The fundamental concepts of open and closed sets are central to the study of real analysis and topology, providing the foundational language for describing the structure of subsets of the real line and higher-dimensional spaces. Understanding these definitions, along with their characterizations in terms of interior points, boundary points, and closure, is essential for rigorously proving theorems about continuity, convergence, and compactness. This article explores these core definitions, their logical equivalences, and their practical implications for classifying common sets like intervals, providing a clear and structured overview for students and practitioners of mathematical analysis.
Fundamental Definitions and Intuitive Understanding
In the context of subsets of the real numbers (\mathbb{R}) or (\mathbb{R}^n), the notions of "open" and "closed" are precise technical terms that do not always align with the everyday meanings of the words. The most critical point in this section is to understand the definitions of open and closed sets and to develop an intuitive feel for what these sets are like. This requires a grasp of the notions of boundary, interior, and closure. While a solid definition using ((\varepsilon, \delta)) arguments is necessary for rigorous proofs, many sets can be recognized as open or closed simply by the nature of their description without always needing to appeal to these arguments.
An open set is defined as a set where every point is an interior point. For a point (\mathbf{a}) to be an interior point of a set (S \subseteq \mathbb{R}^n), there must exist an open ball (B(\mathbf{a}; r)) centered at (\mathbf{a}) with radius (r > 0) that is entirely contained within (S). In the one-dimensional case ((\mathbb{R})), this translates to the requirement that for any (x \in S), there exists an interval ((x - \delta, x + \delta)) that is a subset of (S) for some (\delta > 0). For example, an open interval ((a, b)) is an open set in (\mathbb{R}). This is also true for intervals of the form ((a, \infty)) or ((-\infty, b)). Every open ball (B(\mathbf{a}; r)) in (\mathbb{R}^n) is itself an open set.
A closed set is defined as a set that contains all of its boundary points. The boundary of a set (S), denoted (\partial S), consists of points where every open ball centered at the point intersects both (S) and its complement (S^c). A set (S) is closed if and only if (\partial S \subseteq S). In (\mathbb{R}), a closed interval ([a, b]) is a closed set. Other examples include single points and the entire space (\mathbb{R}) itself.
Characterizations via Interior, Boundary, and Closure
The definitions of open and closed sets are closely intertwined with the concepts of interior, boundary, and closure. The interior of a set (S), denoted (S^{int}), is the set of all interior points of (S). The closure of a set (S), denoted (\overline{S}), is the union of its interior and its boundary: (\overline{S} = S^{int} \cup \partial S).
There is an important equivalence that provides an alternate characterization of open sets: A set (S) is open if and only if (S = S^{int}). The proof is straightforward. If (S) is open, then every point of (S) is an interior point, which implies (S \subseteq S^{int}). Conversely, we know from standard theorems that (S^{int} \subseteq S). Combining these gives (S = S^{int}). Conversely, if every point of (S) is an interior point (i.e., (S \subseteq S^{int})), and since (S^{int} \subseteq S), we conclude (S = S^{int}), meaning (S) is open.
Similarly, there is an alternate characterization for closed sets: A set (S) is closed if and only if (\partial S \subseteq S). The proof begins with the definition: if (S) is closed, then (S = \overline{S} = S^{int} \cup \partial S), which certainly implies (\partial S \subseteq S). Conversely, if we assume (\partial S \subseteq S), and we know (S^{int} \subseteq S), then their union (\overline{S} = S^{int} \cup \partial S) is a subset of (S). Recalling that (S \subseteq \overline{S}) is always true, it follows that (\overline{S} = S), and hence (S) is closed.
An important duality exists: the complement of an open set is closed, and the complement of a closed set is open. Specifically, (S) is closed if and only if its complement (S^c) is open. This follows from the fact that the boundary of (S) is the same as the boundary of (S^c) ((\partial S = \partial S^c)), and the condition that (S) is closed ((\partial S \subseteq S)) is equivalent to the condition that no point of (S^c) is a boundary point, which is the definition of (S^c) being open.
Properties of Open and Closed Sets
The class of open sets and the class of closed sets in (\mathbb{R}^n) have well-defined algebraic properties that are fundamental to analysis.
