Scribd
Upload
53 views

Was There A Beginning

The document discusses arguments for and against the universe having a beginning. It summarizes Leonard Susskind's view that for all practical purposes, the universe was past-eternal. Examples are given of different cosmological models and it is argued that in any inflating model, the odds strongly favor the beginning being so far in the past it is effectively infinite. While a beginning may exist, the probability is unity that it was more than any finite amount of time ago.

Uploaded by

LillyOpenMind
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
53 views

Was There A Beginning

The document discusses arguments for and against the universe having a beginning. It summarizes Leonard Susskind's view that for all practical purposes, the universe was past-eternal. Examples are given of different cosmological models and it is argued that in any inflating model, the odds strongly favor the beginning being so far in the past it is effectively infinite. While a beginning may exist, the probability is unity that it was more than any finite amount of time ago.

Uploaded by

LillyOpenMind
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Was There a Beginning?

arXiv:1204.5385v1 [hep-th] 24 Apr 2012

Leonard Susskind

Stanford Institute for Theoretical Physics and Department of Physics, Stanford


University
Stanford, CA 94305-4060, USA

Abstract
In this note I respond to Mithani and Vilenkin’s claim that there must have been
a beginning.
Mithani and Vilenkin have argued that the universe must have had a beginning [1]. I
will argue the opposite point of view; namely, for all practical purposes, the universe was
past-eternal.
To make the point simply, imagine Hilbertville, a one-dimensional semi-infinite city,
whose border is at x = 0. The population is infinite and uniformly fills the positive axis
x > 0. Each citizen has an identical telescope with a finite power. Each wants to know
if there is a boundary to the city. It is obvious that only a finite number of citizens can
see the boundary at x = 0. For the infinite majority the city might just as well extend to
the infinite negative axis. Thus, assuming he is typical, a citizen who has not yet studied
the situation should bet with great confidence that he cannot detect a boundary. This
conclusion is independent of the power of the telescopes as long as it is finite.
Now let us consider some of the examples in Mithani and Vilenkin’s note. First de
Sitter space. For generality we can consider a landscape of positive cosmological constant
vacua with a bounded vacuum energy greater than zero. We may assume an initial starting
point, for example an initial vacuum on the landscape. As explained in [2] and [3] if we
follow a causal patch it will pass through an infinite number of Boltzmann fluctuations
that will sample all vacua. Most observer’s will be freaks but even if we condition on
normal observers, there will be an infinite number of them. Given any finite time T, all
but a finite number of them will occur later than that time. By later I mean after the
initial starting point. Thus if they think they are typical, the citizens should bet—again
with overwhelming confidence—that they cannot detect a beginning.
Next consider the so-called cyclic universe—eternal to the past and to the future.
Obviously, if such a thing could exist it would have no beginning, but it cannot exist; not
without violating the second law of thermodynamics.
To avoid a heat-death the proponents of a cyclic universe assume that with each cycle
the universe expands by a linear factor L, thereby diluting the entropy [4]. Thus the cyclic
universe expands exponentially. Averaging over a cycle, the cyclic universe is just de Sitter
space in flat-slicing. This is show in Figure 1.
Now consider an observer at r = 0. Such an observer is obviously surrounded by an
event-horizon as in Figure 2. From his point of view the universe is a bounded region with
a finite entropy bound. In the causal patch the entropy does not dilute. It is evidently
a perpetual motion machine that violates the second law. Thermodynamically it must
run down in a finite time. Thereafter it will just be the causal patch of de Sitter space.
Periodicity must give way to the recurrent behavior of the previous case.
Finally let us consider eternal inflation with terminal vacua. Although nothing I say

1
Figure 1: Penrose Diagram of an expanding cyclic universe. The blue curved lines represent
bounces.

will depend on it, we can visualize the multiverse using the tree-model of [2]. There is a
paradox in this case which I explained in [3]. We assume a landscape with eN de Sitter
vacua. Suppose we follow a causal patch from the root of the tree. It starts with vacuum
type n and after about N transitions it will enter a terminal vacuum. The probability for
surviving more than N transitions goes to zero. Thus an observer should bet that he is
within N transitions of the root. With a sufficiently powerful telescope he should be able
to see the beginning.
On the other hand consider the global view. Even though any given causal patch dies in
a terminal, the number of causal patches increases exponentially with time. Therefore the
vast majority of observers are very high up on the tree. At late times the statistics of the
vacuum-types is governed by an attractor called a fractal flow [2] [3] which is completely

2
Figure 2: Causal patch in a cyclic universe. The green area is the region covered by the
analog of static coordinates in de Sitter space. The observer sees perpetual motion.

insensitive to the initial condition. If a multiversal citizen knows these things he will bet
that he is too late to detect any evidence of the root.
Combing the Mithani-Vilenkin’s observations [1] with the ones in this note, we may
conclude that there is a beginning, but in any kind of inflating cosmology the odds strongly
(infinitely) favor the beginning to be so far in the past that it is effectively at minus infinity.
More precisely, given any T the probability is unity that the beginning was more than T
time-units ago.

3
References
[1] A. Mithani and A. Vilenkin, “Did the universe have a beginning?,” arXiv:1204.4658
[hep-th].

[2] D. Harlow, S. H. Shenker, D. Stanford and L. Susskind, “Tree-like structure of eternal


inflation: A solvable model,” Phys. Rev. D 85, 063516 (2012) [arXiv:1110.0496 [hep-
th]].

[3] L. Susskind, “Fractal-Flows and Time’s Arrow,” arXiv:1203.6440 [hep-th].

[4] P. J. Steinhardt and N. Turok, “A cyclic model of the universe,” Science 296, 1436
(2002).

You might also like