← all systems · see it in The System
Clifton-Pohl torus
'In geometry, the Clifton-Pohl torus is an example of a compact Lorentzian manifold that is not geodesically complete. While every compact Riemannian manifold is also geodesically complete (by the Hopf-Rinow theorem), this space shows that the same implication does not generalize to pseudo-Riemannian manifolds.' Construction: M = R^2 \ {0} with g = 2 dx dy/(x^2 + y^2); the group generated by lambda(x, y) = (2x, 2y) acts properly discontinuously, and 'the quotient T = M/Gamma, which is topologically the torus, is a Lorentz surface'.
Properties published scienceour proposal
Sourced fields, and our own readings.
| boundary | Inside: the quotient of the punctured plane R^2 \ {0} with the Lorentz metric 2 dx dy/(x^2 + y^2) by the group generated by the homothety (x, y) -> (2x, 2y): a compact Lorentz surface homeomorphic to a torus, with its geodesics (null geodesics all incomplete), its Killing field and its lack of conjugate points. Outside: the extended Clifton-Pohl plane into which the covering plane extends, the deformed metrics, and any physical spacetime (none is named by the sources). |
| nested level | Rung 11 (math only): a structure held by science, walked as a projected model. Within mathematics it is a kind of Lorentzian (pseudo-Riemannian) manifold of dimension 2, the standard counterexample to compact-implies-complete outside Riemannian geometry. |
| representation | True |
| scale | field |
Measures (9) published science
| homothety ratio of the generating isometry | 2 ((x, y) -> (2x, 2y)) ratio |
| fundamental domain | closed annulus 1 <= rho <= 2, boundaries identified radius |
| parameter at which the null geodesic (1/(1-t), 0) ends | t = 1 affine parameter |
| completeness condition on the 'impulse' of a non-null geodesic (weaker notion, 2002) | 0 < P(alpha, beta, x, y) <= 2 dimensionless |
| year described by Clifton and Pohl (unpublished) | 1962 year |
| Hopf's theorem (Riemannian tori without conjugate points are flat), and in any dimension Burago and Ivanov | 1948; 1994 year |
| dimension of the Mounoud family of projective deformations without conjugate points | 2 dimension |
| topologies of compact connected Lorentzian surfaces | 2 (torus or Klein bottle) count |
| characteristic clock: none (no physical reading); geodesic parameter runs out in finite value | none time |
Cross-scale patterns seen here (1) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Clifton-Pohl torus. Each links to every other system that shows it.
- same toolkit, independent builds (partial)
Two Lorentzian counterexamples built the same way, by dividing a two-dimensional Lorentzian surface by a discrete group of its isometries: Misner space by a boost (giving closed null curves and a non-curvature singularity), the Clifton-Pohl torus by a homothety (giving a compact surface with incompl
Valence: what it trades, and with whom (4) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Four-dimensional Lorentzian manifold (events and world lines) via sys:spacetime · Clifton-Pohl torus gain
- Riemannian manifold (positive-definite geometry; space at one time) via sys:spacetime · Clifton-Pohl torus even
- Pseudo-Riemannian (semi-Riemannian) manifold via sys:spacetime · Clifton-Pohl torus gain
- Geodesic via sys:spacetime · Clifton-Pohl torus gain
Partners and parts published science
Built from / builds (2)
- Punctured plane R^2 \ {0} (the covering surface M)
- Group generated by the homothety lambda(x, y) = (2x, 2y)
Sits inside / holds (3)
- Science (as a human social and cultural system)
- Pseudo-Riemannian (semi-Riemannian) manifold
- Extended Clifton-Pohl plane (its covering plane extends into it)
floor filled by (11)
- Pseudo-Riemannian (semi-Riemannian) manifold
- Pseudo-Riemannian metric
- Four-dimensional Lorentzian manifold (events and world lines)
- Science (as a human social and cultural system)
- Gravitational singularity
- Geodesic
- Punctured plane R^2 \ {0} (the covering surface M)
- Group generated by the homothety lambda(x, y) = (2x, 2y)
- Projective deformations of the Clifton-Pohl torus (Mounoud)
- Compact Lorentzian Klein bottle
- Extended Clifton-Pohl plane (its covering plane extends into it)
represents (1)
Kinds (4)
- Four-dimensional Lorentzian manifold (events and world lines)
- Pseudo-Riemannian (semi-Riemannian) manifold
- Clifton-Pohl tori in the extended sense
- Projective deformations of the Clifton-Pohl torus (Mounoud)
wired (1)
Notes published science
- The Clifton-Pohl torus is compact yet has geodesics that run out in finite parameter, something Hopf-Rinow forbids once the metric is positive-definite: relaxing one sign breaks a whole theorem.
