Sylow's Third Theorem

In the abstract realm of group theory, finite groups are like fundamental particles whose internal structures are hidden from direct view. How do we probe their symmetries and components without a physical microscope? The challenge lies in deducing a group's entire architecture—its gears and clockwork—from a single piece of information: its size, or 'order.' This article explores a powerful set of logical detectors for this task: Sylow's theorems, with a special focus on the third theorem's incredible predictive power.

This exploration is divided into two main parts. In the "Principles and Mechanisms" chapter, we will uncover the simple yet profound arithmetic rules that govern the population of key substructures, known as Sylow $p$-subgroups. We will see how these rules act as a powerful filter, drastically reducing the realm of possibility for a group's structure. Following this, the "Applications and Interdisciplinary Connections" chapter will demonstrate the theorem's impact, showing how it can definitively prove a group is not a fundamental 'atom' of algebra, reveal its internal composition, and even explain the symmetries of physical objects like a cube. By the end, you will understand how two simple conditions on numbers provide a deep understanding of the sophisticated world of finite groups.

Imagine you're a physicist trying to understand the fundamental constituents of matter. You can't see a proton directly, but you can smash things into it and watch what comes out. By analyzing the debris, you deduce its properties, its internal structure, its symmetries. In the world of abstract algebra, we do something remarkably similar. Our "particles" are groups, and their "size" is their order. We "smash" them not with other particles, but with the elegant hammer of pure logic. Our detectors are a set of beautiful theorems laid down by the Norwegian mathematician Ludwig Sylow in the 19th century.

At the heart of this exploration is a process of counting. But we're not just counting any old thing. We are on the hunt for very specific substructures within a group $G$, called ​​Sylow $p$-subgroups​​. For any prime number $p$ that divides the order of our group, $|G|$, these are the largest possible subgroups whose order is a power of just that prime $p$. Think of them as the 'purest' $p$-part of the group's structure. Sylow's Third Theorem gives us two astonishingly simple rules that govern how many of these subgroups, a number we call $n_p$, can exist.

Let's say the order of our group is $|G| = p^k m$, where $p^k$ is the highest power of our prime $p$ that divides the order, and $m$ is the leftover part. The number of Sylow $p$-subgroups, $n_p$, must obey two strict conditions.

1. ​​The Divisibility Constraint:​​ The number $n_p$ must be a divisor of $m$.

2. ​​The Congruence Constraint:​​ $n_p$ must leave a remainder of 1 when divided by $p$. In mathematical shorthand, we write this as $n_p \equiv 1 \pmod p$.

At first glance, these rules might seem a bit arbitrary. A divisibility rule and a funny-looking remainder rule. But their combined power is immense. Let's take a group of order 36. The order is $|G|=36 = 3^2 \cdot 4$. We are interested in the Sylow 3-subgroups (which have order $3^2=9$). Here, $p=3$, $k=2$, and $m=4$.

The first rule says $n_3$ must divide $m=4$. The divisors of 4 are 1, 2, and 4. So, our candidate list for $n_3$ is $\{1, 2, 4\}$. Without any more information, this is a pretty good start.

But now, we apply the second rule: $n_3 \equiv 1 \pmod 3$. Let's test our candidates:

• Does $1$ leave a remainder of 1 when divided by 3? Yes, $1 = 0 \cdot 3 + 1$.
• Does $2$ leave a remainder of 1 when divided by 3? No, it leaves a remainder of 2. So, 2 is out.
• Does $4$ leave a remainder of 1 when divided by 3? Yes, $4 = 1 \cdot 3 + 1$.

Just like that, the list of possibilities for the number of Sylow 3-subgroups has been slashed from three down to just two: $n_3$ must be either 1 or 4. This is not a guess; it is a logical certainty for any group of order 36 that could possibly exist. The two rules, working in tandem, are a powerful filter on reality. You can see how crucial the second rule is. If we were to only use a weaker version of the first rule, say that $n_p$ must divide the total order $|G|$, the possibilities would be much wider and less useful.

This congruence rule has a beautiful, immediate consequence: for any prime $p$, it's impossible for a group to have exactly $p$ Sylow $p$-subgroups. That is, $n_p$ can never be equal to $p$. Why? If we set $n_p=p$, the congruence condition would demand $p \equiv 1 \pmod p$. This means $p$ should divide $p-1$. This is, of course, impossible for any prime $p$, as a positive number cannot divide a smaller positive number. It's a simple observation, but it reveals a deep constraint on the architecture of finite groups.

