{
  "id": "euclidean-geometry",
  "title": "Euclidean Geometry",
  "lang": "en",
  "source_lang": "en",
  "summary": "Euclidean geometry studies figures and spatial relationships in flat spaces through axioms, constructions, distance, angle, and deductive proof.",
  "category": "mathematics",
  "subcategory": "geometry",
  "type": "concept",
  "aliases": [
    "Euclidian geometry",
    "Euclidean plane and solid geometry"
  ],
  "generated_at": "2026-11-02",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/euclidean-geometry/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/euclidean-geometry/",
      "json": "https://aiwiki.page/zh/euclidean-geometry.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Euclidean geometry is the branch of geometry that studies points, lines, planes, and figures in flat space, with relationships governed by distance, angle, incidence, and parallelism. Its classical form concerns the plane and three-dimensional space; its modern formulation extends to Euclidean spaces of any finite dimension. It is named after Euclid, whose Elements, compiled around 300 BCE, organized much of ancient Greek mathematics into a deductive system.\n\nThe subject connects elementary constructions with some of the deepest developments in mathematics. The angle sum of a triangle, the properties of circles, the classification of regular solids, and the measurement of area and volume belong to its classical core. Coordinate methods, vector spaces, and transformation groups provide modern descriptions of the same structures. The investigation of its parallel postulate helped establish that other logically coherent geometries are possible.\n\nEuclidean geometry is both an abstract mathematical theory and a model used in measurement, design, and physical science. Its mathematical validity follows from specified assumptions and proofs, whereas its suitability for describing a physical situation depends on scale and required accuracy."
    },
    {
      "heading": "Historical formation and transmission",
      "text": "Practical geometrical knowledge preceded Euclid by many centuries. Mesopotamian mathematical texts contain procedures for finding areas, volumes, and lengths, including numerical relationships associated with right triangles. Egyptian sources, notably the Rhind Mathematical Papyrus, copied around 1650 BCE, include rules for surveying and mensuration. Such traditions supplied effective procedures, but their surviving texts generally do not present geometry as an extended sequence of propositions derived from explicit axioms.\n\nGreek mathematicians increasingly treated geometrical assertions as statements requiring proof. Traditions connect early developments with Thales and Pythagoras, although evidence for particular discoveries by these figures is often indirect. More securely documented contributions include the theory of proportion associated with Eudoxus and investigations of conic sections and solids in the fourth century BCE. The discovery of incommensurable magnitudes made it necessary to distinguish geometrical ratios from ratios of whole numbers.\n\nEuclid’s Elements brought together earlier results in thirteen books. Books I–IV develop plane geometry, Book V treats proportion, and Book VI applies proportion to similar figures. Books VII–IX concern arithmetic and number theory; Book X studies incommensurable magnitudes; Books XI–XIII develop solid geometry, including the five regular convex polyhedra. Consequently, the Elements is broader than what modern school curricula usually call Euclidean geometry.\n\nOther ancient authors extended the subject beyond this organization. Archimedes established results on areas, volumes, centers of gravity, and equilibrium. His work on the sphere and cylinder and on the quadrature of a parabolic segment displayed the power of geometrical reasoning about limiting processes. Apollonius developed a systematic theory of the conic sections, including ellipses, parabolas, and hyperbolas.\n\nThe text and methods of the Elements circulated through Greek, Arabic, and Latin scholarly traditions. Arabic translations and commentaries supported investigations of proportion, constructions, and the parallel postulate. Medieval Latin translations, including versions associated with Adelard of Bath and Campanus of Novara, helped establish the work in European education. The first printed edition appeared in Venice in 1482.\n\nDuring the seventeenth century, coordinate methods and calculus transformed the range of problems geometry could address. In the nineteenth century, alternative geometries and stricter standards of axiomatization changed its foundational status. Euclidean geometry ceased to be regarded as the only logically possible geometry, but remained a central theory of flat space and a major source of mathematical problems and methods."
