Illustrating Allan Hatcher's Algebraic Topology with AI

Allen Hatcher Homepage

Prompt AI with the Theorem and Related Material:

Example Theorem (plus discussion and proof, not shown):

Theorem 1.10. For every continuous map f : S 2→R 2 there exists a pair of antipodal points x and −x in S 2 with f (x) = f (−x). (Algebraic Topology, 33-34)

Prompt: Illustrate the following proof: Theorem 1.10. For every continuous map f : S 2→R 2 there exists a pair of antipodal points x and −x in S 2 with f (x) = f (−x). 

AI

The provided text is Allen Hatcher's classic algebraic topology proof of the Borsuk–Ulam Theorem for \(n = 2\). Below is an breakdown of the proof's core mechanics alongside a visual simulation of the path lifting argument that drives the contradiction. [1, 2]

1. Conceptual Framework

The proof operates by contradiction via three primary geometric mappings:

    Equator Loop                Odd Map                   Covering Map
  I -------------> S² -------------------------> S¹ <------------------------ ℝ
        η                       g                       p (Path Lift tilde{h})
        
  [0, 1] ───> Equator Circle ───> Unit Circle in ℝ² ───> Unwrapped Spiral

• The Equator (\(\eta \)): A symmetric loop parameterizing the equator of the sphere \(S^{2}\).
• The Normalized Difference (\(g\)): If no antipodal pair maps to the same point (\(f(x) \neq f(-x)\)), we can always construct an odd map \(g(-x) = -g(x)\) by normalizing their difference onto a unit circle \(S^{1}\).
• The Lift (\(\~{h}\)): Because the circle \(S^{1}\) unwraps into its universal covering space \(\mathbb{R}\) via the exponential map \(p(t) = e^{2\pi i t}\), any closed loop \(h = g\eta\) on the circle can be tracking-lifted to a continuous path \(\~{h}\) in \(\mathbb{R}\).

---

2. Step-by-Step Mathematical Visualizer

To see why a contradiction must occur, we trace what happens to the path lift \(\tilde{h}(s)\) as \(s\) journeys from \(0\) to \(1\) (fully circling the equator). Because \(g\) converts antipodal points to opposite sides of the circle (\(h(s + 1/2) = -h(s)\)), its lift in \(\mathbb{R}\) must jump by an odd half-integer \(q/2\) every half-loop.

The plot below illustrates this lift behavior. Notice how traveling the second half of the loop forces the final endpoint \(\tilde{h}(1)\) to terminate exactly \(q\) integer units away from its start point \(\tilde{h}(0)\). This forces the loop to have an odd winding number (\(q \neq 0\)), proving it cannot be continuously shrunk to a point on the circle.

---

3. The Structural Contradiction

The visual breakdown reveals a topological impossibility:

1. Equator Disk Topology: The equator loop \(\eta \) forms the boundary of the upper hemisphere disk \(D^2 \subset S^2\). Because a disk can be shrunk to a single point, the loop \(\eta \) is contractible (nullhomotopic). Therefore, its composition \(h = g\eta\) must also be contractible on \(S^{1}\). [1, 2]
2. The Winding Number Catch: As shown in the graph above, the rigid antipodal structure forces \(\tilde{h}(1) = \tilde{h}(0) + q\). Because \(q\) is an odd integer, the loop explicitly wraps around the circle an odd number of times.
3. The Clash: A loop with a non-zero winding number (\(q \neq 0\)) can never be contracted to a point without breaking the path.

Because a path cannot be simultaneously contractible (\(q = 0\)) and non-contractible (\(q \text{ is odd}\)), the initial assumption must be false.

---

✅ Conclusion

For every continuous map \(f: S^2 \to \mathbb{R}^2\), there must exist at least one pair of antipodal points \(x\) and \(-x\) such that \(f(x) = f(-x)\). [1]

If you would like to explore this theorem further, let me know if you want to look at physical applications (like the meteorology temperature-pressure problem) or see how it logically scales up to prove Brouwer's Fixed-Point Theorem. [1, 2]