The Attaching Map and its Degree

In the field of topology, complex shapes are often understood by building them from simple components—points, lines, and disks. But how are these pieces assembled? The answer lies in a set of instructions known as an ​​attaching map​​, a concept that bridges the gap between intuitive geometric gluing and precise algebraic outcomes. This article addresses a fundamental question: how does the specific way we "glue" a new piece onto an existing structure fundamentally alter its character? We will see that this entire process can often be captured by a single integer—the ​​degree​​ of the attaching map.

This article will guide you through this powerful idea. In the first chapter, ​​"Principles and Mechanisms"​​, we will explore the mechanics of CW complexes and see how the degree of an attaching map directly manipulates a space's algebraic DNA, its fundamental group and homology groups. Subsequently, in ​​"Applications and Interdisciplinary Connections"​​, we will discover how this tool is not merely descriptive but creative, enabling mathematicians to engineer spaces with specific properties and revealing deep connections to cosmology, differential geometry, and abstract algebra.

Imagine you are a sculptor, but instead of clay or marble, your materials are the very fabric of space. You start with the simplest possible objects: points. Then you take one-dimensional threads and connect these points. Then you take two-dimensional patches and glue them onto your thread-like structures. How could you possibly create the rich and complex universe of shapes we see, from simple spheres to bizarre, non-orientable surfaces, using just these elementary building blocks? The secret lies not just in the pieces themselves, but in the instructions for how they are glued together. In the world of topology, these instructions are captured by a beautifully simple and powerful idea: the ​​attaching map​​ and its ​​degree​​.

Topologists have a systematic way of building spaces called a ​​CW complex​​. The name sounds technical, but the idea is wonderfully intuitive. You build a space skeleton-first, dimension by dimension.

1. You start with a collection of points, your ​​0-cells​​. This is the 0-skeleton.
2. You take some line segments, or ​​1-cells​​, and "attach" their endpoints to the points in your 0-skeleton. This gives you a graph-like structure, the ​​1-skeleton​​. For instance, if you start with a single 0-cell (a point) and attach a single 1-cell by gluing both of its ends to that same point, you've created a loop—a space that looks just like a circle, $S^1$.
3. Next, you take two-dimensional disks, or ​​2-cells​​, and attach their circular boundaries to the 1-skeleton you've just built.

This process continues to higher dimensions, but for now, let's focus on this crucial step of attaching a 2-cell to a circle. This is where the magic happens. The instructions for this attachment are encoded in a function called the ​​attaching map​​, $\phi$. This map takes every point on the boundary of our disk (which is a circle) and tells us which point on the target circle to glue it to.

Think of the boundary of your disk as an elastic loop. Your target space is a solid circular wire. How can you glue the loop onto the wire?

You could just wrap it around the wire once, matching point for point. This would be a simple, direct attachment. Or, you could be more creative: you could wrap the elastic loop around the wire twice before gluing it down. Or three times. Or you could wrap it once in the opposite direction. You could even just pinch the entire elastic loop and glue it to a single point on the wire, not wrapping it at all.

Each of these choices creates a profoundly different final shape. Amazingly, the essence of this wrapping, this twisting, can be distilled into a single integer, $k$, called the ​​degree​​ of the attaching map.

• ​​Degree $k=1$​​: You wrap the boundary around the circle once, preserving the direction.
• ​​Degree $k=2$​​: You wrap it twice. Every point on the target circle is "covered" by two points from the disk's boundary.
• ​​Degree $k=0$​​: You take the whole boundary of the disk and collapse it to a single point on the circle.
• ​​Degree $k=-1$​​: You wrap it once, but in the reverse direction.

This number, the degree, is our master control knob. By simply changing this integer, we can dial up a whole universe of different topological spaces.

Here is where the story takes a turn from the visual to the abstract, revealing a deep unity between geometry and algebra. Topologists have developed tools to "listen" to the structure of a space, to distinguish a sphere from a donut without having to look at it. Two of the most powerful tools are the ​​fundamental group​​, $\pi_1$, which describes loops in a space, and ​​homology groups​​, $H_n$, which, in low dimensions, count holes of different dimensions.

The geometric act of attaching a cell with degree $k$ has a precise and predictable echo in this algebraic world. The degree $k$ doesn't just describe the gluing; it dictates the very "sound" of the resulting space.

