# Topology/Differential geometry versus analysis

1. Nov 28, 2006

### jdstokes

Hi all,

I'm a second year student entering 3rd year with an interest in physics and mathematical physics. Foolishly I decided not to enrol in the second year pure mathematics course real and complex analysis''. My current mathematical knowledge comprises the following

First year:

Second year:
Linear Mathematics & Vector Calculus (Advanced)
Algebra (Advanced) [a course in group theory]

I am thinking about majoring in physics and pure mathematics, with the following 3rd year maths courses. The course descriptions can be found in the handbook http://www.maths.usyd.edu.au/u/UG/SM/hbk06.html [Broken]

Rings Fields and Galois Theory (Advanced)

Interestingly, none of these courses require knowledge of analysis. So it is possible to major in pure maths without having done any analysis whatsoever. I can't help but feel that my lack of analysis training will come back to haunt me, which is why I'm also considering the following, less interesting combination of courses

Analysis (Normal)
Rings Fields and Galois Theory (Advanced)

Note that the normal analysis course does not technically satisfy the assumed knowledge for complex analysis, but the lecturers inform me that I might be okay'' if I do very well in the normal course and do some extra work in my own time. Of course, this means dropping differential geometry, which I'm not too keen about due to its obvious connections with general relativity. I guess what it boils down to is whether topology or analysis is considered more important in physics. I would appreciate any advice you may be able to give on this question and/or my course selections.

Thanks,

James.

Last edited by a moderator: May 2, 2017
2. Nov 28, 2006

### dextercioby

The bolded statement is absurd. Metric spaces provide the foundation for analysis. While differential geometry provides the natural link b/w topology, analysis and linear algebra.

As for group representation theory, you gotta be kidding me it doesn't use calculus. Unless there's no Lie group there, thing which would be rather absurd.

Daniel.

Last edited by a moderator: May 2, 2017
3. Nov 28, 2006

### jdstokes

Daniel,

I'm sorry, why is that absurd? You just said yourself that differential geometry provides the natural link b/w topology, analysis and linear algebra''?

4. Nov 28, 2006

### dextercioby

Yes, knowledge of multivariable calculus is essential to diff.geom.

You have to know analysis b4 taclkling geometry.

Daniel.

5. Nov 28, 2006

### ^_^physicist

Personally, I would cut metric spaces and group so I could take the anyalsis courses and diff. geo.

Metric spaces, at least at my school, are part of advanced (riggorous proof based) calculus. From your list I am assuming that you have the background to teach it to yourself.

6. Nov 28, 2006

### Daverz

I really envy you for your choice of courses.

Differential geometry can be taught without recourse to topology and with only as much analysis as in a good calculus text. However "advanced" suggests a little more familiarity may be expected. I'd suggest checking out the text used.

Among those 3rd year courses, the "Modules and Group Representations" one sounds really cool. Most physics grad students are expected to pick this stuff up by osmosis. I wouldn't miss Differential Geometry myself, it's a beautiful subject.

If you have the time, money, and discipline, I'd definitely take real analysis and topology courses.

Last edited: Nov 28, 2006
7. Nov 28, 2006

### mathwonk

i think both topology and analysis are absolutely basic. actually point set topology and metric spaces is merely foundations of analysis.

8. Nov 28, 2006

### jdstokes

Does metric spaces as a stand alone subject have applications to anything other than analysis? For example, does topology help with GR/QM/strings independently of analysis?

From my somewhat naive perspective, it seems that applications of analysis (particularly of the real type) to physics are limited compared to topics such as groups and group representations. I'm not sure what the situation is with topology. Does anyone know of a list of applications of analysis to physics and a corresponding list for topology?

9. Nov 28, 2006

### leon1127

I took topology and analysis simutaneously. I think this they both help me to understand each other. Honestly, I think you will have a basic concept of analysis if you conplete all the course you have listed. Of course, if you really don't want to take a course in analysis, you should still get a book in analysis. I understood my undergrad analysis book before the first time I walk into my class. Knowing analysis makes me to become a more practical person in life

In the end, everything is just topology, analysis, and algebra. For example, functional analysis is a very applicable in mechanic, i.e energy spaces. Operator Theory is also important in many branch of phys. Calculus of Variations is the base of Lagrangian mechanic; one can find application like least action principle in QM, etc. Tensors Analysis is the language of relativity. Exterior Calculus can be applied to E&M and Thermodynamics. For topology, Morse Theory provides a new insight of conjugate point using differential topology. One can also apply algebraic topology to understand n-dimensional circuit. These are all beautiful applicaions of analysis and topology in practical world.
After all, you, as a physicist, also want to do calculus on a manifold, right? That means you want to know how calculus works on real numbers before you do calculus on a manifold. I dont know anyone can skip calculus in reals before jumping into manifold theory. Then concepts of analysis and topology are nice to have with your algebraic knowledge.

Leon

reference:
Bamberg & Stermberg, A course in mathematics for students of physics I & II
Lebedev & Vorovich, Functional analysis in mechaics;
Arnol'd, mathematical methods of classical mechanics;
Hirch, Differential Topology.

Last edited: Nov 28, 2006