- A finite, closed surface you can fall off: the Clifton-Pohl torus is compact, has no edge and no curvature blow-up, yet a light ray on it runs out of parameter at t = 1.
- It is incomplete only because it is a piece of something bigger: change coordinates with tan and the covering plane sits inside the 'extended Clifton-Pohl plane', which is complete.
- Clifton and Pohl found it in 1962 and never published it; it lives on as a textbook counterexample.
- Drop one positivity requirement and a compact space can still be incomplete: the Clifton-Pohl torus is something Riemannian geometry forbids outright.
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
Clifton-Pohl torus (own: compact Lorentz surface; projected model)
- L1: The punctured plane M = R^2 \ {0} with coordinates (x, y) (Meneghini writes (u, v))
- L2: The Lorentz metric g = 2 dx dy/(x^2 + y^2) = (dx (x) dy + dy (x) dx)/(x^2 + y^2); the coordinate axes are null
- L3: Folding: 'Any homothety is an isometry of M'; Gamma generated by (x, y) -> (2x, 2y) acts properly discontinuously; T = M/Gamma is the closed annulus 1 <= rho <= 2 with boundaries identified, a torus, compact
- L4: Geodesics: equations u'' = 2u u'^2/(u^2 + v^2), v'' = 2v v'^2/(u^2 + v^2), first integrals u'v' = A(u^2 + v^2) and u/u' + v/v' = B; sigma(t) = (1/(1-t), 0) is a null geodesic 'not defined at t = 1'; 'every null geodesic on M or T is incomplete'
- L5: Completeness verdicts: geodesically incomplete although compact; incompleteness 'better seen as a direct consequence of the fact that (M, g) is extendable' to the complete extended Clifton-Pohl plane; under complexification every geodesic is complex-complete (Meneghini 2004); in the 2002 weaker notion a non-null geodesic is complete iff 0 < P(alpha, beta, x, y) <= 2
- L6: No conjugate points: 'the first known non-flat Lorentzian tori with no conjugate points' (Bavard and Mounoud); a Killing field with a null orbit; deformations keep the property, 'as stable as possible' (Mehidi); the extended plane 'contains a lot of pairs of conjugate points'
held by science (a counterexample in Lorentzian geometry)
- L1: Described by Yeaton H. Clifton and William F. Pohl in 1962, 'but did not publish their result'; carried in O'Neill's Semi-Riemannian Geometry (1983, example 7.16, p. 193)
- L2: Use as the counterexample: 'this space shows that the same implication does not generalize to pseudo-Riemannian manifolds'; 'a counterexample to show that it is possible to have a geodesically incomplete compact semi-Riemannian manifold' (spacetime catalogue)
- L3: Refinements: a 'weaker definition of geodesic completeness' by analytic continuation (Meneghini 2002) and complex completeness of all geodesics (2004); no conjugate points (Bavard and Mounoud); infinitely many new Lorentzian tori and Klein bottles without conjugate points (Mehidi 2019)
- L4: Brought into the discussion of spacetime singularities as a contrast: an incomplete null geodesic in a compact torus, which is not singular in any physical sense; whether the singularities of the singularity theorems differ in nature from it is little known (Encyclopedia of Mathematics, read only in summary)