What happens when these two rules corner us into a single answer? Specifically, what does it mean when we are forced to conclude that $n_p = 1$?

This is where the theory gets its predictive power. Consider a group of order 35. Let's look at the Sylow 5-subgroups. The order is $|G|=35=5^1 \cdot 7$. Here, $p=5$ and $m=7$. The rules tell us that $n_5$ must divide 7 (so it's 1 or 7) and $n_5 \equiv 1 \pmod 5$. The only number that satisfies both conditions is 1. It must be 1. The same logic applies to a group of order 39; there can only be one Sylow 13-subgroup.

When there is only one of a particular Sylow $p$-subgroup, that subgroup is special. It is a ​​normal subgroup​​. What does that mean intuitively? Think of a perfectly symmetric crystal. You can rotate it in certain ways, and it looks the same. A normal subgroup is like a part of the group's structure that remains unchanged no matter how you "rattle" the group around (an operation called conjugation). It is a fundamental, stable component.

Groups that have no normal subgroups other than the trivial one (just the identity element) and the group itself are called ​​simple groups​​. They are the fundamental building blocks of all finite groups, much like prime numbers are the building blocks of integers. The discovery that $n_p=1$ for some prime $p$ is therefore a monumental one: it is a guarantee that the group in question is not a simple group. It has internal structure. We have found a gear in the clockwork.

The real beauty unfolds when we consider the Sylow subgroups for all the prime factors of a group's order at the same time. Let's stage a little detective story with a group $G$ of order 21, and we are given one extra clue: it is non-abelian (meaning the order of operations matters, like $AB \neq BA$).

The order is $|G|=21=3 \cdot 7$. Let's analyze the Sylow subgroups for both primes.

• For $p=7$: $n_7$ must divide 3, and $n_7 \equiv 1 \pmod 7$. The only number in the universe that works is $n_7 = 1$. So, we know for a fact that there is a unique, normal subgroup of order 7.
• For $p=3$: $n_3$ must divide 7, and $n_3 \equiv 1 \pmod 3$. Our candidates are 1 and 7 (since both $1 \equiv 1 \pmod 3$ and $7 \equiv 1 \pmod 3$).

Now, we use our clue. What if $n_3=1$? If both $n_7=1$ and $n_3=1$, the group would have two normal Sylow subgroups of coprime order. A wonderful result in group theory states that such a group is essentially just the "peaceful coexistence" of its parts (their direct product), and this structure is always abelian. But this contradicts our clue!

Therefore, the assumption that $n_3=1$ must be false. The only remaining possibility is $n_3=7$. With just the group's order and a single bit of information about its commutativity, we have deduced the exact population of its Sylow 3-subgroups. This same kind of logic can be applied to many groups whose order is a product of two primes, like a group of order 55, where we can immediately deduce that the number of Sylow 11-subgroups must be 1, while the number of Sylow 5-subgroups could be either 1 or 11, leading to two different possible group structures.

Sometimes, Sylow's theorems don't hand us a single, neat answer. Instead, they present us with a menu of possibilities. This isn't a failure of the theorem; it's a reflection of the richness of the mathematical world.

Let's examine a group of order 20. We have $|G|=20=2^2 \cdot 5$. How many Sylow 2-subgroups (of order 4) can there be? The rules say $n_2$ must divide 5, and $n_2$ must be odd. The numbers that fit this description are 1 and 5. So, $n_2$ could be 1, or it could be 5.

And it turns out, both scenarios are real! There exists a group of order 20—the cyclic group $C_{20}$, which you can picture as the 20 rotational symmetries of a 20-sided polygon—that has exactly one subgroup of order 4. But there is also a different group of order 20—the dihedral group $D_{10}$, the full symmetry group of a 10-sided polygon (including flips)—which has exactly five subgroups of order 4. Sylow's theorem didn't fail; it correctly predicted the only two possible blueprints for the Sylow 2-structure of a group of this size.

This brings us to the frontier. What about a group of order 120? The prime factorization is $120 = 2^3 \cdot 3 \cdot 5$. Applying Sylow's theorems gives us a whole list of possibilities:

• $n_2$ could be 1, 3, 5, or 15.
• $n_3$ could be 1, 4, 10, or 40.
• $n_5$ could be 1 or 6.