    },
    {
      "heading": "Primitive objects, axioms, and deductive structure",
      "text": "A geometrical theory must specify which objects it discusses, which relationships are fundamental, and what assumptions govern them. Euclid began with definitions, five postulates, and common notions. His descriptions of a point as having no part and a line as having length without breadth express an idealization: geometrical objects are not physical marks with measurable thickness. Modern treatments usually regard points, lines, and certain relations as primitive rather than attempting to define every term.\n\nEuclid’s first three postulates authorize drawing a straight segment between two points, extending a finite straight line, and drawing a circle with a given center and radius. The fourth states that all right angles are equal. The fifth concerns the meeting of two lines when a transversal makes the interior angles on one side sum to less than two right angles. Common notions include principles such as “things equal to the same thing are equal to one another.”\n\nAn axiom is not merely a statement considered visually obvious. Within an axiomatic theory, it is an assumption from which consequences are derived by accepted rules of inference. A theorem therefore depends both on its stated hypotheses and on the underlying axiom system. A diagram can indicate a useful construction or relationship, but appearance alone does not establish that two lengths are equal or that three lines meet at one point.\n\nEuclid’s presentation leaves some assumptions implicit. For example, the construction of an equilateral triangle on a segment uses two circles whose intersection is apparent in the diagram, but the original postulates do not explicitly state an adequate circle-intersection principle. Arguments about points lying between other points, lines separating a plane, and figures being moved without distortion also require more precise treatment than the ancient text provides.\n\nLater axiomatizations distinguish several kinds of structure. Incidence axioms describe which points lie on which lines or planes. Order axioms govern betweenness and separation. Congruence axioms formalize equality of segments and angles. Parallel axioms determine the distinctive behavior of Euclidean lines. Continuity assumptions supply the completeness needed for the usual real-coordinate interpretation.\n\nThese distinctions clarify why apparently simple statements can require substantial foundations. Two distinct points determine a unique line, but that assertion alone says nothing about length or perpendicularity. An affine space supplies incidence and parallelism without a preferred distance. Euclidean geometry adds metric structure, allowing length and angle to be defined consistently with the underlying affine relationships."
    },
    {
      "heading": "Parallelism and its consequences",
      "text": "The parallel postulate distinguishes Euclidean geometry from other geometries that share many of its elementary properties. Euclid’s original formulation describes when two lines must meet. A more familiar equivalent, commonly called Playfair’s axiom, states that through a point not on a given line there passes exactly one line in the same plane that does not meet the given line. Their equivalence is established within an appropriate background of the remaining geometrical axioms.\n\nIn the Euclidean plane, distinct lines either meet at one point or are parallel. This statement changes in three dimensions, where two lines can be skew: they neither intersect nor lie in a common plane. Coplanarity is therefore essential to the usual definition of parallel lines. Parallel planes likewise remain disjoint, while a line can be parallel to a plane without lying in it.\n\nOne basic consequence of Euclidean parallelism is the angle-sum theorem. Through one vertex of a triangle, draw the parallel to the opposite side. Alternate-angle equalities identify the other two interior angles with angles along that parallel, producing\n\n\\[\\alpha+\\beta+\\gamma=\\pi,\\]\n\nwhen angles are measured in radians. The argument depends on the relevant parallel-line theorems, rather than merely on the appearance of a triangular drawing.\n\nParallel lines also underlie proportionality results. If a line through two sides of a triangle is parallel to the third side, it cuts the sides in equal ratios. Such results support similarity, the theory of figures with equal corresponding angles and proportional corresponding lengths. The existence of similar triangles of different sizes is a characteristic Euclidean phenomenon; it does not carry over unchanged to curved geometries.\n\nRectangles provide another important connection. In standard neutral-geometry frameworks, the existence of a rectangle is equivalent to the Euclidean parallel postulate. Likewise, the assertion that every triangle has angle sum equal to two right angles can serve as an equivalent Euclidean condition under suitable background assumptions. These equivalences explain why historical attempts to prove the fifth postulate often unknowingly substituted another statement of comparable strength.\n\nParallelism also supports Cartesian coordinate grids and the familiar properties of parallelograms. Opposite sides of a parallelogram are equal, and its diagonals bisect each other. From the modern affine viewpoint, these facts express the consistency of vector addition. From the classical viewpoint, they arise through congruent triangles and angle relationships created by parallel lines."
