Identity Elements: The Neutral Value in Operations
In the context of a set (or Type) A
and a binary operation •
defined on that set (as discussed for semigroups), an identity element (often called a “unit element”) is a special value within A
that, when combined with any other value from A
using •
, leaves that other value unchanged.
Defining the Identity Element
Section titled “Defining the Identity Element”A value e
(which must be an element of set A
, so e
has type 'a
if A
corresponds to type 'a
) is called an identity element for the operation •
on set A
if, for every element x
in A
:
- Left identity property:
e • x = x
- Right identity property:
x • e = x
This must hold true for all possible values of x
within the set A
.
Identity Elements in Familiar Operations
Section titled “Identity Elements in Familiar Operations”Let’s look at some familiar binary operations and their identity elements:
Addition and Multiplication on Integers
Section titled “Addition and Multiplication on Integers”For the Type int
(the set of integers):
-
With the binary operation
(+)
(addition):- The identity element is
0
(which is of typeint
). - Property: For any integer
x
,0 + x = x
andx + 0 = x
.- Example:
0 + 3 = 3
and3 + 0 = 3
.
- Example:
- The identity element is
-
With the binary operation
(*)
(multiplication):- The identity element is
1
(which is of typeint
). - Property: For any integer
x
,1 * x = x
andx * 1 = x
.- Example:
1 * 5 = 5
and5 * 1 = 5
.
- Example:
- The identity element is
String Concatenation
Section titled “String Concatenation”For the Type string
(the set of all strings):
- With the binary operation
(+)
(string concatenation):- The identity element is
""
(the empty string, which is of typestring
). - Property: For any string
s
,"" + s = s
ands + "" = s
.- Example:
"" + "hello" = "hello"
and"hello" + "" = "hello"
.
- Example:
- The identity element is
Physical Examples: The Challenge of Finding a Natural Identity
Section titled “Physical Examples: The Challenge of Finding a Natural Identity”In our physical analogies, a true identity element in the mathematical sense is often hard to find or doesn’t naturally exist:
-
LEGO blocks (Set: LEGO blocks, Operation: joining two blocks):
- Is there a LEGO block
e
such that joininge
with any blockX
results inX
unchanged (i.e.,e
joined withX
is indistinguishable fromX
alone, andX
joined withe
is also indistinguishable fromX
alone)? - No, any physical LEGO piece, when connected, physically alters the resulting structure or occupies space. There isn’t a “do-nothing” LEGO block that leaves the other block as if no connection was made.
- Is there a LEGO block
-
USB devices (Set: USB devices, Operation: connecting two devices):
- Similarly, there’s no physical USB device
e
that, when connected to another deviceX
(or a hubX
), results in a configuration that is functionally and structurally identical toX
alone. Any connection typically introduces at least a new point in the device tree.
- Similarly, there’s no physical USB device
Uniqueness and Two-Sidedness of Identity Elements
An important property of an identity element, if one exists for a given binary operation •
on a set A
, is that it must be unique and it must be both a left identity and a right identity.
Let’s prove this:
Assume, for a set A
with a binary operation •
, that eL
is a left identity (so eL • x = x
for all x
in A
) and eR
is a right identity (so x • eR = x
for all x
in A
). Both eL
and eR
are elements of A
.
- Since
eL
is a left identity, it holds forx = eR
. So,eL • eR = eR
. - Since
eR
is a right identity, it holds forx = eL
. So,eL • eR = eL
.
From steps 1 and 2, we have eR = eL • eR
and eL = eL • eR
.
Therefore, eL = eR
.
This shows that if a left identity and a right identity both exist, they must be the same element. This also implies that if an identity element exists, it is unique. (If e1
and e2
were both two-sided identities, then e1 = e1 • e2
(since e2
is a right identity) and e1 • e2 = e2
(since e1
is a left identity), thus e1 = e2
.)
So, for a given binary operation on a set:
- You can’t have just a left identity that isn’t also a right identity (if a right identity also exists, and vice-versa).
- If an identity element exists, it’s the only one.
Let’s confirm with our examples:
- For (
int
,+
):0
is the unique two-sided identity. - For (
int
,*
):1
is the unique two-sided identity. - For (
string
,+
):""
is the unique two-sided identity.
This uniqueness and two-sided nature is a fundamental characteristic of identity elements in algebraic structures.