natural number

1. What is a Natural Number?

1. What is a Natural Number?

We can formulate the natural numbers from set construction, or by Peano arithmetic. I will start with the Peano arithmetic formulation. First, we define an immediate successor function \(S:\mathbb{N}\rightarrow\mathbb{N}\) which effectively "adds one" (although we haven't defined addition yet), and a number we call \(0 \in \mathbb{N}\). We then define an axiom:

\begin{align*} \forall n \in \mathbb{N}, \nexists S(n), s.t. S(n) = 0; \\ \forall n \in \mathbb{N}, S(n) \in \mathbb{N}. \end{align*}

which is equivalent to saying: adding one to any natural number makes it not equal to zero, and any natural number's successor is a natural number. Because zero is a natural number, we can define \(1 = S(0)\), and by definition \(1 \in \mathbb{N}\). Note that it doesn't matter what we call \(S(0)\); we just choose to name it one because we like working in the base 10 number system.

In a few lines, we should also try to define equality:

\begin{align*} \forall a \in \mathbb{N}, \; a = a; \\ \forall a, b, c \in \mathbb{N}, \; (a = b) \land (b = c) \rightarrow a = c; \\ \forall a, b \in \mathbb{N}, \; a = b \rightarrow b = a. \end{align*}

which I already explained just sets up equality in the way we're used to. These axioms are probably slightly important for our purposes, and as you can imagine, they generalize past natural numbers. Then we define one more axiom:

\begin{align*} \forall a, b \in \mathbb{N}, \; S(a) = S(b) \Leftrightarrow a = b. \end{align*}

simply saying that if we add one to both sides of an equation the equality remains. And we're almost done! There is one problem: given our current axioms, we can definitely prove propositions like these:

\begin{align*} S(S(0)) \neq S(0) \end{align*}

however, they don't allow for the ability for us to extrapolate properties of natural numbers in general.

1.1.

Let's introduce our last axiom:

\begin{align*} \forall n \in \mathbb{N}, \forall P(x), P(0) \land (P(x) \rightarrow P(S(x))) \rightarrow P(n) \end{align*}

now, this is the principle of specific to natural numbers. What it is saying is that a property \(P(n)\) is true for all \(n\) if there is a "base case" \(P(0)\) which is true, and you can show that \(P(1)\) is true from \(P(0)\), \(P(2)\) is true from \(P(1)\) and so on to infinity, or more generally \(P(S(n))\) is true for every \(P(n)\). This "base case" essentially bootstraps you into proving it for infinite cases. There is also a general version of induction, but the only natural numbers case works for us now.

1.1.1. And so on to ?

Wait a second, so how are we defining "to infinity" here? How do we know that \(P(x)\) is going to work with every \(n\) even though we haven't tried it for every single \(n\)? Well, the answer is we extrapolate. We do the first few loops and we assume the logic carries out to any arbitrarily large loop. It's less of defining things in terms of infinity and more like playing a game where one person dares the other to go \(n\) times, where \(n\) is any natural number. They can say, "calculate that \(P(x)\) is true for \(P(6)\)!", and the claim is that you can always do that, even if they say one million instead of six, or one billion instead of one million. No matter how high the number, you can repeat the process \(n\) times and get the result that \(P(n)\) is true.

1.1.2. = ?

Wait: isn't the idea of a "base case" kind of analogous to the idea of recursion? And comparing \(P(S(n)) = P(n + 1)\) to \(P(n)\) kind of looks suspiciously like a recursive function, only, instead of using the base case in order to stop the program from running infinitely, we use the base case as a starting point to "run the program to infinity". Some connections are beginning to be made…

1.1.3. ~ ?

It seems one can describe many recursive structures as inherently relating to infinity. I posit that recursive structures have a starting point and an ending point – in the case of the factorial, the starting point is a natural number that is an input to the function, and the ending point is when it reaches zero (because the factorial function "iterates down", meaning a number is continually subtracted until it reaches a lower bound, meaning what we call the base case has to be always lower than the input). It is also conceivable that you can have a recursively defined function that has a base case higher than any possible inputs and iterates upwards. In this case, calling zero the "base case" of induction is actually misleading. If you model induction as a function, induction has no base case, and the input is usually evaluated at zero. Meaning, induction is a special case of recursion where no base case is defined. Although, I'm not sure actual career mathematicians would like my wording of this issue.

1.2. Set Construction

Given I've described Peano axioms already, I may as well use them. Although, Peano axioms may also be derived from ZFC set theory axioms.

Set \(S(x) = \{ x \}\) and \(0 = \{\}\). Then set construction describes the process of constructing the natural numbers from the empty set by nesting sets together. For example, \(1 = \{0\} = \{\{\}\}\), and \(2 = \{1\} = \{\{0\}\} = \{\{\{\}\}\}\). Then all natural numbers can be constructed recursively expanding the variables.

1.2.1.

And now there is a clear demonstrated link between Peano axioms and recursive structures.

1.3. Addition

Okay, that's all good, but natural numbers don't have a use case if even simple things like addition are not defined. Let's do that!

1.4. Congrats!

We've just defined a natural number! Every single object that can be described in terms of these axioms is also an instance of a natural number.