Raising and lowering Ricci Tensor

Click For Summary

Discussion Overview

The discussion focuses on the manipulation and understanding of the Ricci tensor, particularly the processes of raising and lowering indices, as well as the implications of the Bianchi identity. Participants explore various methods for contracting indices and express confusion over the notation and steps involved in these tensor operations.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • Some participants reference Hobson's book to discuss the relationship between the Ricci tensor and its indices, particularly the antisymmetry property of the Riemann curvature tensor.
  • There are claims that the notation ##g^{aa}## is meaningless, as it leads to contradictions when contracted with antisymmetric tensors.
  • Participants propose different methods for contracting indices of the Ricci tensor, with some suggesting contracting the first and last indices while others mention different approaches.
  • One participant expresses confusion about the steps involved in deriving the Ricci tensor from the Riemann tensor, particularly in relation to the Bianchi identity.
  • There is mention of raising indices using the metric tensor, with some participants clarifying that one should multiply by ##g^{ad}## rather than using ##g^{aa}##.
  • Some participants discuss the implications of the Bianchi identity on the Ricci tensor and how to manipulate it term by term.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the best methods for manipulating the Ricci tensor or the validity of certain notations. Multiple competing views remain regarding the contraction of indices and the interpretation of the Bianchi identity.

Contextual Notes

Some statements rely on specific definitions and notations that may vary between different texts. There is uncertainty regarding the validity of certain manipulations and the implications of the Bianchi identity on the Ricci tensor.

unscientific
Messages
1,728
Reaction score
13
Taken from Hobson's book:

ricci4.png


How is this done? Starting from:

R_{abcd} = -R_{bacd}

Apply ##g^{aa}## followed by ##g^{ab}##

g^{aa}g^{aa} R_{abcd} = -g^{ab}g^{aa}R_{bacd}
g^{ab}R^a_{bcd} = -g^{ab}g^{aa}R_{bacd}
R^{aa}_{cd} = - g^{ab}g^{aa} R_{bacd}

Applying ##g_{aa}## to both sides:
g_{aa}R^{aa}_{cd} = - g^{ab}g^{aa} g_{aa} R_{bacd}
R^a_{acd} = -g^{ab} R_{bacd}
R^a_{acd} = - R^a_{acd} = 0

Is there a quicker way?
 
Last edited:
Physics news on Phys.org
unscientific said:
Taken from Hobson's book:

ricci4.png


How is this done? Starting from:

R_{abcd} = -R_{bacd}

Apply ##g^{aa}## followed by ##g^{ab}##

g^{aa}g^{aa} R_{abcd} = -g^{ab}g^{aa}R_{bacd}
g^{ab}R^a_{bcd} = -g^{ab}g^{aa}R_{bacd}
R^{aa}_{cd} = - g^{ab}g^{aa} R_{bacd}

Applying ##g_{aa}## to both sides:
g_{aa}R^{aa}_{cd} = - g^{ab}g^{aa} g_{aa} R_{bacd}
R^a_{acd} = -g^{ab} R_{bacd}
R^a_{acd} = - R^a_{acd} = 0

Is there a quicker way?
g^{aa} has no meaning. The metric is symmetric tensor, g^{a b} = g^{b a}, so when you contract it with some antisymmetric tensor like R_{a b c d e f g} = - R_{b a c d e f g}, you get zero:
g^{ab} R_{a b c d} = - g^{b a} R_{b a c d}, \ \Rightarrow \ R^{b}{}_{b c d} = - R^{a}{}_{a c d} .
So, you have something equal to minus itself. It must be zero.
 
samalkhaiat said:
g^{aa} has no meaning. The metric is symmetric tensor, g^{a b} = g^{b a}, so when you contract it with some antisymmetric tensor like R_{a b c d e f g} = - R_{b a c d e f g}, you get zero:
g^{ab} R_{a b c d} = - g^{b a} R_{b a c d}, \ \Rightarrow \ R^{b}{}_{b c d} = - R^{a}{}_{a c d} .
So, you have something equal to minus itself. It must be zero.

Thanks for replying, I think I almost got the hang of it.
Quick but unrelated question:
Suppose I start off with ##R_{abcd}##, how do I get to ##R_{bc}##?

I'm thinking ##g^{aa}R_{abcd} = R^a _{bcd}## then ##-R^a_{bdc}## then ##-g_{de} g^{ea} R^a_{bdc} = -\delta_d^a R^a_{bdc} =- R^a_{bac} =- R_{bc}##.
 
Last edited:
unscientific said:
Quick but unrelated question:
Suppose I start off with ##R_{abcd}##, how do I get to ##R_{bc}##?
This depends on the book you use. I contract the first with the last index R_{a b} = g^{c d} R_{c b a d} = R^{d}{}_{b a d} others do it between the first and the 3rd.
 
samalkhaiat said:
This depends on the book you use. I contract the first with the last index R_{a b} = g^{c d} R_{c b a d} = R^{d}{}_{b a d} others do it between the first and the 3rd.

I don't understand how you can do that... I'm confused.
 
samalkhaiat said:
This depends on the book you use. I contract the first with the last index R_{a b} = g^{c d} R_{c b a d} = R^{d}{}_{b a d} others do it between the first and the 3rd.

Starting with the Bianchi Identity:

\nabla_e R_{abcd} + \nabla_c R_{abde} + \nabla_d R_{abec} = 0

I raise index ##a## by multiplying ##g^{aa}##:

\nabla_e R^a_{bcd} + \nabla_c R^a_{bde} + \nabla_d R^a_{bec} = 0

I'm trying to understand their steps:

riemann1.png
 
samalkhaiat said:
This depends on the book you use. I contract the first with the last index R_{a b} = g^{c d} R_{c b a d} = R^{d}{}_{b a d} others do it between the first and the 3rd.

It's alright, I got it - It can be done by contracting ##a## with ##c## then multiplying ##g^{be}##.
 
unscientific said:
Starting with the Bianchi Identity:

\nabla_e R_{abcd} + \nabla_c R_{abde} + \nabla_d R_{abec} = 0

I raise index ##a## by multiplying ##g^{aa}##:

\nabla_e R^a_{bcd} + \nabla_c R^a_{bde} + \nabla_d R^a_{bec} = 0
I told you, we don't have an object like g^{aa}. If you have a tensor T_{abc} and you want to raise the first index, you multiply your tensor with g^{a d}, i.e. T^{d}{}_{b c} = g^{a d} T_{a b c}.
 
Multiply the BI with g^{ad}. Let us do it term by term:
The first term g^{ad} \nabla_{e} R_{abcd} = \nabla_{e} g^{ad} R_{abcd} = \nabla_{e} R^{d}{}_{bcd} = \nabla_{e} R_{bc} . The second term g^{ad} \nabla_{c} R_{abde} = \nabla_{c} R^{d}{}_{bde} = - \nabla_{c} R_{be} . And the third term is easy, just raise the index on the covariant derivative g^{ad} \nabla_{d} R_{abec} = \nabla^{a} R_{abec} = \nabla_{a} R^{a}{}_{bec} .
 
  • #10
samalkhaiat said:
Multiply the BI with g^{ad}. Let us do it term by term:
The first term g^{ad} \nabla_{e} R_{abcd} = \nabla_{e} g^{ad} R_{abcd} = \nabla_{e} R^{d}{}_{bcd} = \nabla_{e} R_{bc} . The second term g^{ad} \nabla_{c} R_{abde} = \nabla_{c} R^{d}{}_{bde} = - \nabla_{c} R_{be} . And the third term is easy, just raise the index on the covariant derivative g^{ad} \nabla_{d} R_{abec} = \nabla^{a} R_{abec} = \nabla_{a} R^{a}{}_{bec} .

The ricci tensor is defined as ##R_{bc} = R^a_{bac}## in my notes. I'm trying to work with their notation.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 4 ·
Replies
4
Views
1K
  • · Replies 6 ·
Replies
6
Views
1K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 9 ·
Replies
9
Views
719
  • · Replies 16 ·
Replies
16
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 9 ·
Replies
9
Views
2K