Lecture - 1
(Algebraic Topology) (Prepared by ANIRBAN BANERJEE)
Path
Let X be a topological space. A continuous function
f : [0, 1] → X
is called a path in X. Let f (0) = x0 and f (1) = x1 . Then we say f is a path from x0 to
x1 .
Homotopic Paths
Let f, g : [0, 1] → X be two paths such that
f (0) = g(0) = x0 , f (1) = g(1) = x1 .
Then the paths f and g are said to be homotopic if there exists a continuous function
F : [0, 1] × [0, 1] → X
such that:
1. F (s, 0) = f (s), 0≤s≤1
2. F (s, 1) = g(s), 0≤s≤1
3. F (0, t) = x0 , 0≤t≤1
4. F (1, t) = x1 , 0≤t≤1
Note that the existence of such a map F automatically gives a continuous function
G : [0, 1] × [0, 1] → X
such that:
1. G(s, 0) = g(s), 0≤s≤1
2. G(s, 1) = f (s), 0≤s≤1
3. G(0, t) = x0 , 0≤t≤1
4. G(1, t) = x1 , 0≤t≤1
Just by considering
G(s, t) = F (s, 1 − t),
we can say f and g are homotopic to each other, and we write f ≃ g or g ≃ f . We call
F a path homotopy between f and g.
1
Homotopy of Paths
Let also paths f and g be homotopic. Let
F :I ×I →X
be a path homotopy between f and g. For each fixed t, define
F (x, t) = ft (x),
i.e.,
ft : I → X,
is a family of paths such that:
1. f0 = f
2. f1 = g
3. ft (0) = x0
4. ft (1) = x1
The family of paths {ft }0≤t≤1 is called a homotopy of paths, and f (= f0 ) and g(= f1 )
are connected by the homotopy ft or {ft }0≤t≤1 .
To avoid notation, sometimes we take f = f0 and g = f1 . Then it looks good that f0
and f1 are homotopic by the homotopy ft .
Example (Linear Homotopy)
We prove that any two paths f0 and f1 in Rn having the same endpoints are homotopic.
Let x0 and x1 be the endpoints of both paths. Define
ft = (1 − t)f0 + tf1 .
It is easy to see that f0 and f1 are homotopic by the homotopy ft . Here,
F (s, t) = (1 − t)f0 (s) + tf1 (s).
Let (sn , tn ) be a sequence converging to (s, t), i.e., sn → s and tn → t. Since f0 and f1
are continuous,
f0 (sn ) → f0 (s), f1 (sn ) → f1 (s).
Thus,
(1 − tn )f0 (sn ) + tn f1 (sn ) → (1 − t)f0 (s) + tf1 (s) as n → ∞.
i.e.,
F (sn , tn ) → F (s, t) as n → ∞.
Thus F is continuous on I × I.
Now,
1. F (s, 0) = f0 (s)
2. F (s, 1) = f1 (s)
3. F (0, t) = (1 − t)f0 (0) + tf1 (0) = (1 − t)x0 + tx0 = x0
4. F (1, t) = (1 − t)f0 (1) + tf1 (1) = (1 − t)x1 + tx1 = x1
This completes the proof.
2
Theorem
The relation of homotopy on paths with fixed endpoints in any space is an equivalence
relation.
Proof: We check that ≃ is reflexive, symmetric, and transitive.
Reflexive: Let f be a path. Define
F : I × I → X, F (s, t) = f (s).
Then F is continuous and satisfies:
• F (s, 0) = F (s, 1) = f (s)
• F (0, t) = x0 , F (1, t) = x1
Thus f ≃ f by path homotopy F , so ≃ is reflexive.
Symmetric: Let f ≃ g via F . Define
R(s, t) = F (s, 1 − t).
Then R is continuous, and
R(s, 0) = g(s),
R(s, 1) = f (s),
R(0, t) = x0 , R(1, t) = x1 ,
so g ≃ f . Thus ≃ is symmetric.
Transitive: Let f ≃ g via F , and g ≃ h via G. Define
H : [0, 1] × [0, 1] → X
by (
F (s, 2t), 0 ≤ t ≤ 21
H(s, t) =
G(s, 2t − 1), 21 ≤ t ≤ 1
Since F and G are continuous, and the sets involved are closed, H is continuous by the
Pasting Lemma.
Then,
H(s, 0) = F (s, 0) = f (s),
H(s, 1) = G(s, 1) = h(s),
H(0, t) = x0 , H(1, t) = x1 ,
so f ≃ h. Therefore, ≃ is transitive.
From now on, we write [f ] for the equivalence class of f under the homotopy relation,
i.e.,
g ∈ [f ] ⇐⇒ f ≃ g.
3
Product Path
Given two paths f, g : [0, 1] → X such that f (1) = g(0), define the product path f · g
by (
f (2s), 0 ≤ s ≤ 12
f · g(s) =
g(2s − 1), 12 ≤ s ≤ 1
Theorem: For two paths f, g : [0, 1] → X such that f (1) = g(0), the class [f ][g] := [f · g]
is well-defined.
Proof: Suppose f0 ≃ f1 and g0 ≃ g1 . Then, by homotopies, the paths ft and gt
can be combined to form a homotopy between f0 · g0 and f1 · g1 . Hence, the class is
well-defined.
