Thursday, September 11, 2014

An interesting way of looking at the Lorentz Transform

So, this is going to be a rather light posting today.  I was at first going to refer to it as short, but apparently, I can be quite verbose, so, so much for that.  Anyway, this post is really meant as a means for me to get the ball rolling on this blog again after a series of events that lead to me being unwilling to blog.  I actually hope to talk a little about that, at some point, but for now, here is a fun little thing I realized about the Lorentz Transform, which actually stemmed from my desire to find a way of assigning a coordinate to every point in space-time without explicitly relying on velocity.

Let me remind you of the Lorentz Transform, if you don't know what it is:
\begin{equation}
x' = \left (x - vt \right )\gamma
\end{equation}

\begin{equation}
y'=y
\end{equation}

\begin{equation}
z'=z
\end{equation}

\begin{equation}
t' = \left (t -\frac{xv}{c^2}\right )\gamma
\end{equation}

where
\begin{equation}
\gamma = \frac{1}{\sqrt{1-\frac{v^2}{c^2}}}
\end{equation}

If you aren't familiar with what this means, wikipedia has a good article on it.

And of course, the inverse transform is given by:
\begin{equation}
x = \left (x' + vt' \right )\gamma
\end{equation}

\begin{equation}
y=y'
\end{equation}

\begin{equation}
z=z'
\end{equation}

\begin{equation}
t = \left (t' +\frac{x'v}{c^2}\right )\gamma
\end{equation}

So, who cares?

Well, it occurred to me that I was wrong when I said that there was no way of uniquely identifying a space-time point without explicitly referring to velocity.  All it takes is a little adjustment to your point of view, and...well...cheating.

You see, you can actually use the inverse Lorentz Transform to fulfill this exact requirement, and the answer is quite trivial--express it in terms of the origin of the rest frame!  Let me just get into some philosophy before this.  When we describe the universe, at least in the non-general relativistic regime, it is always with respect to some fixed reference point.  Sure, you can express your system in polar coordinates, or spherical, or something completely exotic, but there will always be a point with the label (0,0,0,0).  Often, this point happens to be physically meaningful, and not just convenient.  It might represent some central object like the Sun, or the center of mass.  To show you the significance of this, let's try a thought experiment.

Pick a point--any point in the universe.  The only stipulation is that it must be uniquely defined.  Which did you choose?

I suspect that most people who perform this exercise would pick something that they could easily describe.  They might have chosen "my location", or "the center of Earth", or "the center of mass of Messier 83."  Some particularly creative people might have chosen something like "a point that is located 200 billion km in the direction of where the the geomagnetic field of Earth is pointing at my location."

But, here is what I suspect you didn't do--pick the center of mass of all discrete silicate glass materials in the visible universe with a charge of $-2\pm0.1$ coulombs.  Likewise, I doubt you picked the center of mass of the set of living organisms with the largest volume in the visible universe.

Now, you'll notice the difference between the first set of objects and the second.  In the first, each description, no matter how complicated, relied explicitly on some known physical reference point.  The objects in the second set did not explicitly reference any known physical point.  In fact, it is impossible, given our current knowledge of the universe, to determine what these points are.

We need some physical reference point to anchor our description of physical phenomena.  Even the second set of points, though not explicit, referenced a known location, as the visible universe represents the light cone of Earth, and that's before even talking about the need to express these things against the backdrop of an inertial reference frame to fully define them. (I would like to say, for the record, that this would be a fun, nerd game to play with all your nerd friends.  Specifically, see who can identify the most obscure point in space-time possible, especially with alcohol.  There is no doubt, that eventually, penises would work there way in there some where.  They always do.)

Thus, it seems that only by fixing some origin somewhere that we can begin to describe the universe.

If you think about it, this is exactly what we do with the inverse Galilean transform:
\begin{equation}
x = x' + vt'
\end{equation}

\begin{equation}
y = y'
\end{equation}

\begin{equation}
z = z'
\end{equation}

Namely, as long as everybody agrees where the reference point is, everybody can agree on where the event occurred, and since measurements approximate c as being infinite in speed, then everyone knows when it happened, too.

In fact, we can express equation 10 in another way.  Let $O$ represent the position of the reference point.

Then, we we have
\begin{equation}
x = x' - O
\end{equation}
Of course, this seems obvious.  All we're doing is describing points with respect to $O$.  However, to me, this represents an interesting way of looking at things.  In particular, by rephrasing the Galilean transform in this way, we no longer have to worry about actually measuring relative velocity.  All we need to do is figure out where $O$ was in relation to us, and we would be able to unambiguously identify a point in space that corresponds to every event.

So, the question is, can we use this as inspiration for the Lorentz Transform.  Of course!

So, take equation 6 and replace it with:
\begin{equation}
x = \left (x' - O \right )\gamma
\end{equation}
Now, this makes sense, at least partly.  Somebody moving relative to me at velocity $v$ would say that my trajectory is $-vt'$, assuming our origins coincided at when our clocks both read 0s.  Thus, if they treat me as their reference point, $O$, then this parallels the Galilean so far.
"But hold on!" you might protest, "$\gamma$ isn't just a constant.  It too depends on relative velocity."
Well, algebra's not a problem for us.  $O=-vt'$, so $v=-O/t'$, and this is where things begin to get really kind of interesting.  Let's see what this does to $\gamma$

\begin{equation}
\gamma = \frac{1}{\sqrt{1-\frac{\left( \frac{-O}{t'}\right)^2}{c^2}}}
\end{equation}
\begin{equation}
\gamma = \frac{1}{\sqrt{1-\frac{O^2}{(ct')^2}}}
\end{equation}

Now, notice that $ct'$ would be the position of an electromagnetic wave emitted along the x axis at the time the origins coincided.

So, $\gamma$ becomes an expression involving the position of O and the position of the electromagnetic wave.

We can do something similar with equation 9.
\begin{equation}
t = \left (t' -\frac{x'O}{t'c^2}\right )\gamma
\end{equation}
\begin{equation}
t = \left (\frac{(ct')^2 -x'O}{c(ct')}\right )\gamma
\end{equation}

Again, we see the equation clearly expressed in terms of the positions $O$ and $ct'$.

Now, see what I mean about cheating?  In some sense, this is a mathematical disappointment.  All I did was rewrite the Lorentz transformation, but in a physics sense, it has real ramifications for measurement--perhaps not profound--but real, nonetheless.  Just as with the inverse Galilean transform, we see that there is a way to uniquely assign a space-time coordinate to an event without having to measure velocity, by having everybody agree on a reference point.  In other words, you don't need to differentiate at all.  Just know where the reference point is at the time of the event, and whatever event you encounter, you can say exactly where it would have been for the person at rest with that reference point.  Keep in mind, also, that the fundamental forces are conservative, and therefore rely only position.  This form seems to respect that more than the conventional form.  In all honesty, though, I can't really say this is a useful way of looking at things.  But, then again, I have barely had any time to think about it.

So, what do you think?  Is there any use that can truly come from expressing the transform in this way?  I wonder if this might be useful as a pedagogical tool for talking about closed time-like curves resulting from FTL information transfer.  Anyway, I am signing off, for now.