🔢 mathematics

Separable integer partition classes and Slater's list -- I

This paper applies Andrews' theory of separable integer partition classes to provide natural combinatorial interpretations and parameterized generalizations for the series sides of several Rogers-Ramanujan type identities from Slater's list, subsequently deriving alternative expressions and infinite products through $q$ -hypergeometric transformations.

Original authors: Aritram Dhar, Ankush Goswami, Runqiao Li

Published 2026-03-17

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Aritram Dhar, Ankush Goswami, Runqiao Li

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 have a giant, infinite library of numbers. In this library, there's a special way to organize numbers called partitions. Think of a partition like a stack of Lego bricks. If you have 5 bricks, you can stack them in different ways: a single tower of 5, a tower of 4 with a 1 on top, a tower of 3 with a 2 on top, and so on. Each unique stack is a "partition."

For decades, mathematicians have been trying to find secret rules that connect two different ways of looking at these stacks. One way is to list them out one by one (a series), and the other is to describe the whole collection with a single, elegant formula (a product).

In the 1950s, a mathematician named L. J. Slater compiled a massive list of 130 of these "secret rules" (identities). They are famous and beautiful, but there's a problem: We know the rules work mathematically, but we don't know why they work in a visual, physical sense. It's like knowing a magic trick works, but not seeing how the magician hides the rabbit. The "series" side of these rules is very hard to translate into a picture of Lego stacks.

The New Tool: The "Separable" Lego Set

This paper introduces a new way to look at these stacks, using a concept called Separable Integer Partition (SIP) classes.

Imagine you have a special Lego set with a rule:

You have a Base (a small, fixed set of bricks at the bottom).
On top of that base, you can add a Tail made of bricks that are all multiples of a specific number (like only adding bricks in groups of 4).

The magic of the "Separable" idea is that every valid stack in this specific class can be broken down into exactly one Base and one Tail. It's like having a unique key for every lock. If you can figure out the rules for the "Base" and the rules for the "Tail," you can understand the whole collection.

What the Authors Did

The authors took several of those mysterious rules from Slater's list and asked: "Can we build a special Lego set (an SIP class) where the 'Base' rules match the complicated math on the left side of the equation?"

Here is how they did it, step-by-step:

Finding the Pattern: They looked at a specific identity (a math equation) from Slater's list. The left side was a messy sum of numbers.
Building the Class: They invented a new set of rules for stacking Lego bricks. For example, "The gap between two bricks must be at least 2, but if the brick is even, the gap must be at least 4."
The Breakdown: They showed that any stack following these rules can be separated into a "Base" (the tricky part) and a "Tail" (the easy, repetitive part).
The Translation: Because they understood the "Base" and the "Tail," they could write down a new, generalized formula that represents the messy sum.
The Magic Trick: They used advanced math tools (called q-hypergeometric transformations) to turn that generalized formula into a clean, simple product.

The Results

By doing this, they achieved two big things:

They gave the rules a face: They found a physical, visual way to interpret the "series" side of Slater's identities. Instead of just seeing a string of numbers, we can now see it as a specific type of Lego stack with a "Base" and a "Tail."
They found new secrets: By generalizing the rules (adding variables like $x$ and $y$ to count different types of bricks), they discovered new identities that weren't on Slater's original list. In some cases, these new formulas simplified back into the famous infinite products, proving the old rules were correct in a brand new way.

A Simple Analogy: The Train

Think of the "Series" side of the equation as a long, winding train track with many switches and turns. It's hard to see where it's going.
Think of the "Product" side as a straight, high-speed bullet train.

For a long time, we knew the winding track led to the same destination as the bullet train, but we couldn't see the connection.

This paper builds a transfer station (the SIP class).

They show that the winding track is actually made of a Locomotive (the Base) pulling a long string of identical boxcars (the Tail).
Once you realize the train is just a Locomotive + Boxcars, you can easily calculate where it's going.
They then show that this specific train configuration is mathematically identical to the bullet train (the Product).

Why Does This Matter?

Before this paper, many of Slater's 130 identities were like locked boxes. We knew the combination (the math proof), but we didn't have the key to open them and see the treasure inside (the combinatorial meaning).

This paper provides a master key (the SIP framework). It doesn't just open one box; it gives us a systematic way to open many of them. It turns abstract algebra into a game of building blocks, allowing mathematicians to "see" the structure of these numbers rather than just calculating them.

In short: The authors took a list of 130 mysterious math riddles, built a new type of Lego set to solve them, and in the process, found new riddles and new ways to see the old ones.

1. Problem Statement

The paper addresses a significant gap in the theory of $q$ -series and integer partitions: the lack of combinatorial interpretations for many identities in Slater's list.

Context: Compiled in the 1950s, Slater's list contains 130 Rogers–Ramanujan type identities (series-product identities).
The Issue: While the analytic proofs of these identities rely heavily on Bailey pair techniques, the series sides of many identities lack natural combinatorial interpretations in terms of integer partitions. The product sides are usually easy to interpret (as partitions with specific congruence conditions), but the series sides are often difficult to map to partition constraints.
Goal: The authors aim to provide natural partition-theoretic interpretations for the series sides of selected identities from Slater's list and derive parameterized generalizations using a specific structural framework.

2. Methodology: Separable Integer Partition (SIP) Classes

The core methodology relies on Andrews' theory of Separable Integer Partition (SIP) classes.

