Sequentially Compact: Definition, Examples, and Key Theorems Explained
Sequential compactness is one of those topology terms that sounds abstract until you see it working behind the scenes in calculus, analysis, and optimization. At its core, it asks whether every infinite sequence in a space has a subsequence that settles down to a point inside that same space. That simple question turns out to be a powerful way to detect hidden order in sets that might otherwise seem too wild to control.
For students meeting the concept for the first time, the definition can feel like a cousin of ordinary compactness. In metric spaces, the two ideas are actually equivalent, which is why sequential compactness often appears in proofs of the Bolzano-Weierstrass theorem, existence results for extrema, and convergence arguments. In more general topological spaces, however, compactness and sequential compactness can part ways, making the distinction important for anyone studying modern topology.
The Definition and Intuition Behind Sequential Compactness
Sequential compactness is a property of a topological space that controls the behavior of infinite sequences. The definition is elegantly simple: a space X is sequentially compact if every sequence of points in X has a subsequence that converges to a point in X. That means no matter how erratically a sequence jumps around, there is always some thread within it that settles down to a limit inside the space.
This property is closely related to the idea of compactness, but it is not just a rewording. Compactness talks about open covers, while sequential compactness talks about sequences. The two perspectives often agree in familiar settings such as Euclidean spaces, but they can disagree in more exotic topological spaces. Understanding that difference is essential for moving from calculus-level analysis into general topology.
What It Means for Every Sequence to Have a Convergent Subsequence
Formally, let X be a topological space. X is sequentially compact if for every sequence (x_n) with x_n in X, there exists a subsequence (x_{n_k}) and a point x in X such that x_{n_k} converges to x. Convergence here is topological: every neighborhood of x contains all but finitely many terms of the subsequence. In a metric space, this is equivalent to the distance d(x_{n_k}, x) tending to zero.
A useful way to think about the definition is as an anti-escape condition. A sequence might try to run away to infinity, oscillate forever, or accumulate only at points outside the space. Sequential compactness forbids all of those outcomes. For example, the closed interval [0,1] is sequentially compact, while the open interval (0,1) is not because the sequence 1, 1/2, 1/3, ... converges only to 0, which is not in (0,1).
Subsequences, Cluster Points, and the Role of the Ambient Space
A cluster point of a sequence is a point that every neighborhood visits infinitely often. In metric spaces, a sequence has a convergent subsequence exactly when it has a cluster point in the space. That equivalence makes sequential compactness easier to visualize: every sequence must cluster somewhere inside X. In general topological spaces, cluster points and convergent subsequences can behave differently, which is one reason sequential compactness is studied alongside other compactness notions.
The ambient space matters enormously. The same sequence 1/n has a convergent subsequence in [0,1] but not in (0,1), because the limit point 0 is missing from the smaller space. Similarly, a sequence in a discrete space can only converge if it is eventually constant. This sensitivity to the underlying set and topology is why mathematicians always specify the space when discussing sequential compactness.
Sequential Compactness vs. Compactness
Compactness is usually introduced through open covers: a space is compact if every open cover has a finite subcover. Sequential compactness uses sequences instead. On the surface, these definitions seem to describe different things, and in general topological spaces they really are different. In metric spaces, however, they coincide, which is one of the most useful theorems in analysis.
The relationship is subtle because compactness is a stronger, more set-theoretic condition in general, while sequential compactness is more countable and sequence-oriented. Neither property implies the other in arbitrary topological spaces. That fact surprises many students, especially those who first meet compactness in the comfortable setting of Euclidean space.
Metric Spaces: Where Compactness and Sequential Compactness Coincide
In a metric space, compactness and sequential compactness are equivalent. This theorem is foundational. It means that for metric spaces, you can prove compactness by showing every sequence has a convergent subsequence, and vice versa. The proof typically uses total boundedness and completeness: a metric space is compact if and only if it is complete and totally bounded, and those two conditions also characterize sequential compactness.
This equivalence powers many classical results. The Bolzano-Weierstrass theorem says every bounded sequence in R^n has a convergent subsequence, which is exactly the statement that closed and bounded subsets of R^n are sequentially compact. In a compact metric space, continuous functions are uniformly continuous, attain extrema, and behave predictably on sequences. These results depend on the bridge between open covers and convergent subsequences.
General Topological Spaces: Where the Concepts Diverge
Outside the metric setting, the equivalence breaks down. A compact space need not be sequentially compact. A standard example is the product space [0,1]^[0,1], which is compact by Tychonoff's theorem but contains sequences with no convergent subsequence. Such spaces can be so large that a single sequence cannot explore enough of the space to force a limit.