For each prime, there is a possibility that $n_p > 1$. So, Sylow's theorems alone don't immediately tell us if a group of order 120 must have a normal subgroup. Does this mean we've hit a wall? Not at all! This is where the real work of the mathematician begins. This list of possibilities is not an end, but a beginning. It provides the crucial clues that, when combined with more advanced counting arguments, allow us to prove that no matter how you arrange the pieces, a group of order 120 can never be simple. The path to discovery is not always a straight line. Sometimes, it is a process of eliminating possibilities, a grand game of cosmic sudoku, with Sylow's theorems providing the fundamental rules of play.

Having acquainted ourselves with the machinery of Sylow’s theorems, we might be tempted to view them as a niche tool for the abstract algebraist. But that would be like seeing a master key and thinking it only opens one specific door. In truth, these theorems are a skeleton key, unlocking profound structural truths not just within abstract algebra, but across a surprising landscape of mathematics and science. Their beauty lies in how a few simple arithmetic rules—the conditions on the number of Sylow $p$-subgroups, $n_p$—dictate deep, non-negotiable features of any finite group, no matter how complex or where it arises.

Let us embark on a journey to see this power in action. We will begin with the most striking consequence of the theorems and then venture into subtler territories, discovering connections to geometry, cryptography, and the very "atoms" of group theory itself.

The most dramatic application of Sylow’s Third Theorem occurs when it leaves no room for debate. Sometimes, the dual constraints—that $n_p$ must divide a certain number and also be one more than a multiple of $p$—pin down its value to a single possibility: $n_p=1$.

What does it mean for there to be only one Sylow $p$-subgroup? Recall that conjugating a Sylow $p$-subgroup by an element of the group must yield another Sylow $p$-subgroup. If there is only one to begin with, this subgroup must be sent back to itself under any conjugation. This is the very definition of a normal subgroup! The existence of a single Sylow $p$-subgroup is a certificate, signed by the theorems themselves, guaranteeing that the group is not simple; it possesses a non-trivial, proper normal subgroup.

Consider any group of order 28, which is $2^2 \cdot 7$. How much can we say about its structure without knowing anything else? Let's consult the theorems for the prime $p=7$. The number of Sylow 7-subgroups, $n_7$, must divide $28/7=4$ and satisfy $n_7 \equiv 1 \pmod{7}$. The only number that meets both conditions is $1$. It's inescapable: $n_7=1$. This means that every single group of order 28, no matter how it is constructed, must contain a normal subgroup of order 7. The arithmetic of the number 28 itself forces a specific structural feature.

Sometimes, this logic extends to give us an almost complete picture of the group. Take a group of order 99, or $3^2 \cdot 11$. For $p=11$, $n_{11}$ must divide $99/11=9$ and satisfy $n_{11} \equiv 1 \pmod{11}$. Again, only $n_{11}=1$ is possible. Now let's check $p=3$. The number $n_3$ must divide $99/9=11$ and satisfy $n_3 \equiv 1 \pmod{3}$. The divisors of 11 are 1 and 11, but $11 \not\equiv 1 \pmod 3$. So, we are forced to have $n_3=1$ as well!

This is a remarkable result. The group has a normal subgroup of order 9 and a normal subgroup of order 11. These two subgroups have only the identity element in common and together they account for all the elements of the group. The only way this can happen is if the larger group is simply the "direct product" of its two smaller parts. The group of order 99 behaves just like two independent machines—one of order 9 and one of order 11—working side-by-side with no complicated interaction between them. This kind of decomposition is invaluable in fields like cryptography, where the security of a system can depend on the predictable structure of the underlying group. A similar, unavoidable conclusion holds for any group of order $51 = 3 \cdot 17$, which must also split into a direct product of its Sylow subgroups.

What happens when the theorems are not so decisive? What if they allow for $n_p$ to be greater than one? This is where the real detective work begins. If $n_p > 1$ for every prime $p$, the group has no guaranteed normal subgroups from Sylow's theorem, and it might be one of the elusive "simple groups"—the indivisible building blocks of all finite groups, analogous to the prime numbers for integers.

Sylow’s theorems give us a wonderful tool to hunt these creatures: the counting argument. If a group is simple, it must have $n_p > 1$ for all its prime factors $p$. Let’s take these larger values of $n_p$ and see where they lead.

Imagine we are searching for a simple group of order 30, which is $2 \cdot 3 \cdot 5$. If it were simple, it could not have a unique Sylow subgroup. For $p=5$, $n_5$ must divide $30/5=6$ and be $1 \pmod 5$. The possibilities are $1$ and $6$. To be simple, we must have $n_5=6$. Each of these 6 subgroups has order 5 and thus contains 4 elements of order 5. Since these subgroups can only intersect at the identity, we have $6 \times 4 = 24$ distinct elements of order 5.

