23
$\begingroup$

This question already has an answer here:

Can someone please explain concept of spacetime in simple language? What is it and how it is important in the universe?

Wherever I have tried searching this concept, I have come across most complicated explanations.

A simple example will be appreciated.

$\endgroup$

marked as duplicate by Kyle Kanos, ZeroTheHero, Yashas, John Rennie, sammy gerbil Apr 14 '17 at 12:48

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • 5
    $\begingroup$ It's a long explanation. You can take help from Einstein's own book: Relativity, The Special and the General Theory from archive.org/details/cu31924011804774 $\endgroup$ – Wrichik Basu Apr 10 '17 at 6:30
  • 1
    $\begingroup$ Possible duplicate of physics.stackexchange.com/q/41385 $\endgroup$ – Shufflepants Apr 10 '17 at 15:10
  • 2
    $\begingroup$ There is no simple explanation, if indeed there even IS an explanation. (I mean that we can describe how spacetime behaves, but as for what it actually is...) $\endgroup$ – jamesqf Apr 10 '17 at 18:02
  • 1
    $\begingroup$ You link to the Wikipedia article in English on spacetime, but do you know that there is also a Wikipedia article in the "Simple English" (pseudo)language? $\endgroup$ – user152678 Apr 11 '17 at 15:53
  • 6
    $\begingroup$ Possible duplicate of What really is Spacetime? $\endgroup$ – Alex Robinson Apr 12 '17 at 8:20
31
$\begingroup$

The intuitive and traditional idea of space and time is that objects live in an infinite three-dimensional box, space, and that their motion in space happen in time in such a way that at each definite moment in time all objects have a position, and we can compare those positions because time flows the same for all objects.

Physicists discovered that there is no such box, and there is no such flow of time. This traditional space/time framework somewhat holds but only relatively to an object; it is not the same for all objects.

So there is no universal spatial background, and no universal time flow.

Spacetime is then the notion we use to still have a background after all. By forming a space (in the mathematical sense) combining traditional space and traditional time in an intricate way allowing space to rotate into time and the other way round, we can still get by with the idea that there is some smooth universal scene where everything happens.

The price to pay to see spacetime as a background is that this scene is completely static, sometimes called the block universe. But since all space and all time are intrinsically part of it, it actually cannot be conceived from an external point of view, and indeed Einstein's equations are strictly local and relational: they describe how the distribution of energy defines its own playground and how time and space can be seen in the way we are used to only instant by instant for specific observers, whose mutual perspectives are always shifting and transforming. In that view spacetime is far from static, it is more like a sort of fluid.

$\endgroup$
  • 2
    $\begingroup$ I would like to see some aspects of this albeit decent answer tidied up. The 'allowing space to rotate into time' could be made more rigorous without going beyond the scope. In addition the last paragraph is extremely vague and while there is a relation to fluid dynamics, your entire description I think is not trying to make that point and whatever it is trying to make is unclear. $\endgroup$ – JamalS Apr 10 '17 at 17:38
  • $\begingroup$ @JamalS. The point of the last paragraph is to emphasize the difficulties of a verbal description of spacetime, since it is impossible to do so without using words whose context is deeply Newtonian. So I first talk about the block view (where spacetime is "seen" as monolithic in a cognitively spatial way) and then counterbalance it with a dynamic, local, description. I agree with you that things are eventually left a little unclear, but a clear picture of spacetime may be just wrong (and I do not have one...) $\endgroup$ – Stéphane Rollandin Apr 10 '17 at 18:24
  • $\begingroup$ Well, there is a clear picture of spacetime. What is not clear is the role of spacetime and its applications in quantum gravity and quantum field theory, for reasons illuminated by works by Cachazo and others on scattering amplitudes which hint it may not be the ideal formulation for certain things. $\endgroup$ – JamalS Apr 10 '17 at 19:28
9
$\begingroup$

Spacetime is, like the name suggest, space and time together.

But there's more to it!

3D space isn't just horizontal 2D plane and height together. You can also rotate stuff in it, so for some 3D object, you don't have uniquely specified what is its height - this can change as you rotate it. Similar thing happens with the spacetime.

In 3D space, when you rotate stuff, you don't do much to it, you only change coordinate system. Distances stay the same. Square of the distance $s$ is given by \begin{equation} s^2=\Delta x^2+\Delta y^2+\Delta z^2 \end{equation}

In the spacetime, you can obviously rotate stuff, but you can also do the "rotations" which involve time. They are called "Lorentz transformations" (they are not quite rotations, see gif). So, once you have spacetime, space and time can get mixed up. What remains the same is the following quantity: \begin{equation} s^2=-c^2\Delta t^2+\Delta x^2+\Delta y^2+\Delta z^2 \end{equation} where $c$ is the speed of light.