    },
    {
      "heading": "Distance, angles, and triangle geometry",
      "text": "Length and angle supply the metric content of Euclidean geometry. A segment has a nonnegative length, equal to zero only when its endpoints coincide. Distances are symmetric and satisfy the triangle inequality:\n\n\\[d(A,C)\\leq d(A,B)+d(B,C).\\]\n\nFor three distinct noncollinear points, the inequality is strict. Equality occurs when the intermediate point lies on the segment joining the other two points. This expresses the minimizing property of straight segments.\n\nA triangle is determined, up to congruence, by several standard collections of data. The side–side–side and side–angle–side criteria are sufficient, as are two angles and a corresponding side. Two sides and a nonincluded angle need not determine a unique triangle: in some configurations there are two possibilities. This ambiguity explains why an unrestricted side–side–angle congruence criterion is invalid.\n\nThe Pythagorean theorem states that a right triangle with legs \\(a,b\\) and hypotenuse \\(c\\) satisfies\n\n\\[a^2+b^2=c^2.\\]\n\nIts converse identifies a right angle from this equality. The theorem connects geometric perpendicularity with arithmetic and is the foundation of the Cartesian distance formula. Although associated with Pythagoras in later tradition, particular numerical instances were known in older mathematical cultures.\n\nTrigonometry extends right-triangle ratios to general triangles. For a triangle whose sides \\(a,b,c\\) lie opposite angles \\(A,B,C\\), the cosine rule is\n\n\\[c^2=a^2+b^2-2ab\\cos C.\\]\n\nThe sine rule gives\n\n\\[\\frac{a}{\\sin A} = \\frac{b}{\\sin B} = \\frac{c}{\\sin C} = 2R,\\]\n\nwhere \\(R\\) is the radius of the circumcircle. These relations determine unknown lengths and angles and connect synthetic geometry with periodic mathematical functions.\n\nTriangle geometry also studies distinguished lines and centers. The three medians meet at the centroid, which divides each median in the ratio \\(2:1\\), measured from the vertex. Perpendicular bisectors meet at the circumcenter; internal angle bisectors meet at the incenter; altitudes meet at the orthocenter. Their locations depend on triangle shape: for example, an obtuse triangle has its circumcenter and orthocenter outside the triangle.\n\nMany classical results arise by combining these structures. The centroid, circumcenter, and orthocenter lie on the Euler line for a non-equilateral triangle. Heron’s formula expresses area entirely through side lengths:\n\n\\[K=\\sqrt{s(s-a)(s-b)(s-c)}, \\qquad s=\\frac{a+b+c}{2}.\\]\n\nSuch formulas demonstrate that apparently different descriptions—angles, distances, centers, and areas—are tightly constrained by the same Euclidean metric."
    },
    {
      "heading": "Circles, tangency, and conic sections",
      "text": "A circle is the set of points in a plane at a fixed positive distance \\(r\\) from a center \\(O\\). A disk includes the enclosed region as well as the boundary. This distinction matters in questions of area, intersection, and convexity. The circumference of a circle is \\(2\\pi r\\), while the area of its disk is \\(\\pi r^2\\), where the constant \\pi is independent of the circle’s size.\n\nCircle geometry links distances with angles. An inscribed angle equals half the central angle subtending the same arc. In particular, an angle subtended by a diameter is a right angle. A quadrilateral whose vertices lie on one circle has supplementary opposite angles; conversely, this condition characterizes a cyclic convex quadrilateral. These results provide ways of detecting circles from angle relationships alone.\n\nA tangent meets a circle at exactly one point and is perpendicular to the radius drawn to that point. The two tangent segments from an external point have equal lengths. More generally, if a line through a point \\(P\\) meets a circle at \\(A\\) and \\(B\\), the product of the directed distances to the intersections is independent of the chosen secant. This common value is the power of \\(P\\):\n\n\\[\\operatorname{Pow}(P)=OP^2-r^2.\\]\n\nFor an external tangent point \\(T\\), it equals \\(PT^2\\). Directed lengths extend the relation consistently to points inside the circle.\n\nFor two nonconcentric circles, the points of equal power form a straight line called the radical axis. Subtracting their Cartesian equations cancels the quadratic terms and reveals why the locus is linear. If the circles intersect, their common chord lies on this axis. Three circles with pairwise nonparallel radical axes have a common radical center, giving a useful example of a concurrence theorem derived from an algebraic invariant.\n\nConic sections broaden the study of curved figures. An ellipse can be defined by a constant sum of distances to two foci, and a hyperbola by a constant absolute difference. A parabola consists of points equidistant from a focus and a directrix. These metric definitions connect to the curves obtained by intersecting a plane with a circular cone.\n\nIn suitable Cartesian coordinates, a nondegenerate ellipse and hyperbola have equations\n\n\\[\\frac{x^2}{a^2}+\\frac{y^2}{b^2}=1, \\qquad \\frac{x^2}{a^2}-\\frac{y^2}{b^2}=1,\\]\n\nwhile a parabola can be written \\(y^2=4px\\). Their study belongs simultaneously to Euclidean, analytic, and projective geometry. Euclidean questions emphasize foci, distances, normals, and reflection properties, whereas projective questions concern properties preserved under perspective transformations."
