0 70 5 . 41 41 v 3 [ he p - th ] 2 9 N ov 2 00 7 Good Babies vs . Bad Babies or Inheriting The Arrow of Time

The actual construction of the string Landscape usually relies on nucleation of “baby universes”. One should however ask whether such babies can resemble the Universe we observe. For they may not have an Arrow of time: even if some of them have basic physical laws essentially identical to those we have discovered, they may have the wrong — insufficiently smooth — initial conditions. We argue that all baby universes will be “bad babies” in this sense, unless the Null Energy Condition is violated when they are born. Recent work of Arkani-Hamed et al suggests that NEC violation is sometimes acceptable in string theory, and we show that, in these circumstances, requiring stability against non-perturbative string effects may permit the birth of “good babies”. The latter are [by definition] able to inherit the Arrow which, as we have argued elsewhere, is established when the original universe is “created from nothing”, in accordance with the Ooguri-VafaVerlinde “entropic principle”. 1. Initial Size Does Matter The most promising approach to the problem of the cosmological constant is given by the string theory Landscape [1]. The universes in the Landscape are so numerous — 10 is the standard estimate — and have values of the cosmological constant spaced in such a way, that the value we actually observe ceases to seem surprising; instead we conclude that our Universe is a member of the Landscape. The determination of the cosmological constant works in this relatively straightforward way because all fluxes in the “internal” space contribute to its value. For other quantities [such as the values of couplings in the Standard Model Lagrangian], this is not the case, and this may greatly reduce the number of Landscape universes which resemble our Universe. Such reductions imply that it may be possible to construct a predictive theory; see for example [2]. In fact, there is a danger that the Landscape may be too small to accommodate the observations. There is another sense in which the Landscape runs the risk of being [far] too small: its size certainly cannot account for the extreme “specialness” of the initial conditions of our Universe. For this specialness [or “non-genericity”], manifested as an Arrow of time [3][4][5][6][7], is on a scale — measured by the Penrose number