One might ask "So what? This looks like just a fancy math/set of definitions/whatever...", but there are some interesting consequences. Relativity of simultaneity is one of them. In simple words, it means that (contrary to some philosophies of time) there is no uniquely defined "now" (except "here and now"), just like there is no uniquely defined "this much to the right" in space. Consequence of relativity of simultaneity is that faster than light (FTL) travel could in principle be used for time travel. However, so far we don't know of any laws of physics that would allow for FTL transfer of information, but that's another story.

$\endgroup$
8
$\begingroup$

An experimentalist's answer:

In particle physics the data show that one is not dealing with a three dimensional space that has as a parameter time, but that space and time are united in a specific mathematical way in what is called a "four vector space".

In classical mechanics a vector in space is a three dimensional row or column with values from the field of real numbers that follow eucledian tranformation properties under translations and rotation.

Momentum for example $(p_x,p_y,p_z)$ are the three vector components whose length is $p$ and is given by

$$p=\sqrt{p_x^2+p_y^2+p_z^2}$$

Which is invariant under rotations and translation of the reference system.

Relativistic mechanics which hold for high energies and momenta, and which are necessary in order to make sense of the plethora of particle data, requires four vectors,

The momentum vector above is modified and called a four momentum ( $c=1$ in this):

$$(p_x,p_y,p_z,E)$$ .

In this the "length" of the four momentum becomes the $\sqrt{E^2-p_x^2-p_y^2-p_z^2}$ and is the mass of the particle under consideration, an invariant in all Lorenz transformation frames.

So in analogy to calling three dimensional space , space obeying euclidean rules, we call the four dimensional space of this special construction, which clarifies the transformation rules in particle interactions, space time to denote that one is dealing with the special fourvector quantities.

$\endgroup$
  • 6
    $\begingroup$ "Nice and simple" said the man, after reading the description of spacetime. He then proceeded to cry in the closet for several hours. Unable to cope with the apparent low ceiling of his intellect, the man took a long walk down a short pier and was never seen again. Some say they still feel the air of his melancholy, wafting around in spacetime. $\endgroup$ – bubbleking Apr 11 '17 at 15:20
6
$\begingroup$

Space and time is one single entity; we name it spacetime.

Figuratively speaking, events in spacetime is just like locations on earth; where northward and eastward are just like space and time.

You may ask why space and time have different units? As sometimes northward and eastward sometimes are written in different units (e.g. meters and mile), the same happens in spacetime, and the conversion factor between the meter and the second is light speed.

$\endgroup$
  • 2
    $\begingroup$ But meters and miles are just constant multiples of the same thing: you can convert meters to miles by dividing by roughly 1600. That's not a good comparison with meters and seconds, which are different dimensions (in the sense of dimensional analysis). $\endgroup$ – David Richerby Apr 11 '17 at 9:21
  • 1
    $\begingroup$ @DavidRicherby you can always write $time$ in unit/dimension of $meter$ in relativity. $\endgroup$ – Shing Apr 11 '17 at 17:12
6
$\begingroup$

One of the tenets of relativity is to place both space and time on an equal footing. In special relativity, you will learn that not only like in Galilean relativity can one disagree on say velocities, but there is no absolute time - it is in fact also relative, motivating keeping time as having the same status as a spatial coordinate, rather than a parameter.

So, flat Minkowski spacetime is simply $\mathbb{R}^{1,n-1}$ with one timelike coordinate and $n-1$ spacelike coordinates, for an $n$-dimensional space. It has a special property however. If you think of distances in Euclidean space, we have that,

$$ds^2 = dx^2 + dy^2 + dz^2$$

as you are familiar with, from learning the Pythagorean theorem. However, in spacetime, we have to measure distances slightly differently, according to,

$$ds^2 = dt^2 - dx^2 - dy^2 - dz^2$$

or alternatively,

$$ds^2 = -dt^2 + dx^2 + dy^2 + dz^2$$

rather than taking the naive approach from Euclidean geometry. There are some profound implications from the fact that spacetime has signature $(1,n-1)$ as opposed to $(n,0)$, but this would be beyond the scope of your question.

As for why it is important, the concept of spacetime is used in the theory of gravitation known as general relativity and modern particle physics uses a framework called quantum field theory which is based on spacetime as well. However, recent developments suggest that this formulation may actually have drawbacks, but once again as for why is beyond the scope here.

