P-Regular Elements

The study of symmetry, a cornerstone of modern mathematics and physics, finds its most potent expression in representation theory, which translates the abstract language of groups into the concrete world of linear transformations. In the familiar setting of complex numbers, this theory is remarkably complete and elegant. However, when we shift our perspective to a 'modular' universe, one governed by the arithmetic of a finite field of characteristic $p$, this classical framework encounters profound challenges, particularly when $p$ divides the order of the group. The familiar harmony between representations and group structure appears to be lost.

This article addresses the fundamental question: How can we restore order and build a coherent theory of representations in this modular setting? The answer lies in identifying and focusing on a special class of group elements that behave well even when the classical theory fails. These are the ​​p-regular elements​​. This article provides a comprehensive overview of this pivotal concept. The first chapter, "Principles and Mechanisms," will define p-regular elements and introduce the foundational theorems of Richard Brauer that place them at the center of the theory. The second chapter, "Applications and Interdisciplinary Connections," will demonstrate how this single concept serves as a powerful tool for counting representations, defining new characters, and bridging the gap between the modular and classical worlds.

Imagine you are a physicist studying the fundamental laws of nature. You discover a beautiful, elegant theory that works perfectly, built on the foundation of the real and complex numbers. This is the world of ​​ordinary representation theory​​, a stunningly complete picture of group symmetry where our number system is the familiar, infinite, and forgiving field of complex numbers, $\mathbb{C}$. In this world, we have a powerful dictionary, character theory, that translates the abstract algebra of groups into the concrete language of matrices and numbers. A cornerstone of this dictionary is a miraculous fact: for any finite group, the number of its fundamental, irreducible representations is exactly equal to the number of its conjugacy classes.

But what happens if we are forced to work in a different universe, a "modular" one? What if our numbers are no longer infinite, but are drawn from a finite field of characteristic $p$, like the integers modulo a prime $p$? Suddenly, our beautiful theory seems to crumble. Division by $p$ becomes impossible, and many of the elegant proofs of ordinary theory break down. It's like trying to draw a perfect, smooth circle on a screen made of coarse pixels—some details are lost, and new, strange patterns emerge. This is the world of ​​modular representation theory​​. Our quest is to find the new laws that govern symmetry in this pixelated universe. The key, as we will see, lies in identifying a special set of "well-behaved" group elements that survive the transition.

When we step into the world of characteristic $p$, the prime $p$ itself becomes a central character in our story. It acts like a filter, sorting the elements of our group $G$ into two kinds. Some elements have an order that is a multiple of $p$; these are the ones that cause trouble. But others have an order that is not divisible by $p$. These special elements are called ​​p-regular​​.

Let’s be precise: an element $g$ in a group $G$ is called ​​p-regular​​ if its order, the smallest positive integer $k$ such that $g^k$ is the identity, is not divisible by the prime $p$. These are the elements that, in a sense, don't "feel" the influence of the prime $p$ as strongly. They form a special class of citizens that will be the key to restoring order to our theory.

Let's see this in action. Consider the group $D_{10}$, the symmetries of a regular pentagon, and let's choose our prime to be $p=5$. This group has 10 elements: the identity, four rotations of order 5, and five reflections of order 2.

• The identity element has order 1. Since 5 does not divide 1, it is 5-regular.
• The four non-trivial rotations all have order 5. Since 5 is divisible by 5, these elements are not 5-regular.
• The five reflections all have order 2. Since 5 does not divide 2, they are all 5-regular.

So, in the modular world of characteristic 5, the rotations of $D_{10}$ become problematic, while the identity and the reflections remain "regular". We care not just about individual elements, but about ​​p-regular conjugacy classes​​, which are classes where every element is p-regular. In $D_{10}$, the regular elements fall into two such classes: the class containing just the identity, and the class containing all five reflections.

This filtering process happens in every group. For the alternating group $A_4$ (the rotational symmetries of a tetrahedron) and the prime $p=3$, the elements are the identity (order 1), three double transpositions like $(12)(34)$ (order 2), and eight 3-cycles like $(123)$ (order 3). When we apply the "3-regular" filter, the elements of order 3 are sifted out, leaving only the identity and the double transpositions as the 3-regular elements.

Here is where the magic happens. In ordinary theory, we had a perfect census: the number of irreducible representations equals the number of conjugacy classes. In the modular world, this is false. However, the great mathematician Richard Brauer discovered a new, equally profound law. He proved that:

​​The number of irreducible modular representations of a group $G$ over a field of characteristic $p$ is exactly equal to the number of p-regular conjugacy classes of $G$.​​

This is a breathtaking result! It tells us that even in the strange, pixelated landscape of modular representations, a deep sense of order prevails. The theory doesn't just shatter; it reassembles itself around the p-regular elements. They are precisely the objects we need to count to understand the new building blocks of symmetry.

This theorem has dramatic and beautiful consequences. For instance, what happens if our prime $p$ does not divide the order of the group $G$? Let's consider a cyclic group of order 35 and the prime $p=3$. By Lagrange's theorem, the order of any element must divide 35. Since 3 does not divide 35, the order of every element is coprime to 3. This means that every element is 3-regular! The filter lets everyone through. In this case, the number of p-regular classes is just the total number of classes, and Brauer's theorem tells us that the number of irreducible modular representations is the same as the number of irreducible ordinary ones. The modular world looks identical to the ordinary one. The theory "heals" itself completely when $p$ is not a factor of the group's size.

Now consider the extreme opposite. What if a group has only one p-regular conjugacy class? Since the identity element is always p-regular, this single class must be $\{e\}$. This means every single non-identity element in the group is not p-regular; their orders must all be divisible by $p$. A little thought reveals that this is only possible if the order of every element is a power of $p$. Such a group is called a ​​p-group​​. Brauer's theorem then delivers a stunning punchline: for any p-group, no matter how large or complicated, it has only one irreducible representation in characteristic $p$—the trivial one!. All the rich structure of the group seems to collapse into a single point from the modular perspective.

This theorem isn't just a theoretical curiosity; it's a powerful computational tool. To find the number of irreducible representations of a vast group like $GL_2(\mathbb{F}_3)$ (the group of invertible $2 \times 2$ matrices with entries from the field of 3 elements) in characteristic 2, one doesn't need to construct the representations themselves. Instead, one "simply" needs to embark on a group-theoretic safari to count how many conjugacy classes consist of elements with odd order. This difficult task, which involves analyzing the structure of matrices over finite fields, ultimately reveals that there are only two such classes. By Brauer's theorem, we immediately know there must be exactly two irreducible representations, without ever having to write down a single matrix for them.

So, we have two worlds—the ordinary (characteristic 0) and the modular (characteristic $p$)—and a new law of harmony in the latter. But how are these two worlds connected? Brauer also provided the bridge.

The characters of modular representations, known as ​​Brauer characters​​, are functions that look like ordinary characters but are defined only on the set of p-regular elements. This makes perfect sense; these are the elements that behave well. The most basic one is the ​​principal Brauer character​​, $\phi_1$, which corresponds to the trivial representation (where every element is mapped to 1). Unsurprisingly, its value is $\phi_1(g)=1$ for every p-regular element $g$.

The connection is this: take any ordinary irreducible character $\chi$ (from the world of complex numbers). If you restrict its domain, ignoring all the non-p-regular elements, the resulting function, $\chi|_{p\text{-reg}}$, can be written as a unique sum of irreducible Brauer characters. $\chi|_{p\text{-reg}} = \sum_{j} d_{\chi \phi_j} \phi_j$ The coefficients $d_{\chi \phi_j}$ are non-negative integers called ​​decomposition numbers​​. They form a "decomposition matrix" that acts as the dictionary between the two worlds. It tells us precisely how an ordinary, "high-resolution" representation breaks down or "decomposes" into its fundamental "pixelated" components when viewed through the lens of characteristic $p$.

Let's see this bridge in action with the symmetric group $S_3$ and prime $p=2$. The ordinary characters are well-known. The 2-regular classes are those containing the identity and the 3-cycles. In ordinary theory over complex numbers, the trivial character $\chi_1$ and the sign character $\chi_2$ are distinct. However, when restricted to the 2-regular elements, they become identical, both reducing to the trivial Brauer character $\phi_1$. This demonstrates how distinct ordinary representations can become indistinguishable in the modular world. The 2-dimensional ordinary character $\chi_3$, when restricted, decomposes into a sum of the available Brauer characters, showing how an irreducible object in the ordinary world can become a composite object in the modular one.

But often, the decomposition is more interesting. For the group $S_4$ at prime $p=3$, one of its ordinary 2-dimensional characters, when restricted to the 3-regular elements, is no longer irreducible. Instead, it breaks apart into the sum of two distinct 1-dimensional Brauer characters. An irreducible object in the ordinary world becomes a composite object in the modular world. This process of decomposition, governed by the p-regular elements and Brauer characters, reveals the deep and often subtle relationship between the perfect world of characteristic zero and the fascinating, finite world of characteristic $p$. Far from being a broken version of the old theory, modular representation theory stands as a rich subject in its own right, with its own rules, its own beauty, and its own profound connections to the very heart of group structure.

In the previous chapter, we ventured into what might have seemed like a formal exercise: we took a finite group $G$ and a prime number $p$, and we sorted the group's elements into two piles—those whose order is divisible by $p$, and those whose order is not. The latter, we gave the special name "$p$-regular elements." It may have felt like a simple classification, a bit of abstract bookkeeping. But what I want to show you now is that this single, simple act of sorting is one of the most profound and fruitful ideas in the entire theory of groups. This one distinction turns out to be the master key that unlocks the beautiful and intricate world of "modular representations," the study of how groups act when the numbers we use for measurement have a finite characteristic.

When the characteristic $p$ of our field divides the order of our group, the comfortable world of classical representation theory, where every representation shatters cleanly into a sum of irreducible ones, collapses. The group algebra becomes "non-semisimple," a term that hides a fascinating complexity. It's like looking at a crystal with a flawed lens; everything is a bit blurry. The genius of Richard Brauer was to realize that the way to bring the picture back into focus is to put on a special pair of glasses—glasses that only let you see the $p$-regular elements. When you do that, a hidden, elegantly structured universe reveals itself.

The first and most stunning application is a new census of the fundamental building blocks. In the classical world, the number of irreducible representations is simply the number of conjugacy classes in the group. What is the rule in this new modular world? The answer is as simple as it is beautiful: the number of non-isomorphic simple modules (the irreducible building blocks) is precisely the number of conjugacy classes of ​​$p$-regular​​ elements.

Think about what this means. It tells us that all the complexity of the modular theory, all the information about the fundamental particles of our group representations, is somehow encoded entirely in those elements that manage to avoid having an order divisible by $p$. The other elements, the "$p$-singular" ones, become part of the "scaffolding" or the "mortar" that holds the building blocks together, but they are not the blocks themselves.

For example, if we study the symmetries of a hexagon, the group $D_{12}$ of order 12, using a field of characteristic $p=3$, we find that the number of simple modules is not given by the total number of conjugacy classes. Instead, we must first filter for the 3-regular classes—those whose elements have orders not divisible by 3—and count them. It turns out there are exactly four such classes, and so there are exactly four simple modules. Similarly, for a much larger and more complex group like the alternating group $A_7$ (with 2520 elements), determining the number of simple modules over a field with characteristic 3 would seem a Herculean task. Yet, Brauer's theorem gives us a magnificent shortcut: just count the types of permutations whose cycle structures don't involve multiples of 3. Doing so reveals that there are exactly six such conjugacy classes, and thus precisely six fundamental representations. This is a powerful and predictive piece of magic.

In classical theory, we have characters—functions that assign a complex number (the trace) to each group element, acting as a "tag" or "fingerprint" for a representation. But in the modular world, these ordinary characters behave strangely. Again, the solution is to restrict our vision. We define a new kind of character, a ​​Brauer character​​, which is a function defined only on the set of $p$-regular elements.

It’s on this restricted domain that things become beautiful and well-behaved again. These Brauer characters form a table, much like the ordinary character table, but with its columns indexed only by the $p$-regular conjugacy classes. For a cyclic group of order 6, whose elements have orders 1, 2, 3, and 6, when we work in characteristic $p=2$, we ignore the elements of order 2 and 6. The Brauer character table concerns itself only with the elements of order 1 and 3, which form the 2-regular classes. The resulting table is a small, perfect square, capturing all the essential information about the simple modules in this context.