Killing Loops with the Fundamental Group

Let's start with our circle, $S^1$. Its fundamental group, $\pi_1(S^1)$, is the group of integers, $\mathbb{Z}$. You can think of the integer $n$ as representing a loop that wraps around the circle $n$ times. The generator, let's call it $a$, corresponds to wrapping around once.

When we attach a 2-cell (our disk), we are essentially filling in a region. The boundary of this disk, which we glued on, now becomes a loop that can be contracted to a point within the new space. The loop that this boundary traces out is precisely the loop of degree $k$. In the language of algebra, this means that the loop representing "wrapping around $k$ times," which we can write as $a^k$, becomes trivial in the new space. We are imposing the relation $a^k = 1$.

So, the fundamental group of our new space $X$ is the original group, $\mathbb{Z}$, but with this new relation imposed:

This is the group of integers modulo $k$. Let's see what this means:

• ​​If we choose degree $k=1$ (or $k=-1$):​​ The relation is $a^1=1$. This means our original loop generator is now trivial. The hole has been perfectly plugged! The space becomes ​​simply connected​​ (its fundamental group is the trivial group $\{0\}$). The resulting space is in fact contractible, meaning it is homotopy equivalent to a point. While a 2-sphere, $S^2$, is also simply connected, it is not contractible; our construction yields a space homeomorphic to a 2-disk, $D^2$.

• ​​If we choose degree $k=2$:​​ The relation becomes $a^2=1$. The loop isn't gone. But now, wrapping around it twice is the same as not moving at all. This creates a "2-torsion" element in the group, $\mathbb{Z}/2\mathbb{Z}$. The space we have just built is none other than the famous ​​real projective plane​​, $\mathbb{R}P^2$, a one-sided surface that cannot be embedded in 3D space without self-intersecting! The fact that its fundamental group is $\mathbb{Z}/2\mathbb{Z}$ is one of its defining characteristics, and we have just derived it from first principles by knowing its construction involves a degree-2 attaching map.

• ​​If we choose degree $k=0$:​​ The relation is $a^0=1$, which simply means $1=1$. This adds no constraint at all. The original loop $a$ survives untouched. The fundamental group remains $\mathbb{Z}$. Geometrically, we've attached a sphere (a disk with its boundary pinched to a point) to our circle at a single point. The final space is the ​​wedge sum​​ $S^1 \vee S^2$, and its first Betti number, which counts one-dimensional holes, is still 1.

Sculpting Homology

The effect on homology is just as direct and elegant. In the machinery of ​​cellular homology​​, the boundary map $d_2$ that goes from the group of 2-cells to the group of 1-cells is simply ​​multiplication by the degree of the attaching map​​.

For our space $X_k$ built from one cell in each dimension 0, 1, and 2, the chain complex looks like:

Which, in this case, is just:

The first homology group, $H_1$, which detects 1-dimensional "loop" holes, is calculated as $\ker(d_1) / \operatorname{im}(d_2)$. Here, $\ker(d_1) = \mathbb{Z}$ and $\operatorname{im}(d_2)$ is the subgroup of integers that are multiples of $k$, written $k\mathbb{Z}$.

The result is beautifully clear. The order of the "torsion" in the first homology group is precisely the absolute value of the degree of the attaching map. If we construct a space using an attaching map of degree $k=5$, like the map $\phi(z)=-z^5$, we can immediately predict that its first homology group will be $\mathbb{Z}/5\mathbb{Z}$. This algebraic invariant directly reads out the geometric "twist" of the construction.

What if our sculpture becomes more complex? Suppose we attach two 2-cells to our circle, one with a degree $m$ map, $\phi_a$, and another with a degree $n$ map, $\phi_b$?.

Our algebraic machinery handles this with grace. Now, the group of 2-chains, $C_2$, is $\mathbb{Z}^2$, with one generator for each 2-cell. The boundary map $d_2: \mathbb{Z}^2 \to \mathbb{Z}$ sends the first generator to $m$ and the second to $n$. The image of this map—the set of all resulting boundaries—is the set of all integer linear combinations of $m$ and $n$, which is a subgroup of $\mathbb{Z}$ generated by their ​​greatest common divisor​​, $\gcd(m, n)$.

So, the first homology group becomes:

If we attach one cell with degree 5 and another with degree 15, the resulting "hole" in the space is of order $\gcd(5, 15)=5$. It’s as if the 15-fold wrapping is already accounted for by three 5-fold wrappings, so the only fundamentally new constraint is the 5-fold one.