    },
    {
      "heading": "Straightedge-and-compass constructions",
      "text": "A straightedge-and-compass construction produces points using an ideal unmarked straightedge and an ideal compass. The permitted operations draw a line through two known points and a circle with a known center and an available radius. New points arise from intersections of lines and circles. The straightedge supplies incidence, not measurement; using marked lengths on a ruler changes the class of allowable constructions.\n\nElementary constructions include bisecting a segment, erecting a perpendicular, bisecting an angle, and copying an angle or length. An equilateral triangle can be constructed on a given segment by intersecting equal-radius circles centered at its endpoints. More elaborate tasks are assembled from these operations, with correctness established by congruence, equal radii, and properties of intersections rather than by the accuracy of a physical drawing.\n\nCoordinate algebra explains the limits of the method. A line has a linear equation, and a circle has a quadratic equation. Intersecting allowable pairs therefore requires arithmetic operations and extraction of square roots. Starting from a unit segment, constructible coordinates lie in fields obtained through finite towers of quadratic field extensions. Conversely, quantities obtainable through such towers can be constructed.\n\nThis framework resolves three famous ancient problems. Doubling a cube of unit volume requires a length \\(\\sqrt[3]{2}\\), whose minimal polynomial over the rational numbers has degree three, so it is not constructible. A general method for trisecting arbitrary angles is likewise impossible, although particular angles are trisectable. For example, trisecting \\(60^\\circ\\) would construct \\(20^\\circ\\), and the triple-angle formula yields an irreducible cubic obstruction.\n\nSquaring the circle would require constructing a square of area \\(\\pi\\) from a unit circle, hence a side of length \\(\\sqrt{\\pi}\\). Ferdinand von Lindemann’s 1882 proof that \\(\\pi\\) is transcendental excludes this possibility, since constructible numbers are algebraic. These impossibility results concern exact construction under specified rules; they do not prohibit approximation or constructions using other instruments.\n\nRegular polygons show that the same algebraic viewpoint yields positive results. In 1796, Carl Friedrich Gauss discovered the constructibility of the regular 17-gon. The Gauss–Wantzel criterion states that a regular \\(n\\)-gon is constructible precisely when\n\n\\[n=2^k p_1p_2\\cdots p_m,\\]\n\nwhere the \\(p_i\\) are distinct Fermat primes, each of the form \\(2^{2^j}+1\\). Thus the regular pentagon and 17-gon are constructible, while the regular heptagon is not."
    },
    {
      "heading": "Polygons, polyhedra, and higher-dimensional figures",
      "text": "A polygon is a closed plane figure bounded by a finite sequence of straight segments, subject to conventions about self-intersections and degeneracies. For a simple polygon with \\(n\\) vertices, the interior-angle sum is\n\n\\[(n-2)\\pi.\\]\n\nOne proof divides the polygon into \\(n-2\\) triangles using noncrossing diagonals. For concave polygons, the triangulation must be chosen carefully; drawing all diagonals from an arbitrary vertex need not keep them inside the polygon.\n\nA polygon is convex when the segment joining any two of its points remains within it. This is a special case of a convex set. A regular polygon has equal sides and equal interior angles. Its center, circumcircle, and incircle allow many measurements to be reduced to congruent right triangles. For circumradius \\(R\\), a regular \\(n\\)-gon has side length \\(2R\\sin(\\pi/n)\\).\n\nIn three dimensions, a polyhedron has polygonal faces meeting along edges and vertices. For the boundary of a convex polyhedron, the numbers of vertices \\(V\\), edges \\(E\\), and faces \\(F\\) satisfy Euler’s formula,\n\n\\[V-E+F=2.\\]\n\nThis relationship depends on the surface’s topological form rather than its specific edge lengths. Polyhedral surfaces with handles require a different right-hand side, revealing an early connection between geometry and topology.\n\nThere are exactly five convex regular polyhedra: the tetrahedron, cube, octahedron, dodecahedron, and icosahedron. If each face is a regular \\(p\\)-gon and \\(q\\) faces meet at each vertex, convexity requires\n\n\\[q\\frac{(p-2)\\pi}{p}<2\\pi,\\]\n\nor equivalently \\(1/p+1/q>1/2\\). With integers \\(p,q\\geq3\\), this leaves the five possible pairs corresponding to these solids. The angular condition narrows the possibilities; explicit constructions establish their existence.\n\nHigher-dimensional analogues are called polytopes. A simplex in \\(n\\) dimensions is the convex hull of \\(n+1\\) affinely independent points. A hypercube generalizes the square and cube, while the cross-polytope generalizes the octahedron. Their faces, incidences, symmetries, and metric properties can be studied without requiring physical visualization in the ambient dimension.\n\nPolyhedral geometry separates questions of shape from questions of combinatorial structure. Different polyhedra can have the same pattern of face adjacencies but different lengths and angles. Conversely, metric restrictions can force strong combinatorial conclusions. This interaction connects Euclidean geometry to combinatorics, optimization, and the study of discrete structures."