A beautifully simple, yet crucial, consequence is that the dimension of a simple module is given by its Brauer character's value at the identity element—which is, of course, always $p$-regular. So, by merely glancing at the first column of a Brauer character table, we can instantly read off the dimensions of all the fundamental building blocks.

And just like ordinary characters, these Brauer characters obey their own "rules of the game." They have orthogonality relations, which are incredibly powerful tools for calculation. These relations allow us, for instance, to reconstruct missing parts of a Brauer character table by relating sums of character values to the structure of the centralizers of $p$-regular elements,. This tells us that the theory of Brauer characters is not just a pale imitation of the ordinary theory; it is a rich and complete mathematical structure in its own right.

You might be wondering if this new modular world is completely disconnected from the familiar classical one of characteristic zero. The answer is no, and the bridge between them is, once again, built upon the foundation of $p$-regular elements. If you take an ordinary irreducible character and restrict its domain to just the $p$-regular elements, it turns out that this restricted function can always be written as a sum of irreducible Brauer characters.

The coefficients in this sum are non-negative integers called ​​decomposition numbers​​. We can arrange these numbers in a matrix, the ​​decomposition matrix​​, whose rows are indexed by the ordinary characters and whose columns are indexed by the Brauer characters. This matrix is a veritable "Rosetta Stone" that allows us to translate between the two theories. It tells us precisely how each classical representation "decomposes" or breaks down when viewed through the lens of characteristic $p$.

For the cyclic group $C_6$ and $p=2$, this matrix has a simple, repeating pattern that elegantly shows how the six ordinary characters pair up and correspond to the three Brauer characters. For a more complicated group like $S_4$, calculating this matrix requires a bit more work, but it reveals deep connections between the different representations. The very existence of this matrix is a profound statement about the underlying unity of mathematics: the seemingly disparate worlds of characteristic 0 and characteristic $p$ are locked together in a precise relationship, and the key to this relationship is the set of $p$-regular elements.

The utility of $p$-regular elements goes even deeper, helping us understand the very architecture of the group algebra itself.

When the algebra is not semisimple, it contains a special ideal called the ​​Jacobson radical​​, which you can think of as the collection of all "truly nilpotent" elements—the "mushy" part of the algebra that prevents it from being a clean direct sum of simple pieces. How large is this radical? The concept of $p$-regularity provides a stunningly effective tool for measuring it. For a finite $p$-group (a group whose order is a power of $p$) over a field of characteristic $p$, the only $p$-regular element is the identity. This means there is only one simple module—the trivial one! The consequence is dramatic: the algebra's semisimple part is just one-dimensional, and nearly the entire algebra consists of this Jacobson radical. For the dihedral group $D_8$ over a field of characteristic 2, this principle tells us immediately that the 8-dimensional group algebra has a 7-dimensional radical.

Finally, $p$-regular elements provide the ultimate organizing principle for the entire theory through the concept of ​​blocks​​. The set of all representations (both irreducible and indecomposable) can be partitioned into disjoint families called $p$-blocks. Think of it as sorting all your files into separate folders, where the files in each folder share some deep common properties. What determines this partition? It is, in a deep sense, related to the $p$-regular elements.

Each block is assigned an invariant called a ​​defect group​​, which is a certain $p$-subgroup of $G$. The properties of this defect group dictate almost everything about the representations inside that block. And how do you find this all-important defect group? One way is by examining the centralizers of the $p$-regular elements of the group. By finding a $p$-regular element whose centralizer has the largest possible $p$-part, we can determine the "defect" of the principal block—the most important "folder" containing the trivial representation. This shows how $p$-regular elements are not just useful for counting or character theory, but are fundamental to the highest level of organization in modular representation theory.

So, we see that from a single, simple definition flows a torrent of applications. The idea of a $p$-regular element is the thread that weaves together the entire tapestry of modular representation theory. It gives us a new census for our building blocks, a new language of characters to describe them, a Rosetta Stone to connect to the old world, and a blueprint for understanding the deepest architectural structures of group algebras. It is a perfect example of the power and beauty of a "right" idea in mathematics.