Definition: An SIP class $\mathcal{P}$ with modulus $k$ is a set of partitions where every partition $\lambda$ can be uniquely decomposed into a "basis" part $\mathbf{b}$ and a "tail" $\boldsymbol{\pi}$ consisting of multiples of $k$ . Specifically, $\lambda_i = b_i + \pi_i$ , where $\mathbf{b}$ belongs to a finite basis set $B$ and $\boldsymbol{\pi}$ is a partition into non-negative multiples of $k$ .
Generating Function: The generating function for an SIP class is given by:
$\sum_{\lambda \in \mathcal{P}} q^{|\lambda|} = \sum_{n=0}^{\infty} \frac{B(n; q)}{(q^k; q^k)_n}$
where $B(n; q)$ is the generating function for the basis partitions of length $n$ .
Approach:
1. Identify the series side of a Slater identity.
2. Construct a specific SIP class $\mathcal{P}$ whose generating function matches the series side.
3. Define the "basis" partitions by establishing minimal gap conditions (e.g., differences between parts must be at least 2, or specific parity constraints).
4. Use recurrence relations on the basis generating functions to prove the series form.
5. Apply $q$ -hypergeometric transformations (e.g., Watson's transformation, Sears' transformation, Jacobi's triple product) to convert the generalized series into infinite products, thereby recovering the original Slater identities and their generalizations.

3. Key Contributions and Results

The paper analyzes specific identities from Slater's list, grouping them by the nature of their partition constraints.

A. Partitions with Restricted Gap Mod 4 (Section 3)

Identities Covered: Slater's (4), (25), (51), and (55).
Construction: The authors define a class $\mathcal{P}_4$ of strict partitions with gap conditions dependent on the parity of the parts (e.g., if both parts are odd, the difference is $\equiv 2 \pmod 4$ ; if both even, $\equiv 0 \pmod 4$ ).
Generalization: They derive a 3-variable generating function involving $x$ (odd parts), $y$ (even parts), and $q$ .
$\sum_{\lambda \in \mathcal{P}_4} x^{o(\lambda)} y^{e(\lambda)} q^{|\lambda|} = \sum_{n=0}^{\infty} \frac{(-yq/x; q^2)_n x^n q^{n^2}}{(q^4; q^4)_n}$
Results: By specializing parameters, they recover the product sides of Slater's (4) and (25) and derive new product identities for (51) and (55). They also provide new proofs for related identities using $q$ -hypergeometric transformations.

B. Strict Overpartitions (Section 4)

Identities Covered: Slater's (8) and (13).
Construction: The authors introduce strict overpartitions ( $\mathcal{S}$ ), where parts are distinct, and parts $\ge 2$ can be "overlined" if the gap to the next part is $\ge 2$ .
Generalization: They establish a connection to Lebesgue's identity. The generating function for these overpartitions yields:
$\sum_{n=0}^{\infty} \frac{(-aq; q)_n}{(q; q)_n} z^n q^{n(n+1)/2} = (-aq^2; q^2)_\infty (-zq; q)_\infty \sum_{n=0}^{\infty} \frac{(-aq)^n}{(q)_n (-zq)^n}$
Results: This framework provides a combinatorial interpretation for Slater's (8) and a new proof for Slater's (13), which involves a sum of two infinite products. The authors also discuss the limit behavior connecting these results to the Rogers–Ramanujan identities.

C. Partitions with Positional Gap Conditions (Section 5)

Identities Covered: Slater's (15) and (19).
Construction: The authors define classes $\mathcal{G}$ $G$ and $\mathcal{G}'$ $G^{'}$ based on positional gaps (constraints on $\lambda_{2i} - \lambda_{2i+1}$ $λ_{2 i} - λ_{2 i + 1}$ and $\lambda_{2i-1} - \lambda_{2i}$ $λ_{2 i - 1} - λ_{2 i}$ ) and parity of indices.
- For (15): $\lambda_{2i} - \lambda_{2i+1}$ is even, and specific constraints on the smallest part.
- For (19): $\lambda_{2i-1} - \lambda_{2i} \ge 3$ with parity constraints.
Results: They derive generating functions for these classes.
- For (19), they successfully apply Sears' transformation and Rogers' identity to prove the product side $(q^2, q^3, q^5; q^5)_\infty / (q^2; q^2)_\infty$ .
- For (15), they provide the partition interpretation and the series form but note that the transformation to the product side remains analytically challenging within the current framework.

D. Combinatorial Perspective (Section 6)

The authors discuss the potential for bijective proofs. They illustrate this with Slater's (5), providing a direct bijection between partitions into odd parts and a specific series form using a modified Ferrers diagram (Durfee square analysis).
They propose that for other identities, one can construct involutions on the SIP classes to characterize the difference between even and odd length partitions, potentially leading to full bijective proofs.

4. Significance

Systematic Framework: The paper demonstrates that the SIP framework is a powerful, systematic tool for decoding the combinatorial meaning of Rogers–Ramanujan type series that were previously only understood analytically.
New Generalizations: It does not merely re-prove known identities; it introduces multi-variable generalizations (tracking odd/even parts, overlined parts, etc.) that were previously unknown.
Bridging Analysis and Combinatorics: By converting difficult series into SIP generating functions, the authors bridge the gap between the analytic proofs (Bailey pairs) and combinatorial structures (partition constraints).
Future Directions: The work lays the groundwork for a broader combinatorial understanding of Slater's list, suggesting that many remaining identities can be tackled by constructing appropriate SIP classes and applying standard $q$ -hypergeometric transformations.

In summary, this paper successfully applies the theory of Separable Integer Partition classes to decode the combinatorial structure of several key identities in Slater's list, providing natural partition interpretations, parameterized generalizations, and new proofs for classical $q$ -series identities.