Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $X$ be a normed linear space with norm $||\cdot||$ and $A \neq \emptyset$ is a linear subspace of $X$. Prove that $\bar{A}$ is also a linear subspace of $X$.

share|improve this question
What have you tried? –  Alex Becker Sep 25 '12 at 18:48
I m not able to visualize the additive and multiplicative closure of the new two points added to the subspace closure[A] –  nimesh.iitd Sep 25 '12 at 18:54

2 Answers 2

up vote 1 down vote accepted

Show that the closure $\overline Y$ of a subspace $Y$ of a normed space $X$ is again a vector subspace.

Proof: Suppose $x, y \in \overline Y$. There are three cases we must consider.

  1. If $x, y \in Y$ then it is immediate that $x + y \in Y$ since $Y$ is a subspace of $X$ and thus $x + y \in \overline Y$.
  2. If $x \in \overline Y \setminus Y$ and $y \in Y$ then by definition, given $\varepsilon > 0$ there exists $x_0 \in Y$ with $x_0 \neq x$ such that $d(x_0, x) = \|x_0 - x\| < \varepsilon$. Now observe that since $x_0 \neq x$ then $x_0 + y \neq x + y$ and $x_0 + y \in Y$. Now we have $d(x_0 + y, x + y) = \|x_0 + y - x - y\| = \|x_0 - x\| < \varepsilon$. Hence we can conclude that $x_0 + y$ is an accumulation point of $Y$. Hence $x + y \in \overline Y$.
  3. If $x, y \in \overline Y \setminus Y$ then by definition given $\varepsilon > 0$ there exists $x_0 \in Y$ with $x_0 \neq x$ such that $d(x_0, x) = \|x_0 - x\| < \varepsilon/2$. Similarly, there exists $y_0 \in Y$ with $y_0 \neq y$ such that $d(y_0, y) = \|y_0 - y\| < \varepsilon/2$. If $x_0 + y_0 = x + y$ choose some $x_0 \neq x_0' \in Y$ such that $d(x_0', x) = \|x_0' - x\| < \varepsilon/2$ which is true since there are infinitely many elements in $Y$ with that property. So suppose without loss of generality that $x_0 + y_0 \neq x + y$. It follows that $$d(x_0 + y_0, x + y) = \|x_0 + y_0 - x - y\| \leq \|x_0 - x\| + \|y_0 - y\| < \frac\varepsilon2 + \frac\varepsilon2 = \varepsilon.$$ Hence we have shown that $x + y$ is an accumulation point of $Y$ and hence $x + y \in \overline Y$.

Now suppose $x \in \overline Y$. Suppose $\alpha$ is a scalar. We must consider two cases.

  1. If $x \in Y$ then since $Y$ is a subspace of $X$ it follows immediately that $\alpha x \in Y$ and thus $\alpha x \in \overline Y$.
  2. If $x \in \overline Y \setminus Y$ then $x$ is an accumulation point of $Y$. By definition given $\varepsilon > 0$ there exists $x_0 \in Y$ with $x_0 \neq x$ such that $d(x_0, x) = \|x_0 - x\| < \varepsilon/(\vert \alpha \vert + 1)$. Now since $x_0 \neq x$ then $\alpha x_0 \neq \alpha x$ and observe that $$d(\alpha x_0, \alpha x) = \|\alpha x_0 - \alpha x \| = \vert \alpha \vert \|x_0 - x\| < \vert \alpha \vert \frac{\varepsilon}{\vert \alpha \vert +1} < \varepsilon.$$ Thus we have shown that $\alpha x$ is an accumulation point and we can conclude that $\alpha x \in \overline Y$.

Thus we can conclude that $\overline Y$ is a subspace of $X$.

Note: It is my understanding that the closure of a set is the set along with all its limit points or accumulation points. It is my understanding that it NOT implied that all the elements of a set are limit points of the set. Please correct me if I'm wrong here!

share|improve this answer

It is sufficient to prove that $\alpha x + \beta y \in \overline{X}$ where $\alpha, \beta$ are in the underlying field $\mathbb{F}$ and $x, y \in \overline{X}$. We know that $0 \in \overline{X}$ since $X \subset \overline{X}$. Since $x, y \in \overline{X}$ there exist $x_{j}, y_{j} \in X$ such that $x_j \to x$ and $y_j \to y$. Since multiplication and addition are continuous $\alpha x_j + \beta y_j \to \alpha x + \beta y$. Therefore, $\alpha x + \beta y \in \overline{X}$

share|improve this answer
thanks a lot !!! –  nimesh.iitd Sep 25 '12 at 19:04

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.