This same principle echoes through to the "dual" theory of ​​cohomology​​. The coboundary map $\delta^1$, which plays a parallel role to the boundary map, is simply represented by the transpose of the boundary map's matrix. For our two-cell example with degrees 2 and 3, the boundary map $\partial_2$ is represented by the matrix $\begin{pmatrix} 2 & 3 \end{pmatrix}$, and the coboundary map $\delta^1$ is beautifully represented by its transpose, $\begin{pmatrix} 2 \\ 3 \end{pmatrix}$.

From a simple, intuitive act of gluing, a rich algebraic structure emerges. The degree of the attaching map is the bridge between these two worlds, a single number that translates a geometric twist into an algebraic relation. It is a stunning example of the power and elegance of modern mathematics, turning the art of building shapes into a precise and predictive science.

We have now acquainted ourselves with the machinery of attaching cells and the crucial role played by the degree of the attaching map. At first glance, this might seem like a rather abstract game of topological surgery, a collection of rules for gluing shapes together. But to leave it there would be to miss the entire point. This machinery is not just a descriptive tool; it is a creative one. It is a lens through which we can understand the deep structure of the world, and a toolkit with which we can, in a mathematical sense, build universes. The integer degree of an attaching map, a single number, turns out to be a master key unlocking secrets across a surprising spectrum of scientific thought. Let's see how.

Perhaps the most direct and powerful application of our new tool is in construction. If you were an architect of spaces, how would you build one with a specific, pre-determined property? Suppose you want to create a world where any path that winds around a certain loop exactly $n$ times magically becomes equivalent to a path that didn't move at all. You are, in essence, trying to build a space whose fundamental group is the finite cyclic group, $\mathbb{Z}_n$.

The recipe is astonishingly simple. You start with a simple loop, a circle $S^1$. Its fundamental group is the integers, $\mathbb{Z}$, representing endless, distinct winding numbers. To force an $n$-fold wrap to become trivial, we must "fill it in." We do this by taking a 2-dimensional disk and gluing its boundary circle onto our starting circle. The crucial step is how we glue it. If we attach the disk's boundary by wrapping it around the circle exactly $n$ times, we are declaring, by construction, that a path of "degree $n$" is now the boundary of something, and is therefore contractible. The Seifert-van Kampen theorem confirms our intuition: this single act of gluing a 2-cell with an attaching map of degree $n$ collapses the infinite group of windings $\mathbb{Z}$ into a finite one, leaving us precisely with $\mathbb{Z}_n$. This is a beautiful piece of topological engineering: the algebraic property we desire (a group of order $n$) is directly encoded in the geometric action of wrapping (a map of degree $n$).

This principle extends beyond the fundamental group to the subtler world of homology. Homology groups detect "holes" of various dimensions. When we attach a 2-cell to a circle with a map of degree $k$, we are not quite filling in the 1-dimensional hole. Instead, we are creating a relationship. The original 1-cycle (the circle) is no longer completely free. While it doesn't bound anything itself, $k$ times this cycle does now bound the 2-cell we attached. This creates what is known as ​​torsion​​ in the first homology group. The resulting group, $H_1$, is $\mathbb{Z}_k$, a group whose only non-trivial element has order $k$.

A classic and profound example of this is the ​​real projective plane​​, $\mathbb{R}P^2$. This is the strange, non-orientable surface you get if you take a sphere and identify every point with its exact opposite, its antipode. How can we build this space with our cells? It turns out you can construct it with one cell in each dimension 0, 1, and 2. The 1-skeleton is a circle. The 2-cell is attached in a way that corresponds to identifying the opposite points on its boundary, and the map that achieves this has degree 2. The immediate consequence, as our theory predicts, is that the first homology group must be $\mathbb{Z}_2$. This tiny algebraic fact, $H_1(\mathbb{R}P^2; \mathbb{Z}) \cong \mathbb{Z}_2$, is the algebraic fingerprint of the non-orientability of the projective plane—the reason a two-dimensional painter living on its surface could return to their starting point to find their image mirrored. The same principle works in higher dimensions; attaching a 3-cell to a 2-sphere with a map of degree $m$ doesn't kill the 2-sphere, but it introduces an $m$-torsion element into the second homology group, $H_2 \cong \mathbb{Z}_m$.

