly adhere to a rigidly extrapolation of things beyond our direct experiences.

And, in fact, there is a very natural refinement. Going back to my phone call example, we might consider a counterfactual situation where there is a distant person on the moon I'm chatting with over the phone.

Because of the latency of the communication, my intuitive notion of "now" is a bit fuzzy; now is somewhere between when the other person spoke the sounds I'm hearing and when they'll hear the sounds I'm speaking.

This corresponds exactly with the special relativistic notion of a space-like separation between events.

  • I think it's only frames moving at relativistic relative speeds that really "stress" simultaneity? For small relative speeds, such as between earth and moon, no matter how spatially distant, we can still say "he spoke exactly 3.2s (or whatever) before I heard", preserving a global "now". I'm also reminded of an analogy to the challenge of sound simultaneity for large orchestras in fast passages (a.k.a. ensemble), which is why conductors are needed (the "sound distance" between opposite sides of the orchestra becomes significant relative to the musical notes).
    – Jeff Y
    Jun 9, 2016 at 16:04
  • @JeffY, I think you got SR wrong. the problem of simultaneity arises with distance. I don't need another observer to know there is a problem. it is enough that I contemplate what an imagined observer would observe. Also I do not understand how someone having spoken t seconds ago preserves a global "now". as far as I can tell it does not.
    – nir
    Jun 9, 2016 at 19:07
  • 1
    @Hurkyl Here is an interesting article on how Einstein proposed synchronizing any clocks in the same inertial reference frame: en.wikipedia.org/wiki/Einstein_synchronisation
    – Jeff Y
    Jun 9, 2016 at 20:34
  • 1
    @JeffY, suppose you contemplate what is happening "right now" on Mars, 12 light minutes away. you do not need to have any velocity relative to Mars, nor anyone else in relative velocity to you, to conclude that your postulated "synchronized-clocks-now" on Mars has no metaphysical significance — that it is not a "real" now. the "synchronized-clocks-now" is the relative simultaneity now. you can imagine a super-alien kindly letting you know that they have in fact found the absolute simultaneity plane and it so happens that its reference frame is moving at a terrific speed relative to you
    – nir
    Jun 11, 2016 at 8:43
  • 1
    @JeffY, rendering your "now" at Mars, off by say 10 whole minutes. in short slow relative speeds have nothing to do in rendering your "synchronized-clocks-now" meaningless when it comes to the question at hand.
    – nir
    Jun 11, 2016 at 8:44

An alternative to the block universe is a universe that is local in both space and time. You don't see the moon, your eyes interact with photons whose world lines intersect (and end at) the photoreceptors in your eyes. These events are happening here and now. That there has been, and will be, a moon "out there" is really just a particular pattern over the events at the "heres and nows" of your worldline. If you go down this route, you can probably find flavors of information based physics, idealism, solipsism...

  • are you saying that the moon most certainly exists "out there", but that we cannot say that it exists right now?
    – nir
    Jun 9, 2016 at 19:28
  • I'm saying: "it" exists now, but not "out there" -- with the following thing: the"it" is the photons that hit your eyes that give you the impression of the moon some distance away. Physics is local (in the technical sense). You don't interact with the moon; you interact with photons that are in your local vicinity.
    – Dave
    Jun 9, 2016 at 20:24
  • but we know that there exists an object in the real world that is the source of these photons. or don't we?
    – nir
    Jun 10, 2016 at 8:05
  • @nir If you make that inference you are back to the block universe -- the photons that hit you now are from the past (in the naive sense of what time is) of the moon, from there you can go into all of the relativistic stuff you want. My "answer" is just to point out that you can keep "now" (to a good approximation) if you constrain yourself to only what is local to you.
    – Dave
    Jun 10, 2016 at 13:17
  • @Dave I don't think physics,science, has any business worrying about metaphysical notions of existence and such. It never was necessary and with quantum mechanics isn't even possible (physical properties don't have values outside of interactions/measurements.) Or are you saying basically the same thing? Aug 10, 2016 at 16:17

I believe that there is only one sensible answer to the first part of the question, i.e.

1) In what sense can we ask if the moon exists, if it is 1 light second away?

To answer this question you must first specify a criteria for 'existence' in terms of measurable/observable quantities. It needn't be a completely general criteria just enough for whatever situation is of interest. For what observations or series of observations is the use of the term to be allowed.No matter how vague our use of the term generally is it is still just a term and it's use is set by us. People go about deciding if this that or the other thing exists all the time. Horses exist because we can find them. Unicorns don't exist because we haven't. Yes relativity and it's time delays add some additional complications but ultimately it's word usage that is the issue. Define the term in accordance with your needs and proceed, should be sufficient. If you are concerned with pre-established usage then an empirical study will tell you the range and frequencies of various conventions.


This answer specifically addresses the common temporality of existence of the observer and the moon

In special relativity observers' experience of what is present looks something like this:

enter image description here

source: Is There an Alternative to the Block Universe View?

It doesn't look very common at all. So you might say why should I care about observers' frames. They're all about events and movement relative to the speed of light. The light from the moon is seconds old but we're talking about the 'present' - what's happening on the moon right now.

The frames in the diagram above are based on spacetime diagrams which depend on observers and events set relative to the speed of light

enter image description here

source: A Most Incomprehensible Thing

Surely we're not concerned with things relative to the speed of light. We're talking about the moon in the present instant.

The thing is, one needs to update the idea that there is an 'instantaneously' connected field of points constituting the present by appreciating that the speed of light is itself instantaneous, i.e. at the speed of light no time passes.

One cannot view relativity as if the speed of light is a speed like the speed of sound, beyond which one could go faster, eventually to 'instantaneous' in the limit. The speed of light is absolutely the limit, and so the 'present' is relative. There is no underlying unobservable or superluminally connected present. Mind blown.

So the observer, the moon and everything else exists in temporally simultaneity, but only at the speed of light. Below that speed everything is pulled apart in time and space.

In fact, at the speed of light there is no past or future either, so it wouldn't be right to call it the present - only Now.

Elaboration on the speed of light from someone's blog

Light travels with light velocity “c”. If we do not move relative to the earth, the travel time of the light from the upper atmosphere to the moon for example is about one second. We know this because if it hits a mirror on the moon, it will be back on earth after two seconds. However, these seconds is what it takes us to wait for the light. How long does it take the light to go anywhere?

If we travel along that same path, the time that it takes us to get to the moon becomes shorter and shorter the faster we go. “Of course” you may say, “because you are faster and faster.” But that is not what I mean. What I mean is that we will experience a travel time below one second before we even reach the velocity of light relative to the earth-moon system. This is due to time dilation.

In fact, we would experience about one second of travel time between earth and moon, if we moved with a velocity v that equals light velocity divided by the square root of two: v=c/√2. At 90% light velocity, i.e. at v=9c/10, our travel time will be only a third of a second! At 99.9% of the speed of light, the travel time we would experience has reduced to a thirtieth of a second, or 33.3 milliseconds.

In fact, although we cannot ever reach light velocity, we would soon even by pure experimentation find out that if we could reach light velocity and travel along side by side with the light, time would stop and the travel time between any two points, even between here and the edge of the observable universe, would be exactly zero. Relativity theory tells us that light has no time at all to exist because it moves at the speed of light.

source: http://www.science20.com/alpha_meme/fundamental_nature_light-75861

  • 2
    it is not clear what you are trying to say
    – nir
    Jun 9, 2016 at 16:10
  • I have added statements for clarification; the subject matter itself is quite mysterious though. Jun 11, 2016 at 6:01

Most fundamentally and in specific response to your request..

But alas, special relativity does away with the concept of absolute simultaneity, and as a result it renders physically meaningless the idea of a metaphysical or "real" Now-somewhere-else.

That's the trouble with metaphysical notions, by definition they aren't supportable by empirical results.

1) In what sense can we ask if the moon exists, if it is 1 light second away?

In the same sense we always have, that we have empirical experiences such as we describe. There is no scientific or empirical problem here only worrying about something that was outside of experience to begin with which is why you referred to metaphysics. There is no philosophical problem either if you don't posit such metaphysical constructs with "metaphysical existence."

2) Is there a notable discussion of this problem in philosophy?

Probably but it is unfortunate. The lesson from Special Relativity here is that such speculative notions just don't hold up and can be abandoned.

