Proof of dimension of the tangent space

Click For Summary

Discussion Overview

The discussion revolves around the proof of the dimension of the tangent space of an n-dimensional manifold, as presented in a book on general relativity by Wald. Participants raise questions about specific equations in the proof and explore the implications of differentiability and continuity in the context of tangent spaces.

Discussion Character

  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant questions whether the function in equation 2.2.3 can be arbitrary at point 'a' since the (x-a) term is zero.
  • Another participant suggests that equation 2.2.3 should be an integral if it follows from the gradient theorem, indicating a potential misunderstanding of the equality's nature.
  • A participant references a similar proof in Isham's work, suggesting it may provide additional clarity.
  • One participant explains that the continuity of differentiable functions implies that the value at a specific point is determined by values elsewhere in the domain.
  • Another participant provides a justification for the equality in equation 2.2.5, breaking down the terms involved and their implications.
  • A participant notes that an n-manifold is locally diffeomorphic to R^n, implying that each tangent space is isomorphic to the tangent space of R^n at the corresponding point.

Areas of Agreement / Disagreement

Participants express differing views on the interpretation of specific equations and the nature of the proof, indicating that the discussion remains unresolved with multiple competing interpretations.

Contextual Notes

There are assumptions regarding the definitions of continuity and differentiability that are not fully explored. The discussion also highlights potential dependencies on the specific formulations of the equations in question.

hideelo
Messages
88
Reaction score
15
I am attaching a picture of a proof from the book "general relativity" by wald. This is supposed to show that the tangent space of an n dimensional manifold is also n dimensional. I have two questions.

In equation 2.2.3 couldn't the function be anything at a since the (x-a) term is 0?

How is the equality in equation 2.2.5 justified, I'm just not seeing it
 

Attachments

  • 1432214740404.jpg
    1432214740404.jpg
    16.7 KB · Views: 619
Physics news on Phys.org
I'm sorry, I cannot read the picture attached. Can't you just type it?
 
It's hard to write them out on my phone, so I'll try another picture
 

Attachments

  • 1432215051307.jpg
    1432215051307.jpg
    24.1 KB · Views: 586
  • 1432215072484.jpg
    1432215072484.jpg
    22.1 KB · Views: 620
Is that better?
 
No, it has to be exacly as written. Otherwise 2.2.2. will not be true.
 
I'm assuming 2.2.2 follows from the gradient theorem, in which case 2.2.3 should really be an integral, otherwise it's not an equality but the linear approximation.
 
A proof very similar to Wald's is included in Isham. Start reading on page 79. The proof is on pages 82-84.

Last time I discussed this proof with someone, I posted my version of it. It's in post #14 here.
 
hideelo said:
In equation 2.2.3 couldn't the function be anything at a since the (x-a) term is 0?
The theorem says that the function is smooth, i.e. differentiable an arbitrary number of times. Differentiable functions are continuous. The value of a continuous function at a specific point in its domain is completely determined by its values in the rest of its domain. For example, if ##f:\mathbb R\to\mathbb R## is continuous then ##f(0)=\lim_{n\to\infty}f\big(\frac 1 n\big)##.

hideelo said:
How is the equality in equation 2.2.5 justified, I'm just not seeing it
Assuming that you meant the last equality, this is how:

First term: f(p) is a constant function (the map that takes an arbitrary q to f(p)), so v(f(p))=0.

Second term: The sum is of the form ##\sum_\mu (A_\mu-A_\mu)B_\mu## so every term is zero.

Third term: For all smooth functions f and all constant functions g, we have v(f-g)=v(f)-v(g)=v(f).
 
an n manifold is by definition locally isomorphic (i.e. diffeomorphic)) to R^n, hence (by chain rule) each tangent space is isomorphic to the tangent space to R^n at the image point. Do you know that R^n is the tangent space to R^n at each point? That would do it.
 

Similar threads

  • · Replies 73 ·
3
Replies
73
Views
9K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
5K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 12 ·
Replies
12
Views
5K
  • · Replies 6 ·
Replies
6
Views
7K
  • · Replies 11 ·
Replies
11
Views
4K
  • · Replies 11 ·
Replies
11
Views
7K