Let $f(x_1,\ldots,x_k)$ be a boolean function. There is a
defining formula for $f$ that uses only the $\wedge$, $\vee$ and
$\neg$ operators.
We prove this by induction on $n = k-1$ (i.e. one less than the number of
inputs to $f$. If $n = 0$, $f$ is one of the following four
functions (these are all possible truth tables with one input):
| 1. |
|
which is equivalent to $\neg(x1 \vee \neg
x1)$ |
2. |
|
which is equivalent to $\neg x1$ |
| 3. |
|
which is equivalent to $x1$ |
4. |
|
which is equivalent to $x1\vee\neg x1$ |
Suppose $n > 0$. Then consider the following two functions:
$$
f_0(x_1,\ldots,x_{k-1}) = f(x_1,\ldots,x_{k-1},0) \text{, and }
f_1(x_1,\ldots,x_{k-1}) = f(x_1,\ldots,x_{k-1},1)
$$
Then $f(x_1,\ldots,x_k) = 1$ if any only if
$f_0(x_1,\ldots,x_{k-1}) = 1 \wedge \neg x_k
\vee
f_1(x_1,\ldots,x_{k-1}) = 1 \wedge x_k$.
By the inductive hypothesis, there are formulas $G_0$
defining $f_0$ and $G_1$ defining $f_1$, both of which use
only $\wedge$, $\vee$ and $\neg$. Therefore,
$$
G_0 \wedge \neg x_k \vee G_1 \wedge x_k
$$
is a defining formula for $F$.
So we deduce that the theorem holds for any value of $n$.