The game becomes even more interesting when we attach multiple cells. Imagine we attach two 2-cells to a circle, one with a map of degree $m$ and the other with degree $n$. The 1-cycle of our circle is now constrained in two ways. The set of all cycles that can be "filled in" are those that are integer combinations of an $m$-wrap and an $n$-wrap. From elementary number theory, we know that the set of all numbers of the form $am+bn$ is precisely the set of all multiples of the greatest common divisor, $\gcd(m,n)$. And so, the first homology group of this new space is $\mathbb{Z}_{\gcd(m,n)}$. Topology, it seems, can perform number theory!

We can even use this to play a game of reverse-engineering. The 2-sphere, $S^2$, is simply connected ($\pi_1=0$) and has trivial first homology ($H_1=0$). Could we build something with these same properties—a "homotopy sphere"—from different parts? Let's try attaching two 2-cells to a circle, with degrees 2 and $k$. For the resulting space to have $H_1=0$, we need the group $\mathbb{Z}_{\gcd(2,k)}$ to be trivial. This happens if and only if $\gcd(2,k)=1$. So, any odd integer $k > 1$ will do. The simplest choice is $k=3$. By attaching two disks to a circle, with winding numbers 2 and 3 respectively, we have constructed a space that, from the perspective of homotopy and homology, is indistinguishable from a 2-sphere. We have forged a sphere from the most unlikely of scraps.

The true beauty of a fundamental concept is revealed when it echoes in seemingly unrelated fields. The degree of an attaching map is one such concept, appearing as a cornerstone in differential geometry, cosmology, and abstract algebra.

​​The Shape of the Cosmos:​​ Cosmologists entertain the possibility that our universe may have a non-trivial, finite topology. One of the simplest families of candidates for such a 3-dimensional universe is the ​​lens spaces​​, $L(p,q)$. These spaces can be described as quotients of the 3-sphere, but they can also be built from the ground up as a CW complex with just one cell in each dimension from 0 to 3. When we do this, the fundamental group $\pi_1(L(p,q))$ is found to be $\mathbb{Z}_p$. As our architectural intuition now tells us, this must mean that the 2-cell is attached to the 1-skeleton (a circle) with a map of degree $p$. This is a breathtaking connection. A single integer, the degree of a 2-cell attachment, dictates the fundamental topology of a potential universe. If we lived in such a universe, looking far enough in one direction might let us see the back of our own head, and the number of "ghost images" of distant galaxies would be related to this integer $p$.

​​The Landscape of Functions:​​ In ​​Morse theory​​, we analyze the topology of a smooth manifold by studying a function on it, like height on a hilly terrain. The critical points of the function—the pits (index 0), passes (index 1), and peaks (index 2 for a surface)—correspond to the cells of a CW complex. A minimal Morse function on $\mathbb{R}P^2$, one with the fewest possible critical points, will have one of each type. Morse theory tells us that the way these cells are "attached" to one another is determined by the gradient flow of the function. Because the resulting complex must have the topology of $\mathbb{R}P^2$, its first homology must be $\mathbb{Z}_2$. This forces the boundary map from the 2-cell to the 1-cell to be multiplication by 2. Therefore, any minimal Morse function on the real projective plane must induce a cell attachment of degree 2. The global topology of the manifold dictates the local behavior of smooth functions on it.

​​The Geometry of Algebra:​​ Abstract algebra defines groups by generators and relations, like $\langle a \mid a^p=1, a^q=1 \rangle$. This abstract entity can be given a physical form, the ​​presentation complex​​. We take a wedge of circles, one for each generator (here, just one circle for $a$), and attach a 2-cell for each relation, with the attaching map tracing the path of the relator word. For our example group, we attach two 2-cells to a circle, one with degree $p$ and one with degree $q$. If $p$ and $q$ are coprime, then $\gcd(p,q)=1$, and the resulting complex is simply connected. The Hurewicz theorem then provides a stunning bridge: for such a space, the second homology group $H_2(X_G)$ is isomorphic to the second homotopy group $\pi_2(X_G)$. A calculation reveals that $H_2(X_G)$ is isomorphic to $\mathbb{Z}$, which means the space has a non-trivial 2-dimensional "sphere-like" quality. An abstract algebraic statement about coprime numbers has been translated into a geometric statement about how spheres can be mapped into our constructed space.

From engineering custom-made spaces to modeling the cosmos and giving form to abstract equations, the simple idea of an attaching map and its degree proves to be an exceptionally powerful and unifying concept. It is a perfect illustration of the mathematical ideal: a simple, elegant idea that ripples outward, connecting disparate worlds and revealing the deep, structural beauty that underlies them all.