Eight Faces of Induction: Equivalent Principles on the Natural Numbers

Let ℕ = {0,1,2,…}. (If you start at 1, shift base cases accordingly.) This post proves that eight standard principles are all equivalent—different faces of the well-foundedness of (ℕ, ≤).

Table of Contents (click to expand)
  1. Eight equivalent principles
    1. (1) Ordinary Induction (PMI)
    2. (2) Complete (Strong) Induction (CMI)
    3. (3) Least Number Principle (LNP) / Well-Ordering
    4. (4) Principle of Finite Descent (PFD)
    5. (5) Well-Founded Induction on (&Nopf;, <)
    6. (6) No Infinite Decreasing Sequence
    7. (7) Course-of-Values Recursion / Iteration
    8. (8) Well-Ordering Theorem for ℕ
  2. Implication roadmap
  3. Proofs
    1. (1) ⇒ (2) and (2) ⇒ (1)
    2. (1) ⇒ (3) and (3) ⇒ (1)
    3. (3) ⇔ (4) ⇔ (6)
    4. (2) ⇔ (5)
    5. (3) ⇒ (7)
    6. (7) ⇒ (2)
  4. Conclusion

Eight equivalent principles

(1) Ordinary Induction (PMI)

If S ⊆ ℕ has 0 ∈ S and ∀n (n ∈ S ⇒ n+1 ∈ S), then S = ℕ.

(2) Complete (Strong) Induction (CMI)

If S ⊆ ℕ has 0 ∈ S and ∀n (({0,…,n} ⊆ S) ⇒ n+1 ∈ S), then S = ℕ.

(3) Least Number Principle (LNP) / Well-Ordering

Every nonempty A ⊆ ℕ has a least element.

(4) Principle of Finite Descent (PFD)

No infinite strictly decreasing sequence exists in ℕ.

(5) Well-Founded Induction on (ℕ,<)

If S ⊆ ℕ satisfies (∀m<n, m ∈ S) ⇒ n ∈ S, then S = ℕ.

(6) No Infinite Decreasing Sequence

Same as (4), stated explicitly for sequences.

(7) Course-of-Values Recursion / Iteration

Given X, a ∈ X, and an operator Φ, there is a unique f:ℕ→X with f(0)=a and f(n+1)=Φ(n, f restricted to {0,…,n}).

(8) Well-Ordering Theorem for ℕ

Every nonempty subset of ℕ has a minimum (essentially (3)).

Implication roadmap

(1) ⇔ (2), (1) ⇔ (3), (3) ⇔ (4) ⇔ (6), (2) ⇔ (5), (3) ⇒ (7), (7) ⇒ (2), and (8) ≡ (3).

Proofs

(1) ⇒ (2) and (2) ⇒ (1)

(1) implies (2) by defining T = {n : {0,…,n} ⊆ S} and applying induction. Conversely, if S satisfies PMI, then from {0,…,n} ⊆ S we know n ∈ S, hence n+1 ∈ S, so CMI applies.

(1) ⇒ (3) and (3) ⇒ (1)

(1) ⇒ (3): Assume A ⊆ ℕ nonempty. Let S = {n : no element of A ≤ n}. If 0 ∉ S, then 0 is least. Otherwise, inductively either a least element appears or S = ℕ, contradicting A ≠ ∅.

(3) ⇒ (1): Suppose S ⊆ ℕ with 0 ∈ S and successor-closed. If S ≠ ℕ, then T = ℕ\S is nonempty; by LNP it has a least element t. Minimality forces t ∈ S, contradiction. Thus S = ℕ.

(3) ⇔ (4) ⇔ (6)

(3) ⇒ (4): If an infinite descent exists, its set has no least element. (4) ⇒ (3): If A has no least, we can build an infinite descent. (4) and (6) are two forms of the same principle.

(2) ⇔ (5)

Well-founded induction on (ℕ,<) is exactly strong induction. Each implies the other directly.

(3) ⇒ (7)

Using the least bad index argument: if recursion fails at some n, take least such n, contradiction. So recursion exists and is unique.

(7) ⇒ (2)

Define f:ℕ→{0,1} via recursion to encode “{0,…,n} ⊆ S”. From uniqueness we deduce f(n)=1 always, hence S = ℕ. So recursion yields complete induction.

Conclusion

All eight principles are equivalent. They are different but interchangeable ways of saying that (ℕ, ≤) is well-founded: no infinite descent, every subset has a least element, induction works, and recursion is legitimate. Each tradition (number theory, logic, computer science) prefers one formulation, but logically they all belong to the same equivalence class.


About the Author

Andrew M. Cavallo is a guitarist, composer, and self-produced musician. His debut album, Christendom Reborn, can be found on his YouTube channel.

He is the author of Gödel’s God Theorem, which presents the Leibniz–Gödel System, i.e. four interlinking arguments for God's necessary and unique existence. Cavallo has also published peer-reviewed research, including On Mario Bunge’s Concept of System Boundary, which is indexed by Harvard in the Smithsonian/NASA Astrophysics Data System (ADS).

Andrew M. Cavallo is a math education consultant specializing in logic and proof for college success. Most high school curricula stop at geometry proofs, leaving students unprepared for the rigorous reasoning required in college mathematics, computer science, data science, and rapidly advancing fields such as artificial intelligence and machine learning. His Closing the STEM Gap: Proofs for College Readiness is a 12-week program that closes this gap.

Back to blog