# n-Linear Algebra of Type II - Cover:Layout 1 1/12/2009 3:16 PM Page 1

n-LINEAR ALGEBRA OF TYPE II
W. B. Vasantha Kandasamy
e-mail: vasanthakandasamy@gmail.com
web: http://mat.iitm.ac.in/~wbv
www.vasantha.in
Florentin Smarandache
e-mail: smarand@unm.edu
INFOLEARNQUEST
Ann Arbor
2008
2
This book can be ordered in a paper bound reprint from:
Books on Demand
ProQuest Information & Learning
(University of Microfilm International)
P.O. Box 1346, Ann Arbor
MI 48106-1346, USA
Tel.: 1-800-521-0600 (Customer Service)
http://wwwlib.umi.com/bod/
Peer reviewers:
Professor Diego Lucio Rapoport
Departamento de Ciencias y Tecnologia
Roque Saen Peña 180, Bernal, Buenos Aires, Argentina
Dr.S.Osman, Menofia University, Shebin Elkom, Egypt
Prof. Mircea Eugen Selariu,
Polytech University of Timisoara, Romania.
Copyright 2008 by InfoLearnQuest and authors
Cover Design and Layout by Kama Kandasamy
Digital Library of Science:
http://www.gallup.unm.edu/~smarandache/eBooks-otherformats.htm
ISBN-10: 1-59973-031-6
ISBN-13: 978-1-59973-031-8
EAN: 9781599730318
Printed in the United States of America
3
CONTENTS
Preface 5
Chapter One
n-VECTOR SPACES OF TYPE II AND THEIR
PROPERTIES 7
1.1 n-fields 7
1.2 n-vector Spaces of Type II 10
Chapter Two
n-INNER PRODUCT SPACES OF TYPE II 161
Chapter Three
SUGGESTED PROBLEMS 195
4
INDEX 225
5
PREFACE
This book is a continuation of the book n-linear algebra of type
I and its applications. Most of the properties that could not be
derived or defined for n-linear algebra of type I is made possible
in this new structure: n-linear algebra of type II which is
introduced in this book. In case of n-linear algebra of type II, we
are in a position to define linear functionals which is one of the
marked difference between the n-vector spaces of type I and II.
However all the applications mentioned in n-linear algebras of
type I can be appropriately extended to n-linear algebras of type
II. Another use of n-linear algebra (n-vector spaces) of type II is
that when this structure is used in coding theory we can have
different types of codes built over different finite fields whereas
this is not possible in the case of n-vector spaces of type I.
Finally in the case of n-vector spaces of type II we can obtain n-
eigen values from distinct fields; hence, the n-characteristic
polynomials formed in them are in distinct different fields.
An attractive feature of this book is that the authors have
suggested 120 problems for the reader to pursue in order to
understand this new notion. This book has three chapters. In the
first chapter the notion of n-vector spaces of type II are
introduced. This chapter gives over 50 theorems. Chapter two
introduces the notion of n-inner product vector spaces of type II,
n-bilinear forms and n-linear functionals. The final chapter
6
suggests over a hundred problems. It is important that the reader
should be well versed with not only linear algebra but also n-
linear algebras of type I.
The authors deeply acknowledge the unflinching support of
Dr.K.Kandasamy, Meena and Kama.
W.B.VASANTHA KANDASAMY
FLORENTIN SMARANDACHE
7
Chapter One
n-VECTOR SPACES OF TYPE II AND
THEIR PROPERTIES

In this chapter we for the first time introduce the notion of n-
vector space of type II. These n-vector spaces of type II are
different from the n-vector spaces of type I because the n-vector
spaces of type I are defined over a field F where as the n-vector
spaces of type II are defined over n-fields. Some properties
enjoyed by n-vector spaces of type II cannot be enjoyed by n-
vector spaces of type I. To this; we for the sake of completeness
just recall the definition of n-fields in section one and n-vector
spaces of type II are defined in section two and some important
properties are enumerated.

1.1 n-Fields

In this section we define n-field and illustrate it by examples.

DEFINITION 1.1.1: Let F = F
1
· F
2
· … · F
n
where each F
i
is
a field such that F
i
. F
j
or F
j
. F
i
if i = j, 1 s i, j s n, we call F
a n-field.
8

We illustrate this by the following example.

Example 1.1.1: Let F = R · Z
3
· Z
5
· Z
17
be a 4-field.

Now how to define the characteristic of any n-field, n > 2.

DEFINITION 1.1.2: Let F = F
1
· F
2
· … · F
n
be a n-field, we
say F is a n-field of n-characteristic zero if each field F
i
is of
characteristic zero, 1 s i s n.

Example 1.1.2: Let F = F
1
· F
2
· F
3
· F
4
· F
5
· F
6
, where F
1

= Q( 2 ), F
2
= Q( 7 ), F
3
= Q( 3 , 5 ), F
4
= Q( 11 ), F
5
=
Q( 3 , 19 ) and F
6
= Q( 5 , 17 ); we see all the fields F
1
,
F
2
, …, F
6
are of characteristic zero thus F is a 6-field of
characteristic 0.

Now we proceed on to define an n-field of finite characteristic.

DEFINITION 1.1.3: Let F = F
1
· F
2
· … · F
n
(n > 2), be a n-
field. If each of the fields F
i
is of finite characteristic and not
zero characteristic for i = 1, 2, …, n then we call F to be a n-
field of finite characteristic.

Example 1.1.3: Let F = F
1
· F
2
· F
3
· F
4
= Z
5
· Z
7
· Z
17
·
Z
31
, F is a 4-field of finite characteristic.

Note: It may so happen that in a n-field F = F
1
· F
2
· … · F
n
,
n > 2 some fields F
i
are of characteristic zero and some of the
fields F
j
are of characteristic a prime or a power of a prime.
Then how to define such n-fields.

DEFINITION 1.1.4: Let F = F
1
· F
2
· … · F
n
be a n-field (n >
2), if some of the F
i
’s are fields of characteristic zero and some
of the F
j
’s are fields of finite characteristic i = j, 1 s i, j s n then
we define the characteristic of F to be a mixed characteristic.

9
Example 1.1.4: Let F = F
1
· F
2
· F
3
· F
4
· F
5
where F
1
= Z
2
,
F
2
= Z
7
, F
3
= Q( 7 ), F
4
= Q( 3, 5 ) and F
5
= Q( 3, 23, 2 )
then F is a 5-field of mixed characteristic; as F
1
is of
characteristic two, F
2
is a field of characteristic 7, F
3
, F
4
and F
5

are fields of characteristic zero.

Now we define the notion of n-subfields.

DEFINITION 1.1.5: Let F = F
1
· F
2
· … · F
n
be a n-field (n >
2). K = K
1
· K
2
· … · K
n
is said to be a n-subfield of F if each
K
i
is a proper subfield of F
i
, i = 1, 2, …, n and K
i
. K
j
or K
j
.
K
i
if i = j, 1 s i, j s n.

We now give an example of an n-subfield.

Example 1.1.5: Let F = F
1
· F
2
· F
3
· F
4
where
F
1
= Q( 2, 3 ), F
2
= Q( 7, 5 ),
F
3
=
2
2
Z [x]
x x 1
| |
|
|
+ +
\ .
and F
4
=
11
2
Z [x]
x x 1
| |
|
|
+ +
\ .

be a 4-field. Take K = K
1
· K
2
· K
3
· K
4
= (Q( 2 ) · Q( 7 )
· Z
2
· Z
11
_ F = F
1
· F
2
· F
3
· F
4
. Clearly K is a 4-subfield
of F.

It may so happen for some n-field F, we see it has no n-subfield
so we call such n-fields to be prime n-fields.

Example 1.1.6: Let F = F
1
· F
2
· F
3
· F
4
= Z
7
· Z
23
· Z
2
·
Z
17
be a 4-field. We see each of the field F
i
’s are prime, so F is a
n-prime field (n = 4).

DEFINITION 1.1.6: Let F = F
1
· F
2
· … · F
n
be a n-field (n >
2) if each of the F
i
’s is a prime field then we call F to be prime
n-field.

It may so happen that some of the fields may be prime and
others non primes in such cases we call F to be a semiprime n-
10
field. If all the fields F
i
in F are non prime i.e., are not prime
fields then we call F to be a non prime field. Now in case of n-
semiprime field or semiprime n-field if we have an m-subfield
m < n then we call it as a quasi m-subfield of F m < n.

We illustrate this situation by the following example.

Example 1.1.7: Let
F = Q · Z
7
·
| |
2
2
Z x
x x 1 + +
·
| |
7
2
Z x
x 1 +
·
| |
3
3
Z x
x 9 +
;

be a 5-field of mixed characteristic; clearly F is a 5-semiprime
field. For Q and Z
7
are prime fields and the other three fields are
non-prime; take K = K
1
· K
2
· K
3
· K
4
· K
5
= o · o · Z
2
·
Z
7
· Z
3
c F = F
1
· F
2
· F
3
· F
4
· F
5
. Clearly K is a quasi 3-
subfield of F.
1.2 n-Vector Spaces of Type II

In this section we proceed on to define n-vector spaces of type
II and give some basic properties about them.

DEFINITION 1.2.1: Let V = V
1
· V
2
· … · V
n
where each V
i
is
a distinct vector space defined over a distinct field F
i
for each i,
i = 1, 2, …, n; i.e., V = V
1
· V
2
· … · V
n
is defined over the n-
field F = F
1
· F
2
· … · F
n
. Then we call V to be a n-vector
space of type II.

We illustrate this by the following example.

Example 1.2.1: Let V = V
1
· V
2
· V
3
· V
4
be a 4-vector space
defined over the 4-field Q( 2 ) · Q( 3 ) · Q( 5 ) · Q( 7 ).
V is a 4-vector space of type II.

Unless mention is made specifically we would by an n-vector
space over F mean only n-vector space of type II, in this book.

11
Example 1.2.2: Let F = F
1
· F
2
· F
3
where F
1
= Q( 3 ) ×
Q( 3 ) × Q( 3 ) a vector space of dimension 3 over Q( 3 ). F
2

= Q( 7 ) [x] be the polynomial ring with coefficients from
Q( 7 ), F
2
is a vector space of infinite dimension over Q( 7 )
and
F
3
=
a b
a, b, c, d Q( 5)
c d
¦ ¹

¦ ¦
e
´ `

¦ ¦

¹ )
,
F
3
is a vector space of dimension 4 over Q( 5 ) . Thus F = F
1
·
F
2
· F
3
is a 3-vector space over the 3-field Q( 3 ) · Q( 7 ) ·
Q( 5 ) of type II.

Thus we have seen two examples of n-vector spaces of type II.
Now we will proceed on to define the linearly independent
elements of the n-vector space of type II. Any element o e V =
V
1
· V
2
· … · V
n
is a n-vector of the form (o
1
· o
2
· … ·
o
n
) or (o
1
· o
2
· … · o
n
) where o
i
or o
i
e V
i
, i = 1, 2, … , n
and each o
i
is itself a m row vector if dimension of V
i
is m.

We call S = S
1
· S
2
· … · S
n
where each S
i
is a proper subset
of V
i
for i = 1, 2, …, n as the n-set or n-subset of the n-vector
space V over the n-field F. Any element o = (o
1
· o
2
· … ·
o
n
) = {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } · … ·
{
n
n n n
1 2 n
, ,..., o o o } of V = V
1
· V
2
· … · V
n
where
{
i
i i i
1 2 n
, ,..., o o o } _ V
i
for i = 1, 2, …, n and
i
j
o eV
i
, j = 1, 2, …,
n
i
and 1 s i s n.

DEFINITION 1.2.2: Let V = V
1
· V
2
· … · V
n
be a n-vector
space of type II defined over the n-field F = F
1
· F
2
· … · F
n
,
(n > 2). Let S = {
1 1 1
1 2
, ,...,
i
k
o o o } · {
2
2 2 2
1 2
, ,...,
k
o o o } · … ·
{
1 2
, ,...,
n
n n n
k
o o o } be a proper n-subset of V = V
1
· V
2
· … · V
n

we say the n-set S is a n-linearly independent n-subset of V over
F = F
1
· F
2
· … · F
n
if and only if each subset {
1 2
, ,...,
i
i i i
k
o o o }
12
is a linearly independent subset of V
i
over F
i
, this must be true
for each i, i = 1, 2, … , n. If even one of the subsets of the n-
subset S; say {
1 2
, ,...,
j
j j j
k
o o o } is not a linearly independent
subset of V
j
over F
j
, then the subset S of V is not a n-linearly
independent subset of V, 1 s j s n. If in the n-subset S of V every
subset is not linearly independent subset we say S is a n-linearly
dependent subset of V over F = F
1
· F
2
· … · F
n
.

Now if the n-subset S = S
1
· S
2
· … · S
n
_ V; S
i
c V
i
is a
subset of V
i
, i = 1, 2, …, n is such that some of the subsets S
j
are
linearly independent subsets over F
j
and some of the subsets S
i

of V
i
are linearly dependent subsets of S
i
over F
i
, 1 s i, j s n
then we call S to be a semi n-linearly independent n-subset over
F or equivalently S is a semi n-linearly dependent n-subset of V
over F.

We illustrate all these situations by the following examples.

Example 1.2.3: Let V = V
1
· V
2
· V
3
= {Q( 3 ) × Q( 3 )} ·
{Q( 7 )[x]
5
| this contains only polynomials of degree less than
or equal to 5} ·
a b
a, b, c, d Q( 2)
c d
¦ ¹

¦ ¦
e
´ `

¦ ¦
¹ )

is a 3-vector space over the 3-field F = Q( 3 ) · Q( 7 ) ·
Q( 2 ) .
Take
S = {(1, 0), (5, 7)} · {x
3
, x, 1} ·
1 0 0 1
,
0 2 0 0
¦ ¹

¦ ¦
´ `

¦ ¦

¹ )

= S
1
· S
2
· S
3
_ V
1
· V
2
· V
3
;

a proper 3-subset of V. It is easily verified S is a 3-linearly
independent 3-subset of V over F. Take

13
T = {(1, 3) (0, 2) (5, 1)} · {x
3
, 1, x
2
+ 1, x
3
+ x
3
+ x
5
} ·

0 1 1 0
,
1 0 0 0
¦ ¹

¦ ¦
´ `

¦ ¦

¹ )

= T
1
· T
2
· T
3
_ V
1
· V
2
· V
3

is a proper 3-subset of V. Clearly T is only a semi 3-dependent
3-subset of V over F or equivalently 3-semi dependent 3-subset
of V, over F for T
1
and T
2
are linearly dependent subsets of V
1

and V
2
respectively over the fields Q( 3 ) and Q( 7 )
respectively and T
3
is a linearly independent subset of V
3
over
Q( 2 ) . Take

P = P
1
· P
2
· P
3

= {(1, 2) (2, 5), (5, 4), (–1, 0)} · {1, x
2
+ x, x, x
2
} ·
1 0 0 1 1 1 1 1
, , ,
0 1 1 0 1 1 1 0
¦ ¹

´ `

¹ )

c V
1
· V
2
· V
3

is a 3-subset of V. Clearly P is a 3-dependent 3-subset of V over
F.

Now we have seen the notion of n-independent n-subset; n-
dependent n-subset and semi n-dependent n-subset of V over F.
We would proceed to define the notion of n-basis of V.

DEFINITION 1.2.3: Let V = V
1
· V
2
· … · V
n
be a n-vector
space over the n-field F = F
1
· F
2
· … · F
n
. A n-subset B = B
1

· B
2
· … · B
n
is a n-basis of V over F if and only if each B
i
is
a basis of V
i
for every i = 1, 2, …, n. If each basis B
i
of V
i
is
finite for every i = 1, 2, …, n then we say V is a finite n-
dimensional n-vector space over the n-field F. If even one of the
basis B
i
of V
i
is infinite then we call V to be an infinite
dimensional n-vector space over the n-field F.

Now we shall illustrate by an example a finite n-dimensional n-
vector space over the n-field F = F
1
· F
2
· … · F
n
.
14

Example 1.2.4: Let V = V
1
· V
2
· … · V
4
be a 4-vector space
over the field F = Q( 2 ) · Z
7
· Z
5
· Q( 3 ) where V
1
=
Q( 2 ) × Q( 2 ), a vector space of dimension two over Q( 2 )
V
2
= Z
7
× Z
7
× Z
7
a vector space of dimension 3 over Z
7
,

V
3
=
5
a b c
a, b, c, d, e, f Z
d e f
¦ ¹

¦ ¦
e
´ `

¦ ¦
¹ )
a vector space of dimension 6 over Z
5
and
V
4
= Q( 3 )× Q( 3 ) × Q( 3 )× Q( 3 )
a vector space of dimension 4 over Q( 3 ). Clearly V is of (2,
3, 6, 4) dimension over F. Since the dimension of each V
i
is
finite we see V is a finite dimensional 4-vector space over F.

Now we give an example of an infinite dimensional n-vector
space over the n-field F.

Example 1.2.5: Let V = V
1
· V
2
· V
3
be a 3-vector space over
the 3-field F = Q( 3 ) · Z
3
· Q( 2 ) where V
1
= Q( 3 )[x] is
a vector space of infinite dimension over Q( 3 ), V
2
= Z
3
× Z
3
×
Z
3
is a vector space of dimension 3 over Z
3
and

V
3
=
( )
a b c
a, b, c, d, e, f Q 2
d e f
¦ ¹

¦ ¦
e
´ `

¦ ¦
¹ )

a vector space of dimension 6 over Q( 2 ). Clearly V is a 3-
vector space of infinite dimension over F.

We now proceed on to define the simple notion of n-linear
combination of n-vectors in V.

DEFINITION 1.2.4: Let V = V
1
· V
2
· … · V
n
be a n-vector
space over the n-field F = F
1
· F
2
· … · F
n
. Let ȕ = ȕ
1
· ȕ
2
·
… · ȕ
n
be a n-vector in V where ȕ
i
e V
i
for i = 1, 2, …, n. We
say ȕ is a n-linear combination of the n-vectors
15
Į =
{ }
1
1 1 1
1 2
, ,...,
n
o o o ·
{ }
2
2 2 2
1 2
, ,...,
n
o o o · … ·
{ }
1 2
, ,...,
n
n n n
n
o o o
in V, provided there exists n-scalars
C =
{ }
1
1 1 1
1 2
, ,...,
n
C C C ·
{ }
2
2 2 2
1 2
, ,...,
n
C C C · … ·
{ }
1 2
, ,...,
n
n n n
n
C C C
such that
ȕ = (ȕ
1
· ȕ
2
· … · ȕ
n
) =
{ }
1 1
1 1 1 1 1 1
1 1 2 2
... + + +
n n
C C C o o o ·
{ }
2 2
2 2 2 2 2 2
1 1 2 2
... + + +
n n
C C C o o o · …
·
{ }
1 1 2 2
... + + +
n n
n n n n n n
n n
C C C o o o .

We just recall this for it may be useful in case of representation
of elements of a n-vector Į in V, V a n-vector space over the n-
field F.

DEFINITION 1.2.5: Let S be a n-set of n-vectors in a n-vector
space V. The n-subspace spanned by the n-set S = S
1
· S
2
· …
· S
n
, where S
i
_V
i
, i = 1, 2, …, n is defined to be the
intersection W of all n-subspaces of V which contain the n-set S.
When the n-set S is finite n-set of n-vectors,
S =
{ }
1
1 1 1
1 2
, ,...,
n
o o o ·
{ }
2
2 2 2
1 2
, ,...,
n
o o o · … ·
{ }
1 2
, ,...,
n
n n n
n
o o o
we shall simply call W the n-subspace spanned by the n-vectors
S.

In view of this definition it is left as an exercise for the reader to
prove the following theorem.

THEOREM 1.2.1: The n-subspace spanned by a non empty n-
subset S = S
1
· S
2
· … · S
n
of a n-vector space V = V
1
· V
2
·
… · V
n
over the n-field F = F
1
· F
2
· … · F
n
is the set of all
linear combinations of n-vectors of S.

We define now the n-sum of the n-subsets of V.

DEFINITION 1.2.6: Let S
1
, S
2
, …, S
k
be n-subsets of a n-vector
space V = V
1
· V
2
· … · V
n
where S
i
=
1 2
... · · ·
i i i
n
S S S for i
= 1, 2, …, k. The set of all n-sums

16
{ }
1
1 1 1
1 2
... + + +
k
o o o ·
{ }
2
2 2 2
1 2
... + + +
k
o o o · …
·
{ }
1 2
... + + +
n
n n n
k
o o o
where
i
j i
S o e for each i = 1, 2, …, n and i  j  k
i
is denoted
by
( )
1
1 2
1 1 1
... + + +
k
S S S ·
( )
2
1 2
2 2 2
... + + +
k
S S S · … ·
( )
1 2
... + + +
n
k
n n n
S S S .

If W
1
, …, W
k
are n-subspaces of V = V
1
· V
2
· … · V
n
where
W
i
=
1 2
...
i i i
n
W W W · · · for i = 1, 2, …, n, then

W =
( )
1
1 2
1 1 1
...
k
W W W + + + ·
( )
2
1 2
2 2 2
... + + +
k
W W W · …
·
( )
1 2
... + + +
n
k
n n n
W W W
is a n-subspace of V which contains each of the subspaces W =
W
1
· W
2
· … · W
n
.

THEOREM 1.2.2: Let V = V
1
· V
2
· … · V
n
be an n-vector
space over the n-field F = F
1
· F
2
· … · F
n
. Let V be spanned
by a finite n-set of n-vectors
{ }
1
1 1 1
1 2
, , ...,
m
þ þ þ ·
{ }
2
2 2 2
1 2
, ,...,
m
þ þ þ
· … ·
{ }
1 2
, , ...,
n
n n n
m
þ þ þ . Then any independent set of n-vectors
in V is finite and contains no more than (m
1
, m
2
, …, m
n
)
elements.

Proof: To prove the theorem it is sufficient if we prove it for
one of the component spaces V
i
of V with associated subset S
i

which contains more than m
i
elements, this will be true for
every i; i = 1, 2, …, n. Let S = S
1
· S
2
· … · S
n
be a n-subset
of V where each S
i
contains more than m
i
vectors. Let S
i
contain
n
i
distinct vectors Į
1
, Į
2
, …,
i
n
o , n
i
> m
i
, since
{ }
i
i i i
1 2 m
, ,..., þ þ þ
spans V
i
, there exists scalars
i
jk
A in F
i
such that
i
m
i i i
k jk k
j 1
A
=
o = þ
¯
.
For any n
i
scalars
i
i i i
1 2 n
x , x ,..., x we have
i i
i i i i
1 1 n n
x ... x o + + o
17
=
i
n
i i
j j
j 1
x
=
o
¯

=
i i
n m
i i i
j kj k
j 1 k 1
x A
= =
þ
¯ ¯

=
i i
n m
i i i
kj j k
j 1 k 1
(A x )
= =
þ
¯¯

=
i i
m n
i i i
kj j k
k 1 j 1
(A x )
= =
þ
¯¯
.

Since n
i
> m
i
this imply that there exists scalars
i
i i i
1 2 n
x , x ,..., x not
all zero such that
i
m
i i
kj j
j 1
A x 0
=
=
¯
; 1  k  m
i
.
Hence
i i
i i i i i i
1 1 2 2 n n
x x ... x 0 o + o + + o = . This shows that S
i
is a
linearly dependent set. This is true for each i. Hence the result
holds good for S = S
1
· S
2
· … · S
n
.

The reader is expected to prove the following theorems.

THEOREM 1.2.3: If V is a finite dimensional n-vector space over
the n-field F, then any two n-basis of V have the same number of
n-elements.

THEOREM 1.2.4: Let V = V
1
· V
2
· … · V
n
be a n-vector
space over the n-field F = F
1
· F
2
· … · F
n
and if dim V =
(n
1
, n
2
, …, n
n
), then
1. any n-subset of V which contains more than n-vectors is
n-linearly dependent;
2. no n-subset of V which contains less than (n
1
, n
2
, …, n
n
)
vectors can span V.

THEOREM 1.2.5: Let S = S
1
· S
2
· … · S
n
be a n-linearly
independent n-subset of a n-vector space V. Suppose ȕ = (ȕ
1
·
ȕ
2
· … · ȕ
n
) is a vector in V which is not in the n-subspace
18
spanned by S. Then the n-subset obtained by adjoining ȕ to S is
n-linearly independent.

THEOREM 1.2.6: Let W = W
1
· W
2
· … · W
n
be a n-subspace
of a finite dimensional n-vector space V = V
1
· V
2
· … · V
n
,
every n-linearly independent n-subset of W is finite and is a part
of a (finite) n-basis for W.

The proof of the following corollaries is left as an exercise for

COROLLARY 1.2.1: If W = W
1
· W
2
· … · W
n
is a proper n-
subspace of a finite dimensional n-vector space V, then W is
finite dimensional and dim W < dim V; i.e. (m
1
, m
2
, …, m
n
) <
(n
1
, n
2
, …, n
n
) each m
i
< n
i
for i = 1, 2, …, n.

COROLLARY 1.2.2: In a finite dimensional n-vector space V =
V
1
· V
2
· … · V
n
every non-empty n-linearly independent set
of n-vectors is part of a n-basis.

COROLLARY 1.2.3: Let A = A
1
· A
2
· … · A
n
, be a n-vector
space where A
i
is a n
i
× n
i
matrix over the field F
i
and suppose
the n-row vectors of A form a n-linearly independent set of n-
vectors in
1 2
1 2
... · · ·
n
n n n
n
F F F . Then A is n-invertible i.e., each
A
i
is invertible, i = 1, 2, …, n.

THEOREM 1.2.7: Let W
1
=
1 1 1
1 2
...
n
W W W · · · and W
2
=
2 2 2
1 2
...
n
W W W · · · be finite dimensional n-subspaces of a n-
vector space V, then
W
1
+ W
2
=
1 2 1 2 1 2
1 1 2 2
...
n n
W W W W W W + · + · · +
is finite dimensional and
dim W
1
+ dim W
2
= dim(W
1
ŀ W
2
) + dim(W
1
+ W
2
)
where dim W
1
= (
1
1 1 1
1 2
, ,...,
n
m m m ) and dim W
2
=(
2
2 2 2
1 2
, ,...,
n
m m m )
dim W
1
+ dim W
2
= (
1 2
1 2 1 2 1 2
1 1 2 2
, ,..., + + +
n n
m m m m m m ).

Proof: By the above results we have
19
W
1
ŀ W
2
=
1 2 1 2 1 2
1 1 2 2 n n
W W W W ... W W · · · · · ·
has a finite n-basis.

{
1
1 1 1
1 2 k
, ,..., o o o } · {
2
2 2 2
1 2 k
, ,..., o o o ) · … · {
n
n n n
1 2 k
, ,..., o o o }
which is part of a n-basis.

{
1 1 1
1 1 1 1 1
1 2 k 1 n k
, ,..., , ,...,
÷
o o o þ þ } · {
2 2 2
2 2 2 2 2 2
1 2 k 1 2 n k
, ,..., , , ,...,
÷
o o o þ þ þ }
· {
n n n
n n n n n n
1 2 k 1 2 n k
, ,..., , , ,...,
÷
o o o þ þ þ }
is a n-basis of W
1
and

{
1 1 1
1 1 1 1 1 1
1 2 k 1 2 m k
, ,..., , , ,...,
÷
o o o ¸ ¸ ¸ } ·
{
2 2 2
2 2 2 2 2 2
1 2 k 1 2 m k
, ,..., , , ,...,
÷
o o o ¸ ¸ ¸ } · … ·
{
n n n
n n n n n n
1 2 k 1 2 m k
, ,..., , , ,...,
÷
o o o ¸ ¸ ¸ }

is a n-basis of W
2
. Then the subspace W
1
+ W
2
is spanned by
the n-vectors

{
1 1 1 1 1
1 1 1 1 n 1 2 1
1 k 1 2 n k 1 2 m k
, , , , , , , , , ,
÷ ÷
o o þ þ þ ¸ ¸ ¸ ! ! ! } ·
{
2 2 2 2 2
2 2 2 2 2 2 2 2 2
1 2 k 1 2 n k 1 2 m k
, , , , , , , , , , ,
÷ ÷
o o o þ þ þ ¸ ¸ ¸ ! ! ! } · … ·
{
n n n n n
n n n n n n n n n
1 2 k 1 2 n k 1 2 m k
, , , , , , , , , , ,
÷ ÷
o o o þ þ þ ¸ ¸ ¸ ! ! ! }

and these n-vectors form an n-independent n-set . For suppose

k
k k k k k
i i j j r
r
x y z 0 o + þ + ¸ =
¯ ¯ ¯

true for k = 1, 2, …, n.
k
k k k k k
r i i i i
r
z x y ÷ ¸ = o + þ
¯ ¯ ¯

which shows that
k
k
r
r
z ¸
¯
belongs
k
1
W , true for k = 1, 2, …, n.
k
k
r
r
z ¸
¯
belongs to
k
2
W for k = 1, 2, …, n it follows;

k
k
r
r
z ¸
¯
=
k
k
i
i
C o
¯

20

for certain scalars
k
k k k
1 2 n
C , C ,..., C true for k = 1, 2, …, n.
But the n-set {
i i
i i i i i i
1 2 k 1 2 n
, , , , , , , o o o ¸ ¸ ¸ ! ! }, i = 1, 2, …, n is n-
independent each of the scalars
k
r
z 0 = true for each k = 1, 2, …,
n. Thus
k k k k
i i j j
x y 0 o + þ =
¯ ¯
for k = 1, 2, …, n, and since
{
k k k
k k k k k k
1 2 k 1 2 n k
, , , , , , ,
÷
o o o þ þ þ ! ! } is also an independent set
each
k
i
x 0 = and each
k
j
y 0 = for k = 1, 2, …, n. Thus

{
1 1 1 1 1
1 1 1 1 1 1 1 1 1
1 2 k 1 2 n k 1 2 m k
, , , , , , , , , , ,
÷ ÷
o o o þ þ þ ¸ ¸ ¸ ! ! ! } ·
{
2 2 2 2 2
2 2 2 2 2 2 2 2 2
1 2 k 1 2 n k 1 2 m k
, , , , , , , , , , ,
÷ ÷
o o o þ þ þ ¸ ¸ ¸ ! ! ! } · … ·
{
n n n n n
n n n n n n n n n
1 2 k 1 2 n k 1 2 m k
, , , , , , , , , , ,
÷ ÷
o o o þ þ þ ¸ ¸ ¸ ! ! ! }
is a n-basis for
W
1
+ W
2
= {
1 2
1 1
W W + } · {
1 2
2 2
W W + } · … · {
1 2
n n
W W + }.

Finally
dim W
1
+ dim W
2

= {k
1
+ m
1
+ k
1
+ n
1
, k
2
+ m
2
+ k
2
+ n
2
, …, k
n
+ m
n
+ k
n
+ n
n
}
= {k
1
+ (m
1
+ k
1
+ n
1
), k
2
+ (m
2
+ k
2
+ n
2
), …, k
n
+ (m
n
+ k
n
+
n
n
)}
= dim(W
1
ŀW
2
) + dim (W
1
+ W
2
).

Suppose V = V
1
· V
2
· … · V
n
be a (n
1
, n
2
, …, n
n
) finite
dimensional n-vector space over F
1
· F
2
· … · F
n
. Suppose
B = {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } · …
· {
n
n n n
1 2 n
, ,..., o o o }
and
C = {
1
1 1 1
1 2 n
, , , þ þ þ ! } · {
2
2 2 2
1 2 n
, , , þ þ þ ! } · … · {
n
n n n
1 2 n
, ,..., þ þ þ }
be two ordered n-basis for V. There are unique n-scalars
k
ij
P
such that
i
n
k k k
j ij i
i 1
P
=
þ = o
¯
, k = 1, 2, …, n; 1  j  n
i
.
21
Let {
1
1 1 1
1 2 n
x , x , , x ! } · {
2
2 2 2
1 2 n
x , x , , x ! } · … · {
n
n n n
1 2 n
x , x , , x ! }
be the n-coordinates of a given n-vector Į in the ordered n-basis
C.
Į = {
1 1
1 1 1 1
1 1 n n
x x þ + + þ ! } · {
2 2
2 2 2 2
1 1 n n
x x þ + + þ ! } · …
· {
n n
n n n n
1 1 n n
x x þ + + þ ! }

=
1
n
1 1
j j
j 1
x
=
þ
¯
·
2
n
2 2
j j
j 1
x
=
þ
¯
· … ·
n
n
n n
j j
j 1
x
=
þ
¯

=
1 1
n n
1 1 i
j ij i
j 1 i 1
x P
= =
o
¯ ¯
·
2 2
n n
2 2 2
j ij i
j 1 i 1
x P
= =
o
¯ ¯
· … ·
n n
n n
n n n
j ij i
j 1 i 1
x P
= =
o
¯ ¯

=
1 1
n n
1 1 1
ij j i
j 1 i 1
(P x )
= =
o
¯¯
·
2 2
n n
2 2 2
ij j i
j 1 i 1
(P x )
= =
o
¯¯
· … ·
n n
n n
n n n
ij j i
j 1 i 1
(P x )
= =
o
¯¯
.

Thus we obtain the relation
Į =
1 1
n n
1 1 1
ij j i
j 1 i 1
(P x )
= =
o
¯¯
·
2 2
n n
2 2 2
ij j i
j 1 i 1
(P x )
= =
o
¯¯
· … ·
n n
n n
n n n
ij j i
j 1 i 1
(P x )
= =
o
¯¯
.

Since the n-coordinates (
1
1 1 1
1 2 n
y , y , , y ! ) · (
2
2 2 2
1 2 n
y , y , , y ! ) ·
… · (
n
n n n
1 2 n
y , y , , y ! ) of the n-basis B are uniquely determined
it follows
k
n
k k k
i ij j
j 1
y (P x )
=
=
¯
; 1 s i s n
k
. Let P
i
be the n
i
× n
i
matrix
whose i, j entry is the scalar
k
ij
P and let X and Y be the n-
coordinate matrices of the n-vector in the ordered n-basis B and
C. Thus we get
Y = Y
1
· Y
2
· … · Y
n

= P
1
X
1
· P
2
X
2
· … · P
n
X
n

= PX.

Since B and C are n-linearly independent n-sets, Y = 0 if and
only if X = 0. Thus we see P is n-invertible i.e.
P
-1

= (P
1
)
-1
· (P
2
)
-1
· … · (P
n
)
-1
i.e.
22
X = P
-1
Y
X
1
· X
2
· … · X
n
= (P
1
)
-1
Y
1
· (P
2
)
-1
Y
2
· … · (P
n
)
-1
Y
n
.

This can be put with some new notational convenience as
[Į]
B
= P[Į]
C

[Į]
C
= P
-1
[Į]
B.

In view of the above statements and results, we have proved the
following theorem.

THEOREM 1.2.8: Let V = V
1
· V
2
· … · V
n
be a finite (n
1
, n
2
,
…, n
n
) dimensional n-vector space over the n-field F and let B
and C be any two n-basis of V. Then there is a unique
necessarily invertible n-matrix P = P
1
· P
2
· … · P
n
of order
(n
1
× n
1
) · (n
2
× n
2
) · … · (n
n
× n
n
) with entries from F
1
· F
2

· … · F
n
i.e. entries of P
i
are from F
i
, i = 1, 2, …, n; such that

[Į]
B
= P[Į]
C

and [Į]
C
= P
-1
[Į]
B

for every n-vector Į in V. The n-columns of P are given by
i
j
P =
[
i
j
þ ]
B
; j = 1, 2, …, n and for each i; i = 1, 2, …, n.

Now we prove yet another interesting result.

THEOREM 1.2.9: Suppose P = P
1
· P
2
· … · P
n
is a (n
1
× n
1
)
· (n
2
× n
2
) · … · (n
n
× n
n
) n-invertible matrix over the n-field
F = F
1
· F
2
· … · F
n
, i.e. P
i
takes its entries from F
i
true for
each i = 1, 2, …, n. Let V = V
1
· V
2
· … · V
n
be a finite (n
1
,
n
2
, …, n
n
) dimensional n-vector space over the n-field F. Let B
be an n-basis of V. Then there is a unique n-basis C of V such
that

(i) [Į]
B
= P[Į]
C

and (ii) [Į]
C
= P
-1
[Į]
B

for every n-vector Į e V.

23
Proof: Let V = V
1
· V
2
· … · V
n
be a n-vector space of (n
1
,
n
2
, …, n
n
) dimension over the n-field F = F
1
· F
2
· … · F
n
.
Let B = {(
1
1 1 1
1 2 n
, ,..., o o o ) · (
2
2 2 2
1 2 n
, ,..., o o o ) · …
· (
n
n n n
1 2 n
, ,..., o o o )} be a set of n-vectors in V. If C =
{(
1
1 1 1
1 2 n
, ,..., þ þ þ ) · (
2
2 2 2
1 2 n
, ,..., þ þ þ ) · … · (
n
n n n
1 2 n
, ,..., þ þ þ )} is an
ordered n-basis of V for which (i) is valid, it is clear that for
each V
i
which has (
i
i i i
1 2 n
, , , þ þ þ ! ) as its basis we have
i
n
i i
j kj k
k 1
P
=
þ = o
¯
– I, true for every i, i = 1, 2, …,n.
Now we need only show that the vectors
i
j i
V þ e defined by
these equations I form a basis, true for each i, i = 1, 2, …, n. Let
Q = P
–1
. Then
r r
i i i i i
j j j kj k
j j k
Q Q P þ = o
¯ ¯ ¯

(true for every i = 1, 2, …, n)
=
i i i
kj jr k
j k
P Q o
¯¯

=
i i i
kj jr k
k j
P Q
| |
o
|
\ .
¯ ¯

=
i
k
o
true for each i = 1, 2, …, n. Thus the subspace spanned by the
set {
i
i i i
1 2 n
, ,..., þ þ þ } contain {
i
i i i
1 2 n
, ,..., o o o } and hence equals V
i
;
this is true for each i. Thus
B = {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } · …
· {
n
n n n
1 2 n
, ,..., o o o }
is contained in V = V
1
· V
2
· … · V
n
.

Thus
C = {
1
1 1 1
1 2 n
, ,..., þ þ þ } · {
2
2 2 2
1 2 n
, ,..., þ þ þ } · … · {
n
n n n
1 2 n
, ,..., þ þ þ }
is a n-basis and from its definition and by the above theorem (i)
is valid hence also (ii).

24
Now we proceed onto define the notion of n-linear
transformation for n-vector spaces of type II.

DEFINITION 1.2.7: Let V = V
1
· V
2
· … · V
n
be a n-vector
space of type II over the n-field F = F
1
· F
2
· … · F
n
and W
= W
1
· W
2
· … · W
n
be a n-space over the same n-field F =
F
1
· F
2
· … · F
n
. A n-linear transformation of type II is a n-
function T = T
1
· T
2
· … · T
n
from V into W such that
( )
i i i i i i
i i i
T C C T T o þ o þ + = + for all Į
i
, ȕ
i
in V
i
and for all
scalars C

i
e F
i.
This is true for each i; i = 1, 2, …, n. Thus

T[CĮ + ȕ] = (T
1
· T
2
· … · T
n
)
[C
1
Į
1
+ ȕ
1
] · [C
2
Į
2
+ ȕ
2
] · … · [C
n
Į
n
+ ȕ
n
]
= T
1
(C
1
Į
1
+ ȕ
1
)· T
2
(C
2
Į
2
+ ȕ
2
) · … ·
T
n
(C
n
Į
n
+ ȕ
n
)

for all Į
i
e V
i
or Į
1
· Į
2
· … · Į
n
e V and C
i
e F
i
or C
1
·

C
2
· … · C
n
e F
1
· F
2
· …· F
n
. We say I = I
1
· I
2
· … · I
n

is the n-identity transformation of type II from V to V if

IĮ = I (Į
1
· Į
2
· … · Į
n
)
= I
1

1
) · I
2

2
) · … · I
n

n
)
= Į
1
· Į
2
· … · Į
n

for all Į
1
· Į
2
· … · Į
n
e V = V
1
· V
2
· … · V
n
. Similarly
the n-zero transformation 0 = 0 · 0 · … · 0 of type II from V
into V is given by 0Į = 0Į
1
· 0Į
2
· … · 0Į
n
= 0 · 0 · … · 0, for all Į
1
· Į
2
· … · Į
n
e V.

Now we sketch the proof of the following interesting and
important theorem.

THEOREM 1.2.10: Let V = V
1
· V
2
· … · V
n
be a (n
1
, n
2
, …,
n
n
) dimensional finite n-vector space over the n-field F = F
1

· F
2
· … · F
n
of type II. Let {
1
1 1 1
1 2
, ,...,
n
o o o } · {
2
2 2 2
1 2
, ,...,
n
o o o }
· … · {
1 2
, ,...,
n
n n n
n
o o o } be an n-ordered basis for V. Let W =
25
W
1
· W
2
· … · W
n
be a n-vector space over the same n-field F
= F
1
· F
2
· … · F
n
.
Let {
1
1 1 1
1 2
, ,...,
n
þ þ þ } · {
2
2 2 2
1 2
, ,...,
n
þ þ þ } ·…· {
1 2
, ,...,
n
n n n
n
þ þ þ } be
any n-vector in W. Then there is precisely a n-linear
transformation T = T
1
· T
2
· … · T
n
from V into W such that
i i
i j j
To þ = for i = 1, 2, …, n and j = 1, 2, …, n
i
.

Proof: To prove there is some n-linear transformation T = T
1
·
T
2
· … · T
n
with
i i
i j j
To = þ for each j = 1, 2, …, n
i
and for
each i = 1, 2, …, n. For every Į = Į
1
· Į
2
· … · Į
n
in V we
have for every Į
i
e V
i
a unique
i
i i n
1 2 n
x , x ,..., x such that Į =
i i
i i i i
1 2 n n
x ... x o + + o . This is true for every i; i = 1, 2, …, n. For
this vector Į
i
we define T
i
Į
i
=
i i
i i i i
1 1 n n
x ... x þ + + þ true for i = 1, 2,
…, n. Thus T
i
is well defined for associating with each vector Į
i

in V
i
a vector T
i
Į
i
in W
i
this is well defined rule for T = T
1
· T
2

· … · T
n
as it is well defined rule for each T
i
: V
i
÷ W
i
, i = 1,
2, …, n.
From the definition it is clear that
i i
i j j
To = þ for each j. To
see that T is n-linear. Let ȕ
i
=
i i
i i i i
1 2 n n
y ... y o + + o be in V and let
C
i
be any scalar from F
i
. Now
i i i i i i i
1 1 1
C (C x y ) o +þ = + þ + … +
i i i
i i i i
n n n
(C x y ) + þ
true for every i, i = 1, 2, …, n. On the other hand
i i i
i
T (C ) o +þ =
i i
n n
i i i i i
j j j j
j 1 j 1
C x y
= =
þ + þ
¯ ¯

true for i = 1, 2, …, n i.e. true for every linear transformation T
i

in T.
i i i
i
T (C ) o +þ = C
i
T
i

i
) + T
i

i
) true for every i. Thus
T(CĮ + ȕ) = T
1
(C
1
Į
1
+ ȕ
1
) · T
2
(C
2
Į
2
+ ȕ
2
) · … · T
n
(C
n
Į
n
+
ȕ
n
). If U = U
1
· U
2
· … · U
n
is a n-linear transformation from
V into W with
i i
i j j
U o = þ for j = 1, 2, …, n
i
and i = 1, 2, …, n
then for the n-vector Į = Į
1
· Į
2
· … · Į
n
we have for every Į
i

in Į
26
Į
i
=
i
n
i i
j j
j 1
x
=
o
¯

we have
U
i
Į
i
=
i
n
i i
i j j
j 1
U x
=
o
¯

=
i
n
i i
j i j
j 1
x (U )
=
o
¯

=
i
n
i i
j j
j 1
x
=
þ
¯

so that U is exactly the rule T which we define. This proves TĮ
= ȕ, i.e. if Į = Į
1
· Į
2
· … · Į
n
and ȕ = ȕ
1
· ȕ
2
· … · ȕ
n
then
i i
i j j
To = þ , i  j  n
i
and i = 1, 2, …, n.

Now we proceed on to give a more explicit description of the n-
linear transformation. For this we first recall if V
i
is any n
i

dimensional vector space over F
i
then V
i
~
( )
i
n
i
F =
i
n
i
F . Further
if W
i
is also a vector space of dimension m
i
over F
i
, the same
field then W
i
~
( )
i
m
i
F .
Now let V = V
1
· V
2
· … · V
n
be a n-vector space over
the n-field F = F
1
· F
2
· … · F
n
of (n
1
, n
2
, …, n
n
) dimension
over F, i.e., each V
i
is a vector space over F
i
of dimension n
i

over F
i
for i = 1, 2, …, n. Thus V
i
~ F
i
. Hence V = V
1
· V
2
·
… · V
n
~
1 2 n
n n n
1 2 n
F F ... F · · · . Similarly if W = W
1
· W
2
·
… · W
n
is a n-vector space over the same n-field F and (m
1
,
m
2
, …, m
n
) is the dimension of W over F then W
1
· W
2
· …
· W
n
~
1 2 n
m m m
1 2 n
F F ... F · · · .
T = T
1
· T
2
· … · T
n
is uniquely determined by a
sequence of n-linear n-vectors
(
1
1 1 1
1 2 n
, ,..., þ þ þ ) · (
2
2 2 2
1 2 n
, ,..., þ þ þ ) · … · (
n
n n n
1 2 n
, ,..., þ þ þ )
where
i
j i j
TE þ = , j = 1, 2, …, n
i
and i = 1, 2, …, n. In short T is
uniquely determined by the n-images of the standard n-basis
vectors. The determinate is

27
Į = {(
1
1 1 1
1 2 n
x , x , , x ! ) · (
2
2 2 2
1 2 n
x , x , , x ! ) · …
· (
n
n n n
1 2 n
x , x , , x ! )}.

TĮ = (
1 1
1 1 1 1
1 1 n n
x x þ + + þ ! ) · (
2 2
2 2 2 2
1 1 n n
x x þ + + þ ! ) · …
· (
n n
n n n n
1 1 n n
x x þ + + þ ! ).

If B = B
1
· B
2
· … · B
n
is a (n
1
× n
1
, n
2
× n
2
, …, n
n
× n
n
), n-
matrix which has n-row vector
(
1
1 1 1
1 2 n
, , , þ þ þ ! ) · (
2
2 2 2
1 2 n
, , , þ þ þ ! ) · … · (
n
n n n
1 2 n
, , , þ þ þ ! )
then TĮ = ĮB. In other words if
i
i i i i
k k1 k2 kn
(B , B , , B ) þ = ! ; i = 1,
2, …, n, then
T{(
1
1 1 1
1 2 n
x , x , , x ! ) · … · (
n
n n n
1 2 n
x , x , , x ! )}
= T
1
(
1
1 1 1
1 2 n
x , x , , x ! ) · T
2
(
2
2 2 2
1 2 n
x , x , , x ! ) · … ·
T
n
(
n
n n n
1 2 n
x , x , , x ! ) =

1
1
1 1 1
1 1
11 1n
1 1 1
1 2 n
1 1
m 1 m n
B B
[x , x ,..., x ]
B B

!
# #
!
·

2
2
2 2 2
2 2
11 1n
2 2 2
1 2 n
2 2
m 1 m n
B B
[x , x , , x ]
B B

!
! # #
!
· … ·

n
n
n n n
n n
11 1n
n n n
1 2 n
n n
m 1 m n
B B
[x , x , , x ]
B B

!
! # #
!
.

This is the very explicit description of the n-linear
transformation. If T is a n-linear transformation from V into W
then n-range of T is not only a n-subset of W it a n-subspace of
28
W = W
1
· W
2
· … · W
n
. Let R
T
=
1 2 n
1 2 n
T T T
R R ... R · · · be
the n-range of T = T
1
· T
2
· … · T
n
that is the set of all n-
vectors ȕ = (ȕ
1
· ȕ
2
· … · ȕ
n
) in W = W
1
· W
2
· … · W
n

such that Ǻ = TĮ i.e.,
i i
j i j
T þ = o for each i = 1, 2, …, n and for
some o in V. Let
i
i i i
1 2 T
, R þ þ e and C
i
e F
i
. There are vectors
i i
1 2 i
, V o o e such that
i i
i 1 1
To = þ and
i i
i 2 2
To = þ . Since T
i
is linear
for each i;
i i i
i 1 2
T (C ) o + o =
i i i
i 1 i 2
C T T o + o =
i i i
1 2
Cþ +þ which
shows that
i i i
1 2
Cþ +þ is also in
i
i
T
R . Since this is true for every i
we have
(
1 1 1
1 2
C þ +þ ) · … · (
n n n
1 2
C þ +þ ) e R
T
=
1 2 n
1 2 n
T T T
R R ... R · · · .
Another interesting n-subspace associated with the n-linear
transformation T is the n-set N = N
1
· N
2
· … · N
n
consisting
of the n-vectors Į e V such that TĮ = 0. It is a n-subspace of V
because

(1) T(0) = 0 so N = N
1
· N
2
· … · N
n
is non empty. If TĮ =
Tȕ = 0 then Į, þ e V = V
1
· V
2
· … · V
n
. i.e., Į = Į
1
· Į
2

· … · Į
n
and ȕ = ȕ
1
· ȕ
2
· … · ȕ
n
then
T(CĮ + ȕ) = CTĮ + Tȕ
= C.0 + 0
= 0
= 0 · 0 · … · 0
so CĮ + ȕ e N = N
1
· N
2
· … · N
n.

DEFINITION 1.2.8: Let V and W be two n-vector spaces over the
same n-field F = F
1
· F
2
· … · F
n
of dimension (n
1
, n
2
, …, n
n
)
and (m
1
, m
2
, …, m
n
) respectively. Let T: V ÷ W be a n-linear
transformation. The n-null space of T is the set of all n-vectors Į
= Į
1
· Į
2
· … · Į
n
in V such that T
i
Į
i
= 0, i = 1, 2, …, n.

If V is finite dimensional, the n-rank of T is the dimension of
the n-range of T and the n-nullity of T is the dimension of the n-
null space of T.

The following interesting result on the relation between n-rank
T and n-nullity T is as follows:
29

THEOREM 1.2.11: Let V and W be n-vector spaces over the
same n-field F = F
1
· F
2
· … · F
n
and suppose V is finite say
(n
1
, n
2
, …, n
n
) dimensional. T is a n-linear transformation from
V into W. Then n-rank T + n-nullity T = n-dim V = (n
1
, n
2
, …,
n
n
).

Proof: Given V = V
1
· V
2
· … · V
n
to be a (n
1
, n
2
, …, n
n
)
dimensional n-vector space over the n-field F = F
1
· F
2
· …
· F
n
. W = W
1
· W
2
· … · W
n
is a n-vector space over the
same n-field F. Let T be a n-linear transformation from V into
W given by T = T
1
· T
2
· … · T
n
where T
i
: V
i
÷ W
i
is a
linear transformation, and V
i
is of dimension n
i
and both the
vector spaces V
i
and W
i
are defined over F
i
; i = 1, 2, …, n.

n-rank T = rank T
1
· rank T
2
· … · rank T
n
; n-nullity T =
nullity T
1
· nullity T
2
· … · nullity T
n
. So n-rank T + n-
nullity T = dim V = (n
1
, n
2
, …, n
n
) i.e. rank T
i
+ nullity T
i
= dim
V
i
= n
i
.

We shall prove the result for one T
i
and it is true for every i, i =
1, 2, …, n. Suppose {
i
i i i
1 2 k
, , , o o o ! } is a basis of N
i
; the null
space of T
i
. There are vectors
i i i
i i i
k 1 k 2 n
, , ,
+ +
o o o ! in V
i
such that
{
i
i i i
1 2 n
, , , o o o ! } is a basis of V
i
, true for every i, i = 1, 2, …, n.
We shall now prove that {
i i
i i
i k 1 i n
T ,..., T
+
o o } is a basis for the
range of T
i
. The vector
i
i 1
To ,
i
i 2
To , …,
i
i
i n
To certainly span the
range of T
i
and since
i
i j
T 0 o = for j s k
i
we see that
{
i i
i i
i k 1 i n
T ,..., T
+
o o } span the range. To see that these vectors are
independent, suppose we have scalar
i
r
C such that
i
i
n
i i
r i r
r k 1
C T 0
= +
o =
¯
.
This says that
i
i
n
i i
i r r
j k 1
T C
= +
| |
o
|
|
\ .
¯
= 0
30

and accordingly the vector Į
i
=
i
i
n
i i
r r
j k 1
C
= +
o
¯
is in the null space of
T
i
. Since
i
i i i
1 2 k
, ,..., o o o form a basis for N
i
, there must be scalar
i
i i i
1 2 k
, ,..., þ þ þ in F
i
such that
i
k
i i i
n r
r 1 =
o = þ o
¯
. Thus
i i
i
k n
i i i i
n r j j
r 1 j k 1
C
= = +
þ o ÷ o
¯ ¯
= 0
and since
i
i i i
1 2 n
, ,..., o o o are linearly independent we must have
i i i
i i i i i
1 2 k k 1 n
C C 0
+
þ = þ = = þ = = = = ! ! . If r
i
is the rank of T
i
, the
fact that
i i
i k 1 i n
T , , T
+
o o ! form a basis for the range of T
i
tells us
that r
i
= n
i
– k
i
. Since k
i
is the nullity of T
i
and n
i
is the
dimension of V
i
we have rank T
i
+ nullity T
i
= dim V
i
= n
i
. We
see this above equality is true for every i, i = 1, 2, …, n. We
have rank T
i
+ nullity T
i
= dimV
i
= n
i
; we see this above
equality is true for every i, i = 1, 2, …, n. We have (rank T
1
+
nullity T
1
) · (rank T
2
+ nullity T
2
) · … · (rank T
n
+ nullity
T
n
) = dim V = (n
1
, n
2
, …, n
n
).

Now we proceed on to introduce and study the algebra of n-
linear transformations.

THEOREM 1.2.12: Let V and W be a n-vector spaces over the
same n-field, F = F
1
· F
2
· … · F
n
. Let T and U be n-linear
transformation from V into W. The function (T + U) is defined
by (T + U)Į = TĮ + UĮ is a n-linear transformation from V into
W. If C is a n-scalar from F = F
1
· F
2
· … · F
n
, the n-function
CT is defined by (CT)Į = C(TĮ) is a n-linear transformation
from V into W. The set of all n-linear transformations form V
into W together with addition and scalar multiplication defined
above is a n-vector space over the same n-field F = F
1
· F
2
·
… · F
n
.

Proof: Given V = V
1
· V
2
· … · V
n
and W = W
1
· W
2
· …
· W
n
are two n-vector spaces over the same n-field F = F
1
· F
2

31
· … · F
n
. Let U = U
1
· U
2
· … · U
n
and T = T
1
· T
2
· …
· T
n
be two n-linear transformations from V into W, to show
the function (T + U) defined by (T + U)(Į) =TĮ + UĮ is a n-
linear transformation from V into W.
For Į = Į
1
· Į
2
· … · Į
n
e V = V
1
· V
2
· … · V
n
, C =
C
1
· C
2
· … · C
n
; consider

(T + U)(CĮ + ȕ)
= T(CĮ + ȕ) + U(CĮ + ȕ)
= CT(Į) + Tȕ + CU(Į) + U(ȕ)
= CT(Į) + CU(Į) + Tȕ + Uȕ
= C(T + U)Į + (T + U)ȕ

which shows T + U is a n-linear transformation of V into W
given by
T + U = (T
1
· T
2
· … · T
n
) + (U
1
· U
2
· … · U
n
)
= (T
1
+ U
1
) · (T
2
+ U
2
) · … · (T
n
+ U
n
)

since each (T
i
+ U
i
) is a linear transformation from V
i
into W
i

true for each i, i = 1, 2, …, m; hence we see T + U is a n-linear
transformation from V into W.
Similarly CT is also a n-linear transformation. For

CT = (C
1
· C
2
· … · C
n
) (T
1
· T
2
· … · T
n
)
= C
1
T
1
· C
2
T
2
· … · C
n
T
n
.

Now for (dĮ + ȕ) where d = d
1
· d
2
· … · d
n
and d e F
1
· F
2

· … · F
n
i.e., d
i
e F
i
for i = 1, 2, …, n and Į = Į
1
· Į
2
· … ·
Į
n
and ȕ = ȕ
1
· ȕ
2
· … · ȕ
n
consider

T(dĮ + ȕ) = (T
1
· T
2
· … · T
n
) {[d
1
Į
1
· d
2
Į
2
· … · d
n
Į
n
]
+ (ȕ
1
· ȕ
2
· … · ȕ
n
)}
= T
1
(d
1
Į
1
+ ȕ
1
) · T
2
(d
2
Į
2
+ ȕ
2
) · … · T
n
(d
n
Į
n
+
ȕ
n
)
now
CT = (C
1
· C
2
· … · C
n
)(T
1
· T
2
· … · T
n
)
= C
1
T
1
· C
2
T
2
· … · C
n
T
n

32
(we know from properties of linear transformation each C
i
T
i
is a
linear transformation for i = 1, 2, …, n) so CT = C
1
T
1
· C
2
T
2
·
… · C
n
T
n
is a n-linear transformation as

CT(dĮ + ȕ) = C
1
T
1
(d
1
Į
1
+ ȕ
1
) · C
2
T
2
(d
2
Į
2
+ ȕ
2
) · … ·
C
n
T
n
(d
n
Į
n
+ ȕ
n
).

Hence the claim. CT is a n-linear transformation from V into W.
Now we prove about the properties enjoyed by the
collection of all n-linear transformation of V into W. Let L
n
(V,
W) denote the collection of all linear transformations of V into
W, to prove L
n
(V, W) is a n-vector space over the n-field; F =
F
1
· F
2
· … · F
n
, where V and W are n-vector spaces defined
over the same n-field F. Just now we have proved L
n
(V, W) is
closed under sum i.e. addition and also L
n
(V, W) is closed
under the n-scalar from the n-field i.e. we have proved by
defining (T + U)(Į) = TĮ + UĮ for all Į = Į
1
· Į
2
· … · Į
n
e
V, T + U is again a n-linear transformation from V into W. i.e.
L
n
(V, W) is closed under addition. We have also proved for
every n-scalar C = C
1
· C
2
· … · C
n
e F = F
1
· F
2
· … · F
n

and T = T
1
· T
2
· … · T
n
; CT is also a n-linear transformation
of V into W; i.e. for every T, U e L
n
(V,W), T + U e L
n
(V,W)
and for every C e F = F
1
· F
2
· … · F
n
and for every T e
L
n
(V, W), CT e L
n
(V, W). Trivially 0Į = 0 for every Į e V will
serve as the n-zero transformation of V into W.
Thus L
n
(V, W) is a n-vector space over the same n-field F
= F
1
· F
2
· … · F
n
.
Now we study the dimension of L
n
(V, W), when both the
n-vector spaces are of finite dimension.

THEOREM 1.2.13: Let V = V
1
· V
2
· … · V
n
be a finite (n
1
, n
2
,
…, n
n
)-dimensional n-vector space over the n-field F = F
1
· F
2

· … · F
n
. Let W = W
1
· W
2
· … · W
n
be a finite (m
1
, m
2
, …,
m
n
)-dimensional n-vector space over the same n-field F. Then
the space L
n
(V, W) is of finite dimension and has (m
1
n
1
, m
2
n
2
,
…, m
n
n
n
)-dimension over the same n-field F.

33
Proof: Given V = V
1
· V
2
· … · V
n
is a n-vector space over
the n-field F = F
1
· F
2
· … · F
n
of (n
1
, n
2
, …, n
n
)-dimension
over F. Also W = W
1
· W
2
· … · W
n
is a n-vector space over
the n-field F of (m
1
, m
2
, …, m
n
) dimension over F. Let T = T
1
·
T
2
· … · T
n
be any n-linear transformation of type II i.e. T
i
: V
i

÷ W
i
, this is the only way the n-linear transformation of type II
is defined because V
i
is defined over F
i
and W
i
is also defined
over F
i
. Clearly W
j
(i  j) is defined over F
j
and F
i
 F
j
so one
cannot imagine of defining any n-linear transformation. Let B =
{
1
1 1 1
1 2 n
, , , o o o ! } · {
2
2 2 2
1 2 n
, , , o o o ! } ·… · {
n
n n n
1 2 n
, , , o o o ! }
be a n-ordered basis for V over the n-field F. We say n-ordered
basis if each basis of V
i
is ordered and B
1
= {
1
1 1 1
1 2 m
, , , þ þ þ ! } ·
{
2
2 2 2
1 2 m
, , , þ þ þ ! } · … · {
n
n n n
1 2 m
, , , þ þ þ ! } be a n-ordered basis
of W over the n-field F.
For every pair of integers (p
i
, q
i
); 1 s p
i
s m
i
and 1 s q
i
s m
i
,
i = 1, 2, …, n, we define a n-linear transformation E
p,q
=
1 1 2 2 n n
p ,q p ,q p ,q
1 2 n
E E E · · · ! from V into W by

i i
i
i
p ,q i
i j i i
p
0 if j q
E ( )
if j q
¦ =
¦
o =
´
þ =
¦
¹
=
i i
i
jq p
o þ ,

true for i = 1, 2, …, n
i
, i = 1, 2, …, m
i
. We have by earlier
theorems a unique n-linear transformation from V
i
into W
i

satisfying these conditions. The claim is that the m
i
n
i

transformations
i i
p ,q
i
E form a basis of L (V
i
, W
i
). This is true for
each i. So L
n
(V, W) = L (V
1
, W
1
) · L (V
2
, W
2
) · … · L (V
n
,
W
n
) is a n-vector space over the n-field of (m
1
n
1
, m
2
n
2
, …,
m
n
n
n
) dimension over F. Now suppose T = T
1
· T
2
· … · T
n
is
a n-linear transformation from V into W.
Now for each j, 1 s j s n
i
, i = 1, 2, …, n let
i
i i
kj m j
A , , A ! be
the co ordinates of the vector
i
i j
To in the ordered basis
(
i
i i i
1 2 m
, , , þ þ þ ! ) of B
i
, i = 1, 2, …, n. i.e.
34
i
i i
i
m
i i i
i j
p j p
p 1
T A
=
o = þ
¯
.
We wish to show that T =
i i
i i
i i
i i
m n
i p ,q
i
p q
p 1q 1
A E
= =
¯¯
(I).
Let U be a n-linear transformation in the right hand member of
(I) then for each j
i i
i i
i i
i i p ,q i
i j i j
p q
p q
U A E ( ) o = o
¯¯

=
i i i i
i i
i i
p q jq p
p q
A o þ
¯¯

=
i
i i
j
i
m
i i
p p
p 1
A
=
þ
¯

=T
i
i
j
o

and consequently U
i
= T
i
as this is true for each i, i = 1, 2, …, n.
We see T = U. Now I shows that the E
p,q
spans L
n
(V, W) as
each
i i
p ,q
i
E spans L(V
i
, W
i
); i = 1, 2, …, n. Now it remains to
show that they are n-linearly independent. But this is clear from
what we did above for if in the n-transformation U, we have

i i
i i
i i
i p ,q
i i
p q
p q
U A E =
¯¯

to be the zero transformation then
i
i j
U 0 o = for each j, 1 s j s n
i

so
i
i i
j
i
m
i i
p p
p 1
A 0
=
þ =
¯
and the independence of the
i
i
p
þ implies that
i
j
i
p
A = 0 for every p
i
and j. This is true for every i = 1, 2, …, n.

Next we proceed on to prove yet another interesting result about
n-linear transformations, on the n-vector spaces of type II.

THEOREM 1.2.14: Let V, W and Z be n-vector spaces over the
same n-field F = F
1
· F
2
· … · F
n
. Let T be a n-linear
transformation from V into W and U a n-linear transformation
35
from W into Z. Then the n-composed n-function UT defined by
UT(Į) = U(T(Į)) is a n-linear transformation from V into Z.

Proof: Given V = V
1
· V
2
· … · V
n
, W = W
1
· W
2
· … ·
W
n
and Z = Z
1
· Z
2
· … · Z
n
are n-vector spaces over the n-
field F = F
1
· F
2
· … · F
n
where V
i
, W
i
and Z
i
are vector
spaces defined over the same field F
i
, this is true for i = 1, 2, …,
n. Let T = T
1
· T
2
· … · T
n
be a n-linear transformation from
V into W i.e. each T
i
maps V
i
into W
i
for i = 1, 2, …, n. In no
other way there can be a n-linear transformation of type II from
V into W. U = U
1
· U
2
· … · U
n
is a n-linear transformation
of type II from W into Z such that for every U
i
is a linear
transformation from W
i
into Z
i
; (for W
i
and Z
i
alone are vector
spaces defined over the same field F
i
); this is true for each i, i =
1, 2, …, n. Suppose the n-composed function UT defined by
(UT)(Į) = U(T(Į)) where Į = Į
1
· Į
2
· … · Į
n
e V i.e.

UT = U
1
T
1
· U
2
T
2
· … · U
n
T
n
and
UT(Į) = (U
1
T
1
· U
2
T
2
· … · U
n
T
n
)(Į
1
· Į
2
· … ·Į
n
)
= (U
1
T
1
)(Į
1
) · (U
2
T
2
)(Į
2
) · … · ( U
n
T
n
)(Į
n
).

Now from results on linear transformations we know each U
i
T
i

defined by (U
i
T
i
)(Į
i
) = U
i
(T
i

i
) is a linear transformation from
V
i
into Z
i
; This is true for each i, i = 1, 2, …, n. Hence UT is a
n-linear transformation of type II from V into Z.

We now define a n-linear operator of type II for n-vector spaces
V over the n-field F.

DEFINITION 1.2.9: Let V = V
1
· V
2
· … · V
n
be a n-vector
space defined over the n-field F = F
1
· F
2
· … · F
n
. A n-
linear operator T = T
1
· T
2
· … · T
n
on V of type II is a n-
linear transformation from V into V i.e. T
i
: V
i
÷ V
i
for i = 1, 2,
…, n.

We prove the following lemma.

36
LEMMA 1.2.1: Let V = V
1
· V
2
· … · V
n
be a n-vector space
over the n-field F = F
1
· F
2
· … · F
n
of type II. Let U, T
1
and
T
2
be the n-linear operators on V and let C = C
1
· C
2
· … ·
C
n
be an element of F = F
1
· F
2
· … · F
n
. Then

a. IU = UI = U where I = I
1
· I
2
· … · I
n
is the n-
identity linear operator on V i.e. I
i
(v
i
) = v
i
for every v
i

e V
i
; I
i
; V
i
÷ V
i
, i = 1, 2, …, n.
b. U(T

1
+ T

2
) = UT

1
+ UT

2

(T

1
+ T

2
)U = T

1
U + T
2
U
c. C(UT
1
) = (CU)T

1
= U(CT
1
).

Proof: (a) Given V = V
1
· V
2
· … · V
n
, a n-vector space
defined over the n-field of type II for F = F
1
· F
2
· … · F
n
and
each V
i
of V is defined over the field F
i
of F. If I = I
1
· I
2
· …
· I
n
: V ÷ V such that I
i

i
) = ȣ
i
for every ȣ
i
e V
i
; true for i =
1, 2, …, n then I(Į) = Į for every Į e V
1
· V
2
· … · V
n
in V
and UI = (U
1
· U
2
· … · U
n
) (I
1
· … · I
n
) = U
1
I
1
· U
2
I
2
·
… · U
n
I
n
. Since each U
i
I
i
= I
i
U
i
for i = 1, 2, …, n, we have IU
= I
1
U
1
· I
2
U
2
· … · I
n
U
n
. Hence (a) is proved.

[U(T
1
+ T
2
)](Į) = (U
1
· U
2
· … · U
n
) [(
1 1 1
1 2 n
T T ... T · · · ) +
(
2 2 2
1 2 n
T T ... T · · · )] [Į
1
· Į
2
· … · Į
n
]
= (U
1
· U
2
· … · U
n
)[(
1 2
1 1
T T + ) · (
1 2
2 2
T T + ) · … ·
(
1 2
n n
T T + )] (Į
1
· Į
2
· … · Į
n
)
= U
1
(
1 2
1 1
T T + )Į
1
·U
2
(
1 2
2 2
T T + )(Į
2
) ·… ·U
n
(
1 2
n n
T T + )(Į
n
).

Since for each i we have
U
i
(
1 2
i i
T T + )(Į
i
) =
1 i 2 i
i i i i
U T U T o + o
for every Į
i
e V
i
; i = 1, 2, …, n.
Thus U
i
(
1 2
i i
T T + ) =
1
i i
U T +
2
i i
U T , for every i, i = 1, 2, …, n
hence U(T
1
+ T
2
) = UT
1
+ UT
2
, similarly (T
1
+ T
2
)U = T
1
U +
T
2
U. Proof of (c) is left as an exercise for the reader. Suppose
L
n
(V, V) denote the set of all n-linear operators from V to V of
type II. Then by the above lemma we see L
n
(V, V) is a n-vector
space over the n-field of type II.
37
Clearly L
n
(V, V) = L
n
(V
1
, V
1
) · L
n
(V
2
, V
2
) · … · L
n
(V
n
, V
n
)
we see each L
n
(V
i
, V
i
) is a vector space over the field F
i
, true for
i = 1, 2, …, n. Hence L
n
(V, V) is a n-vector space over the n-
field F = F
1
· F
2
· … · F
n
of type II.
Since we have composition of any two n-linear operators to
be contained in L
n
(V, V) and L
n
(V, V) contains the n-identity
we see L
n
(V, V) is a n-linear algebra over the n-field F of type
II. To this end we just recall the definition of n-linear algebra
over the n-field of type II.

DEFINITION 1.2.10: Let V = V
1
· V
2
· … · V
n
be a n-vector
space over the n-field F = F
1
· F
2
· … · F
n
of type II, where
V
i
is a vector space over the field F
i
. If each of the V
i
is a linear
algebra over F
i
for every i, i = 1, 2, …, n then we call V to be a
n-linear algebra over the n-field F of type II.

We illustrate this by the following example.

Example 1.2.6: Given V = V
1
· V
2
· V
3
· V
4
where
V
1
=
2
a b
a, b, c, d Z
c d
¦ ¹

¦ ¦
e
´ `

¦ ¦
¹ )
,
a linear algebra over Z
2
; V
2
= {All polynomials with
coefficients from Z
7
} a linear algebra over Z
7
. V
3
= Q( 3 )×
Q( 3 ) a linear algebra over Q( 3 ) and V
4
= Z
5
× Z
5
× Z
5
a
linear algebra over Z
5
. Thus V = V
1
· V
2
· V
3
· V
4
is a 4-
linear algebra over the 4-field F = Z
2
· Z
7
· Q( 3 ) · Z
5
of
type II.
All n-vector spaces over a n-field F need not in general be a
n-linear algebra over the n-field of type II.
We illustrate this situation by an example.

Example 1.2.7: Let V = V
1
· V
2
· V
3
be a 3-vector space over
the 3-field F = Z
2
· Q · Z
7
. Here

V
1
=
2
a b c
a, b, c, d, e, f Z
d e f
¦ ¹

¦ ¦
e
´ `

¦ ¦
¹ )

38
a vector space over the field Z
2
. Clearly V
1
is only a vector
space over Z
2
and never a linear algebra over Z
2
. V
2
= {All
polynomials in x with coefficients from Q}; V
2
is a linear
algebra over Q.
V
3
=
7
a b
c d a, b, c, d, e, f Z
e f
¦ ¹

¦ ¦

e
´ `

¦ ¦

¹ )
;

V
3
is a vector space over Z
7
and not a linear algebra. Thus V is
only a 3-vector space over the 3-field F and not a 3-linear
algebra over F.
We as in case of finite n-vector spaces V and W over n-field
F of type II, one can associate with every n-linear
transformation T of type II a n-matrix. We in case of n-vector
space V over the n-field F for every n-operator on V associate a
n-matrix which will always be a n-square mixed matrix. This is
obvious by taking W = V, then instead of getting a n-matrix of
n-order (m
1
n
2
, m
2
n
2
, …, m
n
n
n
) we will have m
i
= n
i
for every i, i
= 1, 2, …, n. So the corresponding n-matrix would be a mixed
n-square matrix of n-order (
2 2 2
1 2 n
n , n ,..., n ) provided n-dim(V) =
(n
1
, n
2
, …, n
n
). But in case of n-linear transformation of type II
we would not be in a position to talk about invertible n-linear
transformation. But in case of n-linear operators we can define
invertible n-linear operators of a n-vector space over the n-field
F.

DEFINITION 1.2.11: Let V = V
1
· V
2
· … · V
n
and W = W
1
·
W
2
· … · W
n
be two n-vector spaces defined over the same n-
field F = F
1
· F
2
· …· F
n
of type II. A n-linear transformation
T = T
1
· T
2
· … · T
n
from V into W is n-invertible if and only
if

1. T is one to one i.e. each T
i
is one to one from V
i
into W
i

such that T
i
Į
i
= T
i
ȕ
i
implies Į
i
= ȕ
i
true for each i, i = 1,
2, …, n.
2. T is onto that is range of T is (all of) W, i.e. each T
i
: V
i

÷ W
i
is onto and range of T
i
is all of W
i
, true for every
i, i = 1, 2, …, n.
39

The following theorem is immediate.

THEOREM 1.2.15: Let V = V
1
· V
2
· … · V
n
and W = W
1
·
W
2
· … · W
n
be two n-vector spaces over the n-field F = F
1
·
F
2
· … · F
n
of type II. Let T = T
1
· T
2
· … · T
n
be a n-linear
transformation of V into W of type II. If T is n-invertible then the
n-inverse n-function
1 1 1 1
1 2
...
n
T T T T
÷ ÷ ÷ ÷
= · · · is a n-linear
transformation from W onto V.

Proof: We know if T = T
1
· T
2
· … · T
n
is a n-linear
transformation from V into W and when T is one to one and
onto then there is a uniquely determined n-inverse function
1 1 1 1
1 2 n
T T T ... T
÷ ÷ ÷ ÷
= · · · which map W onto V such that T
-1
T
is the n-identity n-function on V and TT
-1
is the n-identity n-
function on W. We need to prove only if a n-linear n-function T
is n-invertible then the n-inverse T
-1
is also n-linear. Let ȕ
1
=
1 2 n
1 1 1
... þ ·þ · ·þ , ȕ
2
=
1 2 n
2 2 2
... þ ·þ · ·þ be two n-vectors in W
and let C = C
1
· C
2
· … · C
n
be a n-scalar of the n-field F =
F
1
· F
2
· … · F
n
. Consider

T
-1
(Cȕ
1
+ ȕ
2
) = (
1 1 1 1
1 2 n
T T T ... T
÷ ÷ ÷ ÷
= · · · )
[
1 1 1 2 2 2 n n n
1 2 1 2 1 2
C C ... C þ +þ · þ +þ · · þ +þ ]
=
1 1 1 1 1 2 2 2
1 1 1 2 2 1 1 2
T (C ) T (C ) ...
÷ ÷
þ +þ · þ +þ · ·
1 n n n
n 1 1 2
T (C )
÷
þ +þ .

Since from usual linear transformation properties we see
each T
i
has
1
i
T
÷
and
1 i i i
i 1 1 2
T (C )
÷
þ +þ =
i 1 i 1 i
1 i 1 i 2
C (T T )
÷ ÷
þ + þ true for
each i, i = 1, 2, …, n. Let
i
j
o =
1
i
T
÷ i
j
þ , j = 1, 2, and i = 1, 2, …, n
that is let
i
j
o be the unique vector in V
i
such that
i i
i j j
To = þ ;
since T
i
is linear
i i i
i 1 2
T (C ) o + o =
i i i
i 1 i 2
C T T o + o
=
i i i
1 2
Cþ +þ .
40
Thus
i i i
1 2
C o + o is the unique vector in V
i
which is sent by T
i

into T
i
(
i i i
1 2
Cþ +þ ) and so
1 i i i
i 1 2
T (C )
÷
þ +þ =
i i i
1 2
C o + o =
i 1 i 1 i
i 1 i 2
C T T
÷ ÷
þ + þ
and
1
i
T
÷
is linear. This is true for every T
i
i.e., for i, i = 1, 2, …,
n. So
1 1 1 1
1 2 n
T T T ... T
÷ ÷ ÷ ÷
= · · · is n-linear. Hence the claim.
We say a n-linear transformation T is n-non-singular if T¸ =
0 implies ¸ = (0 · 0 · … · 0); i.e., each T
i
in T is non singular
i.e. T
i
¸
i
= 0 implies ¸
i
= 0 for every i; i.e., if ¸ = ¸
1
· ¸
2
· … ·
¸
n
then T¸ = 0 · … · 0 implies ¸ = 0 · 0 · … · 0.
Thus T is one to one if and only if T is n-non singular.
The following theorem is proved for the n-linear
transformation from V into W of type II.

THEOREM 1.2.16: Let T = T
1
· T
2
· … · T
n
be a n-linear
transformation from V = V
1
· V
2
· … · V
n
into W = W
1
· W
2

· … · W
n
. Then T is n-non singular if and only if T carries
each n-linearly independent n-subset of V into a n-linearly
independent n-subset of W.

Proof: First suppose we assume T = T
1
· T
2
· … · T
n
is n-non
singular i.e. each T
i
: V
i
÷ W
i
is non singular for i = 1, 2, …, n.
Let S = S
1
· S
2
· … · S
n
be a n-linearly, n-independent n-
subset of V. If {
1
1 1 1
1 2 k
, ,..., o o o } · {
2
2 2 2
1 2 k
, ,..., o o o } · … ·
{
n
n n n
1 2 k
, ,..., o o o } are n-vectors in S = S
1
· S
2
· … · S
n
then the
n-vectors {
1
1 1 1
1 1 1 2 1 k
T , T ,..., T o o o } · {
2
2 2 2
2 1 2 2 2 k
T , T ,..., T o o o } · …
· {
n
n n n
n 1 n 2 n k
T , T ,..., T o o o } are n-linearly n-independent; for if
i i
1 i 1
C (T ) o + … +
i
i i
k i k
C (T ) o = 0 for each i, i = 1, 2, …, n. then
T
i
(
i i
i i i i
1 1 k k
C ... C o + + o ) = 0 and since each T
i
is singular we have
i i
i i i i
1 1 k k
C ... C o + + o = 0 this is true for each i, i = 1, 2, …, n. Thus
each
i
t
C = 0; t = 1, 2, …, k
i
and i = 1, 2, …, n; because S
i
is an
independent subset of the n-set S = S
1
· S
2
· … · S
n
. This
shows image of S
i
under T
i
is independent. Thus T = T
1
· T
2
·
… · T
n
is n- independent as each T
i
is independent.
41
Suppose T = T
1
· T
2
· … · T
n
is such that it carriers
independent n-subsets onto independent n-subsets. Let Į = Į
1
· Į
2
· … · Į
n
be a non zero n-vector in V i.e. each o
i
is a non
zero vector of V
i
, i = 1, 2, …, n. Then the set S = S
1
· S
2
· …
· S
n
consisting of the one vector Į = Į
1
· Į
2
· … · Į
n
is n-
independent each Į
i
is independent for i = 1, 2, …, n. The n-
image of S is the n-set consisting of the one n-vector TĮ = T
1
Į
1
· T
2
Į
2
· … · T
n
Į
n
and this is n-independent, therefore TĮ  0
because the n-set consisting of the zero n-vector alone is
dependent. This shows that the n-nullspace of T = T
1
· T
2
· …
· T
n
is the zero subspace as each T
i
Į
i
 0 implies each of the
zero vector alone in V
i
is dependent for i = 1, 2, …, n. Thus
each T
i
is non singular so T is n-non singular.

We prove yet another nice theorem.

THEOREM 1.2.17: Let V = V
1
· V
2
· … · V
n
and W = W
1
· W
2
· … · W
n
be n-vector spaces over the same n-field F = F
1
· F
2
· … · F
n
of type II. If T is a n-linear transformation of type II
from V into W, the following are equivalent.

(i) T is n-invertible
(ii) T is n-non-singular
(iii) T is onto that is the n-range of T is W.

Proof: Given V = V
1
· V
2
· … · V
n
is a n-vector space over
the n-field F of (n
1
, n
2
, … , n
n
) dimension over the n-field F; i.e.
dimension of V
i
over the field F
i
is n
i
for i = 1, 2, … , n. Let us
further assume n-dim(V) = (n
1
, n
2
, … , n
n
) = n-dim W. We
know for a n-linear transformation T = T
1
· T
2
· … · T
n
from
V into W, n-rank T + n-nullity T = (n
1
, n
2
, …, n
n
) i.e., (rank T
1
+
nullity T
1
) · (rank T
2
+ nullity T
2
) · … · (rank T
n
+ nullity
T
n
) = (n
1
, n
2
, …, n
n
). Now T = T
1
· T
2
· … · T
n
is n-non
singular if and only if n-nullity T = (0 · 0 · … · 0) and since
dim W = (n
1
, n
2
, …, n
n
) the n-range of T is W if and only if n-
rank T = (n
1
, n
2
, …, n
n
). Since n-rank T + n-nullity T = (n
1
, n
2
,
… , n
n
), the n-nullity T is (0 · 0 · … · 0) precisely when the
n-rank is (n
1
, n
2
, …, n
n
). Therefore T is n-nonsingular if and
42
only if T(V) = W i.e. T
1
(V
1
) · T
2
(V
2
) · … · T
n
(V
n
) = W
1
·
W
2
· … · W
n
. So if either condition (ii) or (iii) hold good, the
other is satisfied as well and T is n-invertible. We further see if
the 3 conditions of the theorem are also equivalent to (iv) and
(v).
(iv). If {
1
1 1 1
1 2 n
, , , o o o ! } · {
2
2 2 2
1 2 n
, , , o o o ! } · … ·
{
n
n n n
1 2 n
, , , o o o ! } is a n-basis of V then {
1
1 1
1 1 1 n
T , , T o o ! } ·
{
2
2 2
2 1 2 n
T , , T o o ! } · … · {
n
n n
n 1 n n
T , , T o o ! } is a n-basis for
W.
(v). There is some n-basis {
1
1 1 1
1 2 n
, , , o o o ! }· {
2
2 2 2
1 2 n
, , , o o o ! }·
… · {
n
n n n
1 2 n
, ,..., o o o } for V such that {
1
1 1 1
1 1 1 2 1 n
T , T , , T o o o ! }
· {
2
2 2 2
2 1 2 2 2 n
T , T , , T o o o ! } · … · {
n
n n n
n 1 n 2 n n
T , T , , T o o o ! }
is an n-basis for W.
The proof of equivalence of these conditions (i) to (v)
mentioned above is left as an exercise for the reader.

THEOREM 1.2.18: Every (n
1
, n
2
, …, n
n
) dimensional n-vector
space V over the n-field F = F
1
· … · F
n
is isomorphic to
1
1
...
n
n n
n
F F · · .

Proof: Let V = V
1
· V
2
· … · V
n
be a (n
1
, n
2
, …, n
n
)
dimensional space over the n-field F = F
1
· F
2
· … · F
n
of
type II. Let B = {
1
1 1 1
1 2 n
, , , o o o ! } · {
2
2 2 2
1 2 n
, , , o o o ! } · … ·
{
n
n n n
1 2 n
, , , o o o ! } be an n-ordered n-basis of V. We define an n-
function T = T
1
· T
2
·…· T
n
from V into
1 2 n
n n n
1 2 n
F F ... F · · ·
as follows: If Į = Į
1
· Į
2
· … · Į
n
is in V, let TĮ be the (n
1
, n
2
,
…, n
n
) tuple;
(
1
1 1 1
1 2 n
x , x , , x ! ) · (
2
2 2 2
1 2 n
x , x , , x ! ) · … · (
n
n n n
1 2 n
x , x , , x ! )
of the n-coordinate of Į = Į
1
· Į
2
· … · Į
n
relative to the n-
ordered n-basis B, i.e. the (n
1
, n
2
, …, n
n
) tuple such that
Į = (
1 1
1 1 1 1
1 1 n n
x x o + + o ! ) · (
2 2
2 2 2 2
1 1 n n
x x o + + o ! ) · … ·
(
n n
n n n n
1 1 n n
x x o + + o ! ).
43
Clearly T is a n-linear transformation of type II, one to one and
maps V onto
1 2 n
n n n
1 2 n
F F ... F · · · or each T
i
is linear and one to
one and maps V
i
to
i
n
i
F

for every i, i = 1, 2, …, n. Thus we can
as in case of vector spaces transformation by matrices give a
representation of n-transformations by n-matrices.

Let V = V
1
· V
2
· … · V
n
be a n-vector space over the n-
field F = F
1
· F
2
· … · F
n
of (n
1
, n
2
, …, n
n
) dimension over F.
Let W = W
1
· … · W
n
be a n-vector space over the same n-
field F of (m
1
, m
2
, …, m
n
) dimension over F. Let B =
{
1
1 1 1
1 2 n
, , , o o o ! } · {
2
2 2 2
1 2 n
, , , o o o ! } · … · {
n
n n n
1 2 n
, , , o o o ! }
be a n-basis for V and C= (
1
1 1 1
1 2 m
, , , þ þ þ ! ) · (
2
2 2 2
1 2 m
, , , þ þ þ ! ) ·
… · (
n
n n n
1 2 m
, , , þ þ þ ! ) be an n-ordered basis for W. If T is any
n-linear transformation of type II from V into W then T is
determined by its action on the vectors Į
i
. Each of the (n
1
, n
2
,
…, n
n
) tuple vector;
i
i j
To is uniquely expressible as a linear
combination;
i
i i
i
m
i i i
i j k j k
k 1
T A
=
o = þ
¯
. This is true for every i, i = 1, 2,
…, n and 1  j  n
i
of
i
i
k
þ ; the scalars
i i
ij mj
A ,..., A being the
coordinates of T
i
i
j
o in the ordered basis {
i
i i i
1 2 m
, ,..., þ þ þ } of C.
This is true for each i, i = 1, 2, …, n.

Accordingly the n-transformation T = T
1
· T
2
· … · T
n
is
determined by the (m
1
n
1
, m
2
n
2
, …, m
n
n
n
) scalars
i
i
k j
A . The m
i
×
n
i
matrix A
i
defined by
i
i
(k j)
A =
i
i
k j
A is called the component
matrix T
i
of T relative to the component basis {
i
i i i
1 2 n
, ,..., o o o }
and {
i
i i i
1 2 m
, ,..., þ þ þ } of B and C respectively. Since this is true
for every i, we have A =
1 2 n
1 2 n
(k , j) (k , j) (k , j)
A A ... A · · · = A
1
· A
2
· … · A
n
(for simplicity of notation) the n-matrix associated
with T = T
1
· T
2
· … · T
n
. Each A
i
determines the linear
transformation T
i
for i = 1, 2, …, n. If Į
i
=
i i
i i i i
1 1 n n
x ... x o + + o is a
vector in V
i
then
44
T
i
Į
i
= (T
i
i
n
i i
j j
j 1
x
=
o
¯
)
=
i
n
i i
j i j
j 1
x (T )
=
o
¯

=
i i
i i
n m
i i i
j k j k
j 1 k 1
x A
= =
þ
¯ ¯

=
i i
i i
m n
i i i
k j j k
k 1 j 1
(A x )
= =
þ
¯¯
.

If X
i
is the coordinate matrix of Į
i
in the component n-basis of B
then the above computation shows that A
i
X
i
is the coordinate
matrix of the vector
i
i
To ; that is the component of the n-basis C
because the scalar
i
i i
n
i i
k j k
j 1
A x
=
¯
is the entry in the k
th
row of the
column matrix A
i
X
i
. This is true for every i, i = 1, 2, … , n. Let
us also observe that if A
i
is any m
i
× n
i
matrix over the field F
i

then
i i i
i i
n m n
i i i i i
i j j k j j k
j 1 k 1 j 1
T ( x ) ( A x )
= = =
o = þ
¯ ¯ ¯

defines a linear transformation T
i
from V
i
into W
i
, the matrix of
which is A
i
relative to
i
i i
1 n
{ ,..., } o o and
i
i i i
1 2 m
{ , ,..., } þ þ þ , this is
true of every i. Hence T = T
1
· T
2
· … · T
n
is a linear n-
transformation from V = V
1
· V
2
· … · V
n
into W = W
1
·
W
2
· … · W
n
, the n-matrix of which is A = A
1
· A
2
… · A
n

relative to the n-basis B =
1
1 1 1
1 2 n
{ , ,..., } o o o ·
2
2 2 2
1 2 n
{ , ,..., } o o o ·
… ·
n
n n n
1 2 n
{ , ,..., } o o o and C =
1
1 1 1
1 2 m
{ , ,..., } þ þ þ ·
2
2 2 2
1 2 m
{ , ,..., } þ þ þ
· … · {
n
n n n
1 2 m
, ,..., þ þ þ }.

In view of this we have the following interesting theorem.

THEOREM 1.2.19: Let V = V
1
· … · V
n
be a (n
1
, n
2
, … , n
n
)-
dimension n-vector space over the n-field F = F
1
· … · F
n
. Let
45
B =
1
1 1 1
1 2
{ , ,..., }
n
o o o ·
2
2 2 2
1 2
{ , ,..., }
n
o o o · … ·
1 2
{ , ,..., }
n
n n n
n
o o o
= B
1
· B
2
· … · B
n
be the n-basis of V over F and
C =
1
1 1 1
1 2
{ , ,..., }
m
þ þ þ ·
2
2 2 2
1 2
{ , ,..., }
m
þ þ þ · … ·
1 2
{ , ,..., }
n
n n n
m
þ þ þ
= C
1
· C
2
· … · C
n
,

n-basis for W, where W is a n-vector space over the same n-
field F of (m
1
, m
2
, …, m
n
) dimension over F. For each n-linear
transformation T = T
1
· T
2
· … · T
n
of type II from V into W
there is a n-matrix A = A
1
· A
2
· … · A
n
where each A
i
is a m
i

× n
i
matrix with entries in F
i
such that
| |
[ ]
B
C
T A o o = where for
every n-vector Į = Į
1
· Į
2
· … · Į
n
e V we have

1
1
1
...
n
n
n
C C
T T o o · ·

=
1
1 1
...
n
n n
B B
A A o o · ·

.

Further more T ÷ A is a one to one correspondence between
the set of all n-linear transformation from V into W of type II
and the set of all (m
1
× n
1
, m
2
× n
2
, …, m
n
× n
n
) n-matrices over
the n-field F. The n-matrix A = A
1
· A
2
· … · A
n
which is
associated with T = T
1
· T
2
· … · T
n
is called the n-matrix of
T relative to the n-ordered basis B and C. From the equality

2
1 2
...
i n
j J n j
T T T T o o o o = · · · =
1 2
1 2
1 2
1 1 2 2
1 1 1
...
n
n
n
m m m
n n
k j j k j j k j j
k k k
A A A þ þ þ
= = =
· · ·
¯ ¯ ¯

says that A = A
1
· A
2
· … · A
n
is the n-matrix whose n-
columns
1 2
1 1 1 2 2 2
1 2 1 2 1 2
{ , ,..., } { , ,..., } ... { , ,..., }
n
n n n
n n n
A A A A A A A A A · · ·
are given by [ ]
C
A To = i.e. [ ]
i
i i
j i j
C
A To = , i = 1, 2, … , n and j =
1, 2, …, n
i
. If U = U
1
· U
2
· … · U
n
is another n-linear
transformation from V = V
1
· V
2
· … · V
n
into W = W
1
· W
2

· …· W
n
and P =
1 2
1 1 2 2
1 1 1
{ ,..., } { ,..., } { ,..., }
n
n n
n n n
P P P P P P · · is the
n-matrix of U relative to the n-ordered basis B and C then CA +
P is the matrix CT + U where C = C
1
· … · C
n
e F
1
· F
2
· …
· F
n
where C
i
e F
i
, for i = 1, 2, …,n. That is clear because
46
( )
[ ] [ ] [ ]
i i i
i i i i i i i i i
j j i j i j i j i j
C C C
C A P C T U C T U o o o o + = + = + . This is
true for every i, i = 1, 2, … , n and 1  j  n
i
.

Several interesting results analogous to usual vector spaces can
be derived for n-vector spaces over n-field of type II and its
related n-linear transformation of type II.
The reader is expected to prove the following theorem.

THEOREM 1.2.20: Let V = V
1
· V
2
· … · V
n
be a (n
1
, n
2
, … ,
n
n
), n-vector space over the n-field F = F
1
· F
2
· … · F
n
and
let W = W
1
· W
2
· … · W
n
be a (m
1
, m
2
, …, m
n
) dimensional
vector space over the same n-field F = F
1
· … · F
n
. For each
pair of ordered n-basis B and C for V and W respectively the n-
function which assigns to a n-linear transformation T of type II,
its n-matrix relative to B, C is an n-isomorphism between the n-
space L
n
(V, W) and the space of all n-matrices of n-order (m
1
×

n
1
, m
2
×

n
2
, … , m
n
×

n
n
) over the same n-field F.

Since we have the result to be true for every pair of vector
spaces V
i
and W
i
over F
i
, we can appropriately extend the result
for V = V
1
· V
2
· … · V
n
and W = W
1
· W
2
· … · W
n
over
F = F
1
· F
2
· … · F
n
as the result is true for every i.

Yet another theorem of interest is left for the reader to prove.

THEOREM 1.2.21: Let V = V
1
· V
2
· … · V
n
, W = W
1
· W
2
·
… · W
n
and Z = Z
1
· Z
2
· … · Z
n
be three (n
1
, n
2
, …, n
n
), (m
1
,
m
2
, …, m
n
), and (p
1
, p
2
, …, p
n
) dimensional n-vector spaces
respectively defined over the n-field F = F
1
· F
2
· … · F
n
. Let
T = T
1
· T
2
· … · T
n
be a n-linear transformation of type II
from V into W and U = U
1
· U
2
· … · U
n
a n-linear
transformation of type II from W into Z. Let B = B
1
· B
2
· … ·
B
n
=
1
1 1 1
1 2
{ , ,..., }
n
o o o ·
2
2 2 2
1 2
{ , ,..., }
n
o o o · … ·
1 2
{ , ,..., }
n
n n n
n
o o o , C
= C
1
· C
2
·… · C
n
=
1
1 1 1
1 2
{ , ,..., }
m
þ þ þ ·
2
2 2 2
1 2
{ , ,..., }
m
þ þ þ · … ·
{
1 2
, ,..., }
n
n n n
m
þ þ þ and D = D
1
· D
2
·…· D
n
=
1
1 1 1
1 2
{ , , , }
p
¸ ¸ ¸ !
·
2
2 2 2
1 2 1 2
{ , , , } ... { , , , }
n
n n n
p p
¸ ¸ ¸ ¸ ¸ ¸ · · ! ! be n-ordered n-basis
47
for the n-spaces V, W and Z respectively, if A = A
1
· A
2
· … ·
A
n
is a n-matrix relative to T to the pair B, C and R = R
1
· R
2
·
… · R
n
is a n-matrix of U relative to the pair C and D then the
n-matrix of the composition UT relative to the pair B, D is the
product n-matrix E = RA i.e., if E = E
1
· E
2
· … · E
n
= R
1
A
1

· R
2
A
2
· … · R
n
A
n
.

THEOREM 1.2.22: Let V = V
1
· V
2
· … · V
n
be a (n
1
, n
2
, …,
n
n
) finite dimensional n-vector space over the n-field F = F
1
·
F
2
· … · F
n
and let B = B
1
· B
2
· … · B
n
=
1
1 1 1
1 2
{ , , , }
n
o o o !
·
2
2 2 2
1 2
{ , , , }
n
o o o ! · … ·
1 2
{ , , , }
n
n n n
n
o o o ! and C = C
1
· C
2

· …· C
n
=
1
1 1 1
1 2
{ , , , }
m
þ þ þ ! ·
2
2 2 2
1 2
{ , , , }
m
þ þ þ ! · …·
1 2
{ , , , }
n
n n n
m
þ þ þ ! be an n-ordered n-basis for V. Suppose T is a
n-linear operator on V.
If P = P
1
· P
2
· … · P
n
=
1 2
1 1 2 2
1 1
{ ,..., } { ,..., }
n n
P P P P · · …
·
1 2
{ , ,..., }
n
n n n
n
P P P is a (n
1
× n
1
, n
2
× n
2
, … , n
n
× n
n
) n-matrix
with j
th
component of the n-columns [ ]
i i
j j B
P þ = ; 1  j  n
i
, i =
1, 2, …, n then
1
[ ] [ ]
C B
T P T P
÷
= i.e.
1 2
1 2
[ ] [ ] ... [ ]
n
n
C C C
T T T · · ·
=
1 2
1 1 1
1 1 1 2 2 2
[ ] [ ] ... [ ]
n
n n n
B B B
P T P P T P P T P
÷ ÷ ÷
· · · . Alternatively if U
is the n-invertible operator on V defined by
i i
i j j
U o þ = , j = 1, 2,
…, n
i
, i = 1, 2, …, n then
1
[ ] [ ] [ ] [ ]
C B B B
T U B T U
÷
= i.e.
1 2
1 2
[ ] [ ] ... [ ]
n
n
C C C
T T T · · · =
1 1 1 2 2 2
1 1 1
1 1 1 2 2 2
[ ] [ ] [ ] [ ] [ ] [ ] ... [ ] [ ] [ ] .
n n n
n n n
B B B B B B B B B
U T U U T U U T U
÷ ÷ ÷
· · ·

The proof of the above theorem is also left for the reader.

We now define similar space n-matrices.

DEFINITION 1.2.12: Let A = A
1
· A
2
· … · A
n
be a n-mixed
square matrix of n-order (n
1
×

n
1
, n
2
×

n
2
, …, n
n
×

n
n
) over the n-
field F = F
1
· F
2
· … · F
n
i.e., each A
i
takes its entries from
the field F
i
, i = 1,2, …, n. B = B
1
· B
2
· … · B
n
is a n-mixed
48
square matrix of n-order (n
1
×

n
1
, n
2
×

n
2
, …, n
n
×

n
n
) over the
same n-field F. We say that B is n-similar to A over the n-field F
if there is an invertible n-matrix P = P
1
· P
2
· … · P
n
of n-
order (n
1
×

n
1
, n
2
×

n
2
, …, n
n
×

n
n
) over the n-field F such that B
= P
-1
A P i.e.
B
1
· B
2
· … · B
n
=
1 1 1
1 1 1 2 2 2
...
n n n
P A P P A P P A P
÷ ÷ ÷
· · · .

Now we proceed on to define the new notion of n-linear
functionals or linear n-functionals. We know we were not in a
position to define n-linear functionals in case of n-vector space
over the field F i.e. for n-vector spaces of type I.
Only in case of n-vector spaces defined over the n-field of
type II we are in a position to define n-linear functionals on the
n-vector space V.
Let V = V
1
· V
2
· … · V
n
be a n-vector space over the n-
field F = F
1
· F
2
· … · F
n
of type II. A n-linear transformation
f = f
1
· f
2
· … · f
n
from V into the n-field of n-scalars is also
called a n-linear functional on V i.e., if f is a n-function from V
into F such that f(Co + þ) = Cf(o) + f(þ) where C = C
1
· C
2
·
… · C
n
; i = 1, 2, …, n, Į = Į
1
· Į
2
· … · Į
n
and ȕ = ȕ
1
· ȕ
2
·
… · ȕ
n
where Į
i
, ȕ
i
e V
i
for each i, i = 1,2, …, n. f = f
1
· f
2
·
… · f
n
where each f
i
is a linear functional on V
i
; i = 1, 2, …, n.
i.e.,

f(CĮ + ȕ) = (f
1
· f
2
· … · f
n
) (C
1
Į
1
+ ȕ
1
) · (C
2
Į
2
+ ȕ
2
) ·
… · (C
n
Į
n
+ ȕ
n
)
= (C
1
f
1

1
) + f
1

1
))· (C
2
f
2

2
) + f
2

2
)) · … ·
(C
n
f
n

n
) + f
n

n
))]
= f
1
(C
1
Į
1
+ ȕ
1
) · f
2
(C
2
Į
2
+ ȕ
2
) ·…· f
n
(C
n
Į
n
+ ȕ
n
)
= (C
1
f
1

1
) + f
1

1
))· (C
2
f
2

2
) + f
2

2
)) · … ·
(C
n
f
n

n
) + f
n

n
))

for all n-vectors Į,

ȕ e V and C e F.
We make the following observation.

Let F = F
1
· F
2
· … · F
n
be a n-field and let
1
n
1
F ·
2
n
2
F ·
… ·
n
n
n
F be a n-vector space over the n-field F of type II. A n-
49
linear function f = f
1
· f
2
· … · f
n
from
1 2 n
n n n
1 2 n
F F ... F · · · to
F
1
· F
2
· … · F
n
given by f
1
(
1
1 1
1
,...,
n
x x ) · f
2
(
2
2 2
1 n
x ,..., x ) · …
· f
n
(
n
n n
1 n
x ,..., x ) = (
1 1
1 1 1 1
1 1 n n
x ... x o + + o ) · (
2 2
2 2 2 2
1 1 n n
x ... x o + + o ) ·
… · (
2 n
n n n n
1 1 n n
x ... x o + + o ) where
i
j
o are in F
i
, 1  j  n
i
; for each
i = 1, 2, … , n; is a n-linear functional on
1 2 n
n n n
1 2 n
F F ... F · · · . It
is the n-linear functional which is represented by the n-matrix
[
1
1 1 1
1 2 n
, ,..., o o o ] · [
2
2 2 2
1 2 n
, ,..., o o o ] · … · [
n
n n n
1 2 n
, ,..., o o o ]
relative to the standard ordered n-basis for
1 2 n
n n n
1 2 n
F F ... F · · ·
and the n-basis {1} · {1} · … · {1} for F = F
1
· F
2
· … ·
F
n
.
i
j
o =
i
i j
f (E ); j = 1, 2, …, n
i
for every i = 1, 2, …, n. Every
n-linear functional on
1 2 n
n n n
1 2 n
F F ... F · · · is of this form for
some n-scalars (
1
1 1
1 n
o o ! ) · (
2
2 2
1 n
o o ! ) · … · (
n
n n n
1 2 n
o o o ! ).
That is immediate from the definition of the n-linear functional
of type II because we define
i i
j i j
f (E ) o = .

f
1
(
1
1 1
1 n
x , , x ! )· f
2
(
2
2 2
1 n
x , , x ! ) · … · f
n
(
n
n n n
1 2 n
x , x , , x ! )
=
i
n
1 1
1 j j
j 1
f x E
=
| |
·
|
\ .
¯
2
n
2 2
2 j j
j 1
f x E
=
| |
·
|
\ .
¯
… ·
n
n
n n
n j j
j 1
f x E
=
| |
|
\ .
¯

=
1 1
j 1 j
j
x f (e ) ·
¯

2 2
j 2 j
j
x f (e ) ·
¯
… ·
n n
j n j
j
x f (e )
¯

=
1 2 n
n n n
1 1 2 2 n n
j j j j j j
j j 1 j 1
a x a x ... a x
= =
· · ·
¯ ¯ ¯
.

Now we can proceed onto define the new notion of n-dual space
or equivalently dual n-space of a n-space V = V
1
·…· V
n
defined over the n-field F = F
1
· F
2
·…· F
n
of type II. Now as
in case of L
n
(V,W) = L (V
1
,W
1
) · L (V
2
,W
2
) · … · L(V
n
, W
n
)
we in case of linear functional have L
n
(V,F) = L(V
1
,F
1
) · L(V
2
,
F
2
) · …· L(V
n
,F
n
). Now V* = L
n
(V,F)=
* * *
1 2 n
V V ... V · · ·
i.e. each
*
i
V is the dual space of V
i
, V
i
defined over the field F
i
;
i = 1, 2, …, n. We know in case of vector space V
i
, dim
*
i
V =
50
dim V
i
for every i. Thus dim V = dim V
*
= dim
*
1
V · dim
*
2
V ·
… · dim
*
n
V . If B = {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } · … ·
{
n
n n n
1 2 n
, ,..., o o o } is a n-basis for V, then we know for a n-linear
functional of type II. f = f
1
· f
2
· … · f
n
we have f
k
on V
k
is
such that
k k k
i j ij
f ( ) o = o true for k = 1, 2, …, n. In this way we
obtain from B a set of n-tuple of n
i
sets of distinct n-linear
functionals {
1
1 1 1
1 2 n
f , f ,..., f }·{
2
2 2 2
1 2 n
f , f ,..., f }·…·{
k
n n n
1 2 n
f , f ,..., f }
= f
1
· f
2
· … ·f
n
on V. These n-functionals are also n-linearly
independent over the n-field F = F
1
· F
2
· … · F
n
i.e.,
{
i
i i i
1 2 n
f , f ,..., f } is linearly independent on V
i
over the field F
i
; this
is true for each i, i = 1, 2, …, n.
Thus
i
n
i i i
j j
j 1
f c f
=
=
¯
, i = 1, 2, …, n i.e.
f =
1
n
1 1
j j
j 1
c f
=
¯
·
2
n
2 2
j j
j 1
c f
=
¯
· … ·
n
n
n n
j j
j 1
c f
=
¯
.
i
n
i i i i k
j k k j
k 1
f ( ) c f ( )
=
o = o
¯

=
i
n
i
k kj
k 1
c
=
o
¯

=
i
j
c .

This is true for every i = 1, 2, …, n and 1  j  n
i
. In particular
if each f
i
is a zero functional
i i
j
f ( ) o = 0 for each j and hence the
scalar
i
j
c are all zero. Thus
i
i i i
1 2 n
f , f ,..., f are n
i
linearly
independent functionals of V
i
defined on F
i
, which is true for
each i, i = 1, 2, … . n. Since we know
*
i
V is of dimension n
i
it
must be that
*
i
B = {
i
i i i
1 2 n
f , f ,..., f } is a basis of
*
i
V which we
know is a dual basis of B. Thus B
*
=
* * *
1 2 n
B B ... B · · · =
{
1
1 1 1
1 2 n
f , f ,..., f } · {
2
2 2 2
1 2 n
f , f ,..., f } · … · {
n
n n n
1 2 n
f , f ,..., f } is the n-
dual basis of B = {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } · … ·
51
{
n
n n n
1 2 n
, ,..., o o o }. B* forms the n-basis of V
*
=
* * *
1 2 n
V V ... V · · · .

We prove the following interesting theorem.

THEOREM 1.2.23: Let V = V
1
· … · V
n
be a finite (n
1
, n
2
, …,
n
n
) dimensional n-vector space over the n-field F = F
1
· F
2
·
… · F
n
and let B = {
1
1 1 1
1 2
, ,...,
n
o o o } · {
2
2 2 2
1 2
, ,...,
n
o o o } · … ·
{
1 2
, ,...,
n
n n n
n
o o o } be a n-basis for V. Then there is a unique n-
dual basis
B
*
= {
1
1 1 1
1 2
, ,...,
n
f f f } · {
2
2 2 2
1 2
, ,...,
n
f f f } · … ·{
1 2
, ,...,
n
n n n
n
f f f }
for V* =
1 2
...
n
V V V
- - -
· · · such that ( )
k k
i j ij
f o o = . For each n-
linear functional f = f
1
· f
2
· … · f
n
we have
f =
1
( )
i
n
i i i
k k
k
f f o
=
¯
;
i.e.,
f =
1
1 1 1
1
( )
n
k k
k
f f o
=
¯
·
2
2 2 2
1
( )
n
k k
k
f f o
=
¯
· … ·
1
( )
n
n
n n n
k k
k
f f o
=
¯

and for each n-vector Į = Į
1
· Į
2
· … · Į
n
in V we have
Į
i
=
1
( )
i
n
i i i
k k
k
f o o
=
¯

i.e.,
Į =
1
1 1 1
1
( )
n
k k
k
f o o
=
¯
·
2
2 2 2
1
( )
n
k k
k
f o o
=
¯
· … ·
1
( )
n
n
n n n
k k
k
f o o
=
¯
.

Proof: We have shown above that there is a unique n-basis
which is dual to B. If f is a n-linear functional on V then f is
some n-linear combination of
i
j
f , i  j  k
j
and i = 1, 2, …, n;
and from earlier results observed the scalars
i
j
C must be given
by
i i i
j j
C f ( ) = o , 1 j  k
i ,
i =1, 2, …, n. Similarly if
52
Į =
1
n
1 1
i i
i 1
x
=
o
¯
·
2
n
2 2
i i
i 1
x
=
o
¯
· … ·
n
n
n n
i i
i 1
x
=
o
¯

is a n-vector in V then
f
j
(Į) =
1
n
1 1 1
i j i
i 1
x f ( )
=
o
¯
·
2
n
2 2 2
i j i
i 1
x f ( )
=
o
¯
· … ·
n
n
n n n
i j i
i 1
x f ( )
=
o
¯

=
1
n
1 1
i ij
i 1
x
=
o
¯
·
2
n
2 2
i ij
i 1
x
=
o
¯
· … ·
n
n
n n
i ij
i 1
x
=
o
¯

=
1 2 n
j j j
x x ... x ; · · ·

so that the unique expression for Į as a n-linear combination of
k
j
o , k = 1, 2, …, n; 1  j  n
j
i.e.
Į =
1
n
1 1 1
i i
i 1
f ( )
=
o o
¯
·
2
n
2 2 2
i i
i 1
f ( )
=
o o
¯
· … ·
n
n
n n n
i i
i 1
f ( )
=
o o
¯
.

Suppose N
f
=
1
1
f
N ·
2
2
f
N · … ·
n
n
f
N denote the n-null space of f
the n-dim N
f
= dim
n
n
f
N · … · dim
n
n
f
N but dim
i
i
f
N = dim V
i

1 = n
i
–1 so n-dim N
f
= (dim V
1
–1) · dim V
2
–1 · … · dim V
n

1 = n
1
– 1 · n
2
– 1 · … · n
n
– 1. In a vector space of
dimension n, a subspace of dimension n – 1 is called a
hyperspace like wise in a n-vector space V = V
1
· … · V
n
of
(n
1
, n
2
, … , n
n
) dimension over the n-field F = F
1
· F
2
· …
· F
n
then the n-subspace has dimension (n
1
– 1, n
2
– 1, …, n
n

1) we call that n-subspace to be a n-hyperspace of V. Clearly N
f

is a n-hyper subspace of V.

Now this notion cannot be defined in case of n-vector spaces of
type I. This is also one of the marked differences between the n-
vector spaces of type I and n-vector spaces of type II.

Now we proceed onto define yet a special feature of a n-vector
space of type II, the n-annihilator of V.

DEFINITION 1.2.13: Let V = V
1
, · … · V
n
be a n-vector space
over the n-field F = F
1
· F
2
· … · F
n
of type II. Let S = S
1
·
S
2
· … · S
n
be a n-subset of V = V
1
· … · V
n
(i.e. S
i
_ V
i
, i =
53
1, 2, …, n); the n-annihilator of S is S
o
=
1
... · ·
o o
n
S S of n-
linear functionals on V such that f (Į) = 0 · 0 · … · 0 i.e. if f
= f
1
· f
2
· … · f
n
for every Į e S i.e., Į = Į
1
· Į
2
· … · Į
n

e S
1
· S
2
· … · S
n
. i.e., f
i
(o
i
) = 0 for every Į
i
e S
i
; i = 1, 2, …,
n. It is interesting and important to note that
1
... = · ·
o o o
n
S S S
is an n-subspace of V
*
=
* * *
1 2
...
n
V V V · · · whether S is an n-
subspace of V or only just a n-subset of V. If S = (0 · 0 · …
· 0) then S
o
= V
*
. If S = V then S
o
is just the zero n-subspace
of V
*
.

Now we prove an interesting result in case of finite dimensional
n-vector space of type II.

THEOREM 1.2.24: Let V = V
1
· V
2
· … · V
n
be a n-vector
space of (n
1
, n
2
, …, n
n
) dimension over the n-field F = F
1
· F
2

· … · F
n
of type II. Let W = W
1
· W
2
· … · W
n
be a n-
subspace of V. Then dim W + dim W° = dim V (i.e., if dim W is
(k
1
, k
2
, …, k
n
)) that is (k
1
, k
2
, …, k
n
) + (n
1
– k
1
, n
2
– k
2
, …, n
n

k
n
) = (n
1
, n
2
, …, n
n
).

Proof: Let (k
1
, k
2
, …, k
n
) be the n-dimension of W = W
1
· W
2

· … · W
n.
. Let
P = {
1
1 1 1
1 2 k
, ,..., o o o } · {
2
2 2 2
1 2 k
, ,..., o o o } · … · {
n
n n n
1 2 k
, ,..., o o o }
= P
1
· P
2
· … · P
n
be a n-basis of W. Choose n-vectors
{
1 1
1 1
k 1 n
,...,
+
o o } · {
2 2
2 2
k 1 n
,...,
+
o o } · … · {
n n
n n
k 1 n
,...,
+
o o }
in V such that
B = {
1
1 1 1
1 2 n
, ,..., o o o }· {
2
2 2 2
1 2 n
, ,..., o o o }· … ·{
n
n n n
1 2 n
, ,..., o o o }
= B
1
· B
2
· … · B
n
is a n-basis for V.
Let
f =
1
1 1 1
1 2 n
{f , f ,..., f } ·
2
2 2 2
1 2 n
{f , f ,..., f }· … ·
n
n n n
1 2 n
{f , f ,..., f }
= f
1
· f
2
· … · f
n
be a n-basis for V* which is the dual n-
basis for V. The claim is that {
1 1
1 1
k 1 n
f , , f
+
! } · {
2 2
2 2
k 1 n
f , , f
+
! }
· … · {
n n
n n
k 1 n
f , , f
+
! } is a n-basis for the n-annihilator of W°
=
o o o
1 2 n
W W ... W · · · . Certainly
r
i
f belongs to
o
r
W and i  k
r

54
+ 1; r = 1, 2, …, n, 1  i  k
r
, because
r r
i j ij
f ( ) o = o and
ij
0 o = if i
 k
r
+ 1 and j  k
r
+ 1, from this it follows that for i < k
r
.
r
i
f (Į
r
)
= 0 whenever o
r
is a linear combination
r
r r r
1 2 k
a , a ,..., a . The
functionals
r r r
r r r
k 1 k 2 n
a , a ,..., a
+ +
are independent for every r = 1, 2,
…, n. Thus we must show that they span
o
r
W , for r = 1, 2, …, n.
Suppose
r *
r
f V e . Now
r
n
r r r r
i i
i 1
f f ( )f
=
= o
¯
so that if f

r
is in
o
r
W we
have
r r
i
f ( ) 0 o = for i  k
r
and f
r
=
r
r
n
r r r
i i
i k 1
f ( )f
= +
o
¯
.
We have to show if dim W
r
= k
r
and dim V
r
= n
r
then dim
o
r
W =
n
r
– k
r
; this is true for every r. Hence the theorem.

COROLLARY 1.2.4: If W = W
1
· W
2
· … · W
n
is a (k
1
, k
2
,…,
k
n
) dimensional n-subspace of a (n
1
, n
2
, …, n
n
) dimensional n-
vector space V = V
1
· V
2
· … · V
n
over the n-field F = F
1
·
F
2
· … · F
n
then W is the intersection of (n
1
– k
1
) · (n
2
– k
2
) ·
… · (n
n
– k
n
) tuple n-hyper subspaces of V.

Proof: From the notations given in the above theorem in W =
W
1
· W
2
· … · W
n
, each W
r
is the set of vector Į
r
such that
f
i

r
) = 0, i = k
r
+ 1, …, n
r
. In case k
r
= n
r
– 1 we see W
r
is the
null space of
r
n
f . This is true for every r, r = 1, 2, …, n. Hence
the claim.

COROLLARY 1.2.5: If W
1
=
1 2
1 1 1
...
n
W W W · · · and W
2
=
1 2
2 2 2
...
n
W W W · · · are n-subspace of the n-vector space V =
V
1
· V
2
· … · V
n
over the n-field F = F
1
· F
2
· … · F
n
of
dimension (n
1
, n
2
, …, n
n
) then W
1
= W
2
if and only if
1
o
W =
2
o
W
i.e., if and only if
( )
1
o
i
W =
( )
2
o
i
W for every i = 1, 2, …, n.

Proof: If W
1
= W
2
i.e., if
1 2 n
1 1 1
W W ... W · · · =
1 2 n
2 2 2
W W ... W · · ·
55
then each
j j
1 2
W W = for j = 1, 2, …, n, so that
( )
o
j
1
W =
( )
o
j
2
W for
every j = 1, 2, …,n. Thus
o
1
W =
o
2
W . If on the other hand W
1

W
2
i.e.
1 2 n
1 1 1
W W ... W · · · 
1 2 n
2 2 2
W W ... W · · · then one of
the two n-subspaces contains a n-vector which is not in the
other. Suppose there is a n-vector Į
2
= Į
1
· Į
2
· … · Į
n
which
is in W
2
and not in W
1
i.e. Į
2
e W
2
and Į
2
eW
1
by the earlier
corollary just proved and the theorem there is a n-linear
functional f = f
1
· f
2
· … · f
n
such that f(ȕ) = 0 for all ȕ in
W, but f(Į
2
)  0. Then f is in
o
1
W but not in
o
2
W and
o
1
W 
o
2
W .
Hence the claim.

Next we show a systematic method of finding the n-annihilator
n-subspace spanned by a given finite n-set of n-vectors in
1 2 n
n n n
1 1 n
F F ... F · · · . Consider a n-system of n-homogeneous
linear n-equations, which will be from the point of view of the
n-linear functionals. Suppose we have n-system of n-linear
equations.
1 1
1 1 1 1
1 1 1 1
11 1 1n n
1 1 1 1
m 1 1 m n n
A x A x 0
A x A x 0,
+ + =
+ + =
!
# #
!

2
2
2 2 2 2
2 2 2 2
11 1 1n n
2 2 2 2
m 1 1 m n n
A x A x 0
A x A x 0,
+ + =
+ + =
!
# #
!

so on
n n
n n n n
n n n n
11 1 1n n
n n n n
m 1 1 m n n
A x A x 0
A x A x 0
+ + =
+ + =
!
# #
!

for which we wish to find the solutions. If we let
k
i
f , i = 1, 2,
…n, m
k
, k = 1, 2, …,n be the linear function on
k
n
k
F , defined by
56
k
k k k
i 1 n
f (x ,..., x ) =
k k
k k k k
i1 1 in n
A x ... A x + + , this is true for every k, k =
1, 2, …, n; then we are seeking the n-subspace of
1 2 n
n n n
1 2 n
F F ... F ; · · · that is all Į = Į
1
· Į
2
· … · Į
n
such that
k k
i
f ( ) o = 0, i = 1, 2, …, m
k
and k = 1, 2, …, n. In other words
we are seeking the n-subspace annihilated by
1
1 1 1
1 2 m
{f , f ,..., f }·
2
2 2 2
1 2 m
{f , f ,..., f }· … ·
n
n n n
1 2 m
{f , f ,..., f }. Row reduction of each of
the coefficient matrix of the n-matrix provides us with a
systematic method of finding this n-subspace. The (n
1
, n
2
, … .
n
n
) tuple (
1
1 1
i1 in
A ,..., A )· (
2
2 2
i1 in
A ,..., A ) · … · (
n
n n
i1 in
A ,..., A )
gives the coordinates of the n-linear functional
k
i
f , k = 1, 2, …,
n relative to the n-basis which is n-dual to the standard n-basis
of
1 2 n
n n n
1 2 n
F F ... F · · · . The n-row space of the n-coefficient n-
matrix may thus be regarded as the n-space of n-linear
functionals spanned by (
1
1 1 1
1 2 m
f , f ,..., f ) · (
2
2 2 2
1 2 m
f , f ,..., f ) · … ·
(
n
n n n
1 2 m
f , f ,..., f ). The solution n-space is the n-subspace n-
annihilated by this space of n-functionals.

Now one may find the n-system of equations from the n-dual
point of view. This is suppose that we are given, m
i
n-vectors in
1 2 n
n n n
1 1 n
F F ... F · · · ;
Į =
1 n
i i
... o · ·o = (
1
1 1 1
i1 i 2 in
A A ...A ) · … ·(
n
n n n
i1 i 2 in
A A ...A ) and we
find the n-annihilator of the n-subspace spanned by these
vectors. A typical n-linear functional on
1 2 n
n n n
1 1 n
F F ... F · · ·
has the form f
1
(
1
1 1
1 n
x ,..., x )· f
2
(
2
2 2
1 n
x ,..., x )· … · f
n
(
n
n n
1 n
x ,..., x )
= (
1 1
1 1 1 1
1 1 n n
(c x ... c x ) + + ·
2 2
2 2 2 2
1 1 n n
(c x ... c x ) + + · … ·
n n
n n n n
1 1 n n
(c x ... c x ) + + and the condition that f
2
· … ·f
n
be in this
n-annihilator; that is
1 2 n
1 2 n
n n n
1 1 2 2 n n
i j j i j j i j j
j 1 j 1 j 1
A c A c ... A c
= = =
· · ·
¯ ¯ ¯
= 0 ·
0 · … · 0, and 1  i
1
 m
1
,…, 1  i
n
 m
n
that is (
1
1 1
1 n
c , , c ! )
· … · (
n
n n
1 n
c , , c ! ) be the n-solution of the system A
1
X
1
·
A
2
X
2
· … · A
n
X
n
= 0 · 0 · … · 0.
57

As in case of usual vector space we in case of the n-vector
spaces of type II define the double dual. Here also it is
important to mention in case of n-vector spaces of type I we
cannot define dual or double dual. This is yet another difference
between the n-vector spaces of type I and type II.

DEFINITION 1.2.14: Let V = V
1
· V
2
· … · V
n
be a n-vector
space over the n-field F = F
1
· F
2
· … · F
n
(i.e. each V
i
is a
vector space over F
i
). Let V* =
* * *
1 2
...
n
V V V · · · be the n-
vector space which is the n-dual of V over the same n-field F =
F
1
· F
2
· … · F
n
. The n-dual of the n-dual space V* i.e., V**
in terms of n-basis and n-dual basis is given in the following.
Let Į = Į
1
· Į
2
· … · Į
n
be a n-vector in V, then x induces
a n-linear functional L
Į
=
1 2
1 2
...
n
n
L L L
o o o
· · · defined by L
Į
(f)
=
1
1 1
( ) ... ( )
n
n n
L f L f
o o
· · (where f = f

1
· f

2
· … · f

n
) i.e.
L
Į
(f) =
1
1 1
( ) ... ( )
n
n n
L f L f
o o
· · = f(Į) = f

1

1
) · f

2

2
) · …
· f
n

n
), f e V
*
=
* *
1
...
n
V V · · f

i
e
*
i
V for i = 1, 2, …, n. The
fact that each
i
i
L
o

is linear is just a reformulation of the
definition of the linear operators in
i
V
-
for each i = 1, 2, …, n.
L
o
(cf + g) =
1
1 1 1 1
( ) ... ( )
n
n n n n
L c f g L c f g
o o
+ · · +
= ( c
1
f
1
+ g
1
)(Į
1
)· … · (c
n
f
n
+ g
n
)(Į
n
)
= (c
1
f
1

1
) + g
1

1
)) · … · (c
n
f
n

n
) + g
n

n
))
= (c
1

1
1
L
o
(f

1
) +
1
1
L
o
(g
1
)) · … · (c
n

n
n
L
o
(f
n
) +
n
n
L
o
(g
n
))
= cL
Į
( f ) + L
Į
(g).

If V= V
1
· V
2
·...· V
n
is a finite (n
1
, n
2
, … . n
n
) dimensional
and Į  0 then L
Į
 0 , in other words there exits a n-linear
functional f = f

1
· f

2
·...· f

n
such that f(x)  0 i.e. f(Į) =
f
1

1
)· …· f
n

n
) for each f
i

i
)  0, i = 1, 2, …,n.

The proof is left for the reader, using the fact if we choose a
ordered n-basis B = {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } · … ·
58
{
n
n n n
1 2 n
, ,..., o o o } for V = V
1
· V
2
· ... · V
n
such that Į =
1 2 n
1 1 1
, ,..., o o o and let f be a n-linear functional which assigns to
each n-vector in V its first coordinate in the n-ordered basis B.

We prove the following interesting theorem.

THEOREM 1.2.25: Let V = V
1
· V
2
·... ·V
n
be a finite (n
1
, n
2
,
…, n
n
) dimensional n-vector space over the n-field F = F
1
· F
2

· … · F
n
. For each n-vector Į = Į
1
· Į
2
· … · Į
n
in V define
L
o
(f) =
1
1 1
( ) ... ( )
n
n n
L f L f
o o
· · = f (o) = f
1
(o
1
) · … · f
n
(o
n
); f
in V*. The mapping Į ÷ L
Į
, is then an n-isomorphism of V onto
V**.

Proof: We showed that for each Į = Į
1
· Į
2
· … · Į
n
in V, the
n-function L
Į
=
1 n
1 n
L ... L
o o
· · is n-linear. Suppose Į = Į
1
· Į
2
·
… · Į
n
and ȕ = ȕ
1
· ȕ
2
· … · ȕ
n
are in V
1
· V
2
·...· V
n
and
c = c
1
· c
2
· … · c
n
is in F = F
1
· F
2
· … · F
n
and let Ȗ = cĮ
+ ȕ i.e., Ȗ
1
· Ȗ
2
· … · Ȗ
n
=(c
1
Į
1
+ ȕ
1
) · (c
2
Į
2
+ ȕ
2
) · … · (c
n
Į
n

+ ȕ
n
). Thus for each f in V*
L
¸
(f) = f(Ȗ)
= f(cĮ + ȕ)
= f
1
(c
1
Į
1
+ ȕ
1
) · … · f
n
(c
n
Į
n
+ ȕ
n
)
= [c
1
f
1

1
) + f
1

1
)] · … · [c
n
f
n

n
) + f
n

n
)]
=
1 1 n n
1 1 1 1 1 n n n n n
c L (f ) L (f ) ... c L (f ) L (f )
o þ o þ
+ · · +
= cL
Į
(f) + L
ȕ
(f)
L
Ȗ
= cL
Į
+ L
ȕ
.

This proves the n-map Į ÷ L
Į
is an n-linear transformation
from V into V**. This n-transformation is n-nonsingular i.e. L
Į

=0 if and only if Į = 0. Hence dim V** = dim V* = dim V= (n
1
,
n
2
, …, n
n
).
The following are the two immediate corollaries to the
theorem.

COROLLARY 1.2.6: Let V = V
1
· V
2
·... · V
n
be a (n
1
, n
2
, …,
n
n
) finite dimensional n-vector space over the n-field F = F
1
·
59
F
2
· ... · F
n
. If L is a n-linear functional on the dual space V*
of V then there is a unique n-vector Į= Į
1
· Į
2
· … · Į
n
in V
such that L(f) = f (Į) for every f in V*.

The proof is left for the reader to prove.

COROLLARY 1.2.7: If V = V
1
· V
2
·...· V
n
is a (n
1
, n
2
, …, n
n
)
dimensional n-vector space over the n-field F = F
1
· F
2
·...·
F
n
. Each n-basis for V* is the dual of some n-basis for V.

Proof: Given V = V
1
· V
2
·... · V
n
is a (n
1
, n
2
, …, n
n
) vector
space over the n-field F = F
1
· F
2
· ... · F
n
. V* =
* *
1 n
V ... V · · is the dual n-space of V over the n-field F.
Let B
*
= {
1
1 1
1 n
f ,..., f } · ... · {
n
n n
1 n
f ,..., f } be a n-basis for V* by
an earlier theorem there is a n-basis
1
1 1 1
1 2 n
{L , L ,..., L } ·
2
2 2 2
1 2 n
{L , L ,..., L }·…·
n
n n n
1 2 n
{L , L ,..., L } for V** =
** **
1 n
V ... V · ·
such that
k k
i j ij
L (f ) = o for k = 1, 2, …, n, 1  i  n
k
. Using the
above corollary for each i, there is a vector
k
i
o in V
k
such that
k k
i
L (f ) =
k k
i
f ( ) o for every f

k
in
k
V
-
, such that
k
i
L =
k
i
k
L
o
. It
follows immediately that {
1
1 1 1
1 2 n
, ,..., o o o } · {
2
2 2 2
1 2 n
, ,..., o o o } ·
… · {
n
n n n
1 2 n
, ,..., o o o } is a n-basis for V and B* is the n-dual of
this n-basis.

In view of this we can say (W
o
)
o
= W.

Now we prove yet another interesting theorem.

THEOREM 1.2.26: If S is any n-set of a finite (n
1
, n
2
, …, n
n
)
dimensional n-vector space V = V
1
· V
2
· … · V
n
then (S
o
)
o
is
the n-subspace spanned by S.

Proof: Let W = W
1
· W
2
· … · W
n
be a n-subspace spanned
by the n-set S = S
1
· … · S
n
i.e. each S
i
spans W
i
, i = 1, 2, … .
n. Clearly W
o

= S
o
. Therefore what is left over to prove is that
60
W = W
oo
. We can prove this yet in another way. We know dim
W + dim W
o
= dim V. dim W
o
+ dim (W
o
)
o
= dim V* since dim
V = dim V* we have dim W + dim W
o
= dim W
o
+ dim W
oo

which implies dim W = dim W
oo
. Since W is a subspace of W
oo

we see that W = W
oo
.
Let V be an n-vector space over the n-field of type II. We
define an n-hyper subspace or n-hyperspace of V. We assume V
is (n
1
, n
2
, …, n
n
) dimension over F = F
1
· F
2
· … · F
n
. If N is
a n-hyperspace of V i.e. N is of (n
1
– 1, n
2
– 1, …, n
n
– 1)
dimension over F then we can define N to be a n-hyperspace of
V if
(1) N is a proper n-subspace of V
(2) If W is a n-subspace of V which contains N then either
W = N or W = V.
Conditions (1) and (2) together say that N is a proper n-
subspace and there is no larger proper n-subspace in short N is a
maximal proper n-subspace of V.

Now we define n-hyperspace of a n-vector space.

DEFINITION 1.2.15: If V = V
1
· V
2
· … · V
n
is a n-vector
space over the n-field F = F
1
· F
2
· … · F
n
a n-hyperspace in
V is a maximal proper n-subspace of V.

We prove the following theorem on n-hyperspace of V.

THEOREM 1.2.27: If f = f

1
· … · f
n
is a non-zero n-linear
functional on the n-vector space V = V
1
· V
2
·...· V
n
of type
II over the n-field F = F
1
· F
2
·… · F
n
, then the n-hyperspace
in V is the n-null space of a (not unique) non-zero n-linear
functional on V.

Proof: Let f = f

1
· f

2
·...· f

n
be a non zero n-linear functional
on V = V
1
· V
2
· … ·V
n
and N
f
=
1 2 n
1 2 n
f f f
N N ... N · · · its n-
null space. Let Į = Į
1
· Į
2
· … · Į
n
be a n-vector in V = V
1
·
V
2
·...· V
n

which is not in N
f
i.e., a n-vector such that f(Į)  0
· … · 0. We shall show that every n-vector in V is in the n-
subspace spanned by N
f
and Į. That n-subspace consist of all n-
61
vector Ȗ + cĮ where ¸ = Ȗ
1
· … ·Ȗ
n
and c = c
1
· c
2
· ... · c
n
, Ȗ
in N
f
, c in F = F
1
· F
2
· … · F
n
. Let ȕ = ȕ
1
· ȕ
2
· … · ȕ
n
be
in V. Define
1 1 n n
1 1 n n
f ( ) f ( ) f ( )
c ...
f ( ) f ( ) f ( )
þ þ þ
= = · ·
o o o
,
i.e., each
c
i
=
i i
i i
f ( )
f ( )
þ
o
, i = 1, 2, … , n;
which makes sense because each f
i

i
)0, i = 1, 2,…, n i.e., f(Į)
 0. Then the n-vector Ȗ = ȕ – cĮ is in N
f
since f(Ȗ) = f(ȕ – cĮ) =
f(ȕ) – cf(Į) = 0. So ȕ is in the n-subspace spanned by N
f
and Į.
Now let N be the n-hyperspace in V. For some n-vector Į = Į
1
· Į
2
· … · Į
n
which is not in N. Since N is a maximal proper
n-subspace, the n-subspace spanned by N and Į is the entire n-
space V. Therefore each n-vector ȕ in V has the form þ = ¸ +
co, ¸ = ¸
1
· … · ¸
n
in N. c = c
1
· … · c
n
in F = F
1
· … · F
n

and N = N
1
· N
2
· … · N
n
where each N
i
is a maximal proper
subspace of V
i
for i = 1, 2, …, n. The n-vector Ȗ and the n-
scalars c are uniquely determined by ȕ. If we have also ȕ = Ȗ
1
+
c
1
Į, Ȗ
1
in N, c
1
in F then (c
1
– c)Į = Ȗ – Ȗ
1
if c
1
– c  0 then Į
would be in N, hence c = c
1
and ¸ = ¸
1
. i.e., if ȕ is in V there is a
unique n-scalar c such that ȕ – cĮ is in N. Call the n-scalar g(ȕ).
It is easy to see g is an n-linear functional on V and that N is a
n-null space of g.

Now we state a lemma and the proof is left for the reader.

LEMMA 1.2.2: If g and h are n-linear functionals on a n-vector
space V then g is a n-scalar space V then g is a n-scalar
multiple of f if and only if the n-null space of g contains the n-
null space of f that is if and only if f(x) = 0 · … · 0 implies
g(x) = 0 · … · 0.

We prove the following interesting theorem for n-linear
functional, on an n-vector space V over the n-field F.

62
THEOREM 1.2.28: Let V = V
1
· V
2
· … · V
n
be an n-vector
space over the n-field F = F
1
· F
2
· … · F
n
. If
1
1 1 1
1
{ , ,..., }
r
g f f
·
2
2 2 2
1
{ , , , }
r
g f f ! ·…·
1
{ , , , }
n
n n n
r
g f f ! be n-linear functionals
on the space V with respective n-null spaces
1
1 1 1
1
{ , ,..., }
r
N N N ·
2
2 2 2
1
{ , ,..., }
r
N N N ·…·
1
{ , ,..., }
n
n n n
r
N N N . Then (g
1
· g
2
·…· g
n
)
is a linear combination of
1
1 1
1
{ ,..., }
r
f f · … ·
1 2
{ , ,..., }
n
n n n
r
f f f
if and only if N

1
· N

2
· … · N

n
contains the n-intersection
{
1
1 1
1
...
r
N N · · } · {
2
2 2
1
...
r
N N · · } · … · {
1
...
n
n n
r
N N · · }.

Proof: We shall prove the result for V
i
of V
1
· V
2
·... · V
n

and since V
i
is arbitrary, the result we prove is true for every i, i
= 1, 2, ..., n. Let g
i
=
i i
i i i i
1 1 r r
c f ... c f + + and
i i
j
f ( ) 0 o = for each j
the clearly g
i
(Į) = 0. Therefore N
i
contains
i
i i i
1 2 r
N N ... N · · · .
We shall prove the converse by induction on the number r
i
.
The proceeding lemma handles the case r
i
= 1. Suppose we
know the result for r
i
= k
i
-1 and let
i
i i i
1 2 k
f , f , , f ! be the linear
functionals with null spaces
i
i i
1 k
N , , N ! such that
i
i i
1 k
N ... N · ·
is contained in N
i
, the null space of g
i
. Let (g
i
)
'
(
i
1
f )
'

i
i
k 1
(f )
÷
'
be the restrictions of g
i
,
i
1
f , … .
i
i
k 1
f
÷
to the subspace
i
i
k
N . Then
(g
i
)
'
,

(
i
1
f )
'
. … .
i
i
k 1
(f )
÷
' are linear functionals on the vector space
i
i
k
N . Further more if Į
i
is a vector in
i
i
k
N and (
i i
j
(f ( ))' o = 0, j =
1, …, k
i
– 1 then Į
i
is in
i
i i i
1 2 k
N N ... N · · · and so
i i
(g ( ))' o =
0. By induction hypothesis (the case r
i
= k
i
– 1) there are scalars
i
j
c such that
i i
i i i i i
1 1 k 1 k 1
(g ) c (f ) ... c (f )
÷ ÷
' ' ' = + + . Now let h
i
= g
i

i
k 1
i i
j j
j 1
c f
÷
=
¯
. Then h
i
is a linear functional on V
i
and this tells h
i

i
) =
0 for every Į
i
in
i
i
k
N . By the proceeding lemma h
i
is a scalar
multiple of
i
k
f if h
i
=
i i
i i
k k
c f then g
i
=
i
k
i i
j j
j 1
c f
=
¯
.
63
Now the result is true for each i, i = 1, 2, …. n hence the
theorem.

Now we proceed onto define the notion of n-transpose of a n-
linear transformation T = T
n
· … · T
n
of the n-vector spaces V
and W of type II.
Suppose we have two n-vector spaces V = V
1
· V
2
· … ·
V
n
and W = W
1
· W
2
· … · W
n
over the n-field F = F
1
· F
2
·
… · F
n
. Let T = T
1
· … · T
n
be a n-linear transformation from
V into W. Then T induces a n-linear transformation from W*
into V* as follows:

Suppose g = g
1
·... ·g
n
is a n-linear functional on W = W
1
· …
· W
n
and let f(Į) = f
1

1
) · … · f
n

n
) (where Į = Į
1
· Į
2
·
… · Į
n
e V) f(o) = g(TĮ), that is f(Į) = g
1
(T
1
Į
1
) ·...·
g
n
(T
n
Į
n
) ; for each Į
i
e V
i
; i = 1,2, … . n. The above equation
defines a n-function f from V into F i.e. V
1
· V
2
· … · V
n
into
F
1
· … ·F
n
namely the n-composition of T, a n-function from
V into W with g a n-function from W into F = F
1
· F
2
· … ·
F
n
. Since both T and g are n-linear f is also n-linear i.e., f is an
n-linear functional on V. This T provides us with a rule T
t
= T
1
t

· … · T
n
t
which associates with each n-linear functional g on
W = W
1
· … · W
n
a n-linear functional f = T
t
g i.e. f
1
· f
2
· …
· f
n
= T
1
g
1
· … · T
n
g
n
on V = V
1
· V
2
· … · V
n
, i.e., f
i
=
t i
i
T g is a linear functional on V
i
. T
t
= T
1
t
· … · T
n
t
is actually
a n-linear transformation from W
*
=
* *
1 n
W ... W · · into V*=
* *
1 n
V ... V · · for if g
1
, g
2
, are in W i.e., g
1
=
1 n
1 1
g ... g · · and g
2
=
1 n
2 2
g ... g · · in W* and c = c
1
· … · c
n
is a n-scalar
T
t
(cg
1
+ g
2
)(Į) = c(T
t
g
1
)(Į) + (T
t
g
2
)(Į)
i.e.
t 1 1 1 1 t n n n
1 1 2 n 1 2
T (c g g ) ... T (c g g ) + o · · + =
1 t 1 1 t 1 1 n t n n t n n
1 1 1 2 n 1 n 2
(c (T g ) (T g ) ) ... (c (T g ) (T g ) ) o + o · · o + o , so that
T
t
(cg
1
+ g
2
) =
1 2
t t t
1 2 g g
T (cg g ) cT T + = + ; this can be summarized
into the following theorem.

THEOREM 1.2.29: Let V = V
1
· V
2
·... · V
n
and W = W
1
· W
2

· … · W
n
be n-vector spaces over the n-field F = F
1
· F
2

64
·... · F
n
. For each n-linear transformation T = T
1
· … · T
n

from V into W there is unique n-linear transformation T
t
=
1
...
t t
n
T T · · from W* =
* *
1
...
n
W W · · into V* =
* *
1
...
n
V V · ·
such that ( ) ( )
t
g
T g T o o = for every g in W* and Į in V.
We call T

t
=
1
...
t t
n
T T · ·

as the n-transpose of T. This n-
transformation T

t
is often called the n-adjoint of T.

It is interesting to see that the following important theorem.

THEOREM 1.2.20: Let V = V
1
· V
2
·... · V
n
and W = W
1
· W
2

· ... · W
n
be n-vector spaces over the n-field F = F
1
· F
2

· ... · F
n
and let T = T
1
·... · T
n
be a n-linear transformation
from V into W. The n-null space of T

t
=
1
...
t t
n
T T · ·

is the n-
annihilator of the n-range of T. If V and W are finite
dimensional then

(I) n-rank (T

t
) = n-rank T
(II) The n-range of T

t
is the annihilator of the n-null space
of T.

Proof: Let g = g
1
· g
2
· … · g
n
be in W* =
* *
1 n
W ... W · ·
then by definition (T
t
g)Į = g(TĮ) for each Į = Į
1

· …

· Į
n
in
V. The statement that g is in the n-null space of T
t
means that
g(TĮ) = 0 i.e., g
1
T
1
Į
1

· …

· g
n
T
n
Į
n
= (0 · … · 0) for every Į
e
V = V
1
· V
2
· …

· V
n
.
Thus the n-null space of T
t
is precisely the n-annihilator of
n-range of T. Suppose V and W are finite dimensional, say dim
V = (n
1
, n
2
, …, n
n
) and dim W = (m
1
, m
2
, …, m
n
). For (I), let r =
(r
1
, r
2
, …, r
n
) be the n-rank of T i.e. the dimension of the n-range
of T is (r
1
, r
2
, …, r
n
).
By earlier results the n-annihilator of the n-range of T has
dimension (m
1
– r
1
, m
2
– r
2
, …, m
n
– r
n
). By the first statement
of this theorem, the n-nullity of T
t
must be (m
1
– r
1
, …, m
n
– r
n
).
But since T
t
is a n-linear transformation on an (m
1
, …, m
n
)
dimensional n-space, the n-rank of T
t
is (m
1
– (m
1
– r
1
), m
2

(m
2
– r
2
), …, m
n
– (m
n
– r
n
)) and so T and T
t
have the same n-
rank.
65
For (II), let N = N
1
· … · N
n
be the n-null space of T. Every n-
function in the n-range of T
t
is in the n-annihilator of N, for
suppose f = T
t
g i.e. f

1
· … · f

n
= T
t
1
g
1
· … · T
t
n
g
n
for some
g in W* then if Į is in N; f(Į) = f

1

1
) · … · f

n

n
) = (T
t
g)Į =
(T
t
1
g
1

1
· … · (T
t
n
g
n

n
= g(TĮ) = g
1
(T
1
Į
1
) · … · g
n
(T
n
Į
n
) =
g(0) = g
1
(0) · … · g
n
(0) = 0 · 0 · … · 0. Now the n-range of
T
t
is an n-subspace of the space Nº and dim Nº = (n
1
-dim N
1
) ·
(n
2
-dimN
2
) · … · (n
n
-dim N
n
) = n-rank T = n-rank T
t
so that
the n-range of T
t
must be exactly Nº.

THEOREM 1.2.31: Let V = V
1
· V
2
· … · V
n
and W = W
1
·
W
2
· … · W
n
be two (n
1
, n
2
, …, n
n
) and (m
1
, m
2
, …, m
n
)
dimensional n-vector spaces over the n-field F = F
1
· F
2
· …
· F
n
. Let B be an ordered n-basis of V and B* the dual n-basis
of V*. Let C be an ordered n-basis of W with dual n-basis of B*.
Let T = T
1
· T
2
· … · T
n
be a n-linear transformation from V
into W, let A be the n-matrix of T relative to B and C and let B
be a n-matrix of T
t
relative to B*, C*. Then
k k
ij ij
B A = , for k = 1,
2, …, n. i.e.
1 2 n
ij ij ij
A A A · · · ! =
1 2 n
ij ij ij
B B B · · · ! .

Proof: Let B = {
1
1 1 1
1 2 n
, , , o o o ! } · {
2
2 2 2
1 2 n
, , , o o o ! } · … ·
{
n
n n n
1 2 n
, , , o o o ! } be a n-basis of V. The dual n-basis of B, B* =
(
1
1 1 1
1 2 n
f , f , , f ! ) · (
2
2 2 2
1 2 n
f , f , , f ! ) · … · (
n
n n n
1 2 n
f , f , , f ! ). Let C =
(
1
1 1 1
1 2 m
, , , þ þ þ ! ) · (
2
2 2 2
1 2 m
, , , þ þ þ ! ) · … · (
n
n n n
1 2 m
, , , þ þ þ ! ) be an
n-basis of W and C* = (
1
1 1 1
1 2 m
g , g , , g ! ) · (
2
2 2 2
1 2 m
g , g , , g ! ) · …
· (
n
n n n
1 2 m
g , g , , g ! ) be a dual n-basis of C. Now by definition for
Į = Į
1
· … · Į
n

k
m
k k k
k j ij i
i 1
T A
=
o = þ
¯
; j =1, 2, …, n
k
; k =1, 2, …, n
k
n
t k k k
k j ij i
i 1
T g B f
=
=
¯
; j = 1, 2, …, m
k
; k = 1, 2, …, n.

Further
66
t k k
k j i
(T g )( ) o =
k t k
j k i
g (T ) o
k
j
g =
k
m
k k
pi p
p 1
( A )
=
þ
¯

=
k k
m m
k k k
pi j p pi jp
p 1 p 1
A g ( ) A
= =
þ = o
¯ ¯
=
k
ji
A .

For any n-linear functional f = f

1
· … · f

n
on V
k
m
k k k k
i i
i 1
f f ( )f
=
= o
¯
; k = 1, 2, …, n.

If we apply this formula to the functional f

k
= T
t
k
g
k
j
and use the
fact
t k k k
k j i ji
(T g )( ) A , o = we have
k
n
t k k k
k j ji i
i 1
T g A f
=
=
¯
from which it
follows
k k
ij ij
B A = ; true for k = 1, 2, …, n i.e.,
1 2 n
ij ij ij
B B ... B · · ·
=
1 2 n
ij ij ij
A A ... A · · · . If A = A
1
· A
2
· … · A
n
is a (m
1
× n
1
,
m
2
× n
2
, , …, m
n
× n
n
) n-matrix over the n-field F = F
1
· F
2
·
… · F
n
, the n-transpose of A is the (n
1
× m
1
, n
2
× m
2
, …, n
n
×
m
n
) matrix A
t
defined by

1 t 2 t n t
ij ij ij
(A ) (A ) ... (A ) · · · =
1 2 n
ij ij ij
A A ... A · · · .

We leave it for the reader to prove the n-row rank of A is
equal to the n-column rank of A i.e. for each matrix A
i
we have
the column rank of A
i
to be equal to the row rank of A
i
; i = 1, 2,
…, n.

Now we proceed on to define the notion of n-linear algebra over
a n-field of type II.

DEFINITION 1.2.16: Let F = F
1
· F
2
· … · F
n
, be a n-field.
The n-vector space, A = A
1
· A
2
· … · A
n
over the n-field F of
type II is said to be a n-linear algebra over the n-field F if each
A
i
is a linear algebra over F
i
for i = 1, 2, …, n i.e., for Į, ȕ e A
i

we have a vector Įȕ e A
i
, called the product of Į and ȕ in such
a way that
67
a. multiplication is associative Į(ȕȖ) = (Įȕ)Ȗ for Į, ȕ, Ȗ e
A
i
for i = 1, 2, …, n.
b. multiplication is distributive with respect to addition,
Į(ȕ + Ȗ) = Įȕ + ĮȖ and (Į + ȕ)Ȗ = ĮȖ + ȕȖ for Į, ȕ, Ȗ e
A
i
for i = 1, 2, …, n.
c. for each scalar c
i
e F
i
, c
i
(Įȕ) = (c
i
Į)ȕ = Į(c
i
ȕ), true for
i = 1, 2, …, n.

If there is an element 1
n
= 1 · 1 · … · 1

in A such that 1
n
Į =
Į1
n
= Į i.e., 1
n
Į = (1 · 1 · … · 1) (Į
1
· … · Į
n
) = Į
1
· … ·
Į
n
= Į, ĮI
n
= (Į
1
· … · Į
n
) 1
n
= (Į
1
· … · Į
n
) (1 · 1 · … ·
1) = Į
1
· … · Į
n
= Į. We call A an n-linear algebra with n-
identity over the n-field F. If even one of the A
i
’s do not contain
identity then we say A is a n-linear algebra without an n-
identity. 1
n
= 1 · 1 · … · 1 is called the n-identity of A.
The n-algebra A is n-commutative if Įȕ = ȕĮ for all Į, ȕ e
A i.e. if Į = Į
1
· … · Į
n
and ȕ = ȕ
1
· … · ȕ
n
, Įȕ = (Į
1
· … ·
Į
n
) (ȕ
1
· … · ȕ
n
) = Į
1
ȕ
1
· … · Į
n
ȕ
n
. ȕĮ = (ȕ
1
· … · ȕ
n
) (Į
1

· … · Į
n
) = ȕ
1
Į
1
· … · ȕ
n
Į
n
. If each Į
i
ȕ
i
= ȕ
i
Į
i
for i = 1, 2,
…, n then we say Įȕ = ȕĮ for every o, þ e A. We call A in
which oþ = ȕĮ to be an n-commutative n-linear algebra. Even if
one A
i
in A is non commutative we don’t call A to be an n-
commutative n-linear algebra.

1. All n-linear algebras over the n-field F are n-vector
spaces over the n-field F.

2. Every n-linear algebra over an n-field F need not be a n-
commutative n-linear algebra over F.

3. Every n-linear algebra A over an n-field F need not be a
n-linear algebra with n-identity 1
n
in A.

Now we proceed on to define the notion of n-polynomial over
the n-field F = F
1
· F
2
· … · F
n
.

DEFINITION 1.2.17: Let F[x] = F
1
[x] · F
2
[x] · … · F
n
[x] be
such that each F
i
[x] is a polynomial over F
i
, i = 1, 2, …, n,
68
where F = F
1
· F
2
· … · F
n
is the n-field. We call F[x] the n-
polynomial over the n-field F = F
1
· … · F
n
.

Any element p(x)
e F[x] will be of the form p(x) = p
1
(x) · p
2
(x) · … · p
n
(x)
where p
i
(x) is a polynomial in F
i
[x] i.e., p
i
(x) is a polynomial in
the variable x with coefficients from F
i
; i = 1, 2, …, n.
The n-degree of p(x) is a n-tuple given by (n
1
, n
2
, …, n
n
)
where n
i
is the degree of the polynomial p
i
(x); i = 1, 2, …, n.

We illustrate the n-polynomial over the n-field by an example.

Example 1.2.8: Let F = Z
2
· Z
3
· Z
5
· Q · Z
11
be a 5-field;
F[x] = Z
2
[x] · Z
3
[x] · Z
5
[x] · Q[x] · Z
11
[x] is a 5-polynomial
vector space over the 5-field F. p(x) = x
2
+ x + 1 · x
3
+ 2x
2
+1 ·
4x
3
+ 2x+1 · 81x
7
+ 50x
2
– 3x + 1 · 10x
3
+ 9x
2
+ 7x + 1 e F[x].

DEFINITION 1.2.18: Let F[x] = F
1
[x] · F
2
[x] · … · F
n
[x] be
a n-polynomial over the n-field F = F
1
· F
2
· … · F
n
. F[x] is
a n-linear algebra over the n-field F. Infact F[x] is a n-
commutative n-algebra over the n-field F. F[x] the n-linear
algebra has the n-identity, 1
n
= 1 · 1 · … · 1. We call a n-
polynomial p(x) = p
1
(x) · p
2
(x) · … · p
n
(x) to be a n-monic n-
polynomial if each p
i
(x) is a monic polynomial in x for i = 1, 2,
…, n.

The reader is expected to prove the following theorem

THEOREM 1.2.32: Let F[x] = F
1
[x] · F
2
[x] · … · F
n
[x] be a
n-linear algebra of n-polynomials over the n-field F = F
1
· F
2

· … · F
n
. Then

a. For f and g two non zero n-polynomials in F[x] where
f(x) = f
1
(x) · f
2
(x) · … · f
n
(x) and g(x) = g
1
(x) · g
2
(x)
· … · g
n
(x) the polynomial fg = f
1
(x)g
1
(x) · f
2
(x)g
2
(x)
· … · f
n
(x)g
n
(x) is a non zero n-polynomial of F[x]

b. n-deg (fg) = n-deg f + n-deg g where n-deg f = (n
1
, n
2
,
…, n
n
) and n-deg g = (m
1
, m
2
, …, m
n
)

69
c. fg is n-monic if both f and g are n-monic

d. fg is a n-scalar n-polynomial if and only if f and g are
scalar polynomials

e. If f + g  0, n-deg (f + g)  max (n-deg f, n-deg g)

f. If f, g, h are n-polynomials over the n-field F = F
1
· F
2

· … · F
n
such that f = f
1
(x) · f
2
(x) · … · f
n
(x)  0 ·
0 · … · 0 and fg = fh then g = h, where g(x) = g
1
(x) ·
g
2
(x) · … · g
n
(x) and h(x) = h
1
(x) · h
2
(x) · … ·
h
n
(x).

As in case of usual polynomials we see in case of n-polynomials
the following. Let A = A
1
· A
2
· … · A
n
be a n-linear algebra
with identity 1
n
= 1 · 1 · … · 1 over the n-field F = F
1
· F
2
·
… · F
n
, where we make the convention for any Į = Į
1
· Į
2
·
… · Į
n
e A; Į
º
= Į
º
1
· Į
º
2
· … · Į
º
n
= 1 · 1 · … · 1 = 1
n

for each Į e A.
Now to each n-polynomial f(x) e F[x] = F
1
[x] · F
2
[x] · …
· F
n
[x] over the n-field F = F
1
· F
2
· … · F
n
and Į = Į
1
· Į
2

· … · Į
n
in A we can associate an element
F(Į) =
1 2 n
n n n
1 1 2 2 n n
i i i i i i
i 0 i 0 i 0
f f f
= = =
o · o · · o
¯ ¯ ¯
! ;
k
i k
f F e ;
for k = 1, 2, …, n; 1  i  n
k
.

We can in view of this prove the following theorem.

THEOREM 1.2.33: Let F = F
1
· F
2
· … · F
n
be a n-field and A
= A
1
· A
2
· …· A
n
be a n-linear algebra with identity 1
n
= 1
· 1· … · 1 over the n-field F.
Suppose f(x) and g(x) be n-polynomials in F[x] = F
1
[x] ·
F
2
[x]· … · F
n
[x] over the n-field F = F
1
· F
2
· … · F
n
and
that Į = Į
1
· … · Į
n
e A set by the rule

70
f(o) =
1 2
1 2
1 2
0 0 0
( ) ( ) ( )
n
n n n
i i n i
i i i n
i i i
f f f o o o
= = =
· · ·
¯ ¯ ¯
!

and for c = c
1
· c
2
· … · c
n
e F = F
1
· F
2
· … · F
n
, we have
(cf + g)Į = cf(Į) + g(Į); (fg)(Į) = f(Į)g(Į).

Proof:
f(x) =
1 n
m m
1 i n i
i i
i 0 i 0
f x f x
= =
· ·
¯ ¯
!
and
g(x) =
1 n
n n
1 j n j
j j
j 0 j 0
g x g x
= =
· ·
¯ ¯
!
fg =
1 1 i j 2 2 i j n n i j
i j i j i j
i, j i, j i, j
f g x f g x f g x
+ + +
· · ·
¯ ¯ ¯
!
and hence

(fg)Į =
1 1 i j 2 2 i j n n i j
i j 1 i j 2 i j n
i, j i, j i, j
f g f g f g
+ + +
o · o · · o
¯ ¯ ¯
!

(Į = Į
1
· Į
2
· … · Į
n
e A, Į
i
e A
i
; for i = 1, 2, …, n)

=
1 1 2 2
m n m n
1 i 1 j 2 i 2 j
i 1 j 1 i 2 j 2
i 0 j 0 i 0 j 0
f g f g
= = = =
| | | | | | | |
o o · o o · ·
| | | |
\ . \ . \ . \ .
¯ ¯ ¯ ¯
!
n n
m n
n i n j
i n j n
i 0 j 0
f g
= =
| | | |
o o
| |
\ .\ .
¯ ¯

= f(Į)g(Į)
= f
1

1
)g
1

1
) · f
2

2
)g
2

2
) · … · f
n

n
)g
n

n
).

Now we define the Lagranges n-interpolation formula.
Let F = F
1
· F
2
· … · F
n
be a n-field and let
1
1 1 1
0 1 n
t , t , , t !
be n
1
+ 1 distinct elements of F
1
,
2
2 2 2
0 1 n
t , t , , t ! are n
2
+ 1 distinct
elements of F
2
, …,
n
n n n
0 1 n
t , t , , t ! are the n
n
+ 1 distinct elements
of F
n
.
71
Let V = V
1
· V
2
· … · V
n
be a n-subspace of F[x] = F
1
[x]
· F
2
[x] · … · F
n
[x] consisting of all n-polynomial of n-degree
less than or equal to (n
1
, …, n
n
) together with the n-zero
polynomial and let L =
1 2 n
i i i
L , L , , L ! be the n-function from V =
V
1
· V
2
· … · V
n
into F = F
1
· F
2
· … · F
n
defined by L
i
(f)
=
1 n
i i 1 n
1 n
L (f ) L (f ) · · ! =
1 n
1 1 n n
i i
f (t ) f (t ) · · ! , 0  i
1
 n
1
, 0  i
2

 n
2
, …, 0  i
n
 n
n
.
By the property (cf + g)(Į) = cf(Į) + g(Į) i.e., if c = c
1
· …
· c
n
, f = f
1
· … · f
n
; g = g
1
· … · g
n
and Į = Į
1
· … · Į
n

then c
1
f
1

1
) · c
2
f
2

2
) · … · c
n
f
n

n
) + g
1

1
) · g
2

2
) · … ·
g
n

n
) = (c
1
f
1
+ g
1
)(Į
1
) · … · (c
n
f
n
+ g
n
)(Į
n
) each L
i
is a n-linear
functional on V and one of the things we intend to show is that
the set consisting of L
0
, L
1
, …, L
n
is a basis for V* the dual
space of V. L
0
=
0 1 0 n
1 n
L (f ) ... L (f ) · · = f
1
(
1
0
t ) · … · f
n
(
n
0
t )
and so on we know from earlier results (L
0
, L
1
, …, L
n
) that is
{
1
n 0
1 1
L , , L ! } · {
2
n 0
2 2
L , , L ! } · … · {
n
n 0
n n
L , , L ! } is the dual
basis of {
1
n 0
1 1
P , , P ! } · {
2
n 0
2 2
P , , P ! } · … · {
n
n 0
n n
P , , P ! } of
V. There is at most one such n-basis and if it exists is
characterized by
0 0 1 1 2 2 n n
0 0 1 1 2 2 n n
j i j i j i j i
L (P ) L (P ) L (P ) ... L (P ) · · · · =
0 0 1 1 n n 0 0 1 i n n
0 0 1 1 n n
i j i j i j i j i j i j
P (t ) P (t ) P (t ) · · · = o ·o · ·o ! ! . The n-
polynomials P
i
=
1 n
1 n
i i
P ... P · ·
=
1 1 1
1 1 1 1 1 1 1
1 1 1 1
0 i 1 i 1 n
1 1 1 1 1 1 1 1
i 0 i i 1 i i 1 i n
(x t )...(x t )(x t )...(x t )
(t t )...(t t )(t t )...(t t )
÷ +
÷ +
÷ ÷ ÷ ÷
÷ ÷ ÷ ÷

· … ·
n n n
n n n n n n n
n n n n
0 i 1 i 1 n
n n n n n n n n
i 0 i i 1 i i 1 i n
(x t )...(x t )(x t )...(x t )
(t t )...(t t )(t t )...(t t )
÷ +
÷ +
÷ ÷ ÷ ÷
÷ ÷ ÷ ÷

=
1 2 n
1 1 2 2 n n
1 1 2 2 n n
1 2 n
j j j
1 1 2 2 n n
j i j i j i
i j i j i j
x t x t x t
t t t t t t
= = =
| | | | | | ÷ ÷ ÷
· · · | | |
| | |
÷ ÷ ÷
\ . \ . \ .
I I I
!

are of degree (n
1
, n
2
, …, n
n
), hence belongs to V = V
1
· … ·
V
n
. If
f = f
1
· … · f
n

72
=
1
1 1
1
n
1 1
i i
i 1
c P
=
¯
·
2
2 2
2
n
2 2
i i
i 1
c P
=
¯
· … ·
n
n n
n
n
n n
i i
i 1
c P
=
¯

then for each j
k
we have k=1, 2, …, n; 1  j
k
 n
k
.
k k k k k
k
k k k k k k
j i i j j
i
f (t ) c P (t ) c = =
¯

true for each k, k = 1, 2, …, n.
Since the 0-polynomial has the properly 0(t
i
) = 0 for each t
= t
1
· … · t
n
e F
1
· … · F
n
it follows from the above relation
that the n-polynomials {
1
1 1 1
0 1 n
P , P , , P ! } · {
2
2 2 2
0 1 n
P , P , , P ! } · …
· {
n
n n n
0 1 n
P , P , , P ! } are n-linearly independent. The polynomials
{1, x, …,
1
n
x } · {1, x, …,
2
n
x } · … · {1, x, …,
n
n
x } form a
n-basis of V and hence the dimension of V is {n
1
+ 1, n
2
+ 1, …,
n
n
+ 1}. So the n-independent set {
1
1 1
0 n
P , ..., P } · {
2
2 2
0 n
P , , P ! }
· … · {
n
n n
0 n
P , , P ! } must form an n-basis for V. Thus for each
f in V
f =
1
1 1
1
n
1 1 1
i i
i 0
f (t )P
=
¯
· … ·
n
n n
n
n
n n n
i i
i 0
f (t )P
=
¯
(I)

I is called the Lagranges’ n-interpolation formula. Setting f
k
=
k
j
x in I we obtain
1 n
1 k 1 n
1 1 n n
1 n
n n
j j 1 1 n n
i i i i
i 0 i 0
x ... x (t ) P ... (t ) P
o o
= =
· · = · ·
¯ ¯
.
Thus the n-matrix

T=
1 n
1 n
1 n
1 1 1 n n n
n n 1 1 2 1 n n 2 n
0 0 0 0 0 0
n n 1 1 2 1 n n 2 n
1 1 1 1 1 1
n n 1 1 2 1 n n 2 n
n n n n n n
1 t (t ) (t ) 1 t (t ) (t )
1 t (t ) (t ) 1 t (t ) (t )
1 t (t ) (t ) 1 t (t ) (t )

· ·

! !
! !
!
# # # # # # # #
! !
(II)

= T
1
· … · T
n
is n-invertible. The n-matrix in II is called a
Vandermonde n-matrix.

73
It can also be shown directly that the n-matrix is n-invertible
when
1
1 1 1
0 1 n
(t , t , , t ) ! ·
2
2 2 2
0 1 n
(t , t ,..., t ) · … ·
n
n n n
0 1 n
(t , t , , t ) ! are
{n
1
+ 1, n
2
+ 1, …, n
n
+ 1} set of n-distinct elements from the n-
field F = F
1
· … · F
n
i.e., each
i
i i i
0 1 n
(t , t , , t ) ! is the n
i
+ 1
distinct elements of F
i
; true for i = 1, 2, …, n.
Now we proceed on to define the new notion of n-
polynomial function of an n-polynomial over the n-field F = F
1

· F
2
· … · F
n
. If f = f
1
· … · f
n
be any n-polynomial over the
n-field F we shall in the discussion denote by
~ ~ ~ ~
1 2 n
f f f ... f = · · · the n-polynomial n-function from F into
F taking each t = t
1
· … · t
n
in F into f(t) where t
i
e F
i
for i = 1,
2, …, n. We see every polynomial function arises in this way, so
analogously every n-polynomial function arises in the same
way, however if
~ ~
f g = and f = f
1
· … · f
n
and g = g
1
· g
2
·
… · g
n
then
~ ~ ~ ~ ~ ~
1 1 2 2 n n
f g , f g ,..., f g = = = for any two equal n-
polynomial f and g. So we assume two n-polynomials f and g
such that f  g. However this situation occurs only when the n-
field F = F
1
· F
2
· … · F
n
is such that each field F
i
in F has
only finite number of elements in it. Suppose f and g are n-
polynomials over the n-field F then the product of
~
f and
~
g is
the n-function
~ ~
f g from F into F given by
~ ~
f g (t) =
~
f (t)
~
g (t) for every t = t
1
· … · t
n
e F
1
· F
2
· … · F
n
.
Further (fg)(x) = f(x)g(x) hence (
~ ~
f g )(x) =
~
f (x)
~
g (x) for
each x = x
1
· … · x
n
e F
1
· F
2
· … · F
n
.
Thus we see
( )
~ ~
f g = ( )
~
fg and it is also an n-polynomial
function. We see that the n-polynomial function over the n-field
is in fact an n-linear algebra over the n-field F. We shall denote
this n-linear algebra over the n-field F by A
~
=
~ ~
1 n
A ... A · · .

DEFINITION 1.2.19: Let F = F
1
· … · F
n
be a n-field. A = A
1

· A
2
· … · A
n
be a n-linear algebra over the n-field F. Let A
~

be the n-linear algebra of n-polynomial functions over the same
field F. The n-linear algebras A and A
~
are said to be n-
isomorphic if there is a one to one n-mapping Į ĺ Į
~
such that
74
(cĮ + dȕ)
~
= cĮ
~
+ dȕ
~
and (Įȕ)
~
= Į
~
ȕ
~
for all Į, ȕ e A and n-
scalar c, d in the n-field F. Here c = c
1
· … · c
n
and d = d
1
·
… · d
n
, Į = Į
1
· … · Į
n
and ȕ = ȕ
1
· … · ȕ
n
then (cĮ + dȕ)
~

= (c
1
Į
1
+ d
1
ȕ
1
)
~
· (c
2
Į
2
+ d
2
ȕ
2
)
~
· … · (c
n
Į
n
+ d
n
ȕ
n
)
~
= (c
1
Į
~
1
+ d
1
ȕ
~
1
) · (c
2
Į
~
2
+ d
2
ȕ
~
2
) · … · (c
n
Į
~
n
+ d
n
ȕ
~
n
) here (c
i
Į
i
+
d
i
ȕ
i
)
~
= c
i
Į
~
i
+ d
i
ȕ
~
i
where Į
i
, ȕ
i
e V
i
, the component vector
space of the n-vector space and Į
i
, d
i
e F
i
; the component field
of the n-field F, true for i = 1, 2, …, n. Further (Įȕ)
~
= (Į
1
ȕ
1
)
~
·
… · (Į
n
ȕ
n
)
~
= Į
~
1
ȕ
~
1
· … · Į
~
n
ȕ
~
n
. The n-mapping Į ĺ Į
~
is
called an n-isomorphism of A onto A
~
. An n-isomorphism of A
onto A
~
is thus an n-vector space n-isomorphism of A onto A
~

which has the additional property of preserving products.

We indicate the proof of the important theorem

THEOREM 1.2.34: Let F = F
1
· F
2
· … · F
n
be a n-field
containing an infinite number of distinct elements, the n-
mapping f ĺ f
~
is an n-isomorphism of the n-algebra of n-
polynomials over the n-field F onto the n-algebra of n-
polynomial functions over F.

Proof: By definition the n-mapping is onto, and if f = f
1
· … ·
f
n
and g = g
1
· … · g
n
belong to F[x] = F
1
[x] · … · F
n
[x] the
n-algebra of n-polynomial functions over the field F; i.e.
(cf+dg)
~
= cf
~
+ dg
~
where c = c
1
, …, c
n
and d = d
1
, …, d
n
e F =
F
1
· … · F
n
; i.e.,

(cf + dg)
~
=
( )
~ ~
1 1 1 1
c f d g +
·
( )
~ ~
2 2 2 2
c f d g +
· … ·
( )
~ ~
n n n n
c f d g +

for all n-scalars c, d e F. Since we have already shown that (fg)
~

= f
~
g
~
we need only show the n-mapping is one to one. To do
this it suffices by linearity of the n-algebras (i.e., each linear
algebra is linear) f
~
= 0 implies f = 0. Suppose then that f = f
1
·
f
2
· … · f
n
is a n-polynomial of degree (n
1
, n
2
, …, n
n
) or less
such that f

1
= 0 ; i.e., f

1
=
1 1 1
1 2 n
f f f · · · ! = (0 · 0 · … · 0).
Let {
1
0
t ,
1
1
t ,…,
1
1
n
t } · {
2
0
t ,
2
1
t , …,
2
2
n
t } ·…· {
n
0
t ,
n
1
t ,…,
n
n
n
t }
be any {n
1
+ 1, n
2
+ 1, …, n
n
+ 1}; n-distinct elements of the n-
75
field F. Since f
~
= 0, f
k
(t
i
) = 0 for k = 1, 2, …, n; i = 0, 1, 2, …,
n
k
and it is implies f = 0.
Now we proceed on to define the new notion of n-
polynomial n-ideal or we can say as n-polynomial ideals.
Throughout this section F[x] = F
1
[x] · … · F
n
[x] will denote a
n-polynomial over the n-field F = F
1
· … · F
n
.

We will first prove a simple lemma.

LEMMA 1.2.3: Suppose f = f
1
· … · f
n
and d = d
1
· … · d
n

are any two non zero n-polynomials over the n-field F such that
n-deg d  n-deg f, i.e., n-deg d = (n
1
, …, n
n
) and n-deg f = (m
1
,
…, m
n
), n-deg d  n-deg f if and only if each n
i
 m
i
, i = 1, 2, …,
n. Then there exists an n-polynomial g = g
1
· g
2
· … · g
n
in
F[x] = F
1
[x] · … · F
n
[x] such that either f – dg = 0 or n-deg
(f – dg) < n-deg f.

Proof: Suppose f = f
1
· … · f
n

=
1 2
1 1 2 2
1 1 2 2
1 2
m 1 m 1
m i m i 1 1 2 2
m i m i
i 0 i 0
(a x a x ) (a x a x )
÷ ÷
= =
+ · +
¯ ¯

n
n n
n n
n
m 1
m i n n
m i
i 0
(a x a x )
÷
=
· · +
¯
! ;
1 2 n
1 2 n
m m m
(a , a ,..., a )  (0, …, 0) i.e., each
i
i
m
a  0 for i = 1, 2, …, n.
d = d
1
· … · d
n
=
1
1 1
1 1
1
n 1
n i 1 1
n i
i 0
b x b x
÷
=
| |
+
|
|
\ .
¯
·
2
2 2
2 2
2
n 1
n i 2 2
n i
i 0
b x b x
÷
=
| |
+
|
|
\ .
¯
· … ·
n
n n
n n
n
n 1
n i n n
n i
i 0
b x b x
÷
=
| |
+
|
|
\ .
¯
;

with
1 n
1 n
n n
(b , , b ) !  (0, 0, …, 0) i.e.,
i
i
n
b  0, i = 1, 2, …, n.
Then (m
1
, m
2
, …, m
n
) > (n
1
, n
2
, …, n
n
) and

1
1 1
1
1
m
m n
1 1
1
n
a
f x d
b
÷
| |
| |
| ÷ · |
|
|
\ .
\ .

2
2 2
2
2
m
m n
2 2
2
n
a
f x d
b
÷
| |
| |
| ÷ · |
|
|
\ .
\ .
… ·
76

n
n n
n
n
m
m n
n n
n
n
a
f x d
b
÷
| |
| |
| ÷ |
|
|
\ .
\ .
= 0 ·…· 0,
or deg
1
1 1
1
1
m
m n
1 1
1
n
a
f x d
b
÷
| |
| |
| ÷ · |
|
|
\ .
\ .
deg
2
2 2
2
2
m
m n
2 2
2
n
a
f x d
b
÷
| |
| |
| ÷ · |
|
|
\ .
\ .
… ·
n
n n
n
n
m
m n
n n
n
n
a
deg f x d
b
÷
| |
| |
| ÷ |
|
|
\ .
\ .
< degf
1
· degf
2
·…· deg f
n
.

Thus we make take
g =
1 2 n
1 1 2 2 n n
1 2 n
1 2 n
m m m
m n m n m n
1 2 n
n n n
a a a
x x ... x
b b b
÷ ÷ ÷
| | | | | |
· · · | | |
| | |
\ . \ . \ .
.

Using this lemma we illustrate the usual process of long division
of n-polynomials over a n-field F = F
1
·
F
2
·

·
F
n
.

THEOREM 1.2.35: Let f = f
1
· … · f
n
, d = d
1
· … · d
n
be n-
polynomials over the n-field F = F
1
· F
2
· … · F
n
and d = d
1

· … · d
n
is different from 0 · … · 0; then there exists n-
polynomials q = q
1
· q
2
· … · q
n
and r = r
1
· … · r
n
in F[x]
such that

1. f = dq + r; i.e., f = f
1
· f
2
· … · f
n
= (d
1
q
1
+ r
1
) · …
· (d
n
q
n
+ r
n
).
2. Either r = r
1
· … · r
n
= (0 · … · 0) or n-deg r < n-
deg d.

The n-polynomials q and r satisfying conditions (1) and (2) are
unique.

Proof: If f = 0 · … · 0 or n-deg f < n-deg d we make take q =
q
1
· … · q
n
= (0 · … · 0) and r
1
· … · r
n
= f
1
· f
2
· … ·
f
n
. In case f = f
1
· … · f
n
 0 · … · 0 and n-deg f  n-deg d,
then the preceding lemma shows we may choose a n-polynomial
77
g = g
1
· g
2
· … · g
n
such that f – dg = 0 · … · 0, i.e. (f
1

d
1
g
1
) · (f
2
– d
2
g
2
) · … · (f
n
– d
n
g
n
) = 0 · … · 0, or n-deg(f –
dg) < n-deg f. If f – dg  0 and n-deg (f – dg) > n-deg d we
choose a n-polynomial h such that (f – dg) – dh = 0 or [(f
1

d
1
g
1
) – d
1
h
1
] · [(f
2
– d
2
g
2
) – d
2
h
2
] · … · [(f
n
– d
n
g
n
) – d
n
h
n
]

= 0
· … · 0 or n-deg[f – d(g + h)] < nd(f – dg) i.e., deg(f
1
– d
1
(g
1
+
h
1
)) · deg(f
2
– d
2
(g
2
+ h
2
)) · … · deg(f
n
– d
n
(g
n
+h
n
)) < deg(f
1

d
1
g
1
) · deg(f
2
– d
2
g
2
) · … · deg(f
n
– d
n
g
n
).
Continuing this process as long as necessary we ultimately
obtain n-polynomials q = q
1
· q
2
· … · q
n
and r = r
1
· … · r
n

such that r = 0 · … · 0 or n-deg r < n-deg d i.e. deg r
1
· … ·
deg r
n
< deg d
1
· deg d
2
· … · deg d
n
, and f = dq + r, i.e., f
1
·
f
2
· … · f
n
= (d
1
q
1
+ r
1
) · (d
2
q
2
+ r
2
) · … · (d
n
q
n
+ r
n
).
Now suppose we also have f = dq
1
+ r
1
, i.e., f
1
· … · f
n
=
1 1 1 1
1 1 1 n n n
(d q r ) ... (d q r ) + · · + where r
1
= 0 · … · 0 or n-deg r
1
<
n-deg d; i.e.
1 1
1 n
degr ... degr · · < deg d
1
· … · deg d
n
. Then
dq + r = dq
1
+ r
1
and d(q – q
1
) = r
1
– r if q

– q
1
 0 · … · 0 then
d(q – q
1
)  0 · … · 0. n-deg d + n-deg(q – q
1
) = n-deg (r
1
– r);
i.e. (deg d
1
· deg d
2
· … · deg d
n
) + deg(q
1
– q
1
1
) · … ·
deg(q
n
– q
1
n
) = deg (r
1
1

– r
1
) · … · deg (r
1
n

– r
n
).
But as n-degree of r
1
– r is less than the n-degree of d this is
impossible so q

– q
1
= 0
·

·
0 hence r
1
– r = 0
·

·
0.

DEFINITION 1.2.20: Let d = d
1
· d
2
· … · d
n
be a non zero n-
polynomial over the n-field F = F
1
· … · F
n
. If f = f
1
· … · f
n

is in F[x] = F
1
[x] · … · F
n
[x], the proceeding theorem shows
that there is at most one n-polynomial q = q
1
· … · q
n
in F[x]
such that f = dq i.e., f
1
· … · f
n
= d
1
q
1
· … · d
n
q
n
. If such a q
exists we say that d = d
1
· … · d
n
, n-divides f = f
1
· … · f
n

that f is n-divisible by d and f is a n-multiple of d and we call q
the n-quotient of f and d and write q = f/d i.e. q
1
· … · q
n
=
f
1
/d
1
· … · f
n
/d
n
.

The following corollary is direct.

COROLLARY 1.2.8: If f = f
1
· … · f
n
is a n-polynomial over
the n-field F = F
1
· F
2
· … · F
n
and let c = c
1
· c
2
· … · c
n

78
be an element of F. Then f is n-divisible by x – c = (x
1
– c
1
) · …
· (x
n
– c
n
) if and only if f(c) = 0 · … · 0 i.e., f
1
(c
1
) · f
2
(c
2
) ·
… · f
n
(c
n
) = 0 · 0 · … · 0.

Proof: By theorem f = (x–c)q + r where r is a n-scalar
polynomial i.e., r = r
1
· … · r
n
e F. By a theorem proved
earlier we have f(c) = f
1
(c
1
) · … · f
n
(c
n
) = [0q
1
(c
1
) + r
1
(c
1
)] ·
[0q
2
(c
2
) + r
2
(c
2
)] · … · [0q
n
(c
n
) + r
n
(c
n
)] = r
1
(c
1
) · … · r
n
(c
n
).
Hence r = 0 · … · 0 if and only if f(c) = f
1
(c
1
) · … · f
n
(c
n
) =
0 · … · 0.

We know if F = F
1
· … · F
n
is a n-field. An element c = c
1
·
c
2
· … · c
n
in F is said to be a n-root or a n-zero of a given n-
polynomial f = f
1
· … · f
n
over F if f(c) = 0 · … · 0, i.e.,
f
1
(c
1
) · … · f
n
(c
n
) = 0 · … · 0.

COROLLARY 1.2.9: A n-polynomial f = f
1
· f
2
· … · f
n
of
degree (n
1
, n
2
, …, n
n
) over a n-field F = F
1
· F
2
· … · F
n
has
at most (n
1
, n
2
, …, n
n
) roots in F.

Proof: The result is true for n-polynomial of n-degree (1, 1, …,
1). We assume it to be true for n-polynomials of n-degree (n
1

1, n
2
– 1, …, n
n
– 1). If a = a
1
· a
2
· … · a
n
is a n-root of f = f
1

· … · f
n
, f = (x – a)q = (x
1
– a
1
)q
1
· … · (x
n
– a
n
)q
n
where q =
q
1
· … · q
n
has degree n – 1. Since f(b) = 0, i.e., f
1
(b
1
) · … ·
f
n
(b
n
) = 0 · … · 0 if and only if a = b or q(b) = q
1
(b
1
) · … ·
q
n
(b
n
) = 0 · … · 0; it follows by our inductive assumption that
f must have (n
1
, n
2
, …, n
n
) roots.
We now define the n-derivative of a n-polynomial.
Let
f(x) = f
1
(x) · … · f
n
(x)
=
1
1
n 1 1 1
0 1 n
c c x ... c x + + + · … ·
n
n
n n n n
0 1 n
c c x ... c x + + +
is a n-polynomial
f

' =
1 2 n
f f ... f ' ' ' · · ·
=
( )
1
1
n 1 1 1 1
1 2 1 n
c 2c x n c x
÷
+ + + ! · … ·
( )
n
n
n 1 n n n
1 2 n n
c 2c x cn c x
÷
+ + + ! .
79
Now we will use the notation
Df = f' = Df
1
· …· Df
n
=
1 2 n
f f ... f ' ' ' · · · .
D
2
f = fª =
1 2 n
f f ... f '' '' '' · · ·
and so on. Thus
k k k
f 1 n
D f ... f = · · .
Now we will prove the Taylor’s formula for n-polynomials
over the n-field F.

THEOREM 1.2.36: Let F = F
1
· … · F
n
be a n-field of n-
characteristic (0, …, 0); c = c
1
· c
2
· … · c
n
be an element in
F = F
1
· … · F
n
, and (n
1
, n
2
, …, n
n
) a n-tuple of positive
integers. If f = f
1
· … · f
n
is a n-polynomial over f with n-deg f
 (n
1
, n
2
, …, n
n
) then
f =
1 1
1
1
1
1 1
0
1
( )
=
÷
¯
k n
k
k
D f
c x c
k
·
2 2
2
2
2
2 2
0 0
2
( ) ... ( )
= =
÷ · · ÷
¯ ¯
n n
n
n
k k n n
k k n
n n
k k
n
D f D f
c x c c x c
k k
.

Proof: We know Taylor’s theorem is a consequence of the
binomial theorem and the linearity of the operators D
1
, D
2
, …,
D
n
. We know the binomial theorem (a + b)
m
=
m
m k k
k 0
m
a b
k
÷
=
| |
|
\ .
¯
,
where
m
k
| |
|
\ .
=
m! m(m 1)...(m k 1)
k!(m k)! 1.2...k
÷ ÷ +
=
÷
is the familiar
binomial coefficient giving the number of combinations of m
objects taken k at a time.

Now we apply binomial theorem to the n-tuple of polynomials
1
m
x · … ·
n
m
x =
1 2 n
m m m
1 1 2 2 n n
(c (x c )) (c (x c )) ... (c (x c )) + ÷ · + ÷ · · + ÷
=
1
1 1 1
1
m
1 m k k
1 1
k 0
1
m
c (x c )
k
÷
=
| |
÷
|
\ .
¯
·…·
n
n n n
n
m
n m k k
n n
k 0
n
m
c (x c )
k
÷
=
| |
÷
|
\ .
¯

80
=
1 1 1
m m 1 m
1 1 1 1 1
{c m c (x c ) ... (x c ) }
÷
+ ÷ + + ÷ ·
2 2 2
m m 1 m
2 2 2 2 2
{c m c (x c ) ... (x c ) }
÷
+ ÷ + + ÷ · … ·
n n n
m m 1 m
n n n n n
{c m c (x c ) ... (x c ) }
÷
+ ÷ + + ÷

and this is the statement of Taylor’s n-formula for the case
f =
1 n
m m
x ... x · · .
If
f =
1 2 n
1 2 n
1 2 n
1 2 n
n n n
m m m 1 2 n
m m m
m 0 m 0 m 0
a x a x ... a x
= = =
· · ·
¯ ¯ ¯

1 2
1 1 2 2
1 2
1 2
n n
k m k m k 1 2
f m 1 m 2
m 0 m 0
D (c) a D x (c ) a D x (c ) ...
= =
= · · ·
¯ ¯
n
n n
n
n
n
k m n
m n
m 0
a D x (c )
=
¯

and
1 1 1
1
k k m
1 1 1
k 0
1
D f (c )(x c )
k
=
÷
¯
·
2 2 2
2
k k m
2 2 2
k 0
2
D f (c )(x c )
k
=
÷
¯
· … ·
n n n
n
k k m
n n n
k 0
n
D f (c )(x c )
k
=
÷
¯

=
1 1 1
1
1 1
k m k
1
1 1
m
k m
1
D x (c )(x c )
a
k
÷
¯¯
·
2 2 2
2
2 2
k m k
2 2 2
m
k m
2
D x (c )(x c )
a
k
÷
¯¯
·…·
n n n
n
n n
k m k
n
n n
m
k m
n
D x (c )(x c )
a
k
÷
¯¯

=
1 1 1
1
1 1
k m k
1
1 1
m
m k
1
D x (c )(x c )
a
k
÷
¯ ¯
·
2 2 2
2
2 2
k m k
2
2 2
m
m k
2
D x (c )(x c )
a
k
÷
¯ ¯
· … ·
81

n n n
n
n n
k m k
n
n n
m
m k
n
D x (c )(x c )
a
k
÷
¯ ¯

= f
1
· … · f
n
.

If c = c
1
· … · c
n
is a n-root of the n-polynomial f = f
1
· … ·
f
n
the n-multiplicity of c = c
1
· … · c
n
as a n-root of f = f
1
· …
· f
n
is the largest n-positive integer (r
1
, r
2
, …, r
n
) such that
1 n
r r
1 n
(x c ) ... (x c ) ÷ · · ÷ n-divides f = f
1
· f
2
· … · f
n
.

THEOREM 1.2.37: Let F = F
1
· … · F
n
be a n-field of (0, …,
0) characteristic (i.e., each F
i
is of characteristic 0) for i = 1, 2,
…, n and f = f
1
· … · f
n
be a n-polynomial over the n-field F
with n-deg f  (n
1
, n
2
, …, n
n
). Then the n-scalar c = c
1
· … · c
n

is a n-root of f of multiplicity (r
1
, r
2
, …, r
n
) if and only if
(
1
k
D f
1
)(c
1
) · (
2
k
D f
2
)(c
2
) · … · (
n
k
D f
n
)(c
n
) = 0 · 0 · … · 0;
0  k
i
 r
i
– 1; i = 1, 2, …., n. ( )
i
r
i
D f (c
i
)  0, for every, i = 1, 2,
…, n.

Proof: Suppose that (r
1
, r
2
, … , r
n
) is the n-multiplicity of c = c
1

· c
2
· … · c
n
as a n-root of f = f
1
· … · f
n
. Then there is a n-
polynomial g = g
1
·…· g
n
such that
f =
1 n
r r
1 1 n n
(x c ) g ... (x c ) g ÷ · · ÷
and g(c) = g
1
(c
1
) · … · g
n
(c
n
)  0 · … · 0. For otherwise f =
f
1
· … · f
n
would be divisible by
1 n
r 1 r 1
1 n
(x c ) ... (x c )
+ +
÷ · · ÷ .
By Taylor’s n-formula applied to g = g
1
· … · g
n
.

f =
1 1 1 1
1
1
m m n r
r
1 1 1
1
m 0
1
(D g )(c )(x c )
(x c )
m
÷
=

÷
÷

¯
·
2 2 2 2
2
2
m m n r
r
2 2 2
2
m 0
2
(D g )(c )(x c )
(x c )
m
÷
=

÷
÷

¯
· … ·
n n n n
n
n
m m n r
r
n n n
n
m 0
n
(D g )(c )(x c )
(x c )
m
÷
=

÷
÷

¯

82
=
1 1 1 1 1
1
m r m n r
1 1
m 0
1
D g (x c )
m
+ ÷
=
÷
¯
· … ·
n n n n n
n
m r m n r
n n
m 0
n
D g (x c )
m
+ ÷
=
÷
¯
.

Since there is only one way to write f = f
1
· … · f
n
. (i.e., only
one way to write each component f
i
of f) as a n-linear
combination of the n-powers
1 n
k k
1 n
(x c ) ... (x c ) ÷ · · ÷ ; 0  k
i

n
i
; i = 1, 2, …, n; it follows that

i
k
i i
i
(D f )(c )
k
=
i i
i i
k r
i i
i i
i i
0 if 0 k r 1
D g (c )
r k n.
(k r )!
÷
s s ÷
¦
¦
´
s s
¦
÷
¹

This is true for every i, i = 1, 2, …, n. Therefore
i
k
D f
i
(c
i
) = 0 for
0  k
i
 r
i
– 1; i = 1, 2, …, n and
i
r
D f
i
(c
i
)  g
i
(c
i
)  0 ; for every i
= 1, 2, …, n. Conversely if these conditions are satisfied, it
follows at once from Taylor’s n-formula that there is a n-
polynomial g = g
1
· … · g
n
such that f = f
1
· … · f
n
=
( )
1
r
1
x c ÷ g
1
· … ·
( )
n
r
n
x c ÷ g
n
and g(c) = g
1
(c
1
) · … · g
n
(c
n
)
 0 · 0 · … · 0.

Now suppose that
(r
1
, r
2
, …, r
n
) is not the largest positive n-
integer tuple such that
( )
1
r
1
x c ÷ ·
( )
2
r
2
x c ÷ · … ·
( )
n
r
n
x c ÷
divides f
1
· … · f
n
; i.e., each
( )
i
r
i
x c ÷ divides f
i
for i = 1, 2,
…, n; then there is a n-polynomial h = h
1
· … · h
n
such that f
=
( )
1
r 1
1
x c
+
÷ h
1
· … ·
( )
n
r 1
n
x c
+
÷ h
n
. But this implies g = g
1
·
g
2
· … · g
n
= (x – c
1
)h
1
· … · (x – c
n
)h
n
; hence g(c) = g
1
(c
1
)
· … · g
n
(c
n
) = 0 · 0 · … · 0; a contradiction, hence the
claim.

DEFINITION 1.2.21: Let F = F
1
· … · F
n
be a n-field. An n-
ideal in F[x] = F
1
[x] · F
2
[x] · … · F
n
[x] is a n-subspace; m
= m
1
· m
2
· … · m
n
of F[x] = F
1
[x] · … · F
n
[x] such that
when f = f
1
· … · f
n
and g = g
1
· … · g
n
then fg = f
1
g
1
· f
2
g
2

83
· … · f
n
g
n
belongs to m = m
1
· … · m
n
; i.e. each f
i
g
i
e m
i

whenever f is in F[x] and g e m.
If in particular the n-ideal m = dF[x] for some polynomial d
= d
1
· … · d
n
e F[x] i.e. m = m
1
· … · m
n
= d
1
F[x] · … ·
d
n
F[x]; i.e. the n-set of all n-multiples d
1
f
1
· … · d
n
f
n
of d = d
1

· … · d
n
by arbitrary f = f
1
· … · f
n
in F[x] is a n-ideal. For
m is non empty, m in fact contains d. If f, g e F[x] and c is a
scalar then c(df) – dg = (c
1
d
1
f
1
– d
1
g
1
) · … · (c
n
d
n
f
n
– d
n
g
n
) =
d
1
(c
1
f
1
– g
1
) · … · d
n
(c
n
f
n
– g
n
) belongs to m = m
1
· … · m
n
;
i.e. d
i
(c
i
f
i
– g
i
) e m
i
; i = 1, 2, …., n so that m is a n-subspace.
Finally m contains (df)g = d(fg) = (d
1
f
1
)g
1
· … · (d
n
f
n
)g
n
=
d
1
(f
1
g
1
) · … · d
n
(f
n
g
n
) as well; m is called the principal n-ideal
generated by d = d
1
· … · d
n
.

Now we proceed on to prove an interesting theorem about the n-
principal ideal of F[x].

THEOREM 1.2.28: Let F = F
1
· … · F
n
be a n-field and m =
m
1
· … · m
n
, a non zero n-ideal in F[x] = F
1
[x] · … · F
n
[x].
Then there is a unique monic n-polynomial d = d
1
· … · d
n
in
F[x] where each d
i
is a monic polynomial in F
i
[x] ; i = 1, 2, …,
n such that m is the principal n-ideal generated by d.

Proof: Given F = F
1
· F
2
· … · F
n
is a n-field F[x] = F
1
[x] ·
… · F
n
[x] be the n-polynomial over the n-field F. Let m = m
1
·
… · m
n
be a non zero n-ideal of F[x]. We call a n-polynomial
p(x) to be n-monic i.e. if in p(x) = p
1
(x) · … · p
n
(x) every p
i
(x)
is a monic polynomial for i = 1, 2, …, n. Similarly we call a n-
polynomial to be n-minimal if in p(x) = p
1
(x) · … · p
n
(x) each
polynomial p
i
(x) is of minimal degree. Now m = m
1
· … · m
n

contains a non zero n-polynomial p(x) = p
1
(x) · … · p
n
(x)
where each p
i
(x)  0 for i = 1, 2, …, n. Among all the non zero
n-polynomial in m there is a n-polynomial d = d
1
· … · d
n
of
minimal n-degree. Without loss in generality we may assume
that minimal n-polynomial is monic i.e., d is monic. Suppose f =
f
1
· … · f
n
any n-polynomial in m then we know f = dq + r
where r = 0 or n-deg r < n-deg d i.e., f = f
1
· … · f
n
= (d
1
q
1
+
r
1
) · … · (d
n
q
n
+ r
n
). Since d is in m, dq = d
1
q
1
· … · d
n
q
n
e
84
m and f e m so f – dq = r = r
1
· r
2
· … · r
n
e m. But since d is
an n-polynomial in m of minimal n-degree we cannot have n-
deg r < n-deg d so r = 0 · … · 0. Thus m = dF[x] = d
1
F
1
[x] ·
… · d
n
F
n
[x]. If g is any other n-monic polynomial such that
gF[x] = m = g
1
F
1
[x] · … · g
n
F
n
[x] then their exists non zero n-
polynomial p = p
1
· p
2
· … · p
n
and q = q
1
· q
2
· … · q
n

such that d = gp and g = dq i.e., d = d
1
· … · d
n
= g
1
p
1
· … ·
g
n
p
n
and g
1
· …· g
n
= d
1
q
1
· … · d
n
q
n
. Thus d = dpq = d
1
p
1
q
1

· …· d
n
p
n
q
n
= d
1
· …· d
n
and n-deg d = n-deg d + n-deg p +
n-deg q.
Hence n-deg p = n-deg q = (0, 0, …, 0) and as d and g are n-
monic p = q = 1. Thus d = g. Hence the claim.
In the n-ideal m we have f = pq + r where p, f e m i.e. p =
p
1
· p
2
· … · p
n
e m and f = f
1
· …· f
n
e m; f = f
1
· …· f
n

= (p
1
q
1
+ r
1
) · …· (p
n
q
n
+ r
n
) where the n-remainder r = r
1
·
…· r
n
e m is different from 0 · … · 0 and has smaller n-
degree than p.

COROLLARY 1.2.10: If p
1
,

p
2
, …, p
n
are n-polynomials over a n-
field F = F
1
· … · F
n
not all of which are zero i.e. 0 · … · 0,
there is a unique n-monic polynomial d in F[x] = F
1
[x] · … ·
F
n
[x] and d = d
1
· …· d
n
such that

a. d = d
1
· …· d
n
is in the n-ideal generated by p
1
, …, p
n
,
where p
i
= p
i
1
· p
i
2
· … · p
i
n
; i = 1, 2, …, n.

b. d = d
1
· … · d
n
, n-divides each of the n-polynomials p
i

=
1
i
p · … ·
i
n
p i.e. d
j
/
i
j
p for j = 1, 2, …, n true for i =
1, 2, …, n.

Any n-polynomial satisfying (a) and (b) necessarily satisfies

c. d is n-divisible by every n-polynomial which divides
each of the n-polynomials p
1
,
p
2
, …, p
n
.

Proof: Let F = F
1
· … · F
n
be a n-field, F[x] = F
1
[x] · … ·
F
n
[x] be a n-polynomial ring over the n-field F. Let d be a n-
monic generator of the n-ideal
85

{
1
1
p F
1
[x] ·
2
1
p F
1
[x] · … ·
n
1
p F
1
[x]} · {
1
2
p F
2
[x] · … ·
n
2
p F
2
[x]} · … · {
1
n
p F
n
[x] · … ·
n
n
p F
n
[x]}.

Every member of this n-ideal is divisible by d = d
1
· … · d
n
.
Thus each of the n-polynomials p
i
=
i
1
p · … ·
i
n
p is n-divisible
by d. Now suppose f = f
1
· … · f
n
is a n-polynomial which n-
divides each of the n-polynomial
p
1
=
1
1
p · … ·
1
n
p

p
2
=
2
1
p · … ·
2
n
p

and
p
n
=
n
1
p · … ·

n
n
p .

Then there exists n-polynomials g
1
, …, g
n
such that p
i
= fg
i
i.e.,
i
1
p ·…·
i
n
p = f
1
i
1
g ·…· f
n
i
n
g i.e.
i
t
p = f
t
i
t
g for t = 1, 2, …, n;
1  i  n. Also since d = d
1
· … · d
n
is in the n-ideal {
1
1
p F
1
[x]
·
2
1
p F
1
[x] · …·
n
1
p F
1
[x]} · {
1
2
p F
2
[x] ·…·
n
2
p F
2
[x]} · … ·
{
1
n
p F
n
[x] · … ·
n
n
p F
n
[x]} there exists n-polynomials q
1
, …, q
n

in F[x] = F
1
[x] · … · F
n
[x] with q
i
=
i
1
q · … ·
i
n
q ; i = 1, 2,
…, n, such that d = {
1 1
1 1
p q ·
1 1
2 2
p q · … ·
1 1
n n
p q } · … · {
n n
1 1
p q
·
n n
2 2
p q · … ·
n n
n n
p q } = d
1
· … · d
n
. Thus d = f
1
(
1 1
1 1
p q · … ·
1 1
n n
p q ) · f
2
(
2 2
1 1
p q · … ·
2 2
n n
p q ) · … · f
n
(
n n
1 1
p q · … ·
n n
n n
p q ).
We have shown that d = d
1
· … · d
n
is a n-monic
polynomial satisfying (a), (b) and (c). If d
1
=
1 1
1 n
d ... d · · is any
n-polynomial satisfying (a) and (b) it follows from (a) and the
definition of d that d
1
is a scalar multiple of d and satisfies (c) as
well. If d
1
is also n-monic then d = d
1
.

DEFINITION 1.2.22: If p
1
, p
2
,…, p
n
where p
i
=
1
i
p ·
2
i
p ·…·
i
n
p
are n-polynomials over the n-field F = F
1
· F
2
·…· F
n
for i=
1, …, n, such that not all the n-polynomials are 0 · … · 0 the
monic generator

d = d
1
·…· d
n
of the n-ideal {
1
1
p F
1
[x]·
2
1
p F
1
[x]·…·
1
n
p F
1
[x]}· {
1
2
p F
2
[x]·
2
2
p F
2
[x]·…·
2
n
p F
2
[x]}
86
·…· {
1
n
p F
n
[x] ·…·
n
n
p F
n
[x]} is called the greatest common
n-divisor or n-greatest common divisor of p
1
, …, p
n
. This
terminology is justified by the proceeding corollary. We say the
n-polynomials p
1
=
1
1
p ·…·
1
n
p , p
2
=
2
1
p ·…·
2
n
p ,…, p
n
=
1
n
p
·…·
n
n
p are n-relatively prime if their n-greatest common
divisor is (1, 1, … , 1) or equivalently if the n-ideal they
generate is all of F[x] = F
1
[x] ·…· F
n
[x].

Now we talk about the n-factorization, n-irreducible, n-prime
polynomial over the n-field F.

DEFINITION 1.2.23: Let F be an n-field i.e. F = F
1
· F
2
·…·
F
n
. A n-polynomial f = f
1
· …· f
n
in F[x] = F
1
[x] ·…· F
n
[x]
is said to be n-reducible over the n-field F = F
1
· F
2
· … · F
n
,
if there exists n-polynomials g, h e F[x], g = g
1
· … · g
n
and h
= h
1
· … · h
n
in F[x] of n-degree  (1, 1, …, 1) such that f =
gh, i.e., f
1
· … · f
n
= g
1
h
1
· … · g
n
h
n
and if such g and h does
not exists, f = f
1
· … · f
n
is said to be n-irreducible over the n-
field F = F
1
· … · F
n
.
A non n-scalar, n-irreducible n-polynomial over F = F
1
·
… · F
n
is called the n-prime polynomial over the n-field F and
we some times say it is n-prime in F[x] = F
1
[x] · … · F
n
[x].

THEOREM 1.2.39: Let p = p
1
· p
2
· … · p
n
, f = f
1
· f
2
· … ·
f
n
and g = g
1
· g
2
· … · g
n
be n-polynomial over the n-field F
= F
1
· … · F
n
. Suppose that p is a n-prime n-polynomial and
that p n-divides the product fg, then either p n-divides f or p n-
divides g.

Proof: Without loss of generality let us assume p = p
1
· p
2
· …
· p
n
is a n-monic n-prime n-polynomial i.e. monic prime n-
polynomial. The fact that p = p
1
· p
2
· … · p
n
is prime then
simply says that only monic n-divisor of p are 1
n
and p. Let d be
the n-gcd or greatest common n-divisor of f and p. f = f
1
· … ·
f
n
and p = p
1
· … · p
n
since d is a monic n-polynomial which
n-divides p. If d = p then p n-divides f and we are done. So
suppose d = 1
n
= (1 · 1 · … · 1) i.e., suppose f and p are n-
87
relatively prime if f = f
1
· … · f
n
and p = p
1
· … · p
n
, then f
1

and p
1
are relatively prime f
2
and p
2
are relatively prime i.e. (f
i
,
p
i
) = 1 for i = 1, 2, …, n. We shall prove that p n-divides g.
Since (f, p) = (f
1
, p
1
) · … · (f
n
, p
n
) = 1 · … · 1 there are n-
polynomials f
0
=
1
0
f · … ·
n
0
f ; p
0
=
1
0
p · … ·
n
0
p such that 1
· … · 1 = f
o
f + p
o
p i.e. 1 · … · 1 = (
1 1
0
f f +
1 1
0
p p ) · (
2 2
0
f f +
2 2
0
p p ) · … · (
n n
0
f f +
n n
0
p p ) n-multiplying by g = g
1
· … · g
n

we get g = f
0
fg + p
o
pg i.e. (g
1
· … · g
n
) = (
1 1 1
0
f f g +
1 1 1
0
p p g ) ·
… · (
n n n
0
f f g +
n n n
0
p p g ). Since p n-divides fg it n-divides (fg)f
0

and certainly p n-divides pp
0
g. Thus p n-divides g. Hence the
claim.

COROLLARY 1.2.11: If p = p
1
· … · p
n
is a n-prime that n-
divides a n-product f
1
, …, f
n
i.e.
1 1
1 n
f f ! ·
2 2
1 n
f f ! · … ·
1
n n
n
f f ! . Then p n-divides one of the n-polynomials
1 1
1 n
f f ! ·
…·
1
n n
n
f f ! .

The proof is left for the reader.

THEOREM 1.2.40: If F = F
1
· F
2
· … · F
n
is a n-field, a non
n-scalar monic n-polynomial in F[x] = F
1
[x] · … · F
n
[x] can
be n-factored as a n-product of n-monic primes in F[x] in one
and only one way except for the order.

Proof: Given F = F
1
· … · F
n
is a n-field. F[x] = F
1
[x] · …
· F
n
[x] is a n-polynomial over F[x]. Suppose f = f
1
· … ·f
n

is a non scalar monic n-polynomial over the n-field F. As n-
polynomial of n-degree (1, 1, …, 1) are irreducible there is
nothing to prove if n-deg f = (1, 1, …, 1). Suppose f has n
degree (n
1
, n
2
, …, n
n
) > (1, 1, …, 1), by induction we may
assume the theorem is true for all non scalar monic n-
polynomials of n-degree less than (n
1
, n
2
, …, n
n
). If f is n-
irreducible it is already n-factored as a n-product of monic n-
primes and otherwise f = gh = f
1
· f
2
· … · f
n
= g
1
h
1
· … ·
g
n
h
n
where g and h are non scalar monic n-polynomials (i.e. g =
g
1
· … · g
n
and h = h
1
· … · h
n
) of n-degree less than (n
1
, n
2
,
88
…, n
n
). Thus f and g can be n-factored as n-products of monic n-
primes in F[x] = F
1
[x] · … · F
n
[x] and hence f = f

1
· … · f

n

=
1
1 1
1 m
(p , , p ) ! · … ·
n
n n
1 m
(p , , p ) ! =
1
1 1
1 n
(q , , q ) ! · … ·
n
n n n
1 2 n
(q , q ,..., q ) where
1
1 1
1 m
(p , , p ) ! · … ·
n
n n n
1 2 m
(p , p , , p ! ) and
1
1 1
1 n
(q ,..., q ) · … ·
n
n n n
1 2 n
(q , q ,..., q ) are monic n-primes in F[x] =
F
1
[x] · … · F
n
[x]. Then
1 n
1 n
m m
(p ... p ) · · must n-divide some
(
1
i
i
q ·…·
n
n
i
q ). Since both
1 n
1 n
m m
(p ... p ) · · and (
1
i
i
q · … ·
n
n
i
q )
are monic n-primes this means that
t t
t t
i m
q p = for every t = 1, 2,
…, n. Thus we see m
i
= n
i
= 1 for each i = 1, 2, …, n, if either m
i

= 1 or n
i
= 1 for i = 1, 2, …, n. For

n-deg f =
1 1 n n
1 1 n n
1 1 n n
m n m n
1 1 n n
i j i j
i 1 j 1 i 1 j 1
degp degq , , degp degq
= = = =
| |
= =
|
|
\ .
¯ ¯ ¯ ¯
! .

In this case we have nothing more to prove. So we may assume
m
i
> 1, i = 1, 2, …, n and n
j
> 1, j = 1, 2, …, n.
By rearranging q
i
’s we can assume
i i
i i
m n
p q = and that
( )
1 1 n
1 1 1 n
1 m 1 m m 1
p , , p , p , , p
÷ ÷
! ! = (
1 1
1 2
q q , …,
1 1
1 1
n 1 m
q p
÷
, …,
n
1
q , …,
n n
n n
m m
q p ). Thus
( )
1 1 n
1 1 1 n
1 m 1 m m 1
p ,..., p , p ,..., p
÷ ÷
= (
1
1
q , …,
1
1
n 1
q
÷
, …,
n
1
q ,…,
n
n
n 1
q
÷
). As the n-polynomial has n-degree less than (n
1
, n
2
,
…, n
n
) our inductive assumption applies and shows the n-
sequence (
1
1
q ,…,
1
1
n 1
q
÷
,…,
n
1
q ,…,
n
n
n 1
q
÷
) is at most rearrangement
of n-sequence (
1
1
p , …,
1
1
m 1
p
÷
, … ,
n
1
p , …,
n
1
m 1
p
÷
).
This shows that the n-factorization of f = f
1
· … · f
n
as a
product of monic n-primes is unique up to order of factors.
Several interesting results in this direction can be derived. The
reader is expected to define n-primary decomposition of the n-
polynomial f = f
1
· … · f
n
in F[x] = F
1
[x] · … · F
n
[x]. The
reader is requested to prove the following theorems

THEOREM 1.2.41: Let f = f
1
· … · f
n
be a non scalar monic
n-polynomial over the n-field F = F
1
· … · F
n
and let
89
f =
( )
1
1
1 1 1
1
1 1
... ... ...
n
n
k k
n
n
n n
n n
k k
p p p p · ·
be the prime n-factorization of f. For each j
t
, 1  j
t
 k
t
, let
t t
k j
i
n n
t t
j j i
i j
f f p p
=
= =
I
,
t = 1, 2, …, n. Then
1
t
f , …,
t
t
k
f are relatively prime for t = 1, 2,
…, n.
THEOREM 1.2.42: If f = f
1
· … · f
n
is a n-polynomial over the
n-field F = F
1
· F
2
· … · F
n
with derivative f ' =
1
f ' ·
2
f ' ·…
·
n
f ' Then f is a n-product of distinct irreducible n-polynomials
over the n-field F if and only if f and f ' are relatively prime i.e.,
each f
i
and
i
f ' are relatively prime for i = 1, 2, …, n.

The proof is left as an exercise for the reader.
Next we proceed on to define the notion of n-characteristic
value of type II i.e., n-characteristic values for n-linear operator
on the n-vector space V.
Throughout this section we assume the n-vector space is
defined, over n-field; i.e., n-vector space of type II.

DEFINITION 1.2.24: Let V = V
1
· … · V
n
be a n-vector space
over the n-field F = F
1
· F
2
· … · F
n
and let T be a n-linear
operator on V, i.e., T = T
1
· … · T
n
and T
i
: V
i
÷ V
i
, i = 1, 2,
…, n. This is the only way n-linear operator can be defined on
V. A n-characteristic value of T is a n-scalar C = C
1
· … · C
n

(C
i
e F
i
, i = 1, 2, …, n) in F such that there is a non zero n-
vector Į = Į
1
· … · Į
n
in V = V
1
· … · V
n
with TĮ = CĮ i.e.,
T = T
1
Į
1
· … · T
n
Į
n
= C
1
Į
1
· … · C
n
Į
n
i.e., T
i
Į
i
= C
i
Į
i
; i =
1, 2, …, n. If C is a n-characteristic value of T then

a. any Į = Į
1
· … · Į
n
such that TĮ = CĮ is called the n-
characteristic n-vector of T associated with the n-
characteristic value C = C
1
· … · C
n
.
b. The collection of all Į = Į
1
· … · Į
n
such that TĮ =
CĮ is called the n-characteristic space associated with
C.
90

If T = T
1
· … · T
n
is any n-linear operator on the n-vector
space V = V
1
· … · V
n
. We call the n-characteristic values
associated with T to be n-characteristic roots, n-latent roots, n-
eigen values, n-proper values or n-spectral values.
If T is any n-linear operator and C = C
1
· … · C
n
in any
n-scalar, the set of n-vectors Į = Į
1
· … · Į
n
such that TĮ =
CĮ is a n-subspace of V. It is in fact the n-null space of the n-
linear transformation (T – CI) = (T
1
– C
1
I
1
) · … · (T
n
– C
n
I
n
)
where I
j
denotes a unit matrix for j = 1, 2, …, n. We call C
1
· …
· C
n
the n-characteristic value of T if this n-subspace is
different from the n-zero space 0 = 0 · 0 · … · 0, i.e.,(T – CI)
fails to be a one to one n-linear transformation and if the n-
vector space is finite dimensional we see that (T – CI) fails to be
one to one and the n-det(T – CI) = 0 · 0 · … · 0.

We have the following theorem in view of these properties.

THEOREM 1.2.43: Let T = T
1
· … · T
n
be a n-linear operator
on a finite dimensional n-vector space V = V
1
· … · V
n
and let
C = C
1
· … · C
n
be a scalar. The following are equivalent.

a. C = C
1
· … · C
n
is a n-characteristic value of T = T
1

· … · T
n
.
b. The n-operator (T
1
– C
1
I
1
)· … · ( T
n
– C
n
I
n
) = T – CI
is n-singular or (not n-invertible).
c. det(T – CI ) = 0 · 0 · … · 0 i.e., det(T
1
– C
1
I)· … ·
det( T
n
– C
n
I
n
) = 0 · …· 0.

Now we define the n-characteristic value of a n-matrix A = A
1

· … · A
n
where each A
i
is a n
i
× n
i
matrix with entries from
the field F
i
so that A is a n-matrix defined over the n-field F =
F
1
· F
2
· … · F
n
. A n-characteristic value of A in the n-field F
= F
1
· F
2
· … · F
n
is a n-scalar C = C
1
· … · C
n
in F such
that the n-matrix A – CI = (A
1
– C
1
I
1
) · (A
2
– C
2
I
2
) · … · (A
n

– C
n
I
n
) is n-singular or not n-invertible.

Since C = C
1
· … · C
n
is a n-characteristic value of A =
A
1
· … · A
n
, A a (n
1
× n
1
, …, n
n
× n
n
) n-matrix over the n-
91
field, F = F
1
· F
2
· … · F
n
, if and only if n-det(A – CI) = 0 ·
0 · … · 0 i.e., det(A
1
– C
1
I
1
) · … · det(A
n
– C
n
I
n
) = (0 · 0
· … · 0), we form the n-matrix (xI – A) = (xI
1
– A
1
) · … ·
(xI
n
– A
n
).
Clearly the n-characteristic values of A in F = F
1
· F
2
· …
· F
n
are just n-scalars C = C
1
· … · C
n
in F = F
1
· F
2
· … ·
F
n
such that the n-scalars C = C
1
· … · C
n
in F = F
1
· F
2
· …
· F
n
such that f(C) = f
1
(C
1
) · … · f
n
(C
n
) = 0 · 0 · … · 0.
For this reason f = f
1
· f
2
· … · f
n
is called the n-characteristic
polynomial of A. Clearly f is a n-polynomial of different degree
in x over different fields. It is important to note that f = f
1
· f
2

· … · f
n
is a n-monic n-polynomial which has n-degree
exactly (n
1
, n
2
, …, n
n
). The n-monic polynomial is also a n-
polynomial over F = F
1
· F
2
· … · F
n
.

First we illustrate this situation by the following example.

Example 1.2.8: Let

1 0 0 0 0
2 1 0 1
1 0 1 0 1 0 0 0
1 1 0 0 0 4
A 0 1 0 0 4 6 0 1
0 2 2 1 1 0
1 0 0 0 0 0 5 0
0 0 0 1
0 0 0 0 3

= · · ·

= A
1
· A
2
· A
3
· A
4
; be a 4-matrix of order (3 × 3, 4 × 4, 2 ×
2, 5 × 5) over the 4-field F = Z
2
· Z
3
· Z
5
· Z
7
. The 4-
characteristic 4-polynomial associated with A is given by
(xI ÷ A) = (xI
3×3
÷ A
1
) · (xI
4×4
÷ A
2
) · (xI
2×2
÷ A
3
)
· (xI
5×5
÷ A
4
)

x 1 2 0 2
x 1 0 1
2 x 2 0 0
0 x 1 0
0 1 x 1 2
1 0 x
0 0 0 x 2
+

+

+

= + · ·

+

+

92
x 6 0 0 0 0
0 x 6 0 0 0
x 1
0 3 x 1 0 6
4 x
0 0 0 x 2 0
0 0 0 0 x 4
+

+

· +

+

+

is 4-matrix with polynomial entries.
f = f
1
· f
2
· … · f
4

= det(xI ÷ A)
= det (xI
1
÷ A
1
) · det (xI
2
÷ A
2
) · … · det(xI ÷ A)
= {(x + 1)
2
x + (x + 1)} · {(x + 1)(x + 2) × (x + 1)(x + 2)
– 2 × 2(x + 1)(x + 2)} · (x
2
– 4) · {(x + 6)(x + 6)(x +
1)(x + 2)(x + 4)}.

We see det(xI – A) is a 4-polynomial which is a monic 4-
polynomial and degree of f is (3, 4, 2, 5) over the 4-field F = Z
2

· Z
3
· Z
5
· Z
7
.

Now we first define the notion of similar n-matrices when the
entries of the n-matrices are from the n-field.

DEFINITION 1.2.25: Let A = A
1
· … · A
n
be a (n
1
× n
1
, …, n
n

× n
n
) matrix over the n-field F = F
1
· F
2
· … · F
n
i.e., each A
i

takes its entries from the field F
i
, i = 1, 2, …, n. We say two n-
matrices A and B of same order are similar if there exits a n-
non invertible n-matrix P = P
1
· … · P
n
of (n
1
× n
1
, … . n
n
×
n
n
) order such that
B = P
÷1
AP where P
–1
=
1 1 1
1 2
...
÷ ÷ ÷
· · ·
n
P P P .
B = B
1
· B
2
· … · B
n
=
1 1 1
1 1 1 2 2 2
...
n n n
P A P P A P P A P
÷ ÷ ÷
· · ·
then
det(xI ÷ B) = det((xI ÷ P
-1
AP )
= det P
-1
(xI ÷ A) P
= det P
-1
det (xI ÷ A) det P
= det(xI ÷ A)
= det (xI
1
÷ A
1
) · … · det(xI
n
÷ A
n
).

93
Example 1.2.10: Let A = A
1
· A
2
· A
3

=
0 1 0 1
3 1 1
0 1 1 0 1 0
2 2 1
1 0 0 1 0 1
2 2 0
1 0 1 0

÷

÷

· ÷ ·

be a 3-matrix over the 3-field F = Z
2
· Z
5
· Z
3
. det(xI ÷ A) =
det(xI
1
÷ A
1
) · det(xI
2
÷ A
2
) · det(xI
3
÷ A
3
) is a 3-polynomial
of 3-degree(2, 3, 4), can be easily obtained.

We see the 3-polynomial is a monic 3-polynomial.

DEFINITION 1.2.26: Let T = T
1
· T
2
· … · T
n
be a non-linear
operator over the n-space V = V
1
· V
2
· … · V
n
. We say T is
n-diagonalizable if there is a n-basis for V for each n-vector of
which is a n-characteristic vector of T.
Suppose T = T
1
· T
2
· … · T
n
is a n-diagonalizable n-
linear operator. Let {
1
1
C , …,
1
1
k
C } · {
2
1
C , …,
2
2
k
C }· … · {
1
n
C ,
…,
n
n
k
C } be the n-distinct n-characteristic values of T. Then
there is an ordered n-basis B = B
1
· … · B
n
in which T is
represented by a n-diagonal matrix which has for its n-diagonal
entries the scalars
t
i
C each repeated a certain number of times t
= 1, 2, …, n. If
t
i
C is repeated
t
i
d times then (we may arrange
that) the n-matrix has the n-block form

[T]
B
=
| |
1
1
B
T · … ·
| |
n
n
B
T

=
1 1 2 2
1 1 2 2
1 1 1 1
1 1 2 2
2 2 2 2
1 1 2 2
0 0 0 0
0 0 0 0
0 0 0 0
k k k k
C I C I
C I C I
C I C I

· ·

! !
! !
# # # # # #
! !

94
1 1
2 2
0 0
0 0
0 0
n n
n n
n n
n n
k k
C I
C I
C I

· ·

!
!
!
# # #
!

where
t
j
I is the
t t
j j
d d × identity matrix.

From this n-matrix we make the following observations. First
the n-characteristic n-polynomial for T = T
1
· T
2
· … · T
n
is
the n-product of n-linear factors
f = f
1
· f
2
· … · f
n

= (x –
1
1
C )
1
1
d
. (x –
1
1
k
C )
1
n
k
d
· (x –
2
1
C )
2
1
d
. (x –
2
2
k
C )
2
n
k
d

· . · (x –
1
n
C )
1
n
d
. (x –
n
n
k
C )
n
k
n
d
.
If the n-scalar field F = F
1
· F
2
· … · F
n
is n-
algebraically closed i.e., if each F
i
is algebraically closed for i
= 1, 2, …, n; then every n-polynomial over F = F
1
· F
2
· … ·
F
n
can be n-factored; however if F = F
1
· F
2
· … · F
n
is not
n-algebraically closed we are citing a special property of T = T
1

· … · T
n
when we say that its n-characteristic polynomial has
such a factorization. The second thing to be noted is that
t
i
d is
the number of times
t
i
C is repeated as a root of f
t
which is equal
to the dimension of the space in V
t
of characteristic vectors
associated with the characteristic value
t
i
C ; i = 1, 2, ., k
t
, t =
1, 2, …, n. This is because the n-nullity of a n-diagonal n-matrix
is equal to the number of n-zeros which has on its main n-
diagonal and the n-matrix, [T – CI]
B
= [T
1

1
1
i
C I]
1
B
· … · [T
n

n
n
i
C I]
n
B
has (
1
1
i
d …
n
n
i
d ), n-zeros on its main n-diagonal.

This relation between the n-dimension of the n-characteristic
space and the n-multiplicity of the n-characteristic value as a n-
root of f does not seem exciting at first, however it will provide
us with a simpler way of determining whether a given n-
operator is n-diagonalizable.
95

LEMMA 1.2.4: Suppose that TĮ = CĮ, where T = T
1
· … · T
n

is the n-linear operator C = C
1
· … · C
n
where the n-scalar C
i

e F
i
; i = 1, 2, …, n and Į = Į
1
· … · Į
n
is a n-vector. If f = f
1

· f
2
· … · f
n
is any n-polynomial then f(T)Į = f(C)Į i.e.,
f
1
(T
1

1
· … · f
n
(T
n

n
= f
1
(C
1

1
· … · f
n
(C
n

n

.

It can be easily proved by any reader.

LEMMA 1.2.5: Let T = T
1
· T
2
·… · T
n
be a n-linear operator
on the finite (n
1
, n
2
, …, n
n
) dimensional n-vector space V = V
1
·
… · V
n
over the n-field F = F
1
· … · F
n
. If
{ } { } { }
1 2
1 1 2 2
1 1 1 2
n
n n n
k k k
C C C C C C C · · · ! ! ! !
be the distinct n-characteristic values of T and let W
i
=
1
1
i
W ·
2
2
i
W · . ·
n
n
i
W be the n-space of n-characteristic n-vectors
associated with the n-characteristic values C
i
=
1
1
i
C ·
2
2
i
C · .
·
n
n
i
C . If W = {
1
1
W + . +
1
1
k
W } · {
2
1
W + . +
2
2
k
W } · . ·
{
1
n
W + . +
n
n
k
W } the n-dim W = ((dim
1
1
W + . + dim
1
1
k
W ),
(dim
2
1
W + . + dim
2
2
k
W ), ., (dim
1
n
W + . +
n
n
k
W )) = dim W
1
· . · dim W
n
. In fact if
t
t
i
B is the ordered basis of
t
t
i
W , t = 1,
2, …, k
t
and t =1, 2, …, n then B = {
1
1
B , .,
1
1
k
B } · {
2
1
B , .,
2
2
k
B } ·.· {
1
n
B , .,
n
n
k
B } is an n-ordered n-basis of W.

Proof: We will prove the result for a W
t
for that will hold for
every t, t = 1, 2, …, n. Let W
t
=
t
1
W · .·
t
t
k
W , 1 s t s n, be
the subspace spanned by all the characteristic vector of T
t
.
Usually when we form the sum W
t
of subspaces
t
i
W we see that
dim W
t
< dim
t
1
W + . + dim
t
t
k
W because of linear relations
which may exist between vectors in the various spaces. This
result state the characteristic spaces associated with different
characteristic values are independent of one another.
96
Suppose that (for each i) we have a vector
t
i
þ in
t
i
W and
assume that
t
1
þ + . +
t
t
k
þ = 0. We shall show that
t
i
þ = 0 for
each i. Let f be any polynomial. Since
t t t
t i i i
T C þ = þ the
proceeding lemma tells us that 0 = f
t
(T
t
) thus 0 = f
t
(T
t
)
t
1
þ + . +
f
t
(T
t
)
t
t
k
þ = f
t
(
t
1
C )
t
1
þ + . + f
t
(
t
t
k
C )
t
t
k
þ . Choose polynomials
t
1
f
+ . +
t
t
k
f such that
( )
t t t
i j ij
1 i j
f C
0 i j
=
¦
= o =
´
=
¹
.
Then
( )
t t t
i t ij j
j
0 f T , 0 = = o þ
¯
=
t
i
þ .
Now let
t
i
B be an ordered basis for
t
i
W and let B
t
be the
sequence
( )
t
t t t
1 k
B B , , B = ! . Then B
t
spans the subspace W
t
=
t
t t
1 k
W , , W ! . Also B
t
is a linearly independent sequence of
vectors for the following reason.
Any linear relation between the vectors in B
t
will have the
form
t
t t t
1 2 k
0 þ +þ + +þ = ! , where
t
i
þ is some linear combination
of the vectors in
t
i
B . From what we just proved we see that
t
i
0 þ = for each i = 1, 2, …, k
t
. Since each
t
i
B is linearly
independent we see that we have only the trivial linear relation
between the vectors in B
t
.
This is true for each t = 1, 2, …, n. Thus we have
dimW = ((dim
1
1
W +…+ dim
1
1
k
W ), (dim
2
1
W +… + dim
2
2
k
W ),
…, (dim
n
1
W + . + dim
n
n
k
W ))
= (dimW
1
, dimW
2
, ., dim W
n
).

We leave the proof of the following theorem to the reader.

THEOREM 1.2.44: Let T = T
1
· … · T
n
be a n-linear operator
of a finite (n
1
, n
2
, …, n
n
) dimensional n-vector space V = V
1
·
V
2
· … · V
n
over the n-field F = F
1
· … · F
n
. Let
{ }
1
1 1 1
1 2
, , ,
k
C C C ! · … ·
{ }
1 2
, , ,
n
n n n
k
C C C ! be the distinct n-
97
characteristic values of T and let W
i
=
1
1
i
W ·
2
2
i
W · . ·
n
n
i
W
be the n-null space of T ÷ C
i
I =
1 1
( ) ... ( )
i
i i
n k
T C I T C I ÷ · · ÷ .
The following are equivalent:

(i) T is n-diagonalizable.
(ii) The n-characteristic n-polynomial for T = T
1
· … · T
n

is f = f
1
· … · f
n

( ) ( ) ( ) ( )
1 2 1 2
1 1
1 2
1 2
1 1 2 2
1 1
k k
d d d d
k k
x C x C x C x C = ÷ ÷ · ÷ ÷ ! !
( ) ( )
1
1
n n
k
n
n
d d
n n
k
x C x C · · ÷ ÷ ! !
and dim
( )
1 2
1 2
, , , = !
n
n
i i i i
W d d d , i = 1, 2, …, k.
(iii) ((dim
1
1
W +…+ dim
1
1
k
W ), (dim
2
1
W +…+ dim
2
2
k
W ), …,
(dim
1
n
W +…+ dim
n
n
k
W )) = n-dim V = (n
1
, n
2
, …, n
n
).
The n-matrix analogous of the above theorem may be
formulated as follows. Let A = A
1
· … · A
n
be a (n
1
× n
1
, n
2
×
n
2
, …, n
n
× n
n
) n-matrix with entries from the n-field, F = F
1
· …
· F
n
and let
{ } { } { }
1 2
1 1 1 2
1 2 1
, , , , , , · · · ! ! ! !
n
n n
k k k
C C C C C C be
the n-distinct n-characteristic values of A in F. For each i let W
i

=
1
1
i
W + … +
n
n
i
W be the n-subspace of all n-column matrices X
= X
1
· … · X
n
(with entries from the n-field F = F
1
· … · F
n
)
such that (A ÷ C
i
I)X = (A
1
÷
1 1
1
i i
C I )X
1
· … · (A
n
÷
n n
n
i i
C I )X
n
=
(0 · … · 0) = (
1
1
i
B · … ·
n
n
i
B ) and let B
i
be an n-ordered n-
basis for W
i
=
1
1
i
W + . +
n
n
i
W . The n-basis B = B
1
, …, B
n

collectively string together to form the n-sequence of n-columns
of a n-matrix P = [
1
1
P , …,
n
n
P ] = {B
1
, …, B
n
} =
{ }
1
1 1
1
, ,
k
B B ! ·
{ }
2
1 1
2
, ,
k
B B ! · … ·
{ }
1 2
, , ,
n
n n n
k
B B B ! . The n-matrix A over the
n-field F = F
1
· … · F
n
is similar to a n-diagonal n-matrix if
and only if P is a (n
1
× n
1
, …, n
n
× n
n
) n-square matrix. When P
is square, P is n-invertible and P
-1
AP is n-diagonal i.e., each
1
i
P
÷
A
i
P
i
is diagonal in P
-1
AP =
1
1
P
÷
A
1
P
1
· … ·
1
n
P
÷
A
n
P
n
.
98

Now we proceed on to define the new notion of n-annihilating
n-polynomials of the n-linear operator T, the n-minimal n-
polynomial for T and the analogue of the Cayley-Hamilton
theorem for a n-vector space V over n-field of type II and its n-
linear operator on V.
In order to know more about the n-linear operator T = T
1
·
… · T
n
on V
1

· … · V
n
over the n-field F = F
1

· … · F
n
one
of the most useful things to know is the class of n-polynomials
which n-annihilate T = T
1

· … · T
n
.
To be more precise suppose T = T
1

· … · T
n
is a n-linear
operator on V, a n-vector space over the n-field F = F
1

· … ·
F
n
. If p = p
1
· … · p
n
is a n-polynomial over the n-field F = F
1

· … · F
n
then p(T) = p
1
(T
1
) · … · p
n
(T
n
) is again a n-linear
operator on V = V
1
· … · V
n
.
If q = q
1
· … · q
n
is another n-polynomial over the same
n-field F = F
1
· … · F
n
then
(p + q)T = (p)T + (q)T
i.e.,
(p
1
+ q
1
)T
1
· … · (p
n
+ q
n
)T
n

= [p
1
(T
1
) · … · p
n
(T
n
)] + [q
1
(T
1
) · … · q
n
(T
n
)]
and
(pq)T = (p)T(q)T
i.e.,
(p
1
q
1
)T
1
· … · (p
n
q
n
)T
n
= [p
1
T
1
q
1
T
1
· … · p
n
T
n
q
n
T
n
].

Therefore the collection of n-polynomials P = P
1
· … · P
n

which n-annihilate T = T
1

· … · T
n
in the sense that p(T) =
p
1
(T
1
) · … · p
n
(T
n
) = 0 · … · 0 is an n-ideal of the n-
polynomial n-algebra F(x) = F
1
(x) · … · F
n
(x). It may be the
zero n-ideal i.e., it may be, that T is not n-annihilated by any
non-zero n-polynomial. But that cannot happen if the n-space V
= V
1

· … · V
n
is finite dimensional i.e., V is of (n
1
, n
2
, …, n
n
)
dimension over the n-field F = F
1

· … · F
n
.
Suppose T = T
1

· T
2

· … · T
n
is a n-linear operator on the
(n
1
, n
2
, …, n
n
) dimension n-space V = V
1

· V
2

· … · V
n
. The
99
first n, (
2
i
n + 1) operators, (i = 1, 2, …, n) must be n-linearly
dependent i.e., the first (
2
1
n + 1,
2
2
n + 1, .,
2
n
n + 1) n-linear
operators are n-linearly dependent i.e., we have
1
0
C I
1
+
1
1
C T
1
+
. +
2
1
1
n
C
2
1
n
1
T = 0,
2
0
C I
2
+
2
1
C T
2
+ . +
2
2
2
n
C
2
2
n
2
T = 0 and so on,
n
0
C I
n
+
n
1
C T
n
+ … +
2
n
n
n
C
2
n
n
n
T = 0;
That is {
1
0
C I
1
+
1
1
C T
1
+ … +
2
1
1
n
C
2
1
n
1
T } · {
2
0
C I
2
+
2
1
C T
2
+
. +
2
2
2
n
C
2
2
n
2
T } · … · {
n
0
C I
n
+
n
1
C T
n
+ … +
2
n
n
n
C
2
n
n
n
T } = 0 · 0
· … · 0 for some n-scalars;
1
1
i
C ,
2
2
i
C , …,
n
n
i
C not all zero. 1 s
i
p
s n
p
and p = 1, 2, …, n. Thus the n-ideal of n-polynomials
which n-annihilate T contains a non zero n-polynomial of n-
degree (
2
1
n ,
2
2
n , …,
2
n
n ) or less.
We know that every n-polynomial n-ideal consists of all n-
multiples of some fixed n-monic n-polynomials which is the n-
generator of the n-ideal. Thus there corresponds to the n-
operator T = T
1
· … · T
n
a n-monic n-polynomial p = p
1
· …
· p
n
.
If f is any other n-polynomial over the n-field F = F
1
· … ·
F
n
then f(T) = 0 · 0 · … · 0 i.e., f
1
(T
1
) · … · f
n
(T
n
) = 0 · 0
· … · 0 if and only if f = pg where g = g
1
· … · g
n
is some
polynomial over the n-field F = F
1
· … · F
n
i.e., f = f
1
· … ·
f
n
= p
1
g
1
· … · p
n
g
n
.
Now we define the new notion of n-polynomial for the n-
operator T: V ÷ V.

DEFINITION 1.2.27: Let T = T
1
· … ·T
n
be a n-linear operator
on a finite (n
1
, …, n
n
) dimensional n-vector space V = V
1
· …
· V
n
over the field F
1
· … · F
n
. The n-minimal n-polynomial
for T is the (unique) monic n-generator of the n-ideal of n-
polynomials over the n-field, F = F
1
· … · F
n
which n-
annihilate T = T
1
· … · T
n
.
The n-minimal n-polynomial starts from the fact that the n-
generator of a n-polynomial n-ideal is characterised by being
the n-monic n-polynomial of n-minimum n-degree in the n-ideal
that implies that the n-minimal n-polynomial p = p
1
· … · p
n

100
for the n-linear operator T = T
1
· T
2
· … · T
n
is uniquely
determined by the following properties.
1. p is a n-monic n-polynomial over the n-scalar n-field F
= F
1
· … · F
n
.
2. p(T) = p
1
(T
1
) · … · p
n
(T
n
) = 0 · … · 0.
3. No n-polynomial over the n-field F = F
1
· … · F
n

which n-annihilates T = T
1
· T
2
· … · T
n
has smaller
n-degree than p = p
1
· … · p
n
has.
(n
1
× n
1
, n
2
× n
2
, …, n
n
× n
n
) is the order of n-matrix A = A
1

· … · A
n
over the n-field F = F
1
· … · F
n
where each A
i
has
n
i
× n
i
matrix with entries from the field F
i
, i = 1, 2, …, n.
The n-minimal n-polynomial for A = A
1
· … · A
n
is
defined in an analogous way as the unique n-monic generator of
the n-ideal of all n-polynomial over the n-field, F = F
1
· … ·
F
n
which n-annihilate A.
If the n-operator T = T
1
· T
2
· … · T
n
is represented by
some ordered n-basis by the n-matrix A = A
1
· … · A
n
then T
and A have same n-minimal polynomial because f(T) = f
1
(T
1
) ·
… · f
n
(T
n
) is represented in the n-basis by the n-matrix f(A) =
f
1
(A
1
) · … · f
n
(A
n
) so f(T) = 0 · … · 0 if and only if f(A) = 0
· … · 0 i.e., f
1
(A
1
) · … · f
n
(A
n
) = 0 · … · 0 if and only if
f
1
(T
1
) · … · f
n
(T
n
) = 0 · … · 0. So f(P
-1
AP) = f
1
(P
-1
A
1
P
1
) ·
… · f
n
(
1 ÷
n
P A
n
P
n
) =
1
1
÷
P f
1
(A
1
)P
1
· … ·
1 ÷
n
P f
n
(A
n
)
1 ÷
n
P =
P
-1
f(A)P for every n-polynomial f = f
1
· f
2
· … · f
n
.

Another important feature about the n-minimal polynomials
of n-matrices is that suppose A = A
1
· … · A
n
is a (n
1
× n
1
, …,
n
n
× n
n
) n-matrix with entries from the n-field F = F
1
· … · F
n
.
Suppose K = K
1
· … · K
n
is n-field which contains the n-field
F = F
1
· … · F
n
i.e., K _ F and K
i
_ F
i
for every i, i = 1, 2, …,
n. A = A
1
· … · A
n
is a (n
1
× n
1
, …, n
n
× n
n
) n-matrix over F
or over K but we do not obtain two n-minimal n-polynomial
only one minimal n-polynomial.

This is left as an exercise for the reader to verify. Now we
proceed on to prove an interesting theorem about the n-minimal
polynomials for T(or A).

101
THEOREM 1.2.45: Let T = T
1
· … · T
n
be a n-linear operator
on a (n
1
, n
2
, …, n
n
) dimensional n-vector space V = V
1
· … ·
V
n
[or let A be a (n
1
× n
1
, …, n
n
× n
n
)n-matrix i.e., A = A
1
· …
· A
n
where each A
i
is a n
i
× n
i
matrix with its entries from the
field F
i
of F = F
1
· … · F
n
, true for i = 1, 2, ., n].
The n-characteristic and n-minimal n-polynomial for T [for
A] have the same n-roots except for n-multiplicities.

Proof: Let p = p
1
· … · p
n
be a n-minimal n-polynomial for T
= T
1
· … ·T
n
. Let C = C
1
· … · C
n
be a n-scalar of the n-
field F = F
1
· … · F
n
. To prove p(C) = p
1
(C
1
) · … · p
n
(C
n
) =
0 · … · 0 if and only if C = C
1
· … · C
n
is the n-
characteristic value of T. Suppose p(C) = p
1
(C
1
) · … · p
n
(C
n
)
= 0 · … · 0; then p = (x – C
1
)q
1
· (x – C
2
)q
2
· … · (x –
C
n
)q
n
where q = q
1
· … ·q
n
is a n-polynomial, since n-deg q <
n-deg p, the n-minimal n-polynomial p = p
1
· … · p
n
tells us
q(T) = q
1
(T
1
) · … · q
n
(T
n
) = 0 · … · 0. Choose a n-vector þ
= þ
1
· … · þ
n
such that q(T)þ = q
1
(T
1

1
· … · q
n
(T
n

n
= 0
· … · 0. Let Į = q(T)þ i.e., Į = Į
1
· … · Į
n
= q
1
(T
1

1
· …
· q(T
n

n
. Then

0 · … · 0 = p(T)þ = p
1
(T
1

1
· … · p
n
(T
n

n.

= (T – CI)q(T)þ
= (T
1
– C
1
I
1
)q
1
(T
1

1
· … · (T
n
– C
n
I
n
)q
n
(T
n

n

= (T
1
– C
1
I
1
)o
1
· … · (T
n
– C
n
I
n
)o
n

and thus C = C
1
· … · C
n
is a n-characteristic value of T = T
1

· … · T
n
.
Suppose C = C
1
· … · C
n
is a n-characteristic value of T =
T
1
· … · T
n
say To = Co i.e., T
1
o
1
· … · T
n
o
n
= C
1
o · …
· C
n
o with o = 0 · … · 0. From the earlier results we have
p(T)o = p(C)o i.e., p
1
(T
1
)o
1
· … · p
n
(T
n
)o
n.
= p
1
(C
1
)o
1
· …
· p
n
(C
n
)o
n
; since p(T) = p
1
(T
1
) · … · p
n
(T
n
) = 0 · … · 0 and
Į = Į
1
· … · Į
n
= 0 we have p(C) = p
1
(C
1
) · … · p
n
(C
n
) = 0
· … · 0.
Let T = T
1
· … · T
n
be a n-diagonalizable n-linear
operator and let
{ } { } { }
1 2 n
1 1 2 2 n n
1 k 1 k 1 k
C C C C C C · · · ! ! ! ! be
102
the n-distinct n-characteristic values of T. Then the n-minimal
n-polynomial for T is the n-polynomial p = p
1
· … · p
n
= (x –
1
1
C ) … (x –
1
1
k
C ) · (x –
2
1
C ) … (x –
2
2
k
C ) · … · (x –
n
1
C ) …
(x –
n
n
k
C ).
If Į = Į
1
· … · Į
n
is a n-characteristic n-vector then one of
the n-operators {(T
1
÷
1
1
C I
1
), …, (T
1
÷
1
1
k
C I
1
)}, {(T
2
÷
2
1
C I
2
), …,
(T
2
÷
2
2
k
C I
2
)}, …, {(T
n
÷
n
1
C I
n
), …, (T
n
÷
n
n
k
C I
n
)} send Į = Į
1
·
… · Į
n
into 0 · … · 0, thus resulting in {(T
1
÷
1
1
C I
1
), …, (T
1
÷
1
1
k
C I
1
)}, {(T
2
÷
2
1
C I
2
), …, (T
2
÷
2
2
k
C I
2
)}, {(T
n
÷
n
1
C I
n
), …, (T
n
÷
n
n
k
C I
n
)} = 0 · … · 0 for every n-characteristic n-vector Į = Į
1

· … · Į
n
.
Hence there exists an n-basis for the underlying n-space
which consist of n-characteristic vectors of T = T
1
· … · T
n
.
Hence p(T) = p
1
(T
1
) · … · p
n
(T
n
) = {(T
1
÷
1
1
C I
1
), …, (T
1
÷
1
1
k
C I
1
)} · … ·{(T
n
÷
n
1
C I
n
), …, (T
n
÷
n
n
k
C I
n
)} = 0 · … · 0.
Thus we can conclude if T is n-diagonalizable, n-linear
operator then the n-minimal n-polynomial for T is a product of
n-distinct n-linear factors.

THEOREM 1.2.46: (CAYLEY-HAMILTON): Let T = T
1
· … ·
T
n
be a n-linear operator on a finite (n
1
, n
2
, …, n
n
) dimensional
vector space V = V
1
· … · V
n
over the n-field F = F
1
· … ·
F
n
. If f = f
1
· … · f
n
is the n-characteristic, n-polynomial for T
then f(T) = f
1
(T
1
) · … · f
n
(T
n
) = 0 · … · 0; in otherwords the
n-minimal polynomial divides the n-characteristic polynomial
for T.

Proof: Let K = K
1
· … · K
n
be a n-commutative ring with n-
identity 1
n
= (1, …, 1) consisting of all n-polynomials in T; K is
actually a n-commutative algebra with n-identity over the scalar
n-field F = F
1
· … · F
n
.
Let {
1
1
o …
1
1
n
o } · … · {
n
1
o …
n
n
n
o } be an ordered n-
basis for V and let A = A
1
· … · A
n
be the n-matrix which
represents T = T
1
· … · T
n
in the given n-basis. Then
103

i
= T
1
1
1
i
o · … · T
n
n
n
i
o
1 2 n
1 1 1 2 2 2 n n n
1 2 n
n n n
1 1 2 2 n n
j i j j i j j i j
j 1 j 1 j 1
A A A ;
= = =
= o · o · · o
¯ ¯ ¯
!

i
i j
1 j n s s , i = 1, …, n. These n-equations may be equivalently
written in the form

( ) ( )
1 2
1 1 1 1 1 1 2 2 2 2 2 2.
1 2
n n
1 1 2 2
j i 1 j i i j j i 2 j i i j
j 1 j 1
T A I T A I
= =
o ÷ o · o ÷ o
¯ ¯

( )
n
n n n n n n
n
n
n n
j i n j i i j
j 1
T A I
=
· · o ÷ o
¯
!
= 0 · … · 0, 1s i
n
s n.

Let B = B
1
·… · B
n
denote the element of
1 1 n n
n n n n
1 n
K K
× ×
· · !
i.e., B
i
is an element of
i i
n n
i
K
×
with entries
t t t t t t
t
i j i j t i j t
B T A I = o ÷ ,
t = 1, 2, ., n. When n
t
= 2; 1s i
t
, j
t
s n
t
.

t t
t t 11 t 21 t
t
12 t t 22 t
T A I A I
B
A I T A I
÷ ÷
=

÷ ÷

and det B
t
= (T
t

t
11
A I
t
)(T
t

t
22
A I
t
) – (
t
12
A
t
21
A )I
t
= f
t
(T
t
) where
f
t
is the characteristic polynomial associated with T
t
, t = 1, 2, .,
n. f
t
= x
2
-trace A
t
x + det A
t
. For case n
t
> 2 it is clear that det B
t

= f
t
(T
t
) since f
t
is the determinant of the matrix xI
t
– A
t
whose
entries are polynomial
( )
t t
t
t
i j
xI A ÷
t t i j
t t
t
i j
x A = o ÷ .
We will show f
t
(T
t
) = 0. In order that f
t
(T
t
) is a zero
operator, it is necessary and sufficient that (det B
t
)
t
k
o
= 0 for k
t

= 0, 1, …, n
t
.
By the definition of B
t
; the vectors
t
1
o · … ·
t
t
n
o satisfy
the equations;
104
t
t t
t
n
t t
i j jt
j 0
B
=
o
¯
= 0,

1s i
t
s n
t
. When n
t
= 2 we can write the above equation in the
form
22
t t
t
t 11 t 21 t
1
t t
t
21 t t
2
T A I A I 0
A T A I 0
÷ ÷ o
=

÷ ÷
o

.

In this case the usual adjoint B
t
is the matrix

11
t t
t 22 t 21 t
t t
12 t
T A I A I
B
A I T A I
÷
=

÷

and
t
t t
t
det B 0
B B
0 det B

=

.
Hence
( )
t t
t t t 1 1
t t
2 2
det B B B
o o
=

o o

t
t t 1
t
2
B B
o
=

o

t
t t 1
t
2
B B
o
=

o

0
0

=

.

In the general case
t t

. Then
t
t t t t
t
n
t t t
k i i j j
j 1
B B 0
=
o =
¯

,
for each pair k
t
, i
t
and summing on i
t
we have

t t
t
t t t t t
t t
n n
t t
k i i j j
i 1 j 1
0 B B
= =
= o
¯¯

t t
t
t t t t t
t t
n n
t t
k i i j j
i 1 j 1
B B
= =
| |
= o
|
|
\ .
¯ ¯

.

105

Now
i
t
t
B B = (det B
t
)I
t
so that
t
t t t t t t
t
n
t t
k i i j k j
i 1
B B det B
=
= o
¯

.
Therefore
( )
t
t t t
t
n
t t
k j j
j 1
0 det B
=
= o o
¯
( )
t
t t
k
det B = o , 1s k
t
s n
t
.
Since this is true for each t, t = 1, 2, …, n we have 0 · … ·
0 = (det B
t
)
1
1
k
o · … · (det B
n
)
n
n
k
o , 1s k
i
s n
i
, i = 1, 2, …, n.
The Cayley-Hamilton theorem is very important for it is
useful in narrowing down the search for the n-minimal n-
polynomials of various n-operators.
If we know the n-matrix A = A
1
· … · A
n
which
represents T = T
1
· … ·T
n
in some ordered n-basis then we
can compute the n-characteristic polynomial f = f
1
· … · f
n
.
We know the n-minimal polynomial p = p
1
· … · p
n
n-divides
f i.e., each p
i
/f
i
for i = 1, 2, …, n (which we call as n-divides f)
and that the two n-polynomials have the same n-roots.
However we do not have a method of computing the roots
even in case of polynomials so more difficult is finding the n-
roots of the n-polynomials. However if f = f
1
· … · f
n
factors
as
f = (x –
1
1
C )
1
1
d
… (x –
1
1
k
C )
1
k
1
d
· (x –
2
1
C )
2
1
d
… (x –
2
2
k
C )
2
k
2
d
· … ·
(x –
n
1
C )
n
1
d
… (x –
n
n
k
C )
n
k
n
d
{
1
1
C , …,
1
1
k
C } · {
2
1
C , …,
2
2
k
C } · …
· {
n
1
C , …,
n
n
k
C }

distinct n-sets,
t
t
i
d 1 > , t = 1, 2, …, k
t
then
p = p
1
· … · p
n
= (x –
1
1
C )
1
1
r
… (x –
1
1
k
C )
1
k
1
r
· … · (x –
n
1
C )
n
1
r

… (x –
n
n
k
C )
n
k
n
r
; 1s
t
j
r s
t
j
d .

Now we illustrate this by a simple example.

106
Example 1.2.10: Let

A =
1 1 0 0
3 1 1
1 1 0 0 0 1
2 2 1
2 2 2 1 1 0
2 2 0
1 1 1 0

÷

÷ ÷ ÷

· ÷ ·

÷ ÷

÷

be a 3-matrix over the 3-field F = Z
3
· Z
5
· Q. Clearly the 3-
characteristic 3-polynomial associated with A is given by f = f
1

· f
2
· f
3
= x
2
(x – 1)
2
· (x – 1)(x – 2)
2
· x
2
+ 1. It is easily
verified that p = p
1
· … · p
3
= x
2
(x – 1)
2
· (x – 1)(x – 2)
2
·
(x
2
+ 1) is the 3-minimal 3-polynomial of A.

Now we proceed on to define the new notion of n-invariant
subspaces or equivalently we may call it as invariant n-
subspaces.

DEFINITION 1.2.28: Let V = V
1
· … · V
n
be a n-vector space
over the n-field F = F
1
· F
2
· … · F
n
of type II. Let T = T
1
·
… · T
n
be a n-linear operator on V. If W = W
1
· … · W
n
is a
n-subspace of V we say W is n-invariant under T if each of the
n-vectors in W, i.e., for the n-vector Į = Į
1
· … · Į
n
in W the
n-vector TĮ = T
1
Į
1
· … · T
n
Į
n
is in W i.e., each T
i
Į
i
e W
i
for
every Į
i
e W
i
under the operator T
i
for i = 1, 2, …, n i.e., if
T(W) is contained in W i.e., if T
i
(W
i
) is contained in W
i
for i =
1, 2, …, n i.e., are thus represented as T
1
(W
1
) · … · T
n
(W
n
) _
W
1
· … · W
n
.

This simple example is that we can say V the n-vector space
is invariant under a n-linear operator T in V. Similarly the zero
n-subspace is invariant under T.
Now we give the n-block matrix associated with T. Let W = W
1

· … · W
n
be a n-subspace of the n-vector space V = V
1
· …
· V
n
. Let T = T
1
· … · T
n
be a n-operator on V such that W is
n-invariant under the n-operator T then T induces a n-linear
operator T
w
=
1
1
w
T · … ·
n
n
w
T on the n-space W. This n-linear
operator T
w
defined by T
w
(o) = T(o) for o e W i.e., if Į = Į
1
·
107
… · Į
n
then T
w

1
· … · Į
n
) =
1
1
w
T (Į
1
) · … ·
n
n
w
T (Į
n
).
Clearly T
w
is different from T as domain is W and not V. When
V is finite dimensional say (n
1
, …, n
n
) dimensional the n-
invariance of W under T has a simple n-matrix interpretation.
Let B = B
1
· … · B
n
= {
1
1
o …
1
1
r
o } · … · {
n
1
o …
n
n
r
o } be a
chosen n-basis for V such that B' =
1 n
B ... B ' ' · · = {
1
1
o …
1
1
r
o }
· … · {
n
1
o …
n
n
r
o } ordered n-basis for W. r = (r
1
, r
2
, …, r
n
) =
n dim W.
Let A = [T]
B
i.e., if A = A
1
· … · A
n
then A = A
1
· … ·
A
n
= [T
1
]
B
1

· … · [T
n
]
B
n

so that
t
t
t t
j
1
t
n
t t t
i j i
i 1
T A
o
=
= o
¯

for t = 1, 2, …, n i.e.,
To
j
= T
1
1
1
j
o · … · T
n
n
n
j
o
1 2 n
1 1 1 2 2 2 n n n
1 2 n
n n n
1 1 2 2 n n
i j i i j i i j i
i 1 i 1 i 1
A A A .
= = =
= o · o · · o
¯ ¯ ¯
!

Since W is n-invariant under T, the n-vector To
j
belongs to W
for j < r i.e., (j
1
< r
1
, …, j
n
< r
n
).
1 n
1 1 1 n n n
1 n
r r
1 1 n n
j i j i i j i
i 1 i 1
T A A
= =
o = o · · o
¯ ¯
!
i.e.,
k k
k
i j
A 0 = if j
k
< r
k
and i
k
> r
k
for every k = 1, 2, …, n.
Schematically A has the n-block

1 1 n n
1 n
B C B C B C
A
O D
O D O D

= = · ·

!

where B
t
is a r
t
× r
t
matrix. C
t
is a r
t
× (n
t
÷ r
t
) matrix and D is an
(n
t
÷ r
t
) × (n
t
÷ r
t
) matrix for t = 1, 2, …, n. It is B = B
1
· … ·
B
n
is the n-matrix induced by the n-operator T
w
in the n-basis
1 n
B B B ' ' ' = · · ! .

108
LEMMA 1.2.6: Let W = W
1
· … · W
n
be an n-invariant
subspace of the n-linear operator T = T
1
· … · T
n
on V = V
1
·
… · V
n
. The n-characteristic, n-polynomial for the n-restriction
operator T
W
=
1
1
...
n
W nW
T T · · divides the n-characteristic
polynomial for T. The n-minimal polynomial for T
W
divides the
n-minimal polynomial for T.

Proof: We have
1 1 n n
1 n
B C B C B C
A
O D
O D O D

= = · ·

!

where A = [T]
B
=
1 n
1B nB
T ... T · · and
| |
W
B
B T
'
= = B
1
· … ·
B
n
=
1
1
1w
B
T
'

· … ·
n
n
nw
B
T
'

. Because of the n-block form of
the n-matrix
det(xI ÷ A) = det(xI
1
÷ A
1
) · det(xI
2
÷ A
2
) · … · det(xI
n
÷ A
n
)
where
A = A
1
· … · A
n

= det(xI ÷ B) det(xI ÷ D)
= {det (xI
1
÷ B
1
) det(xI
1
÷ D
1
) · … · det(xI
n
÷ B
n
) det(xI
n

÷ D
n
)}.
That proves the statement about n-characteristic
polynomials. Notice that we used I = I
1
· … · I
n
to represent n-
identity matrix of these n-tuple of different sizes.
The k
th
power of the n-matrix A has the n-block form

A
k
= (A
1
)
k
· … · (A
n
)
k
.

( ) ( )
( )
( ) ( )
( )
k k k k
1 1 n n
k
k k
1 n
B C B C
A
O D O D

= · ·

!

where C
k
= (C
1
)
k
· … · (C
n
)
k
is {r
1
× (n
1
÷ r
1
), …, r
n
× (n
n
÷ r
n
)}
n-matrix. Therefore any n-polynomial which n-annihilates A
also n-annihilates B (and D too). So the n-minimal polynomial
for B n-divides the n-minimal polynomial for A.
109
Let T = T
1
· … · T
n
be any n-linear operator on a (n
1
, n
2
,
…, n
n
)finite dimensional n-space V = V
1
· … · V
n
. Let W =
W
1
· … · W
n
be n-subspace spanned by all of the n-
characteristic vectors of T = T
1
· … · T
n
. Let {
1
1
C , …,
1
1
k
C } ·
{
2
1
C , …,
2
2
k
C } · … · {
n
1
C , …,
n
n
k
C } be the n-distinct
characteristic values of T. For each i let W
i
=
1
1
i
W · … ·
n
n
i
W
be the n-space of n-characteristic vectors associated with the n-
characteristic value C
i
=
1
1
i
C · … ·
n
n
i
C and let B
i
= {
1
i
B · …
·
n
i
B } be the ordered basis of W
i
i.e.,
t
i
B is a basis of
t
i
W . B' =
{
1
1
B , …,
1
1
k
B } · … · {
n
1
B , …,
n
n
k
B } is a n-ordered n-basis for
W = (
1
1
W + … +
1
1
k
W ) · … · (
n
1
W + … +
n
n
k
W ) = W = W
1
·
… · W
n
.
In particular n-dimW = {(dim
1
1
W + . + dim
1
1
k
W ),
(dim
2
1
W + … + dim
2
2
k
W ), …, (dim
n
1
W + . + dim
n
n
k
W )}.
We prove the result for one particular W
i
= {
i
1
W + . +
1
i
k
W }
and since W
i
is arbitrarily chosen the result is true for every i, i
= 1, 2, ., n. Let
i
B' = {
i
1
o , …,
i
i
r
o } so that the first few o
i
’s
form the basis of
i
B' , the next few
2
B' and so on.
Then T
i
t
j
o =
i t
j j
t o , j = 1, 2, …, r
i
where (
i
1
t , …,
i
i
r
t ) = {
i
1
C ,
…,
i
1
C , …,
i
i
k
C ,
i
i
k
C , …,
i
i
k
C ) where
i
j
C is repeated dim
i
j
W
times, j = 1, ., r
i
. Now W
i
is invariant under T
i
since for each
o
i
in W
i
, we have
i
o =
i i
1 1
x o + … +
i i
i i
r r
x o
T
i
i
o =
i i i
1 1 1
t x o + … +
i i i
i i i
r r r
t x o .

Choose any other vector
i
i
r 1 +
o , …,
i
i
n
o in V
i
such that B
i
= {
i
1
o ,
…,
i
i
n
o } is a basis for V
i
. The matrix of T
i
relative to B
i
has the
block form mentioned earlier and the matrix of the restriction
operator relative to the basis
i
B' is

110
i
i
1
i
2 i
i
r
t 0 0
0 t 0
B
0 0 t

=

!
!
# # #
!
.

The characteristic polynomial of B
i
i.e., of
i
iw
T is g
i
= g
i
(x –
i
1
C )
i
1
e
… (x –
i
i
k
C )
i
k
1
e
where
i
j
e = dim
i
j
W ; j = 1, 2, …, k
i
.
Further more g
i
divides f
i
, the characteristic polynomial for T
i
.
Therefore the multiplicity of
i
j
C as a root of f
i
is at least
dim
i
j
W . Thus T
i
is diagonalizable if and only if r
i
= n
i
i.e., if
and only if
i
1
e + … +
i
i
k
e = n
i
. Since what we proved for T
i
is
true for T = T
1
· … · T
n
. Hence true for every B
1
· … · B
n
.
We now proceed on to define T-n conductor of o into W =
W
1
· … · W
n
_ V
1
· … · V
n
.

DEFINITION 1.2.29: Let W = W
1
· … · W
n
be a n-invariant n-
subspace for T = T
1
· … · T
n
and let Į = Į
1
· … · Į
n
be a n-
vector in V = V
1
· … · V
n
. The T-n conductor of o into W is
the set S
T
(o; W) =
1
T
S (o
1
; W
1
) · … ·
n
T
S (o
n
; W
n
) which
consists of all n-polynomials g = g
1
· … · g
n
over the n-field F
= F
1
· F
2
· … · F
n
such that g(T)o is in W; that is g
1
(T
1
)o
1
·
g
2
(T
2
)o
2
· … · g
n
(T
n
)o
n
e W
1
· … · W
n
.
Since the n-operator T will be fixed through out the
discussions we shall usually drop the subscript T and write S(o;
W) = S(o
1
;W
1
) · … · S(o
n
; W
n
). The authors usually call the
collection of n-polynomials the n-stuffer. We as in case of vector
spaces prefer to call as n-conductor i.e., the n-operator g(T) =
g
1
(T
1
) · … · g
n
(T
n
); slowly leads to the n-vector Į
1
· … · Į
n

into W = W
1
· … · W
n
. In the special case when W = {0} · …
· {0}, the n-conductor is called the T-annihilator of Į
1
· … ·
Į
n
.

We prove the following simple lemma.

111
LEMMA 1.2.7: If W = W
1
· … · W
n
is an n-invariant subspace
for T = T
1
· … · T
n
, then W is n-invariant under every n-
polynomial in T = T
1
· … · T
n
. Thus for each Į = Į
1
· … · Į
n

in V = V
1
· … · V
n
the n-conductor S(o; W) = S(o
1
;W
1
) · …
· S(o
n
; W
n
) is an n-ideal in the n-polynomial algebra F[x] =
F
1
[x] · … · F
n
[x].

Proof: Given W = W
1
· … · W
n
_ V = V
1
· … · V
n
a n-
vector space over the n-field F = F
1
· … · F
n
. If þ = þ
1
· … ·
þ
n
is in W = W
1
· … · W
n
, then Tþ = T
1
þ
1
· … · T
n
þ
n
is in
W = W
1
· … · W
n
. Thus T(Tþ) = T
2
þ =
2
1
T þ
1
· … ·
2
n
T þ
n
is
in W. By induction T
k
þ =
1
k
1
T þ
1
· … ·
n
k
n
T þ
n
is in W for each
k. Take linear combinations to see that f(T)þ = f
1
(T
1

1
· … ·
f
n
(T
n

n
is in W for every polynomial f = f
1
· … · f
n
.
The definition of S(o; W) = S(o
1
;W
1
) · … · S(o
n
; W
n
) is
meaningful if W = W
1
· … · W
n
is any n-subset of V. If W is
a n-subspace then S(o; W) is a n-subspace of F[x] = F
1
[x] · …
· F
n
[x] because (cf + g)T = cf(T) + g(T) i.e., (c
1
f
1
+ g
1
)T
1
· …
· (c
n
f
n
+ g
n
)T
1
= c
1
f
1
(T
1
) + g
1
(T
1
) · … · c
n
f
n
(T
n
) + g
n
(T
n
). If W
= W
1
· … · W
n
is also n-invariant under T = T
1
· … · T
n
and
let g = g
1
· … · g
n
be a n-polynomial in S(o; W) = S(o
1
;W
1
)
· … · S(o
n
; W
n
) i.e., let g(T)o = g
1
(T
1
)o
1
· g
2
(T
2
)o
2
· … ·
g
n
(T
n
)o
n
be in W = W
1
· … · W
n
is any n-polynomial then
f(T)g(T)o is in W = W
1
· … · W
n
i.e.,
f(T)[g(T)o] = f
1
(T
1
) [g
1
(T
1
)o
1
]

· … · f
n
(T
n
)[g
n
(T
n
)o
n
]
will be in W = W
1
· … · W
n
.
Since (fg)T = f(T)g(T) we have
(f
1
g
1
)T
1
· … · (f
n
g
n
)T
n
= f
1
(T
1
) g
1
(T
1
) · … · f
n
(T
n
)g
n
(T
n
)o
n

be (fg) e S(o; W) i.e., (f
i
g
i
) e S(o
i
; W
i
); i = 1, 2, …, n. Hence
the claim.
The unique n-monic generator of the n-ideal S(o; W) is also
called the T-n-conductor of o in W(the T-n annihilator in case
W = {0} · {0} · … · {0}). The T-n-conductor of o into W is
the n-monic polynomial g of least degree such that g(T)o =
g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
is in W = W
1
· … · W
n
.
A n-polynomial f = f
1
· … · f
n
is in S(o; W) = S(o
1
;W
1
) ·
… · S(o
n
; W
n
) if and only if g n-divides f. Note the n-
112
conductor S(o; W) always contains the n-minimal polynomial
for T, hence every T-n-conductor n-divides the n-minimal
polynomial for T.

LEMMA 1.2.8: Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
)
dimensional n-vector space over the n-field F = F
1
· … · F
n
.
Let T = T
1
· … · T
n
be a n-linear operator on V such that the
n-minimal polynomial for T is a product of n-linear factors p =
p
1
·…· p
n
= (x –
1
1
c )
1
1
r
(x –
1
2
c )
1
2
r
… (x –
1
1
k
c )
1
1
k
r
· (x –
2
1
c )
2
1
r
(x –
2
2
c )
2
2
r
… (x –
2
2
k
c )
2
2
k
r
·…· (x –
1
n
c )
1
n
r
(x –
2
n
c )
2
n
r
… (x –
n
n
k
c )
n
k
n
r
;
i
i
t
c e F
i
, k
1
s t
i
s k
n
, i = 1, 2, ., n.
Let W
1
· … · W
n
be a proper n-subspace of V(V = W)
which is n-invariant under T. There exist a n-vector Į
1
· … ·
Į
n
in V such that

(1) Į is not in W
(2) (T ÷ cI)Į = (T
1
÷ c
1
I
1

1
· … · (T
n
÷ c
n
I
n

n
is in W
for some n-characteristic value of the n-operator T.

Proof: (1) and (2) express that T-n conductor of Į = Į
1
· … ·
Į
n
into W
1
· … · W
n
is a n-linear polynomial. Suppose þ = þ
1
· … · þ
n
is any n-vector in V which is not in W. Let g = g
1
·
… · g
n
be the T-n conductor of þ in W. Then g n-divides p = p
1

· … · p
n
the n-minimal polynomial for T. Since þ is not in W,
the n-polynomial g is not constant. Therefore g = g
1
· … · g
n

= (x –
1
1
c )
1
1
e
… (x –
1
1
k
c )
1
k
1
e
· (x –
2
1
c )
2
1
e
… (x –
2
2
k
c )
2
k
2
e
· … · (x –
n
1
c )
n
1
e
… (x –
n
n
k
c )
n
k
n
e
where at least one of the n-tuple of integers
1
i
e ·
2
i
e · … ·
n
i
e is positive. Choose j
t
so that
t
t
j
e > 0, then
(x – c
j
) = (x –
1
1
j
c ) · (x –
2
2
j
c ) · … · (x –
n
n
j
c ) n-divides g. g =
(x – c
j
)h i.e., g = g
1
· … · g
n
= (x –
1
1
j
c )h
1
· … · (x –
n
n
j
c )h
n
.
But by the definition of g the n-vector Į = Į
1
· … · Į
n
=
h
1
(T
1

1
· … · h
n
(T
n

n
= h(T)þ cannot be in W. But (T ÷ c
j
I)o
= (T ÷ c
j
I)h(T)þ = g(T)þ is in W i.e.,
113
(T
1
÷
1
1
j
c
j
I
1
)o
1
· … · (T
n
÷
n
n
j
c
j
I
n
)o
n

= (T
1
÷
1
1
j
c I
1
)h
1
(T
1

1
· … · (T
n
÷
n
n
j
c I
n
)h
n
(T
n

n

= g
1
(T
1

1
· g
2
(T
2

2
· … · g
n
(T
n

n

with g
i
(T
i

i
e W
i
for i = 1, 2, ., n.

Now we obtain the condition for T to be n-triangulable.

THEOREM 1.2.47: Let V = V
1
· … · V
n
be a finite n
1
, n
2
, …, n
n

dimensional n-vector space over the n-field F = F
1
· … · F
n

and let T = T
1
· … ·T
n
be a n-linear operator on V. Then T is
n-triangulable if and only if the n-minimal polynomial for T is a
n-product of n-linear polynomials over F = F
1
· … · F
n
.

Proof: Suppose the n-minimal polynomial p = p
1
· … · p
n
, n-
factors as p = (x –
1
1
c )
1
1
r
… (x –
1
1
k
c )
1
k
1
r
· (x –
2
1
c )
1
2
r
· (x –
2
1
c )
2
1
r

… (x –
2
2
k
c )
2
k
2
r
· … · (x –
n
1
c )
n
1
r
…(x –
n
n
k
c )
n
k
n
r
. By the repeated
application of the lemma just proved we arrive at a n-ordered n-
basis. B = {
1
1
o …
1
1
n
o } · {
1
2
o …
2
2
n
o } · … · {
1
n
o …
n
n
n
o } =
B
1
· B
2
· … · B
n
in which the n-matrix representing T = T
1
·
… · T
n
is n-upper triangular.

1 n
1 B n B
[T ] ... [T ] · ·

1 2
1 2
1 1 2 2
1 1 1 2 2 2
11 12 1n 11 12 1n
1 1 2 2
22 2n 22 2n
1 2
n n n n
a a a a a a
0 a a 0 a a
0 0 a 0 0 a

= ·

! !
# # # # # #
! !
· … ·

n
n
n n
n n n
11 12 1n
n n
22 2n
n
n n
a a a
0 a a
0 0 a

!
# # #
!
.
114

Merely [T]
B
= the n-triangular matrix of (n
1
× n
1
, …, n
n
× n
n
)
order shows that

j
= T
1
1
1
j
o · … · T
n
n
n
j
o
=
1
1 1
1j 1
a o +…+
1 1 1
1 1
j j j
a o ·
2
2 2
1j 1
a o +…+
2 2 2
2 2
j j j
a o · … ·
n
n n
1j 1
a o + … +
n n n
n n
j j j
a o (a)

1s j
i
s n
i
; i = 1, 2, ., n, that is To
j
is in the n-subspace spanned
by {
1
1
o …
1
1
j
o } · … · {
n
1
o …
n
n
j
o }. To find {
1
1
o …
1
1
j
o } · …
· {
n
1
o …
n
n
j
o }; we start by applying the lemma to the n-
subspace W = W
1
· … · W
n
= {0} · … · {0} to obtain the n-
vector
1
1
o · … ·
n
1
o . Then apply the lemma to
1
1
W ·
2
1
W ·
… ·
n
1
W , the n-space spanned by o
1
=
1
1
o · … ·
n
1
o , and we
obtain o
2
=
1
2
o · … ·
n
2
o . Next apply lemma to W
2
=
1 n
2 2
W ... W · · , the n-space spanned by
1 n
1 1
... o · ·o
and
1 n
2 2
... o · ·o . Continue in that way. After o
1
, o
2
, …, o
i
we
have found it is the triangular type relation given by equation (a)
for j
i
= 1, 2, …, n
i
, i = 1, 2, …, n which proves that the n-
subspace spanned by o
1
, o
2
, …, o
i
is n-invariant under T.
If T is n-triangulable it is evident that the n-characteristic
polynomial for T has the form f = f
1
· … · f
n
= (x –
1
1
c )
1
1
d
… (x

1
1
k
c )
1
k
1
d
·…· (x–
n
1
c )
n
1
d
… (x –
n
n
k
c )
n
k
n
d
. The n-diagonal entries
(
1
11
a …
1
1
1n
a ) · (
2
11
a …
2
2
1n
a ) · … · (
n
11
a …
n
n
1n
a ) are the n-
characteristic values with
t
j
c repeated
t
jt
d times. But if f can be
so n-factored, so can the n-minimal polynomial p because p n-
divides f.

We leave the following corollary to be proved by the reader.
115
COROLLARY 1.2.12: If F = F
1
· … · F
n
is an n-algebraically
closed n-field. Every (n
1
× n
1
, …, n
n
× n
n
) n-matrix over F is
similar over the n-field F to be a n-triangular matrix.
Now we accept with some deviations in the definition of the
n-algebraically closed field when ever F
i
is the complex field C.
THEOREM 1.2.48: Let V = V
1
· … · V
n
be a (n
1
, ., n
n
)
dimensional n-vector space over the n-field F = F
1
· … · F
n
and let T = T
1
· … · T
n
be a n-linear operator on V = V
1
· …
· V
n
. Then T is n-diagonalizable if and only if the n-minimal
polynomial for T has the form
p = p
1
· … · p
n
=
1
1 1
1 1
( ) . . . ( ) . . . ( ) . . . ( ) ÷ ÷ · · ÷ ÷
n
n n
k k
x c x c x c x c
where
1
1 1
1 1
{ , . . ., } . . . { , . . ., } · ·
n
n n
k k
c c c c are n-distinct elements of
F = F
1
· … · F
n
.
Proof: We know if T = T
1
· … · T
n
is n-diagonalizable its n-
minimal polynomial is a n-product of n-distinct linear factors.
Hence one way of the proof is clear.
To prove the converse let W = W
1
· … · W
n
be a
subspace spanned by all the n-characteristic n-vectors of T and
suppose W = V. Then we know by the properties of n-linear
operator that their exists a n-vector o = o
1
· … · o
n
in V and
not in W and the a n-characteristic value
1 n
1 n
j j j
c c . . . c = · · of T
such that the n-vector
þ = (T – c
j
I)o
=
1 n
1 n
1 j 1 1 n j n n
(T c I ) . . . (T c I ) ÷ o · · ÷ o
= þ
1
· … · þ
n
lies in W = W
1
· … · W
n
where each þ
i
e W
i
; i = 1, 2, …, n.
Since þ = þ
1
· … · þ
n
is in W;
i
k 1
i i i
. . . ; þ = þ + + þ for each
i = 1, 2, …, n with
t t t
i i i i
T c þ = þ ; t = 1, 2, …, k
i
, this is true for
every i = 1, 2, …, n and hence the n-vector
h(T)þ =
1 1
1 1 1 1 1 1
1 1 k k
{h (c ) . . . h (c ) } þ + + þ
n n
n n n n n n
1 1 k k
. . . {h (c ) . . . h (c ) } · · þ + + þ
116
for every n-polynomial h. Now
p = (x – c
j
)q
= p
1
· … · p
n
=
1 n
1 n
j 1 j n
(x c )q . . . (x c )q ÷ · · ÷
for some n-polynomial q = q
1
· … · q
n
. Also
q – q(c
j
) = (x – c
j
)h,
i.e.,
1 1 1
1 1 1
1 1 j j j
q q (c ) (x c )h ÷ = ÷ , …,
n n
n n n
n n j j j
q q (c ) (x c )h ÷ = ÷ .
We have
q(T)o – q(c
j
)o = h(T) (T – c
j
I)o = h(T)þ
But h(T)þ is in W = W
1
· … · W
n
and since
0 = p(T)o = (T – c
j
I) q(T)o
= p
1
(T
1
)o
1
· … · p
n
(T
n
)o
n
=
1 n
1 n
1 j 1 1 1 1 n j n n n
(T c I )q (T ) . . . (T c )q (T ) ÷ o · · ÷ o
and the n-vector q(T)o is in W i.e., q
1
(T
1
)o
1
· … · q
n
(T
n
)o
n
is
in W. Therefore
1 n
1 n
j 1 j 1 n j n
q(c ) q (c ) . . . q (c ) o = o · · o
is in W. Since o = o
1
· … · o
n
is not in W, we have q(c
j
) =
1 n
1 n
1 j n j
q (c ) . . . q (c ) · · = 0 · … · 0.
This contradicts the fact that p has distinct roots. Hence the
claim.
We can now describe this more in terms of how the values
are determined and its relation to Cayley Hamilton Theorem for
n-vector spaces of type II. Suppose T = T
1
· … · T
n
is a n-
linear operator on a n-vector space of type II which is
represented by the n-matrix A = A
1
· … · A
n
in some ordered
n-basis for which we wish to find out whether T is n-
diagonalizable. We compute the n-characteristic polynomial f =
f
1
· … · f
n
. If we can n-factor f = f
1
· … · f
n
as
1 n
1 n
k k
1 1 1 n
1 n
d d
d d 1 1 n n
1 k 1 k
(x c ) . . . (x c ) . . . (x c ) . . . (x c ) ÷ ÷ · · ÷ ÷
we have two different methods for finding whether or not T is
n-diagonalizable. One method is to see whether for each i(t) (i(t)
means i is independent on t) we can find a
t
i
d (t = 1, 2, …, n); 1
s i s k
t
independent characteristic vectors associated with the
characteristic value
t
j
c . The other method is to check whether or
not
117
(T – c
1
I) · … · (T – c
k
I) =
1 n
1 1 n n
1 1 1 1 k 1 n 1 n n k n
(T c I ) . . . (T c I ) . . . (T c I ) . . . (T c I ) ÷ ÷ · · ÷ ÷
is the n-zero operator.
Here we have some problems about speaking of n-
algebraically closed n-field in a universal sense; we can only
speak of the n-algebraically closed n-field relative to a n-
polynomial over the same n-field. Now we proceed on to define
or introduce the notion of simultaneous n-diagonalization and
simultaneous n-triangulation. Throughout this section V = V
1
·
… · V
n
will denote a (n
1
, n
2
, …, n
n
) finite n-vector space over
the n-field F = F
1
· … · F
n
and let · = ·
1
· … · ·
n
be the n-
family of n-linear operator on V. We now discuss when one can
simultaneously n-triangularize or n-diagonalize the n-operators
in F. i.e., to find one n-basis B = B
1
· … · B
n
such that all n-
matrices [T]
B
=
1 n
1 B n B
[T ] [T ] · · ! , T in · are n-triangular
(or n-diagonal). In case of n-diagonalization it is necessary that f
be in the commuting family of n-operators UT = TU, i.e., U
1
T
1
· … · U
n
T
n
= T
1
U
1
· … · T
n
U
n
for all T, U in ·. That
follows from the simple fact that all n-diagonal n-matrices
commute. Of course it is also necessary that each n-operator in
· be an n-diagonalizable operator. In order to simultaneously n-
triangulate each n-operator in · we see each n-operator must be
n-triangulable. It is not necessary that · be a n-commuting
family, however that condition is sufficient for simultaneous n-
triangulation (if each T = T
1
· … · T
n
can be individually n-
triangulated).
We recall a subspace W = W
1
· … · W
n
is n-invariant
under · if W is n-invariant under each operator in · i.e., each
W
i
in W = W
1
· … · W
n
is invariant under the operator T
i
in T
= T
1
· … · T
i
· … · T
n
; true for every i = 1, 2, …, n.
LEMMA 1.2.9: Let · = ·
1
· … · ·
n
be a n-commuting family
of n-triangulable n-linear operators on the n-vector space V =
V
1
· … · V
n
. Let W = W
1
· … · W
n
be a proper n-subspace of
V = V
1
· … · V
n
which is n-invariant under ·.
There exists a n-vector o = o
1
· … · o
n
in V such that
a. o = o
1
· … · o
n
is not in W = W
1
· … · W
n
118
b. for each T = T
1
· … · T
n
in · = ·
1
· … · ·
n
the n-
vector To = T
1
o
1
· T
2
o
2
· … · T
n
o
n
is in the n-
subspace spanned by o = o
1
· … · o
n
and W = W
1
·… · W
n
.
Proof: Without loss in generality let us assume that · contains
only a finite number of n-operators because of this observations.
Let {T
1
, …, T
r
} be a maximal n-linearly independent n-subset
of ·; i.e., a n-basis for the n-subspace spanned by ·. If o = o
1
· … · o
n
is a n-vector such that (b) holds for each
i i i
1 n
T T . . . T , 1 i r = · · s s then (b) will hold for every n-
operator which is a n-linear combination of T
1
, …, T
r
.
By earlier results we can find a n-vector
1 1 1
1 n
. . . þ = þ · ·þ
in V (not in W) and a n-scalar
1 1 1
1 n
c c . . . c = · · such that
1 1 1
(T c I) ÷ þ is in W; i.e.,
1 1 1 1 1 1
1 1 1 1 n n n n
(T c I ) . . . (T c ) I ÷ þ · · ÷ þ e
W
1
· … · W
n
i.e., each
1 1 1
r r r r
(T c ) W; ÷ þ e r = 1, 2, …, n. Let
1 1 1
1 n
V V . . . V = · · be the collection of all vector þ in V such
that (T
1
– c
1
I) þ is in W. Then V
1
is a n-subspace of V which is
properly bigger than W. Further more V
1
is n-invariant under ·
for this reason. If T = T
1
· … · T
n
commutes with
1 1 1
1 n
T T . . . T = · · then
1 1 1 1
(T c I)T T(T c I) ÷ þ = ÷ þ. If þ is in
V
1
then (T
1
– c
1
I) þ is in W i.e., Tþ is in V
1
for all þ in V
1
and
for all T in ·. Now W is a proper n-subspace of V
1
. Let
2 2 2
1 n
U U . . . U = · · be the n-linear operator on V
1
obtained by
restricting
2 2 2
1 n
T T . . . T = · · to the n-subspace V
1
. The n-
minimal polynomial for U
2
divides the n-minimal polynomial
for T
2
. Therefore we may apply the earlier results to the n-
operator and the n-invariant n-subspace W. We obtain a n-
vector þ
2
in V
1
(not in W) and a n-scalar c
2
such that (T
2

c
2
I) þ
2
is in W. Note that
a. þ
2
is not in W
b. (T
1
– c
1
I) þ
2
is in W
c. (T
2
– c
2
I) þ
2
is in W.
119
Let V
2
be the set of all n-vectors þ in V
1
such that (T
2
– c
2
I) þ
is in W. Then V
2
is n-invariant under ·. Applying earlier results
to
3 3 3
1 n
U U . . . U = · · the restriction of T
3
to V
2
.
If we continue
in this way we shall reach a n-vector o = þ
r
(not in W)
r r r
1 n
. . . þ = þ · ·þ such that
j j
(T c I) ÷ o is in W; j = 1, 2, …, r.
The following theorem is left as an exercise for the reader.
THEOREM 1.2.49: Let V = V
1
· … · V
n
be a finite (n
1
, …, n
n
)
dimension n-vector space over the n-field F. Let · be a
commuting family of n-triangulable, n-linear operators on V =
V
1
· … · V
n
. There exists an n-ordered basis for V = V
1
· …
· V
n
such that every n-operator in · is represented by a
triangular n-matrix in that n-basis.
In view of this theorem the following corollary is obvious.
COROLLARY 1.2.13: Let · be a commuting family of (n
1
× n
1
,
n
2
× n
2
, …, n
n
× n
n
) n-matrices (square), i.e., (n-mixed square
matrices) over a special algebraically closed n-field F = F
1
·
… · F
n
. There exists a non-singular (n
1
× n
1
, n
2
× n
2
, …, n
n
×
n
n
) n-matrix P = P
1
· P
2
· … · P
n
with entries in F = F
1
· …
· F
n
such that
1 1 1
1 1 1
. . .
÷ ÷ ÷
= · ·
n n n
P A P P A P P A P in n-upper
triangular for every n-matrix A = A
1
· … · A
n
in F.
Next we prove the following theorem.
THEOREM 1.2.50: Let · = ·
1
· … · ·
n
be a commuting
family of n-diagonalizable n-linear operators on a finite
dimensional n-vector space V. There exists an n-ordered basis
for V such that every n-operator in · is represented in that n-
basis by a n-diagonal matrix.
Proof: We give the proof by induction on the (n
1
, n
2
, …, n
n
)
dimension of the n-vector space V = V
1
· … · V
n
. If n-dim V
= (1, 1, …, 1) we have nothing to prove. Assume the theorem
for n-dim V < (n
1
, …, n
n
) where V is given as (n
1
, …, n
n
)
120
dimensional space. Choose any T = T
1
· … · T
n
in · which is
not a scalar multiple of n-identity.
Let
1 2 n
1 1 2 2 n n
1 k 1 k 1 k
{c , . . ., c } {c , . . ., c } . . . {c , . . ., c } · · · be distinct
n-character values of T and for each (i(t)) let
t
i
W be the null
space of the n-null space
i 1 n
i i
W W . . . W = · · ; t = 1, 2, …, n of
(T – c
i
I
i
) = (T
1
– C
1
I
1
) · … · (T
n
– c
n
I
n
). Fix i, then
t
i
W is
invariant under every operator that commutes with T
t
. Let
t
i
{ } · be the family of linear operators on
t
i
W obtained by
restricting the operators in ·
t
where · = ·
1
· … · ·
n
; 1s t s
n to the subspace
t
i
W . Each operator in
t
i
· is diagonalizable
because its minimal polynomial divides the minimal polynomial
for the corresponding operator in ·
t
since dim
t
i
W < dimV
i
, the
operators in
t
i
· can be simultaneously diagonalized.
In other words
t
i
W has a basis
t
i
B which consists of vectors
which are simultaneously characteristic vectors for every
operator
t
i
· . Since this is true for every t, t = 1, 2, …, n we see
T = T
1
· … · T
n
is n-diagonalizable and
1 2 n
1 1 2 2 n n
1 k 1 k 1 k
B {B , . . ., B } {B , . . ., B } . . . {B , . . ., B } = · · ·
is a n-basis for the n-vector space V= V
1
· … · V
n
.
Let V = V
1
· … · V
n
be a n-vector space over the n-field, F =
F
1
· … · F
n
. Let
1 n
1 1 n n
1 k 1 k
W {W , . . . W } . . . {W , . . ., W } = · · be
n-subspaces of the n-vector space V. We say that
t
t t
1 k
W , . . ., W
are independent if
t
1 t t t
1 k i i
. . . 0, in W o + + o = o implies that each
t
i
0 o = i.e., if
1 n
1 1 n n
1 k 1 k
( . . . ) . . . ( . . . ) o + + o · · o + + o = 0 · …
· 0 implies each
t
i t
0, 1 i k ; o = s s t = 1, 2, …, n; then
1
1 1
1 k
{W , . . ., W } · … ·
n
n n
1 k
{W , . . ., W }
is said to be n-independent.
The following lemma would be useful for developing more
121
LEMMA 1.2.10: Let V = V
1
·…· V
n
be a finite dimensional (n
1
,
n
2
, …, n
n
) n-vector space of type II. Let
1
1 1
1
{ , . . ., }
k
W W · … ·
1
{ , . . ., }
n
n n
k
W W be n-subspaces of V and let W = W
1
· … · W
n
=
1 2
1 1 2 2
1 1 1
. . . . . . . . . . . . + + · + + · · + +
n
n n
k k k
W W W W W W . Then the
following are equivalent
a.
1
1 1
1 1
{ , . . ., } . . . { , . . ., } · ·
n
n n
k k
W W W W are n-independent i.e.,
1
{ , . . ., }
t
t t
k
W W are independent for t = 1, 2, …, n.
b. For each j
t
, 2 s j
t
s k
t
, t = 1, 2, …, n we have
1
1
( . . . )
÷
+ +
t t
t t t
j j
W W W = {0} for t = 1, 2, …, n.
c. If
t
i
B is an ordered basis for
t
i
W , 1 s i s k
t
; t = 1, 2, …, n
then the n-sequence
1
1 1
1 1
{ , . . ., } . . . { , . . ., } · ·
n
n n
k k
B B B B is an
n-ordered basis for n-subspace W = W
1
· … · W
n
=
2
1 1
1 1
. . . . . . . . . + + · · + +
n
n n
k k
W W W W .
Proof: Assume (a) Let
t t 1
t t t t
j 1 j
W (W . . . W )
÷
o e + + , then there
are vectors
t 1
t t
1 j
, . . .,
÷
o o with
t t
i i
W o e such that
t 1
t t
1 j
. . .
÷
o + + o +
o
t
= 0 + … + 0 = 0 and since
t
t t
1 k
W , . . ., W are independent it
must be that
t 1
t t t t
1 2 j
. . . 0
÷
o = o = = o = o = . This is true for each t;
t = 1, 2, …, n. Now let us observe that (b) implies (a). Suppose
0 =
t
t t t t
1 k i i
. . . ; W o + + o o e ; i = 1, 2, …, k
t
. (We denote both the
zero vector and zero scalar by 0). Let j
t
be the largest integer i
t
such that
t
i
0. o = Then
t t
t t t
i j j
0 . . . ; 0 = o + + o o = thus
t t 1
t t t
j 1 j
. . .
÷
o = ÷o ÷ ÷ o
is a non zero vector in
t t 1
t t t
j 1 j
W (W . . . W )
÷
+ + . Now that we
know (a) and (b) are the same let us see why (a) is equivalent to
(c). Assume (a). Let
t
i
B be a basis of
t
i t
W ; 1 i k s s , and let
t
t t t
1 k
B { , . . ., B } = B true for each t, t = 1, 2, …, n.
122
Any linear relation between the vector in B
t
will have the
form
t
t t
1 k
. . . þ + + þ = 0 where
t
i
þ is some linear combination of
vectors in
t
i
B . Since
t
t t
1 k
W , . . . , W are independent each of
t
i
þ is
0. Since each
t
i
B is an independent relation. The relation
between the vectors in B
t
is trivial. This is true for every t, t = 1,
2, …, n; so in
B = B
1
· … · B
n
=
1 n
1 1 n n
1 k 1 k
{B , . . ., B } . . . {B , . . ., B } · ·
every n-relation in n-vectors in B is the trivial n-relation.
It is left for the reader to prove (c) implies (a).
If any of the conditions of the above lemma hold, we say
the n-sum W =
1 n
1 1 n n
1 k 1 k
(W . . . W ) . . . (W . . . W ) + + · · + + ; n-
direct or that W is the n-direct sum of
1
1 1
1 k
{W , . . ., W } · … ·
n
n n n
1 2 k
{W , W , . . ., W } and we write
1 n
1 1 n n
1 k 1 k
W (W . . . W ) . . . (W . . . W ) = © © · · © © .
This n-direct sum will also be known as the n-independent
sum or the n-interior direct sum of
1
1 1
1 k
{W , . . . , W } · … ·
n
n n
1 k
{W , . . ., W }.
Let V = V
1
· … · V
n
be a n-vector space over the n-field F
= F
1
·… · F
n
. A n-projection of V is a n-linear operator E = E
1
· … · E
n
on V such that E
2
=
2 2
1 n
E ... E · · = E
1
· … · E
n
=
E.
Since E is a n-projection. Let R = R
1
· … · R
n
be the n-
range of E and let N = N
1
· … · N
n
be the null space of E.
1. The n-vector þ = þ
1
· … · þ
n
is the n-range R if and
only if Eþ = þ. If þ = Eo then Eþ = E
2
o = Eo = þ.
Conversely if þ = Eþ then of course þ, is in the n-range
of E.
2. V = R © N i.e., V
1
· … · V
n
= R
1
1
· … · R
n
N
n
, i.e., each V
i
= R
i
i
; i = 1, 2, …, n.
123
3. The unique expression for o as a sum of n-vector in R
and N is o = Eo + (o – Eo), i.e., o
1
· … · o
n
= E
1
o
1
+
(o
1
– E
1
o
1
) · … · E
n
o
n
+ (o
n
– E
n
o
n
).
From (1), (2) and (3) it is easy to verify. If R = R
1
· … · R
n
and N = N
1
· … · N
n
are n-subspaces of V such that V = R ©
N = R
1
1
· … · R
n
n
there is one and only one n-
projection operator E = E
1
· … · E
n
which has n-range R and
n-null space N. That operator is called the n-projection on R
along N. Any n-projection E = E
1
·…· E
n
is trivially n-
diagonalizable. If
1 n
1 1 n n
1 r 1 r
{ , . . . , } . . . { , . . ., } o o · · o o is a n-basis
of R and
{ } { }
1 1 n n
1 1 n n
r 1 n r 1 n
,..., ... ,...,
+ +
o o · · o o a n-basis for N then
the n-basis B =
1 n
1 1 n n
1 n 1 n
{ , . . . , } . . . { , . . ., } o o · · o o = B
1
· … ·
B
n
, n-diagonalizes E = E
1
· … · E
n
.
1 n
B 1 B n B
[E] [E ] [E ] = · · !
1 n
I 0 I 0
. . . ,
0 0 0 0

= · ·

where I
1
is a r
1
× r
1
identity matrix so on and I
n
is a r
n
× r
n
identity matrix.
Projections can be used to describe the n-direct sum
decomposition of the n-space V = V
1
· … · V
n
. For suppose
V =
1 n
1 1 n n
1 k 1 k
for each j(t) we define
t
j
E on V
t
. Let o be in V = V
1
· … · V
n
say
o =
1 n
1 1 n n
1 k 1 k
( . . . ) . . . ( . . . ) o + + o · · o + + o
with
t
i
o in
t
i
W ; 1 s i s k
t
and t = 1, 2,…, n. Define
t t t
j j
E o = o ;
then
t
j
E is a well defined rule. It is easy to see that
t
j
E is linear
and that range of
t
j
E is
t
j
W and
t 2
j
(E ) =
t
j
E . The null space of
t
j
E is the subspace
t t t t
1 j 1 j 1 k
W . . . W W . . . W
÷ +
+ + + + + for the
statement
t t
j
E o = 0 simply means
t
j
o = 0 i.e., o is actually a
124
sum of vectors from the spaces
t
i
W with i = j. In terms of the
projections
t
j
E we have o
t
=
t t t t
1 k
E . . . E o + + o for each o in V.
The above equation implies;
t t
t 1 k
I E . . . E = + + . Note also that if i
= j then
t t
i j
E E = 0 because the range of
t
j
E is the subspace
t
j
W
which is contained in the null space of
t
i
E . This is true for each
t, t = 1, 2, …, n. Hence true on the n-vector space
V =
1 n
t 1 n n
i k 1 k
Now as in case of n-vector spaces of type I we in case of n-
vector spaces of type II obtain the proof of the following
theorem.
THEOREM 1.2.51: If V = V
1
· … · V
n
is a n-vector space of
type II and suppose
V =
1
1 1
1 1
n
n n
k k
W W W W
then there exists (k
1
, …, k
n
); n-linear operators
1
1 1
1
{ , . . ., }
k
E E ·
… ·
1
{ , . . ., }
n
n n
k
E E on V such that
a. each
t
i
E is a projection, i.e.,
2
( )
t
i
E =
t
i
E for t = 1, 2,
…, n; 1 s i s k
t
.
b.
t t
i j
E E = 0 if i = j; 1 s i, j s k
t
; t = 1, 2, …, n.
c. I = I
1
· … · I
n
=
1
1 1
1 1
( . . . ) . . . ( . . . ) + + · · + +
n
n n
k k
E E E E
d. The range of
t
i
E is
t
i
W i = 1, 2,…, k
t
and t = 1, 2, …, n.
Proof: We are primarily interested in n-direct sum n-
decompositions V =
1 n
1 1 n n
1 k 1 k
= W
1
· … · W
n
; where each of the n-subspaces W
t
is invariant
under some given n-linear operator T = T
1
· … · T
n
.
Given such a n-decomposition of V, T induces n-linear
operators
1 n
i i
{T . . . T } · · on each
1 n
i i
W . . . W · · by restriction.
The action of T is, if o is a n-vector in V we have unique n-
125
vectors
1 n
1 1 n n
1 k 1 k
{ , . . ., } . . . { , . . ., } o o · · o o with
t
i
o in
t
i
W such
that
1 n
1 1 n n
1 k 1 k
. . . . . . . . . o = o + + o · ·o + + o and then
1 1 n n
1 1 1 1 n n n n
1 1 k k 1 1 k k
T T . . . T . . . T . . . T o = o + + o · · o + + o .
We shall describe this situation by saying that T is the n-direct
sum of the operators
1 n
1 1 n n
1 k 1 k
{T , . . ., T } . . . {T , . . ., T } · · . It must
be remembered in using this terminology that the
t
i
T are not n-
linear operators on the space V = V
1
· … · V
n
but on the
various n-subspaces
V = W
1
· … · W
n
.
=
1 n
1 1 n n
1 k 1 k
which enables us to associate with each o = o
1
· … · o
n
in V
a unique n, k-tuple
1 n
1 1 n n
1 k 1 k
( , . . ., ) . . . ( , . . ., ) o o · · o o of vectors
t
i
o e
t
i
W ; i = 1, 2, …, k
t
; t = 1, 2, …, n.
1 n
t 1 n n
i k 1 k
( . . . ) . . . ( . . . ) o = o + + o · · o + + o
in such a way that we can carry out the n-linear operators on V
by working in the individual n-subspaces
1 n
i i i
W W . . . W = + + .
The fact that each W
i
is n-invariant under T enables us to view
the action of T as the independent action of the operators
t
i
T on
the n-subspaces
t
i
W ; i = 1, 2, …, k
t
and t = 1, 2, …, n. Our
purpose is to study T by finding n-invariant n-direct sum
decompositions in which the
t
i
T are operators of an elementary
nature.
As in case of n-vector spaces of type I the following theorem
can be derived verbatim, which is left for the reader.
THEOREM 1.2.52: Let T = T
1
· … · T
n
be a n-linear operator
on the n-space V = V
1
· … · V
n
of type II. Let
1
1 1
1 1
{ , . . ., } . . . { , . . ., } · ·
n
n n
k k
W W W W
and
1
1 1
1 1
{ , . . ., } . . . { , . . ., } · ·
n
n n
k k
E E E E
126
be as before. Then a necessary and sufficient condition that
each n-subspace
t
i
W ; to be a n-invariant under T
t
for 1 s i s k
t
is that =
t t
i t i t
E T E T or TE = ET for every 1 s i s k
t
and t = 1, 2,
…, n.
Now we proceed onto define the notion of n-primary
decomposition for n-vector spaces of type II over the n-field F =
F
1
· … · F
n
.
Let T be a n-linear operator of a n-vector space V = V
1
· …
· V
n
of (n
1
, n
2
, …, n
n
) dimension over the n-field F. Let p = p
1
· … · p
n
be a n-minimal polynomial for T; i.e.,
p =
1 n
1 n
k k
1 1 1 n
1 n
r r
r r 1 1 n n
1 k 1 k
(x c ) . . . (x c )x . . . (x c ) . . . (x c )x ÷ ÷ · · ÷ ÷
where
1 n
1 1 n n
1 k 1 k
{c , . . ., c } . . . {c , . . ., c } · · are distinct elements of F
= F
1
· … · F
n
, i.e.,
t
t t
1 k
{c , . . ., c } are distinct elements of F
t
, t =
1, 2, …, n, then we shall show that the n-space V = V
1
· … ·
V
n
is the n-direct sum of null spaces
s
i
r s
s i s
(T c I ) ÷ , i = 1, 2, …, k
s
and s = 1, 2, …, n.
The hypothesis about p is equivalent to the fact that T is n-
triangulable.
Now we proceed on to give the primary n-decomposition
theorem for a n-linear operator T on the finite dimensional n-
vector space V = V
1
· … · V
n
over the n-field F = F
1
· … ·
F
n
of type II.
THEOREM 1.2.53: (PRIMARY N-DECOMPOSITION THEOREM) Let
T = T
1
· … · T
n
be a n-linear operator on a finite (n
1
, n
2
, …,
n
n
) dimensional n-vector space V = V
1
· … · V
n
over the n-
field F = F
1
· … · F
n
. Let p = p
1
· … · p
n
be the n-minimal
polynomial for T.
1
1
1 1 1
11 1
1 1
. . . . . . . . . = · ·
n
n
k k
n
n
r r
r r
k n nk
p p p p p where
i
t
t
p
are distinct irreducible monic polynomials over F
t
, i = 1, 2, …,
k
t
and t = 1, 2, …, n and the
t
i
r are positive integers. Let
1
. . . = · ·
n
i i i
W W W be the null space of
p(T) =
1
1
1
( ) . . . ( ) ; · ·
n
i i
r r n
i i ni i
p T p T
127
i = 1, 2, …, n. Then
i. V = W
1
· … · W
n
=
1
1 1
1 1
n
n n
k k
W W W W
ii. each
1
. . . = + +
n
i i i
W W W is n-invariant under T; i = 1, 2,
…, n.
iii. if
r
i
T is the operator induced on
r
i
W by T
i
then the
minimal polynomial for
r
i
T is r = 1, 2, …, k
i
; i = 1, 2,
…, n.
The proof is similar to that of n-vector spaces of type I and
hence left as an exercise for the reader. In view of this theorem
we have the following corollary the proof of which is also
direct.
COROLLARY 1.2.14: If
1
1 1
1 1
{ , . . ., } . . . { , . . . , } · ·
n
n n
k k
E E E E are
the n-projections associated with the n-primary decomposition
of T = T
1
· … · T
n
, then each
t
i
E is a polynomial in T; 1 s i
s k
t
and t = 1, 2, …, n and accordingly if a linear operator U
commutes with T then U commutes with each of the E
i
, i.e., each
subspace W
i
is invariant under U.
Proof: We can as in case of n-vector spaces of type I define in
case of n-linear operator T of type II, the notion of n-diagonal
part of T and n-nilpotent part of T.
Consider the n-minimal polynomial for T, which is the
product of first degree polynomials i.e., the case in which each
p
i
is of the form
t t
i i
p x c = ÷ . Now the range of
t
i
E is the null
space of
t
i
W of
t
i
r t
t i t
(T c I ) ÷ ; we know by earlier results D is a
diagonalizable part of T. Let us look at the n-operator
N = T – D
= (T
1
– D
1
) · … · (T
n
– D
n
).
T =
1
1 1
1 1 1 k
(T E . . . T E ) + + ·
2
2 2
2 1 2 k
(T E . . . T E ) . . . + + · ·
n
n n
n 1 n k
(T E . . . T E ) + +
and
128
D =
1 1
1 1 1 1
1 1 k k
(c E . . . c E ) . . . + + · ·
n n
n n n n
1 1 k k
(c E . . . c E ) + +
so
N =
1 1
1 1 1 1
1 1 1 1 1 k 1 k
{(T c I ) E . . . (T c I ) E } ÷ + + ÷ · … ·
1 n
n n n n
n 1 n 1 n k n k
{(T c I ) E . . . (T c I ) E } ÷ + + ÷ .
The reader should be familiar enough with n-projections by now
so that
N
2
=
1 1
1 2 1 1 2 1
1 1 1 1 1 k 1 k
{(T c I ) E . . . (T c I ) E } ÷ + + ÷ · … ·
n n
n 2 n n 2 n
n 1 n 1 n k n k
{(T c I ) E . . . (T c I ) E } ÷ + + ÷ ;
and in general,
N
r
=
1 1
1 1
r r 1 1 1 1
1 1 1 1 1 k 1 k
{(T c I ) E . . . (T c I ) E } ÷ + + ÷ · … ·
n n
n n
r r n n n n
n 1 n 1 n k n k
{(T c I ) E . . . (T c I ) E } ÷ + + ÷ .
When r > r
i
for each i we have N
r
= 0 because each of the
n-operators
t
r t
t i t
(T c I ) ÷ = 0 on the range of
t
i
E ; 1 s t s k
i
and i
= 1, 2, …, n. Thus (T – c I)
r
= 0 for a suitable r.
Let N be a n-linear operator on the n-vector space V = V
1
·
… · V
n
. We say that N is n-nilpotent if there is some n-positive
integer (r
1
, …, r
n
) such that
i
r
i
N = 0. We can choose r > r
i
for i =
1, 2, …, n then N
r
= 0, where N = N
1
· … · N
n
.
In view of this we have the following theorem for n-vector
space type II, which is similar to the proof of n-vector space of
type I.
THEOREM 1.2.54: Let T = T
1
· … · T
n
be a n-linear operator
on the (n
1
, n
2
, …, n
n
) finite dimensional vector space V = V
1
·
… · V
n
over the field F. Suppose that the n-minimal polynomial
for T decomposes over F = F
1
· … · F
n
into a n-product of n-
linear polynomials. Then there is a n-diagonalizable operator D
= D
1
· … · D
n
on V = V
1
· … · V
n
and a nilpotent operator
N = N
1
· … · N
n
on V = V
1
· … · V
n
such that
(i) T = D + N.
(ii) D N = N D.
129
The n-diagonalizable operator D and the n-nilpotent
operator N are uniquely determined by (i) and (ii) and each of
them is n-polynomial in T
1
, …, T
n
.
Consequent of the above theorem the following corollary is
direct.
COROLLARY 1.2.15: Let V be a finite dimensional n-vector
space over the special algebraically closed field F = F
1
· … ·
F
n
. Then every n-linear operator T = T
1
· … · T
n
on V = V
1
·
… · V
n
can be written as the sum of a n-diagonalizable
operator D = D
i
· … · D
n
and a n-nilpotent operator N = N
1
· … · N
n
which commute. These n-operators D and N are
unique and each is a n-polynomial in (T
1
, …, T
n
).
Let V = V
1
· … · V
n
be a finite dimensional n-vector space
over the n-field F = F
1
· … · F
n
and T = T
1
· … · T
n
be an
arbitrary and fixed n-linear operator on V. If o = o
1
· … · o
n
is n-vector in V, there is a smallest n-subspace of V which is n-
invariant under T and contains o. This n-subspace can be
defined as the n-intersection of all T-invariant n-subspaces
which contain o; however it is more profitable at the moment
for us to look at things this way.
If W = W
1
· … · W
n
be any n-subspace of the n-vector
space V = V
1
· … · V
n
which is n-invariant under T and
contains o = o
1
· … · o
n
i.e., each T
i
in T is such that the
subspace W
i
in V
i
is invariant under T
i
and contains o
i
; true for
i = 1, 2, …, n.
Then W must also contain To i.e., T
i
o
i
is in W
i
for each i =
1, 2, …, n, hence T(To) is in W i.e., T
i
(T
i
o
i
) =
2
i i
T o is in W so
on; i.e.,
i
m
i i
T ( ) o is in W
i
for each i so T
m
(o) e W i.e., W must
contain g(T)o for every n-polynomial g = g
1
· … · g
n
over the
n-field F = F
1
· … · F
n
. The set of all n-vectors of the form
g(T)o = g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
with g e F [x] = F
1
[x] · …
· F
n
[x], is clearly n-invariant and is thus the smallest n-T-
invariant (or T-n-invariant) n-subspace which contains o. In
130
view of this we give the following definition for a n-linear
operator on V.
DEFINITION 1.2.30: If o = o
1
· … · o
n
is any n-vector in V =
V
1
· … · V
n
, a n-vector space over the n-field F = F
1
· … ·
F
n
. The T-n-cyclic n-subspace generated by o is a n-subspace
Z(o; T) = Z(o
1
; T
1
) · … · Z(o
n
; T
n
) of all n-vectors g(T)o =
g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
; g = g
1
· … · g
n
in F[x] = F
1
[x] ·
… · F
n
[x]. If Z(o; T) = V then o is called a n-cyclic vector for
T.
Another way of describing this n-subspace Z(o; T) is that
Z(o; T) is the n-subspace spanned by the n-vectors
k
T
o
; k > 0
and thus o is a n-cyclic n-vector for T if and only if these n-
vectors span V; i.e., each
i
i
k
i
T
o
span V
i
, k
i
> 0 and thus o
i
is a
cyclic vector for T
i
if and only if these vectors span V
i
, true for i
= 1, 2, …, n.
We caution the reader that the general n-operator T has no n-
cyclic n-vectors.
For any T, the T n-cyclic n-subspace generated by the n-
zero vector is the n-zero n-subspace of V. The n-space Z(o; T)
is (1, …, 1) dimensional if and only if o is a n-characteristic
vector for T. For the n-identity operator, every non zero n-
vector generates a (1, 1, …, 1) dimensional n-cyclic n-subspace;
thus if n-dim V > (1, 1, …, 1) the n-identity operator has non-
cyclic vector. For any T and o we shall be interested in the n-
linear relations
k
0 1 k
c c T . . . c T 0 o + o + + o = ; where o = o
1
·
… · o
n
so
1
1
k 1 1 1
0 1 1 1 1 k 1 1
c c T . . . c T 0 o + o + + o = ,
2
2
k 2 2 2
0 2 1 2 2 k 2 2
c c T . . . c T 0 o + o + + o =
so on;
n
n
k n n n
0 n 1 n n k n n
c c T . . . c T 0 o + o + + o = ;
between the n-vectors To
j
, we shall be interested in the n-
polynomials g = g
1
· … · g
n
where
i
i
k i i i
i 0 1 k
g c c x . . . c x = + + +
true for i = 1, 2, …, n. which has the property that g(T)o = 0.
131
The set of all g in F[x] = F
1
[x] · … · F
n
[x] such that g(T)o =0
is clearly an n-ideal in F[x]. It is also a non zero n-ideal in F[x]
because it contains the n-minimal polynomial p = p
1
· … · p
n
of the n-operator T. p(T)o = 0 · … · 0, i.e., p
1
(T
1
)o
1
· … ·
p
n
(T
n
)o
n
= 0 · … · 0 for every o = o
1
· … · o
n
in V = V
1
·
… · V
n
).
DEFINITION 1.2.31: If o = o
1
· … · o
n
is any n-vector in V =
V
1
· … · V
n
the T-annihilator (T = T
1
· … · T
n
) of o is the n-
ideal M(o; T) in F[x] = F
1
[x] · … · F
n
[x] consisting of all n-
polynomials g = g
1
· … · g
n
over F = F
1
· … · F
n
such that
g(T)o = 0 · … · 0 i.e., g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
= 0 · … · 0.
The unique monic n-polynomial p
o
= p
1
o
1
· … · p
n
o
n
which n-
generates this n-ideal will also be called the n-T annihilator of
o or T-n-annihilator of o. The n-T-annihilator p
o
n-divides the
n-minimal n-polynomial of the n-operator T. Clearly n-deg (p
o
)
> (0, 0, …, 0) unless o is the zero n-vector.
We now prove the following interesting theorem.
THEOREM 1.2.55: Let o = o
1
· … · o
n
be any non zero n-
vector in V = V
1
· … · V
n
and let p
o
=
1
1
· · !
n
n
p p
o o
be the
n-T-annihilator of o.
(i) The n-degree of p
o
is equal to the n-dimension of the n-
cyclic subspace Z(o; T) = Z(o
1
; T
1
) · … · Z(o
n
; T
n
).
(ii) If the n-degree of p
o
=
1
1
· · !
n
n
p p
o o
is (k
1
, k
2
, …, k
n
)
then the n-vectors o, To, To
2
, …, To
k–1
form a n-basis
for Z(o; T) i.e.,
1
1 2
1 1 1 1 1 1 1
{ , , , . . ., }
÷ k
T T T o o o o ·
2
1 2
2 2 2 2 2 2 2
{ , , , . . ., }
÷ k
T T T o o o o · … ·
1 2
{ , , , . . ., }
÷
n
k
n n n n n n n
T T T o o o o
form a n-basis for Z(o; T) = Z(o
1
; T
1
) · … · Z(o
n
; T
n
)
i.e., Z(o
i
, T
i
) has
1
{ , , . . ., }
÷
i
k
i i i i i
T T o o o as its basis; true
for every i, i = 1, 2, …, n.
132
(iii) If U = U
1
· … · U
n
is a n-linear operator on Z(o, T)
induced by T, then the n-minimal polynomial for U is
p
o
.
Proof: Let g = g
1
· … · g
n
be a n-polynomial over the n-field
F = F
1
· … · F
n
. Write g = p
o
q + r, i.e., g
1
· … · g
n
=
1
1 1
p q
o
+ r
1
·…·
n
n n
p q
o
+ r
n
where p
o
=
1
1
p
o
· … ·
n
n
p
o
for o = o
1
·
… · o
n
, q = q
1
· … · q
n
and r = r
1
· … · r
n
so g
i
=
i
i i
p q
o
+ r
i
true for i = 1, 2, …, n. Here either r = 0 · … · 0 or n-deg r < n-
deg p
o
= (k
1
, …, k
n
).
The n-polynomial p
o
q =
1
1
p
o
q
1
· … ·
n
n
p
o
q
n
is in the T-
n-annihilator of o = o
1
· … · o
n
and so g(T)o = r(T)o. i.e.,
g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
= r
1
(T
1
)o
1
· … · r
n
(T
n
)o
n
. Since r =
r
1
· … · r
n
= 0 · 0 · … · 0 or n-deg r < (k
1
, k
2
, …, k
n
) the n-
vector r(T)o = r
1
T
1
(o
1
) · … · r
n
T
n
(o
n
) is a n-linear
combination of the n-vectors o, To, …, T
k–1
o; i.e., a n-linear
combination on n-vectors o = o
1
· … · o
n
.
1 1 n n
T T . . . T , o = o · · o
2 2 2
1 1 n n
T T . . . T o = o · · o , …,
1 n
k 1 k 1 k 1
1 1 n n
T T . . . T
÷ ÷ ÷
o = o · · o
and since g(T)o = g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
is a typical n-
vector in Z(o; T) i.e., each g
i
(T
i
)o
i
is a typical vector in Z(o
i
; T
i
)
for i = 1, 2, …, n. This shows that these (k
1
, …, k
n
), n-vectors
span Z(o; T).
These n-vectors are certainly n-linearly independent,
because any non-trivial n-linear relation between them would
give us a non zero n-polynomial g such that g(T)(o) = 0 and n-
deg g < n-deg p
o
which is absurd. This proves (i) and (ii).
Let U = U
1
· … · U
n
be a n-linear operator on (Z
o
; T)
obtained by restricting T to that n-subspace. If g = g
1
· … · g
n
is any n-polynomial over F = F
1
· … · F
n
then
p
o
(U)g(T)o = p
o
(T)g(T)o
i.e.,
1 n
1 1 1 1 1 n n n n n
p (U ) g (T ) . . . p (U ) g (T )
o o
o · · o
=
1 n
1 1 1 1 1 n n n n n
p (T ) g (T ) . . . p (T ) g (T )
o o
o · · o
133
= g(T) p
o
(T)o
=
1 n
1 1 1 1 1 n n n n n
g (T ) p (T ) . . . g (T ) p (T )
o o
o · · o
= g
1
(T
1
)0 · … · g
n
(T
n
)0
= 0 · … · 0.
Thus the n-operator p
o
U = p
1
o
1
(U
1
) · … · p
n
o
n
(U
n
) sends
every n-vector in Z(o; T) = Z(o
1
; T
1
) · … · Z(o
n
; T
n
) into 0 ·
0 · … · 0 and is the n-zero operator on Z(o; T). Furthermore
if h = h
1
· … · h
n
is a n-polynomial of n-degree less than (k
1
,
…, k
n
) we cannot have h(U) = h
1
(U
1
) · … · h
n
(U
n
) = 0 · … ·
0 for then h(U)o = h
1
(U
1
)o
1
· … · h
n
(U
n
)o
n
) = h
1
(T
1
)o
1
· …
· h
n
(T
n
)o
n
= 0 · … · 0; contradicting the definition of p
o
.
This show that p
o
is the n-minimal polynomial for U.
A particular consequence of this nice theorem is that if o =
o
1
· … · o
n
happens to be a n-cyclic vector for T = T
1
· … ·
T
n
then the n-minimal polynomial for T must have n-degree
equal to the n-dimension of the n-space V = V
1
· … · V
n
hence by the Cayley Hamilton theorem for n-vector spaces we
have that the n-minimal polynomial for T is the n-characteristic
polynomial for T. We shall prove later that for any T there is a
n-vector o = o
1
· … · o
n
in V = V
1
· … · V
n
which has the
n-minimal polynomial for T = T
1
· … · T
n
for its n-
annihilator. It will then follow that T has a n-cyclic vector if and
only if the n-minimal and n-characteristic polynomials for T are
identical. We now study the general n-operator T = T
1
· … ·
T
n
by using n-operator vector. Let us consider a n-linear
operator U = U
1
· … · U
n
on the n-space W = W
1
· … · W
n
of n-dimension (k
1
, …, k
n
) which has a cyclic n-vector o = o
1
·
… · o
n
.
By the above theorem just proved the n-vectors o, Uo, …,
k 1
U
÷
o i.e.,
k 1
1 1 1 1
{ , U , . . ., U },
÷
o o o
2
k 1
2 2 2 2 2
{ , U , . . ., U }
÷
o o o ,
…,
n
k 1
n n n n n
{ , U , . . ., U }
÷
o o o form a n-basis for the n-space W =
W
1
· … · W
n
and the annihilator p
o
=
1 n
1 n
p . . . p
o o
· · of o =
o
1
· … · o
n
is the n-minimal polynomial for U = U
1
· … ·
U
n
(hence also the n-characteristic polynomial for U).
134
If we let
i i 1
U
÷
o = o; i.e.,
i i i
1 n
. . . o = o · ·o and
i i 1
U
÷
o = o
implies
1 n
i 1 i 1 i i
1 1 1 n n n
U , . . ., U
÷ ÷
o = o o = o ; 1 s i s k
i
– 1 then the
action of U on the ordered n-basis
1
1 1
1 k
{ , . . ., } o o · … ·
n
n n
1 k
{ , . . ., } o o is Uo
i
= o
i+1
for i = 1, 2, …, k – 1 i.e.,
t
i 1 i
t t t
U
+
o = o for i = 1, 2, …, k
t
– 1 and t = 1, 2, …, n. Uo
k
= –
c
0
o
1
– … –c
k–1
o
k
i.e.,
t
k t 1 t k
t t 0 t k 1 t
U c . . . c
÷
o = ÷ o ÷ ÷ o
for t = 1, 2, …, n, where
p
o
=
1 1
1
k 1 k 1 1 1
0 1 k 1
{c c x . . . c x x }
÷
÷
+ + + + · … ·
n n
n
k 1 k n n n
0 1 k 1
{c c x . . . c x x }
÷
÷
+ + + + .
The n-expression for Uo
k
follows from the fact p
o
(U)o = 0
· … · 0 i.e.,
1 n
1 1 1 1 n n
p (U ) . . . p (U )
o o
o · o = 0 · … · 0.
i.e.,
k k 1
k 1 1 0
U c U . . . c U c
÷ o
÷
o + o + + o + = 0 · … · 0,
i.e.,
1 1
1
k k 1 1 1 1
1 1 k 1 1 1 1 1 1 0 1
U c U . . . c U c
÷
÷
o + o + + o + o ·
2 2
2
k k 1 2 2 2
2 2 k 1 2 2 1 2 2 0 2
U c U . . . c U c
÷
÷
o + o + + o + o · … ·
n n
n
k k 1 n n n
n n k 1 n n 1 n n 0 n
U c U . . . c U c
÷
÷
o + o + + o + o
= 0 · … · 0.
This is given by the n-matrix of U = U
1
· … · U
n
in the n-
ordered basis
B = B
1
· … · B
n
=
1 n
1 1 n n
1 k 1 k
{ . . . } . . . { . . . } o o · · o o
=
1
1
0
1
1
1
2
1
k 1
0 0 0 0 c
1 0 0 0 c
0 1 0 0 c
0 0 0 1 c
÷
÷

÷

÷

÷

!
!
!
# # # # #
!
· … ·
135
n
n
0
n
1
n
2
n
k 1
0 0 0 0 c
1 0 0 0 c
0 1 0 0 c
0 0 0 1 c
÷
÷

÷

÷

÷

!
!
!
# # # # #
!
.
The n-matrix is called the n-companion n-matrix of the monic
n-polynomial
1 n
1 n
p p . . . p
o o o
= · · or can also be represented
with some flaw in notation as
1 n
1 n
p . . . p
o o
· · where p = p
1
· …
· p
n
.
Now we prove yet another interesting theorem.
THEOREM 3.2.56: If U = U
1
· … · U
n
is a n-linear operator
on a finite (n
1
, n
2
, …, n
n
) dimensional n-space W = W
1
· … ·
W
n
, then U has a n-cyclic n-vector if and only if there is some n-
ordered n-basis for W in which U is represented by the n-
companion n-matrix of the n-minimal polynomial for U.
Proof: We have just noted that if U has a n-cyclic n-vector then
there is such an n-ordered n-basis for W = W
1
· … · W
n
.
Conversely if there is some n-ordered n-basis
1
1 1
1 k
{ , . . ., } o o ·
… ·
n
n n
1 k
{ , . . ., } o o for W in which U is represented by the n-
companion n-matrix of its n-minimal polynomial it is obvious
that
1 n
1 1
. . . o · ·o is a n-cyclic vector for U.
Now we give another interesting corollary to the above
theorem.
COROLLARY 1.2.16: If A = A
1
· … · A
n
is a n-companion n-
matrix of a n-monic n-polynomial p = p
1
· … · p
n
(each p
i
is
monic) then p is both the n-minimal polynomial and the n-
characteristic polynomial of A.
136
Proof: One way to see this is to let U = U
1
· … · U
n
a n-linear
operator on
1 n
k k
1 n
F . . . F · · which is represented by A = A
1
·
… · A
n
in the n-ordered n-basis and to apply the earlier
theorem and the Cayley Hamilton theorem for n-vector spaces.
Yet another method is by a direct calculation.
Now we proceed on to define the notion of n-cyclic
decomposition or we can call it as cyclic n-decomposition and
its n-rational form or equivalently rational n-form.
Our main aim here is to prove that any n-linear operator T
on a finite (n
1
, …, n
n
) dimensional n-vector space V = V
1
· …
· V
n
, there exits n-set of n-vectors
1
1 1
1 r
{ , . . ., } o o , …,
n
n n
1 r
{ , . . ., } o o in V such that V = V
1
· … · V
n
=
1
1 1
1 1 r 1
Z( ; T ) . . . Z( ; T ) o © © o ·
2
2 2
1 2 r 2
Z( ; T ) . . . Z( ; T ) o © © o ·
… ·
n
n n
1 n r n
Z( ; T ) . . . Z( ; T ). o © © o
In other words we want to prove that V is a n-direct sum of
n-T-cyclic n-subspaces. This will show that T is a n-direct sum
of a n-finite number of n-linear operators each of which has a n-
cyclic n-vector. The effect of this will be to reduce many
problems about the general n-linear operator to similar problems
about an n-linear operator which has a n-cyclic n-vector.
This n-cyclic n-decomposition theorem is closely related to
the problem in which T n-invariant n-subspaces W have the
property that there exists a T- n-invariant n-subspaces W' such
that V = W © W', i.e., V = V
1
· … · V
n
= W
1
1
· … ·
W
n
n
.
If W = W
1
· … · W
n
is any n-subspace of finite (n
1
, …,
n
n
) dimensional n-space V = V
1
· … · V
n
then there exists a n-
subspace
1 1 1
1 n
W W . . . W = + + such that V = W © W'; i.e., V =
V
1
· …· V
n
= (W
1
1
) · … · (W
n
n
) i.e., each V
i
is a
direct sum of W
i
and W'
i
for i = 1, 2, …, n, i.e., V
i
= W
i
n
.
Usually there are many such n-spaces W' and each of these
is called n-complementary to W. We study the problem when a
137
T- n-invariant n-subspace has a complementary n-subspace
which is also n-invariant under the same T.
Let us suppose V = W © W'; i.e., V = V
1
· … · V
n
= W
1
1
· … · W
n
n
where both W and W' are n-invariant
under T, then we study what special property is enjoyed by the
n-subspace W. Each n-vector þ = þ
1
· … · þ
n
in V = V
1
· …
· V
n
is of the form þ = ¸ + ¸' where ¸ is in W and ¸' is in W'
where ¸ = ¸
1
· … · ¸
n
and ¸' = ¸'
1
· … · ¸'
n
.
If f = f
1
· … · f
n
any n-polynomial over the scalar n-field
F = F
1
· … · F
n
then f(T) þ = f(T)¸ + f(T)¸'; i.e., f
1
(T
1

1
· …
· f
n
(T
n

n
= {f
1
(T
1

1
· … · f
n
(T
n

n
} + {f
1
(T
1
)¸'
1
· … · f
n
(T
n
)¸'
n
}
= {f
1
(T
1

1
· … · f
n
(T
n
)¸'
1
} + {f
1
(T
1

n
· … · f
n
(T
n
)¸'
n
}.
Since W and W' are n-invariant under T = T
1
· … · T
n
the n-
vector f(T)¸ = f
1
(T
1

1
· … · f
n
(T
n

n
is in W = W
1
· … · W
n
and f(T)¸' = f
1
(T
1
)¸'
1
· … · f
n
(T
n
)¸'
n
is in W' = W'
1
·… · W'
n
.
Therefore f(T)þ = f
1
(T
1

1
· … · f
n
(T
n

n
is in W if and
only if f(T) ¸' = 0 · … · 0; i.e., f
1
(T
1
)¸'
1
· … · f
n
(T
n
)¸'
n
= 0 ·
… · 0. So if f(T) þ is in W then f(T)þ = f(T)¸.
Now we define yet another new notion on the n-linear operator
T = T
1
· … · T
n
.
DEFINITION 1.2.32: Let T = T
1
· … · T
n
be a n-linear
operator on the n-vector space V = V
1
· … · V
n
and let W =
W
1
· … · W
n
be a n-subspace of V. We say that W is n- T-
1. W is n-invariant under T
2. f(T)þ is in W, for each þ e V,
there exists a n-vector ¸ in W such that f(T)þ = f(T)¸; i.e., if W is
n-invariant and has a n-complementary n-invariant n-subspace
138
Thus n-admissibility characterizes those n-invariant n-
subspaces which have n-complementary n-invariant n-
subspaces.
We see that the n-admissibility property is involved in the n-
decomposition of the n-vector space V = Z(o
1
r
;
T); i.e., V = V
1
·…· V
n
=
1
1 1
1 1 r 1
{Z( ; T ) . . . Z( ; T )} o © © o · … ·
n
n n
1 n r n
{Z( ; T ) . . . Z( ; T )} o © © o .
We arrive at such a n-decomposition by inductively
selecting n-vectors
1
1 1
1 r
{ , . . ., } o o · … ·
n
n 1
1 r
{ , . . ., } o o .
Suppose that by some method or another we have selected
the n-vectors
1 n
1 1 n n
1 r 1 r
{ , . . ., } . . . { , . . ., } o o · · o o and n-subspaces
which is proper say
W
j
=
1 n
j j
W . . . W · ·
=
1
1 1
1 1 j 1
{Z( ; T ) . . . Z( ; T )} o + + o · … ·
n
n n
1 n j n
{Z( ; T ) . . . Z( ; T )} o + + o .
We find the non zero n-vector
1 n
1 n
j 1 j 1
( . . . )
+ +
o · ·o such that
j j 1
W Z( , T) 0 . . . 0
+
o = · ·
i.e.,
1 n
1 1 n n
j j 1 1 j j 1 n
(W Z( ; T )) . . . (W Z( ; T ))
+ +
o · · o
= 0 · … · 0
because the n-subspace.
j 1 j j 1
W W Z( , T)
+ +
i.e.,
1 n
1 n
j 1 j 1 j 1
W W . . . W
+ + +
= · ·
1 n
1 n
j j 1 1 j j 1 n
W Z( ; T ) . . . W Z( ; T )
+ +
would be at least one n-dimension nearer to exhausting V.
But are we guaranteed of the existence of such
1 n
1 n
j 1 j 1 j 1
. . .
+ + +
o = o · ·o .
139
If
1 n
1 1 n n
1 j 1 j
{ , . . ., } . . . { , . . ., } o o · · o o have been chosen so that
W
j
is T-n-admissible n-subspace then it is rather easy to find a
suitable
n
1 n
j 1 j 1
. . .
+ +
o · ·o .
Let W = W
1
· … · W
n
be a proper T-n-invariant n-
subspace. Let us find a non zero n-vector o = o
1
· … · o
n
such that W · Z(o; T) = {0} · … · {0}, i.e., W
1
· Z(o
1
; T
1
)
· … · W
n
· Z(o
n
; T
n
) = {0} · … · {0}. We can choose
some n-vector þ = þ
1
· … · þ
n
which is not in W = W
1
· … ·
W
n
i.e., each þ
i
is not in W
i
; i = 1, 2 ,…, n. Consider the T-n-
conductor S(þ; W) = S(þ
1
; W
1
) ·… · S(þ
n
; W
n
) which consists
of all n-polynomials g = g
1
· … · g
n
such that g(T)þ = g
1
(T
1

1
· … · g
n
(T
n

n
is in W = W
1
· … · W
n
. Recall that the n-
monic polynomial f = S(þ; W); i.e., f
1
· … · f
n
= S(þ
1
; W
1
)
· … · S(þ
n
; W
n
) which n-generate the n-ideals S(þ; W) = S(þ
1
;
W
1
) · … · S(þ
n
; W
n
); i.e., each f
i
= S(þ
I
; W
i
) generate the ideal
S(þ
I
; W
i
) for i = 1, 2, …, n, i.e., S(þ; W) is also the T-n-
conductor of þ into W. The n-vector f(T)þ = f
1
(T
1

1
· … ·
f
n
(T
n

n
is in W = W
1
· … · W
n
. Now if W is T-n-admissible
there is a ¸ = ¸
1
· … · ¸
n
in W with f(T)þ = f(T)¸. Let
o = þ ÷ ¸ and let g be any n-polynomial. Since þ – o is in W,
g(T)þ will be in W if and only if g(T)o is in W; in other words
S(o: W) = S(þ: W). Thus the n-polynomial f is also the T-n
conductor of o into W. But f(T)o = 0 · … · 0. That tells us
f
1
(T
1
)o
1
· … · f
n
(T
n
)o
n
= 0 · … · 0; i.e., g(T)o is in W if
and only if g(T)o = 0 · … · 0 i.e., g
1
(T
1
)o
1
· … · g
n
(T
n
)o
n
=
0 · … · 0. The n-subspaces Z(o; T) = Z(o
1
; T
1
) · … · Z(o
n
;
T
n
) and W = W
1
· … · W
n
are n-independent and f is the T-n-
annihilator of o.
Now we prove the cyclic decomposition theorem for n-linear
operators.
THEOREM 3.2.57: (N-CYCLIC DECOMPOSITION THEOREM): Let
T = T
1
· … · T
n
be a n-linear operator on a finite dimensional
(n
1
, …, n
n
) n-vector space V = V
1
· … · V
n
and let W
o
=
1
o
W ·
… ·
n
o
W be a proper T-n-admissible n-subspace of V. There
140
exists non zero n-vectors
{ }
1
1 1
1
· · ! !
r
o o
{ }
1
!
n
n n
r
o o in V with
respective T-n-annihilators
{ }
1
1 1
1
!
r
p p · … ·
{ }
1
!
n
n n
r
p p such
that
(i) V = W
o
1
r
; T) =
( ) ( ) { }
1
1 1 1
1 1 1
o r
W Z T Z T o o ·
( ) ( ) { }
2
2 2 2
1 2 2
o r
W Z T Z T o o · … ·
( ) ( ) { } 1
n
n n n
o n r n
W Z T Z T o o .
(ii)
t
t
k
p divides
1 ÷
t
t
k
p , k = 1, 2,…, r; t = 1, 2, … , n.
Further more the integer r and the n-annihilators
{ }
1
1 1
1
, ...,
r
p p
· … ·
{ }
1
,...,
n
n n
r
p p are uniquely determined by (i) and (ii) and
infact that no
t
t
k
o is zero, t = 1, 2, …, n.
Proof: The proof is rather very long, hence they are given under
four steps. For the first reading it may seem easier to take W
0
=
{0} · … · {0} =
1
0
W · … ·
n
0
W ; i.e., each
i
0
W = {0} for i =
1, 2, …, n, although it does not produce any substantial
simplification. Throughout the proof we shall abbreviate f(T)þ
to fþ. i.e., f
1
(T
1

1
· … · f
n
(T
n

n
to f
1
þ
1
· … · f
n
þ
n
.
STEP I:
There exists non-zero n-vectors
{ }
1
1 1
1 r
þ þ ! · … ·
{ }
n
n n
1 r
þ þ !
in V = V
1
· … · V
n
such that
1. V = W
0
+ Z(þ
1
; T) + … + Z(þ
r
; T)
=
( ) ( ) { }
1
1 1 1
0 1 1 r 1
W Z ; T ... Z ;T + þ + + þ ·
( ) ( ) { }
2
2 2 2
0 1 2 r 2
W Z ;T ... Z ;T + þ + + þ · … ·
141
( ) ( ) { }
n
n n n
0 1 n r n
W Z ;T ... Z ;T + þ + + þ

2. if 1 s k
i
s r
i
; i = 1, 2, …, n and
W
k
=
n
k k
n
W W + +"
1
1
=
( ) ( ) { }
1
1 1 1
o 1 1 k 1
W Z ;T Z ;T + þ + + þ " ·
( ) ( ) { }
2
2 2 2
o 1 2 k 2
W Z ;T Z ;T + þ + + þ " · … ·
( ) ( ) { }
n
n n n
o 1 n k n
W Z ; T Z ;T + þ + + þ "
then the n-conductor
p
k
=
1 n
1 n
k k
p p · · !
= S
( )
1 1
1 1
k k 1
; W
÷
þ · … · S
( )
n n
n n
k k 1
; W
÷
þ
has maximum n-degree among all T-n-conductors into the n-
subspace W
k–1
=
( )
1 n
1 n
k 1 k 1
W W
÷ ÷
· · ! that is for every (k
1
, …,
k
n
);
n-deg p
k
=
( ) ( )
1
1
1
1 1
k 1
in V
maxdeg S ; W
÷
o
o · … ·
( ) ( )
n
n
n
n n
k 1
in V
maxdeg S ; W
÷
o
o .
This step depends only upon the fact that W
0
=
( )
1 n
0 0
W W · · ! is an n-invariant n-subspace. If W = W
1
· …
· W
n
is a proper T-n-invariant n-subspace then
0 <
( ) ( )
max ndeg S ; W n dimV
o
o s
i.e.,
0 · … · 0 <
( ) ( )
1
1 1
maxdeg S ; W
o
o ·
( ) ( )
2
2 2
maxdeg S ; W
o
o
· … ·
( ) ( )
n
n n
maxdeg S ; W
o
o
s (n
1
, ..., n
n
)
142
and we can choose a n-vector þ = þ
1
· … · þ
n
so that n-
deg(S(þ: W)) = deg(S(þ
1
, W
1
)) · … · deg(S(þ
n
; W
n
)) attains
that maximum. The n-subspace W + Z(þ: T) = (W
1
+ Z(þ
1
; T
1
))
· … · (W
n
+ Z(þ
n
; W
n
)) is then T-n-invariant and has n-
dimension larger than n-dim W. Apply this process to W = W
0
to obtain
1 1
1 1 n
þ = þ · ·þ " . If W
1
= W
0
+ Z(þ
1
; T), i.e.,
( ) ( )
1 1 1 1
1 n 0 1 1
W ... W W Z ; T · · = + þ · … ·
( ) ( )
n n
0 n n
W Z ; T + þ
is still proper then apply the process to W
1
to obtain þ
2
=
1
2
þ ·
… ·
n
2
þ . Continue in that manner.
Since n-dim W
k
> n-dim W
k-1
i.e., dim
1
1
k
W · … · dim
n
n
k
W >
dim
1
1
k 1
W
÷
· … · dim
n
n
k 1
W
÷
we must reach W
r
= V; i.e.,
1
1
r
W ·
… ·
n
n
r
W = V
1
· … · V
n
in not more than n-dim V steps.
STEP 2:
Let
{ }
1
1 1
1 r
, , þ þ ! · … ·
{ }
n
n n
1 r
, , þ þ ! be a n-set of nonzero n-
vectors which satisfy conditions (1) and (2) of Step 1. Fix (k
1
,
…, k
n
); 1 s k
i
s r
i
; i = 1, 2, …, n. Let þ = þ
1
· … · þ
n
be any n-
vector in V = V
1
· … · V
n
and let f = S(þ; W
k–1
) i.e., f
1
· …
· f
n
= S(þ
1
;
1
1
k 1
W
÷
) · … · S(þ
n
;
n
n
k 1
W
÷
).
If fþ = þ
0
+
i i
1 i k
g
s s
þ
¯
i.e., f
1
þ
1
· … · f
n
þ
n
=
1 1 2 2
1 1 2 2
1 1 1 2 2 2
0 i i 0 i i
1 i k 1 i k
g g
s s s s
| | | |
þ + þ · þ + þ
| |
| |
\ . \ .
¯ ¯
· … ·
n n
n n
n n n
0 i i
1 i k
g
s s
| |
þ + þ
|
|
\ .
¯
t t
t
i i
W þ e ; t = 1, 2 ,…, n, then f = f
1
· … · f
n
n-divides each n-
polynomial g
i
=
n
1 n
i i
g ... g · · and þ
0
= f¸
o
i.e.,
1 n
0 0
... þ · ·þ =
1 n
1 0 n 0
f ... f ¸ · · ¸ where ¸
0
=
1 n
0 0
... ¸ · ·¸ e W
0
=
1 n
0 0
W ... W · · .
If each k
i
= 1, i = 1, 2, …, n, this is just the statement that W
o
is
T-n-admissible. In order to prove this assertion for (k
1
, …, k
n
) >
143
(1, 1, …, 1) apply the n-division algorithms g
i
= fh
i
+ r
i
, r
i
= 0 if
n-deg r
i
< n-deg f i.e.,
( )
1 n 1 1
1 n 1 1
i i i i i
g g f h r · · = + " · … ·
( )
n n
n n
n i i
f h r ; +
r
i
= (0 · … · 0) if n-deg r
i
< n-deg f.
We wish to show that r
i
= (0 · … · 0) for each i = (i
1
,…, i
n
).
Let
¸ =
k 1
i i
1
h
÷
þ ÷ þ
¯
;
i.e.,
¸
1
· … · ¸
n
=
1
1 1
k 1
1 1
1 i i
1
h
÷
| |
þ ÷ þ ·
|
\ .
¯
… ·
n
n n
k 1
n n
n i i
1
h
÷
| |
þ ÷ þ
|
\ .
¯
.
Since ¸ – þ is in W
k–1
i.e., (¸
1
– þ
1
) · … · (¸
n
– þ
n
) is in
1
1
k 1
W
÷
· … ·
n
n
k 1
W
÷
. Since
S
( ) ( )
i i
i i
i k 1 i k 1
; W S ; W
÷ ÷
¸ = þ
i.e.,
( ) ( )
1 n
1 n
1 k 1 n k 1
S ; W S ; W
÷ ÷
¸ · · ¸ !
=
( ) ( )
1 n
1 n
1 k 1 n k 1
S ; W S ; W
÷ ÷
þ · · þ "
= f
1
· … · f
n.
S(¸; W
k–1
) = S(þ; W
k–1
) = f.
Further more f¸ =
k 1
0 i i
1
÷
þ + ¸ þ
¯
i.e.,
f
1
¸
1
· … · f
n
¸
n
=
1
1 1
1
k 1
1 1 1
0 i i
i
÷
| |
þ + ¸ þ
|
|
\ .
¯
· … ·
n
n n
n
k 1
n n n
0 i i
i
÷
| |
þ + ¸ þ
|
|
\ .
¯
r
j
=
( )
1 n
1 n
j j
r , , r = ! (0, …, 0) and n-deg r
j
< n-deg f. Let p = S(¸;
W
j-1
); j = (j
1
, …, j
n
). i.e.,
p
1
· … · p
n
= S
( )
1
1
1 j 1
, W
÷
¸ · … · S
( )
n
n
n j 1
, W
÷
¸ .
Since
144
W
k–1
=
1 n
1 n
k 1 k 1
W W
÷ ÷
· · !
contains
W
j–1
=
1 n
1 n
j 1 j 1
W W
÷ ÷
· · ! ;
the n-conductor f = S(¸; W
k-1
) i.e., f
1
· … · f
n
= S(¸
1
;
1
1
k 1
W
÷
)
· … · S(¸
n
;
n
n
k 1
W
÷
) must n-divide p. p = fg i.e., p
1
· … · p
n
=
f
1
g
1
· … · f
n
g
n
. Apply g(T) = g
1
(T
1
) · … · g
n
(T
n
) to both
sides i.e., p¸ = gf¸ = gr
j
þ
j
+ gþ
o
+
i i
1 i j
gr
s <
þ
¯
i.e.,
p
1
¸
1
· … · p
n
¸
n
= g
1
f
1
¸
1
· … · g
n
f
n
¸
n
=
1 1 1 1
1 1
1 1 1 1 1
1 j j 1 0 1 i i
1 i j
g g g r
s <
| |
¸ þ + þ + þ
|
|
\ .
¯
·
2 2 2 2
2 2
2 2 2 2 2
2 j j 2 0 2 i i
1 i j
g g g r
s <
| |
¸ þ + þ + þ
|
|
\ .
¯
· … ·
n n n n
n n
n n n n n
n j j n 0 n j j
1 i j
g g g r
s <
| |
¸ þ + þ + þ
|
|
\ .
¯
.
By definition, p¸ is in W
j–1
and the last two terms on the right
side of the above equation are in W
j-1
=
1 n
1 n
j 1 j 1
W W
÷ ÷
· · ! .
Therefore
gr
j
þ
j
=
1 1 n n
1 j j n j j
g r g r þ · · þ !
is in
W
j-1
=
1 n
1 n
j 1 j 1
W W
÷ ÷
· · ! .
Now using condition (2) of step 1
n-deg(gr
j
) > n-deg(S(þ
j
; W
j-1
)) ;
i.e.,
deg(g
1
1
j
r ) · … · deg(g
n
n
j
r )
> deg(S(
1
j
þ ;
1
1
j 1
W
÷
)) · … · deg(S(
n
j
þ ;
n
n
j 1
W
÷
))
= n-deg p
j
= deg
1
1
j
p · … · deg
n
n
j
p
> n-deg (S(¸; W
j-1
)
145
= deg (S(¸
1
;
1
1
j 1
W
÷
)) · … · deg(S(¸
n
;
n
n
j 1
W
÷
))
= n-deg p
= deg p
1
· … · deg p
n
= n-deg fg
= deg f
1
g
1
· … · deg f
n
g
n
.
Thus n-deg r
j
> n-deg f; i.e., deg
1
j
r · … · deg
n
j
r > deg f
1
· …
· deg f
n
and that contradicts the choice of j = (j
1
, …, j
n
). We
now know that f = f
1
· … · f
n
, n-divides each g
i
=
1
1
i
g · … ·
n
n
i
g i.e.,
t
i
f divides
t
t
i
g for t = 1, 2, …, n and hence that þ
0
= f¸
i.e.,
1 n
0 0
... þ · ·þ = f
1
¸
1
·…· f
n
¸
n
. Since W
0
=
1 n
0 0
W ... W · · is
k
0
W is T
k
-admissible for k = 1, 2, …,
n); we have þ
0
= f¸
0
where ¸
0
=
1 n
0 0
... ¸ · ·¸ e W
0
=
1
0
W · …
·
n
0
W ; i.e.,
1 n
0 0
... þ · ·þ =
1 n
1 0 n 0
f ... f ¸ · · ¸ where ¸
0
e W
0
. We
make a mention that step 2 is a stronger form of the assertion
that each of the n-subspaces W
1
=
1 n
1 1
W ... W · · ,W
2
=
1
2
W ·…
·
n
2
W , …, W
r
=
1 n
r r
W ... W · · is T-n admissible.
STEP 3:
There exists non zero n-vectors (
1
1
o , …,
1
1
r
o ) · … · (
n
1
o , …,
n
n
r
o ) in V = V
1
· … · V
n
which satisfy condition (i) and (ii) of
1
1 1
1 r
, , þ þ ! } · … · {
n
n
þ , …,
n
n
r
þ } as in step 1. Fix k = (k
1
, …, k
n
) as 1 s k
i
s r
i
. We apply step
2 to the n-vector þ = þ
1
· … · þ
n
=
1 n
1 n
k k k
... þ · ·þ = þ and T-
n-conductor f = f
1
· … · f
n
=
1 n
1 n
k k
p p · · ! = p
k
. We obtain
p
k
þ
k
=
k 0 k i i
1 i k
p p h
s <
¸ + þ
¯
;
i.e.,
1 1 n n
1 1 n n
k k k k
p p þ · · þ ! =

1 1 1 1
1 1
1 1 1 1 1
k 0 k i i
1 i k
p p h
s <
| |
¸ + þ
|
|
\ .
¯
· … ·
n n n n
n n
n n n n n
k 0 k i i
1 i k
p p h
s <
| |
¸ + þ
|
|
\ .
¯
146
where ¸
0
=
1 n
0 0
¸ · ·¸ ! is in W
0
=
1 n
0 0
W W · · ! and
{ }
1
1 1
1 k 1
h , , h
÷
! · … ·
{ }
n
n n
1 k 1
h , , h
÷
! are n-polynomials. Let
o
k
=
k 0 i i
1 i k
h
s <
þ ÷ ¸ ÷ þ
¯
i.e.,
{ }
1 n
1 n
k k
o · ·o ! =
1 1 1
1 1
1 1 1 1
k 0 i i
1 i k
h
s <
| |
þ ÷ ¸ ÷ þ · ·
|
|
\ .
¯
!
n n n
n n
n n n n
k 0 i i
1 i k
h
s <
| |
þ ÷ ¸ ÷ þ
|
|
\ .
¯
.
Since þ
k
– o
k
=
( ) ( )
1 1 n n
1 1 n n
k k k k
þ ÷ o · · þ ÷ o ! is in
W
k–1
=
1 n
1 n
k 1 k 1
W W
÷ ÷
· · ! ;
S(o
k
;W
k–1
) = S(þ
k
;W
k–1
)
= p
k
=
( ) ( )
1 1 n n
1 1 n n
k k 1 k k 1
S , W S , W
÷ ÷
o · · o !
=
( ) ( )
1 1 n n
1 1 n n
k k 1 k k 1
S , W S , W
÷ ÷
þ · · þ !
=
1 n
1 n
k k
p p · · !
and since
p
k
o
k
= 0 · … · 0.
1 1 n n
1 1 n n
k k k k
p p o · · o ! = 0 · … · 0.
we have
W
k–1
· Z(o
k
; T) = {0} · … · {0}.
That is
( )
1 1
1 1
k 1 k 1
W Z ;T ...
÷
· o · ·
( )
1 n
n n
k 1 k n
W Z ;T
÷
· o
= {0} · ... · {0}.
Because each o
k
=
1 n
k k
o · ·o ! satisfies the above two
equations just mentioned it follows that W
k
= W
0
1
k
; T); i.e.,
1 n
1 n
k k
W ... W · · =
( ) ( ) { }
1
1 1 1
0 1 1 k 1
147
( ) ( ) { }
2
2 2 2
0 1 2 k 2
( ) ( ) { }
n
n n n
0 1 n k n

and that p
k
=
1 n
1 n
k k
p ... p · · is the T-n-annihilator of o
k
=
1
1
k
o ·
…·
n
n
k
o . In other words, n-vectors
{ }
1
1 1
1 r
, , o o ! ,
{ }
2
2 2
1 r
,..., o o ,
…,
{ }
n
n n
1 r
,..., o o define the same n-sequence of n-subspaces.
W
1
=
{ }
1 n
1 1
W ... W · · , W
2
=
{ }
1 n
2 2
W ... W · · …. as do the n-
vectors
{ } { } { }
1 2 n
1 1 2 2 n n
1 r 1 r 1 r
... , ... , , ... þ · ·þ þ · ·þ þ · ·þ ! and the
T-n-conductors p
k
= S (o
k
; W
k–1
) that is
{ }
1 n
1 n
k k
p ... p · · =
( ) ( )
1 1 n n
1 1 n n
k k 1 k k 1
S ; W S ; W
÷ ÷
o · · o !
have the same maximality properties.The n-vectors
{ }
1
1 1
1 r
, , o o !
{ }
n
n n
1 r
, , o o ! have the additional property that the n-subspaces
W
0
=
{ }
1 n
0 0
W ... W , · ·
Z(o
1
; T) =
( ) ( )
1 n
1 1 1 n
Z ;T ... Z ;T o · · o
Z(o
2
; T) =
( ) ( )
1 n
2 2 2 n
Z , T ... Z , T o · · o , …
are n-independent. It is therefore easy to verify condition (ii) of
the theorem. Since
p
i
o
i
=
( ) ( )
1 1 n n
1 1 n n
i i i i
p ... p o · · o = (0 · … · 0)
we have the trivial relation
p
k
o
k
=
1 1 n n
1 1 n n
k k k k
p ... p o · · o
=
( )
1 1
1 1 1 1
1 1 k 1 k 1
0 p ... p
÷ ÷
+ o + + o ·
( )
2 2
2 2 2 2
1 1 k 1 k 1
0 p ... p
÷ ÷
+ o + + o · …
·
( )
n n
n n n n
1 1 k 1 k 1
0 p ... p
÷ ÷
+ o + + o .
Apply step 2 with
{ } { }
1 n
1 1 n n
1 k 1 k
,..., , , ,..., þ þ þ þ ! replaced by
{ }
1
1 1
1 k
,..., o o
{ }
n
n n
1 k
,..., ,..., o o and with þ = þ
1
· … · þ
n
148
=
1 n
1 n
k k
o · ·o ! ; p
k
n-divides each p
i
, i < k that is (i
1
, …, i
n
) <
(k
1
, …, k
n
) i.e.,
1 n
1 n
k k
p ... p · · n-divides each
1 n
1 n
i i
p ... p · · i.e.,
each
t
t
k
p divides
t
t
i
p for t = 1, 2, …, n.
STEP 4:
The number r = (r
1
, …, r
n
) and the n-polynomials (p
1
, …,
1
r
p ),
…, (p
1
, …,
n
r
p ) are uniquely determined by the conditions of
the theorem. Suppose that in addition to the n-vectors
{ } { } { }
1 2 n
1 1 2 2 n n
1 r 1 r 1 r
,..., , ,..., ,..., ,..., o o o o o o we have non zero n-
vectors
{ } { }
1 n
1 1 n n
1 s 1 s
,..., ,..., ,..., ¸ ¸ ¸ ¸ with respective T-n-
annihilators
{ } { }
1 n
1 1 n n
1 s 1 s
g ,..., g ,..., g ,..., g such that
V =
( ) ( )
0 1 s
i.e.,
V = V
1
· … · V
n
( ) ( ) { }
1
1 1 1
0 1 1 s 1
( ) ( ) { }
n
n n n
0 1 n s n
t
t
k
g divides
t
t
k 1
g
÷
for t = 1, 2, …, n and k
t
= 2, …, s
t
. We shall
show that r = s i.e., (r
1
, …, r
n
) = (s
1
, …, s
n
) and
t t
i i
p g = ; 1 s t s
n. i.e.,
1 n
i i
p ... p · · =
1 n
i i
g ... g · · for each i. We see that p
1
= g
1
=
1 n
1 1
p ... p · · =
1 n
1 1
g ... g · · . The n-polynomial g
1
=
1 n
1 1
g ... g · ·
is determined by the above equation as the T-n-conductor of V
into W
o
i.e., V = V
1
· … · V
n
into
1 n
0 0
W ... W · · . Let S(V;
W
o
) = S(V
1
;
1
0
W ) · … · S(V
n
;
n
0
W ) be the collection of all n-
polynomials f = f
1
· … · f
n
such that fþ = f
1
þ
1
· … · f
n
þ
n
is
in W
o
for every in þ = þ
1
· … · þ
n
in V = V
1
· … · V
n
i.e., n-
polynomials f such that the n-range of f(T) = range of f
1
(T
1
) ·
range of f
2
(T
2
) · … · range of f
n
(T
n
) is contained in W
0
=
1 n
0 0
W ... W · · , i.e., range of f
i
(T
i
) is in
i
0
W for i = 1, 2, …, n.
Then S(V
i
; W
i
) is a non zero ideal in the polynomial algebra so
149
that we see S(V; W
0
) = S(V
1
;
1
0
W ) · … · S(V
n
;
n
0
W ) is a non
zero n-ideal in the n-polynomial algebra.
The polynomial
t
1
g is the monic generator of that ideal i.e.,
the n-monic polynomial g
1
=
1 n
1 1
g ... g · · is the monic n-
generator of that n-ideal; for this reason. Each þ = þ
1
· … · þ
n
in V = V
1
· … · V
n
has the form þ =
( )
1 1
1 1 1 1 1
0 1 1 s s
f ... f þ + ¸ + + ¸ ·
… ·
( )
n n
n n n n n
0 1 1 s s
f ... f þ + ¸ + + ¸ and so
s
1 1 0 1 i i
1
g g g f þ = þ + ¸
¯
i.e.,
1
1 1
s
1 n 1 1 1 1 1
1 1 1 n 1 0 1 i i
1
g ... g g g f
| |
þ · · þ = þ + ¸
|
\ .
¯
· … ·
n
n n
s
n n n n n
1 0 1 i i
1
g g f
| |
þ + ¸
|
\ .
¯
.
Since each
t
i
g divides
t
1
g for t = 1, 2, …, n we have g
1
¸
i
= 0 ·
… · 0 i.e.,
1 n
1 1 n n
1 i 1 i
g ... g ¸ · · ¸ = 0 · … · 0 for all i = (i
1
, …, i
n
)
and g
0
þ = g
1
þ
0
is in W
0
=
1
0
W · … ·
n
0
W . Thus
t
i
g is in S(V
t
;
t
0
W ) for t = 1, 2, …, n, so g
1
=
1
1
g · …·
n
1
g is in
S(V; W
0
) = S(V
1
;
1
0
W ) · …· S(V
n
;
n
0
W ).
Since each
t
i
g is monic; g
1
is a monic n-polynomial of least
n-degree which sends
1
1
¸ into
t
0
W so that ¸
1
=
1 n
1 1
... ¸ · ·¸ into
1 n
0 0 0
W W ... W = · · ; we see that g
1
=
1 n
1 1
g ... g · · is the monic
n-polynomial of least n-degree in the n-ideal S (V; W
0
). By the
same argument p
i
is the n-generator of that ideal so p
1
= g
1
; i.e.,
1 n
1 1
p ... p · · =
1 n
1 1
g ... g · · .
If f = f
1
· … · f
n
is a n-polynomial and W = W
1
· … · W
n
is
a n-subspace of V = V
1
· … · V
n
we shall employ the short
150
hand fW for the set of all n-vectors fo = f
1
o
1
· … · f
n
o
n
with
o = o
1
· … · o
n
in W = W
1
· … · W
n
.
The three facts can be proved by the reader:
1. fZ(o; T) = Z(f
o
; T) i.e., f
1
Z(o
1
; T
1
) · … · f
n
Z(o
n
; T
n
)
=
( ) ( )
1 1 1 n n n
Z f ;T ... Z f ;T o · · o .
2. If V = V
1
k
,
i.e.,
1
1 1
1 k
n
n n
1 k
where each V
t
is n-invariant under T that is each.
i
t
V is
invariant under T
i
; 1 s i < t; t = 1, 2, …, n then fV =
fV
1
n
; i.e., f
1
V
1
· … · f
n
V
n
= f
1
V
1
f
1
1
1
k
V · … ·
n
n n
n 1 n k
3. If o = o
1
· … · o
n
and ¸ = ¸
1
· … · ¸
n
have the same
T-n-annihilator then fo and f¸ have the same T-n-
annihilator and hence n-dim Z(fo; T) = n-dim Z(f¸; T).
i.e., fo = f
1
o
1
· … · f
n
o
n
and f¸ = f
1
¸
1
· … · f
n
¸
n
with
dim Z(f
1
o
1
; T
1
) · … · dim Z(f
n
o
n
; T
n
) = dim Z(f
1
¸
1
;
T
1
) · … · dim Z(f
n
¸
n
; T
n
).
Now we proceed by induction to show that r = s and p
i
= g
i
for i = 1, 2, …, r. The argument consists of counting n-
dimensions in the proper way. We shall give the proof if r = (r
1
,
r
2
, …, r
n
) > (2, 2,…, 2) then p
2
=
1 n
2 2
p ... p · · = g
2
=
1 n
2 2
g ... g · ·
and from that the induction should be clear. Suppose that r =
(r
1
· … · r
n
) > (2, 2, …, 2); then n-dim W
0
+ n-dim Z(o
1
; T) <
n-dim V i.e., (dim
1
0
W · … · dim
n
0
W ) + dim Z
( )
1
1 1
;T o ·… ·
dim Z
( )
n
1 n
;T o < dim V
1
· … · dim V
n
;
dim
1
0
W + dim Z
( )
1
1 1
;T o · … · dim
n
0
W + dim Z
( )
n
1 n
;T o
< dim V
1
· … · dim V
n
.
Since we know p
1
= g
1
we know that Z(o
1
; T) and Z(¸
1
; T)
have the same n-dimension. Therefore n-dim W
0
+ n-dim Z(¸
1
;
T) < n-dim V as before
151
dim
1
0
W + dim Z
( )
1
1 1
;T ¸ · … · dim
n
0
W + dim Z
( )
n
1 n
;T ¸ <
dim V
1
· … · dim V
n
,
which shows that s = (s
1
, s
2
, …, s
n
) > (2, 2, …, 2). Now it makes
sense to ask whether or not p
2
= g
2
,
1 n
2 2
p ... p · · =
1 n
2 2
g ... g · · .
From the two decompositions, of V = V
1
· … · V
n
, we obtain
two decomposition of the n-subspace
p
2
V =
1 n
2 1 2 n
p V ... p V · ·
p
2
V =
( )
1
2 0 2 1
p W Z p ;T © o
i.e.,
1 n
2 1 2 n
p V ... p V · · =
( )
1 1 1 1
2 0 2 1 1
p W Z p ;T ... © o · ·
( )
n n n n
2 0 2 1 n
p W Z p ;T © o
( )
2 2 0 2 1
p V p W Z p ;T ... = © ¸ ©
( )
2 s
Z p ;T © ¸ ;
i.e.,
1 n
2 1 2 n
p V ... p V · · =
( ) ( )
1
1 1 1 1 1 1
2 0 2 1 1 2 s 1
( ) ( )
n
n n n n n n
2 0 2 1 n 2 s n
... p W Z p ;T ... Z p ;T · · © ¸ © © ¸ .
We have made use of facts (1) and (2) above and we have used
the fact
1 n
1 1 n n
2 i 2 i 2 i
p p ... p o = o · · o = 0 · … · 0; i = (i
1
, …, i
n
) >
(2, 2, …, 2). Since we know that p
1
= g
1
fact (3) above tell us
that
( ) ( ) ( )
1 1 n n
2 i 2 1 1 2 1 n
Z p ;T Z p ;T ... Z p ;T o = o · · o
and
( ) ( ) ( )
1 1 n n
2 1 2 1 1 2 1 n
Z p ;T Z p ;T ... Z p ;T ¸ = ¸ · · ¸ , have the same
n-dimension. Hence it is apparent from above equalities that n-
dim
( )
2 i
Z p ;T ¸ = 0 · … · 0.
dim
( ) ( )
1 n
1 1 n n
2 i 1 2 i n
Z p ;T ... dimZ p ;T ¸ · · ¸ =
(0 · … · 0); i = (i
1
, …, i
n
) > (2, …, 2).
We conclude p
2
¸
2
=
( ) ( )
1 1 n n
2 2 2 2
p p 0 0 ¸ · · ¸ = · · ! !
152
and g
2
n-divides p
2
i.e.,
t
2
g divides
t
2
p for each t; t = 1, 2, …, n.
The argument can be reversed to show that p
2
n-divides g
2
i.e.,
t
2
p divides
t
2
g for each t; t = 1, 2, …, n. Hence p
2
= g
2
.
We leave the two corollaries for the reader to prove.
COROLLARY 1.2.17: If T = T
1
· … · T
n
is a n-linear operator
on a finite (n
1
, …, n
n
) dimensional n-vector space V = V
1
· …·
V
n
then T-n-admissible n-subspace has a complementary n-
subspace which is also invariant under T.
COROLLARY 1.2.18: Let T = T
1
· …· T
n
be a n-linear
operator on a finite (n
1
, …, n
n
) dimensional n-vector space V =
V
1
· … ·V
n
.
a. There exists n-vectors o = o
1
· … · o
n
in V = V
1
· … · V
n
such that the T-n-annihilator of o is the n-
minimal polynomial for T.
b. T has a n-cyclic n-vector if and only if the n-
characteristic and n-minimal polynomial for T are
identical.
Now we proceed on to prove the Generalized Cayley-Hamilton
theorem for n-vector spaces of finite n-dimension.
THEOREM 1.2.58:(GENERALIZED CAYLEY HAMILTON THEOREM):
Let T = T
1
· … · T
n
be a n-linear operator on a finite (n
1
, n
2
,
…, n
n
) dimension n-vector space V = V
1
· … · V
n
. Let p and f
be the n-minimal and n-characteristic polynomials for T,
respectively
i. p n-divides f i.e., if p = p
1
· … · p
n
and f = f
1
·…· f
n
then p
i
divides f
i
= i = 1, 2, …, n.
ii. p and f have same prime factors expect for
multiplicities.
153
iii. If p =
1
1
...
k
r r
k
f f is the prime factorization of p then f =
1
1
... · ·
k
d d
k
f f where d
i
is the n-multiplicity of ( )
i
r
i
f T
n-divided by the n-degree of f
i
.
That is if p = p
1
· … · p
n
=
( ) ( ) ( ) ( )
1 1
1 1
1
1
1 1
1 1
... ... ... · ·
n n
k k
n
n
r r r r
n n
k k
f f f f ;
then f =
( ) ( ) ( ) ( )
1 1
1 1
1
1
1 1
1 1
... ... ... · ·
n n
k k
n
n
d d d d
n n
k k
f f f f ;
t
i
d =
( )
1
,...,
t
t t
k
d d
is the nullity of ( )
t
i
r t
i t
f T which is n-divided by the n-degree
t
i
f
i.e.,
t
i
d ; 1s i s k
t
. This is true for each t = 1, 2, …, n.
Proof: The trivial case V = {0} ·… · {0} is obvious. To prove
(i) and (ii) consider a n-cyclic decomposition
V = Z(o
1
r
; T)
=
( ) ( )
1
1 1
1 1 r 1
Z ;T ... Z ;T o © © o · … ·
( ) ( )
n
n n
1 n r n
By the second corollary p
1
= p. Let U
1
=
1 n
i i
U U · · ! be the n-
restriction of T = T
1
· … · T
n
i.e., each
s
i
U is the restriction of
T
s
(for s = 1, 2, …, r
s
) to
( )
s
i s
Z ;T o . Then U
i
has a n-cyclic n-
vector and so p
i
=
1 n
1 n
i i
p p · · ! is both n-minimal polynomial
and the n-characteristic polynomial for U
i
. Therefore the n-
characteristic polynomial f = f
1
· … · f
n
is the n-product f =
1 n
1 1 n n
i r i r
p p p p · · ! ! ! . That is evident from earlier results that
the n-matrix of T assumes in a suitable n-basis.
Clearly p
1
= p n-divides f; hence the claim (i). Obviously
any prime n-divisor of p is a prime n-divisor of f. Conversely a
prime n-divisor of f =
1 n
1 1 n n
i r 1 r
p p p p · · ! ! ! must n-divide one
of the factor
t
i
p which in turn n-divides p
1
.
154
Let p =
( ) ( ) ( ) ( )
1 n 1 h
k k 1 1
1 n
1 n
r r r r
1 1 n n
1 k 1 k
f f f f · · ! ! ! be the prime
n-factorization of p. We employ the n-primary decomposition
theorem, which tell us if
i i i
t 1 n
V V V = · · ! is the n-null space
for
( )
t
i
r
i
t t
f T then
V = V
1
k
=
( )
1
k 1
1 1
( )
n
k 1
n n
and
( )
t
i
r
t
i
f is the minimal polynomial of the operator
i
t
T
restricting T
t
to the invariant subspace
i
t
V . This is true for each t
= 1, 2, …, n. Apply part (ii) of the present theorem to the n-
operator
i
t
T . Since its minimal polynomial is a power of the
prime
t
i
f the characteristic polynomial for
i
t
T has the form
( )
t
i
r
t
i
f where
t t
i i
d r > , t = 1, 2, …, n.
We have
t
i
d =
i
t
t
i
dimV
degf
for every t = 1, 2 ,…, n and dim
i
t
V
= nullity
( )
t
i
r
t
i t
f T for every t = 1, 2, …, n. Since T
t
is the direct
sum of the operators
t
k 1
t t
T , , T ! the characteristic polynomial f
t
is the product, f
t
=
( ) ( )
t t
k i
t
t
d d
t t
1 k
f f " . Hence the claim.
The immediate corollary of this theorem is left as an exercise
COROLLARY 1.2.19: If T = T
1
· …· T
n
is a n-nilpotent
operator of the n-vector space of (n
1
, …, n
n
) dimension then the
n-characteristic n-polynomial for T is
1
· · !
n
n n
x x .
Let us observe the n-matrix analogue of the n-cyclic
decomposition theorem. If we have the n-operator T and the n-
direct sum decomposition, let B
i
be the n-cyclic ordered basis.
{ }
1
1
1 1 1
1
1 1 1
1 1
, , ,
÷
!
i
k
i i i
T T o o o · … ·
{ }
1
1
1
, , ,
÷
!
i
n n n
k
n n n
i n i n i
T T o o o
155
for Z(o
i
; T) = Z
( )
1
1
1
;
i
T o · … ·
( )
;
n
n
i n
Z T o . Here
( )
1 1
, , !
n
i i
k k
denotes n-dimension of Z(o
i
; T) that is the n-degree of the n-
annihilator p
i
=
1
1
... · ·
n
n
i i
p p . The n-matrix of the induced
operator T
i
in the ordered n-basis B
i
is the n-companion n-
matrix of the n-polynomial p
i
. Thus if we let B to be the n-
ordered basis for V which is the n-union of B
i
arranged in order
{ }
1
1 1
1
, , !
r
B B · … ·
{ }
1
, , !
n
n n
r
B B then the n-matrix of T in the
ordered n-basis B will be A = A
1
· A
2
· … · A
n
=
1
1
1
1
2
1
r
A 0 0
0 A 0
0 0 A

"
"
# # #
"
· … ·
n
n
1
n
2
n
r
A 0 0
0 A 0
0 0 A

"
"
# # #
"
where
t
i
A is the
t t
i i
k k × companion matrix of
t
i
p ; for t = 1, 2,
…, n. A (n
1
× n
1
, …, n
n
× n
n
) n-matrix A which is the n-direct
sum of the n-companion matrices of the non-scalar monic n-
polynomial
{ }
1
1 1
i r
p , , p ! · … ·
{ }
n
n n
1 r
p , , p ! such that
t
t
i 1
p
+
divides
t
t
i
p for i
t
= 1, 2 ,… , r
t
– 1 and t = 1, 2, …, n will be said to be
the rational n-form or equivalently n-rational form.
THEOREM 1.2.59: Let F = F
1
· … · F
n
be a n-field. Let B = B
1
· … · B
n
be a (n
1
× n
1
, …, n
n
× n
n
) n-matrix over F. Then B is
n-similar over the n-field F to one and only one matrix in the
rational form.
Proof: We know from the usual square matrix every square
matrix over a fixed field is similar to one and only one matrix
which is in the rational form.
So the n-matrix B = B
1
· … · B
n
over the n-field F = F
1
· … · F
n
is such that each B
i
is a n
i
× n
i
square matrix over F
i
,
is similar to one and only one matrix which is in the rational
form, say C
i
. This is true for every i, so B = B
1
· … · B
n
is n-
156
similar over the field to one and only one n-matrix C which is in
the rational n-form.
The n-polynomials
( ) ( ) { }
1 n
1 1 n n
1 r 1 r
p , , p , , p , , p ! ! ! are called
invariant n-factors or n-invariant factors for the n-matrix B =
B
1
· … · B
n
.
We shall just introduce the notion of n-Jordan form or
Jordan n-form. Suppose that N = N
1
· … · N
n
be a nilpotent n-
linear operator on a finite (n
1
, n
2
,…, n
n
) dimension space V = V
1
· … · V
n
. Let us look at the n-cyclic decomposition for N
which we have depicted in the theorem. We have positive
integers (r
1
, …, r
n
) and non zero n-vectors
{ }
n
n n
1 r
, , o o ! in V
with n-annihilators
{ } { }
1 n
1 1 n n
1 r 1 r
p , , p p , , p · · ! ! ! such that
V =
( ) ( )
1 r
=
( ) ( )
1
1 1
1 1 r 1
Z ; N Z ; N o © © o · · " "
( ) ( )
n
n n
1 n r n
and
t
t
i 1
p
+
divides
t
t
i
p for i
t
= 1, 2 ,…, r
t
– 1 and t = 1, 2, …, n.
Since N is n-nilpotent the n-minimal polynomial is
1 n
k k
x x · · ! with k
t
s n
t
; t = 1, 2, …, n. Thus each
t
t
i
p is of
the form
t
i
t
k t
i
p x = and the n-divisibility condition simply says
t
t t t
1 2 r
k k k > > > ! ; t = 1, 2, …, n. Of course
t
1
k = k
t
and
t
r
k > 1.
The n-companion n-matrix of
1 n
i i
1 n
k k
x x · · ! is the
r r
i i
k k × n-
matrix. A =
1 n
1 n
i i
A A · · ! with
t
t
i
0 0 0 0
1 0 0 0
A 0 1 0 0
0 0 1 0

=

"
!
# # # #
!
; t = 1, 2, …, n.
157
Thus we from earlier results have an n-ordered n-basis for V =
V
1
· … · V
n
in which the n-matrix of N is the n-direct sum of
the elementary nilpotent n-matrices the sizes i
t
of which
decrease as i
t
increases. One sees from this that associated with
a n-nilpotent (n
1
× n
1
, …, n
n
× n
n
), n-matrix is a positive n-tuple
integers (r
1
, …, r
n
) i.e.,
{ }
1
1 1
1 r
k , , k ! · … ·
{ }
n
n n
1 r
k , , k !
such that
1
1 n
1 r 1
k k n + + = !
2
2 n
1 r 2
k k n + + = !
and
n
n n
1 r n
k k n + + = !
and
t t
t t
i i 1
k k
+
> ; t = 1, 2, …, n and 1 s i, i + 1 s r
t
and these n-sets
of positive integers determine the n-rational form of the n-
matrix, i.e., they determine the n-matrix up to similarity.
Here is one thing, we like to mention about the n-nilpotent,
n-operator N above. The positive n-integer (r
1
, …, r
n
) is
precisely the n-nullity of N infact the n-null space has a n-basis
with (r
1
, …, r
n
) n-vectors
i i
1 n
1 n
k 1 k 1
1 n
1 i n i
N N
÷ ÷
o · · o ! . For let o =
o
1
· … · o
n
be in n-nullspace of N we write
o =
( )
1 1
1 1 1 1
1 1 r r
f f o + + o ! · … ·
( )
1 n
n n n n
1 1 r r
f f o + + o !
where
( )
1 n
1 n
i i
f , , f ! is a n-polynomial the n- degree of which we
may assume is less than
1 n
i i
k , , k ! . Since No = 0 · … · 0;
i.e., N
1
o
1
· … · N
n
o
n
= 0 · … · 0 for each i
r
we have
0 · 0 · … · 0 = N(f
i
o
i
)
=
( ) ( )
1 1 n n
1 i i n i i
N f N f o · · o !
=
( )
1 1 n n
1 i 1 i n i n i
N f (N ) N f N o · · o "
=
( )
1 1 n n
i i i i
xf (xf ) o · · o ! .
158
Thus
1 n
i i
xf xf · · ! is n-divisible by
i i
1 n
k k
x x · · ! and since
n-deg
( )
1 n
i i
f , , f ! >
( )
1 2 n
i i i
k , k , , k ! this imply that
1 n
i i
f f · · ! =
i 1 i 1
1 n
1 1
k k
1 n
i i
c x c x
÷ ÷
· · !
where
1 1
1 n
i i
c c · · ! is some n-scalar. But then
o = o
1
· … ·
s
n
o
=
( ) ( )
i 1 i 1
1 1
1 1
k k
1 1 1 1
1 1 r r
c x c x
÷ ÷
o + + o " ·
( ) ( )
i 1 i 1
2 2
2 2
k k
2 2 2 2
1 1 r r
c x c x
÷ ÷
o + + o ! · … ·
( ) ( )
i 1 i 1
n n
n n
k k
n n n n
1 1 r r
c x c x
÷ ÷
o + + o ! ;
which shows that all the n-vectors form a n-basis for the n-null
space of N = N
1
· … · N
n
. Suppose T is a n-linear operator on
V = V
1
· … · V
n
and that T factors over the n-field F = F
1
· … · F
n
as
f = f
1
· … · f
n
=
( ) ( )
1 1
k 1
1
1
d d
1 1
1 k
x c x c ÷ ÷ · · ! !
( ) ( )
1 n
k 1
n
n
d d
n n
1 k
x c x c ÷ ÷ "
where
{ } { }
1 n
1 1 n n
1 k 1 k
c , , c c , , c · · ! ! ! are n- distinct n-element
of F = F
1
· … · F
n
and
t
t
i
d 1 > ; t = 1, 2, …, n.
Then the n-minimal polynomial for T will be
p =
( ) ( )
1 1
k 1
1
1
r r
1 1
1 k
x c x c ÷ ÷ · · ! !
( ) ( )
n n
k 1
n
n
r r
n n
1 k
x c x c ÷ ÷ !
where 1 <
t t
t t
i i
r d s . t = 1, 2, …, n.
If
1 n
1 n
i i
W W · · ! is the n-null space of
( )
i
r
i
T c I ÷ =
( )
1
i
1
1
r
1
1 i 1
T C I ÷ · · !
( )
n
i
n
n
r
n
n i n
T c I ÷
then the n-primary decomposition theorem tells us that
V = V
1
· … ·V
n
=
( )
1
1 1
1 k
( )
n
n n
1 k
and that the operator
t
t
i
T induced on
t
t
i
W defined by
t
i
t
T has n-
minimal polynomial
( )
t
i
t
r
t
i
x c ÷ for t = 1, 2, …, n; 1 s i
t
s k
t
. Let
159
t
t
i
N be the n-linear operator on
t
t
i
W defined by
t t t
t t t
i i i t
N T c I = ÷ .
1 s i
t
s k
t
; then
t
t
i
N is n-nilpotent and has n-minimal polynomial
t
i
t
t
r
i
x . On
t
t
i
W , T
t
acts like
t
t
i
N plus the scalar
t
t
i
c times the
identity operator. Suppose we choose a n-basis for the n-
subspace
1 n
1 n
i i
W W · · ! corresponding to the n-cyclic
decomposition for the n-nilpotent
t
t
i
N . Then the k-matrix
t
t
i
T in
this ordered n-basis will be the n-direct sum of n-matrices.
1 2
1 2
1 2
1 2
c 0 ... 0 0 c 0 ... 0 0
1 c 0 0 1 c 0 0
c c
0 0 1 c 0 0 1 c

·

# # # # # # #
#
! !
· … ·

n
n
n
n
c 0 ... 0 0
1 c 0 0
c
0 0 1 c

!
# # # #
!
each with c =
t
t
i
c for t = 1, 2, …, n. Further more the sizes of
these n-matrices will decrease as one reads from left to right. A
n-matrix of the form described above is called an n-elementary
Jordan n-matrix with n-characteristic values c
1
· … · c
n
.
Suppose we put all the n-basis for
1
1
i
W · … ·
n
n
i
W together
and we obtain an n-ordered n-basis for V. Let us describe the n-
matrix A of T in the n-order basis.
The n-matrix A is the n-direct sum
160
A =
1 n
1 n
1 1
1 n
2 2
1 n
k k
A 0 0 A 0 0
0 A 0 0 A 0
0 0 A 0 0 A

· ·

! !
! "
!
# # # # # #
! "
of the k
i
sets of n-matrices
{ }
1
1 1
1 k
A , ..., A · … ·
{ }
n
n n
1 k
A , ..., A .
Each
( )
( )
( )
t
t
t
t
i
t1
i
t
t 2
i
i
tn
J 0 ... 0
0 J ... 0
A
0 0 ... J

=

#
where each
( )
t
i
jt
J is an elementary Jordan matrix with
characteristic value
t
t
i
c ; 1 < i
t
< k
t
; t = 1, 2, …, n. Also with in
each
t
t
i
A the sizes of the matrices
t
t
i
tj
J decrease as j
t
increase 1 s
j
t
s n
t
; t = 1, 2, …, n. A (n
1
× n
1
, …, n
n
× n
n
) n-matrix A which
satisfies all the conditions described so far for some n-sets of
distinct k
i
-scalars
{ }
1
1 1
1 k
c c ! · … ·
{ }
n
n n
1 k
c c ! will be said to
be in Jordan n-form or n-Jordan form.
The interested reader can derive several interesting
properties in this direction.
In the next chapter we move onto define the notion of n-
inner product spaces for n-vector spaces of type II.
161
Chapter Two
n-INNER PRODUCT
SPACES OF TYPE II
Throughout this chapter we denote by V = V
1
· … · V
n
only a
n-vector space of type II over n-field F = F
1
· … · F
n
. The
notion of n-inner product spaces for n-vector spaces of type II is
introduced. We using these n-inner product define the new
notion of n-best approximations n-orthogonal projection,
quadratic n-form and discuss their properties.
DEFINITION 2.1: Let F = F
1
· … · F
n
be a n-field of real
numbers and V = V
1
· … · V
n
be a n-vector space over the n-
field F. An n-inner product on V is a n-function which assigns to
each n-ordered pair of n-vectors o = o
1
· … · o
n
and þ = þ
1
· … · þ
n
in V a n-scalar (o / þ) = (o
1
/ þ
1
) · … · (o
n
/ þ
n
) in
F = F
1
· … · F
n
i.e., (o
i
/ þ
i
) e F
i
, i = 1, 2, …, n; in such a way
that for all o = o
1
· … · o
n
, þ = þ
1
· … · þ
n
and ¸ = ¸
1
· … · ¸
n
in V and for all n-scalars c = c
1
· … · c
n
in F
1
· …
· F
n
= F.
162
a. (o + þ/¸) = (o/¸) + (þ/¸)
i.e., (o
1

1

1
) · … · (o
n

n

n
) =
( ) ( ) ( ) ( )
1 1 1 1
/ / / / + · · +

"
n n n n
o ¸ þ ¸ o ¸ þ ¸
b. (co/þ) = (co/þ) i.e.,
(c
1
o
1

1
) · … · (c
n
o
n

n
)
= c
1
(o
1

1
) · … · c
n
(o
n

n
)
c. (o/þ) = (þ/o) i.e., (o
1

1
) · … · (o
n

n
)
= (þ
1
/o
1
) · … · (þ
n
/o
n
)
d. (o / o) = (o
1
/ o
1
) · … · (o
n
/o
n
) > (0 · … · 0) if
o
i
= 0 for i = 1, 2, …, n.
From the above conditions we have
(o / cþ + ¸) = c (o/þ) + (o / ¸)
= (o
1
/ c
1
þ
1
+ ¸
1
) · … · (o
n
/ c
n
þ
n
+ ¸
n
)
= [c
1
(o
1
/ þ
1
) + (o
1
/ ¸
1
)] · … · [c
n
(o
n
/ þ
n
) + (o
n
/ ¸
n
)].
A n-vector space V = V
1
· … · V
n
endowed with an n-inner
product is called the n-inner product space. Let F = F
1
· … ·
F
n
for V =
1
1
... · ·
n
n n
n
F F a n-vector space over F there is an n-
inner product called the n-standard inner product. It is defined
for o = o
1
· … · o
n
=
( )
1
1 1
1
!
n
x x · … ·
( )
1
!
n
n n
n
x x
and þ = þ
1
· … · þ
n
=
( )
1
1 1
1
!
n
y y · … ·
( )
1
!
n
n n
n
y y
( )
1 1
1
1 1
/ = · ·
¯ ¯
!
n n
n
n n
j j j j
j j
x y x y o þ .
If A = A
1
· … · A
n
is a n-matrix over the n-field F = F
1
· … ·
F
n
where
×
e
i i
n n
i i
A F is a vector space over F
i
for i = 1, 2, …, n
V =
1 1
1
...
× ×
· ·
n n
n n n n
n
F F over the n-field F = F
1
· … · F
n
, then V
is isomorphic to
2 2
1
1
... · ·
n
n n
n
F F in a natural way. It therefore
follows
(A / B) =
1 1 1 1
1 1
1 1
· ·
¯ ¯
!
n n n n
n n
n n
j k j k j k j k
j k j k
A B A B ,
defines an n-inner product on V. A n-vector space over the n-
field F is known as the n-inner product space.
It is left as an exercise for the reader to verify.
163
THEOREM 2.1: If V = V
1
· … · V
n
is an n-inner product space
then for any n-vectors o = o
1
· … · o
n
, þ = þ
1
· … · þ
n
in V
and any scalar c = c
1
· … · c
n
.
(1) || co|| = | c | ||o||
i.e., || co || = || c
1
o
1
|| · … · ||c
n
o
n
||
= |c
1
| ||o
1
|| · … · |c
n
| ||o
n
||

(2) ||o|| > (0, …, 0) = (0 · … · 0) for o = 0
i.e., || o
1
|| · … · || o
n
|| > (0, 0, …, 0) = (0 · 0 · … · 0)

(3) || (o/þ) || s || o || || þ ||
|| (o
1

1
) || · …· || (o
n
/ þ
n
) ||
s ||o
1
|| ||þ
1
|| · …· || o
n
|| || þ
n
||.
Proof as in case of usual inner produce space.
We wish to give some notation for n-inner product spaces
¸ = þ –
2
( / )
|| ||
þ o
o
o
i.e., ¸
1
· … · ¸
n
=
1 1
1 1
2
1
( / )
|| ||
| | þ o
þ ÷ o
|
o
\ .
· … ·
n n
n n
2
n
( / )
|| ||
| | þ o
þ ÷ o
|
o
\ .
.
It is left as an exercise for the reader to derive the Cauchy –
Schwarz inequality in case of n-vector spaces. We now proceed
on to define the notion of n-orthogonal n-set and n-orthonormal
n-set.
DEFINITION 2.2: Let o = o
1
· … · o
n
, þ = þ
1
· … · þ
n
be n-
vectors in an n-inner product space V = V
1
· … · V
n
. Then o
is n-orthogonal to þ if (o/þ) = 0 · … · 0.
164
i.e., (o
1
/ þ
1
) · …· (o
n
/ þ
n
) = 0 · … · 0.
Since this implies þ = þ
1
· … · þ
n
is n-orthogonal to o =
o
1
· … · o
n
.
We often simply say that o and þ are n-orthogonal. Let S =
S
1
· … · S
n
be n-set of n-vectors in V = V
1
· … · V
n
. S is
called an n-orthogonal n-set provided all n-pair of n-distinct n-
vectors in S are n-orthogonal. An n-orthonormal n-set is an n-
orthogonal n-set with addition property || o || = || o
1
|| · … ·
||o
n
|| = 1 · … · 1 for every o = o
1
· … · o
n
in S = S
1
· … ·
S
n
.
THEOREM 2.2: A n-orthogonal n- set of non-zero n-vectors is n-
linearly independent.
The proof is left as an exercise for the reader.
THEOREM 2.3: Let V = V
1
· … · V
n
be an n-inner product
space and let
{ } { }
1
1 1
1 1
, , ... , , · · ! !
n
n n
n n
þ þ þ þ
be any n-independent vectors in V. Then one way to construct n-
orthogonal vectors
{ } { }
1
1 1
1 1
, , , , · · ! ! !
n
n n
n n
o o o o
in V = V
1
· … · V
n
is such that for each k = 1, 2, …, n the n-set
{ } { }
1
1 1
1 1
, , , , · · ! ! !
n
n n
k k
o o o o
is a n-basis for the n-subspace spanned by
{ } { }
1
1 1
1 1
, , , , · · ! ! !
n
n n
k k
þ þ þ þ .
Proof: The n-vectors
{ } { }
1 n
1 1 n n
1 n 1 n
, , ... , , o o · · o o ! ! can be
obtained by means of a construction analogous to Gram-
Schmidt orthogonalization process called or defined as Gram-
Schmidt n-orthogonalization process. First let o
1
=
1 n
1 1
o · ·o !
and þ
1
=
1 n
1 1
þ · ·þ ! . The other n-vectors are then given
inductively as follows:
165
Suppose
{ } { }
1 n
1 1 n n
1 m 1 m
, , , , o o · · o o ! ! ! (1 s m
i
s n
i
; i = 1, 2,
…, n) have been chosen so that for every k
i
,
{ } { }
1 n
1 1 n n
1 k 1 k
, , , , o o · · o o ! ! ! ; 1 s k
i
s m
i
, i = 1, 2, …, n is
an n-orthogonal basis for the n-subspace of V = V
1
· … · V
n
that is spanned by
{ } { }
1 n
1 1 n n
1 k 1 k
, , , , þ þ · · þ þ ! ! ! . To
construct next n-set of n-vectors,
1 n
1 n
m 1 m 1
, , ;
+ +
o o !
let
( )
1
1 1
1 1
1
1
1 1
m
m 1 k
1 1 1
m 1 m 1 k
1 2
k 1
k
|| ||
+
+ +
=
þ o
o = þ ÷ o
o
¯
and so on
( )
n
n n
n n n
n
n
n n
m
m 1 k
n n n
m 1 m 1 k
n 2
k 1
k
|| ||
+
+ +
=
þ o
o = þ ÷ o
o
¯
.
Thus
( )
1 n 1 n
1 n 1 n
m 1 m 1 j j
|
+ +
o · ·o o · ·o ! ! = (0 · … · 0) for 1 s
j
t
s m
t
; t = 1, 2, …, n. For other wise
( )
n n
1 n
m 1 m 1 + +
þ · ·þ ! will
be a n-linear combination of
{ } { }
1 n
1 1 n n
1 m 1 m
, , ... , , o o · · o o ! ! .
Further 1 s j
t
s m
t
; t = 1, 2, …, n.
Then
(o
m+1
/ o
j
) =
( ) ( )
1 1 n n
1 1 n n
m 1 j m 1 j + +
o o · · o o !
=
( )
( )
( )
1
1 1
1 1 1 1
1
1
1 1
m
m 1 k
1 1 1 1
m 1 j k j
1 2
k 1
k
|| ||
+
+
=

þ o

þ o ÷ o o
o

¯
· … ·
( )
( )( )
n
n n n n
n n
n
n
n n n n
m
m 1 k k j
n n
m 1 j
n 2
k 1
k
|| ||
+
+
=

þ o o o

þ o ÷
o

¯
=
( ) ( )
1 1 1 1
1 1 1 1
m 1 j m 1 j + +

þ o ÷ þ o · ·

!
=
( ) ( )
n n n n
n n n n
m 1 j m 1 j + +

þ o ÷ þ o

= 0 · … · 0.
166
Therefore
{ } { }
1 n
1 1 n n
1 m 1 1 m 1
, , , ,
+ +
o o · · o o ! ! ! is an n-
orthogonal n-set consisting of {m
1
+ 1 · … · m
n
+ 1} non zero
n-vectors in the n-subspace spanned by
{ } { }
1 n
1 1 n n
1 m 1 1 m 1
, , , ,
+ +
þ þ · · þ þ ! ! ! . By theorem we have just
proved it is a n-basis for this n-subspace. Thus the n-vectors
{ } { }
1 n
1 1 n n
1 n 1 n
, , , , o o · · o o ! ! ! may be constructed one after
the other given earlier. In particular when o
1
= þ
1
.
i.e.,
( ) ( )
1 1
1 1 1 1
1 n 1 n
o · ·o = þ · ·þ ! !
( )
2
2 2
1 n
o · ·o ! =
( )
1 1
2 1
2 1
1 1
1 2
1
|| ||

þ o
þ ÷ o · ·
o

!
( )
1 1
n n
2 1
n n
1 2
n
|| ||

þ o
þ ÷ o
o

.
( )
3 3
1 n
o · ·o ! =
( ) ( )
3 1 3 2
1 1 2 2
3 1 2
1 1 2
1 2 2 2
1 2
|| || || ||

þ o þ o
þ ÷ o ÷ o
o o

· … ·
( ) ( )
3 1 3 2
n n n n
3 1 2
n n n
1 2 2 2
n n
|| || || ||

þ o þ o
þ ÷ o ÷ o
o o

.
The reader is expected to derive the following corollary.
COROLLARY 2.1: Every finite (n
1
… n
n
) dimensional n-inner
product space has an n-orthonormal basis.
Next we define the new notion of n-best approximations or best
n-approximations.
DEFINITION 2.3: Let V = V
1
· … · V
n
be a n-inner product
vector space over the n- field F = F
1
· … · F
n
of type II. Let W
= W
1
· … · W
n
be a n-subspace of V. Let þ = þ
1
· … · þ
n
be
167
a n-vector in V. To find the n-best approximation to þ = þ
1
· … · þ
n
by n-vectors in W = W
1
· … · W
n
. This means to
find a vector o = o
1
· … · o
n
for which || þ – o || = || þ
1

o
1
|| · … · || þ
n
– o
n
|| is as small as possible subject to the
restriction that o = o
1
· … · o
n
should belong to W = W
1
· … · W
n
. i.e., to be more precise.
A n-best approximation to þ = þ
1
· … · þ
n
in W = W
1
· … · W
n
is a n-vector o = o
1
· … · o
n
in W such that
|| þ ÷ o || < || þ ÷ ¸ ||
i.e.,
|| þ
1
÷ o
1
|| · … · || þ
n
÷ o
n
|| s || þ
1
÷ ¸
1
|| · … · || þ
n
÷ ¸
n
||
for every n-vector ¸ = ¸
1
· … · ¸
n
in W.
The following theorem is left as an exercise for the reader.
THEOREM 2.4: Let W = W
1
· … · W
n
be a n-subspace of an n-
inner product space V = V
1
· … · V
n
and þ = þ
1
· … · þ
n
be
a n-vector in V = V
1
· … · V
n
.
a. The n-vector o = o
1
· … · o
n
in W is a n-best
approximation to þ = þ
1
· … · þ
n
by n-vectors in W = W
1
· … · W
n
if and only if þ – o = þ
1
– o
1
· … · þ
n
– o
n
is
n-orthogonal to every vector in W. i.e., each þ
i
– o
i
is
orthogonal to every vector in W
i
; true for i = 1, 2, …, n.
b. If a n-best approximation to þ = þ
1
· … · þ
n
by n-vectors
in W = W
1
· … · W
n
exists, it is unique.
c. If W is finite dimensional and
( )
1
1 1
1
, , !
n
o o · … ·
( )
1
, , !
n
n n
n
o o is any n-orthonormal n-basis for W then the
n-vector
o = o
1
· … · o
n
=
( ) ( )
1
1
1
1
1
1
1
1 2 2
|| || || ||
· ·
¯ ¯
!
n
n
n
n
n
n k k
n
k k
n
k k
k k
þ o þ o
o o
o o
is the unique n-best approximation to þ = þ
1
· … · þ
n
by
n-vectors in W.
168

DEFINITION 2.4: Let V = V
1
· … · V
n
be n-inner product
space and S = S
1
· … · S
n
any set of n-vectors in V. The n-
orthogonal complement of S denoted by S
±
= S
1
±
· … · S
n
±
is
the set of all n-vectors in V which are n-orthogonal to every n-
vector in S.

Whenever the n-vector o = o
1
· … · o
n
in the theorem 2.4
exists its is called the n-orthogonal projection to þ = þ
1
· … ·
þ
n
on W = W
1
· … · W
n
. If every n-vector has an n-orthogonal
projection on W = W
1
· … · W
n
the n-mapping that assigns to
each n-vector in V its n-orthogonal projection on W = W
1

· … · W
n
is called the n-orthogonal projection of V on W.

We leave the proof of this simple corollary as an exercise for

COROLLARY 2.2: Let V = V
1
· … · V
n
be an n-inner product
space, W = W
1
· … · W
n
a finite dimensional n-subspace and
E = E
1
· … · E
n
the n-orthogonal projection of V on W. Then
the n-mapping þ ÷ þ – Eþ i.e., þ
1
· … · þ
n
÷ (þ
1
– E
1
þ
1
) ·
… · (þ
n
– E
n
þ
n
) is the n-orthogonal projection of V on W.

The following theorem is also direct and can be proved by the

THEOREM 2.5: Let W be a finite dimensional n-subspace of an
n-inner product space V = V
1
· … · V
n
and let E = E
1
· … ·
E
n
be the n-orthogonal projection of V on W. Then E = E
1
· …
· E
n
is an idempotent n-linear transformation of V onto W, W
±

is the null space of E and V = W © W
±
i.e. V = V
1
· … · V
n
=
W
1
1
±
W · … · W
n
±
n
W .

The following corollaries are direct and expect the reader to
prove them.

COROLLARY 2.3: Under the conditions of the above theorem 1
– E is the n-orthogonal n-projection of V on W
±
. It is an n-
169
idempotent n-linear transformation of V onto W
±
with n-null
space W.
COROLLARY 2.4: Let
1
1 1
1
{{ , , }
n
o o ! ·
2
2 2
1
{ , , }
n
o o ! · … ·
1
{ , , }}
n
n n
n
o o ! be an n-orthogonal set of nonzero n-vectors in
an n-inner product space V = V
1
· … · V
n
. If þ = þ
1
· … · þ
n

is any n-vector in V then

1 2
1 2
1 2
1 2 2 2
1 2
1 2 2 2
| ( | ) | | ( | ) |
|| || || ||
K K
K K
K K
þ o þ o
o o
· · ·
¯ ¯
!
2
2 2
1
2
| ( | ) |
|| || || ||
|| ||
n
n
n
n
n K
n
n
K
K
þ o
þ þ
o
s · ·
¯
!

and equality holds if and only if
1
1
1
1
1
1
1
2
( | )
|| ||
K
K
K
K
þ o
þ o
o
= · ·
¯
!
1
2
( | )
|| ||
n
n
n
n K
n
K n
n
Kn
þ o
o þ þ
o
= · ·
¯
! .

Now we proceed onto define the new notion of linear
functionals and adjoints in case of n-vector spaces over n-fields
of type II.

Let V = V
1
· … · V
n
be a n-vector space over a n-field F =
F
1
· … · F
n
of type II.
Let f = f
1
· … · f
n
be any n-linear functional on V. Define
n-inner product on V, for every o = o
1
· … · o
n
e V; f(o) =
(o | þ) for a fixed þ = þ
1
· … · þ
n
in V.
f
1
(o
1
) · … · f
n
(o
n
) = (o
1
| þ
1
) · … · (o
n
| þ
n
).
We use this result to prove the existence of the n-adjoint of
a n-linear operator T = T
1
· … · T
n
on V; this being a n-linear
operator
* * *
1 n
T T T = · · ! such that (TĮ | ȕ) = (Į | T
*
ȕ) i.e.,
170
(T
1
Į
1
| ȕ
1
) · …· (T
n
Į
n
| ȕ
n
) = (Į
1
|
*
1
T ȕ
1
) · … · (Į
n
|
*
n
T ȕ
n
) for
all Į, ȕ in V.
By the use of an n-orthonormal n-basis the n-adjoint
operation on n-linear operators (passing from T to T
*
) is
identified with the operation of forming the n-conjugate
transpose of a n-matrix.
Suppose V = V
1
· · ! V
n
is an n-inner product space
over the n-field F = F
1
· · ! F
n
and let ȕ = ȕ
1
· … · ȕ
n
be
some fixed n-vector in V. We define a n-function f
ȕ
from V into
the scalar n-field by f ( )
þ
o = (Į|ȕ);
i.e.
1
1
f
þ

1
) · … ·
n
n
f
þ

n
) = (Į
1
| ȕ
1
) · … · (Į
n
| ȕ
n
)
for Į = Į
1
· … · Į
n
e V = V
1
· … · V
n
.

The n-function f
ȕ
is a n-linear functional on V because by its
very definition (Į|ȕ) is a n-linear n-function on V, arises in this
way from some ȕ = ȕ
1
· … · ȕ
n
e V.

THEOREM 2.6: Let V be a finite (n
1
, …, n
n
) dimensional n-inner
product space and f = f
1
· … · f
n
, a n-linear functional on V =
V
1
· … · V
n
. Then there exists a unique n-vector ȕ = ȕ
1
· … ·
ȕ
n
in V such that f(Į) = (Į | ȕ) i.e. f
1

1
) · … · f
n

n
) = (Į
1
| ȕ
1
)
· … · (Į
n

n
) for all Į in V.

Proof: Let
1 n
1 1 n n
1 n 1 n
{ } { } o o · · o o ! ! ! be a n-orthonormal n-
basis for V. Put
ȕ = ȕ
1
· … · ȕ
n

=
1 n
1 1 n n
1 n
n n
1 1 n n
1 j j n j j
j 1 j 1
f ( ) f ( )
= =
o o · · o o
¯ ¯
!
and let f
ȕ
=
1
1
f
þ
·… ·
n
n
f
þ
be the n-linear functional defined by
f
ȕ
(Į) = (Į|ȕ);
1
1
f
þ

1
) ·…·
n
n
f
þ

n
) = (Į
1

1
) ·…· (Į
n

n
); then
f
ȕ

K
) =
K j j K
j
( | f ( ) ) f ( ) o o o = o
¯
i.e., if Į
K
=
1 n
K K
( ) o · ·o !
then
f
ȕ

K
) =
1
1
1 K
f ( )
þ
o · … ·
n
n
n K
f ( )
þ
o
171
=
1 1 n n
1 n
1 1 1 n n n
K 1 j j K n j j
j j
| f ( ) | f ( )
| | | |
o o o · · o o o
| |
| |
\ . \ .
¯ ¯
!
= f
1
1
K
( ) o · … · f
n
n
K
( ) o .

Since this is true for each Į
K
=
1 n
K K
o · ·o ! , it follows that f =
(f
ȕ
). Now suppose ¸ = ¸
1
· … · ¸
n
is a n-vector in V such that
(Į|ȕ) = (Į|¸) for all Į e V; i.e.,(Į
1

1
) · … · (Į
n
| ȕ
n
) = (Į
1
| ¸
1
)
· … · (Į
n
| ¸
n
). Then (ȕ – ¸ | ȕ – ¸) = 0 and ȕ = ¸. Thus there is
exactly one n-vector ȕ = ȕ
1
· … · ȕ
n
determining the n-linear
functional f in the stated manner.

THEOREM 2.7: For any n-linear operator T = T
1
· … · T
n
on
a finite (n
1
, …, n
n
) dimensional n-inner product space V = V
1
·
… · V
n
, there exists a unique n-linear operator T
*
=
*
1
T · … ·
*
n
T on V such that (TĮ|ȕ) = (Į | T
*
ȕ) for all Į, ȕ in V.

Proof: Let ȕ = ȕ
1
· … · ȕ
n
be any n-vector in V. Then Į ÷
(TĮ|ȕ) is a n-linear functional on V. By the earlier results there
is a unique n-vector ȕ' =
1
' þ · … ·
n
'
þ in V such that (TĮ|ȕ) =
(Į|ȕ') for every Į in V. Let T
*
denote the mapping ȕ ÷ ȕ'; ȕ' =
T
*
ȕ. We have (T
1
Į
1
| ȕ
1
) · … · (T
n
Į
n
| ȕ
n
) =
1 *
1 1 1
( | T ) o þ · …
· (Į
n
|
*
n
T ȕ
n
), so must verify that T
*
=
*
1
T · … ·
*
n
T is an n-
linear operator. Let ȕ, ¸ be in V and c = c
1
· … · c
n
be a n-
scalar. Then for any Į,

(Į | T
*
(cȕ + ¸)) = (TĮ | cȕ + ¸)
= (TĮ|cȕ) + (TĮ|¸)
= c(TĮ|ȕ) + (TĮ | ¸)
= c(Į|T
*
ȕ) + (Į | T
*
¸)
= (Į | cT
*
ȕ) + (Į |T
*
¸)
= (Į | cT
*
ȕ + T
*
¸)

i.e. (o
1
|
*
1
T (c
1
ȕ
1
+ ¸
1
)) · … · (o
n
|
*
n
T (c
n
ȕ
n
+ ¸
n
)) =

1
| c
1
*
1
T ȕ
1
+
*
1
T ¸
1
) · … · (o
n
| c
n
*
n
T ȕ
n
+
*
n
T ¸
n
).
172

The uniqueness of T
*
is clear. For any ȕ in V the n-vector T
*
ȕ is
uniquely determined as the vector ȕ' such that (TĮ|ȕ) = (Į|ȕ') for
every Į.

THEOREM 2.8: Let V = V
1
· … · V
n
be a finite (n
1
, …, n
n
)
dimensional n-inner product space and let B =
1
1 1
1
{ }
n
o o ! · …
·
1
{ }
n
n n
n
o o ! be an n-ordered n-orthogonal basis for V. Let T
= T
1
· … · T
n
be a n-linear operator on V and let A = A
1
· …
· A
n
be the n-matrix of T in the ordered n-basis B. Then A
Kj
=
(TĮ
j
| Į
K
) i.e.
1 1
1
n n
n
K j K j
A A · · ! =
1 1
1 1
1
( | ) ( | )
n n
n n
j K n j K
T T o o o o · · ! .

Proof: Since B =
1 n
1 1 n n
1 n 1 n
{ } { } o o · · o o ! ! ! is an ordered n-
basis we have
o =
1 n
1 n
1 n
n n
1 1 n n
1 K K n K K
K 1 K 1
( | ) ( | )
= =
o o o · · o o o
¯ ¯
!
= Į
1
· … · Į
n
.
The n-matrix A is defined by
n
j Kj K
K 1
T A
=
o = o
¯
, i.e.,
T
1
1 n
1 n 1 1 1 n n n
1 n
n n
1 n 1 1 n n
j n j K j K K j K
K 1 K 1
T A A
= =
o · · o = o · · o
¯ ¯
! !
since
n
j j K K
K 1
T (T | )
=
o = o o o
¯
;
i.e.,
1
1 n 1 1 1
1
n
1 n 1 1 1
1 j n j 1 j K K
K 1
T T (T | )
=
o · · o = o o o
¯
!
n
n n n
n
n
n n n
n j K K
K 1
(T | )
=
· · o o o
¯
!
we have A
Kj
= (TĮ
j
| Į
K
);

1 1 n n 1
1 n 1 1
K j K j 1 j1 K
A A (T | ) · · = o o ! · … ·
n
n 1
n jn K
(T | ) o o .
173

The following corollary is immediate and left for the reader to
prove.

COROLLARY 2.5: Let V be a finite (n
1
, …, n
n
) dimensional n-
inner product space over the n-field F and set T = T
1
· … · T
n

a n-linear operator on V. In any n-orthogonal basis for V, the n-
matrix of T
*
is the n-conjugate transpose of the n-matrix of T.

DEFINITION 2.5: Let T = T
1
· … · T
n
be a n-linear operator
on an inner product space V = V
1
· … · V
n
, then we say that T
= T
1
· … · T
n
has an n-adjoint on V if there exists a n-linear
operator T
*
=
*
1
T · … ·
*
n
T on V such that (TĮ|ȕ) = (Į|T
*
ȕ) i.e.
(T
1
Į
1

1
) · … · (T
n
Į
n
| ȕ
n
) = (Į
1
|
*
1
T ȕ
1
) · … · (Į
n
|
*
n
T ȕ
n
) for
all Į = Į
1
· … · Į
n
and ȕ = ȕ
1
· … · ȕ
n
in V = V
1
· … · V
n
.

It is left for the reader to prove the following theorem.

THEOREM 2.9: Let V = V
1
· … · V
n
be a finite dimensional n-
inner product space over the n-field F = F
1
· … · F
n
. If T and
U are n-linear operators on V and c is n-scalar.

i. (T + U)
*
= T
*
+ U
*

i.e., (T
1
+ U
1
)
*
· … · (T
n
+ U
n
)
*

=
* *
1 1
T U + · …· (
* *
n n
T U + ).
ii. (cT)
*
= cT
*
.
iii. (TU)
*
= U
*
T
*
.
iv. (T
*
)
*
= T.
DEFINITION 2.6: Let V = V
1
· … · V
n
and W = W
1
· … · W
n

be n-inner product spaces over the same n-field F = F
1
· … ·
F
n
and let T =
*
1
T · … ·
*
n
T be a n-linear transformation from
V into W. We say that T-n-preserves inner products if (TĮ|Tȕ) =
(Į|ȕ) i.e., (T
1
Į
1
| T
1
ȕ
1
) · … · (T
n
Į
n
| T
n
ȕ
n
) = (Į
1
| ȕ
1
) · … ·

n

n
) for all Į, ȕ in V. An n-isomorphism of V onto W is a n-
vector space n-isomorphism T of V onto W which also preserves
inner products.
174

THEOREM 2.10: Let V and W be finite dimensional n-inner
product spaces over the same n-field F = F
1
· … · F
n
. Both V
and W are of same n-dimension equal to (n
1
,…,n
n
). If T = T
1
·
… · T
n
is a n-linear transformation from V into W, the
following are equivalent.
i. T = T
1
· … · T
n
preserves n-inner products.
ii. T = T
1
· … · T
n
is an n-isomorphism.
iii. T = T
1
· … · T
n
carries every n-orthonormal n-basis
for V into an n-orthonormal n-basis for W.
iv. T carries some n-orthonormal n-basis for V onto an n-
orthonormal n-basis for W.

Proof: Clearly (i) ÷ (ii) i.e., if T = T
1
· … · T
n
preserves n-
inner products, then |TĮ|| = ||Į|| for all Į = Į
1
· … · Į
n
in V,
i.e., || T
1
Į
1
|| · … · || T
n
Į
n
|| = || Į
1
|| · … · || Į
n
||. Thus T is n-
non singular and since n-dim V = n-dim W = (n
1
, …, n
n
) we
know that T = T
1
· … · T
n
is a n-vector space n-isomorphism.

(ii) ÷ (iii) Suppose T = T
1
· … · T
n
is an n-isomorphism. Let
1 n
1 1 n n
1 n 1 n
{ } { } o o · · o o ! ! ! be an n-orthonormal basis for V.
Since T = T
1
· … · T
n
is a n-vector space isomorphism and n-
dim V = n-dim W, it follows that
1
1
1 1
1 n n
{T , , T } o o ! ·
2
1 2
2 2 2 n
{T , , T } o o ! · … ·
n
n n
n 1 n n
{T , , T } o o ! is a n-basis for W.
Since T also n-preserves inner products (TĮ
j
| TĮ
K
) = (Į
j
| Į
K
) =
o
jK
,

i.e.,
(T
1

1 1
1 1
j 1 K
| T ) o o · … ·
n n
1 n
n j n K
(T | T ) o o
1 1 n n 1 1
1 1 n n
j K j K j K
( | ) ... ( | ) o o · · o o =o
n n
j K
· · o ! .

(iii) ÷ (iv) is obvious.
(iv) ÷ (i).
Let
1 n
1 1 n n
1 n 1 n
{ } { } o o · · o o ! ! ! be an n-orthonormal basis for
V such that
1 n
1 1 n n
1 1 1 n n n n n
{T , , T } {T , , T } o o · · o o ! ! ! is an n-
orthonormal basis for W. Then (TĮ
j
| TĮ
K
) = (Į
j
| Į
K
) = o
jK
;
175
i.e.,
1 1 n n
1 1 n n
1 j 1 K n j n K
{T | T } {T | T } o o · · o o !
=
1 1 n n
1 1 n n
j K j K
( | ) ( | ) o o · · o o !
=
1 1 n n
j K j K
o · ·o ! .
For any
Į = Į
1
· · ! Į
n

=
1 1
1 1 1 1
1 1 n n
x x o + + o · · ! !
n n
n n n n
1 1 n n
x x o + + o !
and
þ =
1 1 n n
1 1 1 1 n n n n
1 1 n n 1 1 n n
y y y y o + + o · · o + + o ! ! !
in V we have
(Į|ȕ) =
n
j j
j 1
x y
=
¯

that is
[
1
1 1
1
n
1 1
1 1 n n j j
j 1
( | ) ( | ) ] x y
=
o þ · · o þ =
¯
! · · !
n
n n
n
n
n n
j j
j 1
x y
=
¯
.

(TĮ | Tȕ) = (T
1
Į
1
| T
1
ȕ
1
) · … · (T
n
Į
n
| T
n
ȕ
n
)
=
j j K K
j K
x T y T
| |
o o
|
|
\ .
¯ ¯

=
1 1 1 1
1 1
1 1 1 1
j 1 j K 1 K
j K
x T y T
| |
o o · ·
|
|
\ .
¯ ¯
!
n n n n
n n
n n n n
j n j K n K
j K
x T y T
| |
o o
|
|
\ .
¯ ¯

=
j K j K
j K
x y (T | T ) o o
¯¯

=
1 1 1 1
1 1
1 1
j K 1 j 1 K
j K
x y (T | T ) o o · ·
¯¯
!
n n n n
n n
n n n n
j K n j n K
j K
x y (T | T ) o o
¯¯

176
=
n
j j
j 1
x y
=
¯

=
n
1 1 n n
1 n
n
n
1 1 n n
j j j j
j 1 j 1
x y x y
= =
· ·
¯ ¯
!

and so T-n-preserves n-inner products.

COROLLARY 2.6: Let V = V
1
· … · V
n
and W = W
1
· … · W
n

be finite dimensional n-inner product spaces over the same n-
field F = F
1
· … · F
n
. Then V and W are n-isomorphic if and
only if they have the same n-dimension.

Proof: If
1 n
1 1 n n
1 n 1 n
{ } { } o o · · o o ! ! ! is a n-orthonormal n-
basis for V = V
1
· … · V
n
and
1 n
1 1 n n
1 n 1 n
{ } { } þ þ · · þ þ ! ! ! is
an n-orthonormal n-basis for W; let T be the n-linear
transformation from V into W defined by TĮ
j
= ȕ
j
i.e.
1 n 1 n
1 n 1 n
1 j n j j j
T T o · · o =þ · ·þ ! ! . Then T is an n-isomorphism
of V onto W.

THEOREM 2.11: Let V = V
1
· … · V
n
and W = W
1
· … · W
n

be two n-inner product spaces over the same n-field and let T =
T
1
· … · T
n
be a n-linear transformation from V into W. Then
T-n-preserves n-inner products if and only if ||TĮ|| = ||Į|| for
every Į in V.

Proof: If T = T
1
· … · T
n
, n-preserves inner products then T-
n-preserves norms. Suppose ||TĮ|| = || Į || for every Į = Į
1
· …
· Į
n
in V; i.e., ||T
1
Į
1
|| · … · ||T
n
Į
n
|| = || Į
1
|| · … · || Į
n
||.
Now using the appropriate polarization identity for real space
and the fact T is n-linear we see (Į | ȕ) = (TĮ | Tȕ) i.e. (Į
1

1
) ·
… · (Į
n

n
) = (T
1
Į
1
| T
1
ȕ
1
) · … · (T
n
Į
n
| T
n
ȕ
n
) for all Į = Į
1
·
… · Į
n
and ȕ = ȕ
1
· … · ȕ
n
in V.
Recall a unitary operator on an inner product space is an
isomorphism of the space onto itself.
A n-unitary operator on an n-inner product space is an n-
isomorphism of the n-space V onto itself.
177

It is left for the reader to verify that the product of two n-unitary
operator is n-unitary.

THEOREM 2.12: Let U = U
1
· … · U
n
be a n-linear operator
on an n-inner product space V = V
1
· … · V
n
. Then V is n-
unitary if and only if the n-adjoint U
*
of U exists and UU
*
=
U
*
U = I.

Proof: Suppose U = U
1
· … · U
n
is n-unitary. Then U is n-
invertible and (UĮ | ȕ) = (UĮ|UU
-1
ȕ) = (Į | U
-1
ȕ) for all Į = Į
1

· … · Į
n
and ȕ = ȕ
1
· … · ȕ
n
in V. Hence U
-1
of U for
(U
1
Į
1
| ȕ
1
) · … · (U
n
ȕ
n
| ȕ
n
)
= (U
1
Į
1
| U
1
1
1
U
÷
ȕ
1
) · … · (U
n
o
n
|U
n
1
n
U
÷
ȕ
n
)
= (Į
1
|
1
1
U
÷
ȕ
1
) · · ! (Į
n
|
1
n
U
÷
ȕ
n
).

Conversely suppose U
*
exists and UU
*
= U
*
U = I i.e.,
U
1
*
1
U · … · U
n
*
n
U
=
*
1
U U
1
· … ·
*
n
U U
n

= I
1
· … · I
n
.

Then U = U
1
· … · U
n
is n-invertible with U
-1
= U
*
; i.e.,
1 1
1 n
U U
÷ ÷
· · ! =
*
1
U · … ·
*
n
U . So we need only show that U
= U
1
· · ! U
n
preserves n-inner products. We have (UĮ|Uȕ)
= (Į | U
*
Uȕ) = (Į|Iȕ) = (Į | ȕ);
i.e.,
(U
1
Į
1
| U
1
ȕ
1
) · · ! (U
n
Į
n
| U
n
ȕ
n
)
= (Į
1
|
*
1
U U
1
ȕ
1
) · · ! (Į
n
|
*
n
U U
n
ȕ
n
)
= (Į
1
|I
1
ȕ
1
) · · ! (Į
n
|I
n
ȕ
n
) = (Į
1

1
) · · ! (Į
n

n
)

for all Į, ȕ e V.
We call a real n-mixed square matrix A = A
1
· … · A
n

over the n-field F = F
1
· … · F
n
to be n-orthogonal if A
t
A = I
i.e.,
t
1
A A
1
· … ·
t
n
A A
n
= I
1
· … · I
n
. We say A = A
1
· … ·
178
A
n
to be n-anti orthogonal if A
t
A= – I, i.e.,
t
1
A A
1
· … ·
t
n
A A
n

= –I
1
· –I
2
· … · –I
n
.
Let V = V
1
· … · V
n
be a finite dimensional n-inner
product space and T = T
1
· … · T
n
be a n-linear operator on V.
We say T is n-normal if it n-commutes with its n-adjoint i.e.
TT
*
= T
*
T, i.e., T
1
*
1
T · … · T
n
*
n
T =
*
1
T T · … ·
*
n
T T
n
.

The following theorem speaks about the properties enjoyed by
n-self adjoint n-operators on a n-vector space over the n-field of
type II.

THEOREM 2.13: Let V = V
1
· … · V
n
be an n-inner product
space and T = T
1
· … · T
n
be a n-self adjoint operator on V.
Then each n-characteristic value of T is real and n-
characteristic vectors of T associated with distinct n-
characteristic values are n-orthogonal.

Proof: Suppose c = c
1
· … · c
n
is a n-characteristic value of T
i.e., TĮ = cĮ for some nonzero n-vector Į = Į
1
· … · Į
n
i.e.,
T
1
Į
1
· … · T
n
Į
m
= c
1
Į
1
· … · c
n
Į
n
.

c(Į|Į) = (cĮ|Į) = (TĮ | Į)
= (Į |TĮ) = (Į | cĮ)
= c (Į|Į)
i.e.,
c(Į|Į) = c
1

1

1
) · … · c
n
(Į|Į
n
)
= (c
1
Į
1
| Į
1
) · … · (c
n
Į
n

n
)
= (T
1
Į
1
| Į
1
) · … · (T
n
Į
n

n
)
= (Į
1
| T
1
Į
1
) · … · (Į
n
|T
n
Į
n
)
= (Į
1
| c
1
Į
1
) · … · (Į
n
|c
n
Į
n
)
=
1
c (Į
1

1
) · … ·
n
c (Į
n
| Į
n
).
Since (Į | Į) = 0 · … · 0; i.e., (Į
1
| Į
1
) · … · (Į
n
| Į
n
) = (0 ·
… · 0). We have c = c , i.e., c
i
=
i
c for i = 1, 2, …, n. Suppose
we also have Tȕ = dȕ with ȕ = 0 i.e., ȕ= ȕ
1
· … · ȕ
n
= 0 · …
· 0.

(cĮ|ȕ) = (TĮ|ȕ) = (Į | Tȕ)
179
= (Į | dȕ) = d(Į| ȕ)
= d(Į | ȕ).

If c = d then (Į | ȕ) = 0 · … · 0; i.e., if d
1
· … · d
n
= c
1
· …
· c
n
i.e., c
i
= d
i
for i = 1, 2, …, n then (Į
1
| ȕ
1
) · … · (Į
n
| ȕ
n
) =
0 · … · 0.

We derive the spectral theorem for n-inner product n-vector
space over the n-real field F = F
1
· … · F
n
.

THEOREM 2.14: Let T = T
1
· … · T
n
operator on a finite (n
1
, …, n
n
) dimensional n-inner product
vector space V = V
1
·…· V
n
. Let
1
1 1
1 1
{ } { }
n
n n
K K
c c c c · · ! ! !
be the n-distinct n-characteristic values of T. Let
1
1
n
n
j j j
W W W = · · ! be a n-characteristic space associated with
n-scalar
1
1
n
n
j j j
c c c = · · ! and
1
1
n
n
j j j
E E E = · · ! be the n-
orthogonal projection of V on W
j
i.e., V
t
on
t
j
W for t = 1, 2, …,
n. Then W
i
is n-orthogonal to W
j
if i = j i.e., if W
i
=
1
1
n
n
i i
W W · · ! and W
j
=

1
1
n
n
j j
W W · · ! then
t
t
i
W is orthogonal
to
t
t
j
W if i
t
= j
t
for t = 1, 2, …, n. V is the n-direct sum of
1
1 1
1 1
{ , , } { , , } · · ! ! !
n
n n
K K
W W W W and
T = T
1
· … · T
n

=
1 1
1 1
1 1 1 1
[ ]
K K
c E c E + + · · ! !
1 1
[ ]
n n
K K
n n n n
c E c E + + ! .
Proof: Let Į = Į
1
· … · Į
n
be a n-vector in
W
j
=
1 n
1 n
j j
W W · · ! .
ȕ = ȕ
1
· … · ȕ
n
be a n-vector in
W
i
=
1 n
1 n
i i
W W · · !
and suppose i = j; c
j
(Į|ȕ) = (TĮ|ȕ) = (Į|T
*
ȕ) = (Į|c
i
ȕ) i.e.,

1 n
1 n
j 1 1 j n n
c ( | ) c ( | ) o þ · · o þ !
= (Į
1
|
1 n
1 n
i 1 n i n
c ) ( | c ) þ · · o þ ! .
180

Hence (c
j
– c
i
) (Į|ȕ) = 0 · · ! 0. i.e.,
1 1 n n
1 1 n n
j i 1 1 j i n n
(c c )( | ) (c c )( | ) ÷ o þ · · ÷ o þ ! = 0 · … · 0;
since c
j
– c
i
= 0 · … · 0. It follows (Į|ȕ) = (Į
1
| ȕ
1
) · … ·

n

n
) = 0 · … · 0. Thus W
j
=
1 n
1 n
j j
W W · · ! is n-orthogonal
to W
i
=
1 n
1 n
i i
W W · · ! when i = j. From the fact that V has an
n-orthonormal basis consisting of n-characteristic, it follows
that V = W
1
+ … + W
K
i.e.,
V = V
1
· · ! V
n

=
1 n
1 1 n n
1 K 1 K
(W W ) (W W ) + + · · + + ! ! !
when i
t
= j
t
; 1 s i
t
, j
t
s K
t
or t = 1, 2, …, n. From the fact that V
= V
1
· … · V
n
has an n-orthonormal n-basis consisting of n-
characteristic n-vectors it follows that V = W
1
+ … + W
K
, If
t t
t t
j j t t
W (1 j K ) o e s s and
1 n
1 1 n n
1 K 1 K
( ) ( ) o + +o · · o + +o ! ! !
= 0 · … · 0 then 0 · … · 0
=
i j
j
( | )
| |
o o
|
\ .
¯

=
i j
j
( | ) o o
¯

i.e.
1 1 n n
1
1 1 n n
i j i j
j
( | ) ( | )
| |
| |
o o · · o o
|
|
|
\ .
\ .
¯ ¯
!
1 1 n n
1 n
1 1 n n
i j i j
j j
( | ) ( | ) = o o · · o o
¯ ¯
!
1 n
1 2 n 2
i i
|| || || || = o · · o !

for every 1 < i
t
< K
t
; t = 1, 2, …, n so that V is the n-direct sum
of
1 n
1 1 n n
1 K 1 K
(W , , W ) (W , , W ) · · ! ! ! . Therefore
1 n
1 1 n n
1 K 1 K
E E E E + + · · + + ! ! ! = I
1
· … · I
n

and

T = (T
1
1 n
1 1 n n
1 1 K n 1 n K
E TE ) (T E T E ) + + · · + + ! ! " =
1 1 n n
1 1 1 1 n n n n
1 1 K K 1 1 K K
(c E c E ) (c E c E ) + + · · + + ! ! ! .
181

This decomposition is called the n-spectral resolution of T. The
following corollary is immediate however we sketch the proof
of it.

COROLLARY 2.7: If
1
1
= · · !
n
n
j j j
e e e
=
1
1 1
1 1
1
1 1
=
| | ÷
|
|
÷
\ .
I
i
i j
j i
x c
c c
· … ·
=
| | ÷
|
|
÷
\ .
I
n
n n
n n
n
i
n n
i j
j i
x c
c c

then E
j
=
1
1
· · !
n
n
j j
E E for 1s j
t
s K
t
; t = 1, 2, …, n.

Proof: Since
t t
t t
i j
E E 0 = for every 1 < i
t
, j
t
< K
t
; t = 1, 2, …, n (i
t

= j
t
) it follows that T
2
= (T
1
· … · T
n
)
2
=
2 2
1 n
T T · · ! .
1 1 n n
1 2 1 1 2 1 n 2 n n 2 n
1 1 K K 1 1 K K
(c ) E (c ) E (c ) E (c ) E + + · · + + ! ! !
and by an easy induction argument that
1 1
n 1 n 1 1 n 1
1 1 K K
T (c ) E (c ) E

= + + · ·

! !
n n
n n n n n n
1 1 K K
(c ) E (c ) E

+ +

!
for every n-integer (n
1
, …, n
n
) > (0, 0, …, 0). For an arbitrary n-
polynomial
r
n
n
n 0
f a x
=
=
¯
;
f = f
1
· … · f
n
=

1 n
1 n
1 n
1 n
r r
n n 1 n
n n
n 0 n 0
a x a x
= =
· ·
¯ ¯
! ,
we have
f(T) = f
1
(T
1
) · … · f
n
(T
n
)
=
1 n
1 n
1 n
1 n
r r
n n 1 n
n 1 n n
n 0 n 0
a T a T
= =
· ·
¯ ¯
!
=
1 1
1
1 1 1
1 1
r K
n 1 1
n j j
n 0 j 1
a c E
= =
· ·
¯ ¯
!
n n
n
n n n
n n
r K
n n n
n j j
n 0 j 1
a c E
= =
¯ ¯

182
=
1 1
1
1 1 1
1 1
K r
n 1 1
n j j
j 1 n 0
a c E
= =
| |
· ·
|
|
\ .
¯ ¯
!
n n
n
n n n
n n
K r
n n n
n j j
j 1 n 0
a c E
= =
| |
|
|
\ .
¯ ¯

=
1 n
1 1 n n
1 n
K K
1 1 n n
1 j j n j j
j 1 j 1
f (c ) E f (c ) E
= =
· ·
¯ ¯
! .

Since
1 1 1 1
1 t
j m j m
e (c ) = o for t = 1, 2, …, n it follows
t t
t t
j t j
e (T ) E =
true for t = 1, 2, …, n.
Since
1 n
1 1 n n
1 K 1 K
{E E } {E E } · · ! ! ! are canonically associated
with T and I
1
· … · I
n
= (
1 n
1 1 n n
1 K 1 K
E E ) ( E E ) + + · · + + ! ! !
the family of n-projections
1 n
1 1 n n
1 K 1 K
{E E } {E E } · · ! ! ! is
called the n-resolution of the n-identity defined by T = T
1
· …
· T
n
.

Next we proceed onto define the notion of n-diagonalized and
the notion of n-diagonalizable normal n-operators.

DEFINITION 2.7: Let T = T
1
· … · T
n
be a n-diagonalizable
normal n-operator on a finite dimensional n-inner product
space and
T = T
1
· … · T
n
=
1
1 1
1
1 1
1 1 = =
· ·
¯ ¯
!
n
n n
n
K K
n n
j j j j
j j
c E c E
is its spectral n-resolution.
Suppose f = f
1
· … · f
n
is a n-function whose n-domain
includes the n-spectrum of T that has n-values in the n-field of
scalars. Then the n-linear operator f(T) = f
1
(T
1
) · … · f
n
(T
n
) is
defined by the n-equation
f(T) = f
1
(T
1
) · … · f
n
(T
n
)
=
1
1 1
1
1 1
1
1
( )
=
· ·
¯
!
K
j j
j
f c E
1
( )
=
¯
n
n n
n
K
n n
n j j
j
f c E .

Now we prove the following interesting theorem for n-
diagonalizable normal n-operator.

183
THEOREM 2.15: Let T = T
1
· … · T
n
be a n-diagonalizable
normal n-operator with n-spectrum S = S
1
· … · S
n
on a finite
(n
1
, …, n
n
) dimensional n-inner product space V = V
1
· … ·
V
n
. Suppose f = f
1
· … · f
n
is a n-function whose domain
contains S that has n-values in the field of n-scalars. Then f(T)
is a n-diagonalizable normal n-operator with n-spectrum f(S) =
f
1
(S
1
) · … · f
n
(S
n
). If U = U
1
· … · U
n
is a n-unitary map of V
onto V' and T' = UTU
-1
= U
1
T
1
1 1
1
÷ ÷
· · !
n n n
U U T U then S = S
1

· … · S
n
is the n-spectrum of T' and f(T') = f
1
1
(T ) ' · … ·
f
n
(T'
n
) = Uf(T)U
–1
= U
1
f
1
(T
1
)
1 1
1
( )
÷ ÷
· · !
n n n n
U U f T U .

Proof: The n-normality of f(T) = f
i
(T
i
) · … · f
n
(T
n
) follows by
a simple computation from the earlier results and the fact

f(T)
*
= f
1
(T
1
)
*
· … · f
n
(T
n
)
*

=
1 1 n n
1 n
1 1 n n
j j j j
j j
f (c )E f (c )E · ·
¯ ¯
! .

Moreover it is clear that for every Į = Į
1
· … · Į
n
in E
j
(V) =
1 n
1 n
j 1 j n
E (V ) E (V ) · · ! ; f(T)Į = f(c
j
)Į, i.e., f
1
(T
1

1
· … ·
f
n
(T
n

n
= f
1
1
1
j
(c ) Į
1
· … · f
n
n
n
j n
(c )o . Thus the set f(S) of all
f(c) with c in S is contained in the n-spectrum of f(T).
Conversely, suppose Į = Į
1
· … · Į
n
= 0 · … · 0 and
that f(T)Į = bĮ i.e., f
1
(T
1

1
· … · f
n
(T
n

n
= b
1
Į
1
· … · b
n
Į
n
.
Then Į =
j
j
E o
¯
i.e.,
Į
1
· … · Į
n
=
1
1
j
j
E
¯
Į
1
· … ·
n
n
n
j n
j
E o
¯

and
f(T)Į = f
1
(T
1

1
· … · f
n
(T
n

n

=
1
1 1
j
f (T )
¯
1
1
j 1
E o · … ·
n
n n
j
f (T )
¯
n
n
j n
E o
=
1
1
j
f
¯
1 1
1 1
j j 1
(c )E o · … ·
n
n
j
f
¯
n n
n n
j j n
(c )E o
=
1
1
j
b
¯
1
1
j 1
E o · … ·
n
n
j
b
¯
n
n
j n
E o .
184
Hence
2
j j
j
(f (c ) b) E ÷ o
¯

=
1 1
1
2
1 1
1 j 1 j 1
j
(f (c ) b ) E ÷ o
¯
· … ·
n n
n
2
n n
n j n j n
j
(f (c ) b ) E ÷ o
¯

=
1 1
1
2 2
1 1
1 j 1 j 1
j
f (c b ) E ÷ o
¯
· … ·
n n
n
2 2
n n
n j n j n
j
f (c b ) E ÷ o
¯
.

Therefore f(c
j
) = f
1
1
1
j
(c ) · … · f
n
n
n
j
(c ) = f
1
· … · f
n
or
1
1
j 1
E o
· … ·
n
n
j n
E o = 0 · … · 0; by assumption Į = Į
1
· … · Į
n
=
0 · … · 0 so there exists an n-index tuple i = (i
1
, …, i
n
) such
that E
i
Į =
1 n
i 1 i n
E E o · · o ! = 0 · … · 0. By assumption Į =
Į
1
· … · Į
n
= 0 · … · 0. It follows that f(c
i
) = b i.e., f
1
i
1
(c ) ·
… · f
n
i
n
(c ) = b
1
· … · b
n
and hence that f(S) is the n-
spectrum of f(T) = f
1
(T
1
) · … · f
n
(T
n
). Infact that f(S) =
1
1 1
1 r
{b , , b } ! · … ·
n
n n
1 r
{b , , b } ! = f
1
(S
1
) · … · f
n
(S
n
) where
t t
t t
m n
b b = ; t = 1, 2, …, n; where m
t
= n
t
. Let X
m
=
1
m
X · … ·
n
m
X be the set of indices i = (i
1
, …, i
n
) such that
f(c
i
) = f
1
1
1
i
(c ) · … · f
n
n
n
i
(c ) =
1
m
b · … ·
n
m
b .
Let
P
m
=
1
m
P · … ·
n
m
P
=
i
i
E
¯

=
1 n
1 n
1 n
i i
i i
E E · ·
¯ ¯
!

the sum being extended over the n-indices (i
1
, …, i
n
) in X
m
=
1
1
m
X · … ·
n
n
m
X . Then P
m
=
1
1
m
P · … ·
n
n
m
P is n-orthogonal
projection of V = V
1
· … · V
n
on the n-subspace of n-
185
characteristic n-vectors belonging to the n-characteristic value
b
m
of f(T) i.e.,
1
1
m
b · … ·
n
n
m
b of f
1
(T
1
) · … · f
n
(T
n
) and
f(T) = f
1
(T
1
) · … · f
n
(T
n
)
=
1
1 1
1
r
1 1
m m
m 1
b P
=
¯
· … ·
n
n n
n
r
n n
m m
m 1
b P
=
¯

is the n-spectral resolution of f(T) = f
1
(T
1
) · · ! f
n
(T
n
). Now
suppose U = U
1
· … · U
n
is a n-unitary transformation of V
onto V' and that T' = U TU
–1
, i.e., T'
1
· … · T'
n
= U
1
T
1
1
1
U
÷
·
… · U
n
T
n
1
n
U
÷
. Then the equation TĮ = cĮ i.e., T
1
Į
1
· … ·
T
n
Į
n
= c
1
Į
1
· … · c
n
Į
n
holds good if and only if T'UĮ = cUĮ
i.e.,
1 1 1 n n n
T U T U ' ' o · · o ! = c
1
U
1
Į
1
· … · c
n
U
n
Į
n
. Thus S = S
1
·
… · S
n
is the n-spectrum of T' = T'
1
· … · T'
n
and U = U
1
·
… · U
n
maps each n-characteristic n-subspace for T = T
1
· …
· T
n
onto the corresponding n-subspace for T'. In fact using
earlier results we see that
j j
j
T c E ' ' =
¯

1 n
T T ' ' · · ! =
1 1
1
1 1
j j
j
c (E )' · ·
¯
!
n n
n
n n
j j
j
c (E )'
¯

Here
E' =
1 n
1 n
j j
(E ) (E ) ' ' · · ! .

U
1
1 n
1 1 n 1
j 1 n j n
E U U E U
÷ ÷
· · ! is the n-spectral resolution of T' =
1 n
T T ' ' · · ! . Hence

f(T') = f
1
(T'
1
) · · ! f
n
(T'
n
)
=
1 1 n n
1 n
1 1 n n
1 j j n j j
j j
f (c )(E ) f (c )(E ) ' ' · ·
¯ ¯
!
=
1 1
1
1 1 1
1 j 1 j 1
j
f (c )U E U
÷
¯
· … ·
n n
n
n n 1
n j n j n
j
f (c )U E U
÷
¯

is the n-spectral resolution of T' = T'
1
· … · T'
n
.
Hence
f(T') = f
1
(T'
1
) · … · f
n
(T'
n
)
186
=
1 1
1
1 1 1
1 j 1 j 1
j
f (c ) U E U
÷
¯
· … ·
n n
n
n n 1
n j n j n
j
f (c )U E U
÷
¯

= U
1

1 1
1
1 1 1
1 j j 1
j
f (c ) (E ) U
÷
¯
· … · U
n
n n
n
n n 1
n j j n
j
f (c )(E ) U
÷
¯

= U
1
f
1
(T
1
)
1
1
U
÷
· · ! U
n
f
n
(T
n
)
1
n
U
÷

= Uf(T) U
-1
.

In view of the above theorem we have the following corollary.

COROLLARY 2.8: With the assumption of the above theorem
suppose T = T
1
· … · T
n.
is represented in an n-ordered basis
B = B
1
· … · B
n
=
1
1 1
1 n
{ } o o ! · … ·
n
n n
1 n
{ } o o ! by the n-
diagonal matrix D = D
1
· … · D
n
with entries
1
1 1
1
{ }
n
d d ! · …
·
1
{ }
n
n n
n
d d ! . Then in the n-basis B, f(T) = f
1
(T
1
) · … · f
n
(T
n
)
is represented by the n-diagonal matrix f(D) = f
1
(D
1
) · · !
f
n
(D
n
) with entries
1 1
1 1
1 1
{ ( ), , ( )}
n n
f d f d ! · … ·
1
{ ( )
n
n
f d , …,
( )}
n n
n
n n
f d . If B' =
1 1
1
{( ) , , ( ) }
n
o o ' ' ! · … ·
1
{( ) , , ( ) }
n n
n
o o ' ' !
is any other n-ordered n-basis and P the n-matrix such that
t t t t
t
t t t
j i j i
i
P þ o =
¯

for t = 1, 2, …, n then P
-1
(f (D)) P is the n-basis B'.

Proof: For each n-index i = (i
1
, …, i
n
) there is a unique n-tuple, j
= (j
1
, …, j
n
) such that 1 s j
t
,

n
p
s K
t
; t = 1, 2, …, n.
1 n
1 n
i i i
o = o · ·o ! belongs to
t
j t
E (V ) and
t t
t t
i j
d c = for every t =
1, 2, …, n. Hence f(T)Į
i
= f(d
i

i
for every i = (i
1
, …, i
n
) i.e.
1
1
1 1 i
f (T )o · … ·
n 1 1
n 1 1
n n i 1 i i
f (T ) f (d ) o = o · … ·
n n
n n
n i i
f (d )o and
f(T)
1
1
j 1 1 j
f (T )( ) ' ' o = o · … ·
n
n
n n j
f (T )( )' o .

ij i
i
P f (T) o
¯
=
1 1 1
1
1 1
i j 1 1 i
i
P f (T ) o
¯
· … ·
n n n
n
n n
i j n n i
i
P f (T )o
¯

187
i ij i
i
d P o
¯
=
1 1 1 1
1
1 1 1
i i j i
i
d P o
¯
· … ·
n n n n
n
n n n
i i j i
i
d P o
¯

=
ij i
i
(DP) o
¯

=
1 1 1
1
1 1
i j i
i
(D P ) o
¯
· … ·
n n n
n n
i j i
(D P ) o
¯

=
1
ij Kt K
i K
(DP) P
÷
' o
¯ ¯

=
1 1 1 1 1
1 1
1 1
i j K t K
i K
(DP) P ( )
÷
' o
¯ ¯
· … ·
n n n n n
n n
1 n
i j K t K
i K
(DP) P ( )
÷
' o
¯ ¯

=
1 1 1
1
1 1
1 1 1 K j K
K
(P D P ) ( )
÷
' o
¯
· … ·

n n n
n
1 1 n
n n n K j K
K
(P D P ) ( )
÷ ÷
' o
¯
.

Under the above conditions we have f(A) = f
1
(A
1
) · …· f
n
(A
n
)
= P
–1
f(D) P =
1
1 1 1 1
P f (D )P
÷
· … ·
1
n n n n
P f (D )P
÷
.

The reader is expected to derive other interesting analogous
results to usual vector spaces for the n-vector spaces of type II.
Now we proceed onto define the notion of bilinear n-forms,
for n-vector spaces of type II.

DEFINITION 2.8: Let V = V
1
· … · V
n
be a n-vector space over
the n-field F = F
1
· … · F
n
. A bilinear n-form on V is a n-
function f = f
1
· … · f
n
which assigns to each set of ordered
pairs of n-vectors Į, ȕ in V a n-scalar f(Į, ȕ) = f
1

1
, ȕ
1
) · … ·
f
n

n
, ȕ
n
) in F = F
1
· … · F
n
and which satisfies:
f(cĮ
1
+ Į
2
, ȕ) = cf(Į
1
, ȕ) + f(Į
2
, ȕ)
f
1
(c
1
1 2
1 1 1
, ) o o þ + · … · f
n
(c
n
2
1
, )
n
n n
o o þ +
= c
1
f
1
1 2
1 1 2 1 1
( , ) ( , ) f o þ o þ + · … ·
1 2
( , ) ( , )
n n n n n n n
c f f o þ o þ + .

If we let V × V = (V
1
× V
1
) · … · (V
n
× V
n
) denote the set of
all n-ordered pairs of n-vectors in V
1
this definition can be
rephrased as follows. A bilinear n-form on V is a n-function f
188
from V × V into F = F
1
· … · F
n
i.e., (V
1
× V
1
) · … · (V
n
×
V
n
) into F
1
· … · F
n
i.e., f: V × V ÷ F is f
1
· … · f
n
: (V
1
×
V
1
) · … · (V
n
× V
n
) ÷ F
1
· … · F
n
with f
t
: V
t
× V
t
÷ F
t
for t
= 1, 2, …, n which is a n-linear function of either of its
arguments when the other is fixed. The n-zero function or zero
n-function from V × V into F is clearly a bilinear n-form. It is
also true that any bilinear combination of bilinear n-forms on V
is again a bilinear n-form.
Thus all bilinear n-forms on V is a n-subspace of all n-
functions from V × V into F. We shall denote the space of
bilinear n-forms on V by L
n
(V, V, F) = L(V
1
, V
1
, F
1
) · … ·
L(V
n
, V
n
, F
n
).

DEFINITION 2.9: Let V = V
1
· … · V
n
be a finite (n
1
, …, n
n
)
dimensional n-vector space and let B =
1
1 1
1
{ , , }
n
o o ! · … ·
1
{ , , }
n
n n
n
o o ! = (B
1
· … · B
n
) be a n-ordered n-basis of V. If f
= f
1
· … · f
n
is a bilinear n-form on V then the n-matrix of f in
the ordered n-basis B, is (n
1
× n
1
, …, n
n
× n
n
) n-matrix A = A
1

· … · A
n
with entries ( , )
t t t t
t t t
i j t i j
A f o o = . At times we shall
denote the n-matrix by [f]
B
=
1
1
[ ]
B
f · … · [ ]
n
n B
f .

THEOREM 2.16: Let V = V
1
· … · V
n
be a (n
1
, …, n
n
)
dimensional n-vector space over the n-field F = F
1
· … · F
n
.
For each ordered n-basis B = B
1
· … · B
n
of V, the n-function
which associates with each bilinear n-form on V its n-matrix in
the ordered n-basis B = B
1
· … · B
n
is an n-isomorphism of
the n-space L
n
(V, V, F) = L(V
1
, V
1
, F
1
) · … · L(V
n
, V
n
, F) onto
the n-space of (n
1
× n
1
, …, n
n
× n
n
) n-matrices over the n-field F
= F
1
· … · F
n
.

Proof: We see f ÷ [f]
B
i.e., f
1
· … · f
n
÷
1
1 B
[f ] · … ·
n
n B
[f ]
where f
t
÷
t
t B
[f ] for every t = 1, 2, …, n which is a one to one
n-correspondence between the set of bilinear n-forms on V and
the set of all (n
1
× n
1
, …, n
n
× n
n
) matrices over the n-field F.
That this is a n-linear transformation is easy to see because
(cf + g) (Į
i
, Į
j
) = cf(Į
i
, Į
j
) + cg (Į
i
, Į
j
)
189
i.e., (c
1
f
1
+ g
1
)
1 1
1 1
i j
( , ) o o · … · (c
n
f
n
+ g
n
)
n n
n n
i j
( , ) o o
= [c
1
f
1
1 1
1 1
i j
( , ) o o + g
1
1 1
1 1
i j
( , ) o o ] · … ·
[c
n
f
n
n n
n n
i j
( , ) o o + g
n
n n
n n
i j
( , ) o o ].

This simply says [cf + g]
B
= c[f]
B
+ [g]
B
i.e.,

1
1
1 1 B
[c f g ] + · … ·
n
n
n n B
[c f g ] +
=
1 1
1
1 B 1 B
(c [f ] [g ] ) + · … ·
n n
n
n B n B
(c [f ] [g ] ) + .

We leave the following corollary for the reader to prove.

COROLLARY 2.9: If B =
1
1 1
1
{ , , }
n
o o ! · … ·
1
{ , , }
n
n n
n
o o ! is an
n-ordered n-basis for V = V
1
· … · V
n
and
B
*
=
1
1 1
1
{ , , }
n
L L ! · … ·
1
{ , , }
n
n n
n
L L !
is the dual n-basis for V
*
=
*
1
V · … ·
*
n
V , then
2 2
1
( , , )
n
n n !
bilinear n-forms f
ij
(Į, ȕ) = L
i
(Į) L
j
(ȕ); i.e.,
1 1
1
1 1
( , ) ( , )
n n
n
i j i j n n
f f o þ o þ · · ! =
1 1
1 1
1
( ) ( ) ( ) ( )
n n
n n
i i j i n j n
L L L L o þ o þ · · ! ;
1 s i
t
, j
t
s n
t
; t = 1, 2, …, n, forms a n-basis for the n-space L
n
(V,
V, F) = L(V
1
, V
1
, F
1
) · … · L(V
n
, V
n
, F
n
) is
2 2
1 2 2
( , , , )
n
n n n ! .

THEOREM 2.17: Let f = f
1
· … · f
n
be a bilinear n-form on the
finite dimensional n-vector space V of n-dimension (n
1
, n
2
, …,
n
n
). Let R
f
and L
f
be a n-linear transformation from V into V
*

defined by (L
f
Į) ȕ = f(Į, ȕ) = (R
f
ȕ)Į i.e.,

1
1
1 1
( ) ( )
n
n
f f n n
L L o þ o þ · · ! = f
1

1

1
) · … · f
n

n
, ȕ
n
)
=
1
1
1 1
( ) ( )
n
n
f f n n
R R þ o þ o · · ! .
Then n-rank L
f
= n-rank R
f
.

The reader is expected to give the proof of the above theorem.

190
DEFINITION 2.10: If f = f
1
· … · f
n
is a bilinear n-form on the
finite dimensional n-vectors space V and n-rank of f is the n-
tuple of integers (r
1
, …, r
n
); (r
1
, …, r
n
) = n-rank L
f
= (rank
1
1
f
L ,
…, rank
n
n
f
L ) and n-rank R
f
= (rank
1
1
f
R , …, rank
n
n
f
R ).

We state the following corollaries and the reader is expected to
give the proof.

COROLLARY 2.10: The n-rank of a bilinear n-form is equal to
the n-rank of the n-matrix of the n-form in any ordered n-basis.

COROLLARY 2.11: If f = f
1
· … · f
n
is a bilinear n-form on the
(n
1
, …, n
n
) dimensional n-vector space V = V
1
· … · V
n
the
following are equivalent

a. n-rank f = (n
1
, …, n
n
) = (rank f
1
, …, rank f
n
).
b. For each non zero Į = Į
1
· … · Į
n
in V there is a ȕ =
ȕ
1
· … · ȕ
n
in V such that f(Į, ȕ) = 0 · … · 0; f
1

1
,
ȕ
1
) · … · f
n

n
, ȕ
n
) = 0 · … · 0.
c. For each non zero ȕ = ȕ
1
· … · ȕ
n
in V there is an Į =
Į
1
· … · Į
n
in V such that f(Į,ȕ) = 0 · … · 0 i.e.,
f
1

1
, ȕ
1
) · … · f
n

n
, ȕ
n
) = 0 · … · 0.

Now we proceed onto define the non degenerate of a bilinear n-
form.

DEFINITION 2.11: A bilinear n-form f = f
1
· … · f
n
on a n-
vector space V = V
1
· … · V
n
is called non-degenerate if it
satisfies the conditions (b) and (c) of the above corollary.

We now proceed onto define the new notion of symmetric
bilinear n-forms.

DEFINITION 2.12: Let f = f
1
· … · f
n
be a bilinear n-form on a
n-vector space V = V
1
· … · V
n
. We say that f is n-symmetric if
f(Į,ȕ) = f(ȕ,Į) for all Į, ȕ e V i.e., f
1

1
, ȕ
1
) · … · f
n

n
, ȕ
n
) =
f
1

1
, Į
1
) · … · f
n

n
, Į
n
).
191
If f = f
1
· … · f
n
is a n-symmetric bilinear n-form the
quadratic n-form associated with f is the function q = q
1
· … ·
q
n
from V = V
1
· … · V
n
onto F = F
1
· … · F
n
defined by
q(Į) = q
1

1
) · … · q
n

n
) = f(Į, Į) = f
1

1
, Į
1
) · … · f
n

n
,
Į
n
). If V is a real n-vector space an n-inner product on V is a n-
symmetric bilinear form f on V which satisfies f(Į, Į) > 0 · …
· 0 i.e., f
1

1
, Į
1
) · … · f
n

n
, Į
n
) > (0 · … · 0) where each
Į
i
= 0 for i = 1, 2, …, t. A bilinear n-form in which f
i

i
, Į
i
) > 0
for each i = 1, 2, …, n is called n-positive definite.
So two n-vectors Į, ȕ in V are n-orthogonal with respect to
a n-inner product f = f
1
· f
2
· … · f
n
if f(Į, ȕ) = 0 · … · 0
i.e., f(Į, ȕ) = f
1

1
, ȕ
1
) · … · f
n

n
, ȕ
n
) = 0 · … · 0.
The quadratic n-form q(Į) = f(Į, Į) takes only non negative
values.

The following theorem is significant on its own.

THEOREM 2.18: Let V = V
1
· … · V
n
be a n-vector space over
the n-field F = F
1
· … · F
n
each F
i
of characteristic zero, i =
1, 2, …, n and let f = f
1
· … · f
n
be a n-symmetric bilinear n-
form on V. Then there is an ordered n-basis for V in which f is
represented by a diagonal n-matrix.

Proof: What we need to find is an ordered n-basis B = B
1
· …
· B
n
=
1
1 1
1 n
{ } o o ! · … ·
n
n n
1 n
{ } o o ! such that
f(Į
i
, Į
j
) = f
1
1 1
1 1
i j
( , ) o o · … · f
n
n n
n n
i j
( , ) o o
= 0 · … · 0
for i
t
= j
t
; t = 1, 2, …, n. If f = f
1
· … · f
n
= 0 · … · 0 or n
1
=
n
2
= … = n
n
= 0 the theorem is obviously true thus we suppose f
= f
1
· … · f
n
= 0 · … · 0 and (n
1
, …, n
n
) > (1, …, 1). If f(Į,
Į) = 0 · … · 0 i.e., f
1

1
, Į
1
) · … · f
n

n
, Į
n
) = 0 · … · 0
for every Į
i
e V
i
; i = 1, 2, …, n the associated n-quadratic form
q is identically 0 · … · 0 and by the polarization n-identity
f(Į, ȕ) = f
1

1
, ȕ
1
) · … · f
n

n
, ȕ
n
)
=
1 1 1 1 1 1
1 1
q ( ) q ( )
4 4

o +þ ÷ o ÷þ

· … ·
192

n n n n n n
1 1
q ( ) q ( )
4 4

o +þ ÷ o ÷þ

.

f = f
1
· … · f
n

= 0 · … · 0.
Thus there is n-vector Į = Į
1
· … · Į
n
in V such that f(Į, Į) =
q(Į) = 0 · … · 0 i.e.,
f
1

1

1
) · … · f
n

n
, Į
n
) = q
1

1
) · … · q
n

n
)
= 0 · … · 0.
Let W be the one-dimensional n-subspace of V which is
spanned by Į = Į
1
· … · Į
n
and let W
±
be the set of all n-
vectors ȕ = ȕ
1
· … · ȕ
n
in V such that
f(Į, ȕ) = f
1

1
, ȕ
1
) · … · f
n

n
, ȕ
n
)
= 0 · … · 0.
Now we claim W © W
±
= V i.e.,
1 1 n n
W W W W
± ±
V = V
1
· … · V
n
.

Certainly the n-subspaces W and W
±
are independent. A
typical n-vector in V is cĮ where c is a n-scalar. If cĮ is in W
±

i.e.,
cĮ = (c
1
Į
1
· … · c
n
Į
n
) e W
±
=
1 n
W W
± ±
· · !
then
f
1
(c
1
Į
1
, c
1
Į
1
) · … · f
n
(c
n
Į
n
, c
n
Į
n
)
=
2
1
c f
1

1
, Į
1
) · … ·
2
n
c f
n

n
, Į
n
)
= 0 · … · 0.

But f
i

i
, Į
i
) = 0 for every i, i = 1, 2, …, n thus each c
i
= 0.
Also each n-vector in V is the sum of a n-vector in W and a n-
vector in W
±
. For let ¸ = ¸
1
· … · ¸
n
be any n-vector in V and
put ȕ = ȕ
1
· … · ȕ
n

=
f ( , )
f ( , )
¸ o
¸ ÷ o
o o

=
1 1 1 n n n
1 1 n n
1 1 1 n n n
f ( , ) f ( , )
f ( , ) f ( , )
¸ o ¸ o
¸ ÷ o · · ¸ ÷ o
o o o o
! .
Then
193
f(Į, ȕ) = f(Į, ¸) –
f ( , )
f ( , )
f ( , )
¸ o
o o
o o
;

and since f is n-symmetric, f(Į, ȕ) = 0 · … · 0. Thus ȕ is in
the n-subspace W
±
. The expression

f ( , )
f ( , )
¸ o
¸ = o +þ
o o

shows that V = W + W
±
. Then n-restriction of f to W
±
is a n-
symmetric bilinear n-form on W
±
=
1 n
W W
± ±
· · ! . Since W
±

has (n
1
– 1, …, n
n
– 1) dimension we may assume by induction
that W
±
has a n-basis
1 n
1 1 n n
2 n 2 n
{ , , } { , , } o o · · o o ! ! ! such that
t t
t t
t i j
f ( , ) 0 o o = ; i
t
= j
t
, i
t
> 2, j
t
> 2 and t = 1, 2, …, n. Putting Į
1

= Į we obtain a n-basis
1 n
1 1 n n
1 n 1 n
{ , , } { , , } o o · · o o ! ! ! for V
such that f(Į
i
, Į
j
) = f
1
1 1 n n
1 1 n n
i j n i j
( , ) f ( , ) o o · · o o ! = 0 · … · 0
for i
t
= j
t
; t = 1, 2, …, n.

THEOREM 2.19: Let V = V
1
· … · V
n
be a (n
1
, …, n
n
)
dimensional n-vector space over the n-field of real numbers and
let f = f
1
· … · f
n
be the n-symmetric bilinear n-form on V
which has n-rank (r
1
, …, r
n
). Then there is an n-ordered n-basis
1
1 1
1
{ }
n
þ þ ! ,
2
2 2
1
{ }
n
þ þ ! · … ·
1
{ }
n
n n
n
þ þ ! for V = V
1
· … · V
n
in which the n-matrix of f = f
1
· … · f
n
is n-diagonal and such
that
f(ȕ
j
ȕ
j
) = f
1
1 1
( , ) ( , )
n n
j j n j j
f þ þ þ þ · · ! ;
= ±1 · … · ±1;

1s j
t
s r
t
and t = 1, 2, …, n. Further more the number of n-basis
vectors
t
j
þ for which ( , ) 1
t t
t j j
f þ þ = is independent of the
choice of the n-basis for t = 1, 2, …, n.

The proof of the theorem is lengthy and left for the reader to
prove.
194
Now for complex vector spaces we in general find it
difficult to define n-vector spaces of type II as it happens to be
the algebraically closed field. Further when n happens to be
arbitrarily very large the problem of defining n-vector spaces of
type II is very difficult.
First we shall call the field
1 n
F F F ' ' ' · · = ! to be a
special algebraically closed n-field of the n-field F = F
1
· … ·
F
n
if and only if each
i
F' happens to be an algebraically closed
field of F
i
for a specific and fixed characteristic polynomial p
i
of
V
i
over F
i
relative to a fixed transformation T
i
of T, this is true
of every i.
195
Chapter Three
SUGGESTED PROBLEMS
In this chapter we suggest nearly 120 problems about n-vector
spaces of type II which will be useful for the reader to
understand this concept.
1. Find a 4-basis of the 4-vector space V = V
1
· V
2
· V
3
· V
4
of (3, 4, 2, 5) dimension over the 4-field Q
( )
2
· Q
( )
3
· Q
( )
5
· Q
( )
7
. Find a 4-subspace of V.
2. Given a 5-vector space V = V
1
· V
2
· V
3
· V
4
· V
5
over Q
( )
2
· Q
( ) 3, 5
· Q
( )
7
· Q
( )
11
·
Q
( )
13
of (5, 3, 2, 7, 4) dimension.
a. Find a linearly independent 5-subset of V which is
not a 5-basis.
b. Find a linearly dependent 5-subset of V.
c. Find a 5-basis of V.
d. Does there exists a 5-subspace of (4, 2, 1, 6, 3)
dimension in V?
e. Find a 5-subspace of (4, 3, 1, 5, 2) dimension of V.
196
3. Let V = V
1
· V
2
· V
3
· V
4
be a 4-vector space over the
4-field (Q
( )
2
· Q
( )
7
· Z
11
· Z
2
) of dimension (3,
4, 5, 6).
a. Find a 4-subset of V which is a dependent 4-subset
of V.
b. Give an illustration of a 4-subset of V which is a
independent 4-subset of V but is not a 4-basis of V.
c. Give a 4-subset of V which is a 4-basis of V.
d. Give a 4-subset of V which is semi n-dependent in
V.
e. Find a 4-subspace of V of dimension (2, 3, 4, 5)
over the 4-field F where F = (Q
( )
2
· Q
( )
7
· Z
11
· Z
2
) is a field.
4. Given V = V
1
· V
2
· V
3
is a 3-vector space of type II
over the 3-field F = Q · Z
2
· Z
5
of dimension (3, 4, 5).
Suppose
A = A
1
· A
2
· A
3
=
3 1 0 0 1
2 0 3 0
0 1 6 0 1 0 1 2
1 2 0 1
3 1 0 1 0 2 1 0
0 0 1 3
1 0 1 0 1 0 3 1
1 1 0 0
1 0 1 0 0

÷ ÷

· · ÷

÷ ÷

÷

is the 3-matrix find the 3-linear operator related with A.
a. Is A
3
diagonalizable?
b. Does A give rise to a 3-invertible 3-transformation?
c. Find 3-nullspace associated with A.
5. Let V = V
1
· V
2
· V
3
· V
4
be a 4-vector space over the
4-field F= (Q
( )
3
· Q
( )
2
· Z
5
· Z
2
) of dimension
(3, 2, 5, 4). Let W = W
1
· W
2
· W
3
· W
4
be a 4-vector
space over the same F of dimension (4, 3, 2, 6).
197
Find a 4-linear transformation T from V into W, which
will give way to a nontrivial 4-null subspace of V.
6. Let V = V
1
· V
2
· V
3
· V
4
· V
5
be a 5-vector space
over the 5-field F = Q( ) 3 · Q( ) 2 · Q( ) 5 · Z
5
·
Z
2
of type II of dimension (3, 4, 2, 5, 6).
a. Find a 5-linear operator T on V such that T is 5-
invertible.
b. Find a 5-linear operator T on V such that T is non
5-invertible.
c. Prove 5-rank T + 5 nullity T = (3, 4, 2, 5, 6) for any
T.
d. Find a T which is onto on V and find the 5-range of
T.
7. Let V = V
1
· V
2
· V
3
· V
4
be a (5, 4, 3, 2)-4 vector
space over the 4-field Z
2
· Z
7
· Q
( )
3
· Q
( )
2
. Find
V*. Obtain for any nontrivial 4-basis B its B* explicitly.
8. Given V* is a (3, 4, 5, 7) dimensional dual space over the
4-field F = Z
2
· Z
7
· Z
3
· Z
5
. Find V. What is V** ?
9. Let V = V
1
· V
2
· V
3
· V
4
be a 4-vector space of (3, 4,
5, 6) dimension over the 4-field F = Z
5
· Z
2
· Z
3
· Z
7
.
Suppose W is a 4-subspace of (2, 3, 4, 5) dimension over
the 4-field F. Prove dim W + dim W
o
= dim V. Find
explicitly W
o
. What is W
oo
? Is W
o
a 4-subspace of V or
Give a 4-basis for W and a 4-basis for W
o
. Are these
two sets of 4-basis related to each other in any other way?
10. Let V = V
1
· V
2
· V
3
be a 3-vector space over the 3-field
F = Z
7
· Z
5
· Z
2
of (5, 4, 6) dimension over F. Let S =
{(2 0 2 1 2), (1 0 1 1 1)} · {(0 1 2 3), (4 2 0 4)} · {(1 1
1 0 0 0), (1 1 1 1 1 0), (0 1 1 1 0 0)} = S
1
· S
2
· S
3
be a
3-set of V. Find S
o
. Is S
o
a 3-subspace? Find the basis for
S
o
.
198
11. Given V = V
1
· V
2
· V
3
· V
4
· V
5
is a 5-space of (5, 4,
2, 3, 7) dimension over the 5-field F = Q
( )
3
· Z
2
· Z
5
· Q
( )
2
· Z
7
. Find a 5-hypersubspace of V. Find V*?
Prove W is a 5-hypersubspace of V. What is W
o
? Find a
5-basis for V and its dual 5-basis. Find a 5-basis for W.
What is its dual 5-basis?
12. Let V = V
1
· V
2
· V
3
· V
4
be a 4-space over the 4-field
F = F
1
· F
2
· F
3
· F
4
of dimension (7, 6, 5, 4) over F.
Let W
1
and W
2
be any two 4 subspaces of (4, 2, 3, 1) and
(3, 5, 4, 2) dimensions respectively of V. Find
o
1
W and
o
2
W . Is
o
1
W =
o
2
W ? Find a 4-basis of W
1
and W
2
and their
dual 4-basis. Is W
1
a 4-hypersubspace of V? Justify your
claim! Find a 4-hyper subspace of V.
13. Let V = V
1
· V
2
· V
3
· V
4
· V
5
be a (2, 3, 4, 5, 6)
dimensional 5-vector space over the 5-field F = Z
2
· Z
3
· Z
5
· Z
7
· Q. Find V*. Find a 5-basis and its dual 5-
basis. Define a 5-isomorphism from V into V**. If W =
W
1
· W
2
· … ·W
5
is a (1, 2, 3, 4, 5), 5-subspace of V
find the n-annihilator space of V. Is W a 5-hyper space of
V? Can V have any other 5-hyper space other than W?
14. Let V = V
1
· V
2
· V
3
· V
4
be a 4-space of (4, 3, 7, 2) –
dimension over the 4-field F = Z
3
· Z
5
· Z
7
· Q. Find a
4-transformation T on V such that rank (T
t
) = rank T.
(Assume T is a 4-linear transformation which is not a 4-
isimorphism on V). Find 4-null space of T.
15. Give an example of 5-linear algebra over a 5-field which
is not commutative 5-linear algebra over the 5-field.
16. Give an example of a 6-linear algebra over a 6-field
which has no 6-identity.
199
17. Given A = A
1
· A
2
· A
3
where A
1
is a set of all 3 × 3
matrices with entries from Q
( )
3
. A
1
is the linear
algebra over Q
( )
3
. A
2
= {All polynomials in the
variable x with coefficient from Z
7
}; A
2
is a linear algebra
over Z
7
and A
3
= {set of all 5 × 5 matrices with entries
from Z
2
}. A
3
a linear algebra over Z
2
. Is A a 3-linear
algebra over the 3-field F = Q
( )
3
× Z
7
× Z
2
? Is A a 3-
commutative, 3-linear algebra over F? Does A contain the
3-identity?
18. Define a n-sublinear algebra of a n-linear algebra A over a
n-field F.
19. Give an example of a 4-sublinear algebra of the 4-linear
algebra over the 4-field.
20. Give an example of an n-vector space of type II which is
not an n-linear algebra of type II.
21. Give an example of an n-commutative n-linear algebra
over an n-field F (take n = 7).
22. Give an example of a 5-linear algebra which is not 5-
commutative over the 5-field F.
23. Let A = A
1
· A
2
· A
3
where A
1
is a set of all 3 × 3
matrices over Q, A
2
= all polynomials in the variable x
with coefficients from Z
2
and
A
3
=
n
2i
2i 3
i 0
x Z
=
¦ ¹
o oe
´ `
¹ )
¯
.
Prove A is a 3-linear algebra over the 3-field F = Q · Z
2
· Z
3
. Does A contain 3-identity find a 3-sublinear algebra
of A over F. Is A a commutative 3-linear algebra?
24. Obtain Vandermode 4-matrix with (7 + 1, 6 + 1, 5 + 1, 3
+ 1) from the 4-field F = F
1
· F
2
· F
3
· F
4
= Q · Z
3
· Z
2
· Z
5
.
200
25. Using the 5-field F = Z
2
· Z
7
· Z
3
· Z
5
· Z
11
, construct
the 5-vector space = V = Z
2
[x] · Z
7
[x] · Z
3
[x] · Z
5
[x]
· Z
11
[x]. Verify
(i) (cf + g)o = cf (x) + g (o)
(ii) (fg)o = f(o) g(o)
For C = C
1
· C
2
· C
3
· C
4
· C
5
= 1 · 5 · 2 · 3 · 10;
f = (x
2
+ x + 1) · (3x
3
+ 5x + 1) · (2x
4
+ x +1) · (4x +
2) · (10x
2
+ 5) where x
2
+ x + 1 e Z
2
[x], 3x
3
+ 5x + 1 e
Z
7
[x], 2x
4
+ x + 1 e Z
3
[x], 4x + 2 e Z
5
[x] and 10x
2
+ 5 e
Z
11
[x] and g(x) = x
3
+ 1 · x
2
+ 1 · 3x
3
+ x
2
+ x + 1 · 4x
5
+ x + 1 · 7x
2
+ 4x + 5 where x
3
+ 1 e Z
2
[x], x
2
+ 1 e
Z
7
[x], 3x
3
+ x
2
+ x + 1 e Z
3
[x], 4x
5
+ x + 1 e Z
5
[x] and
7x
2
+ 4x + 5 e Z
11
[x].
26. Let V = Z
2
[x] · Z
7
[x] · Q[x] be a 3-linear algebra over
the 3-field F = Z
2
· Z
7
· Q. Let M = M
1
· M
2
· M
2
be
the 3-ideal generated by {(x
2
+ 1, x
5
+ 2x + 1,} · {(3x
2
+ x + 1, x + 5, 7x
3
+ 1,} · {(x
2
+ 1, 7x
3
+ 5x
2
+ x + 3,}.
Is M a 3-principal ideal of V.
27. Prove in the 5-linear algebra of 5-polynomials A = Z
3
[x]
· Z
2
[x] · Q[x] · Z
7
[x] · Z
17
[x] over the 5-field Z
3
· Z
2
· Q · Z
7
· Z
17
every 5-polynomial p = p
1
· p
2
· …· p
5
can be made monic. Find a nontrivial 5-ideal of A.
28. Let A = A
1
· A
2
· … · A
6
= Z
3
[x] · Z
7
[x] · Z
2
[x] ·
Q[x] · Z
11
[x] · Z
13
[x] be a 6-linear algebra over the 6-
field F = Z
3
· Z
7
· Z
2
· Q · Z
11
· Z
13
. Find a 6-minimal
ideal of A. Give an example of a 6-maximal ideal of A.
Hint: We say in any n-polynomial n-linear algebra A = A
1
· A
2
· … · A
n
over the n-field F = F
1
· ... · F
n
where
A
i
= F
i
[x]; i = 1, 2, …, n. An n-ideal M = M
1
· … · M
n
is said to be a n-maximal ideal of A if and only if each
ideal M
i
of A
i
is maximal in A
i
for i = 1, 2, …, n. We say
the n-ideal N = N
1
· … · N
n
of A is n-minimal ideal of
A if and only if each ideal N
i
of A
i
is minimal in A
i
, for i
201
= 1, 2, …, n. An ideal M = M
1
· … · M
n
is said to be n-
semi maximal if and only if there exists atleast m number
of ideals in M, m s n which are maximal and none of
them are minimal. Similarly we say M = M
1
· … · M
n
is
n-semi minimal if and only if M contains atleast some p-
ideals which are minimal, p s n and none of the ideals in
M are maximal.
29. Give an example of an n-maximal ideal.
30. Give an example of an n-semi maximal ideal which is not
n-maximal.
31. Give an example of an n-semi minimal ideal which is not
an n-minimal ideal.
32. Let A = Z
3
[x] · Z
2
[x] · Z
7
[x] · Z
5
[x] be a 4-linear
algebra over the 4-field F = Z
3
· Z
2
· Z
7
· Z
5
. Give an
example of a 4-maximal ideal of A. Can A have a 4-
minimal ideal? Justify your claim. Does A have a 4-semi
maximal ideal? Can A have a 4-semi minimal ideal?
33. Give an example of a n-linear algebra which has both n-
semi maximal and n-semi minimal ideals.
34. Give an example of a n-linear algebra which has n-ideal
which is not a n-principal ideal? Is this possible if A = A
1
· … · A
n
where each A
i
is F
i
[x] where A is defined over
the n-field F = F
1
· … · F
n
; for i = 1, 2, …, n?
35. Let A = Z
3
[x] · Q[x] · Z
2
[x] be a 3-linear algebra over
the 3-field Z
3
· Q · Z
2
. Find the 3-gcd of {x
2
+ 1, x
2
+ 2x
+ x
3
+ 1} {x
2
+ 2, x + 2, x
2
+ 8x + 16} {x + 1, x
3
+ x
2
+ 1}
= P = P
1
· P
2
· P
3
. Find the 3-ideal generated by P. Is P a
3-principle ideal of A? Justify your claim. Can P generate
an n-maximal or n-minimal ideal? Substantiate your
202
36. Find the 4-characteristic polynomial and 4-minimal
polynomial for any 4-linear operator T = T
1
· T
2
· T
3
·
T
4
defined on the 4-vector space V = V
1
· V
2
· V
3
· V
4
where V is a (3, 4, 2, 5) dimensional over the 4-field F =
Z
5
· Q · Z
7
· Z
2
.
a. Find a T on V so that the 4-characteristic polynomial
is the same as 4-minmal polynomial.
b. Define a T on the V so that T is not a 4-
diagonializable operator on V.
c. For a T in which the 4-chracteristic polynomial is
different from 4-minimal polynomial find the 4-ideal
of polynomials over the 4-field F which 4-annihilate
T.
d. For every T can T be 4-diagonalizable. Justify your
claim.
37. Let
A =
1 0 0 0 1
4 0 1 1
0 1 0 0 1 0 1 0
0 1 1 3
1 0 1 1 0 1 0 1
1 0 0 0
0 0 2 0 0 1 1 0
2 1 0 0
1 1 0 0 1

· · ·

0 1 0 2 3 5
7 0 0 1 0 0
1 0 1 0 2 0 0 1
2 6 0 0 1 1 0 0
0 1 0 0 1 0
1 0 0 1 1 1

÷

÷

·

÷ ÷

be a 5-matrix over the five field; F = Z
3
· Z
5
· Z
2
· Z
7
· Q. Find the 5-characteristic value of A. Find the 5-
characteristic vectors of A. Find the 5-characteristic
polynomial and 5-minimal polynomial of A. Find the 5-
203
characteristic space of all o such that To = co, T related
to A.
38. Let A = A
1
· … · A
5
be a (5 × 5, 3 × 3, 2 × 2, 4 × 4, 6 ×
6), 5-matrix over the 5-field F = Z
2
· Z
3
· Z
5
· Z
7
· Q.
If A
2
= A i.e., A
i
2
= A
i
for i = 1, 2, …, 5. Prove A is 5-
similar to a 5-diagonal matrix. What can you say about
the 5-characteristic vectors and 5-characteristic values
associated with A.
39. Let V = V
1
· … · V
6
be a 6-vector space over a 6-field,
Z
2
· Q · Z
7
· Z
5
· Z
3
· Z
11
. Let T be a 6-diagonalizable
operator on V. Let W = W
1
· … · W
6
be a 6-invariant
subspace of V under T. Prove the 6-restriction operator
T
W
is 6-diagonalizable.
40. Let V = V
1
· V
2
· V
3
be a 3-vector space over the 3-
field, F = Q · Z
2
· Z
5
of (5, 3, 4) dimension over F. Let
W = W
1
· W
2
· W
3
be a 3-invariant 3-subspace of T
(Hint: Find T and find its 3-subspace W which is invariant
under T). Prove that the 3-minimal 3-polynomial for the
3-restriction 3-operator T
W
divides the 3-minimal
polynomial for T. Do this without referring to 3-matrices.
41. Let V = V
1
· … · V
n
be a (3 × 3, 2 × 2, 4 × 4, 5 × 5), 4-
space of matrices over the 4-field F = Z
2
· Z
3
· Z
5
· Q.
Let T and U be 4-linear operators on V defined by
T (B) = AB
i.e., T(B
1
· B
2
· B
3
· B
4
) = A
1
B
1
· … · A
4
B
4
where A = (A
1
· A
2
· A
3
· A
4
) is a fixed chosen (3 × 3,
2 × 2, 4 × 4, 5 × 5) matrix over
F = Z
2
· Z
3
· Z
5
· Q.
U(B) = AB – BA
= (A
1
B
1
· A
2
B
2
· A
3
B
3
· A
4
B
4
)
– (B
1
A
1
· B
2
A
2
· B
3
A
3
· B
4
A
4
)
= (A
1
B
1
– B
1
A
1
) · (A
2
B
2
– B
2
A
2
) ·
(A
3
B
3
– B
3
A
3
) · (A
4
B
4
– B
4
A
4
).
204
a. If A = (A
1
· A
2
· A
3
· A
4
), choose fixed 4-matrix
4-diagonalizable; then T is 4-diagonalizable. Is this
statement true or false?
b. If A is 4-diagonalizable then U is 4-diagonalizable.
Is this statement true or false?
42. Suppose A = A
1
· A
2
· A
3
· A
4
· A
5
is a 5-triangular n-
matrix similar to a 5-diagonal matrix then is A a 5-
43. Let V = V
1
· V
2
· V
3
· V
4
be a (3, 4, 5, 6) dimensional
vector space over the four field F = F
1
· F
2
· F
3
· F
4
.
Let T = T
1
· T
2
· T
3
· T
4
be a 4-linear operator on V.
Suppose there exists positive integers (k
1
, k
2
, k
3
, k
4
) such
that
1
k
1
T 0 = ,
2
k
2
T 0 = ,
3
k
3
T 0 = and
4
k
4
T 0 = . Will T
(3, 4, 5, 6)
=
3 4 5 6
1 2 3 4
T T T T · · · = 0 · 0 · 0 · 0 ?
44. Let V = V
1
· … · V
n
be a n-vector space over the n-field
F = F
1
· F
2
· … · F
n
. What is the n-minimal polynomial
for the n-identity operator on V? What is the n-minimal
polynomial for the n-zero operator?
45. Find a 3-matrix A = A
1
· A
2
· A
3
of order (3 × 3, 2 × 2,
4 × 4) such that the 3-minimal polynomial is x
2
· x
2
· x
3
over any suitable 3-field F.
46. Let A = A
1
· … · A
n
be a (n
1
× n
1
, n
2
× n
2
, …, n
n
× n
n
)
matrix over the n-field F = F
1
· … · F
n
with the n-
characteristic n-polynomial f = f
1
· … · f
n
=
( ) ( ) ( ) ( )
1 n
k k 1
1 n
1 n
d d d
1 1 n n
1 k 1 k
x c x c x c x c ÷ ÷ · · ÷ ÷ ! ! ! .
Show that trace A = trace A
1
· … · trace A
n
=
1 1 n n
1 1 1 1 n n n n
1 1 k k 1 1 k k
c d c d c d c d + + · · + + ! ! ! .
47. Let A = A
1
· … · A
n
be a n-matrix over the n-field F =
F
1
· … · F
n
of n-order (n
1
× n
1
, …, n
n
× n
n
) with the n-
characteristic polynomial
205
( ) ( ) ( ) ( )
1 n 1 n
k k 1 1
1 n
1 n
d d d d
1 1 n n
1 k 1 k
x c x c x c x c ÷ ÷ · · ÷ ÷ ! ! !
where
( ) ( )
1 n
1 1 n n
1 k 1 k
c , , c c , , c · · ! ! ! are n-distinct n-
characteristic values. Let V = V
1
· … · V
n
be a n-vector
space of (n
1
× n
1
, …, n
n
× n
n
) matrices B = B
1
· … · B
n
be such that AB = BA. Prove
n dimV =
( ) ( ) ( ) ( )
( )
1 n
2 2 2 2
1 1 n n
1 k 1 k
d d , , d d + + + + ! ! ! .
48. Let V = V
1
· … · V
n
be the n-space of (n
1
× n
1
, …, n
n
× n
n
), n-matrices over the n-field F = F
1
· … · F
n
. Let A
= A
1
· …· A
n
be a fixed n-matrix of n-order (n
1
× n
1
,
…, n
n
× n
n
). Let T be a n-linear operator ‘n-left
multiplication by A’ on V. Is it true that A and T have the
same n-characteristic values?
49. Let A = A
1
· … · A
n
and B = B
1
· … · B
n
be two n-
matrices of same n-order over the n-field F = F
1
· … ·
F
n
. Let n-order of A and B be (n
1
× n
1
, …, n
n
× n
n
). Prove
if (I – AB) is n-invertible then I-BA is n-invertible and
(I – BA)
-1
= I + B(I – AB)
-1
A. Using this result prove both
AB and BA have the same n-characteristic values in F =
F
1
· … · F
n
.
50. Let T is a n-linear operator of a (n
1
× n
1
, …, n
n
× n
n
)
dimensional n-vector space over the n-filed F = F
1
· … ·
F
n
and suppose T has (n
1
, …, n
n
) distinct n-characteristic
values. Prove T is n-diagonalizable.
51. Let A = A
1
· … · A
n
be a (n
1
× n
1
, …, n
n
× n
n
) triangular
n-matrix over the n-field F. Prove that the n-characteristic
values of A are the diagonal entries
( )
1 2 n
ii ii ii
A , A ,..., A .
206
52. Let A = A
1
· … · A
n
be a n-matrix which is n-diagonal
over the n-field F = F
1
· … · F
n
of (n
1
× n
1
, …, n
n
× n
n
)
order i.e., if A
k
=
( )
k
ij
A ;
k
ij
A = 0 if i = j for k = 1, 2, …, n.
Let f = f
1
· …· f
n
be the n-polynomial over the n-field F
defined by F =
( ) ( ) ( ) ( )
1 1 2 2
1 1 2 2
11 n n 11 n n
x A x A x A x A ÷ ÷ · ÷ ÷ · · ! ! !
( ) ( )
n n
n n
11 n n
x A x A ÷ ÷ ! .
What is the n-matrix of f (A) = f
1
(A
1
) · ... · f
n
(A
n
)?
53. Let F = F
1
· … · F
n
be a n-field F[x] = F
1
[x] · … ·
F
n
[x] be the n-polynomial in the variable x. Show that the
intersection of any number of n-ideals in F[x] is a
minimal n-ideal.
54. Let A = A
1
· … · A
n
be a n-matrix of (n
1
× n
1
, n
2
× n
2
,
…, n
n
× n
n
) order over a n-field F = F
1
· … · F
n
. Show
that the set of n-polynomials f = f
1
· … · f
n
in F[x] is
such that f(A) = f
1
(A
1
) · … · f
n
(A
n
) = 0 · 0 · … · 0.
55. Let F = F
1
· … · F
n
be a n-field. Show that the n-ideal
generated by a finite number of n-polynomial f
1
, …, f
n
.
where f
i
=
i i
1 n
f ... f · · ; i = 1, 2, …, n in F[x] = F
1
[x] · …
· F
n
[x] is the intersection of all n-ideals in F[x] is an n-
ideal.
56. Let (n
1
, …, n
n
) be a n-set of positive integers and F = F
1
·
… · F
n
be a n-field, let W be the set of all n-vectors
( ) ( ) ( )
1 2 n
1 1 2 2 n n
1 n 1 n 1 n
x x x x x x · · · ! ! ! !
in
1 2 n
n n n
1 2 n
F F F · · · ! such that
( )
1
1 1
1 n
x x 0 + + = ! ,
( )
2
2 2
1 n
x x 0 + + = ! , … ,
( )
n
n n
1 n
x x 0 + + = ! .
a. Prove W
o
=
o o o
1 2 n
W W ... W · · · consists of all n-
linear functionals f = f
1
· … · f
n
of the form
207
( ) ( )
1 n
1 n
n n
1 1 n n 1 n
1 1 n n 1 n i j n j
j 1 j 1
f x x f x ...x c x c x .
= =
· · = · ·
¯ ¯
! ! !
b. Show that the n-dual space W* of W can be naturally
identified with n-linear functionals.
( ) ( )
1 n
1 1 n n
1 1 n n 1 n
f x x f x x · · ! ! !
=
1 1 n n
1 1 1 1 n n n n
1 1 n n 1 1 n n
c x ... c x ... c x ... c x + + · · + +
on
1 2 n
n n n
1 2 n
F F ... F · · · which satisfy
i
i i
1 n
c ... c + + = 0
for i = 1, 2, …, n.
57. Let W = W
1
· … · W
n
be a n-subspace of a finite (n
1
,
…, n
n
) dimensional n-vector space over V = V
1
· … ·
V
n
and if
{ } { } { }
1 2 n
1 1 2 2 n n
1 r 1 r 1 r
g g g g g g · · · ! ! ! !
is a basis for W
o
=
o o
1 n
W ... W · · then
W =
1 n
i i i
1 n
1 n
r r
1 n
g g g
i i 1 i 1
N N N
= =
= · · !
where
{ } { }
1 n
1 1 n n
1 r 1 r
N N N N · · ! ! ! is the n-set of n-null
space of the n-linear functionals
f = f
1
· … · f
n
=
{ } { }
1 n
1 1 n n
1 r 1 r
f f f f · · ! ! !
and

{ } { }
1 n
1 1 n n
1 r 1 r
g g g g · · ! ! !
is the n-linear combination of the n-linear functions f = f
1
· … · f
n
.
58. Let S = S
1
· … · S
n
be a n-set, F = F
1
· … · F
n
a n-
field. Let V(S; F) = V
1
(S
1
; F
1
) · V
2
(S
2
; F
2
) · … · V
n
(S
n
;
F
n
) the n-space of all n-functions from S into F;
i.e., S
1
· … · S
n
into = F
1
· … · F
n
. (f + g) x = f (x) +
g(x) and (cf)(x) = cf(x).
If f = f
1
· … · f
n
and g = g
1
· … · · g
n
; then (f + g)(x)
= f
1
(x) + g
1
(x) · … · f
n
(x) + g
n
(x),
208
cf(x) = c
1
f
1
(x) · … · c
n
f
n
(x) where c = c
1
· … · c
n
e F
= F
1
· … · F
n
. Let W = W
1
· … · W
n
be any (n
1
, …,
n
n
) dimensional space of V(S; F) = V
1
(S
1
; F
1
) · … ·
V
n
(S
n
, F
n
).
Prove that there exists (n
1
, …, n
n
) points
( ) ( ) ( )
1 2 n
1 1 2 2 n n
1 n 1 n 1 n
x x x x x x · · · ! ! ! !
in S = S
1
· … · S
n
and (n
1
, …, n
n
) functions
( ) ( ) ( )
1 2 n
1 1 2 2 n n
1 n 1 n 1 n
f f f f f f · · · ! ! ! !
in W such that
( )
t t t
i j ij
f x = o ; i = 1, 2, …, n.
59. Let V = V
1
· … · V
n
be a n-vector space over the n-field
F = F
1
· … · F
n
and let T = T
1
· … · T
n
be a n-linear
operator on V. Let C = C
1
· … · C
n
be a n-scalar
suppose there is a non zero n-vector o = o
1
· … · o
n
in
V such that To = Co i.e., T
1
o
1
= C
1
o
1
, …, T
n
o
n
= C
n
o
n
.
Prove that there is a non-zero n-linear functional f = f
1
·
… · f
n
on V such that T
t
f = Cf i.e.,
t t
1 1 n n
T f ... T f · · = C
1
f
1
· … · c
n
f
n
.
60. Let A = A
1
· … · A
n
be a (m
1
× m
1
, …, m
n
× m
n
) n-
mixed rectangular matrix over the n-field F = F
1
· … ·
F
n
with real entries. Prove that A = 0 · …· 0 (i.e., each
A
i
= (0) for i = 1, 2, …, n) if and only if the trace A
t
A =
(0) i.e., A
1
t
A
1
· … · A
n
t
A
n
= 0 · … · 0.
61. Let (n
1
, …, n
n
) be n-tuple of positive integers and let V be
the n-space of all n-polynomials functions over the n-field
of reals which have n-degrees atmost (n
1
, n
2
, …, n
n
) i.e.,
n-functions of the form
f(x) = f
1
(x) · …· f
n
(x) =
1 n
1 n
n n 1 1 1 n n n
0 1 n 0 1 n
c c x c x c c x c x + + + · · + + + ! ! ! .
Let D = D
1
· … · D
n
be the n-differential operator on V
= V
1
· … · V
n
over the n-field F = F
1
· … · F
n
. Find
an n-basis for the n-null space of the n-transpose operator
D
t
.
209
62. Let V = V
1
· … · V
n
be a n-vector space over the n-field
F = F
1
· … · F
n
. Show that for a n-linear transformation
T = T
1
· … · T
n
on T ÷ T
t
is an n-isomorphism of
L
n
(V, V) onto L
n
(V*, V*);
i.e., T = T
1
· … · T
n
÷ T
t
=
t t
1 n
T ... T · · is an n-
isomorphism of L
n
(V, V) = L(V
1
, V
1
) · … · L(V
n
, V
n
)
onto L
n
(V
*
, V
*
) =
( ) ( )
* * * *
1 1 n n
L V , V L V , V · · ! .
63. Let V = V
1
· … · V
n
be a n-vector space over the n-field
F = F
1
· … · F
n
, where V is a n-space of (n
1
× n
1
, n
2
× n
2
, …, n
n
× n
n
); n-matrices with entries from the n-field
F = F
1
· … · F
n
.
a. If B = B
1
· … · B
n
is a fixed (n
1
× n
1
, n
2
× n
2
, …, n
n
× n
n
) n-matrix define a n-function f
B
= f
1
B
1
· … · f
n
B
n
on V by f
B
(A) = trace(B
t
A);
f
B
(A) =
1 n
B 1 B n
f (A ) f (A ) · · !
= trace
( ) ( )
t t t
1 1 n n
B A trace B A · · !
where A = A
1
· … · A
n
e V = V
1
· … · V
n
. Show
f
B
is a n-linear functional on V.
b. Show that every n-linear functional on V of the above
form i.e., is f
B
for some B.
c. Show that B ÷
1 n
1B nB
f f · · ! i.e., B
1
· … · B
n
÷
1 n
1B nB
f f · · ! i.e., B
i
÷
i
iB
f ; i = 1, 2, …, n is an n-
isomorphism of V onto V*.
64. Let F = F
1
· … · F
n
be a n-field. We have considered
certain special n-linear functionals F[x] = F
1
[x] · … ·
F
n
[x] obtained via “evaluation at t” t = t
1
· … · t
n
given
by L(f) = f(t).
L
1
(f
1
) · … · L
n
(f
n
) = f
1
(t
1
) · … · f
n
(t
n
).
Such n-functional are not only n-linear but also have the
property that
L(fg) = L
1
(f
1
g
1
) · … · L
n
(f
n
g
n
) =
210
L
1
(f
1
) L
1
(g
1
) · … · L
n
(f
n
) L
n
(g
n
).
Prove that if L = L
1
· … · L
n
is any n-linear functional
on F[x] = F
1
[x] · … · F
n
[x] such that L(fg) = L(f)L(g)
for all f, g then L = 0 · … · (0) or there is a t = t
1
· …
· t
n
such that L(f) = f(t) for all f. i.e., L
1
(f
1
) · … · L
n
(f
n
)
= f
1
(t) · … · f
n
(t).
65. Let K = K
1
· … · K
n
be the n-subfield of a n-field F = F
1
· … · F
n
. Suppose f = f
1
· … · f
n
and g = g
1
· … ·
g
n
be n-polynomials in K[x] = K
1
[x] · … · K
n
[x]. Let
M
K
=
1 n
K K
M M · · ! be the n-ideal generated by f and g
in K[x]. Let M
F
=
1 n
F F
M M · · ! be the n-ideal, n-
generated in F[x] = F
1
[x] · … · F
n
[x]. Show that M
k
and
M
F
have the same n-monic generator. Suppose f, g are n-
polynomial in F[x] = F
1
[x] · … · F
n
[x], if M
F
is the n-
ideal of F[x] i.e., M
F
=
1 n
F F
M M · · ! . Find conditions
under which the n-ideal M
k
of K[x] = K
1
[x] · … · K
n
[x]
can be formed.
66. Let F = F
1
· … · F
n
be a n-field. F[x] = F
1
[x] · … ·
F
n
[x]. Show that the intersection of any number of n-
ideals in F[x] is again a n-ideal.
67. Prove the following generalization of the Taylors formula
for n-polynomials. Let f, g, h be n-polynomials over the
n-subfield of complex numbers with n-deg f = (deg f
1
, …,
deg f
n
) s (n
1
, …, n
n
), where f = f
1
· … · f
n
.
Then f(g) = f
1
(g
1
) · … · f
n
(g
n
)
( ) ( )
1 n
1 n
1 n
1 n
n n
k k
(k ) (k )
1 1 1 1 n n n n
k 0 k 0
1 n
1 1
f (h ) g h f (h ) g h
k k
= =
= ÷ · · ÷
¯ ¯
!
68. Let T = T
1
· … · T
n
be a n-linear operator on (n
1
, …, n
n
)
dimensional space and suppose that T has (n
1
, …, n
n
)
distinct n-characteristic values. Prove that any n-linear
operator which commutes with T is a n-polynomial in T.
211
69. Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
) dimensional n-
vector space. Let W
1
=
( )
1 1
1 n
W W · · ! be any n-
subspace of V. Prove that there exists a n-subspace W
2
=
( )
1 n
2 2
W W · · ! of V such that V = W
1
2
i.e.,
V = W
1
2
=
( ) ( ) ( )
1 1 2 2 n n
1 2 1 2 1 2
70. Let V = V
1
· … · V
n
be a (n
1
, …, n
n
) finite dimensional
n-vector space and let W
1
, …, W
n
be n-subspaces of V
such that V = W
1
+ … + W
n
i.e., V = V
1
· …· V
n
=
( ) ( ) ( )
1 2 n
1 1 2 2 n n
1 k 1 k 1 k
W W W W W W + + · + + · · + + ! ! ! !
and n dim V = (n dim W
1
+ … + n dim W
n
)
= (dim
1
1
W + … + dim
1
1
k
W , dim
2
1
W + … + dim
2
2
k
W , …,
dim
n
1
W + … + dim
n
n
k
W ). Prove that V = W
1
+ … + W
n
=
( ) ( ) ( )
1 2 n
1 1 2 2 n n
1 k 1 k 1 k
71. If E
1
and E
2
are n-projections E
1
=
( )
1 n
1 1
E E · · ! and E
2
=
( )
1 n
2 2
E E · · ! on independent n-subspaces then is E
1
+ E
2
72. If E
1
, …, E
n
are n-projectors of a n-vector space V such
that E
1
+ … + E
n
= I.
i.e.,
( ) ( )
1 n
k k 1 1
1 1 n n
E E E E + + · · + + ! ! ! = I
1
· … · I
n
.
Then prove
i j
t t
E .E 0 = if i = j for t = 1, 2, …, n then
2
t t
p p
t t
E E = for every t = 1, 2, …, n where p
t
= 1, 2, …, k
t
.
73. Let V = V
1
· … · V
n
be a real n-vector space and E = E
1
· … · E
n
be an n-idempotent n-linear operator on V i.e.,
n-projection. Prove that 1 + E = 1 + E
1
· … · 1 + E
n
is
n-invertible find (1 + E)
-1
= (1 + E
1
)
-1
· … · (1 + E
n
)
-1
.
212
74. Obtain some interesting properties about n-vector spaces
of type II which are not true in case of n-vector spaces of
type I.
75. Let T be a n-linear operator on the n-vector space V of
type II which n-commutes with every n-projection
operator on V. What can you say about T?
76. If N = N
1
· … · N
n
is a n-nilpotent linear operator on a
(n
1
, …, n
n
) dimensional n-vector space V = V
1
· … · V
n
then the n-characteristic polynomial for N is (
1
n
x · … ·
1
n
x ).
77. Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
) dimensional n-
vector space over the n-field F = F
1
· … · F
n
and T = T
1
· … · T
n
be a n-linear operator on V such that n-rank T
= (rank T
1
, …, rank T
n
) = (1, 1, …, 1). Prove that either T
is n-diagonalizable or T is n-nilpotent, not both
simultaneously.
78. Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
) dimensional n-
vector space over the n-field F = F
1
· … · F
n
. Let T =
T
1
· … · T
n
be a n-operator on V. Suppose T commutes
with every n-diagonalizable operator on V, i.e., each T
i
commutes with every n
i
-diagonalizable operator on V
i
for
i = 1, 2, …, n then prove T is a n-scalar multiple of the n-
identity operator on V.
79. Let T be a n-linear operator on the (n
1
, n
2
, …, n
n
)
dimensional n-vector space V = V
1
· … · V
n
over the n-
field F = F
1
· … · F
n.
Let p = p
1
· … · p
n
be the n
minimal polynomial for T i.e.,
k k k
1 1 2 2 n n
1 2 n
r r r
r r r
1 11 1k 21 2k n1 nk
p p p p p p p = · · · ! ! ! !
be the n-minimal polynomial for T.
Let V =
( ) ( )
1 n
1 1 n n
1 k 1 k
1
· … · W
n
be the n-primary decomposition for T, i.e.,
213
t
j
W is the null space of
j
r
tj t
p (T ) , j = 1, 2, …, k
t
. and t = 1,
2, …, n. Let W
r
=
r r
1 n
W W · · ! be any n-subspace of V
which is n-invariant under T. Prove that
W
r
=
( ) ( ) ( )
r r r
1 1 2 2 n n
80. Define some new properties on the n-vector spaces of
type II relating to the n-linear operators.
81. Compare the n-linear operators on n-vector spaces V of
type II and n-vector space of type I. Is every n-linear
transformation of type I always be a n-linear operator of a
type I n-vector space?
82. State and prove Bessel’s inequality in case of n-vector
spaces.
83. Derive Gram-Schmidt orthogonalization process for n-
vector spaces of type II.
84. Define for a 3-vector space V = (V
1
· V
2
· V
3
) of (7, 2,
5) dimension over the 3-field F = Q · Z
3
· Z
2
two
distinct 3-inner products.
85. Let V = V
1
· V
2
· V
3
be a 3-spaces of (3, 4, 5)
dimension over the 3-field F = Q
( )
2
· Q
( )
3
·
Q
( )
7
. Find L
3
(V, V, F) = L(V
1
, V
1
, Q
( )
2
) · L(V
2
,
V
2
, Q
( )
3
) · L (V
3
, V
3
, Q
( )
7
).
86. Does their exists a skew-symmetric bilinear 5-forms on
5 1 2
n n n
R R R · · · ! ; n
i
= n
j
, 1 s i, j s 5? Justify your
claim.
87. Prove that 5-equation (Pf)(o. þ) = ½ [f(o. þ) – f(þ.o)]
defines a 5-linear operator P on L
5
(V,V,F) where V = V
1
· V
2
· … · V
5
is a 5-dimension vector space over a
214
special 5-subfield F = F
1
· F
2
· F
3
· F
4
· F
5
of the
complex field (such that F
i
= F
j
if i = j, 1 s i, j s 5.
a. V is of (7, 3, 4, 5, 6) dimension over F = F
1
· … ·
F
5
.
b. Prove P
2
= P is a 5-projection.
c. Find 5-rank P and 5-nullily P. Is 5 rank P = (21, 3, 6,
10, 15) and 5-nullity P = (28, 6, 10, 15, 21).
88. Let V = V
1
· … · V
4
be a finite (4, 5, 3, 6) dimension
vector space over the 4-field Q
( )
2
· Q
( )
3
· Q
( )
5
· Q( ) 7 and f a symmetric bilinear 4-form on V. For
each 4-subspace W of V. Let W
±
be the set of all 4-vector
o = o
1
· o
2
· o
3
· o
4
in V such that f(o. þ) = 0 · 0 · 0
· 0 for þ in W. Show that
a. W
±
is a 4-subspace.
b. V = {0}
±
· {0}
±
· {0}
±
· {0}
±
.
c. V
±
= {0} · {0} · {0} · {0} if and only if f is a non
degenerate! Can this occur?
d. 4-rank f = 4-dim V – 4-dim V
±
.
e. If 4-dim V = (n
1
, n
2
, n
3
, n
4
) and 4-dim W = (m
1
, m
2
,
m
3
, m
4
) then 4-dim W
±
> (n
1
– m
1
, n
2
– m
2
, n
3
– m
3
, n
4
–m
4
).
f. Can the 4-restriction of f to W be a non-degenerate if
W · W
±
= {0} · {0} · {0} · {0}?
89. Prove if U and T any two normal n operators which
commute on a n-vector space over a n-field of type II
prove U + T and UT are also normal n-operators (n > 2).
90. Define positive n-operator for a n-vector space of type II.
Prove if S and T are positive n-operators every n-
characteristic value of ST is positive.
91. Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
) inner n-product
space over a n-field F = F
1
· … · F
n
. If T and U are
positive linear n-operators on V prove that (T + U) is
positive. Show by an example TU need not be positive.
215
92. Prove that every positive n-matrix is the square of a
positive n-matrix.
93. Prove that a normal and nilpotent n-operator is the zero n-
operator.
94. If T = T
1
· … · T
n
is a normal n-operator prove that the
n-characteristic n-vectors for T which are associated with
distinct n-characteristic values are n-orthogonal.
95. Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
) dimensional n-
inner product space over a n-field F = F
1
· … · F
n
for
each n-vector o, þ in V let T
o.þ
=
1 1 n n
1 n
, ,
T T
o þ o þ
· · !
(where T = T
1
· …· T
n
, o = o
1
· … · o
n
and þ = þ
1
·
… · þ
n
) be a linear n-operator on V defined by T
o.þ
(¸) =
(¸ / þ)o i.e., T
o.þ
(¸) =
1 1 n n
1 n
, 1 , n
T ( ) T ( )
o þ o þ
¸ · · ¸ ! = (¸
1
/ þ
1
)o
1
· … · (¸
n
/ þ
n
)o
n
. Show that
a.
*
, ,
T T
o þ þ o
= .
b. Trace (T
o.þ
) = (o/þ).
c. (T
o.þ
) (T
¸.o
) = T
o. (þ / ¸)o
.
d. Under what conditions is T
o.þ
96. Let V = V
1
· … · V
n
be a (n
1
, n
2
, …, n
n
) dimensional n-
inner product space over the n-field F = F
1
· … · F
n
and
let L
n
(V, V) = L(V
1
, V
1
) · … · L(V
n
, V
n
) be the n-space
of linear n-operators on V. Show that there is a unique n-
inner product on L
n
(V, V) with the property that ||T
o.þ
||
2
=
||o||
2
||þ||
2
for all o, þ e V i.e.,
1 1 n n
2 2
1 n
T ... T
o þ o þ
· · =
2 2 2 2
1 1 n n
... o þ · · o þ .
T
o.þ
is an n-linear operator defined in the above problem.
Find an n-isomorphism between L
n
(V,V) with this n-inner
space of (n
1
× n
1
, …, n
n
× n
n
), n-matrix over the n-field F
= F
1
· … · F
n
with the n-inner product (A/B) = tr(AB*)
i.e., (A
1
/ B
1
) · … · (A
n
/ B
n
) = tr(A
1
B
1
*) · … ·
216
tr(A
n
B
n
*) where A = A
1
· ...· A
n
and B = B
1
· ...· B
n
are (n
1
× n
1
, …, n
n
× n
n
) n-matrix.
97. Let V = V
1
· … · V
n
be a n-inner product space and let
E = E
1
· … · E
n
be an idempotent linear n-operator on
V. i.e., E
2
= E prove E is n-self adjoint if and only if EE*
= E*E i.e., E
1
E
1
*
· … · E
n
E
n
*
= E
1
*
E
1
· … · E
n
*
E
n
.
98. Show that the product of two self n-adjoint operators is
self n-adjoint if and only if the two operators commute.
99. Let V = V
1
· … · V
n
be a finite (n
1
, n
2
, …, n
n
)
dimensional n-inner product vector space of a n-field F.
Let T = T
1
· … · T
n
be a n-linear operator on V. Show
that the n-range of T* is the n-orthogonal complement of
the n-nullspace of T.
100. Let V be a finite (n
1
, n
2
, …, n
n
) dimensional inner product
space over the n-field F and T a n-linear operator on V. If
T is n-invertible show that T* is n-invertible and (T*)
-1
=
(T
-1
)*.
101. Let V = V
1
· … · V
n
be a n-inner product space over the
n-field F = F
1
· … · F
n
. þ and ¸ be fixed n-vectors in V.
Show that To = (o/þ)¸ defines a n-linear operator on V.
Show that T has an n-adjoint and describe T* explicitly.
102. Let V = V
1
· … · V
n
be an n-inner product space of n-
polynomials of degree less than or equal to (n
1
, …, n
n
)
over the n-field F = F
1
· … · F
n
i.e., each V
i
= F
i
[x] for i
= 1, 2, …, n. Let D be the differentiation on V. Find D*.
103. Let V = V
1
· … · V
n
be a n-real vector space over the
real n-field F = F
1
· … · F
n
. Show that the quadratic n-
form determined by the n-inner product satisfies the n-
parallelogram law.
||o + þ||
2
+ ||o – þ||
2
= 2||o||
2
+ 2||þ||
2
217
For o = o
1
· … · o
n
and þ = þ
1
· … · þ
n
in V.
i.e., ||o
1
+ þ
1
||
2
+ ||o
1
– þ
1
||
2
· … · ||o
n

n
||
2
+ ||o
n
– þ
n
||
2
= 2||o
1
||
2
+ 2||þ
1
||
2
· … · 2||o
n
||
2
+ 2||þ
n
||
2
.
104. Let V = V
1
· … · V
n
be a n-vector space over the n-field
F = F
1
· … · F
n
. Show that the sum of two n-inner
product on V is an n-inner product on V. Is the difference
of two n-inner products an n-inner product? Show that a
positive multiple of an n-inner product is an n-inner
product.
105. Derive n-polarization identity for a n-vector space V = V
1
· … · V
n
over the n-field F = F
1
· … · F
n
for the
standard n-inner product on V.
106. Let A = A
1
· … · A
n
be a (n
1
× n
1
, …, n
n
× n
n
) n-matrix
with entries from n-field F = F
1
· … · F
n
. Let
{ } { }
1 n
1 1 n n
1 n 1 n
f f f f · · ! ! ! be the n-diagonal entries of
the n-normal form of xI – A = xI
1
– A
1
· … · xI
n
– A
n
.
For which n-matrix A is
( )
1 n
1 1
f , , f ! = (1, 1, …1)?
107. Let T = T
1
· … · T
n
be a linear n-operator on a finite
(n
1
, …, n
n
) dimensional vector space over the n-field F =
F
1
· … · F
n
and A = A
1
· … · A
n
be a n-matrix
associated with T in some ordered n-basis. Then T has a
n-cyclic vector if and only if the n-determinants of
(n
1
– 1) × (n
1
– 1), …, (n
n
– 1) × (n
n
– 1) n-submatrices of
xI – A are relatively prime.
108. Derive some interesting properties about n-Jordan forms
or Jordan n-form (Just we call it as Jordan n-form or n-
Jordan forms and both mean one and the same notion).
109.
a. Let T = T
1
· … · T
n
be a n-linear operator on the n
space V of n-dim (n
1
, …, n
n
) . Let R = R
1
· … · R
n
be the n-range of T. Prove that R has a n-
218
complementary T-n-invariant n-subspace if and only
if R is n-independent of the n-null space N = N
1
· …
· N
n
of T.
b. Prove if R and N are n-independent N is the unique T-
n-variant n-subspace complementary to R.
110. Let T = T
1
· … · T
n
be a n-linear operator on the n-
space V = V
1
· … · V
n
. If f = f
1
· … · f
n
is a n-
polynomial over the n-field F = F
1
· … · F
n
and o
1
· …
· o
n
= o and let f(o) = f(T)o
i.e., f
1
(o
1
) · … · f
n
(o
n
)
= f
1
(T
1
) o
1
· ...· f
n
(T
n
) o
n.
If
{ } { }
1 n
1 1 n n
1 k 1 k
V V , , V V ! ! ! are T-n-invariant n- sub
space of V=
( ) ( )
1 n
1 1 n n
1 k 1 k
that fV =
( ) ( )
1 n
1 1 n n
1 1 1 k n 1 n k
111. Let T, V and F are as in the above problem (110).
Suppose o
1
· … · o
n
and þ = þ
1
· … · þ
n
are n-vectors
in V which have the same T n-annihilator. Prove that for
any n-polynomial f = f
1
· … · f
n
the n-vectors fo = f
1
o
1
· … · f
n
o
n
and fþ = f
1
þ
1
· … · f
n
þ
n
have the same n-
annihilator.
112. If T = T
1
· … · T
n
is a n- diagonalizable operator on a n-
vector space then every T n-invariant n-subspace has a n-
complementary T-n-invariant subspace.
113. Let T = T
1
· … · T
n
be a n-operator on a finite (n
1
, n
2
,
…, n
n
) dimensional n-vector space V = V
1
· … · V
n
over the n-field F = F
1
· … · F
n
. Prove that T has a n
cyclic vector if and only if every n-linear operator U = U
1
· … · U
n
which commutes with T is a n-polynomial in
T.
114. Let V = V
1
· … · V
n
be a finite (n
1
, n
2
, …, n
n
) n-vector
space over the n-field F = F
1
· … · F
n
and let T = T
1
·
219
… · T
n
be a n-linear operator on V. When is every non
zero n-vector in V a n-cyclic vector for T? Prove that this
is the case if and only if the n-characteristic n-polynomial
for T is n-irreducible over F.
115. Let T = T
1
· … · T
n
be a n-linear operator on the (n
1
, n
2
,
…, n
n
) dimensional n-vector space over the n-field of type
II. Prove that there exists a n-vector o = (o
1
· … · o
n
)
in V with this property. If f is a n-polynomial i.e., f = f
1
·
… · f
n
and f(T)o = 0 · … · 0 i.e., f
1
(T
1
)o
1
· … ·
f
n
(T
n
)o
n
= 0 · … · 0 then f(T) = f
1
(T
1
) · … · f
n
(T
n
) = 0
· … · 0. (such a n-vector is called a separating n-vector
for the algebra of n-polynomials in T). When T has a n-
cyclic vector give a direct proof that any n-cyclic n-vector
is a separating n-vector for the algebra of n-polynomials
in T.
116. Let T = T
1
· … · T
n
be a n-linear operator on the n-
vector space V = V
1
· … · V
n
of (n
1
, n
2
, …, n
n
)
dimension over the n-field F = F
1
· … · F
n
suppose that
a. The n-minimal polynomial for T is a power of an
irreducible n-polynomial.
b. The minimal n-polynomial is equal to the
characteristic n-polynomial. Then show that no
nontrivial T-n-invariant n-subspace has an n-
complementary T-n-invariant n-subspace.
117. Let A = A
1
· … · A
n
be a (n
1
× n
1
, …, n
n
× n
n
) n-matrix
with real entries such that A
2
+ I =
2 2
1 1 n n
A I A I + · · + "
= 0 · … · 0. Prove that (n
1
, …, n
2
) are even and if (n
1
,
…, n
n
) = (2k
1
, 2k
2
, …, 2k
n
) then A is n-similar over the n-
field of real numbers to a n-matrix of the n-block form B
= B
1
· … · B
n
where B =
n
1
n
k
k
k
k
0 I
0 I
I 0
I 0
÷
÷

· ·

"
where
t t
k k
I
×
is a
t t
k k × identity matrix for t = 1, 2, …, n.
220
118. Let A = A
1
· … · A
n
be a (m
1
× n
1
, …, m
n
× n
n
) n-
matrix over the n-field F = F
1
· … · F
n
and consider the
n-system of n-equation AX = Y i.e., A
1
X
1
· … · A
n
X
n
=
Y
1
· … · Y
n
. Prove that this n-system of equations has a
n-solution if and only if the row n-rank of A is equal to
the row n-rank of the augmented n-matrix of the n-
system.
119. If P
n
=
n
n n
1 n
P P · · ! is a stochastic n-matrix, is P a
stochastic n-matrix? Show if A = A
1
· … · A
n
is a
stochastic n-matrix then (1, …, 1) is an n-eigen value of
A.
120. Derive Chapman Kolmogorov equation for
1 n
1 1 n n
(n ) (n ) n
ij i j i j
p p p = · · !
=
1 n
1 1 1 1 n n n n
1 1 n n
(n 1) (n 1)
i j k j i j k j
k s k s
p p p p
÷ ÷
e e
· ·
¯ ¯
"
S = S
1
· … · S
n
is an associated n-set.
221
1. ABRAHAM, R., Linear and Multilinear Algebra, W. A.
Benjamin Inc., 1966.
2. ALBERT, A., Structure of Algebras, Colloq. Pub., 24, Amer.
Math. Soc., 1939.
3. BERLEKAMP, E.R., Algebraic Coding Theory, Mc Graw Hill
Inc, 1968.
4. BIRKHOFF, G., and MACLANE, S., A Survey of Modern
Algebra, Macmillan Publ. Company, 1977.
5. BIRKHOFF, G., On the structure of abstract algebras, Proc.
Cambridge Philos. Soc., 31 433-435, 1995.
6. BRUCE, SCHNEIER., Applied Cryptography, Second Edition,
John Wiley, 1996.
7. BURROW, M., Representation Theory of Finite Groups,
Dover Publications, 1993.
8. CHARLES W. CURTIS, Linear Algebra – An introductory
Approach, Springer, 1984.
9. DUBREIL, P., and DUBREIL-JACOTIN, M.L., Lectures on
Modern Algebra, Oliver and Boyd., Edinburgh, 1967.
10. GEL'FAND, I.M., Lectures on linear algebra, Interscience,
New York, 1961.
222
11. GREUB, W.H., Linear Algebra, Fourth Edition, Springer-
Verlag, 1974.
12. HALMOS, P.R., Finite dimensional vector spaces, D Van
Nostrand Co, Princeton, 1958.
13. HAMMING, R.W., Error Detecting and error correcting
codes, Bell Systems Technical Journal, 29, 147-160, 1950.
14. HARVEY E. ROSE, Linear Algebra, Bir Khauser Verlag,
2002.
15. HERSTEIN I.N., Abstract Algebra, John Wiley,1990.
16. HERSTEIN, I.N., and DAVID J. WINTER, Matrix Theory and
Linear Algebra, Maxwell Pub., 1989.
17. HERSTEIN, I.N., Topics in Algebra, John Wiley, 1975.
18. HOFFMAN, K. and KUNZE, R., Linear algebra, Prentice Hall
of India, 1991.
19. HUMMEL, J.A., Introduction to vector functions, Addison-
Wesley, 1967.
20. JACOB BILL, Linear Functions and Matrix Theory ,
Springer-Verlag, 1995.
21. JACOBSON, N., Lectures in Abstract Algebra, D Van
Nostrand Co, Princeton, 1953.
22. JACOBSON, N., Structure of Rings, Colloquium
Publications, 37, American Mathematical Society, 1956.
23. JOHNSON, T., New spectral theorem for vector spaces over
finite fields Z
p
, M.Sc. Dissertation, March 2003 (Guided by
Dr. W.B. Vasantha Kandasamy).
24. KATSUMI, N., Fundamentals of Linear Algebra, McGraw
Hill, New York, 1966.
223
25. KOSTRIKIN, A.I, and MANIN, Y. I., Linear Algebra and
Geometry, Gordon and Breach Science Publishers, 1989.
26. LANG, S., Algebra, Addison Wesley, 1967.
27. LAY, D. C., Linear Algebra and its Applications, Addison
Wesley, 2003.
28. MAC WILLIAM, F.J., and SLOANE N.J.A., The Theory of
Error Correcting Codes, North Holland Pub., 1977.
29. PETTOFREZZO, A. J., Elements of Linear Algebra, Prentice-
Hall, Englewood Cliffs, NJ, 1970.
30. PLESS, V.S., and HUFFMAN, W. C., Handbook of Coding
Theory, Elsevier Science B.V, 1998.
31. ROMAN, S., Advanced Linear Algebra, Springer-Verlag,
New York, 1992.
32. RORRES, C., and ANTON H., Applications of Linear
Algebra, John Wiley & Sons, 1977.
33. SEMMES, Stephen, Some topics pertaining to algebras of
linear operators, November 2002.
http://arxiv.org/pdf/math.CA/0211171
34. SHANNON, C.E., A Mathematical Theory of
Communication, Bell Systems Technical Journal, 27, 379-
423 and 623-656, 1948.
35. SHILOV, G.E., An Introduction to the Theory of Linear
Spaces, Prentice-Hall, Englewood Cliffs, NJ, 1961.
36. THRALL, R.M., and TORNKHEIM, L., Vector spaces and
matrices, Wiley, New York, 1957.
37. VAN LINT, J.H., Introduction to Coding Theory, Springer,
1999.
224
38. VASANTHA KANDASAMY and RAJKUMAR, R. Use of best
approximations in algebraic bicoding theory, Varahmihir
Journal of Mathematical Sciences, 6, 509-516, 2006.
39. VASANTHA KANDASAMY and THIRUVEGADAM, N.,
Application of pseudo best approximation to coding theory,
Ultra Sci., 17, 139-144, 2005.
40. VASANTHA KANDASAMY, W.B., Bialgebraic structures and
Smarandache bialgebraic structures, American Research
Press, Rehoboth, 2003.
41. VASANTHA KANDASAMY, W.B., Bivector spaces, U. Sci.
Phy. Sci., 11, 186-190 1999.
42. VASANTHA KANDASAMY, W.B., Linear Algebra and
Smarandache Linear Algebra, Bookman Publishing, 2003.
43. VASANTHA KANDASAMY, W.B., SMARANDACHE, Florentin
and K. ILANTHENRAL, Introduction to bimatrices, Hexis,
Phoenix, 2005.
44. VASANTHA KANDASAMY, W.B., SMARANDACHE, Florentin
and K. ILANTHENRAL, Introduction to Linear Bialgebra,
Hexis, Phoenix, 2005.
45. VASANTHA KANDASAMY, W.B., SMARANDACHE, Florentin
and K. ILANTHENRAL, Set Linear Algebra and Set fuzzy
Linear Algebra, Infolearnquest, Ann Arbor, 2008.
46. VASANTHA KANDASAMY, W.B., and SMARANDACHE,
Florentin, New Classes of Codes for Cryptologists and
computer Scientists, Infolearnquest, Ann Arbor, 2008.
47. VASANTHA KANDASAMY, W.B., and SMARANDACHE,
Florentin, n-Linear Algebra of Type I and its Applications,
Infolearnquest, Ann Arbor, 2008.
48. VOYEVODIN, V.V., Linear Algebra, Mir Publishers, 1983.
Press, 1973.
225
INDEX
B
Bilinear n-form, 187
C
Cayley Hamilton theorem for n-linear operator over n-vector
space of type II, 102
Cyclic n-decomposition, 136-7
F
Finite n- dimensional n-vector space, 13
G
Generalized Cayley Hamilton theorem for n-vector spaces of
type II, 152-3
I
Infinite n-dimensional n-vector space, 13
Invertible n-matrix, 47-8, 92
J
Jordan n-form, 159-160
226
M
m-subfield, 9-10
N
n-algebraically closed n-field relative to a n-polynomial, 94
n-algebraically closed n-field, 93-4
n-annihilator of a n-subset of a n-vector space of type II, 52-3
n-best approximation, 166-7
n-characteristic value of a n-linear operator of a n-vector space
of type II, 89-90
n-commutative n-linear algebra of type II, 66-7
n-companion n-matrix of the n-polynomial, 134-5
n-conjugate transpose of a n-matrix of T, 173
n-cyclic decomposition, 136-7
n-cyclic n-decomposition , 136-7
n-cyclic n-subspace, 130
n-cyclic n-vector for T, 152
n-diagonalizable n-linear operator, 93
n-diagonalizable normal n-operator, 182
n-divisible, 77-8
n-dual of a n-vector space of type II, 49, 57
n-equation, 182
n-field of characteristic zero, 8
n-field of finite characteristic, 8-9
n-field of mixed characteristic, 8-9
n-field, 7
n-hypersubspace of a n-vector space of type II, 52, 54
n-idempotent n-linear transformation, 168
n-independent n-subset, 11-2
n-inner product n-vector space of type II, 179
n-inner product space, 161-2
n-invertible n-linear transformation, 38-9
n-irreducible n-polynomial over a n-field, 86
n-Jordan form, 159-160
n-linear algebra of n-polynomial functions, 67-8
n-linear algebra of type II, 37
227
n-linear algebra with identity of type II, 66-7
n-linear functional on a n-vector space of type II, 48, 57
n-linear n-combination of n-vectors, 14-5
n-linear operator on a n-vector space of type II, 35-6
n-linear transformation of type II n-vector spaces, 24
n-linearly dependent n-subset in n-vector space of type II, 11-2
n-linearly dependent, 11-2
n-linearly independent n-subset of type II n-vector space, 11-2
n-mapping, 168
n-monic n-polynomial, 68
n-normality, 183
n-nullity of T, 28
non degenerate, 190
n-ordered pair, 161-2
n-orthogonal projection of a n-subspace, 167-8
n-orthogonal, 163, 178, 190-1
n-orthonormal, 163-4
n-positive definite, 190-1
n-range of T, 28
n-rank of T, 28
n-rational form, 136-7
n-reducible n-polynomial over a n-field, 86
n-root or n-zero of a n-polynomial, 68
n-semiprime field, 9-10
n-similar, 47-8
n-spectrum, 182
n-standard inner product, 161-2
n-subfield, 9
n-subspace spanned by n-subspace, 15
n-symmetric bilinear n-form, 190-1
n-T-annihilator of a n-vector, 131, 152
n-vector space of type I, 7
n-vector space of type II, 7-10
P
Prime n-field, 9
Principal n-ideal generated by a n-polynomial, 83
228
Q
Quasi m-subfield, 9-10
S
Semi n-linearly dependent n-subset, 11-2
Semi n-linearly independent n-subset, 11-2
Semi prime n-field, 9-10
Spectral n-resolution, 182
T
T-n-annihilator of a n-vector, 131, 139-140
T-n-conductor, 110
229
Dr.W.B.Vasantha Kandasamy is an Associate Professor in the
Department of Mathematics, Indian Institute of Technology
scholars in the different fields of non-associative algebras,
algebraic coding theory, transportation theory, fuzzy groups, and
applications of fuzzy theory of the problems faced in chemical
industries and cement industries.
She has to her credit 646 research papers. She has guided
over 68 M.Sc. and M.Tech. projects. She has worked in
collaboration projects with the Indian Space Research
Organization and with the Tamil Nadu State AIDS Control Society.
This is her 39
th
book.
On India's 60th Independence Day, Dr.Vasantha was
conferred the Kalpana Chawla Award for Courage and Daring
Enterprise by the State Government of Tamil Nadu in recognition
of her sustained fight for social justice in the Indian Institute of
Technology (IIT) Madras and for her contribution to mathematics.
(The award, instituted in the memory of Indian-American
astronaut Kalpana Chawla who died aboard Space Shuttle
Columbia). The award carried a cash prize of five lakh rupees (the
highest prize-money for any Indian award) and a gold medal.
She can be contacted at vasanthakandasamy@gmail.com
You can visit her on the web at: http://mat.iitm.ac.in/~wbv
Dr. Florentin Smarandache is a Professor of Mathematics and
Chair of Math & Sciences Department at the University of New
Mexico in USA. He published over 75 books and 150 articles and
notes in mathematics, physics, philosophy, psychology, rebus,
literature.
In mathematics his research is in number theory, non-
Euclidean geometry, synthetic geometry, algebraic structures,
statistics, neutrosophic logic and set (generalizations of fuzzy
logic and set respectively), neutrosophic probability
(generalization of classical and imprecise probability). Also, small
contributions to nuclear and particle physics, information fusion,
neutrosophy (a generalization of dialectics), law of sensations and
stimuli, etc. He can be contacted at smarand@unm.edu
n-Linear Algebra of Type II - Cover:Layout 1 1/12/2009 3:16 PM Page 1