Conversely, a sequentially compact space need not be compact. The ordinal space [0, omega_1) with the order topology is sequentially compact but not compact. Every sequence of countable ordinals is bounded below omega_1 and has a convergent subsequence, yet the open cover consisting of initial segments [0, alpha) for alpha < omega_1 has no finite subcover. This example shows that sequences are not always enough to capture the full strength of compactness.
Examples That Build Intuition
Examples are the best way to internalize sequential compactness. Simple intervals in the real line reveal the importance of closedness and boundedness. More exotic spaces show why the concept cannot be reduced to a single intuitive picture. Working through both compact and non-compact cases helps clarify what the definition demands.
A good strategy is to test sequences that try to escape through a missing endpoint, jump between infinitely many isolated points, or wander in a large product space. Each failure of sequential compactness has a distinct flavor, and each success reveals a different mechanism that forces a convergent subsequence.
Compact and Sequentially Compact: The Closed Interval [0,1]
The closed interval [0,1] is the prototypical sequentially compact space. Any sequence in [0,1] is bounded, so the Bolzano-Weierstrass theorem guarantees a convergent subsequence. Because [0,1] is closed, the limit of that subsequence must remain in [0,1]. Thus every sequence has a subsequence converging inside the space.
This example also illustrates why both closedness and boundedness matter in Euclidean space. The sequence 1/2, 1/3, 1/4, ... lies in (0,1] and converges to 0, which is outside the space. Even though the sequence is bounded, the missing endpoint prevents sequential compactness. In [0,1], the same sequence has a subsequence converging to 0, which is included, so the definition is satisfied.
Non-Examples: Open Intervals and Infinite Discrete Spaces
The open interval (0,1) is not sequentially compact. The sequence 1/n has no subsequence converging to a point in (0,1) because its only limit in the real line is 0. Similarly, the sequence 1 - 1/n converges only to 1, which is also missing. These examples show that removing limit points can destroy sequential compactness.
An infinite discrete space is another non-example. In the discrete topology, a sequence converges only if it is eventually constant. A sequence that lists infinitely many distinct points has no constant subsequence and therefore no convergent subsequence. So an infinite discrete space is never sequentially compact, even though every finite discrete space is sequentially compact.
Why Sequential Compactness Matters
Sequential compactness matters because it guarantees the existence of limits, and limits are the foundation of analysis. When a space is sequentially compact, you can take a sequence that approximates a solution and extract a convergent subsequence whose limit is also in the space. This is the engine behind many existence theorems in mathematics.
The concept also helps organize the many notions of compactness in topology. It sits near countable compactness, limit point compactness, and compactness itself, and understanding its relationships sharpens your intuition about how topology behaves when sequences are no longer sufficient.
Applications in Analysis and Optimization
In analysis, sequential compactness is the hidden assumption behind the extreme value theorem. A continuous real-valued function on a sequentially compact space attains its maximum and minimum: take a sequence approaching the supremum, extract a convergent subsequence, and use continuity to show the limit achieves the supremum. This argument works in metric spaces and many other settings.
In optimization and the calculus of variations, sequential compactness appears in the direct method. A minimizing sequence is chosen, and one hopes to extract a convergent subsequence whose limit is the minimizer. In infinite-dimensional problems, compactness may be too strong, so mathematicians use weak sequential compactness. Even then, the same basic idea applies: control the sequence, extract a convergent subsequence, and pass to the limit.
Connections to Countable Compactness and Limit Point Compactness
Sequential compactness implies countable compactness, which says that every countable open cover has a finite subcover. Equivalently, in many spaces, every sequence has a cluster point. In first-countable spaces, countable compactness and sequential compactness are equivalent. Metric spaces are first-countable, which helps explain why the two notions merge there.
Limit point compactness is another relative: every infinite subset has a limit point. In T1 spaces, limit point compactness is equivalent to countable compactness. In metric spaces, all of these properties—compactness, sequential compactness, countable compactness, and limit point compactness—are equivalent. In general topology, they form a family of related but distinct conditions, and sequential compactness is one of the most useful when sequences are the main tool.
From the familiar Bolzano-Weierstrass theorem to the wilder reaches of ordinal spaces and uncountable products, sequential compactness offers a precise language for asking whether a space keeps its sequences under control. The next time a proof calls for a convergent subsequence, look for the compactness assumption nearby; it is often the quiet reason the argument works.