# Spin-statistics theorem

1. Mar 12, 2006

### coolguy

could anyone post a link to Pauli's original proof of spin-statistics theorem.have been looking for a long time.could'nt find any.
thanks.

2. Mar 12, 2006

### masudr

3. Mar 13, 2006

### dextercioby

I'm sure there has to be a free copy of it, somewhere...

Daniel.

4. Mar 13, 2006

### CarlB

Drop by your local physics library and download it for free onto your USB hard drive (sometimes called memory stick). Failing that, you can always photocopy it out of the stacks.

Carl

5. Mar 14, 2006

### coolguy

i have a sick library around.wish i could get it on the net.even a modified version will be good for a start,heard its pretty deep.

6. Mar 14, 2006

### dextercioby

I got a .PDF file of ~622KB with a scanned version. Too bad i can't attach it on this forums...

Daniel.

7. Mar 15, 2006

### coolguy

could u mail it to me.it would be great.

8. Mar 31, 2008

### iSlak

If you email it to me I will host it.

9. Apr 1, 2008

### RandallB

Guys; aren’t there copyright issues on things like that?

Don’t the mentors here have to remove posts and known links to stolen material that violates copyright ownership.

It is one thing to get a library copy for private use, but sharing between members here and providing free access to everyone across the entire internet is not a valid version of private use.

10. Apr 1, 2008

### CarlB

11. Apr 1, 2008

### Hans de Vries

The Connection Between Spin and Statistics
W. Pauli Princeton, New Jersey (Received August 19, 1940)

Amazingly, I now see that Pauli comes up here with the Feynman propagator
for the KG equation (18) in position space, in 1940(!) after first presenting the
causal Klein Gordon propagator (15).

He then goes on to dismiss it with the words:

"Theories which would make use of the D1 function in their quantization
would be very much different from the known theories in their consequences."

This because of the propagation outside the light cone....

Regards, Hans

PS:

D = Klein Gordon Causal propagator in 4d position space
F = Klein Gordon Causal propagator in 2d position space
D1 = Feynman KG propagator in 4d position space
F1 = Feynman KG propagator in 2d position space

PPS: The domains (x0>r) etcetera are mixed up in (18)

Last edited: Apr 1, 2008
12. Apr 1, 2008

### Hans de Vries

Just for reference, see also this thread:
https://www.physicsforums.com/showthread.php?t=217846

1+1 dimensional Feynman KG propagator in position space:

$$D^2_F(x,y) = \lim_{\epsilon \to 0} \frac{1}{(2 \pi)^2} \int d^2p \, \frac{e^{-ip(x-y)}}{p^2 - m^2 + i\epsilon} = \left \{ \begin{matrix} \ \ \ \frac{1}{4} H_0^{(1)}(ms) & \textrm{ if }\, s^2 \geq 0 \\ -\frac{i }{ 2 \pi} K_0(ms) & \textrm{if }\, s^2 < 0 \end{matrix} \right. \qquad \qquad s^2\ =\ |x^0 - y^0|^2 - |\vec{x} - \vec{y}|^2$$

3+1 dimensional Feynman KG propagator in position space:

$$D^4_F(x,y) = \lim_{\epsilon \to 0} \frac{1}{(2 \pi)^4} \int d^4p \, \frac{e^{-ip(x-y)}}{p^2 - m^2 + i\epsilon} = \left \{ \begin{matrix} -\frac{1}{4 \pi} \delta(s^2) + \frac{m}{8 \pi s} H_1^{(1)}(ms) & \textrm{ if }\, s^2 \geq 0 \\ -\frac{i m}{ 4 \pi^2 s} K_1(ms) & \textrm{if }\, s^2 < 0 \end{matrix} \right.$$

And the relation between the two is:

$$D^4_F(s)\ =\ \frac{1}{\pi}\ \frac{\partial}{\partial (s^2)}\ D^2_F(s) \qquad \left(= \frac{1}{2\pi s}\ \frac{\partial}{\partial s}\ D^2_F(s)\ \right)$$

Regards, Hans

13. Apr 2, 2008

### Hans de Vries

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?