Our definition of function
involved the undefined word ``rule". If I define
by
the rule is perfectly clear. I will often want to define functions by ``rules"
of the following sort:
is given by
(3.50)
It is not quite so clear that this is a rule, since the right side of (3.50)
involves the function I am trying to define. However, if I try to use this rule to
calculate , I get
(3.51)
and by this example, you recognize that (3.50) defines the familiar factorial
function. In fact, I make this my definition of the factorial function.
3.52Definition (Factorial function.)
We define
by the rules.
We call the factorial function, and denote by . By definition,
I could use the same rule (3.50) to define a factorial function
. The calculation (3.51) shows that then
and
but in
, so I have , contradicting . So I see that
(3.50) is not a ``rule". How do I know that (3.50) is a
``rule" when considered as a function from
?; i.e., how do I know
that no contradiction arises when I use (3.50) to calculate values for
? I have decided not to worry about this question, and to treat definitions
analogous to (3.50) where functions on
are defined by giving
explicitly, and expressing in terms of and for values of , as valid ``rules". Such defintions are called definitions by
recursion. A discussion of, and justification for definitions by recursion can be found in
[27].
3.53Definition (Powers.)
Let be a field, and let . Define a function
by
(3.54)
Thus,
We denote the value of by .
Then we can rewrite (3.54) as
Note that and .
3.55Theorem.
Let be a field and let . Then for all
,
Proof: Define a proposition form on
by
Then
which is true, since both sides of the equation are equal to .
For all
,
and
Hence for all
,
By induction, is true for all
, i.e.
3.56Exercise.
Let be a field, and let be elements of . Show that