Big Bang Models in String Theory
Ben Craps
TL;DR
These proceedings survey how general relativity's spacetime singularities can be addressed within string theory, focusing on static orbifold/ALE singularities and cosmological singularities such as the Milne orbifold and the matrix big bang. The analysis combines perturbative string methods (twisted strings, D-branes, and tachyon effects) with non-perturbative tools (AdS/CFT, matrix theory, and DLCQ) to understand whether singularities can be resolved or whether spacetime itself can emerge from underlying degrees of freedom. A key finding is that Milne-type cosmologies exhibit divergences in tree-level amplitudes signaling breakdown of perturbation theory, while non-perturbative mechanisms, including winding condensates, tachyon phases, and holographic descriptions, offer potential resolutions. The matrix big bang proposal provides a concrete non-perturbative description in which late-time dynamics imply emergent spacetime as off-diagonal modes decouple, highlighting a pathway from singular early-times to classical spacetime and raising important questions about higher-loop effects, observables, and large-$N$ limits in cosmological string theory $g_s$-dependent settings.
Abstract
These proceedings are based on lectures delivered at the "RTN Winter School on Strings, Supergravity and Gauge Theories", CERN, January 16 - January 20, 2006. The school was mainly aimed at Ph.D. students and young postdocs. The lectures start with a brief introduction to spacetime singularities and the string theory resolution of certain static singularities. Then they discuss attempts to resolve cosmological singularities in string theory, mainly focusing on two specific examples: the Milne orbifold and the matrix big bang.