    },
    {
      "heading": "Area, volume, and limiting measurement",
      "text": "Area and volume assign numerical sizes to regions while respecting geometrical structure. Congruent regions have equal measure, and a region decomposed into finitely many nonoverlapping measurable pieces has measure equal to the sum of their measures. Choosing a unit square or unit cube fixes the scale. Classical formulas then follow from decompositions, comparisons, and limiting arguments rather than from units alone.\n\nFor a triangle, area is half the product of a base and its corresponding altitude. Parallelograms with the same base and height have equal areas even when their angles differ. A general polygon can be triangulated, so its area reduces to a finite sum of triangle areas. In coordinates, the signed area of a triangle with vertices \\((x_i,y_i)\\) is\n\n\\[\\frac12 \\det \\begin{pmatrix} x_1&y_1&1\\\\ x_2&y_2&1\\\\ x_3&y_3&1 \\end{pmatrix}.\\]\n\nThe absolute value gives ordinary area, while the sign records orientation.\n\nCurved boundaries require more than a single finite polygonal dissection. The ancient method of exhaustion compares a curved region with increasingly close polygonal approximations. Applied to a circle, inscribed and circumscribed polygons constrain its area and circumference. Archimedes used such methods to bound \\(\\pi\\), obtaining the celebrated inequalities \\(223/71<\\pi<22/7\\).\n\nThe corresponding solid formulas include\n\n\\[V_{\\text{prism}}=Bh, \\qquad V_{\\text{pyramid}}=\\frac13 Bh,\\]\n\nand, for a sphere,\n\n\\[V=\\frac43\\pi r^3, \\qquad S=4\\pi r^2.\\]\n\nCross-sectional comparison gives a systematic way to understand many volume identities. In modern terms, an integral accumulates cross-sectional areas, but ancient exhaustion arguments achieved some of the same exact results without the symbolic machinery of calculus.\n\nDimension controls scaling. Multiplying every length by \\(s>0\\) multiplies area by \\(s^2\\) and volume by \\(s^3\\). In \\(n\\)-dimensional Euclidean space, volume scales by \\(s^n\\). For a general invertible linear transformation represented by \\(A\\), \\(n\\)-dimensional volume is multiplied by \\(|\\det A|\\), connecting geometrical measurement to the determinant.\n\nFinite dissection has different powers in two and three dimensions. Polygons of equal area can be cut into finitely many polygonal pieces and reassembled into one another. Equal-volume polyhedra need not have this property. In 1900, Max Dehn introduced an additional invariant and used it to resolve Hilbert’s third problem: a cube and a regular tetrahedron of equal volume are not equidecomposable by finite polyhedral dissection. Volume is therefore not the only invariant governing such rearrangements."
