Difference between revisions of "Power set"

m (added all- now states: set of all subsets)
m (Size comparison)
Line 18: Line 18:
  
 
Now, note that <math>y \in T</math> by definition if and only if <math>y \not\in f(y)</math>, so <math>y \in T</math> if and only if <math>y \not \in T</math>.  This is a clear contradiction.  Thus the bijection <math>f</math> cannot really exist and <math>|\mathcal P (S)| \neq |S|</math> so <math>|\mathcal P(S)| > |S|</math>, as desired.
 
Now, note that <math>y \in T</math> by definition if and only if <math>y \not\in f(y)</math>, so <math>y \in T</math> if and only if <math>y \not \in T</math>.  This is a clear contradiction.  Thus the bijection <math>f</math> cannot really exist and <math>|\mathcal P (S)| \neq |S|</math> so <math>|\mathcal P(S)| > |S|</math>, as desired.
 +
 +
 +
Note that this proof does not rely upon either the [[continuum hypothesis]] or the [[axiom of choice]].  It is a good example of a [[diagonal argument]], a method pioneered by the mathematician [[Georg Cantor]].
  
 
==See Also==
 
==See Also==

Revision as of 10:55, 7 September 2006

The power set of a given set $S$ is the set $\mathcal{P}(S)$ of all subsets of that set.

Examples

The empty set has only one subset, itself. Thus $\mathcal{P}(\emptyset) = \{\emptyset\}$.

A set $\{a\}$ with a single element has two subsets, the empty set and the entire set. Thus $\mathcal{P}(\{a\}) = \{\emptyset, \{a\}\}$.

A set $\{a, b\}$ with two elements has four subsets, and $\mathcal{P}(\{a, b\}) = \{\emptyset, \{a\}, \{b\}, \{a, b\}\}$.

Similarly, for any finite set with $n$ elements, the power set has $2^n$ elements.

Size comparison

Note that for any nonnegative integer $n$, $2^n > n$ and so for any finite set $S$, $|\mathcal P (S)| > |S|$ (where absolute value signs here denote the cardinality of a set). The analogous result is also true for infinite sets (and thus for all sets): for any set $S$, the cardinality $|\mathcal P (S)|$ of the power set is strictly larger than the cardinality $|S|$ of the set itself.

Proof

There is a natural injection $S \hookrightarrow \mathcal P (S)$ taking $x \mapsto \{x\}$, so $|S| \leq |\mathcal P(S)|$. Suppose for the sake of contradiction that $|S| = |\mathcal P(S)|$. Then there is a bijection $f: \mathcal P(S) \to S$. Let $T \subset S$ be defined by $T = \{x \in S \;|\; x \not\in f(x) \}$. Then $T \in \mathcal P(S)$ and since $f$ is a bijection, $\exists y\in S \;|\; T = f(y)$.

Now, note that $y \in T$ by definition if and only if $y \not\in f(y)$, so $y \in T$ if and only if $y \not \in T$. This is a clear contradiction. Thus the bijection $f$ cannot really exist and $|\mathcal P (S)| \neq |S|$ so $|\mathcal P(S)| > |S|$, as desired.


Note that this proof does not rely upon either the continuum hypothesis or the axiom of choice. It is a good example of a diagonal argument, a method pioneered by the mathematician Georg Cantor.

See Also

This article is a stub. Help us out by expanding it.