When I state $(a\cdot b) \cdot c = (a \cdot c) \cdot b$ , am I using only the commutative property or both the commutative and associative property?
4 Answers
If you mean whether the statement you're posting relies upon commutativity and associativity, yes: You rely upon both. I am assuming $a, b, c \in \mathbb R$ and that by "$\cdot$" you mean "standard" multiplication:
$$\begin{align}(a\cdot b)\cdot c & = a\cdot (b\cdot c) \tag{by associativity}\\ \\ & = a\cdot (c \cdot b) \tag{by commutativity}\\ \\ & = (a \cdot c) \cdot b \tag{by associativity}\end{align}$$
- 209,954
-
@Peter Please see the OP's first question before you interpret my interpretation as wrong. – amWhy Jul 30 '13 at 22:57
-
Amy: I am sorry, but there was no way yo tell what the OP was asking. Maybe you can add a disclaimer before the answer "If you mean..." – Pedro Jul 30 '13 at 22:58
-
-
-
1The fact that in your "personal" proof you used both laws does not imply that both laws are really necessary. – Christian Blatter Aug 04 '13 at 18:40
-
The identity
$(a \cdot b) \cdot c = (a \cdot c) \cdot b$
can be proved using both commutativity and associativity.
If your structure has an identity, then the statement implies both commutativity and associativity.
For commutativity. Let $b$ and $c$ be any element. Let $a$ be identity for $\cdot$, then
$ b \cdot c = (a \cdot b) \cdot c = (a \cdot c) \cdot b = c \cdot b$
Now to show associativity:
$(a \cdot b) \cdot c$
By commutativity shown above
$ = (b \cdot a) \cdot c$
By the statement
$= (b \cdot c) \cdot a$
By commutativity
$= a \cdot (b \cdot c)$
Hence associativity follows.
In conclusion if there is an identity for $\cdot$, then commutativity and associativity follows.
In fact, if there is no identity, then you can not prove commutativity.
Consider the two element set $\{x,y\}$ with multiplication defined by $x \cdot x = x$, $y \cdot y = y$, $x \cdot y = x$ and $y \cdot x = y$. Note that the multiplication is not commutative and has no identity. However, the value of the product depends only on what the first element of the product is. Hence it easy to verify that the statement holds in this structure.
However associativity does hold in this structure. I feel that the statement should not be able to prove associativity. If I can come up with a counter example, I will post it later.
- 19,935
So far nobody has given counterexamples showing that associativity alone or commutativity alone are not sufficient to guarantee $(a\cdot b)\cdot c=(a\cdot c)\cdot b$.
Multiplication of $(n\times n)$-matrices is certainly associative, but not commutative when $n\geq2$. Choose any two non-commuting $(2\times2)$-matrices $b$ and $c$, and let $a$ be the $(2\times2)$ identity matrix. Then $(a\cdot b)\cdot c\ne(a\cdot c)\cdot b$.
For complex numbers $z$, $w$ define $$z\ *\ w:=\overline{z\cdot w}\ .$$ This product is certainly commutative, but $(1*1)*i\ne(1*i)*1$.
- 226,825
Another set of examples satisfying $(ab)c=(ac)b$ is given on any set $S$ by defining $xy=x$ for all $x,y \in S$. If $S$ has more than one element, this is not commutative (nor does it have a two sided identity element). It is however associative.
For a nonassociative example, take $x \cdot y=x-y$ on $\mathbb{R}$, and then the relation $(ab)c=(ac)b$ becomes $(a-b)-c=(a-c)-b.$
Just noticed: subtraction is not commutative either, so gives a single example where the relation holds, but it is neither commutative nor associative.
- 29,884
- 2
- 31
- 52
All in all, though, the OP is probably actually asking whether his equality requires only commutativity or both commutativity and associativity of (real) numbers, in which case, amWhy's answer is as good as it gets.
– Branimir Ćaćić Jul 30 '13 at 22:49amWhy's answer is sufficient. – Bentley4 Jul 30 '13 at 23:23