Although it is not specified what T looks like in the theorem, it is basically defined through the convergence of sequences in dense subspaces during the proof process. By extending this way, the operator T can be naturally denoted as T.
When there are bounded linear operators T:W→V and w∈W∖W, we wish to define Tw, but since the domain of T is limited to W, it’s not possible. However, the sequence {wk} converging to w exists within W, so Tw can be ‘indirectly’ defined in the following plausible manner:
T(w):=k→∞limT(wk)
To paraphrase,
Since w is originally not part of the domain of T, Tw cannot be defined. However, there exists a sequence {wk} that converges to w within the domain of T. Therefore, when wk→w, it will be Twk→Tw, so defining it like Tw:=k→∞limTwk is natural and plausible. And this is indeed possible.
Thus, it is guaranteed from the above theorem that a bounded linear operator T:W→V defined on a Banach space W can expand its domain in the same way up to the closureW of W while maintaining the mapping to V.
Proof
Let’s denote by v∈V1. Assuming that W is dense in V1, there exists a sequence {vk} that converges to v.
k→∞limvk=v
Since T is bounded and linear, the following holds for ℓ,k∈N:
∥Tvk−Tvℓ∥=∥T(vk−vℓ)∥≤∥T∥∥vk−vℓ∥
Now, since {vk} is a Cauchy sequence, by the above equation, {Tvk} also becomes a Cauchy sequence. Therefore, since V2 is a Banach space, {Tvk} will converge to some element within V2. From this, define T as follows:
Tv:=k→∞limTvk
Then, due to the properties of the bounded linear operator for all v∈W, Tvk→Tv holds, thus satisfying Tv=Tv. Now, consider the case of v∈V1∖W.
Part 1. The irrelevance of sequence choice
Consider two sequences {vk},{uk} of W that converge to v. If v were an element within W, since T is a bounded linear operator, when vk,uk→v∈W, Tvk and Tuk would converge to the same value Tv. However, in the case of vk,uk→v∈V1∖W, this property cannot be used, so it must be directly confirmed whether they converge to the same value.
Now consider the following sequence {wk}:
{wk}={v1,u1,v2,u2,…}
Then k→∞limwk=v holds, and Twk also converges to some value Tv=k→∞limTwk. Since Tvk and Tuk are subsequences of Twk, they both must have the same limit value. Therefore, regardless of which sequence converging to v is taken, Tv is uniquely determined.
Therefore, T is bounded, and ∥T∥≤∥T∥ is valid. Now let’s show that the opposite inequality holds. For vectors that are v∈W, the following holds:
∥Tv∥=∥Tv∥≤∥T∥∥v∥
However, for vectors that are v∈V1∖W, there can be vectors that are not bound as above by ∥T∥. Hence, the validity of ∥T∥≥∥T∥ can be known. Therefore, since both sides of the inequality hold, we obtain:
∥T∥=∥T∥
■
Ole Christensen, Functions, Spaces, and Expansions: Mathematical Tools in Physics and Engineering (2010), p55 ↩︎
Erwin Kreyszig, Introductory Functional Analysis with Applications (1989), p99-101 ↩︎