Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $U$ and $W$ be subspaces of an inner product space $V$. If $U$ is a subspace of $W$, then $W^{\bot}$ is a subspace of $U^{\bot}$?.

I don't find the above statement intuitively obvious. Could someone provide a proof?

share|cite|improve this question
up vote 4 down vote accepted

If you're orthogonal to everything in a set, then you're also orthogonal to everything in every subset of that set.

Put another way: Elements of $W^\perp$ have to be orthogonal to more vectors than elements of $U^\perp$; they have to be orthogonal to the vectors in $U$ and the vectors in $W\setminus U$ (if any). Therefore there are less of them than if we took the vectors that only have to satisfy the property of being orthogonal to vectors in $U$. (More restrictions$\implies$ Fewer vectors satisfying the restrictions.)

share|cite|improve this answer
And that since you know that $U^\perp,W^\perp$ are subspaces (hopefully) of $V$ then it suffices to show containment. – Alex Youcis Nov 6 '11 at 22:30
So shouldn't it be $U^{\bot}$ is a subspace of $W^{\bot}$ ? That is my confusion. – Mark Nov 6 '11 at 22:33
@Mark: No. If you're orthogonal to everything in $W$, i.e., if you're in $W^\perp$, then you're also orthogonal to everything in a subset of $W$ such as $U$, i.e., you're in $U^\perp$. For a concrete example, think of $\mathbb R^3$ with standard dot product, let $U$ be the span of $(1,0,0)$, and let $W$ be the span of $\{(1,0,0),(0,1,0)\}$. If you have to be orthogonal to both $(1,0,0)$ and $(0,1,0)$, then you are in particular orthogonal to $(1,0,0)$, so $W^\perp\subset U^\perp$. – Jonas Meyer Nov 6 '11 at 22:35
Oh right, I had this at the back of my head, but was unable to explain it clearly. Now I understand, thanks. – Mark Nov 6 '11 at 22:54

It should be intuitive, already at the level of logic:

To be in $W^\perp$, you have to satisfy a certain condition $P(w)$ (namely: 'be orthogonal to $w$') for each and every element $w\in W$.

So given a subset $U\subseteq W$, to be in $U^\perp$ means you have to satisfy $P(w)$ merely for all $u\in U$.

Thus you have to satisfy less properties to be in $U^\perp$, thus it is easier to be in $U^\perp$, thus $U^\perp$ is larger: $W^\perp\subseteq U^\perp$.

It should also be intuitive geometrically: consider $\mathbb{R}^3$, let $U$ be the $x$-axis, and $W$ the $x,y$-plane. Then $U^\perp$ is the $y,z$-plane, and $W^\perp$ is the $z$-axis.

//Edit: I was slow so I missed Joans Meyer's edit, which kind of makes my answer redundant.

share|cite|improve this answer

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.