  • You write: "That's the trouble with metaphysical notions, by definition they aren't supportable by empirical results." how's that? the Newtonian intuition of having a snapshot-Now or metaphysical-Now across the universe seems to be "contradicted" by SR. and being contradicted is as good as being supported. Imagine someone in 1900 saying that what we call space is not euclidean by nature. would he not be right in saying 18 years later that GR supports his claim? or consider Chomsky: "Science is metaphysics. It's talking about what the world is made of" - youtu.be/cQd6QGQIxmQ?t=7m15s
    – nir
    Aug 8, 2016 at 5:58
  • You write: "In what sense can we ask if the moon exists, if it is 1 light second away? In the same sense we always have, that we have empirical experiences such as we describe." I do not understand what you mean. Does a galaxy 1 billion light years away exists because we can Now see the light that it emitted so many years ago? what sense is that?
    – nir
    Aug 8, 2016 at 6:00
  • You write: "Is there a notable discussion of this problem in philosophy? Probably but it is unfortunate. The lesson from Special Relativity here is that such speculative notions just don't hold up and can be abandoned." Given that from my perspective your answer at this point does not make sense philosophically, this final claim of yours appears somewhat unjustified and condescending. Before you throw philosophy of science and metaphysics down the dustbin, please try to make your own answer clearer.
    – nir
    Aug 8, 2016 at 6:03
  • @nir Newton's formulation of assigning a uniform "t" (time) throughout space was found to be inadequate for empirical predictions. There is no need to attribute any metaphysical notion of "existence" of anything here. The shift from euclidean to Riemann geometry is the same. The person in 1900 trying to claim non Euclidean geometry wouldn't have much evidence to support the claim. If they had formulated empirical claims they they might have scooped Einstein. Chomsky was wrong or rather naive. Science builds models of how the world works, predicts future observations from earlier ones. Aug 9, 2016 at 11:10
  • @nir To the extent scientists refer to physical objects etc, these are just part of the formulation involved in predictions. Fundamentally physical properties don't have actual values outside of observations/interactions so metaphysical claims of 'existence' aren't supported. If a 'metaphysical' claim can be settled by experiment then it's a physical theory and should be presents as such. Aug 9, 2016 at 11:23

A more fitting example for the question of existence would be a rainbow, I think. Do rainbows exist? I believe they don't, as their characteristics only occur if an observer is in the right position, and looking at the right spot (pedantically this is also true for the moon, but as the tides also occur in daylight, it could be argued that it exists 'more' than a rainbow).

Anyway, when we ask about existence, it tends to be tied to a given combination of concepts (whether they themselves exist or not), and I believe that is all existence ultimately is - a series of concepts, nested within and linked to one another like a spider web. Is there a tangible link between two concepts, or not? Can something exist in one scenario, but not another? Can something stop existing for some people, but not for others? A rainbow can do these things, and I think it's a solid example of the fluidity of existence.

If most of space is nothingness, does nothingness exist? Within the concept of 'space', sure it does, but by definition it cannot. Seems like a paradox, unless we are saying that the concept of nothingness exists, when tied to space, but that it defines something that represents the antithesis of colloquial existence, when tied to most other things: If I have no apples, those apples do not exist. If I have nothingness through the majority of space, that nothingness does exist.

A strange set of rules, for sure, but I believe it comes down to what makes sense to our minds, and fits within the scope of our collections of concepts. The moon exists as it fits nicely into our concept of reality, and until it fails to do so any longer, it will continue to exist for us in reality.

The same applies to stories, I think. As long as they are internally consistent, we're happy to assume their aspects exist 'within the story', but if something is too distant from the concepts being told, our immersion will be broken as we won't believe such a thing 'could exist' in the story being told.

This is a bit of a ramble, but I guess I'm saying that I think existence is whether or not there are links between concepts in our mind, and that as long as there are, whatever those concepts are, they exist at least for us, until we don't believe in them anymore.

As to existence in the physical realm, I think that's a bit more concrete, but as always, we have no way of really knowing. As long as our worldviews still operate successfully, with any assumptions we make about existence, our mind is doing its job correctly. Assuming anything further than that leads into unquantifiable speculation, and no real payoff.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .