The weak topology of locally convex spaces and

the weak-* topology of their duals

Jordan Bell
jordan.bell@gmail.com
Department of Mathematics, University of Toronto
April 3, 2014

1 Introduction
These notes give a summary of results that everyone who does work in functional
analysis should know about the weak topology on locally convex topological
vector spaces and the weak-* topology on their dual spaces. The most striking
of the results we prove is Theorem 9, which shows that a subset of a locally
convex space is bounded if and only if it is weakly bounded. It is straightforward
to prove that if a set is bounded then it is weakly bounded, but to prove that if
a set is weakly bounded then it is bounded we use the Hahn-Banach separation
theorem, the Banach-Alaoglu theorem, and the uniform boundedness principle.
If X is a topological vector space then we will see that the weak topology
on it is coarser than the original topology: any set that is open in the original
topology is open in the weak topology. From this it follows that it is easier
for a sequence to converge in the weak topology than in the original topology:
for a sequence to converge to a point means that it is eventually contained in
every neighborhood of the point, and a point has fewer neighborhoods in the
weak topology than it does in the original topology. The weak topology encodes
information we may care about, and we may be able to establish that certain
sets are compact in the weak topology that are not compact in the original
topology.
In these notes I first define the weak topology on a topological vector space
X, and show that if X is locally convex then X with the weak topology is also
a locally convex space. Indeed a normed space is locally convex, but there are
function spaces that we care about that are not normed spaces. For example,
the set of holomorphic functions on the open unit disc is a Fréchet space1 that is
not normable. If U is an open subset of Rn , then C k (U ) is a Fréchet space that
1 A Fréchet space is a complete metrizable locally convex space. It is a fact that a locally

convex space is metrizable if and only its topology is induced by countably many of its semi-
norms. See John B. Conway, A Course in Functional Analysis, second ed., p. 105, chapter
IV, Proposition 2.1.

1
is not normable, and the set of Schwartz functions on U is also a Fréchet space
that is not normable. Moreover, none of the theorems stated for topological
vector spaces and locally convex spaces is much easier to prove in the case of
normed spaces, and thus it is not a great waste of time to digest their statements
in a larger category of spaces.
Except for the results about normed spaces, the hypotheses of all the the-
orems that we present are satisfied for a separable metrizable locally convex
space. The hypotheses of every result in this note are satisfied for separable
reflexive Banach spaces, for example, Lp (Rn ) for 1 < p < ∞.
If every instance of C in this text is replaced with R, the resulting text does
not require any further changes to be correct.

2 Topological vector spaces

If X is a topological space and x ∈ X and B is a set of open sets, we say that
B is a local basis at x if each element of B includes x and if for every open set
U that includes x there is some V S ∈ B such that V ⊆ U . If for each x ∈ X the
set Bx is a local basis at x, then x∈X Bx is a basis for the topology of X. If X
is a vector space, B is a set of subsets of X, and x ∈ X, we define

x + B = {x + N : N ∈ B}, x + N = {x + y : y ∈ N }.

If X is a vector space over C with a topology O(X) such that (X, O(X))
is Hausdorff and such that addition X × X → X and scalar multiplication
C × X → X are continuous, we say that X is a topological vector space. It is
straightforward toSprove that if B is a local basis at 0 then x + B is a local
basis at x, and so x∈X {x + B} is a basis for the topology O(X). To specify a
topology on a vector space it suffices to specify a local basis at 0: This gives a
basis by taking the union of the translates of the local basis over all x ∈ X, and
then this basis generates a topology. However, X might not be a topological
vector space with the topology thus generated. (That is, if we define a topology
on a vector space by declaring certain sets including the origin to be open, the
vector space need not be a topological vector space with this topology.)
A topological vector space (X, O(X)) is said to be locally convex if there is a
local basis at 0 each element of which is convex. If να is a family of seminorms
on X, we define the seminorm topology induced by this family to be the coarsest
topology on X such that for all x0 ∈ X and α ∈ I, the map x 7→ να (x − x0 )
is continuous. We say that a set of seminorms να is a separating family if
x 6= 0 implies that there is some α such that να (x) > 0. One can prove that a
topological vector space is locally convex if and only if its topology is induced
by a separating family of seminorms.2
Let (X, O(X)) be a topological vector space over C. The dual space X ∗ is
the set of all continuous linear maps (X, O(X)) → C. The weak topology on X,
which we denote by Ow (X), is the initial topology for X ∗ . That is, Ow (X) is the
2 Walter Rudin, Functional Analysis, second ed., p. 27, Theorem 1.36 and Theorem 1.37.

2
coarsest topology on X so that each element of X ∗ is continuous (X, Ow (X)) →
C. Equivalently, the weak topology on X is the seminorm topology induced by
the seminorms |α|, α ∈ X ∗ . The topologies O(X) and Ow (X) are comparable,
and O(X) is at least as fine as Ow (X). That is, Ow (X) ⊆ O(X). A vague rule
is that the smaller X ∗ is compared to the set of all linear maps X → C, the
smaller Ow (X) will be compared to O(X). If X ∗ separates X then (X, Ow (X))
is a locally convex topological vector space. It is locally convex because Ow (X)
is induced by the separating family of seminorms |α|, α ∈ X ∗ .

• We say that xk → x weakly if xk → x in (X, Ow (X)): for every neighbor-

hood N of x in (X, Ow (X)), the sequence xk is eventually in N .
• We say that A is weakly bounded if A is a bounded subset of (X, Ow (X)):
for every neighborhood N of 0 in (X, Ow (X)) there is some c ≥ 0 such
that A ⊆ {cx : x ∈ N } = cN .

3 Locally convex spaces

The Hahn-Banach theorem separation theorem states the following.3

Theorem 1 (Hahn-Banach separation theorem). If

• X is a locally convex topological vector space over C
• A, B are disjoint, nonempty, closed, convex subsets of X
• A is compact

then there is some λ ∈ X ∗ and γ1 , γ2 ∈ R such that, for all a ∈ A and b ∈ B,

Re λ(a) < γ1 < γ2 < Re λ(b).

It follows that if X is locally convex then X ∗ separates X. We can also use

the Hahn-Banach separation theorem to prove that in a locally convex space,
the weak closure of a convex set is equal to its original closure.4
Theorem 2. If E is a convex subset of a locally convex space (X, O(X)), then
the closure E of E in (X, O(X)) is equal to the closure E w of E in (X, Ow (X)).

3 Walter Rudin, Functional Analysis, second ed., p. 59, Theorem 3.4.

4 Walter Rudin, Functional Analysis, second ed., p. 66, Theorem 3.12.

3
Proof. For A ⊆ X, denote Ac = X \ A. We have, as Ow (X) is coarser than
O(X),
 c
[
E =  U
U ∈O(X),E⊆U c
 c
[
⊆  U
U ∈Ow (X),E⊆U c

= Ew.
In the other direction, let x0 6∈ E. If we can show that x0 6∈ E w , this will
show that E w ⊆ E and hence that E = E w . Let A = {x0 } and B = E, which
satisfy the conditions of the Hahn-Banach theorem. Thus there is some λ ∈ X ∗
and γ1 , γ2 ∈ R such that for all x ∈ E,
Re λ(x0 ) < γ1 < γ2 < Re λ(x).
Let U = {x ∈ X : Re λ(x) < γ1 }. The set V = {z ∈ C : Re z < γ1 } is an
open subset of C and U = λ−1 (V ), so U ∈ Ow (X). For every x ∈ E we have
Re λ(x) > γ2 > γ1 , and for every x ∈ U we have Re λ(x) < γ1 , so E ∩ U = ∅ and
in particular E ∩ U = ∅. The three facts x0 ∈ U, U ∈ Ow (X), and E ∩ U = ∅
imply that x0 6∈ E w , completing the proof.
If a sequence converges weakly, it need not converge in the original topology.
Mazur’s theorem shows that if a sequence in a metrizable locally convex space
converges weakly then there is a sequence in the convex hull of the original
sequence that converges to the same limit as the weak limit of the original
sequence.5
Theorem 3 (Mazur’s theorem). Let X be a metrizable locally convex space. If
xn → x weakly, then there is a sequence yi ∈ X such that each yi is a convex
combination of finitely many xn and such that yi → x.
Proof. The convex hull of a subset A of X is the set of all convex combinations
of finitely many elements of A. The convex hull of a set is convex and contains
the set. Let H be the convex hull of the sequence xn , and let K be the weak
closure of H. Since xn → x weakly and xn ∈ H, it follows that x ∈ K. As H
is convex, Theorem 2 tells us that K = H, so x ∈ H. But X is metrizable, so
x being in the closure of H implies that there is a sequence yi ∈ H such that
yi → x. This sequence yi satisfies the claim.

4 Weak-* topology
If X is a vector space over C and F is a set of linear maps X → C that separates
X, and we give X the initial topology for F, then one can prove that with this
5 Walter Rudin, Functional Analysis, second ed., p. 67, Theorem 3.13.

4
topology X is a locally convex space whose dual space is F.6
Let X be a topological vector space over C, and for x ∈ X, define fx :
X ∗ → C by fx (λ) = λ(x). fx is linear. If λ1 , λ2 ∈ X ∗ are distinct, then
λ1 −λ2 6= 0 so there is some x ∈ X such that (λ1 −λ2 )(x) 6= 0, which tells us that
fx (λ1 ) 6= fx (λ2 ). Therefore the set {fx : x ∈ X} separates X ∗ . Let OX (X ∗ ) be
the initial topology for {fx : x ∈ X}, and by the previous paragraph we have
that (X ∗ , OX (X ∗ )) is a locally convex space whose dual space is {fx : x ∈ X}.
The topology OX (X ∗ ) is called the weak-* topology on X ∗ . We record what
we’ve just said as a theorem to make it easier to look up.
Theorem 4. If X is a topological vector space, then its dual X ∗ with the weak-*
topology OX (X ∗ ) is a locally convex space, and the dual space of (X ∗ , OX (X ∗ ))
is the set of fx : X ∗ → C, where fx (λ) = λ(x), x ∈ X, λ ∈ X ∗ .
The Banach-Alaoglu theorem7 shows that certain subsets of X ∗ are weak-*
compact, i.e. they are compact subsets of (X, OX (X ∗ )). The set K in the
statement of the theorem is called the polar of the set V .

Theorem 5 (Banach-Alaoglu theorem). If X be a topological vector space and

V ⊆ X is a neighborhood of 0, then

K = {λ ∈ X ∗ : if x ∈ V then |λ(x)| ≤ 1}

is a compact subset of (X ∗ , OX (X ∗ )).

The following theorem shows that in a separable topological vector space, a

weak-* compact set is weak-* metrizable.8 This is useful because if possible we
would like to characterize a topology by its convergent sequences rather than
by its open sets.

Theorem 6. If X is a separable topological vector space and K is a weak-

* compact subset of X ∗ , then K with the subspace topology inherited from
(X ∗ , OX (X ∗ )) is metrizable.
Proof. Let {xn } be a countable dense set in X, and define fn : X ∗ → C by
fn (λ) = λ(xn ). For each n, fn : (X ∗ , OX (X ∗ )) → C is linear and continuous.
For distinct λ1 , λ2 ∈ X ∗ , the set U = {x ∈ X : λ1 (x) 6= λ2 (x)} is an open subset
of X. As U is a nonempty open set and {xn } is a dense subset of X, there is
some xn ∈ U , giving fn (λ1 ) 6= fn (λ2 ). Therefore {fn : X ∗ → C} separates X ∗ .
It is a fact that if Y is a compact topological space and there is a countable set
of continuous functions Y → C that separates Y then Y is metrizable.9 K is a
compact topological space with the subspace topology inherited from OX (X ∗ ),
and hence this topology on K is metrizable.
6 Walter Rudin, Functional Analysis, second ed, p. 64, Theorem 3.10.
7 Walter Rudin, Functional Analysis, second ed., p. 68, Theorem 3.15.
8 Walter Rudin, Functional Analysis, second ed., p. 70, Theorem 3.16.
9 Walter Rudin, Functional Analysis, second ed., p. 63, §3.8, (c). In Rudin it is stated for
real valued functions.

5
 We can combine Theorem 5 and Theorem 6 to get the following, which states
that the polar of a neighborhood of 0 in a separable topological vector space is
weak-* sequentially compact.10
Theorem 7. Let X is a separable topological vector space. If V is a neighbor-
hood of 0 and if the sequence λn ∈ X ∗ satisfies

|λn (x)| ≤ 1, n ≥ 1, x ∈ V,

then there is a subsequence λa(n) and some λ ∈ X ∗ such that for all x ∈ X,

lim λa(n) (x) = λ(x).

n→∞

Proof. The Banach-Alaoglu Theorem implies that the polar

K = {λ ∈ X ∗ : if x ∈ V then |λ(x)| ≤ 1},

is weak-* compact. K with the subspace topology inherited from OX (X ∗ )

is compact, hence by Theorem 6 it is metrizable. Since the sequence λn is
contained in K, it has a subsequence λa(n) that converges weakly to some λ ∈ K.
For each x ∈ X, the function fx : (X ∗ , OX (X ∗ )) → C defined by fx (λ) = λ(x)
is continuous, hence for all x ∈ X we have fx (λa(n) ) → fx (λ), which is the
claim.
The following lemma gives a tractable characterization of weakly bounded
sets.11
Lemma 8. If X is a topological vector space and E is a subset of X, then E
is weakly bounded if and only if for all λ ∈ X ∗ there is some γ(λ) such that if
x ∈ E then |λ(x)| ≤ γ(λ).
The following theorem12 shows that in a locally convex space, the bound-
edness of a set is equivalent to its weak boundedness. Its proof involves much
of what we have talked about so far, and actually requires introducing some
new terms with which one should probably have at least an acquaintance. The
statement of the theorem does not involve the weak-* topology on X ∗ , but the
proof uses the Banach-Alaoglu theorem.
Theorem 9 (Weak boundedness is equivalent to boundedness). If X is locally
convex and E ⊆ X, then E is bounded in (X, O(X)) if and only if E is bounded
in (X, Ow (X)).
Proof. Suppose that E is bounded in (X, O(X)) (for every neighborhood of 0,
the set E is contained in some dilation of that neighborhood). Let N be a
neighborhood of 0 in Ow (X). There is some U ∈ Ow (X) with 0 ∈ U ⊆ N .
Now, U ∈ O(X), so N is a neighborhood of x in O(X). As E is bounded, there
10 Walter Rudin, Functional Analysis, second ed., p. 70, Theorem 3.17.
11 Walter Rudin, Functional Analysis, second ed., p. 66.
12 Walter Rudin, Functional Analysis, second ed., p. 70, Theorem 3.18.

6
is some α such that E ⊆ αN = {αx : x ∈ N }, which is what it means for E to
be bounded in (X, Ow (X)).
Suppose that E is bounded in (X, Ow (X)). Let N be a neighborhood of 0 in
(X, O(X)). We want to show that there is some α such that E ⊆ {αx : x ∈ N }.
If B is a local basis at a point in a topological vector space, then every element
of B contains the closure of some element of B.13 A subset A of a vector space
is said to be balanced if for every α ∈ C with |α| ≤ 1 we have αA ⊆ A. It is a
fact that in a topological vector space, every convex neighborhood of 0 contains
a convex balanced neighborhood of 0.14 The purpose of having said all of this
is the following: since N is a neighborhood of 0 in O(X), there is a balanced
convex V that is a neighborhood of 0 in (X, O(X)) such that V ⊆ N , where
V is the closure of V in (X, O(X)). Moreover, it is a fact that the closure of a
balanced set is itself balanced, and the closure of a convex set is itself convex,15
so V is a balanced convex set that is contained in N . Thus to prove E ⊆ αN it
suffices to prove that E ⊆ αV , and this is what we will do.
Let
K = {λ ∈ X ∗ : if x ∈ V then |λ(x)| ≤ 1},
the polar of V . If x ∈ V and λ ∈ K then |λ(x)| ≤ 1, so
\
V ⊆ {x ∈ X : if λ ∈ K then |λ(x)| ≤ 1} = {x ∈ X : |λ(x)| ≤ 1}.
λ∈K

The right-hand side is an intersection of closed sets in (X, O(X)), so it is closed.

Therefore V is contained in the right-hand side. Furthermore, it is a consequence
of the Hahn-Banach separation theorem16 that if B is a convex balanced closed
set in a locally convex space and x0 6∈ B, then there is some λ ∈ X ∗ such that
λ(x0 ) > 1 and |λ(x)| ≤ 1 for x ∈ B. Thus, if x0 6∈ V , then there is some λ ∈ X ∗
such that λ(x0 ) > 1 and if x ∈ V then |λ(x)| ≤ 1. From this it follows that
λ ∈ K, and then because λ(x0 ) > 1 it follows that

x0 6∈ {x ∈ X : if λ ∈ K then |λ(x)| ≤ 1}.

We have shown that if x0 6∈ V then x0 is not an element of the polar of K, and

therefore
V = {x ∈ X : if λ ∈ K then |λ(x)| ≤ 1}.
By the Banach-Alaoglu theorem, K is a compact set in (X ∗ , OX (X ∗ )), and
one checks that K is convex. The uniform boundedness principle17 states that
if
• W and Z are topological vector spaces
• K is a compact convex set in W
13 Walter Rudin, Functional Analysis, second ed., p. 11, Theorem 1.11.
14 Walter Rudin, Functional Analysis, second ed., p. 12, Theorem 1.14.
15 Walter Rudin, Functional Analysis, second ed., p. 11, Theorem 1.13.
16 Walter Rudin, Functional Analysis, second ed., p. 61, Theorem 3.7.
17 Walter Rudin, Functional Analysis, second ed., p. 46, Theorem 2.9

7
 • Γ is a set of continuous linear maps W → Z
• For each w ∈ K, the set Γ(w) = {g(w) : g ∈ Γ} is a bounded subset of Z
then there is a bounded set B in Z such that if g ∈ Γ then g(K) ⊆ B. For
x ∈ E, define gx : X ∗ → C by gx (λ) = λ(x), and define Γ = {gx : x ∈ E}.
Because E is weakly bounded, for all λ ∈ X ∗ there is some γ(λ) such that if
x ∈ E then |λ(x)| ≤ γ(λ); this is by Lemma 8. Hence for all λ ∈ X ∗ there
is some γ(λ) such that the set Γ(λ) is contained in the closed disc in C with
radius γ(λ). Thus Γ(λ) is bounded in C. We apply the uniform boundedness
principle using W = (X ∗ , OX (X ∗ )) and Z = C. We thus obtain that there is
some 0 ≤ γ < ∞ such that if g ∈ Γ then g(K) is contained in the closed disc of
radius γ. That is, for all x ∈ E and λ ∈ K, we have
|gx (λ)| ≤ γ,
i.e. for all x ∈ E and λ ∈ K we have
|λ(x)| ≤ γ.
Therefore, if x ∈ E and λ ∈ K, then
  
λ 1 · x  = 1 |λ(x)| ≤ 1,
 
 γ  γ

which means that γ1 · x ∈ V . This is true for all x ∈ E, so γ1 · E ⊆ V ⊆ N , and

hence E ⊆ γ · N . Since N was an arbitrary neighborhood of 0 in (X, O(X)), we
have satisfied the definition of the set E being bounded in (X, O(X)), completing
the proof.
The final result we state in this section gives a condition for the dual of a
locally convex space to be weak-* separable.18 We already stated in Theorem
4 that the dual with the weak-* topology of a topological vector space is itself
a locally convex space.
Theorem 10. If X is a separable metrizable locally convex space, then (X ∗ , OX (X ∗ ))
is a separable locally convex space.

5 Normed spaces
Let X be a normed space with norm k·k. The topology on X is the coarsest
topology such that for each x0 ∈ X, the map x 7→ kx − x0 k is continuous. A
normed vector space is locally convex and is metrizable, with metric d(x, y) =
kx − yk.
If X and Y are topological vector spaces and T : X → Y is a linear map, we
say that T is bounded if E being bounded in X implies that T (E) is bounded
in Y .19
18 WalterRudin, Functional Analysis, second ed., p. 90, chapter 3, Exercise 28.
19 Astatement close to it is proved in Walter Rudin, Functional Analysis, second ed., p. 24,
Theorem 1.32.

8
Theorem 11. Let X and Y be normed vector spaces and let T : X → Y be
linear. The following three statements are equivalent:
• T is continuous.
• T is bounded.
• There is some γ such that if x ∈ X then kT xk ≤ γ kxk.

If X and Y are normed spaces and T : X → Y is a bounded linear map,

the operator norm kT k of T is defined to be the infimum of those γ such that
if x ∈ X then kT xk ≤ γ kxk. kT k = supkxk≤1 kT xk. The set of bounded linear
maps X → Y is denoted B(X, Y ). B(X, Y ) is a normed space, and if Y is a
Banach space then B(X, Y ) is a Banach space.20 X ∗ = B(X, C), and since C
is a Banach space, with the operator norm X ∗ is a Banach space.
The following theorem rewrites Theorem 9 in the terminology of norms.21
Theorem 12 (Weak boundedness is equivalent to boundedness). If X is a
normed space and E is a subset of X, then

sup |λ(x)| < ∞

x∈E

for all λ ∈ X ∗ if and only if there is some γ such that x ∈ E implies

kxk ≤ γ.

Proof. Suppose that supx∈E |λ(x)| < ∞ holds for all λ ∈ X ∗ . By Lemma 8,
this means that the set E is weakly bounded. But by Theorem 9 this implies
that E is bounded. The closed unit ball is a neighborhood of 0 in (X, O(X)),
so, as E is bounded, there is some γ such that E ⊆ {γx : kxk ≤ 1}. Hence, if
x ∈ E then kxk ≤ γ.
Suppose that there is some γ such that x ∈ E implies that kxk ≤ γ, and let
λ ∈ X ∗ . Because λ is continuous, for x ∈ E we have

|λ(x)| ≤ kλk kxk ≤ kλk γ < ∞.

Theorem 13 (Banach-Alaoglu theorem). If X be a normed vector space, then

B = {λ ∈ X ∗ : kλk ≤ 1} is a compact subset of (X, OX (X ∗ )).
Proof. It is a fact22 (proved using the Hahn-Banach extension theorem) that
for every x ∈ X,
kxk = sup{|λ(x)| : λ ∈ B}.
From this and Theorem 5, it follows that B is weak-* compact in X ∗ .
20 Walter Rudin, Functional Analysis, second ed., p. 92, Theorem 4.
21 Walter Rudin, Functional Analysis, second ed., p. 71.
22 Walter Rudin, Functional Analysis, second ed., p. 94, Theorem 4.3.

9
 The Eberlein-Smulian theorem states that a set in a normed space is weakly
compact if and only if the set is weakly sequentially compact.23
Theorem 14 (Eberlein-Smulian theorem). If X is a normed space, then a
subset A of X is compact in Ow (X) if and only if every sequence in A has a
subsequence that converges in Ow (X) to an element of A.

6 Banach spaces
We say that a Banach space X is reflexive if X ∗∗ = X, where X ∗ = B(X, C)
and X ∗∗ = B(X ∗ , C). (Although it makes sense to talk about X ∗ for a normed
space, X ∗ is itself a Banach space and so too is X ∗∗ , hence if a normed space
were reflexive then it would have to be a Banach space.)
Kakutani’s theorem relates the property of a Banach space being reflexive
with weak compactness.24
Theorem 15 (Kakutani’s theorem). The closed unit ball in a Banach space is
weakly compact if and only if the space is reflexive.
Thus, combining the Eberlein-Smulian theorem and Kakutani’s theorem we
get that a Banach space is reflexive if and only if the closed unit ball is weakly
sequentially compact.
The following theorem states that a Banach space is separable if and only if
the closed unit ball in the dual space is weak-* metrizable.25
Theorem 16. Let X be a Banach space and let B = {λ ∈ X ∗ : kλk ≤
1}. X is separable if and only if B with the subspace topology inherited from
(X ∗ , OX (X ∗ )) is metrizable.

7 Hilbert spaces
Let H be a Hilbert space with inner product h·, ·i : H × H → C, linear in the
first argument. The Riesz representation theorem26 states that if λ ∈ H ∗ , then
there is a unique hλ ∈ H such that if h ∈ H then

λ(h) = hh, hλ i ,

and kλk = khλ k. On the other hand, if h0 ∈ H then λ(h) = hh, h0 i satisfies, by
the Cauchy-Schwarz inequality,

|λh0 (h)| = | hh, h0 i | ≤ khk kh0 k ,

hence, by Theorem 11, λh0 ∈ H ∗ . Thus H ∗ = {λh0 : h0 ∈ H}. The weak

topology Ow (H) on H is the coarsest topology on H such that each λ ∈ H ∗
23 Robert E. Megginson, An Introduction to Banach Space Theory, p. 248, Theorem 2.8.6.
24 JosephDiestel, Sequences and Series in Banach Spaces, p. 18, chapter III.
25 John B. Conway, A Course in Functional Analysis, second ed., p. 134, Theorem 5.1.
26 John B. Conway, A Course in Functional Analysis, second ed., p. 13, Theorem 3.4.

10
is continuous (H, Ow (H)) → C, hence such that for each h0 ∈ H, the funtion
h 7→ hh, h0 i is continuous (H, Ow (H)) → C. Thus, a net hα ∈ H converges
to h ∈ H in Ow (H) if and only if for all h0 ∈ H we have that hhα , h0 i ∈ C
converges to hh, h0 i ∈ C, and this characterizes the weak topology on H.

11