🔢 mathematics

A symmetry approach to number tricks

This paper generalizes the classical "1089-number trick" by demonstrating that any pair of zero divisors in the group ring of the symmetric group induces a partition of n-digit numbers into subsets where the associated operations yield constant results, thereby creating a broad class of number tricks.

Original authors: Håkon Kolderup

Published 2026-04-07

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Håkon Kolderup

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are a magician standing on a stage. You ask a volunteer to pick a three-digit number, say 593. You tell them to reverse the digits to get 395, subtract the smaller from the larger (593 - 395 = 198), reverse that result (891), and add them together (198 + 891).

The volunteer does the math, and you confidently shout, "The answer is 1089!"

They are stunned. It feels like magic. But in this paper, mathematician Håkon Kolderup pulls back the curtain to show that this isn't magic at all—it's symmetry wearing a disguise.

Here is the paper explained in simple terms, using everyday analogies.

1. The Core Idea: The "Zero Divisor" Magic Trick

In the world of numbers, there are special pairs of "operators" (think of them as instructions like "reverse the digits" or "rotate the digits") that, when you do one after the other, cancel each other out completely. In math speak, they are called zero divisors.

The Analogy: Imagine you have a dance move called "Spin Left" and another called "Spin Right." If you do "Spin Left" followed by "Spin Right," you end up exactly where you started. The net movement is zero.
The Paper's Insight: Kolderup realized that if you take a number, apply a "Spin Left" instruction, then a "Spin Right" instruction, the digits of the number should theoretically cancel out to zero.
The Catch: Numbers aren't just abstract symbols; they are written in base-10 (our standard counting system). When you subtract or add, you have to deal with carrying (like when 5 + 7 = 12, you write down 2 and carry the 1). This "carrying" is the only thing that prevents the result from being zero.

2. The "Carrying" is the Secret Ingredient

The paper argues that the reason the 1089 trick always works is that the carrying process is predictable.

The Analogy: Think of the "carrying" like water flowing through a series of pipes. No matter how much water (digits) you pour in at the start, the pipes have a specific shape. If you pour water in a certain way, the pipes will always overflow in the exact same spots, creating a specific pattern of "splashes" (the final number).
The Discovery: Kolderup proved that if you group all possible starting numbers into different "buckets" (based on simple rules like "is the first digit bigger than the last?"), the water flow (the carrying) inside each bucket is identical.
The Result: Because the carrying is identical within a bucket, the final answer is always the same constant number (like 1089, 1998, or 990), regardless of which specific number you picked from that bucket.

3. The "Spectator Friendly" vs. "Formal" Tricks

The paper makes a fun distinction between tricks that work for a live audience and those that only work on a chalkboard.

Spectator Friendly: The math stays within the realm of normal, positive numbers that a human can easily write down.
- Example: The classic 1089 trick. You start with a positive number, subtract, get a positive number, and end with 1089.
Formal (Not Spectator Friendly): The math works perfectly in the abstract, but it might involve negative numbers or numbers that don't look like standard digits during the middle steps.
- Example: If you try the 1089 trick but start with a number where the last digit is bigger than the first (e.g., 395 - 593), you get a negative number. A magician can't easily show a negative number to a crowd without confusing them, even though the math still "works" in the background.

4. How to Invent Your Own Tricks

The most exciting part of the paper is that it gives you a recipe to invent your own number tricks. You don't need to be a genius; you just need to look at symmetries.

The Recipe:
1. Pick a set of digits (like 3 digits).
2. Find a "symmetry" (a way to shuffle them, like reversing them or rotating them).
3. Find a "zero divisor" (a combination of shuffles that cancel out).
4. Apply the math.
5. The "carrying" will force the digits to cancel out, leaving you with a constant number.

Real Examples from the Paper:

The Rotation Trick: Instead of reversing the number, rotate the digits (e.g., 123 becomes 231). If you subtract the rotated version and add up all the rotations of the result, you always get 1998.
The Swap Trick: Swap just the first two digits. If you follow the right steps, you can always get 990.

5. Why This Matters

This paper takes a fun party trick and turns it into a powerful mathematical tool. It shows that:

Math is Predictable: Even when things look chaotic (like different starting numbers), the underlying rules of how we write numbers (base-10) create rigid patterns.
Symmetry is Powerful: By understanding how things can be shuffled and canceled out, we can predict outcomes that seem impossible.
Anyone Can Do It: The author suggests that even school students learning about symmetry (like the symmetries of a triangle) could discover these number tricks just by playing with the rules of shuffling.

Summary

Think of the number 1089 not as a random lucky number, but as the echo of a specific symmetry. When you shuffle the digits of a number and then try to undo that shuffle, the "noise" of the digits cancels out. The only thing left standing is the "echo" of the carrying process, which Kolderup has proven is always the same for specific groups of numbers.

He has essentially given us a blueprint for building infinite new magic tricks, simply by finding new ways to shuffle and cancel out digits.

1. Problem Statement

The paper addresses the mathematical mechanism behind "number tricks," specifically the classical 1089 trick. In this trick, a three-digit number $abc$ (where $a > c$ ) is reversed to $cba$, subtracted ($abc - cba$), and the result is added to its reverse. The outcome is always 1089.

While previous research (e.g., Almirantis & Li, Behrends, Webster) has analyzed specific instances of these tricks or generalized them to $n$ -digit numbers, there was no unified formal framework explaining why these tricks work or a systematic method to generate new ones. The core problem is to formalize the cancellation of digits in these operations and to determine the conditions under which a sequence of permutations and arithmetic operations yields a constant result.

2. Methodology

Kolderup proposes a framework based on Group Ring Theory and Symmetric Groups. The methodology involves the following steps:

Group Ring Representation: The set of permutations of $n$ digits is modeled by the symmetric group $\Sigma_n$ . The operations performed on numbers (reversing, rotating, swapping digits) are treated as elements of the integral group ring $\mathbb{Z}[\Sigma_n]$ .
Zero Divisors as the Mechanism: The paper posits that number tricks arise from zero divisors in $\mathbb{Z}[\Sigma_n]$ . Specifically, if two elements $f, g \in \mathbb{Z}[\Sigma_n]$ satisfy $f \circ g = 0$ , the linear combination of permutations defined by $f$ and $g$ causes the input digits to cancel out algebraically.
Handling Carrying (Normalization): A major technical hurdle is that standard arithmetic involves "carrying" (e.g., $9+1=10$ ), which breaks simple linearity. The author defines a normalization map $N: \mathbb{Z}^n \to V_n$ (where $V_n = \{0, \dots, 9\}^n$ ). This map recursively processes a vector of integers from right to left, applying the standard carry operation to produce a valid $n$ -digit vector.
Evaluation Homomorphism: The map $\Phi: \mathbb{Z}^n \to \mathbb{Z}$ converts a digit vector back into an integer value ( $\Phi(x_1, \dots, x_n) = \sum x_i 10^{n-i}$ ).
Partitioning the Input Space: The author proves that the "carry" behavior is locally constant. The space of all $n$ -digit numbers ( $V_n$ ) can be partitioned into subsets (cells) defined by linear inequalities. Within each cell, the carry vector produced by the normalization map is constant.

3. Key Contributions

A. Formal Definition of Number Tricks

The paper introduces a rigorous definition of a Formal Number Trick as a triple $(f, g, U)$ , where:

$f, g \in \mathbb{Z}[\Sigma_n]$ are zero divisors ( $f \circ g = 0$ ).
$U \subseteq V_n$ is a subset of input vectors.
The output $\Phi(f \cdot N(g \cdot v))$ is constant for all $v \in U$ .

The paper distinguishes between Formal Tricks (mathematically valid but potentially yielding negative numbers or non-standard digit vectors) and Spectator Friendly Tricks (where the intermediate steps remain within the set of valid $n$ -digit numbers, and the final result is non-zero).

B. The Main Theorem (Theorem 3.6)

The central result establishes that for any pair of zero divisors $f \circ g = 0$ :

Partition: The set of $n$ -digit numbers $V_n$ can be partitioned into a finite number of disjoint cells $U_e$ , where each cell is defined by a specific constant carry vector $e$ .
Constancy: On each cell $U_e$ , the normalization map acts as $N(g \cdot v) = g \cdot v + e$ .
Constant Output: The final result is given by $\Phi(f \cdot e)$ . Since $e$ is constant on the cell, the output is independent of the specific digits of the input number, depending only on the carry vector generated by the region of the input space.

C. Generalization of Existing Tricks

The framework successfully generalizes:

The classical 1089 trick (using transposition $\tau=(13)$ ).
The "Rotation Trick" (using rotation $\rho=(123)$ ).
Generalized $n$ -digit tricks found in previous literature.
New tricks derived from subgroups and characters of $\Sigma_n$ .

4. Results and Examples

The paper provides several concrete examples derived from the theory:

Transposition Tricks:
- Using $\tau = (13)$ $τ = (13)$ (reversing first and last digit):
  - If $a > c$ : Output is 1089 (Spectator friendly).
  - If $a < c$ : Output is 1010 (Not spectator friendly due to negative intermediate steps).
  - If $a = c$ : Output is 0.
- Using $\tau = (12)$ $τ = (12)$ (swapping first two digits):
  - If $a > b$ : Output is 990.
- Using $\tau = (23)$ $τ = (23)$ (swapping last two digits):
  - If $b > c$ : Output is 99.
Rotation Trick:
- Using $\rho = (123)$ (cyclic shift) with $f = 1+\rho+\rho^2$ and $g = 1-\rho$ .
- For inputs satisfying $c < b \le a$ , the output is 1998.
- The paper identifies multiple cells with different outputs (e.g., 999, 1110, 2109) depending on the ordering of digits.
Subgroup Generalizations:
- The paper demonstrates how to construct tricks using any non-trivial subgroup $H \subseteq \Sigma_n$ and characters $\chi$ . For example, using the Klein four-subgroup in $\Sigma_4$ yields a spectator-friendly trick with an output of 1782.

5. Significance

Unification: The paper unifies disparate number tricks under a single algebraic framework, showing they are all manifestations of zero divisors in group rings acting on digit vectors.
Algorithmic Discovery: It provides a constructive method for discovering new number tricks. One need only find a zero divisor relation in $\mathbb{Z}[\Sigma_n]$ and analyze the resulting carry vectors to find spectator-friendly regions.
Educational Value: The approach connects abstract algebra (group rings, symmetric groups) with elementary arithmetic, offering a way for students to "discover" their own tricks by exploring symmetry relations (as illustrated in Example 4.2 regarding the symmetries of a triangle).
Mathematical Rigor: It resolves the ambiguity of "carrying" in number tricks by formalizing it as a normalization map and proving that the output depends solely on the carry vector, not the specific input digits.

In summary, Kolderup transforms the "magic" of number tricks into a predictable, algebraic phenomenon governed by the structure of the symmetric group and the arithmetic of carries.