Properties of Open Sets: - The empty set (\emptyset) and the entire space (\mathbb{R}^n) are open. - The union of any collection of open sets is open. This is a key property: if (G = \bigcup{\alpha \in I} G{\alpha}) where each (G{\alpha}) is open, then for any point (a \in G), there exists some (\alpha0) with (a \in G{\alpha0}). Since (G{\alpha0}) is open, there is a ball (B(a; \delta)) contained in (G{\alpha0}), and hence in (G), proving (G) is open. - The intersection of a finite number of open sets is open. If (G1, ..., Gn) are open and (a \in \bigcap{i=1}^n Gi), then for each (i), there is a (\deltai > 0) such that (B(a; \deltai) \subset Gi). Letting (\delta = \min{\delta1, ..., \delta_n} > 0), the ball (B(a; \delta)) is contained in the intersection. Note that this does not necessarily hold for infinite intersections.
Properties of Closed Sets: - The empty set (\emptyset) and the entire space (\mathbb{R}^n) are closed. - The intersection of any collection of closed sets is closed. If (S = \bigcap{\alpha \in I} S{\alpha}) where each (S{\alpha}) is closed, then the complement (S^c = \bigcup{\alpha \in I} S{\alpha}^c) is a union of open sets (since each (S{\alpha}^c) is open), and thus (S^c) is open, implying (S) is closed. - The union of a finite number of closed sets is closed. This follows from the de Morgan's law: the complement of a finite union is the intersection of the complements, which is open.
These properties allow for the quick identification of many sets as open or closed based on their description. For instance, any finite set in (\mathbb{R}) is closed, as it can be expressed as the finite union of single points, each of which is closed.
Classification of Common Sets in (\mathbb{R})
Applying these definitions and properties allows for the classification of various subsets of the real line.
- Open Intervals: The set ((a, b) = {x \in \mathbb{R} : a < x < b}) is open. For any (x \in (a, b)), choosing (\delta = \min{x-a, b-x} > 0) gives the interval ((x-\delta, x+\delta) \subset (a, b)). Similarly, ((a, \infty)) and ((-\infty, b)) are open.
- Closed Intervals: The set ([a, b] = {x \in \mathbb{R} : a \leq x \leq b}) is closed. Its boundary is ({a, b}), which is a subset of ([a, b]). The set (\mathbb{R}) itself is both open and closed (clopen), as is the empty set (\emptyset).
- Half-Open Intervals: The interval ([0, 1)) is neither open nor closed. It is not open because the point (0) is in the set but is not an interior point (any open ball around 0 contains negative numbers not in ([0, 1))). It is not closed because the point (1) is a boundary point of ([0, 1)) (every neighborhood of 1 intersects both ([0, 1)) and its complement) but (1 \notin [0, 1)).
- Sets Defined by Inequalities: The set (A = (0, 1)) is open. Its limit points are ([0, 1]), and it has no isolated points. The set (B = [0, 1)) has limit points ([0, 1]) and an isolated point at 0? (Note: 0 is a limit point because every neighborhood contains points of B other than 0). The set of rational numbers (\mathbb{Q}) is neither open nor closed in (\mathbb{R}); its closure is (\mathbb{R}), and its boundary is (\mathbb{R}).
Compactness and Further Concepts
While not the primary focus of the definitions of open and closed, these concepts are essential for understanding compactness. A set (A \subseteq \mathbb{R}) is compact if every open cover of (A) has a finite subcover. An equivalent definition for (\mathbb{R}^n) is sequential compactness: every sequence in (A) has a subsequence that converges to a point in (A). A fundamental theorem states that a subset of (\mathbb{R}^n) is compact if and only if it is closed and bounded. For example, the interval ([0, 1]) is compact, while ((0, 1)) and ([0, \infty)) are not. The union of two compact sets is compact, and the intersection of any collection of compact sets is compact.
Conclusion
The rigorous definitions of open and closed sets, based on interior and boundary points, provide the essential framework for real analysis. Understanding that a set is open if it equals its interior and closed if it contains its boundary allows for precise classification. The algebraic properties of unions and intersections of open and closed sets are powerful tools for constructing and identifying these sets. While concepts like compactness build upon these foundations, the core ideas of openness and closedness remain the bedrock for describing the topological structure of subsets of (\mathbb{R}) and (\mathbb{R}^n).