For $p=3$, $n_3$ must divide $30/3=10$ and be $1 \pmod 3$. The possibilities are $1$ and $10$. Again, for a simple group, we must have $n_3=10$. Each of these 10 subgroups contains 2 elements of order 3, giving us $10 \times 2 = 20$ distinct elements of order 3.

Now, let's step back. Our hypothetical simple group must contain at least 24 elements of order 5 and 20 elements of order 3. But $24 + 20 = 44$. This is a catastrophe! We have already counted 44 distinct elements, but our group only has 30 elements in total. This is an absurdity, like trying to fit an elephant in a birdcage. The contradiction is inescapable. Our initial assumption—that a simple group of order 30 exists—must be false. No such group can exist. This beautiful argument, using nothing but counting, rules out an entire class of groups from being fundamental building blocks. The counting argument is a powerful tool to rule out simplicity. For other orders, like $255 = 3 \cdot 5 \cdot 17$, the conclusion is even more direct: Sylow's theorem alone forces the number of Sylow 17-subgroups to be one, which immediately proves the group is not simple.

The power of group theory, and Sylow's theorems in particular, is most dazzling when it connects the abstract world of symbols to tangible, physical reality.

Let’s consider an object we can all hold in our hands (or at least in our minds): a perfect cube. The set of all rotational symmetries that leave the cube looking unchanged forms a group, called the chiral octahedral group, $O$. A quick count reveals it has 24 distinct rotations, so its order is $24=2^3 \cdot 3$. What can Sylow's theorems tell us about the symmetries of a cube?

For $p=3$, $n_3$ must divide $24/3=8$ and be $1 \pmod 3$. The possibilities are $1$ and $4$. For $p=2$, $n_2$ must divide $24/8=3$ and be $1 \pmod 2$. The possibilities are $1$ and $3$.

It turns out that for this group, $n_3=4$ and $n_2=3$. These are not just abstract numbers! Look at a cube. It has four axes of 3-fold rotation, passing through opposite vertices. These correspond to the four Sylow 3-subgroups. It also has three axes of 4-fold rotation, passing through the centers of opposite faces. These correspond to the three Sylow 2-subgroups. The abstract arithmetic of Sylow’s theorems predicts the geometric arrangement of the cube's symmetries! This is a profound link between pure algebra and the spatial reality we inhabit. In fact, this symmetry group is isomorphic to the group of permutations of four objects, $S_4$, where the Sylow subgroups also have clear interpretations related to how elements are permuted.

This reach extends beyond geometry and into the realm of linear algebra, which forms the bedrock of modern physics and computer science. Consider the group of invertible $2 \times 2$ matrices whose entries are from the finite field $\mathbb{Z}_3$ (the integers modulo 3). This group, denoted $GL(2, \mathbb{Z}_3)$, represents all the ways one can stretch, shear, and reflect a two-dimensional grid of points. Its order is 48, or $2^4 \cdot 3$. Applying Sylow’s theorem for $p=2$, we find that the number of Sylow 2-subgroups (hulking structures of order 16) must divide $48/16=3$ and be $1 \pmod 2$. The only possibilities are 1 or 3. This simple fact provides a crucial starting point for dissecting the intricate structure of this matrix group, which is a fundamental object in areas like error-correcting codes and cryptography.

We've seen how Sylow's theorems can prove a group is not simple. But what happens when the counting argument fails to produce a contradiction? This failure is, in itself, a momentous discovery.

Consider the group of even permutations of 5 elements, $A_5$, whose order is $60 = 2^2 \cdot 3 \cdot 5$. Let's apply our techniques. For $p=5$, we find $n_5=6$. For $p=3$, we find $n_3=10$. For $p=2$, we get $n_2=5$. None of these are 1, so the door to simplicity is open. If we count the elements, we find $6 \times 4=24$ elements of order 5, and $10 \times 2=20$ elements of order 3. That's 44 elements. Adding the elements from the Sylow 2-subgroups and the identity, the count comes out just right—no contradiction!.

Sylow's theorems have led us to the threshold. They have ruled out countless orders, but here, for order 60, they go silent. They cannot prove non-simplicity. And it is precisely because this group withstands the onslaught of the Sylow tests that it becomes a prime candidate for a simple group. In fact, $A_5$ is simple—the smallest non-abelian simple group. It is a true fundamental particle of finite group theory. The quest to find and classify all such "atoms" was one of the greatest mathematical odysseys of the 20th century. And on that grand journey, Sylow's theorems served as the first, most trusted, and most illuminating compass.