$\endgroup$
  • $\begingroup$ A "tenant" is a renter. A "tenet" is a fundamental belief. I'm pretty sure you meant the latter. $\endgroup$ – Monty Harder Apr 10 '17 at 19:30
  • $\begingroup$ @MontyHarder Haha yes, thank you - auto-correct (more like auto-incorrect!) on my iPhone :) $\endgroup$ – JamalS Apr 10 '17 at 19:31
3
$\begingroup$

Spacetime is a system in which neither the three yardsticks of space, nor the clock of time are invariant with respect to all observers and observed objects.

The appearance of yardsticks and clocks varies depending on the relative velocities of observers and observed objects, or on their acceleration, or on the position of an observer or an observed object within a gravitational field.

The four vectors (space and time) of spacetime are unified as events in each observer's reference frame, rather than as locations and times within a universal reference frame.

$\endgroup$
0
$\begingroup$

Whenever you draw a curve of a trajectory over time you are using spacetime as a concept. It is just a space with n+1 dimensions, where n is the number of space dimensions.

When we draw such diagrams on a piece of paper we usually omit one or two of the space dimensions (e.g. recording only the x position in space; y and z may be less important). That leaves room for using one axis for the time values of events.

That is, a 2-dimensional piece of paper can represent spacetime with n=1, in which you draw (x, t) curves. Similarly, a 3D diagram on paper or on a computer can represent subsets of spacetime with two explicit space dimensions.

Mathematically, this is just a Cartesian space of tuples of n+1 numbers. In these tuples you write the values of space coordinates, plus the time position, of events, and the set of these events forms a curve or surface in spacetime.

$\endgroup$
  • 2
    $\begingroup$ This answer doesn't really address what spacetime is or how it differs from normal Euclidean space with a dimension of time. $\endgroup$ – Shufflepants Apr 10 '17 at 15:07
  • $\begingroup$ @Shufflepants, If you're not worried about GR, then it can be a Euclidean space. The point is, you're folding time into the mathematical idea of space. You can model the evolution of some system over time as a curved path in a space, and that's something that you can analyze and understand with mathematics. $\endgroup$ – Solomon Slow Apr 10 '17 at 15:35
  • 1
    $\begingroup$ @jameslarge But the whole point of space-time as a distinct concept is because of GR and its deviation from Euclidean space with a distinct and separate time dimension. And yes, of course, you can always view some warped space as a space embedded in a Euclidean space with a possibly higher number of dimensions with some parameterization, but that's not the point. $\endgroup$ – Shufflepants Apr 10 '17 at 15:40
  • 2
    $\begingroup$ And just plotting things against time is absolutely not what the concept of space-time is about. People in highschool and Physics 101 doing 2-d kinematic problems about balls flying through the air are not dealing with space-time. They're doing Newtonian physics. $\endgroup$ – Shufflepants Apr 10 '17 at 15:45
  • 1
    $\begingroup$ @Zardos But it's necessary to say something about the structure because, otherwise, the reader is likely to believe that spacetime is nothing at all. If you don't say anything about the structure, then spacetime is just notation that condenses "location $(x,y,z)$ at time $t$" to "spacetime co-ordinates $(x,y,z,t)$". And then you get the followup question, "OK, so I completely understand what spacetime is but it's completely pointless, right?" And, of course, the answer is that the person doesn't understand at all. $\endgroup$ – David Richerby Apr 11 '17 at 9:26
-1
$\begingroup$

You can think of spacetime as a map. If you want to meet someone in this universe you have to agree an exact location and a time on the map. If you show up at correct location but at the wrong time then you are in the wrong place on the spacetime map.

$\endgroup$
  • 2
    $\begingroup$ If you think of spacetime in these terms, it's just concise notation that lets you write "location $(x,y,z)$ at time $t$" as "$(x,y,z,t)$", which is really nothing at all. Furthermore, a fundamental point in relativity is that two people who agree to meet at 3pm don't necessarily agree when 3pm actually happens. This "spacetime map" doesn't actually exist. $\endgroup$ – David Richerby Apr 11 '17 at 17:25
  • 1
    $\begingroup$ The age of the universe is the same for all observers, so it could be used as a time that everyone agrees. They could theoretically agree to meet at e.g. 13.998837373 billion years after the Big Bang, at the agreed location. $\endgroup$ – Paul Apr 12 '17 at 0:04

Not the answer you're looking for? Browse other questions tagged or ask your own question.