    },
    {
      "heading": "Coordinates, vectors, and algebraic representation",
      "text": "Analytic geometry represents geometrical objects by equations. Its seventeenth-century development is particularly associated with René Descartes and Pierre de Fermat. Choosing an origin, perpendicular coordinate axes, and a unit assigns each point of the plane an ordered pair of real numbers. A spatial point similarly receives three coordinates. The coordinate system is additional descriptive machinery, not part of the figure itself.\n\nIn an orthonormal coordinate system, the Euclidean distance between \\(x,y\\in\\mathbb{R}^n\\) is\n\n\\[d(x,y)=\\sqrt{\\sum_{i=1}^{n}(x_i-y_i)^2}.\\]\n\nThe formula extends the Pythagorean theorem to any finite dimension. A circle has equation \\((x-a)^2+(y-b)^2=r^2\\), and a sphere has the corresponding three-coordinate equation. Straight lines and planes are described by linear equations or parametrizations.\n\nThe vector formulation distinguishes points from displacements. Differences of points are vectors, and a point plus a vector gives another point; adding two points has no origin-independent interpretation. A vector space with an inner product supplies the metric structure:\n\n\\[\\langle u,v\\rangle=\\sum_i u_iv_i, \\qquad \\|u\\|=\\sqrt{\\langle u,u\\rangle}.\\]\n\nFor nonzero vectors, their angle is determined by\n\n\\[\\cos\\theta= \\frac{\\langle u,v\\rangle}{\\|u\\|\\|v\\|}.\\]\n\nPerpendicularity is the condition \\(\\langle u,v\\rangle=0\\).\n\nThis formulation turns many constructions into algebra. A line through \\(p\\) in direction \\(v\\ne0\\) is \\(p+tv\\), with \\(t\\in\\mathbb{R}\\). A hyperplane with normal vector \\(n\\ne0\\) has equation \\(\\langle n,x\\rangle=c\\). The perpendicular projection of a point onto a line or subspace follows by requiring the residual displacement to be orthogonal to every allowed direction. This is the geometric basis of orthogonal projection and least-squares approximation.\n\nNot every coordinate grid is orthonormal. Relative to an oblique basis \\(e_1,\\ldots,e_n\\), the squared length of a vector with coordinate column \\(v\\) is\n\n\\[\\|v\\|^2=v^{\\mathsf T}Gv, \\qquad G_{ij}=\\langle e_i,e_j\\rangle.\\]\n\nThe Gram matrix \\(G\\) records the lengths and mutual angles of the basis vectors. A positive-definite matrix therefore represents a Euclidean inner product, even when its coordinate formula is not a sum of unweighted squares.\n\nAlgebraic and synthetic methods answer complementary needs. Coordinates make computation systematic and generalize easily to high dimensions. Synthetic arguments often reveal why a result is independent of coordinate choices, or expose a simple invariant hidden by a lengthy calculation."
    },
    {
      "heading": "Congruence, similarity, and transformation groups",
      "text": "A transformation-based view studies geometry through the mappings that preserve specified properties. An isometry preserves distances. Two figures are congruent when an isometry of the ambient Euclidean space maps one onto the other. In the plane, every isometry is a translation, rotation, reflection, or glide reflection, with the identity included as a special case.\n\nIn Cartesian coordinates, every Euclidean isometry has the form\n\n\\[x\\longmapsto Qx+b, \\qquad Q^{\\mathsf T}Q=I.\\]\n\nHere \\(Q\\) is an orthogonal matrix, and \\(b\\) is a translation vector. The orthogonality equation guarantees preservation of inner products of displacement vectors and hence of distances and angles. Matrices with determinant \\(1\\) preserve orientation; those with determinant \\(-1\\) reverse it.\n\nReflections are especially fundamental. In \\(n\\)-dimensional Euclidean vector space, every orthogonal transformation can be expressed as a product of at most \\(n\\) reflections in hyperplanes through the origin. In the plane, composing reflections in intersecting lines produces a rotation through twice the angle between the lines. Composing reflections in distinct parallel lines produces a translation. These constructions connect classical geometric operations with matrix multiplication.\n\nA similarity preserves shape while allowing uniform change of scale. Its coordinate form is\n\n\\[x\\longmapsto sQx+b, \\qquad s>0.\\]\n\nIt multiplies every distance by \\(s\\) and preserves angles and ratios of lengths. Congruence is the special case \\(s=1\\). Similarity differs from a general affine map, which preserves straightness, parallelism, and ratios along a line but need not preserve angles or ratios of lengths in different directions.\n\nTransformations preserving a figure form its symmetry group under composition. A regular \\(n\\)-gon has \\(n\\) rotational symmetries and \\(n\\) reflection symmetries. The symmetries of a cube include rotations about axes through opposite faces, opposite vertices, and opposite edges. Such examples provide concrete models for group theory and group actions.\n\nIn 1872, Felix Klein proposed organizing geometries by transformation groups, an approach known as the Erlangen program. Euclidean geometry concerns invariants under Euclidean motions, similarity geometry allows uniform scaling, affine geometry allows arbitrary invertible affine transformations, and projective geometry admits a still larger class. This hierarchy explains precisely which familiar properties survive a change of geometric viewpoint: incidence survives broadly, while metric length requires the more restrictive Euclidean group."
    },
    {
      "heading": "Modern foundations, continuity, and logical scope",
      "text": "Nineteenth-century developments exposed the need to separate intuitive pictures from explicit logical assumptions. In Foundations of Geometry, published in 1899, David Hilbert presented an influential axiomatization based on incidence, order, congruence, parallels, and continuity. Its purpose was not simply to rephrase Euclid but to identify dependencies among assumptions and establish which conclusions follow from which groups.\n\nContinuity is crucial in recovering the usual geometry over the real numbers. Incidence and metric axioms alone can admit coordinate systems over ordered fields smaller than \\(\\mathbb{R}\\), provided those fields possess the algebraic operations required by the axioms. Such models can satisfy many elementary geometrical statements while lacking some points expected in the real plane. Completeness assumptions rule out these gaps in sufficiently strong formulations.\n\nTwo notions of completeness must be distinguished. A complete ordered field has the least-upper-bound property, which characterizes the real numbers up to isomorphism. A complete logical theory, by contrast, decides every sentence in its specified language. These are different claims. Hilbert’s continuity and completeness principles concern the intended geometrical continuum, whereas later logical investigations examine what can be expressed and proved in formal languages.\n\nAlfred Tarski developed a particularly economical first-order approach to elementary Euclidean geometry using points, betweenness, and segment congruence. Through coordinates, statements in this framework can be related to polynomial equations and inequalities over real-closed fields. The first-order theory of real-closed fields admits quantifier elimination and is decidable. Consequently, the corresponding elementary geometrical questions can, in principle, be settled algorithmically.\n\nThis decidability has a restricted scope. It does not mean that every problem involving Euclidean space has an effective solution. Statements quantifying over arbitrary sets, arbitrary functions, or sufficiently rich arithmetic structures go beyond the elementary language. Even within a decidable theory, a general decision procedure may be computationally expensive and much less informative than a short geometrical proof.\n\nModels also permit relative consistency arguments. Interpreting points as real coordinate tuples and lines as suitable sets defined by linear equations shows that Euclidean axioms hold if the background theory of real numbers is accepted. Conversely, coordinatization theorems recover algebraic structure from geometric assumptions. This two-way connection makes Euclidean geometry an important example in model theory, illustrating how different mathematical presentations can describe the same structure while emphasizing different primitive concepts."
    },
    {
      "heading": "Non-Euclidean alternatives and curvature",
      "text": "For many centuries, mathematicians attempted to derive the parallel postulate from the other assumptions of geometry. Work by Giovanni Girolamo Saccheri, Johann Heinrich Lambert, and others instead revealed elaborate consequences of denying it. In the nineteenth century, Nikolai Lobachevsky and János Bolyai independently developed systematic hyperbolic geometry; Gauss also investigated related ideas, largely without publishing them.\n\nIn hyperbolic geometry, through a point outside a line there is more than one line that does not meet the given line. Triangles have angle sums less than \\(\\pi\\). For a hyperbolic plane of constant curvature \\(-1/R^2\\), a triangle of area \\(K\\) satisfies\n\n\\[K=R^2(\\pi-\\alpha-\\beta-\\gamma).\\]\n\nThe angle defect is thus a measure of area, unlike the Euclidean case, where angle sum is independent of size.\n\nSpherical geometry supplies a contrasting example. Taking great circles as geodesics on a sphere of radius \\(R\\), a suitable spherical triangle satisfies\n\n\\[K=R^2(\\alpha+\\beta+\\gamma-\\pi).\\]\n\nIts angle sum exceeds \\(\\pi\\). Great circles meet, so Euclidean parallelism fails. Spherical geometry also changes other global assumptions: antipodal points lie on more than one great circle, for example. It is therefore not obtained merely by replacing one sentence while leaving every familiar Euclidean incidence property intact.\n\nModels made the logical significance of these alternatives precise. Eugenio Beltrami’s 1868 work, followed by models associated with Klein and Henri Poincaré, represented hyperbolic geometry within familiar mathematical structures. In a disk model, hyperbolic “lines” may be represented by Euclidean chords or certain circular arcs, depending on the model. The hyperbolic distance differs from ordinary Euclidean distance, so visual appearance must not be confused with the modeled metric.\n\nBernhard Riemann’s 1854 lecture opened a broader study of spaces with variable metric structure. In Riemannian geometry, Euclidean space is the standard flat example. Its curvature vanishes, and its metric can be represented globally in Cartesian coordinates by\n\n\\[ds^2=dx_1^2+\\cdots+dx_n^2.\\]\n\nA general curved space resembles Euclidean space infinitesimally but need not share its global geometry.\n\nLocal flatness and global Euclidean structure are distinct. A flat torus has zero intrinsic curvature but has different topology and global distance properties from the Euclidean plane. In physical science, the geometry assigned to space or spacetime is an empirical and theoretical question. Euclidean geometry remains exact as a mathematical theory even when another geometry better models a particular physical setting."
