1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Books to Prepare for Differential Geometry

  1. Jun 9, 2012 #1
    What books should I read as prerequisites for Spivak's Differential Geometry Series? Trying to pick up Diff Geom for graduate physics but right now it is pretty daunting. I've got single variable calculus and linear algebra under my belt. Just looking for bare minimum requirements here to jump into rigorous differential geometry.
  2. jcsd
  3. Jun 9, 2012 #2
    Real Analysis is a prerequisite for most differential geometry courses. How comfortable are you with mathematical proofs?
  4. Jun 9, 2012 #3
    I am pretty comfortable with proofs and prefer rigor to hand-waving. The university course in linear algebra that I took was proof-based and used Halmos. So I should probably work through Rudin for real analysis background, right?

    Would I need anything else after Rudin? I've heard that the first volume in Spivak's Differential Geometry series covers calculus on manifolds, so would it be necessary to first read his smaller book Calculus on Manifolds?
  5. Jun 10, 2012 #4
    Working through calculus on manifolds is recommended, but I don't think it's necessary. You do need to be very familiar with multi-variable calculus.

    Further, you need to be familiar with things like metric spaces. Familiarity with topology is even better.
  6. Jun 10, 2012 #5
    I once checked that small book by Spivak but seemed too trivial after having had analysis courses. So I ended up not reading it. I would say that if you have a comprehensible understanding of analysis, you should be able to read Spivak's diff. geom. straight ahead. However, it's always good to *really* master the basics before going into that kind of stuff, so Rudin might be a good idea, yes.

    I would comment about Spivak's differential geometry; but you seem to consider only Spivak from your OP. I personally find Spivak too verbose (I ended up not learning diff. geom. from him), and preferred Subrovin/Fomenko/Novikov's modern geometry for a physics' approach. Otherwise Lee's books (Intro to topological manifolds/smooth manifolds) or Guillemin & Pollacks' are as good.
  7. Jun 10, 2012 #6
    micromass, would the topology in Rudin be sufficient? I was scared after reading the first Amazon review (of Spivak's DG 1 text) which listed differential topology as well as abstract algebra as prerequisites. Also, could you recommend a book for multivariable calculus? Do you mean something like Stewart?

    From what you and Redsummers are saying it sounds like I will be fine with just single variable analysis and familiarity with multivariable calculus. This comes as music to my ears.

    The reason I'm so set on Spivak's Differential Geometry books is that my first physics professor told me to that they are true gems and should be digested thoroughly.
  8. Jun 10, 2012 #7
    Calculus on manifolds is actually a multivariable calc text. But it's a little compact.
    Stewart would of course also be good, but it's not really rigorous.

    As for the prerequisities of diff topology and abstract algebra. It's certainly not needed. But as with everything in math: the more math you're familiar with, the easier it will be for you. So if you know abstract algebra, then it will likely be a bit easier. But don't learn abstract algebra just to read Spivak, that's overkill.

    Yeah, but you'll need more than familiarity with multivariable calculus. You need to know it inside out. Differential geometry is actually a generalization of multivariable calculus to more abstract spaces (=manifolds). So you must know the former really, really well.

    As for analysis: you need to be acquainted with metric spaces and the topology of metric spaces. If you insist on using Rudin (which I think is a horrible book, but other people think it's a gem), you will need to know chapter 2 really well. The other chapters are things you will probably remember from calculus, but it's good to go through to them anyway. Chapter 9-11 are horrible in Rudin, so I would suggest another book for that :biggrin:
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook