Chapter 2

Finite-Dimensional
Vector Spaces

In the last chapter we learned about vector spaces. Linear algebra

focuses not on arbitrary vector spaces, but on finite-dimensional vector
spaces, which we introduce in this chapter. Here we will deal with the
key concepts associated with these spaces: span, linear independence,
basis, and dimension.
Let’s review our standing assumptions:

Recall that F denotes R or C.

Recall also that V is a vector space over F.

✽✽ 21
22 Chapter 2. Finite-Dimensional Vector Spaces

Span and Linear Independence

A linear combination of a list (v1 , . . . , vm ) of vectors in V is a vector
of the form

2.1 a1 v1 + · · · + am vm ,

where a1 , . . . , am ∈ F. The set of all linear combinations of (v1 , . . . , vm )

Some mathematicians is called the span of (v1 , . . . , vm ), denoted span(v1 , . . . , vm ). In other
use the term linear words,
span, which means the
same as span. span(v1 , . . . , vm ) = {a1 v1 + · · · + am vm : a1 , . . . , am ∈ F}.

As an example of these concepts, suppose V = F3 . The vector

 
(7, 2, 9) is a linear combination of (2, 1, 3), (1, 0, 1) because

(7, 2, 9) = 2(2, 1, 3) + 3(1, 0, 1).

 
Thus (7, 2, 9) ∈ span (2, 1, 3), (1, 0, 1) .
You should verify that the span of any list of vectors in V is a sub-
space of V . To be consistent, we declare that the span of the empty list
() equals {0} (recall that the empty set is not a subspace of V ).
If (v1 , . . . , vm ) is a list of vectors in V , then each vj is a linear com-
bination of (v1 , . . . , vm ) (to show this, set aj = 1 and let the other a’s
in 2.1 equal 0). Thus span(v1 , . . . , vm ) contains each vj . Conversely,
because subspaces are closed under scalar multiplication and addition,
every subspace of V containing each vj must contain span(v1 , . . . , vm ).
Thus the span of a list of vectors in V is the smallest subspace of V
containing all the vectors in the list.
If span(v1 , . . . , vm ) equals V , we say that (v1 , . . . , vm ) spans V . A
Recall that by vector space is called finite dimensional if some list of vectors in it
definition every list has spans the space. For example, Fn is finite dimensional because
finite length.  
(1, 0, . . . , 0), (0, 1, 0, . . . , 0), . . . , (0, . . . , 0, 1)

spans Fn , as you should verify.

Before giving the next example of a finite-dimensional vector space,
we need to define the degree of a polynomial. A polynomial p ∈ P(F)
is said to have degree m if there exist scalars a0 , a1 , . . . , am ∈ F with
am = 0 such that

2.2 p(z) = a0 + a1 z + · · · + am zm
 Span and Linear Independence 23

for all z ∈ F. The polynomial that is identically 0 is said to have de-

gree −∞.
For m a nonnegative integer, let Pm (F) denote the set of all poly-
nomials with coefficients in F and degree at most m. You should ver-
ify that Pm (F) is a subspace of P(F); hence Pm (F) is a vector space.
This vector space is finite dimensional because it is spanned by the list
(1, z, . . . , zm ); here we are slightly abusing notation by letting zk denote
a function (so z is a dummy variable).
A vector space that is not finite dimensional is called infinite di- Infinite-dimensional
mensional. For example, P(F) is infinite dimensional. To prove this, vector spaces, which
consider any list of elements of P(F). Let m denote the highest degree we will not mention
of any of the polynomials in the list under consideration (recall that by much anymore, are the
definition a list has finite length). Then every polynomial in the span of center of attention in
this list must have degree at most m. Thus our list cannot span P(F). the branch of
Because no list spans P(F), this vector space is infinite dimensional. mathematics called
The vector space F∞ , consisting of all sequences of elements of F, functional analysis.
is also infinite dimensional, though this is a bit harder to prove. You Functional analysis
should be able to give a proof by using some of the tools we will soon uses tools from both
develop. analysis and algebra.
Suppose v1 , . . . , vm ∈ V and v ∈ span(v1 , . . . , vm ). By the definition
of span, there exist a1 , . . . , am ∈ F such that

v = a1 v1 + · · · + am vm .

Consider the question of whether the choice of a’s in the equation

above is unique. Suppose â1 , . . . , âm is another set of scalars such that

v = â1 v1 + · · · + âm vm .

Subtracting the last two equations, we have

0 = (a1 − â1 )v1 + · · · + (am − âm )vm .

Thus we have written 0 as a linear combination of (v1 , . . . , vm ). If the

only way to do this is the obvious way (using 0 for all scalars), then
each aj − âj equals 0, which means that each aj equals âj (and thus
the choice of a’s was indeed unique). This situation is so important
that we give it a special name—linear independence—which we now
define.
A list (v1 , . . . , vm ) of vectors in V is called linearly independent if
the only choice of a1 , . . . , am ∈ F that makes a1 v1 + · · · + am vm equal
0 is a1 = · · · = am = 0. For example,
24 Chapter 2. Finite-Dimensional Vector Spaces

 
(1, 0, 0, 0), (0, 1, 0, 0), (0, 0, 1, 0)

is linearly independent in F4 , as you should verify. The reasoning in the

previous paragraph shows that (v1 , . . . , vm ) is linearly independent if
and only if each vector in span(v1 , . . . , vm ) has only one representation
as a linear combination of (v1 , . . . , vm ).
Most linear algebra For another example of a linearly independent list, fix a nonnegative
texts define linearly integer m. Then (1, z, . . . , zm ) is linearly independent in P(F). To verify
independent sets this, suppose that a0 , a1 , . . . , am ∈ F are such that
instead of linearly
independent lists. With 2.3 a0 + a1 z + · · · + am zm = 0
that definition, the set
for every z ∈ F. If at least one of the coefficients a0 , a1 , . . . , am were
{(0, 1), (0, 1), (1, 0)} is
nonzero, then 2.3 could be satisfied by at most m distinct values of z (if
linearly independent in
you are unfamiliar with this fact, just believe it for now; we will prove
F2 because it equals the
it in Chapter 4); this contradiction shows that all the coefficients in 2.3
set {(0, 1), (1, 0)}. With
equal 0. Hence (1, z, . . . , zm ) is linearly independent, as claimed.
our definition, the list A list of vectors in V is called linearly dependent if it is not lin-
(0, 1), (0, 1), (1, 0) is
early independent. In other words, a list (v1 , . . . , vm ) of vectors in V
not linearly
is linearly dependent if there exist a1 , . . . , am ∈ F, not all 0, such that
independent (because 1  
a1 v1 + · · · + am vm = 0. For example, (2, 3, 1), (1, −1, 2), (7, 3, 8) is
times the first vector
linearly dependent in F3 because
plus −1 times the
second vector plus 0 2(2, 3, 1) + 3(1, −1, 2) + (−1)(7, 3, 8) = (0, 0, 0).
times the third vector
equals 0). By dealing As another example, any list of vectors containing the 0 vector is lin-
with lists instead of early dependent (why?).
sets, we will avoid You should verify that a list (v) of length 1 is linearly independent if
some problems and only if v = 0. You should also verify that a list of length 2 is linearly
associated with the independent if and only if neither vector is a scalar multiple of the
usual approach. other. Caution: a list of length three or more may be linearly dependent
even though no vector in the list is a scalar multiple of any other vector
in the list, as shown by the example in the previous paragraph.
If some vectors are removed from a linearly independent list, the
remaining list is also linearly independent, as you should verify. To
allow this to remain true even if we remove all the vectors, we declare
the empty list () to be linearly independent.
The lemma below will often be useful. It states that given a linearly
dependent list of vectors, with the first vector not zero, one of the
vectors is in the span of the previous ones and furthermore we can
throw out that vector without changing the span of the original list.
 Span and Linear Independence 25

2.4 Linear Dependence Lemma: If (v1 , . . . , vm ) is linearly depen-

dent in V and v1 = 0, then there exists j ∈ {2, . . . , m} such that the
following hold:

(a) vj ∈ span(v1 , . . . , vj−1 );

(b) if the j th term is removed from (v1 , . . . , vm ), the span of the
remaining list equals span(v1 , . . . , vm ).

Proof: Suppose (v1 , . . . , vm ) is linearly dependent in V and v1 = 0.

Then there exist a1 , . . . , am ∈ F, not all 0, such that

a1 v1 + · · · + am vm = 0.

Not all of a2 , a3 , . . . , am can be 0 (because v1 = 0). Let j be the largest

element of {2, . . . , m} such that aj = 0. Then

a1 aj−1
2.5 vj = − v1 − · · · − vj−1 ,
aj aj

proving (a).
To prove (b), suppose that u ∈ span(v1 , . . . , vm ). Then there exist
c1 , . . . , cm ∈ F such that

u = c1 v1 + · · · + cm vm .

In the equation above, we can replace vj with the right side of 2.5,
which shows that u is in the span of the list obtained by removing the
j th term from (v1 , . . . , vm ). Thus (b) holds.

Now we come to a key result. It says that linearly independent lists

are never longer than spanning lists.

2.6 Theorem: In a finite-dimensional vector space, the length of Suppose that for each
every linearly independent list of vectors is less than or equal to the positive integer m,
length of every spanning list of vectors. there exists a linearly
independent list of m
Proof: Suppose that (u1 , . . . , um ) is linearly independent in V and vectors in V . Then this
that (w1 , . . . , wn ) spans V . We need to prove that m ≤ n. We do so theorem implies that V
through the multistep process described below; note that in each step is infinite dimensional.
we add one of the u’s and remove one of the w’s.
26 Chapter 2. Finite-Dimensional Vector Spaces

Step 1
The list (w1 , . . . , wn ) spans V , and thus adjoining any vector to it
produces a linearly dependent list. In particular, the list

(u1 , w1 , . . . , wn )

is linearly dependent. Thus by the linear dependence lemma (2.4),

we can remove one of the w’s so that the list B (of length n)
consisting of u1 and the remaining w’s spans V .

Step j
The list B (of length n) from step j −1 spans V , and thus adjoining
any vector to it produces a linearly dependent list. In particular,
the list of length (n + 1) obtained by adjoining uj to B, placing it
just after u1 , . . . , uj−1 , is linearly dependent. By the linear depen-
dence lemma (2.4), one of the vectors in this list is in the span of
the previous ones, and because (u1 , . . . , uj ) is linearly indepen-
dent, this vector must be one of the w’s, not one of the u’s. We
can remove that w from B so that the new list B (of length n)
consisting of u1 , . . . , uj and the remaining w’s spans V .

After step m, we have added all the u’s and the process stops. If at
any step we added a u and had no more w’s to remove, then we would
have a contradiction. Thus there must be at least as many w’s as u’s.

Our intuition tells us that any vector space contained in a finite-

dimensional vector space should also be finite dimensional. We now
prove that this intuition is correct.

2.7 Proposition: Every subspace of a finite-dimensional vector

space is finite dimensional.

Proof: Suppose V is finite dimensional and U is a subspace of V .

We need to prove that U is finite dimensional. We do this through the
following multistep construction.

Step 1
If U = {0}, then U is finite dimensional and we are done. If U =
{0}, then choose a nonzero vector v1 ∈ U .

Step j
If U = span(v1 , . . . , vj−1 ), then U is finite dimensional and we are
 Bases 27

done. If U = span(v1 , . . . , vj−1 ), then choose a vector vj ∈ U such

that
vj ∉ span(v1 , . . . , vj−1 ).

After each step, as long as the process continues, we have constructed

a list of vectors such that no vector in this list is in the span of the
previous vectors. Thus after each step we have constructed a linearly
independent list, by the linear dependence lemma (2.4). This linearly
independent list cannot be longer than any spanning list of V (by 2.6),
and thus the process must eventually terminate, which means that U
is finite dimensional.

Bases
A basis of V is a list of vectors in V that is linearly independent and
spans V . For example,
 
(1, 0, . . . , 0), (0, 1, 0, . . . , 0), . . . , (0, . . . , 0, 1)

is a basis of Fn , called the standard basis of Fn . In addition to the

 
standard basis, Fn has many other bases. For example, (1, 2), (3, 5)
 
is a basis of F2 . The list (1, 2) is linearly independent but is not a
 
basis of F2 because it does not span F2 . The list (1, 2), (3, 5), (4, 7)
spans F2 but is not a basis because it is not linearly independent. As
another example, (1, z, . . . , zm ) is a basis of Pm (F).
The next proposition helps explain why bases are useful.

2.8 Proposition: A list (v1 , . . . , vn ) of vectors in V is a basis of V

if and only if every v ∈ V can be written uniquely in the form

2.9 v = a1 v1 + · · · + an vn ,

where a1 , . . . , an ∈ F.

Proof: First suppose that (v1 , . . . , vn ) is a basis of V . Let v ∈ V . This proof is

Because (v1 , . . . , vn ) spans V , there exist a1 , . . . , an ∈ F such that 2.9 essentially a repetition
holds. To show that the representation in 2.9 is unique, suppose that of the ideas that led us
b1 , . . . , bn are scalars so that we also have to the definition of
linear independence.
v = b1 v1 + · · · + bn vn .
28 Chapter 2. Finite-Dimensional Vector Spaces

Subtracting the last equation from 2.9, we get

0 = (a1 − b1 )v1 + · · · + (an − bn )vn .

This implies that each aj − bj = 0 (because (v1 , . . . , vn ) is linearly inde-

pendent) and hence a1 = b1 , . . . , an = bn . We have the desired unique-
ness, completing the proof in one direction.
For the other direction, suppose that every v ∈ V can be written
uniquely in the form given by 2.9. Clearly this implies that (v1 , . . . , vn )
spans V . To show that (v1 , . . . , vn ) is linearly independent, suppose
that a1 , . . . , an ∈ F are such that

0 = a 1 v 1 + · · · + an vn .

The uniqueness of the representation 2.9 (with v = 0) implies that

a1 = · · · = an = 0. Thus (v1 , . . . , vn ) is linearly independent and
hence is a basis of V .

A spanning list in a vector space may not be a basis because it is not

linearly independent. Our next result says that given any spanning list,
some of the vectors in it can be discarded so that the remaining list is
linearly independent and still spans the vector space.

2.10 Theorem: Every spanning list in a vector space can be reduced

to a basis of the vector space.

Proof: Suppose (v1 , . . . , vn ) spans V . We want to remove some

of the vectors from (v1 , . . . , vn ) so that the remaining vectors form a
basis of V . We do this through the multistep process described below.
Start with B = (v1 , . . . , vn ).

Step 1
If v1 = 0, delete v1 from B. If v1 = 0, leave B unchanged.

Step j
If vj is in span(v1 , . . . , vj−1 ), delete vj from B. If vj is not in
span(v1 , . . . , vj−1 ), leave B unchanged.

Stop the process after step n, getting a list B. This list B spans V
because our original list spanned B and we have discarded only vectors
that were already in the span of the previous vectors. The process
 Bases 29

insures that no vector in B is in the span of the previous ones. Thus B

is linearly independent, by the linear dependence lemma (2.4). Hence
B is a basis of V .

Consider the list

 
(1, 2), (3, 6), (4, 7), (5, 9) ,

which spans F2 . To make sure that you understand the last proof, you
 
should verify that the process in the proof produces (1, 2), (4, 7) , a
basis of F2 , when applied to the list above.
Our next result, an easy corollary of the last theorem, tells us that
every finite-dimensional vector space has a basis.

2.11 Corollary: Every finite-dimensional vector space has a basis.

Proof: By definition, a finite-dimensional vector space has a span-

ning list. The previous theorem tells us that any spanning list can be
reduced to a basis.

We have crafted our definitions so that the finite-dimensional vector

space {0} is not a counterexample to the corollary above. In particular,
the empty list () is a basis of the vector space {0} because this list has
been defined to be linearly independent and to have span {0}.
Our next theorem is in some sense a dual of 2.10, which said that
every spanning list can be reduced to a basis. Now we show that given
any linearly independent list, we can adjoin some additional vectors so
that the extended list is still linearly independent but also spans the
space.

2.12 Theorem: Every linearly independent list of vectors in a finite- This theorem can be
dimensional vector space can be extended to a basis of the vector space. used to give another
proof of the previous
Proof: Suppose V is finite dimensional and (v1 , . . . , vm ) is linearly corollary. Specifically,
independent in V . We want to extend (v1 , . . . , vm ) to a basis of V . We suppose V is finite
do this through the multistep process described below. First we let dimensional. This
(w1 , . . . , wn ) be any list of vectors in V that spans V . theorem implies that
the empty list () can be
Step 1 extended to a basis
If w1 is in the span of (v1 , . . . , vm ), let B = (v1 , . . . , vm ). If w1 is of V . In particular, V
not in the span of (v1 , . . . , vm ), let B = (v1 , . . . , vm , w1 ). has a basis.
30 Chapter 2. Finite-Dimensional Vector Spaces

Step j
If wj is in the span of B, leave B unchanged. If wj is not in the
span of B, extend B by adjoining wj to it.

After each step, B is still linearly independent because otherwise the

linear dependence lemma (2.4) would give a contradiction (recall that
(v1 , . . . , vm ) is linearly independent and any wj that is adjoined to B is
not in the span of the previous vectors in B). After step n, the span of
B includes all the w’s. Thus the B obtained after step n spans V and
hence is a basis of V .

As a nice application of the theorem above, we now show that ev-

ery subspace of a finite-dimensional vector space can be paired with
another subspace to form a direct sum of the whole space.

Using the same basic 2.13 Proposition: Suppose V is finite dimensional and U is a sub-
ideas but considerably space of V . Then there is a subspace W of V such that V = U ⊕ W .
more advanced tools,
this proposition can be Proof: Because V is finite dimensional, so is U (see 2.7). Thus
proved without the there is a basis (u1 , . . . , um ) of U (see 2.11). Of course (u1 , . . . , um )
hypothesis that V is is a linearly independent list of vectors in V , and thus it can be ex-
finite dimensional. tended to a basis (u1 , . . . , um , w1 , . . . , wn ) of V (see 2.12). Let W =
span(w1 , . . . , wn ).
To prove that V = U ⊕ W , we need to show that

V =U +W and U ∩ W = {0};

see 1.9. To prove the first equation, suppose that v ∈ V . Then,

because the list (u1 , . . . , um , w1 , . . . , wn ) spans V , there exist scalars
a1 , . . . , am , b1 , . . . , bn ∈ F such that

v = a1 u1 + · · · + am um + b1 w1 + · · · + bn wn .
   
u w

In other words, we have v = u+w, where u ∈ U and w ∈ W are defined

as above. Thus v ∈ U + W , completing the proof that V = U + W .
To show that U ∩ W = {0}, suppose v ∈ U ∩ W . Then there exist
scalars a1 , . . . , am , b1 , . . . , bn ∈ F such that

v = a1 u1 + · · · + am um = b1 w1 + · · · + bn wn .

Thus
 Dimension 31

a1 u1 + · · · + am um − b1 w1 − · · · − bn wn = 0.

Because (u1 , . . . , um , w1 , . . . , wn ) is linearly independent, this implies

that a1 = · · · = am = b1 = · · · = bn = 0. Thus v = 0, completing the
proof that U ∩ W = {0}.

Dimension
Though we have been discussing finite-dimensional vector spaces,
we have not yet defined the dimension of such an object. How should
dimension be defined? A reasonable definition should force the dimen-
sion of Fn to equal n. Notice that the basis
 
(1, 0, . . . , 0), (0, 1, 0, . . . , 0), . . . , (0, . . . , 0, 1)

has length n. Thus we are tempted to define the dimension as the

length of a basis. However, a finite-dimensional vector space in general
has many different bases, and our attempted definition makes sense
only if all bases in a given vector space have the same length. Fortu-
nately that turns out to be the case, as we now show.

2.14 Theorem: Any two bases of a finite-dimensional vector space

have the same length.

Proof: Suppose V is finite dimensional. Let B1 and B2 be any two

bases of V . Then B1 is linearly independent in V and B2 spans V , so the
length of B1 is at most the length of B2 (by 2.6). Interchanging the roles
of B1 and B2 , we also see that the length of B2 is at most the length
of B1 . Thus the length of B1 must equal the length of B2 , as desired.

Now that we know that any two bases of a finite-dimensional vector

space have the same length, we can formally define the dimension of
such spaces. The dimension of a finite-dimensional vector space is
defined to be the length of any basis of the vector space. The dimension
of V (if V is finite dimensional) is denoted by dim V . As examples, note
that dim Fn = n and dim Pm (F) = m + 1.
Every subspace of a finite-dimensional vector space is finite dimen-
sional (by 2.7) and so has a dimension. The next result gives the ex-
pected inequality about the dimension of a subspace.
32 Chapter 2. Finite-Dimensional Vector Spaces

2.15 Proposition: If V is finite dimensional and U is a subspace

of V , then dim U ≤ dim V .

Proof: Suppose that V is finite dimensional and U is a subspace

of V . Any basis of U is a linearly independent list of vectors in V and
thus can be extended to a basis of V (by 2.12). Hence the length of a
basis of U is less than or equal to the length of a basis of V .

The real vector space To check that a list of vectors in V is a basis of V , we must, according
R has dimension 2;
2
to the definition, show that the list in question satisfies two properties:
the complex vector it must be linearly independent and it must span V . The next two
space C has results show that if the list in question has the right length, then we
dimension 1. As sets, need only check that it satisfies one of the required two properties.
R2 can be identified We begin by proving that every spanning list with the right length is a
with C (and addition is basis.
the same on both
spaces, as is scalar 2.16 Proposition: If V is finite dimensional, then every spanning
multiplication by real list of vectors in V with length dim V is a basis of V .
numbers). Thus when
we talk about the Proof: Suppose dim V = n and (v1 , . . . , vn ) spans V . The list
dimension of a vector (v1 , . . . , vn ) can be reduced to a basis of V (by 2.10). However, every
space, the role played basis of V has length n, so in this case the reduction must be the trivial
by the choice of F one, meaning that no elements are deleted from (v1 , . . . , vn ). In other
cannot be neglected. words, (v1 , . . . , vn ) is a basis of V , as desired.

Now we prove that linear independence alone is enough to ensure

that a list with the right length is a basis.

2.17 Proposition: If V is finite dimensional, then every linearly

independent list of vectors in V with length dim V is a basis of V .

Proof: Suppose dim V = n and (v1 , . . . , vn ) is linearly independent

in V . The list (v1 , . . . , vn ) can be extended to a basis of V (by 2.12). How-
ever, every basis of V has length n, so in this case the extension must be
the trivial one, meaning that no elements are adjoined to (v1 , . . . , vn ).
In other words, (v1 , . . . , vn ) is a basis of V , as desired.

As an example of how the last proposition can be applied, consider

 
the list (5, 7), (4, 3) . This list of two vectors in F2 is obviously linearly
independent (because neither vector is a scalar multiple of the other).
 Dimension 33

Because F2 has dimension 2, the last proposition implies that this lin-
early independent list of length 2 is a basis of F2 (we do not need to
bother checking that it spans F2 ).
The next theorem gives a formula for the dimension of the sum of
two subspaces of a finite-dimensional vector space.

2.18 Theorem: If U1 and U2 are subspaces of a finite-dimensional This formula for the
vector space, then dimension of the sum
of two subspaces is
dim(U1 + U2 ) = dim U1 + dim U2 − dim(U1 ∩ U2 ). analogous to a familiar
counting formula: the
Proof: Let (u1 , . . . , um ) be a basis of U1 ∩ U2 ; thus dim(U1 ∩ U2 ) = number of elements in
m. Because (u1 , . . . , um ) is a basis of U1 ∩ U2 , it is linearly independent the union of two finite
in U1 and hence can be extended to a basis (u1 , . . . , um , v1 , . . . , vj ) of U1 sets equals the number
(by 2.12). Thus dim U1 = m + j. Also extend (u1 , . . . , um ) to a basis of elements in the first
(u1 , . . . , um , w1 , . . . , wk ) of U2 ; thus dim U2 = m + k. set, plus the number of
We will show that (u1 , . . . , um , v1 , . . . , vj , w1 , . . . , wk ) is a basis of elements in the second
U1 + U2 . This will complete the proof because then we will have set, minus the number
of elements in the
dim(U1 + U2 ) = m + j + k intersection of the two
= (m + j) + (m + k) − m sets.

= dim U1 + dim U2 − dim(U1 ∩ U2 ).

Clearly span(u1 , . . . , um , v1 , . . . , vj , w1 , . . . , wk ) contains U1 and U2

and hence contains U1 + U2 . So to show that this list is a basis of
U1 + U2 we need only show that it is linearly independent. To prove
this, suppose

a1 u1 + · · · + am um + b1 v1 + · · · + bj vj + c1 w1 + · · · + ck wk = 0,

where all the a’s, b’s, and c’s are scalars. We need to prove that all the
a’s, b’s, and c’s equal 0. The equation above can be rewritten as

c1 w1 + · · · + ck wk = −a1 u1 − · · · − am um − b1 v1 − · · · − bj vj ,

which shows that c1 w1 + · · · + ck wk ∈ U1 . All the w’s are in U2 , so this

implies that c1 w1 + · · · + ck wk ∈ U1 ∩ U2 . Because (u1 , . . . , um ) is a
basis of U1 ∩ U2 , we can write

c1 w1 + · · · + ck wk = d1 u1 + · · · + dm um
34 Chapter 2. Finite-Dimensional Vector Spaces

for some choice of scalars d1 , . . . , dm . But (u1 , . . . , um , w1 , . . . , wk )

is linearly independent, so the last equation implies that all the c’s
(and d’s) equal 0. Thus our original equation involving the a’s, b’s, and
c’s becomes

a1 u1 + · · · + am um + b1 v1 + · · · + bj vj = 0.

This equation implies that all the a’s and b’s are 0 because the list
(u1 , . . . , um , v1 , . . . , vj ) is linearly independent. We now know that all
the a’s, b’s, and c’s equal 0, as desired.

The next proposition shows that dimension meshes well with direct
sums. This result will be useful in later chapters.

Recall that direct sum 2.19 Proposition: Suppose V is finite dimensional and U1 , . . . , Um
is analogous to disjoint are subspaces of V such that
union. Thus 2.19 is
analogous to the 2.20 V = U1 + · · · + U m
statement that if a
and
finite set B is written as
A1 ∪ · · · ∪ Am and the
2.21 dim V = dim U1 + · · · + dim Um .
sum of the number of
elements in the A’s Then V = U1 ⊕ · · · ⊕ Um .
equals the number of
elements in B, then the Proof: Choose a basis for each Uj . Put these bases together in
union is a disjoint one list, forming a list that spans V (by 2.20) and has length dim V
union. (by 2.21). Thus this list is a basis of V (by 2.16), and in particular it is
linearly independent.
Now suppose that u1 ∈ U1 , . . . , um ∈ Um are such that

0 = u 1 + · · · + um .

We can write each uj as a linear combination of the basis vectors (cho-

sen above) of Uj . Substituting these linear combinations into the ex-
pression above, we have written 0 as a linear combination of the basis
of V constructed above. Thus all the scalars used in this linear combina-
tion must be 0. Thus each uj = 0, which proves that V = U1 ⊕ · · · ⊕ Um
(by 1.8).
 Exercises 35

Exercises
1. Prove that if (v1 , . . . , vn ) spans V , then so does the list

(v1 − v2 , v2 − v3 , . . . , vn−1 − vn , vn )

obtained by subtracting from each vector (except the last one)

the following vector.

2. Prove that if (v1 , . . . , vn ) is linearly independent in V , then so is

the list
(v1 − v2 , v2 − v3 , . . . , vn−1 − vn , vn )
obtained by subtracting from each vector (except the last one)
the following vector.

3. Suppose (v1 , . . . , vn ) is linearly independent in V and w ∈ V .

Prove that if (v1 + w, . . . , vn + w) is linearly dependent, then
w ∈ span(v1 , . . . , vn ).

4. Suppose m is a positive integer. Is the set consisting of 0 and all

polynomials with coefficients in F and with degree equal to m a
subspace of P(F)?

5. Prove that F∞ is infinite dimensional.

6. Prove that the real vector space consisting of all continuous real-
valued functions on the interval [0, 1] is infinite dimensional.

7. Prove that V is infinite dimensional if and only if there is a se-

quence v1 , v2 , . . . of vectors in V such that (v1 , . . . , vn ) is linearly
independent for every positive integer n.

8. Let U be the subspace of R 5 defined by

U = {(x1 , x2 , x3 , x4 , x5 ) ∈ R5 : x1 = 3x2 and x3 = 7x4 }.

Find a basis of U.

9. Prove or disprove: there exists a basis (p0 , p1 , p2 , p3 ) of P3 (F)

such that none of the polynomials p0 , p1 , p2 , p3 has degree 2.

10. Suppose that V is finite dimensional, with dim V = n. Prove that

there exist one-dimensional subspaces U1 , . . . , Un of V such that

V = U1 ⊕ · · · ⊕ Un .
36 Chapter 2. Finite-Dimensional Vector Spaces

11. Suppose that V is finite dimensional and U is a subspace of V

such that dim U = dim V . Prove that U = V .

12. Suppose that p0 , p1 , . . . , pm are polynomials in Pm (F) such that

pj (2) = 0 for each j. Prove that (p0 , p1 , . . . , pm ) is not linearly
independent in Pm (F).

13. Suppose U and W are subspaces of R 8 such that dim U = 3,

dim W = 5, and U + W = R 8 . Prove that U ∩ W = {0}.

14. Suppose that U and W are both five-dimensional subspaces of R9 .

Prove that U ∩ W = {0}.

15. You might guess, by analogy with the formula for the number
of elements in the union of three subsets of a finite set, that
if U1 , U2 , U3 are subspaces of a finite-dimensional vector space,
then

dim(U1 + U2 + U3 )
= dim U1 + dim U2 + dim U3
− dim(U1 ∩ U2 ) − dim(U1 ∩ U3 ) − dim(U2 ∩ U3 )
+ dim(U1 ∩ U2 ∩ U3 ).

Prove this or give a counterexample.

16. Prove that if V is finite dimensional and U1 , . . . , Um are subspaces

of V , then

dim(U1 + · · · + Um ) ≤ dim U1 + · · · + dim Um .

17. Suppose V is finite dimensional. Prove that if U1 , . . . , Um are

subspaces of V such that V = U1 ⊕ · · · ⊕ Um , then

dim V = dim U1 + · · · + dim Um .

This exercise deepens the analogy between direct sums of sub-

spaces and disjoint unions of subsets. Specifically, compare this
exercise to the following obvious statement: if a finite set is writ-
ten as a disjoint union of subsets, then the number of elements in
the set equals the sum of the number of elements in the disjoint
subsets.