    },
    {
      "heading": "Applications, computation, and geometric reasoning",
      "text": "Euclidean geometry supplies a working language for measurement and design. Surveying uses triangles, angle observations, and distance measurements to determine positions. Architecture and engineering use perpendicularity, parallelism, similarity, and solid geometry to specify shapes and dimensions. Scale drawings preserve angles and length ratios, while area and volume scaling must be handled separately.\n\nThe approximation involved in surveying depends on spatial scale. Small regions of the Earth can often be treated as planar to a specified tolerance, whereas large-scale geodesy must account for curvature. For a sphere of radius \\(R\\), the departure of a short arc from a tangent-plane approximation grows approximately quadratically with horizontal distance. Thus the appropriateness of a Euclidean model is not an all-or-nothing judgment but a quantitative matter of error.\n\nIn classical mechanics, positions and displacements are usually represented in three-dimensional Euclidean space. Dot products express work and orthogonal components; rotations describe rigid-body orientation; moments of inertia depend on distances from axes. Robotics uses related structures to model linked rigid bodies, compute poses, and transform coordinates between reference frames. The combination of rotations and translations is especially important because a pose contains both orientation and position.\n\nComputer graphics uses Euclidean coordinates for objects, surface normals, distances, and lighting calculations, while perspective projection introduces projective methods. Computer-aided design represents curves and surfaces whose intersections, tangencies, and clearances must be computed. Euclidean constraints such as equal radii or perpendicular edges can become systems of algebraic equations, connecting geometric design to numerical methods and constraint solving.\n\nComputational geometry develops algorithms for geometric data. A Voronoi diagram partitions space into regions nearest to specified sites. Its associated Delaunay triangulation connects sites according to an empty-circle condition in the plane. These structures support interpolation, mesh generation, spatial search, and facility-location models. Euclidean distance also underlies nearest-neighbor methods and many forms of clustering, although changing units or coordinate scaling changes the resulting distances.\n\nExact mathematical predicates require care in computation. A tiny numerical error can change whether three points are classified as collinear or whether a point lies inside a circumcircle, potentially altering an entire geometric data structure. Floating-point arithmetic, error bounds, adaptive precision, and exact predicates therefore play a central role in robust implementations. The underlying geometry may be simple, while reliable computation of its incidence and ordering relations is not.\n\nIn education, Euclidean geometry provides a setting for learning deductive reasoning through explicit hypotheses, constructions, and proofs. Dynamic geometry software can display patterns and suggest conjectures, but a moving diagram does not establish a universal statement. The distinction between experimental evidence and proof remains essential when a configuration approaches a degenerate case, such as coincident points, parallel limiting lines, or a triangle whose area tends to zero."
    }
  ],
  "links": [
    "geometry",
    "euclidean-space",
    "euclid",
    "euclids-elements",
    "mathematics",
    "mathematical-proof",
    "pythagoras",
    "archimedes",
    "conic-section",
    "calculus",
    "circle",
    "axiom",
    "theorem",
    "affine-space",
    "parallel-postulate",
    "triangle-inequality",
    "pythagorean-theorem",
    "trigonometry",
    "pi",
    "straightedge-and-compass-construction",
    "field-extension",
    "carl-friedrich-gauss",
    "polygon",
    "convex-set",
    "polyhedron",
    "topology",
    "convex-hull",
    "combinatorics",
    "method-of-exhaustion",
    "integral",
    "determinant",
    "analytic-geometry",
    "pierre-de-fermat",
    "euclidean-distance",
    "vector-space",
    "inner-product",
    "hyperplane",
    "orthogonal-projection",
    "gram-matrix",
    "isometry",
    "orthogonal-matrix",
    "affine-map",
    "group-theory",
    "group-action",
    "felix-klein",
    "projective-geometry",
    "david-hilbert",
    "real-number",
    "first-order-logic",
    "model-theory",
    "bernhard-riemann",
    "architecture",
    "classical-mechanics",
    "voronoi-diagram",
    "delaunay-triangulation",
    "floating-point-arithmetic",
    "education"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}